The 47^th Annual Meeting of the European Society for Blood and Marrow Transplantation: Physicians - Oral Sessions (O010 – O169)

14 – 17 March, 2021 • Virtual 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.

Acute Leukaemia

O010. Impact of Co-Occurring Cytogenetic Abnormalities on Transplant Outcome of TP53 Mutant Acute Myeloid Leukaemia: Report From The ALWP of The EBMT

Justin Loke¹, Myriam Labopin², Charles Craddock¹, Jan J Cornelissen³, Hélène Labussière-Wallet⁴, Eva Maria Wagner-Drouet⁵, Gwendolyn Van Gorkom⁶, Nicolaas Schaap⁷, Nicolaus Kröger⁸, Joan Hendrik Veelken⁹, Montserrat Rovira¹⁰, Herve Tilly¹¹, Hans Martin¹², Gesine Bug¹², Ali Bazarbachi¹³, Sebastian Giebel¹⁴, Eolia Brissot², Arnon Nagler¹⁵, Jordi Esteve¹⁰, Mohamad Mohty²

¹ Queen Elizabeth Hospital, Birmingham, United Kingdom, ² Hospital Saint-Antoine, Paris, France, ³ Erasmus MC Cancer Institute, Rotterdam, Netherlands, ⁴ Hopital Lyon Sud, Hospices Civils de Lyon, Lyon, France, ⁵ University Medical Center Mainz, Mainz, Germany, ⁶ University Hospital Maastricht, Maastricht, Netherlands, ⁷ Radboud University Medical Centre, Nijmegen, Netherlands, ⁸ University Hospital Eppendorf, Hamburg, Germany, ⁹ Leiden University Hospital, Leiden, Netherlands, ¹⁰ Hospital Clinic, Barcelona, Spain, ¹¹ Centre Henri Becquerel, Rouen, France, ¹² Goethe-Universitaet, Frankfurt Main, Germany, ¹³ American University of Beirut Medical Center, Beirut, Lebanon, ¹⁴ Maria Sklodowsk-Curie Memorial Cancer Centre & Institute of Oncology, Gliwice, Poland, ¹⁵ Chaim Sheba Medical Center, Tel Hashomer, Israel

Background: The presence of a TP53 mutation in patients with acute myeloid leukaemia (AML) has been reported to be associated with an adverse outcome after allogeneic stem cell transplantation (allo-SCT). However, the determinants of transplant outcome in patients with mutant TP53 AML have not been studied in a large patient cohort.

Methods: We analysed the outcomes of patients with AML, who had intermediate or adverse risk cytogenetic abnormalities, who received an allo-SCT in CR1 between 2015-2019, according to their TP53 mutation status. Patients were only included in this analysis if their TP53 mutation status was reported to the EBMT registry. There was no restriction to inclusion with regards to age, conditioning intensity or donor source.

Results: We identified 179 patients with AML with mutated TP53 and compared them to 601 patients with AML without TP53 mutations. We identified a statistically significant enrichment of chromosomal losses in 17p (31% vs. 3%), 5q (49% vs. 5%), and 7q (34% vs. 8%) as well as complex (65% vs. 12%) and monosomal karyotypes (41% vs. 8%) in the cohort of patients with mutated TP53, in comparison to those without. The overall survival (OS) at 2 years in patients with mutated TP53 was significantly lower than in the cohort without (mutated TP53, 35.1% [95% CI: 26.7-43.7]; without TP53 mutations, 64% [95%CI:59.1-68.4], p = 0.001). This was due to a cumulative incidence of relapse (CIR) at 2 years in patients with mutated TP53 which was significantly higher than in the cohort without (mutated TP53, 55% [95%CI:45.2-63.8]; without TP53 mutations, 25.2% [95%CI:21.2-29.3], p = 0.001). The adverse prognostic impact of TP53 mutations on OS and CIR remained significant in a multivariate analysis that accounted for other adverse prognostic variables. Neither differences in conditioning intensity, nor GVHD prophylaxis impacted on outcomes in patients with mutated TP53 AML. However, in this cohort further prognostic discrimination was provided by the presence of either 17p abnormalities (abn17p) or complex karyotype (CK). In patients with mutated TP53 AML without abn17p or CK the 2-year OS was 65.2% [95%CI:48.4-77.6], in comparison, patients with mutated TP53 AML with either abn17p or CK, the OS was significantly lower, 24.6% [95%CI:16.2-34], p = 0.001) (figure). The adverse effect of abn17p and CK on the OS of patients with mutated TP53 AML was retained in a multivariate analysis. In patients with mutated TP53 AML, without abn17p or CK, the CIR was 27.5% [95%CI:13.4-43.7], compared with a CIR of 65.4% ([95%CI:53.9-74.8], p = 0.001) in patients with mutant TP53 AML with either abn17p or CK.

Conclusions: Our data are practice informing and assist in both risk stratification of patients allografted for mutated TP53 AML and in identification of patients likely to benefit from transplant. The adverse prognostic effect of this mutation is only evident in patients with co-occurring abn17p or CK. Even in patients with mutated TP53 AML and a co-occurring abn17p or CK, a significant proportion of patients achieved long-term survival. The frequent co-occurrence of abn17p with TP53 mutations, alongside their additive poor prognostic implication, underlines the biological importance of biallelic loss of TP53 activity in AML.

Disclosure: There are no conflicts of interest to declare.

O011. Impact of Cytogenetic Classification in ALLO-HCT in Patients With Acute Myeloid Leukemia Secondary to Myelodysplastic Syndrome: A Join Study by The ALWP and The CMWP

Eolia Brissot¹, Myriam Labopin², Arnold Ganser³, Johan Maertens⁴, Matthias Stelljes⁵, Dietrich Beelen⁶, Nicolaus Kröger⁷, Jean-Henri Bourhis⁸, Didier Blaise⁹, Adrian Bloor¹⁰, Jürgen Finke¹¹, Riitta Niittyvuopio¹², Gerard.G. wulf¹³, Hélène Labussière¹⁴, Ibrahim Yakoub-Agha¹⁵, Arnon Nagler¹⁶, Jordi Esteve¹⁷, Mohamad Mohty¹

¹ Hôpital Saint-Antoine, Sorbonne University, Paris, France, ² ALWP, Paris, France, ³ Hannover Medical School, Hannover, Germany, ⁴ University Hospital Gasthuisberg, Leuven, Belgium, ⁵ University of Muenster, Muenster, Germany, ⁶ University Hospital, Essen, Germany, ⁷ Bone Marrow Transplantation Centre, University Hospital Eppendorf, Hamburg, Germany, ⁸ Gustave Roussy Cancer Campus, Paris, France, ⁹ Institut Paoli-Calmettes, Marseille, France, ¹⁰ CHU Grenoble Alpes, Grenoble, France, ¹¹ Freiburg University Medical Center, Freiburg, Germany, ¹² HUCH Comprehensive Cancer Center, Stem Cell Transplantation Unit, Hensinski, Finland, ¹³ Clinic for Hematology and Medical Oncology, Georg-August University, Göttingen, Germany, ¹⁴ Hôpital Lyon Sud, Hospices Civils de Lyon, Pierre Bénite, Lyon, France, ¹⁵ CHU de Lille, Univ Lille, INSERM U1286, Infinite, Lille, France, ¹⁶ Hematology Division and BMT, Chaim Sheba Medical Center, Tel Hashomer, Israel, ¹⁷ Hospital Clínic de Barcelona, Barcelona, Spain

Background: Acute myeloid leukemia with prior history of myelodysplastic syndrome (sAML-MDS) represents an identified subgroup of AML within the ELN (European LeukemiaNet) classification. Allogeneic hematopoietic cell transplantation (allo-HCT) represents the only curative treatment for these patients. The cytogenetic pathophysiological continuum between MDS and sAML-MDS prompted us to investigate, in sAML-MDS allo-HCT patients, the prognostic impact of cytogenetics by comparing the MDS Cytogenetic Scoring System (R-IPSS) (Greenberg et al, Blood 2012)) and the ELN classification (Döhner et al, Blood 2017).

Methods: All patients in this retrospective study fulfilled the following criteria: age ≥ 18 years; diagnosed with sAML-MDS and undergoing first allo-HCT in first complete remission (CR1); from a MSD (matched sibling donor) or UD (unrelated donor). Graft source was peripheral blood (PBSC) or bone marrow (BM). All patients underwent transplantation between January 2000 and December 2018. Genetic risk groups were classified, at transplant, according to both R-IPSS and ELN cytogenetic classifications. For the analysis, the R-IPSS was changed to 3 grades by pooling “very good” and “good” groups into a single “good” group, while “poor” and “very poor” groups formed a single “adverse” group. Endpoints of the study were leukemia-free survival (LFS), overall survival (OS), refined GVHD-free, relapse-free survival (GRFS), cumulative incidence of relapse (RI), non-relapse mortality (NRM), cumulative incidence of acute GVHD (aGVHD) grade II-IV and chronic GVHD (cGVHD).

Results: Overall, 1030 patients fulfilled the inclusion criteria. The median follow-up period was 36.5 (IQR 15.4-78.6) months. The 2 year-probability of LFS and OS was 48.1% (95%CI: 44.7-51.4) and 55.4% (95% CI: 52.1-58.7), respectively. The 2-year RI was 33.3% (95% CI: 30.2-36.4. In multivariate analysis, cytogenetic risk was a major prognostic factor for LFS, OS and RI. The R-IPSS was more selective than the ELN classification and was able to discriminate significantly between the good, intermediate, and adverse groups. Incremental age (per 10 year) and reduced intensity conditioning (RIC) regimen were associated with lower LFS. Transplant with a UD was associated with lower RI and, as expected, RIC was associated with higher RI. NRM at 2 years was 18.6% (95% CI: 16.2- 21.2); incremental age (per 10 year) was the only risk factor associated with higher NRM. At 180 days, CI of aGVHD grade II-IV was 23.9% and grade III-IV was 8.3%. In multivariate analysis, two factors were associated with higher aGVHD grade II-IV: transplant with a UD and female donor to a male recipient. A RIC regimen was associated with lower aGVHD grade II-IV. The 2-year chronic GVHD CI was 38.1% and extensive GVHD CI was 14.2%. In multivariate analysis, the two prognostic factors associated with higher cGVHD were year of transplant and female donor to male recipient. In vivo T cell depletion was associated with lower extensive cGVHD.

Conclusions: Cytogenetics is a major prognostic factor for sAML-MDS patients undergoing allo-HCT. Using the R-IPSS cytogenetic classification to evaluate prognosis in this subtype of high-risk patients was more discriminative than the ELN classification.

Disclosure: The authors have no disclosure.

O012. Trends And Predictive Factors for Outcome of Relapsed Philadelphia Positive Acute Lymphoblastic Leukemia After Allogeneic Hematopoietic Cell Transplantion: Improvement of Survival in Recent Years

Ali Bazarbachi¹, Myriam Labopin², Mahmoud Aljurf³, Riitta Niittyvuopio⁴, Hélène Labussière-Wallet⁵, Didier Blaise⁶, Ibrahim Yakoub-Agha⁷, Anna Grassi⁸, Christian Reinhardt⁹, Stig Lenhoff¹⁰, Pavel Jindra¹¹, Jakob Passweg¹², Iman Abou dalle¹, Michael Stadler¹³, Bruno Lioure¹⁴, Patrice Ceballos¹⁵, Eolia Brissot¹⁶, Sebastian Giebel¹⁷, Arnon Nagler¹⁸, Christoph Schmid¹⁹, Mohamad Mohty²

¹ American University of Beirut Medical Center, Beirut, Lebanon, ² Hôpital Saint Antoine, APHP, UMR-S938, Paris, France, ³ King Faisal Specialist Hospital & Research Centre Oncology (Section of Adult Haematolgy/BMT), Riyadh, Saudi Arabia, ⁴ HUCH Comprehensive Cancer Center, Stem Cell Transplantation Unit, Helsinki, Finland, ⁵ Hopital Lyon Sud, Hospices Civils de Lyon, Pierre Bénite, France, ⁶ Programme de Transplantation & Therapie Cellulaire, Centre de Recherche en Cancérologie de Marseille, Institut Paoli Calmettes, Marseille, France, ⁷ CHU de Lille, Univ Lille, INSERM U 1286, Infinite, Lille, France, ⁸ Hematology and Bone Marrow Transplant Unit, Bergamo, Italy, ⁹ University Hospital, Essen, Germany, ¹⁰ Skanes University Hospital, Lund, Sweden, ¹¹ Charles University Hospital, Pilsen, Czech Republic, ¹² University Hospital, Basel, Switzerland, ¹³ Hannover Medical School, Hannover, Germany, ¹⁴ Nouvel Hopital Civil, Strasbourg, France, ¹⁵ CHU Lapeyronie, Montpellier, France, ¹⁶ Service d’Hématologie Clinique et de Thérapie Cellulaire, Hôpital Saint Antoine, APHP, UMR-S938, Paris, France, ¹⁷ Maria Sklodowska-Curie Institute, Oncology Center, Gliwice Branch, Wybrzeze Armii Kr, Gliwice, Poland, ¹⁸ Chaim Sheba Medical Center, Tel Hashomer, Israel, ¹⁹ Universitätsklinikum Augsburg, Augsburg, Germany

Background: Despite the increasing use of tyrosine kinase inhibitors (TKI), allogeneic hematopoietic cell transplantation (allo-HCT) remains an important and potentially curative treatment modality for patients with Philadelphia positive (Ph+) acute lymphoblastic leukemia (ALL) in first complete remission (CR1). However, patients who progress after allo-HCT were historically reported to have a dismal prognosis and a short survival. Standard treatment modalities for this indication include reduction/withdrawal of immunosuppressive therapy, chemotherapy, TKI, donor lymphocyte infusion (DLI), a second allo-HCT or even palliative care. More recently, monoclonal antibodies such as blinatumomab and inotuzumab as well as CAR-T cells have been introduced in that setting. However, little information is available about the global impact of the current standard of care for relapsed Ph+ ALL after allo-HCT and about the predictive factors for outcome.

Methods: We compared outcomes at EBMT participating centers of 899 adult patients with Ph+ ALL who relapsed between 2000 and 2019 after allo-HCT performed in CR1. We included patients who received transplants from matched sibling donors (MSD) or matched unrelated donors (MUD). Median follow up for alive patients was 56 months.

Results: Median age at transplant and at relapse was 44 and 45.4 years, respectively. Overall, 116 patients relapsed between 2000-2004, 225 between 2005-2009, 294 between 2010-2014, and 264 between 2015-2019. Patient and transplant characteristics were similar over the 4 time periods except for a progressive increase in the use of MUD (from 34.5% between 2000-2004 to 55.7% between 2015-2019; p = 0.0002), peripheral blood stem cells (from 60.3% to 84.5%; p < 0.0001), reduced intensity conditioning (RIC) (from 16.4% to 34.5%; p = 0.004), and in vivo T cell depletion (TCD) (from 27.9% to 62.4%; p = 0.0001) as well as a progressive decrease in the use of total body irradiation (TBI) (from 73.3% to 53%; p = 0.0002), respectively. The 2-year overall survival (OS) after relapse was 41.5 % (95% CI: 38 - 44.9). Original disease was the cause of death in 68.5% of patients. Importantly, the 2-year OS after relapse increased from 27.8% for patients relapsing between 2000-2004 to 31.7% for 2005-2009, 44.5% for 2010-2014 and 54.8% for 2015-2019 (p = 0.001) (Figure 1). A second allo-HCT within 2 years after relapse was performed in 13.9% of patients resulting in a 2-year OS of 35.9%. The incidence of second allo-HCT was 21.7%, 12.8%, 9.9% and 15.7% for the same time periods (p = 0.027). In multivariate analysis, OS from relapse was positively affected by an increased time from transplant to relapse above the median of 7.1 months (hazard ratio [HR] 0.71; p = 0.0007) and the year of relapse (reference 2000-2004; HR 0.68, p < 0.013 for patients relapsing from 2005-2009; HR 0.72; p < 0.008 for patients relapsing from 2010-2014, and HR = 0.72; p = 0.016 for patients relapsing from 2015-2019). Other patient, donor and transplant characteristics had no significant effect on outcomes.

Conclusions: We observe a major progressive improvement in OS from post-transplant relapse for patients with Ph+ -ALL, likely reflecting the efficacy of post-transplant salvage. These large-scale real-world data can serve as a benchmark for future studies in this setting.

Figure 1.

Disclosure: The authors declare no conflict of interest.

O013. Comparison of Haploidentical Peripheral Blood Cell Transplantation Using Post-Transplant Cyclophosphamide With Double Cord Blood Transplantation in Adults With Acute Leukemia

Annalisa Ruggeri¹, Jacques Emmanuel Galimard², Myriam Labopin², Didier Blaise³, Jose-Luiz Diez-Martin⁴, Fabio Ciceri¹, Yener Koc⁵, Patrice Chevallier⁶, Jan J Cornelissen⁷, Emma Nicholson⁸, Montserrat Rovira⁹, Zafer Gulbas¹⁰, Jiri Pavlu¹¹, Edouard Forcade¹², Jan Vydra¹³, Jurgen Kuball¹⁴, Luca Castagna¹⁵, Fermin Sanchez Guijo¹⁶, Frederic Baron¹⁷, Jaime Sanz¹⁸, Eliane Gluckman¹⁹, Alexandros Spyridonidis²⁰, Bipin Savani²¹, Arnon Nagler²², Mohamad Mohty²

¹ San Raffaele Scientific Institute, Milano, Italy, ² Saint Antoine Hospital, Paris, France, ³ Institute Paoli Calmettes, Marseille, France, ⁴ Madrid University, Madrid, Spain, ⁵ Istanbul University, Istanbul, Turkey, ⁶ CHU Nantes, Nantes, France, ⁷ Erasmus University, Rotterdam, Netherlands, ⁸ Royal Marsden Hospital, Leukaemia Myeloma Units, London, United Kingdom, ⁹ Hospital Clinic, Institute of Hematology & Oncology, Barcelona, Spain, ¹⁰ Anadolu Medical Center Hospital, Anadolu, Turkey, ¹¹ Imperial College, London, United Kingdom, ¹² CHU Pessac, Bordeaux, France, ¹³ Hematology Unit, Prague, Czech Republic, ¹⁴ Medical Center, Utrecht, Netherlands, ¹⁵ Humanitas Cancer Center, Milano, Italy, ¹⁶ Salamanca Hospital, Salamanca, Spain, ¹⁷ Liege University, Liege, Belgium, ¹⁸ La Fe Hospital, Valencia, Spain, ¹⁹ Eurocord Hôpital Saint Louis, Paris, France, ²⁰ Patras University, Patras, Greece, ²¹ Vanderbilt Hospital, Nashville, United States, ²² Sheba University, Tel Hashomer, Israel

Background: Study aim was to compare outcomes of Haploidentical hematopoietic-cell transplantation (Haplo-HCT) using post-transplant cyclophosphamide (PT-Cy) with double-cord-blood-transplantation (dCBT).

Methods: We report a retrospective study on adults with de-novo acute myeloid leukemia or acute lymphoblastic leukemia (AML or ALL) transplanted either with Haplo-HCT with peripheral blood stem cell (PBSC) or dCBT (as first allo-HCT) in Acute Leukemia Working Party (ALWP) of EBMT centers from 2013-2018.

Results: 981 patients (median age 50 years), 230 dCBT and 751 Haplo-HCT, were included. Median follow-up was 2 years. There was no difference between the two groups, except dCBT recipients were transplanted less recently (p < 0.01) using reduced-intensity-conditioning (RIC) regimen (p < 0.01). Diagnosis was AML in 72% of cases and majority of patients (67%) were transplanted in first complete remission (CR1). Conditioning regimen varied according to type of donor; in dCBT, total-body-irradiation (TBI) with fludarabine-cyclophosphamide was used in 95% of RIC and 87% of myeloablative conditioning (MAC) respectively, while in Haplo-HCT busulfan-thiotepa-fludarabine was used in 38% of RIC and 34% of MAC. Graft-versus-host disease (GVHD) prophylaxis was antithymocyte-globulin(ATG) free, cyclosporine-A (CSA) and mycophenolate-mofetil (MMF) was reported in 88% of dCBT while Haplo-HCT received PT-Cy mainly with CSA+MMF (53%) or with MMF+tacrolimus (29.6%).

Cumulative incidence function (CIF) of 60-days neutrophil-recovery was 93% (95%CI 89-96) and 95% (95%CI 93-97) for dCBT and Haplo-HCT. Day-100 CIF of grade II-IV acute-GVHD was 36% for Haplo-HCT and 42% for dCBT, p = 0.051, with no significant difference for grade III-IV aGVHD, p = 0.16. The 2-year CIF of chronic-GVHD was 35% and 29% for Haplo-HCT vs dCBT, p = 0.16. In the multivariate analysis (MVA) no factors were significantly associated with the risk of acute or chronic GVHD, dCBT recipients had a lower risk of extensive chronic GVHD (hazard ratio [HR] 0.47, p = 0.03) compared to haplo-HCT. The CIF of 2-year relapse was 20% in both donor groups, while 2-year NRM was 21.9% (95%CI 17-28) and 24% (95%CI 21-27), p = 0.66, for dCBT and Haplo-HCT, respectively. Diagnosis of ALL in CR2 was the only factor independently associated with relapse incidence in MVA (ALL in CR2, HR 2.09 (95%CI 1.77-3.48, p < 0.01) as compared with AML in CR1, CR2 or ALL in CR1. Disease status was also associated with NRM (AML in CR2, HR 1.56 (95% CI1.1-2.2), p = 0.01; ALL in CR2, HR 2.82 (95%CI 1.75-4.54), p < 0.01), as well as older age (HR 1.26 (95%CI 1.13-1.42), p < 0.01).

Two-year leukemia-free-survival (LFS) was 58% and 56%, p = 0.92, for dCBT and haplo-HCT, respectively. LFS was 51.7% for ALL and 58.3% for AML (p = 0.03), and 59.3% vs 50.3% for CR1 and CR2 (p < 0.01), respectively. In MVA, older age at HCT (HR 1.1 (95%CI 1.01-1.19), p = 0.03), and ALL transplanted in CR2 (HR 2.48 (95%CI 1.77-3.48), p < 0.01) were associated with reduced LFS. Donor type was not associated with OS (HR 0.97 (95%CI 0.71-1.34), p = 0.87) or with LFS (HR 0.97 (95%CI 0.72-1.30), p = 0.83).

Conclusions: Our results show comparable outcomes among Haplo-HCT PBSC with PT-Cy and dCBT in patients with acute leukemia in CR1 and CR2. Results were not different for AML and ALL. Age and disease status remain the most important prognostic factors for outcome.

Disclosure: The authors declare no COI.

O014. The Impact of Cytogenetic Risk on Outcome of Non-T Cell Depleted Haploidentical Stem Cell Transplantation in Relapsed/Refractory AML: A Study From The ALWP / EBMT

Arnon Nagler¹, Myriam Labopin², Fabio Ciceri³, Johanna Tischer⁴, Didier Blaise⁵, Renato Fanin⁶, Benedetto Bruno⁷, Edouard Forcade⁸, Jan Vydra⁹, Patrice Chevallier¹⁰, Claude Eric Bulabois¹¹, Pavel Jindra¹², Friedrich Stölzel¹³, Jonathan Canaani¹, Jaime Sanz¹⁴, Bipin Savani¹⁵, Alexandros Spyridonidis¹⁶, Sebastian Giebel¹⁷, Eolia Brissot¹⁸, Ali Bazarbachi¹⁹, Jordi Esteve²⁰, Mohamad Mohty²

¹ Sheba Medical Center, Ramat Gan, Israel, ² Saint Antoine Hospital, Paris, France, ³ Ospedale San Raffaele, Milano, Italy, ⁴ Klinikum Grosshadern, Med. Klinik III, Munich, Germany, ⁵ Therapie Cellulaire, Centre de Recherche en Cancérologie de Marseille, Institut Paoli Calmettes, Marseille, France, ⁶ Azienda Ospedaliero Universitaria di Udine, Udine, Italy, ⁷ S.S.C.V.D Trapianto di Cellule Staminali A.O.U Citta della Salute e della Scienza di Torino, Torino, Italy, ⁸ CHU Bordeaux, Hôpital Haut-Leveque, Pessac, France, ⁹ Servicio de Hematología, Institute of Hematology and Blood Transfusion, Prague, Czech Republic, ¹⁰ CHU Nantes, Nantes, France, ¹¹ CHU Grenoble Alpes - Université Grenoble Alpes, Service d`Hématologie, Grenoble, France, ¹² Charles University Hospital, Pilsen, Czech Republic, ¹³ Universitaetsklinikum Dresden,Medizinische Klinik und Poliklinik I, Dresden, Germany, ¹⁴ University Hospital La Fe, Valencia, Spain, ¹⁵ Vanderbilt University Medical Center, Nashville, United States, ¹⁶ University Hospital of Patras, Patras, Greece, ¹⁷ Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice, Poland, ¹⁸ Hôpital Saint Antoine, Paris, France, ¹⁹ American University of Beirut Medical Center, American University of Beirut Medical Center, Beirut, Lebanon, ²⁰ Hospital Clínic de Barcelona, Barcelon, Spain

Background: Baseline cytogenetics remains the single most important determinant of outcome in patients (pts) with AML and in conjunction with disease status, is a key factor predicting transplantation outcome, including haploidentical stem cell transplantation (HaploSCT). Less is known about its value in pts with primary refractory or relapsed (Rel-Ref) AML undergoing HaploSCT.

