23 – 26 April, 2023 Hybrid Meeting

Copyright: Modified and published with permission from https://www.ebmt.org/annual-meeting

Sponsorship Statement: Publication of this supplement is sponsored by the European Society for Blood and Marrow Transplantation. All content was reviewed and approved by the EBMT Committee, which held full responsibility for the abstract selections.

19 - Acute Leukaemia

P001 CORD BLOOD TRANSPLANTATION FOR AML: COMPARABLE LFS IN PATIENTS WITH DE NOVO VERSUS SECONDARY AML

Frederic Baron 1, Arnon Nagler2, Jacques-Emmanuel Galimard3, Jaime Sanz4, A.E.C. Broers5, Edouard Forcade6, Patrice Chevalier7, Anne Sirvent8, Emma Nicholson9, Jürgen Kuball10, Sabine Furst11, Alessandro Rambaldi12, Jorge Sierra13, Peter A. von dem Borne14, Maria Pilar Gallego Hernanz15, Thomas Cluzeau16, Stephen Robinson17, Anna Maria Raiola18, Hélène Labussière-Wallet19, Jenny L. Byrne20, Jean-Valère Malfuson21, Annalisa Ruggeri22, Mohamad Mohty23, Fabio Ciceri22

1University of Liege, Liege, Belgium, 2The Chaim Sheba Medical Center, Tel Hashomer, Israel, 3EBMT Paris Office, Paris, France, 4University Hospital La Fe, Valancia, Spain, 5Erasmus MC Cancer Institute, Rotterdam, Netherlands, 6CHU Bordeaux, Pessac, France, 7CHU Nantes, Nantes, France, 8CHU Montpellie1,0r, Montpellier, France, 9Royal Marsden Hospital, London, United Kingdom, 10University Medical Centre, Dept. of Haematology – Utrecht, Utrecht, Netherlands, 11Institut Paoli Calmettes, Marseille, France, 12ASST Papa Giovanni XXIII, Bergamo, Italy, 13Hospital Santa Creu i Sant Pau, Barcelona, Spain, 14Leiden University, Leiden, Netherlands, 15Hopital La Miletrie, Poitiers, France, 16CHU Nice, Nice, France, 17Bristol Royal Hospital for Children, Bristol, United Kingdom, 18IRCCS Ospedale San Martino, Genova, Italy, 19CHU Lyon, Lyon, France, 20Nottingham University, Nottingham, United Kingdom, 21Hôpital D’Instruction des Armées (HIA) PERCY, Clamart, France, 22Ospedale San Raffaele s.r.l., Milano, Italy, 23Hôpital Saint-Antoine, Sorbonne University, INSERM UMRs 938, Paris, France

Background: We previously reported inferior leukemia-free survival (LFS) following allogeneic hematopoietic stem cell transplantation (allo-HCT) for secondary versus de novo acute myeloid leukemia (AML) in first complete remission (CR1) among patients given grafts from either HLA-identical sibling, unrelated donor or HLA-haploidentical donor1. Given that unrelated cord blood transplantation (CBT) has been associated with high graft-versus-leukemia effects2 and with good transplantation outcomes in secondary AML (sAML)3, we investigated whether secondary versus de novo AML would be also a risk factor for poor outcomes in adult AML patients in CR1 receiving unrelated CBT.

Methods: This is a retrospective study from the acute leukemia working party (ALWP) of the EBMT. Inclusion criteria included adult at first allo-HCT between 2000 and 2021, unrelated single or double unit CBT, AML in CR1, no ex vivo T-cell depletion and no post-transplant cyclophosphamide. The primary endpoint of the study was leukemia-free survival (LFS). Multivariate Cox models were adjusted for cytogenetic risk, single versus double CBT, anti-thymocte globuline (ATG) or not, TBI or not, myeloablative versus reduced-intensity (RIC) conditioning, age at CBT, time from diagnosis to transplantation and year of transplantation.

Results: A total of 879 patients with de novo (n = 696) or secondary (n = 183) AML met the inclusion criteria. In comparison with de novo AML patients, those with sAML were older (55 versus 47 years old, P < 0.001), were transplanted sooner after diagnosis (median of 5 months versus 5.6 months, P = 0.003), received more frequently double CBT (62 versus 47%, P < 0.001) and were transplanted more frequently following RIC (63% versus 50%, P = 0.002). Two-year LFS in de novo versus secondary AML patients were 48 (95% CI: 44-52%) and 44 (95% CI: 36-51%), respectively. In multivariable analyses, sAML patients had non significantly different LFS (HR = 0.98, P = 0.86), overall survival (HR = 1.07, P = 0.58), relapse incidence (HR = 0.74, P = 0.09) and incidence of nonrelapse mortality (HR = 1.26, P = 0.13) than those with de novo AML. Factors associated with worse LFS in multivariate analysis included poor risk cytogenetic (HR = 1.43, P = 0.002), the use of ATG(HR = 1.5, P = 0.007), and older age at transplantation (HR per ten-year increment = 1.11, P = 0.008).

Conclusions: Our results demonstrate comparable LFS following CBT in adult patients with secondary versus de novo AML. A potential negative impact on LFS of ATG emerged in this specific disease context.

Clinical Trial Registry: References:

1. Schmaelter A-K, Labopin M, Socié G, Itälä-Remes M, Blaise D, Yakoub-Agha I et al. Inferior outcome of allogeneic stem cell transplantation for secondary acute myeloid leukemia in first complete remission as compared to de novo acute myeloid leukemia. Blood Cancer J 2020; 10: 26.

2. Milano F, Gooley T, Wood B, Woolfrey A, Flowers ME, Doney K et al. Cord-Blood Transplantation in Patients with Minimal Residual Disease. N Engl J Med 2016; 375: 944–953.

3. Baron F, Labopin M, Ruggeri A, Volt F, Mohty M, Blaise D et al. Cord blood transplantation is associated with good outcomes in secondary Acute Myeloid Leukaemia in first remission. J Intern Med 2019; 285: 446–454.

*FB and AN contributed equally to this abstract

Disclosure: No COI to report

19 - Acute Leukaemia

P002 SECOND ALLOGENEIC STEM CELL TRANSPLANTATION (HSCT2) FOR PRIMARY GRAFT FAILURE IN PATIENTS WITH ACUTE LEUKEMIA IN REMISSION: ON BEHALF OF THE ALWP OF THE EBMT

Arnon Nagler 1, Myriam Labopin2, Aleksandr Kulagin3, Andrea Velardi4, Jaime Sanz5, Hélène Labussière-Wallet6, Victoria Potter7, Jürgen Kuball8, Simona Sica9, Elena Parovichnikova10, Wolfgang Bethge11, Xavier Leleu12, Uwe Platzbecker13, Friedrich Stölzel14, Fabio Ciceri15, Mohamad Mohty2

1Sheba MC, Ramat-Gan, Israel, 2Hôpital Saint-Antoine, Paris, France, 3First State Pavlov Medical University of St. Petersburg, Raisa Gorbacheva Memorial Research Institute for Paediatric Oncology, Hematology, and Transplantation, St Petersburg, Russian Federation, 4Università di Perugia, Ospedale Santa Maria della, Perugia, Italy, 5University Hospital La Fe, Valencia, Spain, 6Centre Hospitalier Lyon Sud, Lyon, France, 7Kings College Hospital NHS Foundation Trust, London, United Kingdom, 8University Medical Centre, Utrecht, Netherlands, 9Universita Cattolica S. Cuore, Rome, Italy, 10National Research Center for Hematology, Moscow, Russian Federation, 11Universitaet Tuebingen, Medizinische Klinik, Tuebingen, Germany, 12Hopital La Miletrie, Poitiers, France, 13Medical Clinic and Policinic 1, Hematology and Cellular Therapy, University hospital Leipzig, Leipzig, Germany, 14Universitaetsklinikum Dresden, Medizinische Klinik und Poliklinik I, Dresden, Germany, 15Ospedale San Raffaele s.r.l. Haematology and BMT, Milano, Italy

Background: Primary graft failure (GF) is a devastating complication of allogeneic transplantation (HSCT). Second transplantation (HSCT2) is a potential treatment.

Methods: We assessed the outcome of adult patients (pts) with acute leukemia failing to engraft (ANC < 0.5 x 109/L) ≥ day 28 after the first transplant (HSCT1) and received a second transplant between 2000 and 2021.

Results: 243 pts were included: 182 (74.9%) with AML and 61(25.1%) with ALL. Median age was 44.8 (range, 18.4-75.2) years, 71.6% and 28.4% of pts were in CR1 and CR2, respectively. Donors of HSCT1 were matched siblings in 9.1%, unrelated in 39.9%, haploidentical in 35%, and cord blood in 16%. At HSCT1, conditioning was myeloablative (MAC) in 58.4% of the cases. GVHD prophylaxis was mainly calcineurin inhibitor (CNI)-based (70.9%). 46.1% and 14.8% of the HSCTs were with in vivo and ex vivo T-cell depletion, respectively. The time from first to second transplantation was 48 (range, 28-120) days. Donors at HSCT2 were the same as in HSCT1 in 49% while they differ in 51%. Engraftment post HSCT2 was achieved by 78.6% vs. 59% of the pts with AML and ALL, respectively (p = 0.034.) Incidence of acute (a) GVHD II-IV and III-IV was 23.2% and 8.2%, and 5-year total and extensive chronic (c) GVHD was 22.3%, and 10.1%, respectively. The 5-year NRM was 51.6% with infections being the main cause. The 5 -year RI, LFS, OS and GRFS were 18.8%, 29.6%, 30.7% and 22.4%, respectively. Being transplanted at CR2 vs. CR1, increased age (per 10-years), lower KPS ( < 90) at HSCT2 and receiving MAC at HSCT1 were adverse prognostic factors for NRM, LFS and OS. Hazard ratio (HRs) for NRM were 2.26 (95% CI 1.41-3.63, p = 0.0007), HR = 1.31 (95% CI 1.09-1.54, p = 0.005), HR = 2.17 (95% CI 1.30-3.70, p = 0.003) and HR = 2.63 (95% CI 1.52-4.55, p = 0.0006), respectively. HRs for LFS were 2.21 (95% CI 1.43-3.33, p = 0.0002), HR = 1.2 (95% CI 1.03-1.41, p = 0.02), HR = 1.92 (95% CI 1.23-3.03, p = 0.004) and HR = 2.0 (95% CI 1.27-3.13, p = 0.003), respectively. HRs for OS were 2.38 (95% CI 1.57-3.6, p < 0.0001), HR = 1.26 (95% CI 1.07-1.48, p = 0.005), HR = 2.04 (95% CI 1.30-3.23, p = 0.002) and HR = 2.38 (95% CI 1.47-3.85, p = 0.0003), respectively. Being transplanted at CR2 vs CR1, lower KPS ( < 90) and receiving MAC at HSCT1 were adverse prognostic factors for GRFS with HR of 1.82 (95% CI 1.23-2.71, p = 0.003), HR = 1.49 (95% CI 1-2.22, p = 0.047), and HR = 1.59 (95% CI 1.02-2.5, p = 0.039), respectively. In AML patients, adverse cytogenetics was an adverse prognostic factor for RI, LFS, OS, GRFS, and aGVHD II-IV with a HR of 8.22 (95% CI 3.01-22.46, p < 0.0001), HR = 2.48 (95% CI 1.43-4.31, p = 0.001), HR = 2.06 (95% CI 1.16-3.65, p = 0.014), HR = 2.34 (95% CI 1.38-3.97, p = 0.002) and HR = 3.32 (95% CI 1.36-8.13, p = 0.008), respectively.

Conclusions: HSCT2 can rescue about a third of AL pts with GF. However, transplant-related mortality is very high with a 5–year mortality of about 50%. The outcomes of HSCT2 for GF in AML and ALL are similar. A different donor at HSCT2 seems not to improve results. GF remains a grave complication of transplantation that awaits novel ideas and strategies.

Disclosure: Nothing to declare

19 - Acute Leukaemia

P003 AUTOLOGOUS VERSUS HAPLOIDENTICAL DONOR STEM CELL TRANSPLANTATION IN ADULT PATIENTS WITH AML IN FIRST COMPLETE REMISSION WITH UNDETECTABLE MRD: A GLOBAL REAL WORLD STUDY

Jia Chen 1, Myriam Labopin2, Michael Daskalakis3, Xi Zhang4, Erlie Jiang5, Alessandra Tucci6, A.E.C Broers7, Ellen Meijer8, Emmanuelle Polge2, Irma Khevelidze2, Depei Wu1, Mohamad Mohty2, Norbert-Claude Gorin2

1National Clinical Research Center for Hematologic Diseases, the First Affiliated Hospital of Soochow University, Suzhou, China, 2Saint-Antoine Hospital, AP-HP, Sorbonne University, Paris, France, 3University Hospital Bern, Bern, Switzerland, 4Xinqiao Hospital of Army Medical University, Chongqing, China, 5Institute of Hematology, Chinese Academy of Medical Sciences, Tianjin, China, 6Spedali Civili – Brescia, Brescia, Italy, 7Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, Netherlands, 8VU University Medical Center, Amsterdam, Netherlands

Background: For intermediate-risk AML patients in CR1, the optimal choice of consolidation therapy is still questioned. Recent Existing data have demonstrated the value of autologous stem cell transplantation (ASCT) in CR1 patients with undetectable MRD (uMRD). However allogeneic stem cell transplantation is the standard approach for the majority of transplant teams, including the use of haploidentical donor transplants (HAPLO) in the absence of a matched donor. There are presently no comparative data on transplant outcomes following a HAPLO or an ASCT for AML patients in CR1 with uMRD.

Methods: The retrospective study was conducted by the Global Committee and the ALWP of the EBMT. All patients reported to the EBMT registry, older than 18 years of age, diagnosed as intermediate-risk AML according to the ELN 2010 classification, and underwent a first auto or haplo-HCT while in CR1 with uMRD in the period of 2010-2021 were included. The endpoints were overall survival (OS), leukemia-free survival (LFS), relapse incidence (RI) and Non relapse Mortality (NRM). A particular attention was devoted to molecular markers to enable detailed stratification.

Results: 1122 patients from 196 centers were recruited, including 547 patients who received a ASCT and 575 patients receiving a HAPLO (Table 1). When compared to HAPLO, ASCT patients were older and transplanted earlier. ASCT patients less frequently carried a FLT3-ITD mutation and more frequently a NPM1 mutation. Besides, they had inconspicuously shorter interval from diagnosis and better performance status at transplant.

Table 1 Comparison of characteristics of enrolled patients

  

Auto-HCT (n = 547)

Haplo-HCT (n = 575)

P value

Patient age

median years (range)

53.8 (18.7-77.4)

49.1 (18.1-78.7)

<0.0001

Patient sex

Male

278 (50.8%)

328 (57%)

0.037

 

Female

269 (49.2%)

247 (43%)

 

Year of HCT

median (range)

2016 (2010-2021)

2018 (2010-2021)

<0.0001

FLT3-ITD mutation

FLT3-wt

418 (76.4%)

325 (56.5%)

<0.0001

FLT3-ITD

129 (23.6%)

250 (43.5%)

 

NPM1 mutation

NPM1 absent

210 (39.3%)

327 (63.5%)

<0.0001

 

NPM1 presence

324 (60.7%)

188 (36.5%)

 
 

missing

13

60

 

Patient CMV

Pat. CMV neg.

64 (36.8%)

209 (37.9%)

0.8

 

Pat. CMV pos

110 (63.2%)

343 (62.1%)

 
 

missing

373

23

 

Karnofsky score

<90

82 (15.9%)

188 (35.9%)

<0.0001

 

≥90

433 (84.1%)

335 (64.1%)

 
 

missing

32

52

 

Time from diagnosis to HCT

median months (range)

4.7 (1.2-52.6)

5.2 (1.4-61.4) [4-6.8]

0.0003

The median follow-up was 37.5 months. All survival events were censored at 3 years. Due to a significant interaction between the FLT3-ITD status and the type of transplant, comparisons were performed separately for patients with wild type FLT3 (FLT3-wt) and those bearing a FLT3-ITD mutation (FLT3-ITD).

For FLT3-wt patients, the RI was lower in HAPLO than in ASCT (16.9% versus 32.6%; HR = 0.40 [95% CI: 0.27-0.60], P < 0.001), but contrarily the NRM was higher (17.2% vs 3.5%; HR = 7.02 [95% CI: 3.26-15.08, p < 0.001). Although no significant difference for LFS (65.9% vs 63.8%; HR = 0.86 [95% CI: 0.62-1.2], p = 0.37), the OS was significantly lower following HAPLO (73.2% vs 80.6%; HR = 1.69 [95% CI: 1.09-2.61], p = 0.018).

For FLT3-ITD patients, HAPLO was associated with a lower RI (8.2% vs 47.8%; HR = 0.14 [95% CI: 0.07-0.28], p < 0.001) and a higher NRM (20.2% vs 5.6%; HR = 3.43 [95% CI: 1.55-7.61], p = 0.002). The LFS was better (71.5% vs 46.6%; HR = 0.53 [95% CI: 0.34-0.85], p = 0.007) for HAPLO, but OS was comparable (73.5% vs 61.9%; HR = 0.83 [95% CI: 0.52-1.33], p = 0.44). Of note, more ASCT patients received a subsequent allogeneic HCT (25.2% vs 3.3%, P < 0.0001).

In both the FLT3-wt and FLT3-ITD group, NPM1 mutation was identified as a favorable factor for RI, LFS and OS.

Conclusions: For intermediate-risk AML patients without FLT3-ITD, ASCT is a valid consolidation option in CR1 with uMRD. In patients with FLT3-ITD, results of HAPLO could be better. In these patients, the role of maintenance with targeted agents post-transplant deserves further investigation.

Clinical Trial Registry: N.A.

Disclosure: No conflict to disclose.

19 - Acute Leukaemia

P004 NON MYELOABLATIVE VERSUS REDUCED INTENSITY CONDITIONING FOR ALLO-HCT FROM SIBLING OR UNRELATED DONORS IN PATIENTS WITH AML ≥ 65YEARS. A STUDY FROM THE EBMT ALWP

Alexandros Kanellopoulos 1,2, Myriam Labopin3, Alexandros Spyridonidis4, Emmanouil Nikolousis5, Uwe Platzbecker6, Didier Blaise7, Henrik Sengeloev8, David Burns5, Thomas Schroeder9, Goda Choi10, Eleni Tholouli11, Peter Dreger12, Victoria Potter13, Régis Peffault de Latour14, AEC Broers15, Jakob Passweg16, Matthias Eder17, Gitte Olesen18, Jean-Henri Bourhis19, Gwendolyn Van Gorkom20, Matthias Stelljes21, Jacques-Olivier Bay22, J.L. Byrne23, Anne Sirvent24, JA Snowden25, Eolia Brissot3, Fabio Ciceri26, Arnon Nagler27, Mohamad Mohty3

1Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, United Kingdom, 2University of Sheffield, Sheffield, United Kingdom, 3Hôpital Saint Antoine, Sorbonne University and INSERM UMRs 938, Paris, France, 4University of Patras, Patras, Greece, 5University Hospitals Birmingham, Birmingham, United Kingdom, 6University hospital Leipzig, Leipzig, Germany, 7Centre de Recherche en Cancérologie de Marseille, Institut Paoli Calmettes, Marseille, France, 8National University Hospital, Copenhagen, Denmark, 9University Hospital, Essen, Germany, 10University Medical Center Groningen (UMCG), Groningen, Netherlands, 11Manchester Royal Infirmary, Manchester, United Kingdom, 12University of Heidelberg, Heidelberg, Germany, 13Kings College Hospital, London, United Kingdom, 14Hopital St. Louis, Paris, France, 15Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, Netherlands, 16University Hospital, Basel, Switzerland, 17Hannover Medical School, Hannover, Germany, 18Aarhus University Hospital, Aarhus, Denmark, 19Gustave Roussy Cancer Campus, Villejuif, France, 20University Hospital Maastricht, Maastricht, Netherlands, 21University of Muenster, Muenster, Germany, 22Chu Estaing, Clermont, France, 23Nottingham University, Nottingham, United Kingdom, 24CHU Lapeyronie, Montpellier, France, 25Sheffield Teaching Hospitals NHS Foundation trust, Sheffield, United Kingdom, 26Ospedale San Raffaele, Milano, Italy, 27Sheba Medical Center, Tel Aviv, Israel

Background: Acute myeloid leukemia patients ≥65 years old have higher transplant failure risks and should be considered for reduced intensity (RIC) or non-myeloablative (NMA) conditioning. The optimal type of conditioning in allogeneic hematopoietic stem cell transplantation (allo-HCT) for elderly AML remains unknown. The aim of this study was to compare NMA versus RIC allo-HCT outcomes in AML patients aged ≥65 years.

Methods: Retrospective analysis of transplant outcomes for AML patients ≥65 years old (n = 2900) who had first allo-HCT in first or second complete remission (CR1/CR2) between years 2004-2021 from either matched sibling, HLA-10/10 or HLA-9/10 matched unrelated, or haploidentical unrelated donor receiving RIC Fludarabine/Busulfan iv 6.4mg/kg (Flu/Bu2) (35.6%, n = 1033), or Fludarabine/Melphalan 110-140mg/m2(Flu/Mel) (22.2%, n = 643), or Fludarabine/Treosulfan 30g/m2(Flu/Treo) (10.8%, n = 313), versus NMA Fludarabine/Total Body Irradiation 2Gy (Flu/TBI2Gy) (31.4%, n = 911). Patients who had cord blood units or ex vivo T cell depletion (TCD) were excluded.

Results: Median age of NMA and RIC cohorts was 68.6 and 68.0 years respectively (p < 0.0001). The RIC arm had more Karnofsky score ≥90 (70% vs 63.4%, p < 0.0006), female donor to male recipient combination (44.7% vs 16.6%, p < 0.0001), and in vivo T cell depletion (82.5% vs 10.2%, p,0.0001) whereas NMA patients more frequently received a haploidentical graft (19.9% vs 4.3%, p < 0.0001), bone marrow graft (6.4% vs 3.1%, p < 0.0001), and post-transplant cyclophosphamide (23.8% vs 7.3%, p < 0.0001). Secondary AML incidence, recipient sex, cytogenetic risk group, pre-transplant measurable residual disease (MRD), and FLT3/NPM1 status were not statistically different between the two arms. With a median follow-up 35.6 and 50.4 months for RIC and NMA (p < 0.0001) respectively, overall survival (OS), leukemia free survival (LFS) and non-relapse mortality (NRM) at 2 years was 53.4% v 56.5% (HR 0.87, p = 0.19), 51% vs 50.7% (HR 0.97, p = 0.78) and 23.8% vs 20.7% (HR 0.97, p = 0.83) respectively. Despite higher cumulative incidence (CI) of grade II-IV graft versus host disease (GVHD) for RIC (26.3% vs 25.5%, HR = 1.45, p = 0.01), the NMA arm had higher CI of chronic extensive GVHD (25.3% vs 12.9%, HR = 1.59, p = 0.02) whilst GVHD-free/relapse free survival (GRFS) was equivalent [HR (RIC vs NMA) = 0.94, p = 0.46] between both arms. Adverse factors for OS were HLA-9/10 donors (HR 1.5, p = 0.002), secondary AML (HR 1.19, p = 0.01), and complex cytogenetics (HR 1.64, p = 0.014). KPS ≥ 90 (HR 0.84, P = 0.006) and use of peripheral blood stem cell over bone marrow grafts (HR 0.66, p = 0.001), exerted a favourable OS effect.

Subgroup sensitivity analysis between NMA n = 242 and RIC n = 737 patients with known MRD pre-transplant, demonstrated no significant benefit of RIC over NMA, in terms of LFS (HR 0.95, p = 0.74), relapse (HR 0.88, p = 0.58) and OS (HR 0.85, p = 0.38).

Conclusions: We conclude that conditioning intensity (NMA vs RIC) did not impact on 2-year NRM, LFS and OS in elderly ( ≥ 65 years) AML. Higher risk of acute gr II-IV GVHD and lower risk of extensive chronic GVHD using RIC compared to NMA was identified. Subgroup sensitivity analysis did not reveal significant interaction between conditioning intensity and pre-transplant MRD status in terms of transplant outcomes.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P005 ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANT IN ADULT PATIENTS WITH IDH MUTATED AML AND MDS: A STUDY FROM THE GRUPO ESPAÑOL DE TRASPLANTE HEMATOPOYÉTICO Y TERAPIA CELULAR

Ana Pérez-González 1, Víctor Navarro2, Manuel Guerreiro3, Izarbe Delgado4, Ana Pilar González5, Alejandro Avendaño6, Alberto Musseti7, Mi Kwon8, Irene García Cadenas9, María Teresa Zudaire Ripa10, Oriana Lopez Godino11, David Valcárcel Ferreiras1, Guillermo Ortí Pascual1

1University Hospital Vall d’Hebron, Barcelona, Spain, 2Oncology Data Science (ODysSey) Group, Vall d’Hebron Institute of Oncology (VHIO), Barcelona, Spain, 3University Hospital La Fe, Valencia, Spain, 4University Hospital Arnau de Vilanova, Lleida, Spain, 5University Hospital of Cabueñes, Gijón, Spain, 6University Hospital of Salamanca, Salamanca, Spain, 7Institut Catalá d’Oncologia, Hospital Duran i Reynals, Barcelona, Spain, 8University Hospital Gregorio Marañón, Madrid, Spain, 9University Hospital Santa Creu i Sant Pau, Barcelona, Spain, 10Complejo Hospitalario de Navarra, Pamplona, Spain, 11University Hospital Morales Meseguer, Murcia, Spain

Background: Data regarding the prognostic relevance of IDH mutations (IDHm) in AML and MDS patients undergoing allogeneic hematopoietic cell transplant (allo-HCT) are limited. We sought to investigate the outcomes and prognostic factors of these patients in a retrospective manner.

Methods: This was a multicentre, retrospective study approved by Grupo Español de Trasplante Hematopoyético y Terapia Celular (GETH-TC) and by the Vall d’Hebron University Hospital ethics committee. Inclusion criteria were: 1) diagnosis of AML/MDS as per WHO criteria in patients >18 years; 2) mutation of IDH1/IDH2 detected by Sanger, PCR or NGS; and 3) undergoing first allo-HCT with MRD/MMRD/MUD/MMUD and myeloablative/reduced-intensity conditioning. The primary endpoint was overall survival. Secondary endpoints were PFS, NRM, cumulative incidence of relapse and GRFS.

Results: A total of 101 patients allografted from 2012 to 2021 in 9 GETH-TC institutions were identified and constitute the objective of this analysis; 88 (87.1%) and 13 (12.9%) patients were diagnosed with AML and MDS, respectively. Clinical and allo-HCT characteristics are shown in table 1. Of note, only one patient with IDH1m-AML received an IDH-inhibitor prior to allo-HCT in our cohort. Normal karyotype was observed in 66 patients (71%). The vast majority of patients (99%) carried additional mutations, with 43 patients (43%) harbouring ≥2 mutations. The most frequent co-mutations comprised NPM1 (31%) and DNMT3A (27%); FLT3-ITD mutations (12%) and TP53 (2%) were less represented.

With a median follow up of 21.2 months (CI95% 17.5 - 29.6), the estimated 2-year OS was 73% (IQR 63-84); 2-year OS was 77% and 70% in IDH1 and IDH2 cohorts, respectively (p = 0.76). In univariate analysis non-intensive treatment (p = 0.03) was associated with lower OS and MRD positivity (p = 0.07) showed a trend towards impaired outcomes. The 2-year PFS was 70% (IQR 61-81), with 2-year PFS of 71% and 69% in IDH1 and IDH2 cohorts, respectively (p = 0.86). MVA showed that non-intensive induction treatment (p = 0.05) and FLT3 mutations (p = 0.01) were associated with a lower PFS. The 2-year RI was 16% and 23% in IDH1 and IDH2 cohorts, respectively (p = 0.7). In the MVA, active disease at allo-HCT (p = 0.1) and detectable MRD (p = 0,08) showed a trend towards a higher RI. The 2-year NRM was 12% (95% CI 4-19), with no differences between IDH1 and IDH2 cohorts.

The 100-day CI of aGVHD was 39% (IQR 29-48); with grade III-IV of 9% (IQR 2-12). In univariate analysis, IDH1m patients showed a trend to a higher CI of 100-day grade I-IV/grade III-IV aGVHD compared to IDH2m patients, 46.1%/12% vs 32.6%/5%, respectively (p = 0.06 / p = 0.1). The 1-year CI of cGVHD was 31% (IQR 20-41) with no differences between IDH1 and IDH2 cohorts. The 1-year GRFS was 54% (40-65%), with 1-year GRFS 31% for IDH1 and 58% for IDH2 cohort (p = 0.08).

Table 1. Patients’ characteristics

Variables

Total (n = 101)

IDH1 (n = 46)

IDH2 (n = 54)

Median age (range)

55 (23 - 69)

56 (27 - 69)

55 (23 - 70)

Male gender, n (%)

70 (69%)

34 (74%)

35 (65%)

Diagnosis AML/MDS, n (%)

88 (87%)/13 (13%)

41 (89%)/5 (11%)

46 (85%)/8 (15%)

Abnormal cytogenetics, n (%)

27 (29%)

10 (24%)

16 (32%)

ELN 2017 cytogenetic risk in AML (n = 88), n (%)

   

Favourable

20 (23%)

9 (22%)

11 (24%)

Intermediate

46 (52%)

23 (56%)

23 (50%)

Adverse

22 (25%)

9 (22%)

12 (26%)

Treatment prior to allo-HCT, n (%)

   

Intensive/

83 (82%)/

36 (78%)/

46 (85%)/

hypomethylating & other non-intensive treatments

18 (18%)

10 (22%)

8 (15%)

Disease status at allo-HCT (CR+CRi) CR1/CR2, n (%)

66 (65%)/20 (20%)

28 (61%)/13 (28%)

37 (69)/7 (13)

Conditioning, MAC/RIC n (%)

41 (41%)/60 (60%)

13 (28%)/33 (72%)

28 (52%)/26 (48%)

Type of donor, n (%)

   

MRD/MUD

38 (38%)/37 (37%)

21 (46%)/15 (33)

16 (30%)/22 (40%)

Haplo

26 (26%)

10 (22%)

16 (30%)

GvHD prophylaxis, n (%)

   

CNI-based/ PT-Cy

37 (37%)/54 (53%)

18 (39%)/25 (54%)

18 (33%)/29 (54%)

None (TCD ex-vivo)

10 (10%)

3 (7%)

7 (13%)

  1. N number, AML acute myeloid leukemia, MDS myelodysplastic syndrome, ELN European Leukemia Net, CR1 first complete response, CR2 second complete response, CRi complete response without haematological recovery, MAC myeloablative conditioning, RIC reduced intensity conditioning, MRD matched related donor, MUD matched unrelated donor, Haplo haploidentical donor, GVHD graft-versus-host disease, CNI-based calcineurin inhibitor based, PT-Cy Post-Transplant Cyclophosphamide, TCD ex-vivo T-cell depleted ex-vivo.

Conclusions: In this cohort, the prognosis of IDH1 and IDH2m patients undergoing allo-HCT is comparable, with a 2-year OS exceeding 70% in both cohorts. FLT3 mutation was an independent prognostic factor for PFS. Studies on post allo-HCT maintenance with FLT3 or IDH inhibitors in this setting are warranted.

Disclosure: Nothing to declare

19 - Acute Leukaemia

P006 COMBINATION OF BENDAMUSTINE AND CYCLOPHOSPHOMIDE FOR GRAFT-VERSUS-HOST DISEASE PROPHYLAXIS IN REFRACTORY MYELOID NEOPLASMS

Ivan Moiseev 1, Yulia Vlasova1, Elena Morozova1, Olga Epifanovskaya1, Ksenia Afanasyeva1, Anastasia Beynarovich1, Dmitrii Zhogolev1, Mikhail Kanunnikov1, Yulia Rogacheva1, Tatyana Rudakova1, Nikita Volkov1, Sergey Bondarenko1, Alexander Kulagin1

1RM Gorbacheva Research Institute, Pavlov University, Saint Petersburg, Russian Federation

Background: Efficacy of salvage allogeinic hematopoietic stem cell transplantation (HSCT) in myeloid neoplasms not responding to chemotherapy and targeted therapies remains limited. Our group have recently demonstrated augmented graft-versus-leukemia (GVL) effect with substituting cyclophosphamide with bendamustine in graft-versus-host disease (GVHD) prophylaxis regimen (Moiseev et al., TCT, 2021), but significant toxicity due to poorly controlled cytokine release syndrome (CRS). To overcome this limitation we conducted a pilot single-center study of GVHD prophylaxis with a combination of cyclophosphamide with bendamustine in refractory myeloid neoplasms.

Methods: The prospective (NCT04943757) Phase I/II single-arm study evaluated GVHD prophylaxis regimen consisting of bendamustine 50 mg/kg/day on days +3, +4, cyclophosphamide 25 mg/kg/day on days +3, +4 (PTCBCy), tacrolimus 0.03 mg/kg from day+5 to day+100 and mycophenolate mofetil 30 mg/kg/day on days 5-35. Patients received reduced intensity FB2 or FB3 conditioning according to performance status. Forty patients evaluable for response were included into the interim safety and efficacy analysis. Median age was 43 years (range 18-69). AML was an indication for HSCT in 29, MDS in 9 and 2 patients had CML and aCML. All patients had >5% of blasts in a bone marrow at the time of conditioning. Donors were matched related siblings in 23%, matched unrelated for 50% and haploidentical for 27% of patients. None of the patients received DLI for relapse prevention.

Results: Engraftment with hematological complete remission was documented in 85% of patients. No detectable minimal residual disease after engraftment was documented in 75%. Median time to engraftment was 17 days (range 12-35). One-year overall survival was 47% (95%CI 24-67%), while event-free survival (EFS, including graft failure as event) was 24% (95%CI 6-48%). Disease progression or relapse was the major cause of failure and was documented in 71% of patients (95%CI 47-86%). The only predictor of better EFS was either unrelated or haploidentical donor compared to matched related (36% vs 11%, p = 0.0060). On the other hand non-relapse mortality was very low (7.5%, 95%CI 2-18%) as well as the incidence of other complications. Cumulative incidence of Grade II-IV acute GVHD was 10%, moderate and severe chronic GVHD – 19%. CRS was documented in 27% of patients and only in 2 it was grade 4-5. Most common manifestations of CRS involved liver (in 25%) and skin (7%). CRS was effectively controlled by tocilizumab, ruxolitinib and high dose steroids in all, but one patient. All CRS cases were associated with increased serum ferritin (median 15541 ng/ml, range 6150-206500). Preliminary flow cytometry analysis demonstrated the same pattern of early immunological recovery, preservation of memory T-cells and induction of tolerance by PD-1L positive monocytes as in the single-agent bendamustine study.

Conclusions: This pilot trial demonstrated that PTBCy combination prophylaxis provides the level of safety compared to conventional GVHD prophylaxis regimens with maintenance of GVL patterns. Optimal type of donor and post-transplant relapse prevention are still to be determined.

Clinical Trial Registry: NCT04943757, clinicaltrials.gov

Disclosure: The authors declare no conflicts of interest. The study was supported by RSF grant 23-15-00327.

19 - Acute Leukaemia

P007 OUTCOMES IN PATIENTS WITH B-CELL PRECURSOR ACUTE LYMPHOBLASTIC LEUKEMIA WHO RECEIVED INOTUZUMAB OZOGAMICIN AND PROCEEDED TO HEMATOPOIETIC STEM CELL TRANSPLANTATION: A REGISTRY-BASED, OBSERVATIONAL STUDY

David Marks 1, Marcos de Lima2, Partow Kebriaei3, Francesco Lanza4, Christina Cho5, Gizelle Popradi6, Mei-Jie Zhang7, Fan Zhang8, Verna Welch9, Saara Tikka10, Erik Vandendries11, Matthias Stelljes12, Wael Saber7

1University Hospitals Bristol NHS Trust, Bristol, United Kingdom, 2Ohio State University, Blood and Marrow and Cellular Therapy Program, Columbus, United States, 3MD Anderson Cancer Center, Houston, United States, 4Ospedale di Ravenna, Ravenna, Italy, 5Memorial Sloan Kettering Cancer Center, New York, United States, 6McGill University Health Centre, Montreal, Canada, 7Center for International Blood and Marrow Transplant Research, Medical College of Wisconsin, Milwaukee, United States, 8Pfizer Inc, Shanghai, China, 9Pfizer Inc, New York, United States, 10Pfizer Oy, Helsinki, Finland, 11Pfizer Inc, Cambridge, United States, 12Universitätsklinikum Münster, Münster, Germany

Background: Inotuzumab ozogamicin (InO) is a CD22-directed antibody–drug conjugate indicated for the treatment of relapsed/refractory (R/R) B-cell precursor acute lymphoblastic leukemia (ALL). This study examined outcomes following hematopoietic stem cell transplantation (HSCT) in patients who received InO prior to HSCT.

Methods: We report outcomes in patients (aged ≥18 years) with ALL and R/R ALL who received InO and proceeded to first allogeneic HSCT (United States observational, post-authorization safety study). Outcomes included disease status, overall survival, non-relapse (transplant-related) mortality (NRM), non-transplant-related mortality, and adverse events (AEs) of interest, including veno-occlusive disease/sinusoidal obstruction syndrome (VOD/SOS). Multivariate analyses examined prognostic factors of NRM at 1 year and VOD/SOS incidence at 100 days post HSCT. This analysis is based on 4-year interim data.

Results: In total, 196 adults (median age 39 years, 54% male, 75% White) received InO and proceeded to first allogeneic HSCT: 32% in first complete remission (CR1), 46% in CR2, 12% in CR ≥ 3, 5% in first relapse, 2% in third or subsequent relapse, and 3% with primary induction failure. Prior to HSCT, 32%, 45%, and 17% received 1, 2, and ≥3 cycles of InO, respectively. Post HSCT outcomes were analyzed in 177 adults with ALL and 120 adults with R/R ALL. The most common causes of NRM at 12 months were VOD/SOS (10/35; 29%) and graft-versus-host disease (GVHD; 8/35; 23%). AEs occurring in ≥30% of patients ≤100 days post HSCT were bacterial infection (46%), viral infection (41%), and acute GVHD (grades II–IV, 38%). VOD/SOS incidence ≤100 days post HSCT was 15% and 19% in patients with ALL and R/R ALL; corresponding mortality rates were 38% and 36% in patients who developed VOD/SOS ≤ 100 days (6% and 7% as a percentage of all patients). Median (range) time from HSCT to VOD/SOS was 0.4 (0.2–2.6) months in both groups. Of 26 adults with ALL who developed VOD/SOS, 10 cases were mild, 16 severe; 5/26 patients received prophylactic defibrotide; 12/26 patients received defibrotide treatment alone or in combination. In those who died, median (range) time to death was 4.5 (0.4–39.2) months.

In total, 143 adults with ALL (alive at last contact and reported ≥1 year of follow-up or who died at any time after HSCT) were included in multivariate analyses. Total serum bilirubin ≥ the upper limit of normal before HSCT (hazard ratio [HR], 5.72; P = 0.0001) and 4 cycles of InO (HR, 6.81; P = 0.0009) were negative prognostic factors for NRM at 1 year. A conditioning regimen containing dual alkylators was the only negative prognostic factor for VOD/SOS incidence at 100 days (odds ratio, 8.3; P = 0.0033).

Conclusions: In this real-world population of adults with ALL who received InO before HSCT, including heavily pretreated patients, the incidence of VOD/SOS after HSCT was similar to that observed in INO-VATE (phase 3 study) and a pooled analysis of 2 clinical trials of InO-treated patients with R/R ALL. Although the VOD/SOS mortality rate was higher (36% vs 26%), NRM at 12 months was lower (24% vs 38%) here vs the pooled clinical trial population.

Disclosure: David Marks: consultancy at Kite and Pfizer, honoraria from Amgen, Kite, Novartis, and Pfizer, and speakers bureau with Kite, Novartis, and Pfizer; Marcos de Lima: consultancy at Amgen, Celgene, Incyte, and Pfizer, and research funding from Celgene and Pfizer; Partow Kebriaei: consultancy at Jazz, Kite, and Pfizer, and research funding from Amgen and Ziopharm; Francesco Lanza: consultancy at AbbVie, Alexion, and Amgen, and research funding from Pfizer; Christina Cho: consultancy at AlloVir; Gizelle Popradi: consultancy at, and honoraria from, AbbVie, Kite, Gilead, Kyowa Kirin, Merck, Novartis, Pfizer, Servier, Seattle Genetics, and Taiho, research funding from Jazz, Novartis, and Syndax Pharmaceutical, and speakers bureau with AbbVie, Gilead, Jazz, Kyoma Kirin, Merck, Novartis, PeerVoice, Pfizer, Servier, and Seattle Genetics; Mei-Jie Zhang: nothing to declare; Fan Zhang: employment at Pfizer; Verna Welch, Saara Tikka, and Erik Vandendries: employment and equity ownership at Pfizer; Matthias Stelljes: consultancy at, speakers bureau with, and research funding and honoraria from, Pfizer; Wael Saber: nothing to declare.

19 - Acute Leukaemia

P008 ACUTE LYMPHOBLASTIC LEUKEMIA PH POSITIVE VS PH NEGATIVE - REAL WORLD EXPERIENCE WITH ALLOGENEIC HCT OUTCOME ON BEHALF OF POLISH ADULT LEUKEMIA GROUP (PALG)

Jarosław Dybko 1, Agnieszka Piekarska2, Alicja Sadowska-Klasa2, Anna Płotka3, Beata Piątkowska-Jakubas4, Donata Urbaniak-Kujda5, Małgorzata Sobczyk-Kruszelnicka6, Sebastian Giebel6, Lidia Gil3

Lower Silesian Oncology Center, Wrocław, Poland, University Clinical Center, Medical University of Gdańsk, Gdańsk, Poland, Poznań University of Medical Sciences, Poznań, Poland, Jagiellonian University Hospital, Kraków, Poland, Blood Neoplasms and Bone Marrow Transplantation, Wroclaw Medical University,, Wrocław, Poland, Maria Skłodowska-Curie National Research Institute of Oncology (MSCNRIO), Gliwice, Poland

Background: Patients with acute lymphoblastic leukemia Philadelphia positive (ALL Ph+) have been reported to have a better transplant outcome comparing to Ph negative cases (ALL Ph-). The routine pretransplant and posttransplant use of tyrosine kinase inhibitors (TKIs) has improved outcomes in Ph+ cases but the optimal conditioning regimen, donor type, and TKI remain undefined although the role of pre-transplant positive minimal residual disease (MRD) has beem established. We analyzed retrospective PALG data of 256 ALL patients to compare this outcome in Polish population.

Methods: We analyzed a “historic” group from pre-antibodies (anti-CD19 and anti-CD22) era. The whole cohort consisted of 82 Ph+ and 169 Ph- cases. The Ph+ group was older (median age 40 vs 29, p < 0.001) but there were no differences in sex or demographics. Majority of Ph+ patients expressed B-cell phenotype (95.1% vs 76.9% in Ph-, p < 0.001) and most of them in both groups were transplanted in CR1 (87.8% vs 78.1, p = NS) from matched unrelated donor (61.2% vs 52.0%, p = NS; only 10 haploidentical transplants in Ph- group, all the rest transplanted form matched sibling donors) based on MAC conditioning (80.5% vs 89.3%, p = NS). In Ph+ group 66 patients were treated with imatinib, 16 - with dasatinib before alloHCT, 50 and 9 after transplantation, respectively.

Results: Patienst were transplanted form matched unrelated and sibling donors in Ph+ group (50 and 32, respectively) and from matched unrelated, sibling and haplidentical donors in Ph- group (89, 71 and 10, respectively) with no signifcant difference between groups. MAC/RIC conditioning ratio was 66/16 in Ph+ and 151/18 in Ph- negative group. There were no differences in CMV staus of patients and donors and median number of CD34+ cells transplanted. Acute GvHD rate was 46.3% vs 41.4% (p=NS), chronic GvHD - 65.9% vs 66.9% (p=NS) with no difference for higher grades. The only significant difference was for liver acute GvHD with Ph- group domination (3.7% vs 13.0%, p = 0.02). The survival rate was significantly higher for Ph+ group (78.0% vs 52.1%, p < 0.001, Fig.1) with no difference between groups for non relapse mortality. The relapse rate was significantly lower in Ph+ group (9.7% vs 20.1%, p < 0.01).

Conclusions: Our historic data confirmed better survival rate in Ph+ group with no significant differences in NRM or GvHD but significantly higher relapse rate in Ph- group. These results are just another voice in discussion on precise control of minimal residual disease in pre- and post-transplant setting in ALL patients and modern therapies with bispecific antibodies and antibody-drug conjugates as new tools to eliminate MRD are of course another chapter.

Clinical Trial Registry: No

Disclosure: Nothing to declare

19 - Acute Leukaemia

P009 ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION FOR T-CELL ACUTE LYMPHOBLASTIC LEUKEMIA AND LYMPHOBLASTIC LYMPHOMA: OUTCOMES AT THE NATIONAL CANCER CENTER HOSPITAL (NCCH) OF JAPAN

Masanori Sakata 1, Mizuki Watanabe1, Sung Won Kim1, Asako Usui1, Wataru Takeda1, Takashi Tanaka1, Ayumu Ito1, Yoshihiro Inamoto1, Takahiro Fukuda1

1National Cancer Center Hospital, Tokyo, Japan

Background: There are limited data on outcomes and prognostic factors in allogeneic hematopoietic stem cell transplantation (allo-HSCT) for T-cell acute lymphoblastic leukemia (T-ALL) and T-cell lymphoblastic lymphoma (T-LBL). Herein, we retrospectively examined the clinical outcomes of patients with T-ALL/T-LBL receiving allo-HSCT at National Cancer Center Hospital of Japan.

Methods: Between 2000 and 2022, 54 patients received a first allo-HSCT for T-ALL(n = 24) and T-LBL(n = 30). Overall survival (OS) and relapse free survival (RFS) were assessed using the Cox proportional hazard model and progression and transplant-related mortality (TRM) were assessed using Fine and Gray’s proportional subhazards model. Variables selected manually in the preceding univariate analysis with P < 0.05 were put in the multivariate models.

Results: Median age at transplantation was 31 years (range, 7-64) and median follow-up for survivors was 3 years. Disease status at allo-HSCT was 1st complete remission (CR) in 19 patients, 2nd CR in 12 patients and not in remission (NIR) in 23 patients. Myeloablative conditioning (MAC) regimens utilizing total body irradiation (TBI) ≥ 12Gy was used in 37 patients, non TBI-MAC regimens in 5 and reduced intensity conditioning (RIC) regimens in 12. A total of 16 patients (T-ALL, n = 3; T-LBL, n = 13) had received focal irradiation at mediastinal or sanctuary involvements prior to allo-HSCT. Twelve patients were with early T-cell precursor type, and other subtypes were pre-T in 21, cortical-T in 9, mature-T in 5, and pro-T in 2. Three-years overall survival (OS) was 46.9% and 3-years relapse/progression-free survival (RFS) was 37.8 %. Cumulative incidence of relapse/progression and non-relapse mortality (NRM) at 3 years was 54.3 % and 7.9 %, respectively, In multivariable analysis, status of NIR at allo-HSCT was significantly associated with inferior outcomes (1st CR vs NIR: OS, HR 3.33, P = 0.038; PFS, HR 4.72, P = 0.006) with higher relapse/progression risk (1st CR vs NIR: relapse, HR 3.37, P = 0.01), whereas patients with 2nd CR and 1st CR showed comparable outcomes (1st CR vs 2nd CR : OS, HR 0.826, P = 0.795; PFS, HR 1.07, P = 0.919; relapse/progression, HR 1.02, P = 0.97). ETP type was associated with inferior OS (HR 4.97, P = 0.0038). TBI-MAC had no superior outcomes compared to non TBI-MAC (OS, HR 1.42, P = 0.68; PFS, HR 0.55, P = 0.479) and RIC (OS, HR 1.07, P = 0.899; PFS, HR 0.52, P = 0.224). Conditioning regimen showed no impact on NRM, either (P = 0.138, in univariable analysis).

Conclusions: In our analysis, no difference was observed among the conditioning groups, contrary to the previous reports. Although our study was a retrospective analysis in a small cohort, it suggested that conditioning regimens other than TBI-MAC potentially could be taken into consideration in certain situations. We observed no difference in clinical outcomes between patients with 1st and 2nd CR at transplantation, although NIR status significantly affected the outcomes. Patients who achieved 2nd CR might be good candidates for allo-HSCT. Overall, high frequency of relapse after transplant remains a major issue, especially in those at NIR (3-year relapse: 65.2%), where efforts to achieve better remission before transplantation should be made.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P010 EARLY LYMPHOCYTE IMMUNE RECONSTITUTION AS PREDICTOR FOR OUTCOMES AFTER ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANTATION FOR MALIGNANT INDICATIONS: A TRI-INSTITUTIONAL ANALYSIS

Alexandre Troullioud Lucas 1,2, Caroline Lindemans2,3, Senthil Velan Bhoopalan4, Rana Dandis2, Susan Prockop5, Swati Naik4, Dinesh Keerthi4, Coco de Koning3, Akshay Sharma4, Stefan Nierkens2,3, Jaap Jan Boelens1

1Memorial Sloan Kettering Cancer Center, New York, United States, 2Princess Máxima Center for Pediatric Oncology, Utrecht, Netherlands, 3University Medical Center Utrecht, Utrecht, Netherlands, 4St. Jude Children’s Research Hospital, Memphis, United States, 5Boston Children’s Hospital/Dana-Farber Cancer Institute, Boston, United States

Background: Early CD4 immune reconstitution (IR) after allogeneic hematopoietic cell transplant (allo-HCT) correlates with lower non-relapse mortality (NRM) and a cutoff of >50/mL CD4+ cells has been found to be a good predictor for reduced NRM. The impact of CD4 and other lymphocyte IR on relapse of malignancy is less clear, especially in children.

To study the correlation of CD4, CD8 and natural killer (NK) cell IR with NRM and relapse in a large cohort of children and young adults after HCT for malignant indications.

Methods: We retrospectively analyzed data from consecutive patients receiving their first allo-HCT for a malignant indication at three large academic institutions between 2008-2019. Immune phenotyping analyses were performed regularly as standard of care after HCT in these centers. Statistical analyses involved linearity evaluation using martingale residuals plots. Cox proportional hazard models were used to study correlations where linearity assumptions were met. In case of non-linearity, maximally selected log-rank statistics were used to identify cutoffs related to outcomes. Fine-Gray competing risk methods were used to calculate cumulative incidence. All statistical analyses were done using R statistical software, version 4.2.1.

Results: Up to 503 patients were included in our analyses, median age 11.9 years (range 0.5 – 32.2). Diagnoses included acute lymphoblastic leukemia (n = 219), acute myeloid leukemia (n = 186), myelodysplastic syndrome (n = 42), non-Hodgkin lymphoma (n = 17), mixed phenotype acute leukemia (n = 11), and others (n = 28). There was non-linearity between maximum CD4 cell count before day 100 after HCT and NRM n = 503). Log-rank statistics demonstrated a CD4 count of 57 cells/µL as the best cut-off to discriminate for risk of NRM (hazard ratio [HR] 0.27, 95% confidence interval [CI] 0.17-0.44, P < 0.001). There was linear correlation between maximum CD8 and NK cell counts within 100 days and NRM, but Cox proportional hazard models did not reveal any clinically relevant association (CD8: hazard ratio [HR] 0.999, 95% CI 0.998–1, P = 0.046, n = 491; NK: HR 0.9997, 95% CI 0.999 – 1, P = 0.381, n = 481). We found linear correlation between CD4, CD8 and NK cells within 100 days and relapse, but once again Cox proportional hazard models showed no statistical correlation (CD4: HR 1, 95% CI 0.9996–1.001, P = 0.407, n = 503; CD8: HR 1, 95% CI 0.9997–1, P = 0.943, n = 491; NK: HR 1, 95% CI 0.9996–1, P = 0.926, n = 481).

Conclusions: We confirm the strong association between early CD4 IR (of ~50/mL) and decreased NRM, but did not find an association between early CD8 or NK cell IR and NRM. Also, no component of IR (CD4, CD8 or NK cell) correlated with relapse in this analysis. To further elucidate cellular anti-leukemic effects of allo-HCT, more detailed studies on reconstitution of more specific lymphocyte subsets after allo-HCT need to be performed.

Disclosure: Caroline Lindemans reports honoraria for a lecture from Genzyme and on a data safety monitoring board of ExCellThera (related to this topic).

Susan Prockop reports support for the conduct of clinical trials: Jasper, Atara, AlloVir, Consulting/Advisory Board Participation CellEvolve, Smart Immune, Regeneron, and IP related to VSTs licensed to Atara with all rights assigned to MSKCC.

Akshay Sharma has received consultant fee from Spotlight Therapeutics, Medexus Inc. and Vertex Pharmaceuticals. He has also received research funding from CRISPR Therapeutics and honoraria from Vindico Medical Education. Akshay Sharma is the St. Jude Children’s Research Hospital site principal investigator of clinical trials for genome editing of sickle cell disease sponsored by Vertex Pharmaceuticals/CRISPR Therapeutics, Novartis Pharmaceuticals and Beam Therapeutics. The industry sponsors provide funding for the clinical trial, which includes salary support paid to his institution. Akshay Sharma has no direct financial interest in these therapies. None of these conflicts are related to the work presented here.

Jaap Jan Boelens reports honoraria from Avrobio, BlueRock, Bluebird Bio, Sanofi, Sobi, SmartImmune, Advanced Clinical consulting (not related to this topic).

All other authors report no conflicts of interest.

19 - Acute Leukaemia

P011 FREQUENCY AND IMPACT OF PRE-TRANSPLANT SOMATIC MUTATIONS ON CLINICAL OUTCOMES OF AML PATIENTS RECEIVING ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION

Khalid Halahleh1, Eman Alnaimi1, Ebtisam Danoun1, Isra Muradi2, Malek Horani1, Mohammad Makoseh1, Iyad Sultan1, Waleed Da’na 1

1King Hussein Cancer Center, Amman, Jordan, 2Al-Ahliyya Amman University, Amman, Jordan

Background: Acute myelogenous leukemia is characterized by recurrent somatic variants that have implications in management algorithm. ELN-AML 2022 included ASXL1, BCOR, EZH2, RUNX1, SF3B1, SRSF2, STAG2, U2AF1, or ZRSR2 and TP53 mutations. In the context allo-HSCT, the frequency and prognostic value of different gene-gene interactions has not been studied. We aim to assess the frequency and impact of different somatic mutations, either alone or in combination, on prediction of relapse and survival in patients receiving allo-HSCT regardless of MRD status.

Methods: This is a retrospective, single center KHCC registry-based analysis. Adult patients aged ≥18 years with a diagnosis of AML who received an allo-HSCT, Jan 2018- dec 2021 with an available pre-transplant genetic profile by next generation sequencing (NGS) were included.

Results: 149 patients were identified. 133 patients had de novo AML (89%). 94 patients (63%) were males, median age of 41 y (range: 18-78 y). 134 patients (90%) had pathogenetic mutation, -2 VUS (1.5%), and 13 (8.5%) had no detectable mutations. The most frequent mutations were DNMT3A (20%), NPM1 (20%), FLT3 (16%), NRAS (13.5%), TET2(9%), TP53(7.5%), SRSF2(7%), IDH1(6%), RUNX1, IDH2, KIT, and KRAS (5.4%); GATA2, ASXL1, CEBPA (5%). By multiple correspondence analysis, two independent groups of co-occurring mutations were identified, group 1 includes DNMT3A, NPM1 and FLT3, group 2 includes ASXL1, SRSF2, RUNX1.

Outcome analysis was performed on the subset of 74 patients allografted with available data for the six genes.29 subjects had de novo AML (39%), median age 41y (range:18-65 y). 53 patients (35.6%) received myeloablative conditioning, blood grafts (100%) from MRD 63(42.3%), MMRD 1 (0.7%), and haplo-donors 10(6.7%). 35 patients (47%) were in CR1, and 39(53%) ≥ CR2. 16/35 patients classified as intermediate and 19 as adverse risk.

Median follow up calculated by reverse Kaplan-Meier 38.4 mo(2- 78 mo). Median LFS was 29mo (18.5-42.5mo) and OS 25 mo (17-34mo) with 2-year LFS and OS of 54% and 50%. The 2-year LFS and OS were 70%, and 82%, respectively, for patients with DNMT3 mutation; 44% and 68% with NPM1 mutation; 44% and 67% with FLT-3 mutation, 100% and 100 % with ASXL mutation; 67% and 67% with RUNX1 mutation; 25% and 25 % with TP53 mutation; 75% and 74% with SRSF2 mutation; 50%, and 67% with NRAS; 67% and 100% with KRAS; 50% and 74% with TET2; 67% and 67% with IDH1; 80%, and 80% for patients with C-KIT mutation. When mutations were investigated in groups, the 2-year LFS and OS for group 1 and 2 were 50%, and 42.5% compared with 60%, and 29% respectively. 2-year LFS and OS for patients in CR1 84% and 84% compared with 22% and 69% allografted in CR2. 2-year LFS according to ELN-AML-2022, were 44%,43%, 61% and 90%,76% and 68% for OS.

Tale-1: Description of most frequent mutations (N-12)

Gene

Frequency (N)

Frequency

(%)

Median VAF

Max VAF

Min

FAV

DNMT3A

30

20%

40.0%

80.0%

3.0%

NPM1

30

20%

39.5%

69.0%

5.0%

FLT-3

24

16%

22.5%

56.0%

1.4%

NRAS

13

13.5%

28.0%

52.0%

3.0%

TET2

`3

9%

43.5%

81.0%

3.0%

TP53

11

7.5%

45.0%

92.0%

6.0%

SRSF2

10

7%

49.0%

77.0%

5.0%

IDH1

9

6%

42.0%

55.0%

16.0%

IDH2

8

5.4%

41.0%

45.0%

9.0%

C-KIT

8

5.4%

28.5%

53.0%

7.0%

RUNX1

8

5.4%

29.5%

67.0%

7.0%

KRAS

8

5.4%

23.0%

36.0%

7.0%

LFS and OS for allografted patients(N-74)

Conclusions: Next generation sequencing is a pivotal important too for risk stratification of patients with AML receiving allo-HSCT and determining post-transplant consolidation.

Disclosure: All authers have nothing to disclose

19 - Acute Leukaemia

P012 ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANTATION SUB-ANALYSIS OF THE GIMEMA AML1310, A RISK ADAPTED, MRD-DIRECTED THERAPY FOR YOUNG ADULTS WITH AML

Francesca Bonifazi 1, Alfonso Piciocchi2, Valentina Arena2, Enrico Maffini1, Anna Candoni3, Giovanni Martinelli4, Valeria Calafiore5, Lorella Melillo6, Paolo de Fabritiis7, Maria Paola Martelli8, Francesco Fabbiano9, Roberto Cairoli10, Prassede Salutari11, Agostino Tafuri12, Saveria Capria13, Mario Luppi14, Francesco Buccisano15, Maria Ilaria Del Principe15, Luca Maurillo15, Paola Fazi16, Marco Vignetti16, Fabio Ciceri17, Adriano Venditti15

1IRCCS Azienda Ospedaliero-Universitaria di Bologna, Bologna, Italy, 2GIMEMA Foundation, Roma, Italy, 3Azienda Sanitaria Universitaria Integrata di Udine, Udine, Italy, 4Istituto Tumori della Romagna, Meldola, Italy, 5Ospedale Ferrarotto, Catania, Italy, 6Fondazione IRCCS Casa Sollievo della Sofferenza, UO di Ematologia, San Giovanni Rotondo, Italy, 7Ospedale S. Eugenio, Roma, Italy, 8Ospedale S. Maria della Misericordia, Perugia, Italy, 9Ospedali Riuniti Villa Sofia-Cervello, Palermo, Italy, 10Ospedale Niguarda Ca Granda, Milano, Italy, 11Azienda USL di Pescara, Pescara, Italy, 12Azienda Ospedaliera Sant’ Andrea, Roma, Italy, 13Università degli Studi di Roma Sapienza, Roma, Italy, 14Università degli Studi di Modena e Reggio Emilia, Modena, Italy, Università di Tor Vergata, Roma, Italy, 16GIMEMA Foundation Rome, Roma, Italy, 17University Vita-Salute San Raffaele, Milano, Italy

Background: GIMEMA AML1310 was a prospective, MRD-driven clinical trial, designed to offer a risk-adapted treatment strategy for de novo young adults with acute myeloid leukemia (AML), including allogeneic hematopoietic cell transplantation (HCT) for patients with significant disease relapse risk.

Methods: The aim of this specific sub-analysis was to describe post-HCT outcomes of the patients included in the study who underwent allogeneic HCT. In the original protocol, patients with high-risk cytogenetics/genetics and those belonging to the intermediate-risk (IR) category with a MRD positive status after the first consolidation, were candidate to HCT.

Results: A total of 142 patients (median age 45 years, range 18-61) underwent an allogeneic HCT and were included in the analysis. Sixty-three percent of patients belong to HR group. Median follow-up for the entire population was 6.6 years. Donors were HLA-matched sibling (MSD) (n = 40), unrelated donors (UD) (n = 68), and haploidentical donors (n = 34). Graft source was peripheral blood in 56%, and bone marrow in the remaining patients. Most of the patients (91%) received myeloablative regimens. Patients’ characteristics stratified by donor type were superimposable across the cohorts. The overall 6-year rates of overall survival (OS) and disease-free survival (DFS) were: 50.6% (95% CI 43.0 - 59.5) and 43.7% (95% CI 35.7 – 53.3) respectively, for the entire population. The cumulative incidence of disease relapse (RI) and non-relapse mortality (NRM) were 33.9% (95% CI 25.4 - 42.4) and 22.3% (95% CI 14.9 – 29.8) respectively. There was a statistically not significant trend for better survival rates for IR respect to HR patients (6-year OS 58.4% vs. 46.6%, p = 0.16). Patients transplanted from haploidentical donors (24.1%, 95% CI 8.1 – 40.1) and UD (27.4%, 95% CI 15.4 – 39.3) showed a lower RI in comparison to MSD (51.4%, 95% CI 34.9 – 67.8). Conversely, patients grafted from MSD experienced lower rates of NRM (5.6%, 95% CI 2.1 – 13.3) opposed to UD (30.9%, 95% CI 18.5 – 43.3) and haploidentical (27.6%, 95% CI 11 – 44.2). Six-year OS was superimposable between MSD (54.8%, 95% CI 41.3 – 72.7) haploidentical (52.9%, 95% CI 38.6 – 72.7) and UD (47.1%, 95% CI 36.6 – 60.6). The analysis failed to demonstrate any influence of MRD on relapse incidence and survival outcomes. The univariate analysis did not identify sufficient statistically significant variables, to perform a multivariate analysis.

Conclusions: The protective effect of an allogeneic HCT observed among recipients of UD and haploidentical donors, respect to MSD, was offset by higher rates of NRM, nullifying the relative advantage and making survival outcomes superimposable between different donors. Although the original AML1310 was not conceived as a transplantation study, MRD was not informative respect to clinical outcomes.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P013 EFFICACY OF FLAG-MITOXANTRONE (M) COMBINED TO SHORT-ADMNINISTRATION OF VENETOCLAX (VEN) FOLLOWED BY ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION (AHSCT) IN RELAPSED/REFRACTORY (R/R) ACUTE MYELOBLASTIC LEUKEMIA (AML)

Ahmad Ibrahim 1, Abir Harmouch1, Ahmad Khalil1, Pamela Sfeir1, Kamal Zahran1, Charbel Khalil1, Rima Moghnieh1, Ali Youssef1, Tamima Jisr1

1Lebanese University, Middle East Institute of Health, Beirut Arab University, Makassed Hospital, Beirut, Lebanon

Background: Ven combined with FLAG-Ida (G-CSF, Fludarabine, Cytarabine, Idarubicin) represents an effective therapy for R/R AML patients (pts), associated with deep remissions and high rate of transition to successful AHSCT. Lowering the cytarabine dose (to 1.5g/m²/d for 5 d) and shortening the duration of Ven (to 14 days) led to lower toxicities (DiNardo, JCO 2021). FLAG -M is highly effective and well tolerated salvage regimen for R/R AML (Hassan, Int. J. Hematol. 2018). We report the results of salvage regimen combining shortened duration of Ven (7d) and dose reduced FLAG-M followed by AHSCT when feasible in R/R AML pts aged ≥ 18y.

Methods: This is a pilot study conducted between 7/2020 and 7/2022. Salvage therapy combined G-CSF d1-7, Fludarabine (30mg/m²/d IV) and Cytarabine (1.5g/m²/d IV) d2-5, M (12mg/m²/d IV) d2,4 and Ven (100mg PO daily) given with voriconazole d2-8. Anti-FLT3 therapy was included in 3 pts. For refractory pts after 1 cycle, no more cycles were given. Pts planned for AHSCT received 2 cycles of salvage and responsive pts not planned for AHSCT received 3 more cycles as consolidation. The objectives included rate of composite CR (CRc: CR,CRh,Cri), OS and RFS after salvage, and safety. Safety profile will not presented in this report.

Results: Median age at relapse was 40 y (25-67). There were 13 females and 7 males. First line therapy was 7 + 3 in 12 pts followed by AHSCT in 4, FLAG-Ida in 6 pts followed by AHSCT in 2, and 5-azacytidine+Ven in 2 pts. Anti FLT3 therapy was included in 3 pts. Median interval between first induction and salvage was 27.5m (2-175). Salvage therapy was given at a median of 2 cycles (1-4). Sixteen pts (80%) obtained CRc. Eleven pts (55%) underwent AHSCT in 2nd CR. Median follow up after salvage was 7m (1-34). Of the 20 pts, 8(40 %) are alive in CR between 4 and 34m, 2 are alive with AML, 8 died of AML and 2 pts died of infection in CR after AHSCT. The 3y-OS and 3y-RFS were 35% and 40%, respectively. Of the 9 pts who did not undergo AHSCT, only 1 pt is alive in CR at 4m, 1 pt is alive with AML at 14m, and 7 pts died of AML. The 3y-OS and 3y-RFS for these pts were 20% and 25%, respectively. Of the 11 pts who underwent AHSCT, 7 (64%) are alive in CR between 5 and 34m, 1 pt is alive with AML, 1 pt died of AML, and 2 pts died of infection at 1 and 5m of AHSCT. The 3y-OS and 3y-RFS for these pts were 70% and 85%, respectively, significantly higher than those for the pts who did not undergo AHSCT (p = 0.025 and p = 0.002,respectively).

Conclusions: Reduced dose FLAG- M with shortened duration of Ven (7d) is an active regimen in R/R AML capable of producing high remission rate and enabling transition to AHSCT when appropriate, with significant improvement in survival.

Disclosure: Nothing to declare

19 - Acute Leukaemia

P014 VOLATILE PROFILING USING AN ENOSE ALLOWS DIFFERENTIATION OF HEALTHY AND LEUKEMIC BREATH SAMPLES

Tobias Baudrexler1, Tobias Boeselt2, Micheal Atzler1, Anne Hartz1, Ursel Boas2, Christoph Schmid3, Andreas Rank3, Jörg Schmohl4, Rembert Koczulla2, Helga Schmetzer 1

1Ludwig-Maximilians-University, Munich, Germany, 2Phillipps-University of Marburg, Marburg, Germany, 3University Hospital of Augsburg, Augsburg, Germany, Diaconia Hospital Stuttgart, Stuttgart, Germany

Background: Volatile organic compounds (VOCs) reflecting the human metabolism can be collected easily in a noninvasive matter, directly measured by electronical nose (eNose) and might qualify as a systemic tool to monitor biomarkers related to disease. Myeloid leukemic blasts can be transformed into leukemia derived dendritic cells (DCleu) being able to improve (anti-leukemic) immune responses.

Methods: 3 therapy refractory AML were treated with/without rescue therapy (for 2-4 weeks) using Kit M (Granulocyte macrophage colony stimulating factor and Prostaglandin E1: GM-CSF and PGE1) (in consensus the local Ethic Committee of LMU in Munich, Vote-No 339-05). The impact of Kit-M on several immune cells and blasts (control vs. Kit-M treatment) was evaluated by culturing patients’ blood cells ex vivo (dendritic cell-culture (DC), mixed lymphocyte-culture (MLC), functional assays (cytokine secretion- (CSA), intracellular- (InCyt) and cytotoxicity-assay (CTX)). Patients were monitored clinically, hematologically and immunologically over the whole treatment phase. VOC monitoring: VOCs were collected with earloop masks containing exhaled air from AML patients in the course of the disease and from healthy probands (n = 15) and were measured by eNose. To profile the immunological changes in acute myeloid leukemia (AML) patients, we correlated clinical- and immunological- with the VOC results of breath samples collected in the course of the disease. Clinical and immunological monitoring: 3 refractory AML patients were observed in the course of the disease with different therapies: P1511 received chemotherapy and served as a control. P1482 and P1601 were treated with Kit-M (GM-CSF and PGE1) for 2-4 weeks. These patients were monitored clinically, hematologically and immunologically over the whole treatment phase. Blood samples were taken in the course of the observation time to monitor (leukemia specific) immune cells (flowcytometry, CSA, InCyt).

Results: Ex vivo: Kit-M was shown to give rise to DC/DCleu as well as to improve antileukemic functionality after (T cell enriched) mixed lymphocyte culture with Kit-M treated blood. In vivo: Patients’ treatment with Kit-M was shown to be safe, to improve clinical parameters (neutrophils, thrombocytes, and blast counts) and to induce (leukemia specific) immune activation, although effects decreased after discontinuation of Kit-M treatment. Healthy and leukemic VOC results from P1511, P1482 and P1601 could be differentiated during the whole observation time. Moreover, VOC profiles collected from healthy vs. AML breath donors (%sensitivity: 100; %specificity: 100; p = .0001) and VOC prints during chemotherapy vs. during Kit-M therapy (%sensitivity: 100; %specificity: 100; p = .0006) were significantly different.

Conclusions: Kit-M was a safe clinical drug, that produced DC/DCleu and improved anti-leukemic responses with an activating effect on adaptive and innate (leukemia specific) immunoreactive cells ex vivo and in vivo in patients with AML. Breath profiling using an eNose might qualify as a diagnostic tool to detect residual disease and to deduce a VOC based, disease-related monitoring strategy- without need of collecting VOCs directly via eNose.

Disclosure: Modiblast Pharma GmbH (Oberhaching, Germany) holds the European Patent 15 801 987.7-1118 and US Patent 15-517627 ‘Use of immunomodulatory effective compositions for the immunotherapeutic treatment of patients suffering from myeloid leukemias’, with whom H.M.S. is involved with.

19 - Acute Leukaemia

P015 VENETOCLAX AND HYPOMETHYLATING AGENTS FOLLOWED BY ALLOGENEIC STEM CELL TRANSPLANTATION IN PATIENTS WITH ACUTE MYELOID LEUKEMIA

Christina Rautenberg 1, Matthias Stelljes2, Fabian Acker3, Udo Holtick4, Edgar Jost5, Christoph Schliemann2, Anja Ragette1, Jennifer Kaivers1, Nina Neuendorff1, Maher Hanoun1, Laura K. Schmalbrock6, Krishan Braitsch7, Mareike Verbeek7, Katharina Goetze7, Gesine Bug3, Christian Reinhardt1, Nicolaus Kröger8, Thomas Schroeder1

1University Hospital Essen, Essen, Germany, 2University Hospital Muenster, Muenster, Germany, 3University Hospital Frankfurt, Frankfurt, Germany, 4University Hospital Cologne, Cologne, Germany, 5University Hospital Aachen, Aachen, Germany, 6Charite Berlin, Berlin, Germany, 7TU Munich, Munich, Germany, 8University Hospital Hamburg, Hamburg, Germany

Background: So far, it was a treatment paradigm to induce remission in patients with either newly diagnosed (ND) or relapsed/refractory (R/R) AML prior allo-SCT using intensive therapies with anthracyclins in combination with low- or intermediate to high-dose cytarabine. For AML patients ineligible for intensive therapy, the combination of HMA and Venetoclax (VEN) has become the new treatment standard due to high and rapidly achieved remission rates and a favorable toxicity profile. These properties make this combination also attractive for frontline induction of ND-AML or as salvage therapy in R/R-AML patients with borderline fitness to tolerate and/or limited chance to achieve remission by intensive therapy. Considering the limited evidence from the literature so far, we here report our multi-center experience of allo-SCT after HMA/VEN-based therapy.

Methods: We retrospectively analyzed ND and R/R AML patients, who had received HMA/VEN therapy followed by first allo-SCT at 8 German centers from 2021 to 2022, regarding outcome after transplant. The study was approved by the ethics committee of University Duisburg/Essen (approval number: 22-10708-BO) and all patients gave written informed consent for scientific use of their data.

Results: Eigthy-six patients (median 64 years, 15 to 76) with ND (n = 33, 38%) or R/R (n = 53, 62%) AML (ELN 2022 genetic risk int/adv 86%) were treated with a median of 2 cycles HMA/VEN (1 to 5). Overall response rate (CR/CRi, MLFS) in the entire cohort was 69% with no difference between patients with ND and R/R AML (76% vs. 63%, p = 0.22). MRD was not detectable prior transplant in 50% of responders. Allo-SCT using PBSC (n = 83, 99%) or BM (n = 1, 1% missing n = 2) from related (n = 14, 17%), unrelated (n = 56, 66%) or haploidentical donor (n = 14, 17%, missing n = 2) was performed in median 88 days (17 to 481) after start of HMA/VEN. RIC was used in 91% including 24% receiving sequential conditioning. Median follow-up from start of HMA/VEN was 15 months and from allo-SCT 11 months, respectively. The estimated 1-year-OS probability after allo-SCT is 60% with no significant difference between patients with ND and those with R/R AML (50% vs. 67%, p = .54). However, 1-year-OS after transplant was significantly higher in patients transplanted in remission (72% vs. 33%, p = .001), mainly driven by R/R patients achieving remission prior transplant (1-year-OS 92%). In patients with remission prior transplant OS did not differ between MRD negative and positive patients (85% vs. 71%, p = .98). A total of 25 patients (29%) relapsed after allo-SCT and a total of 31 patients (36%) have died due to relapse (n = 21) or treatment-related toxicity (n = 10). The estimated 1-year-RFS probability was 54% with no significant difference between ND and R/R patients (46% vs. 60%, p = 0.46). Again, RFS significantly differed between patients in remission and with active disease at transplant (66% vs. 28%, p = 0.002), but not between MRD negative and positive patients (73% vs. 66%, p = 0.69).

Conclusions: Treatment with HMA/VEN either as induction or salvage therapy in ND or R/R AML leads to reasonable remission rates facilitating allo-SCT with promising survival in this elderly high-risk population. This warrants confirmatory prospective investigation in larger cohorts with longer follow-up.

Clinical Trial Registry: not applicable

Disclosure: no COI

19 - Acute Leukaemia

P016 OUTCOMES OF ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANT IN PATIENTS WITH WT1 MUTATED MYELOID NEOPLASMS

Anmol Baranwal 1, Rami Basmaci1, Rong He1, David Viswanatha1, Patricia Greipp1, James Foran2, Jeanne Palmer3, William Hogan1, Mark Litzow1, Abhishek Mangaonkar1, Mithun Shah1, Aref Al-Kali1, Hassan Alkhateeb1

1Mayo Clinic, Rochester, United States, 2Mayo Clinic, Jacksonville, United States, 3Mayo Clinic, Phoenix, United States

Background: Mutation in Wilms’ Tumor 1 (WT1) gene in patients with acute myeloid leukemia (AML) has been associated with a poor prognosis. The primary objective of this study was to describe allogeneic stem cell transplant (alloSCT) outcomes and assess prognostic factors in patients with WT1 mutated (mWT1) myeloid neoplasms.

Methods: We retrospectively reviewed patients with mWT1 myeloid neoplasms who underwent alloSCT. Kaplan–Meier and log-rank tests were used to estimate OS. Post-transplant non-relapse mortality (NRM) and relapse incidence (RI) was calculated using competing risk analysis. Factors with P < 0.10 in univariate analysis were included in multivariate analysis. Genes/co-mutations were analyzed if they were present in at least 10% (n ≈ 3) patients. R 4.2.0 (R Foundation for Statistical Computing) was used for statistical analyses.

Results: Thirty-two patients (24 (75%) AML, 3 (9.4%) MDS and 5 (15.6%) others) were evaluated. Median age at alloSCT was 45 years (IQR 33-62 years). Among patients with AML, 4 (16.7%) were favorable, 9 (37.5%) intermediate, and 11 (45.8%) adverse risk disease. Twelve (37.5%) patients had multiple WT1 mutations (mmWT1). Two hotspots in exon 4 were found: 1. codons 301-303 [7 (21.9%) patients], and 2. codons 312-314 [13 (40.6%) patients]. Compared to single mWT1, a higher proportion of patients with mmWT1 had a mutation in one of the hotspot regions (35% vs. 91.7%, P = 0.006): 6 (50%) had mutation among codons 301-303 (P = 0.01), while 5 (41.7%) had mutation among codons 312-314 (P = 0.23). Median WT1 VAF was 31% (IQR 12-42%).

The cumulative incidence of NRM was 6.3% at 100 days, 13.2% at 1 year and 17.9% at 3 years after alloSCT. Cumulative RI was 18.8% at 100 days, 49.2% at 1 year and 55.8% at 3 years after alloSCT.

Patients with mmWT1 had higher RI at 100 days (33.3% vs. 10%, P = 0.12), 1 year (75% vs. 35.6%, P = 0.04) and 3-years (87.5% vs. 35.6%, P = 0.01) post-transplant. Multivariate analysis showed that mmWT1 was associated with increased risk of post-transplant relapse (HR 6.12, 95% CI 2.03-18.4, P = 0.001). Among the 9 patients with mmWT1 who relapsed, 2 were alive at 3-years post-alloSCT (Table 1).

OS at 3-years post-transplant was similar among patients with AML vs. other, non-AML, diseases (median OS 1.45 vs. 1.20 years, P = 0.59). Among the patients with mutations in codons 312-314, 3-year RI was 61.5%, compared to 53.9% for the rest of the cohort (P = 0.64). However, all the patients with codon 312-314 mutations who relapsed, died within 3 years of alloSCT, and codon 312-314 mutations were associated with worse 3-year post-alloSCT survival (median 0.8 vs. 1.9 years, P = 0.03). Multivariate analysis identified the presence of a mWT1 in codons 312-314 as an independent risk factor for 3-year post-alloSCT OS (HR 4.15, 95%CI 1.43-12.01, P = 0.009).

Table 1. Characteristics and outcomes of patients with WT1 mutated myeloid neoplasms.

Variable

Multi-mutated WT1 (mmWT1)

 
 

No

(N = 20)

Yes

(N = 12)

P-value

Complete remission at alloSCT

 No

2 (10.0%)

2 (16.7%)

1

 Yes

17 (85.0%)

10 (83.3%)

 

Conditioning intensity

 

 MAC

12 (60.0%)

7 (58.3%)

1

 RIC

7 (35.0%)

5 (41.7%)

 

Grade 2-4 acute GVHD

 No

17 (85.0%)

10 (83.3%)

1

 Yes

3 (15.0%)

2 (16.7%)

 

Grade 3-4 acute GVHD

 No

13 (65.0%)

12 (100%)

0.06

 Yes

7 (35.0%)

0 (0%)

 

Chronic GVHD

 None

17 (85.0%)

10 (83.3%)

0.83

 Mild

1 (5.0%)

0 (0%)

 

 Moderate

1 (5.0%)

1 (8.3%)

 

 Severe

1 (5.0%)

1 (8.3%)

 

Conclusions: mWT1 MN is associated with high risk of relapse post-transplant, especially in patients with multi-mWT1. Moreover, patients with mutations in hotspot codon 312-314 of exon 4 have inferior survival post-alloSCT. Further studies are needed to validate these findings.

Disclosure: Nothing to declare

19 - Acute Leukaemia

P017 VENETOCLAX COMBINED WITH LOW DOSE CYTARABINE AND ACTINOMYCIN D FOR RELAPSED ACUTE MYELOID LEUKEMIA OR MYELODYSPLASTIC SYNDROME WITH EXCESS BLASTS-2 AFTER ALLOGENEIC STEM CELL TRANSPLANTATION

Andrius Žučenka1, Dominika Vasilevska1, Vilmantė Vaitekėnaitė1, Igoris Trociukas1, Regina Pileckytė1, Rita Čekauskienė 1, Valdas Pečeliūnas1, Artūras Slobinas1, Renata Jucaitienė1, Inga Šlepikienė1, Linas Davainis1, Adomas Bukauskas1, Mindaugas Stoškus1, Eglė Gineikienė1, Vaidas Dirsė1, Rimvydas Norvilas1, Ugnė Ringelevičiūtė1, Kazimieras Maneikis1, Lukas Kevličius1, Birutė Davainienė1, Rūta Čepulytė1, Roberta Petrauskaitė1, Agnė Šiaudinienė1, Tomas Daukšas1, Karolis Šablauskas1, Emilė Žučenkienė1, Rūta Semaškevičienė1, Indrė Klimienė1, Andrius Degulys1, Skirmantė Černauskienė1, Austėja Dapkevičiūtė-Purlienė1, Sima Pakštytė-Danėlienė1, Ilma Tavorienė1, Lina Kryžauskaitė1, Guoda Daukėlaitė1, Laimonas Griškevičius1

1Vilnius University Hospital Santaros Klinikos, Vilnius, Lithuania

Background: Despite the recent advancement in acute myeloid leukemia (AML) therapy, almost half of the allotransplanted patients relapse and face poor prognosis. Promising results of various Venetoclax-based salvage therapies for relapsed AML or Myelodysplastic Syndrome with Excess Blasts-2 (MDS-EB2) in the post-allogeneic stem cell transplantation (alloSCT) setting have been published. However, the highest response rates and improved survival were demonstrated mainly after intensive chemotherapy and Venetoclax combination regimens which are highly myelosuppressive and can result in excessive toxicity, especially in the post-alloSCT patients.

Methods: We performed an observational, retrospective, single-center study. The patients were older than 18 years of age and had a morphologically relapsed AML or MDS-EB2 post-alloSCT. The salvage ACTIVE regimen consisted of either 1 or 2 cycles of Venetoclax 600mg/d from D1 up to D28 + Cytarabine 20mg/m2 D1-10 + Actinomycin D 12.5mcg/kg D1-3. The number of Venetoclax days per cycle was adapted individually based on the early bone marrow evaluations and toxicity but could not exceed 28 days per cycle. Concomitant Gilteritinib or Trametinib was administered in 2 cases harboring FLT3 or KRAS mutations. Responders proceeded to either second alloSCT or continued maintenance therapy with Venetoclax 600mg/d (D1-D14), Cytarabine 20mg/m2 (D1-5, or D1-10), and concomitant DLI in the absence of GVHD. We evaluated patients’ characteristics, composite CR (CRc = CR + CRi + CRp), overall response (ORR = CRc + MLFS), measurable disease negativity (MRD) rates, overall survival (OS), relapse-free survival (RFS) for responders, Grade 4-5 CTCAE v5.0 non-hematological toxicity, mortality rates within 60 days.

Results: 27 patients had been enrolled of whom 24 (89%) had AML, whereas 3 (11%) had MDS-EB2. 52% (14/27) were male, the median age was 59 (20-74) years and the median ECOG was 1 (0-3). 59% (16/27) of patients had secondary AML or MDS-EB2. 78% (21/27) of patients had been stratified to the adverse ELN2022 risk group. 41% (11/27) of cases had adverse cytogenetics of which 26% (7/27) were identified as complex-monosomal karyotypes. 81% (22/27) of patients had received intensive chemotherapy, 15% (4/27) had prior HMA exposure, and 19% (5/27) had previously received Venetoclax. The median number of previous therapy lines was 1 (1-5). RIC was used in 85% (23/27) of patients. The median time from alloSCT to relapse was 203 days (48-1215). The ORR of the ACTIVE regimen was 70% (19/27). The CRc rate was 67% (18/27), and the CR rate was 59% (16/27). MRD negativity was achieved in 56% (10/18) of CRc cases. 74% (14/19) of responders proceeded to maintenance therapy, 15% (3/19) underwent the second alloSCT, whereas 11% (2/19) relapsed prior to subsequent therapy. The median OS was 10.1 months (Figure 1A), and the median RFS was 8 months (Figure 1B). D30 and D60 mortality rates were 4% (1/27) and 7% (2/27), respectively. Both deaths were related to AML progression. Efficacy and toxicity results are summarized in Table 1.