Methods: Study aim was to investigate the impact of cytogenetic risk (defined by the Medical Research Council as intermediate or adverse risk group) in adult pts (≥18 years) with de novo AML undergoing first HaploSCT with Rel-Ref AML (>5% of residual morphologic blasts) in 2006-2019, using the EBMT/ALWP registry. Multivariate analysis (MVA) adjusting for differences between the groups was performed using Cox’s proportional hazards regression model for main outcomes.

Results: 460 pts were included: 299 (65%) with intermediate risk and 161 (35%) with adverse risk cytogenetics. Median F/U was 24 (20-38) and 37 [34-56] months, respectively (p = 0.64). Median age was 53.5 (18.1-77.8) and 54.3 (19.4-74) y (p = 0.51), respectively. 52.8% and 57.8% were male (p = 0.31). Pre-HaploSCT disease status was Rel 1/2 in 53.8% and 28.6% and Ref in 46.2% and 71.4% of the pts, respectively (p < 0.0001). Conditioning was myeloablative in 40.6% and 34%, respectively (p = 0.22). Graft sources were peripheral blood in 69.9% and 69.6% of transplants, respectively. Karnofsky performance score was > 90 in 52.8% and 57.3% of pts (p = 0.37), respectively. The most frequent GVHD prophylaxis was post-transplant cyclophosphamide in 74.9% and 72.7%, respectively (p = 0.73). Engraftment was achieved by 91% and 92.3% of pts respectively, (p = 0.64). Day 180 incidence of acute (a) GVHD II-IV and III-IV was 30.0% and 32.6% (p = 0.51); 15.4% and 11.3 % (p = 0.3) and 2-year total and extensive chronic (c) GVHD was 22.3% and 15.3 % (p = 0.09), 11.1% and 6.6% (p = 0.04), respectively. Complete response (CR) was achieved by 76.3% and 71.9% of intermediate and adverse risk groups respectively (p = 0.32). Two-year relapse rate was inferior while leukemia free survival (LFS) and overall survival (OS) rates were superior for pts with intermediate vs poor risk cytogenetics, 46.6% vs 65.9% (p = 0.001), 28.3% vs 12.9% (p = 0.004) and 31.5% vs 17% (p = 0.005), respectively. Non-relapse mortality (NRM) and GVHD-free, relapse-free survival (GRFS) did not differ between the groups with 25.1% vs 21.2% (p = 0.38) and 20.2% vs 10.8 % (p = 0.13), respectively. In MVA, Rel was significantly higher and LFS and OS significantly lower for pts with adverse risk compared to intermediate risk cytogenetics, hazard ratio (HR) = 1.5 (95% CI 1.11-2.02, p = 0.008), HR = 1.29 (95% CI 1.01-1.66, p = 0.041) and 1.32 (95% CI 1.02-1.71, p = 0.037), respectively. NRM and GRFS did not differ, HR = 0.92 (0.59-1.44, p = 0.71) and HR = 1.1 (0.87-1.4, p = 0.42). Risk of aGVHD was similar, HR = 1.06 (0.72-1.56, p = 0.75); as was risk of cGVHD, HR 0.88 (0.5-1.55, p = 0.66).

Conclusions: Cytogenetic risk is significantly associated with transplantation outcome in Rel-Ref AML pts undergoing non-T cell depleted haploidentical transplantation while also maintaining its important prognostic role in pts with active leukemia. HaploSCT cannot overcome the deleterious effect of adverse-risk cytogenetics attesting to a comparable graft versus leukemia effect to that observed in other types of allogeneic transplantation.

Disclosure: Nothing to declare.

O015. Prevalence and Outcome of CNS Relapse after Allogeneic Hematopoietic Stem Cell Transplantation In Patients Suffering from AML and all. A Study from the ALWP-EBMT

Sabine Blum¹, Yves Chalandon², Myriam Labopin³, Jürgen Finke⁴, Tobias Gedde-Dahl⁵, Tarek Ben Othman⁶, Jan J Cornelissen⁷, Pavel Jindra⁸, Hélène Labussière-Wallet⁹, Matthew Collin¹⁰, Stig Lenhoff¹¹, Guido Kobbe¹², Dolores Caballero¹³, Arnon Nagler¹⁴, Mohamad Mohty¹⁵

¹ University Hospital and University of Lausanne, Lausanne, Switzerland, ² University Hospital Geneva, Geneva, Switzerland, ³ EBMT Paris Study Office, Saint Antoine Hospital, INSERM UMR 938, Sorbonne University, Paris, France, ⁴ University Hospital of Freiburg, Freiburg, Germany, ⁵ University Hospital Oslo, Oslo, Norway, ⁶ Centre National de Greffe de Moelle, Tunis, Tunisia, ⁷ Erasmus MC Cancer Institute, Rotterdam, Netherlands, ⁸ Charles University Hospital, Pilsen, Czech Republic, ⁹ Hôpital Lyon Sud, Pierre Bénite, Lyon, France, ¹⁰ Freeman Hospital, Newcastle_Upon Tyne, United Kingdom, ¹¹ Skanes University Hospital, Lund, Sweden, ¹² University Hospital Duesseldorf, Duesseldorf, Germany, ¹³ Hospital Clínico, Salamanca, Spain, ¹⁴ Chaim Sheba Medical Center, Tel Hashomer, Israel, ¹⁵ University Hospital Pierre et Marie Curie, Paris, France

Background: Central nervous system (CNS) relapses in acute leukemia (AL) are a serious form of disease relapse, especially after hematopoietic stem cell transplantation (HSCT). The prevalence of this complication and the overall survival (OS) after treatment as well as the best treatment strategy remains largely unknown.

Methods: Patients suffering from acute myeloid leukemia (AML) or acute lymphoblastic leukemia (ALL) with or without CNS relapse undergoing HSCT between 1996 and 2016 within the EBMT database were identified. A specific questionnaire was sent through the Acute Leukemia Working Party (ALWP)-EBMT data office to transplant centers that then reported patients with or without CNS relapse, together with relapse type, treatment and outcome.

Results: From the participating centers, a total of 7991 adult patients transplanted for either AML or ALL between 1996 and 2016 were included in this study. Ninety-one patients had a CNS relapse; median age was 41 years (range (r), 18-75.2). No difference could be detected between the groups according to donor type, sex of the patient or donor, conditioning regimen, time from diagnosis to transplant, stem cell source, CMV viral status or prior CNS involvement. The incidence of CNS relapse at 5 years was 0.9% in AML and 1.9% in ALL (p = 0.002). Relapse was more frequent after 2010 (the median year of HSCT), than before (1.6% vs 0.9%, p = 0.013). The median follow-up after CNS relapse was 88.3 months (IQR, 35.2-119.9). The median time from HSCT to CNS relapse was 396 days (r, 18-3644). The CNS relapse was isolated in 36 (39.6%) patients, it was combined (CNS and bone marrow (BM)) in 30 (33%). The CNS relapse occurred after the hematological relapse in 23 (25.3%) and after molecular relapse in 2 (2.2%) patients. Before CNS relapse, 66 (72.5%) patients did not have GvHD, 25 (27.5%) did.

A complete remission (CR) was achieved in 47 (52.8%) patients after CNS relapse, while 42 patients (47.2%) did not achieve CR, data was missing for two patients. There was a trend for more CR in ALL than in AML (64.1% vs 44%, p = 0.06). Nineteen (20.9%) patients had a second CNS relapse after the first.

For patients with CNS relapse as first relapse post transplant (n = 66), the OS at two years after CNS relapse was 28.8% in AML and 19.4% in ALL patients (ns), 8.2% and 15.5% at 5 years, respectively. The only factor associated with a significantly better OS was isolated CNS relapse as compared to CNS and BM combined (37.5% vs 6.7% at 2 years, p = 0.004).

Conclusions: CNS relapse after HSCT is a rare event (1.14%), more frequent in ALL than in AML. More CNS relapses were detected after 2010, possibly due to information bias with better reporting of CNS relapse in recent years. Half of the patients could achieve CR, but with a low OS in patients with CNS as first relapse (8.2% in AML and 15.5% in ALL, at five years). Patients with an isolated CNS relapse had better OS at two years (37.5% vs 6.7%).

Disclosure: No conflict of interests to declare.

O016. Prognosis of Patients with Acute Lymphoblastic Leukemia Relapsing After An Allogeneic Stem Cell Transplantation. Study of 132 Patients

Christelle Ferra Coll¹, Mireia Morgades de la Fe¹, Laura Prieto García², Antonio Campos Júnior³, María Inmaculada Heras Fernando⁴, Rebeca Bailén Almorox⁵, Irene García Cárdenas⁶, Marisa Calabuig Muñoz⁷, Teresa Zudaire Ripa⁸, Joud Zanabili Al-Sibai⁹, Pere Barba Suñol¹⁰, Beatriz Aguado Bueno¹¹, Anna Torrent Catarineu¹, Oriana López-Godino⁴, Rodrigo Martino Bufarull⁶, Mi Kwon⁵, Carlos Pinho Vaz³, Dolores Caballero Barrigón¹², Josep-Maria Ribera Santasusana¹

¹ Institut Català d’Oncologia-Badalona, Badalona, Spain, ² Hospital Universitario de Salamanca. IBSAL (Instituto Biosanitario de Salamanca), Salamanca, Spain, ³ Instituto Português de Oncologia, Porto, Portugal, ⁴ Hospital General Universitario Morales Meseguer, Murcia, Spain, ⁵ Hospital General Universitario Gregorio Marañón, Madrid, Spain, ⁶ Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, ⁷ Hospital Universitario Clínico de Valencia, Valencia, Spain, ⁸ Complejo Hospitalario de Navarra, Pamplona, Spain, ⁹ Hospital Universitario Central de Asturias, Oviedo, Spain, ¹⁰ Hospital Universitari Vall d’Hebrón, Barcelona, Spain, ¹¹ Hospital Universitario La Princesa, Madrid, Spain, ¹² Hospital Universitario de Salamanca, IBSAL (Instituto Biosanitario de Salamanca), Salamanca, Spain

Background: Allo-SCT is indicated in patients with high-risk ALL. However, the prognosis is poor for those who relapse after transplant. There are relatively few data reported in this setting. A retrospective evaluation of outcome for patients with ALL who have relapsed after allo-HSCT is presented.

Methods: One hundred and thirty two patients with relapsed ALL have been identified after a first allo-SCT performed between March 1998 and February 2019, in 11 centres belonging to the Spanish Group for Hematopoietic Transplantation and Cell Therapy (GETH).

Results: The median time [range] between allo-SCT and relapse was 6.87 months [0.93-116.50]. Patient characteristics are listed in Table 1. One hundred and six pts received myeloablative conditioning and 114 pts peripheral blood progenitors. Forty six (44%) and 59 (56%) pts showed complete and mixed chimerism on relapse. In 22 pts (17%), palliative treatment was adopted. Among 110 pts (83%) that received a rescue treatment, 73 pts achieved CR and 40 pts relapsed again. Treatment was chemotherapy (n = 82), TKi (n = 26 pts, all Phi+ ALL) or immunotherapy (n = 19) (inotuzumab and/or blinatumumab). Cell therapy consisted of: donor lymphocyte infusion (29 pts), 2nd allo-SCT (37 pts [34%]) and CAR T 19 therapy (14 pts [13%]). The causes of death were ALL progression (n = 80), toxicity related to rescue treatment (n = 8) and TRM (n = 12) after a 2nd allo-SCT. 30 pts are alive in CR and 1 in relapse. The probabilities of survival at 1 and 5 years were 44% (95%CI: 36%; 52%) and 19% (95% CI: 11%; 27%) respectively with 3.16 years [0.57; 19.34] median follow-up (Figure 1). For the 37 pts who received a 2nd allo-SCT, 5y OS probability was 40% [22%; 58%]. For pts with ALL relapsed after allo-SCT: age >30 yrs, Ph negative ALL, early period for 1st SCT (1998-2013), early relapse (<12 months between 1st allo-SCT and relapse), status at 1st allo-SCT (>1st CR) and chronic GVHD showed a significant influence in OS in the univariable analysis. Younger age (HR: 1.996 [1.314 ; 3.032]), recent period for allo-SCT (HR: 2.184 [1.428 ; 3.340]), late relapse (HR: 2.221 [1.366 ; 3.612]), 1st CR at 1st allo-SCT (HR: 1.962 [1.262 ; 3.050]) and presence of chronic GVHD (HR: 1.761 [1.046 ; 2.965]) confirmed their impact in the multivariable analysis.

Table 1: Characteristics of the patients from the series. N = 132 (%)

-Age at SCT (years). Median [range]	32 [14 - 64]
-B cell ALL/T cell ALL.	104 (79)/28 (21)
-Time interval (months) between ALL diagnosis and SCT. Median [range]	6.15 [2.63 – 21.77]
-Type of relapse: overt cytologic/extra-medullary/MRD Positive	96 (73)/17 (13)/19 (14)
-Disease status at allo-SCT (1st CR/2nd CR/ >2nd CR or active disease).	87 (66)/32 (24) /13 (10)
-Donor characteristics: MSD/MUD/UCB/Haploidentical-RD.	77 (58)/38 (29)/11 (8)/6 (5)
-Immunosuppression: GVHD treatment/GVHD prophylaxis/Tappering.	28 (30)/44 (48)/20(22)
-Acute GVHD: I/II/III/IV	5/3/0/0
-Chronic GVHD: Limited/Extensive	5/9

Conclusions: Although pts with ALL who relapsed after a first allo-SCT have an unfavourable prognosis, those with younger age, late relapse and, allo-SCT performed in 1st CR can be satisfactorily rescued.

Clinical Trial Registry: NA

Disclosure: Founding: Supported in part by grants from CERCA Program, Generalitat de Catalunya, Spain and “La Caixa” Foundation.

O017. Venetoclax Combined with Hypomethylating Agent (HMA) is Safe And Effective May Be A Good Bridge to Transplant in High-risk Acute Myeloid Leukemia

Vincenzo Federico¹, Michelina Dargenio¹, Rosella Matera¹, Angelodonato Canaris¹, Valentina Bozzoli¹, Vincenza Caretto¹, Daniela Carlino¹, Maria Rosaria De Paolis¹, Maria Paola Fina¹, Giuseppina Loglisci¹, Annarita Messa¹, Giovanni Reddiconto¹, Stefania Scardino¹, Davide Seripa¹, Annarosa Soda¹, Carolina Vergine¹, Nicola Di Renzo¹

¹ ’Vito Fazzi’ Hospital, ASL Lecce, Lecce, Italy

Background: High-risk acute myeloid leukemia (HR-AML), as well older patients are characterized by high frequency of cytogenetic adverse-risk features or comorbidities, being them less responsive to induction therapy with poor outcome as R/R AML patients. Venetoclax, an oral highly selective BCL-2 inhibitor, demonstrated a synergistic anti-leukemia activity when combined with HMA without additional significant toxicity. Here we report the outcome of patients with de novo-AML or R/R AML treated with venetoclax and HMA aimed to evaluate efficacy and safety of this combination and its role as a bridge to transplant.

Methods: From October 2018 to October 2020, a total of 24 patients, median age was 69 years (range 27-80), 14 (60%) with de novo-AML and adverse/complex cytogenetics features according to ELN criteria not eligible for intensive induction therapy, and 10 (40%) with R/R AML, were included in the study. Among patients with R/R AML, 3 patients relapsed after HMA, 5 after induction therapy, and 2 after allogenic transplant. After rum-up, all patients received Venetoclax 400 mg/daily orally in 28-day cycles combined with decitabine 20 mg/m² days 1-5 of each 28-day cycle, 15 patients (63%) or azacitidine 75 mg/m² days 1-7 of each 28-day cycle, 9 patients (37%). All patients received a median of 3 cycles (range 1-19) of venetoclax in combination with HMA.

Results: Composite complete remission (CR+iCR) rate was 60%, 75% and 50% for all patients, de novo-AML, and R/R AML, respectively. In both groups, the median time to response was 2 months (range 1-5). No induction death was observed. At 5 months of median follow-up (range 2-22), 14 out of 24 patients (58%) are alive, including 5 patients (21%) still on therapy and in CR, while 10 patients (42%) died, mostly because progressive disease. After a median of 3 cycles (range 3-4) of venetoclax plus HMA, 9 patients, 6 R/R (60%) and 3 de novo-AML (40%) underwent to allogenic HSCT (6 haploidentical, 1 matched-related donor and 2 MUD). The most common grade 3/4 hematological toxicity was anemia (58%), thrombocytopenia (45%), leukopenia (62%) and febrile neutropenia (31%). No grade 3/4 non-hematological toxicity was observed. There was no statistically significant difference in OS between patients with de novo AML or R/R AML (11.7 vs 11.9 months), [p = 0.527]). In contrast, median OS was better for transplanted patients compared to did not, 9.7 months, (range 5.5-14.0) vs 3.0 months, (range 0.8-5.2), [p = 0.001]. After a median follow-up of 13 months (range 2.2-22.0) all patients transplanted are alive and in CR.

Conclusions: The treatment of HR-AML, R/R-AML, as well as ederly AML, is challenging for hematologist. The combination of venetoclax and HMA has shown to be safe and effective in this setting of patients with high response rate without significant toxicity. In our series about one half of patients underwent to transplant making this combination a good option as bridge to transplant.

Disclosure: no disclosure to declare.

O018. Abstract already published

O019. Comprehensive Geriatric Assessment, Allogeneic HSCT and Survival in AML Patients 65-75 Years Old

Gabriele Magliano¹, Luana Fianchi^1,2, Patrizia Chiusolo^1,2, Federica Sora’^1,2, Luca Laurenti^1,2, Elisabetta Metafuni², Sabrina Giammarco², Idanna Innocenti², Francesco Autore², Simona Sica^1,2, Marianna Criscuolo², Livio Pagano^1,2, Andrea Bacigalupo^1,2

¹ Università Cattolica del Sacro Cuore, Roma, Italy, ² Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Roma, Italy

Background: Acute myeloid leukemia (AML) in patients over the age of 65 carries a poor prognosis, due to unsatisfactory control of the disease with chemotherapy alone. Allogeneic hematopoietic stem cell transplantation (HSCT) can provide significant anti-leukemic effect in eligible patients. The aim of our study was to assess the feasibility and outcome of allogeneic HSCT in a prospective cohort of AML patients aged 65-75.

Methods: 44 consecutive de novo or secondary AML patients, aged 65-75 years, were diagnosed in our Unit between September 2018 and August 2020. Patients were risk-stratified according to ELN 2017 criteria and classified as FIT, UNFIT and FRAIL according to a comprehensive geriatric assessment (CGA), which included ECOG, CIRS, ADL and IADL scores. Only fit patients with intermediate-high ELN 2017 risk score were eligible for HSCT.

Results: The median age was 70 years (range 65-75). 9 patients presented with favorable, 16 with intermediate and 15 with unfavorable ELN2017 risk. 19 patients were classified as FIT, 12 as UNFIT and 13 as FRAIL. All FRAIL patients were treated with best supportive care, except for one with hypomethylating agents (HMAs). UNFIT patients were treated with HMAs (n = 8), citarabine+Flt-3 inhibitors (n = 1), CPX-351 (n = 1), FLA regimen (n = 1), one refused treatment.

In the FIT group, induction was performed with standard chemotherapy (n = 9), CPX-351 (n = 5), HMAs (n = 1), low dose citarabine (n = 1), and allo HSCT upfront (n = 3). Median CIRS score was 1 (1-4). 12/19 had ECOG score 0, whereas 18/19 had IADL over 5/8.

A complete remission (CR) was achieved in 15/44 patients overall (39%) and in 15/19 FIT patients (79%). At last follow-up, 14/44 patients were alive (median 135 days, range 1-813), with an actuarial 2 year survival of 21% . Survival was 0% for FRAIL and UNFIT patients and 52% for FIT patients (Figure 1).

Allogeneic HSCT. 13/19 FIT patients were allografted. Reasons for not grafting were early relapse (n = 2), refusal (n = 2) and waiting list (n = 2). The donor type was HLA haploidentical (n = 6), MUD (n = 4), MSD (n = 2), cord blood (n = 1). The stem cell source was unmanipulated peripheral blood cells (n = 7), bone marrow (n = 5), or cord blood (n = 1). The conditioning was non myeloblative (n = 2), reduced intensity (n = 9) or myeloablative (n = 2). GvHD prophylaxis consisted of cyclosporin, mofetil mycophenolate and post transplant cyclophosphamide in 9/13 patients. Acute GvHD grade (II-IV) developed in 5 patients (38%) and chronic GvHD in 3 (23%). The actuarial 2 year survival is 60%, and the median survival from diagnosis is 365 days (60-813) ; the cause of death was transplant related (n = 2, 15%) and relapse (n = 2, 15%).

Figure 1.

Conclusions: CGA has a strong influence on treatment strategies in elderly AML. An allogeneic HSCT was performed in 68% of FIT patients with promising results: bed availability was the single most important factor delaying transplantation. New therapies are required for UNFIT and FRAIL patients.

Disclosure: Nothing to declare.

Aplastic Anaemia

O020. HLA Evolutionary Divergence Influences Characteristics and Outcomes of Patients with Aplastic Anemia and Paroxysmal Nocturnal Hemoglobinuria

Simona Pagliuca^1,2, Carmelo Gurnari¹, Bhumika Patel¹, Hassan Awada¹, Ashwin Kisthagari¹, Cassandra Kerr¹, Sunisa Kongkiatkamon¹, Laila Terkawi¹, Misam Zawit¹, Thomas LaFramboise³, Valeria Visconte¹, Jaroslaw Maciejewski¹

¹ Cleveland Clinic Foundation, Cleveland, United States, ² University of Paris, Paris, France, ³ Case Western University, Cleveland, United States

Background: HLA evolutionary divergence (HED) is a continuous metric that quantifies the breadth of the immunopeptidome presented by an individual’s HLA allotype. High HED scores have been shown to be associated with better outcomes after immune check point inhibitor therapy in solid tumors. However, the role of HED has never been assessed in autoimmune disorders, and may have an impact on disease susceptibility and outcomes.

Here we explored the impact of HED on clinical features of aplastic anemia (AA) and paroxysmal nocturnal hemoglobinuria (PNH). The immunological landscape of those diseases is charecterized by an HLA-restricted CD8+ and CD4+ mediated autoimmunity which leads to hematopoietic stem cell distruction, with a crucial role played by immune-dominant PNH clones and immune escape mechanisms likely associated to the down-regulation of the HLA-antigen presentation. Since higher HED scores reflect a wider antigenic spectrum and potentially more diverse peptide presentation capabilities, we hypothesize that this context not only predisposes to the elicitation of T-cellular response eventually by auto-antigen presentation but also that HLA structural diversity can impact on disease phenotype and response to therapy.

Methods: Using an NGS-based platform targeting the entire HLA locus, we have sequenced DNA samples from a well annotated study cohort of 216 AA, PNH, or AA/PNH patients from Cleveland Clinic Foundation. We then calculated HED at the classical HLA class I and II loci (A, B, C, DRB1, DQA1, DQB1, DPB1), and investigated its impact on disease course and outcomes.

Results: Analysis of HED metrics across several disease parmeters at diagnosis, and binary or time-dependent endpoints, demonstrated that higher HED at class I and II HLA loci was generally associated with unfavorable disease characteristics and response outcomes.

When we examined the configuration of HED across patient groups we observed a significant difference in distribution of class II HED, with higher values in patients with pure AA phenotype vs primary hemolytic PNH (p = 0.0041). Interestingly, this configuration was seen for most of class II loci in study (DQB1, DRB1, DQA1). Also, high mean HED in class I correlated with a more severe phenotype at diagnosis in patients with AA [OR = 1.2 (95%CI: 1.01-1.42), p = 0.0008] while increased mean in class II was associated with earlier age at onset (inverse correlation between class II mean HED values and age at presentation: R²= 0.0467, p = 0.0014 especially for DQB1 and DPB1 loci). Looking at disease outcomes, higher global mean HED negatively influenced response to immunosuppression [OR = 1.18 (95%CI: 1.03-1.42), p = 0.0016], however, interestingly, higher scores were also associated with a lower risk of progression to MDS/AML [HR = 0.5 (95%CI: 0.2-0.99)], indicating that, although an increased immunopeptidomic variability may elicit more pervasive auto-immune responses, this environment could be beneficial for the anti-tumor surveillance also in a context of increased immune pressure.

Conclusions: In summary, our results indicate that higher HLA divergence clinically correlates with severity, early disease onset and refractoriness to immune-suppression in AA patients and, more in general, that an increased structural diversity of MHC complexes may dictate the capability to present more self-peptides eliciting dominant autoimmune T cell responses.

Disclosure: Nothing to declare.