Figure 1. Overall and relapse-free survival

Table 1. Efficacy and toxicity

Response evaluation

Overall response rate

70% (19/27)

Cumulative CR rate

67% (18/27)

CR

59% (16/27)

CRp

7% (2/27)

MLFS

4% (1/27)

Refractory/progressive disease

30% (8/27)

MRD negativity in CRc patients

56% (10/18)

Toxicity

Grade 4 non-hematological toxicity

7% (2/27)

Treatment-related deaths

0

Deaths within 60 days due to progressive disease

7% (2/27)

Conclusions: In the real-life setting, the ACTIVE regimen demonstrates promising anti-leukemic efficacy with manageable toxicities in AML and MDS-EB2 patients relapsing post-alloSCT.

Clinical Trial Registry: Observational trial No.2019/2-1088-591 approved by Vilnius Regional Biomedical Research Ethics Committee.

http://bioetika.sam.lt/index.php?1444926469

Disclosure: Funding: none.

Disclosures:

Žučenka: Abbvie: Consultancy, Honoraria, Travel Expenses; Astellas: Consultancy, Honoraria; Novartis: Consultancy, Honoraria, Travel Expenses; Pfizer: Consultancy.

Pileckytė: Abbvie: Consultancy, Honoraria, Travel Expenses; Griškevičius: Miltenyi Biomedicine: Membership on an entity’s Board of Directors or advisory committees.

19 - Acute Leukaemia

P018 INCREASED RELAPSE OF AML WITH HIGHER CYCLOSPORINE-A CONCENTRATION FIRST MONTH AFTER ALLOGENEIC STEM CELL TRANSPLANTATION WHEN COMBINED WITH ANTI-THYMOCYTE GLOBULIN

Mikael Lisak 1, Malin Niklasson1, Róbert Pálmason2, Per-Ola Andersson1, Jan-Erik Johansson1, Stig Lenhoff2, Markus Hansson1

1University of Gothenburg, Gothenburg, Sweden, 2University of Lund, Lund, Sweden

Background: Cyclosporine-A (CsA) is a well established prophylaxis against acute graft-versus-host disease (aGvHD) after allogeneic stem cell transplantation (HSCT). Higher CsA concentration, especially at time of engraftment, has been shown to reduce presence and severity of aGvHD. Some studies have shown correlation between higher CsA concentration and relapse of hematological malignancies.

There is no consensus on how early CsA concentration affects incidence of relapse of acute myeloid leukemia (AML) when combined with antithymocyte globulin (ATG). European Society for Blood and Marrow transplantation (EBMT) recommends CsA target serum concentration 200-300 µg/L during the first month post-HSCT as aGvHD prophylaxis, without distinction concerning parallell immunosuppressive drugs, e.g. ATG.

The aim of this study was to investigate whether a median CsA concentration the first month post-HSCT of >200 µg/L (CsAhigh) compared to ≤200 µg/L (CsAlow), increases the risk of AML relapse, when combined with ATG.

Methods: We collected data from 157 consecutive patients with AML who underwent HSCT 2010-2016 at three Swedish transplant centers. All patients were transplanted with unrelated donors (URD) and received ATG, besides a a short course of intravenous methotrexate.

CsA exposure was based on median concentration the first month post-HSCT of each patient.

Exclusion criterias were haploidentical donor, cord blood cells, conditioning with total lymphoid irradiation, alemtuzumab in conditioning or CsA treatment shorter than 30 days.

The risk of relapse at transplantation was categorized into low, intermediate or high risk according to the Swedish National Guidelines for AML.

The primary endpoint was the cumulative incidence of relapse at 60 months after HSCT.

Secondary endpoints were acute (aGVHD), chronic GvHD (cGVHD) and OS.

The study was approved by the Regional Ethic Review Board of Gothenburg (Dnr 144-18).

Results: The cumulative incidence of relapse up to 60 months in the CsAhigh and CsAlow group was 50% (95% CI, 38 – 62) and 32% (95% CI, 14 – 36; p = 0.016), respectively. In univariable analysis, CsAhigh vs. CsAlow (p = 0.028), 10-unit increase of CsA as a continuous variable (p = 0.017) and high risk disease (p = 0.003) were associated with cumulative incidence of relapse up to 60 months. The results remained after adjusting for disease risk. The mean CsA concrentation was lower amongst patients without relapse compared to those with relapse; 190.5 µg/l vs. 200.9 µg/l, (SE 3.78; 95% CI, -17.86 – -2.93; p = 0.0066).

When analyzing the CsA concentration as a non-linear risk factor, the chosen cut-off, 200 µg/L, seemed to be appropriate.

Relapse as cause of death was more frequent in the CsAhigh group (p = 0.0051). No significant differences were seen in rates of aGvHD, cGvHD or overall survival. Severe and life threatening aGvHD and GvHD-related deaths were more frequent in the CsAlow group, but the numbers to few to render significancy.

Table. Background characteristics (n = 157)

No (%)

CyA conc

≤200µg/L

n = 87

CyA conc

>200µg/L

n = 70

P value

Female gender

38 (44)

32 (46)

p = 0.80

Disease risk

Intermediate risk

26 (30)

21 (30)

p = 0.99

High risk

61 (70)

49 (70)

Stem cell source

Bone marrow

8 (9)

4 (6)

p = 0.67

Peripheral blood stem cells

79 (91)

66 (94)

HLA ≤ 7/8

8 (9)

7 (10)

p = 0.88

CMV IgG-pos recipients

65 (74)

48 (69)

p = 0.33

Conditioning

Reduced conditioning

52 (60)

32 (46)

p = 0.079

Myeloablative conditioning

35 (40)

38 (54)

Age at alloSCT,

median (range), yrs

56 (19 – 71)

51.5 (18 – 71)

 

Follow-up time, median, months

57.5 (95% CI, 52.9-64.6)

Conclusions: Our study, only including AML patients, shows that higher CsA exposure the first month post-HSCT is associated with increased relapse incidence, when combined with ATG. This has earlier been shown with alemtuzumab as T-cell depletion. The findings suggest that the EBMT recommendations concerning intended CsA concentration the first month should be nuanced, at least when treating T-cell depleted AML patients.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P019 GENOMIC LANDSCAPE OF MYELOID SARCOMA: A PLEA FOR BETTER MOLECULAR CHARACTERIZATION AND A ROAD MAP FOR TARGETED THERAPIES

Clément Collignon 1, Anne-Béatrice Notarantonio1, Gabrielle Roth Guepin1, Caroline Bonmati1, Maud D’Aveni1, Arnaud Campidelli1, Charline Moulin1, Caroline Jacquet1, Cécile Pochon1, Marc Muller1, Marion Divoux1, Francesca Ferraro2, Pierre Feugier1, Marie Thérèse Rubio1, Francesco Grimaldi3, Simona Pagliuca1

1CHRU de Nancy, Vandœuvre-lès-Nancy, France, 2Washington University in St. Louis, St. Louis, United States, 3University Federico II of Naples, Naples, Italy

Background: Myeloid sarcoma (MS) constitutes a rare, high-risk form of acute myeloid leukemia (AML), characterized by infiltration and destruction of otherwise normal tissues by leukemic blasts. This entity could occur de novo or at leukemia relapse, after chemotherapy or allogeneic hematopoietic cell transplantation, and may be isolated or concomitant to bone marrow (BM) infiltration. Small case series exploiting paired high-throughput sequencing of solid lesions vs BM show the presence of complex clonal architectures, sometimes divergent from the genetic structure of BM leukemia and, although some recurrent somatic lesions have been identified, clear genomic correlations are only putative and not investigated in larger cohorts with comparative studies.

Methods: Here we assembled a large metanalytic cohort of 70 patients with paired MS/BM molecularly profiled samples, recruited from our institutions (Nancy University Hospital and Federico II of Naples) and from 10 public studies. https://www.ncbi.nlm.nih.gov/sites/myncbi/simona.pagliuca.1/collections/62441058/public/

Results: In patients with MS, solid lesions, compared to BM, displayed a higher complexity due to an increased mutational burden (average mutational rate of 2.75 vs 1.9, p = 0.011), and a higher variant allele frequency (Median 33 vs 19%, p = 0.008). When paralleled with BM, mutational landscape of MS was characterized by a different configuration of myeloid lesions with enrichment in NPM1 (36 vs 17%), FLT3 (29 vs 15%), NRAS (19 vs 12%) and KRAS (9 vs 2%) mutations. The acquisition of some of these lesions in extramedullary localizations, initially not present in BM, occurred in 53% of the patients and concerned particularly NPM1 and FLT3 mutations. Intriguingly, some patients acquiring NPM1 in MS, presented with a complex karyotype AML, underlying the genomic instability of these disorders as compared to classical NPM1 mutated AML, usually associated with normal cytogenetics.

When comparing this mutational background with an assembled non-MS multi-study AML cohort,https://bit.ly/3hAncqy the enrichment in NPM1 mutations in MS lesions was more evident (p = 0.003). We also noticed a slight increment of co-occurrence of mutations in NPM1 and RAS genes (20% in MS vs 14% in non-MS AML vs 11% in BM of MS patients), while lesions in FLT3 and RAS genes were in all cases mutually exclusives, likely for overlapping deleterious effects on cellular functions.

Since leukemia sanctuaries in solid organs are supposed to exist because of failure of immune control, we reasoned that altered NPM1/FLT3/RAS pathways could share some molecular features prone to tissue invasion and immune escape. We explored this possibility at genomic level, by investigating the patterns of co-mutations in patients with myeloid neoplasms harboring lesions in these genes and profiled with whole exome sequencing (N = 8606). https://bit.ly/3W8CCB8 Gene enrichment analysis of the “altered” vs “unaltered” group showed an increased occurrence of mutations in genes involved in mTOR signaling, interferon alpha and gamma response, hypoxia and unfolded protein response that could play a role in the escape from anti-tumor surveillance in these molecular subsets of AML.

Conclusions: This comparative study sheds light on the unique molecular features of MS, pleading for the need to molecularly characterize all solid lesions occurring in AML context, and to explore targeted therapies in this high-risk category of patients.

Disclosure: No conflict of interest to disclose

19 - Acute Leukaemia

P020 CHIDAMIDE MAINTENANCE THERAPY FOLLOWING ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION FOR T CELL ACUTE LYMPHOBLASTIC LEUKEMIA

Xing-Yu Cao 1, De-Yan Liu1, Yue Lu1, Jian-Ping Zhang1, Jia-Rui Zhou1

1Hebei Yanda Lu Daopei Hospital, Langfang, China

Background: Relapse following allogeneic hematopoietic stem cell transplantation (allo-HSCT) remains a major challenge for T cell acute lymphoblastic leukemia (T-ALL) patients. Epigenetic abnormalities are common and histone deacetylase (HDAC)1 and HDAC4 are frequently overexpressed in T-ALL tumor cells compared to normal bone marrow samples. Chidamide, approved in China in 2014, is an HDAC inhibitor that has anti-tumor activity and the ability to enhance immune cell-mediated tumor cell cytotoxicity. Here, we explored whether Chidamide maintenance therapy after allo-HSCT for T-ALL patients is safe, could reduce relapse, and improve survival.

Methods: From June 2017 to May 2020, 12 patients with T-ALL who underwent transplantation at the Hebei Yanda Lu Daopei Hospital and received Chidamide as maintenance therapy post-HSCT were analyzed. Maintenance treatment with Chidamide was initiated when patients met the following criteria: 1) complete remission (CR) status prior to transplantation; 2) post-transplant hematological recovery and minimal residual disease (MRD) negativity.

Results: One patient was female (8.3%) and the median patient age was 14 years (range: 7-37). Five patients (41.7%) had positive fusion genes, including 1 (8.3%) with P2RY8-CRLF2, 1 (8.3%) with SET-CAN and 5 (25%) with STIL-TAL1 fusion genes. Before receiving allo-HSCT, 1 patient (8.3%) was in MRD-positive CR and the other 11 (91.7%) were in MRD-negative CR. Nine patients (75%) received haplo-identical HSCT and the other 3 (25%) underwent sibling-identical HSCT. All patients received total body irradiation (TBI)-based conditioning regimens. The median time of neutrophil and platelet engraftment was 12 days (11-19) and 12 days (7-37), respectively. Post transplantation, the median time to Chidamide administration was 183 days (range: 30-532). The dose of Chidamide ranged from 5 mg once a week to 15 mg twice a week, depending on the patient’s weight, hematopoietic functions and other conditions. The median duration of Chidamide therapy was 358 days (range: 74-762). The most common adverse event was a decrease in blood count. The 6-month cumulative relapse incidence (RI) was only 8.3% (95%CI,1.3-54.4%), and the 1-year, 2-year and 5-year RI were 16.7% (95%CI,4.7-59.1%), 25% (95%CI,9.4-66.6%) and 25% (95%CI,9.4-66.6%), respectively. The 2-year overall survival (OS) for all patients was 83.3% (95% CI,58.2-98.1%) and the 5-year OS was 66.7% (95%CI,38.8-89.3%). Two years post-transplant, no relapses occurred. The 2-year and 5-year leukemia free survival (LFS) were 66.7% (95% CI,38.8-89.3%).

The SET-CAN or STIL-TAL1 fusion genes are indicators of poor prognosis based on published papers. The 2 patients with STIL-TAL1 remained leukemia-free up to the last follow-up, and the patient with SET-CAN relapsed on day 389 post-transplant. Four patients died from diffuse alveolar hemorrhage (n = 1) and relapse (n = 3).

Conclusions: Our preliminary study showed that Chidamide maintenance therapy post allo-HSCT was safe for T-ALL patients, which did not increase adverse event frequency. Chidamide has the potential to improve T-ALL patient survival following allo-HSCT. Given the small cohort enrolled, long-term survival follow up is needed as well as additional studies. As we saw an increase in RI after six months and an RI plateau at 2 years after transplantation, we recommend initiation of Chidamide within six months following transplantation.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P021 AUTOLOGOUS STEM CELL TRANSPLANTATION FOR ADULTS WITH PHILADELPHIA-NEGATIVE ACUTE LYMPHOBLASTIC LEUKEMIA IN FIRST COMPLETE REMISSION. A STUDY BY THE ACUTE LEUKEMIA WORKING PARTY OF THE EBMT

Ryszard Swoboda 1, Myriam Labopin2, Sebastian Giebel1, Johan Maertens3, Elena Parovichnikova4, Annoek E.C. Broers5, Jane Apperley6, Grzegorz Helbig7, Anna Paola Iori8, Ludek Raida9, Alessandro Rambaldi10, Denis Caillot11, Jiri Mayer12, Dominik Wolf13, Mustafa Ozturk14, Mahmoud Aljurf15, Marie Thérèse Rubio16, Norbert Claude Gorin17, Francesco Lanza18, Mohamad Mohty19, Fabio Ciceri20

1Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice Branch, Gliwice, Poland, 2ALWP of the EBMT Paris Office; Sorbonne University, Saint Antoine Hospital, Paris, France, 3University Hospital Gasthuisberg, Leuven, Belgium, 4National Research Center for Hematology, Bone Marrow Transplantation, Moscow, Russian Federation, 5Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, Netherlands, 6Imperial College, Hammersmith Hospital, London, United Kingdom, 7Silesian Medical Academy, Katowice, Poland, 8Univ. La Sapienza, Rome, Italy, 9Olomouc University Hospital, Olomouc, Czech Republic, 10University of Milan and ASST Papa Giovanni XXIII, Bergamo, Italy, 11Hopital d`Enfants, Dijon, France, 12University Hospital Brno, Brno, Czech Republic, 13University Hospital Innsbruck, Innsbruck, Austria, 14GATA BMT Center, Gülhane Military Medical Academy, Ankara, Turkey, 15King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia, 16CHRU Nancy, Vandoeuvre les Nancy, Nancy, France, 17Sorbonne University, Hôpital Saint-Antoine, Paris, France, 18Romagna Transplant Network, Ravenna, Italy, 19Sorbonne University, Hospital Saint Antoine, Paris, France, 20San Raffaele Scientific Institute; Vita-Salute San Raffaele University, Milan, Italy

Background: The role of autologous stem cell transplantation (ASCT) in patients with Philadelphia-negative acute lymphoblastic leukemia (Ph- ALL) remains controversial. According to results of the prospective UKALLXII/ECOG E2993 study, ASCT should not substitute consolidative chemotherapy [Blood 2008:111:1827-33]. However, its role as late intensification in the era of routine testing of minimal residual disease has not been evaluated. The aim of this study was to analyze results of ASCT for adults with ALL and to identify prognostic factors.

Methods: Overall, 700 adult patients with Ph- ALL transplanted in first complete remission (CR1) between years 1999-2020 were included in this retrospective, multicenter study. Comparative analysis was performed between patients transplanted in 1999-2009 (n = 490) and 2010-2020 (n = 210).

Results: In the whole study population median patient age was 31.9 years (68% male). B-ALL and T-ALL was diagnosed in 35% and 65%, respectively. The median time from diagnosis to ASCT was 6.6 months. Total body irradiation (TBI)-based conditioning was used in 43%. Among 190 patients with available data, negative MRD status was reported in 167 (88%) cases.

The graft failure was observed in 1.3% of subjects. The probabilities of overall survival (OS), leukemia-free survival (LFS) at 2 years were 67% and 56%; relapse incidence (RI) and non-relapse mortality (NRM) were 39% and 5%, respectively. The most frequent causes of death were original disease (66%) and infections (13%). T-ALL was associated with lower RI (35% vs. 47%, p = 0.008) and higher LFS (60% vs. 49%, p = 0.01) at 2 years when compared to B-ALL. The older patient age ( ≥ 32 years) was associated with increased NRM (6.6% vs. 2.9%, p = 0.03) and decreased OS (63% vs. 71%, p = 0.02). In the multivariate analysis, the risk of relapse was reduced for T-ALL compared to B-ALL (hazard ratio (HR) = 0.72, p = 0.02) and with longer interval from diagnosis to ASCT (per month, HR = 0.95, p = 0.03). The risk of NRM was increased with patient age (per 10 years, HR = 1.38, p = 0.01). A chance of LFS was improved with longer interval from diagnosis to ASCT (HR = 0.95, p = 0.02), while a chance of OS was reduced with increasing patient age (HR = 1.14, p = 0.01). No impact of the study period (1999-2009 vs. 2010-2020) as well as the type of conditioning (TBI vs. chemotherapy alone) could be demonstrated with regard to any of the study end-points. MRD status prior to transplantation could not be included in the analysis due to insufficient data. Better results of T-ALL compared to B-ALL were demonstrated in the analysis restricted to patients treated between years 2010-2020 in terms of reduced risk of relapse (HR = 0.51, p = 0.03) and improved LFS (HR = 0.54, p = 0.02). LFS and OS rates for T-ALL in this period were 68% and 75% at 2 years, respectively.

Conclusions: Results of autologous stem cell transplantation in adults with Ph- ALL, especially for those with T-ALL are encouraging. Late intensification may be a valuable option for T-ALL as no humoral or cellular immunotherapies have been approved so far for this population. The latter, however, requires verification in prospective trials.

Disclosure: Nothing to declare

19 - Acute Leukaemia

P022 POTENTIAL MODIFICATION OF DYSREGULATED EXPRESSIONS OF INHIBITORY CHECKPOINT (ICM) AND THEIR LIGANDS ON T CELLS AND BLASTS IN AML RELAPSES AFTER STEM CELL TRANSPLANTATION (SCT)

Xiaojia Feng1, Giuliano Filippini Velázquez2, Sophia Bohlscheid1, Christoph Schmid2, Helga Schmetzer 1

1Munich University Hospital, Munich, Germany, 2Augsburg University Hospital, Augsburg, Germany

Background: Dysregulation of inhibitory checkpoint molecule (ICM) and ICM ligand (ICML) expression on T cells and leukemic blasts may represent a mechanism of immune-escape and clinical relapse of acute myeloid leukemia (AML) blasts after allogeneic stem cell transplantation (SCT). In vitro, culture of AML blasts in the presence of GM-CSF and PGE1 (KitM) can improve presentation of leukemic antigens and blast lysis by generation of dendritic cells of leukemic origin (DCleu). The effect of KitM based DCleu on ICM expression and cellular functions is unknown.

Methods: Flow cytometry analyses of expressions of ICM (PD-1/CTLA-4) and ICML (CD86/PD-L1/PD-L2) were performed on uncultured mononuclear cells (MC) in peripheral blood (PB) and bone marrow (BM) samples of 11 AML-patients at relapse after SCT and from PB of 7 healthy individuals (H).

Dendritic cells (DC) and DCleu were generated in culture in presence or absence of Kit-M (GM-CSF + PGE-1) followed by mixed lymphocyte culture (MLC) with patients’ uncultured T-Cells. After MLC, immune-activation and functionality (degranulation, intracellular cytokine production, lysis of the original blasts) were assessed. Clinically, all patients had received hypomethylating agents/Venetoclax (HMA/VEN) as salvage therapy. ICM/ICML expressions on T-cells and blasts were correlated with ex-vivo blast-lysis and patients’ clinical response.

Results: High expressions of ICM/ICML on uncultured blasts and T cells in AML: Patients presented with varying, but high frequencies of blasts co-expressing CD86/CTLA-4/PD-1(mean%(range): 44 (5.2-72)/49.7(2.73-99.67)/45.7(1.94-97.3)) and high frequencies of T-cells co-expressing CTLA-4/PD-1 (mean%(range): 68.9(47.2-86.7)/54.5(16.5-96.9)), whereas frequencies of PD-L1/PD-L2 co-expressing blasts/T-cells were low. T-cells of H-PB showed low expressions of CTLA-4/PD-1/PD-L1/PD-L2. We observed a negative correlation of ICM-expressions on uncultured patient T-cells with clinical response to HMA/VEN therapy.

Downregulated expressions of ICM on T-cells and improved anti-leukemic activities after MLC with Kit-M pretreated vs.untreated PBMC: Whereas there was no difference in the expression of ICM/ICML on blasts/DCs after culture with Kit-M when compared to blasts/DCs cultured without Kit-M, co-cultivation of Kit-M pretreated PBMC enriched with PTs’ T-cells resulted in reduced frequencies of ICM-expression on T-cells (CTLA-4 in 8/11 cases) after MLC. Increased leukemia specific degranulation, intracellular IFN production and improved blast lysis was found after MLC with vs. without Kit-M-pretreated PBMC. Nevertheless, improvement of anti-leukemic activity was negatively influenced by high CTLA-4/PD-1-expression of uncultured T-cells. Interestingly, this correlation was not observed after MLC with Kit-M-pretreated PBMC.

Conclusions: Uncultured T-cells and blasts of AML-PTs relapsed after SCT regularly co-express ICM (CTLA-4/PD-1). ICM expression on T-cells correlated negatively with PTs’ response to HMA/VEN and improvement of blast-lysis in a stimulation-based cell culture. Culture of patients’ PBMC with KitM did not alter ICM/ICML expression on blasts, whereas expression of CTLA-4 expression on T cells was downregulated after MLC. Functionally, culture with Kit-M improved specific blast lysis and seemed to overcome the negative influence on cellular immune reaction by ICM/ICML expression on blasts and T-cells.

Disclosure: H.M.S. is involved with Modiblast Pharma GmbH (Oberhaching, Germany) that holds the European Patent 15 801 987.7-1118 and US Patent 15-517627 ‘Use of immunomodulatory effective compositions for the immunotherapeutic treatment of patients suffering from myeloid leukemias’,

19 - Acute Leukaemia

P023 COMPARISON BETWEEN TRANSPLANT OUTCOMES AFTER CPX-351 VS OTHER INDUCTION REGIMENS IN NEWLY DIAGNOSED SECONDARY ACUTE MYELOID LEUKEMIA

Gabriele Facchin 1, Eleonora Toffoletti1, Marta Lisa Battista1, Michela Cerno1, Carla Filì1, Giulia Battaglia1, Giuseppe Petruzzellis1, Martina Pucillo1, Umberto Pizzano1, Antonella Geromin1, Renato Fanin1, Francesca Patriarca1

1Azienda Sanitaria Universitaria Friuli Centrale, Udine, Italy

Background: Allogeneic transplant (HSCT) remains the only curative therapy for secondary acute myeloid leukemia (sAML). Even if there are many evidence that CPX-351 induction has increased the HSCT rate, less is known regarding the impact of this induction regimen on post-transplant outcome.

Methods: In this study we compared outcome of 26 patients with diagnosis of sAML underwent HSCT after CPX-351 induction with an historical cohort of 27 patients receiving other induction regimen.

Results: Between January 2016 and June 2022, 53 patients with newly diagnosis of sAML underwent HSCT at our Center. Median age at transplant was 63 (23-74) years and the male/female sex ratio was 22/31. According to WHO2016 the sAML subtypes were AML with myelodysplasia-related changes/with prior myelodysplasia-MDS-(69%), AML therapy related (12%) or AML secondary to CMML/MPC (19%). Nine patients (17%) had received prior treatment with hypomethylating agents (HMAs) for MDS. Twenty-six patients (46%) received CPX-351 as induction therapy while 27 other regiments (12/27 fludarabine based, 7/27 HMAs ± venetoclax, 6/27 “3 + 7” standard chemotherapy, 1/27 other regimens). Thirteen (24%) patients received HMAs as bridge to transplant. Most patients received a graft from an unrelated donor (58%) following a myeloablative conditioning (58%). Patient and transplant characteristics were well balanced between CPX-351 and historical cohort except for higher proportion of HCT-CI > 2 (p = 0,0254) and higher median CD3+ infused cells from historical cohort (p = 0,0122). Numerically more patients treated with CPX-351 induction achieve a complete remission (CR) before HSCT (65% vs 44%). With a median follow-up of 11 months for the CPX-351 cohort and 23 months for the historical cohort, the median OS since transplant was not reached in the CPX-351 group while it was 16 months (p = 0.11, HR 2.052, 95% CI 0.8685-4.846) in the historical cohort. The 2-year OS is 73% vs 44% (p = 0,1152), respectively. There are no differences in terms of 2-years-progression free survival-PFS-(51% vs 35%; p = 0,2274), 2-years-transplant relate morality-TRM- (19% vs 16%, P = 0,9140), cumulative incidence of acute GVHD (54% vs 61%, p = 0,6967) and relapse rate (31% vs 55%, p = 0,1621). The most common causes of death were disease relapse (CPX-351: 1/26 [4%] vs 12/27 [44%] other regimens) and GVHD complications (4/26 [15%] vs 3/ 27 [11%]). Age >65 years, hyperleukocytosis at diagnosis and disease status prior to transplant were the only factors that influence 2-yr-OS in univariate analysis. In multivariate analysis only age and disease status influence survival.

Conclusions: In our real-life experience, 2-yr-OS for HSCT after CPX-351 induction was 73% vs 43% for historical cohort (p = 0,1152). No differences were found also in terms of 2-yr-PFS, TRM, incidence of aGVHD or relapse. We confirm that age and pre-transplant disease status are the main factors that impact on survival after HSCT in sAML. A multicenter real-life study, with a larger number of cases, is necessary to better understand the impact of CPX-351 on transplant outcomes.

Disclosure: no one.

19 - Acute Leukaemia

P024 LONG TERM OUTCOME OF ACUTE LYMPHOBLASTIC LEUKEMIA FOLLOWING ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANTATION: A 22 YEARS’ SINGLE CENTER EXPERIENCE

Mohammad Vaezi 1, Maryam Barkhordar1, Tannaz Bahri1, Mohammad Biglari1, Soroush Rad1, Sahar Tavakoli1, Ardeshir Ghavamzadeh1

1Hematology, Oncology and Stem Cell Transplantation Research Center, Tehran University of Medical Sciences, Tehran, Iran, Tehran, Iran, Islamic Republic of

Background: Allogeneic hematopoietic cell transplantation (allo-HCT) is a potentially curative treatment for acute lymphoblastic leukemia (ALL). As part of the conditioning regimen for ALL patients receiving allo-HCT, total body irradiation (TBI) has been proposed in particular. However, in many countries with insufficient funding, allo-HSCT patients commonly receive conditioning without TBI. This may cause concern for ALL individuals with a higher recurrence risk following allo-HCT. Therefore, the purpose of this study was to evaluate the outcomes of ALL patients who underwent myeloablative TBI-free allo-HCT throughout the previous two decades.

Methods: We retrospectively analyzed the trends of allo-HCT outcomes in 1120 patients who underwent their first allo-HCT for ALL between 2000 and 2022 at our single high-volume tertiary referral center. All recipients of allo-HCT received the same non-TBI myeloablative conditioning (MAC) regimen consisting of Bu (Busilvex) at 3.2 mg/kg i.v. on days − 6 through − 3, and cyclophosphamide 60 mg/kg on days – 3, – 2, followed by a unmanipulated granulocyte-colony stimulating factor (G-CSF)-mobilized peripheral blood stem cell source. The recipients of Haplo and unrelated allo-HCT received rabbit ATG 2.5 mg/kg (on days − 3, − 2, − 1) as a part of conditioning regimen. Post-transplant cyclophosphamide (PTCy) 40 mg/kg was given on days + 3 and + 4, in the haplo-HSCT. The combining ATG and PTCy with adjusted dosages was employed in haplo-HSCT after 2010.

Results: The period was divided into two intervals of before 2010 (D1) and after 2010 (D2) for analysis. The median follow-up times was 116 and 43.7 months for D1 and D2. Five-year overall survival (OS) improved significantly from 38.0% (95% CI: 33.23-42.69) before 2010 to 48.8% (95% CI: 44.28% to 53.21%) after 2010; (P = 0.0002). The three-year relapse incidence improved over the two decades from 53.5% (45.43% to 63.14%) to 38.9% (33.04% to 45.83%); (P = 0.009). However, we did not find any notable differences in non-relapse mortality (NRM) during the two decades. Additionally, in terms of HLA-match status, there was no statistically significant difference in the five-year OS between full matched (44.81%) and mismatched (33.74%) HCT; P = 0.093. In the first decade, relapse was the main cause of mortality, but in the second decade, infectious complications were the leading reason for death.

Patient characteristics by decade

Before 2010

After 2010

Total

Total (Number,%)

424

696

1120

37.90%

62.10%

100.00%

Remission status

CR1

260

437

697

37.30%

62.70%

100.00%

CR2

107

177

284

37.70%

62.30%

100.00%

CR > = 3

57

80

137

41.60%

58.40%

100.00%

Match status

Full match

408

593

1001

40.80%

59.20%

100.00%

Mismatch

16

103

119

13.40%

86.60%

100.00%

Patients Age

<16

83

175

258

32.20%

67.80%

100.00%

16-40

311

424

735

42.30%

57.70%

100.00%

> = 40

30

97

127

23.60%

76.40%

100.00%

Conclusions: We discovered that the whole cohort’s 5-year OS, when employing a non-TBI preparative regimen, remained at 44.01% (40.68% to 47.25%), which is within the range of 25 to 50% reported in previous research on the 5-year OS of ALL patients undergoing TBI-based allo-HSCT. Our research showed that the 5-y OS for ALL patients treated with the Bu-based, non-TBI MAC regimen has improved over the past 20 years as a result of a decline in the incidence of relapse. Importantly, to improve NRM and improve outcomes going forward, more efficient infectious complication prevention will likely be needed.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P025 SORAFENIB MAINTENANCE AFTER ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANT FOR FLT3+ ACUTE MYELOID LEUKEMIA: A SINGLE CENTER EXPERIENCE

Giuseppe Sapienza 1, Roberto Bono1, Stefania Tringali1, Marta Castronovo1, Lucia Sbriglio1, Chiara Russo2, Giulia Maria Camarda2, Valeria Calafiore1, Antonino Mulè1, Caterina Patti1, Alessandra Santoro1, Anna Marfia1, Cecilia Agueli1, Valentina Randazzo1, Luca Castagna1

1Azienda Ospedaliera Ospedali Riuniti Villa Sofia Cervello, Palermo, Italy, 2Azienda Ospedaliera Universitaria Policlinico ‘Paolo Giaccone’, Palermo, Italy

Background: Acute myeloid leukemia (AML) relapse is the main cause of death after allogeneic stem cell transplant (allo-SCT). In AML FLT3+, it was shown that Sorafenib (SOR) used as maintenance therapy after allo-SCT, significantly reduces the risk of relapse and death.

Methods: This is a retrospective analysis aimed to evaluate the feasibility, safety and outcome of maintenance with SOR in patients with FLT3+ AML who underwent allo-SCT. The majority of patients received a myeloablative conditioning regimen and peripheral blood stem cells. Donor types were equally distributed between HLAid sibling MUD and haploidentical donors. SOR was started at different time points after allo-SCT based on hematological reconstitution, performance status, infectious complications and disease evaluation. The starting dose was 400 mg/day. For survival analysis we used the Kaplan Meier method. All statistical analyses were performed using NCSS 2019 software.

Results: From 2017 to 2022, 43 patients with FLT3+ AML received allo-SCT. 95% (n = 41) were FLT3 ITD mutated and 5% (n = 2) TKD mutation. Median age was 53 (range 19-71). 62% (n = 27), 14% (n = 6) and 2% (n = 1) of patients had NPM1, IDH2 and DNMT3A mutations, respectively. 95% of patients were in CR (CR1 83%, CR2 9%, CR3 2%) and 80% were MRD negative at time of allo-SCT. 2 patients underwent allo-SCT with active disease. Grade 2-4 acute GVHD incidence was 38%. 51% (n = 22) of patients received SOR. The most frequent reasons for not starting SOR were relapse (19%) and GVHD (9%). Considering patients in CR before allo-SCT (n = 41), 8 patients (19%) relapsed before the start of SOR and median day of relapse was 93 days (range 30-153). Median day to start SOR was 120 days (range 59-279). SOR was discontinued because of adverse events in 18% (2 developed heart toxicity, 1 gastrointestinal toxicity and 1 skin toxicity). With a median follow up of 27 months, 3-year overall survival of the whole cohort was 65% and 90% for patients treated with SOR. No patients relapsed during SOR treatment. 10 patients completed the treatment and the median of SOR exposition was 731 days (range 575-752). 5 patients are still on treatment (median time of SOR exposition 422 days (27-638).

Conclusions: Post-transplantation SOR reduces the risk of relapse in FLT3+ AML. In our experience SOR was well tolerated even if 4 patients stopped the treatment because of toxicity. Early relapse, observed in 19% of patients, mostly in the first 3 months after allo-SCT, was the most important factor that impacted treatment feasibility. This underlines that the timing of SOR start should be as early as possible after allo-SCT.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P026 REDUCING THE DOSE OF BUSULFAN IN TBF CONDITIONIG, FOR AML OVER THE AGE OF 60: IS IT ENOUGH TO REDUCE TRANSPLANT RELATED MORTALITY?

Andrea Bacigalupo 1, Federica Sora2, Carmen di Grazia3, Anna Maria Raiola3, Alida Dominietto3, Francesca Gualandi3, Sabrina Giammarco4, Simona Sica4, Elisabetta Metafuni4, Eugenio Galli4, Emanuele Angelucci3

1Lazio, Roma, Italy, 2Andrea Bacigalupo, Roma, Italy, 3IRCCS Policlinico San Martino IST, Genova, Italy, 4andrea bacigalupo, Roma, Italy

Background: Transplant related mortlity (TRM) remains an important issue in patients with AML above the age of 60 undergoing an allogeneic hemopoietic stem cell transplant (HSCT). Reduced intensity regimens (RIC) have been used, although the problem is an increased incidence of leukemia relapse. One is therefore confronted with the opposing requirement of reduced intensity conditioning to control toxicity, but maintain some degree of myeloablation to control leukemia. We have used the combination of thiotepa, busulfan, fludarabine (TBF) with 3 days of busulfan (total dose 9.6 mg/kg) (TBF3) for patients under the age of 60. Over the age of 60, and for patients considered unfit, we have reduced the dose of busulfan to 3.2 mg/kgx2 (TBF2).

Methods: We are now reporting 107 patients with remission AML receiving TBF2 compared with 175 remission AML receiving TBF3. Donors were matched siblings, matched unrelated, 7/8 unrelated, and haploidentical.

Patients: All patients had acute myeloid leukemia in first (CR1) or second (CR2) remission. The proportion of CR1 was 75% for TBF3 and 69% for TBF2 (p = 0.5). The median age of TBF3 was 46 years (range 18-64), and for TBF2 it was 61 years (31-73) (p < 0.00001). HAPLO donors comprised 61% in both groups (p = 0.9). And post transplant cyclophosphamide GvHD prophylaxis was used in 71% and 72% of patients respectively (p = 0.7).

Results: The 8 years disease free survival (DFS) was 78% and for TBF3 and 42% for TBF2 (p < 0.0001). The Overal survival was 82% and 41% (p < 0.0001)- You will then ask the question: is this due to more relapse or more TRM in older patients receiving TBF2. The cumulative incidence of relapse at 8 years. was 12% for TBF3 and 16% for TBF2 (p = 0.5). The cumulative incidence of TRM at 8 years was 9% for TBF3 and 41% for TBF2 (p = 0.00001).

Causes of death were respectively for TBF3 and TBF2 as follows: relapse 7% and 10%; GvHD 3% and 7%; infections 3% and 15%; multiorgan failure 1% and 4%.

Conclusions: We conclude that the reduction of TBF3 to TBF2 for older patients is insufficient to control the toxicity of the conditioning regimen, whereas control of leukemia is satisfactory, and not statistically different from TBF3. The regimen needs to be further reduced in AML patients over the age of 60. Alternative options could be one day of busulfan (TBF1) with the addition of total marrow irradiation (Pieri et al, Blood Advance 2021).

Disclosure: The Authors declare no conflict of interest.

19 - Acute Leukaemia

P027 FACTORS ASSOCIATED WITH ACUTE MYELOID LEUKEMIA RELAPSING AFTER ALLOGENEIC STEM CELL TRANSPLANTATION

Juan José Domínguez-García 1, Guillermo Martín Sánchez1, María de las Mercedes Colorado Araujo1, Miriam Sánchez Escamilla1, Juan Manuel Cerezo Martín1, Carmen Montes Gaisán1, Enrique María Ocio San Miguel1, María Aránzazu Bermúdez Rodríguez1

1Hospital Universitario Marqués de Valdecilla, Santander, Spain

Background: Acute myeloid leukemia (AML) relapsing after first allogeneic hematopoietic stem cell transplantation (ASCT) has a dismal prognosis. The aim of our study is to identify risk factors associated with decreased overall survival (OS) and higher incidence of relapse post-ASCT and how can we manage this group of patients after relapsing.

Methods: We analyzed 135 patients who underwent first ASCT between January 2011 and December 2021. Patient demographic characteristics, AML and transplant related outcomes were retrospectively collected. We used cox regression for the statistical analysis.

Results: Patients’ characteristics and univariable overall survival (OS) and progression free survival (PFS) analysis are shown in Table 1.

With a median follow-up 92.6 months, two-year PFS and OS were 44.7% (95%CI: 36.6-53.6%) and 64.9% (95%CI: 56.0%-72.4%) respectively. Incidence of acute and chronic GVHD were 30.6% (95%CI: 18.9-47.1%) and 60.1% (95%CI: 34.4-86.6%), correspondingly.

Factors significantly associated with a decreased overall survival were age at ASCT, longer time from diagnosis to ASCT, higher disease burden at ASCT, absence of cGVHD (p < 0.001), sequential transplant (p < 0.001), monosomic or complex karyotype leukemia (p = 0.008) and early relapse ( < 1 year after ASCT). In the multivariate analysis, only age at transplant, absence of cGVHD development, and sequential ASCT maintained their significance.

Forty-five patients (33.3%) relapsed after ASCT. Four were molecular relapses, 10 were detected by flow cytometry and 31 morphologic relapses. Among them, 27 (60%) were very early-relapses ( < 6 months post-ASCT), 6 patients (13.3%) relapsed between 6 to 12 months post-ASCT and 12 (26.7%) beyond one year post-ASCT.

Factors associated with a PFS decreased (Table 1) were AML status at ASCT, peripheral blood as source, lack of acute or chronic GVHD development and monosomic or complex karyotype. Only control of disease and high DRI pre-ASCT, MUD and MMUD SCT, sequential conditioning, and lack of acute and chronic GVHD, maintained the statistical significance in the multivariable analysis

Patients with an overt morphologic relapse and those with an early relapse had significantly worse prognosis. With regards to the management of these relapses, 38% of patients received hypomethylating agents (65% in early-relapse), chemotherapy in 33% (47% in early-relapse), DLI infusion in 44% (50% in early-relapse), and a second transplant in 29% (XX in early relapse). The use of DLI or 2nd transplant was associated with a significantly better prognosis. In fact, in the multivariate analysis, use of 2nd ASCT (p = 0.001), early (p = 0.003) and morphological relapse (p < 0.001) were associated with worse OS.

Table 1.

Variables

SAMPLE

(n = 135)

PFS

HR (CI95%) p-value

OS

HR (CI95%) p-value

Age (y) at dx, mn (SD)

53.8 (13.96)

1.02 (0.99-1.03) 0.083

1.02 (1.01-1.04) 0.026*

Age (y) at ASCT, mn (SD)

54.4 (13.95)

1.02 (0.99-1.03) 0.068

1.02 (1.01-1.04) 0.020*

Time dx-ASCT, mn (SD)

235.5 (349.39)

1.00 (0.99-1.01) 0.052

1.01 (1.01-1.02) 0.002*

Female Sex, n (%)

59 (43.7)

0.69 (0.43-1.10) 0.114

0.60 (0.35-1.02) 0.061

ELN22

Favorable, n (%)

18 (13.4)

1.12 (0.54-2.33) 0.767

1.19 (0.54-2.60) 0.667

Intermediate, n (%)

74 (55.2)

Reference category

Reference category

Adverse, n (%)

42 (31.3)

2.09 (1.27-3.45) 0.004*

1.63 (0.93-2.86) 0.091

MRD pre-ASCT

MRD -, n (%)

56 (41.5)

Reference category

Reference category

MRD +, n (%)

65 (48.2)

2.29 (1.35-3.88) 0.001*

1.90 (1.05-3.45) 0.033*

PR, n (%)

5 (3.7)

3.73 (1.27-10.93) 0.017*

4.30 (1.44-12.87) 0.009*

REFRACT, n (%)

9 (6.7)

4.47 (1.89-10.60) 0.001*

5.29 (2.18-12.84) < 0.001*

PB source (vs. BM), n (%)

62 (45.9)

1.70 (1.07-2.71) 0.024*

1.47 (0.88-2.45) 0.143

ASCT

MSD, n (%)

31 (23.0)

Reference category

Reference category

Haplo, n (%)

40 (30.0)

1.55 (0.85-2.84) 0.154

1.14 (0.60-2.20) 0.685

MUD, n (%)

48 (35.6)

0.68 (0.36-1.31) 0.250

0.62 (0.31-1.23) 0.169

MMUD, n (%)

16 (11.9)

0.85 (0.37-1.93) 0.694

0.55 (0.21-1.45) 0.230

Relapse

≤6m Relapse, n (%)

27 (20.00)

NA

6.19 (3.38-11.33) < 0.001*

6m-1y Relapse, n (%)

6 (4.44)

NA

3.39 (1.29-8.90) 0.013*

>1y Relapse, n (%)

12 (8.89)

NA

0.64 (0.20-2.12) 0.471

Conclusions: In our 10-year series, it seems that a third of AML relapses after transplant and survival of these patients is highly related to their age, the very early relapse, absent of cGVHD, or high risk cytogenetic/molecular abnormalities. Achieving negative residual disease at transplant seems to improve the results and it must be the primary objective.

AML-relapsed patients, influenced mostly by type and moment of relapse, continues to be a major challenge. Prognosis is poor, and it seems that better results are offered with DLI or 2nd ASCT instead of hypomethylants or chemotherapy.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P028 TREATMENT WITH AZACITIDINE AND VENETOCLAX IS AN EFFECTIVE BRIDGE TO SECOND ALLOGENEIC TRANSPLANTATION IN PATIENTS WITH RELAPSED ACUTE MYELOID LEUKEMIA AFTER FIRST ALLOGENEIC TRANSPLANTATION

Nika Premuž1, Ljubica Mežnarić1, Lana Desnica2, Zinaida Perić1,2, Ranka Serventi Seiwerth2, Viktor Blaslov3, Hrvoje Holik4, Radovan Vrhovac1,2, Nadira Duraković 1,2

1University of Zagreb School of Medicine, Zagreb, Croatia, 2University Hospital Centre Zagreb, Zagreb, Croatia, 3University Hospital Centre Split, Split, Croatia, 4General Hospital Dr. Josip Benčević, Slavonski Brod, Croatia

Background: Venetoclax in combination with azacitidine (AZA-VEN) is approved for first line therapy of AML in unfit patients, while initial studies with this combination also showed good results in refractory/relapsed setting. Prognosis of patients in relapse after allogeneic transplantation (allo-HSCT) is dismal, and therapeutic options include DLI, subsequent chemotherapy and second allogeneic transplantation. Here we present outcomes of using AZA-VEN combination therapy in relapse post allo-HSCT in a single-center study.

Methods: We performed a retrospective analysis of 25 AML patients who were treated with AZA-VEN for relapse after allo-HSCT at University Hospital Centre Zagreb in the period between April 2019 and September 2022. Patient data was collected from the medical records and included: demographic data, treatment details, response, and subsequent treatment including second allo-HSCT. Kaplan-Meier method was used for survival analysis, and log-rank test was used for group comparison. All statistical results were obtained using Prism 7 program.

Results: Patient median age was 48 years (22-71), 11 (44%) were female and 14 (56%) male. All patients received prior allo-HSCT, and 23 (92%) were in CR prior to transplant. Nine (36%) patients received prior transplant from an unrelated, 10 (40%) from a related and 6 (24%) from a haploidentical donor. The median number of days from transplantation to relapse was 217 (range 26-1042) while median number of days from transplant to the start of AZA-VEN was 257 (range 57-1151). Median number of AZA-VEN cycles received was 2 (range 1-5). One patient died during first cycle so 24/25 patients were evaluable for response: 19 (79%) responded to treatment, of which 17 (71%) achieved CR/CRi. Best response was noted after a median of 1.67 months (range 1-6.6). Median survival for the entire group was 8.5 months. We compared survival proportions between patients that subsequently received second alloHSCT and those that did not, and significantly better survival was found in the group that was treated with 2nd transplantation, with median survival in that group reaching 14.2 months (p = 0.0003). Of 16 patients that did not reach or were not candidates for second transplantation 9 (56%) died due to AML, 1 (6%) of infection and 1 (6%) of other reasons (bleeding). Of 9 patients that reached second allo-HSCT 2 (22%) died of AML, 2 (22%) of infection and 1 (11%) due to transplant toxicity.

Conclusions: In this retrospective analysis we have shown that AZA-VEN is a good treatment option for patients with AML in relapse after allo-HSCT with response rates and time to response comparable to results published for AZA-VEN as the upfront therapy. It can be administered in most cases in the outpatient setting so it relieves some of the strain on inpatient services. However, difference in survival proportions between patients that were able to proceed to second transplantation timely and those that did not suggest that the duration of response is short and unless patients received subsequent transplantation their survival is rather limited.

Disclosure: Nothing to declare

19 - Acute Leukaemia

P029 IMPAIRED ACCESS OF UKRAINIAN PATIENTS TO HEMATOPOIETIC CELL TRANSPLANTATION (HCT) DURING THE RUSSIAN AGGRESSION IN 2022

Kateryna Filonenko 1, Ewa Paszkiewicz-Kozik2, Maciej Majcherek3, Olha Kylivnyk4, Maria Bieniaszewska5, Anna Czyz3, Wojciech Legiec4, Yana Stepanishyna6, Iryna Korenkova7, Oleksandr Lysytsia8, Iryna Kriachok6, Jan Zaucha5

1University Clinical Center, Gdansk, Poland, 2Maria Sklodowska-Curie National Research Institute of Oncology, Warsaw, Poland, 3University Clinical Hospital, Wroclaw, Poland, 4Center of Oncology of the Lublin Region St. Jana z Dukli, Lublin, Poland, 5Gdansk Medical University, Gdansk, Poland, 6National Cancer Institute, Kyiv, Ukraine, 7Kyiv City Center of Bone Marrow Transplantation, Kyiv, Ukraine, 8National Specialized Children’s Hospital “OHMATDYT”, Kyiv, Ukraine

Background: Ukraine, with 40 million (mln) citizens in 2021 performed about 250 (including 44 allogeneic) procedures of hematopoietic cell transplantation (HCT) for adult and pediatric patients, which covered 10% of the expected needs of 300 HCT/10 mln. For 11 months of 2022 the number of SCT procedures dropped-off by 30% (at least 204 procedures were performed). Despite that fact the number of allogeneic SCT procedures increased to 55. The full-scale war since February 24th 2022, significantly impaired the transplant activity in Ukraine. About 8,48 mln Ukrainian citizens entered Poland, from which 1,4 mln remained and received equal rights to Polish citizens, including rights to free healthcare. Very robust calculations indicate that at least 45 HCT should be performed to cover the needs of Ukrainian refugees.

Methods: We collected information regarding the HCT procedures performed in Ukrainian refugees in Poland in the period 24.02-31.11.2022 in 24 transplant centers in Poland. The questionary with basic data on patients’ demographic, family, clinical characteristics, treatment details, and patients’ needs was sent to the centers.

Results: Fifteen patients received HCT in 4 Polish transplant centers over the last nine months. Among them, 3 were male and 12 female, with median age of 32 (19-53) years. At least three patients were from the occupied territories or zones of active military actions, 5 patients were alone (without any family member to support them). Two patients had the diagnosis of AML, 2 - T-ALL, 1 patient had SAA, 7 - HL, 2 - DLBCL, and 1 – PTCL. One patient (with AML) was diagnosed in Poland, other 11 patients continued the treatment, which was started in Ukraine. 6 patients were after at least 2 lines of previous treatment in Ukraine, 8 patients had PD at the moment of the 1st hospitalization to Polish hospitals. Treatment that is unavailable in Ukraine received 3 patients. Auto-HCT was performed in 10 patients, and allo-HCT in 5 patients, including 3 from MUD, 1 from haploidentical donor, 1 from MRD. The biggest issue during the treatment was the language barrier. In 9 patients, the translator’s support was needed, including translations of documents. In 3 cases, hospitals helped with the accommodation and caring of dependent family members during the hospitalization. The median time of treatment in Poland was five months (3-9 months). In 9 patients full treatment plan was completed in Poland, six patients returned to Ukraine, and 2 patients died due to complications of allo-HCT.

Conclusions: The Russian aggression against Ukraine significantly impaired the access of the Ukrainian patient to HCT. Our data indicate that the number of HCT performed in Poland, the only European country offering refugee treatment equal to its citizens, does not cover the expected needs. All Ukrainian patients in Poland received treatment either unavailable or limitedly available in their home country. There is a great unmet need to improve the access of Ukrainian patients to HCT in Europe.

Disclosure: No COI

19 - Acute Leukaemia

P030 ADVERSE PROGNOSIS OF LOW AR FLT3-ITD IN AML AND CLINICAL BENEFIT OF ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANTATION IN AML PATIENTS WITH LOW AR FLT3-ITD AND NPM1

Hyeoung-Joon Kim 1, TaeHyung Kim2, Hoon Kook3, Ga-Young Song1, Seo-Yeon Ahn1, Sung-Hoon Jung1, Mihee Kim1, Deok-Hwan Yang1, Je-Jung Lee1, Seung Hyun Choi4, Mi Yeon Kim4, Chul Won Jung5, Jun-Ho Jang5, Hee Je Kim6, Joon Ho Moon7, Sang Kyun Sohn7, Jong-Ho Won8, Seong-Kyu Park8, Sung-Hyun Kim9, Zhaolei Zhang2, Jae-Sook Ahn1, Dennis Dong Hwan Kim10

Chonnam National University Hwasun Hospital, Hwasun, Korea, Republic of, University of Toronto, Toronto, Canada, Chonnam National University Hwasun Hospital, Hwansun, Korea, Republic of, 4Genomic Research Center for Hematopoietic Diseases, Chonnam National University Hwasun Hospital, Hwasun, Korea, Republic of, 5Division of Hematology-Oncology, Samsung Medical Center, Seoul, Korea, Republic of, The Catholic University of Korea, Seoul, Korea, Republic of, Kyungpook National University Hospital, Daegu, Korea, Republic of, Soon Chun Hyang University Hospital, Seoul, Korea, Republic of, Dong-A University College of Medicine, Busan, Korea, Republic of,, Princess Margaret Cancer Centre, University of Toronto, Toronto, Canada

Background: There have been questions about whether the acute myeloid leukemia (AML) with FLT3-ITD with low AR have a really favorable prognosis. This study analyzed the clinical impact of low AR FLT3-ITD and whether the patients with low AR FLT3-ITD would benefit from receiving allogeneic hematopoietic stem cell transplantation (HCT).

Methods: Totally, 624 AML patients who received intensive induction therapy from November 1996 to May 2019 were included in the study. Patients who achieved CR received consolidation chemotherapy with or without allogeneic HCT, depending on the availability of a matched donor. Genetic factors were not considered in choosing allogeneic HCT. About the AR of FLT3-ITD, low AR (ARlow) was defined AR < 0.5 and high AR (ARhigh) was defined AR ≥ 0.5.

Results: Among the 624 patients, 48 patients (7.7%) were ARhigh FLT3-ITD and 45 patients (7.2%) were ARlow FLT3-ITD. According to the ELN 2017 risk stratification criteria, 99 patients (15.9%) were adverse risk group, 276 patients (44.2%) were intermediate risk group, and 249 patients (39.9%) were favorable risk group.

In patients who have intermediate cytogenetics risk and did not received allogeneic HCT, overall survival (OS) and relapse-free survival (RFS) were inferior in patients with ARlow FLT3-ITD than the patients without FLT3-ITD (5yr OS 0.0% vs. 25.0%; 5yr RFS 0.0% vs. 22.8%) (OS, HR 2.009, 95% CI 0.939-4.296, p = 0.066; RFS, HR 2.848, 95% CI 1.178-6.883, p = 0.015). Cumulative incidence of relapse (CIR) was higher in patients with ARlow FLT3-ITD than the patients without FLT3-ITD (5yr CIR 100.0% vs. 36.9%) (HR 4.073, 95% CI 1.983-8.368, p = 0.001) There was no difference in cumulative incidence of non-relapse mortality (NRM) between the patients with ARlow FLT3-ITD and FLT3-ITD wild-type (5yr NRM 22.2% vs. 20.5%) (HR 1.122, 95% CI 0.251-5.025, p = 0.880). OS, RFS, CIR and cumulative incidence of NRM were similar between the patients with ARlow FLT3-ITD and with ARhigh FLT3-ITD (5yr OS 0.0% vs. 7.1%; 5yr RFS 0.0% vs. 7.1%; 5yr CIR 100.0% vs. 58.3%; 5yr NRM 22.2% vs. 7.1%) (OS, HR 1.301, 95% CI 0.510-3.317, p = 0.582; RFS, HR 1.267, 95% CI 0.433-3.704, p = 0.666; CIR, HR 1.444, 95% CI 0.551-3.782, p = 0.517; NRM, HR 3.273, 95% CI 0.344-31.120, p = 0.312).

Then we analyzed the clinical outcome according to allogeneic HCT in the patients with ARlow FLT3-ITD and NPM1 in patients with intermediate risk cytogenetics. Twelve patients who received allogeneic HCT showed superior OS, RFS and lower CIR than 6 patients who received only chemotherapy consolidation (5yr OS 75.0% vs. 0.0%; 5yr RFS 75.0% vs. 0.0%; 5yr CIR 8.3% vs. 100.0%) (OS, HR 0.157, 95% CI 0.028-0.888, p = 0.036; RFS, HR 0.126, 95% CI 0.023-0.697, p = 0.018; CIR, HR 0.045, 95% CI 0.005-0.422, p < 0.001). All 6 patients who received chemotherapy consolidation experienced relapse, but only 2 of the 12 patients who received allo HCT relapsed. Cumulative incidence of NRM was not statistically different (5yr NRM 16.7% vs. 0.0%) (p = 0.304).

Conclusions: AML patients with mutated NPM1 and ARlow FLT3-ITD have a worse prognosis than mutated NPM1 and FLT3-ITD wild-type, and the adverse prognosis of ARlow FLT3-ITD could be overcome by allogeneic HCT.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P031 IMPACT OF CYTOGENETIC ABNORMALITIES DEFINING AML MYELODYSPLASIA-RELATED (WHO AND ICC 2022) IN FIRST LINE DECITABINE IN UNFIT AML PATIENTS

Adolfo de la Fuente 1, Paola Beneit2, Pau Montesinos3, Ana Muñoz Gama4, Antonia Sampol Mayol5, Lucia Castilla Garcia6, Karla Javier Gonzalez7,8, Julio Davila Valls9, Alicia Roldan Perez10, Angela Figuera11, Javier Cornago12, Cristina Barrenetxea Lekue13, Itziar oiartzabal14, Sandra Suarez Ordoñez15, Susana Vivez Polo16, Carmen Martinez Chamorro17, Rosa Coll Jordá18, Montserrat Arnan Sangerman19, Alfonso Garcia de Coca20, Maria de los Angeles Foncillas21, Maria José Otero22, José María Grassa Ulrich23, Lara María Gomez Garcia24, Juan Antonio Vera Goñi25, Maria Perez Rios1, Antonio Diaz Lopez1, Jorge Sierra26, Mar Tormo27

1MD Anderson Cancer Center Madrid, Madrid, Spain, 2H. U. San Juan de Alicante, Alicante, Spain, 3H. U. Politecnico La Fe, Valencia, Spain, 4H. U. Virgen de la Salud Toledo, Toledo, Spain, 5H. U. Son Espases, Palma de Mallorca, Spain, 6H. U. Principe de Asturias, Alcala de Henares, Spain, 7H. General Universitario de Valencia, Valencia, Spain, 8The Hong Kong Politecnic University, Hong Kong, Hong Kong, SAR of China, 9Complejo Asistencial de Avila, Avila, Spain, 10H. Infanta Sofia, Madrid, Spain, 11H. U. La Princesa, Madrid, Spain, 12H. Fundación jiménez Díaz, Madrid, Spain, 13H. U. Basurto, Bilbao, Spain, 14H. U. de Araba Txagorritxu, Vitoria, Spain, 15H. U. Alvaro Cunqueiro, Vigo, Spain, 16ICO Hospital Germans Trias i Pujol, Barcelona, Spain, 17H. Quiron Salud Madrid, Madrid, Spain, 18ICO Gerona, Gerona, Spain, 19ICO Duran i Reynals, Barcelona, Spain, 20H. Clinico U. de Valladolid, Valladolid, Spain, 21H. Infanta Leonor, Madrid, Spain, 22H. Central de la defensa, Madrid, Spain, 23Quiron Salud, Zaragoza, Spain, 24H. Nuestra Señora de Prado, Talavera de la Reina, Spain, 25H. U. Virgen de la Macarena, Sevilla, Spain, 26Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 27H. U. Clínico de Valencia, Valencia, Spain

Background: HMAs in monotherapy have been the standard front line for unfit AML patients for a decade (Kantarjian H et al JCO 2012 and Dombret H et al Blood 2015). Risk factors for HMA frontline treatment have been suggested (De la Fuente et al EHA 2014 and EHA 2018). In 2022 the WHO and the ICC classifications for Myeloid Neoplasm have been published updating biological risk factors for AML. The aim of this study is to analyze the impact of the new 2022 Cytogenetic Abnormalities Defining AML Myelodysplasia-related (AML-MRC by WHO and by ICC) in the outcome of unfit AML patients treated with decitabine (Dec) as first line.

Methods: We carried out the analysis on patients with previously untreated AML included in the MDA-AML-2017-05 study from 23 Spanish sites. Inclusion criteria were as follows Age >18, diagnosis of AML under WHO criteria, treated with Dec during the period 01/09/2014 to 31/12/2016. We evaluated impact of the new 2022 Cytogenetic Abnormalities Defining AML Myelodysplasia-related (WHO and ICC). OS by Kaplan-Meier and the mortality within the first 8 weeks (M8wks). The MDA-AML-2017-05 study was approved by the Spanish Medicines Agency AEMPS.

Results: A total of 126p (77M, 49F) were analyzed. Average age 75.7 (48-91), 80 yrs and above 57p, WBC pre-Dec >15.000/µL: 21p, Creatinine>1.3 mg/dL: 23p, ECOG ≥ 2: 38p, adverse cytogenetic: 39p, AML-MRC by WHO 38p, AML-MRC by ICC 48p. A total of 716 cycles were analyzed, median 4 (1-31) per patient. No cases of treatment related mortality. One hundred and three patients were studied for effectiveness ORR 49% (CR 20p, PR 21p, ED 28p). With a mean follow up of 8 mths 77 died. The M8wks was 24.5% and the median OS 8 months.

AML-MRC defined as 2016 WHO classification resulted in non-significant differences for OS (p0.08). The new 2022 AML-MRC definition, both WHO and ICC resulted in significant differences in OS (p 0.005 and p0.03) and both predict M8wks 27% vs 12.3% and 26% vs 10.7% respectively.

The update of the DecLAM scale with the following risk factors: WBC pre-Dec >15.000/µL (p < 0.01) creatinine >1.3 mg/dL (p0.02), ECOG ≥ 2 (p < 0.01) and AML-MRC WHO 2022 (p < 0.01) allows us to identify risk groups with 0 vs 1-2 vs 3-4 risk factors with differences for OS (19 vs 5.5 vs 3.5 m p < 0.01). Age >80 revealed no OS differences.

Conclusions: The results of this study confirm AML-MRC 2022 by WHO and by ICC as adverse risk factor in unfit AML patients treated with Dec. The updated DecLAM scale is useful to identify risk groups with differences in OS,

Disclosure: The MDA-AML-2017-05 study was founded by Janssen in 2017. Present post-hoc analysis and abstract is an independent work carried out by co-authors without any financing.

19 - Acute Leukaemia

P032 FOLLOW-UP UPDATE OF THE FLAG-IDA REGIMEN AS BRIDGE THERAPY TO ALLOTRANSPLANTATION IN REFRACTORY/RELAPSED ACUTE MYELOID LEUKEMIA PATIENTS: 10 YEAR-RESULTS FROM A SINGLE CENTRE EXPERIENCE

Mario Delia 1, Vito Pier Gagliardi1, Paola Carluccio1, Corinne Contento2, Daniela Di Gennaro2, Wanda Strafella2, Immacolata Attolico1, Francesco Albano2,1, Pellegrino Musto2,1

1Hematology and Stem Cell Transplantation Unit, Azienda Ospedaliero-Universitaria-Consorziale (AOUC) Policlinico, Bari, Italy, “Aldo Moro” University – School of Medicine, Bari, Italy

Background: The outcome of refractory/relapsed acute myeloid leukemia (AML) is still dismal, and its treatment represents an unmet clinical need. Our group previously reported (Delia M., et al. Clin Lymphoma Myeloma Leuk.2017;17:767-773) the efficacy of the FLAG-Ida regimen in patients with refractory/relapsed AML as a bridge to transplantation demonstrating a median overall survival (OS) of 60 months for patients undergoing allogeneic hematopoietic stem cell transplantation (allo-HSCT) in complete remission (CR). In this study we performed a landmark analysis starting observation from the time of allotransplant updating the follow-up on June 30th, 2022, with regard to the post-allotransplant outcomes in terms of OS, disease free survival (DFS), non-relapsed mortality (NRM) and cumulative incidence of relapse (CIR).

Methods: The baseline characteristics of allotransplanted patients are reported in Table 1. In addition, time from diagnosis and FLAG-Ida to allotransplant was of 236 (r, 63-1061) and of 133 days (31-365), respectively. Seventeen CR-patients (33%) performed allotransplants in first CR (CR1). Source of stem cells was peripheral blood for 38 allotranplanted patients (73%). All patients received a myeloablative conditioning regimen: busulfan+cyclophosphamide; busulfan+fludarabine; thiothepa+busulfan+fludarabine (TBF) and TBF + post-transplant cyclophosphamide in 15 (29%), 11 (21%),16 (31%) and 10 (19%) allotransplants, respectively. A multivariate analyses of risk factors (disease status at FLAG-Ida (refractory to “3 + 7” or relapsed after CR), disease status at allotransplant, type of donor, recipient age, sex match) associated with OS, DFS, NRM and relapse was also performed.

Table 1. Baseline patients’ characteristics at allotransplantation n = 52

 

%

Age, median value, years (range)

47 (16-63)

 

Disease status at allotransplant

 compete response (CR)

30

57

 partial response (PR)

18

35

 active disease

4

8

Disease status at FLAG-Ida

 refractory to “3 + 7”

31

60

 relapsed after CR

21

40

Molecular-cytogenetics risk§

 poor

20

38

 intermediate

32

62

ITD

 mutated

12

23

 unmutated

40

77

Donor type

 Matched Related

18

35

 Matched unrelated

24

46

 Aploidentical

10

19

Results: The median OS significantly correlated with receipt of allo-HSCT (5 vs 23 months, p < 0.001). According to landmark analysis, the 10-year OS and DFS were 33 and 32%, respectively. With a median follow-up of 8 years (r, 0.5-15) for post-allotransplant surviving patients, the median OS and DFS for all series were 37 and 19 months, respectively. According to disease status at allotransplant, the 5-year OS and DFS rates in active disease-, PR- and CR-patients were 23, 12, 51% and 25, 5, 50%, respectively. The 10-year CIR and NRM rate were 57 and 25%, respectively. The 5-year CIR and NRM in active disease-, PR- and CR-patients were 72, 93 and 36% and 25, 29 and 22%, respectively. The 10-year CIR rates of CR-allotransplanted patients in CR1 and CR2 were 35 and 39%, respectively. The 10-year CIR rates of CR-allotransplanted patients belonging to the high and intermediate risk group were 55 and 26% (p=ns), respectively. The CIR of CR-patients was not conditioned by FLT3-ITD mutation status: 33 vs 37% of 10-year CIR in FLT3 unmutated and mutated-patients, respectively. In multivariate analysis, the disease status at allotransplant (CR vs not CR) was confirmed as the factor impacting OS (HR = 0.34, CI95%:0.171-0.685; p = 0.002), DFS (HR = 0.330, CI95%:0.164-0.664; p = 0.002), and relapse rates (HR = 0.262, CI95%:0.113-0.610; p = 0.002).

Conclusions: Our data confirm the value of FLAG-Ida as salvage therapy in relapsed/refractory AML and subsequent allotransplantation if CR is obtained. Furthermore, allotransplant performed in CR seems to counteract the impact of FLT3-ITD mutation in these patients.

Clinical Trial Registry: na

Disclosure: nothing to disclose.

19 - Acute Leukaemia

P033 ACUTE MYELOID LEUKEMIA AND ALLOGENEIC TRANSPLANTATION: SINGLE-CENTER EXPERIENCE

Rafael Colmenares1, Pilar Martínez-Sánchez1, José María Sánchez-Pina 1, Magdalena Corona1, Adolfo J Sáez1, Daniel Gil-Alós1, Rodrigo Gil-Manso1, María Calbacho1, Joaquín Martínez-López1

1Hospital 12 de Octubre, Madrid, Spain

Background: Acute myeloid leukemia (AML) was the disease suffered by 39% of allogeneic transplant patients in Europe, according to EBMT data in 2020. In fit patients, allogeneic transplant is the single curative therapy for AML, except classical favorable AML which could be cured only with chemotherapy if negative minimal residual disease (MRD). There is not an established MRD cut-off to receive an allogeneic transplant or more chemotherapy trying to reduce this MRD before transplantation.

Methods: A single-center retrospective analysis was carried out on 40 patients diagnosed with AML who went into an allogeneic transplant between January 2018 and October 2022. We used the 2017 ELN risk stratification and the ELN recommendations for the management of AML. We considered negative MRD if <0.1% (using multiparametric flow cytometry). We determined post-transplant MRD between day 25 and day 40.

Results: 23 (57.5%) patients were women, with a median age of 50 (range 19-67) years. 14 (35%) patients were referred from other centers. There were 5 (12.5%) secondary AML cases, and 9 (22.5%) patients had an HCT-CI/Age score greater than 2.

36 patients (90%) patients had achieved first complete response (CR1), although 5 (12.5%) were favorable-risk AML with positive MRD and 4 (10%) needed 2 chemotherapy-induction cycles; 4 (10%) had achieved second complete response (CR2). Regarding MRD, only 5 (12.5%) had positive pre-transplant MRD.

67.5% of patients received myeloablative conditioning (MAC) and 32.5% received reduced-intensity conditioning (RIC). There were 12 HLA-matched siblings, 18 haploidentical and 10 HLA-matched unrelated donors.

Only 1 patient (2.5%) had positive post-transplant MRD and, after a follow-up of 18 (range 2-57) months from the transplant there were 5 (12.5%) relapses (and all of them had negative pre-transplant and post-transplant MRD; 3 of patients had an adverse cytogenetic risk). 9 (22.5%) patients died; 6 patients with transplant-related mortality (TRM) (5 of them due to respiratory infection and 1 due to sinusoidal obstruction syndrome) and 3 patients with relapse.

Median overall survival (OS) was not reached in the cohort, and OS at 1 year and 2 years was 80%. No statical significative differences were found comparing patients regarding the type of donor or conditioning, but an HCT-CI/Age score of more than 2 had a worse OS (27 months; Log Rank p = 0,012).

Regarding patients with positive pre-transplant MRD, none of them relapsed during the follow-up, but one of patients is receiving azacytidine because of post-transplant positive MRD (Table 1). 4 (10%) patients needed a second allogeneic transplantation (3 because of graft failure and 1 due to relapse).

Patient

Pre-transplant MRD

Donor and conditioning

Post-transplant MRD

Outcome

#1

0.70%

Haploidentical-RIC

0.02%

Dead (respiratory sepsis, day 153)

#2

0.22%

Haploidentical-MAC

Not reached (death)

Dead (respiratory sepsis, day 35)

#3

0.50%

HLA-matched siblings-MAC

0%

Alive without relapse (follow-up, 28 months)

#4

0.90%

Haploidentical-RC

1.50%

Alive, receiving azacytidine, blasts < 5% (follow up, 4 months)

#5

2.00%

HLA-matched siblings-MAC

0%

Alive without relapse (follow-up, 4 months)

Table 1. Patients with positive pre-transplant MRD.

Conclusions: In our small sample, the vast majority of patients have a negative pre-transplant MRD, and none of the patients with a positive MRD relapsed during the transplant, although the results are not significant. The most frequent cause of mortality is TRM, so it may be important to apply comorbidity scales (such as HCT-CI/Age). More studies are needed to clarify the role of pre-transplant MRD in optimizing the timing of the allogeneic transplant.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P034 EARLY ALLOGENEIC STEM CELL TRANSPLANTATION AFTER MELPHALAN-INDUCED APLASIA LEADS TO LONG-TERM SURVIVAL IN PATIENTS WITH PRIMARY REFRACTORY AML

Claudius Söhn 1, Kerstin Schäfer-Eckart1, Knut Wendelin1, Wolfgang Hitzl2, Stefan Knop1

1Klinikum Nürnberg, Nürnberg, Germany, 2Paracelsus Medizinische Privatuniversität Salzburg, Salzburg, Austria

Background: Survival rates of younger patients with newly diagnosed acute myeloid leukemia (AML) have improved, but still up to 30% of patients (pts) fail to achieve complete remission after induction therapy. Allogeneic blood stem cell transplantation is the only option for long-term remission in eligible patients. Salvage chemotherapy before allogenic hematopoietic stem cell transplantation (HSCT) is considered as a standard of care by most centers. Yet this approach is associated with considerable treatment-related complications and only about 50% of patients will achieve CR before transplantation. Here we present our results of an immediate allogeneic transplantation in refractory disease after chemotherapy-induced aplasia.

Methods: All pts from our center with refractory AML after induction chemotherapy who received an allogeneic transplantation in active disease between 2012 and 2021 were retrospectively analyzed. Overall survival was evaluated in regard to time from diagnosis to HSCT using Cox proportional hazard model. The use of one induction cycle vs. two induction cycles after diagnosis was analyzed in regard to overall survival using Kaplan-Meier analysis.

Results: 41 pts were included. Their median age was 56 (range 24-75) years. Melphalan (100-140 mg/m2) for induction of aplasia was used in 36/41 pts. The conditioning regimen consisted of Treosulfan (30g/m2) and Fludarabine (150mg/m2) in most pts (68%). GvHD-prophylaxis was CSA and MTX or CSA and MMF. 34 pts (83%) additionally received ATG-prophylaxis. Mean time from diagnosis to HSCT was 3 (range 0.93-11.03) months (mos). Induction chemotherapy was daunorubicin and cytarabine (DA 3 + 7) in all pts. 17 pts received 1 cycle; 2 induction cycles were given to 24 pts, mostly because of study protocol. 22 of those 24 pts were given high-dose AraC as second induction because of refractory disease after first induction. After a median follow-up of 8.9 (range 0.3-122) mos 14 pts (34.1 %) were alive without relapse. Median observation time for those event free at the last follow up (14/41 patients) was 41 mos (range 7-122 mos). This was considerably longer than the median relapse-free survival of 9.2 mos. The last relapse took place 18 mos after

transplantation. The main cause of death was early treatment-related mortality (59 %) mainly because of septicemia (44 %). Only 7 pts (17 %) relapsed and 2 pts (7 %) had refractory disease. Cox proportional hazard model showed no significant correlation between time from diagnosis to HSCT and overall survival (p = 0.19). There was no significant difference of 1 vs. 2 Induction cycles in regard to OS (p = 0.51).

Conclusions: In this retrospective analysis, we showed that an immediate allogeneic transplantation after induction of aplasia is feasible in pts with primary refractory AML. With a two-year survival rate of 37%, this approach offers a realistic chance to patients with an otherwise dismal prognosis. The effectivity is compromised by the high rate of early toxicity mainly because of infectious reasons.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P035 OUTCOMES OF ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANTATION IN ADULT ACUTE MYELOID LEUKEMIA PATIENTS: A 30-YEAR SINGLE CENTER EXPERIENCE

Tanaz Bahri 1, Maryam Barkhordar1, Hossein Kamranzadeh1, Soroush Rad1, Sahar Tavakoli1, Ardeshir Ghavamzadeh2, Mohammad Biglari1, Mohammad Vaezi1

1Cell Therapy and Hematopotic Stem Cell Transplantation Research Center, Research Institute for Oncology, Hematology and Cell Therapy, Tehran University of Medical Sciences, Tehran, Iran., Tehran, Iran, Islamic Republic of, 2Cancer & Cell Therapy Research Center, Tehran University of Medical Sciences, Tehran, Iran, Tehran, Iran, Islamic Republic of

Background: Acute Myeloid Leukemia (AML) is a heterogenous disease with until recently a standard treatment which was unchanged for decades. We began allogeneic transplantation from 1991 in our center with full matched and 1 locus mismatched donors, with the introduction of haplo-donors in 2009-2010 there was a alternative source for patients with no matched donor. We performed this study to verify the efficacy of our treatment and identify further challenges.

Methods: Patients with AML who were transplanted at our center between 1991 and January 2022 were included in the study. Unfortunately due to lack of access to cytogenetic and molecular laboratory services we were unable to complete the risk stratification of the patients. We divided the patients in 2 groups of before and after 2010. The conditioning regimen was myeloablative and consisted of Busulfan 0.8 mg/kg QID for 4 consecutive days (12.8 mg/kg in total) and cyclophosphamide 60 mg/kg per day for two consecutive days. GVHD prophylaxis consisted of cyclosporine in combination with a short course of methotrexate. Patients transplanted from a matched unrelated or mismatch donor received also ATG. Patients with a haplo-HCT received Cyclosporine, ATG and post-CY on days +3 and +4 as GVHD prophylaxis.

Results: We retrospectively analyzed 1337 patients with AML, from 16 to 70 years of age. The median age of patients was 34 years (range 25-44), median follow-up was 61,6 months. The 5-year overall survival (OS) was respectively 61.5% in CR1 and 47.4% and 37.2% in patients transplanted in CR2 and CR³3. In the first complete remission the 5-year overall survival (OS) did not improve in the second timeline (64.1% (D1) vs 60.2% (D2)), but in CR2 the OS did improve from 36.7% (D1) to 54.8% (D2) and in CR ³ 3 from 28.3% (D1) to 47.0% (D2). Relapse incidence was less frequent in patients who transplanted in CR1 25.8% compared to CR2 (51.9%) and CR³ 3 (62.4%). Disease free survival did not improve in patients transplanted in the first CR (60.9% (D1) vs 54.6% (D2)) but it did improve in the second 34.4% (D1) vs 50.1% (D2)) and third CR (29.0% (D1) vs 43.0% (D2)). NRM differed significantly by match status: 26.1% in MRD and 42.3% in MUD and 46.7% in haplo-donors.

Conclusions: Allogeneic hematopoietic cell transplantation has evolved considerably since the introduction of haplo-transplantation. Our data suggests that the OS and DFS of patients with AML who received haplo-HSCT were significantly worse from MRD- and MUD-HSCT. NRM is significant higher in haplo-transplants due to more infections post-transplant and more GVHD. The patients transplanted in CR1 have the best outcome and the longest OS. We are aware that a better and more precise risk assessment with cytogenetic and molecular studies and detection of Minimal Residual Disease will help to identify the high risk patients more accurately and provide a better treatment outcome by choosing the right treatment strategy for each patient.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P036 POST ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION MAINTENANCE WITH AZACITIDINE IN COMBINATION WITH VENETOCLAX IN T-CELL ACUTE LYMPHOBLASTIC LEUKEMIA

Nour Moukalled 1, Mona Hassan1, Jean El Cheikh1, Ali Bazarbachi1, Iman Abou Dalle1

1American University of Beirut Medical Center, Beirut, Lebanon

Background: Relapsed/refractory (R/R) T-cell acute lymphoblastic leukemia (T-ALL), is an aggressive disease with few salvage options. Little progress has been seen, where nelarabine is the only approved drug (overall response rate of 50%; associated hematologic/neurological toxicities). Allogeneic-hematopoietic-stem-cell-transplantation(allo-HSCT)continues to be a potentially curative strategy in this setting, however up to half of patients still relapse post-transplant. Post-transplant maintenance treatment has long been debated, and has not been extensively explored in T-ALL, despite previous preclinical/clinical data suggesting a potential role for hypomethylating agents (HMAs) such as 5-azacitidine, in addition to suggested sensitivity to BCL2-inhibition exerted by venetoclax, paving the way for the rationale of using HMAs/venetoclax as post-transplant maintenance in high-risk T-ALL.

Methods: Case series: Four adult patients diagnosed with high-risk T-ALL received post allo-HSCT maintenance-5 days of 5-azacitidine (32 mg/m2 daily) with venetoclax 400 mg daily every 28 days, with a median duration of treatment of 18 months with a very good toxicity profile. After a median follow-up of 22 months from transplant, all patients remain in complete remission (CR).

Results: Patient 1: Mature T-ALL, with relapsed disease post hyperCVAD and 2 years maintenance (POMP protocol), received induction (augmented-BFM), achieved CR with negative measurable residual disease (MRD) by T-cell-receptor-(TCR)-rearrangements, underwent a full-matched-related-allo-HSCT (clofarabine/total body irradiation (TBI)-4 Gy conditioning). Maintenance treatment started day 42 post-transplant. Bone marrow 100 days post-transplant showed CR, undetectable MRD. Transplant course complicated by grade-1 skin graft-versus-host-disease (GVHD), managed by topical steroids. Patient is currently 32 months post-transplant, continuing maintenance therapy, remains in CR.

Patient 2: ETP-ALL (mutant NOTCH1/EZH2), received induction with hyperCVAD with persistent MRD (TCR-rearrangements), received re-induction with augmented-BFM/venetoclax, had undetectable TCR-rearrangements, received a haploidentical-HSCT (fludarabine/TBI-8Gy conditioning). Maintenance started day 53 post-transplant; day 100 post-transplant evaluation revealed detectable TCR rearrangements, received donor lymphocyte infusion (DLI), complicated by grade 2 gastro-intestinal and skin GVHD responsive to systemic steroids. Patient is currently 24 months post-transplant with chronic skin GVHD, continuing his maintenance treatment, last disease evaluation showing CR with detectable MRD.

Patient 3: Mature T-ALL (complex cytogenetics/EZH2/KIT/TET2 mutations), received induction hyperCVAD, achieved CR with detectable MRD, then received Capizzi regimen/venetoclax followed by matched-related allo-HSCT (clofarabine and TBI-8GY conditioning). Post-transplant course was complicated by mild skin-GVHD. Maintenance treatment started at day 115 post-allo-HSCT, continuing for 20 months to date with very good tolerance. Last bone marrow evaluation showed CR but detectable TCR MRD, received two DLI doses complicated with liver GVHD responsive to systemic corticosteroids.

Patient 4: Near-ETP ALL (normal karyotype-ASXL1/NOTCH1/TET2 mutations), received asparaginase-based treatment (GRAALL regimen), achieved CR with negative MRD, then completed two consolidation cycles, early-intensification followed by a matched-related allo-HSCT (clofarabine/TBI-8Gy conditioning). Maintenance started at day 46 post-transplant without significant toxicity. Patient is currently 18 months post-transplant and still in CR with undetectable MRD.

Two additional patients with T-ALL post-allo-HSCT in CR2, MRD positive, have been recently started on maintenance (1 cycle), remain in CR with detectable TCR.

Conclusions: Encouraging results presenting a foundation for further studies that could assess the efficacy of HMAs combined with BCL2 inhibitors not only in the post-transplant setting but also for high-risk T-ALL patients.

Clinical Trial Registry: NA

Disclosure: No conflict of interest.

19 - Acute Leukaemia

P037 POST-INDUCTION AKT EXPRESSION DOES NOT IMPACT OVERALL SURVIVAL IN PATIENTS WITH ACUTE MYELOID LEUKEMIA

Safa Al-Amrani 1, Fahad Al-Zadjali1, Yasir Jeelani1, Zaaima AL-Jabri1, Mahdiya Al-Bulushi2, Mohammed AlRawahi1, Adhari Al Zaabi1, Jalila AlShekaili1, Mohammed Al-Huneini1, Murtadha Al-Khabori1

1Sultan Qaboos University, Muscat, Oman, 2Royal Hospital, Muscat, Oman

Background: The activation of the AKT signaling pathway is crucial for cellular survival, proliferation, and regulation of apoptosis in leukemias and other types of cancers. Therefore, this pathway may contribute to the survival of cancer cells. The impact of differences in AKT expression between diagnosis and remission states on patients’ survival has not yet been investigated. In the current study, we aim to assess the impact of AKT expression at diagnosis and the difference in expression between diagnosis and remission on overall survival.

Methods: Patients with acute myeloid leukemia (AML) were included in this study. Cell lysates of peripheral blood and bone marrow of these patients were collected at diagnosis and remission. The expression of AKT in these samples was estimated using ELISA assay and normalized to the total protein content. Overall survival was estimated using the Kaplan-Meier curve and groups compared with the COX regression model. All analyses were performed using the R program.

Results: A total of 17 patients (a total of 68 peripheral blood and bone marrow samples) were included with a median age of 47 years (IQR: 23 – 57). At diagnosis, the median hemoglobin, white blood cells, and platelet counts were 8.0 g/dL (IQR: 6.0 – 9.0), 56*109/L (36 – 68), and 33*109/L (8 – 53), respectively. The median blast percentage was 50% (IQR: 26 – 77) in peripheral blood and 64% (IQR: 53 – 83) in bone marrow samples. In the peripheral blood, the median AKT expression was 564 (IQR: 499 – 1347) at diagnosis and 903 (IQR: 117 – 1270) at remission whereas, in bone marrow, it was 520 (IQR: 388 – 698) at diagnosis and 816 (IQR: 489 – 20290) at remission.

During follow-up, nine patients died. The overall survival was not affected by the AKT expression at diagnosis in peripheral blood (HR 0.8, p = 0.83) or bone marrow cell lysates (p = 0.999). In six out of the 17 patients, the AKT expression decreased in peripheral blood samples at remission. The reduction did not predict the overall survival in these patients (HR 0.5, p = 0.67).

Conclusions: Although AKT expression may be high in patients with AML, the level of expression does not predict overall survival. In addition, the reduction of the expression in post-induction samples does not impact overall survival and hence quantification using ELISA assay may not be a useful candidate to test a minimal residual disease. The study is limited in sample size and needs to be confirmed in future studies with larger sample sizes.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P038 ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION IN ACUTE LYMPHOBLASTIC LEUKEMIA PATIENTS WITHOUT TOTAL BODY IRRADIATION: A SINGLE-CENTER EXPERIENCE

Fabio Serpenti 1, Giorgia Saporiti1, Maria Goldaniga1, Giulia Galassi1, Filippo Bagnoli1, Francesca Cavallaro1, Kordelia Barbullushi1, Alessandro Bosi1, Maria Rita Sciumè1, Nicola Stefano Fracchiolla1, Wilma Barcellini1, Francesco Onida1

1Hematology Unit, Fondazione IRCCS Ca Granda Policlinico di Milano, Milan, Italy

Background: Allogeneic stem cell transplantation (HSCT) is the only curative option for very-high risk or relapsed/refractory acute lymphoblastic leukemia/lymphoma (ALL). Total body irradiation (TBI) is a pivotal component of conditioning regimens for ALL patients <50 yo, especially for its greater disease control and lower relapse risk, but its toxicity and logistic organization can limit its use. We aimed to report outcomes of ALL patients treated with non-TBI conditioned HSCT at our center.