O021. The Genomic Landscape of Myeloid Neoplasms Arising from Aplastic Anemia and Paroxysmal Nocturnal Hemoglobinuria

Carmelo Gurnari^1,2, Simona Pagliuca^1,3, Bhumika Patel¹, Hassan Awada¹, Cassandra Kerr¹, Wenyi Shen¹, Sunisa Kongkiatkamon¹, Laila Terkawi¹, Misam Zawit¹, Jibran Durrani¹, Valeria Visconte¹, Seth Corey¹, Maria Teresa Voso², Hetty Carraway¹, Jaroslaw Maciejewski¹

¹ Cleveland Clinic, Cleveland, United States, ² University of Rome, Tor Vergata, Rome, Italy, ³ University of Paris, Paris, France

Background: Up to 20% of aplastic anemia (AA) patients treated with immunosuppression will evolve to myeloid neoplasia (MN) over a median time of 10 years. Myeloid progression is characterized by dismal outcomes (Fig.1A). The pathogenesis of MN post-AA is diverse and will often include antecedent clonal facilitating events that herald progression. In addition, progression to MN may reflect an immune escape due to selection pressure e.g., through acquisition of HLA mutations. Here, we studied the molecular landscape of MN arising from AA, to better understand its pathogenesis and ultimately to develop measures of early detection, prevention, and therapeutic strategies.

Methods: An integrative mutational analysis of myeloid/germline (GL) genes was performed to comprehensively evaluate their role within the scenario of AA/paroxysmal nocturnal hemoglobinuria (PNH) clonal evolution. By using a deep targeted-sequencing panel covering HLA classical loci, and applying an in-house newly developed pipeline for the study of the HLA region, we also detected class I/II HLA somatic mutations in a subset (n = 24) of these patients.

Results: Among 350 AA/PNH patients, 38 (11%, median age 61 years) developed a secondary MN (sMN; 77% MDS, 21% AML, 2% MPN). Evolution was less common in patients with moderate AA (7% vs. 14% in severe) or in the presence of a PNH clone (21% vs. 52% in non-progressors, p = 0.0003). Cytogenetics at evolution revealed abnormalities in 83% of patients, with chromosome 7 alterations in 47% of cases (-7, 35%; del(7q),12%). By comparison, -7/del(7q) are present in 7.5% of patients with primary MN (p = 0.0001, Fig.1B). First, we investigated GL alterations classified as Tier1 (9/38 patients) and Tier2 (11/38 patients). Tier1 variants included NF1, CBLC, SBDS and SAMD9L, were enriched in del(7q) patients (76%, p = 0.0001) and were detected in 24% vs. 8% in non-evolved cases (p = 0.008). A total of 148 somatic variants were found at evolution in 34/38 sMN patients with no differences in patients with or without chromosome 7 abnormalities (Fig.1C). ASXL1 (29% vs. 14%, p = 0.02) and SETBP1 (15% vs. 3%, p = 0.005) were more frequent in evolved cases (Fig.1D). When comparing patients with chromosome 7 abnormalities with de novo counterpart, sMN appeared most commonly mutated in ASXL1 (p = 0.02), SETBP1 (p = 0.0007), ETV6 (p = 0.02) and NF1 (p = 0.02) and had a shorter median survival time (12 vs. 48 months in de novo cases, p = 0.0002, Fig.1E, F). In a cross-sectional analysis (17 cases at onset/ 35 at progression), hits in TET2, DNMT3A and ASXL1 were found respectively in 5, 1 and 3 cases at baseline, with a higher clonal burden at sMN onset (median VAF 24% vs 43%, p = 0.0001; Fig.1.G-H). The HLA mutational analysis showed the presence of somatic class I/II loci variants in 21% of progressors. By comparison, in non-progressing AA patients HLA class I/II variants were found in 13% of cases (Fig.1I).

Conclusions: We demonstrate that AA progression to MN has distinct molecular characteristics. The presence of HLA mutations suggests that immune escape/selection may play a role, while the presence of GL predisposition variants shows that they not only may facilitate AA but also clonal evolution as described from classic congenital BMF.

Disclosure: Nothing to disclose.

O022. HLA Mutations in Paroxysmal Nocturnal Hemoglobinuria: A Possible Alternative Mechanism of Immune Escape

Carmelo Gurnari^1,2, Simona Pagliuca^1,3, Cassandra Kerr¹, Hassan Awada¹, Sunisa Kongkiatkamon¹, Laila Terkawi¹, Neetha Parameswaran¹, Misam Zawit¹, Adam Wahida⁴, Valeria Visconte¹, Jaroslaw Maciejewski¹

¹ Cleveland Clinic, Cleveland, United States, ² University of Rome, Tor Vergata, Rome, Italy, ³ University of Paris, Paris, France, ⁴ Munich Leukemia Laboratory, Munich, Germany

Background: Unlike leukemic driver mutations, PIGA mutations produce an escape phenotype in the context of immune-mediated bone marrow failure (BMF) such as aplastic anemia (AA). Another way to create clinical advantage will be to disable HLA-mediated cytotoxic T-cell recognition. Somatic hits in HLA genomic region have been previously assessed in AA patients. We stipulated that HLA mutations may contribute to the intrinsic expansion of PNH clones being: i) additive to the effects of PIGA mutations in creating immune escape or ii) redundant and thus not needed for the expansion of PNH clones.

Methods: Using a deep targeted-sequencing panel covering HLA classical loci, and applying an in-house newly developed pipeline for the study of the HLA region, we detected class I/II HLA somatic mutations of an initial cohort of 25 patients with hemolytic PNH (PNH granulocytes clone size >20% and LDH>x2.5 ULN). To further increase the sensitivity of HLA-mutations detection, we also performed the same analysis on additional 10 patients whose samples were previously sorted by flow cytometry to purity of glycosylphosphatidylinositol anchor proteins (GPI-AP) (-) cells. An integrative mutational analysis of PIGA and myeloid genes (63 genes) was then performed in order to comprehensively evaluate the scenario of PNH clonal evolution.

Results: Overall, median age at diagnosis (n = 35, M:F ratio 0.66) was 38 years (11-84) while median PNH granulocyte clone size was 88% (20.1-99). A total of 3 HLA somatic mutations were found in class-II loci in 25 PNH patients (12%) at a mean variant allele frequency (VAF) of 5.43%. Additional 20 samples from 10 PNH patients were analyzed from sorted GPI-AP(+) and GPI-AP(-) myeloid fractions (mean purity >95%). Six somatic mutations of HLA class I (N = 3) and class II (N = 3) were found in 4 patients (67% detected on GPI-AP(+) and 33% on GPI-AP(-) fractions) at a low VAF (mean 3.5%). Overall, mutated patients (7/35) were characterized by a male predominance (M:F ratio 2.5:1) and in 42% of cases, they had a previous history of AA.To further unveil PNH genomic landscape, we then looked at PIGA and myeloid genes mutations. A total of 104 PIGA mutations (Fig.1A) were detected, 62 in the whole blood samples (n = 25 patients) and 41 solely in the GPI-AP(-) fraction (mean VAF 29%) of the 10 sorted patients. Missense (44%, n = 46) and frameshift (33%, n = 34) mutations were the most common, followed by splice sites (13%, n = 14), non-sense (6%, n = 6) and non frameshift insertion/deletions (4%, n = 4) with 68% of cases harboring clonal mosaicism. Myeloid genes analysis revealed the presence of mutations in 17% (6/35) of patients with a mean VAF of 25%. Of note, PIGA was the ancestral hit in 67% of cases, none of the patients progressed to a myeloid disorder and in 2 cases we did not find any PIGA, myeloid or HLA mutations.

Conclusions: In summary, somatic HLA class-I/II mutations can be found in patients with PNH, occurring in GPI-AP(+) cells in subclonal fashion but also in GPI-AP(-) cells. Subclonal HLA mutations may impact the immune pressure on PNH clone dynamics, reflecting an alternative immune escape pathway (Fig.1B).

Disclosure: Nothing to disclose.

Autoimmune Diseases

O023. Early Morbidity of Autologous Haematopoietic Stem Cell Transplantation in Multiple Sclerosis- A Real World Experience

Varun Mehra¹, Maria Cuadrado¹, Elijah Rhone¹, Alice Mariottini¹, Victoria Williams², Sarah Ware¹, Paolo Muraro³, Antonio Pagliuca¹, Victoria Potter¹, Eli Silber¹, Majid Kazmi¹

¹ Kings College Hospital London, London, United Kingdom, ² Guys & St Thomas’ Hospital NHS Foundation Trust, London, United Kingdom, ³ Imperial College Healthcare NHS Trust, London, United Kingdom

Background: Autologous Haematopoietic Stem Cell Transplantation (aHSCT) is increasingly being used in treatment of patients with active inflammatory form of Multiple sclerosis (MS). The efficacy from this therapy is superior to some highly active disease modifying therapies (MIST study) and although transplant related mortality risk is low, potential toxicity from such treatments remain a significant barrier. We present data of a single centre experience focussing on early morbidity and complications.

Methods: We retrospectively analysed MS patients who underwent HSCT (n-84) in our centre at Kings College Hospital NHS Foundation Trust, London between 2012-2019 for relapsing remitting, RRMS(n-38); primary progressive, PPMS(n-22) and secondary progressive SPMS(n-24) disease. Patients had Cyclophosphamide (4gm/m2 or 2gm/m2 post 2018)-GCSF primed stem cell harvest followed by Cyclophosphamide (200 mg/Kg)-rabbit ATG (7.5mg/kg) conditioning and stem cell return. We collected detailed data on complications experienced by patients during mobilisation and after aHSCT analysed in pseudo-anonymous format using standard descriptive statistics.

Results: Baseline patient characteristics were compared between RRMS, SPMS and PPMS, with some differences noted in median age, previous treatments, median EDSS (Expanded disability symptom scale), median time from diagnosis to aHSCT between the groups. Median time to neutrophil engraftment was similar across all groups. Median length of hospitalisation was 26 days (range 6-103), with 5 patients staying longer than 28 days and one death pre-HSCT infusion from suspected ATG related cardiac event. One patient died due to conditioning related acute respiratory distress. Overall transplant related mortality was 2.4%. A large proportion of patients (95%) developed febrile episodes following conditioning, 45% had ATG related fevers & rashes. A significant number of patients developed Gram negative sepsis (35%) post aHSCT, although a similar proportion had no source identified despite extensive investigations. More concerningly, 80% of patients experienced significant fluid overload during aHSCT (defined by >5% weight gain or pulmonary oedema/hypoxia requiring diuretics), with 41% of them developing symptomatic mild-severe pulmonary oedema and 9% patients required intensive care unit admission. Interestingly, 31% patients also suffered from transient worsening of neurology symptoms (pseudo-MS), mostly related to infection episodes and EBV viraemia. 13% patients developed hematuria, almost all happened related to 4gm/m2 cyclophosphamide dose as priming and none after dose reduction to 2gm/m2. Patients were subsequently analysed for morbidity events based on the EDSS group (≤5.0 vs >5.0) (Fig. 1b). Patients with higher EDSS (>5.0) had significantly prolonged hospitalization post cell return (median 21d vs 15d, p-0.03). Although there was no significant difference in morbidity events, but all cardiac events and deaths were unfortunately noted in higher EDSS group. Overall efficacy of treatment was 80% in terms of relapse free survival at 3 years.

Conclusions: Our experience describes in detail some of the key early morbidity issues faced by MS patients undergoing aHSCT, namely cardio-pulmonary events, gram negative septicaemia, fluid overload with a potential impact on underlying MS symptoms, which highlights the challenges in appropriate patient selection, close monitoring and specialist input required to safely manage these transplants.

Disclosure: Nothing to declare.

O024. High-Dose Immunosuppressive Therapy with Autologous Hematopoietic Stem Cell Transplantation (AHSCT) in Multiple Sclerosis: Changing Approaches in Single-Center Experiences

Alexey Polushin¹, Natalia Totolyan¹, Yuri Zalyalov¹, Andrei Gavrilenko¹, Aleksandr Tsynchenko¹, Evgenia Kakoulina¹, Iaroslav Skiba¹, Lilia Stelmakh¹, Sergei Bondarenko¹, Aleksandr Volosyuk², Daria Plotnikova², Artem Smolin², Alexander Kulagin¹

¹ First Pavlov State Medical University of St. Petersburg, Saint-Petersburg, Russian Federation, ² St. Petersburg Electrotechnical University, ITMO University, Saint-Petersburg, Russian Federation

Background: AHSCT is used for multiple sclerosis (MS) more than 20 years. In recent years, most transplant centers have used low-intensity HDIT with cyclophosphamide, which has led to a reduction in mortality and other complications. Another trend is a change in the inclusion criteria based on current concept of early neurodegeneration in MS and the lower effectiveness of therapy in patients with a high degree of disability. If earlier the aHSCT was regarded as “salvage-therapy”, now it is proposed to use it at earlier stages of MS, in cases of high activity of the disease with unfavorable prognostic parameters.

Objective: To compare the characteristics of transplanted MS patients and the aHSCT procedure in the periods 2000-2012 and 2018-2020.

Methods: A retrospective-prospective study in 2 groups of MS patients with different aHSCT protocols. The main characteristics of patients were evaluated taking into account anamnestic and clinical data, as well as the intensity of conditioning regimens.

Results: During the first period aHSCT was performed in 25 patients, median age 34 years. BEAM-ATG (n = 15), FM-ATG (n = 9), Cy-ATG (n = 1) were used as conditioning regimens. RMS, SPMS and PPMS was in 4 (16%), 10 (40%) and 11 (44%) patients, respectively. The median EDSS score was 6 (≤4.0 in 3 patients, 4.5-6.0 in 13 and 6.5-8.0 in 9). The median time from symptoms onset to the MS diagnosis and SCT was 1.37 and 6.92 years, respectively. The number of disease-modifying therapy (DMT) lines before SCT averaged 1.5. The median hospital stay was 39 days without HSC harvest step. The median dose of CD34+ cells was 2.8 x 106/kg.

A total of 35 patients (median age 35 years) received aHSCT during the second period. Conditioning regimens were Cy-ATG (n = 11), Cy-R (n = 21) and FluCy-R (n = 3). There were 22 (63%) patients with RMS, 9 (26%) patients with SPMS and 4 (11%) with PPMS. The median EDSS score was 3.5 points (≤4.0 in 18, 4.5-6.0 in 13, 6.5 – in 4). The median time from symptoms onset to the MS diagnosis and SCT was 2.62 and 9.87 years, respectively. The number of DMT lines before SCT was 2.46. The median hospital stay was 39 days including HSC harvest step. The median dose of CD34+ cells was 3.75x106/kg.

Conclusions: Comparison of two periods of aHSCT using in MS demonstrated: 1. Decrease in the EDSS score with lengthening the time before transplantation, which may indicate a more aggressive course of MS in patients transplanted in the first period. 2. More DMTs prior to aHSCT. 3. Identical duration of hospitalization despite the inclusion of HSC harvest step. Our experience indicates compliance with the global trend to use SCT for the patients with less severe neurological deficits, suggesting preventative method in order to preserve the quality of life of patients with MS.

Clinical Trial Registry: -

Disclosure: No conflict of interest.

O025. Impact of Ahsct on Fertility In Multiple Sclerosis Patients

Alice Mariottini¹, Riccardo Boncompagni¹, Ilaria Cutini¹, Maria Di Cristinzi¹, Antonella Gozzini¹, Chiara Innocenti¹, Luca Massacesi¹, Chiara Nozzoli¹, Giulia Rastrelli¹, Anna Maria Repice¹, Linda Vignozzi¹, Riccardo Saccardi¹

¹ Careggi Hospital, Florence, Italy

Background: Autologous haematopoietic stem cell transplantation (AHSCT) is a treatment option for aggressive multiple sclerosis (MS) with acceptable safety profile, but paucity of data is available so far about impact of AHSCT on fertility.

Methods: Data on clinical and endocrine assessment (luteinizing hormone, Follicle-Stimulating Hormone, oestradiol, progesterone, prolactin, testosterone, anti-müllerian hormone, AMH) were collected in patients with MS who underwent AHSCT at our centre. Patients who carried out at least one endocrine assessment after transplant were included; occurrence of spontaneous menstrual cycle at baseline was also recorded.

Results: Twenty-three females (17 RR-MS; 6 SP-MS) and 6 males (3 RR-MS; 3 SP-MS) were included. Median age at transplant for females was 36 years (range 24–50). Twenty-seven patients were conditioned with full dose BEAM/ATG; 2 females received a reduced intensity conditioning regimen due to safety issues (carmustine/etoposide/cytarabine or cyclophosphamide 200 mg/Kg+ATG). Seventeen patients had received chemotherapy-based immunosuppression (cyclophosphamide, azathioprine, mitoxantrone) prior to AHSCT. Nine females had a baseline endocrine assessment and at least one post-transplant assessment. AMH levels at baseline were in line with those expected according to patients’ age in 8/9 cases (89%). After transplant, 19/23 (83%) experienced a persistent amenorrhea. Four females (3 BEAM-ATG, 1 carmustine-etoposide-cytarabine) showed spontaneous menses recovery within 6 months after transplant; these patients were younger compared to those with prolonged amenorrhea: median age at AHSCT was 26 years (24 – 39) vs 38 years (26 – 50), p = 0.027. Out of 9 patients with AMH assessment at baseline, 6 (78%) showed a drop below normal level; a further reduction in 6/7 evaluable cases was observed at 24 months of follow-up. Post-transplant AMH was normal in 2/14 (14%) females for whom only post-treatment assessment was available. No pregnancies were reported following transplant. However, desire of pregnancy was reported only by one patient who was 42 years-old at transplant and showed persistent amenorrhea up to last follow-up. In males Post-AHSCT testosterone levels were within normal range in 5 out 6 available cases. Testosterone levels at follow-up were not reduced compared to baseline in 2/3 (67%) cases who had the baseline assessment.

Conclusions: The rate of spontaneous menstrual recovery in our study was lower than that reported in women treated with the same conditioning regimen for lymphoma, where 68% of women experienced spontaneous recovery of menses; however, no differences were shown after stratification for patients’ age. Our data confirm that AHSCT deeply affects gonadal function in females, with more severe impact among older women. It should be pointed out that most of the patients had received cytotoxic treatments before transplant that could have affected gonadal function at baseline and/or might have enhanced AHSCT-related toxicity. Prospective, multidisciplinary studies on larger cohorts of patients will provide a more definite evaluation of this important issue.

Disclosure: Nothing to declare.

CAR-based Cellular Therapy – Clinical

O026. CD19 Car T Therapy In Children with R/R All: Adaptive Split Dosing Improves Safety and Maintains Efficacy of the Approach

Olga Molostova¹, Larisa Shelikhova¹, Yakov Muzalevsky¹, Alexey Kazachenok¹, Rimma Khismatullina¹, Julia Abugova¹, Elena Kurnikova¹, Pavel Trakhtman¹, Dmitry Balashov¹, Dmitriy Pershin¹, Viktoria Zubachenko¹, Maria Fadeeva¹, Alexander Popov¹, Olga Illarionova¹, Georg Rauser², Regina Alex², Boro Dropulic³, Rimas Orentas³, Dina Shneider³, Natalia Miakova¹, Dmitry Litvinov¹, Galina Novichkova¹, Alexey Maschan¹, Michael Maschan¹

¹ Dmitriy Rogachev National Medical Center of Pediatric Hematology, Oncology and Immunology, Moscow, Russian Federation, ² Miltenyi Biotec GmbH, Bergisch Gladbach, Germany, ³ Lentigen, A Miltenyi Biotec Company, Gaithersburg, United States

Background: CD19 CAR-T cells manufactured at the point-of-care based on the automatic bioreactor platform were tested in a prospective academic trial. Despite prophylactic tocilizumab, severe CRS and ICANS developed in a minority of patients. In the amended part of the trial, protocol of CAR-T application was modified to adapt the starting CAR-T dose to the leukemia burden and implement a split-dose strategy in high-risk patients.

Methods: A cohort of 30 children with BCP-ALL received CAR-T therapy in dose-escalation regimen (cohort 1) and 24 pts in the risk-adapted regimen (cohort 2). Final non-cryopreserved product was administered to all patients after lymphodepletion and prophylactic tocilizumab.

In cohort 1, four CAR-T dose levels were used: 0.1, 0.5, 1, and 3*106/kg. Bone marrow leukemia burden was high (>20%) in 19 patients. In cohort 2 a flat dose of 1*106/kg was used for 10 pts with low leukemia burden (<20%). Among 14 patients with high leukemia burden 0.1*106/kg was infused on day 0, a second infusion (0.9*106/kg) was scheduled at day 7. Criteria for second CAR-T infusion included the absence of CRS or ICANS > grade 2 and reduction of leukemia burden to <20%. The CliniMACS Prodigy T-cell transduction process with lentiviral second generation CD19.4-1BB zeta vector (Lentigen, Miltenyi Biotec) was used.

Results: Interim analysis in cohort 1 showed that severe (grade 3-5) CRS and neurotoxicity were associated exclusively with large leukemia burden (>20% blasts in the bone marrow) at the enrollment (p = 0,002). In cohort 2 there were no cases of grade IV-V toxicity in patients with high leukemia burden, four patients received the second CAR-T infusion, thereafter 1 pt developed CRS grade III and ICANS grade I. One pt with low leukemia burden developed ICANS grade IV with brain edema. In cohort 1, twenty-seven patients were evaluable for response at day 28, and 24 (89%) of them had MRD-negative remission. Relapse after initial response was registered in 18 (75%) cases, median time to relapse was 184 days. With the risk-adapted strategy in the group with low leukemia burden 8 patients (80%) achieved CR at day 28, and 4 pts (50%) relapsed (median time to relapse was 127 days). In the group with high leukemia burden all patients (100%) achieved complete remission on day 28, 7 relapses (50%) occurred with a median of 93 days.

Conclusions: All patients with high leukemia burden, who received only one CAR-T dose at 0.1*106/kg, achieved MRD-negative CR, confirming high potency of the fresh CD19 CAR-T product. Risk-adapted CAR-T dosing strategy allowed to maintain a high rate of remission and to reduce the risk of severe toxicity among patients with high initial disease burden.

Clinical Trial Registry: CT: NCT03467256

Disclosure: Nothing to disclare.

O027. Contribution of Age and The HCT-CI to Predict CRS And Icans Severity In NHL Patients Receiving CD19 Car T-cell Therapy

Jordan Gauthier¹, Aisling Cearley¹, Paula Perkins¹, Angela Kirk¹, Mazyar Shadman¹, David Maloney¹, Cameron Turtle¹, Mohamed Sorror¹

¹ Fred Hutchinson Cancer Research Center, Seattle, United States

Background: CD19-targeted chimeric antigen receptor-engineered (CD19 CAR) T-cell therapy lead to high response rates in patients with relapsed or refractory (R/R) B-cell malignancies, yet significant toxicities impact its use and broad dissemination. Severe cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) have been reported in a subset of patients after CD19 CAR T-cell therapy. To date, predictive models of CRS and ICANS severity are lacking, and the impact of age and comorbidities on these toxicities remains poorly characterized. Thus, we assessed the role of age and the hematopoietic cell transplantation comorbidity index (HCT-CI) to predict CRS and ICANS severity in B-cell non-Hodgkin lymphoma (B-NHL) patients undergoing CD19 CAR-T cell therapy.

Methods: We analyzed 93 patients treated at our institution with relapsed or refractory B-NHL. CRS and ICANS were graded using the 2019 ASTCT and CTCAE4.03, respectively. Patients received cyclophosphamide and fludarabine lymphodepletion followed by the infusion of axicabtagene ciloleucel (axi-cel; up to 2 x 10⁸ CAR T cells) or defined-composition CAR T cells (JCAR014; 2 x 10⁶/kg) on a phase I/II clinical trial (NCT01865617). We applied multivariable proportional odds logistic regression to model peak CRS and ICANS grade as a function of predictors evaluated at baseline prior to lymphodepletion. The base models were specified after testing interaction effects using an analysis of variance (ANOVA) table of a model including the following predictors: age, HCT-CI, CAR T-cell product type, prelymphodepletion LDH, platelet count, absolute lymphocyte count, and albumin. The predictive value of age and HCT-CI was assessed using the likelihood ratio (LR) test, discrimination (C-index), and calibration (smooth calibration curves).

Results: Fifty four patients (58%) received axi-cel, while 39 (42%) received JCAR014. The ANOVA table using all predictors identified the CAR T-cell product (p = 0.001) and the prelymphodepletion LDH (p = 0.005) as the most important predictors of CRS severity. Age and HCT-CI did not add predictive information to a base model including CAR T-cell product and prelymphodepletion LDH (LR test, p = 0.72), and did not improve discrimination (C-index = 0.71 and 0.72 without and with age + HCT-CI) or calibration.The ANOVA table showed the CAR T-cell product (p = 0.09), age (p = 0.07), and the prelymphodepletion LDH (p = 0.06) were the most important predictors of ICANS severity.Age improved a model including CAR T-cell product and prelymphodepletion LDH (LR Chi2, 2.89; p = 0.09). Models with or without age had comparable discrimination (C-index=0.65 and 0.66, respectively), but the addition of age improved model calibration. The HCT-CI (LR Chi2, 1.47, p = 0.69) did not add predictive value when added to the base model including age with similar discrimination (C-index = 0.66 with or without HCT-CI). In a model including age, CAR T-cell product and prelymphodepletion LDH, the odds ratio for age was 1.03 per year increase (95%CI, 0.99-1.06).