Methods: We retrospectively analyzed all consecutive non-TBI conditioned HSCT performed at our center for ALL patients between 2014 and 2021. Data on disease risk, treatment lines, disease status at transplant and transplant outcomes were recorded. A written consent was given for the use of medical records for research in accordance with the Declaration of Helsinki. Multivariate analysis using Cox regression model was used to detect variables impact on overall survival. Variables included in the model were age, diagnosis (T-ALL vs Ph- B-ALL vs PH+ B-ALL), donor (HLA-id vs MUD vs haplo) and conditioning intensity (MAC vs RIC).

Results: Seventeen patients received non-TBI HSCT for ALL between January 2014 and December 2021 at our center. Patient characteristics are summarized in table 1. Median age was 48 years old (range 31 – 68). Median Comorbidity Index score was 2 (range 0-5). Only one patient had not reached complete remission before transplant. Reasons for avoiding TBI were age >50 yo in 53%, comorbidities in 47% and logistics in 17,6% of cases. Conditioning regimen was myeloablative in 15 cases, with two alkylating agents in 14. GvHD prophylaxis was ATG-based in 11 patients, ptCy-based in 5 and only-Cya based in 1. Median time to neutrophil engraftment was 19 days (12-30). Median time to platelet engraftment was 14 days (11-45), with two patients never achieving it. Febrile neutropenia occurred in all but one patient, viral infection/reactivation in 7 and fungal invasive infection in 3. There was no case of veno-occlusive disease or acute GvHD and there were three cases of moderate chronic GvHD.

After a median follow-up of 28 months (3-79), 2y-overall survival was 61,9% and relapse-free survival was 32,4% (Fig. 1). TRM was observed in one patient and was due to a post-transplant lymphoproliferative disorder. Relapse occurred in 10 patients, six of whom died due to progressive disease. Median time from transplant to relapse was 5 months (1-15). Donor lymphocyte infusions were used in three cases of mixed chimerism, leading to full chimerism reconstitution, and in combination with other agents in five relapsed patients, leading to long-lasting disease control in two.

Multivariate analysis indicated RIC as the only variable highly associated with worse OS (HR = 14, p < 0,1).

Sex (M/F)

9/8

Disease

• T-ALL

3

• B-ALL Ph- / very high risk B-ALL Ph-

7 / 1

• B-ALL Ph+

7

Disease status at transplant

• CR1 / CR 2 / > CR2

7 / 5 / 4

• Progressive disease

1

Previous InO / Blina

5 / 7

Donor

• HLA-id

4

• MUD 10/10 / 9/10

6 / 2

• HAPLO

5

Conditioning regimen

• TEC

3

• TTCy

3

• TTF

5

• TFM

3

• TreoFlu

1

• FluMel

2

Conclusions: Older high-risk ALL patients can still benefit from HSCT despite TBI avoidance; in our series, around one third of patients was alive and relapse-free at 2 y from transplant and relapsed patients could benefit of salvage treatment coupled with DLI, which led to long-term disease control in 2 cases. Toxicity was low with only one TRM case (5%). RIC was the only variable associated to poor outcome.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P039 MYELOID SARCOMA OF THE BREAST IN POST-ALLOGENEIC STEM CELL TRANSPLANT RELAPSE: RETROSPECTIVE ANALYSIS FROM A REFERENCE TRANSPLANT CENTER

Evelyn Zapata 1, Pablo Garcia1, Carlos Puerta1, Ana Garcia1, Monica Cabrero1, Monica Baile1, Alejandro Avendaño1, Cristina De Ramon1, Almudena Cabero1, Ana Martin1, Fermin Sanchez1, Estefania Perez1, Lucía Corral1, Lourdes Vazquez1

1Hospital Universitario de Salamanca, Salamanca, Spain

Background: Allogeneic stem cell transplantation (AlloSCT) is a recognized curative therapy for higher risk acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS). Unfortunately, post-transplant relapse is still a challenge. Relapse as myeloid sarcoma is a rare condition which most common locations are skin, lymph nodes or gastrointestinal. Breast involvement is a rare location for a post-AlloSCT relapse, presenting sometimes as an isolated location that could be misdiagnosed as a breast carcinoma. Further understanding of these extramedullary relapses and its mechanism of escape from donor immune system is mandatory to a better management.

Methods: Here we present a series of 5 patients with AML relapsing post-AlloSCT with breast involvement, selected from a total of 603 patients receiving and AlloSCT in our center from 2015, resulting in an incidence of 0.83%.

Results: All breast relapses occurred in women, with a median age of 38 (24-57) years. Diagnosis was AML in 2 and MDS in 3 patients. None of them had previous diagnosed breast pathology or extramedullary myeloid disease. All patients were in complete remission (CR) at AlloSCT and received myeloablative conditioning regimen. Donor was related in 4 and GVHD prophylaxis was based on methotrexate plus tacrolimus. All patients were in CR at day +100 and 3 of them maintained CR 1 year after AlloSCT. Only 1 patient developed chronic GVHD after AlloSCT and she achieved CR with first line steroids treatment.

Relapse occurred after a median of 40 months (7-99) after AlloSCT. Breast infiltration occurred in 1st relapse in 3 patients, and 2nd and 3rd relapse in 1 each. Two patients were diagnosed after initial suspicion of breast carcinoma with no other extramedullary locations affected, whereas 3 patients had blast infiltration in pericardium, muscle and bone. Two patients had bone marrow infiltration with 0.89 and 15% blasts respectively.

Treatment at relapse was chemotherapy based on schema Flag-Ida (n = 2), 5-azacitidine (n = 1) or Vyxeos® (n = 1) combined with radiotherapy in 2 of them. One patient received palliative care. All 4 treated patients achieved CR; 3 of them underwent a 2nd AlloSCT 3, 5 and 13 months after relapse, and it is planned for the 4th one at the moment of the analysis. One patient relapse 7 months after 2nd transplant and died from progression disease, whereas 3 patients are alive after 7, 64 and 75 months.

We retrospectively collected stored samples from patients at diagnosis and at relapse to evaluated clonal evolution. By now, one sample have been analyzed showing that clonal infiltration of the breast is related with the one at diagnosis in bone marrow with acquisition of NRAS y PTPN11 that could be responsible for extramedullary relapse (this patient had an isolated breast relapse).

Conclusions: Myeloid sarcoma of the breast in a rare condition after AlloSCT and differential diagnosis with breast carcinoma is mandatory since it can be the only relapse location.

Treatment based on chemotherapy combined or not with radiotherapy followed by a 2nd AlloSCT can lead to long-term survival in selected patients.

Disclosure: Nothing to declare

19 - Acute Leukaemia

P040 SEQUENTIAL REGIMEN AND REDUCED-INTENSITY CONDITIONED TRANSPLANTATION FOR ACUTE MYELOBLASTIC LEUKEMIA AND MYELODYSPLASTIC SYNDROMES. UNICENTRIC EXPERIENCE

Alejandro Avendaño Pita 1, Monica Cabrero Calvo1, María Cortés Rodríguez1, Pablo García Jaen1, Evelyn Zapata Tapia1, Carlos Puerta Vázquez1, Ana Garcia Bacelar1, Marta Fonseca Santos1, Cabero Martínez Almudena1, Mónica Baile González1, Ana África Martín López1, Estefania Pérez López1, Fermín Sánchez Guijo1, Lourdes Vázquez López1, Belén Vidriales Vicente1, María Diéz Campelo1, Lucía López Corral1

1University Hospital of Salamanca, Salamanca, Spain

Background: Acute myeloblastic leukemia (AML) and myelodysplastic neoplasms (MDS) are the two main indications of allogeneic stem cell transplantation (ASCT) in the present time. Refractory disease is a poor scenario and even with new drugs, in some patients, we don’t reach any response so sequential conditioning is an option to proceed to transplant.

Methods: We retrospectively analysed 25 consecutive patients collected during the last 15 years in our institution. Statistics analysis were performed with IBM SPSS statistics v.26.

Results: Clinical and ASCT characteristics are shown in table 1 and table 2. Three patients (12%) were allocated as therapy related and another 3 patients (12%) evolved from a previous MDS or MPN as diagnostic qualifiers according European Leukemia Net (ELN 2022).

 

n (%) / median (range)

Median age

53 (25-61)

Diagnosis

AML

22 (88%)

MDS

3 (12%)

AML ELN 2022 (n = 22)

Favorable risk

1 (4%)

Intermediate risk

10 (40%)

Adverse risk

11 (44%)

Treatment

HMA

5 (20%)

HMA + venetoclax

1 (4%)

AML induction regimen

16 (64%)

ASCT as first line

3 (12%)

Bone marrow blast pre ASCT

31.3 (7-83)

Status at ASCT

Primary refractory

11 (44%)

1st relapse

9 (36%)

2nd relapse

2 (8%)

ASCT as first line

3 (12%)

ECOG < 2

19 (76%)

Cytoreductive treatment

Fludarabine based

8 (32%)

Clofarabine based

16 (64%)

Others

1 (4%)

Conditioning platform

 

Fludarabine + Busulfan

23 (92%)

Others

2 (8%)

Post-transplant Cyclofosfamide

10 (40%)

Stem cell source

Peripheral blood

24 (96%)

Bone marrow

1 (4%)

Type of donor

MSD

15 (60%)

MUD

5 (20%)

Haploidentical

5 (20%)

GVHD prophylaxis

CNI + MMF

9 (36%)

Tacro + PT-Cy

4 (16%)

Tacro + Sirolimus

12 (48%)

Acute GVHD (n = 13)

Grade I-II

9 (36%)

Grade III-IV

4 (16%)

Chronic GVHD

Limited

1 (4%)

Extensive

3 (12%)

We have Next Generation Sequencing (NGS) data of 10 patients (40%). Median number mutation per patient were 3 (0-9). Most recurrent mutations were DNMT3 and FLT3 in 30% and KRAS, PTPN11, RUNX1and SF3B1 in 20% of patients with available data.

With a median follow up of 19.5 months (1-103) the median, the estimated 1-year and 5-year OS was 48% [31%-74%, 95% CI] and 21% [9%-50%, 95% CI] respectively (graph 1)

In univariate analysis, those patients with ³ 2 treatment had worse survival (p = 0.039) and on the contrary, to reach complete remission at day +100 improves outcomes (p = 0.0012). Sixteen patients (64%) reached profound response consistent with CR MRDneg and full donor chimerism at some point but on the formal evaluation at day +100, 9 patients (36%) experienced relapse and we lost 4 patients (16%) due to early mortality. The main causes of dead were progression in 10 cases (40%), infection in 5 cases (20%), bleeding in 2 cases (8%) and other reasons in 1 case (4%).

Conclusions: Our data shows the dismal prognosis of those patients with refractory disease but at the same time reveals that there are a small subset of long term survivors. Identification of who are the patients who benefits the most, the correct timing to transplant and optimization of cytoreductive and conditioning platforms require further studies.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P041 LONG-TERM OUTCOMES OF ALLOGENEIC STEM CELL TRANSPLANTATION IN PATIENTS WITH ACUTE LYMPHOBLASTIC LEUKEMIA

Anna Płotka 1, Anna Wache1, Anna Łojko-Dankowska1, Magdalena Matuszak1, Dominik Dytfeld1, Jolanta Kiernicka-Parulska1, Anna Mierzwa1, Ewa Bembnista1, Krzysztof Lewandowski1, Lidia Gil1

1Poznan University of Medical Sciences, Poznan, Poland

Background: Allogeneic hematopoietic cell transplantation (alloHCT) is a standard treatment of patients (pts) with Philadelphia positive (Ph+) acute lymphoblastic leukemia (ALL), especially with detectable minimal residual disease (MRD+) or with relapsed disease. Long term survival after alloHCT is influenced by relapse and non-relapse mortality (NRM). Herein, we present a retrospective analysis of the long-term outcomes of alloHCT in pts with ALL in one institution with respect on MRD status and HCT-CI index prior to the procedure.

Methods: The study group consisted of 92 de novo ALL pts, median age 33 (range, 18-65) years, transplanted at our institution between 2007-2021. 35% pts underwent alloHCT form HLA-identical sibling donor, while 65% from 9-10/10 matched unrelated donor. The stem cells were collected from peripheral blood (71%) or bone marrow (29%). Graft versus host disease (GvHD) prophylaxis consisted of calcineurin inhibitor combined with methotrexate, plus ATG in 68% of pts. Only first transplantations were analyzed.

Results: In the whole cohort 82% of pts were transplanted in the first complete remission (CR1), while 18% beyond CR1. 84% of pts were conditioned with myeloablative therapy, including 77% based on total body irradiation (TBI). 70 pts (76%), 19 pts (21%) and 3 pts (3,2%) had low, median and high HCT-CI index, respectively. 60% of pts were MRD+ prior to the procedure. 8% of pts were older than 60 years.

The median follow-up time was 34 month. The median time of neutrophil recovery was 20 (range, 11-36) days. Acute GvHD was diagnosed in 23% of pts, while chronic in 15% of pts. Relapse of disease occurred in 23 pts (25%), mostly within first year after alloHCT (79%).

During follow-up the median overall survival (OS) was 36 months, and 35 (38%) pts died. The main reason for death was the relapse of the disease (18 pts), followed by infectious complications (12 pts). The 2-year estimated OS was 68% and the relapse free survival (RFS) was 20%. In the analysed cohort, MRD status prior to alloHCT, HCT-CI index and older age of pts did not influenced OS, RFS or NRM significantly.

Conclusions: In long-term observations of pts with ALL after alloHCT, relapse of the disease remains the main cause of death. Most relapses occur within first 1 year after alloHCT. Infectious complications contributes to substantial number of deaths.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P042 POST-INDUCTION IDH2 EXPRESSION DOES NOT IMPACT OVERALL SURVIVAL IN PATIENTS WITH ACUTE MYELOID LEUKEMIA

Safa Al-Amrani 1, Fahad Al-Zadjali1, Yasir Jeelani1, Zaaima AL-Jabri1, Mahdiya Al-Bulushi2, Mohammed AlRawahi1, Adhari Al Zaabi1, Jalila AlShekaili1, Mohammed Al-Huneini1, Murtadha Al-Khabori1

1Sultan Qaboos University, Muscat, Oman, 2Royal Hospital, Muscat, Oman

Background: Isocitrate dehydrogenases (IDHs) are enzymes that play important roles in numerous cellular metabolic and epigenetic processes. About 20% of patients with acute myeloid leukemia (AML) have mutations in either IDH1 or IDH2. To our knowledge, the impact of differences in IDH2 expression between diagnosis and remission states on patients’ survival has not been investigated. In this study, we aim to assess the impact of IDH2 expression at diagnosis and compare the difference in the expression between diagnosis and remission on overall survival.

Methods: Patients with acute myeloid leukemia (AML) were included in this study. Cell lysates of peripheral blood and bone marrow of these patients were collected at diagnosis and at remission. The expression of IDH2 in these samples was estimated using ELISA assay and normalized to the total protein content. Overall survival was estimated using the Kaplan-Meier curve and groups were compared with the COX regression model. All analyses were performed using the R program.

Results: A total of 19 patients (a total of 76 peripheral blood and bone marrow samples) were included with a median age of 47 years (IQR: 27 – 59). At diagnosis, the median hemoglobin, white blood cells, and platelet counts were 8.3 g/dL (IQR: 6.3 – 9.4), 24*109/L (8 – 49), and 53*109/L (33 – 63), respectively. The median blast percentage was 50% (IQR: 27 – 68) in peripheral blood and 70% (IQR: 53 – 83) in bone marrow samples. In peripheral blood, the median IDH2 expression was 4.4 (IQR: 2.0 – 17.5) at diagnosis and 19.0 (IQR: 6.6 – 42.1) at remission whereas, in bone marrow, it was 5.3 (IQR: 3.3 – 5.6) at diagnosis and 6.7 (IQR: 4.6 – 12.5) at remission.

During follow-up, nine patients died. The overall survival was not affected by the IDH2 expression at diagnosis in peripheral blood (HR 1.3, p = 0.755) or bone marrow cell lysates (HR = 0.7, p = 0.68). In one out of the 19 patients, the IDH2 expression decreased in remission peripheral blood samples.

Conclusions: The level of IDH2 expression does not predict overall survival in peripheral blood and bone marrow samples. In addition, the reduction of the expression occurs in a small minority in post-induction samples and hence quantification using ELISA assay may not be a useful candidate as a minimal residual disease test. The study is limited in sample size and needs to be confirmed in future studies with larger sample sizes.

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P043 CD33+ TARGETING, CONSOLIDATIVE HSCT AND INTEGRATION OF MIDOSTAURIN IN FLT3 POSITIVE AML PEDIATRIC PATIENTS: A SINGLE CENTER EXPERIENCE

Manuela Tumino1, Maria Gabelli 1, Chiara Mainardi1, Marta Pillon1, Antonio Marzollo1, Elilsabetta Calore1, Giulia Reggiani1, Maria Caterina Putti1, Martina Pigazzi2, Anna Leszl1, Barbara Buldini1, Alessandra Biffi1

1Pediatric Onco-Hematology and Hematopoietic Stem Cell Transplantation Division, Azienda Ospedale-Università Padova, Padova, Italy, 2Pediatric Onco-Hematology and Hematopoietic Stem Cell Transplantation Unit, Padova University, Padova, Italy

Background: FMS-like tyrosine kinase 3 internal tandem duplication (FLT3-ITD) characterizes a high-risk form of acute myeloid leukemia (AML). FLT3-ITD mutations are associated to an increased expression of CD33. In 2017, Midostaurin was approved by the FDA as first FLT3 inhibitor and as first targeted therapy in adult AML. Gemtuzumab ozogamicin (GO) is an anti-CD33 monoclonal antibody approved by EMA in 2018 for the treatment of patients aged 15 years and above with previously untreated de novo CD33-positive AML, in combination with daunorubicin and cytarabine.

Methods: We report two pediatric patients affected by relapsed/refractory AML FLT3-ITD. Case 1 is a 9-years-old male with diagnosis of M1 AML positive for FLT3-ITD and t(3;5)(NPM1-MLF1); blasts expressed CD45, CD7, CD33 and CD38, detected by flow cytometry. Cytogenetic analysis confirmed t(3;5) translocation. The patient received four cycles of chemotherapy according to the AIEOP-AML2013 Protocol, high-risk group (ICE, FLA-My, AVE, HAM). Morphological and cytofluorimetric complete remission (CR) occurred at the end of the first cycle of induction, with persistency of FLT3-ITD and t(3;5) signal at high level. Disease relapse occurred after the 4th cycle, immediately before allogeneic hematopoietic stem cell transplantation (alloHSCT). Two additional cycles (Clofarabin-daunorubicin-Cytarabin and high dose Etoposide) were administered, without response. The patient then received additional treatment based on the association of GO, Doxorubicin, Aracytin and Midostaurin. Haploidentical HSCT with peripheral TCRalpha/beta-CD19 depleted cells from father was then performed in aplasia, preceded by a myeloablative conditioning regimen based on TBI-Melphalan. Case 2 is a 12-year-old female with diagnosis of AML positive for FLT3-ITD and t(5;11), with positive expression of CD33, CD45, CD99, CD38, CLL1, CD11a, CD44 detected by flow cytometry. Cytogenetic analysis confirmed t(5;11) translocation. The patient received the first two treatment cycles according to the AIEOP-AML2013 Protocol, high-risk group (ICE, ICE), without disease remission. Additional treatment with Fludarabin-Aracytin-Myocet-G CSF could not prevent disease progression. The patient then received one cycle based on the association of GO, Clofarabin, Aracytin and Midostaurin. Thereafter, alloHSCT from a HLA-matched sibling donor was performed in aplasia, preceded by a myeloablative conditioning regimen based on TBI-Melphalan.

Results: Post-HSCT course of both patients was complicated by pneumonia (atypical mycobacteria in the case 1, fungal in case 2) that required prolonged hospitalization. No GvHD occurred. Bone marrow evaluation at neutrophil engraftment showed morphological, cytofluorimetric and molecular CR, with full donor chimerism in both patients. Midostaurin was restarted in patient 1, at day +40 days post HSCT and continued up to 2 years post HSCT. Both patients are still alive and in persistent complete remission at + 3 years and +1 year after HSCT respectively.

Conclusions: GO integration into the FLT3-AML therapy before HSCT resulted in improved outcomes for pediatric patients with refractory AML. Combination of targeted inhibition of FLT3 and CD33 followed by allogeneic HSCT could be a promising therapeutic opportunity for refractory/relapsed pediatric AML.

Disclosure: nothing to declare.

19 - Acute Leukaemia

P044 THE OUTCOME OF SECONDARY AML PATIENTS WITH PRIOR BREAST CANCER UNDERGOING ALLOGENEIC STEM CELL TRANSPLANTATION IS AS GOOD AS DE NOVO AML

Merve Aydoğan1, Bülent Karakaya1, Gül Yavuz Ermiş1, Güldane Cengiz Seval1, Sinem Civriz Bozdağ1, Selami Koçak Toprak1, Pervin Topçuoğlu1, Meltem Kurt Yüksel 1

1Ankara University, Ankara, Turkey

Background: Treatment-related acute myeloid leukemia (t-AML) may develop in patients exposed to DNA-damaging agents, including alkylating agents, topoisomerase-II-inhibitors and antimetabolites and ionizing radiation. 71% patients with t-AML had a previous solid cancer and breast cancer was the most common neoplasm. Compared to de novo AML, patients with t-AML tend to exhibit worse clinical outcomes, including significantly inferior complete remission rates, relapse-free survival, and overall survival. Based on this, 5 cases of t-AML followed in our center will be presented.

Methods: The data of 5 female who undergoing allogeneic stem cell transplantation (ASCT) at Ankara University, Stem Cell Transplantation Unit, with a history of breast cancer, were analyzed retrospectively.

Results: Case Series

Case-1: A 48-year-old female patient. 5 years ago, she had breast cancer. The patient received 6 courses of chemotherapy (CT) and radiotherapy (RT) after mastectomy. While she was in remission with tomoxifen treatment, she was diagnosed with AML(FLT3-ITD+, WT1+). After receiving 7 + 3+midostaurin and 3 courses of azasitidine+ venetoclax, ASCT was performed from his fully matched brother. She is in second month and has poor graft function, now.

Case-2: A 63-years-old female patient. She was diagnosed with breast cancer 35 years ago. The patient who received bilateral mastectomy and then CT + RT, was cured. She had AML-M5 (FLT-ITD+, NPM1A+) one year ago. After 7 + 3+midostaurin and 2 courses of IDAC+midostaurin, relapsed 4 months later. ASCT was performed from a female donor. After 4 months, she had gastrointestinal(GI) graft-versus-host disease(GvHD).

Case-3: A 57-year-old female patient. When she was 47 years old, she had breast cancer and mastectomy, CT and RT was applied. After 9 years, she had AML-M5 (NPM1A+). She received 3 courses of HiDAC after 7 + 3 induction therapy. She had haploidendic ASCT from his brother. She had hemorrhagic cystitis in the 2nd month after transplantation

Case-4: 35-year-old female patient.7 years ago, she was diagnosed with, breast cancer. After mastectomy, she received 8 courses of CT and RT. 4 years ago, she had AML-M5 (11q23, MLL+). After 7 + 3 induction therapy ve 2 courses of HiDAC, she was relapsed 2 months later. She had ASCT from a female donor. Afterwards, she had acute skin and GI-GvHD and thrombotic microangiopathy(TMA). The disease relapsed 2 years after the transplant and the patient died at the age of 35.

Case-5: A-58-year-old female patient. At age 45, 13 years ago, she was diagnosed breast cancer. After receiving 6 courses of CT and RT, she received tamoxifen. At age 50, she had AML M4/5. After 7 + 3 induction therapy, 4 courses of FLAG therapy were given, as she did not has remission. Finally, ASCT was performed from an unrelated donor. At the first month of transplantation, she had skin-GvHD. The patient continues to be followed in the 7th year of transplantation.

Conclusions: In our center, 5 cases diagnosed with t-AML after breast cancer, followed by ACST are presented. The outcomes of ASCT of the patients are as good as de novo AML. Overall survival is 80%

Disclosure: Nothing to declare.

19 - Acute Leukaemia

P045 OUTCOMES IN SURVIVAL IN PATIENTS WITH ACUTE LEUKEMIA IN PEDIATRIC AND YOUNG ADULTS WITH ALLOGENIC HSCT AND HAPLOIDENTICAL HSCT IN A MEXICAN PUBLIC HEALTH HOSPITAL

Guillermo Sotomayor Duque1, Severiano Baltazar Arellano 1, Guadalupe Gonzalez Villarreal1, Juan Daniel Garza Escobar1, Rosa Elva De León Cantu1, Roberto Hernández Valdez1

1IMSS UMAE 25 Monterrey, Monterrey, Mexico

Background: Hematopoetic Stem Cell Transplantation (HSCT) in acute leukemias is an indication in those with high-risk disease in first remission or those who achieve a second response after relapse. The first option is to find a match related donor; in our country due to the limited access of a match unrelated donor, Haplo HSCT has become a feasible option in our center.

Methods: Objective: Describe the results of patients undergoing allogeneic HSCT and HaploHSCT with acute leukemia.

Retrospective analysis of patients diagnosed with acute leukemia who underwent transplantation from 2017 to 2022 at the Hospital IMSS UMAE 25 Monterrey Mexico. Descriptive analysis with measures of central tendency, analysis of overall survival (OS) and relapse-free survival (RFS) with the Kaplan Mier method.

Results: A total of 45 patients were included, 29 men (65%) and 16 women (35%) with a median age of 17 years (1-44). AML were 7 (15.6%) ALL 38 (84.4%). Seven patients had detectable minimal residual disease prior to transplantation. The type of transplant was allogenic HSCT in 21 and Haplo HSCT in 24 patients. The conditioning regimen was myeloablative in 97.8% of patients (Flu/Bu protocol). By type of donor, the most common was father 35.6%(16), mother 2.2%(1), brother 42.2%(19), sister 20%(20). Disparity in the AB0 group was presented in 11.1%. The median CD 34 was 10.6/kg (3.0-60.3) obtained from peripheral blood replete with T lymphocytes. The median engraftment of the myeloid series was 14 days (10-32) and platelets 14 days (9-33).

The median post-transplant follow-up was 11 months (1-31 months). OS for ALL was 45% and for AML 85% at 20-month follow-up. For allogeneic transplant, ALL had an OS of 25% and for AML an OS of 100% at 20 months; for Haplo HSCT in ALL the OS was 80% and AML 75% at 18 months of follow-up.

The RFS was 100% for AML and 57% for ALL at 20 months. For allogeneic transplant in ALL the RFS was 42%, LMA was 100%. In the Haploc HSCT in ALL the RFS was 82% and in LMA 100% at 20 months.

Conclusions: Our retrospective analysis has limitations such as the lack of characterization of genetic risk of leukemia and the number of patients.

In our analysis, the HSCT Haplo compared to the match allogeneic HSCT shows a trend of better results in high-risk ALL and refractory behavior with a greater OS in the 20-month follow-up of this analysis; while the LMA found similar OS rates. The use of myeloablative conditioning with the Bu/Flu protocol demonstrates a tendency to reduce relapse in our center in both transplant modalities.

Our results show that both Haplo HSCT and allogeneic HSCT have comparable results in relapse-free survival in acute leukemias as reported in recent studies by CIMBTR.

Disclosure: Nothing to declare.

20 - Aplastic Anaemia

P046 COMPARISON OF THE OUTCOMES OF IMMUNOSUPPRESSIVE THERAPY WITH RABBIT - VS. HORSE-ANTI-THYMOCYTE GLOBULIN FOR THE TREATMENT OF SEVERE APLASTIC ANEMIA IN CHILDREN

Ayami Yoshimi 1, Peter Noellke1, Jan Starý2, Krisztián Kállay3, Owen Smith4, Franco Locatelli5, Jochen Buechner6, Ivana Bodova7, Julian Sevilla8, Markus Schmugge9, Marc Bierings10, Tania Masmas11, Michael Dworzak12, Veerle Labarque13, Katarzyna Pawelec14, Kirsi Jahnukainen15, Sophia Polychronopoulou16, Paula Kjollerstrom17, Marko Kavcic18, Miriam Erlacher1, Charlotte M Niemeyer1, Brigitte Strahm1

1University of Freiburg, Freiburg, Germany, 2Charles University and University Hospital Motol, Prague, Czech Republic, 3Central Hospital of Southern Pest - National Institute of Hematology and Infectious Diseases, Budapest, Hungary, 4Our Lady’s Children’s Hospital, Dublin, Ireland, 5IRCCS Ospedale Pediatrico Bambino Gesù; Catholic University of the Sacred Hearth, Rom, Italy, 6Oslo University Hospital, Oslo, Norway, 7Detská Fakultná Nemocnica s Poliklinikou v Bratislave, Bratislava, Slovakia, 8Hospital Infantil Universitario Niño Jesús, Madrid, Spain, 9University Children’s Hospital, Zurich, Switzerland, 10Princess Máxima Center for Pediatric Oncology and Wilhelmina Children’s Hospital, Utrecht, Netherlands, 11Center for Hemoglobinopathies, Herlev University Hospital, Herlev, Denmark, 12St. Anna Children’s Hospital, Medical University of Vienna, and St. Anna Children’s Cancer Research Institute (CCRI), Vienna, Austria, 13University Hospital Leuven Gasthuisberg, Leuven, Belgium, 14Warsaw Medical University, Warsaw, Poland, 15Children’s Hospital, University of Helsinki and Helsinki University Hospital, Helsinki, Finland, 16Aghia Sophia Children’s Hospital, Athens, Greece, 17Hospital Dona Estefânia, Centro Hospitalar Universitário de Lisboa Central, Lisboa, Portugal, 18University Children’s Hospital, Ljubljana University Medical Centre, Ljubljana, Slovenia

Background: Immunosuppressive therapy (IST) with anti-thymocyte globulin (ATG) and cyclosporine A (CSA) is an effective treatment for many patients with severe aplastic anaemia (SAA). Most previous studies including a randomized control trial comparing horse (h-ATG: Lymphoglobulin® or Atgam®) and rabbit ATG (r-ATG: Thymoblobulin®) indicated superiority of h-ATG. However, there is limited data comparing the outcomes of Atgam® and Thymoblobulin® outside US. Here, we report on the outcomes of IST with ATG and CSA in 150 children with SAA registered in the European Working Group of Severe Aplastic Anemia (EWOG-SAA) and consecutively treated with Atgam® or Thymoblobulin®.

Methods: Among the 150 children (M/F = 92/58, median age 10.2 years [1.1-18]), 110 patients received h-ATG (ATGAM®, 40 mg/kg x 4 days), while 40 patients received r-ATG (Thymoglobulin®, 3.5 mg/kg x 5 days). Granulocyte-colony stimulating factor was administered to patients with an absolute neutrophil count (ANC) < 0.5.10*9/L. None of the patients received eltrombopag at start of IST, but it was added in 5 non-responders at a median time of 192 days (122-671) after IST start. Remarkably, 85% of patients had very SAA with <0.2.10*9/LANC and 41% had zero neutrophils in peripheral blood prior to IST. There were no significant differences in patient characteristics between two ATG groups. The median time from diagnosis to IST was 36 (7–305) days and the median follow-up time 3.1 (0.1-8.9) years.

Results: The overall (complete + partial) response rate at 6 months was significantly higher in the h-ATG compared to the r-ATG group (42% vs. 22%, p = 0.03). In univariate and multivariate analyses, lower ANC prior to IST and use of r-ATG were associated with an inferior response to IST at 6 months. Due to late responses > day 180 (28% vs. 20%) and fewer relapses (0/9 vs. 9/46 of responders at day 180) in the r-ATG group compared to the h-ATG cohort, there was no difference in the response rates between two groups at time of last follow-up (39% vs. 35%). To note, 55% and 53% of patients in the h-ATG and r-ATG groups received allogeneic hematopoietic stem cell transplantation during their treatment course, and 20% and 28% before day 180, respectively. Only two patients had clonal evolution (1 in each group). There were no differences in the overall survival (OS: 93% vs. 88%) and failure-free-survival (33% vs. 38%) between h-ATG and r-ATG groups. Causes of deaths were transplant-related complications (n = 9), lymphoma after HSCT (n = 1), and infection prior to HSCT (n = 1). Responders at 6 months (n = 55) showed 100% OS. Importantly, non-responders with <0.2 x10*9/L ANC at day 90 had inferior OS compared to those with > 0.2 x10*9/L ANC (74% vs. 95%, p < 0.01).

Conclusions: In conclusion, there was no difference in response rate and survival at time of last follow-up between the patients receiving h-ATG or r-ATG. However, earlier response in patients treated with h-ATG may be a great benefit, because prolonged severe neutropenia is associated with increased mortality in patients with SAA.

Disclosure: Nothing to declare.

20 - Aplastic Anaemia

P047 OUTCOMES AFTER CORD BLOOD TRANSPLANTATION FOR INHERITED BONE MARROW FAILURE SYNDROMES: A REPORT FROM THE JSTCT INHERITED DISEASE WORKING GROUP

NAO YOSHIDA 1, Miharu Yabe2, Hiromasa Yabe2, Katsutsugu Umeda3, Shinya Osone4, Takashi Koike2, Shoji Saito5, Katsuyoshi Koh6, Hiroyuki Ishida7, Maho Sato8, Koji Kato9, Atsushi Sato10, Yoshiko Hashii11, Yoshiko Atsuta12

1Children’s Medical Center, Japanese Red Cross Aichi Medical Center Nagoya First Hospital, Nagoya, Japan, 2Tokai University School of Medicine, Isehara, Japan, 3Graduate School of Medicine, Kyoto University, Kyoto, Japan, 4Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto, Japan, 5Shinshu University School of Medicine, Matsumoto, Japan, 6Saitama Children’s Medical Center, Saitama, Japan, 7Kyoto City Hospital, Kyoto, Japan, 8Osaka Women’s and Children’s Hospital, Izumi, Japan, 9Central Japan Cord Blood Bank, Seto, Japan, 10Miyagi Children’s Hospital, Sendai, Japan, 11Osaka University Graduate School of Medicine, Suita, Japan, 12Japanese Data Center for Hematopoietic Cell Transplantation, Nagakute, Japan

Background: Recent advances in cord blood transplantation (CBT) have increased options available in hematopoietic cell transplantation. However, studies describing the outcomes of CBT for inherited bone marrow failure syndromes (IBMFS) are limited.

Methods: Using data from the Japanese Society for Transplantation and Cellular Therapy registry, we assessed the outcomes of 52 children ( < 16 years old) with IBMFS (14 Fanconi anemia [FA], 5 dyskeratosis congenita [DC], 14 severe congenital neutropenia, 11 Diamond-Blackfan anemia and 8 others) who underwent CBT (5 related and 47 unrelated) between 1995 and 2018. Eight of the patients with FA were transplanted after malignant transformation (2 myelodysplastic syndrome [MDS] and 6 acute leukemia). Cord blood units were HLA-A, -B, -C, and -DR serologically matched (n = 12) or had 1 (n = 23) or 2-3 (n = 17) disparities with the recipient, and the median total number of nucleated cells (TNC) infused was 7.4 x 107/kg. Fludarabine-based reduced intensity conditioning was used in 39 patients (75%).

Results: The probability of neutrophil engraftment at day 60 was 83%; limited HLA disparity was the only favorable factor for engraftment (91% in the matched or 1 mismatched vs. 65% in the 2 or more mismatched; P = 0.034). The 5-year overall survival (OS) and event-free survival (EFS) were 77% and 58%, respectively, which varied depending on the underlying disease. Patients with FA or DC had inferior OS (56% vs. 90%; P = 0.002) and EFS at 5 years (41% vs. 68%; P = 0.034) than other patients. The main cause of treatment failure was graft failure (n = 11; 6 primary and 5 secondary), 9 of which were salvaged by retransplantation; limited to FA, leukemia relapse or development of a second malignancy was most common (n = 5), all of which occurred in patients who had progressed to MDS/acute leukemia before CBT. The cumulative incidence of grade II-IV acute GVHD at day 100 and of chronic GVHD at 5 years was 34% and 14%, respectively. Other than HLA disparity and underlying disease, no other factors were associated with transplant outcomes, including the TNC infused.

Conclusions: The current study reveals that a significant proportion of children with IBMFS, especially other than FA and DC, can be cured with CBT. Given that graft failure was a major cause of treatment failure and was more associated with the use of CB units with greater disparities in HLA, better HLA matching may further improve the outcomes. For patients with FA, transplantation before progression to MDS/acute leukemia may be a good option.

Disclosure: Nothing to declare.

20 - Aplastic Anaemia

P048 PREGNANCY OUTCOMES AND MANAGEMENT IN ACQUIRED BONE MARROW FAILURE SYNDROME AND PAROXYSMAL NOCTURNAL HAEMOGLOBINURIA: A SINGLE CENTRE EXPERIENCE

Marta Bortolotti 1,2, Roochi Trikha1, Sarah Salter1, Joanna Large1, Shreyans Gandhi1, Austin Kulasekararaj1

1King’s College Hospital NHS Foundation Trust, London, United Kingdom, 2University of Milan, Milan, Italy

Background: Onset and relapse of acquired aplastic anaemia (AA), hypoplastic myelodysplastic syndrome (hypo-MDS) and paroxysmal nocturnal haemoglobinuria (PNH) during pregnancy are rare events and little is known about foeto-maternal outcomes and management of cytopenias during pregnancy.

Methods: We conducted a single-centre retrospective study evaluating impact of AA, hypo-MDS and PNH on pregnancy, focusing on disease severity, treatment need and foeto-maternal outcomes.

Results: Seventy pregnancies occurring in 52 women were registered between 1989 and 2022 (Table 1). Thirty-three women had a previous diagnosis (21 AA; 4 AA/PNH; 5 PNH; 3 hypo-MDS) while 19 had an onset during pregnancy (14 AA; 2 PNH; 3 hypo-MDS). Mean age at diagnosis of disease and of pregnancy was 24(3-38) and 30 (20-40) years, respectively. Among the 21 with previous AA diagnosis, 8 were non-severe, 8 severe and 5 very-severe; median time between diagnosis and pregnancy was 10 years (0.5-26). Twenty-one women had received at least 1 therapy line (12 ATG; 3 cyclosporin (CSA); 1 tacrolimus; 5 eculizumab), ≥2 lines in 3 and 3 patients were CSA-dependent. Twenty-two were on treatment at pregnancy onset (3 CSA; 1 eltrombopag; 18 eculizumab). PNH-clone was present in 32 patients with median granulocyte clone of 16.3% (0.08-99). Considering only pregnancies with previous AA/hypoMDS diagnosis (N = 38), relapse occurred in 9 (24%), with 1 severe case. Overall, onset/relapse occurred in 11, 7 and 4 pregnancies during the 1st, 2nd and 3rd trimester respectively (not known for 6). At the time of onset/relapse, median haemoglobin was 80 g/l (49-113), reticulocytes 59 x109/l (32-66), neutrophils 2.27 x109/l (0.1-6.61) and platelets 32 x109/l (2-87). Transfusions were necessary in 61% of cases and 45 women required treatment during pregnancy. CSA was used in 12 cases; 8 patients started CSA during pregnancy, with response in 5 and of the 4 patients, who were on CSA before pregnancy (1 switched from Eltrombopag at pregnancy onset), 2 relapsed. Eculizumab was administered during 33 pregnancies, pre-emptively (granulocyte clone >20% and minimal haemolysis LDH < 2 x ULN) in 13 and for haemolysis in 20. Eculizumab dose was increased in 17/33 (52%) and thromboprophylaxis was performed in 31/33 (94%). Spontaneous improvement in blood counts after delivery/miscarriage occurred in 29 pregnancies (55%). Post-partum therapy was required in 14 (6 ATG, 6 CSA, 1 transplant and 1 androgens) with 75% response rate. Five (7%) spontaneous miscarriages were reported (4/5 in the 1sttrimester), all in PNH women. Maternal adverse events occurred in 24% of pregnancies, including a Budd-Chiari syndrome after delivery, 6 infections, 4 peripartum bleedings, 5 placental complications (2 pre-eclampsia, 2 retained placenta, 1 premature rupture of the membrane) and 6 urgent caesarean sections/preterm vaginal delivery. Seven women needed peripartum transfusions. Foetal complications occurred in 10% of pregnancies, including 5 premature births, 1 foetal growth restriction and a baby with a talipes foot.

Table 1

N° of women, n

52

N° of pregnancies, n

70*

Age at AA/hypo-MDS/PNH diagnosis, mean (range)

24 (3-38)

Age at pregnancy, mean (range)

30 (20-40)

Diagnosis before pregnancy, n

33

• AA (NSAA; SAA/VSAA; na), n

21 (8; 5; 8)

• AA/PNH, n

4

• Hypo-MDS, n

3

• PNH, n

5

Treatment before pregnancy, n

21

• ATG-based regimen, n

12

• CSA, n

3

• Tacrolimus, n

1

• Eculizumab, n

5

Treatment ongoing at pregnancy onset, n

22

• CSA, n

3

• Eltrombopag, n

1**

• Eculizumab, n

18

AA/hypo-MDS relapse during pregnancy, n (%)

9 (24)§

Onset during pregnancy, n

19

• AA (NSAA; SAA/VSAA; na), n

14 (5; 4; 5)

• Hypo-MDS, n

3

• PNH, n

2

Transfusion needs during pregnancy, n (%)

40 (61%)

• for onset/relapse;

24

• without frank relapse or for PNH-related haemolysis

16

Treatment needs during pregnancy, n

45

• CSA, n

12

• Eculizumab, n

33

• Steroids, n

3

Spontaneous improvement after delivery, n (%)

29 (55)

Treatment needs after pregnancy, n (%)

39 (58)

• ATG-based, n

6

• CSA ± Eltrombopag, n

6

• Androgens, n

1

• Eculizumab, n

25

• HSCT, n

1 (3 further patients were transplanted afterwards)

Maternal complications, n (%)

16 (24)*

• Thrombotic complications

1

• Infective complications

6

• Bleeding complications

4

• Placental complications (pre-eclampsia; retained placenta; PROM)

5 (2; 2; 1)

• Needs of urgent caesarean section or delivery before full term (for any causes)

6

• Needs of transfusion support during and after delivery

7

Foetal complications, n (%)

7 (10)*

• Prematurity

5

  1. Extremely preterm (less than 28 weeks)

2

  2. Very preterm (28-32 weeks)

1

  3. Moderate-late preterm (32-37 weeks)

2

• Foetal growth restriction

1

• Other (talipes foot)

1

Miscarriage, n (%)

5 (7)*

• - I trimester

4

• - II trimester

1

  1. *Two pregnancies are still ongoing.
  2. **Switched to CSA after positive pregnancy test.
  3. §Percentage calculated considering only pregnancies with a previous diagnosis of AA/hypo-MDS (N = 38).
  4. AA, aplastic anaemia; ATG, anti-thymocyte globulin; CSA, cyclosporin; hypo-MDS, hypoplastic myelodysplastic syndrome; NSAA, non-severe aplastic anaemia; PNH, paroxysmal nocturnal haemoglobinuria; PROM, premature rupture of membrane; SAA, severe aplastic anaemia; VSAA, very-severe aplastic anaemia.

Conclusions: At our best knowledge, this is one of the largest cohorts reported so far and suggest that, even though pregnancy should not be discouraged in these women, it surely requires a high level of awareness and multidisciplinary approach.

Disclosure: Nothing to declare.

20 - Aplastic Anaemia

P049 SHORT TELOMERES BEFORE TRANSPLANT ARE ASSOCIATED WITH GRAFT FAILURE FOLLOWING HEMATOPOIETIC CELL TRANSPLANT FOR CHILDREN WITH APLASTIC ANEMIA

Atsushi Narita 1, Hideki Muramatsu1, Masayuki Imaya1, Daiki Yamashita1, Daichi Sajiki1, Ryo Maemura1, Yusuke Tsumura1, Ayako Yamamori1, Manabu Wakamatsu1, Kotaro Narita1, Shinsuke Kataoka1, Rieko Taniguchi1, Nobuhiro Nishio1, Yoshiyuki Takahashi1

1Nagoya University Graduate School of Medicine, Nagoya, Japan

Background: Short telomeres are associated with aplastic anemia (AA) as well as telomere biologic disorders. Hematopoietic stem cell transplantation (HSCT) in patients with telomere biologic disorders is associated with a high risk of transplant-related mortality. In children with AA, however, the correlation between pre-transplant telomere length (TL) and outcomes following HSCT is still not clear.

Methods: Of the patients enrolled in the Central Review of Bone Marrow Morphology for children in Japan, 101 children with AA who underwent HSCT were retrospectively studied. We measured TL by flow-fluorescence in situ hybridization using the Telomere PNA Kit (DakoCytomation, Glostrup, Denmark) from peripheral blood (PB) lymphocytes for all patients. Data of patients with AA were expressed as “delta relative TL” to compare with TL of patients with age-matched healthy controls. To distinguish between engraftment and graft failure, receiver operating characteristic (ROC) curves were generated.

Results: The median age at HSCT was 11.0 (1.3–27.2) years. The etiology was idiopathic in 94 and hepatitis associated in 7 cases. The severity was moderate in 19, severe in 42, and very severe in 40 cases. The median TL at diagnosis was -1.12 SD ( − 4.60 to +2.06 SD). The median time from the diagnosis to HSCT was 10 months (1–261 months), and 51 patients received immunosuppressive therapy including anti-thymocyte globulin before undergoing HSCT. Graft sources included 88 bone marrows (BM), 10 cord blood, 2 BM plus PB, and 1 PB. Fifty-nine patients were completely matched at HLA-A, HLA-B, HLA-C, and HLA-DR allele; 25 were mismatched at 1 HLA allele; 10 were mismatched at 2 HLA alleles, and 7 were mismatched at ≥ 3 alleles. Primary graft failure was observed in one patient. Since 8 patients experienced secondary graft failure, sustained engraftment was seen in 92 patients. Acute graft-versus-host-disease (GVHD) of grade II–IV was observed in 29 patients, while chronic GVHD was in 15 patients. The three-year overall survival (OS) rate was 96.0% (95% CI, 89.8%–98.5%). The median TL was significantly shorter in the patients who had graft failure than in those who sustained engraftment. According to the ROC curve, 29 patients were assigned to the group with shorter TL ( < − 1.73 SD) and the others to the group with longer TL. The incidence of graft failure was significantly higher in the group with shorter TL than in the group with longer TL (27.6% vs. 1.4%, p < 0.001). Meanwhile, there was no significant correlation found between TL and acute/chronic GVHD or OS. In multivariate analysis, TL shortening was an independent prognostic predictor of graft failure (OR, 37.1; 95% CI, 3.9–358.0; p = 0.002).

Conclusions: This study indicates that short TL is a prognostic predictor of graft failure after allogeneic HSCT in children with AA. Patients with shorter TL before undergoing transplantation may clinically behave like patients with telomere biologic disorders rather than idiopathic AA and may need to be managed accordingly. Our results suggest a potential role for telomere length in the risk stratification of children with AA in regard to their HSCT outcomes.

Disclosure: Nothing to declare.

20 - Aplastic Anaemia

P050 PEDIATRIC PATIENTS WITH VERY SEVERE APLASTIC ANEMIA AND HIGH-RISK TRANSPLANT-ASSOCIATED THROMBOTIC MICROANGIOPATHY CAN BE SUCCESSFULLY TREATED WITH ECULIZUMAB

Maria Isabel Benitez Carabante 1, David Bueno Sanchez2, Jose Maria Fernandez Navarro3, Maria Luz Uria Oficialdegui1, Laura Alonso García1, Melissa Panesso1, Luisa Sisinni2, Antonio Pérez Martínez2, Cristina Díaz de Heredia Rubio1

1Hospital Universitario Vall d’Hebron, Barcelona, Spain, 2Hospital Infantil Universitario La Paz, Madrid, Spain, 3Hospital Universitari i Politècnic la Fe, Valencia, Spain

Background: Transplant-associated thrombotic microangiopathy (TA-TMA) is a fatal complication of hematopoietic stem cell transplantation (HSCT) associated with high morbidity and mortality.

Patients with severe aplastic anemia (SAA), especially those who received pre transplant immunosuppression, are known to be under special risk to develop this complication. Eculizumab, a complement C5 inhibitor, has significantly increased the survival of high-risk TA-TMA (hrTA-TMA). Its use in this subgroup of patients has been reported in few case reports.

Methods: We conducted a national multicenter retrospective study. Children (from birth to 18 years of age) diagnosed of SAA and treated with Eculizumab for hrTA-TMA until December 31, 2021, were included.

All patients included presented hrTA-TMA features.

Results: Seven patients with SAA were diagnosed of hrTA-TMA after a first allogeneic HSCT. Four patients were male and 3 female. Median age at transplantation was 11 years (9 - 15).

All but 1 patient (n = 6;85.7%) had received immunosuppression prior to transplantation because a suitable donor was not available at diagnosis; however, none of them achieved sustained complete remission and subsequently underwent HSCT (n = 1 bone marrow (BM) matched unrelated donor (MUD); n = 1 BM 9/10 MMUD; n = 2 9/10 MMUD with CD34+/CD3 add-back; n = 2 CD3+CD45RA+ haploidentical HSCT). One patient received an upfront HSCT from a matched family donor. Most patients (n = 6;85%) received peripheral blood as stem cell source.

Four patients presented grade II-IV acute graft versus host disease (3 were grade IV) and four patients presented either CMV, adenovirus, EBV or BK virus reactivation.

Median time from HSCT to TA-TMA was 154 days (130-387). All patients fulfil the characteristic triad (hypertension, proteinuria and elevated LDH). sC5b-9 was increased in 4/5 where measured. Three patients presented extrarenal involvement (1 pulmonary hypertension; 1 posterior reversible encephalopathy and gastrointestinal bleeding; 1 serositis). Renal (n = 3) and pulmonary (n = 1) biopsy confirmed the diagnosis in 2/4 (50%) patients. Three patients required intensive care unit admission.

Median time from hrTA-TMA diagnosis to the initiation of eculizumab treatment was 18 days (1-56). CH50 was monitored to ensure complement blockade. Five patients initiated eculizumab weekly with a standard body weight adjustment and 2 required intensified treatment for the first two doses. Overall, 6(85%) patients responded to eculizumab of whom 5 (83%) are alive and off treatment. The median duration of eculizumab in patients with resolution of hrTA-TMA was 145 days (80-198).

With a median follow up of 35 months, 5(71%) patients were alive at 12 months after initiation of eculizumab. Two of the three patients with extrarenal involvement died (one due to severe pulmonary hypertension and right cardiac failure and one of idiopathic pneumonia syndrome).

Conclusions: To our knowledge, this is the largest cohort of hrTA-TMA patients with very SAA treated with eculizumab. Despite the high-risk profile, the response rate and overall survival were high.

To conclude, special attention should be taken to very SAA who received pre-transplant immunosuppression, as they are at particular risk of developing hrTA-TMA: early detection strategies, aggressive treatment of concomitant complications, intensifying the initial induction dose and eculizumab pharmacokinetic monitoring could be of utmost importance to achieve the best outcomes.

Disclosure: Nothing to declare.

20 - Aplastic Anaemia

P051 COMPARABLE OUTCOMES OF PTCY AND BUCY REGIMENS FOR FIRST-LINE HEMATOPOIETIC STEM CELL TRANSPLANTATION FROM UNRELATED AND MATCHED SIBLING DONORS IN PATIENTS WITH SEVERE APLASTIC ANEMIA

Xiaowei Chen 1, Cunte Chen1, Ming Zhou1, Yuping Zhang1, Wenjian Mo1, Shunqing Wang1

1Guangzhou First People’s Hospital, South China University of Technology, Guangzhou, China

Background: To explore the feasibility of a posttransplant cyclophosphamide (PTCy) regimen for the unrelated donor (URD) allogeneic hematopoietic stem cell transplantation (allo-HSCT) in the treatment of patients with severe aplastic anemia (SAA), we conducted a retrospective single-center study to compare it with the busulfan and cyclophosphamide (BuCy) regimen for matched sibling donor (MSD) allo-HSCT.

Methods: The clinical outcomes were collected from 28 SAA patients treated with PTCy regimen for URD allo-HSCT and 43 patients treated with BuCy regimen for MSD allo-HSCT between May 2020 and June 2022.

Results: The overall survival (OS) and restricted mean survival time (RMST) were not significantly different between SAA patients treated with PTCy regimen for URD allo-HSCT and patients treated with BuCy regimen for MSD allo-HSCT (P = 0.174, 2-year OS: 100% vs. 93%). PTCy regimen for URD allo-HSCT had an increased cumulative incidence of cytomegalovirus (CMV) viremia (82.1% vs. 48.8%, P = 0.010), but similar rates of grade II acute graft-versus-host disease (aGVHD) (P = 1.000), chronic GVHD (cGVHD) (P = 0.704), CMV disease (P = 0.422), cystitis (P = 0.501), Epstein-Barr virus (EBV) viremia (P = 1.000) and EBV positive post-transplant lymphoproliferative disease (EBV-PTLD) (P = 1.000) compared with patients treated with BuCy regimen for MSD allo-HSCT during the same period. After using propensity score matching to reduce the influence of potential confounders before transplantation, the OS and 2-year RMST was similar (P = 0.352, 2-year OS: 100% vs. 93.3%), and no significant differences in the aforementioned outcomes were observed between the two groups.

Conclusions: PTCy may be an effective conditioning regimen for URD allo-HSCT in SAA patients.

Disclosure: Nothing to declare.

20 - Aplastic Anaemia

P052 LIVER AND KIDNEY IRON OVERLOAD IN PATIENTS WITH PAROXYSMAL NOCTURNAL HEMOGLOBINURIA

Vitalii Latyshev 1, Kira Lukina1, Alexandra Kitsenko1, Elena Lukina1

1National Medical Research Center for Hematology, Moscow, Russian Federation

Background: Paroxysmal nocturnal hemoglobinuria (PNH) is a rare clonal hematopoietic disease characterized by intravascular hemolysis, prothrombotic state, and bone marrow insufficiency. Bone marrow aplasia often precedes or accompanies PNH. Iron metabolism in patients with PNH differs from other hemolytic anemias due to intensive iron losses with urine in non-treated patients. On the other hand, patients receiving anti-complement therapy may develop liver iron overload, which can be detected by MRI in T2* (T2-star) sequence. This study investigates liver and kidney iron content in patients with PNH.

Methods: Between August 2016 and October 2022, overall 70 MRI-R2* investigations of liver and 62 of kidneys were performed. Liver iron concentration (LIC, mg/g) was measured by MRI-R2* and converted according to formulas as described elsewhere. For kidneys values were given in relaxation time (ms). Median age of patients was 39 years (20-75). Patients were divided into a treatment-naïve cohort and receiving therapy (either eculizumab or ravulizumab). Thirty-two patients (32/70) had a history of bone marrow aplasia (BMA+ group). Statistic analysis was performed using an unpaired t-test.

Results: Overall, 62 MRI-T2* investigations of kidney were available for analysis. All treatment-naïve patients (n = 45) had iron overload of renal cortex with median T2*-time of 4,0 ms versus 42,0 ms in the group receiving anti-complement therapy (p < 0,01). There was no significant difference in kidney iron overload between BMA+ and BMA- groups (p = 0,62).

Total 70 MRI-T2* investigations of liver were available for analysis. LIC values were significantly lower in treatment-naïve group with a median of 2,68 mg/g versus 6,3 mg/g in group receiving anti-complement therapy (p < 0,01). In treatment-naïve group only 13/57 (25%) had an iron overload of liver (LIC>2mg/g). There was no significant difference in liver iron overload between BMA+ and BMA- groups (p = 0,8).

Conclusions: Iron overload of renal cortex is common in treatment-naïve patients with PNH. After treatment initiation PNH-patients may develop liver iron overload. Performing liver and kidneys MRI in R2* sequence may be useful for tissue iron overload detection and subsequent decision if chelation therapy is indicated. History of bone marrow aplasia does not significantly affect values of tissue iron content measured by MRI-R2*.

Disclosure: Nothing to declare.

20 - Aplastic Anaemia

P053 SUCCESSFUL OUTCOMES OF UPFRONT HAPLOIDENTICAL TRANSPLANT USING 4GY TBI AND POST TRANSPLANT CYCLOPHOSPHAMIDE IN PEDIATRIC SEVERE APLASTIC ANEMIA

Esha Kaul 1, Rohit Kapoor1, Gagan Saini1, Nivedita Dhingra2, Akriti Khare1, Nitin Dayal3, Ritika Pahwa1, Titus Philip1

1Max SuperSpeciality Hospital, Vaishali, Vaishali, India, 2Max SuperSpeciality Hospital, Patparganj, India, 3Max SuperSpeciality Hospital, Saket, India

Background: Immuno-suppressive Therapy (IST) is considered standard treatment for management of severe aplastic anemia(SAA) in children in the absence of a full match. There have been concerns of high rates of primary and secondary graft failure with upfront haploidentical stem cell transplant (hapSCT). Increasing the dose of total body irradiation (TBI) can decrease this risk. We present our outcomes of upfront hapSCT in SAA in using 4Gy TBI instead of 2Gy TBI with post-transplant cyclophosphamide (PTCy).

Methods: We retrospectively analysed data of 4 consecutive children who underwent haplo SCT with PTCy and 4Gy at our center.

Results: Four patients were transplanted using increased radiation dose using 4Gy TBI with PTCy. The median age was 5.5 years (4 –14years). The Female : male ratio was 3:1. Diagnosis was very severe aplastic anemia -1,very severe aplastic anemia with celiac disease-1, congenital amegakaryocytic thrombocytopenia with severe aplastic anemia-1,severe aplastic anemia with LIG4 deficiency-1. Stem cell source was peripheral blood stem cells(PBSC) in 3 patients and bone marrow in one. The conditioning regimen was Flu(150mg/m2, Cy,29mg per kg, ATG and TBI 4Gy. In two patients cumulative dose of rabbit ATG was 4.5mg per kg and in other 2 patients cumulative dose of horse ATG was 45mg per kg (used due to non-availability of rabbit ATG). The donor specific antibody (DSA) was weakly positive with a MFI of 1500 in 1 and negative in the other 3. Three -fourth transplants were ABO mismatched. GVHD prophylaxis was PTCy, MMF, Tac in all. All of the patients received a single dose of Rituximab pre-transplant to decrease the risk of antibody mediated rejection. The mean cell dose was 8.1 million CD34 positive cells/kg of the recipient (Range- 6.4-10 Million/kg). All patients engrafted. The median day of neutrophil engraftment was 14 days. The median day of platelet engraftment was 17 days. Chimerism was fully donor in all. One patient developed grade II mucositis. Incidence of acute GVHD was 25%(1/4) and was Grade I acute GVHD of skin. Two patients had asymptomatic CMV viremia which responded to oral valganciclovir. One patient with LIG4 deficiency developed post-transplant immune cytopenias which resolved with a short course of steroids. There was no chronic GvHD. All patients are alive until the last follow-up with a median follow-up of 358 days (Range 307-1004 days) and all of them are off immunosuppression with full donor chimerism and normal blood counts.

Conclusions: Upfront haploSCT using 4Gy TBI and PTCy in pediatric severe aplastic anemia is feasible and practical without risk of primary or secondary graft failure. Given the dismal response rates with IST in this population, upfront haploSCT should be considered over immunosuppressive therapy especially in resource limited settings where patients often have just one shot at definitive treatment.

Acknowledgement: Our sincere thanks to Mr Vinay Kumar and Ms Diksha to organizing and retrieving data for analysis.

Disclosure: We do not have any conflicts of interest.

20 - Aplastic Anaemia

P054 THERAPEUTIC MODALITIES AND EVOLUTION PATTERNS IN 597 PATIENTS WITH INHERITED AND ACQUIRED BONE MARROW FAILURE DISORDERS: A LARGE AND COMPREHENSIVE SINGLE CENTRE EXPERIENCE

Marta Bortolotti 1,2, Roochi Trikha1, Sarah Salter1, Joanna Large1, Shreyans Gandhi1, Austin Kulasekararaj1

1King’s College Hospital NHS Foundation Trust, London, United Kingdom, 2University of Milan, Milan, Italy

Background: Inherited (IBMF) and acquired bone marrow failure syndromes are a group of rare disorders characterised by pancytopenia and hypoplastic bone marrow, with risk of transformation to myeloid neoplasms. Data about the natural history of IBMF, aplastic anaemia (AA) and hypoplastic myelodysplastic syndromes (hypo-MDS) are increasing and provide help in their better management.

Methods: We conducted a single-centre retrospective study, including patients with IBMF, acquired AA and hypo-MDS referred to our institute in the last 15 years. Data about clinical features, management, evolution and outcomes were compared among the three groups with the aims of describe their natural history, management and disease evolution.

Results: Overall, 597 patients (459 AA; 82 hypo-MDS; 56 IBMF) were included (Table 1). Among IBMF, 11 were Fanconi Anaemia, 3 Diamond-Blackfan Anaemia, 2 Shwachman-Diamond syndrome, 18 telomeropathies and 22 had other mutation related to IBMF or, even without the pathogenic gene identified, presenting with classical clinical features and strong family history. Among AA, according to Camitta’s Criteria, 22% were very severe, 34% severe, 28% non severe while 16% had no severity specification. Mean age at diagnosis was 43 years (0.1-86) and with older patients in hypo-MDS group. A correct diagnosis before the first visit at our institute was performed more often in AA and IBMF patients but it could take up to 8 years for IBMF diagnosis. A concomitant autoimmune disease was present in 16% in AA versus 5% in IBMF. Overall, 81% of patients required medical treatment (86% AA, 79% hypo-MDS, 45% IBMF) and 26% were transplanted. With a median follow-up of 4.5 years (0-24), an abnormal karyotype at first visit was found in 13% of patients with an increase to 28% at last visit. Myeloid gene mutations were detected in 21% overall and in 30% of hypo-MDS patients. Paroxysmal nocturnal haemoglobinuria (PNH) clone was present in 52% of AA patients versus 11% of IBMF, with a higher PNH clone (2.4% vs 0.65%) and an increasing trend in AA. Increase in disease severity overtime was observed in 1.8% while clonal evolution was reported in 24% overall. In particular, IBMF patients showed progression to MDS in 21% and to AML in 3.5%, comparing to 10% and 1.5%, respectively, in AA. Evolution to overt PNH occurred in 11% of AA patients versus a 3.6% in hypo-MDS. Time to evolution was shorter in hypo-MDS patients comparing with AA (1.7 years vs 5.4) and, in 5 IBMF patients, a MDS or AML diagnosis was concomitant with that of IBMF. Overall, 24% patients dead during the observation period, reaching 39% in hypo-MDS patients.

Table 1

 

Overall

AA

Hypo-MDS

IBMF

N° patients

597

459

82

56

M/F

309/287

230/229

44/38

35/21

Age at diagnosis (years), mean (range)

43 (0.1-86)

41.5 (2.4 - 85)

57 (13 - 86)

30 (0.1-71)

Time diagnosis-first visit at KCH (years), mean (range)

3.5 (-8 - 47.7)

3.9 (-6.7 - 47.7)

1.3 (-1.1 - 16.6)

3.7 (-8 - 36.5)

Associated autoimmune disease, n (%)

87 (15)

72 (16)

10 (12)

5 (9)

Treated, n (%)

483 (81)

393 (86)

65 (79)

25 (45)

Transplanted, n (%)

158 (26)

125 (27)

20 (24)

14 (25)

Abnormal karyotype at first visit, n (%)“n” evaluable = 449

60 (13)

35 (11)

15 (20)

11 (25)

Abnormal karyotype at last FU, n (%)“n” evaluable = 262

73 (28)

54 (28)

12 (28)

7 (27)

Abnormal myeloid panel NGS, n (%)“n” evaluable = 238

51 (21)*

32 (19)

10 (30)

9 (25)

PNH clone at first visit, n (%)

284 (48)

241 (52)

37 (45)

6 (11)

Granulocyte PNH clone, median (range)

2.2 (0.02 - 99.9)

2.4 (0.02 - 99.9)

1.6 (0.03 - 97)

0.65 (0.02 - 3.8)

PNH clone at last FU, n (%)

284 (48)

237 (52)

42 (51)

5 (9)

Granulocyte PNH clone, median (range)

3.8 (0.01 - 100)

4.3 (0.02 - 100)

2.9 (0.02 - 98)

0.2 (0.01 - 3.8)

Increase in severity overtime, n (%)

10 (1.8)

8 (1.7)

3 (3.6)

-

Evolution, n (%)

    

• overall

144 (24)

114 (25)

16 (19)

14 (25)

• MDS

70 (12)

47 (10)

11 (13)

12 (21)

• AML

10 (1.7)

7 (1.5)

1 (1.2)

2 (3.5)

• overt PNH

54 (10)

51 (11)

3 (3.6)

-

• other

10 (1.7)

9 (1.9)

1 (1.2)

-

Time to evolution (years), median (range)

4.7 (-1 - 38.2)

5.4 (0.2 – 32.5)

1.7 (0.2 – 38)

1.9 (-1 - 14)§

31.7 (23 - 71)*

Outcome, n (%)

    

• alive

331 (55)

265 (58)

33 (40)

33 (59)

• death

142 (24)

99 (21)

32 (39)

11 (20)

• lost at FU

124 (21)

95 (21)

17 (21)

12 (21)

  1. § Median time to evolution calculated from date of diagnosis
  2. * Median time to evolution calculated from date of birth
  3. AA, aplastic anaemia; hypo-MDS, hypoplastic myelodysplastic syndrome; IBMF, inherited bone marrow failure; KCH, King’s College Hospital; MDS, myelodysplastic syndrome; PNH, paroxysmal nocturnal haemoglobinuria.

Conclusions: We report the largest single institute series of cases with various BMF syndromes, including hypo-MDS and IBMFs in adult patients. Bone marrow failure syndromes are a heterogeneous group of disorders with a majority requiring treatment, including transplantation in 26% of patients, and increased risk of clonal evolution.

Disclosure: Nothing to declare.

20 - Aplastic Anaemia

P055 SINGLE CENTRE STUDY OF ROMIPLOSTIM ADDED TO IMMUNOSUPPRESSIVE THERAPY AS A FIRST-LINE TREATMENT IN SEVERE APLASTIC ANAEMIA: A CASE SERIES

Ashok Kumar Rajoreya1,1, Gaurav Dingra2

1RJN Apollo Spectra Hospital, Gwalior, India, 2All India Institute of Medical Sciences, Rishikesh, India

Background: Romiplostim, a thrombopoietin (TPO) mimetic protein, has been shown to promote trilineage haematopoiesis in patients with acquired aplastic anaemia (AA) that are refractory to immunosuppressive therapy (IST). However, its efficacy in the combination therapy with IST, i.e. anti-thymocyte globulin (ATG) plus cyclosporine (CSA) as a first-line treatment remains unexplored.

Aim: To evaluate the efficacy and safety of Romiplostim in combination with ATG and CSA as first-line treatment in patients with aplastic anaemia.

Methods: A Single centre, retrospective study of aplastic anaemia patients, every patient was given ATG + CSA + Romiplostim as a first-line treatment. Romiplostim 5µg/kg weekly for 1 month, post that dose was increased to 10µg/kg weekly for next 5 months. Data was evaluated at baseline, after 3rd and 6th month.

Results: Data of 12 patients with Median age of 18 years was evaluated. At a median follow-up of six months, the hematological response rate was 66.6% (25% achieved complete response, 41.6% achieved partial response) and 16.7% had no response. Two mortalities within 3 months of treatment were due to sepsis with pneumonia and cerebral haematoma with infection. Patients Haemoglobin level, total leukocyte count (TLC), Absolute Neutrophil count and Platelet count showed statistical significant improvement.

Table 15 Table 1: Mean values of Hb, TLC, ANC and platelet count at baseline, 3 months and 6 months

Conclusions: Although a larger number of patients and a longer follow-up period are needed to confirm our findings, our results show the efficacy of Romiplostim with ATG plus CSA as first-line treatment in patients with aplastic anaemia.

Disclosure: no conflict of interest.

20 - Aplastic Anaemia

P056 QUALITY OF LIFE AFTER IMMUNE SUPPRESSIVE TREATMENT FOR APLASTIC ANEMIA

Iris Nicola Lommerse 1, Constantijn Jozef Marie Halkes1, Chris Hinnen1, Jennifer May-Ling Tjon1

1Leids Universitair Medisch Centrum, Leiden, Netherlands

Background: Acquired aplastic anemia (AA) leads to progressive bone marrow failure and is treated with either immune suppressive therapy (IST) or allogeneic stemcell transplantation (alloSCT). Quality of live (QoL) after alloSCT can be diminished due to chronic complications like Graft Versus Host Disease (GVHD) but little is known about QoL after IST. We investigated the QoL in AA patients after successful IST.

Methods: For our research we used an AA/PNH specific QoL questionnaire developed by a German research group (Niedeggen et al. 2019; Annals of Hematology). Fifty-four multiple choice questions were translated in Dutch and send to 48 AA patients who were treated with IST (anti-thymocyte globulin (ATG), Ciclosporin and/or Eltrombopag). Four answer options could be chosen: ‘not at all’, ‘a little’, ‘a lot’ or ‘very often’. The questions represent possible issues within 12 different domains. Four domains were physically oriented (fatigue (FA), infections (IN), physical functioning (PF) and other symptoms (OS)), eight were more psychologically oriented (body image (BI), cognitive functioning (CF), emotional functioning (EF), illness intrusiveness (II), fear of progression (PAF), role functioning (RF), social support (SS) and stigmatization (ST)). Results were analyzed for all patients who responded to IST and grouped based on type of response (complete response (CR, normalization of blood counts) or partial response (PR, transfusion independent, but no normalization).

Results: Thirty-nine patients returned the questionnaire of whom 36 were successfully treated with IST (15 CR and 21 PR). Median age of these 36 patients was 54 years (range 21-71) Median time between last IST and the survey was 5 years, with a maximum of 41 years. Fatigue was experienced by 83% of patients and was scored ‘very often’ by 8%. Other physical-related domains such as PF, IN and OS were scored ‘not at all’ or ‘a little’ in the majority of patients. In contrast, most patients experienced psychological effects of their (past) illness. Specific items within PAF and BI were a major concern (‘a lot’ or ‘very often’) for up to respectively 31% and 36% of patients. When stratified based on hematological response, patients with PR scored worse than patients with CR in almost all domains. The difference was most clear in PAF: ‘very often’ was scored by up to 24% of patients with PR versus 7% of patients with CR.

QoL Domain

Mean % of patients with ‘a lot’ or ‘very often’ problems within this domain

 

Patients after successful IST (n = 36)

PR (n = 21)

CR (n = 15)

FA

24

26

21

IN

7

8

7

PF

20

16

25

OS

15

15

15

BI

27

34

17

CF

11

5

20

EF

23

28

16

II

19

23

15

PAF

24

29

15

RF

29

34

21

SS

10

10

11

ST

20

25

12

Conclusions: Patients with AA after successful IST still experience psychological and physical effects. Physical complaints appear to be less prominent, though fatigue is experienced by a large majority of patients. Partial responders to IST seem to experience more impact on their quality of life than complete responders. We have shown that QoL is affected also after successful IST and that the level of blood count normalization further influences QoL, especially in psychological domains.

In conclusion, evaluating QoL in AA patients might identify specific domains requiring extra attention and can be used to improve personalized supportive measures. In the future, it could be insightful to include QoL as a parameter in studies evaluating AA treatment effects, too.

Disclosure: Nothing to declare.

21 - Autoimmune Diseases

P057 EXPERIENCE IN THE USE OF ABATACEPT FOR THE MANAGEMENT OF REFRACTORY AUTOIMMUNE HEMOLYTIC ANEMIA IN PEDIATRIC PATIENTS AFTER ALLOGENIC STEM CELL TRANSPLANTATION

Carlos Echecopar Parente 1, Maria Dolores Corral Sánchez1, David Bueno Sánchez1, Luisa Sisinni1, Yasmina Mozo Del Castillo1, Gema Casado Abad2, Antonio Pérez Martínez1,3,4

1Pediatric Hemato-Oncology, University Hospital La Paz, Madrid, Spain, Hospital La Paz, Madrid, Spain, 3University Hospital La Paz Research Institute (IdiPAZ), Madrid, Spain, Universidad Autónoma de Madrid, Madrid, Spain

Background: Autoimmune hemolytic anemia (AIHA) is a rare complication after hematopoietic stem cell transplantation (HSCT) with significant morbidity and mortality. Response to standard therapy in this patients is worse than in cases not associated with HSCT, so it’s necessary to seek for alternative therapies.

Methods: We describe our experience using abatacept for the treatment of refractory or corticodependant AIHA after HSCT since July 2019 to April 2020.

Results: We used abatacept in five patients with AIHA after HSCT, four cases with refractory AIHA after standard therapy and one case with corticodependant AIHA. All patients had clinical response associated to the therapy and a complete withdrawal of steroids was possible in all cases. To date, 4 patients completed treatment with abatacept, one of them had a relapse after withdrawal of treatment. We did not see any side effects associated with the use of this therapy.

Table 17 Table 1. Patients characteristics and treatment of AIHA

Conclusions: Abatacept can be an alternative therapy for refractory or corticodependent AIHA after HSCT.

Disclosure: Nothing to declare.

21 - Autoimmune Diseases

P058 SARS-COV-2 VACCINATION IN SYSTEMIC SCLEROSIS PATIENTS TREATED BY AUTOLOGOUS HEMATOPOIETIC STEM CELL TRANSPLANTATION, ON BEHALF OF MATHEC-SFGM-TC GROUP

Lucie Biard1, Grégory Pugnet2, Carlotta Cacciatore 1, Ingrid Munia1, Pauline Lansiaux1, Christine Le Maignan1, Benjamin Crichi1, Fanny Urbain3, Zora Marjanovic4, Sabine Berthier5, Doron Rimar6, Dominique Farge1

1AP-HP Saint-Louis Hospital, Paris, France, 2CHU Rangueil, Toulouse, France, 3AP-HP Bicetre Hospital, Paris, France, 4AP-HP Saint-Antoine Hospital, Paris, France, 5CHU Dijon, Dijon, France, 6Bnai Zion Medical Centre, Haifa, Israel

Background: On December 27th 2020, the implementation of SARS-CoV-2 vaccination in France first became available amongst all immunosuppressed patients, specifically Hematopoietic Stem Cell Transplant (HSCT) recipents or other fragile immunosupressed patients. In July 2021, the ADWP-EBMT guidelines recommended vaccination against SARS-CoV2 as early as 3 months after HSCT, but only one Israeli study so far analyzed the efficacy of SARS-CoV-2 vaccination after autologous HSCT (AHSCT) for Systemic Sclerosis (SSc). We designed this study to assess the acceptance and effectiveness of such vaccination program in SSc patients treated by AHSCT as compared to other fragile SSc individuals.

Methods: This retrospective case-control cohort analysis included SSc patients above 18 years old, diagnosed according to the 1980 ARA or the 2013 ACR/EULAR criteria, followed in two French expert centers and who were offered SARS-CoV-2 vaccination according to the French Good Clinical Practices (with at least either a two-dose mRNA vaccine or single-dose adenoviral vector) between January 1st 2021 to June 30th 2022. AHSCT SSc patients (cases) were matched 1:1 with non-AHSCT SSc patients (controls) according to the following criteria on January 1st 2021: sex, age (+/- 5 years), disease duration since first non-Raynaud phenomenon (+/-5 years). Nearest-neighbor Mahalanobis distance matching was performed.

Primary endpoint: cumulative incidence of asymptomatic (positive PCR on routine testing) or symptomatic (positive PCR plus COVID-19 clinical symptoms) infection.

Secondary endpoints: proportion of patients refusing vaccination, cumulative incidence of symptomatic COVID-19 infection, time of onset of first COVID-19 infection (symptomatic or asymptomatic) during the study after completing at least 2 vaccines injections, COVID-19 severity, proportion of patients with at least 1 grade≥3 adverse event (CTC-AE v5.0) and humoral response after vaccination.

Results: 36 SSc patients treated by AHSCT and 36 controls (22 females in each group, overall cohort median age 53 (46;63) years, prior median SSc duration 11 (7;15) years) were retrospectively studied from January 1st 2021 until June 30th 2022. Five (14%) of the AHSCT patients refused vaccination while the others received either 1 (n = 1, 3%), 2 (n = 4, 11%) or 3 (n = 26, 72%) vaccine injections, essentially BNT162b2-Pfizer (94% for 1st injection, 93% for 2nd injection and 88% for 3rd injection). Among the 36 AHSCT patients, 11 experienced one COVID-19 infection during the study period, of whom 1 had refused vaccination. All were symptomatic: 9 (82%) with mild symptoms and 2 (18%) with moderate symptoms which required hospitalization outside ICU. Among AHSCT patients, the cumulative incidence of symptomatic COVID-19 starting from 1st January 2021 was 8% (95%CI 2-20), 14% (95%CI 5-28) and 33% (95%CI 18-49) at 6, 12 and 18 months respectively (figure 1). There was no death from COVID-19 infection in the controls.

Figure 1

Conclusions: Out of 11 COVID-19 infections occurring in SSc patients treated by AHSCT once the vaccination campaign had started, only 2 required hospitalization that occurred in the first quarter of 2021, amongst patients who had not received a complete vaccination. In this unique cohort, most COVID-19 infections occurred by the end of 2021– early 2022, during the 5th epidemic wave in France.

Disclosure: Nothing to declare.

21 - Autoimmune Diseases

P059 THE FIRST AUTOLOGOUS HEMATOPOIETIC STEM CELL TRANSPLANT FOR A MULTIPLE SCLEROSIS PATIENT IN THE UNITED ARAB EMIRATES

Inas El-Najjar 1, Ruqqia Mir1, Yendry Ventura-Carmenate1, Fatema Al-Kaabi1, Maysoon AlKaram1, Yandy Castillo-Aleman1, Hiba Abdelrahman1, May Martinez1, Dina El-Mouzain1, Antonio Bencomo-Hernandez1, Alzahraa Abouhelal1, Tarek Diab1, Pierre Krystkowiak1, Nameer Al-Saadawi1

1Abu Dhabi Stem Cell Centre, Abu-Dhabi, United Arab Emirates

Background: Autologous hematopoietic stem cell transplant (AHSCT) is recommended for patients with certain types of multiple sclerosis (MS). We endeavored to make AHSCT locally accessible to patients living in the UAE, to avoid the need to travel for treatment. We report clinical lessons learnt with the first AHSCT in a MS patient done in the Gulf region, Abu Dhabi, UAE.

Methods: The patient is a 24 year old female diagnosed with aggressive, highly active MS in November 2021 with Expanded Disability Status Scale (EDSS) 8.5. She had treatment with Ocrelizumab on 02/02/2022, despite this treatment, a MRI in three months revealed new active MRI lesions, in keeping with disease progression. She was assessed in the Hematology clinic in August 2022 and deemed fit for AHSCT.

She was mobilized on 24/09/2022, utilizing Cyclophosphamide priming 2.0g/m2, followed five days later by granulocyte colony stimulating factor (G-CSF) 5 mcg/kg/day for five days. A single leukapheresis procedure achieved a dose of 17.7 x 106 CD34+ cells/Kg. She was admitted on 01/11/2022 to start conditioning. A non-myeloablative protocol was utilized includes cyclophosphamide 50 mg/kg/iv over 2 hours daily for four days(D-5 to D-2) along with MESNA 50 mg/kg/day continuous infusion. Rabbit antithymocyte globulin (ATG) in increasing doses was infused over five days (D-5 to D-1). We infused 8.6 x 106 CD34+ cells/Kg on D0 (07/11/2022). She was on our standard anti-infective prophylaxis.

Results: Engraftment: Patient engrafted on D + 9, with neutrophil count > 0.5 x 109/L.

Transfusions: No platelet transfusion required, platelets did not drop below 45 x 109/L, she had two Packed Red Blood Cells (PRBC) transfusions during admission on D + 9 and D + 10, that were irradiated and leukodepleted.

Fever and Infections: ATG associated fever on D0, this abated on restarting methylprednisolone(MP) 500 mg once daily(OD) for two days. She had further fever D + 2 treated with 2 further days of MP 250 mg OD for two days. During this time she had had a full septic work up, including blood cultures, urine culture, and a chest X-ray She developed fever again on D + 10 but was clinically well. In addition to another septic screen, CMV and EBV PCR were sent. She was on broad spectrum antibiotics and her central line was removed. A CT chest/abdomen/pelvis revealed only splenomegaly on D + 17. CMV PCR report positive 3050 IU/ml (CMV log 3.48, she was CMV PCR negative on admission). She was started on valganciclovir on D + 17. Patient’s fever settled within 48 hours. She was discharged on D + 20.

Neurology: She had mild neurological improvement whilst still an inpatient D + 15 improving in functional scores in the cerebellar and pyramidal systems.

Conclusions: Unlike AHSCT in adult malignant diseases, MS patients may not require platelet transfusions at all. In addition, monitoring and suspicion of CMV infection is essential as it may cause fever at D + 10, which we didn’t expect. Early neurological improvement is feasible post AHSCT, the patient will be followed up to assess for further sustained improvement. We believe this is the first step in making AHSCT accessible for MS patients in the UAE.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P060 TREATMENT OF SYSTEMIC SCLEROSIS AND ANTI-SYNTHETASE SYNDROME WITH CD19-CAR T CELLS

Fabian Müller 1, Sebastian Böltz1, Christina Bergmann1, Johannes Knitza1, Simon Völkl1, Michael Aigner1, Soraya Kharboutli1, Arnd Kleyer1, Hannah Reimann1, Sascha Kretschmann1, Jochen Wacker1, Wolf Rösler1, Bernd Spriewald1, Gerhard Krönke1, Gerhard Schett1, Andreas Mackensen1

1University Hospital of Erlangen, Erlangen, Germany

Background: Using CD19 CAR T cells, we recently achieved lasting remission in 5 of 5 patients with severe and treatment refractory systemic lupus erythematosus (SLE, Mackensen et al. NatMed; 2022). Similar to lupus, other autoimmune diseases may depend on pathologic B cell clones including anti-synthetase syndrome (SAA), a form of inflammatory myositis or systemic sclerosis (SSc). In line, B cell directed treatment often induces remission in SAA or SSc even though refractory or rapidly relapsing cases show insufficient B cell depletion. Thus, we hypothesized that patients with severe, refractory forms of SAA or SSc may achieve durable responses following CD19 CAR T cells.

Methods: In this pilot study we used autologous CD19 CAR T cells produced in house with the Miltenyi prodigy. Patients received lymphodepletion with 3x25 mg/m² Fludarabin, days-5 to -3 and Cyclophosphomaid 1g/m² d-3 followed by 1 mio/kgBW CAR T cells on day 0. CAR T cells were analyzed using flow cytometry. SAA was assessed by MRI and by serum creatinine kinase (CK) and severity of SSc by fibroblast activation protein inhibitor (FAPI)-PET CT scan. Both diseases were also staged by clinical scores.

Results: A heavily pretreated patient with SAA (prednisolone (250 mg/day), rituximab, IVIG, tacrolimus, and cyclophosphamide) proved refractory with progressive, severe muscle weakness, pulmonary involvement with increasing oxygen supplementation and progressive alveolitis by CT. At CAR-infusion, CK was 9305 U/l (normal <190 U/l). CK increased further following lymphodepletion and before Car expansion to a maximum of 13,600 U/l. CAR T cells expanded by more than 600-fold. Apart from fever (CRS grade 1) treated with tocilizumab, the patient did not show relevant side effects. Then, CK started to slowly decline over weeks to finally normal values. Manual muscle test score of 115/150 at baseline improved to full muscle strength at day +180 (149/150). Myositis by MRI was abrogated and maximum walking distance increased from 10 m to more than 5 km. Alveolitis mostly disappeared.

A second patient with SSc showed anti-RNAP antibodies and progressive systemic sclerosis by skin biopsies. Furthermore, the patient presented with joint pain, diffuse lung fibrosis, and heart-involvement as demonstrated by MRI. Despite several treatment lines (steroids, MTX, and mycofenolat mofetil), disease progressed with increasing skin lesions and worsening of lung and heart function. Before CAR T, the patient received 50% lymphodepletion due to reduced kidney function. CAR T cells expanded well and peaked at day 9 with >1200 CAR T cells/nl. Except for mild fever which was treated symptomatically, the patient had no side effects. More than 3 months after CAR T therapy, the patient was treatment-free, skin sclerosis remained stable, FAPI-PET showed minimal enhancement, joint pain was greatly reduced, and clinical tests gradually improved.