Conclusions: While older age was associated with higher risk of severe ICANS after CD19 CAR T-cell therapy, the HCT-CI did not improve CRS or ICANS risk prediction. We plan to validate our findings and refine our models in a larger cohort including other commercial CAR T-cell products.

Clinical Trial Registry: NCT01865617

Disclosure: D.G.M. has received research funding from Kite Pharma, Juno Therapeutics, a Celgene company, and Celgene; has received honoraria for participation in advisory boards meetings with Kite Pharma, Gilead, Genentech, Novartis and Eureka. C.J.T. receives research funding from Juno Therapeutics, Nektar Therapeutics, Minerva, TCR2 and AstraZeneca; is a member of scientific advisory boards and has options in Precision Biosciences, Eureka Therapeutics, Caribou Biosciences, Myeloid Therapeutics, and ArsenalBio; serves on scientific advisory boards for T-CURX and Century Therapeutics; has served on advisory boards for Nektar Therapeutics, Allogene, Kite/Gilead, Novartis, Humanigen, PACT Pharma, Amgen, and Astra Zeneca; and has patents licensed to Juno Therapeutics. The remaining authors declare no competing financial interests. M.L.S was advisory committee member and received honorarium from JAZZ pharmaceuticals in 2019. MS: Consulting, Advisory Boards, steering committees or data safety monitoring committees: Abbvie, Genentech, AstraZeneca, Sound Biologics, Pharmacyclics, Verastem, ADC Therapeutics, Beigene, Cellectar, Bristol Myers Squibb, Morphosys, TG Therapeutics, Innate Pharma, and Atara Biotherapeutics; Research Funding: Mustang Bio, Celgene, Bristol Myers Squibb, Pharmacyclics, Gilead, Genentech, Abbvie, TG Therapeutics, Beigene, AstraZeneca, Sunesis, Beigene.

O028. Real World of Experience Axicabtagene Ciloleucel for The Treatment of Relapsed or Refractory Large B-Cell Lymphoma in Spain

Mi Kwon¹, Rebeca Bailen¹, Lucia Lopez Corral², Juan Luis Reguera³, Gloria Iacoboni⁴, Rafael Hernani Morales⁵, Valentin Ortiz Maldonado⁶, Mariana Bastos Oreiro¹, Alejandro Martin Garcia², Nuria Martinez Cibrian³, Pere Barba⁴, Julio Delgado⁶, Maria Jose Terol⁵, Jose Luis Diez Martin¹

¹ Hospital G. Univ. Gregorio Marañon, Madrid, Spain, ² Hospital Univ. Salamanca, Salamanca, Spain, ³ Hospital Univ. Virgen del Rocio, Sevilla, Spain, ⁴ Hospital Univ. Vall d´Hebron, Barcelona, Spain, ⁵ Hospital Clinico Universitario, Valencia, Spain, ⁶ Hospital Clinic de Barcelona, Barcelona, Spain

Background: Axicabtagen Ciloleucel (axi-cel) is approved in Europe for the treatment of adults with R/R large B-cell lymphoma (LBCL), primary mediastinal B-cell lymphoma (PMBCL) and transformed follicular lymphoma (tFL). In Spain, nationwide CAR-T administration requests are reviewed centrally within the Ministry of Health. We analyzed the real-world outcomes of patients treated with axi-cel under the commercial label in Spanish centers.

Methods: Six designated centers for commercial CAR-T administration collected data on behalf of GETH-GELTAMO. Data were collected retrospectively from consecutive patients in whom apheresis was performed for axi-cel treatment from Febrruary-2019 to November-2020. CRS and ICANS were graded with the ASTCT consensus criteria. Response was assessed according to the Lugano criteria.

Results: 106 patients with R/R lymphoma underwent apheresis for axi-cel. At data cutoff, 92 (87%) received infusion. The reasons for not undergoing infusion were progression-related death in 12 (86%), tumor lysis syndrome in 1 (7%) and complete response after bridging therapy in 1 (7%). Of note, 14 patients were conditioned and infused during the peak of COVID-19 epidemic in Spain (March-April 2020).

Median time from Ministry approval to infusion was 54 days. Histology consisted of 74% DLBCL with 11% tFL, and 15% PMBCL. Disease status at lymphodepletion was PD in 69%, SD in 22% and PR in 9%. All patients received lymphodepletion. Median time from leukapheresis to start of lymphodepletion was 34 days. Median time from leukapheresis to infusion was 39 days. Median hospitalization period was 21 days. Any grade of CRS occurred in 86% of pts (18% grade 2, 6.5% grades 3-4). Tocilizumab was used in 58% of patients who developed CRS, corticosteroids in 19%. ICANS was diagnosed in 42.5% of pts (10% grade 2, 15% grade 3-4). Treatment for ICANS included corticosteroids in 78%, tocilizumab in 31%, siltuximab in 15%, and anakinra in 21%. ICU admission was needed in 20 patients (22%). 4 patients died in the context of ICANS, 1 due to CRS, and 1 due to infection. Of 80 patients evaluable and restaged at day 30, ORR was 78% with 40% CR, 38% PR, 11% PD and 11% SD or indeterminate. Of 58 patients evaluable at day 100, 66% had ongoing response (CR 48%, PR 18%). Of 23 patients evaluable at day 180, 65% presented CR. Of 39 patients who showed PR/SD at day 30, 9 (23%) converted to CR. After a median follow-up of 6.3 months, EFS and OS were 55.5% and 78%, respectively in the infused population, with an estimated median EFS and OS of 13.1 and 7.3 months. In the intention-to-treat analysis for all patients who underwent apheresis, median estimated OS and EFS were 12.3 (95%CI 8.9-15.7) and 6.6 months (95%CI 4.6-8.5), respectively (Figure 1).

Conclusions: This Spanish multicenter retrospective analysis shows encouraging results of axi-cell treatment in patients with R/R aggressive B-cell lymphoma in the real-world setting. Significant toxicity events were less frequent than those reported in the pivotal trial, however events of mortality associated to toxicity occurred. With a limited follow-up time, response outcomes are favorable.

Clinical Trial Registry:

Disclosure: Mi Kwon, advisory honoraria Novartis, BMS, Gilead. Juan Luis Reguera, advisory honoraria Novartis, Celgene, Gilead. Alejandro Martin, advisory honoraria Roche, Celgene, Janssen, Servier, Gilead. Gloria Iacoboni, advisory honoraria Celgene, Gilead, Novartis, Roche. Mariana Bastos, advisory honoraria Roche, Celgene, Takeda, Gilead.

O029. Humoral Response to Pneumococcal Antigens Declines After CD19-targeted Car T Cell Therapy

Dasom Lee¹, Michael Jain², Julio Chavez², Marco Davila², Farhad Khimani², Aleksandr Lazaryan², Javier Pinilla², Bijal Shah², Frederick Locke²

¹ University of South Florida, Tampa, United States, ² H. Lee Moffitt Cancer Center & Research Institute, Tampa, United States

Background: We evaluated Large B cell Lymphoma (LBCL) patients treated with CD19-targeted Chimeric Antigen Receptor T cell therapy (CAR-T) to determine pre and post serum pneumococcal antibody levels and the impact of post CAR-T vaccination on these levels.

Methods: We retrospectively identified LBCL patients that had serum pneumococcal IgG titers within 30 days prior to CAR-T (baseline), day+90 (range: 75-110) or day+180 (range: 170-200) after CAR-T. We next identified patients that received pneumococcal conjugate vaccine (PCV13) on day+90. Serotypes tested include those specific (1, 4, 5, 7F, 9V, 18C) and non-specific (8, 9N, 12F) to the vaccine. Pneumococcal IgG titers, stratified by serotype and/or vaccination status, were compared with paired nonparametric t-test. An established IgG concentration ≥1.3μg/mL was considered protective. Patients with ≥4/6 tested vaccine specific serotypes meeting this threshold were deemed successfully vaccinated.

Results: We identified 66 patients with pneumococcal IgG titers drawn on at least one of the following timepoints: baseline, day+90 or day+180. Protective immunity against pneumococcus was observed in 17% (5/29), 21% (7/33), and 17% (5/29) of patients at baseline, day+90, and day+180 respectively. 10 patients had paired data at baseline and day+90, and 12 patients had paired data on day+90 and day+180. 1 patient had IgG titers at all three timepoints.

In the 10 patients with paired data at baseline and day+90, vaccine specific serotype IgG titers significantly decreased from baseline to day+90 (p = 0.002) (Figure 1). Non-vaccine specific serotype titers decreased but did not meet statistical significance (p = 0.08). When pooling all serotypes together, there was a significant decrease in IgG titers from baseline to day+90 (p = 0.002). Only 1 out of these 10 patients met criteria of protective immunity at baseline, which was lost on day+90.

12 patients had paired samples on day+90 and day+180, and 5 of those patients received PCV13 on day+90 after titers were drawn. The unvaccinated patients (n = 7) had a significant decrease in IgG titers of all vaccine specific serotypes (p = 0.04) (Figure 2). Non-vaccine specific serotypes titers decreased but did not meet statistical significance (p = 0.08). There was a significant decrease in IgG titers of all serotypes from day+90 to day+180 (p = 0.006). 1 out of the 7 unvaccinated patients met the criteria of protective immunity on day+90, which persisted on day+180.

In the 5 patients vaccinated on day+90, both vaccine specific and non-vaccine specific serotypes titers decreased significantly on day+180 (p = 0.016 and p = 0.031, respectively) (Figure 3). None of the vaccine specific titers converted from non-protective to protective immunity. 2 out of the 5 patients met the criteria of protective immunity on day+90, which persisted on day+180. However, no titers increased more than 2-fold on day+180 after vaccination suggesting vaccination was ineffective.

Conclusions: Our results demonstrate few LBCL patients have protective immunity against pneumococcus at baseline before CAR-T, and existing titers may decrease after CAR-T. Data in several patients who received PCV13 on day+90 suggests humoral immunity cannot be restored by vaccination at this timepoint. Clinical trials are needed to determine the optimal timing of vaccination, before or after CAR-T, to develop protective immunity against pneumococcus.

Disclosure: Nothing to declare.

O030. Decreased Neuroprotective Analytes on Day of Infusion of Chimeric Antigen Receptor (Car) T Cells are Associated with Neurotoxicity

Madhavi Lakkaraja^1,2, Kinga Hosszu¹, Devin McAvoy¹, Audrey Mauguen¹, Terence J Purdon¹, Yasmin Khakoo^,2, Bianca D. Santomasso¹, Isabelle Riviere¹, Michele Sadelain¹, Kevin J Curran¹, Jae H Park¹, Renier J Brentjens¹, Jaap Jan Boelens^1,2

¹ Memorial Sloan Kettering Cancer Center, New York, United States, ² Weill Cornell Medical College, New York, United States

Background: Neurotoxicity is a severe and potentially fatal complication associated with Chimeric Antigen Receptor (CAR-T cell) therapy - a promising curative treatment option for hematological malignancies. In children and adults with leukemia/lymphoma, elevated cytokines such as IL2, sIL-2R Aα, IL-6, IL-8, IL-10, IL13, IL-15, IFN-γ, MCP-1 and others were identified in plasma on day 3 after infusion of CAR-T cells. While in CSF, elevated protein and cytokines -IL1a, IL6, IL10, GCSF, TNF-α, IFN-γ have been noted. Identifying early predictors/biomarkers, could help identifying patients at risk of developing neurotoxicity. In this pilot study, we aim to detect analytes or signatures that could potentially be biomarkers to predict neurotoxicity in patients receiving CAR-T cells. To achieve this, we used the Olink proteomic platform.

Methods: 15 patients (6 adults and 9 children), who were treated with CD19 directed CAR-T cells (MSKCC and commercial), and had serum samples cryopreserved and banked at various time points (baseline to 2 weeks after infusion) were included in the study. We analyzed plasma by targeted proteomics with proximity extension assays (PEAs; Olink) for 92 markers focused on neurological diseases and biological processes such as axon development, neurogenesis and synapse assembly. To investigate correlation in analytes values, heatmaps of the normalized biomarker values were drawn based on euclidean distance and complete linkage at baseline. Baseline biomarker values were compared between patients with and without neurotoxicity using Mann-Whitney-Wilcoxon tests. The analysis was repeated using the slope of biomarker change during the first week or up to the time to neurotoxicity. A p value of <0.05 was considered to be significant.

Results: 15 patients received CAR T cells for B-cell Acute Lymphoblastic Leukemia(ALL) and were included in the study. Median age was 19.9 (Range:2.8 to 46.3) years, 8(53%) patients were female. Conditioning regimen was cyclophosphamide and fludarabine. 8/15(53%) patients had prior hematopoietic cell transplantation. Median fludarabine exposure was 20.4 (Range:10.5–27.5)mg*h/L. 10 patients(67%) had neurotoxicity (7:grade 1, 1:grade 3, and 2:grade 4). Median time to neurotoxicity was 7.5 (Range:5 to 18) days. At baseline,13/92 analytes were significantly decreased in patients who developed neurotoxicity compared to patients who did not develop neurotoxicity - ECE1, CDH15, KLB, GPNMB, KIR2DL3, IMPA1, CLSTN1, LEPR, FGFR2, NPM1, IL32, IL3RA, and GGT5. All these 13 markers identified at baseline have been suggested to be neuroprotective. In addition, 5 analytes increased from baseline before development neurotoxicity: BST2, NPM1, ASGR1, PHOSPHO1 and AOC1 - Four analytes have been associated with cellular stress and neuroinflammation in neurological diseases and one (NMP1) is neuroprotective.

Conclusions: Using Olink proteomics platform we identified 13 analytes for which lower value at baseline was associated with development of neurotoxicity after CAR-T cells. These analytes are known to be neuroprotective. The 4 analytes,which increased from baseline in patients who developed neurotoxicity,have been associated with cellular stress and one is neuroprotective. These findings suggest that patients at risk for neurotoxicity may be identified at infusion of CAR-T cells. If these results are confirmed in larger dataset,these markers can be used to take preventive measures in patients at risk.

Disclosure: Madhavi Lakkaraja, MD, MPH, Kinga Hosszu, PhD, Devin McAvoy, Terence J. Purdon,Audrey Mauguen,PhD,Yasmin Khakoo,MD: No disclosures.

Bianca D. Santomasso, MD, PhD: Acted as a consultant for Kite/Gilead, Juno/Celgene, and Novartis.

Isabelle Riviere, PhD: Fate Therapeutics Inc.: Consultancy, Other: Ownership interest, Research Funding; FloDesign Sonics: Consultancy, Other: Ownership interest; Juno Therapeutics: Other: Ownership interest, Research Funding; Takeda: Research Funding; Atara: Research Funding.

Michele Sadelain, MD,PhD: Atara: Patents & Royalties, Research Funding; Fate Therapeutics: Patents & Royalties, Research Funding; Minerva: Other: Biotechnologies, Patents & Royalties; Mnemo: Patents & Royalties; Takeda: Patents & Royalties, Research Funding.

Kevin J. Curran, MD: Novartis: Consultancy, Research Funding; Mesoblast: Consultancy; Celgene: Research Funding.

Jae H. Park, MD: Consulting : Amgen, Novartis, Kite Pharma, SERVIER, Autolus, Takeda, AstraZeneca, InnatePharma, Scientific advisory board :Artiva.

Renier J. Brentjens, MD, PhD: Bristol Myers Squibb : Research Funding, royalties, consultant and speaker; Gracell Biotechnologies, Inc.: consultant/advisor.

Jaap Jan Boelens, MD, PhD: Consulting Takeda, Bluebird Bio, Omeros, Advanced clinical, Magenta, Bluerock.

O031. Comparison of Two Kinds of Anti-Thymocyte Globulin on Outcomes of Haploidentical Stem Cell Transplantation For Refractory/Relapsed and High-Risk B-Cell Acute Lymphoblastic Leukemia After CAR-T Therapy

Zhao-Yanli Zhao^1,2, Jian-Ping Zhang^1,2, De-Yan Liu^1,2, Zhi-Jie Wei^1,2, Min Xiong^1,2, Rui-Juan Sun^1,2, Jia-Rui Zhou^1,2, Yue Lu^1,2, Xing-Yu Cao^1,2, Xian Zhang^1,2, Jun-Fang Yang^1,2, Pei-Hua Lu^1,2, Dao-Pei Lu^1,2

¹ Lu Daopei Hospital, Langfang, China, ² Lu Daopei Institute of Hematology, Beijing, China

Background: Anti-thymocyte globulin (ATG) plays an important role in preventing graft-versus-host disease (GVHD) after unmanipulated haploidentical stem cell transplantation (haplo-SCT). However, the optimal type and dose of ATG remain undetermined, especially in the relatively distinctive population of refractory/relapsed (R/R) and high-risk B-cell acute lymphoblastic leukemia (B-ALL) after CAR-T therapy.

Methods: The clinical outcomes of 87 R/R and high-risk B-ALL patients achieved complete remission (CR) after anti-CD19 CAR-T cell therapy subsequently bridged to unmanipulated haplo-SCT between July 2015 and March 2018 were retrospectively analyzed. Conditioning regimens were administered with total body irradiation (TBI, 2Gyx5~6) plus either anti-T lymphocyte globulin (ATLG, Grafalon®) 20 mg/kg (n = 63) or rabbit anti-thymocyte globulin (rATG, Thymoglobuline®) 5~7.5 mg/kg (n = 24). Cyclosporine, short-term methotrexate, and mycophenolate mofetil were used for GVHD prophylaxis. As of August 2020, the median survival time for this cohort was 39 months (29 - 58 months).

Results: The patients in the ATLG group were relatively older with median age of 15 (2-49 years old) versus 8 (4-41 years old, p = 0.004) in the rATG group. In the ATLG group, total mononuclear cell count (MNC, median 8.7 [5.0-15.5] x10⁸/kg versus 12.0 [6.8-17.5] x10⁸/kg, p = 0.005) and infused CD34 cell count (median 4.46 [1.4-10.2] x10⁶/kg versus 5.95 [2.8-8.8] x10⁶/kg, p = 0.000) were relatively lower than those in the rATG group. In this cohort, the incidence of acute GVHD were comparable among the two ATG groups for Grade II-IV (29.0% [95% CI 19.0-39.0%] in the ATLG group vs. 37.5% [95% CI 18.1-56.9%] in the rATG group, p = 0.598) and Grade III-IV (6.5% [95% CI 0.4-12.6%] in the ATLG group vs. 16.7% [95% CI 1.8-31.6%] in the rATG group, p = 0.164). The occurrence of EBV-DNAemia (15.9% vs. 29.2%, p = 0.162), CMV-DNAemia (47.6% vs. 58.3%, p = 0.372), and hemorrhagic cystitis (27.0% vs. 25.0%, p = 0.851) were similar between the ATLG group and rATG group. Chronic GVHD in both ATG groups were comparable (74.1% [95% CI 61.2-87.0%] and 69.8% [95% CI 49.2-90.4%] for the ATLG and rATG groups respectively (p = 0.526). The rate of extensive chronic GVHD in the ATLG group and rATG group were 12.7% [95% CI 3.9-21.5%] and 15.4% [95% CI -0.9-31.7%] (p = 0.959), respectively. The incidence of 3-year relapse rate were 13.8% [95% CI 4.8-22.8%] in the ATLG group versus 4.5% [95% CI -4.1-13.1%] in the rATG group (p = 0.248). The 3-year non-relapse mortality (NRM) was 20.5% [95% CI 10.1-30.9%] in the ATLG group versus 16.7% [95% CI 1.8-31.6%] in the rATG group (p = 0.746). In the ATLG group, 3-year leukemia-free survival (LFS) was 68.3% [95% CI 56.7-79.9%] versus 79.2% [95% CI 62.9-95.5%]in the rATG group (p = 0.349). The 3-year overall survival (OS) in the ATLG group was 69.8% [95% CI 58.4-81.2%] versus 83.3% [95% CI 68.4-98.2%] in the rATG group (p = 0.234).

Conclusions: Both types of ATG (ATLG versus rATG) led to comparable clinical outcomes with respect to GVHD prophylaxis and viral infection in this cohort of relatively homogenous patient population treated with CD19 CAR-T followed by bridging to haplo-SCT with TBI. Larger, randomized clinical studies are currently underway to further confirm these results.

Clinical Trial Registry: ChiCTR-oic-16009259, ChiCTR-OON-16009143, ChiCTR-IIh-16008711, ChiCTR-ONC-17012829 (http://www.chictr.org.cn)

Disclosure: Nothing to declare.

O032. Impact of SARS-COV-2 on Delivery of Car T Cell Therapy in Europe: A Survey from The Cellular Therapy and Immunobiology Working Party of The EBMT

Sara Ghorashian¹, Florent Malard², Meltem Kurt Yüksel³, Jorinde Hoogenboom⁴, Katya Mauff⁵, Alvaro Urbano-Ispizua⁶, Jürgen Kuball⁷, Rafael de la Camara⁸, Stephen Mielke⁹, Annalisa Ruggeri¹⁰, Christian Chabannon¹¹

¹ UCL, London, United Kingdom, ² Saint Antoine Hospital, AP-HP, Sorbonne Université, Paris, France, ³ Ankara University Faculty of Medicine, Ankara, Turkey, ⁴ EBMT Leiden Study Unit, Leiden, Netherlands, ⁵ EBMT Statistical Unit, Leiden, Netherlands, ⁶ Hospital Clinic Barcelona, Barcelona, Spain, ⁷ University Medical Center, Utrecht University, Utrecht, Netherlands, ⁸ Hospital de la Princesa, Madrid, Spain, ⁹ Karolinska University Hospital and Karolinska Institutet, Stockholm, Sweden, ¹⁰ IRCCS Ospedale Pediatrico Bambino Gesù, Rome, Italy, ¹¹ Institut Paoli-Calmettes, Marseille, France

Background: The SARS-COV-2/COVID-19 pandemic had a great impact on cancer therapy. Given the complexity of CAR T cell therapy (products manufactured on a bespoke basis; treatment centres also providing procurement; transportation to and from manufacturing sites; requirement for in-patient and/or ICU beds), we sought to define the impact of the pandemic on delivery of CAR T cell therapies in Europe.

Methods: We carried out a snapshot survey of EBMT centres to define the impact of the pandemic and associated socio-economic restrictions on CAR T cell activity during the interval Q1 to Q4 2020.

Results: 49 out of 118 centres in 12 countries (42%) responded to the survey. 49% of responding centres deliver licensed CAR T cell therapy for B-ALL and B-NHL with a further 25% delivering these licensed therapies and academic CAR T studies. The median number of CAR T cell infusions per centre in 2019 was 8 (range 0-68), with the majority (59%) carrying out 0-10.

At survey completion, 25/49 (51%) of centres were subject to partial lockdown, 4% a complete lockdown and 45% were without restriction. 25/49 (51%) of centres reported reduced patient capacity, most commonly due to patient factors (14/25 (56%), e.g. too unwilling or unable to travel), reduction in in-patient beds (13/25 (52%)), reduced ICU access (12/25 (48%)), reduced clinical trial activity (12/25 (48%)), proven or unproven COVID-19 infection (12/25 (48%)) and reduced out-patient provision (11/25 (44%) NB more than one reason was allowed. More rare causes included delayed manufacture (6 (24%)) or reduced tocilizumab access (3 (12%)). 14/49 (29%) of centres reported a delay for at least one patient up to survey completion. 8/14 centres (57%) had 1-2 patients delayed; 5/14 centres (36%) had 3-5 and 1 centre (7%) had 7 patients delayed.

Of 31 delayed patients, the median age was 62 years, (range 8-75), 19/31 (61%) had DLBCL, 8/31 (26%) had B-ALL and 4/31 (13%) had another indication. 13/31 (42%) were considered for axicabtagene ciloleucel, 8/31 (26%) were considered for tisagenlecleucel and 10 (32%), another product.

Most commonly cited reasons for delay included reduced ICU access, reduction in rostered haematology medical team, reduction in in-patient beds (22, 7 and 5 reports respectively). 27/31 (87%) of patients were delayed in bridging. Patients were delayed up to 1 month (18/31, 58%) 1-3 months (8/31, 26%) or 2/31 (6%) for 4-12 months. 3/31 (10%) of patients were delayed indefinitely. 25/31 (81%) were subsequently infused, 5/31 (16%) were cancelled. In 15 cases, delay was associated with additional therapy prior to CAR T and in 12 cases, with disease progression.

Conclusions: COVID-19 had a significant impact on CAR T therapy across Europe. 51% of centres reported decreased patient capacity. Just under a third reported a delay to delivery of CAR T therapy. The majority of centres with delayed patients reported 1-2 patients affected, mainly in the bridging period, most patients (58%) proceeded to infusion by one month. Delay to CAR T cell therapy necessitated additional therapy and resulted in disease progression in 15 and 12 instances respectively.

Disclosure: Sara Ghorashian: Patents/royalties UCLB, Honoraria Novartis, Amgen.