Conclusions: As in SLE, CAR T cell therapy is safe in two cases of SAA and SSc. While in established SSc disease progression was arrested, disease activity and clinical symptoms in SAA were abrogated with a single shot of car T. These promising early response signals prompt further clinical evaluation.

Disclosure: FM and AM have received travel support and honoraria from Miltenyi.

All other authors do not declare relevant confilicts of interest.

4 - CAR-based Cellular Therapy – Clinical

P061 THIRD-GENERATION CHIMERIC ANTIGEN RECEPTOR (CAR) T CELLS IN ADULT ALL PATIENTS

Maria-Luisa Schubert1, Anita Schmitt1, Angela Hückelhoven-Krauss1, Brigitte Neuber1, Philip Waldhoff1, Dominik Vonficht2,3, Schayan Yousefian4,5,6, Lea Jopp-Saile2,3, Felix Korell1, Simon Haas3,2,5,4,6,7, Anthony D. Ho1,7, Carsten Müller-Tidow1,7, Peter Dreger1,7, Michael Schmitt 1,7

1Heidelberg University Hospital, Heidelberg, Germany, 2Deutsches Krebsforschungszentrum (DKFZ), Heidelberg, Germany, 3Institute for Stem Cell Technology and Experimental Medicine, Heidelberg, Germany, 4Max Delbrück Center, Berlin, Germany, 5Berlin Institute of Health, Berlin, Germany, 6Charité University Medicine, Berlin, Germany, 7German Cancer Consortium, Heidelberg, Germany

Background: Third-generation chimeric antigen receptor (CAR)-engineered T-cells (CARTs) might improve clinical outcome of patients with B-cell malignancies. Here we report a phase-1/2 investigator-initiated trial (HD-CAR-1) on a third-generation CART dose-escalating, investigator-initiated trial treating adult patients with acute lymphoblastic leukemia (ALL).

Methods: Refractory and/or relapsed (r/r) adult ALL patients received escalating doses of CD19-directed third-generation CARTs comprising the costimulatory domains CD28 and CD137 (4-1BB) after lymphodepletion with fludarabine and cyclophosphamide. Leukapheresis, manufacturing and administration of CARTs were performed in-house.

Results: 15 patients with r/r ALL were enrolled. Median age of patients was 41 (range 21 to 67) years. Median time from initial diagnosis to CART administration was 22 (range 5 to 117) months and patients had received a median of 4 (range 2 to 9) prior treatment lines, including allogeneic stem cell transplantation (alloSCT) in 12 patients (80%). For all patients, CART manufacturing was feasible. Median duration of CART manufacturing was 10 (range 10 to 14) days. Median transduction efficiency was 52.7% (range 39.3% to 66.9%) with a viability of CARTs of > 85%. Six patients received bridging therapy between leukapheresis and lymphodepleting therapy. Thirteen patients were treated with HD-CAR-1 CARTs (1×106 CARTs/m2 (n = 3), 5×106 CARTs/m2 (n = 3), 2x107 CARTs/m2 (n = 4), 5×107 CARTs/m2 (n = 3)). Two patients did not receive CARTs due to progressive disease during CART manufacturing. None of the thirteen treated patients developed any grade of immune effector cell-associated neurotoxic syndrome (ICANS) or a higher-grade ( ≥ grade III) cytokine release syndrome (CRS).

CART expansion and long-term CART persistence were evident in the peripheral blood of evaluable patients. At end-of-study (EOS) on day 90 after CART administration, ten patients were evaluable for disease response: eight patients (80%) achieved a complete remission (CR), including five patients (50%) with MRD-negative CR. Response and outcome were associated with the administered CART dose. Persistent, high-grade ( ≥ III) neutropenia beyond day 90 was observed in two patients. At EOS, all evaluable patients had ongoing B-cell aplasia, even though recovered levels of immunoglobulins were detectable in six patients. At 1-year-follow-up, median overall survival (OS) was not reached and progression-free survival (PFS) was 38%. Median PFS was reached on day 120. Responders displayed higher frequencies of a specific memory-like T cell subset within the CART-product. Also, a physiological pattern of immune cells and lower monocyte counts in the PB were associated with response.

Conclusions: Third-generation HD-CAR-1 CARTs were remarkably safe and of promising efficacy. A specific subset of memory T cells within the CART product could predict response to treatment. Overall, HD-CAR-1 appears to be a promising step towards safe and effective ALL eradication.

Clinical Trial Registry: NCT03676504 (www.clinicaltrials.gov)

Disclosure: AS: Travel grants from Hexal and Jazz Pharmaceuticals. Research grant from Therakos/Mallinckrodt. Consultancy BMS, Janssen-Cilag. Co-founder and part-time employee of TolerogenixX LtD. of TolerogenixX Ltd.

CMT: research support from Bayer AG. Advisory board member Pfizer, Janssen-Cilag GmbH. Grants and/or provision of investigational medicinal products from Pfizer, Daiichi Sankyo, BiolineRx.

MS: research grants from Apogenix, Hexal and Novartis. Travel grants from Hexal and Kite. Financial support for educational activities and conferences from bluebird bio, Kite and Novartis. Advisory board member of MSD. (Co-)PI of clinical trials of MSD, GSK, Kite and BMS. Co-Founder and shareholder of TolerogenixX Ltd.

PD: consultancy AbbVie, AstraZeneca, Gilead, Janssen, Novartis, Riemser, Roche; speakers bureau AbbVie, Gilead, Novartis, Riemser, Roche; research support from Neovii and Riemser.

ADH, AHK, BN, DV, LJS, MLS, PW, SH, SY: none.

4 - CAR-based Cellular Therapy – Clinical

P062 TORQUE TENO VIRUS PLASMA DNA LOAD: A NOVEL PROGNOSTIC BIOMARKER IN CAR-T THERAPY

Rafael Hernani 1, Estela Giménez2, Gloria Iacoboni3,4, Manuel Guerreiro5, Ana Benzaquén1, Eliseo Albert2, Cecilia Carpio3, Aitana Balaguer5, Ariadna Perez1, Mario Sánchez-Salinas3, Pedro Chorão5, José Luis Piñana1, Francisco Beas3, Juan Montoro5, Juan Carlos Hernández-Boluda1,6, Ana Facal5, Blanca Ferrer1, Marta Villalba5, Paula Amat1, María José Terol1,6, Jaime Sanz5,6, Pere Barba3,4, David Navarro2,7, Carlos Solano1,6

Hospital Clínico Universitario, INCLIVA Research Institute, Valencia, Spain, Hospital Clínico Universitario, INCLIVA Research Institute, Valencia, Spain, Hospital Universitari Vall d’Hebron, Barcelona, Spain, 4Universitat Autònoma de Barcelona, Bellaterra, Spain, Hospital Universitari i Politècnic La Fe, Valencia, Spain, University of Valencia, Valencia, Spain, University of Valencia, Valencia, Spain

Background: Disease-refractory and immunological toxicities are associated with significant morbidity and failure in chimeric antigen receptor T-cell (CAR-T) therapy. Novel biomarkers could be of value to predict these complications. Torque teno virus (TTV) has been identified as a surrogate marker of immunocompetence both in solid organ (SOT) and hematopoietic stem cell transplantation (HSCT).

Methods: Adult patients who received commercial CAR-T (tisa-cel or axi-cel) for B-cell non-Hodgkin lymphoma (B-NHL) in three Spanish centers between January 2020 and June 2022 were included. TTV DNA plasma loads were quantified by real-time PCR at different timepoints before (prelymphodepletion [TTVpreLD], preinfusion [TTVd0]) and after CAR-T infusion (days +1, +3, +5, +7, +14, +21, +28, +60 and +90). Cytokine-release syndrome (CRS) and immune-effector cells associated neurotoxicity (ICANS) were graded following ASCTC criteria. The association between baseline factors and overall response (ORR), CRS and ICANS was estimated by univariate logistic regression model. Overall survival (OS) was calculated using the Kaplan–Meier method and the log-rank test was used for statistical comparison. Cumulative incidences were analyzed by Gray test. Preliminary data of preinfusion plasma samples is provided here.

Results: Seventy-nine patients were included. CRS and ICANS were developed in 63 (79.7%) and 24 (30%) patients, being severe in 7 (9%) and 18 (22.7%) patients, respectively. Seven (9%) patients died of toxicity (infection, n = 5; ICANS, n = 1; undetermined, n = 1) at a median time of 46 days (7 – 772). The 1-year cumulative incidence of therapy-related mortality (TRM) was 6% (1 – 12). ORR was achieved in 54 (68%) patients (complete response, n = 39; partial response, n = 15) at a median time of 38 days (25 – 351). ORR was maintained in 32 patients at last follow-up with a median duration of response (DOR) of 14 months. Forty-five (57%) patients remained alive at a median follow-up of 371 days (91 – 882). OS at 3 months and 1 year was 85% (77 – 93) and 56% (44 – 68), respectively. In multivariate analysis, lower TTVd0 was associated with OS (HR 1.15, 95% CI, 1.006 – 1.322, P = 0.041). OS at 1 year was 75% (61 – 94) and 40% (25 – 66) for patients with TTVd0 under 5 log10 copies/mL or equal/above 5 log10 copies/mL (P = 0.041). TTVpreLD and TTVd0 were also related to ORR (P = 0.068 and 0.013, respectively), but no association was found in multivariate analysis. Neither CRS, ICANS, DOR nor TRM were related to TTV plasma load.

Conclusions: In this study, preliminary analyses show that preinfusion TTV plasma load may be associated with OS in adult patients receiving CAR-T therapy for B-NHL. Lower TTV could be linked to a better immunocompetence, which may translate into an improved CAR-T effect. These results suggest that TTV monitoring could be useful in predicting CAR-T therapy efficacy and may have a potential for the development of personalized strategies for CAR-T patients, as previously suggested in SOT and HSCT receptors. To our knowledge, this is the first time TTV is used as a biomarker in CAR-T therapy. More data is warranted to confirm these preliminary findings.

Clinical Trial Registry: The Institut Paoli-Calmettes (Marseille, France) was also included as a collaborator center. The study was considered as a clinical trial in France (NCT04822974). Unfortunately, covid pandemic avoid the implementation of the study in France. All the authors would like to express our thanks to our colleagues from Marseille, especially to Pr Didier Blaise and Dr Raynier Devillier.

Disclosure: A Gilead grant (Beca FEHH-GILEAD para Formación en Investigación en Terapia Celular en un centro internacional Convocatoria 2019) was conceded to Ana Benzaquén. Rafael Hernani was the principal investigator of the grant-associated study. The rest of authors declare no conflict of interest related to this study.

4 - CAR-based Cellular Therapy – Clinical

P063 IMMUNE RECONSTITUTION AND INFECTIONS AFTER TREATMENT WITH COMMERCIAL CAR-T CELL THERAPY

Eva Catalá 1,2, Diego Carbonell1,2, Gillen Oarbeascoa1,2, Mariana Bastos1,2, Rebeca Bailén1,2, Ana Pérez Corral1, Paula Fernández-Caldas1,2, Ana Alarcón1,2, Ignacio Gómez-Centurión1,2, Diego Conde1,2, Jose Luis Díez1,2,3, Jorge Gayoso1,2, Javier Anguita1,2, Mi Kwon1,2

Hospital General Universitario Gregorio Marañón, Madrid, Spain, 2Instituto de Investigación Sanitaria Gregorio Marañón, Madrid, Spain, School of Medicine, Complutense University of Madrid, Madrid, Spain

Background: Commercial anti-CD19 CAR-T cell therapy with tisa-cel and axi-cel has improved outcomes in refractory/relapsed DLBCL patients, but infection rates are significant after receiving this therapy. Our aim was to analyze differences in terms of immune reconstitution dynamics and infection rates between both commercial products.

Methods: We performed a single center retrospective study in 50 patients receiving commercial CAR-T cell products from August-2019 to July-2022. Immune reconstitution dynamics were monitored at +3, +7, +14, +30, +90, +180 and +360 days after infusion. Peripheral blood lymphocyte subsets were analyzed by flow cytometry. Infections were classified as early (<30 days) and late ( > 30 days) and by microbial etiology. Mann Whitney U test was used to analyze differences in cell population levels and immunoglobulins between tisa-cel and axi-cel. ROC curves were used to obtain the best cut-off point of cell number to subsequently perform cumulative incidence (CI) analysis. Fine-Gray test was used to calculate the CI of infection from the day of obtaining the variable analyzed in each case, considering death as competitive event.

Results: Characteristics of patients and complications associated to CAR-T cell therapy are shown in Table 1. 50 patients received commercial CAR-T cell therapy, with a minimum follow-up of 3 months. A total of 48 infections were detected during de follow-up, 22 were early infections: 12 (54%) were bacterial, 9 (41%) viral and 1 (5%) fungal. Four (18%) patients required hospitalization and none of them required intensive care. 26 were late infections: 10 (38%) bacterial, 15 (58%) viral and 1 (4%) fungal, with a median of appearance at day 102 (range 32-489) after infusion. 13 (50%) required hospitalization and 1 (4%) required intensive care. No differences were found in the rate of infections between tisa-cel or axi-cel, either total incidence, early and late infections, or bacterial or viral infection.

Total T cell count depleted after LD chemotherapy, showing a better expansion of total T cells (specifically CD4) after infusion in tisa-cel compared to axi-cel from +3 to +180 days after infusion (p < 0,001). CD8 T cells were significantly higher in tisa-cel from day +3 to +14. 14 (28%) patients presented serum IgG levels <400 mg/dL after LD chemotherapy, 9 (18%) of them received endovenous immunoglobulins. By day +90, 11 (22%) patients did not reach >400 mg/dL IgG, with no differences between both products. Finally, values of 643 T cells/uL and 424 CD8 T cells/uL were identified as cut-off points that stratify patients into low and high-risk of infection after day +90 (p < 0.01).

Table 1. Patient characteristics

 

Tisa-cel

Axi-cel

 

N = 18

N = 32

Sex, male, n (%)

11 (61)

14 (44)

Age, median (range)

63 (41-73)

63 (33-79)

Bridging therapy, n (%)

12 (67)

31 (97)

ECOG, n (%)

0-1

18 (100)

32 (100)

2-3

0 (0)

0 (0)

Histology, n (%)

DLBC

16 (89)

26 (81)

Transformed FL

2 (11)

2 (6)

Primary mediastinal lymphoma

0 (0)

4 (13)

Disease status at apheresis, n (%)

Progressive disease

8 (44)

19 (60)

Stable disease

6 (33)

4 (13)

Partial response

2 (11)

6 (19)

Complete response

2 (11)

3 (10)

Primary refractory, n (%)

8 (44)

20 (63)

Previous lines, median (range)

2 (2-4)

2 (2-5)

Prior ASCT, n (%)

9 (50)

11 (34)

Prior Allo-SCT, n (%)

1 (6)

0 (0)

CRS 2, n (%)

2 (11)

12 (38)

Tocilizumab

2 (11)

12 (38)

Corticosteroids

0 (0)

10 (31)

ICANS, n (%)

1 (6)

6 (19)

Corticosteroids

1 (6)

6 (19)

Siltuximab

0 (0)

4 (13)

Anakinra

0 (0)

4 (13)

Conclusions: Infections after CAR-T cell therapy are frequent and may be severe. Low rates of total T cell counts and CD8 T cells are associated to an increase risk of infections after day +90 after commercial CAR-T cell therapy. Tisa-cel shows a better CD4 T cell reconstitution although this does not seem to influence in presenting significant higher infection rates compared to axi-cel. Microbial etiology is similar in both products, with bacterial infections predominant in the first 30 days and later viral infections.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P064 OUTCOMES AFTER ANTI-BCMA CAR T-CELLS FOR MULTIPLE MYELOMA, A RETROSPECTIVE ANALYSIS ON BEHALF OF THE CTIWP OF THE EBMT

Florent Malard 1, Edouard F. Bonneville2, Francesca Bonifazi3, Johan Maertens4, Tobias Gedde-Dahl5, Michael Daskalakis6, Goda Choi7, Hermann Einsele8, Alessandro Busca9, Christof Scheid10, Jacques-Olivier Bay11, Paolo Corradini12, Lucía López Corral13, Anne Sirvent14, Ibrahim Yakoub-Agha15, Nico Gagelmann16, Patrick Hayden17, Laurent Garderet18, Jorinde D. Hoogenboom19, Jurgen Kuball20, McLornan Donal P.21, Annalisa Ruggeri22

1Sorbonne Université, Paris, France, 2Leiden University Medical Center, Leiden, Netherlands, 3IRCCS Azianda Ospedaliero-Universitaria di Bologna, Bologna, Italy, 4University Hospital Gasthuisberg, Leuven, Belgium, 5Oslo University Hospital, Oslo, Norway, 6University Hospital Bern, Bern, Switzerland, 7University Medical Center Groningen (UMCG), Groningen, Netherlands, 8Universitaetsklinikum Wuerzburg, Wuerzburg, Germany, 9S.S.C.V.D Trapianto di Cellule Staminali, Torino, Italy, 10University of Cologne, Cologne, Germany, 11CHU ESTAING, Clermont-Ferrand, France, 12University of Milano, Milano, Italy, 13Hospital Clínico, Salamanca, Spain, 14CHU Lapeyronie, Montpellier, France, 15CHU de Lille, Univ Lille, INSERM U 1286, Infinite, Lille, France, 16University Medical Center Hamburg-Eppendorf, Hamburg, Germany, 17St. James’s Hospital, Dublin, Ireland, 18Hôpital Pitié-Salpêtrière, Paris, France, 19EBMT Leiden Study Unit, Leiden, Netherlands, 20University Medical Centre, Utrecht, Netherlands, University College London Hospital, London, United Kingdom, 22San Raffaele Scientific Institute, Hematology and Bone Marrow Transplantation Unit, Milan, Italy

Background: Idecabtagene vicleucel (ide-cel) and Ciltacabtagene Autoleucel (cilta-cel) are autologous B-cell maturation antigen (BCMA)-directed chimeric antigen receptor (CAR) T-cell therapies, approved for relapsed/refractory multiple myeloma (RRMM). We aimed to evaluate outcomes of patients treated with standard of care (SOC) ide-cel and cilta-cel and reported to the cellular therapy registry of the EBMT.

Methods: The EBMT Cellular Therapy and Immunobiology Working Party (CTIWP) conducted a retrospective analysis amongst EBMT registered centers to evaluate outcomes of patients with MM receiving anti-BCMA CAR T-cells. Patient-, disease- and treatment characteristics were collected. Patient outcomes with a focus on disease response rates and rates/grades of cytokine release syndrome (CRS) were also evaluated. Descriptive statistics were utilized to report patient, disease and treatment characteristics as well as patient outcomes for both cohorts (ide-cel and cilta-cel).

Results: Overall, 53 patients with MM were reported with available data (33 patients infused with ide-cel and 20 with cilta-cel by time of data cut-off). Median age was 63 years for patients treated with ide-cel and 59 years for cilta-cel. A total of 25% of the patients had light chain multiple myeloma. The majority of patients had a history of autologous HCT (ide-cel, 73%; cilta-cel 85%) and had received an immunomodulatory drug (ide-cel, 84%; cilta-cel 87%), a proteasome inhibitor (ide-cel, 84%; cilta-cel 80%) and anti-CD38 monoclonal antibody (ide-cel, 74%; cilta-cel 53%).

Median time between leukapheresis and CAR T-cells infusion was 1.76 months for ide-cel and 2.53 months for cilta-cel. The majority of patients were treated within a clinical trial setting (ide-cel, 84%; cilta-cel 100%).

Regarding CRS, data was available from 15 patients treated with cilta-cel and 9 patients treated with ide-cel, that all developed CRS. For patients treated with cilta-cel, 53% developed grade 1 CRS, 40% grade 2 and 7 % grade 3, none developed grade 4 CRS. For patients treated with ide-cel, 44% developed grade 1 CRS, 56 % grade 2 and none developed grade 3-4 CRS. Best overall response rate ( ≥ partial response) and ≥ complete response (CR) at day 100 post infusion were 88% and 38% respectively for ide-cel and 94% and 67% for cilta-cel, respectively. With a median follow-up of 9.2 months, 10 patients had died (ide-cel, n = 8, cilta-cel, n = 2).

Conclusions: Therapy with the anti-BCMA CAR T-cell products ide-cel and cilta-cel translate into high response rates in patients with highly pretreated MM, including previous lines of treatment including autologous HCT, immunomodulatory drugs, proteasome inhibitors and anti-CD38 monoclonal antibodies. Furthermore, in patients that developed CRS, the majority developed grade 1-2 CRS and severe CRS appeared very uncommon. Overall, analysis of real-life data is important, regarding response, but also toxicities, in order to guide treatment choice; not only between available anti-BCMA CAR T-cells but also with regard to bispecific monoclonal antibodies that are increasingly available for patients with MM.

Disclosure: Florent Malard received honoraria from BMS, Ibrahim Yakoub-Agha received honoraria from BMS and Janssen. Others authors did not disclose conflict of interest.

4 - CAR-based Cellular Therapy – Clinical

P065 REAL-WORLD EXPERIENCE OF TISAGENLECLEUCEL FOR CHILDREN, ADOLESCENTS AND YOUNG ADULTS WITH ALL IN JAPAN

Itaru Kato 1, Shinsuke Hirabayashi2, Daisuke Tomizawa3, Atsushi Manabe2, Masahiro Irie4, Yoji Sasahara4, Yuki Arakawa5, Katsuyoshi Koh5, Hirotoshi Sakaguchi3, Masanaka Sugiyama6, Chitose Ogawa6, Takahiro Kamiya7, Motohiro Kato8, Shoji Saito9, Yozo Nakazawa9, Nobuhiro Nishio10, Yoshiyuki Takahashi10, Takako Miyamura11, Naoko Iwai1, Souichi Adachi1, Junko Takita1, Satomi Yokoyama12, Wakako Kato12, Hiroaki Ono12, Yuhki Koga12, Hidefumi Hiramatsu1

1Kyoto University, Kyoto, Japan, 2Hokkaido University Graduate School of Medicine, Sapporo, Japan, 3National Center for Child Health and Development, Tokyo, Japan, 4Tohoku University Graduate School of Medicine, Sendai, Japan, 5Saitama Children’s Medical Center, Saitama, Japan, 6National Cancer Center Hospital, Tokyo, Japan, 7Tokyo Medical and Dental University Hospital, Tokyo, Japan, 8The University of Tokyo, Tokyo, Japan, 9Shinshu University School of Medicine, Matsumoto, Japan, 10Nagoya University Graduate School of Medicine, Nagoya, Japan, 11Osaka University Graduate School of Medicine, Osaka, Japan, 12Kyushu University, Fukuoka, Japan

Background: Although the majority of pediatric B-cell acute lymphoblastic leukemia (ALL) patients have great outcomes, about 15% of individuals will experience a relapse. In the case of relapse, one major new treatment option is chimeric antigen receptor (CAR) T cells that target CD19. The pivotal phase 2, multicenter, global trial of a CAR T-cell therapy (ELIANA; NCT02435849) included 9 patients (median, 11 years old [5-24 years old]) in Japan (Int J Hematol. 2020;111:303–310). Tisagenlecleucel was approved in 2019 in Japan, and more than 3 years have passed since then.

Methods: The Japan CAR-T Consortium developed a database for children, adolescents, and young adults with relapsed and/or refractory B-ALL who received tisagenlecleucel to investigate the real-world (RW) experience. Eleven centers in Japan contributed to the retrospective RW registry analysis after May 22, 2019, and a total of 42 infused patients (median, 10 years old [1-23 years old]) were analyzed as of Feb 1, 2022. The best overall response of complete remission (CR), event-free survival (EFS), and overall survival (OS) were evaluated as efficacy, and cytokine release syndrome (CRS) was graded using the American Society for Transplantation and Cellular Therapy criteria as a safety profile.

Results: In RW registry, the median follow-up period of the survivors was 312 days (63-813 days). The best overall response rate (CR/CRi) was 92.9%. MRD after infusion was evaluated in 36 patients and 35 (97.2%) were negative. The 1-year OS and EFS rates after infusion were 81.6% and 56.3%, respectively. Twenty-seven patients (64%) had low-disease burden (LB, defined as <5% bone marrow (BM) lymphoblasts) pre-tisagenlecleucel infusion, and LB was associated with superior outcomes, with 1-year EFS rate of 80% compared with that of 24% in cases with a high-disease burden (5% BM lymphoblasts) (p < 0.0001), that is the same observation as previously reported (J Clin Oncol. 2022;40(9):945-955). Notably, multivariate analyses revealed an association of prior hematopoietic stem cell transplantation (HSCT) (n = 23, 55%) with superior outcomes (HR 0.084, 95%CI 0.02 to 0.3, p < 0.001), with a 1-year EFS rate of 75% compared with that of 24% in patients without prior HSCT.

Compared with ELIANA trial data, the blast percentage in BM before infusion were significant lower in RW registry (median, 1.3% [0-100%]) than ELIANA cohort (93%, [2.3-98%]). Although similar efficacy was observed (the 1-year OS and EFS rates after infusion were 67%, and 44% in the ELIANA cohort), CRS grades were significantly lower in RW registry (median; grade 1), compared with the ELIANA cohort (median; grade 4) and the death within 100 days after infusion were lower in RW registry (0% vs 22%).

Conclusions: This first analysis of patients with ALL undergoing commercial tisagenlecleucel in Japan confirmed LB and prior HSCT to be associated with superior EFS and showed more favorable safety profile compared with pivotal trials. Under RW setting, tisagenlecleucel tended to be administered at LB status, which contributed to the decrease of severe CRS. Long-term follow-up data will provide further insights into the use of tisagenlecleucel in RW setting.

Disclosure: Itaru Kato, Speakers Bureau(Novartis);

Daisuke Tomizawa, Membership on an entity’s Board ofDirectors or advisory committees and Speakers Bureau (Novartis);

Takahiro Kamiya,Speakers Bureau(Novartis);

Motohiro Kato,Speakers Bureau (Novartis);

Shoji Saito, Research Funding (Toshiba Corporation);

Hidefumi Hiramatsu, Speakers Bureau (Novartis).

4 - CAR-based Cellular Therapy – Clinical

P066 MODIFIED EASIX PREDICTS DISSEMINATED INTRAVASCULAR COAGULOPATHY, CRS AND PROGNOSIS IN PATIENTS TREATED WITH ANTI-CD19 CAR-T CELLS

Eugenio Galli 1,2, Federica Sorà1,2, Ilaria Pansini1, Francesco Autore2, Elisabetta Metafuni2, Alberto Fresa1,2, Maria Assunta Limongiello2, Idanna Innocenti2, Sabrina Giammarco2, Luca Laurenti1,2, Andrea Bacigalupo1,2, Patrizia Chiusolo1,2, Stefan Hohaus1,2, Valerio De Stefano1,2, Simona Sica1,2

1Sezione di Ematologia, Dipartimento di Scienze Radiologiche ed Ematologiche, Università Cattolica del Sacro Cuore, Rome, Italy, 2Dipartimento di Diagnostica per Immagini, Radioterapia Oncologica ed Ematologia, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy

Background: Cytokine release syndrome (CRS) and consumptive coagulopathy can complicate the treatment with chimeric antigen receptors T (CAR-T) cells. The modified version of Endothelial Activation and Stress Index EASIX (mEASIX), a score derived from hematopoietic stem cell transplantation, combines platelets, CRP, and LDH and has been correlated with CRS and endothelial biomarkers.

Methods: In 38 consecutive patients with aggressive lymphoproliferative disease we measured a coagulative laboratory panel at baseline and early after infusion of anti-CD19 CAR-T. The panel was investigated also in the presence of CRS graded ≥2, or immune effector cell-associated neurotoxicity syndrome (ICANS).

ISTH score (platelets, PT, FDP, fibrinogen) for diffuse intravascular coagulopathy was applied.

Results: CRS and ICANS

All patients experienced CRS during the first 10 days after CAR-T cells reinfusion (max CRS of grade 2 or more in 79%). ICANS occurred in 31% patients (Grade 3-4 in 10% patients).

Coagulopathy and CRS/ICANS

No patient experienced a major bleeding; one patient had a symptomatic catheter-related deep vein thrombosis (DVT) with pulmonary embolism (PE).

Patients with CRS grade ≥ 2 had prolonged PT and aPTT values, a lower platelet count, and decreased antithrombin levels in comparison with those with CRS mild or absent. In moderate-to-severe CRS we have observed fibrinogen, D-dimer, FVIII, vWF antigen levels higher than in mild or absent CRS. Similar results were found for ICANS.

At baseline, immediately before CAR-T cells reinfusion, all but one patient had an ISTH score lower than 5. Conversely, if assessed while experiencing CRS grade 2 or more, 30% of the evaluable patients presented an ISTH score of 5 or more, consistent with DIC. Overall, 13/38 (34%) patients fulfilled ISTH criteria for DIC during the first two weeks after CAR-T cells infusion.

Endothelial activation and Coagulopathy

Overt DIC was strongly associated with simultaneous mEASIX scoring. Moreover, baseline mEASIX higher than 6.89 predicted DIC during the two weeks after infusion. Coagulopathy defined as per ISTH criteria occurred in 29%, 33% and 40% of patients treated with axi-cel, brexu-cel and tisa-cel, respectively (p = 0.8).

Higher mEASIX was associated with laboratory findings suggesting an active coagulative process and endothelial involvement (Higher aPTT, fibrinogen, D-dimer, factor VIII, and vWF, and lower antithrombin).

We have also found a positive correlation with sST2 (Soluble Suppression of tumorigenicity 2, secreted in response to stress or injury by a plethora of cell types, including immune-cells, fibroblasts, and endothelial cells) and suPAR (Soluble urokinase-type plasminogen activator receptor, widely associated with markers of endothelial damage and activation).

mEASIX and prognosis

PFS at day 180 after CAR-T cells infusion was 60% vs 33% for patients with a baseline mEASIX lower or higher than 6.89, respectively (p = 0.031). OS at day 180 was 90% vs 50% for patients with a baseline mEASIX lower or higher than 6.89, respectively (p = 0.024).

Conclusions: Overall, the role of endothelium is emerging as pivotal in the pathophysiology of CRS and ICANS. Further studies are needed to explore the role of endothelial activation and coagulopathy in order to improve the safety and efficacy of CAR-T cell therapies.

Disclosure: Authors disclose no competing conflict of interests.

4 - CAR-based Cellular Therapy – Clinical

P067 ADOPTION OF BENDAMUSTINE LYMPHODEPLETION AS STANDARD OF CARE IN SETTING OF DRUG SHORTAGES

Richard Maziarz 1, Andy Chen1, Levanto Schachter1, Peter Ferreira-Gandolfo1, Brandon Hayes-Lattin1, Jessica Leonard1, Staci Williamson1, Amber Diaz1

1Oregon Health & Science University, Portland, United States

Background: Cyclophosphamide and fludarabine (Cy/Flu) lymphodepleting chemotherapy (LDC) has been adopted universally prior to administration of CAR T-cell therapy. A prior comparative analysis suggested that acceptable outcomes can be achieved, substituting bendamustine (Benda) for Cy/Flu prior to administration of Tisagenlecleucel (Tisa-cel) (Ghidelli, Ann Onc, 2022). Due to an unanticipated absence of availability of Flu, Benda was adopted as a standard of care for all CAR-T administrations of commercially available product (Maziarz, JTCT, 2022) including CAR T-cell products using the CD28 activation domain.

Methods: For calendar year 2022, all subjects receiving CAR T-cell therapy for standard indications (ALL, lymphoma, myeloma) were reviewed and separated into 2 cohorts, those receiving Cy/Flu vs Benda as LDC. The retrospective analysis was limited to the current year to provide a contemporary comparative cohort. Data were collected from institutional data bases and were assessed for disease response, adverse events, and depth of lymphodepletion. Ongoing data analysis for quality assurance of safety has been performed consistent with institutional regulatory guidelines.

Results: Thirty-five subjects (19 NHL; 11 ALL; 5 MM) received CAR T-cell therapy after Cy/Flu LDC during 2022; 23 subjects (20 NHL; 3 MM) were treated with CAR T-cell after Benda LDC since Aug 2022. For those receiving Cy/Flu LDC, best observed response included 19 CR, 8 PR, 1 Stable, 4 Prog, 3 NE due to early death). Restricted assessments to the MM and NHL cohort, 19/26 had gained CR/PR status as best response. Sixteen of 35 developed CRS, with 9/16 with Gr II-IV CRS observed. Fourteen of 35 developed ICANS with 12/14 with Gr II-IV observed. Absolute lymphocyte count (ALC) nadir was assessed with a median ALC of 20 (range 0-240).

Subjects receiving Benda LDC had shorter follow-up, given the timing of LDC transition. For those receiving Benda LDC, best observed response included 5 CR, 7 PR, 2 Stable, 7 Prog, 2 NE due to not yet reach first evaluable date). Thirteen of 23 developed CRS, with 2/13 with Gr II-IV CRS observed. Five of 23 developed ICANS with 2/5 with Gr II-IV observed. Absolute lymphocyte count (ALC) nadir was assessed with a median ALC of 100 (range 0-1460). No subjects with ALL received Benda LDC.

Conclusions: Bendamustine has been shown in a comparative institutional analysis to be an acceptable agent to be utilized for LDC prior to CAR T-cell therapy. Due to an unforeseen drug shortage with limited availability, strict institutional guidelines for use were developed impacting leukemia, transplant and immune effector cell programs. Ongoing assessment of impact occurs via oversight of a newly established Oncology Stewardship Committee. Current review suggests adequate clinical outcomes are achieved with Benda LDC. CRS and ICANS rates are observed to be lower than with Cy/Flu. Continued evaluation will be required to assure long term goals are achieved.

Clinical Trial Registry: Not applicable

Disclosure: No conflicts of interest are reported re: the contents of this abstract.

Thus, nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P068 EFFICACY AND TOXICITY FOR CD7 CHIMERIC ANTIGEN RECEPTOR T-CELL THERAPY IN PATIENTS WITH RELAPSED/REFRACTORY T-CELL LYMPHOMA

Guoai Su1, Fan Yang 1, Hui Shi1, Rui Liu1, Ruiting Li1, Teng Xu1, Shaoma Feng1, Peihao Zheng1, Yuelu Guo1, Lixia Ma1, Biping Deng2, Xiaoyan Ke1, Kai Hu1

Beijing Gobroad Boren Hospital, Beijing, China, 2Cytology Laboratory, Beijing Gobroad Boren Hospital, Beijing, China

Background: The prognosis of refractory/relapsed T-cell lymphoma is extremely poor, especially for the patients who failed to allogeneic hematopoietic stem cell transplantation (alloHSCT).

Methods: From August 2017 to December 2022, 20 patients were enrolled. The median age was35 (18-72) years old. The diagnosis included T cell lymphoblastic leukemia/lymphoma (T-LBL) (n = 17), hepatosplenic T-cell lymphoma (HSTL, n = 1), monomorphic epitheliotropic intestinal T-cell lymphoma (MEITL, n = 1) and cutaneous T-cell lymphoma (CTCL, n = 1). The disease status was progressive disease (PD) in all patients who failed to multi-line therapies, including autologous HSCT (n = 5), and alloHSCT (n = 9). Six patients (6/20, 30%) had central nervous system involvement. In order to further reduce the tumor burden, 8/20 (40%) patients were treated with bridging therapy before CAR-T cell infusion. Before the trial, the expression of CD7 antigen in tumor tissue was positive confirmed by pathology. Infusion of donor-derived CD7 CAR-T cells in patients who have relapsed after alloHSCT, whereas infusion of autologous CAR-T cells in other patients. Patients received fludarabine and cyclophosphamide regimens before CAR-T cell infusion. The kinetics and function of CAR-T cells was monitored by quantitative PCR and flow cytometry. The efficacy was evaluated by PET-CT as well as bone marrow puncture after CAR-T infusion.

Results: The median CAR-T cells infused were 1×105/kg (range,0.06-34×105/kg). For CAR T cell expansion,the peak time in vivo was on median 14 (range, 11-29) days after CAR-T cell infusion. The median peak kinetics of CAR-T cells in peripheral blood of individual patients measured by flow cytometry was 23.65 (range,1.61-49)×106/L, which was no correlation with the number of CAR-T infused (P = 0.0818). Peak CAR-T amplification is also independent of whether CART cells are sufficiently donor-derived(P = 0.4692). Levels of CAR-T cells were very low after the first 1 months postinfusion. The incidence of grade 3 cytokine release syndrome (CRS) was 15% (3/20), and the incidence of grade 1-2 neurotoxicity was 10% (2/20). Although CD7-positive normal T cells were depleted, CD7-negative T cells expanded in all patients. Seventeen patients had occurred cytopenias. Nine patients (9/20,45%) had prolonged cytopenias (1 month). Viremia occurred in 12/20 (60%) patients. 2/20 (10%) patient developed post transplant lymphoproliferative disorders (PTLD) associated with EBV infection.1/9 (11%) patients after allogeneic HSCT were found to have grade IV aGVHD (intestinal).

With a median follow-up of 5.96 months (95% CI: 0.56-18.6), one-year overall survival (OS) and progression-free survival (PFS) were 65.5% and 56.6%, respectively. The overall response (ORR) was 60% (CR 55% and PR 5%). However, patients treated with donor-derived CAR-T had significantly worse PFS (33.3% versus 100%) than other patients, which may be due to the very high incidence of infections. Data from scRNA-seq of PBMCs from patients treated with donor-derived CAR-T indicated a fulfilled T lymphocytes immune function by CD7-negative T cells. The major challenge on prevent post-treatment infection may be the protection from loss of monocytes.

Conclusions: Our study showed promising efficacy of CD7 CAR-T cell therapy in r/r T-cell lymphoma. CRS is manageable. For patients who have relapsed after allogeneic transplantation, bridge second transplantation is recommended if CD7 CAR-T therapy is effective.

Clinical Trial Registry: ChiCTR2200058969 and http://www.chictr.org.cn/index.aspx

Disclosure: The authors confirm that there are no conflicts of interest.

4 - CAR-based Cellular Therapy – Clinical

P069 EARLY POST-CAR-T MARKERS OF PROGRESSION, DESPITE CONCOMITANT PET REMISSION: A MONOCENTRIC EXPERIENCE IN THE SETTING OF AGGRESSIVE B-CELL LYMPHOMA

Massimiliano Gambella 1, Anna Maria Raiola1, Chiara Ghiggi1, Livia Giannoni1, Riccardo Varaldo1, Anna Ghiso1, Silvia Luchetti1, Alberto Serio1, Monica Centanaro1, Alessandra Bo1, Angelo Gratarola1, Emanuele Angelucci1

1IRCCS Ospedale Policlinico San Martino, Genova, Italy

Background: Metabolic remission of lymphoma 1 month after CAR-T therapy (PET-1) is challenging: some patients will remain lymphoma-free; a proportion will progress over time. Clinical factors associated with the risk of early progression, as depicted concomitantly with PET-1, lack and could help the decisional process (observation vs pre-emptive strategies). We sought to identify these factors.

Methods: Twenty-five consecutive patients affected by aggressive B cell lymphoma, treated at our center from October 2020 to July 2022 with anti-CD19 CAR-T (tisagenlecleucel or axicabtagene ciloleucel) were analyzed. Patients with PET-1 metabolic remission (i.e., complete or partial remission, as per Cheson criteria) were considered; primary mediastinal lymphomas were excluded. Clinical/laboratory variables considered relevant for progression were collected 1 month after CAR-T infusion; their association with subsequent progression was evaluated through the two-tailed Fisher’s exact test. Variables with a p < 0.25 for association with progression entered a dichotomic score (two groups: 0-1 vs ≥2 points). Progression-free survival (PFS) was calculated for the two groups from PET-1 through the Kaplan-Meier estimator. Log-rank test was employed for comparisons.

Results:

Tab.1a

Variables

No Progression

within 6 months

(N = 8)

Progression

within 6 months

(N = 6)

p-value

CRP, n (%)

 > ULN

1 (12.5)

0 (-)

1.0

 < UNL

7 (87.5)

6 (100)

 

PLT, n (%)

 < 25x10^9/L

0 (-)

3 (50)

0.05

 > 25x10^9/L

8 (100)

3 (50)

 

LDH, n (%)

 > 1.5 ULN

2 (25)

4 (66.7)

0.24

 < 1.5 ULN

6 (75)

2 (33.3)

 

fibrinogen, n (%)

 < LLN

2 (25)

5 (83.3)

0.1

 > LLN

6 (75)

1 (16.7)

 

dexamethasone, n (%)

 > 40mg

0 (0)

3 (50)

0.05

 < 40mg

8 (100)

3 (50)

 

Tab.2a

Risk-Group, n (%)

 High-Risk ( ≥ 2 points)

0 (-)

4 (66.7)

0.015

 Low-Risk (0-1 points)

8 (100)

2 (33.3)

 
  1. CRP C-Reactive Protein; UNL Upper Limit of Normal; PLT platelets; LDH lactate dehydrogenase; LLN Lower Limit of Normal.

Fourteen patients fulfilled the pre-specified criteria at PET-1 evaluation, 10 CR and 4 PR; 6/14 (42%) progressed within 6 months from infusion, 4 from PET-1 PR and 2 from CR. Among variables evaluated one month post-infusion, four (i.e., fibrinogen < lower limit of normal, lactate dehydrogenase > 1.5 upper limit of normal, platelets count < 25x10^9/L, total dexamethasone dose > 40mg, administered for inflammatory complications) were considered relevant for subsequent progression (p < 0.25; table 1a) and gathered together in a dichotomic score, each variable accounting for 1 point: low-risk (0-1 points); high-risk ( ≥ 2 points). Ten patients were classified as low-risk and 4 patients as high-risk; among the high-risk patients, 2/4 had a concomitant PET-1 CR. The high-risk score resulted significantly associated with progression (p = 0.015). With a median follow-up of 17 months from PET-1 (95% CI: 14.6 – 19.4), 4 patients (100%) in the high-risk group progressed, as compared to 3/10 (30%) in the low-risk group (p = 0.003).

Conclusions: Simple clinical factors evaluated following infusion, likely representing a “hostile” environment for CAR-T cells, emerge as risk factors for progression. Their use, gathered together in a risk score, might be of particular interest in PET-1 responding patients, allowing prophylactic/pre-emptive strategies before progression, such as allogeneic transplant. We recognize that the index probably identifies patients inherently with a major risk of relapse due to a high tumor burden at CAR-T. However, it is still difficult to foresee whether and when to treat patients who achieve an early, major response.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P070 BREXUCABTAGENE AUTOLEUCEL FOR ADULT PATIENTS WITH REFRACTORY/RELAPSED ACUTE LYMPHOBLASTIC LEUKEMIA: COST-EFFECTIVENESS IN THE CLINICAL SUBSETS

Monia Marchetti1

1Azienda Ospedaliera SS Antonio e Biagio e Cesare Arrigo, Alessandria, Italy

Background: Acute Lymphoblastic Leukemia (ALL) is a rare aggressive neoplasm incurring in about 2/100,000 adults per year, but less than half of them are projected to survive 5 years after diagnosis. Expected overall survival (OS) is particularly poor (i.e. less than 12 months) in patients relapsing or not responding after/to frontline therapy (R/R) despite being offered standard of care (SOC) with targeted agents (Blinatumomab (BLIN), Inotuzumab Ozogamicin (INO)) or chemotherapy (CIT), +/- tyrosin-kinase inhibitors (TKI), possibly followed by allogeneic stem cell-transplantation (aSCT). Brexucabtagene autoleucel is a chimeric antigen receptor (CAR) T-cell therapy that allowed 56% of R/R LLA patients to achieve a durable complete response and a median OS of 25.4 months. The present study aimed at comparing the clinical and economic outcomes of R/R LLA patients treated with Brexucabtagene autoleucel versus the standard of care (SOC) in the Italian Healthcare System.

Methods: We developed a partitioned-survival model to extrapolate event-free, overall survival and healthcare costs of R/R LLA patients and intended to receive KTE-X19 or SOC. The source of safety and survival data for Brexucabtagene autoleucel was the mITT cohort of ZUMA-3 trial (16.4 month median follow-up data; 77% received CAR-T infusion), while historical trial data (e.g. INO-VATE, PACE and TOWER) were used for for SOC. Long-term OS and EFS were estimated using mixture cure modelling methods: log-logistic and log-normal parametrization of OS and EFS curves was usually employed, respectively. Moreover, consolidation aSCT was modelled in 23.48% of patients achieving a response to SOC and in 18% of the patients treated with Brexucabtagene autoleucel, based on trial data. Patients whose disease had not progressed after 2 years were assumed to experience long-term remission.

Patients’ quality of life (i.e. utilities) were driven from ZUMA-3 trial and the published literature. Resource consumption was based on trial data and published literature. Unit costs were estimated from an Italian National Healthcare System perspective and based on national charges, ex-factory drug costs and published economic analyses. Costs and health outcomes were discounted at 3% per year. Sensitivity analyses were performed to test model robustness.

Results: Median estimated survival was 9.68 life years for Brexucabtagene autoleucel, while it was 2.78-4.98 LYs for SOC. Discounted and quality-adjusted life expectancy was 4.96-5.51 quality-adjusted years (QALY) for Brexucabtagene autoleucel versus 1.34-3.04 for SOC. Cumulative discounted costs in the 50-year time horizon were €387,904-405,023 for patients intended-to-receive Brexucabtagene autoleucel, versus €111,976-271,651 for SOC, which corresponds to a cost of €46,415-76,384 per QALY gained. At a willingness-to-pay threshold of 100,000/QALY probabilistic sensitivity analysis demonstrated that Brexucabtagene autoleucel has a probability higher than 85% of being cost-effective versus SOC. The most influential model parameters were patients’ quality of life, Brexucabtagene autoleucel acquisition cost, and the proportion of patients receiving HSCT. The results were also sensitive to the time horizon of the analysis.

Conclusions: Brexucabtagene autoleucel is a cost-effective alternative to SOC for adult patients with R/R LLA, both Ph+ and Ph-.

Disclosure: MM (Gilead consultancy fees).

4 - CAR-based Cellular Therapy – Clinical

P071 OPTIMIZING MULTITARGET APPROACH IN THE PEDIATRIC PATIENT: NEW PERSPECTIVES OF DUAL CD19/CD22 CAR-T

Víctor Galán-Gómez 1, Berta González Martínez1, Isabel Mirones Aguilar2, Carlos Echecopar Parente1, Adela Escudero3, Cristina Ferreras4, Alfonso Navarro-Zapata4, Carmen Mestre-Durán4, Marta Cobo2, Alicia Pernas2, Susana Vallejo2, Raquel de Paz Arias5, Jordi Minguillón4, Isabel Martínez-Romera1, Pilar Guerra-García1, Sonsoles San Román1, Antonio Pérez-Martínez1

Hospital La Paz, Madrid, Spain, 2Advanced-Production Therapy Unit; University Hospital La Paz, Madrid, Spain, University Hospital La Paz, Madrid, Spain, 4La Paz Research Institute; University Hospital La Paz, Madrid, Spain, La Paz University Hospital, Madrid, Spain

Background: CD19 CAR-T tisagenlecleucel has been proved to be effective in B-cell precursor acute lymphoblastic leukemia (B-ALL). However, 30-50% of patients will suffer a relapse after CD19 CAR-T infusion, sometimes CD19 negative. New CAR-T targets have been supposed to be considered, such as CD22 or multitarget approach as dual CD19/CD22.

Methods: To evaluate the safety and efficacy of the use of an in-house manufactured dual CD19/22 CAR-T through a compassionate use program in patients diagnosed with a relapsed or refractory ALL-B.

Five batches of the drug under investigation for advanced therapy CD19/22 CAR-T cells have been manufactured under Good Manufacturing Practices (GMP), authorized by the Spanish Agency for Medicines and Health Products (AEMPS) under a compassionate use program. The starting material in all cases was immobilized apheresis. After immunoselection of the populations of CD3+ T lymphocytes (CD4/CD8) and activation with CD3/CD28, they were transduced with a CD19/CD22 lentiviral vector provided by Miltenyi Biotec. Cell expansion was performed in the CliniMACS Prodigy® device with TexMACs culture medium supplemented with IL-7 and IL-15. During manufacturing, process and quality controls were carried out and the finished product was fresh-available or cryopreserved in vapor phase liquid nitrogen until it was administered to the patient.

Results: A total of 6 dual CD19/22 CAR-T infusions have been performed in a total of 5 patients. Three patients (60%) were diagnosed with a relapsed B-ALL after tisagenlecleucel infusion. Of them, 2 patients (67%) had also previously received hematopoietic stem cell transplantation (HSCT). These two patients did not present CD19 expression at relapse. One of them also expressed low CD22. One patient (20%) was diagnosed with a refractory B-ALL, and the last patient (20%) with a relapsed B-ALL. Both were no eligible for tisagenlecleucel treatment.

Table 20

The age range of the patients was 6-17 years. The time range between the start of manufacturing and the availability of the CAR-T for infusion was 9-11 days. The pre-infusion minimal residual disease range was 0.01-87%. Bridging therapy was individualised in each patient, and lymphodepletion regimen consisted on fludarabine and cyclophosphamide. The infused dose range was 7.5x1e5-3x1e6 cells/kg.

Regarding toxicity, a total of 4 patients (80%) presented cytokine release syndrome (CRS): 2 patients (50%) grade 3 (ASBMT guidelines). Two patients (20%) presented grade 3 immune effector cell-associated neurotoxicity syndrome (ICANS).

On day +28, 3 (60%) of the patients were in complete remission (CR). Two (40%) suffered disease progression and received palliative treatment.

Conclusions: The manufacturing of dual CD19/22 CAR-T therapy is feasible and reproducible under GMP conditions. The infusion of the product seems to be well tolerated without severe toxicities in the majority of patients, with an adequate safety profile. Data from our experience suggest the efficacy of its use. A larger number of patients is needed, so the use of the CAR-T in the context of a clinical trial should be necessary.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P072 RETROSPECTIVE MULTICENTRIC COMPARISON OF OUT-OF-SPECIFICATION AND STANDARD-OF-CARE TISAGENLECLEUCEL INFUSION FOR DIFFUSE LARGE B-CELL LYMPHOMA

Daniele Mannina1, Chiara De Philippis 1, Cristina Zucchinetti1, Flavio Pistolese1, Mauro Krampera2, Francesca Bonifazi3, Annalisa Chiappella4, Alice Di Rocco5, Enrico Orciuolo6, Maria Chiara Tisi7, Laura Giordano1, Armando Santoro1,8, Stefania Bramanti1

1IRCCS Humanitas Research Hospital- Humanitas Cancer Center, Rozzano (Milano), Italy, 2University of Verona, Verona, Italy, 3Istituto di Ematologia L. e A. Seràgnoli, Bologna, Italy, 4Fondazione IRCCS Istituto Nazionale dei Tumori, Milano, Italy, 5Sapienza University of Rome, Rome, Italy, 6University of Pisa, Pisa, Italy, 7Ospedale San Bortolo, Vicenza, Italy, 8Humanitas University, Pieve Emanuele (Milano), Italy

Background: CAR-T treatment requires a complex manufacturing process that includes multiple functional tests, to guarantee strict quality criteria and ensure the safety and the efficacy of the product1,2. A final product that complies with all the quality criteria is acknowledged as “standard-of-care” (SOC), while the ones presenting at least one feature not fitting the benchmark are considered “out-of-specification” (OOS)2. CAR-T infusion is often urgent because of disease aggressiveness, and infusion of OOS product could be the only chance to cure some patients3,4. Recently published literature suggests that OOS occurrence is not negligible and it may occur more frequently in highly pre-treated patients ( ≥ 4 prior therapies or bendamustine exposure).5 It is still unclear whether the outcome of OOS tisa-cel is significantly different from SOC.

Methods: This retrospective multicentric study aims at evaluating the outcomes of the administration of OOS tisa-cel in comparison with SOC product, both in term of toxicity and efficacy. We enrolled 40 DLBCL patients from 6 italian centers, 11 treated with OOS tisa-cel, 29 with SOC tisa-cel in a real life-setting, according to Italian Medicines Agency criteria. Patients treated within an experimental trial were excluded.

Results: Table 1 summarizes patient characteristics, homogeneously distributed across the cohorts. Reason for defining OOS was viability below range in 1 case, inadequate transgene copies in 2 case, insufficient CAR expression by cytofluorimetry in 2 cases, high IFN-gamma secretion in 4 cases, T-cell expansion below range in 1 case, inadequate reporting of microbiological contamination (mycoplasma) in 1 case. After a median follow-up of 11 months, one-year PFS was 54.5% and 39.5% for OOS and SOC patients, respectively (p = 0.99). One-year OS was 55.5% and 64.5% for OOS and SOC patients, respectively (p = 0.31) Complete remission was achieved in 6/11 OOS patients (54.5%), and in 18/29 SOC patients (62.1%), without any statistically significant difference (p = 0.664). CRS, ICANS and infections occurred in similar proportions across the groups, with no statistically significant difference. Prolonged cytopenia, defined as persistence of at least one grade 3-4 cytopenia at 45 days after infusion, was absent in the OOS group, while it was found in 11/29 (39.3%) patients in the SOC group (p = 0.017).

Table 1. Patients characteristics

  

OOS (n = 11)

SOC (N = 29)

Total (n = 40)

p. value

Sex

Male

6 (54.5%)

13 (44.8%)

19

0.727

Female

5 (45.5%)

16 (55.2%)

21

Age

Median (range)

56 (25-69)

59 (28-72)

58 (25-72)

 

Diagnosis

de novo DLBCL

10 (90.9%)

27 (93.1%)

37

1.000

transformed DLBCL (t-FL)

1 (9.1%)

2 (6.9%)

3

MYC/BCL2/BCL6 status

Double-hit

1 (9%)

3 (10.3%)

4

0.328

Double/triple expressor

5 (45.5%)

8 (27.6%)

13

Negative

5 (45.5%)

18 (62.1%)

23

Stage at diagnosis

I-II-III

2 (18.2%)

14 (48.3%)

16

0.148

IV

9 (81.8%)

15 (51.7%)

24

Prior bendamustin

Yes

2 (18.2%)

2 (6.9%)

4

0.300

No

9 (81.8%)

27 (93.1%)

36

Prior ASCT

Yes

3 (27.3%)

5 (17.2%)

8

0.660

No

8 (72.7%

24 (82.8%)

32

Previous lines n.

2

6 (54.5%)

22 (75.9%)

28

0.264

3

2 (18.2%)

4 (13.8%)

6

>3

3 (27.3%)

3 (10.3%)

6

Stage at apheresis

I-II-III

5 (45.4%)

12 (41.4%)

17

1.000

IV

6 (54.6%)

17 (58.6%)

23

Conclusions: Our data confirm that OOS infusion is a feasible option, without any increase in terms of inflammatory and neurological toxicity. In consideration of the rapid progressivity of r/r DLBCL, a second manufacturing attempt with prolongation of apheresis-to-infusion time in case of OOS is not justifiable. Further data collection is desirable in order to confirm the non-inferiority OOS CAR-T in terms of efficacy, due to the small number of patients. Moreover, quality criteria of tisa-cel are still subject to continuous update, and larger cohorts of patients are required to establish which type of alteration from SOC may actually impact the outcome, possibly leading to further revision of OOS criteria.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P073 MANAGEMENT OF CRS AND ICANS WITH SILTUXIMAB AND ANAKINRA: A SYSTEMATIC REVIEW OF EVIDENCE TO INFORM SIES-GITMO-SIDEM RECOMMENDATIONS

Monia Marchetti 1, Annalisa Chiappella2, Maria Guarino3, Giorgia Gobbi2, Massimo Martino4, Andres JM Ferreri5, Luca Fumagalli6, Riccardo Saccardi7, Mario Montanari8, Maurizio Marchetti9, Pierluigi Zinzani10

1Hematology and Transplant Unit, Azienda Ospedaliera SS Antonio e Biagio e Cesare Arrigo, Alessandria, Italy, 2Division of Hematology & Stem Cell Transplant, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy, 3Neuro AOU, IRCCS Policlinico S’Orsola, Bologna, Italy, 4Cell Transplantation and Cellular Therapies Unit, Grande Ospedale Metropolitano “Bianchi-Melacrino-Morelli”, Reggio Calabria, Italy, 5Lymphoma Unit, Fondazione Centro San Raffaele, Milan, Italy, 6Intensive Care Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy, 7Azienda Ospedaliera Careggi, Florence, Italy, 8OspedaliRiuniti Marche, Ancona, Italy, 9Hematology, Azienda Ospedaliera Policlinico Umberto I, Sapienza University, Rome, Italy, 10Lymphoma Unit, Istituto Seragnoli, Policlinico S’Orsola, University of Bologna, Bologna, Italy

Background: Patients with aggressive B-cell lymphomas may achieve prolonged survival after treatment with autologous chimeric antigen receptor T-cells (CAR-T). Nevertheless, after CAR-T infusion a relevant portion of the patients develop cytokine-related syndrome (CRS) and cytokine-mediated neurotoxicity (ICANS). Monoclonal agents anainst Interleukin-6 (IL6) receptor (namely tocilizumab) are standard of care for treating CRS, however, ICANS might be increased due to blood-brain bareeer blocking tocilizumab entrance. For the above reasons Siltuximab, a direct anti-IL6 monoclonal agent, and Anakinra, a monoclonal neutralizing antibody against interleukin-1 (IL1), have been tested in patients developing cytokine-related toxicities after CAR-T infusion. The present study is aimed at revising the available literature regarding Siltuximab and Anakinra use for the prevention or treatment of CRS and/or ICANS in order to support the development of evidence-based recommendations for the CAR-T guideline program started by three national scientific societies, namely SIE, GITMO and SIDEM.

Methods: On 1 December 2022 EMBASE, PubMed and Cochrane Library databases were scanned with standard queries in order to identify overall recordspublished since Jan 2019 and reporting the clinical outcomes of adult patients trated with CAR-T cell therapy for aggressive B-cell lymphomas. Out of 830identified records, those reporting Siltuximab or Anakinra use after CAR-T therapy were identified. The search also included studies presented uniquely in abstract form

Results: Fifteen studies reported the use of Anakinra in 132 patients with severe or steroid-refractory ICANS after CAR-T cell therapy (52% axi-cel), while 57 patients were enrolled into 3 prospective studies aimed at ICANS prevention. Grade 5 ICANS was reported in 4 out of 84 (4.7%) patients treated with Anakinra and amelioration of the neurological status was reported in 46-100% of the patients. Higher response rates were acieved with Anakinra doses higher than 200 mg per day and early mortality declined from 84% to 7% in patients receiving high vs low doses in one study. The rate of ICANS in patients receiving prophylaxis with Anakinra was 14% and only transient grade 3 ICANS developed.

Siltuximab use was detailed in overall 50 patients reported by 15 studies, but also in 4.4% of the 291 patients reported by the CART Consortium and CARICU studies. Most of the patients were treated with Siltuximab for tocilizumab-refractory CRS or concurrent CRS and ICANS. Unfortunately, specific outcomes of patients receiving Siltuximab were rarely reported. Overall no specific toxicity was alerted.

Conclusions: A growing body of evidence has been reporting the outcomes of Siltuximab and Anakinra in lymphoma patients who developed CRS or ICANS after CAR-T. Preliminary studies also tested an early adoption of Anakinra for the prevention of severe ICANS. The above systematic review confirms the need of randomized clinical trials to assess the effectiveness of Siltuximab and Anakinra in different patients subgroups, including those at high-risk of developing ICANS (see Rubin et al 2020).

Disclosure: MM (Gilead consultancy fees).

4 - CAR-based Cellular Therapy – Clinical

P074 EFFICACY OF EARLY LYMPHOPHERESIS TO IMPROVE CAR-T FITNESS AND PATIENT’S OUTCOMES IN HIGH RISK LARGE B CELL LYMPHOMA- A SINGLE CENTER, OPEN-LABEL PROSPECTIVE STUDY

Ron Ram 1, Odelia Amit1, Chava Perry1, David Hagin1, Tal Freund1, Yair Herishanu1, Gil Fridberg1, Nadav Sarid2, Yakir Moshe1, Uri Abadi3, Noa Lavi4, Shirley Shapira3, Ofrat Beyar-Katz4, Luiza Akria5, Yafit Segman6, Arie Apel7, Shimrit Ringelstein-Harlev4, Katrin Herzog Tzarfati7, Tamir Shragai1, Ronit Gold1, Rinat Eshel1, Irit Avivi1

1Tel Aviv Sourasky Medical Center, Tel Aviv, Israel, 2Wolfson Medical Center, Holon, Israel, 3Meir Medical Center, Kefar Saba, Israel, 4Rambam Medical Center, Heifa, Israel, 5Galilee Medical Center, Nahariya, Israel, 6Assuta Ashdod Hospital, Ashdod, Israel, 7Shamir Medical Center, Rishon Leziyon, Israel

Background: Approximately 60%-70% of patients with relapse/refractory LBCL experience post anti CD-19 CAR-T disease progression that is partially related to the lack of CAR-T cell persistence and T cell exhaustion, attributed to T cell fitness at lymphopheresis. We hypothesized that early lymphopheresis (EL) in patients with high risk LBCL may improve CAR-T fitness and outcomes.

Methods: Patients with newly diagnosed high risk LBCL (defined as IPI ≥ 4, MYC + LBCL, a positive interim PET scan, evidence of progression of disease after 2 courses of first line treatment, or Richter’s transformation) who signed informed consent were eligible to this study. Collection of cells was performed after ≤3 courses of first line therapy. Cells were sent for production of tisagenlecleucel, once patients were refractory to 2 lines of therapy and eligible to treatment. Prespecified interim analysis of collection and CAR-T product’s characteristics was planned after the first 30 patients. We compared collection materials (T cell subpopulations and expression of exhaustion markers – HLA-DR and PD-1) and CAR-T products’ characteristics between patients who underwent EL and patients who underwent standard lymphopheresis (SL) during same years.

Results: Between March, 2020 and November, 2022, 34 patients were recruited (IPI ≥ 4 (n = 6), MYC + LBCL (n = 25), positive interim PET scan/progression within 2 months (n = 1), and Richter’s transformation (n = 2)). Median age was 66 (range, 33-80) years and in 31 (86%) patients collection was performed after 2 cycles of chemotherapy. Collection product was analyzed in all 34 patients. Median CD4/CD8 ratio and median % of naïve CD4 cells were both statistically significant higher in patients who underwent EL compared to those in the SL group (1.53 vs..59, p = .018, and 16.6% vs. 4.89%, p = .042, respectively), Table. Median expression of HLA-DR was statistically significant lower in the EL group in both the CD4 and the CD8 compartments (16.95% vs. 51.3%, p < .001, and 17.75% vs. 44.25%, p < .001, respectively) while median expression of anti PD-1 was statistically significant lower only in the CD4 compartment (32.46% vs. 52.8%, p = .021), Table. At a median follow-up of 22 (range, 3-35) months, 13 patients (38%) had a first disease progression with a median PFS of 30 (95% CI-22-38) months. Eight patients after second relapse/progression were given CAR-T. Characteristics of the CAR-T product of patients in the EL group showed statistically significant better transduction efficiency, higher % of cell viability, and increased CAR-positive viable cells, compared to patients in the SL group (p = .009, p = .042, and p = .042, respectively), Table. Incidence of grade 1-3 CRS was 100% (grade 3, 25%) and immune effector cell-associated neurotoxicity was documented in 25% of the patients (all grade 1). At 1 month post CAR-T infusion, overall response rate was 63% (CR = 4; PR = 1), and 2 patients had progressive disease. In 1 patient assessment of disease is pending.

Conclusions: Early lymphocyte collection in patients with high-risk LBCL is associated with high percentage of naiive T cells and low expression of exhaustion markers. This translates into superior specifications of the CAR-T product. This trial is still recruiting patients to test the toxicity profile and effectiveness of this approach.

Domain

Early Apheresis

Standard Apheresis

p Value

Median

IQR

Median

IQR

Apheresis material

 CD4/CD8

1.53

.93-2.3

.59

.34—1.2

.018

 Naïve CD4, %

16.6

11.7-31.2

4.89

1.7-11.6

.042

 HLA-DR CD4, %

16.95

9.7-22.1

51.3

36.5-63.3

<.001

 PD1 CD4, %

32.46

22.8-39.6

52.8

32.6-68.5

.021

 Naïve CD8, %

11.42

3.4-35.0

4.03

0.6-5.0

.052

 HLA-DR CD8, %

17.75

7.2-26.4

44.25

32.9-71.2

<.001

 PD1 CD8, %

28.40

20.9-33.6

31.3

29.3-35.5

.45

CAR-T product

 % of viable T cells

99.9

99.7-99.9

99.5

99.4-99.7

.33

 Transduction efficiency

.7

.35-.99

.25

.15-.39

.009

 % cell viability

93.3

91.2-96.7

86.5

82.4-88.8

.042

 Total cell count (106/ml)

97.1

84.9-120

90.2

43.8-97.9

1

 Number of viable cells (x106)

1700

1300-3300

1800

1050-2400

1

 CAR-pos viable cells (x106)

450

34-495

280

170-380

.042

 % CAR expression by flow

20.2

15.1-31.5

15.7

10.3-20.9

.16

 Release of IFNg in response to CD19 (fg/transduced cells)

352

180-477

186

122-349

.16

Disclosure: Ron Ram. - Honoraria and speakers bureau: Novartis, Gilead; Ofrat Beyar-Katz - Honoraria: Novartis.

4 - CAR-based Cellular Therapy – Clinical

P075 A REAL-WORLD ANALYSIS OF CAR T-CELL CANCELLATIONS FROM A CANADIAN CELL THERAPY REFERRAL CENTRE

Eshrak Al-Shaibani 1, Rhida Bautista1, Eshetu Atenafu2, Andrew Winter1, Anca Prica1, Michael Crump1, John Kuruvilla1, Vishal Kukreti1, Robert Kridel1, Arjun Law1, Ivan Pasic1, Sam Saibil1, Wilson Lam1, Sita Bhella1, Christine Chen1

1Division of Medical Oncology and Hematology, Princess Margaret Cancer Centre, Toronto, Canada, Princess Margaret Cancer Centre, Toronto, Canada

Background: CAR T-cell therapy is a potentially curative treatment for aggressive large cell lymphoma. However, CAR T success in the real world can be constrained by restricted resources, systemic delays, manufacturing inefficiencies, and with few CAR T centres in Canada, the need to provide services across regional borders (out of province; OOP). These realities could contribute to CAR T delays and cancellations. We reviewed patient data from time of referral, initial assessment, cell collection and infusion, identifying causes for patient attrition.

Methods: We performed our first commercial CAR T in June 2020. At that time, there were only 3 CAR T centres in Canada using two CD19 CAR T products axicabtagene ciloleucel (axi-cel) and tisagenlecleucel (tisa-cel). We analyzed 101 patients referred for CAR T to our centre from June 2020 to December 2021; 28 (28%) from OOP. Chi-square/Fisher test was used for categorical covariates and Kruskal-Wallis test for continuous variables comparing between patients receiving and not receiving CAR T infusion.

Results: Of the 101 patients referred for CAR T, median age was 61 years (22-81), lymphoma histologies included: DLBCL (51%), high grade lymphoma (21%), transformed lymphoma (18%), PMBCL (5%), other/missing (6%). 24% patients had double/triple-hit lymphoma, 54% refractory disease.

Of the 101 patients, 33 patients (33%) did not reach CAR T-cell infusion. 4/33 patients (12%) were declined at triage even before initial assessment due to: poor ECOG (1), CNS disease (1), unconfirmed relapse histology (1), alternative therapy favored (1). Of the remaining 29 patients who underwent initial assessment, 17 patients proceeded to T-cell collection (59%), 12 (41%) did not. Reasons for not proceeding at this point included: patient declined (4), disease progression (5), including 2 active CNS disease, severe comorbidity (1), and alternative therapy chosen (2).

Of the 17 patients who proceeded to T-cell collection, reasons for not proceeding to infusion included: failed manufacturing (n = 8; 7 tisa-cel, 1 axi-cel), disease progression (n = 8, 4 with active CNS disease), 1 declined. The 8 patients unable to reach cell infusion due to disease progression had high risk features at relapse (all stage IV, IPI score>2, high LDH), 50% double-hit disease, 75% refractory, 50% CNS disease. To identify a-priori predictors for cancellation amongst collected patients, we compared patient demographics, disease factors, and timeline metrics in patients who proceeded to CAR-T infusion to those who did not (Table 1). Aggressive disease features such as CNS involvement (active or previous) and double/triple-hit status predicted for failure to proceed to infusion. Lower absolute lymphocyte counts and elevated inflammatory markers at apheresis were increased in cancelled patients. Timeline metrics and proportion of OOP referrals were similar between the two groups.

Table 1: Comparison between collected patients who proceeded and did not proceed to CAR T-cell infusion

 

CAR-T infusion N = 68

No CAR-T infusion N = 17

P value

Demographics

Median age (range) at time of CAR T referral

60.6 (22-81)

64 (50.8-73.9)

0.22

Gender, n (%)

  

1

Female

26 (38.2%)

7 (41.2%)

 

Male

42 (61.8%)

10 (58.8%)

 

Referring center, n (%)

  

0.52

Within province

54 (79.4%)

12 (70.6%)

 

Out of province (OOP)

14 (20.6%)

5 (29.4%)

 

Diagnosis, n (%)

  

0.096

DLBCL

38 (55.9%)

8 (47.1%)

 

High grade lymphoma

11 (16.2%)

8 (47.1%)

 

Transformed follicular lymphoma

12 (17.6%)

1 (5.9%)

 

Primary mediastinal B cell lymphoma (PMBCL)

5 (7.4%)

0

 

Others

2 (2.9%)

0

 

Factors at time of relapse

IPI score at relapse, n (%)

  

0.17

0-2

29 (42.7%)

4 (23.5%)

 

>2

39 (57.4%)

13 (76.5%)

 

CNS IPI at relapse, n (%)

  

0.28

Low

11 (16.4%)

2 (12.5%)

 

Intermediate

44 (65.7%)

8 (50.0%)

 

High

12 (17.9%)

7 (37.5%)

 

Disease stage at relapse, n (%)

  

0.72

I-II

13 (19.1%)

2 (11.8%)

 

III-IV

55 (80.9%)

15 (88.2%)

 

High serum LDH at relapse, n (%)

55 (80.9%)

17 (100%)

0.06

Bulky disease at relapse, n (%)

35 (51.5%)

13 (76.5%)

0.099

Extra-nodal disease at relapse, n (%)

41 (60.3%)

9 (52.9%)

0.58

Prior CNS disease, n (%)

0

2 (11.8%)

0.04

Factors at time of CAR T referral

Active CNS disease, n (%)

0

4 (23.5%)

0.001

Prior lines of therapy, median (range)

3 (2-5)

3 (1-4)

0.28

Disease status at referral for CAR-T, n (%)

  

0.16

Relapsed

33 (48.5%)

5 (29.4%)

 

Refractory

35 (51.5%)

12 (70.6%)

 

Double or triple hit lymphoma, n (%)

11 (16.2%)

8 (47.1%)

0.006

Karnofsky Performance Status (KPS), n (%)

  

0.11

≥80

53 (77.9%)

10 (58.8%)

 

<80

15 (22.1%)

7 (41.2%)

 

ECOG, n (%)

  

0.15

0-1

55 (80.9%)

11 (64.7%)

 

≥2

13 (19.1%)

6 (35.3%)

 

Hematopoietic cell transplant comorbidity index (HCT-CI), n (%)

  

0.21

0-1

48 (70.6%)

12 (70.6%)

 

3-Feb

16 (23.5%)

2 (11.8%)

 

>3

4 (5.9%)

3 (17.7%)

 

Factors at time of apheresis cell collection

Elevated ferritin level, n (%)

63 (92.7%)

11 (64.7%)

0.0003

Elevated serum C-reactive protein, n (%)

34 (50%)

8 (47%)

0.016

Median peripheral blood absolute lymphocyte count, 109/L (range)

0.9 (0.1-3.4)

0.5 (0.1-1.6)

0.028

Median peripheral blood CD3 counts/UL, median (range)

685.5 (16-3662)

308 (22- 4916)

0.057

Bridging therapy (after apheresis), n (%)

53 (77.9%)

12 (75%)

0.75

Timeline metrics

Median days from salvage treatment to apheresis (range)

19.5 (5-45)

19.5 (6-30)

0.77

Median days from initial visit to apheresis (range)

12 (2-35)

13 (1-33)

0.88

Conclusions: We report that one-third of referrals do not receive CAR T-cell infusion. Half of these cancel before cell collection, many due to aggressive disease progression and active CNS disease. CNS disease was also a primary reason for cancellation after collection, emphasizing the unmet need for improved CNS disease control through the CAR T process. Cancellations did not appear due to delays in timeline metrics for OOP referrals.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P076 EASIX AS A PREDICTOR INDEX FOR NEUROLOGICAL COMPLICATIONS AFTER ANTI-CD19 CAR-T CELL THERAPY IN AGGRESSIVE B-CELL LYMPHOMAS: A STUDY FROM THE GETH-TC GROUP

Marta Peña Domingo 1,2, Alberto Mussetti1,2, Blanca Rius Sansalvador2, Lucía Lopez-Corral3, Mario Sanchez-Salinas4, Pere Barba4, Ana Benzaquén5, Rafael Hernani5, Manuel Guerreiro6, Maria Perera7, Anna Torrent8, Antonio Perez-Martinez9, Anna Sureda1,2,10

1Hospital Duran i Reynals-Institut Català d’Oncologia-Hospitalet, Hospitalet de Llobregat, Spain, 2Institut d’Investigació biomèdica de Bellvitge (IDIBELL), Hospitalet de Llobregat, Spain, 3Hospital Clínico Universitario de Salamanca, Salamanca, Spain, 4University Hospital Vall d’Hebron, Barcelona, Spain, 5Hospital Clínico Universitario de Valencia, INCLIVA Research Institute, Valencia, Spain, 6Hospital La Fe, Valencia, Spain, 7Hospital Universitario Doctor Negrin, Las Palmas de Gran Canaria, Spain, 8Hospital Germans Trias i Pujol Institut Catala d’Oncologia-Badalona, Badalona, Spain, 9Hospital Universitario La Paz, Madrid, Spain, 10University of Barcelona, Barcelona, Spain

Background: Cytokine release syndrome (CRS) and immune-effector cells associated neurotoxicity (ICANS) due to chimeric antigen receptor T (CAR-T) cell therapy are associated to endothelial damage. This study evaluated the Endothelial Activation and Stress Index (EASIX) as a predictor for CRS and ICANS in patients treated with commercial CD19-directed CAR-T cells.

Methods: We retrospectively analyzed clinical data of 126 patients affected by aggressive B-cell lymphomas who received commercial-CD19 CAR-Ts (tisacel, axicel) from 7 centers in Spain. Inclusion period started from February 2019 to August 2022. Data were collected into the Grupo Español de Trasplante y Terapia Celular (GETH-TC) registry. CRS and ICANS were graded according to the ASTCT grading system. All patients received lymphodepleting chemotherapy with fludarabine and cyclophosphamide in doses determined according to the manufacturer. EASIX was calculated before the start of lymphodepletion (EASIX-PRE) and on the day of CAR-T administration (EASIX-d0). Log2 transformation was applied to reduce skew. Predictors of CRS and ICANS were identified using logistic regression analysis. Variables found to be significant (p-value < 0.05) in the univariate analysis and/or those considered clinically relevant for the study were included in the logistic multivariate analysis. Other outcome variables were overall survival (OS) and progression-free survival (PFS), calculated using the Kaplan-Meier estimator method, and non-relapse mortality (NRM) and relapse incidence/progression of disease (RI/POD) calculated as cumulative incidences.

Results: A total of 101 (80.2%) patients developed CRS of any grade. Of those, grade 3-4 CRS occurred in 6 (5.9%). A total of 36 (28.6%) patients developed ICANS, which happened to be grade 3-4 in 16 (44.4%). For CRS, ECOG score >1 was associated with a higher incidence of CRS grade 2-4 (Odds ratio [OR] 3.16, 95% confidence interval [CI] 1.08-9.27, p = 0.036) and day 0 C-reactive protein was associated with a higher incidence of CRS grade 3-4 (OR 1.09, 95% CI 1.02-1.17, p = 0.015). Neither Log2-EASIX-PRE nor log2-EASIX-d0 were found to be associated with grade 2-4 or 3-4 CRS. The use of axicel (OR 8.59, 95% CI 2.45-30.18, p = 0.001) and Log2-EASIX-d0 (OR = 1.66, 95% CI 1.23-2.25, p = 0.001) were associated to a higher incidence of ICANS grade 2-4. The same two factors maintained their negative prognostic role also for ICANS grade 3-4. Six patients died before day +30 (refractory disease, n = 4; CRS, n = 1; ICANS, n = 1). Median follow-up among survivors was 728 (range 84-1164) days. At 2-years, OS, PFS, NRM and RI/POD were 45% (95% CI 35-54%), 36% (95% CI, 28-45%), 8% (95% CI, 3-13%) and 55% (95% CI, 36-64%), respectively.

Conclusions: In our study, no association was found between Log2-EASIX-PRE or Log2-EASIX-d0 and CRS. However, Log2-EASIX-d0 was a strong prognostic factor for ICANS underlying the endothelial involvement in such complication. This index might help guiding preemptive strategies for those individuals at higher risk for ICANS.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P077 OUTCOMES OF CAR-T CELL THERAPY FOR LARGE B CELL LYMPHOMA IN PATIENTS OF 70 YEARS AND OLDER: MULTICENTRIC REAL WORLD EXPERIENCE

Rebeca Bailen 1,2, Mi Kwon1,2, Gloria Iacoboni3,4, Juan Luis Reguera5, Lucia Lopez-Corral6, Rafael Hernani7, Valentin Ortiz-Maldonado8, Manuel Guerreiro9, Ana Carolina Caballero10, Maria Luisa Guerra Dominguez11, Jose Maria Sanchez Pina12, Alberto Mmussetti13, Juan Manuel Sancho14, Mariana Bastos-Oreiro1,2, Eva Catala1, Javier Delgado5, Hugo Luzardo Henriquez11, Jaime Sanz9, Maria Calbacho Robles12, Cecilia Carpio3, Jose Maria Ribera14, Anna Sureda13, Javier Briones10, Juan Carlos Hernández-Boluda7, Nuria Martinez-Cibrian8, Jose Luis Diez-Martin1,15,2, Alejandro Martin6, Pere Barba4,3

1Hospital General Universitario Gregorio Maranon, Madrid, Spain, 2Instituto de Investigacion Sanitaria Gregorio Maranon, Madrid, Spain, 3Hospital Vall d’Hebron, Barcelona, Spain, 4Universidad Autonoma de Barcelona, Barcelona, Spain, 5Hospital Universitario Virgen del Rocio, Sevilla, Spain, 6Hospital Clinico Universitario de Salamanca, Salamanca, Spain, 7Hospital Clínico Universitario de Valencia, Valencia, Spain, 8Hospital Clinic de Barcelona, Barcelona, Spain, 9Hospital Universitario La Fe, Valencia, Spain, 10Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 11Hospital Universitario Doctor Negrin, Gran Canaria, Spain, 12Hospital Universitario Doce de Octubre, Madrid, Spain, 13Hospital Duran i Reynals, Instituto Catalán de Oncología, Barcelona, Spain, 14Hospital Universitari Germans Trias i Pujol, Instituto Catalán de Oncología, Barcelona, Spain, 15Universidad Complutense de Madrid, Madrid, Spain

Background: Median age at diagnosis of diffuse large B cell lymphoma (DLBCL) patients is 66 years; 40% of patients are diagnosed at an age greater than 70 years. CAR-T cell therapy is approved for the treatment of relapsed/refractory DLBCL patients; however, outcomes in patients older than 70 years are poorly reported. Our aim was to report outcomes of CAR-T cell therapy in this population as compared to those obtained in younger patient in the real world setting.

Methods: A subgroup analysis of our prior real life experience report (Kwon, Iacoboni et al. Haematologica 2022) was performed. Data from consecutive patients treated in Spain with commercial CAR-T products were retrospectively collected on behalf of GETH (Spanish Group of Stem Transplantation and Cell Therapy)-GELTAMO (Spanish Group of Lymphoma and Autologous Stem Cell Transplantation). Patients included were infused between November-2018 to August-2021. Last update of the cohort was performed in December-2021. Cytokine release syndrome (CRS) and Immune effector cell-associated neurotoxicity syndrome (ICANS) were graded with the ASTCT consensus criteria. Response was assessed according to the Lugano criteria.

Results: A total of 307 patients underwent apheresis for CAR-T cell therapy as 3rd of subsequent line. Fifty-four (18%) were 70 years or older. There were no differences between groups regarding product selection, proportion of patients infused, production failure and patients receiving bridging therapy. There were more patients with HCT-CI score ≥3 and a trend of higher proportion of patients with ECOG > 2 at apheresis in the older group (31% vs. 19%, p = 0.047; 9% vs. 5%, p = 0.054); the remaining baseline characteristics including gender, histology, disease stage, status, R-IPI and bulky disease, primary refractory disease, prior lines and prior transplant did not differ.

Among the infused population (n = 261), median time from apheresis to infusion was 49 days for patients ≥70 years (n = 45) and 47 for younger patients (n = 216) (p = 0.824). Regarding toxicity, the proportion of patients developing CRS and ICANS did not differ between groups (88% vs. 80%, p = 0.132 and 31% vs. 29%, p = 0.843). Similarly, the incidence of grade 3-4 CRS and ICANS was similar (11% vs. 6.5%, p = 0.277 and 18% vs. 10%, p = 0.146). Median duration of CRS (4 and 5 days, p = 0.150), ICANS (4 and 5 days, p = 0.540) and admission length (20 days for both groups, p = 0.995) did not differ between groups. There were not significant differences in the proportion of patients admitted to ICU (24% vs. 18%, p = 0.284); however, the proportion of patients developing infection in the first 6 months after infusion showed a trend to be higher in the older group (42% vs. 29%, p = 0.099). With a median follow-up of 9.2 months, estimated 12-m PFS was similar between groups (27% vs. 34%, p = 0.303); OS showed a trend to be lower in the older group (31 vs. 50%, p = 0.059).

Conclusions: In our real-life experience, CAR-T cell therapy in patients older than 70 years showed similar efficacy and safety than that observed in younger patients. Consequently, these patients should receive CAR-T cell therapy if treatment criteria are met.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P078 CORRELATION OF CAR-T EXPANSION WITH EARLY TUMOR RESPONSE AND TOXICITY OF CD19 CAR T THERAPY AMONG CHILDREN WITH RELAPSED/REFRACTORY B-ALL

Mariia Fadeeva1, Dmitriy Pershin 1, Alexander Popov1, Olga Molostova1, Elena Kulakovskaya1, Ekaterina Malahova1, Viktoriya Vedmedskaya1, Arina Rahteenko1, Yakov Muzalevskiy1, Alexey Kazachenok1, Larisa Shelihova1, Michael Maschan1

1Dmitry Rogachev National Medical Research Center of Pediatric Hematology, Oncology and Immunology, Moscow, Russian Federation

Background: Chimeric antigen receptor T-cell therapy (CAR-T therapy) is a novel immunotherapy for relapsed/refractory CD19+ lymphomas and leukemias. While clinical trials of CAR T cells have shown clear efficacy in CD19+ malignancies, life-threating toxicities are possible. Different CAR T products are used in a variety of dosing regimens, product formulations and lymphodepletion schedules. Here we report on the correlation of early response and CAR-T toxicity with CAR-T expansion, based on a trial of CD19 CAR-T cells prepared with Prodigy platform.

Methods: From February 2018 through August 2020, we enrolled 57 pediatric patients with refractory B-cell ALL. All the patients had relapsed disease in response to their previous treatment. The purpose of the study was to assess the safety and the efficacy of CD19-CAR-T cells. Four CAR-T dose levels were used: 0.1, 0.5, 1, and 3*10^6/kg. Early response was defined as a negative status of minimal residual disease (MRD), assessed by flow cytometry on day 14 after CAR-T cells infusion. Treatment-related toxic effects were assessed according to ASTCT Consensus grading for CRS and neurologic toxicity. CAR-T cells persistence in peripheral blood (day 7, 11, 14, 28) and their functional differentiation were evaluated by flow cytometry.

Results: All the patients survived 21 days or more. One patient died on day 26 from severe CRS and multiorgan failure. 46 of 57 had early response on day 14. A status of complete remission with negative MRD was significantly associated with a higher peak CAR T-cell expansion (p < 0.001). Ten patients had grade 3-4 CRS and 18 had neurotoxic effects. The CAR T-cell expansion values on day 7, 11, 14 and 28 were significantly higher in patients with neurotoxic effects of grade 3-5 and with CRS of grade 2-4 than in those with toxic effects of lower grade 2 (P = 0.042). We found a significant association of severe neurotoxicity with predominance of central memory CD8 + CAR-T (CD197 + CD45RA-) cell phenotype (p = 0.027).

Conclusions: Our analyses show that high CAR-T expansion associated with greater toxicity. The main question is how the CD19 CAR-T cell therapy can be improved to prevent life-threating toxicities while maintaining effectiveness. One of possible way is to use split dose of CAR-T cells according to leukemia burden.

Disclosure: M.Maschan recieved lecturer’s fee from Miltenyi Biotec.

4 - CAR-based Cellular Therapy – Clinical

P079 CELL AND GENE THERAPY FOR SOLID TUMORS

Patrizia Comoli 1, George Pentheroudakis2, Annalisa Ruggeri3, Ulrike Koehl4, Florian Lordick4, Jarl E. Mooyaart5, Jorinde Hoogenboom6, Alvaro Urbano-Ispizua7, Solange Peters8, Jurgen Kuball9, Nicolaus Kroeger10, Anna Sureda11, Christian Chabannon12, John Haanen13, Paolo Pedrazzoli1

1Fondazione IRCCS Policlinico San Matteo, Pavia, Italy, 2University Hospital, Ioannina, Greece, 3San Raffaele Hospital, Millan, Italy, 4Universitatsklinikum, Leipzig, Germany, 5EBMT Statistical Unit, Leipzig, Netherlands, 6EBMT Leiden Study Unit, Leipzig, Netherlands, 7Clinic University Hospital, Barcelona, Spain, 8Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland, 9Center for Translational Immunology, UMC, Utrecht, Netherlands, 10University Medical Center, Hamburg, Germany, 11Institut Català d’Oncologia, Barcelona, Spain, 12Institut Paoli-Calmettes, Marseille, France, 13Netherlands Cancer Institute, Amsterdam, Netherlands

Background: Among novel targeted treatment strategies, advanced therapy medicinal products (ATMPs) have gained new momentum, as encouraging results were observed in patients with hematologic malignancies who were treated with industry-manufactured gene therapies such as CAR-T Cells. So far, available information on ongoing studies in solid tumors (ST) is limited, due to the variety of programs and infrastructures involved in ATMP manufacturing and delivery.

Methods: The aim of this study was to describe the current landscape of ATMP developments for the treatment of ST from January 2018 to December 2020, by means of a web-based questionnaire circulated within the EBMT and the European Society of Medical Oncology (ESMO) centers.

Results: 149 questionnaires were returned from 53 countries, 23% of the respondents were involved in ATMP trials during the study period, and 15% indicated their intention to start a cell/gene therapy program in the future. The majority of centers involved in ATMP trials treated an approximate number of 1-5 adult patients, while a minority exclusively or partially treated children. Only 23% of the centers enrolled more than 20 pts. Among targeted tumors, melanoma and lung cancer were the most common, but GI tract tumors, gynecological cancers, bone sarcomas, head & neck and breast cancer were also targeted. Although increasingly used, T-cells gene-modified either with CAR sequence or TCR transgene represented no more than 51% of ATMPs employed in patients with ST, differently from the overwhelming prevalence of CAR gene therapy in the setting of hematological malignancies worldwide. Tumor-infiltrating lymphocytes were the most frequently used non-gene modified products. In 56% of the centers, ATMPs were combined with other treatment modalities, largely represented by immunotherapeutic or immunomodulatory agents. Small-scale cell culture and gene engineering was largely performed in academic institutions, mainly by Point of Care manufacturing facilities. When looking at the sources of financial support, only a minority of the studies were supported by EU funding.

Conclusions: Our survey evidences that the development of ATMPs in medical oncology has not reached the same level of maturity than in hematology. Many clinical trials are still based on ATMP production by academic centers, although industry-sponsored trials are present in at least half of the centers. While waiting for breakthrough cellular products to treat solid tumors, that may have widespread distribution as observed for CD19-CART cells, the field may benefit from network models for ATMP production in academic centers.

Clinical Trial Registry: NA

Disclosure: All authors have no conflict of interest to declare.

4 - CAR-based Cellular Therapy – Clinical

P080 PREDICTIVE ROLE OF PRE-LYMPHODEPLETION ALC/AMC RATIO IN PEDIATRIC PATIENTS WITH ACUTE LYMPHOBLASTIC LEUKEMIA RECEIVING CAR T-CELL THERAPY

Hannah Lust 1, Jenna Rossoff1

1Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, United States

Background: The ratio of peripheral absolute lymphocyte count (ALC) to absolute monocyte count (AMC) can predict treatment response and outcomes in adult malignancies, including breast, colon, and lung cancer (Tibaldi et al, J. Can. Res. & Clin. Onc 2008; Stotz et al, Br. J. Canc 2014). Low ALC/AMC ratio prior to CD-19 chimeric antigen receptor T-cell therapy (CAR-T) for relapsed/refractory (R/R) non-Hodgkin lymphoma was associated with shorter EFS and OS (Zhang et al, JCO 2021). ALC/AMC ratio has not been assessed in pediatric patients with R/R acute lymphoblastic leukemia (ALL) receiving CAR-T.

Methods: We performed a retrospective review of 30 pediatric patients who received 32 infusions of CD19-directed CAR-T for R/R B-cell ALL at our institution between 2018 and 2022. Primary endpoints were relapse, loss of B-cell aplasia (BCA), death, and CAR-T-related toxicity in the form of cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS). Receiver operator curve was generated using nominal logistic regression to predict toxicity and survival. Survival curves were calculated using Kaplan-Meier method.

Results: High ALC/AMC ratio ≥4.6 prior to lymphodepleting chemotherapy (LDC) (range D-25 to D-8) was associated with moderate to severe toxicity (grade ≥2) with a likelihood ratio of 1.62 (AUC 0.65, 95% CI 0.46-0.85, P = 0.14, Log-rank). High ALC/AMC ratio ≥4.6 prior to LDC was associated with improved OS (median survival low ratio 580 days, high ratio not reached, P = 0.11, Log-rank). Between patients with low and high ratio, there was no difference in 6-month relapse rate, development of ICANS, EFS, duration of BCA, or duration of remission. Though there was no significant difference in duration of BCA between the groups, duration plateaued after D + 100 in patients with a low ratio. Patients who did not achieve a response by D + 30 were excluded from analysis of duration of BCA. All patients, including those who received stem cell transplant in remission after CAR-T, were included in EFS analysis.

Table 1

Variable

All infusions (n = 32)

ALC/AMC < 4.6 (n = 15)

ALC/AMC > 4.6 (n = 17)

Age (yr), median (range)

13 (3-21)

13 (3-21)

13 (3-19)

High disease burden, n (%)( > 5% blasts on bone marrow, peripheral blasts, or CNS3 disease)

15 (47)

4 (27)

11 (65)

Relapse at any point, n (%)

11 (34)

5 (33)

6 (35)

Deaths, n (%)

10 (31)

7 (47)

3 (18)

CRS ≥2, n (%)

17 (53)

6 (40)

11 (65)

ICANS any grade, n (%)

12 (38)

5 (33)

7 (41)

Conclusions: High pre-LDC ALC/AMC ratio may be a useful tool in predicting pediatric patients with ALL who are more likely to experience moderate to severe CRS after CAR-T. Though we did not examine lymphocyte subsets, this ratio may suggest a more activated phenotype that is primed for inflammation. This may identify patients who would benefit from CRS prophylaxis. High ALC/AMC ratio was also associated with improved OS. This may reflect decreased impact of tumor-associated macrophages, derived from monocytes, in leukemic states as compared to the solid tumor microenvironment. Calculations are limited by small sample size, with 7 of 10 total deaths occurring in the low ratio group. However, the causes of death were not different between the low and high ratio groups. While the distribution of patients with high-disease burden may help explain the relationship between high ratio and increased toxicity, this does not account for the difference in survival. Given the availability of pre-LDC ALC/AMC ratio, these findings warrant further investigation in a larger population to determine validity as a clinical tool in risk-stratifying pediatric patients receiving CAR-T.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P081 CAR-T CELL DYNAMICS ARE ASSOCIATED WITH CLINICAL OUTCOME IN PATIENTS TREATED WITH ANTI-CD19 CAR-T CELL THERAPY FOR RELAPSED/REFRACTORY B-CELL MALIGNANCIES: A REAL-LIFE COHORT STUDY

Maddalena Noviello 1, Francesco De Felice1, Gregorio Maria Bergonzi2, Veronica Valtolina1, Piera Angelillo1, Raffaella Greco1, Elisabetta Xue1, Edoardo Campodonico1, Maria Teresa Lupo-Stanghellini1, Francesca Somai2, Laura Falcone1, Valeria Beretta1, Elena Tassi1, Jacopo Peccatori1, Monica Casucci1, Matteo Carrabba1, Fabio Ciceri2, Chiara Bonini2

1IRCCS Ospedale San Raffaele, Milan, Italy, 2Università Vita-Salute San Raffaele, Milan, Italy

Background: Although anti-CD19 CAR-T cell therapy is highly effective in treating recurrent/refractory B-cell malignancies, the high rate of treatment related toxicities, including cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS), and the significant rate of disease recurrence, still limit its overall impact. The purpose of this study was to shed light on the biological determinants associated with the short- and long-term efficacy and toxicity of anti-CD19 CAR-T cell therapy in a real life setting.

Methods: The infused product (IP) was phenotypically characterized using a 24-color panel. In addition, fresh blood samples collected at several timepoints (+3, 7, 10, 14, 21, 30, 45, 60, 90, 180, 365 days) after the infusion were analyzed to assess the pharmacokinetics (PK) of CAR-T cells.

Results: Twenty-two patients were enrolled. Of these, 20 had aggressive B-cell lymphomas and 2 B-cell precursor acute lymphoblastic leukemia, all relapsed/refractory after previous therapies. Eight patients received axicabtagene ciloleucel (axi-cel), 4 brexucabtagene autoleucel (brexu-cel) and 10 tisagenlecleucel (tisa-cel). The Overall Response Rate at day +90, evaluable for all the patients enrolled, was 50%. Among the major toxicities, CRS was observed in 20 patients (n = 15 grade 1, n = 5 grade 2, none were grade 3 or 4), whereas ICANS was observed in 7 patients (n = 4 grade 1-2, n = 3 grade 3-4).

Despite distinct costimulatory domains and manufacturing process, no significant differences were observed between tisa-cel and axi-cel/brexu-cel products in terms of both T-cell differentiation and PK. Of note, the infusion of CAR-T cells at early stages of differentiation correlated with a more favorable clinical outcome, in terms of long-term (1 year) clinical response and toxicity. Interestingly, a higher proportion of central memory CD8+ CAR-T cells within IP was associated to a lower risk to develop severe neurotoxicity. Moreover, a high expression of the inhibitory receptor LAG-3 on infused CAR-T cells correlated with disease recurrence. The PK curves showed a significant association between higher counts of circulating CAR-T cells in the first 30 days and a favorable response to therapy evaluated for up to 1 year, threatened by a higher frequency of adverse events, mainly ICANS. Finally, a longitudinal in-depth phenotypic T-cell and myeloid characterization on frozen PBMC is ongoing in order to evaluate both CAR+ T cells and non-manipulated immune cells potentially involved in the development of adverse events and in the response to therapy.

Conclusions: Although preliminary, this study indicates the presence of biological biomarkers in the infusion product and in circulating CAR-T cells at early timepoints able to predict the clinical response, including the susceptibility to develop severe neurotoxicity.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P082 MANUFACTURING OF CAR T CELLS FROM PATIENTS WITH AUTO-IMMUNE DISEASES

Michael Aigner 1, Simon Völkl1, Sascha Kretschmann1, Hannah Reimann1, Gloria Lutzny-Geier1, Susanne Achenbach1, Fabian Müller1, Georg Schett1, Andreas Mackensen1

1University Hospital Erlangen, Erlangen, Germany

Background: We have recently shown that robust manufacturing of autologous CD19 CAR T cells is possible for patients with systemic lupus erythematosus (SLE), despite a history of heavy immune-suppressive treatments. To broaden this approach, we have extended the manufacturing efforts to other auto-immune indications like systemic sclerosis (SSc) and dermatomyositis (DM), where patients could potentially also profit from CAR treatments.

Methods: We utilized aphereses of patients with SLE, SSc and DM and subsequently generated CAR T cell products under full GMP conditions in clinical scale using a closed, semi-automatic system. The cellular composition of the source materials, in-process controls and CAR T cell products were analyzed with special focus on T cells and B cell content.

Results: Despite heterogeneity of patients’ aphereses, successful production of GMP grade CAR T cells in sufficient numbers for clinical application was achieved in all cases (N = SLE = 7, SSc=2, DM = 2). All CAR T products showed excellent viability and comparable, homogenous properties (cell numbers, product composition) regardless of the underlying disease, demonstrating both the T cell quality in the source materials as well as the robustness of the manufacturing process.

Conclusions: We were able to demonstrate that CAR T cell manufacturing is feasible for auto-immune patients beyond SLE and thus could offer a promising therapeutic option.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P083 LISOCABTAGENE MARALEUCEL FOR RELAPSED OR REFRACTORY LARGE B-CELL LYMPHOMA: FEASIBILITY, SAFETY AND EFFICACY IN A REAL-WORLD SETTING

Andrew Portuguese 1, Emily Liang1, Aya Albittar2, Alexandre Hirayama1, Brian Till1, Ryan Cassaday1, Ryan Lynch1, Christina Poh1, Filippo Milano1, Aude Chapuis1, Hans-Peter Kiem1, David G. Maloney1, Jordan Gauthier1

1Fred Hutchinson Cancer Center, Seattle, United States, 2University of Washington Medical Center, Seattle, United States

Background: CD19-targeted chimeric antigen receptor (CAR) therapy with lisocabtagene maraleucel (liso-cel) has demonstrated impressive efficacy with manageable toxicity in patients with relapsed or refractory (R/R) large B-cell lymphoma (LBCL) enrolled on the TRANSCEND clinical trial (NCT02631044; Abramson, Lancet, 2020). This led to the approval of liso-cel by the FDA for LBCL patients with R/R disease after 2 or more lines of prior therapy. We report outcomes of LBCL patients referred to our institution for liso-cel treatment in the non-trial setting.

Methods: We retrospectively analyzed the outcomes of all patients who arrived to the Bezos Family Immunotherapy Clinic at the Fred Hutchinson Cancer Center for planned treatment with liso-cel between 1/1/21 and 11/17/22 (intention-to-treat [ITT] group). Disease response was assessed by PET-CT imaging per Lugano 2014 criteria. Cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) were graded using ASTCT criteria.

Results: Of 26 ITT patients, 25 (96.2%) underwent leukapheresis, 24 (92.3%) received lymphodepleting chemotherapy, and 23 (88.5%) received liso-cel. Six of 23 infused patients (26.1%) received an out-of-specification product on an expanded access protocol. The median time from leukapheresis to liso-cel infusion was 33 days (interquartile range [IQR] 30.5-39.5). In the ITT group, the median patient age was 67.6 years (IQR 62.2-72.3), ECOG performance status was 1 (IQR 1.0-1.0), and hematopoietic cell transplantation-specific comorbidity index (HCT-CI) was 1.5 (IQR 1.0-3.0, range 0-7). Fourteen of 26 ITT patients (53.8%) met eligibility criteria for the TRANSCEND trial. Reasons for ineligibility included: coronary artery disease (n = 3), ECOG ≥ 2 (n = 3), significant central nervous system pathology (n = 3), recent history of prior malignancy (n = 1), and pulmonary dysfunction (n = 1). LBCL types included diffuse LBCL (DLBCL; 80.8%, n = 21), transformed DLBCL from indolent histologies (tDLBCL; 11.5%, n = 3), and high-grade B cell lymphoma (HGBCL; 7.7%, n = 2). Bulky and extranodal disease were present in 7 (26.9%) and 13 patients (50.0%), respectively.

Liso-cel was administered in the outpatient setting in 20 of 23 infused patients (87.0%). Among these, 15 (75%) required admission at a median of 4.5 days (IQR 2.0-6.0) with a median duration of hospitalization of 7.0 days (IQR 5.3-11.3). We observed CRS and ICANS after liso-cel infusion in 16 (69.6%; grade ≥3, none) and 7 patients (30.4%; grade ≥3, 13.0%), respectively. Tocilizumab and steroids were administered to 5 (21.7%) and 7 (30.4%), respectively.

Among infused patients (n = 23), the best overall response (ORR) and complete response rates were 78.3% (ITT, 69.2%) and 56.5% (ITT, 50.0%), respectively. After a median follow-up of 310 days (IQR 105-383) among infused patients, the 1-year duration of response (DOR), disease-free survival, and overall survival were 63.0%, 52.7%, and 90.3%, respectively. The median DOR was not reached. In an exploratory univariate logistic regression model, increasing vein-to-vein time was associated with a lower likelihood of response (odds ratio 0.72, 95% confidence interval 0.54 to 0.96, p = 0.02).

Table 25 Table. Patient characteristics (n = 26, all arrivals) and disease outcomes (n = 23, received CAR-T therapy).

Conclusions: Our ITT analysis in an older patient population referred to our center for CD19 CAR-T therapy for R/R LBCL in the non-trial setting showed high rates of durable response after liso-cel with an acceptable safety profile.

Disclosure: Alexandre Hirayama: honoraria from Bristol Myers Squibb, Novartis; research funding from Juno Therapeutics, NanoString Technologies, Nektar Therapeutics.

Ryan Cassaday: honoraria from Amgen, Jazz, Kite/Gilead, Pfizer; consultancy or advisory role for Amgen; research funding from Pfizer, Amgen, Servier, Kite, and Vanda; DSMB for Pepromene Bio; independent response review committee for Autolus; and his spouse has been employed by and owned stock in Seagen. Brian Till: patent/royalties, research funding, and consulting from Mustang Bio; consulting for Proteios Technology, Inc.; research funding from BMS/Celgene.

Ryan Lynch: research funding from TG Therapeutics, Incyte, Bayer, Cyteir, Genentech, SeaGen, Rapt; consulting for Cancer Study Group, SeaGen, Foresight Diagnostics.

Christina Poh: research funding from Incyte and MorphoSys; advisory role for Acrotech.

David Maloney: Honoraria from BMS, Caribou Biosciences, Inc., Celgene, Genentech, Incyte, Juno Therapeutics, Kite, Lilly, Mustang Bio, Novartis, Umoja; research funding from Kite Pharma, Juno Therapeutics, Celgene, Legend Biotech, BMS; royalties from Juno Therapeutics/BMS; stock options from A2 Biotherapeutics, Navan Technologies; consultancy or advisory role for A2 Biotherapeutics, Navan Technologies, Chimeric Therapeutics, Genentech, BMS, ImmPACT Bio, Gilead Sciences, Interius.

4 - CAR-based Cellular Therapy – Clinical

P084 COMMERCIAL MANUFACTURING EXPERIENCE OF AXICABTAGENE CILOLEUCEL DELIVERY IN EUROPE: FROM THE FIRST 2 YEARS TO THE LATEST 2 YEARS

Didier Hallard1, Louis van de Wiel1, Jonathan Tsang1, Suresh Vunnum1, Lisa Mazzoni1, Clare Spooner1, Harry W. Smith 1, Jurjen Velthuis1

1Kite, a Gilead Company, Santa Monica, United States

Background: Axicabtagene ciloleucel (axi-cel), is a CD19-directed genetically modified autologous T-cell immunotherapy with a CD28 costimulatory domain that provides rapid and strong expansion and reprograms T cells to trigger target-specific cytotoxicity of cancer cells. Axi-cel was granted European marketing authorisation in August 2018 for the third-line treatment of adults with relapsed/refractory diffuse large B-cell lymphoma (R/R DLBCL) and primary mediastinal large B-cell lymphoma. Axi-cel is manufactured at 3 sites (El Segundo, CA, USA; Frederick, MD, USA; and Hoofddorp, Netherlands), and has a treatment network of over 300 qualified centres worldwide, including over 180 centres in Europe. The manufacturing process has previously been described (Better M, et al. Cell Gene Ther Insights. 2017) as has the first 2 years’ experience for patients in Europe (Van de Wiel L, et al. EBMT 2021. Poster 017). Here, we discuss commercial manufacturing experience for European patients with R/R DLBCL from the first 2 years to the latest 2 years.

Methods: This analysis includes data from 3701 European patients, including those from the European Union, Great Britain, Switzerland, and Israel, who were registered on Kite Konnect and leukapheresed during the 4-year period from 6 September 2018 to 5 September 2022 (Table). The manufacturing experience of the first 2 years was compared with the latest 2 years. If additional leukapheresis was needed, the first leukapheresis was considered for each patient and is subsequently referred to as that patient’s lot. Delivery success rate is defined as the percentage of patient lots shipped (dispositioned as Qualified Person–released or Physician-released) out of the total number of patients leukapheresed in the time period (excluding those patient lots in process and patients withdrawn). Turnaround time is defined as time from date of leukapheresis to Qualified Person–release for lots using fresh material (excluding starts from re-leukapheresed lots, or for the latest 2 years’ frozen peripheral blood mononuclear cells).

Results: For European patients who underwent leukapheresis between September 2020 and September 2022, median turnaround time was 19 days (range, 16-38). In total, 2398 patient lots were delivered to the treatment centres, resulting in a delivery success rate of 99%. Compared with data for European patients leukapheresed between September 2018 and September 2020, the current median turnaround time has reduced from 25 days to 19 days, respectively. An improvement in delivery success rate since the initial 2-year period (99% vs 96%) was also observed.

Conclusions: Patient outcomes are dependent on rapid and reliable manufacturing capability, as real-world experience has shown (Locke FL, et al. ASH 2022. Abstract 3345). Results from the latest 2 years’ experience demonstrate a consistent and robust commercial axi-cel manufacturing capability with high delivery success rates and improved turnaround time for European patients with R/R DLBCL.

Table

 

First 2 years’ experience

Latest 2 years’ experience

(as of 30Nov22)

Date range

(with final lot disposition available)

6 September 2018 – 5 September 2020

6 September 2020 – 5 September 2022

Patients registered on Kite Konnect and leukaphereseda

1155

2546

Median turnaround timeb

25 days

19 days

Delivery success rate

96% (1072/1115)

99% (2398/2432)

  1. a Includes patients from the European Union, Great Britain, Switzerland, and Israel. b Based on primary peripheral blood mononuclear cell (PBMC) process. A frozen PBMC process was used in the first 2 years’ experience, and a fresh PBMC process was used in the latest 2 years’ experience.

Clinical Trial Registry: N/A

Disclosure: DH: employment with and travel support from Kite, a Gilead Company; and stock or other ownership in Gilead. LvdW: employment with and travel support from Kite Pharma EU B.V.; and stock or other ownership in Gilead. JT: employment with Kite, a Gilead Company; stock or other ownership in Bayer, BioMarin, Bristol Myers Squibb, and Gilead; and travel support from Gilead. SV: employment with and travel support from Kite, a Gilead Company; and stock or other ownership in Gilead Sciences. LM: employment with Kite, a Gilead Company; and stock or other ownership with Baxter, Gilead, and Roche. CS: employee with Kite, a Gilead Company; and stock or other ownership in Gilead Sciences. HWS: employment with and stock or other ownership in Kite, a Gilead Company. JV: employment with and travel support from Kite Pharma EU B.V.; and stock or other ownership in Gilead (self) and Johnson and Johnson (spouse).

4 - CAR-based Cellular Therapy – Clinical

P085 DEVELOPING IN-HOUSE, POINT-OF-CARE MANUFACTURED ANTI-CD19 CHIMAERIC ANTIGEN RECEPTOR (CAR)-T CELLS

Nelson Chun Ngai Chan 1, Sandra Loaiza1, Shab Uddin1, Edward Aina1, Lucy Cook2,3, Renuka Palanicawandar2, Anna Bara4, Eduardo Olavarria2,3, Jane Apperley3,2, Anastasios Karadimitris3,2

1John Goldman Centre for Cellular Therapies, Imperial College Healthcare NHS Trust, London, United Kingdom, Imperial College Healthcare NHS Trust, London, United Kingdom, 3Centre for Haematology, Imperial College London, London, United Kingdom, 4National Institute of Health Research Imperial Clinical Research Facility, London, United Kingdom

Background: Multiple anti-CD19 CAR-T cells have been granted marketing approval and established as the standards of care for the treatment of relapsed/refractory B cell lineage haematological neoplasms. Nevertheless, the timely administration of CART-T cells remains a clinical challenge. Experience from the United Kingdom national CAR-T program revealed the median time from CAR-T approval to infusion was 57 days while 87% of patients required bridging therapy and 26% of patients were unable to receive infusion due to disease progression. In part, this is attributable to intercontinental transport of cellular products between leukapheresis facilities and central current good manufacturing practice (cGMP) compliant cell processing facilities. Here, we propose and validate an alternative model of manufacturing, where clinical grade CAR-T cells are produced at the point-of-care (POC).

Methods: Peripheral blood mononuclear cells (PBMC) were harvested on-site from three healthy volunteers as starting materials. Each harvest was processed in-house using CliniMACS® Prodigy with a 12-day manufacturing protocol which involves T cell enrichment, transduction with lentiviral vectors and cell culture to generate anti-CD19 CAR-T cells in accordance with cGMP standards in a closed system. The starting materials, and the final drug products were characterized and validated against pre-defined product release criteria.

Results: Three anti-CD19 CAR-T cell products were harvested after 12 days of production. Characteristics of the starting materials and final drug products were summarized in Table 1. In summary, all pre-defined product release criteria including transduction frequency, CAR-T cell dose, safety and purity parameters were achieved and clinical grade CAR-T cells were manufactured.

Table 1: Characterization of starting materials and final drug products in POC CAR-T manufacturing

Validation

1

2

3

Starting material

Total viable T cells

4.20 x109

3.54 x109

2.88 x109

Percentage of CD3+ T cellsamong CD45+ viable events

49.8%

49.1%

49.7%

T cell subsets

• CD4+ T cells

58.9%

54.8%

46.9%

• CD8+ T cells

35.9%

32.9%

47.4%

• CD4+CD8+ T cells

2.5%

1.7%

2.0%

• CD4+CD8+ T cells

2.7%

10.5%

3.7%

Multiplicity of infection

20

20

20

Final drug product

Total viable T cells

3.29 x109

4.48 x109

3.92 x109

Percentage of CD3+ T cells

among CD45+ viable events*

(Acceptance ≥ 80%)

93.3%

97.1%

94.7%

Total CD3+ CAR-T cells

1.96 x109

2.5 x109

1.59 x109

Transduction frequency (%)*

(Acceptance: ≥ 10%)

54.02

58.43

42.67

Dose (CD3+CAR+ cells/kg) *

(Acceptance: ≥ 2.0 x106/kg)

2.61 x107

3.58 x107

1.98 x107

CAR+ T cell subsets

• CD4+ T cells

39.8%

66.0%

22.0%

• CD8+ T cells

53.5%

28.6%

71.7%

• CD4+CD8+ T cells

6.6%

5.3%

6.2%

• CD4-CD8- T cells

0.1%

0.1%

0.1%

Vector copy number (Copies) *

(Acceptance: < 5 Copies)

2.28

1.95

1.54

Sterility*†

(Acceptance: Negative)

Negative

Negative

Negative

Mycoplasma* ‡

(Acceptance: Negative)

Negative

Negative

Negative

Endotoxin (EU/mL) *

(Acceptance: < 5 EU/mL)

< 5

< 5

< 5

  1. * Product release criteria
  2. † Detection of microbial growth after 14 days culture
  3. ‡ Detection of mycoplasma by cell culture method
  4. EU, Endotoxin units; PBMC: Peripheral blood mononuclear cells

Conclusions: This validation provides a proof of concept for POC manufacturing of clinical grade CAR-T cells using a 12-day manufacturing protocol within the capacity of an Academic Health Science Centre. Moreover, it enables further clinical studies to evaluate the efficacy of POC manufactured CAR-T cells, the effect of cryopreservation on CAR-T cells and the impact of shortened leukapheresis to infusion lead time in the treatment of patients with advanced B cell neoplasms.

To address the clinical safety of this investigational anti-CD19 CAR-T therapy, a phase I clinical study is in the pipeline. Specific challenges were encountered which underpin the complexity for clinical academics to navigate through the regulatory approval process of Advanced Therapy Investigational Medicinal Product (ATIMP). More importantly, they highlighted the potential of an enhanced Trust and College ATIMP clinical trial approval system and the invaluable input from pharmacy and clinical services. Early engagement of regulatory authorities cannot be overemphasized.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P086 OUTCOMES OF IDECABTAGENE VICLEUCEL IN RELAPSED AND REFRACTORY MULTIPLE MYELOMA PREVIOUSLY TREATED WITH BELANTAMAB MAFODOTIN

Megan Melody 1, Daniel Przybylski1, Adam Robinson1, Adam Lin1, Leo Gordon1, Kehinde Adekola1, Seema Singhal1, Jayesh Mehta1, Jonathan Moreira1

1Northwestern University Feinberg School of Medicine, Chicago, United States

Background: Belantamab mafodotin is an immunoconjugate comprised of an anti-BCMA monoclonal antibody conjugated to a microtubule-disrupting agent. Idecabtagene Vicleucel (Ide-cel) is a BCMA-directed chimeric antigen receptor (CAR) T-cell therapy. Both are used for management of relapsed/refractory (R/R) multiple myeloma (MM) after 4 lines of therapy. The pivotal study of Ide-cel for R/R MM excluded patients previously treated with BCMA-targeted therapy. We studied the outcome of patients treated with Ide-cel and the impact of prior exposure to Belantamab.

Methods: Utilizing the NMH Stem Cell Transplantation and Cell Therapy Program database, we identified 28 patients treated with Ide-cel between June 2021 and May 2022. Group A patients (n = 9) were treated with Belantamab prior to Ide-cel, and Group B (n = 19) had not received Belantamb prior to Ide-cel therapy. Treatment response, disease progression, and survival were assessed. Fisher’s exact test of independence and two-sample t-test with equal variances were utilized to determine the statistical significance of the difference between variables.

Results: Group A had received a median of 8 (range, 5-10) prior lines of therapy and a median of 4 (1-13) dose of Belantamab. Eight out of nine patients responded to Belantamab therapy. Median time from last dose of Belantamab to Ide-cel infusion was 225 (19-484) days. Group B had received a median of 6 (3-13) prior lines of therapy. There was no difference in the incidence and severity of CRS and ICANS between the groups. Response rates in Groups A and B were 33% vs 53% (CR), 44% vs 42% (PR) and 22% vs 5% (NR/PD) respectively. Median PFS of group A patients was 127 (51-175) days versus 291 (25-501) in group B (p = 0.020). There was no difference in OS between patients in Group A vs Group B with a median OS of 277 (179-452) vs 291 (67-516), respectively (p = 0.77).

Conclusions: Our data suggest that prior exposure to BCMA-directed therapy adversely impacts the duration of response to Ide-cel therapy but does not appear to affect overall survival. Given the limited patient numbers, it is difficult to draw definitive conclusions regarding response to therapy. With the advent of new BCMA-directed therapies such as teclistamab, further studies are needed to evaluate the sequencing of therapy with BCMA-directed CAR-T cells.

Disclosure: None.

4 - CAR-based Cellular Therapy – Clinical

P087 DURABLE RESPONSES FOLLOWING CHIMERIC ANTIGEN RECEPTOR T-CELL THERAPY IN PATIENTS WITH RELAPSED/REFRACTORY LARGE B-CELL LYMPHOMA AFTER MULTIPLE LINES OF TREATMENT: THE HELLENIC REAL-WORLD EXPERIENCE

Vasiliki Bampali 1, Panagiotis Dolgyras2, Angelos Matheakakis3, Alexandros Alexandropoulos4, Ioannis Tsonis1, Eleni Gavriilaki2, Ifigeneia Tzannou1, Ioannis Batsis2, Tatiana Tzenou1, Despina Mallouri2, Maria Bouzani1, Anna Vardi2, Achilles Anagnostopoulos2, Stavros Gigantes1, Damianos Sotiropoulos2, Dimitrios Karakasis1, Ioannis Baltadakis1, Ioanna Sakellari2

Evangelismos Hospital, Athens, Greece, George Papanikolaou General Hospital, Thessaloniki, Greece, University Hospital of Heraklion, Heraklion-Crete, Greece, Division of Hematology, Laikon General Hospital, Athens, Greece

Background: Chimeric Antigen Receptor (CAR)-T cells are an approved therapy for relapsed/refractory aggressive B-cell lymphomas. However, their implementation into clinical practice is associated with several challenges regarding proper selection of patients, optimization of efficacy and amelioration of toxicity. The aim of study was to analyze the experience with the application of CAR-T therapy at two national referral centers and to identify areas for improvement.

Methods: Enrolled in the study were consecutive adult patients who received CAR-T infusion as 3rd or greater line of treatment for Large B-Cell Lymphoma (LBCL) at the first two accredited centers in Greece. All patients received lymphodepleting conditioning with cyclophosphamide/fludarabine. The severity of cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) was assessed according to the American Society for Transplantation and Cellular Therapy Consensus Grading. For toxicity management, the European Blood and Marrow Transplantation Group guidelines were followed. Responses were assessed as per Lugano 2014 classification. Progression-free (PFS) and overall survival (OS) were estimated by Kaplan-Meier analysis and were correlated with patient and disease characteristics by log-rank test.

Results: Between 01/2020-11/2022, 40 patients (female/male: 15/25) underwent CAR-T therapy at a median age of 46 (range, 20-69) years. Histologic subtypes were diffuse large B-cell lymphoma (DLBCL) in 30 (75%) patients, transformed follicular lymphoma (TFL) in 6 (15%), primary mediastinal B-cell lymphoma (PMBCL) in 3 (7.5%) and high-grade B-cell lymphoma (HGBCL) in 1 (2.5%). Patients were administered either axicabtagene ciloleucel (n = 22) or tisagenlecleucel (n = 18). The median number of prior lines of therapy was 3 (range, 2-9). Seven (17.5%) patients had received autologous stem cell transplantation. CRS occurred in 33 (82.5%) patients and was severe (grade 3) in 9 (22.5%). Ten (25%) patients developed ICANS, which was severe (grade 3) in 4 (10%). Tocilizumab was administered in 28, and steroids in 17 patients. Transfer to the Intensive Care Unit was required in 8 cases with a median duration of stay of 9.5 (range, 4-14) days. An objective response was observed in 25 (65.8%) of 38 evaluable patients, with complete response (CR) in 17 (44.7%). With a median follow-up of 15 months (range, 1.5-35), PFS and OS at 1 year were 36.1% and 58.7%, respectively. A single treatment-related death was noted 4.7 months after infusion due to infection. Elevated baseline LDH levels were associated with inferior OS (p = 0.005) and PFS (p = 0.038). In addition, 3 or more prior lines of treatment were also correlated with worse OS (p = 0.039) and PFS (p = 0.029). Treatments administered for relapse/progression after CAR-T infusion included chemoimmunotherapy, lenalidomide/rituximab, bispecific T-cell engager (glofitamab), pembrolizumab, and selinexor. Notably, 2 of 6 patients who received glofitamab, achieved complete remission and are scheduled for an allogeneic transplant.

Conclusions: CAR-T cell therapy is a viable option in relapsed/refractory LBCL, and results in durable remissions in a considerable subgroup of patients. The major limitation of the procedure is disease progression following CAR-T infusion, which is associated with pre-treatment disease burden and the number of prior lines of therapy. Therefore, moving CAR-T cells to earlier lines of treatment may enhance their therapeutic potential.

Disclosure: Ioannis Tsonis: Honoraria, Eleni Gavriilaki: Honoraria, Ifigeneia Tzannou: Honoraria, Ioannis Batsis: Honoraria, Maria Bouzani: Honoraria, Dimitrios Karakasis: Honoraria/Advisory Board, Ioannis Baltadakis: Honoraria/Advisory Board, Ioanna Sakellari: Honoraria/Advisory Board.

4 - CAR-based Cellular Therapy – Clinical

P088 BREXUCABTAGENE AUTOLEUCEL FOR ADULT PATIENTS WITH B-ACUTE LYMPHOBLASTIC LEUKEMIA - THE ITALIAN REAL LIFE FROM THE COMPASSIONATE USE PROGRAM

Federica Sorà1,2, Eugenio Galli 1,2, Elisabetta Xue3, Fabio Giglio3, Francesco Restuccia4, Francesco Autore2, Alberto Fresa1,2, Maria Assunta Limongiello2, Elisabetta Metafuni2, Idanna Innocenti2, Sabrina Giammarco2, Ilaria Pansini1, Luca Laurenti1,2, Patrizia Chiusolo1,2, Andrea Bacigalupo1,2, Prassede Salutari4, Fabio Ciceri3, Simona Sica1,2

1Sezione di Ematologia, Dipartimento di Scienze Radiologiche ed Ematologiche, Università Cattolica del Sacro Cuore, Rome, Italy, 2Dipartimento di Diagnostica per Immagini, Radioterapia Oncologica ed Ematologia, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy, 3UO Ematologia e Trapianto Midollo Osseo, IRCCS Ospedale San Raffaele, Milan, Italy, 4Dipartimento di Oncologia ed Ematologia, UOC Ematologia, Ospedale Civile Spirito Santo, Pescara, Italy

Background: Chimeric antigen receptor (CAR) T-cell therapy, approved in the pediatric setting for B-cell acute lymphoblastic leukemia (B-ALL), currently represents one of the most promising immunotherapeutic approaches for hematologic malignancies. In January 2021 Kite/Gilead started a Brexucabtagene autoleucel (brexu-cel) compassionate use program dedicated to adults affected by B-ALL relapsed or resistant to all lines of therapy according to national current practice.

Methods: We report the Italian experience with the compassionate use program of brexu-cel in 10 R/R B-ALL adults treated from February 2021 to October 2022. This observational study based on retrospective data collection and analysis was approved by the ethical committee.

Results: The median age was 40.5 years (range 32-65), 5 were male, 6 showed a common B phenotype, 6 were Philadelphia positive, median number of prior lines were 3 (range 2-6), and 7 patients had received a previous allogeneic HSCT.

Five out of 10 (50%) patients were refractory to the last treatment at the time of CAR-T eligibility. Patients had been exposed to blinatumomab and inotuzumab before leukapheresis in 4 (40 %) and 5 (50%) cases, two had received both. All patients had CD19+ blasts at the enrolment. ECOG performance status score was 0 in 6 patients, and 1 in the others. All patients received bridging therapy.

Lymphodepletion was administered with Flu-Cy regimen, according to ZUMA-3 trial schedule; none experienced progression during bridging therapy and all received CAR-T infusion. The median time from leukapheresis to CAR-T reinfusion was 68 days (range 37-122). Two patients (20%) experienced a manufacturing failure, and a second leukapheresis was then successfully performed.

CRS was observed in 9 patients after 7 median days (range 4-9): CRS was usually mild, with grade 1 in 8 patients (80%) and grade 2 in 1 (10%) patient. ICANS was observed in 2 (20%) patients after 7.5 median days (range 6-9), graded 1 in one patient and 2 in the other. Tocilizumab was administered in 3 patients and high-dose steroids was required in 1 patient. No patients developed consumptive coagulopathy, and macrophage activation syndrome was recorded in 1 patient. Other adverse events included transient transaminase elevation (n = 2) and deep venous thrombosis (n = 1). None required Intensive Care Unit admission.

Median follow-up was 178 days (32-424); nine out of 10 patients were in MRD negative CR at day +30, while one patient progressed before day 30. Of six patients evaluable 6 months after treatment, three were in MRD negative CR, two had progressed and received salvage therapy with blinatumomab, and one had positive MRD with persistent morphological CR, received preemptive TKI and subsequent allo-HSCT. At last FU, one patient deceased due to relapse.

Conclusions: In our series of 10 infused heavily pretreated adult B-ALL patients, brexu-cel was well tolerated and effective.

The incidence of CRS, ICANS, and other toxicities was very low. ORR at day+30 after infusion is very promising, despite a longer follow-up and further study are necessary to confirm these data.

Disclosure:

4 - CAR-based Cellular Therapy – Clinical

P089 VIRAL REACTIVATION AFTER CAR-T CELL THERAPY: ACTIVE SURVEILLANCE, PROPHYLAXIS AND TREATMENT STRATEGIES

ELISABETTA XUE1, Edoardo Campodonico1, Maria Teresa Lupo Stanghellini1, Chiara Oltolini1, Piera Angelillo1, Elisa Diral1, Francesca Farina1, Daniela Clerici1, Consuelo Corti1, Jacopo Peccatori1, Fabio Ciceri1, Matteo Giovanni Carrabba1, Raffaella Greco 1

1San Raffaele Hospital, Milan, Italy

Background: CD19-targeted chimeric antigen receptor (CAR)-T cell therapy provides a valid treatment option for B-cell haematological malignancies, with substantial disease remission rate. However, CAR-T cells on-target effects can determine B-cell aplasia and hypogammaglobulinemia which expose patients to prolonged infectious risk.

Methods: We retrospectively collected data on viral infections and immune-reconstitution in patients treated with CD19 CAR-T cell therapy. Fludarabine-cyclophosphamide based lymphodepletion was given to all patients. Routine viral PCR monitoring on peripheral blood after CAR-T cell therapy included CMV, HSV6 and EBV. Patients received prophylactic acyclovir. Starting on May 2022, all patients received antii-SarsCov2 prophylaxis with tixagevimab/cilgavimab. Lymphocyte counts were monitored on +30, +90 and +180 days. Patients with a minimum of 3 months of follow-up were analysed; data were censored at disease relapse.

Results: Twenty-eight CAR-T cell recipients treated between December 2019 and July 2022 were analysed (Table1). Median follow-up was 362 days (range 91-1005). Twenty-one and three patients achieved complete and partial response, respectively; of these, ten eventually relapsed after a median of 124 days (range 74-466). Four never responded.

Median CD4 values at 1 months were 196/mcl (range 21-1127), at 3 months were 215/mcl (range 11-481), at 6 months were 284/mcl (range 9-535). Median CD8 values at 1 months were 365/mcl (range 81-1047), at 3 months were 447/mcl (range 14-1518), at 6 months were 587/mcl (range 62-1418). Median NK values at 1 months were 95 (range 6-462), at 3 months were 91 (range 36-152), at 6 months were 85 (range 29-394). B-cell aplasia was still present at 6 months in more than 50% of evaluable cases.

CMV, EBV and HSV6 reactivations occurred in 8, 2 and 1 patient, respectively. CMV reactivated after a median of 26 days (range 10-167); in five cases the viremia spontaneously resolved, whereas in three the reactivation was clinically significant and occurred within the first month; one developed CMV-related pneumoniae and one developed concomitant hepatitis. EBV and HSV6 reactivation were non-clinically significant and spontaneously resolved. None had VZV reactivation.

Nine patients developed Covid-19 after a median of 15 months (range 2-32); six had received post-CAR-T vaccination, one had received tixagevimab/cilgavimab prophylaxis. Seven had minor respiratory symptoms: of these, three received nirmatrelvir/ritonavir and one sotrovimab. Two developed Covid-19-related pneumoniae at 32 and 17 months after CAR-T, respectively; both were Covid-19 vaccinated and in disease remission; one received nirmatrelvir/ritonavir, the other received remdesivir plus sotrovimab, with clinical resolution. One patient had persistent viral shedding. Both had B-cell aplasia and CD4 < 200/mcl at the time of infection.

Patients characteristics N = 28

Primary Disease

DLBCL n = 15, PMBCL n = 2, HG BCL n = 2, MCL n = 4, ALL n = 5

Prior treatment lines

2 lines n = 8, 3 lines or more n = 20

Prior transplant

Auto n = 8, Allo n = 5, None n = 16

Disease status at CAR T

PD n = 17, CR/PR n = 9, SD n = 2

CAR T Product

Axi-Cel n = 6, Tisa-Cel n = 15, Brexu-Cel n = 7

Cytokine release syndrome

Grade 1-2 n = 23

Grade 3 or more – none

Neurotoxicity

Grade 1-2 n = 3

Grade 3 or more n = 3

Conclusions: Despite not frequent, clinically significant viral reactivations can occur after CAR-T cell therapy, confirming the importance of active monitoring of herpes-viruses reactivation including CMV and EBV within the first six months after treatment. Role of additional viruses like HHV-6 needs further investigation in this setting. Covid-19 can have severe presentations also long-time after CAR-T administration; several studies suggested CAR-T recipients had an attenuated immune response to SARS-CoV-2 vaccination compared with healthy individuals. In this setting, the risk of severe Covid-19 presentation can be minimized by prophylaxis strategies and early treatment.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P090 OUTCOMES WITH AXICABTAGENE CILOLEUCEL IN RELAPSE/REFRACTORY FOLLICULAR LYMPHOMA: PILOT STUDY OF A NOVEL COMPOSITE ENDPOINT OF TOXICITY AND PROGRESSION-FREE SURVIVAL

Razan Mohty 1, Julio Chavez1, Sameh Gaballa1, Sepideh Mokhtari1, Rawan Faramand1, Hayder Saeed1, Christina Bachmeier1, Bijal Shah1, Taiga Nishihori1, Frederick Locke1, Michael Jain1, Aleksandr Lazaryan1

1H. Lee Moffitt Cancer Center, Tampa, United States

Background: Axicabtagene Ciloleucel (axi-cel) is an FDA approved anti-CD19 chimeric antigen receptor T cell (CAR T) therapy for relapsed/refractory (R/R) follicular lymphoma (FL) after 2 or more lines of therapy. This approval was based on the results of ZUMA-5 trial (Neelapu et al. ASH 2022) showing high overall response rate (ORR, 94%), complete remission (CR, 79%), and prolonged progression free survival (median PFS = 40.2 months; 3-years PFS = 54.4%).

Methods: In this single-center retrospective study, we examined the outcomes (safety, responses, and survival) following axi-cel for R/R FL and focused on a novel composite endpoint of toxicity and progression-free survival (TPFS) defined as absence of severe (Grade 3+) cytokine release syndrome (CRS), immune effector cell-associated neurotoxicity syndrome (ICANS), lymphoma progression/no response, and non-relapse mortality within 6 months after CAR T cell infusion. Both CRS and ICANS were graded according to ASTCT Consensus grading.

Results: A total of 21 consecutive patients (June 2018-Dec 2021) were included in this study. Baseline patient characteristics are summarized in Table 1 and divided in 2 groups: 1) patients with TPFS at 6-months (N = 12, 57%); 2) patients with toxicity and/or progression at 6 months (N = 9, 43%). The median age was 61 years. Most patients had high-risk disease with 62% (N = 13) of the patients having FL international prognostic index (FLIPI) > 2 and 76% (N = 16) had POD24. With a median follow-up of 36 months for the entire population, 2-year overall survival (OS) and PFS were 84.6% and 48.5%, respectively (Fig 1A). At 6 months, the ORR and CR rates were 81% and 76%, respectively. The best ORR and CR rates were 95% and the median duration of response was 15.9 months. Any grade and severe CRS occurred in 18 (85.7%) and 1 (4.8%) patients, respectively. Any grade and severe ICANS occurred in 14 (66.7%) and 5 (23.8%) patients, respectively. The 6-months TPFS was 57% (Fig 1B). Of patients with 6 months toxicity and/or progression, 1 had severe CRS and progressive disease, 5 had severe ICANS, and 3 had progressive or stable disease (Fig 1B). There was no non-relapse mortality within 6 months. In the landmark analysis of long-term PFS, 6 months TPFS did not appear to have prognostic significance in this pilot analysis albeit limited by number of survival endpoints (only 2 deaths and 9 progression events) (p = 0.52, Fig 1C).

 

All patients

N = 21

Toxicity/Progression-free at 6 months

N = 12

Toxicity and/or progression at 6 months

N = 9

Age, median (range)

61 (42-79)

61 (42-79)

61 (49-79)

Male, N (%)

12 (57%)

5 (42%)

5 (56%)

Histologic grade

   

Grade 1-2

17 (81%)

9 (75%)

8 (89%)

Grade 3a

4 (19%)

3 (25%)

1 (11%)

Follicular Lymphoma International Prognostic Index

   

Low risk (0-1)

2 (10%)

0

2 (22%)

Intermediate Risk (2)

6 (29%)

4 (33%)

2 (22%)

High Risk (≥3)

13 (62%)

8 (67%)

5 (56%)

Bulky disease (≥10 cm)

2 (10%)

1 (8%)

1 (11%)

Previous lines of therapy, median (range)

4 (2-9)

4 (2-6)

4 (2-9)

≥3 lines of therapy

15 (71%)

8 (67%)

7 (78%)

Previous Pi3K inhibitor

7 (33%)

5 (42%)

2 (22%)

Previous bendamustin

12 (57%)

6 (50%)

6 (67%)

Previous lenalidomide

7 (33%)

4 (33%)

3 (33%)

Previous autologous stem cell transplantation

2 (10%)

0

2 (22%)

Refractory to last line of therapy at time of CAR T

14 (67%)

9 (75%)

5 (56%)

POD24 from first anti-CD20 therapy

16 (76%)

8 (67%)

8 (89%)

Conclusions: Our safety and efficacy outcomes with axi-cel in R/R FL were largely aligned with the results of the pivotal ZUMA-5 trial. We analyzed a novel composite end point, TPFS, which reflects the major complications of CAR T therapy. Using this new end point, 57% of axi-cel recipients with R/R FL survived at 6 months without experiencing a TPFS-defining event. Since each TPFS component is clinically meaningful, this endpoint may represent an ideal recovery outcome after CAR T therapy as it measures initial success without progression, major morbidity or mortality. Thus, TPFS can serve as a uniform safety and efficacy endpoint for comparing different CAR T products in future larger studies among patients with R/R FL.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P091 CYTOKINE-BASED MODELS FOR EFFICIENT DIFFERENTIATION OF INFECTION AND CYTOKINE RELEASE SYNDROME IN PATIENTS WITH HEMATOLOGICAL MALIGNANCIES

Linqin Wang 1,2, Yuqi Lv1,2, Linghui Zhou1,2, Shenghao Wu3, Yuanyuan Zhu1,3, Shan Fu1,2, Shuyi Ding1, Ruimin Hong1,2, Mingming Zhang1,2, Hanjing Yu4, Alex H. Chang5, Guoqing Wei1,2, Yongxian Hu1,2, He Huang1,2

1The First Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China, 2Zhejiang University Medical Center, Hangzhou, China, 3Wenzhou Central Hospital, Dingli Clinical College of Wenzhou Medical University, Wenzhou, China, 4Hangzhou Medical College, Hangzhou, China, 5Shanghai YaKe Biotechnology Ltd., Shanghai, China

Background: Although the promising efficacy of chimeric antigen receptor (CAR)-T cell therapy has been demonstrated widely in clinical trials, the complexity and severity of toxicities largely hampered the widespread clinical application. Cytokine release syndrome (CRS) is the most common toxicity after CAR-T cell infusion, characterized by fever and various symptoms of multiple organ involvements. Nonetheless, these profiles have substantial overlaps with infection. Worse still, routine blood culture and radiological examinations show limited efficiency in quick diagnosis, which would influence the accurate clinical intervention. Therefore, it’s necessary to build an efficient and feasible diagnosis model.

Methods: This study included 95 patients with relapsed or refractory B cell hematologic malignancies who developed CRS-related fever after CAR-T cell infusion (ChiCTR1800017404, ChiCTR-ORN-16008948, ChiCTR1800017402, ChiCTR1800015575), as well as 80 patients who suffered from infective fever in our center between January, 2017 and June, 2022. After fully informed consent, blood were collected from patients during the episode of fever ( > 38°C).

To build the clinical model, patients were chronologically divided into training cohort (n = 130) and external validation cohort (n = 45) approximately at the proportion of 3:1. In training cohort, forty-four serum cytokines during fever were detected and log-transformed before analysis. Two cytokine-based models were obtained via classification tree algorithm and stepwise logistic regression analysis, respectively. Based on that, two new cytokine panels containing targeted cytokines were tested in validation cohort. Sensitivity, specificity and receiver operating characteristic (ROC) curve were used to validate the accuracy of models.

Results: A feasible decision tree model was obtained based on three cytokines with moderate accuracy. It identified that high levels of IFN-β, GRO-α, and low IP-10 were predominant indicators for infection. The area under curve (AUC) reaches around 0.94, and the sensitivity is no less than 90% both in training and validation cohort.

In contrast, a seven-cytokine based predictive model demonstrated higher accuracy through stepwise regression analysis. The equation model converts the logistic regression score into a modeled probability. Patients with high cumulative score (å 0.610) would be recognized as a “infection” case. Consistent with decision tree model, this model identified high level of IP-10 as an indicator for CRS, accompanied with MIP-3b, MIP-1a and VEGF. Moreover, it indicated that high level of Eotaxin, MIP-3a and IL-4 were indicators for infection. Of note, in validation cohort, this could completely divide infective and CRS case in a accurate way.

Conclusions: Based on the profiles of serum cytokine during fever, our study developed feasible and accurate clinical models for efficiently differentiate infection from CRS, which could prompt the early diagnose and early intervention of infective cases. Moreover, the observation of wide-spectrum serum cytokines could facilitate the understanding of the mechanism of CRS or infection initiated fever.

Clinical Trial Registry: ChiCTR1800017404, ChiCTR-ORN-16008948, ChiCTR1800017402, ChiCTR1800015575

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P092 CAR-T CELL EXPANSION DYNAMICS AND THEIR CLINICAL IMPLICATIONS IN LYMPHOMA PATIENTS TREATED WITH AXICABTAGEN CILOLEUCEL AND TISAGENLECLEUCEL

Diego Carbonell1,2, Eva Catalá 1,2, Ana Pérez-Corral1,2, Gillen Oarbeascoa1,2, Rebeca Bailén1,2, Ignacio Gómez-Centurión1,2, José Luis Díez Martín1,2,3, Jorge Gayoso1,2, Mi Kwon1,2, Javier Anguita1,2

1Hospital General Universitario Gregorio Marañón, Madrid, Spain, 2Instituto de Investigación Sanitaria Gregorio Marañón, Madrid, Spain, School of Medicine, Complutense University of Madrid, Madrid, Spain

Background: CD19-targeted chimeric antigen receptor (CAR)-T cell is a treatment for B-cell lymphoma patients. CAR-T cell monitoring is important to ensure a correct follow-up, being multiparametric flow cytometry (MFC) the actual gold standard technique. However, clinical implications of CAR-T dynamics are not fully elucidated. Here, we aimed to analyze potential associations between CAR-T cell levels and post-infusion complications.

Methods: Seventy-seven patients affected by B-cell lymphoma treated with axicabtagen ciloleucel (axi-cel -n = 48-) and tisagenlecleucel (tisa-cel -n = 29-) between September 2019 and July 2022 were included in the study. MFC analysis was performed on a DxFLEX cytometer (Beckman Coulter), using CD19 (20-291) protein-FITC (ACRO Biosystems). Comparison between axi-cel and tisa-cel CAR-T cell levels at different time points was performed using Mann-Whitney U test. Associations between main post-CAR-T therapy complications (cytokine release syndrome -CRS-, immune effector cell-associated neurotoxicity syndrome -ICANS-, relapse and exitus) and CAR-T cell levels (on different post-CAR-T days and at maximum peak of expansion) were analyzed by cumulative incidence using Fine-Gray test, based on cut-off points obtained from ROC curves. Statistical tests were performed with R (3.3.2 version) and Graphpad Prism (8.0.1 version).

Results: Median values of CAR-T cells at different days post-infusion of axi-cel and tisa-cel were different only on day 3 post-infusion (Mann-Whitney U test, <0.001). No significant differences were found between axi-cel and tisa-cel in terms of clinical complications, except for ICANS, which was greater in axi-cel group (CI 53% vs. 18%, p = 0.006). Respecting CAR-T complications and CAR-T cell expansion dynamics, absence of CAR-T cells at day +3 was associated with an increased risk of developing ICANS in patients treated with axi-cel (CI 73% vs. 18%, p = 0.007.

Conclusions: Based on the results obtained, levels of CAR-T cells in peripheral blood at day 3 after infusion could be a potential biomarker to predict ICANS in patients treated with axi-cel, a common complication in this CAR-T cell therapy. Further studies are needed to confirm these results.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P093 FROM MNC COLLECTION TO CAR T CELL INFUSION: EXPERIENCE OF A CELLULAR THERAPY CENTRE

Susana Roncon 1, Catarina Pinho1, Sérgio Lopes1, Filipa Bordalo1, Sara Ferreira1, Fátima Amado1

1Instituto Português de Oncologia do Porto, Porto, Portugal

Background: T cells engineered to express chimeric antigen receptors (CAR) are capable of inducing clinical responses in selected patients with relapsed/refractory hematologic malignancies. However, not all are eligible to receive these cell therapies, as manufacture presents unique challenges and patients aggressive diseases. Our goal was to characterize the population of patients treated with CAR T cells in our hospital and to analyze the real time of the circuit from collection to infusion.

Methods: We retrospectively studied 52 adult patients proposed for CAR T therapy (April/2019-November/2022), with statistical analysis of clinical data. All patients performed collection of mononuclear cells (MNC) by leukapheresis using Spectra Optia with a continuous method and citrate as single anticoagulant (ratio 1:12-1:14). MNC were shipped to manufacturing site and CAR T final product was received and stored at -150ºC until infusion.

Results: Nineteen female and 33 male patients with a median age of 50(18-70)years, weight of 80(47-185)kg, volemia of 5.390(3.290-12.950)ml, diagnosis of Diffuse Large B-Cell Lymphoma (n = 48), Acute Lymphoblastic Leukemia (n = 3) or Mantle Cell Lymphoma (n = 1), followed at our hospital (n = 26) or referred by another (n = 26), were proposed to therapy with Tisa-cel (n = 29), Axi-cel (n = 22) or Brexu-cel (n = 1).

During pre-leukapheresis screening, we noted that patients had previously received 2 or more ( ≤ 6) chemotherapy lines. Nineteen collected peripheral blood stem cells, but only 8 were transplanted with an autologous graft 28(6-118)months before; 7 with progressive disease and 3 poor mobilizers were no longer considered eligible for transplantation; 1 was transplanted 20 months after CAR T infusion. Two young adults performed allogeneic transplant (one familiar, one unrelated) 16 months before. Regarding infectious disease marker reactivity, we reported that 6 were positive for Hepatitis B and C and syphilis, but with no active infection.

Following EBMT criteria, all but one had >1.000 nucleated cells/ul [4.900(900-17.000)/ul] and all presented >200 CD3+ cells/ul in the peripheral blood [900(240-2.890)/ul] immediately before leukapheresis.

Patients processed 13.919(5.762-32.749)ml of total blood, equivalent to 2.3(1.2-7.9) of blood volemias, with a flow rate of 55(35-79)ml/min during 226 (119-783)minutes. All of them performed a one-day collection, except one that required two days, by peripheral vein (n = 28) or central venous catheter (n = 24).

Leukapheresis products released for Tisa-cel production were shipped cryopreserved (n = 28) with a median number of 12.8(5.0-76.6)x109 nucleated cells and 5.8(1.6-19.4)x109 CD3+ cells; other products (n = 23) were sent “fresh”; one is stored waiting for new clinical indication and further manufacturing.

We received and stored 42 compliant CAR T products and 3 out-of-specification (1 low viability, 1 low dose and 1 with residual beads) but with no contraindication for administration; 6 productions were cancelled due to patient death. We infused 42 and 3 are still cryopreserved (2 patients are in complete remission, 1 is waiting for infusion).

Median (minimum-maximum) time between the main steps of CAR T circuit

1st clinical evaluation → MNC collection: 8 (1-46) days

MNC collection → CAR T storage: 32 (21-177) days

CAR T storage → CAR T infusion: 11 (0-116) days

  1. End of CAR T thawing → start of CAR T infusion: 12 (7-55*) minutes (* only one patient exceeded 30 minutes due to catheter malfunction).

Conclusions: In our department, CAR T circuit is fully implemented; traceability and chain of custody are well guaranteed. However, some improvements should be made to simplify logistical issues. In order to offer these innovative therapies to a greater number of patients in a timely manner, academic CAR T cells should also be considered.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P094 ACCESS BARRIERS FOR ANTI-CD19+ CHIMERIC ANTIGEN RECEPTOR T (CAR-T) CELL THERAPY FOR NON-HODGKIN LYMPHOMA (NHL) ACROSS A LARGE COMMUNITY TRANSPLANT AND CELLULAR THERAPY NETWORK

Minoo Battiwalla 1, Michael Tees2, Ian W. Flinn1, Jeremy Pantin1, Jesus Berdeja1, Tara Gregory2, Michael Maris2, Vikas Bhushan3, Estil Vance III3, John Mathews3, Carlos Bachier4, Paul J. Shaughnessy4, Aravind Ramakrishnan5, Shahbaz Malik5, Shahram Mori6, Rocky Billups7, Betsy Blunk7, C. Fred LeMaistre7, Navneet Majhail7

1Sarah Cannon Transplant and Cellular Therapy Program at TriStar Centennial Medical Center, Nashville, United States, 2Colorado Blood Cancer Institute at HealthONEs Sarah Cannon Cancer Institute at Presbyterian/St. Luke’s Medical Center, Denver, United States, 3Texas Oncology, PA, Dallas, United States, 4Adult Blood and Marrow Transplant, Sarah Cannon Transplant and Cellular Therapy Program at Methodist Hospital, San Antonio, United States, 5Sarah Cannon Transplant and Cellular Therapy Program at St. David’s South Austin Medical Center, Austin, United States, 6Sarah Cannon Transplant and Cellular Therapy Program at Mountain View Hospital, Las Vegas, United States, 7Sarah Cannon Cancer Institute, Nashville, United States

Background: Anti-CD19+ CAR-T cell utilization has accelerated in the US after FDA-approval for several indications in B-cell NHL. The current generation of CAR-T therapies is complicated by high cost, a lengthy manufacturing process, and acute toxicities requiring administration at specialized centers. The Sarah Cannon Transplant and Cellular Therapy Network (SCTCTN) implemented a coordinated approach to streamline the utilization of these complex therapies. We conducted a retrospective review to identify barriers to delivery of anti-CD19+ CAR-T cells for NHL.

Methods: All patients referred to SCTCTN for CAR-T were tracked in our prospective registry (Stafa-CT). We identified 254 patients referred to 5 SCTCTN centers who were intended recipients of FDA-approved anti-CD19+ CAR-T through 9/5/22.

Results: The 254 patients were categorized as follows:

  1. 1.

    96 patients (38%) were screened but did not proceed or were not eligible because of disease progression (14), decline in clinical status (7), insurance denial (8), other treatment including CAR-T clinical trials (17), or other reasons (50).

  2. 2.

    22 patients (9%) were collected but did not infuse because of disease progression (4), decline in clinical status (7), manufacturing issues (5), other treatment (2), other reasons (4).

  3. 3.

    29 patients (11%) are approved and currently awaiting collection or infusion.

  4. 4.

    107 patients (42%) were collected and received infusion of IECT.

For the whole cohort, median age at referral was 61 years (range 23-87), 65% were male, referral year was prior to 2019 (5%), 2019 (12%), 2020 (16%), 2021 (34%), and 2022 (33%). The diagnosis was diffuse large B-cell (68%), mantle cell (18%), follicular (10%), or other (4%) lymphoma. 59% were privately insured, 33% had Medicare, 3% had Medicaid, 1% had other (eg. VA), 4% missing. Median time from referral to consultation was 4 days. For eligible patients, the median time from referral to infusion was 143 days. Among the 107 infused, product was Axi-cel (56%), Brexu-cel (18%), Liso-cel (13%) and Tisa-cel (13%). Infused patients had a median age of 61 (range 23-82), 70% were male, and 60% had private insurance.

Conclusions: Despite prompt evaluation at the time of referral, 47% of patients did not ultimately receive CAR-T therapy for NHL. Disease-related factors were the primary reason, while age and financial coverage were not major contributors. Ongoing analyses will review the impact of other health determinants such as race and socioeconomic vulnerability. The complexity of the current generation of CAR-T involves authorization, apheresis, and manufacturing steps which practically result in a duration of >3 months till CAR-T infusion. Implications of our findings include the necessity for reducing manufacturing time, education efforts to support early referral, and routine planning for interim bridging strategies.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P095 ADDITION OF NIVOLUMAB TO ANTI-CD-19 CAR-T CELLS IN PATIENTS WITH STABLE/PROGRESSIVE DLBCL AT LYMPHODEPLETION – A SINGLE ARM, PHASE 2, PROSPECTIVE INTERVENTIONAL STUDY

Ron Ram 1, Odelia Amit1, Chava Perry1, Yair Herishanu2, Irit Avivi1, Nadav Sarid2, Arie Apel3, Meir Price4, Ariel Aviv5, Ronit Gold1, Chen Glait-Santar1, Sigi Kay1, Rivi Hertz-Levi6, Dina Rosenberg6, Ofrat Beyar-Katz6

1Tel Aviv Sourasky Medical Center, Tel Aviv, Israel, 2Wolfson Medical Center, Holon, Israel, 3Shamir Medical Center, Rishon Leziyon, Israel, 4Carmel Medical Center, Heifa, Israel, 5Haemek Medical Center, Afula, Israel, 6Rambam Medical Center, Heifa, Israel

Background: Progression of DLBCL is the major cause of failure of CAR-T therapy and those with stable or progressive disease (SD/PD) at the time of lymphodepletion have a dismal 6 month PFS of 20-30%. Improved in-vivo expansion of CAR-T cells may overcome this grave prognosis and may result in better PFS. Based on our recent published analyses, day +7 CAR-T concentrations analyzed by flow cytometry can predict response in this population. We thus hypothesized that the addition of nivolumab will improve clinical response by “switching on” the immune system and inducing greater CAR-T cell expansion.

Methods: Patients with SD/PD DLBCL documented by PETCT prior to lymphodepletion were recruited to this prospective phase 2 trial (NCT05385263). Patients who did not present active CRS/ICANS were assigned to receive nivolumab (3mg/kg) on day +5. Those with <100 CAR-T cells/microL on day +7 were then given an additional dose of nivolumab on day +19. Endpoints included assessment of disease response at 1 and 3 months, response duration, and overall safety of the protocol. This trial is planned to recruit 20 patients and the current analysis aimed to test feasibility.

Results: As of November, 2022, 8 patients were recruited and received anti-CD19 CAR-T (Axicabtagene ciloleucel, n = 5 and tisagenlecleucel, n = 3). Median age was 63 (range, 47-77) years. Of them, 4 were not eligible to receive nivolumab due to ongoing active CRS (n = 4) and ICANS (n = 1). Day +7 CAR-T concentrations were higher in patients not eligible for nivolumab infusion compared to those eligible (range, 87-724 cells/microL vs. range 1-26 cells/microL). Three patients received the full 2 doses of nivolumab, while 1 patient withdrew consent after the first dose. Six non-hematologic AEs were noted (CRS grade 1-2, n = 3 and CRS grade 3, n = 1; fever, n = 1 and folliculitis, n = 1) and 2 SAEs (neutropenia grade 4, n = 1 and pseudomonas skin infection, n = 1). Of these, 5 AEs were attributed to nivolumab. Three patients received tociluzumab and 2 patients, in addition, received systemic steroids. No patients required ICU admission. There was no difference in the median duration of severe neutropenia and severe thrombocytopenia between patients who received nivolumab and those who were excluded (p = .34 and p = .37, respectively). In 2 patients who received the 2 doses of nivolumab, serial PCR CAR-T concentrations showed a shift from the standard day +7 peak concentrations to day +21, Figure. Disease response, documented by PETCT at 1-month showed PR, n = 2, SD, n = 1, and PD, n = 1. Both PETCT PRs converted to CR in subsequent 3-month PETCT. PETCT re-evaluation of the 3rd patient with SD is pending.

Conclusions: Interim data from this risk-for-relapse-adopted administration of nivolumab suggest feasibility with a reasonable toxicity profile. Assessment of disease response at 1 month may be premature at this time. Recruitment for this trial is ongoing.

Clinical Trial Registry: NCT05385263

Disclosure: Ron Ram. - Honoraria and speakers bureau: Novartis, Gilead; Ofrat Beyar-Katz - Honoraria: Novartis

4 - CAR-based Cellular Therapy – Clinical

P096 COST-EFFECTIVENESS AND BUDGET IMPACT ANALYSES OF AXICABTAGENE CILOLEUCEL VS SALVAGE CHEMOTHERAPY IN ADULTS WITH RELAPSED/REFRACTORY LARGE B-CELL LYMPHOMA FROM SINGAPORE’S HEALTHCARE SYSTEM PERSPECTIVE

Francesca Lorraine Wei Inng Lim1, Tza-Kai Diong 2, Graeme Ball3, Soon Thye Lim4, Hian Li Esther Chan5, Colin Phipps Diong6, Zi Yi Lim7, David Champion8, Devi Raman8, William Ying Khee Hwang9

1Singapore General Hospital, Singapore, Singapore, 2Kite Gilead Science, Singapore, Singapore, 3Gilead Sciences Canada Inc, Mississaga, Canada, 4National Cancer Centre, Singapore, Singapore, 5National University Cancer Institute, National University Health System, Singapore, Singapore, 6Mount Elizabeth Medical Centre, Parkway Cancer Center, Singapore, Singapore, 7Center for Clinical Haematology, Singapore, Singapore, 8Ansea Consultants Pte Ltd, Singapore, Singapore, 9National University Cancer Institute, Singapore, Singapore

Background: Patients with relapsed or refractory large B-cell lymphoma (R/R LBCL) have limited treatment options and a poor prognosis in Singapore. Axicabtagene ciloleucel (axi-cel) has demonstrated improved clinical outcomes in the ZUMA-1 trial but its clinical and economic value to Singapore’s multi-payer healthcare system is currently unknown. Hence, in this study, our objective is to evaluate the cost-effectiveness and budget impact of axi-cel versus salvage chemotherapy (SC) for treating R/R LBCL patients who have failed at least 2 lines of systemic therapies from Singapore’s public healthcare system perspective.

Methods: Over a life-time horizon, a mixture-cure partition survival model was developed to evaluate the cost-effectiveness of axi-cel vs. SC. Clinical data for axi-cel and SC were derived from ZUMA-1 (5-year follow-up) and the SCHOLAR-1 retrospective cohort study, respectively. Outcomes included quality-adjusted life-years (QALYs), health-related quality of life (HRQoL), costs and incremental cost-effectiveness ratios (ICERs). Direct costs for conditioning chemotherapy, hospitalisation, drugs and management of adverse events were included. Sensitivity analysis and scenario analyses were conducted. The financial implication of introducing axi-cel in Singapore was analysed comparing the current treatment pathway (without axi-cel) with a future scenario (with axi-cel) over 5 years.

Results: In the base case analysis over a lifetime horizon, compared with SC, axi-cel generated 4.86 incremental quality adjusted life year (QALYs) at an incremental cost of S$454,358 (US$326,876), resulting in an ICER of S$93,440 (US$67,223) per QALY gained. The projected annual incremental budget impact ranged from S$2.29 million (US$1.65 million) to S$10.91 million (US$7.85 million) during the first 5 years of introduction of axi-cel.

Conclusions: At commonly accepted ICER thresholds, results of this analysis suggest that axi-cel can be considered a cost-effective allocation of resources with manageable budget impact compared with SC for the treatment of adult patients with R/R LBCL after 2 or more lines of systemic therapy in Singapore.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P097 POSSIBLE EFFECT OF PRE CAR-T RADIOTHERAPY ON TIME-TO-RELAPSE: A MONOCENTRIC EXPERIENCE

Massimiliano Gambella 1, Anna Maria Raiola1, Chiara Ghiggi1, Livia Giannoni1, Alessandra Bo1, Salvina Barra1, Emanuele Angelucci1

1IRCCS Ospedale Policlinico San Martino, Genova, Italy

Background: Metabolic persistence of lymphoma 1 month after CAR-T therapy (PET-1) is challenging: some patients will achieve complete remission thereafter; the majority will progress in an unpredictable time. We sought to analyse the impact of pre CAR-T radiotherapy in patients still with evidence of disease 1 month following treatment.

Methods: Twenty-five consecutive patients affected by aggressive B cell lymphoma, treated at our center from October 2020 to July 2022 with anti-CD19 CAR-T therapy (tisagenlecleucel or axicabtagene ciloleucel) were analyzed. Patients with complete metabolic response or with disease progression at PET-1 were excluded. Involved-field radiation therapy (IFRT) was delivered as part of the bridging-therapy with remission purposes. Criteria for IFRT were mediastinal and/or bulky disease. For each patient, areas of active disease were still present before CAR-T in addition to the irradiated ones. Progression-free survival was calculated from PET-1 thorugh the Kaplan Meier estimator, with progression and death considered as events; patients still in remission at last follow-up were censored; Log-rank test was employed for comparison among groups.

Results: Nine patients fulfilled the pre-specified criteria, 6 with partial remission and 3 with stable disease at PET-1; 5 patients received radiotherapy before CAR-T therapy (2/5 for mediastinal, 3/5 for non-mediastinal bulky masses). Eight of nine (89%) patients progressed following PET-1. Median progression-free survival was 127 days (95% CI: 0 - 444) for patients who received pre CAR-T radiotherapy and 33 days (95% CI: 31.04 – 34.9) for those who did not (p = 0.036). Among the former, one patient is still in remission and two relapsed beyond 6 months from CAR-T; among the latter, all progressed after PET-1, three of whom within 1 month. No significative differences were depicted among the two groups for variables associated with tumor burden or inflammation, as evaluated before CAR-T (Table). Relapse in patients who received radiotherapy occurred in areas other than those irradiat.

Table. Pre CAR-T variables associated with tumor burden or inflammation, comparison among groups.

Variable

No Radiation Therapy

(N = 5)

Radiation Therapy

(N = 5)

Significance

(p)

IPI int-high / high

2/4

2/5

0.7

Bulky ≥ 10 cm

1/4

1/5

0.8

Refractory to last line

3/4

4/5

0.8

B symptoms

1/4

2/5

0.1

LDH > UNL

2/4

4/5

0.3

LDH (U/L), median

235

244

0.9

Ferritin (mcg/L), median

233

310

0.5

CRP (mg/L), median

3,3

4,9

0.5

  1. IPI international prognostic ndex; LDH lactate dehydrogenase; UNL upper normal limit; CRP C-reactive protein.

Conclusions: Almost all patients who still retain metabolically active disease at PET-1 subsequently progress, representing an ultra-high-risk group. Radiotherapy before CAR-T may prolong the progression free survival and consent additional salvage treatments (i.e. bispecific antibodies, allogeneic transplant). We conclude that radiation therapy is a per-se useful strategy to limit progression in relevant areas of disease (metabolically active, bulky masses) and should be systematically integrated in the bridging-strategy. Lack of relapse in irradiated areas also suggests biological interactions with CAR-T activity that should be explored.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P098 CD19 CAR-T-CELL THERAPY IN POST HSCT RELAPSE OF MIXED PHENOTYPE ACUTE LEUKEMIA (MPAL) WITH PHENOTYPIC LINEAGE SWITCH FROM MYELOID TO LYMPHOID CLONE

Aleksandra Oszer 1, Bartłomiej Pawlik1,2, Paweł Marschollek3, Natalia Ćwilichowska4, Julia Kołodrubiec1, Justyna Jakubowska1, Maciej Zdunek1, Karolina Liszka3, Joanna Trelińska1, Szymon Janczar1, Krzysztof Kałwak3, Marcin Poręba4, Wojciech Młynarski1

1Medical University of Lodz, Lodz, Poland, 2Medical University of Warsaw, Warsaw, Poland, 3Wroclaw Medical University, Wroclaw, Poland, 4Wroclaw University of Science and Technology, Wroclaw, Poland

Background: This is questionable whether CD19 CAR-T-cell therapy is effective in mixed phenotype acute leukemia (MPAL) because of the clonal plasticity of the disease. Here we report a case of an 11-years-old girl with MPAL who was successfully treated with CAR-T cells.

Methods: Evaluation of the phenotypic lineage switch from myeloid blasts to lymphoid blasts was shown by flow cytometry. The percentage of CAR T cells was counted at apheresis time, 7, 28, and 60 days after CAR-T cells infusion by cytometry by time of flight (CyTOF), using Maxpar Direct Immune Profiling Assay (Standard BioTools) for deep immunophenotyping and in-house 169Tm-conjugated monoclonal anti-FMC63 scFv antibody (ACROBIOSYSTEMS Inc.). Additionally, quantitative polymerase chain reaction was used to enumerate integrated CAR transgene in CAR-T cells in peripheral blood at leukapheresis time, 7 and 28 days after CAR-T cells infusion. The study was reviewed and approved by the human research ethics committee of the Medical University of Lodz. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. No personal or identifiable data were included in the paper.

Results: MPAL was diagnosed based on immunophenotyping of bone marrow showing 94.7% of monocytic cells with abnormal phenotype: CD45dim+, CD34+, cytMPO+, CD7-, cytCD3-, CD3-,CD19+, CD79a-, HLA-DR+, CD117-, CD10-, CD11b+, CD13+, CD14+, IREM2-, CD15+, CD16-, CD33+, CD36-, CD64+, CD4-, CD56-, CD71-, NG2-, CD41-, CD42b-. Genetic diagnostics revealed TCF3-ZNF384 fusion and FLT3 (NM_004119.3):c.2505T>A (p. Asp835Glu) mutation. The patient was treated with AML-BFM 2019 protocol, subsequently underwent an allogeneic hematopoietic stem cell transplantation but relapsed within 9 months after HSCT with phenotypic lineage switch to lymphoid clone with persistent TCF3-ZNF384 fusion and FLT3 mutation. She started treatment according to IntreALL HR 2010 protocol, achieving cytologic remission within 5 weeks, but with remaining positive PCR-MRD < 5 x 10-4. Then, after leukapheresis she was treated with sorafenib and 6-MP/MTX as a bridging therapy and was subjected to CD19-targeted CAR-T cells (tisagenlecleucel). CyTOF-based immunophenotyping allowed us to count the percentage of CAR-T cells in peripheral blood, which reached 21.14% on day 7 after CAR-T infusion and sustained concentration on day 28 and 60 after CAR-T infusion – 0.05-0.07%. The presence of CAR-T cells in peripheral blood was also confirmed using qPCR. Persistent complete molecular remission at 90 days post-CAR T cells infusion was confirmed (PCR-MRD negative with sensitivity 10-4).

Conclusions: This case study shows an example of a successful CD19 CAR-T-cell therapy for relapsed mixed phenotype acute leukemia (MPAL) with a phenotypic lineage switch from myeloid clone to lymphoid. Further follow-up is needed to confirm the sustained effect of the therapy.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P099 VALIDATION OF DIGITAL PCR-BASED MONITORING OF TISAGENLECLEUCEL EXPANSION AND PERSISTENCE IN CLINICAL SAMPLES - A SINGLE CENTRE EXPERIENCE IN SINGAPORE

Gee Fung How1, Aloysius Ho1, Yeh Ching Linn1, William Hwang1, Jeffrey Quek1, Yunxin Chen1, Lawrence Ng1, Melinda Tan1, Yeow Tee Goh1, Chandramouli Nagarajan1, Shin Yeu Ong1, Christopher Tham1, Jing Jing Lee1, Angeline Tan1, Francesca Lim1, Hein Than 1

1Singapore General Hospital, Singapore, Singapore

Background: Chimeric antigen receptor (CAR) T-cell therapy has now been established in the treatment of relapsed/refractory B-cell acute lymphoblastic leukaemia (ALL) and diffuse large B-cell lymphoma (DLBCL). Durability of clinical response after infusion of tisagenlecleucel (Tisa-cel) has been linked to expansion and persistence of CAR in the blood and marrow samples of patients. Quantification of CAR has been done via flow cytometry or real time quantitative polymerase chain reaction (qPCR) in clinical trials, but is not standardized in real-world clinical practice.

Methods: CAR quantification was performed in the peripheral blood samples collected from adult relapsed/refractory DLBCL patients treated with Tisa-cel in Singapore General Hospital from January 2021 to October 2022. Genomic DNA samples were extracted using QIAamp DNA blood mini kits (QIAGEN). Probe-based assays targeting the junction of the 4-1BB costimulatory domain and the CD3-zeta signaling region of Tisa-cel transgene were developed. The ubiquitously expressed housekeeping CDKN1A gene was used as a control for variability in DNA quality and quantity in each clinical sample. Digital PCR (dPCR) reactions consisting of 150-300 ng DNA as template per reaction were performed on two platforms for comparison: QIAcuity Digital PCR System (QIAGEN) and QuantStudio Absolute Q Digital PCR System (ThermoFisher Scientific). Both systems were plate-based with similar workflows. Copies of transgene and CDKN1A gene were determined in each reaction using the respective system software. Results of dPCR were expressed as ratio percentages of transgene copies to CDKN1A copies, and correlated with flow cytometric data and clinical response.

Results: Forty-six clinical samples longitudinally collected from twelve patients at different time points were available for analysis. CAR transgene percentages detected by the two dPCR platforms, the QIAcuity and Absolute QTM showed an excellent agreement (R2 = 0.998, p < 0.01). Both sets of dPCR results also exhibited a strong correlation with flow cytometric data (R2 = 0.9473 for QIAcuity, and R2 = 0.9771 for Absolute QTM). Longitudinal dPCR results from the two methods showed similar trends with an expansion of CAR-T cells peaking around day 7 post-infusion. Persistence of CAR-T cells beyond 6 months after infusion of Tisa-cel was consistently detected by dPCR in the patients with durable clinical response.

Conclusions: The results from our study show that the current dPCR protocol provides an accurate and reliable means of monitoring kinetics of CAR-T cells. The dPCR platform is potentially superior to real time qPCR which is restricted by the standard curve for its reproducibility, and also to flow cytometry which is limited by the target population size, total event count and sample quality. With further refinement and detailed analytical validation, the dPCR protocol can be established for routine clinical monitoring of CAR-T cells.

Clinical Trial Registry: Not applicable

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P100 BREXUCABTAGENE AUTOLEUCEL THERAPY FOR ADULT PATIENTS AFFECTED BY RELAPSED/ REFRACTORY B-ACUTE LYMPHOBLASTIC LEUKEMIA: THE ITALIAN REAL LIFE EXPERIENCE IN A CONTEXT OF THE NAMED PATIENT USE

Federica Sorà 1, Eugenio Galli2, Elisabetta Xue3, Francesco Restuccia4, Ilaria Pansini1, Sabrina Giammarco2, Idanna Innocenti2, Elisabetta Metafuni2, Fabio Giglio3, Francesco Autore2, Alberto Fresa2, Andrea Bacigalupo1, Patrizia Chiusolo1, Luca Laurenti1, Prassede Salutari4, Fabio Ciceri3, Simona Sica1

1Università Cattolica del Sacro Cuore, Rome, Italy, 2Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy, 3San Raffaele Hospital, Milan, Italy, 4Ospedale Civile Spirito Santo, Pescara, Italy

Background: Relapsed/refractory (R/R) B-cell acute lymphoblastic leukemia (B-ALL) in adult patients is associated with poor prognosis after salvage chemotherapy. Chimeric antigen receptor (CAR) T-cell therapy, approved in the pediatric setting, currently represents one of the most promising immunotherapeutic approaches for hematologic malignancies.

Methods: Since January 2021 10 adult patient affected by B-ALL relapsed or resistant to all lines of therapy according to national current practice were treated with Kite/Gilead Brexucabtagene autoleucel (brexu-cel) in a context of a named patient use in three different italian transplant centres. This is a observational study based on retrospective data collection and analysis approved by the ethical committe.

Results: Their clinical characteristics were collected: median age was 40.5 years (range 32-65), 5 were male, 6 showed a common B phenotype, 6 were Philadelphia positive, median number of prior lines were 3 (range 2-6), and 7 patients had received a previous allogeneic HSCT.

Five out of 10 (50%) patients were refractory to the last treatment at the time of CAR-T eligibility. Patients had been exposed to blinatumomab and inotuzumab before leukoapheresis in 4 (40%) and 5 (50%) cases, two had received both. All patients had CD19+ blasts at the enrolment[EX1]. ECOG performance status score was 0 in 6 patients, and 1 in the others.

All received lymphodepleting Flu-Cy according to ZUMA-3 trial schedule; none experienced progression during bridging therapy and all received CAR-T infusion. Median time from leukapheresis to CAR-T reinfusion was 68 days (range 37-122). Two patients (20%) experienced a manufacturing failure, and a second leukapheresis was then successfully performed.

CRS was observed in 9 patients after 7 days (range 4-9): CRS was usually mild, with grade 1 in 8 patients (80%) and grade 2 in 1 (10%) patient. ICANS was observed in 2 (20%) patients after 7.5 days (range 6-9), graded 1 in one patient and 2 in the other. Tocilizumab was administered in 3 patients and high-dose steroids was required in 1 patient. None developed consumptive coagulopaty, and macrophage activation syndrome (MAS) was recorded in 1 patient. Other adverse events included transient transaminase elevation (n = 2) and deep venous thrombosis (n = 1). None required Intensive Care Unit admission.