Florent Malard: Astellas: Honoraria; JAZZ pharmaceuticals: Honoraria; Sanofi: Honoraria; Janssen: Honoraria; Keocyt: Honoraria; Theralos/Mallinckrodt: Honoraria; Biocodex: Honoraria.

Jurgen Kuball: Shareholder and cofounder of Gadeta. Inventor on multiple patents dealing with gdT receptors and CAR T. Research support Novartis, Miltenyi Biotech and Gadeta.

Christian Chabannon: Kite/Gilead, Novartis, Celgene/BMS, Janssen, Bellicum, Terumo BCT: honoraria, speakers bureau, advisory board.

O033. Building of The G0-CAR-T Initiative to Promote Use of Data Collected from Patients Treated With CAR-T Cells AT EU OR EBMT Affiliated Programs

Christian Chabannon¹, Sofie Terwel², Jorinde D Hoogenboom², Gregorz Basak³, Selim Corbacioglu⁴, Rafeal de la Camara⁵, Harry Dolstra⁶, Bertram Glass⁷, Raffaella Greco⁸, Arjan Lankester⁹, Mohamad Mohty¹⁰, Régis Peffault de la Tour¹¹, John Snowden¹², Ibrahim Yakoub-Agha¹³, Francisco Cerisoli¹⁴, Anja van Biezen², Marianne Mol², Katya AL Mauff², Debra Gordon², Breggie Verhoeven¹⁵, Isabel Sanchez-Ortega², Chiara Bonini⁸, John Gribben¹⁶, Nicolaus Kröger¹⁷, Jurgen Kuball¹⁸

¹ Institut Paoli-Calmettes, Marseille cedex 9, France, ² EBMT, Leiden, Netherlands, ³ Medical University of Warsaw, Warsaw, Poland, ⁴ UniversitätsKlinikum Regensburg, Regensburg, Germany, ⁵ Hospital de la Princesa, Madrid, Spain, ⁶ Radboud University Medical Center, Nijmegen, Netherlands, ⁷ HELIOS Klinikum Berlin-Buch GmbH, Berlin, Germany, ⁸ Università Vita-Salute San Raffaele, Milan, Italy, ⁹ Leiden University, Leiden, Netherlands, ¹⁰ Sorbonne Université, Paris, France, ¹¹ Hôpital Saint-Louis - AP-HP, Paris, France, ¹² Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, United Kingdom, ¹³ Centre Hospitalier Universitaire de Lille, Lille, France, ¹⁴ EHA, Den Haag, Netherlands, ¹⁵ Hematon, Utrecht, Netherlands, ¹⁶ Cancer Research UK Barts Centre, London, United Kingdom, ¹⁷ University of Hamburg, Hamburg, Germany, ¹⁸ Utrecht University Medical Center, Utrecht, Netherlands

Background: Because Hematopoietic Cell Transplants (HCTs) carry risks of specific complications, require a peculiar hospital organization that entails tight collaborations between various specialties, and were never granted a marketing authorization, EBMT pioneered long-term follow-up in real-world conditions of these procedures. This lead to the establishment of a registry that today collects information from more than 600 transplant programs established in Europe and beyond, and contains data on more than 700.000 transplants. Innovative hematopoietic cellular therapies that are regulated as medicinal products (MP) are now entering the market at an accelerated pace, raising new hopes for patients and families. These include Immune Effector Cells (IEC), the most publicized being Chimeric Antigen Receptor (CAR)-T cells. These gene therapy MPs induce responses in a significant proportion of patients affected with relapsed/refractory lymphoid malignancies at advanced stages, but also carry significant risks to the patients. They are living drugs with effects that can persist for extended periods of time. They are manufactured through viral engineering of autologous T cells isolated from blood mononuclear cells, and thus qualify as genetically modified organisms. The manufacturing process takes advantage of infrastructures and expertise that are similarly used for HCTs, including cell collection and cell processing facilities. CAR-T cells are among the most expensive MP ever marketed. In view of these features, the EBMT registry appears ideally suited to contribute to the long-term follow-up of patients treated with CAR-T cells, as mandated by health authorities.

Methods: A “Cellular Therapy Form” (CTF) was created, taking into account requirements expressed by the European Medicines Agency (EMA) as part of the Registry initiative; EMA released a positive qualification opinion of the EBMT CTF in February 2019. Soon thereafter, the EBMT started to collect information on CAR-T cell treated patients and publish monthly activity reports from July 2019 onwards. Harmonization with the CIBMTR data collection toolbox is continuously being pursued.

Results: As of Dec 1st and 27 months after EMA approved the first two CAR-T cells, 1,235 treated patients have been registered, 84% of them received a commercial MP, while the remainder received investigational CAR-T cells. Activity is growing in all EU countries. Registration and follow-up are collected from treating centers on a voluntary basis, similar to what has been done with the Med-A form for HCTs; this information has allowed to produce the first reports to the two Marketing Authorization Holders (MAH) that currently market CAR-T cells in Europe. The Post-Authorization Safety Studies (PASS) now launched with these MAHs will create conditions for improved reporting and quality checks. At the time of writing, approximately 10% of registered patients have a follow-up beyond one year; this asset will expand quantitatively and qualitatively in the future, as more patients reach this milestone, and additional follow-up is captured.

Conclusions: The Go-CAR-T initiative aims to bring together different categories of stakeholders in order to define rules for revision of the CTF, access to data, design studies and ultimately take full advantage of this repository to measure the medical value of this new class of MPs.

Disclosure: CC: Kite/Gilead, Novartis, Celgene / BMS, Janssen, Bellicum, Terumo BCT: honoraria, speakers bureau, advisory board.

JK: Shareholder and cofounder of Gadeta. Inventor on multiple patents dealing with gdT receptors and CAR T. Research support Novartis, Miltenyi Biotech and Gadeta.

CAR-based Cellular Therapy – Preclinical

O034. Strategies To Increase CAR T-Cell Safety and to Prevent Epitope Masking In CAR+ B-Cell Leukemia Blasts

Concetta Quintarelli¹, Marika Guercio¹, Simona Manni¹, Iolanda Boffa¹, Matilde Sinibaldi¹, Stefano Di Cecca¹, Simona Caruso¹, Zeinab Abbaszadeh¹, Antonio Camera¹, Biancamaria Cembrola¹, Roselia Ciccone¹, Alberto Orfao², Lourdes Martin-Martin², Sara Gutierrez-Herrero², Maria Herrero-Garcia², Giovanni Cazzaniga³, Vittorio Nunes³, Simona Songia³, Paolo Marcatili⁴, Marco Ruella⁵, Valentina Bertaina¹, Luciana Vinti¹, Francesca Del Bufalo¹, Mattia Algeri¹, Pietro Merli¹, Biagio De Angelis¹, Franco Locatelli¹

¹ Ospedale Pediatrico Bambino Gesù, Roma, Italy, ² Institute for Biomedical Research of Salamanca (IBSAL), Salamanca, Spain, ³ Centro Ricerca Tettamanti, Fondazione Tettamanti, Università degli Studi di MilanoBicocca, Monza, Italy, ⁴ Technical University of Denmark, Kgs., Lyngby, Denmark, ⁵ Cancer Center, Perelman School of Medicine of the University of Pennsylvania, Philadelphia, United States

Background: Chimeric antigen receptor T-cells (CAR T-cells) used for the treatment of relapsing/refractory B-cell precursor acute lymphoblastic leukemia have led to exciting clinical results. However, CAR T-cell approaches revealed a potential risk of CD19-/CAR+ leukemic relapse related to an inadvertent transduction of leukemia blasts. In this study, we addressed the hypothesis that the CAR design could be crucial for regulating the ability of CAR.CD19 T-cells to recognize and kill CAR+ leukemia cells.

Methods: We performed real-time quantitative PCR for Ig-rearrangements and high-sensitivity EuroFlow flow-cytometry analysis to evaluate the impact of a high percentage of leukemia blast contamination in patient-derived starting material (SM) on CAR T-cell drug product (DP) manufacturing. Moreover, in vitro and in vivo experiments were conducted to investigate whether CAR design may affect the control of CAR+ leukemia cells. We used 4 different retroviral CAR constructs carrying anti-human CD19-scFv from FMC63 clone and 4.1bb and CD3ζ cytoplasmic domain and the safety switch inducible-caspase-9 (iC9): CAR construct in which VL and VH fragments were joined by a linker represented by three GSSSS repetitions (3xG4S, long linker, LL); CAR construct in which VL and VH fragments were joined by a linker represented by three GSSSS repetitions (3xG4S, long linker, LL), in frame with 16aa sequence derived from human CD34 antigen (ΔCD34, long hinge, LH); CAR construct in which VL and VH fragments were joined by a linker represented by one GSSSS repetition (G4S, short linker, SL), in frame with CD8 stalk domain (short hinge, SH; CAR construct in which VL and VH fragments were joined by a linker represented by one GSSSS repetition (G4S, short linker, SL), in frame with 16aa sequence derived from human CD34 antigen (ΔCD34, long hinge, LH).

Results: The presence of large amounts of CD19+ cells in SM did not affect the transduction level of DPs, as well as the rate of expansion of CAR T-cells at the end of standard production lasting 14 days. DPs were deeply characterized by flow-cytometry and molecular biology for Ig-rearrangements, showing that the level of B-cell contamination in DPs did not correlate with the percentage of CD19+ cells in SM in the studied patient cohort. Moreover, we demonstrated either in vitro or in vivo that CD19+CAR+ leukemic cells were killed by CAR.CD19 T-cells, although T-cells genetically modified with a CAR construct characterized by a short linker (SL) between the VH-VL region and a long hinge (LH) were able to better control in vivo expansion of CAR-positive leukemia cells in a xenograft mouse model than T-cells genetically modified with CAR with long linker (LL) and a short hinge (SH). Furthermore, the presence of the suicide gene iC9 in the construct allowed to control CAR+ leukemic cells in both in vitro and in vivo experiments.

Conclusions: Taken together, these data suggest that a VH-VL short linker and the inclusion of iC9 suicide gene may result in a safe CAR-T cell product, representing a step forward on leukemia cell control in case inadvertent B-cell precursor acute lymphoblastic leukemia blast cell transduction occurs.

Disclosure: None.

O035. Engineered CAR.CD123 NK Cells As An Innovative “Off The Shelf” Strategy To Treat CD123^POS Childhood Acute Myeloid Leukaemia

Simona Caruso¹, Roselia Ciccone¹, Zeinab Abbaszadeh¹, Simona Manni¹, Marika Guercio¹, Biancamaria Cembrola¹, Stefano Di Cecca¹, Antonio Camera¹, Angela Pitisci¹, Matilde Sinibaldi¹, Giuseppina Li Pira¹, Francesca Del Bufalo¹, Pietro Merli¹, Mattia Algeri¹, Luisa Strocchio¹, Luciana Vinti¹, Biagio De Angelis¹, Concetta Quintarelli¹, Franco Locatelli¹

¹ Bambino Gesù Children Hospital, IRCCS, Rome, Italy

Background: Childhood acute myeloid leukaemia (AML) is a life-threatening malignancy, with current survival rates of ∼70%. The standard of care for the treatment of this disease is represented by repeated courses of intensive chemotherapy followed by allogeneic hematopoietic stem cell transplantation in patients with high-risk features. Despite improvement in patient outcome observed in the last years, disease recurrence occurs in 25-30% of patients and represents the main cause of treatment failure. Although therapy with human T cells genetically engineered with Chimeric Antigen Receptors (CARs) targeting CD19 has produced unprecedented successful results in the treatment of B-cell malignancies, CAR T-cell manufacturing remains an expensive, time-consuming approach for the generation of autologous drug products. Therefore, an “off-the-shelf” CAR-cell product based on Natural Killer (NK) cells could offer several advantages, mainly represented by the lack of alloreactivity and high cytotoxic potential against tumour targets.

In AML, the high CD123 expression levels in leukaemia stem cells provides the biological basis for considering this target in the development of CAR-cell therapies. Thus, we developed an innovative “off the shelf” third-party strategy based on the use of NK cells engineered through a second-generation CAR.CD123.

Methods: Retroviral plasmid was designed to carry the cassette of a second-generation CAR.CD123 including 4.1bb as costimulatory domain. NK cells were generated from peripheral blood of healthy donors (HD) and ex vivo expanded by a feeder-free FBS-free approach in the presence of IL2 (C. Quintarelli et al., Leukaemia. 2020 Apr;34(4):1102-1115). CAR.CD123 NK cell functionality was evaluated both in vitro and in vivo.

Results: After genetic modification with retroviral CAR construct, NK cells display a stable in vitro CAR.CD123 expression (58%±21,9%) and show significant anti-leukaemia activity towards CD123+ tumour cell lines (6,5%±10%, 5,1%±5% and 7,7%±9% of residual THP1, MOLM-13 and OCI-AML3 cells after co-culture with CAR.CD123 NK cells at an effector target ratio of 1:1, respectively. Figure 1A). Importantly, CAR.CD123 NK cells exert high anti-leukaemia activity towards CD123+ primary AML blasts collected from pediatric patients at diagnosis, compared to un-transduced NK cells (p = 0,0005. Figure 1B). To test on-target off-tumour effects of our approach, we performed short-term colony forming assay of CD34+ bone marrow-derived cells from HD pre-cultured with un-modified or CAR.CD123-engineeered NK cells. We did not observe significant reduction of BFU-E and CFU-GM colonies in our assay. Finally, we have developed a xenograft immunodeficient mouse model of human AML CD123+ THP-1 cell line, treated with CAR.CD123 NK cells. We clearly demonstrate that two consecutive infusions of CAR.CD123 NK cells significantly improved overall survival (OS) of the mice, with 81% of the treated mice alive at day 125 (vs 0% of the mice receiving un-modified NK cells. Figure 2).

Conclusions: These in vitro and in vivo data prove the feasibility of feeder-free generation of stably transduced CAR.CD123-NK cells. This third-party approach represents an innovative ‘off-the-shelf’ strategy for targeting CD123+ leukemic cells in the absence of significant toxicity on haematopoietic precursors, offering to AML patients an innovative therapeutic approach in case of relapse or resistance to standard therapies.

Disclosure: The authors declare that they have no conflict of interest.

O036. Third-Generation GD2.CAR T-Cells Incorporating CD28.4-1BB Costimulatory Domains for Treatment of GD2+ Sarcomas

Antonio Camera¹, Roselia Ciccone¹, Marika Guercio¹, Sofia Reddel¹, Domenico Orlando¹, Biancamaria Cembrola¹, Stefano Di Cecca¹, Matilde Sinibaldi¹, Simona Caruso¹, Zeinab Abbaszaddeh¹, Simona Manni¹, Francesca Ferrandino¹, Francesca Del Bufalo¹, Angela Di Giannatale¹, Marta Colletti¹, Ida Russo¹, Giuseppe Maria Milano¹, Rossella Rota¹, Marco Pezzullo¹, Ezio Giorda¹, Rita Alaggio¹, Rita De Vito¹, Concetta Quintarelli¹, Biagio De Angelis¹, Franco Locatelli¹

¹ Ospedale Pediatrico Bambino Gesù, Roma, Italy

Background: Sarcomas are a subgroup of cancers which include a variety of different bone and soft-tissue tumors. They account for approximately 10-15% of all childhood solid tumors and are characterized by a significant mortality rate in patients of 0–19 years of age. GD2 is a disialoganglioside, which is abundantly expressed in various types of cancers and some subtypes of sarcomas, including Rhabdomyosarcoma (RMS), Ewing’s Sarcoma (EWS) and Osteosarcoma (OS). Therefore, GD2 represents an ideal candidate for targeted cancer therapy. Genetically modified T cells expressing chimeric antigen receptors (CARs) represents a new class of therapeutics drugs that has shown encouraging results for the treatment of GD2+ cancers.

Methods: A clinical grade γ-retroviral vector carrying a third-generation CAR targeting GD2 (CAR.GD2.CD28.4-1BB.CD3z) is currently used in our Institution in a phase I/II academic clinical study for patients with high-risk and/or relapsed/refractory Neuroblastoma (NCT03373097). The cytotoxic activity of CAR.GD2 T cells was assessed in vitro by 6-day long-term co-culture assay using RMS (RD) and OS (143B) cell lines. In a RMS metastatic mouse model, 6-week old NSG mice were engrafted intravenously (i.v.) with RD.eGFP-FF-luciferase tumor cell line. After tumor engraftment, the mice received i.v. injection of effector T cells (either NT or CAR.GD2 T cells). Furthermore, an orthotopic mouse model of OS was developed. NSG mice were engrafted with 143B.eGFP-FF-luciferase tumor cells in the tibia periosteum and, after tumor engraftment, mice were treated with effector NT or CAR.GD2 T cells. Tumor growth was evaluated using IVIS imaging system.

Results: CAR.GD2 T cells exert a significant tumor control against both RD and 143B cell lines at target:effector ratio of 1:1 (10.6%±9.4% and 2.6.%±3.8% residual tumor cell after co-culture with CAR.GD2 T cells, respectively) as compared to control NT T cells (83.2%±7.9% and 75.6%±11.5%, respectively; in both case p = 0.00001). In in vivo RMS model, CAR.GD2 T cells showed a superior anti-tumor activity compared to NT T cells, which resulted in a significant reduction of tumor bioluminescence after 50 days (5,7x109 vs 2.6x107, respectively, p = 0.05). Median overall survival for RMS-tumor-bearing mice treated with CAR.GD2 T cells was significantly longer (69.5 days) compared to mice treated with NT T cells (50.0 days) (p = 0.012). Moreover, an enrichment of CD8+CAR+ T cells with Central Memory and Effector Memory phenotype was observed. In an orthotopic mouse model of OS, CAR.GD2 T cells failed to control tumor growth. However, genetically modified CAR T cells slowed down metastatic spread of the primary tumor, increasing significantly the overall survival of mice treated with CAR.GD2 T cells compared to NT T cells (46% vs 0% at day +46, p = 0.0052).

Conclusions: Overall, these pre-clinical data show that CAR T cells redirected against GD2 antigen may be considered as an alternative therapeutic option for selected GD2 highly-expressing pediatric sarcomas patients.

Disclosure: Nothing to declare.

Cellular Therapies other than CARs

O037.Overall Survival by Best Overall Response with Tabelecleucel In Patients with Epstein-Barr Virus-Driven Post-Transplant Lymphoproliferative Disease After Allogeneic Hematopoietic Cell Transplant

Susan Prockop¹, Laurence Gamelin², Rajani Dinahavi², Yan Sun³, Norma Guzman-Becerra³, Hema Parmar²

¹ Memorial Sloan Kettering Cancer Center, New York, United States, ² Atara Biotherapeutics, South San Francisco, United States, ³ Atara Biotherapeutics, Thousand Oaks, United States

Background: Tabelecleucel is an investigational, off-the-shelf, allogeneic Epstein–Barr virus (EBV)-specific T-cell immunotherapy being studied in patients (pts) with serious EBV-driven diseases, including post-transplant lymphoproliferative disease (EBV⁺ PTLD). Allogeneic hematopoietic cell transplantation (HCT) recipients with EBV⁺ PTLD that is relapsed or refractory (R/R) to rituximab treatment decline rapidly with a median overall survival (OS) of ~1.7 months (Socié EBMT 2020), demonstrating an urgent unmet need for effective, well-tolerated therapies. We have previously shown clinical benefit in pts with EBV⁺ PTLD after HCT who responded (complete response [CR] or partial response [PR]) to tabelecleucel, including an 86% 2-year survival rate (Prockop ASH 2019 and JCI 2019). Here, we report aggregate OS in patients with EBV⁺ PTLD after HCT with CR or PR with tabelecleucel treatment.

Methods: Treatment response and OS were assessed in three studies (NCT00002663, NCT01498484 and NCT02822495). All pts received tabelecleucel at ≈2 x 10⁶ cells/kg on Days 1, 8 and 15 in a 35-day treatment cycle. Pts received a median (range) of 2.0 (1–5) cycles.

Results: Fifty HCT recipients with EBV⁺ PTLD R/R to rituximab were treated with tabelecleucel. The investigator assessed objective response rate (PR+CR) was 62% (31/50) with a best overall response (BOR) of CR (n = 24) or PR (n = 7; Table 1, Figure 1). Two-year survival rates were 81.6 and 85.7% for patients with CR and PR, respectively (Table 1). Treatment was well tolerated with no confirmed evidence for graft vs host disease, cytokine release syndrome, or neurotoxicity attributable to tabelecleucel in these very sick, treatment refractory, and immunocompromised pts.

Table 1. OS by BOR

BOR	CR (n = 24)	PR (n = 7)
1-year OS rate	86.7%	85.7%
(95% CI)	(64.2, 95.5)	(33.4, 97.9)
2-year OS rate	81.6%	85.7%
(95% CI)	(57.9, 92.7)	(33.4, 97.9)
Median follow-up	28.2	25.3
(min, max) months	(1.4, 88.9)	(5.1, 52.4)

1. CI=confidence interval.

Conclusions: Tabelecleucel is well tolerated, and the data reported here show that across studies, pts with EBV+ PTLD following HCT that is R/R to rituximab who responded to tabelecleucel experienced long-term survival, which contrasts with the short median OS of <2 months historically reported in this high-risk, treatment refractory pt population (Socié EBMT 2020). Importantly, patients who achieved a PR with tabelecleucel derived similar OS benefit to those who achieved a CR.

Clinical Trial Registry: NCT00002663, NCT01498484 and NCT02822495. Clinicaltrials.gov.

Disclosure: S Prockop holds a consulting/advisory role with Mesoblast and has received research funding from Mesoblast, Atara Biotherapeutics and Jasper Pharmaceuticals. S Prockop also holds IP related to the development of third party viral specific T cells, with all interests assigned to MSK. L Gamelin, R Dinavahi,

Y Sun, N Guzman-Becerra, and H Parmar are employees of Atara Biotherapuetics.

O038. Crispr-Based Replacement of The Endogenous T-Cell Repertoire With Novel, High-Avidity, Natural T Cell Receptors To WT1 For The Treatment of Acute Myeloid Leukemia

Eliana Ruggiero¹, Erica Carnevale¹, Aaron Prodeus², Zulma Magnani¹, Barbara Camisa¹, Claudia Politano¹, Lorena Stasi¹, Alessia Potenza¹, Beatrice Claudia Cianciotti¹, Francesco Manfredi¹, Mattia Di Bono¹, Luca Vago¹, Michela Tassara¹, Sara Mastaglio¹, Maurilio Ponzoni¹, Francesca Sanvito¹, Daniel O’Connell², Ivy Dutta², Stephanie A Yazinski², Mark McKee², Mohamed S Arredouani², Birgit Schultes², Fabio Ciceri¹, Chiara Bonini¹

¹ San Raffaele Scientific Institute, Milan, Italy, ² Intellia Therapeutics, Cambridge, United States

Background: By targeting tumor antigens with high sensitivity, and by promoting T cell survival signals, T cell receptor (TCR)-based therapy has the potential to induce potent and durable clinical responses in cancer patients. However, the need of high-avidity TCRs specific for shared oncogenic antigens and of manufacturing protocols able to completely redirect T-cell specificity while preserving T-cell fitness, remain relevant limiting factors. Here, we aim at the identification of a panel of novel tumor-specific TCRs, to be exploited in adoptive T-cell therapy.

Methods: We focused on Wilms’ Tumor 1 (WT1), a zinc finger transcription factor overexpressed by a wide range of hematological and solid tumors, of proven immunogenicity, with restricted expression on healthy tissues and a strong correlation with oncogenesis, all characteristics that prioritize this antigen for cancer immunotherapy purposes. We designed and implemented an innovative protocol for the rapid isolation of WT1-specific T cells and for the characterization of a library of tumor-specific TCRs restricted to different human leukocyte antigen (HLA) alleles. T cell recognition was assessed by flow cytometry in terms of CD107a expression and IFNγ production. Recognized peptides were mapped by a deconvoluting grid and their HLA restriction assessed by using a panel of cell lines harboring the HLA alleles of interest. Tumor-specific TCRs were identified by TCR αβ sequencing.

Results: By longitudinal monitoring of T cell functionality and dynamics in 14 healthy donors, we isolated 19 WT1-specific TCRs which recognize several peptides restricted by 5 HLA alleles, and display a wide range of functional avidities, including some in the low nM range. TCR αβ sequencing at different time points enabled the longitudinal clonal tracking of the tumor-specific T cells and the correct pairing of TCR α and β chains, without the need of cell cloning nor single cell analysis. We selected two high avidity HLA-A*02:01-restricted TCRs specific for the less explored immunodominant WT137-45 epitope, naturally processed by primary AML blasts. With tailored, high precision CRISPR/Cas9 genome editing tools, we combined TCR targeted integration into the TRAC locus with TRBC knock-out, thus avoiding the risk of TCR mispairing and maximizing TCR expression and function. The new cellular product, generated with a clinic-ready protocol in a 4-day span followed by a 6-10-day expansion, was enriched in memory stem T cells. Of note, one TCR showed antigen-specific responses in both CD4⁺ and CD8⁺ T cells, and efficiently eliminated primary AML blasts, harvested from leukemic patients, in vitro and in vivo in the absence of off-tumor toxicity.

Conclusions: T cells engineered to express this receptor are being advanced into clinical development for AML immunotherapy and potentially other WT1-expressing tumors.

Disclosure: A. Prodeus, I. Dutta, S.A.Yazinski, M. McKee, M.S. Arredouani and B. Schultes: employees and share holders of Intellia Therapeutics.

C. Bonini: research funding from Intellia Therapeutics.

L. Vago: research funding from GenDx and Moderna Therapeutics.

The other authors declare no competing interests.

O039. Exogenous M-CSF Protects From Lethal A. Fumigatus Pulmonary Infection Early After Hematopoietic Stem Cell Transplantation

Dalia Sheta^1,2, Zeinab Mokhtari¹, Marlene Strobel^1,2, Yidong Yu¹, Jorge Amish³, Nora Trinks^4,2, Sina Thusek¹, Hermann Einsele¹, Katrin Heinze⁵, Ulrich Terpitz⁴, Andreas Beilhack^1,2

¹ Würzburg University Hospital, Würzburg, Germany, ² Graduate School of Life Sciences, Würzburg, Germany, ³ School of Biological Sciences, University of Manchester, Manchester, United Kingdom, ⁴ Biocenter, University of Würzburg, Würzburg, Germany, ⁵ Rudolf-Virchow-Center, University of Würzburg, Würzburg, Germany

Background: In the time window of bone marrow engraftment and immune reconstitution, patients undergoing hematopoietic cell transplantation (HCT) particularly are vulnerable to life-threatening opportunistic infections, such as invasive fungal infections of the lung. Alveolar macrophages (AMs) form a first-line defense in the lung and are tissue resident and not bone marrow derived. Here, we addressed how cytokines impact AMs function after allogeneic HCT (allo-HCT) to improve protection from lethal lung infections.

Methods: To investigate the effect of various cytokines, including M-CSF and IL-34 on primary AMs, we utilized super resolution and dynamic confocal microscopy to study AM phenotype, migration and phagocytic behavior in vitro. To study the outcome of exogenous cytokine stimulation on the pulmonary immune response in vivo we employed mouse models of allo-HCT (8 Gy TBI, C57Bl/6àBALB/c) and invasive A. fumigatus infection (survival, clinical performance, flow cytometry and 3D light sheet fluorescence microscopy of the lungs).

Results: Allo-HCT recipients survived an intratracheal A. fumigatus infection when infected 6 days after allo-HCT but succumbed to lethal invasive aspergillosis when infected as early as 4 days after allo-HCT. 6 days after allo-HCT, tissue resident AMs showed the highest frequency, proliferation and phagocytic activity in the lung, when compared to neutrophils and monocytes, suggesting that AMs were responsible for protecting mice from lethal A. fumigatus infection. Therefore, we asked whether cytokines could boost AMs to achieve earlier protection after allo-HCT. Consequently, we screened the effects of various cytokines on AM function in vitro and observed that M-CSF increased AM’ migration speed (0.52 μm2/min) and versatility (diffusion co-efficient of 0.78) when compared to IL-34 (0.3 μm/min and 0.6, respectively) or mock controls. Notably, M-CSF mediated this functional gain specifically on lung-derived AM but not peritoneal macrophages.

Next, we investigated whether M-CSF therapy can enhance pulmonary immune protection in vivo and prevent invasive aspergillosis early after allo-HCT. To test this, allo-HCT-recipients were treated with M-CSF and subsequently infected with A. fumigatus 4 days after allo-HCT. 3D microscopy and flow cytometry revealed that exogenous M-CSF not only boosted myelopoiesis by 2-fold but also expanded the local tissue-resident AM population in the lung by 1.5-fold. M-CSF improved the clinical performance of mice and protected 100% of allo-HCT recipients from an early A. fumigatus infection on day+4 after allo-HCT.

Conclusions: M-CSF treatment enhances lung-resident AM proliferation and function and protects from early A. fumigatus lung infection, which may benefit patients undergoing HCT.

Disclosure: No conflicts of interest relevant to this study.

O040. A Phase I Dose-Escalation Single Center Study To Evaluate The Safety Of Allogenic Memory T Cells Containing SARS-COV-2 Specific Lymphocytes As Adoptive Therapy in COVID19

Antonio Pérez-Martínez¹, Cristina Ferreras², Marta Mora-Rillo¹, Pilar Guerra¹, Barbara Pascual-Miguel², Carmen Mesrtre-Durán², Alberto Borobia¹, Antonio Carcas¹, Irene Garcia¹, Elena Zapardiel¹, Mercedes Gasior¹, Raquel de Paz¹, Antonio Marcos¹, José Luis Vicario¹, Antonio Balas³, Cristina Eguizabal⁴, Carlos Solano⁵, Rocio Montejano¹, José Ramón Arribas¹, Bernat Soria⁶

¹ Hospital Universitario La Paz, Madrid, Spain, ² Instituto de Investigación La Paz (idiPAZ), Madrid, Spain, ³ Centro de Transfusiones de la Comunidad de Madrid, Madrid, Spain, ⁴ Centro Vasco de Transfusion y Tejidos Humanos (CVTTH), Bilbao, Spain, ⁵ Hospital Clínico, Valencia, Spain, ⁶ Instituto de Investigacion Universidad Miguel Hernández de Alicante, Alicante, Spain

Background: We are in the second wave created by the severe acute respiratory syndrome coronavirus 2 (SARS-Cov-2) with no effective treatments for moderate/severe hospitalized COVID-19 patients. Memory T cells defined as CD45RA- T cells obtained from COVID-19 convalescent donors, may be useful to treat SARS-CoV-2 pneumonia and/or lymphopenia because they retain SARS-CoV-2 specific T cells. In addition, the broad donor memory T cell repertoire may protect these vulnerable patients from other common viral co-infections. Adoptive infusion of memory T lymphocytes has been previously successfully reported improving pathogen-specific immune response after HSCT. We previously reported the existence of a SARS-CoV-2 specific T cell population within the CD45RA- T memory cells of the blood from the convalescent donors. These cells can be stored and be immediately available as an “off-the-shelf” COVID-19 convalescent donor-derived biobank. A wide number of doses can be easily manufactured using the CliniMACS Miltenyi device without the need for a GMP facility. With this approach, we can generate a biobank of "living drugs" containing SARS-CoV-2 specific T cells that can cover the country population based on the human leukocyte antigen (HLA) genotype. In this study, we conducted a first-in-human phase 1 clinical trial, dose-escalation study to evaluate the safety of a single infusion of CD45RA- memory T cells containing SARS-CoV-2 specific T cells from a COVID-19 convalescent donor as adoptive cell therapy against moderate/severe cases of COVID-19.

Methods: Eligible participants were hospitalized patients with confirmed COVID-19 by PCR suffering from pneumonia and lymphopenia related to COVID-19. Patients were enrolled based on the matched with the HLA genotype of the convalescent donor and following the protocol inclusion/exclusion criteria. Participants were sequentially enrolled to receive a single infusion in a dose-escalating manner. All patients received the standard of care. Primary outcomes were to determine the safety of a single infusion of memory T cells from a healthy donor and the dose-limiting toxicity. Secondary outcomes were to evaluate time to lymphopenia recovery and immune dysregulation.

Results: Nine patients were enrolled. The first 3 patients received 1x105 cells /kg, the next 3 patients received 5x105 cells/Kg and the last 3 patients received 1x106 cells/kg of CD45RA- memory T cells. Patients’ clinical status measured by NEWS and 7-category point ordinal scales showed an improvement 6 days after infusion. The median duration of hospitalization after the infusion was 8 days in the low dose group, 7 in the intermediate dose group, and 4 days in the high dose group. The inflammatory parameters were stabilized and they all showed lymphocyte recovery two weeks after infusion. Donor microchimerism was observed at least for 2 weeks after infusion.

Conclusions: Despite the small sample size, our study supports the idea that treatment of COVID-19 patients with moderate/severe symptoms using CD45RA- memory T cells is feasible, safe, and it is associated with quick clinical improvement and short hospitalization stays. In our study, neither patient had an infusion reaction, inflammatory impairment, or other serious adverse reaction. We are now opening a multicenter phase 2 study to show treatment efficacy.

Clinical Trial Registry: NCT04578210

Disclosure: Antonio Pérez- Martínez, Cristina Ferreras and Bernat Soria filed a patent on this topic.

O041. Open Label Dose Escalation Trial of AB-205 (E-CEL® Cells) In Adults With Systemic Lymphoma Undergoing High-Dose Therapy And Autologous Hematopoietic Cell Transplantation (HDT-AHCT)

Carolyn Mulroney¹, Michael Scordo², Mehrdad Abedi³, Lihua E. Budde⁴, Bita Fakhri⁵, Attaphol Pawarode⁶, Bhagirathbhai Dholaria⁷, Geoff Shouse⁴, Edward Kavalerchik⁸, Sanjay Aggarwal⁸, Muzaffar Qazilbash⁹, Paul Finnegan⁸, Sergio Giralt²

¹ UCSD Moores Cancer Center, San Diego, United States, ² Memorial Sloan Kettering Cancer Center, New York, United States, ³ UC Davis Medical Center, Sacramento, United States, ⁴ City of Hope National Medical Center, Duarte, United States, ⁵ University of California San Francisco, San Francisco, United States, ⁶ University of Michigan, Ann Arbor, United States, ⁷ Vanderbilt University, Nashville, United States, ⁸ Angiocrine Bioscience, San Diego, United States, ⁹ The University of Texas MD. Anderson Cancer Center, Houston, United States

Background: HDT-AHCT is a standard curative therapy for patients with relapsed, high risk lymphomas. Severe regimen-related toxicities (SRRT) occur frequently after AHCT, leading to high symptom burden and potentially limiting wider use. The cause of SRRT is thought to be diffuse injury to the organ vascular endothelial niches caused by off-target cytotoxic effects of HDT. Organs with high cell turnover such as oral-gastrointestinal (GI) tract are most affected. The cytotoxic effect impairs the renewal of mucosa, leading to mucositis, nausea, vomiting, and diarrhea. With loss of mucosal integrity, translocation of gut flora into the systemic circulation may lead to life-threatening infections.

AB-205 is an experimental engineered-cell therapy designed to repair damaged tissues and reduce SRRT. AB-205 contains E-CEL® cells: allogeneic E4+ human umbilical vein endothelial cells.

Methods: AB-205 was administered intravenously 4 hours after AHCT in dose-escalated cohorts: 5, 10 and 20x10⁶ cells/kg, either as single dose or divided (D0, D2). Objectives included safety and assessment of grade (G)≥3 AEs (NCI-CTCAE v5.0), oral/GI SSRT (G≥3: mucositis, nausea, vomiting and diarrhea) and time to engraftment. Supportive care was administered per local practice.

A retrospective chart review (n = 45, 2019-2020) at a participating site served as a contemporary control.

Results: As of 18-August 2020, 28 systemic lymphoma subjects were treated with AB-205 with a median (range) follow up of 104 (12, 399) days (Table).

Table 3 Baseline Characteristics (Systemic Lymphoma)

No MTD has been reached. AEs were generally mild/moderate and as expected with HDT-AHCT. One subject relapsed (D101).

AB-446p205 therapy demonstrated dose-dependent reduction of oral/GI SRRT (Figure) and all-grade oral/GI toxicities. The highest AB-205 dose (20 x 10⁶ cells/kg) eliminated oral/GI SRRT. The G≥3 oral/GI SRRT event rate in the control cohort was 60%. A trend for a dose-dependent decrease in febrile neutropenia (FN) was observed (80%, 67% and 53% with 5, 10 and 20 x 10⁶ cells/kg respectively). 71% of subjects had accelerated platelet engraftment, occuring within 1 day after neutrophil engraftment vs 16% in the control.

Rate of Severe (CTCAE Grade ≥ 3) Regimen-Related Oral/GI Toxicities

G≥3 infection AEs within 28 days of AHCT were reported for only 1 subject. As of the above follow-up date, there have been no reports of severe regimen related toxicities affecting the pulmonary, cardiac, renal or hepatic systems.

Conclusions: Highest dose AB-205 eliminated oral/GI SRRT compared to 50-60% event rate seen in contemporary control cohort and previously published reports (Olivieri J, BBMT 2018; Perales MA, BBMT 2017). Severe infection rate was low which may reflect the accelerated repair of oral/GI mucosal lining. AB-205 will be assessed in a forthcoming pivotal randomized clinical trial for potential registration.

Clinical Trial Registry: NCT03925935

Disclosure: Carolyn Mulroney has nothing to declare.

Michael Scordo declares McKinsey and Company consultancy, Angiocrine Bioscience, Inc consultancy and research funding, Omeros corporation consultancy, Kite, A Gilead Sciences company, ad-hoc advisory board.

Mehrdad Abedi declares Abbvie speakers bureau, Settle Genetics speakers bureau, Takeda speakers bureau,BMS speakers bureau and research funding, Gilead Sciences speakers bureau and research funding.

Lihua E. Budde declares Gilead Sciences consultancy, Roche consultancy, Kite, a Gilead Sciences company, consultancy, honoraria, speakers bureau, Astra Zeneca research funding and speakers bureau, Mustang Therapeutics research funding, Merck research funding, Amgen research funding.

Bita Fakhri has nothing to declare.

Attaphol Pawarode has nothing to declare.

Bhagirathbhai Dholaria has nothing to declare.

Geoff Shouse declares Kite, a Gliead Sciences company, speakers bureau.

Edward Kavalerchik declares Angiocrine Bioscience current employment and equity holder, Abbvie equity holder.

Sanjay Aggarwal declares Angiocrine Bioscience current employment and equity holder, Kadmon Corporation current equity holder.

Muzaffar Qazilbash declares Janssen, Bioline, Angiocrine Bioscience, Amgen research funding; Bioclinica consultancy.

Paul Finnegan declares Angiocrine Bioscience current employment.

Sergio Giralt declares Celgene and Novartis consultancy, honoraria and research funding; Jazz consultancy and honoraria; Amgen, Actinium and Miltenyi consultancy and research funding; Miltenyi consultancy and research funding; Kite, consultancy; Takeda Research Funding.

O042. Immunotherapy In Acute Leukemia: A Gitmo (Gruppo Italiano Trapianto Midollo Osseo) Survey on Efficacy And Toxicity of Donor Lymphocyte Infusions After Allogeneic Stem Cell Transplantation

Francesca Patriarca¹, Alessandra Sperotto¹, Francesca Lorentino², Elena Oldani³, Sonia Mammoliti⁴, Miriam Isola⁵, Alessandra Picardi⁶, William Arcese⁷, Giorgia Giorgia Saporiti⁸, Roberto Sorasio⁹, Nicola Mordini⁹, Irene Cavattoni¹⁰, Maurizio Musso¹¹, Carlo Borghero¹², Caterina Micò³, Renato Fanin¹, Benedetto Bruno¹³, Fabio Ciceri², Francesca Bonifazi¹⁴

¹ ASUFC, University of Udine, Udine, Italy, ² IRCSS Ospedale San Raffaele, Milano, Italy, ³ ASST Papa Giovanni XXIII, Bergamo, Italy, ⁴ Trial Office GITMO, Genova, Italy, ⁵ University of Udine, Udine, Italy, ⁶ AO Cardarelli, Napoli, Italy, ⁷ AOU Policlinico Tor Vergata, Roma, Italy, ⁸ Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milano, Italy, ⁹ AO S. Croce, Cuneo, Italy, ¹⁰ AO di Bolzano, Bolzano, Italy, ¹¹ Ospedale LaMaddalena, Palermo, Italy, ¹² Ospedale San Bortolo, Vicenza, Italy, ¹³ AOU Città della Salute e della Scienza, Torino, Italy, ¹⁴ AOU di Bologna, Policlinico S. Orsola-Malpighi, Bologna, Italy

Background: Donor lymphocyte infusions (DLIs) have provided a way to enhance the graft-versus-leukemia (GVL) effect after allogeneic stem cell transplantation (HSCT) for decades, but their role in the treatment strategy of acute leukemias (AL) is still poorly defined. We conducted a retrospective multicentre study including all consecutive pediatric and adult patients with AL who received DLIs after HSCT between January 1, 2010 to December 31, 2015, in order to determine the efficacy and the toxicity of the treatment.

Methods: 252 patients, median age 45.1 years (1.6-73.4) were enrolled from 35 Italian Transplant Centres. Median follow-up was 878 days (55-6754) after HSCT. The main characteristics of patients, donors and transplants were provided by the Italian Registry; other DLI-specific information were requested to the centre in a study-specific database. OS was calculated from HSCT and first DLI, and the impact of prognostic factors was evaluated.

Results: The underlying disease was AML in 180 cases (71%), ALL in 68 patients (27%), biphenotipic AL in 4 patients (2%). Donors were HLA identical sibling (39%), unrelated (40%) or haploidentical (21%). Conditioning regimens were myeloablative in 72% of patients. The first DLI was administered at a median of 258 days (55-6754) after HSCT. The main indication was leukemia relapse or persistence (73%), followed by mixed chimerism (17%) and preemptive/prophylactic use (10%). 40% of patients received no treatment before the first DLI, while radiotherapy, conventional chemotherapy or targeted treatment were administered in 3%, 39% and 18% of patients, respectively. 156 (62%), 91 (36%), 49 patients (19%) received 2, 3 or ≥ 4 infusions, respectively, with a median of 31 days between 2 subsequent DLI. An escalating schedule was mainly chosen, ranging from a median dose of 1 to 10 x106/kg CD3+ lymphocytes. Severe adverse events were a few, including 3% grade III-IV GVHD, 11% grade III-IV haematological toxicity and 3 DLI-related deaths. 46 patients (18%) performed a second HSCT. After a median follow-up of 232 days (32-1390) from the first DLI, 1-year, 3-year and 5-year OS of from start of DLI treatment was 55%, 39%, and 33%, respectively. In multivariate analysis older recipient age and haploidentical donors significantly reduced OS (HR 1.02; p = 0.001 and HR 2.80; p = 0.000, respectively), whereas the following factors were significantly associated with a prolonged OS: DLI administration caused by mixed chimerism or preemptive/prophylactic treatment in comparison with AL relapse (HR 0.44; p = 0.002 and HR 0.22; p = 0.000); an escalating schedule in comparison with a single dose (HR 0.87;p = 0.050) and time longer than 2 years between HSCT and first DLI (HR 0.41; p = 0.003). Higher rate of acute GVHD and lower AL remissions were reported after haploidentical DLIs in comparison with DLIs from conventional donors.

Conclusions: This GITMO survey confirms that DLI administration in absence of overt hematological relapse and multiple infusions are associated with a favorable outcome in AL patients. The poorer outcome after DLI from haploidentical donors should be interpreted with caution due to the heterogeneity of GVHD prophylaxis, however it may represent an area of further investigation.

Clinical Trial Registry: retrospective study approved by the Ethic Committee Friuli Venezia Giulia on 2017, 3 October (Protocon number 26522)

Disclosure: No conflicts of interest to declare.

O043. Monitoring Patient SARS-COV-2-T-Cell Immune Responses And Generating SARS-COV-2-specific T Cells From Convalescent Donors; Implications For The Identification of High-Risk Patients and Treatment With T-Cell Immunotherapy

Anastasia Papadopoulou¹, Penelope-Georgia Papayanni^1,2, Dimitris Hasiotis^1,2, Kiriakos Koukoulias^1,2, Elisavet Vlachonikola³, Glykeria Gkoliou³, Eleni Gavriilaki¹, Militsa Bitzani¹, Eleni Geka⁴, Polichronis Tasioudis⁴, Adamantios Chloros¹, Asimina Fylaktou⁵, Maria Triantafyllidou¹, Afroditi K Boutou¹, Ioannis Kioumis¹, Achilles Anagnostopoulos¹, Anastasia Chatzidimitriou³, Evangelia Yannaki^1,6

¹ “George Papanikolaou” Hospital, Thessaloniki, Greece, ² Aristotle University of Thessaloniki, Thessaloniki, Greece, ³ Centre for Research and Technology Hellas (CERTH), Thessaloniki, Greece, ⁴ AHEPA University Hospital, Thessaloniki, Greece, ⁵ Hippokration General Hospital, Thessaloniki, Greece, ⁶ University of Washington, Seattle, United States

Background: The SARS-CoV-2 pandemic poses an urgent need for the development of effective therapies for COVID-19. We here aimed to assess i) the kinetics of endogenous SARS-CoV-2-specific T cells (CoV-2-STs) in COVID-19 patients and evaluate their role in disease outcome, ii) the development and persistence of CoV-2-STs in convalescent donors, and iii) the feasibility of generating convalescent donor-derived CoV-2-STs.

Methods: Immunological monitoring and virus-specific immune recovery of COVID-19 patients (non-ICU=14, ICU-recovered=7, ICU-critical=16) was assessed at 2 time-points (week0-1, week2-3) post hospitalization or ICU admission, by flow cytometry (FC) and Elispot. Control groups were convalescent and healthy/unexposed donors. To generate CoV-2-STs, PBMCs were exposed to viral pepmixes (NCAP, Spike) and cultured in the presence of IL-4/IL-7 for 10 days. Cells were immunophenotypically characterized by FC, functionally by IFN-γ/TNF-α Elispot and FC and molecularly for T cell receptor (TR) gene repertoire by NGS.

Results: We observed T-cell lymphopenia in COVID-19 patients, particularly in the ICU critical patients (p < 0.0001), whose T cells were functionally impaired, demonstrating higher levels of exhaustion markers (PD-1/CTLA-4/Lag3/TIM3; p < 0.001) and a varying differentiation status over control group-derived T cells (p ≤ 0.0025). A strong SARS-CoV-2 specific immunity was observed in convalescents and patients, suggesting that SARS-CoV-2 boosts T-cell immunity. A mild SARS-CoV-2-specific response was also detected in healthy/uninfected donors (p ≤ 0.0006), probably due to previous expose to other coronaviruses. Robust expansion of CoV-2-STs was associated with a favorable outcome of infected patients and discharge, whereas patients who couldn’t mount an immune response against SARS-CoV-2 and did not expand CoV-2-STs failed to effectively control the infection and had a prolonged/complicated stay in ICU or succumbed (p ≤ 0.0002). A net increase >35 and >101 of IFN-γ- and TNF-α-secreting CoV-2-STs/5x105 PBMCs, respectively, at weeks 2-3 post-infection, was predictive of a favorable outcome (p ≤ 0.019). SARS-CoV-2-specific T-cell immunity in convalescents, tested at early (month 1-2) or later (month 8) time-points post-infection (n = 6), proved to be long-lasting. Towards the development of adoptive immunotherapy for COVID-19, we expanded and produced in vitro clinical doses of polyclonal CoV-2-STs expressing memory markers and low levels of exhaustion markers while presenting strong cytolytic activity against SARS-CoV-2 and a non-alloreactive safety profile. Convalescent-derived CoV-2-STs showed a broad pattern of clonotypes, with the 10 major clonotypes accounting for 30% of the overall TR repertoire of each product. Interestingly, 133 TR clonotypes were shared among cell products, 47 of which were highly shared (≥3/12 tested products). The majority of these identified clonotypes were context-specific, as revealed by extensive comparisons with sequences of known specificity in public databases, implicating potential association with the recognition of SARS-CoV-2 epitopes. CoV-2-STs generated by healthy/uninfected donors (although cross-reactive to SARS-CoV-2) were not cytolytic against CoV-2-pulsed PHA-blasts, indicating that they are unlikely to provide protection from severe COVID-19. Finally, we explored the feasibility of ex vivo generating an autologous T-cell immunotherapy product from a COVID-19 critical patient; despite its specificity, autologous-origin SARS-CoV-2-STs presented an exhausted profile hampering their proliferation.

Conclusions: Overall, we provide a predictive tool to identify high-risk patients and a robust, clinical scale, recovered donor-derived T-cell immunotherapy product.

Disclosure: Nothing to declare.

O044. Generation of Hspc-Derived CD16A-Engineered NK Cells With Advanced Antibody-Dependent Cellular Cytotoxicity

Paulien van Hauten¹, Paul de Jonge¹, Willemijn Hobo¹, Anniek van der Waart¹, Jianming Wu², Joop Jansen¹, Nicole Blijlevens¹, Bruce Walcheck², Nicolaas Schaap¹, Harry Dolstra¹

¹ Radboud University Medical Center Nijmegen, Nijmegen, Netherlands, ² University of Minnesota, Saint Paul, United States

Background: Adoptive cell therapy with allogeneic natural killer cells (NK cells) is a promising therapeutic option for hematologic malignancies as it is relatively safe, does not induce graft-versus-host disease (GvHD) and can exert an anti-tumor effect. This effect can be further potentiated by combining NK cell therapy with tumor-directing antibodies exerting antibody-dependent cell-mediated cytotoxicity (ADCC) that is facilitated by the Fc-receptor CD16a. Here, we investigated whether CD16a-engineering of CD34+ progenitor-derived NK cells would result in improved responses against NK-resistant cancers.