Median follow-up [EX3] was 178 days (32-424); nine out of 10 patients were in MRD negative CR at day +30, while one patient progressed before day 30. Of six patients evaluable 6 months after treatment, three were in MRD negative CR, two had progressed and received salvage therapy with blinatumomab, and one had positive MRD with persistent morphological CR, received preemptive TKI and subsequent allo-HSCT. At last FU, one patient deceased due to relapse.

Conclusions: In our series of 10 infused heavily pretreated adult B-ALL patients, brexu-cel was effective and well tolerated.

The incidence of CRS, ICANS, and other toxicities was very low. ORR at day+30 after infusion is very promising, despite a longer follow-up and further study are necessary to confirm these data.

Disclosure: “Nothing to declare”.

4 - CAR-based Cellular Therapy – Clinical

P101 CAR-T THERAPY IN LATIN AMERICA: REALITY AND EXPECTATIONS OF HEMATOPOIETIC STEM CELL TRANSPLANTATION CENTERS

Fernando Barroso Duarte 1, Luciana Maria Barros Carlos2, Nelson Hamerschlak3, Lucila Kerbauy3, Carmem Bonfim4, Marco Aurelio Salvino5, Amado Karduss6, Alvaro Urbano Ispizua7, Manel Juan Otero7, Andrés Gómez De León8, Juan Ramon Navarro Cabrera9, Ana Lisa Basquiera10, Juan Carlos Serrano Casas11, Mariana Stevenazzi12, Talyta Ellen Jesus Santos13

1Hospital Universitário Walter Cantídio, Fortaleza, Brazil, 2Centro de Hematologia e Hemoterapia do Ceara-HEMOCE, Fortaleza, Brazil, 3Hospital Israelita Albert Einstein, Sao Paulo, Brazil, 4Hospital Pequeno Príncipe, Curitiba, Brazil, 5Universidade Federal da Bahia, Salvador, Brazil, 6Instituto de Cancerologia Las Americas-AUNA, Medellin, Colombia, 7Hospital Clínic de Barcelona, Barcelona, Spain, 8Universidad Autónoma de Nuevo León, Hospital Universitario Dr Jose Eleuterio González, Monterrey, Mexico, 9Hospital Edgardo Rebagliati Marints, Jesús María, Peru, 10Hospital Privado Universitario de Cordoba, Cordoba, Argentina, 11Clinica Cancerologica del Norte de Santander, Norte de Santander, Colombia, 12Integral Medical Service, Montevideo, Uruguay, 13Laboratório Central de Análises Clínicas, Fortaleza, Brazil

Background: Recently, CAR-T cell has emerged as a necessary and well-established treatment strategy, especially for refractory haematological diseases. Latin America (LA) comprises 20 countries: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Ecuador, El Salvador, Guatemala, Haiti, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Dominican Republic, Uruguay, and Venezuela and the estimated population is 569 million inhabitants who primarily speak Spanish and Portuguese. A very important aspect is the availability of Hematopoietic Stem Cell Transplantation (HSCT) centres in this region. Currently, there are 16 active HSCT countries and no reports of HSCTs in Nicaragua, El Salvador, Honduras, and Guatemala. In October 2021, the Latin American Bone Marrow Transplantation Group (LABMT) reported a total of 230 centres in LA, with the highest number in Brazil (120) and the lowest in Ecuador (1). In this region, some peculiarities exist, including limited access to unrelated allogeneic HSCT and the availability of paediatric HSCT centres. Due to the complexity of the CAR-T cell procedure, there is a premise that teams with experience in allogeneic HSCT are better prepared to take care of the patients. The chain of this process ranges from lymphocyte collection, cryopreservation, and infusion of CAR-T cells to clinical complications such as Cytokine-Release Syndrome (CRS) or Immune Effector Cell-Associated Neurotoxicity Syndrome (ICANS), besides the need for Good Manufacture Practices experience.

Methods: We conducted an informal poll, including some groups of haematologists and transplant HSCT physicians, about possible CAR-T cell initiatives.

Results: No CAR-T cell development activities exist in Colombia, Venezuela, Peru, and Uruguay. In Mexico, there is one university initiative. In Argentina, three centres are participating in the CARTITUDE-5 study. There are ongoing industry and academic projects in Brazil. In São Paulo, the following initiatives are active: CAR T Prodigy Platform for CD19 haematological diseases, CD19 CAR-T for B cell Lymphomas, and CARTITUDE-5, all at the Albert Einstein Hospital; Hub for CAR-T production Prodigy System at the Butantan Institute; Belinda Phase III trial at the Albert Einstein Hospital and San Raphael Hospital (Bahia). In the Northeast of Brazil (Fortaleza), CAR T Prodigy Platform is in development at the Federal University of Ceara (HEMOCE/HUWC). An Agreement among the Brazilian Society of Bone Marrow Transplantation and Cellular Therapy, LABMT, and the Clinical Hospital of Barcelona was created to improve access to CAR-T cell therapies in this region.

Conclusions: CAR-T cell indications are increasing in LA, but access is still the biggest problem. Initiatives by specialists from medical societies are fundamental for implementing this treatment in countries with limited resources. It is necessary to join efforts to serve our patients better and define specific roles for each centre within this process. These efforts are more relevant in countries with continental dimensions, such as Brazil, where HSCT activity is not evenly distributed across all regions. The emergence of rapidly growing CAR-T cell activity in restricted- resources countries will need support from international HSCT societies to ensure the best use of the limited resources available in this region.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P102 CHALLENGES IN OPTIMIZING CELL COLLECTION FOR MANUFACTURING OF CART-T CELLS IN RELAPSED ACUTE LYMPHOBLASTIC LEUKEMIA

Sanja Mazić1, Ines Bojanić 1, Maja Pavlović1, Koraljka Gojčeta1, Vladimira Rimac1, Milica Liker1, Ernest Bilić1, Branka Golubić Ćepulić1

1UHC Zagreb, Zagreb, Croatia

Background: Chimeric antigen receptor modified T-cells (CAR-T) offer a therapeutic possibility in pediatric relapsed acute lymphoblastic leukemia (ALL). Prerequisite to production of CAR-Ts is a collection of the target number of autologous T-cells that comply with manufacturer’s requirements ( > 1x109 CD3+ cells, >2x109 TNC, >3% CD3+ cells for tisagenlecleucel - Kymriah®, Novartis). However, a leukapheresis procedure in active relapsed disease presents a challenge and here we describe our center’s first experience with such a patient.

Methods: At our pediatric hematology department 10 years old girl was treated for pre-B ALL with translocation 4:11. She received induction phase and 2 high risk (HR) cycles according to ALL IC BFM 2009 protocol. Before HR1 cycle she was in complete remission, and before HR2 she had 1% of leukemic cells in bone marrow. Just before third HR cycle, she presented with hyperleukocitosys (156x109/L) with 71% blasts and thrombocytopenia (43x109/L). Bone marrow aspiration confirmed relapse of the disesase with 81% blasts. Femoral central line was placed before her transfer to apheresis unit. CD3+ percentage in peripheral blood was 1,01% and count 1579x106/L.

Results: Leukapheresis was performed using Spectra Optia, cMNC procedure, AC:WB ratio was increased from 12:1 to 24:1. The patient weighted 44 kg with haemoglobin 92 g/L and hematocrit 0.274 L/L and a blood prime was not required. Due to the hyperleukocytosis, the procedure started with collect pump rate set at 3.0 mL/min and was lowered to 1.0 mL/min after product was sampled. During 180 minutes a total of 7360 mL of blood was processed (2,7 TBV). Product was sampled after 73 minutes (V 180 mL, CD3+ 1,57%, count 5271x106/L, total CD3+ 0,95x109) and 120 minutes (V 280 mL, CD3+ 1,92%, count 5907x106/L, total CD3+ 1,65x109) which estimated that we will be able to reach target numbers. Apheresis procedure was well tolerated and uneventful. Since it was an evening procedure going into a night shift cryopreservation, leukapheresis was completed after 180 min. Final volume of apheresis product was 296 mL, TNC 80,24x109, CD3+ 2,25%, count 6060x106/L, total CD3+ 1,81x109 that was cryopreserved in 4 bags. The product met the target cell numbers for TNC and CD3+cells but the percentage of CD3+ was lower then required. After consultation with Novartis team, apheresis product was accepted for the manufacturing. CAR-T product passed quality control and the patient was infused with 2,6x106 CAR positive viable T-cells. The CAR-Ts were successfully engrafted, but severe case of immune effector cell-associated neurotoxicity syndrome occurred.

Conclusions: Collection of target CD3+ cells is challenging in relapsed pediatric ALL patients due to need to balance the urgency of the collection, delay of chemotherapy in case of disease rapid progression and organisational issues such as central line placement and availability of trained apheresis and cryopreservation staff, all happening, very likely, outside regular working hours. Optimizing cell collection in the case when patient’s peripheral blood is highly burdened with blasts is difficult since the algorithm normally used to determine the target blood volume to process is not reliable in such a setting.

Disclosure: Nothing to declare.

4 - CAR-based Cellular Therapy – Clinical

P103 LYMPHOCYTE APHERESIS IN HEALTHY DONORS FOR VALIDATING CHIMERIC ANTIGEN RECEPTOR (CAR) T CELLS PRODUCTION

Yandy Marx Castillo-Aleman 1,2, Loubna Abdel-Hadi1, John Dearborn Philips1, Yendry Ventura-Carmenate1,2, May Ann Martinez1, Antonio Alfonso Bencomo-Hernandez1, Samira Hasan-Sheikh1, Carlos Agustin Villegas-Valverde1, Rene Antonio Rivero-Jimenez1, Nameer Abdul Raheem Al-Saadawi1,2, Inas El-Najjar1,2, Fatema M. Al-Kaabi1,2

1Abu Dhabi Stem Cells Center, Abu Dhabi, United Arab Emirates, 2Yas Clinic Khalifa City Hospital, Abu Dhabi, United Arab Emirates

Background: The emerging chimeric antigen receptor T cell (CAR-T) therapies require strict validation before they can be introduced into the relevant clinical studies that support their marketing authorization. Lymphocyte apheresis provides the starting material for CAR-T cell manufacturing; therefore, this collection is critical for the success of those studies or treatments in clinical settings. Interestingly, apheresis parameters showed high variability, and clinical guidelines are currently lacking. We report the first lymphocyte aphereses undertaken in healthy donors for CAR-T cell validation studies in the United Arab Emirates performed at Abu Dhabi Stem Cells Center (ADSCC).

Methods: Apheresis procedures were performed on the Amicus Blue™ Separator System running software v6.0 (Fresenius Kabi, Germany) using peripheral venous access. A 12:1 whole blood (WB) to ACD-A anticoagulant ratio was used, with an average citrate infusion rate of 1.25 mg/kg/min, maximum WB draw rate of 55 mL/min, and 8 cycles. The mononuclear cells (MNC) offset was set at 1.5 mL, red blood cells (RBC) offset to 6.8 mL, with a mean of cycle volume of 900 mL, and plasma storage fluid of 50 mL. Lymphocyte collection was considered based on absolute lymphocyte count of peripheral blood cells that was higher than 1,500/μL in nonmobilized, healthy volunteers with normal complete blood counts, coagulation, electrolytes, metabolic, renal, and liver profiles, along with negative infectious disease screening within one week of apheresis. The collection goal was set at a minimum of 1.0 x 109 lymphocytes and 0.6 x 109 CD3+ cells with a target of 3.0 x 109 lymphocytes and 2.0 x 109 CD3+ cells required for the validation trials.

Results: Five healthy male donors (34–61 years old) were eligible for apheresis and signed an informed consent form for CAR-T clinical validation. The procedures were performed between September 2021 and September 2022 by a qualified apheresis team.

The five cell products were each collected in single procedures, received by the ADSCC Stem Cells Processing Laboratory, and met the requirements for manufacturing CAR-T cells. All cell products exceeded the collection target for lymphocyte and CD3+ cell thresholds. Table 1 summarizes the means of peripheral blood/collection cell counts and apheresis parameters.

Table 1. Means of peripheral blood/collection cell counts and procedural parameters (n = 5)

Parameters

Peripheral blood * (range)

Collection (range)

HGB: Hemoglobin (g/dL)

14.7 (13.5–16.6)

2.9 (2.1–3.8)

PLT: Platelets (109/L)

268 (200–356)

1019 (389–2125)

WBC: White blood cells (109/L)

7.9 (5.4–9.2)

67.8 (56.0–79.6)

NE: Neutrophils (109/L)

4.8 (3.0–6.0)

5.4 (1.4–12.2)

LY: Lymphocytes (109/L)

2.3 (1.6–3.5)

45.4 (34.9–54.6)

MO: Monocytes (109/L)

0.5 (0.4–0.7)

16.4 (12.2–18.4)

MNC: Mononuclear cells (109/L)

2.9 (2.0–4.2)

61.8 (50.9–71.6)

Procedure parameters

Mean (range)

 

TBV: Total blood volume (mL)

5000 (4436–6823)

 

WB processed (mL)

6546 (4693–8441)

 

TBV processed (ratio)

1.31 (1.02–1.90)

 

AC used (mL)

518 (336–655)

 

Duration (minutes)

247 (109–356)

 

Product volume (mL)

129 (102–161)

 

LY-CE2 (%)

45.2 (24.0–67.7)

 

LY-FE (ratio)

20.9 (14.9–32.8)

 

LY-CT (mL/min)

13.8 (4.2–22.8)

 

LY collected (109)

5.9 (4.0–8.4)

 

CD3+ T cells collected (109)

4.6 (2.1–7.9)

 
  1. * Pre-apheresis counts
  2. CE2 Collection efficiency 2; FE fold enrichment; CT collection throughput

All donors tolerated apheresis without undergoing any serious adverse events or technical incidents. Prophylactic intravenous calcium gluconate (1,000 mg) was administered in all procedures per institutional protocol with no reports of citrate-related events.

The most common adverse events (CTCAE v.5.0) were grade 1 pain at the needle insertion site in 3/5 donors (60%) followed by grade 1 diminished inlet flow rate (lower than 30 mL/min) in 2/5 donors (40%), grade 1 chills in one volunteer (20%), and grade 1 headache in one donor (20%).

Conclusions: Despite the heterogeneity of procedures and cell products collected from healthy donors, the Amicus Blue™ Separator System showed a favorable safety profile, and the collection efficiency in a single low-volume apheresis was adequate for the CAR-T cell validation studies.

Clinical Trial Registry: Not applicable.

Disclosure: Nothing to declare.

3 - CAR-based Cellular Therapy–Preclinical

P104 INHIBITION OF MITOCHONDRIAL ISOCITRATE DEHYDROGENASE ENHANCES CAR-T CELL IMMUNOTHERAPY BY PROMOTING ANTIOXIDANT ACTIVITY

Mi Shao 1, Xiaohui Si1, Xinyi Teng1, Yue Huang1, Gang Xiao2, He Huang1

1The First Affiliated Hospital of Zhejiang University of Medicine, Hangzhou, China, 2Zhejiang University Medical Center, Hangzhou, China

Background: Chimeric antigen receptor (CAR)-T cells are engineered to initiate immune response to specific antigens that mark target cells. CAR-T cell therapy was approved for clinical practice for its remarkable efficacy particularly in treating B-cell malignancies. However, several obstacles are limiting the long-term efficacy and restraining the application in broader cancer types of CAR-T therapy. Besides antigen escape, poor tumor infiltration and toxicity, CAR-T cell exhaustion and short persistence are major challenges affect treatment success. In addition, nutrient-deprived tumor microenvironment (TME) has been found to render CAR-T cell dysfunction through impairment of metabolic fitness.

Methods: we performed mitochondria-related compound library screen in CAR-T cells and CAR-T cells post killing via expansion assay and flow cytometry. Further analysis were performed to detect cell phenotypes including cell differentiation (CD62L, CD45RO), cell exhaustion (PD-1, LAG-3, TIM-3) and cell apoptosis (Annexin-V). Luciferase-based cytotoxicity assays was employed to evaluate the cytolytic ability. We used NCG mice to validate the antitumor of CAR-T in vivo. Metabolomic and transcriptomic analyses were used to reveal the effect of IDH2 inhibitors on metabolic reprogramming of CAR-T cells.

Results: Firstly, we found that the IDH2 inhibitor enasidenib could significantly increase the proportion of its naïve and central memory population without affecting the proliferation of CAR-T cells by compound screening. Further in vitro experiments demonstrated that enasidenib treatment reduced exhaustion (measured by expression of inhibitory receptors PD-1, TIM-3 and LAG-3) and terminal differentiation of CAR-T cells and increased their killing capacity, both in the co-stimulatory domain of 4-1BB and CD28. Secondly, we further demonstrated in NCG mice that enasidenib-pretreated CAR-T cells exhibited better antitumor function in vivo, as evidenced by lower tumor burden and longer survival in mice. More interestingly, enasidenib pretreatment combined with in vivo gavage administration group showed optimal outcome. What is more, by combined transcriptomic and metabolomic analysis, we found the suppression of IDH2 results in a reduction of the TCA cycle activity and dramatically rewired glucose utilization into pentose phosphate pathway (PPP) that provides anti-oxidant capacity.

Conclusions: We screened out enasidenib, a mutant and wild-type IDH2 inhibitor approved by the US FDA, from 157 mitochondria-related compound library. Enasidenib was able to enhance anti-tumor functionality of CAR-T cells via metabolic reprogramming. Enasidenib treatment during CAR-T cell culture process not only augmented long-lived memory CAR-T cell production, but also promoted survival and sustained cytotoxicity. Our studies also provide insights into the metabolic regulation of CAR-T cells. The PPP plays a pivotal role in CAR-T cell memory formation and persist function. We simultaneously developed a potential combination therapy using IDH2 inhibition and CAR-T cell immunotherapy.

Clinical Trial Registry: Not involved

Disclosure: Nothing to declare.

3 - CAR-based Cellular Therapy–Preclinical

P105 THE CHOICE OF CO-STIMULATION DETERMINES THE FATE OF CD5 TARGETING CAR-NK CELLS

May Daher 1, Rafet Basar1, Nadima Uprety1, Emily Ensley1, Sunil Acharya1, Mayra Shanley1, Pinaki Banerjee1, Ana Karen Nunez Cortes1, Ye Li1, Bin Liu1, Enli Liu1, Vakul Mohanty1, Merve Dede1, Jinzhuang Dou1, Ken Chen1, Richard Champlin1, David Marin1, Elizabeth Shpall1, Katy Rezvani1

1UT MD Anderson Cancer Center, Houston, United States

Background: Relapsed/refractory T-cell malignancies have a particularly poor prognosis and novel therapies are direly needed. CD5 is a great candidate for adoptive cellular therapy to target T-cell malignancies since it is ubiquitously expressed on T cells with restricted expression on other hematopoietic cells. NK cells are an attractive platform for CAR engineering to target CD5 since, unlike T cells, they do not express CD5 on their surface, which eliminates the risk of fratricide. Another advantage of NK cells for CAR engineering is their safety profile; in contrast to T cells, they do not cause cytokine release syndrome (CRS) or immune effector cell-associated neurotoxicity syndrome (ICANS) and they are not associated with graft-versus-host disease (GVHD) in the allogeneic setting, opening the potential for a completely off-the-shelf cellular product to be used at point of care. Therefore, we sought to develop CD5 targeting CAR-NK cells for the treatment of T-cell malignancies.

Methods: Multiple studies have demonstrated that the choice of co-stimulatory domain in CAR-T cells influences their function, persistence and metabolic profile. In the field of CAR-NK cells, the first constructs tested in the clinic have incorporated CD28, a T cell specific co-stimulatory domain, borrowing from the design of CAR-T cells. There is a sparsity of data regarding the impact of co-stimulatory domains on the proliferation, transcriptomic and proteomic profile, polyfunctionality, metabolism and fitness of CAR-NK cells. In this study, we aimed to do so, by comprehensively studying the impact of co-stimulatory domains including some that are more relevant for NK cell biology, namely DNAX-activating Protein 10 (DAP10), a major adaptor protein and the exclusive signaling intermediate of NKG2D in human NK cells, DNAX-activating Protein 12 (DAP12), an important adaptor molecule that associates with multiple activating receptors (e.g. NKG2C, NKp44, activating KIRs) and NKG2D, one of the most potent NK cytotoxicity receptors which is essential for anti-tumor immunity.

Results: Our results show that CD5 CAR-NK cells with DAP10 co-stimulatory domain show enhanced cytotoxicity against CD5+ T-cell leukemia targets even after multiple tumor rechallenges in an Incucyte live-cell imaging assay. They also show augmented polyfunctionality compared to CD5 CAR-NK cells with other co-stimulatory domains in an Isoplexis single-cell secretome assay. Moreover, DAP10 co-stimulation endowed CD5 CAR-NK cells with enhanced metabolic fitness as evidenced by increased oxidative phosphorylation compared to other co-stimulatory molecules. At the epigenetic level, CD5 CAR-NK cells with DAP10 co-stimulation interrogated by sc-ATACseq show enrichment in AP-1 complex and BATF transcription factors related to memory formation and exhaustion resistance. This translates to better in vivo performance as CD5 CAR-NK cells with DAP10 co-stimulatory domain significantly improve tumor control and survival in an NSG mouse model of CD5+ T-cell leukemia (CCRF-CEM) and show evidence of recall response following tumor rechallenge.

Conclusions: In conclusion, our data show that DAP10 co-stimulation induces epigenetic reprogramming of CD5 CAR-NK cells leading to enhanced cellular fitness and memory formation ensuing better anti-tumor potential. Based on these preclinical data, a Phase I/II clinical trial evaluating the safety and efficacy of CD5 CAR-NK cells for the treatment of CD5+ malignancies is in preparation.

Clinical Trial Registry: NCT05110742

https://clinicaltrials.gov/ct2/show/NCT05110742?term=CD5+CAR+NK&draw=2&rank=1

Disclosure: M.Daher, R.Basar, P.Banerjee, E.Liu, E.Shpall, D.Marin, and K.Rezvani and The University of Texas MD Anderson Cancer Center (MDACC) have an institutional financial conflict of interest with Takeda Pharmaceutical for the licensing of CAR-NK cell technology.

3 - CAR-based Cellular Therapy – Preclinical

P106 INHIBITION OF CD38 ENZYMATIC ACTIVITY ENHANCES CAR-T CELL IMMUNOTHERAPEUTIC FUNCTIONS

Yue Huang 1, Mi Shao1, Xinyi Teng1, Xiaohui Si1, Longyuan Wu1, Yongxian Hu1, Pengxu Qian2, He Huang1

1Bone Marrow Transplantation Center The First Affiliated Hospital Zhejiang University School of Medicine, Hangzhou, China, 2Institute of Hematology Zhejiang University, Hangzhou, China

Background: The impressive efficacy of CAR-T therapy for treating acute lymphoblastic leukemia, non-Hodgkin lymphoma, and multiple myeloma leads to FDA marketing approval of CAR-T products. Nonetheless, less than 50% of patients achieve sustainable disease control and others experience inadequate T cell potency to eradicate solid tumor cells. T cell exhaustion hinders the efficacy of CAR-T cells driven by excessive CAR signaling due to a high antigen burden or by constant signaling resulting from the CAR receptor aggregation in an antigen-independent manner. Omics-based analyses of in vitro exhausted T cell models and patient-derived tumor-infiltrating lymphocytes (TILs) have identified a series of genes that regulate T cell exhaustion. Genome-wide CRISPR-Cas9 knockout and knockin screening have also demonstrated potential targets to regulate CAR-T cells’ durability and long-term cytotoxic function.

Methods: We pretreated CAR-T cells with small molecule CD38 inhibitors for 72h after CAR-T cell sorting. Flow cytometry was performed to detect cell phenotypes including cell differentiation (CD62L, CD45RO), cell activation (CD25, CD69), and cell exhaustion (PD-1, LAG-3, TIM-3). Luciferase-based cytotoxicity assay was employed to evaluate the cytolytic ability of CAR-T. We established the B-ALL NSG mice model with the inoculation of 1 × 106 luciferase/GFP/Nalm6 cells by tail vein injection followed by intravenous injection of 1 × 106 mCherry/CAR-T cells 5 days later. Leukemia progression was measured weekly by bioluminescence imaging.

Results: First, we re-analyzed our recently published scATAC-seq data from two patient-derived CAR-T cells at the expansion peak stage and the later declining stage. CD38 is highly expressed together with other widely defined key factors including TOX, CTL4, BATF, and IRF4 in exhausted CD8+ cells, and is positively correlated with exhaustion score. We employed two different in-vitro culture systems to mimic CAR-T exhaustion induced by tonic signaling or tumor antigens. Expression of CD38 was significantly increased both during the natural culture procedure and after two times of consecutive tumor stimulation.

Then we treated CD19-41BBζ, CD19-28ζ, and HA-28ζ CAR-T cells with small molecule CD38 inhibitors. All three inhibitors enable CAR-T cells to maintain the naïve state (Tn) and central memory state (Tcm). CD38 inhibitors also endowed CAR-T cells with lower expression of activation markers CD25 and CD69, and inhibitory receptors PD-1, TIM-3, and LAG-3. When facing an exhaustion-inducing condition (low E:T ratio at 1:10 and multiple rounds of tumor challenge), CD38-inhibited CAR-T cells demonstrated significantly marked enhancement in cytotoxicity.

In a model wherein NSG mice were inoculated with Nalm6-GFP leukemic cells and 5 days later infused with CD38-inhibited CD19-41BBζ CAR-T cells, we observed enhanced tumor control and prolonged survival. The superiorly expanded CAR-T cells were endowed with fewer exhaustion characteristics, representing diminished co-expression of multiple inhibitory receptor markers. Finally, we found that the CD38 inhibitor could encounter CAR-T cell exhaustion and improve therapeutic efficacy via CD38-cADPR-Ca2+ signaling downregulation.

Conclusions: We present evidence that CD38 may serve as a potential molecular target for regulating CAR-T cell exhaustion. CD38 inhibition encounters CART cell exhaustion and significantly boosts the efficacy of CAR-T cells against hematological malignancies in vitro and in vivo, which constitutes a promising option for CAR-T function enhancement.

Disclosure: Nothing to declare.

3 - CAR-based Cellular Therapy – Preclinical

P107 GENERATION OF EX VIVO AUTOLOGOUS HEMATOPOIETIC STEM CELL-DERIVED T LYMPHOCYTES FOR OFF-THE-SHELF CANCER IMMUNOTHERAPY

Kajornkiat Maneechai1, Wannakorn Khopanlert1, Panarat Noiperm1, Phakaporn Udomsak1, Pongtep Viboonjuntra1, Jakrawadee Julamanee 1

1Prince of Songkla University, Hat Yai, Thailand

Background: Hematopoietic stem cells (HSCs) contain self-renewal and differentiative capacities. The hematopoietic stem cell transplantation is currently a standard of care to cure various malignancies. Recently, adoptive T-cell therapy, especially anti-CD19CAR T-cell which demonstrated promising results from various clinical trials, becomes an alternative curative approach for B-cell malignancies. Despite widespread enthusiasm for CAR-T cells, there are still unmet needs in the current autologous CAR-T cell therapy such as the poor T-cell function in heavily-pretreated patients which expresses exhausted phenotypes or inadequate T-cell number collection for therapeutic purposes. To overcome such limitations, we investigated the feasibility of “off-the-shelf” HSC-derived T-cells as an alternative T-cell source for CAR-T cell-based immunotherapy.

Methods: The autologous mobilized peripheral blood HSCs from myeloma or lymphoma patients were collected and isolated for CD34+ and CD3+ cells. CD34+ cells were cultured using a StemSpanTM T-cell generation kit composed of 2 steps: lymphoid progenitor differentiation phase (day 0–14) and T-cell progenitor maturation phase (day 15–42). On day 42, progenitor T-cells were stimulated with anti-CD3/CD28 beads and cultured with IL-2 to induce the final stage of differentiation and functional maturity. HSC-derived T-cells (HSC-T) were harvested on day 51 for downstream functional assays to compare with pre-isolated CD3+ cells (CD3-T). T-cell viability was assessed by trypan blue staining and the immunophenotypes were analyzed using flow cytometry at indicated time points. The CD3-derived and HSC-derived CD19CAR-T cells were also generated to assess T-cell applications.

Results: HSC-derived T-cells were successfully generated using HSCs from one myeloma and two lymphoma patients. Notably, robust HSC-T cell generation was observed for 3735 times fold expansion (Figure 1A). The purity of isolated CD34+ and CD3+ before expansion were 91.8% and 86.1%, respectively. After 14 days of culture, CD34+ cells significantly decreased; whereas, CD5+ and CD7+ cells dramatically increased (65–84%). CD3+ cells were significantly matured for 40% and 90% on days 42 and 51, sequentially (Figure 1B). The CD8:CD4 ratio was 2:1 and the naïve T-cell phenotype was the majority population of HSC-T (73%) compared to CD3-T cells (34%). However, TIM-3+ cells were predominantly observed in HSC-T cells. To evaluate T-cell functions, T-cells were stimulated with anti-CD3/CD28 beads. Using intracellular cytokine staining, we observed higher IFN-γ, IL-2, granzyme B, and perforin in HSC-T compared to CD3-T cells (Figure 1C). For T-cell proliferation, we did not observe the difference among T-cell types after culture for 2 weeks. To assess cancer immunotherapy application, HSC-derived and CD3-derived CD19CAR-T cells were generated. The preliminary data demonstrated the higher cytokine production capacity of HSC-derived CD19CAR-T cells after being stimulated with CD19-K562 cells. In terms of cytotoxicity, we observed a trend of superior specific cytolysis against primary mantle cell lymphoma and NALM-6 cells in HSC-derived compared to CD3-derived CD19CAR-T cells at all E:T ratios. In the chronic antigen stimulation assay, we did not observe the difference in T-cell proliferation, T-cell subsets, or T-cell exhaustion phenotypes among CAR-T cell types.

Conclusions: HSC-derived T-cells were successfully generated with preserved T-cell efficacy. The autologous HSC-T cells are a potential off-the-shelf cellular product for cancer immunotherapy.

Disclosure: Nothing to declare.

3 - CAR-based Cellular Therapy – Preclinical

P108 BLOCKADE OF CD95/CD95L PATHWAY ENHANCES CAR T CELL PERSISTENCE AND ANTI-TUMOR EFFICACY IN VITRO AND IN VIVO

Bailin He1, Marina Scheller1, Lei Wang1, Brigitte Neuber1, Anita Schmitt1, Susanne Hofmann1, Carsten Müller-Tidow1, Michael Schmitt 1

1Heidelberg University Hospital, Heidelberg, Germany

Background: CAR T cell persistence remains a major clinical challenge. Activation-induced cell death (AICD) is a programmed cell death caused by the interaction of CD95 and CD95L. Through specific blocking of the CD95-CD95L pathway, the CD95L inhibitor APG101 (asunercept®, obtained by Apogenix AG, Heidelberg) could prevent activated T cells from AICD. Therefore, we evaluated whether a blockade of the CD95L pathway through APG101 can improve CAR T cell persistence and enhance antitumor efficacy.

Methods: Human CAR T cells generated from were co-cultured with tumor cells at a 1:1 E:T ratio (round I) in the presence of APG101. Additional tumor cells were supplied to the co-culture every 24 hours. After 3 rounds (72 hr) of stimulation, tumor cells (CD3-CD19+) and CAR T cells (CD3+CD19-) were harvested for FACS analysis. To assess the antigen-induced CAR T cell proliferation, CAR T cells were preloaded with Cell Trace Violet cytosolic dye and cocultured with tumor cells for 72 hours. NSG mice were inoculated with CD19+ Nalm6 cells 0.5 x 105/mouse and an equal number of CD19.CART cells, followed by treatment of mice by APG101: 1mg/mouse, twice a weeks, i.v.

Results: Significant AICD of CAR T cells was observed after repeated antigenic stimulation, accompanied by an increased CD95L expression. CD4+ CAR T cells were more susceptible to AICD compared with CD8+ CAR T cells, although there was no difference in the expression of CD95L between CD4+ and CD8+ CAR T cells. Interestingly, addition of APG101 significantly inhibited CD95L expression and resulted in a lower level of CAR T cell death. Importantly, APG101 did not hamper the activation and proliferation of CAR T cells but was able to restore CAR T cell viability. The expression of PD1, TIM3 and LAG3 were also up-regulated after successive stimulation, however, their expression on CAR T cells were not influenced by APG101. After 3 days of co-culture, the number of CAR T cells were increased in the presence of APG101 (7.9 x 105 vs 6.0 x 105, P = 0.01) and residual tumor cells were dramatically reduced (1.7 x 105 vs 2.7 x 105, P = 0.02). Of note, APG101 itself showed no impact on CAR T cells or tumor cells when cultured separately. Moreover, the central memory CAR T (TCM) cell subset showed higher CD95L expression after coculturing which could be inhibited by APG101. Therefore, the addition of APG101 to the coculture resulted in a significant accumulation of TCM subset after APG101 treatment. In mice, addition of APG to CAR T cell treatment of Nalm6 inoculated mice resulted in a significantly better survival of the animals: all mice receiving solely CAR T cells were dead by day 32, whereas 50% of the animals obtaining additionally APG101 survived day 60 (P < 0.01).

Conclusions: Upregulation of CD95L after repeated antigen stimulation was reversed by APG101. CD95L blockade enhanced CAR T cell survival and promoted killing of tumor cells in vitro. Combining CAR T cell therapy with CD95L inhibitor improved CAR T cell persistence in vivo and thus enhanced the effect of CAR T cell therapy.

Disclosure: Asunercept®, obtained by Apogenix AG, Heidelberg.

3 - CAR-based Cellular Therapy – Preclinical

P109 SUPERIOR ANTI-TUMOR EFFICACY OF CD4+ CAR-T CELLS IN HEMATOLOGICAL MALIGNANCIES

Qian Chen 1, Lei Wang1, Min Lin1, David Sedloev1, Hao Yao1, Marina Scheller1, Volker Eckstein1, Stefanie Hofmann1, Huixiu Han1, Yuntian Ding1, Anna Keib1, Genqiao Jiang1, Dominik Hoffmann1, Hannes Heilmann1, Julia Unglaub1, Anita Schmitt1, Carsten Müller-Tidow1, Michael Schmitt1, Tim Sauer1

1Heidelberg University, Heidelberg, Germany

Background: Chimeric antigen receptor T-cell (CART) therapy has shown promising therapeutic efficacy in several hematological malignancies. However, current CART therapies still face major hurdles leading to primary or secondary resistance and relapse. The optimization of CART treatment requires a better understanding of the characteristics of the administered cell products, including the appropriate ratio between CD4+ and CD8+ T-cell subsets. In most reported trials, CART products contain a random, donor-dependent composition of CD4+ and CD8+ T cells, with only a few clinical trials implementing products with a defined 1:1 ratio of CD4+ and CD8+ T cells which were proposed to be the most beneficial composition. The aim of our study was to further investigate the functionality of CD4 and CD8 expressing CARTs targeting different hematologic malignancies.

Methods: Healthy donor derived, activated T cells were transduced with CAR constructs targeting CD33 (CD33.CD28.41BB.zeta), CD19 (CD19.CD28.41BB.zeta) and CD70 (CD27.zeta) using a gamma retroviral vector system. Fluorescence Activated Cell Sorting (FACS) was used to purify CD4+ and CD8+ CARTs before they were mixed at various CD4/CD8 ratios and co-cultured with target antigen expressing tumor cell lines. Serial re-stimulation with tumor cells was performed every 72 hours[TS1]. Cytokine secretion was measured by flow cytometry (LEGENDplex kit) and CART proliferation using CSFE staining. For in-vivo studies, NSG mice were injected with luciferase expressing tumor cells, that could be detected by bioluminescence imaging, and were subsequently treated with CART products containing different CD4/CD8 ratios.

Results: In our serial co-culture screen, we observed an incremental anti-tumor-efficacy of all CART products with increasing CD4/CD8 for all ratios which was mainly related to an enhanced proliferative capacity of CART products with higher CD4 + T cell content. Indeed, pure CD4+ CARTs exhibited a superior anti-tumor efficacy compared to a mixture of CD4+ and CD8+ CARTs even though the mixed product contained a higher total number of T cells suggesting that the addition of CD8+ CARTs even diminished the functionality of CD4+ CARTs. On the other hand, CD4+ CARTs could improve the proliferation of CD8+ CARTs, in particular at high CD4/CD8 ratios. We measured cytokine secretion upon antigen stimulation and found that CART products with increasing CD4/CD8 ratios secreted higher levels of Th1 and Th2 cytokines compared to products containing a higher percentage of CD8+ CARTs, namely IL-2, TNF-α, IL-4, IL-6, and IL-10. Further mechanistic studies revealed that cytokines secreted by CD4+ CARTs could enhance the functionality of CD8+ CARTs whereas the coapplied CD8+ CARTs hampered the cytolytic and proliferative potential of CD4+ CARTs. Using NSG xenograft models, we confirmed the superior anti-tumor efficacy of pure CD4+ over mixed CD4+/CD8+ or CD8+ CART products in-vivo.

Conclusions: Our data suggest that the CD4/CD8 ratio of CART products is crucial for their functionality and that CD4+ containing T cell products have superior anti-tumor efficacy over CD8+ CARTs. Our results warrant further exploration of pure CD4+ CAR T-cell products within future clinical trials.

Disclosure: Nothing to declare.

3 - CAR-based Cellular Therapy – Preclinical

P110 EXTRACORPOREAL PHOTOPHERESIS CONSTITUTES A PROMISING STRATEGY FOR THE TREATMENT OF GVHD AFTER CAR-T CELL THERAPY

Huixiu Han 1, Lei Wang1, Brigitte Neuber1, Yuntian Ding1, Min Lin1, Hao Yao1, Qian Chen1, Bailin He1,2, Sanmei Wang1,3, Tim Sauer1, Maria-Luisa Schubert1, Alexander Kunz1, Angela Hückelhoven-Krauss1, Anna Keib1, Christian Kleist1, Carsten Müller-Tidow1,4, Peter Dreger1,4, Michael Schmitt1,4, Anita Schmitt1

1Heidelberg University, Heidelberg, Germany, 2Southern Medical University, Guangzhou, China, 3Nanjing Medical University, Nanjing, China, 4German Cancer Consortium (DKTK), Heidelberg, Germany

Background: Anti-CD19 chimeric antigen receptor transduced T (CAR-T) cell therapy has attracted much attention in the treatment of CD19+ leukemia or lymphoma. Next to their potent antitumor activity, however, CAR-T cells give rise to significant side effects as well. Besides the well-known adverse events such as cytokine release syndrome (CRS) and immune-effector cell-associated neurotoxicity syndrome (ICANS), graft-versus-host disease (GvHD) also occurred in 10-30% of the patients receiving “autologous” CAR-T cells after preceding allogeneic transplantation due to the substantial amount of untransduced and alloreactive T cells.

Extracorporeal photopheresis (ECP), a cell-based photoimmunotherapy, shows very promising clinical outcomes in the treatment of GvHD and organ rejection. Since it selectively modulates alloreactive T cells without hampering anti-tumor and anti-virus effects, an interesting question arises whether ECP might constitute a new way to treat patients with GvHD after CAR-T cell therapy without having a negative influence on CAR-T cell function.

Methods: The 3rd generation CD19-specific CAR-T cells were generated using healthy donor PBMCs. To establish an in vitro ECP protocol according to ECP treatment the BIO-LINK crosslinkerTM was used and for this treatment medium, cell concentration and UV-A dose were investigated.

The phenotype of CAR-T cells after ECP was evaluated by a multi-color flow cytometry. Moreover, the quantity and quality in terms of cell apoptosis, killing capacity, cytokine release, proliferation and persistence were assessed in CAR-T cells after ECP in two different models. Besides a “non-dilution model“ (NDM) mimicking the situation of GvHD in the early period after CAR-T cell infusion, a ”dilution model“ (DM), i.e., CAR-T cells were diluted with B cell depleted auto-PBMCs, was established to imitate GvHD at a later time period.

Results: Principal components analysis based on cell components and the inhibitory marker expression showed no distinct difference between different UV-A dosages, while cell viability decreased in an UV-A dose dependent manner. Since 2 J/cm2 is used clinically, this UVA dose is chosen for the ECP treatment protocol.

In the NDM, ECP hampered short-term cytotoxicity of CAR-T cells. The average killing efficiency decreased from 79% to 53.9%. Moreover, proinflammatory cytokines decreased significantly 24 hours after the ECP. The IL-6 decreased from 151 ± 55 to 82 ± 15 pg/mL. The IL-17A decreased from 87 ± 41 to 34 ± 20 pg/mL.

In contrast, in the DM, CAR-T cells diluted with auto-PBMCs maintained a high CAR-T cell viability after ECP, when compared to the NDM (76.4 ± 0.8 in DM and 69.6 ± 3.5 in NDM). Moreover, ECP in the DM had an immunemodulating effect by increasing cytokines as IL2 and IL10 with immunemodulating function and decreasing the proinflammatory cytokines such as IL-6 and IL-17A without negative effect on the short- and long-term cytotoxicity of CAR-T cells.

Conclusions: Our data suggest that ECP constitutes an effective promising treatment strategy for patients suffering from GvHD also in the situation after allo-HSCT and CAR-T cell transfusion, as ECP does obviously not affect the quantity and quality of CAR-T cells.

Disclosure: This study was financially supported by Mallinckrodt Pharmaceuticals through an external collaborative research grant to A.S and M.S.

3 - CAR-based Cellular Therapy – Preclinical

P111 PROTEASOMAL INHIBITION SENSITIZES AML CELLS TO CAR NK-CELL MEDIATED KILLING

David Sedloev 1, Qian Chen1, Dominik Hoffmann1, Anna Keib1, Anita Schmitt1, Michael Schmitt1, Carsten Müller-Tidow1, Tim Sauer1

1University Clinic Heidelberg, Heidelberg, Germany

Background: The prognosis of patients with relapsed and refractory acute myeloid leukemia (AML) remains dismal. Natural killer (NK) cell-based immunotherapy is emerging as a promising strategy for cancer treatment. However, in AML, patients’ responses to activated allogeneic NK cells have been only moderate.

Proteasome inhibitors (PIs), such as Bortezomib (BTZ) and Carfilzomib (CFZ) have been shown to downregulate the expression of HLA class I molecules and increase the presentation of NKG2D ligands on tumor cells, hereby potentially enhancing the functionality of NK cells.

The goal of this study was to determine the potential of BTZ and CFZ to improve the susceptibility of AML cells to NK-cell mediated killing.

Methods: After establishing their IC50 concentration against a panel of AML cell lines, the effects of BTZ and CFZ treatment on the expression of HLA class I molecules and stress-induced proteins by AML cells were determined using multiparameter flow cytometry. Activated NK cells derived from healthy donors were retrovirally transduced with two AML-specific CARs (CD33/CD70). AML cells were treated with BTZ and CFZ either prior to or simultaneously with the addition of CAR- or non-CAR-transduced NK cells. We measured NK-cell mediated short term toxicity against PI treated target cells using a 24-hour co-culture assay as well as their long-term proliferation and killing capabilities in a serial co-culture assay.

Results: Ten of twelve cell lines measured, as well as all six primary AML samples were highly susceptible to proteasomal inhibition, with IC50 values in the low nanomolar range. Resistance to Venetoclax/Azacytidin did not confer resistance to proteasome inhibitors. BTZ and CFZ treatment at the IC50 concentration decreased HLA class I molecule expression by a median of 34% and increased NKG2D ligand and death receptor expression in the majority of tested AML cell lines. We show that these effects are mediated by different mechanisms and do not occur concurrently or peak at the same dose. Pre-treatment of AML cells with PIs significantly enhanced the anti-tumor efficacy of NK cells against AML cells in a short-term cytotoxicity assay (p < 0.001). Transduction of NK cells with AML-specific CARs further improved their functionality in vitro, reducing the number of surviving targets by over 80%, as compared to non-CAR transduced, but activated NK cell controls[SD3]. The combinatorial treatment of PIs and CAR-NK cells effectively eliminated AML cell lines resistant to Venetoclax/Azacitidin treatment, as well as primary AML samples. We investigated the mechanism of action of proteasomal inhibitors in AML and found that both NFkB inhibition and cFLIP downregulation as well as transcriptional upregulation of ULBPs are responsible for the enhanced tumor elimination.

Conclusions: Pre-treatment of AML cells with PIs significantly sensitizes them to NK-cell mediated killing and CAR-expression further enhances the anti-tumor efficacy of activated NK cells. AML cell lines, refractory to conventional chemotherapy treatment are still susceptible to both proteasomal inhibition and NK-mediated killing. Studies aiming to confirm these promising results in-vivo are ongoing.

Disclosure: No conflict of interest disclosed.

3 - CAR-based Cellular Therapy – Preclinical

P112 PD-1 CHECKPOINT INHIBITION IMPROVES FUNCTIONALITY OF TOLEROGENIC CD19-CAR-INKT CELLS AGAINST CD19 + MALIGNANCIES

Emmanuelle Moraes Ribeiro1, Anton Wesle1, Kathy-Ann Secker1, Anita Schmitt2, Hildegard Keppeler1, Corina Schneidawind1, Michael Schmitt2, Dominik Schneidawind 1

1University Hospital Tübingen, Tübingen, Germany, 2University Hospital Heidelberg, Heidelberg, Germany

Background: Invariant natural killer (iNKT) T cells are a small T-cell population with robust immunoregulatory and antitumor properties. We recently demonstrated that human culture-expanded iNKT cells hinder allo-T-cell activation and proliferation while exerting tumor control. Hence, iNKT cells expressing a chimeric antigen receptor (CAR) are interesting cytotherapeutic candidates to treat relapse after allogeneic HCT since they also prevent GVHD. Here, we focused on evaluating the immunoregulatory and cytotoxic properties of CD19-CAR-iNKT cells in combination with the PD-1 inhibitor nivolumab to further enhance functionality.

Methods: iNKT cells were immunomagnetically isolated from PBMCs, transduced with an anti-CD19-CAR retrovirus and expanded in vitro. Cytotoxicity and proliferation of CD19-CAR-iNKT cells were assessed by flow cytometry, image stream analysis and multiplex analysis in single- or repeated-stimulation assays. Moreover, immunoregulatory properties of CD19-CAR-iNKT cells were analyzed in apoptosis and transactivation assays and in mixed lymphocyte reactions (MLR). The effect of checkpoint inhibition through nivolumab was analyzed in both settings.

Results: Robust cytotoxicity could be observed against both CD19 + CD1d- and CD19 + CD1d+ engineered K562 cells. Also, acute lymphocytic leukemia (ALL) and Burkitt lymphoma (BL) cells were highly susceptible to CD19-CAR-iNKT cells. Cell interactions and apoptosis of cancer cells could be confirmed by image stream analysis. Multiplex analysis revealed a robust release of cytotoxic factors such as IFN-γ and granzyme A. The interaction with BL cells induced a higher expression of PD-1 by CD19-CAR-iNKT cells in comparison to non-transduced iNKT cells. CD19-CAR-iNKT-cell function was impaired when challenged with lymphoma cells engineered to overexpress PD-L1/2, which could be partially reversed through nivolumab-treatment. Alloreactivity assays revealed that CD19-CAR-iNKT cells also maintain their ability to induce DC apoptosis and hinder allo-T-cell activation and proliferation even upon treatment with nivolumab.

Conclusions: We showed that CD19-CAR-iNKT cells promote effective lysis of CD19+ tumor cells through their CAR while preventing alloreactive T cell responses. Checkpoint inhibition may further enhance their cytotoxic activity without exacerbating the risk of GVHD after allogeneic HCT making them an ideal cytotherapeutic to treat relapse in this challenging clinical setting.

Disclosure: No conflict of interest to disclosure

3 - CAR-based Cellular Therapy – Preclinical

P113 INDIGENOUS DEVELOPMENT OF NEXT-GENERATION CAR MOLECULES TO OVERCOME ANTIGEN LOSS AND LIMITED IN-VIVO PERSISTENCE OF CONVENTIONAL CAR-T CELLS, CHALLENGES AND SUCCESSES: INITIAL EXPERIENCE FROM INDIA

Gaurav Kharya 1,2, Sivaprakash Ramalingam3, Aditya Iyer3, Pragya Gupta3, Vinod Sarvanakumar3, Md Imam Faizan4, Areej Akhtar4, Iqbal Azmi4, Sufyan Ansari4, Tanveer Ahmed4

1Centre for Bone Marrow Transplant & Cellular Therapy, Indraprastha Apollo Hospital, Delhi, India, 2Cellogen Therapeutics Pvt Ltd, Delhi, India, 3CSIR – IGIB, Counsel of Industrial and Scientific Research – Institute of Genomics and Integrative Biology, Delhi, India, 4Multidisciplinary Center for Advanced Research and Studies, Jamia Millia Islamia, Delhi, India

Background: CD19 directed chimeric antigen receptor (CAR)-T cells have shown impressive results against B-cell related malignancies. However, CD19 antigen loss (30 - 70% of B-ALL patients; Maude SL, et al. N Engl J Med. 2018) and lower persistence of conventional 2nd generation CAR T cells is a common mechanism leading to relapse observed in a subset of patients.

Methods: To overcome these underlying limitations, we have designed and functionally screened 42 CARs consisting of a single (mono) or combinatorial (tandem) scFVs ectodomains (CD19, CD20 and/or CD22) with co-stimulatory domains based on CD28 and/or 4-1BB and/or ICOS and/or OX40 in various combinations. We screened these CAR designs in T cells to find CARs with optimal anti-tumor killing activity.

Results: Based on this initial screening, we found ICOS/4-1BB co-stimulatory domains with tandem CD20/CD19 or CD22/CD19 exhibit the most robust anti-tumor activity against CD19/20/22 triple positive cells as well as cells with either CD19 negative, CD20 negative or CD22 negative antigens, mimicking in vivo antigen loss. Upon comparing with the conventional 2nd generation CARs with either 4-1BB or CD28, we found that ICOS/4-1BB based co-stimulatory domains CAR-T (CAR-TICOS+4-1BB) cells (either CD20/19 or CD22/19 scFv ectodomain) have comparable effector and activation function but persisted longer, with better proliferation rate, and cytokine response. Notably, CAR-TICOS+4-1BB displayed an increased memory phenotype than conventional CARs, a feature critical for the long-term in-vivo persistence of these cells. Currently, these CAR-TICOS+4-1BB with bi-specific or tri-specific tandem ectodomains are being evaluated in-vivo in xenograft models which also mimic the respective tumor antigen loss.

Conclusions: Our pre-clinical work demonstrates that bi-specific (20-19/22-19) tandem CAR-TICOS+4-1BB displayed better proliferation and persistence. Owing to its bi-specificity, they are expected to mitigate tumor relapse by antigen loss/escape as well. Based on our ongoing in-vivo work, the best performing construct shall be taken forward to Phase-1 trial. This is the 1st known indigenous bi-specific 3rd generation CAR molecule development from India.

Clinical Trial Registry: This work was funded by Cellogen Therapeutics Pvt ltd (Cell & Gene Therapy Start-up).

Disclosure: Nothing to declare.

3 - CAR-based Cellular Therapy – Preclinical

P114 CHANGES IN PHENOTYPE OF T CELLS / CD19 CAR-T CELLS IN THE COURSE OF PATIENT TREATMENT

Jonas Wißkirchen 1, Anna-Lena Seebald1, Diana Wolff1, Beate Hauptrock1, Matthias Theobald1, Eva Wagner-Drouet1

1Universitätsmedizin Mainz, Mainz, Germany

Background: Relapse after CAR-T-cell therapy occurs in almost half of all patients and can be divided into 2 modes: CD19-negative and CD19-positive relapse. Mechanisms for the latter remain unclear.

High proliferative and cytotoxic capacity of CAR-T-cells are fundamental for tumor control and may correlate with differentiation status and expression of exhaustion markers.

We examined initial T-cells and CAR-T-cells to determine how differentiation status and expression of exhaustion markers influence clinical outcome.

Methods: We analyzed blood samples from 22 lymphoma patients before lymphapheresis, 1 and 2 weeks after CAR-T-cell infusion and a backwash of the CAR-T-cell bag by flow cytometry.

T-cells were divided into CD4 and CD8 subsets as well as naive (CD45RA + CCR7 + ), central memory (CD45RA-CCR7 + ), effector memory (CD45RA-CCR7-) and effector (CD45RA + CCR7-) T-cells.

T-cell exhaustion was examined by expression of CD160, LAG-3, PD-1, TIGIT and TIM-3.

Results: 20 patients suffered from DLBCL, 2 had mantle-cell lymphoma. 12 received axi-cel, 8 tisa-cel and 2 brexu-cel. Before apheresis CD45RA + CCR7- T-cells were the major subset counting for > 50% of all T-cells in most patients. Within the final product almost all T-cells were CD45RA-CCR7- effector memory type (CAR-T-cells and non-CAR-T-cells). At d + 7 about 20% of CAR-T-cells are CD45RA + CCD7- and this increased further on d + 14. At apheresis in all patients less than 20% of T-cells expressed TIM-3. At least 60% of CAR-T-cells obtained from the bag expressed TIM-3. Expression of PD-1 and TIGIT was more prevalent on T-cells before apheresis but here we also observed significant increases especially within the CAR-T-cell population.

There was no significant difference between the different CAR-T-cell products, patient age or sex. Previous chemotherapy had no influence on the initial T-cell phenotype (no patient received Bendamustine).

In 21 patients we had sufficient data to compare CAR-T-cell frequency in patients with and without relapse respectively. Patients without relapse had a greater percentage of CAR-T-cells of their total T-cells at all timepoints (figure 1).

We observed a trend of higher expression of some checkpoint receptors in patients with relapse.

Patients with relapse had a higher expression of LAG-3 before apheresis (24,4% vs 15,1%) and on day

+8 (23,6% vs 9,4%) on their CD4 T-cells. Expression of TIM-3 and TIGIT was increased on both CD4

(71,0% vs 59,5% and 89,1% vs 83,1 respectively) and CD8 T-cells (86,2% vs 76,1% and 84,9% vs 67,8% respectively) in patients with relapse.

figure 1

Conclusions: We observed that the production process of CAR-T-cells itself causes a change from predominately CD45RA + CCD7- effector T-cells to effector memory T-cells. In vivo some of these develop into terminally differentiated effector T-cells.

Our data suggests a greater percentage of CAR-T-cells in the product and a better expansion of CAR-T-cells in patients without relapse. This may be the consequence of the differential upregulation of 3 checkpoint receptors and downregulation of two other checkpoint receptors.

Further phenotyping for these checkpoints may lead to prediction of clinical outcomes, or new therapeutic options in the course of CAR-T-cell therapy or optimizing the production process.

Disclosure: Advisory Board Celgene and Kite/Gilead.

3 - CAR-based Cellular Therapy – Preclinical

P115 A PERSONALIZED BIOMARKER-TARGETED CAR-T THERAPY IS POSSIBLE BASED ON IMMUNOPHENOTYPE FOR ACUTE MYELOID LEUKEMIA PATIENTS

Meiwei Gong1, Hui Wang 1

1Hebei Yanda Lu Daopei Hospital, Langfang, China

Background: Chimeric antigen receptor T-cell (CAR-T) therapy has resulted in remarkable efficacy for certain patients with hematological malignancies. However, it is difficult to find an high-expression biomarker with good specificity for acute myeloid leukemia (AML) due to the strong heterogeneity of AML tumor cells. Here, we explored biomarker expression among AML patients to facilitate a personalized target CAR-T therapy approach.

Methods: From August 1, 2022 to November 12, 2022, of 671 AML patients tested for minimal residual disease (MRD) by flow cytometry (FCM) at the Hebei Yanda Lu Daopei Hospital, 174 were MRD positive. Among these MRD+ cases, there were 98 males and 76 females. The median age was 32 years (range: 1-76 years) and the median tumor blast proportion was 7.85% (range: 0.01%-97.83%). Biomarkers analyzed included CD34, CD117, CD45, HLA-DR, CD33, CD13, CD11b, CD7, CD56, CD38, and other biomarkers were selected according to the patient’s subtype. In some cases, CD96, CD19, CD371(CLL1), LILRB4(CD85k), CD123, CD15, CD64, CD14, Tim-3,were also tested. Expression intensity of tumor cells was defined as follows: Bright = Score 5. Medium intensity expression = Score 4; Weak expression = Score 3, Partial bright/ partial expression =Score 2; Part dim = Score 1; No expression = Score 0. Each marker has its own positive control.

Results: The overall expression rate and the rate of patients with expression intensity Score ≥3 for CD7, CD38, CD371, CD123, CD96, LILRB4, Tim3 and CD19 were 36.21%(63/174)/16.09%(28/174), 96.27%(155/161)/93.78%(151/161), 88.35%(91/103)/65.04%(67/103), 94.11%(96/102)/90.2%(92/102), 55.28%(68/123)/44.72%(55/123), 24.51%(25/102)/12.75%(13/102), 72.28%(73/101)/48.51%(49/101), and 11.21%(13/116)/5.17%(6/116).

We assumed that patients with expression Score ≥3 could receive CAR-T therapy. In terms of expression intensity, the sequence was CD38, CD123, CD371, tim3, CD96, CD7, LILRRB4, and CD19. However, CD38 and CD371 were expressed in various hematopoietic cells, may lead to greater off target effects and side effects. targeting CD123 is not likely to lead to high efficacy because of its weak expression intensity. If considering using dual targeted CAR-T for AML, the expression rates of CD7/CD38, CD7/CD371, CD7/CD123, CD38/CD123 and CD371/CD123 were 90.80% (158/174), 45.40% (79/174), 36.78% (67/174), 98.78% (163/165), 95.19% (99/104), respectively. Combining these dual targeted CAR T therapies with CD96/Tim-3/CD85k/CD19 CAR-T, the expression rate could be further improved.

Conclusions: Combined with the advantages of multi-parameter flow cytometry immunotyping, we expect that AML can be divided into various subtypes according to the expression of potential CAR-T targets and select individualized CAR-T.

Clinical Trial Registry: no.

Disclosure: Nothing to declare.

3 - CAR-based Cellular Therapy – Preclinical

P116 NK CELLS ENGINEERING USING ADVANCED LENTIVIRAL VECTORS QUALITY AND DESIGN

Regis Gayon1, Florence Borot2, Christine Duthoit 1, Nicolas Martin1, Florine Samain1, Alexandra Iche1, Pascale Bouille1

1Flash Therapeutics, Toulouse, France, 2Columbia, New York, United States

Background: The emergence of targeted immunotherapy, especially chimeric antigen receptor (CAR) T-cell therapy, has opened new possibilities, demonstrating tremendous success for patients with lymphoblastic leukemia. Natural Killer (NK) cells are other key immune effector cells which, contrary to T cells, do not require antigen priming and are at a low risk of Graft-Versus-Host Disease (GVHD), therefore offering the potential of an allogenic “off-the-shelf” therapeutic product. Nevertheless, primary NK cells exhibit a high resistance to lentiviral transduction, hampering transgene expression and consequently the generation of CAR-NK cells.

Methods: Pre-clinical and clinical CAR-NK cells manufacturing requires to be performed using the same process. Here we show a manufacturing process of highly purified and concentrated third generation lentiviral vectors, available for a continuum going from Discovery to Clinic phases, integrating a control plan, allowing for the development and for the good manufacturing practices (GMP) production of custom lentiviral batches dedicated to clinical applications. Human primary NK cells, previously activated by artificial antigen-presenting cells (APC) are transduced with such highly purified and concentrated Lentiviral vectors prior to be evaluated by in vitro cytotoxic assays and in vivo engraftments into mouse model.

Results: In this collaborative work, we show that the use of highly purified and concentrated self-inactivating lentiviral vectors in combination with an optimized transduction protocol, allows up to 65% of trasnduced human cord blood derived NK cells. Also, we show that transduction does not lead to viability nor phenotypic alterations of the transduced NK cells. Our approach not only achieves high transduction efficiency, leading to strong and stable transgene expression, but also preserves the cytotoxic function of the NK cells, in vitro and in vivo.

Conclusions: Many obstacles exist for clinical development of a CAR-based cellular therapy product, which requires efficient and safe delivery technologies, as well as gene expression level and duration tailoring. It’s possible to achieve this, through the use of delivery tools, which allow highly efficient gene transfer while maintaining transduced cell viability and phenotype. Here we propose a novel method allowing for the generation and production of lentiviral vector engineered primary NK cells, thus circumventing the problem of poor autologous CAR-T cell efficiency and gamma retrovirus associated risks. This work lays the groundwork for novel cellular therapies based on lentivirally transduced primary NK cells. All these factors, as well as the ability to produce lentiviral vectors using Flash Therapeutics’ GMP compliant production platform, offer additional safety considerations for clinical development and human use.

Disclosure: NONE.

3 - CAR-based Cellular Therapy – Preclinical

P117 A NOVEL MULTICOLOR FLUORESCENT SPOT ASSAY FOR THE FUNCTIONAL ASSESSMENT OF CAR T CELL PRODUCTS

Thierry Iraguha1, Destiny Omili 1, Etse Gebru1, Nancy Hardy1, Tim Luetkens1, Aaron Rapoport1, Djordje Atanackovic1

1University of Maryland Greenebaum Comprehensive Cancer Center, Baltimore, United States

Background: Chimeric antigen receptor (CAR) T cell therapies are powerful anti-tumor immunotherapies and have revolutionized the treatment of B cell lymphomas. Unfortunately, recurrent disease after CAR T cell therapy remains a clinical problem with up to 50% early relapses after CD19-targeted CAR T cells. Therefore, there is a critical need for novel assays that are able to predict durable responses and will help to guide the optimization of CAR T cell therapies.

Methods: We developed a novel multicolor fluorescent spot assay (MFSA) for the functional assessment of CAR T Cell products on a single-cell level combining the numerical assessment of CAR T cell products with their functional characterization.

Results: We first used a standard single-cell interferon (IFN)g ELISPOT assay to measure CD19-targeted CAR T cell responses to CD19-coated beads or to beads coated with an irrelevant target. Beads coated with anti-CD3/anti-CD28 antibodies were used as a positive control. We then developed, optimized, and validated an MFSA that simultaneously measures the secretion of combinations of different cytokines on a single-CAR T cell level. We identified IFNg/TNFa/Granzyme B as the most relevant combination of cytokines and we used our novel multicolor ELISPOT assay to functionally and numerically characterize two clinical-grade CAR T cell products.

Conclusions: We have developed a novel multicolor fluorescent spot assay (MFSA) for the quantitative and functional assessment of CAR T cell products. The clinical value of our novel assay will be assessed in clinical studies correlating the pre-infusion assessment of CAR T cell products with the patients’ outcome.

Clinical Trial Registry: N/A

Disclosure: Nothing to declare.

3 - CAR-based Cellular Therapy – Preclinical

P118 PROCESSING AND CRYOPRESERVATION OF PERIPHERAL BLOOD LEUKAPHERESIS FOR DLI ADMINISTRATION OR CAR-T MANUFACTURING

Serena Urbani1, Julien Camisani2, Oliver Link2, Rita Antonini1, Lucia Bianchi1, Valeria Calossi1, Carmela De Sanzo1, Francesca Felicioni1, Valentina Fulgido1, Gaetano Mobilio1, Michela Santosuosso1, Irene Sodi1, Riccardo Saccardi 1

1Cell Therapy and Transfusion Medicine Unit, Careggi University Hospital, Firenze, Italy, 2Cytiva, Eysins, Switzerland

Background: Autologous Mononuclear cells from Apheresis [MNC (A)] products contain mononuclear cells (MNC) collected from peripheral blood by an apheresis procedure and are intended for clinical use other than Hematopoietic Stem Cells Transplantation (HSCT), such as Donor Lymphocytes Infusions (DLIs) or Genetically modified Immune Effector Cells (IECs).

We carried out a validation of MNC (A) automated processing by Cytiva, aimed to cryopreservation, storage and thawing/washing for subsequent manipulation. The following processes were assessed:

  1. 1.

    Cryopreservation

  2. 2.

    Thawing

  3. 3.

    Washing

  4. 4.

    Density-Gradient-Based Separation (DGBS).

Methods: Peripheral Blood Leukapheresis were cryopreserved by a standardized controlled temperature freezing technique and stored in freezing bags in liquid nitrogen. The bags were thawed using a SmartMax device (CellThaw Protocol) and washed using SmartWash Protocol v314 on Sepax 2 S-100.

Finally, the washed product was stratified on a density gradient in order to increase the MNC purity, using NeatCell Protocol v319.

A sample was drawn after any step to evaluate WBC, GRC, RBC, MNC, CD3+ cells content and viability by flow-cytometer analysis.

Results: Characterization and yields of TNC, MNC, and CD3+ cells of the thawed units, after density gradient separation (NeatCell output), and comparison with waste fraction of density gradient separation, are reported here: the product composition in terms of MNC components is very similar between the fraction obtained after NeatCell and its waste fraction; NeatCell allows a 36% yield of WBC and 24,7% of CD3 of the thawed product, while depleting unwanted GRC and RBC cells to provide a purer cellular product. Viability of all fractions is >97%.

Process performance (from washed product):

NeatCell output

NeatCell waste

MNC %

96,1

96,9

MNC yield %

36,4

62,3

WBC yield %

36,2

61,5

CD3 yield %

24,7

77

GRC depletion %

64,6

 

RBC depletion %

75,4

 

Overall process performance (from fresh product):

Without NeatCell

With NeatCell

MNC yield %

93,1

33,8

CD3+ yield %

76,3

18,9

GRC depletion %

78,7

93,5

RBC depletion %

43,9

86,1

Conclusions: Sepax technology allows a highly automated processing of Autologous MNC in a closed system. High Lymphocytes yield and viability after thawing/washing process were shown, therefore allowing further manipulation such as CAR-T manufacturing.

DGBS, carried out after the washing procedure, resulted in a loss of both WBC and Lymphocytes, but allows a purer cellular product by further depleting undesirable GRC and RBC cells.

DGBS shall be avoided if the primary goal is to increase MNC recoveries and shall be considered if the primary goal is to maximize cellular product purity.

Disclosure: Nothing to declare.

5 - Cellular Therapies other than CARs

P119 SARS-COV-2 T-CELLS FOR ADOPTIVE CELL THERAPY IN IMMUNOCOMPROMISED CANCER PATIENTS

Rafet Basar 1, David Marin2, Nadima uprety2, May Daher2, Ye li2, Christina Ganesh2, Kimberly Klein2, Mayra Shanley2, Fernando Martinez2, Fleur Aung2, Ana Karen Nunez Cortes2, Sunil Acharya2, Pinaki Banerjee2, Indreshpal Kaur2, Mariam Ammari2, Sheetal Rao2, Richard Champlin2, Elizabeth J. Shpall2, Katayoun Rezvani2

1MD Anderson Cancer Center, Hosuton, United States, 2MD Anderson Cancer Center, Houston, United States

Background: SARS-CoV-2 causes serious infections in immunocompromised cancer patients with reported mortality up to 20%. Furthermore, a proportion of immunocompromised patients with COVID19 develop persistent or recurrent infection in spite of receiving novel antiviral therapies. Adoptive cell therapy with off-the-shelf, partially HLA matched, allogeneic cytotoxic T-lymphocytes (CTLs) has been successfully used to treat immunosuppressed patients with other severe viral infections such as adenovirus, cytomegalovirus and BK virus.

Methods: We manufactured 11 SARS-CoV-2 specific CTL cell lines (COVID-CTLs) from the peripheral blood of 11 healthy donors who had recovered from COVID19. Briefly, cells were cultured with a combination of peptide libraries spanning the entire sequence of the SARS-CoV-2 in the presence of IL-2/4/7 for 14 days, cryopreserved and banked. For each cell line, SARS-CoV-2 reactive T-cells were enumerated. The trial opened in December 2020 and 26 patients have been enrolled so far. Table-1 shows the patients characteristics. Cell lines was selected based on the degree of HLA matching between the line and the patient. Consideration for the frequency of SARS-CoV-2 reactive T-cells was also given. Patients received one infusion of 2x105 cell/kg. Infusions could be repeated at 2 weeks intervals if the patient failed to respond. Patients could not receive steroid treatment immediately prior or after infusion of the CTLs. Responses were defined as improvement by at least one category in the WHO COVID severity scale within 28 days of the infusion. In vivo CTL expansion was assessed using an HLA-based flow cytometry chimerism assay.

Table 1. Patients’ characteristics

Variable

n (%)

Age median (range)

54.5 (23-82)

Sex

female

8 (30.8)

male

18 (69.2)

Recipient of allogeneic SCT

no

12 (46.2)

yes

14 (53.8)

Underlaying disease

AML/MDS

8(30.8)

CML/MF

4 (15.4)

ALL

2 (7.7)

NHL

9 (34.6)

Clinical characteristics

Requiring Oxygen

5 (19.2)

Multifocal pneumonia

26 (100)

Number of CTL infusions

1

24 (92.3)

2

2 (7.7)

Results: We did not observe any toxicity (any grade) attributable to the CTLs, including cytokine release syndrome (CRS) or graft vs host disease (GVHD). Twenty-six out of the 26 patients treated (100%) responded. The median time to response was 17 days (range 5-28 days). No patient had a recurrence of the COVID19 infection during the follow-up. We confirmed in vivo expansion of donor-derived COVID-CTLs in the peripheral blood of patients, with the peak expansion observed 7-14 days post-infusion.

Conclusions: COVID-CTLs are safe. Although the design of the study is not adequate to assess the efficacy of the CTLs, the fact that 25/26 severely immunocompromised patients improved, none requiring mechanical ventilation or ICU admission, gives a strong signal in favor of the efficacy of the CTLs. The trial is being amended to include contemporaneous matched controls.

Clinical Trial Registry: 2020-0759,https://clinicaltrials.gov/ct2/show/NCT04742595

Disclosure: • Rafet Basar and The University of Texas MD Anderson Cancer Center have an institutional financial conflict of interest with Takeda Pharmaceutical for the licensing of the CAR-NK cell technology

• Rafet Basar and The University of Texas MD Anderson Cancer Center have an institutional financial conflict of interest with Affimed Pharmaceutical.

5 - Cellular Therapies other than CARs

P120 PRECISION-ENGINEERED CELL THERAPY ORCA-T DEMONSTRATES HIGH RELAPSE-FREE SURVIVAL AT 1 YEAR WHILE REDUCING GRAFT-VERSUS-HOST DISEASE AND TOXICITY

Caspian Oliai1, Rasmus Hoeg2, Anna Pavlova3, Arpita Gandhi4, Lori Muffly3, Rohtesh Mehta5, Samer Srour5, Edmund Waller6, Robert Lowsky3, Sagar Patel7, Bhagirathbhai Dholaria8, Jeremy Pantin9, Amandeep Salhotra10, Nathaniel Fernhoff11, J. Scott McClellan 11, Mehrdad Abedi2, Robert Negrin3, Everett Meyer3

1UCLA Jonsson Comprehensive Cancer Center, Los Angeles, United States, 2University of California, Davis, Sacramento, United States, 3Stanford University, Stanford, United States, 4Oregon Health and Science University, Portland, United States, 5MD Anderson Cancer Center, Houston, United States, 6Emory University, Atlanta, United States, 7University of Utah, Salt Lake City, United States, 8Vanderbilt University Medical Center, Nashville, United States, 9Sarah Cannon Center for Blood Cancer at TriStar Centennial Medical Center, Nashville, United States, 10City of Hope National Medical Center, Duarte, United States, 11Orca Bio, Menlo Park, United States

Background: Allogeneic hematopoietic stem cell transplants (alloHSCT) remain the only curative treatment for many hematologic malignancies; however, they are associated with significant toxicity. Strategies to maintain the graft-vs-tumor and graft-vs-infection effects while eliminating GvHD have long been a key goal in the field.

Orca-T is a high-precision cell therapy consisting of stem and immune cells, derived from allogeneic donors, that leverages highly purified, polyclonal donor regulatory T cells to control alloreactive immune responses.

Methods: As of 25 Oct 2022, 151 patients had received Orca-T and had ≥ 100 days of follow-up (f/u). These patients had a diagnosis of acute leukemia in CR [AML (44%), ALL (31%), MPAL (3%), CML w/prior blast crisis (5%), or high-risk MDS (15%). Patients received Orca-T as part of a single-center Phase 2 study (n = 34) or a multi-center Phase 1b study (n = 117). Median f/u was 15 months (range 3-43); 146 pts had > 1 year and 86 had >18 months of f/u. Patients were aged 19-71 (median 48) and 57% male. Donors were HLA-matched related (52%) or unrelated (48%). Patients received investigator’s choice of myeloablative conditioning (MAC) regimens (busulfan-based (77%); TBI-based (23%)) prior to Orca-T, followed by single-agent GvHD prophylaxis with either tacrolimus (n = 148) or sirolimus (n = 3). For comparison, an independent CIBMTR-based cohort was identified which consisted of patients with AML, ALL, or MDS who received MAC alloHSCT with a PBSC source between 2016-2018 followed by tacrolimus/methotrexate prophylaxis.

Results: Orca-T was successfully manufactured in a single, centralized GMP manufacturing facility, distributed, and infused throughout the U.S.

Orca-T outperformed standard of care alloHSCT. GVHD and relapse-free survival (GRFS) and overall survival was 70% and 88% at 1 year in the overall Orca-T study population compared to 21% and 68% at 1 year in the CIBMTR comparator cohort, respectively. Non-relapse mortality (NRM) was lower with Orca-T at 4% at 1 year compared to 10% at 1 year in the CIBMTR cohort.

Median times to neutrophil and platelet engraftment for all Orca-T recipients were 13 and 16 days, respectively; graft failure was 1.6%.

Clinical outcomes with Orca-T appeared to be enhanced with a conditioning regimen of busulfan, fludarabine, and thiotepa (“BFT”, n = 71 patients, median f/u 14 months). Relapse free survival (RFS) was 87% at 1 year in this group, 93% among MRD- patients and 68% among MRD+ (Figure). NRM was 0% at 1 yr. Severe infections (Grade 3 + MOP infections per the BMT-CTN grading scale) occurred in 9% of these patients. Grade ≥ 3 aGvHD and moderate to severe cGvHD rates were low in this group at 1.5% and 5%, respectively. Additionally, GRFS was 81% and overall survival was 94% at one year.

Conclusions: At more than 1 year of median f/u, outcomes with Orca-T demonstrate robust graft-vs-tumor and graft-vs-infection effects while markedly reducing GvHD and NRM despite MAC. These outcomes were accomplished with reliable cell manufacturing and distribution of Orca-T nationally. A multi-center randomized controlled phase 3 trial comparing Orca-T to SOC is currently enrolling across the US (NCT05316701).

Clinical Trial Registry: https://clinicaltrials.gov/ct2/show/NCT05316701

Disclosure: Caspian Oliai | Research Funding: Pfizer, Jazz Pharmaceuticals, Orca Bio, Arog, Seagen

Rasmus Hoeg | Research Funding: Orca Bio

Anna Pavlova | Current Employment/Equity: Orca Bio

Arpita Gandhi | Research Funding: Orca Bio; Consultancy: Care Dx; Honoraria: Gamida

Lori Muffly | Consultancy: Amgen, Pfizer, Kite, Medexus, CTI Biopharma, Astellas, UpToDate; Research Funding: Kite, Astellas, Jasper, Adaptive, BMS, Novartis, Orca Bio; Honoraria: Adaptive, UpToDate

Rohtesh Mehta | Research Funding: Syndax, Orca Bio

Samer Srour | Research Funding: Orca Bio

Edmund Waller | Research Funding: Verastem Oncology, Orca Bio; Consultancy, BoD or Advisory Committee: Verastem Oncology

Robert Lowsky | Research Funding: Orca Bio

Sagar Patel | Honoraria: Kite, Medexus Pharma, CareDx; Research Funding: Orca Bio

Bhagirathbhai Dholaria | Research Funding: Poseida, Takeda, Pfizer, Janssen, Wugen, Molecular Templates, Orca Bio, Angiocrine, BMS, MEI Pharma; Consultancy: BEAM Therapeutics, Gamida Cell, Arivan, Jazz Pharmaceuticals; Honoraria: MJH Biosciences

Jeremy Pantin | Consultancy: NKARTA, Omeros; Speakers Bureau: Omeros; Honoraria: Cardinal Health; Research Funding: Orca Bio

Amandeep Salhotra | Advisory Board: Kadmon; Research Funding: BMS

Nathaniel Fernhoff | Current Employment/Equity: Orca Bio

J. Scott McClellan | Current Employment/Equity: Orca Bio

Mehrdad Abedi | Speakers Bureau: BMS, AbbVie, Kite, Celgene; Consultancy: Celgene; Equity Holder: CytoDyn; Research Funding: Orca Bio

Robert Negrin | Consultancy: Garuda, Kuur, Amgen, Novartis, Magenta; Equity Holder: Magenta, CoImmune, BioEclipse Therapuetics; Honoraria: UpToDate

Everett Meyer | BoD/Advisor: indee labs; Co-founder & Scientific Advisor: GigaGen, Triursus Therapeutics; Research Funding: Orca Bio

This study was sponsored by Orca Bio

5 - Cellular Therapies other than CARs

P121 MULTISTATE MODEL ANALYSIS OF PROPHYLACTIC/PRE-EMPTIVE DONOR LYMPHOCYTE INFUSION FOR HIGH-RISK ACUTE MYELOID LEUKEMIA AND MYELODYSPLASTIC SYNDROME IN COMPLETE HEMATOLOGIC REMISSION AFTER ALLOGENEIC STEM CELL TRANSPLANTATION

Giuliano Filippini Velázquez 1, Jan Frederic Weller2, Anna Rubeck3, Tobias Arndt3, Stefan Schiele3, Markus Mezger2, Gernot Müller3, Claudia Lengerke2, Martin Trepel1, Maximilian Christopeit2, Christoph Schmid1

1Augsburg University, Augsburg, Germany, 2Tübingen University Hospital, Tübingen, Germany, 3Augsburg University, Institute of Mathematics, Computational Statistics and Data Analysis, Augsburg, Germany

Background: Donor lymphocyte infusion (DLI) is given in complete hematologic remission (CHR) after allogeneic stem cell transplantation (alloSCT) to prevent relapse in AML/MDS, either as prophylaxis in high-risk patients without evidence of imminent relapse (proDLI), or pre-emptively in patients with incomplete chimerism (preDLI-IC) or minimal residual disease (preDLI-MRD). The antileukemic efficacy is counterbalanced by the risk of DLI-induced Graft-versus-Host disease (GvHD) and its potentially associated morbidity/mortality.

Methods: We retrospectively analysed clinical outcomes and GvHD-incidence in all consecutive patients with AML/MDS receiving unmodified (CD3 + ) proDLI/preDLI in two German centres between 2007 and 2021 according to institutional standards. No ex-vivo T cell depletion had been used for SCT. For inclusion, minimal follow-up after first DLI had to be 100 days to assure the detection of post-DLI acute GvHD. Criteria for DLI were: CHR, no immunosuppression for ≥4 weeks, no active GvHD, no history of GvHD ≥III, and no active infection. As per institutional protocol, proDLI/preDLI was to be given in 4-6 weeks’ intervals at escalating doses, which differed according to donor type and time since alloSCT. Beyond standard outcome analysis, a multistate model (allowing for non-absorbable states) was fitted to evaluate the probability of final treatment success, i.e., being alive and in remission without or with ongoing low dose immunosuppression (IS) for DLI-induced GvHD. Further, an exploratory analysis correlating cytogenetics at diagnosis and long-term survival after DLI1 was performed.

Results: Eighty-three patients (AML/MDS, n = 75/8, Table 1) were identified, representing a high-risk cohort with high-risk ELN/IPSS (58%), TP53mut/del (10%), SCT in active disease (59%) and second alloSCT (10%). ProDLI/preDLI-IC/preDLI-MRD was given in 43%/33%/24%. Median number of DLI/patient was 3 (1-5).

Following DLI1, median follow-up of survivors was 41 months. Response to preDLI (n = 47) was 83% (preDLI-IC: 22/27, preDLI-MRD: 17/20). Two-year overall/leukemia-free survival (OS/LFS) from DLI1 was 82%/70% for proDLI, 81%/74% for preDLI-IC and 61%/48% for preDLI-MRD. Two-year cumulative incidence (CI) of aGvHD I-IV/III-IV was 51%/18%. Two-year CI of cGvHD/extensive cGvHD was 46%/7%. Two-year non-relapse mortality/relapse incidence/leukemia-associated death (NRM/RI/LAD) were 8%/28%/17%. Two-year GvHD-relapse-free survival (GRFS) was 63%. Using the multistate model, probability of final treatment success was 63% (alive/in remission/without IS, n = 45; alive/in remission/with low dose IS, n = 7). 34/52 patients transiently requiring IS could discontinue the drugs later. In univariate analysis, the following cytogenetic/molecular abnormalities were associated with long term survival: DNMT3A (6/6, of which preDLI=5), FLT3-TKD (4/4; preDLI=2), Trisomy 8 (8/9, preDLI=3), del7 (7/8, preDLI=3), TP53mut/del (6/8, preDLI=5).

Conclusions: High rates of long-term survival were observed in this high-risk cohort receiving pro/pre DLI in CHR after alloSCT. Rates of DLI-induced GvHD were considerable, however, severe grades were rare, and most patients requiring IS could discontinue it over time. Hence, as shown in the multistate model, final treatment success was achieved in >60%, with 54% being alive/in remission/without IS by end of study. The data provide an improved estimate of risks and benefits of pro/preDLI. Beyond, exploratory analysis identified genetic subtypes associated with favourable outcome after SCT + DLI.

Table 1. Patients’ characteristics

  

Total (%)

Diagnose

AML

75 (90%)

MDS

8 (10%)

Risk group

AML, according to ELN

low

6 (7%)

intermediate

29 (35%)

high*

40 (48%)

high*

8 (10%)

MDS, according to IPSS

Number of allo-SCT

First alloSCT

75 (90%)

Second alloSCT

8 (10%)

Stage at this alloSCT

CR

34 (41%)

Active disease

49 (59%)

Donor type

MSD

20 (24%)

MUD 10/10

47 (57%)

MUD 9/10

10 (12%)

Haploidentical

6 (7%)

Conditioning intensity (TCI) according to Spyridonidis et al, BMT 2020

low

8 (10%)

intermediate

25 (30%)

high

50 (60%)

Extramedullary disease before alloSCT

yes

5 (6%)

no

78 (94%)

Type of DLI

proDLI

36 (43%)

preDLI-IC

27 (33%)

preDLI-MRD

20 (24%)

Interval from alloSCT to first DLI in median (range)

all patients

182 (73-1306) days

proDLI

203 (118-498) days

preDLI-IC

173 (121-470) days

preDLI-MRD

184 (73-1306) days

Initial cell dose in median (range)

MSD

1 (0.5-10) x 106 /Kg CD3+

MUD 10/10

0.2 (0.1-10) x 106 /Kg CD3+

MUD 9/10

0.2 (0.2-1) x 106 /Kg CD3+

Haploidentical

0.2 (0.2-1) x 106 /Kg CD3+

Number of DLI/patient

1

24 (29%)

2

16 (19%)

3

31 (37%)

4

8 (10%)

5

4 (5%)

  1. ELN European Leukemia Network, DLI donor lymphocyte infusion, proDLI prophylactic DLI, preDLI preemptive DLI, IC incomplete chimerism, MRD minimal residual disease, alloSCT allogeneic stem cell transplantation, CR complete remission, MSD matched sibling donor, MUD matched unrelated donor, 9/10 and 10/10 refers to HLA-match, TCI: transplant conditioning intensity
  2. *Totally 8 patients (10%) with TP53 mutation/deletion

Disclosure: Nothing to declare

5 - Cellular Therapies other than CARs

P122 T-ALLO10 INFUSION AFTER AΒDEPLETED-HSCT IN CHILDREN AND YOUNG ADULTS WITH HEMATOLOGIC MALIGNANCIES: IMPROVED IMMUNE RECONSTITUTION IN THE ABSENCE OF SEVERE GVHD

Alice Bertaina 1,2, Rosa Bacchetta1,2, David Shyr1,2, Gopin Saini1, Jennifer Lee1, Karen Kristovich1,2, Rajni Agarwal-Hashmi1,2, Orly Klein1,2, Kathryn Melsop1, Keri Tate1, Giulia Barbarito1, Linda Oppizzi1, Pauline Chen1, Alma-Martina Cepika1, Maria Grazia Roncarolo1,2

1Stanford University School of Medicine, Stanford, United States, 2Lucile Packard Children’s Hospital, Palo Alto, United States

Background: The use of allogeneic hematopoietic stem cell transplantation (HSCT) for treating pediatric patients with acute leukemia has been limited by the availability of HLA-matched donors. TCRαβ+ T-cell and CD19+ B-cell depletion (αβdepleted) of the graft has significantly broadened the use of HLA-mismatched related and unrelated donors. However, the incidence of viral reactivations ( ~ 50%) and leukemic relapse (25-30%) after αβdepleted-HSCT remains significant, mostly because of the suboptimal immune reconstitution (IR) due to the extensive ex vivo TCRαβ+ T-cell depletion and the use of pre-HSCT serotherapy.

Methods: We have developed an innovative T-cell immunotherapy, the T-allo10 drug product. T-allo10 product is generated in vitro from donor CD4+ T cells and is enriched in type 1 regulatory T (Tr1) cells, which are host alloantigens specific and therefore suppress host-reactive TCRαβ+ T cells causing GvHD. It also contains polyclonal naive and memory TCRαβ+ T cells able to respond to pathogens and tumor antigens. We hypothesized that the infusion of T-allo10 after αβdepleted-HSCT will a) expedite IR by providing a source of TCRαβ+ T cells, which support immune responses and facilitate the generation of donor-derived naive T cells, and b) modulate anti-host immune responses thanks to the function of alloantigen specific Tr1 cells. The improved IR will reduce the risk of infections and leukemic relapse, without increasing the risk of GvHD.

Results: We report the preliminary results of a single center, non-randomized, non-controlled open-label Phase I/Ib trial in children and young adults with hematologic malignancies receiving αβdepleted-HSCT (NCT 04640987).

At present, we enrolled 10 patients (Table 1) with high-risk disease (i.e., 3 patients received a 2nd HSCT) in the Phase I portion of the study: 9/10 received the T-allo10 infusion. No DLT (grade IV acute GvHD, grade 3 or 4 treatment-related adverse events) have been observed. Two patients developed grade II acute GvHD (22%), and 1 of these 2 patients developed chronic GvHD (11%). Four out of the 6 evaluable patients (66%) achieved the IR efficacy endpoint (Figure 1), reaching the threshold of 50 CD3+CD4+ T cells/mcl by Day +60 post αβdepleted-HSCT. These data compare favorably to our historical control Cohort in which only 31% of patients achieved this endpoint (Figure 1). Interestingly, TCRαβ+ CD3+CD4+ T cells with a memory phenotype increase following the T-allo10 infusion. In all patients, Tr1 cells are detected in the peripheral blood shortly after T-allo10 infusion (within the first 7 days). In Cohort 2 patients, the % of Tr1 cells reached 19.9%, 11.5%, and 5.2% of the memory CD4+ T cell population, with the highest value observed at day +7 post-T-allo10 infusion. CD4+CD25hiCD127lo Treg cells were also detected in Cohort 2 patients, at high frequencies early after T-allo10 cell infusion, reaching 8-17% of the CD4+ T cell subset 1 to 7 days post-T-allo10 infusion.

Table 36 Table 1. Patients and Transplants Characteristics

Conclusions: These early results support our hypothesis that the adoptive transfer of T-allo10 cells boosts T-cell immune reconstitution without increasing the risk of GvHD and show that Tr1 cells and CD4+CD25hiCD127lo Treg cells are detected in the peripheral blood of T-allo10 infused patients.

Clinical Trial Registry: NCT04640987, https://clinicaltrials.gov/ct2/show/NCT04640987

Disclosure: Nothing to declare.

5 - Cellular Therapies other than CARs

P123 FIRST-IN-HUMAN STUDY OF MDG1011, A TCR-T CELL THERAPY DIRECTED AGAINST HLA-A*02:01-RESTRICTED PRAME, FOR HIGH-RISK MYELOID AND LYMPHOID NEOPLASMS (CD-TCR-001)

Simone Thomas 1, Martin Wermke2, Vladan Vucinic3, Eva Wagner-Drouet4, Andreas Mackensen5, Robert Zeiser6, Gesine Bug7, Michael Schmitt8, Petra Prinz9, Dolores Schendel9, Kai Pinkernell9,10, René Goedkoop9

1University Hospital Regensburg, Regensburg, Germany, 2University Hospital Carl Gustav Carus Dresden, Dresden, Germany, 3University Hospital Leipzig, Leipzig, Germany, 4University Medical Center Mainz, Mainz, Germany, 5University of Erlangen-Nuernberg, Erlangen, Germany, 6University of Freiburg Medical Center, Freiburg, Germany, 7Goethe University Frankfurt, University Hospital Frankfurt, Frankfurt, Germany, 8University Hospital Heidelberg, Heidelberg, Germany, 9Medigene AG, Martinsried, Martinsried, Germany, 10Glycostem, Oss, Netherlands

Background: Preferentially expressed antigen in melanoma (PRAME) is a cancer-testis antigen present in several solid and hematological malignancies. This multicenter Phase I study (NCT03503968) evaluated the safety and feasibility of escalating doses of MDG1011 in patients with refractory/relapsed (r/r) AML, MDS or multiple myeloma (MM).

Methods: Treatment was conducted with MDG1011 after lymphodepletion with fludarabine (25 mg/m² x 3d) and cyclophosphamide (300 mg/m² x 3d) in patients with HLA-A*02:01- and PRAME-positive r/r disease. Dose escalation included 3 dose levels (DL), 0.1x106 (DL1), 1x106 (DL2) and 5x106 (DL3) TCR-T cells/kg BW, respectively. The primary objectives were to assess safety and tolerability of MDG1011 and establish maximum tolerated dose and/or recommended Phase II dose. Secondary objectives were to evaluate efficacy and correlation with PRAME expression through immune monitoring (quantification of PRAME mRNA levels in bone marrow (BM) and/or peripheral blood (PB) and monitoring of TCR-T persistence in PB).

Results: 13 patients with advanced, myeloid and lymphoid neoplasms, and a median age of 65 years were enrolled and underwent leukapheresis (n = 10 with AML, n = 1 with MDS/MPN and n = 2 with MM). MDG1011 manufacturing feasibility was high with release criteria met for 12/13 patients (92.3%).

9 patients were treated with MDG1011 at the three escalating DLs (n = 3, 4 and 2 patients, respectively); 4 patients succumbed to their disease before TCR-T infusion. All 13 patients experienced adverse events (AEs), of which 54/124 AEs were ≥ grade 3 toxicities (NCI CTCAE v4.01), 31/124 AEs related to lymphodepletion and 21/124 AEs related to MDG1011. 12 SAEs were reported in 7/9 treated patients. Grade 1 cytokine release syndrome (CRS) occurred in 1 patient at DL2, grade 2 CRS in 1 patient at DL3 that was manageable with tocilizumab. Neurotoxicity (ICANS) or dose-limiting toxicity were not reported.

In patients receiving MDG1011, 4 died from their disease (none in DL3, none considered related to MDG1011) and 4 experienced disease progression. 1 patient with AML with extramedullary disease (DL1) experienced complete remission at week 4 but had progressed by month 3. In addition, 1 patient with multilineage MDS/MPN (DL3) remained without progression to secondary AML through month 12. In this patient, TCR-T cells were detected in PB at week 4 through month 12, whereas PB PRAME levels were no longer detected at week 4 but gradually increased thereafter while remaining clearly below the baseline level. In concordance, blast counts both in PB and/or BM remained well below baseline.

TCR-T cells were present in 6 of 8 patients within 4 weeks. PRAME (BM) decreased in 4 patients (3 AML, 1 MM) while a slight increase occurred in 1 patient (MM). PRAME (PB) decreased at week 4 for 2 patients treated at the highest dose but increased thereafter.

Conclusions: In heavily pre-treated patients with advanced myeloid and lymphoid neoplasms, treatment with MDG1011 was generally safe and well tolerated up to 5x106 PRAME-specific TCR-T cells/kg. Clinical observations were corroborated by persistence of MDG1011 cells in PB and reductions in PRAME mRNA levels in PB and/or BM.

Clinical Trial Registry: https://clinicaltrials.gov/ct2/show/NCT03503968

Disclosure: Simone Thomas, Medigene is the Sponsor of the clinical trial

Martin Wermke, Nothing to declare

Vladan Vucinic, Nothing to declare

Eva Wagner-Drouet, Travel grant: MSD,Janssen, Medac; advisory board: Kite Gilead, Novartis, BMS

Andreas Mackensen, Honoraria by Gilead/KITE, BMS/Celgene, Novartis, Miltenyi Biomedicine

Robert Zeiser, Honoraria by Novartis, Sanofi, Incyte and MNK

Gesine Bug, Honoraria by Novartis, Pfizer, Eurocept, Kite/Gilead, BMS/Celgene and Jazz

Michael Schmitt, Apogenix: Funding for collaborative research; BMS: Financial support of educational activities and conference, travel grants. Co-PI of clinical trials on CAR-T cells; Hexal: Financial support for research on biosimilars, travel grants; Kite/Gilead: Financial support of educational activities and conference, travel grants. Co-PI of clinical trials on CAR-T cells; MSD: Ad board member, PI of clinical trials on letermovir; Novartis: Collaborative research grant. Co-PI of clinical trials on CAR-T cells; TolerogenixX, Heidelberg: Co-Founder and shareholder.

Petra Prinz, Medigene is the Sponsor of the clinical trial

Dolores Schendel, Medigene is the Sponsor of the clinical trial

Kai Pinkernell, Medigene is the Sponsor of the clinical trial

René Goedkoop, Medigene is the Sponsor of the clinical trial.

5 - Cellular Therapies other than CARs

P124 HIGH-THROUGHPUT ASSESSMENT OF T CELL RECEPTOR SEQUENCES WITH SPECIFICITY TO THE MOST COMMON PATHOGENS AFTER ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION

Chrysoula Pantazi 1,2,3, Elisavet Vlachonikola2,3, Anna Vardi1, Eleni Zotou1,3, Kyriakos Koukoulias1, Ioanna Vallianou1, Maria Giannaki1, Zoe Boussiou1, George Karavalakis1, Apostolia Papalexandri1, Minas Yiangou3, Ioanna Sakellari1, Anastasia Chatzidimitriou2,4, Anastasia Papadopoulou1, Evangelia Yannaki1,5

1Gene and Cell Therapy Center, “George Papanikolaou” Hospital, Thessaloniki, Greece, 2Institute of Applied Biosciences (INAB), Centre for Research and Technology Hellas (CERTH), Thessaloniki, Greece, 3School of Biology, Development and Molecular Biology, Aristotle University of Thessaloniki, Thessaloniki, Greece, Karolinska Institute, Stockholm, Sweden, 5Hematology, University of Washington, Seattle, United States

Background: Adoptive immunotherapy with pathogen-specific T cells (pSTs) recognizing a broad range of pathogen-derived peptides through their native T-cell receptor (TR) has successfully emerged as an alternative treatment of opportunistic infections post allogeneic hematopoietic cell transplantation. However, the TR repertoire of ex vivo generated pSTs for immunotherapy has not yet been fully elucidated.

Methods: Herein, we analysed the TR beta-chain gene repertoire by next-generation sequencing (NGS) of two distinct immunotherapy products, i) tri-virus specific T cells (tri-VSTs) targeting cytomegalovirus (CMV), Epstein Barr virus (EBV) and BK virus (BKV) and ii) pentavalent-specific T cells (penta-STs) additionally targeting adenovirus (ADV) and the fungus Aspergillus fumigatus (AF), which are under evaluation in phase I/II clinical trials (EudraCT: 2014-004817-98/2020-004725-23, respectively). GMP-grade tri-VSTs and penta-STs were manufactured after exposure of peripheral blood mononuclear cells (PBMCs) from immunocompetent donors to CMV, EBV, BKV ± AdV and AF overlapping peptides and a 10-day culture. Specificity of donor-derived cell products and patient-derived PBMCs was assessed by IFN-γ Elispot. TCRB sequencing was performed in the whole T cell products (tri-VSTs n = 2, penta-STs n = 4), their corresponding pathogen-specific cell subsets (n = 26) post immunomagnetic IFN-γ enrichment and in patient PBMCs (n = 26). Immunogenetic analysis was performed by RT-PCR amplification of TRBV-TRBD-TRBJ rearrangements according to the BIOMED-2 protocol and paired-end NGS (Miseq/NextSeq). The NGS sequences after length and quality filtering were submitted to IMGT/HighVQUEST for annotation. Metadata analysis and clonotype computation (TRB rearrangements using the same TRBV gene and identical CDR3 amino acid sequence) were based on a validated purpose-built bioinformatics platform (tripr).

Results: Tri-VSTs and penta-STs provided a diverse TR repertoire consisting of 6,580-33,863 unique clonotypes/sample (median: 22,530) and demonstrated high clonality levels with the median frequency of the major clonotype being 6.48% (range: 2.23%-20.7%). The clonotypes identified in the enriched subpopulations were subjected to strict filtering: i) ≥10 read counts, ii) higher frequency in the enriched subpopulations over the unselected product as defined by the greater frequency (fold increase) of each post-enrichment clonotype than the median fold increase of all enriched clonotypes. The applied criteria resulted in 9,799 clonotypes, of which 8,749 were present in a single specific-cell fraction, arguably suggesting that they are pathogen-specific (CMV-specific: 2,768, EBV-specific: 1,763, BKV-specific: 2,003, ADV-specific: 1,363, Asp-specific: 852). Indeed, several of those identified virus-specific clonotypes could be tracked in vivo in two patient-derived PBMCs up to 15 weeks post tri-VST infusion with ranging frequencies 0.002%-12.7% (Pt1: 32 CMV-specific, 25 EBV-specific and 29 BKV-specific & Pt2: 20 CMV-specific, 31 EBV-specific and 31 BKV-specific clonotypes). Importantly, their presence in vivo was correlated with a decrease in the corresponding viral load, an increase in the frequency of circulating virus-specific T cells and ultimately, clinical response.

Conclusions: Overall, our findings elucidate the diverse immunogenetic profile of ex vivo generated pSTs, identify potential pathogen-specific clonotypes conferring protection against infections and provide a novel method to track in vivo the pSTs. Prospectively, the identification of optimal TRs that mediate clinical responses may serve as a stratification tool for patients at risk for suboptimal responses and help to select the best candidates for adoptive immunotherapy.

Clinical Trial Registry: EudraCT: 2014-004817-98, 2020-004725-23

Disclosure: Funding for this project was provided by Research, Technology Development and Innovation (RTDI) State Aid Action “RESEARCH - CREATE - INNOVATE” (T2EΔK-02437).

5 - Cellular Therapies other than CARs

P125 A CLINICAL TRIAL OF ADOPTIVE IMMUNOTHERAPY OF STEROID-REFRACTORY CHRONIC GVHD WITH MULTIPLE INFUSIONS OF CRYOPRESERVED DONOR-DERIVED REGULATORY T (TREG) CELLS

Mario Arpinati 1, Gabriella Chirumbolo2, Martina Barone2, Simonetta Rizzi1, Barbara Sinigaglia1, Tiziana Montemurro3, Cristiana Lavazza3, Emanuela Sergio4, Marie-Laure Yaspo5, Francesco Barbato1, Michele Dicataldo1, Enrico Maffini1, Michele Cavo1,2, Francesca Bonifazi1

1IRCCS Azienda Ospedaliero-Universitaria di Bologna, Bologna, Italy, 2Università di Bologna, Bologna, Italy, 3IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milano, Italy, 4AUSL Bologna, Bologna, Italy, 5Max Planck Institute for Molecular Genetics, Berlin, Germany

Background: Steroid refractory chronic GVHD has a poor prognosis, with only about 10-20% patients stopping immunosuppressive within 4 years. Chronic GVHD has been associated with selectively delayed Treg cell recovery, leading to the hypothesis that the infusion of healthy Treg cells may improve their recovery and possibly ameliorate chronic GVHD symptoms.

Methods: In the current trial (NCT02749084) we treated patients with steroid refractory chronic GVHD with multiple infusions of cryopreserved GMP-purified Treg cells. Each patient received three infusions of Treg cells each one month apart. The study had a dose escalating 3 + 3 design, with three dose levels (0.5, 1 and 2x10e6/kg total dose, respectively) with safety as the primary end point. Treg cells were isolated in a two-step procedure by immunomagnetic depletion of CD8+ and CD19+ cells followed by enrichment of CD25+ cells.

Results: Eleven Treg products have been prepared. Median purity was 94% (IQ 93-96%), 90% (IQ 90-94%) and 71% (IQ 66-77%) as based on identification of CD4 + CD25 + , CD4 + CD25 + CD127- and CD4 + CD25 + CD127-foxp3+ cells, respectively. No contamination (percentage <0.1%) was observed by CD19 + , CD8+ and CD56+ cells. Median contamination by effector T cells (defined as CD4 + CD25 + CD127+ cells) and by Th17 cells (defined as CD161 + CD196 + ) was 11% and 1.6%, respectively, resulting in the infusion of 2.8x10e4/kg effector T cells and 3.9x10e3/kg Th17 cells. Nine out of 11 products have received at least one infusion of purified T regs, 3 at the first dose level (0,5x10e5 Treg/kg total dose), and 6 at the second dose level (10e6 Treg/kg total dose). No infusion related events were observed. One patient developed a DLT (CMV pneumonia) one month after the last infusion. 5 more SAEs were observed during the 12 months predetermined observation period but were considered unrelated to the infusion. Importantly, no acute GVHD or early flares of chronic GVHD were observed. Disease responses observed in the 8 evaluable patients who received all three Treg infusions are reported in Table 1. Interestingly, 5 out patients were able to reduce the dose of prednisone at the 12 month time point, with two patients being able to stop prednisone. While Treg numbers and percentages did not change significantly during the study, NGS analysis of TCR sequences confirmed the persistence of the infused Treg clones for up to 12 months after treatment.

Table 1: NIH-based evaluation of response in the 8 patients who received three T reg infusions.

3 months

 

CR

PR

SD

Prog

Global

0

7 (87%)

1 (13%)

0

Skin

0

2 (33%)

4 (67%)

0

Mouth

0

2 (33%)

4 (67%)

0

Eyes

0

2 (33%)

3 (67%)

0

Lung

0

0

6 (100%)

0

12 monthsa

 

CR

PR

SD

Prog

Global

0

5 (71%)

0

2 (29%)

Skin

0

3 (50%)

3 (50%)

0

Mouth

1 (25%)

2 (50%)

2 (25%)

0

Eyes

1 (15%)

1 (15%)

5 (70%)

0

Lung

0

0

4 (71%)

2 (29%)

  1. aone of the eight patients missed the 12 month follow up visit due to worsening clinical conditions
  2. abbreviations: CR: complete response; PR: partial response; SD: stable disease; Prog: progression.

Conclusions: Treatment of patients with steroid-refractory chronic GVHD with purified cryopreserved donor T regs appears feasible and safe and may improve disease severity in some patients.

Clinical Trial Registry: NCT02749084

Disclosure: The study has received funding from the European Union H2020 research and innovation programme under grant agreement No 643776 (“Tregeneration”)

The authors have no conflict of interest related to the present work.

5 - Cellular Therapies other than CARs

P126 CLINICAL APPLICATION OF DIRECTLY SORTED HAPLOIDENTICAL T REGULATORY AND MEMORY T CELLS IN PATIENTS WITH REFRACTORY GVHD AND VIRAL INFECTIONS

Viktoria Vedmedskia 1, Dmitry Pershin1, Mariya Fadeeva1, Elena Kulakovskaya1, Tatiana Sozonova1, Anna Vasilieva1, Maria Klimentova1, Svetlana Radygina1, Maria Dunaikina1, Larisa Shelikhova2, Dmitry Balashov1, Michael Maschan1

1Dmitry Rogachev National Medical Research Center of Pediatric Hematology, Oncology and Immunology, Moscow, Russian Federation, 2Dmitry Rogachev National Medical Research Center Of Pediatric Hematology, Oncology and Immunology, Moscow, Russian Federation

Background: The most significant non-relapse cause of morbidity and mortality after HSCT are graft-versus-host disease (GvHD) and viral infections, commonly associated with prolonged pharmacological immune suppression (IST). Direct sorting of donor Treg cells and memory T cells for adoptive transfer is being developed as a cell therapy approach to treat patients with GVHD, refractory to IST and concomitant viral infections. Combination of immunomagnetic and flow-based sorting can be used to create highly pure populations with defined composition. We report here on the first compassionate use of this approach.

Methods: Six patients after allogenic HSCT (TCRab/CD19 depletion) with IST-refractory chronic GvHD received infusions of Treg cells. Five of them had ongoing viral infections and received Infusions of memory T cells.

Cells products were obtained from haploidentical donors according to good manufacturing practice standard. Sorting of Treg cells was performed by MACSQuant® Tyto® Cell Sorter after pre-enrichment of CD25+ cells with the CliniMACS Prodigy from the apheresis product. Treg cell was defined by CD4 + CD25 + CD127- markers. Sorting of memory T cells (all CD4 + /CD8+ cells except CD45RA + CD197+ population) was performed from PBMC obtained with BD Vacutainer® CPT™ Mononuclear Cell Preparation Tube. Infusion dose information and composition of cells products are given in the table. The median time after HSCT to the moment of infusion was 9 month (min 120 days, max 1 year). Viremia was monitored by PCR. Detection and monitoring of virus-specific donor T cells was performed by IFNγ ELISpot assay. CD4+FoxP3+cells persistence in the blood was monitored every week by FACS. Additionally, expression of CTLA-4 and heterogeneity within the memory Treg compartment (CD197/CD45RA) were assessed. Before the infusion IST was withdrawn and one infusion of cyclophosphamide at 400 mg/m2 was given.

Results: Clinical response to the treatment was observed in all 6 patients. One patient succumbed to septic event four weeks after the infusion of Treg/Tmemory. No adverse effects and decline of clinical condition because of the alloreactive lymphocytes infusion were observed in 5 patients. FoxP3+cells appearance was detected in 4 patients on day 7 and maximum level of FoxP3+cells (7-32 cells/mkl) was achieved on the 21st day after infusion in all patients. The Treg population was represented mainly by highly proliferating T effector memory phenotype (CD197-CD45RA-). Distinguishing feature of Treg cells persistence was stable high levels of FoxP3 expression (MFI median in patients was 6903 (4503–9318), MFI median in healthy controls was 2797 (2051–4139), p < 0.0001) and CTLA-4 expression (MFI median in patients was 2740 (1677-16033), MFI median in healthy controls was 1275 (810-2213), p = 0.0022).

Clinical and laboratory indicators are presented in the Table:

Table 38

Conclusions: Our first experience confirms the feasibility of direct sorting in the GMP-grade conditions of highly pure populations of regulatory and memory T cells by means of immunomagnetic and flow-based approach. Infusions were associated with clear increment of Treg cells in the blood and meaningful clinical responses. Reduction of GVHD manifestations were correlated with stable Treg cells persistence and high level of FoxP3 expression throughout the period of monitoring.

Disclosure: M.Maschan recieved lecturer’s fee from Miltenyi Biotec.

5 - Cellular Therapies other than CARs

P127 PATIENT-TAILORED ADOPTIVE IMMUNOTHERAPY WITH EBV-SPECIFIC T CELLS FROM STEM CELL AND THIRD PARTY DONORS

Britta Maecker-Kolhoff 1, Agnes Bonifacius1, Britta Lamottke1, Sabine Tischer-Zimmermann1, Rebecca Schultze-Florey1, Lilia Goudeva1, Hans-Gert Heuft1, Lubomir Arseniev1, Rita Beier1, Gernot Beutel1, Gunnar Cario2, Birgit Fröhlich3, Johann Greil4, Leo Hansmann5, Justin Hasenkamp6, Michaela Höfs7, Patrick Hundsdoerfer5, Edgar Jost8, Kinan Kafa9, Oliver Kriege10, Nicolaus Kröger11, Stephan Mathas5, Roland Meisel12, Michaela Nathrath13,14, Mervi Putkonen15, Sarina Ravens1, Hans Christian Reinhardt7, Elisa Sala16, Martin Sauer1, Clemens Schmitt5, Roland Schroers17, Nina K. Steckel7, Ralf Ulrich Trappe18, Mareike Verbeek14, Daniel Wolff19, Rainer Blasczcyk1, Britta Eiz-Vesper1

1Hannover Medical School, Hannover, Germany, 2University Hospital Schleswig-Holstein, Kiel, Germany, 3University Children’s Hospital Muenster, Muenster, Germany, 4University Children’s Hospital Heidelberg, Heidelberg, Germany, 5Charité - Universitätsmedizin Berlin, Berlin, Germany, 6University Medicine Göttingen, Göttingen, Germany, 7University Hospital of Essen, Essen, Germany, 8University Medical Center RWTH Aachen, Aachen, Germany, 9Martin Luther University Halle-Wittenberg, Halle, Germany, 10Johannes Gutenberg-University Medical Centre, Mainz, Germany, 11University Medical Center Hamburg Eppendorf, Hamburg, Germany, 12Heinrich Heine University, Düsseldorf, Germany, 13Klinikum Kassel, Kassel, Germany, 14Technical University Munich, Munich, Germany, 15Turku University Hospital, Turku, Finland, 16University Hospital of Ulm, Ulm, Germany, 17Knappschaftskrankenhaus University Hospital, Bochum, Germany, 18Ev. Diakonie-Krankenhaus, Bremen, Germany, 19University Hospital of Regensburg, Regensburg, Germany

Background: Immunosuppressed patients are at risk of virus-induced complications, among which EBV-associated lymphoproliferative disorders represent the most frequent malignant complication. Adoptive transfer of EBV-specific T cells can restore specific immunity, leads to control of EBV replication and regression of EBV-associated lymphoproliferation. Here, we provide data of consecutive runs of a personalized T-cell manufacturing program evaluating donor, patient, T-cell product and outcome data.

Methods: Patient-tailored clinical-grade EBV-specific cytotoxic T-lymphocyte (EBV-CTL) products from stem cell donors (SCD), related third party donors (TPD) or unrelated TPD from the allogeneic T-cell donor registry (alloCELL) established at Hannover Medical School were manufactured by immunomagnetic selection using CliniMACS Plus or Prodigy and EBV PepTivators EBNA-1 and Select. Consecutive manufacturing processes were evaluated from quality control data and patient outcome and side effects were retrieved by retrospective chart analysis. EBV load was quantified by whole blood qPCR analysis according to local standards. EBV-specific T cells in patient blood were measured by interferon-γ ELISpot analysis.

Results: Forty clinical-grade EBV-CTL products from SCDs, related TPDs or unrelated TPDs were generated for 37 patients with and without transplantation history within a median timeframe of 5 days from donor identification to product manufacturing. Regarding quality of T-cell product no significant differences were observed between SCD and TPD products regarding cell number, product composition, and cell purity. 34 patients received 1-14 EBV-CTL products (fresh and cryopreserved) with a median cell number of 2.5x10e4 CD3+ cells/kg body weight. EBV-CTL transfer led to complete response in 19 of 30 patients who were evaluated for clinical response. In HSCT patients, complete response (CR) rate was slightly higher in patients receiving EBV-CTL from SCD (9/10 CR) than from TPD (5/11 CR). The majority of patients showed a decline in EBV load after transfer. No infusion-related toxicity was reported. Three of 10 patients with SCD-derived CTLs developed de novo GvHD after transfer, two of which in context with unmanipulated donor lymphocyte infusion (n = 1) and forced reduction of immunosuppression (n = 1). No de novo GvHD occurred in the TPD CTL group. EBV-specific T cells in patients’ blood were detectable in 16/18 monitored patients (89%) after transfer and detection correlated with clinical response.

Conclusions: In conclusion, personalized clinical-grade manufacturing of EBV-CTL products from SCD, related TPDs or unrelated TPD is feasible and their adoptive transfer is effective and safe in the majority of patients. While SCD-derived CTL appeared to be slightly more effective in inducing CR, TPD CTL exhibited no de novo GvHD development. Prospective clinical trials are needed to further explore the potential of SCD and TPD derived EBV-CTL adoptive therapy.

Clinical Trial Registry: Not applicable.

Disclosure: Nothing to declare.

5 - Cellular Therapies other than CARs

P128 COMBINED CAR-NK AND CAR-GD T CELLS AS A THERAPEUTIC TOOL POST ALLO-SCT

Nassim Salem 1,2,3, Anne-Sophie Chretien1,2,4, Marie-Sarah Rouviere1,2, Amira Ben Amara1, Alexandrine Faivre1, Anne-Charlotte Le Floch1,2,4, Laurent Gorvel1, Armelle Goubard1, Rémy Castellano1,2, Geoffrey Guittard1, Yves Collette1,2, Jacques Nunès1, Daniel Olive1,2,4, Raynier Devillier1,2,4

1Centre de Recherche en Cancerologie de Marseille, Marseille, France, 2Institut Paoli Calmettes, Marseille, France, 3Faculté de Pharmacie, Marseille, France, 4Faculté de Médecine, Marseille, France

Background: Allogeneic hematopoietic stem cell transplantation (Allo-SCT) is the main curative treatments for Acute Myeloid Leukemia (AML). While procedure related mortality related has significantly decreased over the 2 past decades, relapse of AML remains the major challenge, occurring in 20-50% according to disease risk. While T cells have a major role in the (graft versus leukemia) effect, post allo-SCT T cell-based immunomodulation is appealing to decrease relapse. However, T cells are also highly involved in graft-versus host disease (GVHD), limiting the clinical use of such strategies. As an alternative source for post allo-SCT immunomodulation, NK cells and Tγδ lymphocytes (γδT) are appealing because they can mediate a GVL effect without triggering GVHD. We propose here an expansion process, to generate a clinically compatible cell therapy product composed of both NK cell and gd T cells for future application after allo-SCT as well as CAR generation.

Methods: From healthy donor peripheral blood mononucleate cells, NK cells and γδT are expanded during 14 days using feeder cell support. The phenotypic changes between are assessed using mass cytometry. Following a co-culture with leukemic blasts, cytotoxicity, degranulation and cytokine production were measured by flow cytometry. As for the in vivo study a luciferase-expressing leukemic cell line was injected to NSG mice, and a bioluminescence and clinical follow-up was performed. Lentiviral transduction was used to generate CAR-NK/CAR-γδT.

Results: The process allowed us to achieve a 1000-fold expansion for both NK cells and γδT within 14 days. Conventional ab T cells represented systematically less than 5%. Both NK cells and γδT acquired a hyperactivated phenotype especially, with an increased expression of NKG2D and BCL-2 in both cell types and NKp30 in NK cells, which led to major degranulation and cytokine production abilities toward leukemic blaststogether with a significant cytotoxicity leading up to 80% of apoptotic blasts in 1h in vitro. In vivo, we showed a tumor charge decrease together with less circulating blasts (29 vs 5 blasts/µL of blood). Also, mice receiving combined NK and γδT cells displayed a significant prolonged survival compared with untreated controls, without developing GVHD.

Moreover, the genetic editing led to a combined αCD19 CAR-NK and CAR-γδTproduct with enhanced degranulation capacities against CD19pos targets.

Conclusions: We set up a cellular immunotherapy product with highly activated NK and γδT cells in large amount, possibly compatible with future clinical application in the context of allo-SCT, by inducing a high anti leukemic effect without triggering GVHD. Moreover, this expansion process is compatible with genetic manipulation, opening perspectives of future improvement of the cell product.

Disclosure: D.O. is a cofounder and shareholder of Imcheck Therapeutics, Alderaan Biotechnology, Emergence Therapeutics, and Stealth IO.

5 - Cellular Therapies other than CARs

P129 FIRST IN HUMAN CLINICAL RESPONSES AND PERSISTENCE DATA ON TEG001: A NEXT GENERATION OF ENGINEERED ALPHA/BETA T-CELLS TARGETING AML AND MM WITH A GAMMA9DELTA2 TCR

Moniek de Witte 1, Jojanneke Scheepstra1, Naomi Weertman1, Annet Daudeij1, Lotte van der Wagen1, Rimke Oostvogels1, Colin de Haar1, Henk-Jan Prins1, Carli Bartels-Wilmer1, Willem Dohmen1, Inez Johanna1, Zsolt Sebestyen1, Jurgen Kuball1, Trudy Straetemans1

1University Medical Center Utrecht, Utrecht, Netherlands

Background: γδT cells target malignant cells by sensing cancer mediated metabolic changes via their γδTCR. We have developed the TEG001 cell product, which are αβ T cells expressing the high affinity Vγ9Vδ2 TCR clone 5. TEG001 showed tumor reactivity towards a range of hematological malignancies. Here we present the results of dose level 1 and 2, addressing the safety and tolerability of TEG001 in patients with acute myeloid leukemia (AML) or multiple myeloma (MM).

Methods: The TEG001 study is a single center, investigator-initiated study, with a standard 3 + 3 dose escalation design. Cohorts receive a single infusion of 1x106 TEG001 cells/kg (dose level 1), 3x106 TEG001 cells/kg (dose level 2) and 1x107 TEG001 cells/kg (dose level 3). Relapsed/refractory AML/high risk MDS and MM patients were eligible. AML patients with > 30% blasts or circulating blasts required bridging therapy. TEG001 cells were produced as previously described (doi: 10.3389/fimmu.2018.01062). Conditioning consisted of fludarabine i.v. 25 mg/m2 (day -4 to -2) and cyclophosphamide i.v. 900 mg/m2 (day -2). Pamidronate (PAM) (30 mg i.v.) was administered on day 0 and 28 to propagate TEG001 activity. TEG001 cells were measured in peripheral blood by flowcytometry. The primary endpoint was the development of dose-limiting toxicities (DLTs) to determine the maximum tolerated dose. Main criteria for reaching DLTs were TEG001 related adverse events (AEs) requiring ventilator support and grade 4 AEs not related to the underlying disease. AEs contributable to cytokine release syndrome (CRS) and/or immune effector cell-associated neurotoxicity syndrome (iCANS) grade 3 responding to tocilizumab and/or steroids were not defined as DLT.

Results: Fourteen patients were included. Eight patients were not evaluable for the primary endpoint (insufficient response to bridging therapy (N = 3); abrogated TEG001 production (N = 2); TEG001 product not meeting the target dose (N = 2); COVID-19 hospital restrictions (N = 1)). In the six patients infused in dose level 1 and 2 (AML N = 5; MM N = 1), no DLTs were observed. No AEs ≥ grade 3 related to TEG001 were observed. Patient 16 (dose level 2), who received PAM according to the protocol on day 0 and 28, developed seizures at day 29 and 30. The seizure responded well to levetiracetam and no other neurological signs associated with iCANS were observed. No infections or abnormalities in the spinal fluid or brain imaging were identified and connection of the seizure to TEG001 remained inconclusive. In all patients TEG001 cells were detectable in peripheral blood. In dose level 1, patient 7 had stable disease (SD, duration 3M). In dose level 2, patient 9 had SD (duration 12M) and patient 16, who had 28% bone marrow blasts prior to TEG001 infusion showed full hematological recovery and achieved a complete remission (CR, duration 6M).

Conclusions: TEG001 cells are well-tolerated at the first 2 dose levels tested. TEG001 cells show persistence up to day 56. In the 5 AML patients, we observed 1 complete remission and 2 patients with stable disease. The TEG001 study is recruiting patients for dose level 3.

Clinical Trial Registry: NL6357

Disclosure: Research Grants 2019 KWF 12586 to MdW, JK, TS; KWF 13493 to MdW, JK.

JK is inventor on multiple patents dealing with gdTCRs, ligands, isolation strategies of engineered immune cells. JK is cofounder and shareholder of Gadeta (www.gadeta.nl). JK received research support from Gadeta.

5 - Cellular Therapies other than CARs

P130 DONOR-DERIVED ANTI-LEUKEMIA CTLS FOR THE CONTROL OF LEUKEMIA RELAPSE IN HIGH-RISK PEDIATRIC PATIENTS GIVEN HAPLOIDENTICAL HSCT

Matteo Tanzi1, Enrica Montini1, Agnese Rumolo1,2, Patrizia Comoli1, Antonia Moretta1, Giovanna Giorgiani1, Santina Recupero1, Tommaso Mina1, Gloria Acquafredda1, Jessica Rotella1, Stella Boghen1, Marco Zecca1, Daniela Montagna 1,2

1Foundation IRCCS Policlinico San Matteo, Pavia, Italy, 2University of Pavia, Pavia, Italy

Background: Adoptive cell therapy based on the infusion of donor-derived cytotoxic T lymphocytes (CTLs) able to recognize patients’ leukemia blasts (LB) is a promising approach to prevent or control leukemia relapse after allogeneic hematopoietic stem cell transplantation (HSCT). The success of this approach mainly depends on the ability to ex vivo generate high quality donor-derived anti-leukemia CTLs in compliance with Good Manufacturing Practice.

Methods: We have developed a procedure for generating large numbers of donor-derived CTLs directed against LB, through stimulation of CD8-enriched T lymphocytes with IFN-Dendritic Cells (DC) pulsed with apoptotic LB in the presence of IL-12, IL-7, IL-15 and IFN alfa2b during priming phase. After priming phase, CTLs underwent one or two rounds rapid antigen independent expansion in the presence of IL-2, IL-15, OKT3 and irradiated allogeneic feeder cells. Using this approach, 47 batches of PMTC were produced, in the Cell Factory of Fondazione IRCCS Policlinico San Matteo, for 20 pediatric patients given haploidentical HSCT for high-risk acute leukemia. The quality of each batch was evaluated in terms of microbiological safety and immunological efficacy.

Results: Variable numbers of CTLs were produced, depending from patients’ weight and numbers of infusions they have received. Expansion rate was very homogeneous and there were no significant differences between CTLs obtained after the first rapid expansion round (105 ± 21 fold the number of cells seeded at the beginning of the culture) or after the second round (118 ± 15 fold). Microbiological quality controls (CQ) demonstrated that all batches were sterile, were free of mycoplasma and adventitious viruses, and conformed to acceptable endotoxin levels. Biological CQ, including cell viability, identity, phenotype and potency were in compliance with the defined cut off. Fresh anti-leukemia CTLs displayed high percentages of cell viability (mean: 89% ± 6%), that were maintained after thawing (mean 80% ± 5%). The majority of effector cells were CD3 + /CD8+ (mean 73%, ± 5%), and functional evaluation demonstrated that all batches of CTLs were able to lyse patients’ LB (range 19-74% of specific lysis at effector:target ratio of 25:1), while the lysis against patients’ derived normal cells was lower (range 9-28% of specific lysis at the same ratio). Anti-leukemia CTLs were also able to secrete sizeable amounts of IFN gamma and TNF alfa in response to patients’ LB. So far, six patients received CTLs, on a compassionate base, for the treatment of post-HSCT molecular or hematological relapse, while six patients were enrolled in a phase I/II trial based on infusion of escalating dose of CTLs (starting from 5x10e4/kg up to 8x10e6/kg) every three weeks, for the prevention of leukemia recurrence. No severe adverse reactions, and no grade 2-4 toxicities, including occurrence of severe GVHD were recorded during follow-up. Preliminary results demonstrated that anti-leukemia CTLs might have a role in both prevention and treatment of post-haplo-HSCT recurrence, leading to long-term remission.

Conclusions: These results demonstrated that our protocol is highly reproducible and allows the generation of large numbers of safe and functional donor-derived anti-leukemia CTLs for the prevention/treatment of leukemia recurrence in high-risk pediatric patients.

Disclosure: Nothing to declare.

5 - Cellular Therapies other than CARs

P131 A SIX-YEAR RETROSPECTIVE ANALYSIS OF THE USE OF DONOR LYMPHOCYTE INFUSION IN A NATIONAL ALLOGENEIC TRANSPLANT CENTRE

Conor Browne 1, Christopher Armstrong1, Fiona Lynott1, Greg Lee1, Nicola Gardiner1, Carmel Waldron1, Robert Henderson1, Larry Bacon1, Elizabeth Vandenberghe1, Nina Orfali1, Catherine M. Flynn1, Eibhlin Conneally1, Patrick J. Hayden1

1St James’ Hospital, Dublin, Ireland

Background: Donor lymphocyte infusion (DLI) can be given prophylactically (Pro-DLI) in the setting of complete donor chimerism (DC) to prevent relapse, pre-emptively (Pre-DLI) for mixed chimerism (MC) or positive minimal residual disease (MRD), or therapeutically (T-DLI) as treatment for relapsed disease post allogeneic stem cell transplant (AlloSCT).

Methods: The records of consecutive AlloSCT recipients who received DLI between 1/1/2015 and 31/12/2020 were retrieved from the St. James’s Hospital transplant database. Responses to DLI, rates of graft versus host disease (GVHD), disease free survival (DFS) and overall survival (OS) were assessed. Statistical analysis was performed using SPSS version 28 and STATA.

Results: 472 AlloSCT’s occurred and 106 (22%) patients subsequently received DLI. Clinical details are shown in Table 1. Of these 106 patients, 16 (15%) received Pro-DLI; 8 for high risk disease with complete DC and 8 as part of the FLAMSA transplant protocol, 57 (54%) received Pre-DLI; 55 for MC and 2 for positive MRD assays and 33 (31%) received T-DLI. 204 DLI doses were infused. The median follow-up (FU) from first DLI was 29 months. Patients received a median of 2 (1-8) doses. The median interval from AlloSCT to the first DLI dose was 222 days. The median first dose of DLI was 0.32 (0.1-0.59) x107 CD3+ cells/kg. The median overall dose of DLI was 0.5 (0.1-5.16) x107 CD3 cells/kg. The median interval between DLI doses was 73 days.

48 (45%) DLI recipients developed GVHD following DLI. 41 (39%) patients developed acute GVHD (aGVHD); 12 (11%) Grade I and 29 (27%) Grade II-IV aGVHD. The median time to onset of aGVHD post-first DLI was 56 days. 22 (21%) developed chronic GVHD.

In the Pro-DLI cohort (n = 16), 5 (33%) patients relapsed post DLI at a median FU of 29 months.

In the Pre-DLI cohort (n = 57), DC rose in 51 (89%) patients. The median DC prior to DLI was 60% and 48 (84%) achieved complete DC post-DLI. 10 (18%) patients relapsed after Pre-DLI. The median time to relapse in both the Pro-DLI and the Pre-DLI cohorts was 13 months. In the T-DLI cohort (n = 33), 22 (67%) responded to DLI (complete remission (CR), partial remission or complete DC) of which 16 (48%) achieved CR. Of DLI responders, 11 (50%) relapsed at a median of 9 months.

DFS in the Pro-DLI and the Pre-DLI cohorts (Graph 1(A)) is estimated at 87.6% at 2 years and didn’t differ significantly between these groups (p = 0.09). OS in all 106 DLI recipients was estimated at 76.8% at 2 years (Graph 1(B)), survival being poorest at 44% in T-DLI at 30 months (p < 0.001).

In a univariate Cox regression analysis, response to DLI was a positive predictor for OS (p < 0.001), relapsed disease a negative predictor (p < 0.001) and aGVHD was not significantly associated with OS (p = 0.434).

Table 1. Patient Characteristics

 

N = 106

Transplant Characteristics

N = 106

Gender no. (%)

 

Conditioning Regimen no (%)

 

Male

61 (58)

Flu/Bu/ATG

46 (43)

Female

45 (42)

Flu/Mel/Alemtuzumab

19 (18)

  

BEAM Alemtuzumab

14 (13)

Median Age at DLI 1 years (range)

57 (22-70)

Bu/Cy

9 (9)

Disease Group: no. (%)

 

Cy/TBI

3 (3)

AML

41 (39)

FLAMSA-Bu

14 (13)

MDS/MPD

26 (25)

Other

5 (5)

ALL

   

NHL

4 (4)

  

HL

28 (26)

  

CLL

6 (6)

Disease Status at Transplant no. (%)

 
 

1 (1)

CR 1/2

62 (58)

  

CR 3/4

6 (6)

  

Partial Remission

22 (21)

Co-Morbidity Index no. (%)

 

Relapse/Refractory

4 (4)

HCT-CI 0-1

65 (61)

Unknown

12 (11)

HCT-CI > / = 2

41 (39)

  

Associated Treatment with DLI

 

GVHD Prophylaxis no. (%)

 

Pro/Pre-DLI no. (% of N = 73)

7 (10)

Ciclosporin

78 (74)

T-DLI no. (% of N = 33)

31 (94)

Tacrolimus

28 (26)

Conclusions: Pro-DLI and Pre-DLI resulted in complete DC in >80% of patients and a DFS estimate of 87.6% at 2 years. T-DLI, in combination with other agents, resulted in CR in 48% and an OS estimate of 44% at 2.5 years.

Disclosure: Nothing to Declare.

5 - Cellular Therapies other than CARs

P132 VIRUS SPECIFIC T-CELLS AGAINST ADV, EBV AND CMV IN INFECTED MULTI-TREATED POST-HSCT PATIENTS

Esther Hazane Leroyer 1, Danièle Bensoussan1, Maud D’Aveni1, Jean Hugues Dalle2, Nadine Petitpain1

1CHRU Nancy, Vandoeuvre Les Nancy, France, 2Hôpital Robert Debré APHP, Paris, France

Background: Post-transplant viral (adenovirus-ADV, cytomegalovirus-CMV or Epstein Barr virus-EBV) replication/disease remains a major cause of mortality in the context of allogeneic stem cell transplantation (allo-SCT). Despite the adequate use of antiviral drugs, Virus Specific T cells (VST) still represent a promising efficient immunotherapy.

Methods: Since 2016, we obtained the advanced-therapy medical products (ATMP) authorization to treat 29 recipients of allo-SCT in France with polyclonal ADV-VST (n = 13), CMV-VST (n = 7), EBV-VST (n = 8) and both ADV-VST and CMV-VST (n = 1). VSTs were generated in Nancy, by a 6-hour ex vivo stimulation with an appropriate peptide pool (PepT-ADV5 Hexon, -CMV pp65, -EBV Select, Miltenyi Biotec), and isolated by a CliniMACS immunomagnetic selection using the IFN-γ capture system (Miltenyi Biotec). Infused VST derived from the original stem cell donor (n = 13) or a third party with a haploidentical donor (n = 16). The clinical and biological follow-up was collected with frequent questionnaires addressed to physicians. The efficacy at 1-month has been evaluated. We defined complete response (CR) as complete clearance of the virus and partial response (PR) as viral load decrease ≥1log. We compared recipient outcomes with Student t, Mann-Whitney or Chi² tests where appropriated.

Results: From May 2016 to January 2022, we report 33 VST infusions for 29 patients (4 patients received 2 infusions). Median patients’age was 14 years old (range 0,3 to 68). The mean number of previous therapeutic lines was 2, 3 and 1 for ADV, CMV and EBV patients, respectively. The median CD3-IFN-γ T-cells (10e4/kg) infused was 0,25 (range 0,086 to 1,052) for ADV-VST, 1,01 (range 0,482 to 5,059) for CMV-VST and 0.79 (range 0,268 to 1,63) for EBV-VST (no significant difference). Four patients died within the first month. Responders (PR and CR) at 1 month represented 64% of the 25 patients still alive (p < 0.001). The mean viral load at the day of VST infusion (D0) and the median CD3 + T cells were not significantly different between CR/PR and non-responders (NR) patients. Interestingly, 61% responders benefited from haploidentical donor’s VSTs, different from the original stem cell donor. In vivo expansion data were available for 11 patients with 8 of them experiencing specific immune reconstitution. Six de novo cases of GVHD (23%) were reported during or after the first month post infusion and up to 3 months. Overall survival was 54% and 35% at 3 and 12 months post-infusion, respectively, and was significantly higher when D0 viral load was ≤3.75 log (p < 0.05). The first cause of death in the 3 months following VST infusion was the initial virus related replication/disease (n = 8). After 3 months, causes of death were HSCT toxicity related mortality (n = 4), septic shock (n = 2), mixed hepatopathy (ADV replication and hepatic GVHD) (n = 1), and relapse from initial disease (n = 2).

Conclusions: In this retrospective single-production-center cohort, almost 2/3 of patients responded to the last resort treatment, VST infusion. Adoptive transfer of VST cells seems to be a rapidly efficient and safe therapeutic option. Our results question its place in the early therapeutic lines, when viral load is low and expansion time greater.

Disclosure: Danièle Bensoussan, startup Seminov.

5 - Cellular Therapies other than CARs

P133 DONOR LYMPHOCYTE INFUSIONS (DLI) IN MYELOFIBROSIS: TREATMENT PATTERNS, EFFICACY AND SAFETY. A EUROPEAN MULTICENTRE RETROSPECTIVE REAL-WORLD STUDY

Alex Rampotas 1, Katja Sockel2, Catrin Theuser2, Martin Bornhaeuser2, Rafa Hernani3, Juan Carlos Hernandez Boluda3, Albert Esquirol Sanfeliu4, Daniele Avenoso5, Panagiotis Tsirigotis6, Marie Robin7, Tomasz Czerw8, Grzegorz Helbig9, Claire Roddie1, Jonathan Lambert1, Fotios Panitsas10, Donal McLornan1

1University College London Hospital NHS Foundation Trust, London, United Kingdom, 2University Hospital Dresden, Dresden, Germany, 3Hospital Clínico Universitario de Valencia, Valencia, Spain, 4Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 5King’s College Hospital NHS Foundation Trust, London, United Kingdom, 6Attikon University Hospital, Athens, Greece, 7Hôpital Saint-Louis, Paris, France, 8National Institute of Oncology, Gliwice, Poland, 9Medical University of Silesia, Katowice, Poland, 10Laikon University Hospital, Athens, Greece

Background: Allogeneic haematopoietic cell transplantation (allo-HCT) remains the only curative option for Myelofibrosis (MF). Relapse remains a significant problem in up to 20-30% of cases. Approaches to relapse prevention and management vary greatly including use of Donor Lymphocyte Infusions (DLI) representing one of the most effective strategies. Utilisation of DLI post allo-HCT is predominantly either an escalating dose regimen (EDR) or ‘bulk salvage’ therapy. Optimal timing and regimen choice remains undetermined. We hereby report on a multicentre MF allo-HCT cohort who received DLI.

Methods: Patient selection included MF allo-HCT patients from 8 European transplant centres who received DLI between 2005-2022. Response was assessed by IWG-MRT defined response (albeit this is unvalidated post-transplant). Kaplan-Meier estimator and log-rank test were used to estimate survival endpoints.

Results: Patient, disease and transplant characteristics are shown in Table 1. Median age was 58 years (IQR:53 – 62.5) and 19 (68%) were male. Karnofsky performance status was >80 in all patients and the majority had DIPPS+ Int-2 or High-risk disease at time of allo-HCT. Median time from diagnosis to allo-HCT was 23.4 months (IQR:8-44). Driver mutation status was JAK2; n = 18(64.3%), CALR; n = 2(7.1%), ‘Triple Neg’ n = 2(7.1%) and unknown n = 6(21.4%); 57.1% received a JAK inhibitor prior to allo-HCT. T cell depletion was utilised in 20(71.2%) regimens. Regarding donor type, 14 (50%) had a Matched Sibling Donor and 14(50%) an Unrelated Donor. The majority received PBSC (n = 26;92.9%), 11 patients had a history of acute and 8 patients had chronic GVHD, respectively.

Indication for DLI was a decrease in recipient chimerism (n = 13) or clinician defined relapse (n = 15; haematological (n = 11) and molecular relapse (n = 4). For the entire cohort, median time to DLI administration was 10.4 months (IQR:5.5-23.6 months). Median DLI doses administered was 2 (range, 1-5) with a median 1st dose of 1x10^6/kg (range 0.5 – 10). Of 16 patients receiving >1 dose of DLI, 12 were part of an EDR. Median follow up from 1st DLI was 55.4 months (IQR:27.7-96.5). Regarding response 15/28(54%) patients had a IWG-MRT defined response (unvalidated post-transplant); CR (n = 5), PR (n = 1) and clinical improvement (n = 9). Stable disease was reported in 6 patients with and no response or progression in 7. Chimerism levels improved in 16 patients. Where DLI was administered for falling chimerism, it was increased in 9/13 after a median of 2 infusions (range 1-4). 5 patients remained in remission with a median follow up 55 months. For clinical relapse, 9/15 patients had a response and remained in remission (median follow up: 42 months). DLI induced aGVHD was reported in 11 cases, grade 3/4 (n = 7).

Median OS from time of 1st DLI, censored at time of 2nd allo-HCT(N = 4), was 62.6 months (IQR:10- NR). Cumulative incidence of relapse/ progression after 1st DLI (death as competing) was 30.8% (95% CI 14.4-48.9%) at 6-months. Regarding cause of death, this was due to progressive disease (n = 8/14) or infectious complications (n = 3/14).

Table 40

Conclusions: Albeit a heterogeneous cohort, clear efficacy of DLI is evident in this challenging cohort of relapsed MF patients post allo-HCT. More prospective studies are warranted to identify the optimal DLI regimen and timing.

Disclosure: AR: Conference fees (Gilead).

5 - Cellular Therapies other than CARs

P134 INB-100: RELAPSE PROPHYLAXIS POST-HAPLOIDENTICAL BONE MARROW TRANSPLANTATION AND CYCLOPHOSPHAMIDE (HAPLO/CY) BY INFUSION OF DONOR-DERIVED EXPANDED/ACTIVATED GAMMA-DELTA (ΓΔ) T CELLS: A PHASE I TRIAL

Joseph McGuirk 1, Sunil Abhyankar2, Trishna Goswami3, Rupal Soder4, Mariska ter Haak3, Tyce Bruns5, Samantha Langford-Youngblood3, Lawrence Lamb3

1University of Kansas, Westwood, United States, 2University of Kansas Cancer Center, Westwood, United States, 3In8bio, Inc, New York, United States, 4Midwest Stem Cell Therapy Center, Kansas City, United States, 5University of Kansas Medical Center, Fairway, United States

Background: Effector gd T cells, are nature’s anti-cancer cell that can recognize and directly kill malignant cells in a non-MHC restricted manner. Relapse post Haplo/Cy bone marrow transplantation (BMT) occurs in ~51% of patients within 1 year and may be mitigated by infusing expanded, activated donor-derived haploidentical gd T cells (EAGD) post-transplant. Additionally, data suggest post-transplant Cy may lead to poor immune reconstitution and higher risk of infections. We report data on the first two dose levels (DL) of a Phase 1 trial assessing the safety and efficacy of allogeneic EAGD cells. Increasing doses of EAGD are administered to patients undergoing haploidentical transplants for underlying hematologic malignancies to prevent relapse, with corresponding preliminary clinical and biologic correlative findings.

Methods: Adult patients with newly diagnosed or relapsed ALL, CML, AML undergoing consolidative haploidentical transplant with reduced-intensity flu/cy/TBI conditioning received EAGD intravenously within 7 days of neutrophil engraftment. Peripheral blood was collected at EAGD infusion and monthly thereafter through day +90, with additional collections every 6 months through 1 year. Primary endpoints include dose-limiting toxicities (DLT), grade 3-4 adverse events including graft-versus-host disease (GvHD) while secondary endpoints include relapse and overall survival. Biologic parameters included multiparameter flow cytometric immunophenotyping and single-cell cytokine analysis of the EAGD graft. Peripheral blood analysis includes leukocyte count and differential, immunophenotyping, and serum Th1/Th2/Th17 cytokine analysis.

Results: 10 patients have been enrolled with four treated at DL1 of 1 x 106 EAGD/kg and two additional patients treated at DL2 (3 x 106 EAGD/kg). One screen failure, one manufacturing failure, one patient died prior to dosing, and one received an out of study specification product. All four patients in DL1 remain in morphologic complete remission (CR) at 32.2, 29.8, 18.1 and 3.8 months post-BMT at data cut-off of 18Dec2022. One patient received intermittent hypomethylating therapy for occurrence of recipient chimerism. Majority of toxicities related to EAGD cells were grade (Gr) 1-2 skin and Gr2 GI acute GvHD. Other toxicities include constipation, CMV reactivation, emesis, fatigue, and hypomagnesaemia. One patient sustained chronic mild GvHD of her oral surface. No DLTs, treatment-related ≥ Gr3 adverse events, neurotoxicity or cytokine release syndrome were reported. Significant peripheral lymphodepletion persisted through the first 100 days post-BMT followed by slow recovery of CD3 + CD4 + , CD3 + CD8+ and gd T cells. Interestingly, B cell counts recovered in the first 30 days and NK cells remained within the low normal range throughout. T cells transitioned from a CD45 + CD27-effector phenotype to CD45RA CD27 central to effector memory phenotype as recovery progressed. CD3, CD4, CD25, and FoxP3 Treg cells remained <3% of circulating T cells. Preliminary serum cytokine and chemokine analysis revealed predominant expression of IFNγ, IL-12p70, IP-10, RANTES and TNFα.

Conclusions: The preliminary results of this Phase 1 study of donor-derived, EAGD cells has demonstrated the therapy to have a robust safety profile, with no infusional toxicity, treatment associated SAEs, or grade 3-4 acute or extensive chronic GVHD. The absence of relapse suggest the possibility that this therapy will be an effective measure in mitigating relapse after HaploBMT.

Clinical Trial Registry: NCT03533816

Disclosure: Nothing to declare.

5 - Cellular Therapies other than CARs

P135 ROLE OF HISTOCOMPATIBILITY, ACCESSORY CELLS AND FREEZE-THAWING FOR IMMUNOSUPPRESSION BY PLACENTA-DERIVED DECIDUA STROMAL CELLS

Olle Ringden 1, Myrese Witkamp1, Dominik Schefberger1, Anna Arbmann1, Behnam Sadeghi1

1Karolinska Institutet, Huddinge, Sweden

Background: Mesenchymal stromal cells (MSCS) are very rare precursor cells ( < 1;10000nucleated cells) in all tissues of the body. MSCs from all tissues, even fibroblasts are immunosuppressive. We introduced bone marrow derived (BM)-MSCs for treatment of severe acute GVHD. Some, but not all patients responded. The fetus is protected from the mothers HLA haploidentical immune system by the placenta. Fetal membrane derived so called decidua stromal cells (DSCs) gave a stronger immunosuppression in vitro in Mixed Lymphocyte Culture (MLC) and responses in acute GVHD than BM-MSCs or other sources of MSCs. DSCs differ in many ways from BM-MSCs, they are half the size and do not differentiate well to bone and fat. DSCs inhibit MLC by direct contact and not mainly by soluble factors as do BM-MSCs. Blocking experiments suggest that interferon-gamma, prostaglandin E-2, IDO and PD-L1 are involved in the immunosuppressive mechanism. DSCs have a stronger effect on coagulation and hemostasis compared to BM-MSCs. DSCs have been successfully used to treat acute GVHD, hemorrhagic cystitis, ARDS and radiculomyelopathy.

Methods: Placentas were obtained from caesarean sections. DSCs were isolated and expanded from the fetal membranes in a GMP-lab. Mouse(m)- MSCs were obtained from femur and tibia. In MLC, m-splenocytes or human(h) PBL were used as responder cells. Stimulatory cells were m-splenocytes or h- PBL pooled from five donors. Con A was also used for stimulation. In trans-well experiments, DSCs were seeded in the upper chamber. CD3 + T-cells, CD14+monocytes and CD56 + NK cells were isolated by positive selection. Cells were stored frozen in liquid nitrogen. Viability was measured in room temperature.

Results: When 10% m-MSCs were added to MLCs, being autologous, allogeneic, or haploidentical (F1), > 95% inhibition was seen. Using h-DSCs(xeno), inhibition was median 68%. Using h-PBL as stimulators of m-solenocytes-hDSCs showed a median inhibition of 88%. Trans-well experiments indicated that h-DSCs needed direct contact for inhibition to occur. CD14+cells reduced the immunosuppressive effect by DSCs. NK cell activation by IL-2 was increased by a median of 58% when DSCs were added. Fresh or frozen-thawed DSCs had the same viability and similar immunomodulatory effects.

Conclusions: Histocompatibility, direct contact and CD14+cells had impact on DSCs immunomodulation, but freeze-thawing did not.

Disclosure: No conflict of interest.

5 - Cellular Therapies other than CARs

P136 RAPID IMMUNE RECONSTITUTION AND ELEVATED REGULATORY T CELL FREQUENCIES IN PATIENTS TREATED WITH ORCA-T

Arpita Gandhi1, M. Scott Killian2, Anna Pavlova3, Cameron Bader3, Sean Summers2, Fernando Teque2, J. Scott McClellan 2, Nathaniel Fernhoff2, Everett Meyer3

1Oregon Health and Science University, Portland, United States, 2Orca Bio, Menlo Park, United States, 3Stanford University, Stanford, United States

Background: Immune reconstitution following myeloablative allogeneic hematopoietic stem cell transplant (MA-alloHSCT) is significantly delayed for T cell depleted allografts when compared to T-cell replete allografts, a feature that has been implicated in higher rates and grades of infection, less GVT, and worse overall survival. Orca-T is a high precision cell therapy currently being investigated for the treatment of certain hematological malignancies otherwise treated with MA-alloHSCT. The cellular drug products of Orca-T (HSPCs, Tregs, and Tcons) are administered at high purity, in controlled doses, and on an established schedule with the intent to reconstitute the blood and immune system while controlling GVHD. Here, we present data on the immune reconstitution in 100 adult patients who received Orca-T.

Methods: In the context of an ongoing multicenter Phase Ib clinical trial of Orca-T in recipients with hematologic malignancies (NCT04013685), we performed longitudinal measurements (days -28, 28, 56, 100, 180, and 365 post-transplant) of immune reconstitution in the first 100 consecutive patients. With fresh whole blood, clinical 5-part leukocyte differentials were performed at clinical sites, and lymphocyte subset frequencies were measured by flow cytometry in a central lab. Principal component analysis (PCA) was performed to investigate potential associations with recipient sex (male vs. female) and donor relation (related vs. unrelated).

Results: T cell and B cell counts were readily observed at days 28 and 56 respectively, and increased with each subsequent time point. Median NK cell levels were observed to be in the normal range at all post-transplant time points. CD4 + T cell and Treg cell counts exhibited similar post-transplant patterns with both being appreciably present at d28 and increasing with each subsequent time point. Strikingly, relative to the level measured in 75 corresponding PBSC donors, the Treg frequency among CD4 + T cells was significantly elevated at all time points post-transplant. Median CD8 + T cell counts increased for the first 6 months post-transplant and then plateaued in the normal range. Upon PCA, very few significant differences were observed in 2-group comparisons of recipient sex and donor relation.

Conclusions: Orca-T patients exhibit early immune reconstitution of each of the major leukocyte and lymphocyte subsets hypothesized to control relapse and infection. Concomitantly, elevated Treg frequencies were also observed. This feature of immune reconstitution profiles of Orca-T recipients may be correlated to the reduced occurrence and severity of acute and chronic GVHD in these patients (Oliai et al., ASH 2022). Similar immune reconstitution profiles were observed in patients of disparate sex and regardless of donor relation. Prospective comparisons of immune reconstitution between Orca-T and standard-of-care patients will be performed in our ongoing phase 3 clinical trial (NCT05316701).

Clinical Trial Registry: https://clinicaltrials.gov/ct2/show/NCT04013685

Disclosure: Arpita Gandhi | Consultancy: Care Dx; Honoraria: Gamida; Research Funding: Orca Bio

M. Scott Killian | Current Employment/Equity: Orca Bio

Anna Pavlova | Current Employment/Equity: Orca Bio

Cameron Bader | Nothing to declare

Sean Summers | Current Employment/Equity: Orca Bio

Fernando Teque | Current Employment/Equity: Orca Bio

J. Scott McClellan | Current Employment/Equity: Orca Bio

Nathaniel Fernhoff | Current Employment/Equity: Orca Bio

Everett Meyer | BoD/Advisor: indee labs; Co-founder & Scientific Advisor: GigaGen, Triursus Therapeutics; Research Funding: Orca Bio

This study was sponsored by Orca Bio.

5 - Cellular Therapies other than CARs

P137 MESENCHYMAL STROMAL CELLS FOR THE TREATMENT STEROID REFRACTORY GRAFT-VERSUS-HOST DISEASE IN CHILDREN

Krzysztof Czyzewski 1,2, Robert Debski1,2, Monika Richert-Przygonska1,2, Agnieszka Majk2, Kamila Jaremek2, Oliwia Grochowska2, Joanna Stankiewicz1,2, Lukasz Ledzinski3, Sylwia Koltan1,2, Jan Styczynski1,2

Medicum, Nicolaus Copernicus University in Torun, Bydgoszcz, Poland, Antoni Jurasz University Hospital No 1, Bydgoszcz, Poland, 3Central Pharmacy, Antoni Jurasz University Hospital No 1, Bydgoszcz, Poland

Background: Mesenchymal stromal (MSC) cells are multipotent non-hematopoietic stem cells, that can differentiate into other cells. MSC have immunoregulatory and immunosuppressive activity by secretion of chemokines, cytokines and extracellular vesicles. The first-line treatment of acute and chronic graft versus host disease (GVHD) are corticosteroids. However 50% of acute GVHD cases are resistant to high doses of corticosteroids and 50-60% moderate/severe chronic GVHD require “steroid-sparing” treatment. MSC is an option for next lines treatment of steroid refractory (SR) GVHD.

The objective of the study was the analysis of safety and efficacy of MSC use in pediatric patients with SR GVHD in single center experience.

Methods: We retrospectively analyzed the use of cryopreserved commercially available third-party donor Wharton’s jelly derived mesenchymal stromal cells in patients with SR GVHD after allogeneic hematopoietic stem cell transplantation (allo-HSCT).

Results: Fifteen children (11 boys, 4 girls) with median age 4 years (age range 8,5 months-16,5 years), diagnosed for ALL (n = 5), primary immunodeficiencies (n = 5), AML (n = 3), SAA (n = 2) recipients of allo-HSCT (9 MUD-PB, 4 MUD-BM, 2 MSD-PB) were included into the study. In all patients SR GVHD was diagnosed (12 acute GVHD, grade 2-4: 4-skin+gastro-intestinal,3- skin, 2-gastro-intestinal, 1-skin+lungs, 1-hepatic+gastro-intestinal, 1-skin+hepatic+gastro-intestinal; 3 chronic GVHD, grade moderate-severe: 1-skin, 1-sclerodermatous, 1-skin+eye). The total number of 45 MSC intravenous infusions were performed (median: 2, range: 1-8). The median number of infused MSCs was 1.39x106/kg b.w. (range 0.85-2.99) in 1-4 weeks intervals. In 14/15 patients before MSC administration combined therapy with corticosteroids was performed including 1-3 agent/therapy (ciclosporine, tacrolimus, mycophenolate mofetil, vedolizumab, ruxolitinib, basiliximab, extracorporeal photopheresies). The median time from GVHD diagnosis to first infusion of MSC was 105 days (range 15-305 days); 73% (11/15) achieved a response (53% complete response, 20% partial response), while 27% (4/15, all acute GVHD) did not respond to MSC therapy. Overall, 91% (10/11) of patients with complete/partial response are alive (1 patient died due to leukemic relapse). Among 4 non-respond patients, 2 died due to GVHD progression, 1 due to infection and 1 due to toxic encephalopathy. No adverse event related to MSC administration was observed in all patients.

Conclusions: Administration of MSC is a safe and effective therapeutic option in our experience in pediatric patients with steroid refractory acute or chronic GVHD. Further studies concerning optimal dose, interval between doses and number of doses are needed.

Disclosure: Nothing to declarate.

5 - Cellular Therapies other than CARs

P138 FAMILIAL CD45RA¯ CELLS CONTAINING PATHOGEN-SPECIFIC MEMORY T-CELLS AS COMPLEMENTARY STRATEGY TO TREAT SEVERE REFRACTORY INFECTIONS AND EBV LYMPHOPROLIFERATIVE DISEASES IN IMMUNOCOMPROMISED PATIENTS

Carlos Echecopar Parente 1, Karima Al-Akioui Sanz2, Maria Dolores Corral Sánchez1, Luisa Sisinni1, David Bueno Sánchez1, Cristina Ferreras2, Raquel De Paz3, Antonio Balas4, Eduardo López-Granados5, Berta González Martínez1, Víctor Galán Gómez1, Pilar Guerra García1, Sonsoles San Román Pacheco1, Isabel Martínez Romera1, Antonio Pérez Martínez1,2,6

1Pediatric Hemato-Oncology, University Hospital La Paz, Madrid, Spain, 2University Hospital La Paz Research Institute (IdiPAZ), Madrid, Spain, University Hospital La Paz, Madrid, Spain, 4Histocompatibility Unit, Centro de Transfusión Comunidad de Madrid, Madrid, Spain, 5Immunology Unit, University Hospital La Paz, Madrid, Spain, Universidad Autónoma de Madrid, Madrid, Spain

Background: Immunocompromised patients are susceptible to high-risk opportunistic infections and malignant diseases. Most antiviral and antifungal drugs are quite toxic, relatively ineffective, and induce resistance in the long term. The adoptive passive transfer of pathogen-specific cytotoxic T-cells has shown a minimal toxicity profile and effectiveness in treating several infections, but this therapy have the main limitations of regulatory issues, high cost, and absence of public cell banks. However, CD45RA¯ cells containing pathogen-specific memory T-cells involve a less complex manufacturing and regulatory process and are cheaper, feasible, safe, and potentially effective.

Methods: We present preliminary data from six immunocompromised patients: four who had severe infectious diseases and two who had EBV lymphoproliferative disease. All of them underwent multiple safe familial CD45RA¯ T-cell infusions as adoptive passive cell therapy, containing pathogen-specific memory T-cells.

Table 41

Results: The infusions were safe, there was no case of graft-versus host disease and they showed a clear clinical benefit. The patients treated for BK virus nephritis, Cytomegalovirus encephalitis, Cytomegalovirus reactivation, and pulmonary aspergillosis experienced pathogen clearance, complete resolution of symptoms in 4-6 weeks and a lymphocyte increase in 3 of 4 cases after 3–4 months. Donor T cell transient microchimerism was detected in one patient. The two patients treated for EBV lymphoproliferative disease underwent chemotherapy and CD45RA¯ memory T-cells containing EBV cytotoxic lymphocytes. Donor T-cell microchimerism was observed in both patients. The viremia cleared in one of the patients, and in the other, despite the viremia not clearing, hepatic lymphoproliferative disease remained stable and was ultimately cured with EBV-specific cytotoxic T Lymphocytes.

Conclusions: The use of familial CD45RA¯ T cells containing specific cytotoxic T-lymphocytes could be a feasible, safe and potential effective approach for treating severe pathogen infections in immunocompromised patients through a third party donor. Furthermore, this approach might be of universal use with fewer institutional and regulatory barriers.

Disclosure: Nothing to declare.

5 - Cellular Therapies other than CARs

P139 ADD-BACK NK CELLS FOLLOWING HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANT IN PEDIATRIC HIGH-RISK B CELLS ACUTE LYMPHOBLASTIC LEUKEMIA (B-ALL)

Daniela D’Angelo1, Fabiana Cacace2, Emanuela Rossitti1, Valeria Caprioli2, Maria Rosaria D’Amico2, Giuseppina De Simone2, Mario Toriello2, Maisto Giovanna2, Donato Madalese2, Alessandra Cianflone2, Peppino Mirabelli2, Roberta Penta de Vera d’Aragona2, Francesco Paolo Tambaro 2

1University of Naples “Federico II”, Napoli, Italy, 2AORN Santobono Pausilipon, Napoli, Italy

Background: Natural killer lymphocytes (NKs) are cells of innate immunity that protect against tumor and infections, and recognize self and non-self HLA class I ligands. Killer-cell Immunoglobulin-like Receptors (KIRs) are inhibitory molecules expressed on NK surface. They act “educating” NKs in effective defense against unhealthy self/non-self cells and tolerance towards healthy self cells. NKs play a key role in preventing early relapse of hematologic malignancies, virus infection with no graft versus host disease (GvHD) after haploidentical hematopoietic stem cell transplantation (haplo-HSCT).

In B-/T-cell alpha/beta depleted haplo-HSCT, donor-versus-recipient alloreactive NKs exert an efficient graft versus leukemia (GvL) effect, mediated by inhibitory KIRs where the HLA ligand is missing on recipient cells.

NKs after haplo-HSCT are immature, CD56bright, derived from CD34+ stem cells. They are not effective as the mature NKs with a CD56dim phenotype, characterized by CD16, CD57 and KIRs in late stages.

We evaluated the alloreactive effect of donor-derived NKs in 3 pediatric high-risk B-ALL who underwent B-/T-cell alpha/beta depleted haplo-HSCT followed by donor NK infusion.

Methods: Patients’ characteristics are summarized in the Table. Median follow-up was 180 days after transplantation. All 3 patients were male; median age was 13 months. Patients 1 and 2 had infant ALL, patient 3 had refractory-relapsed (r/r) ALL. All patients received a median of 3 lines chemotherapy. Patient 1 received a prior allo-HSCT. Two patients received chemo-based conditioning regimens. Patient 3 received Total Body Irradiation (TBI)-based conditioning regiment plus chemotherapy. All patients received rituximab and letermovir for EBV and CMV prophylaxis, respectively. Mother was Donor for all patients. Median CD34+, CD19+, CD3+-alpha/beta cell dose was 8.8, 0.05, 0.41 x 106/kg, respectively. NKs alloreactivity was tested by evaluating KIR genotyping and KIR/KIR-ligand mismatch in graft-versus-host direction. All three patients present KIR/KIR-ligand mismatch. NK reconstitution was analyzed in all three transplanted patients.

Results: Engraftment was day +13, +14 and +19 for ANC, and +14, +18 and +17 for PLT, for patients 1, 2 and 3, respectively. All three patients showed a fast reconstitution of NKs at day +14. Patients received NKs infusion at median day +36. No infusion reactions were observed. Median NK infused were 256x106/kg. None of the patients developed EBV reactivation. Patient 1 developed CMV and ADV reactivation at day +19 and +288, respectively, without needing to treat. Patient 2 developed Grade 2 acute GvHD (skin) at day +48, treated by steroid, and Score 3 chronic GvHD (skin) 5 months after transplantation, resolved with cyclosporin treatment. No Grade GvHD 3-4 was observed. Currently all patients are in CR with full donor chimerism.

Conclusions: Alloreactive NKs add-back infusion after B-/T-cell alpha/beta depleted haplo-HSCT could represent an effective option to improve transplant outcome. NKs may improve efficacy by enhancing GvL and contribute to maintain “disease control” through immunologic surveillance in high-risk ALL. Infusion of phenotypic mature donor NKs could be more useful to maintain the immunologic control of MRD and virus reactivation, without increase aGvHD incidence. Our preliminary observations should be confirmed in clinical trial of a larger number of patients and extended to other haplo-HSCT platforms.

Disclosure: Nothing to declare.

5 - Cellular Therapies other than CARs

P140 EVALUATION OF THE SINGLE NEEDLE/DOUBLE NEEDLE AMICUS EXTRACORPOREAL PHOTOPHERESIS (ECP) SYSTEM WITH SOFTWARE VERSION 6.1 IN HEALTHY HUMAN SUBJECTS

Axel G. Stover1, Danielle Badgley1, Cheryl Huber1, Adrienne Karpiel1, Gregory Coultas1, Carrie Pineda1, Zahra Ali1, Jennifer Weitgenant1, Barbara Brantigan1, Divya Srivastava1, Thomas Gniadek 1

1Fresenius Kabi USA, LLC, Lake Zurich, United States

Background: The Amicus Blue Separator System for ECP (Fresenius Kabi), performs on-line photoactivation and reinfusion of mononuclear cells (MNCs). This study evaluated safety and performance of a new disposable kit enabling single- (SN) or double-needle (DN) vascular access and new device software (SW).

Methods: This Institutional Review Board approved study enrolled healthy adults who met safety guidelines for whole blood and apheresis donations. MNCs were collected with the Amicus Separator (SW 6.1), sampled from the Amicus return line during the collection as well as before (untreated) and after addition of 8-MOP/photoactivation (treated) with the Phelix device (SW 2.1); for this study, treated cells were not reinfused to the donors. Whole-blood-to-process was set at 500, 2000 or 5000 mL. Conversions (SN-to-DN or DN-to-SN) were also performed. Testing included cell counts (KX-21, Sysmex), complement activation (C3a, C5a immunoassays, Quidel), plasma free hemoglobin (modified Harboe method), Factor VIII activity (ACL Elite, Werfen), and lymphocyte proliferation (phytohemagglutinin culture/carboxyfluorescein succinimidyl ester labeling). Apoptosis was assessed from cultured cells based on phosphatidylserine expression by Annex V/CD2/CD19/7AAD labeling and flow cytometry.

Results: 22 subjects enrolled (16 male/6 female, mean age 52.0 [range 22-74] years) for 20 evaluable procedures. Two procedures could not be completed: one due to vein infiltration and another pre-procedure hemoglobin below the cutoff for volunteer blood donation (a study-specific requirement). No device errors or alarms prevented procedure completion. Mean procedure time was 88.1 minutes (SD 32.27). Lymphocyte apoptosis after 72-hour culture was 83.59% (SD 7.124%) for treated vs. 10.48% (SD 2.881%) for untreated cells, p = 0.0002. Lymphocyte proliferation over 72 hours post-treatment was 0.90% (SD 0.532%) for treated cells vs. 73.14% (SD 11.003%) for untreated cells, p < 0.0001. Free hemoglobin measured in the return line as well as treated cells was below safety limits and activated complement was not significantly different between treated cells and return line samples. Mean Factor VIII activity was lower in the treated samples vs. the return line: 46.1% (SD 15.88%) vs. 126.6% (SD 59.61%), p < 0.0001. Subject Factor VIII activity did not differ significantly pre vs. post collection procedure (146.9%, SD 53.77 vs. 135.5%, SD 48.38). All parameters met the acceptance criteria. Overall, MNC collection efficiency (CE1) was equivalent between DN (75.31%, SD 9.702%) and SN procedures (79.24%, SD 7.576%), p = 0.5321; as well as between DN and conversion procedures (78.93%, SD 6.996%), p = 0.4902.

Parameter

Procedure type

N

Mean (SD)

Lower 95% confidence limit of the mean

Lower bound, tolerance interval (95% confidence, 90% population capture)

Lymphocyte proliferation inhibition (%)

All procedures

20

98.77

98.66

97.75

Single needle

8

98.84

98.68

97.81

Double needle

6

98.53

98.31

97.22

Conversion

6

98.91

98.74

97.96

Lymphocyte apoptosis (%

All procedures

20

83.59

80.26

69.87

Single needle

8

80.80

74.43

61.13

Double needle

6

87.38

81.19

69.66

Conversion

6

83.52

76.31

62.88

ParameterProcedure type

N

Mean (SD)

Median

Min, Max

MNC collection efficiency (CE1, %)

All procedures

20

77.96 (7.873)

80.32

62.0, 87.8

500 mL

7

84.89 (2.980)

85.91

80.5, 87.8

2,000 mL

7

76.61 (6.853)

78.78

64.2, 83.5

5,000 mL

6

71.46 (6.928)

70.59

62.0, 81.6

Conclusions: This study evaluated 20 simulated ECP procedures with the Amicus ECP system (SN, DN and conversion). Only one subject experienced an adverse event (vein infiltration). Overall, the MNC collection efficiency was satisfactory and comparable to published reports. Photoactivation consistently reduced lymphocyte proliferation and induced apoptosis over the course of 72 hours. Pre- and post-procedure subject laboratory evaluation indicated no clinically significant abnormalities. Taken together, this study does not suggest any new procedural risks or hazards while demonstrating the safety and performance of this new option for single-needle vascular access as well as updated device software.

Clinical Trial Registry: n/a

Disclosure: All authors are employed by the device manufacturer.

29 - Chronic Leukaemia and Other Myeloproliferative Disorders

P141 TKI IN RELAPSE PROPHILAXIS AFTER ALLOGENEIC TRANSPLANTATION IN CML/ADVANCED PHASES

Antonella Gozzini 1, Riccardo Boncompagni1, Chiara Nozzoli1, Ilaria Cutini1, Scappini Barbara1, Riccardo Saccardi1

1AUO Careggi, Florence, Italy

Background: The revolutionary clinical activity of tyrosine kinase inhibitors (TKIs) in chronic myeloid leukemia (CML) has transformed patient outcome. Consequently, allogeneic stem cell transplantation (allo-SCT) is no longer the only treatment modality with the ability to deliver long-term survival. In contrast to the central position it held in the treatment algorithm 20 years ago, allografting is now largely reserved for patients with either chronic-phase disease resistant to TKI therapy or advanced-phase disease. The major causes of transplant failure in patients allografted for CML are transplant toxicity and disease relapse. The role of TKIs in relapse preventing after allo-SCT is still to be defined.

Methods: Between 2016 and 2022, 11 patients affected by CML advanced phases received allo-SCT. 10/11 were blast crisis, one was a chronic phase, primary cytogenetic resistant to second and third generation TKis. We report the main clinical characteristics of patients at time to transplant in Table 1. All 10 CB patients treated with TKis obtained CP2, no t315I mutation or compound mutations were identified. The transcript level was <1% for 9/11 patients at the transplant. 8/11 patients received myloablative conditioning regimen. 6/11 patients received TKIs prophylaxis with Ponatinib 15 mg/die: the therapy started within 60 days after SCT for 2 years, if well tolerated.

AGE (at HCT)

median (c.i.)

44,8 (27 - 63)

 

TKI pre transplant

Imatinib

2

18%

 

Dasatinib

6

54%

 

Ponatinib

4

36%

Status at transplant

CP2

10

90%

 

CP1

1

10%

Conditioning regimen

TBF MAC

8

72%

 

TBI Cy

1

6%

 

TBIFlu200

1

6%

 

TBF RIC

1

6%

HCT-CI

low

7

63%

 

intermediate

3

27%

 

high

1

10%

MATCH HLA

ID sibling

3

28%

 

haploidentical

4

36%

 

MUD

4

36%

TKI post SCT

Ponatinib

6

55%

Results: NRM was 27%. 1 patient died because of relapse, 1 patient died for infectious event and one for acuteGVHD. No patients had molecular relapse and no patients received DLI infusion. With a median follow up of 796 days, 72% (8/11) of patients are alive in deep molecular response with fully donor chimerism. Acute GVHD incidence was 45%; chronic GVHD (only grade 1-2) developed in 30% of patients.

Conclusions: Thirty to 70% relapse are reported after Allo-SCT for CML. Several factors determine the risk of disease recurrence after a transplant, including AP disease at SCT and the use of a RIC regimen. DLI remains the most effective salvage therapy in patients relapsing after allograft, even though the most of data originated from patients allografted in first CP.

Post-transplant administration of TKIs has the potential to reduce disease relapse, but selection of drug and administration scheduling is still to be clarified.

Disclosure: Nothing to declare.

29 - Chronic Leukaemia and Other Myeloproliferative Disorders

P142 FEASIBILITY OF EARLY DOSE REDUCTION OF FRONTLINE DASATINIB IN CHRONIC MYELOID LEUKEMIA PATIENTS: A 3-YEAR FOLLOW UP RESULTS OF DAS-CHANGE STUDY

Dong-Yeop Shin 1, Sung-Eun Lee2, Jee Hyun Kong3, Sukjoong Oh4, Hawk Kim5, Sung Hyun Kim6, Young Rok Do7, Won Sik Lee8, Jae-Yong Kwak9, Dae Young Zang10, Seong Nam Im11, Dong-Wook Kim12

1Seoul National University hospital, Seoul, Korea, Republic of, 2Seoul St. Mary’s Hospital, The Catholic University, Seoul, Korea, Republic of, 3Wonju College of Medicine, Yonsei University, Wonju, Korea, Republic of, 4Hanyang University College of Medicine, Seoul, Korea, Republic of, 5Gachon University Gil Hospital, Incheon, Korea, Republic of, 6Dong-A University College of Medicine, Busan, Korea, Republic of, 7School of Medicine, Keimyung University, Daegu, Korea, Republic of, 8Inje University College of Medicine, Inje University Busan Paik Hospital, Busan, Korea, Republic of, 9Jeonbuk National University medical School, Jeonju, Korea, Republic of, 10Hallym University College of Medicine, Anyang, Korea, Republic of, 11Haeundae Paik Hospital, Inje University College of Medicine, Busan, Korea, Republic of, 12Uijeongbu Eulji Medical Center, Eulji University, Uijeongbu, Korea, Republic of

Background: Dasatinib is a potent second generation tyrosine kinase inhibitor (TKI) used for a first line treatment option for patients with chronic myeloid leukemia (CML). Dasatinib has high rates of deep molecular responses (MRs) compared to the first generation TKI imatinib. However, dose reduction due to adverse events (AEs) including cytopenia and pleural effusion is common in patients treated with dasatinib.

Methods: We performed a prospective clinical trial which enrolled patients with newly diagnosed chronic phase CML who had initiated dasatinib as the frontline treatment, experienced dasatinib-related AEs within 12 weeks from the start of dasatinib treatment, and CHANGEd their treatment with reduced dose of 80mg daily were enrolled (DAS-CHANGE study, NCT04150471). For efficacy evaluation, BCR-ABL1/ABL1 levels were measured at screening and every 3 months during treatment by internationally standardized real-time quantitative polymerase chain reaction assay in a central laboratory. The primary endpoint was to evaluate MMR at 12 months. Data of molecular response rates and AEs were collected during a 3 year follow-up period.