Methods: CD34+ hematopoietic stem and progenitor cells (HSPC), derived from umbilical cord blood or G-CSF mobilized blood, were transduced with high-affinity cleavable or non-cleavable CD16a receptor containing retroviral vectors. Subsequently, these HSPC were expanded and differentiated towards NK cells during a five week culture protocol using SR1, IL-15 and IL-12. During this culture period the HSPC-NK cell product was enriched for CD16a-transduced cells using flowcytometry-based sorting on the FACSAria (BD Biosciences). After five weeks, CD16a-transduced HSPC-NK cells were compared to non-transduced HSPC-NK cells for expansion and differentiation using multicolor flowcytometry. In addition, cells were stimulated with different stimuli evaluating the shedding capacity of the cleavable and non-cleavable CD16a variants on the HSPC-NK cells. Furthermore, 4-hour CD107a-based degranulation and IFNγ-production assays as well as overnight killing assays targeting various tumor cell lines were performed to evaluate ADCC of these CD16-HSPC-NK cell products.

Results: Viral transduction of CD34+ progenitor-derived NK cells results in stable transduction resulting in up to 69% for UCB-derived and 30% for G-CSF mobilized CD16a-expressing HSPC-NK cells at the end of the culture, increasing up to 91% and respectively 88% after enrichment. Expansion and differentiation into CD56-expressing NK cells was comparable to non-transduced cells. Stimulation of cleavable CD16a-transduced NK cells with multiple myeloma cell line L363 in combination with Elotuzumab resulted in a 18% decrease of CD16-expressing NK cells and stimulation with PMA with Ionomycin in a 86% decrease, whereas stimulation of non-cleavable CD16a-transduced NK cells showed no decrease in CD16 expression. Four hour stimulation of non-transduced NK cells with B lymphoma cell line Raji either in combination with IgG1 control versus Rituximab showed 1.2 fold increase in CD107a expressing NK cells and 1.1 fold increase in CD107a+IFNγ+ HSPC-NK cells. Interestingly, we observed a 2.2 fold increase in CD107a+IFNγ+ HSPC-NK cells for enriched cleavable CD16-transduced NK cells and even a 3.8 fold increase for enriched non-cleavable CD16-transduced NK cells.

In addition, overnight killing assays targeting L363 multiple myeloma cells with either IgG1 control or Elotuzumab markedly improved ADCC for both non-enriched and enriched CD16a-transduced HSPC-NK cells in contrast to no ADCC in non-CD16a-transduced HSPC-NK cells.

Conclusions: We can successfully transduce HSPC-derived NK cells with high-affinity CD16a receptors resulting in enhanced ADCC activity upon stimulation of these NK cells with lymphoma and multiple myeloma targets in combination with tumor-directing antibodies. This results in a highly interesting engineered HSPC-NK cell therapy with improved ADCC properties against NK-cell resistant malignancies. Future experiments need to ascertain the functional advantage or disadvantage of the shedding of CD16a upon activation for NK cell survival and serial killing.

Disclosure: Nothing to declare.

Conditioning Regimens

O045. Microbiota Injury is Conditioning Regimen-Dependent in Allogeneic Hematopoietic Cell Transplantation Recipients

Roni Shouval¹, Antonio Gomes¹, Chi Nguyen¹, Kate Ann Markey¹, Anqi Dai¹, Corrado Zuanelli Brambilla¹, Ying Taur¹, Mike Scordo¹, Sergio A. Giralt¹, Miguel-Angel Perales¹, Marcel R. M. van den Brink¹, Jonathan U. Peled¹

¹ Memorial Sloan Kettering Cancer Center, New York, United States

Background: Loss of intestinal microbial diversity and bacterial monodomination are associated with and potentially drive allogeneic hematopoietic cell transplantation (allo-HCT) complications. Although multivariate analyses of microbiome associations with clinical outcome are often adjusted for conditioning-intensity categories, individual regimens’ contribution to the severity of microbiome injury patterns is not known. We, therefore, aimed to characterize the relationship between specific conditioning regimens and the intestinal microbiota.

Methods: This retrospective single-center cohort included 1,190 adult recipients of allo-HCT following a conditioning regimen that at least 50 patients received (A). Conditioning intensity was classified according to standard guidelines. 7,697 stool samples whose microbial communities had been previously characterized by V4-V5 16S rRNA sequencing were evaluated.

Results: The median age was 57 years (IQR: 57 – 64), and acute myeloid leukemia (35%) followed by myelodysplastic syndrome (22%) and non-Hodgkin lymphoma were the leading transplant indications. The majority of patients received myeloablative (MA) conditioning regimens (68%). Diversity loss was greater in MA and reduced-intensity (RI) conditioning groups compared to the nonmyeloablative group (NMA). When considering the individual regimens, we observed considerable variation within the MA regimens (B). The decrease in diversity was consistent after excluding samples collected following exposure to empiric antibiotics for neutropenic fever or C. difficile diarrhea. In a pair-matched analysis (C) of the earliest samples collected before (days -20 to -1) and shortly after conditioning (days 0 to 10), the NMA regimen fludarabine-cyclophosphamide-TBI[200 cGy] (Flu/Cy/TBI200) was associated with the least reduction in diversity (34%), while the MA regimens clofarabine-thiotepa-melphalan and TBI[1375 cGy]-thiotepa-cyclophosphamide (TBI1375/Tt/Cy) were associated with the greatest reduction (69%-79%). Recipients of the RI regimen fludarabine-melphalan (Flu/Mel) had a similar diversity reduction as those receiving regimens with busulfan at a myeloablative dose. Finally, the incidence of monodomination by day 30 was over 90% with all regimens, defined here as the abundance of any genus ≥ 30%. However, the taxonomic composition of domination events differed by regimen, with striking blooms of Akkermansia in patients receiving Flu/Cy/TBI200 (D). Akkermansia abundance has been associated with various clinical phenotypes, including graft-versus-host disease, obesity, and tumor immunotherapy responses.

Conclusions: Microbiota injury following conditioning vary between regimens and is independent of empiric-antibiotic exposure. TBI dose may be an important mediator of injury, as regimens with low and high-dose TBI were at the opposite extremes of the microbiota injury scale. Finally, we observed a bloom of Akkermansia following Flu/Cy/TBI200.

Disclosure: MS: consultation fees/ research support/ advisory boars/ speaker - McKinsey & Company; Angiocrine Bioscience; Omeros; Kite; i3 Health. SAG: research funding/ scientific advisory board: Actinnum, Celgene, BMS, Sanofi, Amgen, Pfizer, Janssen, GSK, Jazz; MAP: honoraria, DSMB, scientific advisory board, research support, board of directors:Abbvie, Bellicum, Celgene, Bristol-Myers Squibb, Incyte, Kite/Gilead, Merck, Novartis, Nektar Therapeutics, Omeros,Takeda, Cidara Therapeutics, Servier, Medigene, MolMed, NexImmune, Incyte, Kite/Gilead, Miltenyi Biotec, NMDP, CIBMTR. MRMVDB: honoraria, IP, research funding, stock options, board, royalties: Seres,Jazz, Therakos, Amgen, Merck & Co, Inc, MagentaTherapeutics, WindMILTherapeutics, Rheos, Frazier Healthcare Partners, Nektar Therapeutics, Notch Therapeutics, Forty-Seven, Inc, Priothera, Wolters Kluwer, Juno Therapeutics, DKMS, Pharmacyclics (spouse in advisory board), Kite Pharmaceuticals (spouse in advisory board). JUP: Intellectual Property Rights, Research Funding, Travel fees, Consulting Fee: Seres, DaVolterra.

O046. Thiotepa Busulfan And Fludarabine (TBF) Patients With Acute Myeloid Leukemia in Complete Remission: A Retrospective Analysis on 221 Patients and 369 Controls

Federica Sora¹, Carmen Di Grazia², Anna Maria Raiola², Stefania Bregante², Sabrina Giammarco¹, Idanna Innocenti¹, Riccardo Varaldo², Elisabetta Metafuni¹, Francesca Gualandi², Francesco Autore³, Eugenio Galli¹, Patrizia Chiusolo¹, Luca Laurenti¹, Simona Sica¹, Andrea Bacigalupo¹, Emanuele Angelucci²

¹ Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy, ² UOC Ematologia e Trapianto di Midollo Osseo, IRCCS Ospedale Policlinico San Martino, Genova, Italy

Background: We have recently preported that a conditioning regimen including thiotepa, busulfan, fludarabine (TBF) reduces the risk of relapse in patients with AML in first and second remission (CR1/CR2), as compared to busulfan fludarabine (BUFLU) (Sora F et al. Biol Blood Marrow Transplant. 2020).

We hypothesized that TBF would be superior also to other conditioning regimens in terms of anti-leuykemic effect.

Methods: We therefore studied 590 AML patients who were prepared wither with TBF (n = 221) or other regimens (n = 369). The latter comprised full dose TBI (n = 190), or other nonTBI-nonTBF regimens (n = 179) (including BUCY, BUFLU, Thiotepa-FLU, FLU MEL). The diagnosis was de novo AML (n = 516) and secondary AMLor AML MRC (n = 74).

The clinical characteristics of the 3 groups (nonTBF-nonTBI, TBI, TBF) are as follows: median age 50, 37, 52 years; proportion in CR1 68%, 64%, 74% (p = 0.054). The proportion of patients grafted from an alternative donor was 45%, 57%, 86% (p < 0.0001).

Results: TBF vs TBI. The 5 year disease free survival (DFS) of TBF vs TBI patients was 70% vs 55% (p = 0.01)(Figure 1) and the cumulative incidence (CI) of relapse 13% vs 23% (p = 0.054).

TBF vs nonTBI-nonTBF. The 5 year DFS of TBI vs nonTBI-nonTBF was 70% vs 41% (p < 0.001) (Fig.1), and the Ci of relapse 13& vs 30% (p = 0.0005).

In a Cox analysis significant negative predictors of DFS were age >60 (RR1,9, p = 0.003), mismatched unrelated donor (RR 2.1, p = 0.0007), conditioning with TBI vs TBF (RR 1.8, p = 0.01) and conditioning with nonTBI-nonTBF (RR 2.9, p < 0.000001). In a Cox analysis on relapse the only negative predictor was a conditioning with TBI vs TBF (RR 1.6 p = 0.08 and nonTBI-nobnTBF vs TBF (RR 2.4, p = 0.002). Negative predictors of TRM were age >60 (RR 2.25, p = 0.01) and mismatched UD (RR 2.9, p = 0.0007).

Conclusions: Conclusion the combination of thiotepa, busulfan and fludarabine (TBF) improves disease free survival in CR1/CR2 AML patients, by reducing the risk of relapse, when compared to TBI regimens and nonTBI-nonTBF regimens. TRM is comparable in the 3 groups.

Disclosure: Conflict of interest “Nothing to declare”.

O047. Impact of The Addition of Antithymocyte Globulin to Post-Transplant Cyclophosphamide in Haploidentical Transplantation With Peripheral Blood Compared to Post-Transplant Cyclophosphamide Alone

Giorgia Battipaglia¹, Myriam Labopin^2,3,4, Didier Blaise⁵, Jose Luis Diez-Martin⁶, Ali Bazarbachi⁷, Antonin Vitek⁸, Patrice Chevallier⁹, Luca Castagna¹⁰, Giovanni Grillo¹¹, Eric Deconinck¹², Javier Lopez-Jimenez¹³, Yener Koc¹⁴, Annalisa Ruggeri¹⁵, Arnon Nagler^2,16, Mohamad Mohty^2,3,4

¹ Federico II University of Naples, Naples, Italy, ² EBMT Paris Study Office, Paris, France, ³ Hôpital Saint Antoine, Service d’Hématologie et Thérapie Cellulaire, Paris, France, ⁴ Sorbonne Universités, UPMC Univ Paris 06, INSERM, Centre de Recherche Saint-Antoine (CRSA), Paris, France, ⁵ Programme de Transplantation & Therapie Cellulaire, Institut Paoli-Calmettes, Aix Marseille Univ, CNRS, INSERM, CRCM, Marseille, France, ⁶ Hospital General Universitario Gregorio Marañón, Instituto de Investigación Sanitaria Gregorio Marañon, Madrid, Spain, ⁷ American University of Beirut, Beirut, Lebanon, ⁸ Institute of Hematology and Blood Transfusion, Prague, Czech Republic, ⁹ CHU, Nantes, Nantes, France, ¹⁰ Humanitas Clinical and Research Center, IRCCS, Rozzano, Milan, Italy, ¹¹ ASST Grande Ospedale Metropolitano Niguarda, Milan, Italy, ¹² Centre Hospitalier Regional Universitaire Besançon, INSERM UMR 1098, Université de Franche-Comté, Besançon, France, ¹³ Hospital Ramón y Cajal, Madrid, Spain, ¹⁴ Medicana International, Istanbul, Turkey, ¹⁵ Ospedale San Raffaele s.r.l., Milan, Italy, ¹⁶ Chaim Sheba Medical Center, Tel Hashomer, Israel

Background: Haploidentical transplantation (Haplo-HCT) with peripheral blood stem cells (PBSC) is increasingly used. Due to the higher risk of graft-versus-host disease (GVHD) in the presence of multiple HLA mismatches together with PBSC, addition of antithymocyte globulin (ATG) to post-transplant cyclophosphamide (PTCY) may hypothetically better prevent GVHD. We aimed to determine whether the addition of ATG to PTCY allows better outcomes than PTCY alone in Haplo-HCT with PBSC.

Methods: Included were 441 adult patients undergoing a first Haplo-HCT with PBSC for acute myeloid leukemia (AML) in first or second complete remission; GVHD prophylaxis contained either PTCY alone (n = 364) or PTCY+ATG (n = 67). Only cyclosporine A (CsA) and mycophenolate mofetil (MMF) as adjuvant immunosuppressors were included. Transplant period was 2011-2019.

Results: No major imbalances were observed between the two groups.Median age was 56 years and median year of Haplo-HCT was 2017 for both groups. Secondary AML was reported in 21% and 15% of PTCY and PTCY+ATG (p = 0.23). Poor performance status was observed in 21% in PTCY and 16% in PTCY+ATG (p = 0.35). Median interval from diagnosis to Haplo-HCT was 5 months in both groups. Karyotype was favorable, intermediate or unfavorable in 5% versus 3%, 54% versus 63%, 24% versus 25% in PTCY or PTCY+ATG, respectively. NPM1 and FLT3-ITD were mutated in 22% versus 19% (p = 0.61), 28% versus 16% (p = 0.82) in PTCY and PTCY+ATG, respectively. Total ATG dose was 2.5, 5 and 7.5 mg/kg in 21, 25 and 6 patients; information was missing in 15 cases. A female donor for a male recipient was used in 21% and 18% in PTCY and PTCY+ATG, (p = 0.58). Median donor age was 38 and 37 years for PTCY and PTCY+ATG, respectively (p = 0.98). A reduced-intensity conditioning regimen was more frequently used (57% and 61% in PTCY and PTCY+ATG, respectively; p = 0.54), mainly based on a thiotepa-busulfan-fludarabine schedule (44% and 63% in PTCY and PTCY+ATG). Median follow-up was 19 versus 15 months in PTCY and PTCY+ATG, respectively (p = 0.59). Neutrophil engraftment occurred in 97% and 98% of PTCY and PTCY+ATG, respectively. In univariate analysis, no statistical differences in transplant outcomes were observed between PTCY and PTCY+ATG (relapse incidence 25% versus 24%, p = 0.41; non-relapse mortality 22% versus 26%, p = 0.51; leukemia-free survival 53% versus 50%, p = 0.9; overall survival 58% for both, p = 0.89; GVHD-free/relapse-free survival 44%versus 43%, p = 0.71; grade II-IV acute GVHD 34% versus 30%, p = 0.58; chronic GVHD 33% versus 22%, p = 0.06). The main cause of death was infections in 34% and 35%, followed by death due to AML in 32% and 25% in PTCY and PTCY+ATG, respectively. In multivariate analysis no statistical differences were observed for the main transplant outcomes according to the GVHD prophylaxis regimen used except for a lower risk of chronic GVHD of all grades in patients receiving PTCY+ATG as compared to PTCY alone (HR 0.46, 95% CI 0.23-0.93; p = 0.03).

Conclusions: In Haplo-PB, addition of ATG to PTCY with CsA and MMF is feasible, better at preventing chronic GVHD and has comparable survival and transplant outcomes to PTCY alone, without increasing transplant toxicity or mortality and without increasing relapse incidence.

Clinical Trial Registry: N/A

Disclosure: No conflict of interest to disclose.

O048. Hematopoietic Cell Transplantation Comorbidity Index (HCT-CI) And Outcome After Non-Myeloablative Allogeneic Stem Cell Transplantation in Acute Myeloid Leukemia (AML) Patients Older Than 60 Years

Donata Backhaus¹, Madlen Jentzsch¹, Dominic Brauer¹, Julia Schulz¹, Dietger Niederwieser¹, Uwe Platzbecker¹, Sebastian Schwind¹

¹ Universitätsklinikum Leipzig, Leipzig, Germany

Background: Hematopoietic stem cell transplantation (HSCT) offers the highest chance of long-term cure for most acute myeloid leukemia (AML) patients. Age and comorbidities may restrict this option due to toxic intensive preparation regimens. For these patients non-myeloablative conditioning (NMA) was developed to allow HSCT. The HSCT comorbidity index (HCT-CI) predicts outcome after HSCT in unselected patients, but there is no separate evaluation of its prognostic significance in older (>60 years) AML patients undergoing NMA-HSCT. Here we analyzed a large, homogenously treated AML patient cohort that received NMA-HSCT for the prognostic impact of the HCT-CI.

Methods: We retrospectively analyzed 289 AML patients older than 60 years subjected to NMA-HSCT. The conditioning regimen consisted of fludarabine 30 mg/m2 for 3 consecutive days and 2 (n = 278) or 3 Gy (n = 6) total body irradiation (TBI) while 5 patients received 2Gy TBI alone. Graft versus host disease prophylaxis contained cyclosporine and mycophenolate mofetil. AML disease risk was assessed according to European LeukemiaNet (ELN) 2017 classification. Measurable residual disease (MRD) at HSCT was assessed based on NPM1 mutations and BAALC, MN1, and WT1 expression. We calculated the pretransplant HCT-CI to stratify outcomes. Median follow up after HSCT was 3.8 years.

Results: 36% of patients had a low risk HCT-CI score (0 points), 31% an intermediate risk (1-2 points), and 33% a high risk (≥3 points). The nonrelapse mortality (NRM) did not differ significantly between HCT-CI risk groups after NMA-HSCT (P = 0.56, Figure 1A, at 5 years HCT-CI low 24% vs. HCT-CI intermediate 20% vs. HCT-CI high 27%, respectively). Likewise, cumulative incidence of relapse (P = 0.88, Figure 1B, at 5 years 46% vs. 45% vs. 43%) and overall survival (P = 0.70, Figure 1C, at 5 years 40% vs. 44% vs. 41%) did not differ between HCT-CI risk groups. The HCT-CI also did not impact outcomes in separate analyses according to ELN2017 risk at diagnosis (OS, P = 0.20, P = 0.30, and P = 0.70 for favorable, intermediate, and adverse, respectively) or MRD status prior to HSCT (OS, P = 1 and P = 0.30 for MRD-negative and MRD-positive patients, respectively). In the favorable risk subgroup of MRD-negative patients prior to NMA-HSCT, the median overall survival reached 49% at 5 years after NMA-HSCT, irrespective of the HCT-CI.

Conclusions: In this older AML patient cohort NRM following NMA-HSCT was relatively low. A higher HCT-CI did not have a negative impact on NRM or survival. Our data indicates that comorbidities per se reflected by a higher HCT-CI should not impede NMA-HSCT. Independently from the HCT-CI score MRD-negative patients notably showed good survival of 49 % at 5 years after NMA-HSCT. These data aid in informed clinical decisions towards HSCT in older AML patients. HSCT with this reduced toxicity regimen represents a feasible treatment option in older and comorbid AML patients.

Disclosure: Nothing to declare.

O049. Sequential Flamsa-Bu Versus FLUBU2 in Patients With Active Relapsed or Refractory Acute Myeloid Leukemia: A Study From The Acute Leukemia Working Party of The EBMT

Eduardo Rodríguez-Arbolí¹, Myriam Labopin², Matthias Eder³, Arne Brecht⁴, Igor Wolfgang Blau⁵, Anne Huynh⁶, Noël Milpied⁷, Johanna Tischer⁸, Wolfgang Bethge⁹, Sergey Bondarenko¹⁰, Mareike Verbeek¹¹, Claude Eric Bulabois¹², Hermann Einsele¹³, Martin Bornhäuser¹⁴, Bipin Savani¹⁵, Alexandros Spyridonidis¹⁶, Ali Bazarbachi¹⁷, Sebastian Giebel¹⁸, Eolia Brissot¹⁹, Arnon Nagler²⁰, Mohamad Mohty¹⁹

¹ Hospital Universitario Virgen del Rocío, Instituto de Biomedicina de Sevilla (IBIS/CSIC/CIBERONC), University of Seville, Seville, Spain, ² Hôpital Saint-Antoine, AP-HP, Sorbonne University, INSERM, Service d’Hématologie Clinique et Thérapie Cellulaire UMRS 938, Paris, France, ³ Hannover Medical School, Hannover, Germany, ⁴ Deutsche Klinik für Diagnostik, KMT Zentrum, Wiesbaden, Germany, ⁵ Medizinische Klinik m. S. Hämatologie, Onkologie und Tumorimmunologie, Charité Universitätsmedizin Berlin, Berlin, Germany, ⁶ CHU Institut Universitaire du Cancer Toulouse, Oncopole, Toulouse, France, ⁷ CHU Bordeaux, Hôpital Haut-Lévêque, Pessac, France, ⁸ University Hospital of Munich-Grosshadern, LMU, Munich, Germany, ⁹ Universität Tübingen, Medizinische Klinik, Tübingen, Germany, ¹⁰ RM Gorbacheva Research Institute, Pavlov University, Saint Petersburg, Russian Federation, ¹¹ Klinikum Rechts der Isar, III Med Klinik der TU, Munich, Germany, ¹² CHU Grenoble Alpes, Université Grenoble Alpes, Service d’Hématologie, CS 10217, Grenoble, France, ¹³ Universitätsklinikum Würzburg, Med. Klinik und Poliklinik II, Würzburg, Germany, ¹⁴ Universitätsklinikum Dresden, Dresden, Germany, ¹⁵ Vanderbilt University Medical Center, United States, ¹⁶ University Hospital of Patras, Patras, Greece, ¹⁷ American University of Beirut, Beirut, Lebanon, ¹⁸ Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice Branch, Gliwice, Poland, ¹⁹ Hôpital Saint-Antoine, AP-HP, Sorbonne University, INSERM UMRS 938, Paris, France, ²⁰ Chaim Sheba Medical Center, Tel Hashomer, Israel

Background: Optimal conditioning choice in patients with active relapsed/refractory (R/R) acute myeloid leukemia (AML) remains unsettled. FLAMSA chemotherapy followed by busulfan-based reduced-intensity conditioning (RIC) (FLAMSA-Bu) has shown promising results and is widely used in this setting. However, comparative outcome data for FLAMSA-Bu versus standard RIC regimens are lacking. In the present study, we analysed outcomes after FLAMSA-Bu or FluBu2 in a large cohort of patients with R/R AML who underwent allogeneic hematopoietic stem cell transplantation (alloSCT) with active disease.

Methods: This was a registry analysis performed by the Acute Leukemia Working Party (ALWP) of the European Society for Blood and Marrow Transplantation (EBMT). Patient eligibility criteria included the diagnosis of primary refractory or first or second relapsed AML with active disease at the time of transplant, first alloSCT from a matched sibling donor (MSD) or a 10/10 matched unrelated donor (MUD) performed between 2005 and 2019, and FLAMSA-Bu or FluBu2 conditioning (Bu oral dosing at 8 mg/kg or IV at 6.4 mg/kg). The study endpoints were overall survival (OS), relapse incidence (RI), leukemia-free survival (LFS), non-relapse mortality (NRM), acute and chronic graft-versus-host disease (aGVHD and cGVHD), and refined GVHD, relapse-free survival (GRFS). Cox proportional hazards regression models were constructed for all outcome measures. Cause-specific models were used to account for competing risks.

Results: A total of 476 patients fulfilled the inclusion criteria. Median follow-up was 41 months (IQR 16-71). Median age at the time of alloSCT was 60 years (IQR 51-65), and the predominant donor type was 10/10 MUD (62%). Two-hundred fifty-seven patients received conditioning with FluBu2 and 219 with sequential FLAMSA-Bu. Seventy-five percent of patients were administered in vivo T cell depletion after FluBu2 and 95% after FLAMSA-Bu (p < 0.001). A higher proportion of patients harbored adverse risk cytogenetics in the FLAMSA-Bu group (38% vs 25%, p = 0.04). Other relevant covariates were well-balanced between conditioning arms. The underlying reasons for the selection of FLAMSA-Bu versus FluBu2 were unknown.