Results: A total of 90 patients were screened, and 81 patients were enrolled into the study. Eight patients failed screening because BCR-ABL1 transcript was not decreased below 10% at 3 months and 1 patient withdrew consent. Median age was 52 years, and male genders were 54.3%. Every patient has Korean ethnicity. 18.5% of patients had high Sokal risk score. BCR-ABL1/ABL1 was dropped to a median 0.032% by international scale (IS) at 12 months. Although patients who experienced dose reduction due to AEs within 3 months were enrolled in this study, most patients (96.3%) were able to complete 12 cycles of dasatinib treatment, which suggested that early dose reduction strategy in patients treated with dasatinib could improve the tolerability of continuous dasatinib treatment. Cumulative major molecular response (MMR) rates were 71% at 12 months, and reached up to 85% and 94% at 24 and 36 months, respectively. MR4.0 (BCR-ABL1/ABL1 ratio ≤ 0.01% by IS) and MR4.5 (BCR-ABL1/ABL1 ratio ≤ 0.0032% by IS) were 24% and 11% at 12 months, 46% and 23% at 24 months, 65% and 44% at 36 months, respectively. Most common grade 3/4 AEs were thrombocytopenia (27.2%), and neutropenia was followed by (11.1%). Pleural effusion was occurred in 35.8% of patients, grades were 1 or 2 in most cases. Only one patient (1.2%) experienced grade 3 pleural effusion in our study.

Conclusions: Our results suggests that early dose reduction of dasatinib due to adverse events is safe and feasible. Early dose reduction does not compromise efficacy in patients with newly diagnosed CML. A proactive reduction of dasatinib in the early period in newly diagnosed CML could improve the tolerability of dasatinib without compromising the efficacy.

Clinical Trial Registry: NCT04150471

Disclosure: Nothing to declare.

29 - Chronic Leukaemia and Other Myeloproliferative Disorders

P143 THE CLINICAL VALUE OF NEUTROPHIL-TO-LYMPHOCYTE RATIO AT THE DIAGNOSIS OF MYELOPROLIFERATIVE NEOPLASM

Jong-Ho Won 1, Seug Yun Yoon1, Min-Young Lee1, Kyoung Ha Kim1, Namsu Lee1

1Soonchunhyang University Seoul Hospital, Seoul, Korea, Republic of

Background: Philadelphia chromosome-negative myeloproliferative neoplasms (MPNs), along with polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF), are clonal disorders of hematopoietic stem cells. MPNs are cancers in which malignant clones trigger cytokines that sustain inflammatory drive. Neutrophile-to-lymphocyte ratio (NLR), calculated as the ratio of absolute neutrophil count to absolute lymphocyte count, is a fast and simple method in assessing inflammatory status and has been reported to be associated with various diseases. In a previous study, we confirmed that NLR in patients with MPN was higher than that in the normal population. We also suggested that NLR would be more beneficial than EPO in diagnosing PV. In this study, we aimed to investigate clinical significance of NLR at the time of MPN diagnosis.

Methods: We retrospectively analyzed electronic medical records of patients who visited Soonchunhyang University Hospital Seoul or Soonchunhyang University Hospital Bucheon. Patients with PV, ET, and MF who met the 2016 WHO criteria were included.

Results:

Among 186 MPN patients, the most common diagnosis was ET (40.9%, 76/186), followed by PV (39.2%, 73/186) and MF (19.9%, 37/186). The median NLR was higher in PV group (6 ± 2.92) than in ET group (2.97 ± 1.22) and MF group (4.47 ± 2.29). The median NLR in the JAK2 positive group was significantly higher than that in the JAK2 negative group (5.79 vs. 2.85, respectively, p < 0.001). Most patients with PV were proven to be JAK2 positive, and the NLR was high. In patients with PV, there was no change in NLR according to risk. In patients with ET, the NLR was also higher in the JAK2-positive group compared to the other group. In addition, it was established that the NLR was greater in ET patients at high risk compared to those at low risk (4.001 ± 0.34 vs 2.752 ± 0.26, p = 0.0004). (Figure) As the disease advanced in patients with MF, the amount of peripheral blood cells other than neutrophils and lymphocytes increased, diminishing the clinical significance of NLR.

Conclusions: NLR was shown to be elevated in JAK2-positive MPN patients. This can help in the diagnostic suspicion of PV/ET in patients.

Disclosure: The authors have no conflicts of interest to declare.

16 - Conditioning Regimens

P144 FLUDARABINE EXPOSURE IMPACTS OUTCOMES OF LARGE B-CELL LYMPHOMA PATIENTS RECEIVING TREATMENT WITH CHIMERIC ANTIGEN RECEPTOR T-CELLS

Mario Andres Sanchez-Salinas 1,2, Javier Varela3, Victor Navarro4, Cecilia Carpio1,2, Norberto Nuñez5, Carla Alonso-Martinez3, Maria Jose Carreras3, Iñaki F. Troconiz6, Anna Farriols3, Maria Guerra3, Lucas Rivera3, Carol Valdivia3, Francesc Bosch1,2, Gloria Iacoboni1,2, Marta Miarons3, Pere Barba1,2

1Deparment of Hematology, Vall D’Hebron Institute of Oncology (VHIO), Hospital Universitari Vall D’Hebron., Barcelona, Spain, Universitat Autonoma de Barcelona, Barcelona, Spain, 3Pharmacy Service, Vall d’Hebron University Hospital, Barcelona, Spain, 4Oncology Data Science Group, Vall d’Hebron Institute of Oncology, Barcelona, Spain, 5Vall d’Hebron Institut de Reserca, Barcelona, Spain, 6Pharmacometrics and Systems Pharmacology Research Unit, School of Pharmacy and Nutrition, University of Navarra, Pamplona, Spain

Background: Lymphodepleting chemotherapy (LDC) has a key role in the treatment success of chimeric antigen receptor (CAR) T-cell therapy. Pharmacokinetic (PK) parameters of fludarabine in the context of LDC have been associated with outcomes in patients with acute lymphoblastic leukaemia but there is scarce data on large B-cell lymphoma (LBCL) patients. Thus, we aimed to study the impact of fludarabine PK on outcomes in this population.

Methods: We conducted a prospective study including adult patients with relapsed/refractory LBCL receiving CAR-T therapy with commercially available axicabtagene ciloleucel (axi-cel) or tisagenlecleucel (tisa-cel) at our institution, between March 2021 and September 2022. All patients received LDC with fludarabine and cyclophosphamide (planned fludarabine total dose of 90mg/m² for axi-cel and 75mg/m² for tisa-cel). In total, seven blood samples were taken on days -5, -4, -3 and 30 minutes before CAR-T infusion for each patient to determine fludarabine levels through liquid chromatography MS/MS. We calculated fludarabine exposure as area under the curve (AUC, mg×h/L) applying a population PK model (Langenhorst. ClinPharmacokinetic2019), implemented in NONMEMv7.4. Patients were divided according to 3 exposure threshold levels based on Scordo.ASH22[O657]. We aimed to identify the impact of AUC and pre-infusion fludarabine levels on efficacy and safety outcomes after CAR T-cell therapy.

Results: Fifty-four consecutive patients were included. Baseline characteristics are summarized in Table1. Complete (CR) and overall (ORR) response rates were 63% and 81%, respectively. Median PFS and OS were 14.6 months (CI95% 6.21–not reached [NR]) and NR (CI95% 13-NR), respectively. Incidence of grade ≥3 CRS and ICANS were 14.8% and 9.2%, respectively. Median number of samples analyzed per patient were 7 (IQR = 6-7), 90% (n = 339/378) of the planned samples. Despite patients treated with axi-cel received a higher fludarabine dose than tisa-cel recipients, median exposure was similar in both groups (AUC 25.3 and 25.2). Median fludarabine levels before CAR-T infusion were 5.3ng/mL, also similar in both groups (5 vs 6.7 ng/mL). Patient distribution according to AUC was: <18 (n = 18), 18-25 (n = 27), >25 (n = 9). CR and ORR were higher in patients with intermediate fludarabine exposure (53% vs. 78% vs. 50% and 71% vs. 100% vs. 75%, respectively). Median PFS (5.4, 14.6 and 5 months,p = 0.012) and OS (13 vs. NR vs. 5 months, p = 0.04) were longer in the intermediate group as well. Also, patients with high fludarabine levels at time of CAR-T infusion ( > 12ng/mL) had a shorter median PFS (2.6 months vs. NR) and OS (5.1 months vs. NR). Pre-infusion fludarabine levels assessed as a continuous variable were associated with a shorter PFS (HR 1.05 (95%CI 1-1.11, p = 0.03) and OS (HR 1.14 (95%CI 1.06-1.24, p = 0.001). Finally, fludarabine exposure and pre-infusion levels were not associated with a higher risk of CRS or ICANS.

Conclusions: To our knowledge this is the first study evaluating the impact of fludarabine PK on outcomes in patients with LBCL receiving CAR-T cell therapy. Over- and underexposure, as well as high fludarabine levels before CAR T-cell infusion, were associated with worse efficacy outcomes. Dose modulation strategies could potentially improve efficacy results of this promising therapy.

Disclosure: Pere Barba: declares having received honoraria from Allogene, Amgen, BMS, Kit/Gilead, Incyte, Jazz Pharmaceuticals, Miltenyi Biomedicine, Nektar Novartis and Pierre Fabre.

Gloria Iacoboni: Consultancy and Honoraria: Novartis, Roche, Kite/Gilead, Bristol-Myers Squibb, Abbvie, Janssen, Sandoz, Miltenyi, AstraZeneca.

The rest of the authors declare no conflict of interest related to this study.

16 - Conditioning Regimens

P145 AGE IMPACTS RISK OF MIXED CHIMERISM FOLLOWING RIC HCT FOR NON-SCID INBORN ERRORS OF IMMUNITY

Taylor Fitch 1,2, Adam Lane1,2, John McDonnell1,2, Jack Bleesing1,2, Michael Jordan1,2, Ashish Kumar1,2, Pooja Khandelwal1,2, Ruby Khoury1,2, Rebecca Marsh1,2, Sharat Chandra1,2

1Cincinnati Children’s Hospital Medical Center, Cincinnati, United States, 2University of Cincinnati College of Medicine, Cincinnati, United States

Background: Alemtuzumab, fludarabine and melphalan containing reduced intensity conditioning (RIC) was commonly used in patients undergoing allogeneic hematopoietic cell transplantation (HCT) for definitive treatment of high-risk inborn errors of immunity (IEI). Although survival is favorable, there is increased risk of mixed chimerism leading to secondary graft failure and subsequently secondary interventions (donor lymphocyte infusion (DLI), CD34+ stem cell boost or second HCT). Our aim was to evaluate if patient age impacts the risk of developing mixed chimerism with this regimen.

Methods: We retrospectively reviewed records of patients who underwent HCT for non-SCID IEIs with an uniform RIC regimen that included intermediate schedule alemtuzumab (1 mg/kg divided over days -14 to - 10), fludarabine (150 mg/m2 or 5 mg/kg if weight <10 kg divided over days -9 to -4), and melphalan (140 mg/m2 or 4.7 mg/kg if weight <10 kg on day-3) between 2010 and 2020 at our institution. Mixed chimerism was defined as <95% donor on 2 or more consecutive occasions on whole blood.

Results: Median age was 3.6 years (range, 0.35-27.2 years). Patients were categorized into 3 age groups <1, 1-5, and > 5 years of age. Forty-nine patients (52.7%) developed mixed chimerism at a median of 34 days post HCT (range, 10-1396 days). Mixed chimerism developed in 88.9 % (n = 16/18) for <1 years of age, 57.1% (n = 20/35) for years 1-5, and 35% (n = 14/40) for patients >5 years. Patients <5 years of age were significantly more likely to develop mixed chimerism (X2 (3, N = 93) = 14.8, p = 0.001). Analysis of the cumulative incidence function demonstrated significantly increased incidence of developing mixed chimerism if <1, p = 0.0002. Twenty-seven patients (29.0%) required one or more secondary intervention(s).

Ninety-three patients underwent HCT as shown in table 1.

Number of Patients

Number of patients

93

Age at HCT, years

< 1

18, (19.4%)

 

1-5

35, (37.6%)

 

> 5

40, (43.0%)

Diagnosis

HLH

49, (52.6%)

 

Other

44, (47.3%)

HLA Match

Matched Related Donor (MRD)

21, (22.6%)

 

Matched Unrelated Donor (MUD)

44, (47.3%)

 

Mismatched Unrelated Donor

28, (30.1%)

Graft Source

Bone Marrow

78, (83.9%)

 

PBSC-No manipulation

6, (6.4%)

 

PBSC-CD34 selected or alpha/beta depleted

9, (9.7%)

Mixed Chimerism

Mixed chimerism

49, (52.7%)

GVHD

All acute GVHD

31, (33.3%)

 

Grade II-IV GVHD

21, (22.6%)

 

Chronic GVHD

6, (6.5%)

Secondary Interventions

Total number of patients requiring Secondary Interventions

27, (29.0%)

 

DLI

18, (19.4%)

 

CD34 Boost

14, (15.1%)

 

Second HCT

9, (9.7%)

Patients <1 years with mixed chimerism were significantly more likely to require secondary intervention than patients > 5 years of age (X2 (3, N = 93) = 15.46, p = 0.004). Competing risk regression analysis to estimate the odds of development of mixed chimerism as a function of age category, estimated an increase in odds of development of mixed chimerism for ages <1 (OR 3.72, p = 0.006, 95% CI 1.46-19.46) and 1-5 years (OR 2.18, p = 0.022, 95% CI 1.12-4.24) compared to age >5 years. There was no significant association between mixed chimerism and graft source, graft type, CD34 /CD3 dose, underlying disease (HLH vs non-HLH), or if the patient developed GVHD.

Conclusions: Our study demonstrates that young age ( < 5 years), especially age <1 year is associated with increased risk of developing mixed chimerism in patients undergoing RIC-HCT with intermediate schedule alemtuzumab, fludarabine and melphalan for non-SCID IEIs. In addition, children < 1 year of age required more secondary interventions. Our data suggests to tailor regimen intensity based on age to reduce incidence of mixed chimerism. Children <5 years, particularly those <1 year of age, would benefit from increasing the intensity of RIC regimen. Possible strategies include adding thiotepa to the RIC regimen or using a busulfan based reduced toxicity regimen.

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P146 INAPPROPRIATE ANTIDIURETIC HORMONE SECRETION AFTER HIGH-DOSE THIOTEPA AND CONCURRENT PHENYTOIN

Maria Assunta Limongiello 1, Elisabetta Metafuni1, Sabrina Giammarco1, Federica Sorà1, Francesco Autore1, Idanna Innocenti1, Alberto Fresa2, Eugenio Galli1, Andrea Bacigalupo2, Patrizia Chiusolo1,2, Simona Sica1,2

1Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Dipartimento di Diagnostica per Immagini, Radioterapia Oncologica e Ematologia, Rome, Italy, 2Università Cattolica del Sacro Cuore, Rome, Italy

Background: Thiotepa is an alkylating compound with immunosuppressive properties which holds favourable characteristics as the capability of penetrating the blood-brain barrier combined with reduced non-haematologic toxicity. This feature led to a widespread use of this compound within transplant preparative regimens. In 1999, our centre described a first report of inappropriate secretion of antidiuretic hormone syndrome (SIADH) in a patient with primary central nervous system lymphoma who underwent high-dose thiotepa as conditioning for autologous stem cell transplantation (ASCT). The conditioning protocol consisting of thiotepa, busulfan and fludarabine (TBF), initially designed for cord blood transplant, is actually widely used as standard myeloablative conditioning regimen for allogeneic stem cell transplantation (HSCT).

Methods: This is a monocentric analysis on 259 patients who underwent HSCT after TBF conditioning regimen between May 2018 and November 2022.

Results: During this period, 18 patients developed SIADH with an incidence of 7%. Baseline patients’ characteristics are listed in Table 1. Sex was male in 8 cases (44 %) and female in 10 cases (56 %). Median age was 59 years (range 30 – 70). Seven patients had acute myeloid leukemia (AML) (38 %), 8 patients had myelofibrosis (MFI) (44%), 1 patient had mixed phenotype acute leukemia (MPAL) (6%), 1 patient had lymphoplasmacytic lymphoma (LPL) (6 %), 1 patient had systemic mastocytosis (SM) (6 %). Five patients (28%) underwent standard TBF, 9 patients (50%) TBF2 (busulfan for 2 days), 4 patients (22 %) TBF1 (1 day busulfan). Hyponatremia was detected at 48 hours from thiotepa infusion in 8 patients (44 %), at 72 hours in 8 patients (44 %), and at 96 hours in 2 patients (12 %). Laboratory tests showed the following median values: sodium (Na) 121.5 mmol/l (range 112-130), cortisol 136.5 ng/ml (range 121-175), antidiuretic hormone (ADH) 2 pg/ml (range 0.6-72.7), plasma osmolarity 259.5 mOsm/l (range 250-282), urine osmolarity 371.5 mOsm/kg (range 341-378), urinary sodium 103 mmol/l (range 53-147). Clinically, 13 patients (72 %) had nausea and among them, 9 had also vomiting, 2 had also profuse sweating and confusion, 1 had also nystagmus and subjective vertigo. One patient (6 %) had headache only, 1 (6 %) had disorientation and tremors. Three patients (16 %) had no symptoms.

Table 1. Patients’ characteristics

Total patients number

18

Gender n (%)

Male 8 (44 %)

Female 10 (56 %)

Median age at transplant

 

Years (range)

59 years (30-70)

Hematological disease n (%)

AML

7 (38%)

MFI

8 (44 %)

MPAL

1 (6 %)

LPL

1 (6 %)

SM

1 (6%)

Conditioning n (%)

TBF

5 (28 %)

TBF2

9 (50 %)

TBF1

4 (22 %)

Median sodium value (range, mmol/l)

121.5 (112-130) (NV 135-145)

Median cortisol value (range, ng/l)

136.5 (121-175) (NV 60-220)

Median ADH value (range, pg/ml)

2 (0.6-72.7) (NV < 8)

Median plasma osmolarity value (range, mOsm/l)

259.5 (250-282) (NV 280- 295)

Median urine osmolarity value (range, mOsm/kg)

371.5 (341-378) (NV 50-1200)

Median urinary sodium value (range, mmol/24h)

103 (53-147) (NV 40-220)

Symptoms n (%)

Headache

1 (6%)

Disorientation and tremors

1 (6%)

Nausea and vomiting

13 (72%)

No symptoms

3 (16%)

Conclusions: Despite the limits of a retrospective study, our real-life analysis shows that the presence of hyponatremia with corresponding serum hypo-osmolality and continuing urinary sodium excretion are suggestive of SIADH diagnosis in the presence of normovolemia. Administration of normal saline, diuretics and fluid restriction was followed by a rapid normalization of the hyponatremia in all patients together with normalization of the neurological condition. Moreover, hyponatremia and neurological symptoms developed 48-72 hours after thiotepa infusion while on phenytoin prophylaxis for busulfan. The role of phenytoin as a potential inhibitor of ADH secretion seems also relevant, as shown in a recent Swedish study. In conclusion, strict monitoring of fluid balance and electrolytes should be undertaken during conditioning in order to promptly detect SIADH onset, allowing immediate treatment of patients receiving concomitant thiotepa and phenytoin.

Clinical Trial Registry: Not applicable

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P147 CHIMERISM ANALYSIS POST-HSCT WITH A TREOSULFAN-BASED CONDITIONING REGIMEN FOR PEDIATRIC NONMALIGNANT DISEASES

Ehud Even-Or 1,2, Ziv Cohen2, Irina Zaidman1,2, Yael Dinur Schejter1,2, Adeeb Naser Eddin1,2, Polina Stepensky1,2

1Hadassah Medical Center, Jerusalem, Israel, 2The Hebrew University, Jerusalem, Israel

Background: Allogeneic Hematopoietic stem cell transplantation (HSCT) offers cures for a wide variety of pediatric nonmalignant diseases by providing healthy stem cells, which may produce the missing enzymes or replace malfunctioning cells. Unlike HSCT for malignant diseases, in transplants for nonmalignant diseases, full-donor chimerism (FD) is not necessarily required and a state of stable mixed donor-recipient chimerism (MC) may suffice for a cure. The clinical significance of MC and the factors affecting the evolvement of this condition are poorly understood. In recent years, the use of treosulfan-based conditioning regimens for HSCT in nonmalignant diseases became popular due to the myeloablative quality and low toxicity profile of these regimens. Rates of MC with treosulfan-based regiments are relatively high. In this study, we aimed to shed light on the phenomenon of MC by analyzing clinical data and post-HSCT chimerism results of pediatric patients who received a treosulfan-based conditioning regimen in our center.

Methods: In this retrospective study, we collected and analyzed clinical and transplant data from medical charts of pediatric patients who underwent HSCT for nonmalignant diseases with a treosulfan-based conditioning regimen at Hadassah Medical Center. The collected clinical data included patient demographics, primary disease, post-HSCT clinical course, and outcomes. Transplant data included donor and graft parameters, time to engraftment, and chimerism at several time points post-HSCT.

Results: Out of the 92 patients who were included in the study, 27 (29.3%) developed MC, and 65 (70.7%) achieved full-donor chimerism. Survival rates were similar between the two groups (88.9% vs 90.8% overall survival in the MC and FD groups respectively, p = 0.882). Acute GvHD rate was significantly lower in the MC group (14.8% vs 41.3% GvHD rate in the MC and FD groups respectively, p = 0.016). Graft cellularity in the MC group was higher than in the FD group (mean TNC 5.3x108 vs 4.1x108 in the MC and FD groups respectively, p = 0.047). No correlation was found between engraftment time and the development of mixed chimerism. Patient age at transplant was significantly lower in the MC group (mean age 1.41 years vs 4.14 years in the MC and FD groups respectively, p < 0.001). Looking at primary disease, patients with SCID developed significantly more MC (9 out of 18 SCID patients, 50%) than patients with other diseases (p = 0.033). Other factors associated with donor match such as family vs unrelated donors, degree of HLA match, blood type, donor and recipient pre-HSCT CMV status, were not found to be correlated to the development of MC in our cohort.

Conclusions: Stable MC is a common condition in post-HSCT pediatric nonmalignant patients treated with a treosulfan-based conditioning regimen. In our cohort, acute GvHD rate was significantly lower in the MC group, suggesting that a state of MC may pose as an advantage in nonmalignant diseases where FD chimerism is not mandatory. Younger age was associated with developing MC, possibly due to differences in chemotherapy bioavailability and metabolism in younger patients. Larger studies are required to better understand the factors associated with the development of MC and its possible advantages.

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P148 FLUDARABINE, BUSULFAN AND MELPHALAN BASED CONDITIONING REDUCED RELAPSE AFTER ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION IN ADULT PATIENTS WITH MYELOID MALIGNANCIES: A MULTICENTER RETROSPECTIVE ANALYSIS

Jieling Jiang1, Sanbin Wang2, Xiaojin Wu3, Xiaolin Yin4, Xiaofan Li5, Dong Wu6, Quanyi Lu7, Kourong Miao8, Houcai Wang9, Jiong Hu 10

1Ruijin Hospital affiliated to Shanghai JiaoTong university School of Medicine, Shanghai, China, 2The 920th Hospital of Joint Logistics Support Force of People’s Liberation Army of China, Kunming, China, 3The First Affiliated Hospital of Soochow University, Soohow, China, 4The 923th Hospital of Joint Logistics Support Force of People’s Liberation Army of China, Nanning, China, 5Fujian Medical University Union Hospital, Fuzhou, China, 6Shanghai JiaoTong University Affiliated sixth People’s Hospital, Shanghai, China, 7Zhongshan Hospital Affialted to Xiamen University, Xiamen, China, 8The First Affilated Hospital of Nanjing Medical University, Nanjing, China, 9Shanghai Tenth People’s Hospital, Shanghai, China, 10Ruijin Hospital Affiliated to Shanghai JiaoTong university School of medicine, Shanghai, China

Background: Fludarabine in combination with full dose of busulfan has become the most popular myeloablative conditioning regimen for allogeneic hematopoietic stem cell transplantation (allo-HSCT) in patients with myeloid malignancies, while relapse was till the main obstacle for cure. We observed a particular low incidence of relapse in patients prepared with double alkylating agents of busulfan and melphalan in our previous single center phase II study (NCT04269811). To further evaluate the efficacy of the regimen, we designed a multicenter retrospective study and tried to verify our previous results in large samples.

Methods: The inclusion criteria were : 1) Adult patients (16 to 70 years old) with myeloid malignancies including AML, MDS-EB-1, MDS-EB-2 and CMML; 2) Patients received first allo-HSCT from HLA matched sibling donors(MSDs), matched unrelated donor(MUDs), related haplo-identical donor(HIDs) or cord blood(CB) during Jan.2020 to Mar. 2022; 3) Conditioning regimen included busulfan ( ≥ 3.2mg/kg) and melphalan( ≥ 100mg/m2); fludarabine, cladribine, cytarabine, etopodide or cyclophosphamide could be used in combination, while venetoclax, thiotepa and total body irradiation(TBI) were excluded. Clinical data were collected and the data for patients who were alive were censored at last follow-up on May 31, 2022. The statistics was performed by SPSS and R software. The overall survival(OS) and disease free survival (DFS) were calculated using the Kaplan-Meier method and compared by log-rank tests. The cumulative incidence of relapse(CIR) and non-relapse mortality(NRM) were calculated using a competing-risk setting.

Results:

Characteristics

Values/n

Characteristics

Values/n

All eligible patients

202

Donor type

 

Age, median (range) years

45 (16 ~ 67)

MSD (1 homogeneic)

45 (22.3%)

Sex

 

MUD

10 (4.9%)

Male

116 (57.4%)

HID

143 (70.8%)

Female

86 (42.6%)

CB

4(2.0%)

Diagnosis

 

GVHD prophylaxis

 

AML

153 (75.7%)

No prophylaxis

1(0.5%)

MDS

43(21.3%)

PTCy based

167 (82.7%)

CMML

6(3.0%)

ATG based

34 (16.8%)

A total of 202 patients from nine HSCT centers were enrolled in this study. Most patients(193/202) received conditioning only consisted of different dose of fludarabine, busulfan and melphalan. Post transplantation cyclophosphamide (PTCy) based graft-versus-host disease (GVHD) prophylaxis strategy was used in 167 out of all patients. The median follow-up time was 314 days (range, 8~841days) for the whole cohort and the baseline data were summarized in Table-1. Three patients developed primary engraft failure. The median time for neutrophil and platelet recovery were 13 days (range, 9 ~ 22 days) and 14 days (range, 9 ~ 121 days), respectively. The 100-day cumulative incidence of grade II-IV acute GVHD was 11.4 ± 2.2% and the 2-year cumulative incidence of moderate/severe chronic GVHD was 6.4 ± 1.7%. At last follow-up, eighteen patients died of non-relapse reasons and sixteen patients relapsed including eight patients were still alive. The 1-year and 2-year NRM were both 10.3% ± 2.4% and the CIR at 1-year was 6.9% ± 2.2%, while the CIR at 2-year was 17.8% ± 1.7%. The 1-year OS and DFS were 86.7% ± 2.7% and 83.4% ± 3.0%, while the 2-year OS and DFS were 79.9% ± 4.3% and 72.4% ± 5.5%, respectively. Subgroup analysis showed no significance in OS and DFS when grouping by sex, age(<50y vs. ≥50y), diagnosis (AML vs. MDS), disease risk index (DRI, low risk vs. intermediate risk vs high/very high risk) and transplantation conditioning intensity(TCI, low vs. intermediate vs. intensive).

Conclusions: These results identified with our previous study and supported that busulfan and melphalan based conditioning was associated with low relapse rate and acceptable NRM in adult patients with myeloid malignancies. Randomized controlled clinical trial should be warranted to evaluate this regimen further.

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P149 REDUCED INTENSITY ALLOGENEIC TRANSPLANTATION COMBINED WITH STRATEGIC DONOR LYMPHOCYTE INFUSION USE OFFERS AN EFFECTIVE AND TOLERABLE CURATIVE THERAPY FOR OLDER PATIENTS WITH MYELOID DISEASE

Chris Armstrong 1, Eibhlin Conneally1, Catherine Flynn1, Nicola Gardiner1, Hayley Foy-Stones1, Mairead Ní Chongháile1, Nina Orfali1

1St. James’ Hospital, Dublin, Ireland

Background: Reduced-intensity conditioning (RIC) regimens offer potentially curative transplants to older patients with aggressive myeloid disease including AML, MDS and MPN, but are associated with an increased incidence of relapse compared to myeloablative regimens. Donor lymphocyte infusions (DLIs) administered strategically in high-risk patients (prophylactic) or in cases of mixed donor chimerism (pre-emptive) can attenuate this increased relapse risk in those who have not experienced de novo GVHD. Increasingly DLI is also used as an adjunct to salvage treatment after relapse (therapeutic). We here examine the effectiveness and potential burden of this approach in older patients.

Methods: We performed a national retrospective review of 174 patients with myeloid disease who received peripheral blood stem cell grafts between 2015 and 2022, following reduced intensity Flu-Bu-ATG conditioning (Fludarabine 30/m2 D-9 to D-4, Busulfan 3.2mg/kg D-5 to D-4, ATG Grafalon® 10mg(SIB)/20mg(MUD)/kg D-3 to D-1). Patients received CNI/MTX as GVHD prophylaxis. Statistical analysis was performed using STATA.

Results: The median age of our cohort was 60yrs (range 27-74) (Table 1). 162 had intermediate or high-risk CIBMTR DRI scores. The median age-adjusted HCT-CI was 2. Neutrophil engraftment was universal and platelet engraftment was 97.1% at a median of 20 and 22 days respectively. The cumulative incidence (CI) of acute GVHD at 1- and 2-years was 28.2% and 55.4% respectively. The CI of chronic GVHD at 1- and 2-years was 16% and 24%.

Overall survival (OS) at 1 and 2 years was estimated at 80.7% and 71.1%, and similar across disease groups (p = 0.45) and age groups (p = 0.38). The cumulative incidence of NRM at 2 years was 12.5% and was not affected by age (p = 0.63). The cumulative incidence of relapse at 1 and 2 years was 21.7% and 31.4%, occurring at an overall median of 197 days (64.8% in year 1). Reduced relapse was seen in aGVHD and cGVHD (p = 0.02, p = 0.04) but increased in those with a high DRI (p = 0.04) or recipients of a cryopreserved graft (p = 0.01).

At a median follow up of 610 days, 61 patients had received a DLI. Median time to first infusion was 183 days. 44 patients received DLIs in remission. 3 received prophylactic infusions due to high-riskdisease despite full donor chimerism and subsequently remained in remission. Mixed chimerism was detected at D90 in 113 patients, of whom 41 received a pre-emptive DLI. Subsequent relapse rates were 19.5% vs. 37.7% in those with mixed chimerism who did not receive a pre-emptive DLI (p = 0.05). Therapeutic DLIs were administered to 17 patients with confirmed relapse after D90. Median survival in this group was 600 days, compared to 351 days in patients who relapsed beyond D90 but did not receive a therapeutic DLI (n = 27). Survival in those receiving a DLI for any indication was significantly greater than non-recipients (p = 0.004).

Table 1: Demographics

Sex

Male

106 (60.9%)

Female

68 (39.1%)

Age

Median: 60 Range: 27-74

Disease

AML

91 (52%)

MDS

43 (25%)

MPN

40 (23%)

CIBMTR Disease Risk Index

Low

12 (7%)

Intermediate

108 (62%)

High

54 (31%)

Age-adjusted HCT-CI

Median: 2 Range: 1-8

Donor

Sibling

77 (44%)

Unrelated

97 (56%)

GVHD

Acute

103/173 (59%)

Chronic

25/164 (15%)

Cryopreservation of graft pre-infusion

20 (11%)

Mortality

Non-relapse

19

Relapse

30

Donor Lymphocyte Infusion

Prophylactic

3/44 Full donor chimerism

Pre-emptive

41/113 Mixed chimerism

Therapeutic

17/44 Relapsed

Conclusions: We validate the high clinical efficacy and tolerability of RIC HSCT in older patients with myeloid disease and confirm the potent immunomodulatory role of DLI to help safely abrogate the inherent increased relapse risk.

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P150 TOTAL BODY IRRADIATION PROCEDURES IN THE FORUM TRIAL

Bianca Hoeben 1,2, Chiara Julita3, Valeria Tremolada3, Peter Bader4, Jochen Buechner5, Jean-Hugues Dalle6, Marianne Ifversen7, Franco Locatelli8, Petr Sedláček9, Jerry Stein10, Ulrike Pötschger11, Karin Dieckmann12, Detlef Imhoff4, Rudolf Schwarz13, Alexandra Jensen14,15, Valentine Martin16, Sylvie Helfre17, Claire Charra-Brunaud18, Elisa Meldolesi19, Eugenio Vinés20, Hans Theodor Eich21, Natalia Pinto22, Michaela Jirkovska9, Örs Ferenczi23, Agnieszka Dybko24, Mogens Bernsdorf7, Claudia Linsenmeier25, Charalampos Zampatis26, Alexander Fosså5, Monika Sandorova27, Jordi Giralt28, Christina Peters29,11, Adriana Balduzzi30

1University Medical Center Utrecht, Utrecht, Netherlands, 2Princess Máxima Center for Pediatric Oncology, Utrecht, Netherlands, 3ASST Monza, Ospedale San Gerardo, Monza, Italy, 4Goethe-University Frankfurt, Frankfurt, Germany, 5Oslo University Hospital, Oslo, Norway, 6Robert Debré Hospital, GHU APHP Nord Université Paris Cité, Paris, France, 7Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark, 8IRCCS Bambino Gesù Children’s Hospital, Rome, Italy, 9Teaching Hospital Motol, 2nd Medical School, Charles University Motol, Prague, Czech Republic, 10Schneider Children’s Medical Center, Tel Aviv, Israel, 11Children’s Cancer Research Institute, Vienna, Austria, 12Medical University Vienna, Vienna, Austria, 13University Medical Center Hamburg-Eppendorf, Hamburg, Germany, 14University Hospitals Gießen and Marburg (UKGM), University Hospitals Gießen and Marburg (UKGM), Germany, 15Philipps University, Marburg, Germany, 16Gustave Roussy, Paris, France, 17Institut Curie, Paris, France, 18Institut de Cancérologie de Lorraine Alexis Vautrin, Nancy, France, 19IRCCS Fondazione Policlinico Universitario ‘A.Gemelli’, Rome, Italy, 20School of Medicine, Pontificia Universidad Católica de Chile, Red de Salud UC-Christus, Santiago, Chile, 21University Hospital Muenster, Muenster, Germany, 22Hospital de Pediatría ‘Prof. Dr. Juan P. Garrahan’, Buenos Aires, Argentina, 23National Institute of Oncology, Budapest, Hungary, 24Lower Silesian Oncology, Pulmonology and Hematology Center, Wroclaw, Poland, 25University Hospital Zürich, Zürich, Switzerland, 26St. Savvas Anticancer Hospital, Athens, Greece, 27Oncological Institute of St. Elizabeth, Bratislava, Slovakia, 28Vall d’Hebron University Hospital, Barcelona, Spain, 29St. Anna Children’s Hospital, Vienna, Austria, 30Università degli Studi di Milano Bicocca, Fondazione MBBM / Ospedale San Gerardo, Monza, Italy

Background: In the ALL SCTped FORUM study (NCT01949129, 2013-2019), TBI during conditioning for HSCT in children with high-risk ALL was performed in many international radiotherapy centers. We aimed to evaluate the variability of the TBI procedures.

Methods: An electronic survey investigating TBI delivery technique characteristics was distributed among the FORUM Principal Investigators and forwarded to their respective radiation oncologists.

Results: Responses were collected from 54 radiotherapy centers in 25 countries, serving 53 of the 88 FORUM centers. Thirty-nine radiotherapy centers treated ≤10 patients annually. Forty-three had treated ≤10 patients with TBI within the FORUM trial, and few 26-30 (n = 3) or >40 (n = 3) patients. Fractionation was mainly 12 Gy in 6 fractions of 2 Gy, usually twice-daily (n = 52), and once 1 fraction daily over 6 days. Twelve Gy in 4 daily fractions was given in 1 center, and 9.9 Gy in 3 fractions was preferred in case of sedation in 2 centers. Lung dose was reduced in 44 centers, mostly to 8-10 Gy. Other organs, especially kidneys, received dose-reduction as well in 10 centers. Patient positions were lateral decubitus (n = 15), supine (n = 17), supine and prone (n = 25), and sitting (n = 5). Beam setup was: conventional TBI with large fields at extended distance (n = 32); sweeping beam (n = 8), adjacent fields or moving couch (n = 11) with the patient underneath the gantry; and conformal modulated TBI at extended distance (n = 1), or with a conformal modulated rotational technique (n = 10). Some centers used multiple setups or switched techniques during the trial period. Conventional TBI beam direction was anterior-posterior/posterior-anterior for 28 centers, bilateral for 9, and from 4 sides in 7. Distance to the gantry was 5-6m (n = 3), 4-5m (n = 11), 3-4m (n = 11), 2-3m (n = 10), 1-2m (n = 15), or 1m (n = 6). Thirty-nine centers performed planning CT scans. Doses were calculated with a treatment planning system (n = 28), and/or with spreadsheet calculation (n = 30). Machines used were cobalt (n = 2), linear accelerator (n = 49), and/or tomotherapy (n = 4). Beam energies were ≤6 MV in 33, 6-15 MV in 16, and >15 MV in 6 centers. Dose rates at patient midplane varied from 0-5 cGy/min (n = 8), to 6-10 cGy/min (n = 13), 11-15 cGy/min (n = 7), 16-25 cGy/min (n = 13), or 60-80 cGy/min (n = 4), and variable high dose rates in conformal techniques (n = 12). Interventions to improve dose homogeneity over the body were performed in 38 centers, beam spoilers to increase surface doses in 33. Dose measurements on the patient during treatment occurred in 43 centers.

Conclusions: In the FORUM trial, radiotherapy centers used multiple known variations in setup of conventional TBI techniques. Conformal rotational TBI was used in a fifth of the centers, and its implementation is increasing. Fractionation and lung dose reduction were mainly consistent with the study protocol. To which extent different radiobiological effects due to variations in TBI setup, technique, fractionation, and timing may have led to potential differences in outcome or toxicities will be the object of further investigation in this homogeneous FORUM cohort.

Clinical Trial Registry: This is not a clinical trial, but a survey among centers who performed TBI during conditioning of children in the trial: NCT01949129

Disclosure: Conflict of interest: nothing to declare.

16 - Conditioning Regimens

P151 RETROSPECTIVE COMPARATIVE ANALYSIS OF TWO DIFFERENT FORMS OF ANTI THYMOCYTE GLOBULIN (ATG) IN PEDIATRIC AND YOUNG ADULT PATIENTS UNDERGOING ALLOGENIC HEMATOPOIETIC STEM CELL TRANSPLANT

Gaurav Kharya 1, Ravi Shankar1, Saksham Singh1, Varsha Mishra1, Garima Nirmal1, Atish Bakane1, Bharti Sharma1, Manju Joseph1, K K Swathymon1, Mohit Chaudhary2, Himshikha Yadav1

1Centre for Bone Marrow Transplant & Cellular Therapy, Indraprastha Apollo Hospital, Delhi, India, Indraprastha Apollo Hospital, Delhi, India

Background: Anti-thymocyte Globulin (ATG) is an integral part of conditioning chemotherapy for pediatric and young adult patients undergoing HSCT. With increasing use of alternate donor sources, role of ATG to prevent graft failure and graft versus host disease (GvHD) is becoming even more relevant. Although used widely but the data on comparative analysis of different forms of ATG is sparse, thus leaving the transplant physicians with multiple unanswered questions. In a retrospective analysis, we tried to look at the comparative efficacy and safety of two different forms of ATG namely Thymoglobulin® and Grafalon®. Thymglobulin® is obtained after administration of human thymocytes in pathogen free rabbits whereas Grafalon® is produced by immunizing the rabbits with Jurkat T-cell leukemia line.

Methods: We retrospectively analyzed a total of 155 patients who underwent allogenic-HSCT at our center, for both benign and malignant conditions, from August 2019 to November 2022. The primary objectives of the study were to compare the overall survival (OS), event free survival (EFS), engraftment kinetics, graft failure (GF) and GVHD between the two different cohort. The secondary objective was to assess the immune reconstitution and to compare the complications in the two groups.

Results:

Patient variables

Thymoglobulin n = 63

Grafalon n = 92

Graft failure

3

1

Acute GVHD

18 (8)

10 (4)

Chronic GVHD

8

6

CMV reactivation

32

45

Bacterial infections

15

19

PRES

6

1

Engraftment syndrome

13

6

Median time to neutrophil engraftment (days)

13 (9-20)

15 (9-28)

Median time to platelet engraftment (days)

13 (9-35)

16 (6-48)

A total of 63 patients received Thymglobulin® and 92 received Grafalon®. Patient and donor characteristics have been highlighted in table 1. There was no significant difference in OS (p = 0.09), EFS (p = 0.1), grade III-IV acute GVHD (p = 0.06) and chronic GVHD (p = 0.25) in between the two groups. However, GF (p = 0.04), grade II-IV acute GVHD (p = 0.02) were significant less in Grafalon® cohort. Neutrophil and platelet engraftment were however better in Thymoglobulin cohort (p = 0.0001 and 0.0055) respectively. Reconstitution of CD3, CD4, CD8, CD19 and CD 56 subsets at day +100 were not statistically different between the two groups. Immunoglobulin profile at day +100 was in favor of Grafalon® for IgM (p = 0.02), however for IgG and IgA there was no significant difference. There was no significant difference in bacterial, fungal or viral (CMV, BKV, adenovirus) infections between the two group. Among other complications, we observed PRES and engraftment syndrome occurred more in Thymoglobulin® cohort (p = 0.01).

Conclusions: We observed Grafalon® cohort to have significantly less graft failure and overall grade II-IV GVHD whereas Thymoglobulin® cohort showed better engraftment kinetics. However, there was no significant difference in OS, EFS and grade III-IV GVHD in between the two groups. Also, there was no significant difference for immune reconstitution at day +100 in between the two groups except for IgM reconstitution, which was in favor of Grafalon®. There was no difference in the two groups for any infections. PRES was seen more in Thymoglobulin cohort but that might be confounded by concurrent use of CNI’s.

Clinical Trial Registry: Not applicable, retrospective analysis.

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P152 TREOSULFAN BASED CONDITIONING IN PEDIATRIC PATIENTS WITH NON-MALIGNANT HAEMATOLOGICAL INDICATIONS FOR TRANSPLANT IS ASSOCIATED WITH VERY LOW TREATMENT RELATED MORTALITY AND HIGH OVERALL SURVIVAL

Daniel Farrugia 1, Salah Ali1, Persis J. Amrolia2, Catriona Brook1, Ben Carpenter3, Rachael Hough4, Anna-Maria Ewins5, Oana Mirci-Danicar6, Brenda Gibson7, Katharine Patrick8, Geoff Shenton9, Sanjay Tewari10, Robert F. Wynn11, Robert Chiesa2, Kanchan Rao2, Mary Slatter9, Su Han Lum9, Beki James1

1Leeds Children’s Hospital, Leeds, United Kingdom, 2Great Ormond Street Hospital for Children, London, United Kingdom, 3University College, London, United Kingdom, 4University College London, London, United Kingdom, 5University of Glasgow, Glasgow, United Kingdom, 6Bristol Royal Hospital for Children, Bristol, United Kingdom, 7Royal Hospital for Children, Glasgow, United Kingdom, 8Sheffield Children’s NHS Foundation Trust, Sheffield, United Kingdom, 9Great North Children’s Hospital, Newcastle, United Kingdom, 10Royal Marsden Hospital, London, United Kingdom, 11Royal Manchester Children’s Hospital, Manchester, United Kingdom

Background: Treosulfan (L-treitol-1,4-bis-methanesulfonate) is a myeloablative alkylating agent with a comparatively favorable toxicity profile and more predictable pharmacokinetics than busulfan, which historically formed the backbone of many conditioning regimens. It is increasingly used in pediatric haematopoietic stem cell transplantation (HSCT) although there remains a paucity of data regarding toxicity and outcomes according to disease-specific pediatric cohorts.

Methods: A retrospective multi-center analysis was performed for consecutive pediatric patients who underwent a treosulfan conditioned HSCT in 9 BSBMTCT centers in the UK (Bristol; Glasgow; Great Ormond Street; Leeds; Manchester; Newcastle; Sheffield; Royal Marsden Hospital and University College, London) for non-malignant haematological indications between 2015 and 2021 inclusive, to determine the incidence of key treatment related toxicities, as well as treatment related mortality (TRM) and overall survival (OS).

Results: 96 patients met the criteria, with haemoglobinopathy being the most common indication: 32 (33%) had thalassaemia major; 22 (30%) had sickle cell disease; followed by bone marrow failure in 32 (33%). The majority, 57 (60%), had a matched family donor, the source was bone marrow in 70 (93%) and the most common combination was fludarabine/ treosulfan/thiotepa in 87(90%). The OS was 95% with a median follow up of was 4 years, and EFS was similarly high at 87%. There were no significant predictors on univariate analysis. The 5 deaths were due to: respiratory failure (n = 2); thrombotic microangiopathy with multi-organ failure (MOF; n = 1); sepsis with MOF (n = 1); and cerebral fungal infection associated with GVHD (n = 1). Only 1% had Grade II-IV GVHD, and 6% chronic GVHD. VOD occurred in 2 (2%). Five patients (5%) had second procedures: 3 had an unconditioned stem cell boost; 1 had donor lymphocyte infusions and 1 patient with thalassaemia had a second transplant for primary aplasia. They are alive and well.

Conclusions: These results for treosulfan based conditioning are very encouraging, with high overall survival and very low treatment related mortality and low graft failure. The poorer outcomes in osteopetrosis are not dissimilar to other cohorts, reflecting the need for timely diagnosis and transplantation. A prospective randomized comparison with busulfan containing regimens is unlikely to happen, but this real world data suggests a superiority over historical busulfan conditioned cohorts. Further, as the data matures it will inform understanding of late effects, including on fertility and development.

Disclosure: No disclosures.

16 - Conditioning Regimens

P153 COMPARISON OF FLUDARABINE/MELPHALAN(FM140) WITH FLUDARABINE/MELPHALAN/BCNU(FBM110) IN PATIENTS WITH RELAPSED/REFRACTORY AML UNDERGOING ALLOGENEIC TRANSPLANTATION – A REGISTRY STUDY ON BEHALF OF THE EBMT ACUTE LEUKEMIA WORKING PARTY

Jesus Duque-Afonso 1, Jürgen Finke1, Maud Ngoya2, Jacques-Emmanuel Galimard2, Martin Bornhäuser3, Matthias Eder4, Wolf Rösler5, Gesine Bug6, Andreas Neubauer7, Matthias Edinger8, Gerald G. Wulf9, Pavel Jindra10, Herman Einsele11, Matthias Stelljes12, Dominik Selleslag13, Eva Maria Wagner-Drouet14, Donald Bunjes15, Alexandros Spyridonidis16, Eolia Brissot2, Arnon Nagler17, Fabio Ciceri18, Mohamad Mohty2

1University of Freiburg, Freiburg, Germany, 2Hôpital Saint Antoine, Sorbonne University, Paris, France, 3University of Dresden, Dresden, Germany, 4University of Hannover, Hannover, Germany, 5University of Erlangen, Erlangen, Germany, 6University of Frankfurt, Frankfurt, Germany, 7University of Marburg, Marburg, Germany, 8University of Regensburg, Regensburg, Germany, 9University of Göttingen, Göttingen, Germany, 10University of Pilsen, Pilsen, Czech Republic, 11University of Würzburg, Würzburg, Germany, 12University of Münster, Münster, Germany, 13University of Brugge, Brugge, Belgium, 14University of Mainz, Mainz, Germany, 15University of Ulm, Ulm, Germany, 16University of Patras, Patras, Greece, 17Chaim Sheba Medical Center, Tel Hashomer, Israel, 18IRCCS San Raffaele Scientific Institute, Milan, Italy

Background: The treatment of relapsed/refractory AML is associated with a dismal prognosis. The allogeneic hematopoietic cell transplantation (allo-HCT) is frequently performed as salvage therapy in this scenario. Conditioning protocols have been developed with the aim to reduce the leukemia burden without increasing their toxicity, especially in older patients or those with comorbidities. In our previous studies, we showed that AML patients in complete remission (CR) had better outcomes including overall survival (OS) after conditioning based on two alkylating agents (FBM110/FTM110) compared to conditioning with FM140 from the intermediate transplantation conditioning index (TCI) score.

Methods: In the present study, we compared the conditioning protocol FM140 (fludarabine, median 150 mg/m2; melphalan, median 140mg/m2) with a conditioning protocol based on FM140 with an additional alkylating agent i.e. FBM110 (fludarabine, median 150mg/m2; BCNU/carmustine 300-400mg/m2; and melphalan, median 110 mg/m2). From the registry of the EBMT Acute Leukemia Working Party, we identified 538 adult patients (288 patients with FM140 and 250 patients with FBM110) with acute myeloid leukemia (AML) with active disease (primary induction failure, relapsed or progressive disease), and transplanted with unmanipulated peripheral blood grafts from related or unrelated donors. Impacts of these two regimens on the outcomes were evaluated using Cox multivariable models.

Results: Patients in the FBM110 group were older (63.4 years vs. 58.9 years, p < 0.001) and had a worse Karnofsky performance score (KPS < 90, 50.4% vs. 41.7%, p < 0.05). Patients conditioned with FBM110 received more often in vivo T-cell depletion (TCD, 89.2% vs 63.9%, p < 0.001) of which they received more often ATG (75.6% vs. 46.9%) compared to patients conditioned with FM140, who received more frequently alemtuzumab (17% vs. 13.6%). No differences were observed between FBM110- compared to FM140-treated patients regarding OS (2y OS: 46.7% vs. 40.8%, hazard ratio (HR) for FM140 1.05, p = 0.74), progression-free survival (PFS) (2y LFS: 38.7% vs 36.8%, HR 1.06, p = 0.66), non-relapse mortality (NRM) (2y NRM: 25.8% vs 25.9%, HR: 1.07, p = 0.8) and relapse incidence (RI) (2y RI: 35.5% vs. 37.2%, HR: 1, p = 0.98). Despite FBM110 patients received more frequently in vivo TCD, there were no significant differences in aGvHD II-IV incidence (100d aGvHD II-IV: 34.1% vs 28.7%, HR 0.87, p = 0.53), aGvHD III-IV (100d aGvHD III-IV: 17% vs. 11.3%, HR 0.66, p = 0.13) and cGvHD (2y cGvHD: 31.7% vs. 32.7%, HR 0.94, p = 0.73)

Conclusions: In conclusion, despite differences on age and KPS, AML patients with active disease undergoing allo-HCT after FBM110 and FM140 conditioning show similar outcomes. We speculate that the reduced dose of melphalan is compensated by the addition of the second alkylating agent (BCNU) in FBM110 without increasing toxicity. Future studies should address the efficacy of melphalan containing regimens in combination with other drugs and/or using different dosing.

Disclosure: JD-A has received speaker’s honoraria from Roche, Amgen, AstraZeneca, Riemser, Lilly, Ipsen and Sobi and travel support from AstraZeneca, Gilead and Sobi. JF has received research support and speaker’s honoraria from Medac, Neovii, and Riemser.

16 - Conditioning Regimens

P154 OUTCOMES OF TBI 8-GRAY CONDITIONING IN ADULT PATIENTS WITH ACUTE LYMPHOBLASTIC LEUKAEMIA IN FIRST COMPLETE REMISSION

Sarah Tan1, Henry Ngu2, Lauren Child1, Nicole Chien1, Claire Hemmaway1, Peter Browett3, Timothy Hawkins1, Leanne Berkahn1, Richard Doocey1, Clinton Lewis 1

1Auckland City Hospital, Auckland, New Zealand, 2BC Cancer Agency, Vancouver, Canada, 3University of Auckland, Auckland, New Zealand

Background: Allogeneic haematopoietic stem cell transplantation (allo-HSCT) remains the primary curative option for adults with acute lymphoblastic leukaemia (ALL). Total body irradiation (TBI) based conditioning regimens are preferred but balancing the intensity of the preparative regimen to achieve optimal disease control with non-relapse mortality remains challenging.

Methods: The bone marrow transplant unit at Auckland City Hospital introduced a reduced toxicity 8-Gray (Gy) TBI-based conditioning regimen for ALL in 2017. We report our initial experience by conducting a retrospective review including adult patients ≥ 18 years with ALL in first complete remission (CR1) who underwent allo-SCT from sibling, matched unrelated (MUD) or haploidentical donors using different intensity conditioning regimens from 1 January 2015 and 31 December 2021. All patients provided consent for data submission and use in local and international data registries.

Results: Thirty patients met inclusion criteria. The median age was 51 years for both 8-Gy and Fludarabine/Melphalan (Flu/Mel) cohorts, while the median age was 26 years in 12-Gy cohort. There was a sex imbalance between the 8-Gy and Flu/Mel cohorts. There were significantly more patients with an elevated comorbidity index (HCT-CI) of 3 or more in the 8-Gy cohort. The majority of patients achieved flow MRD negativity pre-transplant.

With a median follow up of 40 months, the 2-year overall survival was 50% in TBI 8-Gy cohort, and 55% in Flu/Mel cohort (Figure 1a). The myeloablative TBI 12-Gy cohort had an improved 2-year overall survival numerically at 80%, although it did not reach statistical significance. The cumulative incidence of relapse (CIR) is the highest in TBI 8-Gy group (Figure 1b) with 2 out of 9 patients relapsing at 1 year of follow up. The 2-year GVHD and relapse free survival (GFRS) were higher at 69% in TBI 12-Gy cohort, whereas TBI 8-Gy and Flu/Mel have similar GRFS at 50% and 62%, respectively (Figure 1c). Non-relapse mortality (NRM) was highest in Flu/Mel cohort at 32% and no NRM event was noted in TBI 8-Gy cohort at 2 years (Figure 1d).

 

12-Gy

8-Gy

Flu/Mel

p value

n

10

9

11

 

Median Age, y (range)

26 (20-38)

51 (42-55)

p = 0.6 between 8-Gy vs Flu/Mel

p = <0.01 between 12-Gy vs Flu/Mel

 

Female, no. (%)

5 (50)

2 (22)

9 (82)

p = 0.025

ALL Subtype, no. (%)

    

B-ALL

10 (100)

7 (78)

10 (100)

p = 0.081

T-ALL

 

2 (22)

  

Comorbidity Index, no. (%)

    

0

4 (40)

0

3 (30)

p = 0.112

1-2

6 (60)

5 (55)

8 (70)

p = 0.285

3+

0

4 (45)

 

p = 0.002

Donor Type, no. (%)

    

Matched sibling

4 (40)

6 (67)

4 (36)

P = 0.376

Matched unrelated

5 (55)

1 (11)

7 (64)

P = 0.054

Haploidentical

 

2 (22)

 

P = 0.085

Mismatched MUD

1 (5)

  

P = 0.381

Pre-transplant MRD, no. (%)

    

Positive (>/= 0.01%)

4 (40)

2 (22)

5 (45)

P = 0.570

Negative ( < 0.01%)

6 (60)

7 (78)

6 (55)

P = 0.570

Conclusions: For patients with high-risk ALL in CR1, TBI 8-Gy conditioned allo-SCT resulted in a lower treatment related mortality rate, no early in-hospital death with similar OS and GFRS as the RIC conditioning cohort. While the cohorts are small, this data supports our clinical experience that the TBI 8-Gy is a well-tolerated and effective regimen for patients with ALL not fit for full dose myeloablative TBI.

The CIR at 1 year in older and more co-morbid patients using TBI 8-Gy is higher than those who received myeloablative TBI 12-Gy. The combination of reduced TBI dose and increased comorbidities may account for the difference in overall survival. Interestingly, the NRM in the 8-Gy cohort was much lower than that seen in the RIC cohort. Prospective studies comparing the efficacy of 8-Gy TBI to a comparable chemotherapy regimen (e.g., Flu/Mel or FluBu3) should be pursued to understand the benefit of reduced toxicity radiotherapy-based conditioning for ALL.

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P155 COMBINATION OF FLUDARABINE AND CYCLOPHOSPHAMIDE AS A CONDITIONING REGIMEN IN SECOND HSCT FOR GRAFT FAILURE

Tatiana Rudakova 1, Ivan Moiseev1, Elena Morozova1, Julia Vlasova1, Alina Vitrishchak1, Alexander Kulagin1

1RM Gorbacheva Research Institute, Pavlov University, St. Petersburg, Russian Federation

Background: Primary (PGF) and secondary (SGF) graft failure is associated with poor outcomes after allogeneic HSCT. The second allo-HSCT (2HSCT) may be only therapeutic option for GF. However, there are no established conditioning regimens for the 2HSCT. This study aimed to analyze the efficacy and safety adopted from aplastic anemia reduced-intensity conditioning regimen based on fludarabine (Flu) and cyclophosphamide (Cy) for salvage 2HSCT.

Methods: The study evaluated conditioning regimen consisting of Flu (30 mg/m2) (days−5 to−2) and Cy (300 mg/m2/day) (days−5 to−2). Thirty-five patients (pts) (F43%, M57%) with acute leukemia (66%), myeloproliferative (29%) and non-malignant (5%) diseases received the 2HSCT for GF from 2014 to 2022. Median age was 31 years 18-62). Median follow up was 77 (range 4-2531) days. GVHD prophylaxis was Cy-based (28 pts, 80%) and other protocols (7 pts, 20%). Donors were haploidentical siblings or parents in 20 (57%) cases, matched unrelated donors 9 (26%), mismatched unrelated 4 (11%), matched related 2 (6%) cases. Twenty-two (62%) pts were re-transplanted from the same donor, a new donor was used for 13 (38%) pts. PGF was defined as failure to achieve an absolute neutrophil count >0,5x109/l by 30 days after allo-HSCT and absence of donor chimerism. SGF was defined as cytopenias after initial engraftment with loss of donor chimerism less than 5%. Donor-specific anti-HLA-antibodies were detected in 4 pts. Indication for the 2HSCT was PGF in 21 (60%) pts, SGF - in 14 (40%) pts. BM and PBSC was used in 18 (49%) and 16 (48%) pts, respectively, and 1 pt (3%) received BM with PBSC.

Results: Median time from the first and second HSCT was 42 (30-237) days. Neutrophil engraftment was documented in 16 (46%) pts with median time of 30 (18-55) days. A total of 23 pts died (9 of them before D + 30 after 2HSCT), causes of death were infections (20) and relapse (3). Non-relapse mortality rate (NRM) was 37% (95% CI, 21-53) at 100 days and 61% (95%CI, 42-76) at two years. Two-year event free and overall survival (OS) was 21% (95% CI, 9–38) and 31% (95% CI, 18-52), respectively. The only factor associated with better OS and lower NRM was Cy-based GVHD prophylaxis in the 2d HSCT compared to other protocols (55% vs 87%, p = 0.013). Eleven pts received the 3d HSCT (PGF in 8 cases, and SGF in 3 cases). Infectious episodes were documented in 31 cases; 26 pts had active infection at the time of the 2HSCT. Toxicity included mucositis 22 (63%) pts, cystitis 8 (23%) cases, cytokine release syndrome 2 (6%) cases, VOD 3 (9%) cases, TMA 2 (6%) cases, hemorrhagic complications 14 (40%) cases. CI of aGVHD grade III-IV was 16% (95%CI 5-31).

Conclusions: The 2HSCT with FluCy conditioning might be an option as a salvage therapy after GF. Infectious complications due to prolonged cytopenia remain the major problem in the setting of patients with PGF.

Disclosure: None.

16 - Conditioning Regimens

P156 COMPARISONS OF ATG-THYMOGLOBULIN (ATG-T) WITH ATG-FRESENIUS (ATG-F) TO OPTIMIZE CONDITIONING REGIMENS FOR HAPLOIDENTICAL HEMATOPOIETIC STEM CELL TRANSPLANTATION (HSCT) IN T-CELL ACUTE LYMPHOBLASTIC LEUKEMIA

Xing-Yu Cao 1, De-Yan Liu1, Jian-Ping Zhang1, Min Xiong1, Zhi-Jie Wei1, Jia-Rui Zhou1, Yue Lu1, Rui-Juan Sun1, Yan-Li Zhao1

1Hebei Yanda Lu Daopei Hospital, Langfang, China

Background: Adding anti-human T lymphocyte or thymocyte immunoglobulin (ATG) to haploidentical HSCT (haplo-HSCT) conditioning regimens can decrease the incidence of graft-versus-host disease (GVHD) and promote implantation in T-cell acute lymphoblastic leukemia (T-ALL) patients. ATG-F (formerly called ATG-Fresenius, Grafalon, Neovii Biotech GmbH) and ATG-T (Thymoglobulin, Genzyme Polyclonals S.A.S.) are commonly used ATGs with different metabolic properties. ATG-F has been shown to have the anti-leukemic activity in vitro. We conducted a study comparing the anti-leukemic activities of two doses of ATG-T vs ATG-F, used as part of the conditioning regimens in T-ALL haplo-transplantation.

Methods: From July 2012 to June 2020, 152 patients with T-ALL in Hebei Yanda Lu Daopei Hospital who underwent haplo-HSCT with either a ATG-T 5mg/kg (n = 18), ATG-T 7.5mg/kg (n = 37) or ATG-F 20mg/kg (n = 97) conditioning regimen were enrolled. The total ATG dose was equally divided over 4 days (from the day -2 to day -4).

Results: The median age was 15 years (range: 2-61) and 107 (70.4%) of patients were male. All patients achieved complete remission (CR) before transplantation (first CR [CR1] =106; ≥ second CR [CR2] = 46) and received total body irradiation (TBI)-based conditioning regimens. There were no significant differences in age, gender, time from diagnosis to transplantation, number of chemotherapy cycles to obtain remission, disease status (CR1 or ≥CR2) before transplant among the three ATG groups. The median time to neutrophil and platelet engraftment in the ATG-T 5 mg, ATG-T 7.5mg and ATG-F 20mg groups was 15 days, 14 days and 14 days (p = 0.075) and 15 days, 12 days and 13 days, (p = 0.193), respectively.

The median follow-up time was 700 days (range:24-3126). Patients who received ATG-F 5mg had superior overall survival (OS) and leukemia free survival (LFS) compared to the ATG-T 7.5mg and ATG-F 20mg groups (5-year OS: 80.8% vs 53.4% vs 67.3%, p = 0.052; 5-year LFS: 80.8% vs 50.7% vs 65.5%, p = 0.046). The ATG-T 5mg group had a lower 5-year relapse incidence (RI) but this difference was not statistically significant compared to the ATG-T 7.5 mg and ATG-F 20 mg groups (0% vs. 8.1% vs.10.8%, p = 0.401). The 5-year non-relapse mortality (NRM) of ATG-T 5mg group was lower compared to the ATG-T 7.5mg and ATG-F 20mg groups (19.2% vs. 23.8% vs. 43.9%, p = 0.027).

There was no difference in grade 3-4 acute GVHD incidence and extensive chronic GVHD among the ATG-T 5mg, ATG-T 7.5mg and ATG-F 20mg groups (8.11% vs 10.31% vs 27.78%, p = 0.068; 29.4% vs 22.2% vs 21.6%, p = 0.369). No statistical difference was observed in the 100-day cytomegalovirus (p = 0.231) or Epstein-Barr virus reactivation (p = 0.262). By multivariate analysis, ≥CR2 before transplant, ATG-T 7.5mg, and the use of tacrolimus for GVHD prophylaxis were independent prognostic factors in reducing LFS. Infections in the ATG-T 7.5mg group resulted in an increased risk of mortality.

Conclusions: Our results demonstrate that despite its anti-leukemic activity in vitro, ATF-F did not reduce recurrence rate for T-ALL patients undergoing haplo-HSCT, compared to ATG-T. ATG-T 5mg resulted in better OS and LFS by reducing NRM compared to ATG-T 7.5mg and ATG-F 20mg.

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P157 SEQUENTIAL ALLOGENEIC TRANSPLANTATION IN ACUTE MYELOID LEUKAEMIA AND MYELODYSPLASTIC SYNDROME: EXPERIENCE FROM A SINGLE CENTRE

Rebeca Jurado1, Mireia Morgades1, Alejandro de Jaureguizar1, Marta Canelo1, Blanca Xicoy1, Susana Vives1, María Josefa Jiménez1, Maria Huguet 1, Christelle Ferrà1, Josep Maria Ribera1, Juan Manuel Sancho1, Anna Torrent1

1Institut Català d’Oncologia Badalona, Hospital Germans Trias i Pujol, Josep Carreras Research Institute., Badalona, Spain

Background: Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a curative option for acute myeloid leukemia (AML) and high-risk myelodysplastic syndromes (HR-MDS). A proportion of patients fail to achieve a satisfactory response to undergo allo-HSCT, compromising the response to the procedure. In these cases, there is the possibility of sequential treatment schemes, including a first phase of chemotherapy to reduce the leukaemia cell burden, followed by allo-HSCT with reduced-intensity conditioning to take advantage of the graft-versus-leukaemia effect. We show the results of allo-HSCT with sequential conditioning (allo-SEQ) regimen from a single centre.

Methods: This study included all 15 patients with relapsed/refractory (R/R) AML or HR-MDS who underwent an allo-SEQ between July 2007 and September 2021 at the ICO-Hospital Universitari Germans Trias i Pujol. Patient’s characteristics and outcomes after allo-HSCT with a sequential conditioning regimen were analysed.

Results: The characteristics of the patients and allo-HSCT data are summarized in Table 1. There were 8/15 females (53%), and the median age was 60 years old (range 28-69). The reasons for allo-SEQ were: early relapse ( < 6 months) after high dose chemotherapy (n = 6); lack of complete response ( > 5% bone marrow blasts) after two induction regimens, including high-dose cytarabine (n = 3); <50% reduction of blasts after the first induction (n = 3); HR-SMD ( > 10% blasts or unfavourable cytogenetics) after two cycles of hypomethylating agents (n = 1); to avoid adding toxicity after a first partial response (PR) with slow hemoperipheral recovery post-induction therapy (n = 1); and to avoid toxicity in an MRD-positive patient with multiple complications in induction and consolidations (n = 1). In 67% of patients the allo-SEQ were HLA identical.

Seven out 13 (54%) patients presented acute graft-versus-host disease (GVHD), of grade ≥3 in 4 of them, predominantly affecting the skin and gastrointestinal tract. Of the 7 patients who had received post-transplant cyclophosphamide, only 2 presented acute GVHD. Likewise, a high percentage of chronic GVHD was observed in 4/8 patients, being severe in 2/4.

Overall survival probabilities at 100 days and 5 years were 60% (95% CI: 32-80%) and 20% (95% CI: 5-42%), respectively (Figure 1). Twelve patients died, 9 transplant-related (infections (n = 3) GVHD (n = 2), graft failure (n = 2), hepatic obstructive sinusoidal syndrome (n = 1) and cerebral haemorrhage (n = 1)) and 3 due to disease relapse. The cumulative incidence of relapse (CIR) at one year was 20% (95% CI: 5%-42%).

TABLE 1: Patients and transplant characteristics.

Patients and Allo-HSCT characteristics

N = 15

Diagnosis

Acute myeloid leukaemia

13 (86%)

Myelodysplastic syndrome

1 (7%)

Acute panmyelosis with myelofibrosis

1 (7%)

Previous neoplasm

Yes

2 (13%)

ECOG score

0/1

12 (80%)/ 3 (20%)

Sorror score

≥4

3 (20%)

EBMT score

≥5

5 (33%)

Number of prior lines

1/2

10 (67%) / 5 (33%)

Disease status at allo-HSCT

Relapse / NR / PD / PR / MRD

6 (40%) / 4 (27%) / 1 (7%) / 3 (20%) / 1 (7%)

% Blasts at allo-HSCT

Median [range]

12.5% [1% - 35%]

Donor

MRD / Haploidentical / MUD

8 (53%) / 5 (33%)1 / 2 (13%)

Stem cells source

PB / BM

14 (93%) / 1 (7%)

Type of conditioning

Myeloablative / Non myeloablative

5 (33%)/10 (67%)

Conditioning scheme

Fludarabine+Cytarabine+ Busulfan

9 (60%)

Fludarabine+Cytarabine+Idarubicin+ Melphalan

4 (27%)

Fludarabine+Cytarabine+Idarubicin+ Busulfan

2 (13%)

Graft versus host disease prophylaxis

CNI + others

7 (47%)

PT-Cy based

7 (47%)

CsA+MMF+Thymoglobulin

1 (7%)

  1. 1: Of which four were 5/10 and one 9/10; NR: non-response; PD: progression disease; PR: partial response; MRD: positive minimal residual disease. MRD: matched related donor; HSCT: hematopoietic stem cell transplantation; MUD: matched unrelated donor; PB: peripheral blood; BM: bone marrow; CNI: Calcineurin inhibitors; others: methotrexate or mycophenolate; PT-Cy: posttransplantation cyclophosphamide; CsA: cyclosporine; MMF: mycophenolate; MTX: methotrexate.

Conclusions: Although allo-SEQ is a valid therapeutic option for some selected high-risk patients diagnosed with R/R-AML or HR-MDS, it was associated to high mortality. The adequate selection of patients and the implementation of measures to reduce the toxicity are essential to improve the results.

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P158 RECIPIENT STEM CELL MOBILIZATION WITH PLERIXAFOR IN PRE-HSCT CONDITIONING REGIMEN FOR NON-MALIGNANT CONDITIONS

Dmitry Balashov 1, Alexandra Laberko1, Elvira Sultanova1, Aishat Idarmacheva1, Svetlana Radygina1, Svetlana Kozlovskaya1, Yulia Skvortsova1, Larisa Shelikhova1, Michael Maschan1, Alexei Maschan1

1Dmitriy Rogachev National Center for Pediatric Hematology, Oncology, and Immunology, Moscow, Russian Federation

Background: Graft failure (GF) is one of the most common problems after allogeneic HSCT in non-malignant inherited diseases, which may reach up to 15%. Treosulfan-based conditioning and T-cell graft depletion are effective options to reduce risk of transplant-related toxicity and GVHD, which, however, may lead to increased incidence of GF. Addition of hematopoietic stem cell mobilizer - plerixafor to pre-HSCT preparative regimen may enhance myeloablation without escalation of chemo-associated organ toxicity. Here we present an experience of pre-HSCT plerixafor usage in non-malignant conditions.

Methods: Between May 2016 and January 2022, 81 patients (median age 2,5 years; range 0.5-13.4) with non-malignant diseases (Wiskott-Aldrich syndrome (WAS) 48, chronic granulomatous disease (CGD) 23, severe combined immunodeficiency (SCID) 4, severe congenital neutropenia 2, Х- linked adrenoleukodystrophy 1, thalassemia 1, ataxia-telangiectasia (AT) 1, WHIM syndrome 1) underwent 84 allo-HSCT with plerixafor in conditioning regimen. 67 patients received plerixafor before first, 11 before second, 3 before first and second HSCTs.

The source of stem cells was PB (n = 79) or BM (n = 5); all PB grafts were TCRab/CD19 depleted. MMRD were used in 43 transplants, MUD in 34, and MSD in 7. Most patients received myeloablative doses of treosulfan (n = 76) or busulfan (n = 3) with fludarabine and melphalan (or thiothepa) in conditioning regimen. Other 5 patients (4 SCID, 1 AT) received minimally intensive conditioning (MIC) due to severe clinical status. For serotherapy, 79 of 81 patients received anti-thymocyte globulin. In all cases were used G-CSF 10 μg/kg (from day −8 to −4) and plerixafor 240 μg/kg (from day −6 to −4). 45 patients received post-transplant GVHD prophylaxis.

Results: Engraftment was achieved in 82 (97,6%) cases with a median time of 14 days (range 6-28) for neutrophils and 12 days (range 9-33) for platelets.

Acute GVHD was observed in 26 of 84 cases, but probability of acute GVHD grade ≥III was only 8,4% (95%CI 3,5-16,6%). 6 patients (7%) had chronic GVHD (limited, n = 4; extensive, n = 2).

GF was observed in 8 (9,7%) cases (non-engraftment, n = 2; graft rejection, n = 6) with a median time of 4 months post-HSCT (range 1-7). GF was observed in 2 (4,1%) of 48 WAS patients, in 2 (8,6%) of 23 CGD patients, and in 4 (40%) of 10 other patients. No difference in GF rates after first and second HSCT with plerixafor was seen (p = 0,6).

Median follow-up time was 37 months, range 1-79. OS was 78,6% (95%CI 68,3-86,8%) and varied between the diseases: 90% (95%CI 78,2-96,7%) in WAS, 75% (95%CI 53,3-90,2%) in CGD, and 30% (95%CI 6,7-65,2%) in other. The causes of death were bacterial sepsis (n = 4), GVHD or associated viral infections (n = 9) and TA-TMA (n = 2).

Conclusions: Most patients received TCRab/CD19-depleted graft after treosulfan-based conditioning, and the incidence of severe GVHD and transplant-related toxicity was low. Addition of plerixafor to conditioning regimen led to decreased incidence of GF in WAS (4,1%) and CGD (8,6%). Apparently, high incidence of GF in other patients might be caused by inadequate myeloblation of MIC regimens used in half of the patients.

Disclosure:

There are no conflicts of interest to report.

16 - Conditioning Regimens

P159 INCIDENCE OF HEMORRHAGIC CYSTITIS USING CYCLOPHOSPHAMIDE VERSUS ETOPOSIDE COMBINED WITH TOTAL BODY IRRADIATION AS CONDITIONING FOR HEMATOPOIETIC STEM CELL TRANSPLANT IN PAEDIATRIC ACUTE LYMPHOBLASTIC LEUKAEMIA

Sushil Gunaseelan 1, Arina Lazareva1, Maria Gabelli1, Khushnuma Mullanfiroze1, Robert Chiesa1, Persis Amrolia1, Giovanna Lucchini1, Kanchan Rao1

1Great Ormond street hospital NHS Trust, London, United Kingdom

Background: Total body irradiation (TBI) is commonly used as conditioning in Hematopoietic stem cell transplant (HSCT) for acute lymphoblastic leukaemia (ALL) in children. TBI is most frequently administered in combination with either cyclophosphamide (Cy/TBI) or etoposide (VP16/TBI). While both regimens are reported in literature to be well tolerated and with equivalent survival (Gassas et al. Bone marrow transplantation 2008); we observed an excess of troublesome Haemorrhagic cystitis (HC) in the Cy/TBI arm and sought to study this is more detail.

Methods: We conducted a retrospective analysis of patients with ALL, between ages 0-18 years, undergoing an HSCT in CR1 or CR2 at Great Ormond Street Hospital between 2009 and 2018. Primary objective was to compare the incidence of HC in both arms. The secondary objective was to analyse the outcomes following HSCT in the Cy/TBI arm compared to the VP16/TBI arm. The dose of Cyclophosphamide was 120mg/kg and TBI was administered to majority (69%) of the patients at 14.4Gy in 8 fractions and remaining (31%) at 12Gy in 6 fractions. Etoposide was given at 60mg/kg along with 12Gy TBI in 6 fractions. Patients were identified in the department database and relevant demographic and clinical data were extracted, these were compared using Chi-square test or Fisher’s exact test as appropriate. Event-free survival (EFS), overall survival (OS), Transplant-related mortality (TRM) and relapse were analysed using Kaplan Meier method and Log rank test.

Results: A total of 75 patients who underwent Allogeneic HSCT for ALL were identified. 8 patients were excluded as they had different conditioning regimens. Out of 67 remaining patients, 32 were conditioned with VP16/TBI and 35 received Cy/TBI. The median age at HSCT in both groups were 7.7 and 7.8 years respectively. Majority of the patients had B-ALL and were transplanted in CR2 in both groups. There were no significant differences in incidence of GvHD and AKI in both the groups. However, 60% of patients who received Cy/TBI developed HC compared to 12% in those who received VP16/TBI (p < 0.0001). The incidence of BK virus in urine was also significantly higher in the Cy/TBI group (60%) as compared to VP16/TBI (16%)(p < 0001). At a median onset of 19.5 days (Range 3-106) after HSCT, 28 patients (41%) developed HC and 17(60%) were of grade 1-2 and were managed conservatively. Grade 3-4 HC were observed in 11 patients (40%) with 6 patients requiring surgical intervention. The 3 year event-free survival and overall survival were 51.5 % and 54% for those receiving VP16/TBI, and 48% and 49.5 % for the Cy/TBI group (p value not significant). There was no significant difference in the TRM between the two groups (13% in VP16/TBI and 23% in Cy/TBI).

Conclusions: We were able to identify a higher incidence of Hemorrhagic cystitis and BK viruria in patients conditioned with Cy/TBI compared to VP16/TBI. This did not determine the survival outcomes and transplant related mortality of the patients. If large prospective studies show equivalent survival with both these conditioning regimens; VP16/TBI may be preferable in order to minimise the incidence of troublesome HC.

Disclosure: 1. Sushil Gunaseelan- Nothing to declare.

2. Arina Lazareva- Nothing to declare

3. Maria Gabelli- Nothing to declare

4. Khushnuma Mullanfiroze-Nothing to declare

5. Robert Chiesa- Nothing to declare

6. Persis Amrolia- ADC Therapeutics -Patents & Royalties: named inventor WO2022063853A1

Autolus- Patents & Royalties and Research Funding

Bluebird Bio- Research Funding

Pierre Fabre- Consultancy

UCLB- Patents & Royalties

7. Giovanna Lucchini- Nothing to declare

8. Kanchan Rao- Nothing to declare

16 - Conditioning Regimens

P160 ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION USING REDUCED TOXICITY VS MYELOABLATIVE CONDITIONING FOR THE TREATMENT OF CHILDREN WITH MYELOID MALIGNANCIES

Asaf D Yanir 1,2, Yoad Prokocimer1, Yotam Dizitzer1, Shlomit Barzilai Birenboim1,2, Anat Yahel1, Aviva Krauss1,2, Jerry Stein1,2

1Schneider Children’s Medical Center of Israel, Petach Tikva, Israel, 2Tel Aviv University, Tel Aviv, Israel

Background: Allogeneic hematopoietic stem cell transplantation (aHSCT) is the definitive treatment for children with high-risk myeloid malignancies, but there is no consensus regarding the optimal conditioning regimen in this clinical setting. The addition of melphalan to the busulfan-cyclophosphamide regimen (Bu-cy-mel), introduced in the early nineties, has been adopted by many groups to reduce relapse rates. Nevertheless, concerns regarding the high toxicity of this regimen, especially in adolescents, has compelled the search for a less toxic regimen that does not sacrifice efficacy for tolerability. Most studies to date have focused on comparison between reduced intensity (RIC) regimens to fully myeloablative (MAC) regimens in these patients; the use of what is often termed “reduced toxicity conditioning” (RTC) has not been studied rigorously. Here we report results with a RTC using fludarabine, thiotepa and either melphalan or treosulfan (Flu-TT-Mel or Flu-Treo-TT) versus the use of the traditional Bu-CY- Mel conditioning for 1st aHSCT in children with myeloid malignancies.

Methods: This retrospective cohort included all children who received a 1st aHSCT for myeloid malignancies at Schneider Children’s Medical Center of Israel between 2010 and 2021. The default conditioning regimen was MAC for patients < 12 years and RTC for patients ≥12 years, but was ultimately left to HSCT physician discretion, based on clinical evaluation of toxicity versus relapse risk. In a few cases, Treosulfan was used due to non-availability of melphalan. Serotherapy with ATG (Grafalon)was added for matched unrelated (MUD) or haploidentical donors. GVHD prophylaxis consisted of cyclosporin A alone (matched sibling donors, MSD), or in combination with low dose Methotrexate (MUD).

Univariate statistical analyses were performed to compare demographic and clinical characteristics, including outcomes of treatment. Kaplan-Meier estimates were utilized to assess cumulative incidence of relapse (CIR), treatment-related mortality (TRM), event-free survival (EFS) and overall survival (OS). Results are presented by log-rank and hazard ratio (HR) with 95% confidence interval (CI), and depicted using survival graphs.

Results: Twenty patients were transplanted using MAC and 24 using RTC. The MAC group included significantly fewer patients with AML and more JMML cases; patients in this group were significantly younger. All other pre aHSCT variables were comparable. Neutrophil and platelet engraftment occurred earlier in the RTC group, but the difference reached statistical significance only for platelet engraftment. Relapse rate was significantly higher in the MAC group, resulting in worse EFS. TRM and long-term organ toxicity were comparable between the 2 groups. OS was better in the RTC group, though the difference didn’t reach statistical significance.

 

MAC

N = 20 (45.5%)

RTC

N = 24 (54.5%)

p-value

Total

N = 44 (100%)

AML/MDS/JMML

6/10/4 (30/50/20%)

15/9/0 (63/37%)

0.03/0.68/0.03

21/19/4 (47/43/10%)

Age (years)2;

2.4 [1.4-9.1], 0.3-15.2

13.6 [9.9-15.6], 0-19.2

<0.001

9.9 [1.8-14.7], 0-19.2

Age < 12 (years)1,3; n(%)

17 (85.0%)

8 (33.3%)

0.001

25 (56.8%)

Sex (male)1; n(%)

13 (65.0%)

12 (50.0%)

0.317

25 (56.8%)

MSD/MUD/Haplo

9/10/1 (45/50/5%)

11/8/5(46/33/21%)

0.95/0.29/0.12

20/18/6 (45/40/15%)

CR1/CR2/Active disease

5/3/12(25/15/60%)

12/2/10 (50/8/42%)

0.09/1.0/0/22

18/4/22 (38/12/50%)

Neutrophil engraftment day2,3;

17 [14-22], 8-53

14 [12-17], 8-32

0.056

15 [13-19], 8-53

Platelet engraftment day2,3;

29 [17.5-41.5], 14-171

17.5 [13-22], 10-234

0.016

19 [16-33], 10-234

VOD1; n(%)

7 (35.0%)

5 (20.8%)

0.293

12 (27.3%)

aGVHD Grade 1-2/ 3-4/None

6/2/12 (25/15/60%)

2/8/14(8/32/60%)

0.145

7/11/26 (16/25/60%)

cGVHD :None/Limited/Extensive

16/4/0 (80/20%)

18/4/2 (75/17/8)

1.000

34/8/2 (77/19/4%)

Relapse1; n(%)

7 (35.0%)

2 (8.3%)

0.029

9 (20.5%)

Time to relapse (months)2,3;

9 [2-17], 2-47

4 [1-7], 1-7

0.242

7 [2-14], 1-47

Treatment related mortality1;n (%)

3 (15.0%)

2 (8.3%)

0.646

5 (11.4%)

Long term organ toxicity

6 (30.0%)

9 (37.5%)

0.601

15 (34.1%)

Total follow-up time2,3;

29 [15-55], 1-98

28 [13-52], 2-87

0.906

29 [14-54], 1-98

Alive at last follow-up1,3; n(%)

13 (65.0%)

21 (87.5%)

0.076

34 (77.3%)

  1. 1Pearson’s chi-square or Fishers exact test
  2. 2Mann-Whitney U test
  3. 3Continuous variables are presented as median [IQR], range

Conclusions: In this small cohort, the use of 3 alkylating agents in a MAC protocol did not lead to better disease control, and EFS was worse compared to a RTC regimen. There was no improvement even in patients with MDS who received no therapy prior to HSCT. The optimal conditioning regimen for high-risk pediatric myeloid malignancies should be investigated in large prospective clinical trials.

Disclosure: Nothing to declare.

16 - Conditioning Regimens

P161 PREVIOUS BUSULFAN IS NOT A CONTRAINDICATION FOR SECOND ALLOGRAFT WITH REDUCED INTENSITY TREOSULFAN-BASED CONDITIONING REGIMEN FOLLOWED BY ALLOGENEIC HAEMATOPOIETIC STEM-CELL TRANSPLANT FROM UNRELATED DONORS

Adrian Maraj1, Daniele Avenoso 1, Michelle Kenyon1, Pramila Krishnamurthy1, Varun Mehra1, Austin Kulasekararaj1, Shreyans Gandhi1, Francesco Dazzi1, Ye Ting Leung1, Sandra Anteh1, Maria Cuadrado1, Mili Naresh Shah1, Hayat Ullah1, Styliani Bouziana1, Christianne Bourlon1, Oana Diana Dragoi1, Antonio Pagliuca1, Victoria Potter1

1King’s College Hospital NHS Foundation Trust, London, United Kingdom

Background: Despite the improvement in the treatment of relapsed myelodysplastic syndromes with excess