A complete remission (CR) after alloSCT was achieved by 72% and 75% of patients in the FluBu2 and FLAMSA-Bu groups, respectively. In the univariate analysis, 2-year NRM (21% [95% CI: 17-25]), GRFS (24% [95% CI: 20-29%]) and cGVHD (29% [95% CI: 25-34%]) were similar in both cohorts. On the other hand, FLAMSA-Bu was associated with lower 2-year RI (38% vs 49% after FluBu2, p = 0.004), as well as with increased LFS (42% vs 29%, p = 0.001), OS (47% vs 39%, p = 0.008), and grades II-IV aGVHD (36% vs 20%, p = 0.001). In the multivariate analysis, FLAMSA-Bu remained associated with lower RI (HR 0.69, p = 0.042), increased LFS (HR 0.74, p = 0.048) and a higher risk of aGVHD (HR 2.06, p = 0.005). Yet, no statistical association remained between conditioning regimen choice and OS (HR 0.81, p = 0.18). There was no significant interaction between conditioning and age.

Conclusions: Sequential FLAMSA-Bu was associated with lower RI and higher LFS as compared to FluBu2 in patients with active R/R AML. There were no significant differences in terms of NRM between conditioning regimens. Both strategies achieved CR rates above 70%, resulting in 2-year OS over 30%.

Disclosure: Nothing to declare.

O050. Bendamustine-Based Conditioning Prior to Autologous Hematopoietic Cell Transplant Improves Outcomes in Patients With Elapsed-Refractory Non-Hodgkin Lymphoma

Sylvie Lachance¹, Alexandre Bourguignon², Josie-Anne Boisjoly³, Imran Ahmad¹, Jean-Sébastien Delisle⁴, Nadia Bambace¹, Léa Bernard¹, Sandra Cohen¹, Thomas Kiss¹, Philippe Bouchard⁵, Guy Sauvageau⁶, Denis-Claude Roy¹, Justine Zehr⁷, Miguel Chagnon⁷, Jean Roy¹

¹ Hôpital Maisonneuve-Rosemont, Université de Montréal, Montreal, Canada, ² Université de Montréal, Montréal, Canada, ³ Université de Sherbrooke, Sherbrooke, Canada, ⁴ Hôpital Maisonneuve-Rosemont, Université de Montréal, Montréal, Canada, ⁵ Hôpital Maisonneuve-Rosemont, Montreal, Canada, ⁶ Institut de Recherche en Immunologie et Cancer, Université de Montréal, Montreal, Canada, ⁷ Université de Montréal, Montreal, Canada

Background: Carmustine (BCNU) based conditioning is regarded as a standard prior to autologous hematopoietic cell transplantation (aHCT) for treatment of relapsed/refractory non-Hodgkin lymphomas (r/R-NHL). After carmustine drug shortage in Canada, followed by a drastic price increase, we decided in May 2015, to substitute BCNU for bendamustine (Benda) in the preparative regimen prior to aHCT after initial report of its safety and efficacy. We elected to study the outcomes of patients conditioned with Benda-EAM and compare their results to our BCNU standard (BEAM/BEAC) as part of a quality control project.

Objective: The objectives of this retrospective study was to confirm the safety and efficacy of Benda-EAM conditioning prior to aHCT for relapsed/refractory lymphomas.

Methods: After institutional ethic board approval, the chart of 131 consecutive patients who underwent aHCT with Benda-EAM conditioning, from May 2015 to May 2018, were reviewed and compared to a historical cohort of 96 consecutive patients treated with BCNU-containing preparative regimen between January 2012 and May 2015. Except for conditioning, indications for aHCT, eligibility criteria and supportive care measures were identical for the 2 groups according to SOPs of our FACT accredited transplant program. Baseline characteristics, toxicities and post-transplant outcomes were analyzed.

Results: Clinical characteristics of patients receiving BCNU (n = 96) and bendamustine conditioning (n = 131). Both groups were well matched except for an older age in Bendamustine group. Benda-EAM conditioning was found to be more toxic. Development of mucositis, acute renal failure and ICU admission were more common, leading to a longer hospital stay. However, no toxic death associated with Benda-EAM was reported. With a median follow-up of 36 and 64 months for bendamustine and BCNU groups respectively, the disease-free and overall survival at 48 months are 71% and 86% following Benda-EAM and 61% and 73% for BCNU (p = 0.006).

Conclusions: The substitution of carmustine to bendamustine in the conditioning regimen prior to aHCT improves outcomes in patients with r\R-NHL and should be considered a new standard.

Disclosure: The authors declare no conflict of interest related to this work.

O051. The Impact of Pulmonary Function in Patients Undergoing Autologous Stem Cell Transplantation

Jesus Duque-Afonso¹, Sophie Ewald¹, Gabriele Ihorst¹, Miguel Waterhouse¹, Robert Zeiser¹, Ralph Wäsch¹, Hartmut Bertz¹, Joachim Müller-Quernheim¹, Jürgen Finke¹, Reinhard Marks¹, Monika Engelhardt¹

¹ University of Freiburg Medical Center, Freiburg, Germany

Background: High dose chemotherapy followed by autologous hematopoietic stem cell transplantation (auto-HSCT) is an established therapy for patients with hematological malignancies. The age of patients undergoing auto-HSCT, and therefore the co-morbidities, have increased during the last decades. However, the assessment of organ dysfunction prior auto-HSCT has not been well characterized in order to predict complications and outcome after auto-HSCT. Therefore, we analyzed retrospectively the association of clinical factors, lung and cardiac function parameters and early complications with outcome of patients undergoing auto-HSCT after conditioning with BEAM (BCNU/carmustin, etoposide, cytarabine, melphalan) and high-dose melphalan.

Methods: In this study, we included 754 patients treated with auto-HSCT (334 patients conditioned with BEAM and 425 patients conditioned with high-dose melphalan) at our institution between 2007 and 2017. The median follow up for patients conditioned with BEAM was 52 months (range: 0.2-152) and with high-dose melphalan was 46 months (range: 0.5-149). The median age of patients was 56 years (range: 19-78) and 61 years (range: 29-80) conditioned with BEAM and high-dose melphalan, respectively.

Statistical analysis of patient characteristics, pulmonary and cardiac function tests, cumulative incidences and hazard ratios was conducted using STATA v11.0 (College Station, Texas, USA). The Cox proportional hazards regression model was used to estimate hazard ratios (HR) and confidence intervals (CI) for OS and PFS. We applied the Fine and Gray model to compare cumulative incidence rates in the presence of competing risks and presented subdistribution hazard ratios (SHR) for relapse incidence and non-relapse mortality (NRM). Multivariate analyses for OS were conducted using Cox proportional hazards regression model.

Results: Several clinical factors, lung and cardiac function parameters were associated with decreased OS in patients treated with BEAM and high-dose melphalan prior auto-HSCT in the univariate analysis using a Cox regression model. In the multivariate analysis, we identified progressive disease status, CO diffusion capacity corrected for hemoglobin (DLCOcSB)<60% of predicted, Karnofsky performance status ≤80%, cardiac diseases prior auto-HSCT and age >70 years to be associated with decreased OS in patients treated with BEAM. Similarly, we identified age >60 years, DLCOcSB<80% of predicted, maximal expiratory flow at 25% (MEF25)<20% of predicted and lung diseases prior auto-HSCT to be associated with higher mortality in patients treated with high-dose melphalan prior auto-HSCT. Of note, DLCOcSB<60% of predicted was also associated with decreased OS in univariate analysis in this patient population. Abnormalities in above mentioned pulmonary function tests were associated with increased NRM and pulmonary causes of death in both groups, patients treated with BEAM and high-dose melphalan.

Conclusions: In summary, we have identified several pulmonary function abnormalities as risk factors associated with decreased OS and increased NRM in patients conditioned with BEAM and high-dose melphalan prior auto-HSCT. These findings may lead to important preventive and early therapeutic interventions in patients undergoing auto-HSCT in order to decrease morbidity and mortality further.

Disclosure: Nothing to declare.

Experimental Stem Cell Transplantation

O052. Mutational Consequences of Hematopoietic Stem Cell Transplantation in Humans

Mirjam Belderbos¹, Jurrian de Kanter^2,1, Flavia Peci^2,1, Eline J.M. Bertrums^2,1, Axel Rosendahl Huber^2,1, Markus J. van Roosmalen^2,1, Freek Manders^2,1, Mark Verheul^2,1, Rurika Oka^2,1, Arianne M. Brandsma^2,1, Marc Bierings¹, Ruben van Boxtel^2,1

¹ Princess Maxima Center for Pediatric Oncology, Utrecht, Netherlands, ² Oncode Institute, Utrecht, Netherlands

Background: Genetic instability is a major concern for the successful application of stem cells in regenerative medicine. However, the mutational consequences of the most commonly applied stem cell therapy in humans, hematopoietic stem cell transplantation (HSCT) remain unknown. Here, we aimed to systematically assess the mutational consequences of HSCT in human recipients, using whole genome sequencing of single hematopoietic stem and progenitor cells (HSPCs) prior and after transplantation.

Methods: We performed whole genome sequencing (WGS) of single hematopoietic cell cultures of human HSCT recipients and their donors, to catalogue all the mutations that were present in the clonally expanded parental HSPCs. We included six pediatric HSCT recipients, who were transplanted with either bone marrow cells of an HLA-identical sibling donor (n = 3) or with an anonymous umbilical cord blood (UCB) donor (n = 3). All recipients were transplanted for hematologic malignancy, after chemotherapy-based myeloablative conditioning. HSPCs were isolated from residual donor graft material at the time of HSCT and from blood of the HSCT recipients, collected up to 40 months after HSCT. Of each time point, we selected 3-8 clones for WGS. We compared the number and type of mutations in these cells to HSPCs obtained from healthy donors, with ages ranging across the entire human lifespan.

Results: In total, we identified 6338 base substitutions and 377 indels in 32 assessed HSPCs. In the majority of recipients HSPCs, the number of base substitutions was within the predicted range of normal hematologic aging (ratio observed/expected 0.6-1.2). However, we identified multiple HSPCs isolated from two HSCT recipients, in which the number of mutations was up to five fold higher than predicted based on their age (mean observed/expected 3.4, range 1.6-5.0). Remarkably, this increase could be attributed exclusively to a novel mutational process signature, SBStrans. SBStrans is characterized by C>A transversions at CpA dinucleotides, Watson-versus-Crick strand lesion segregation and a strong replication strand bias. These features indicate that the causative process interferes with DNA replication during a short period of time, likely only one cell division. Using a machine-learning approach, we detected SBStrans in tumor genomes of two therapy-related leukemias that occurred after HSCT and in two solid tumors of patients who underwent kidney transplantation earlier in life.

Conclusions: Our study demonstrates that HSCT in human recipients can increase mutation accumulation in the transplanted HSPCs, which may contribute to carcinogenesis. Our study emphasizes the clinical relevance of stem-cell therapy associated mutagenesis in humans, and urges for careful surveillance of HSCT recipients, in order to detect and prevent long-term morbidity.

Clinical Trial Registry: NL7585; www.trialregister.nl.

Disclosure: Nothing to declare.

O053. Over-Expression of HIF-1Α In Mesenchymal Stromal Cells Increases Human CD34+ Cells Clonogenic Capacity in Vitro And Engraftment Ability in A Xenotransplantation Model

Silvia Preciado¹, Salomé Sirerol², Sandra Muntión¹, Gerardo J Martí-Chillón¹, Lika Osugui¹, Pilar Sepúlveda³, Fermín Sánchez-Guijo¹

¹ Hospital Universitario de Salamanca, Salamanca, Spain, ² Universidad de Valencia, Valencia, Spain, ³ Instituto de Investigación Sanitaria La Fe, Valencia, Spain

Background: Graft failure or poor graft function and their consequences are important complications after allogeneic stem cell transplantation (allo-SCT). The administration of mesenchymal stromal cells (MSC) has been shown to increase the engraftment ability and the hematopoietic function in preclinical models of xenotransplantation. Since it has been shown that hypoxia and, more specifically, overexpression of hypoxia-inducible factor-1 (HIF-1) by genetic engineering potentiates the therapeutic properties of MSCs, in the current study we wanted to analyze if the co-transplantation of human MSC overexpressing HIF-1α increases the engraftment ability of human cord blood CD34+ cells in a standard xenotransplantation murine model compared to wildtype MSC.

Methods: Human MSC for this study were obtained from dental pulp. Lentiviral overexpression of HIF-1α was performed transducing cells with pWPI-green fluorescent protein (GFP) (wt-MSC) or pWPI-HIF-1α-GFP (HIF-MSC) expression vectors. CD34+ cells were isolated from human cord blood. For this purpose, mononuclear cells were isolated by Ficoll-Paque density gradient centrifugation and then CD34+ cells were purified by immunomagnetic sorting in an AutoMACS. 2x105 CD34+ cells were co-cultured with 5x104 wt-MSC or HIF-MSC (4:1) for 72 hours. Then CD34+ cells were recovered, and subsequent experiments were performed as follows. Viability (Annexin V and 7AAD), some molecules involved in engraftment (CD44, CXCR4, CD34, ITGA4, cKIT) and ROS expression were evaluated by flow cytometry. In addition, CD34+ cells clonogenic capacity was also assessed. For this purpose, 2.5x103 CD34+ cells recovered from each experimental condition were cultured in MACSMedia Stem MACS HSC-CFU complete without human Epo. After 14 days, CFU were scored. Finally, in vivo engraftment ability was analyze after the injection of 5x105 MSC via intrafemorally in the right femur of NOD-SCID mice undergoing 3.5Gy of total body irradiation. Four hours later, 1x105 CD34+ cells were injected intravenously. Three experimental groups were established: 1) CD34 cells alone, 2) CD34+cells + wt-MSC and 3) CD34+cells + HIF-MSC. Four weeks after the xenotransplantation, human hematopoietic engraftment was determined by flow cytometry in both femurs and spleen.

Results: Transduction efficiency was always higher than 95%. A significant increase in CD34, CXCR4 and ITGA4 expression (p = 0.019; p = 0.003; p = 0.019, respectively) was observed in CD34+ cells co-cultured with HIF-MSC in comparison with wt- MSC. We did not observe significant differences in their viability and ROS expression. In addition, CD34+ cells cultured with HIF-MSC displayed a higher CFU-GM clonogenic potential (241 [137-311] CFU) than those cultured with wt-MSC (210 [94-267] CFU). Finally, regarding their engraftment ability in NOD-SCID mice, we observed a significant increase in CD34+ cells engraftment ability when they were co-transplanted with HIF-MSC (27.83%±5.78) in comparison with CD34+ co-transplanted with wt-MSC (13.74±5.51) (p = 0.016) o alone (7.23±2.51) (p = 0.015) both in right and left femur (p = 0.024 and p = 0.008 respectively). No differences were found in spleen among groups.

Conclusions: HIF-1α overexpression significantly enhances MSC therapeutic effect on increasing CD34, CXCR4 and ITGA4 expression on CD34+ cells, increasing their clonogenic capacity and improving their engraftment ability in a xenotransplantation model.

Disclosure: Nothing to declare.

Gene Therapy

O054. Interim Results of Betibeglogene Autotemcel Gene Therapy in Pediatric Patients With Transfusion-dependent Β-Thalassemia (TDT) Treated in The Phase 3 Northstar-2 (HGB-207) And Northstar-3 (HGB-212) Studies

Isabelle Thuret¹, Alexis Thompson², Janet Kwiatkowski³, John Porter⁴, Suradej Hongeng⁵, Evangelia Yannaki⁶, Andreas Kulozik⁷, Martin Sauer⁸, Adrian Thrasher⁹, Ashutosh Lal¹⁰, Ruiting Guo¹¹, Weijian Liu¹¹, Richard Colvin¹¹, Mark Walters¹⁰, Franco Locatelli¹²

¹ Hôpital de la Timone, Marseille, France, ² Ann & Robert H. Lurie Children’s Hospital of Chicago and Northwestern University Feinberg School of Medicine, Chicago, United States, ³ Children’s Hospital of Philadelphia and Perelman School of Medicine of the University of Pennsylvania, Philadelphia, United States, ⁴ University College London Hospitals, London, United Kingdom, ⁵ Mahidol University, Ramathibodi Hospital, Bangkok, Thailand, ⁶ G. Papanikolaou Hospital, Thessaloniki, Greece, ⁷ University of Heidelberg, Heidelberg, Germany, ⁸ Medizinische Hochschule Hannover, Hannover, Germany, ⁹ University College London Great Ormond Street Hospital Institute of Child Health and Great Ormond Street Hospital NHS Trust, London, United Kingdom, ¹⁰ UCSF Benioff Children’s Hospital, Oakland, United States, ¹¹ bluebird bio, Inc., Cambridge, United States, ¹² IRCCS Bambino Gesù Children’s Hospital, Rome, Italy

Background: Phase 3 studies evaluating betibeglogene autotemcel (beti-cel; LentiGlobin for β-thalassemia) gene therapy in patients with TDT, HGB-207 (non-β⁰/β⁰ genotypes; NCT02906202) and HGB-212 (β⁰/β⁰, β⁰/β^+IVS-I-110, β^+IVS-I-110/β^+IVS-I-110 genotypes; NCT03207009), demonstrated positive outcomes in adults. Therefore, the studies were expanded to include patients <18 years of age. We report interim results in these pediatric patients.

Methods: Autologous CD34+ cells were transduced ex vivo with BB305 lentiviral vector to produce beti-cel. Patients underwent pharmacokinetic-adjusted busulfan myeloablation and beti-cel infusion. Transfusion independence (TI; weighted average hemoglobin [Hb] ≥9 g/dL without packed red blood cell transfusions for ≥12 months) was the HGB-207 primary endpoint and an HGB-212 secondary endpoint (primary endpoint: transfusion reduction). Data are presented as median (min–max).

Results: As of 3 March 2020, 24 pediatric patients were treated and followed for 15.5 (1.1–29.5) months, including 13 patients <12 years (HGB-207: n = 8; HGB-212: n = 5) and 11 patients ≥12‒<18 years at assent (HGB-207: n = 6; HGB-212: n = 5). The youngest patients were two 4-year-olds. TI was achieved in 4/6 evaluable patients <12 years and 9/9 evaluable patients ≥12–<18 years. The median duration of ongoing TI was 14.9 (12.2–21.6) months. Pediatric patients achieved TI at a similar rate as adults (87% [13/15] vs 83% [10/12]).

Weighted average Hb during TI in patients <12 years and ≥12–<18 years was 10.1 (9.4‒10.3) g/dL and 11.4 (9.5‒12.8) g/dL, respectively. At last visit, gene therapy-derived HbA^T87Q levels in these patients were 6.8 (5.1–9.2) g/dL (n = 4) and 9.3 (8.0–10.9) g/dL (n = 9), respectively. In patients ≥18 years who achieved TI, weighted average Hb was 12.7 (11.5 – 13.4) g/dL and HbA^T87Q at last visit was 10.1 (7.3 – 12.6) g/dL (n = 10).

Non-hematologic ≥ Grade 3 adverse events in ≥3 patients <18 years were stomatitis (n = 14), febrile neutropenia (n = 12), decreased appetite (n = 5), epistaxis (n = 4), increased alanine aminotransferase (n = 3), hypoxia (n = 3), and pyrexia (n = 3). Grade 4 veno-occlusive liver disease occurred in 2 patients ≥12–<18 years and one grade 2 event occurred in a patient <12 years; all events resolved with defibrotide. No replication-competent lentivirus was reported. All patients had polyclonal vector integration; no integration site contributed >3% of all integration sites at last assessment.

Conclusions: Pediatric patients with diverse TDT genotypes achieved TI rates comparable to adults after beti-cel gene therapy, suggesting that beti-cel is a viable treatment option for patients with TDT across ages. The treatment regimen had a safety profile consistent with busulfan myeloablation.

Table 1. Patient, drug product, engraftment characteristics (median [min‒max])

	<12 years (N = 13)	≥12‒<18 years (N = 11)
Age, years	8 (4‒11)	15 (12‒17)
Genotype
β0/β+	8 (62)	3 (27)
β0/β0	3 (23)	4 (36)
β+/β+	1 (8)	4 (36)
βE/β0	1 (8)	0
Liver iron concentration, mg Fe/g dw	3.8 (1.2‒12.7)	5.6 (1.0‒13.2)
Drug product
Vector copies/diploid genome	2.6 (1.9‒4.7)	3.3 (1.2‒6.0)
Transduced cells, %	70 (34‒85)	82 (34‒90)
Cell dose, x 106 cells/kg	9.9 (6.1‒19.9)	7.4 (5.0‒19.4)
Neutrophil engraftment	26 (17‒36)	26 (16‒38)
Platelet engraftment	50 (20‒94)*	50.5 (25‒84)†

1. *2 patients with <3 months and ^†1 patient with 1-month follow-up had not yet achieved platelet engraftment.

Clinical Trial Registry: NCT02906202; https://clinicaltrials.gov/ct2/show/NCT02906202

NCT03207009; https://clinicaltrials.gov/ct2/show/NCT03207009

Disclosure: Thuret, Isabelle: investigator for clinical trials, participation on scientific/medical advisory board (bluebird bio, Inc., Novartis, Celgene), consultancy (Apopharma). Thompson, Alexis: consultant (bluebird bio, Inc.), consultant, research funding (Celgene, Novartis) research funding (Baxalta, Biomarin). Kwiatkowski, Janet: consultancy (bluebird bio, Inc., Agios, Celgene, Imara), research funding (bluebird bio, Inc., Apopharma, Novartis, Terumo Corporation, Sangamo). Porter, John: honoraria, consultancy (Celgene, Agios, bluebird bio, Inc.), honoraria (Protagonism, Vifor, La Jolla Pharmaceuticals, Silence Therapeutics). Hongeng, Suradej; Yannaki, Evangelia; Sauer, Martin: none. Kulozik, Andreas: consultancy, honoraria, membership on an entity’s Board of Directors or advisory committees (bluebird bio, Inc., Novartis). Thrasher, Adrian: co-founder and consultant (Orchard Therapeutics), consultant (Rocket Pharmaceuticals, Generation bio, bluebird bio, 4Bio Capital Partners, and Sana Biotechnology). Lal, Ashutosh: research funding (bluebird bio, Inc., Insight Magnetics, La Jolla Pharmaceutical Company, Novartis, Terumo Corporation), membership on an entity’s Board of Directors or advisory committees and research funding (Protagonist Therapeutics, Celgene).

Guo, Ruiting; Liu, Weijian; Colvin, Richard: employee, ownership interest, and salary (bluebird bio, Inc.). Walters, Mark: consultancy (Editas Medicine, AllCells, Veevo Medicine). Locatelli, Franco: honoraria, membership on board of directors or advisory committees (Amgen); consultancy, membership on board of directors or advisory committees (Bellicum Pharmaceuticals, Novartis); consultancy (bluebird bio, Inc.); honoraria (Miltenyi Biotec).

O055. Hematopoietic Stem Cell Gene Therapy For Mucopolysaccharidosis Type I, Hurler (MPS-IH): Biological Efficacy and Preliminary Evidence of Early Clinical Response

Maria Ester Bernardo¹, Bernhard Gentner¹, Francesca Tucci¹, Francesca Fumagalli¹, Francesca Ciotti¹, Marina Sarzana¹, Silvia Pontesilli¹, Cristina Baldoli¹, Maurizio De Pellegrin¹, Paolo Silvani¹, Silvia Darin¹, Giulia Consiglieri¹, Chiara Filisetti¹, Simona Miglietta¹, Erika Zonari¹, Rossella Parini², Attilio Rovelli², Giancarlo La Marca³, Fabio Ciceri¹, Luigi Naldini¹, Alessandro Aiuti¹

¹ San Raffaele Scientific Institute, Milano, Italy, ² San Gerardo Hospital, Monza, Italy, ³ Meyer Children’s University Hospital, Firenze, Italy

Background: Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is the current standard of care for patients with mucopolysaccharidosis type I Hurler (MPSIH). Despite good transplant outcomes in the short-term, allo-HSCT efficacy remains limited at the level of some organs, e.g. skeleton and central nervous system (CNS), severely affecting patients’ quality of life.

Methods: We are conducting a first-in-human phase I/II clinical trial (NCT03488394) to test whether transplantation of autologous hematopoietic stem and progenitor cells (HSPCs) transduced ex-vivo with a lentiviral vector coding for iduronidase (IDUA) protein is safe and efficacious in restoring enzymatic activity up to supraphysiologic levels in MPSIH patients’ blood and tissues. We have enrolled 8 patients (6 M, 2 F; median age at treatment=24 months) who lacked a non-heterozygous HLA-matched sibling or cord blood donor and displayed an IQ/DQ>70. Seven out of 8 patients received enzyme replacement therapy before gene therapy (GT), 5 of whom had a positive anti-IDUA IgG titer at baseline. A high number of autologous HSPCs were collected from all patients by leukapheresis following mobilization with lenograstim and plerixafor. The primary efficacy endpoint is IDUA activity in dried blood spot up to supraphysiologic levels at 1 year post-GT. Treatment impact on CNS and bone is assessed by measurement of IDUA and glycosaminoglycans (GAGs) in cerebrospinal fluid (CSF), DQ/IQ, range of motion, growth velocity and multiple radiologic parameters.

Results: Following busulfan/fludarabine/rituximab-based myeloablative conditioning, patients received autologous HSPCs transduced ex-vivo with a 36h transduction protocol with prostaglandin E2 as transduction enhancer (median dose=20.7 CD34+/Kg, median vector copy Number=2, median transduction efficiency >80%). All patients showed prompt and sustained engraftment of gene corrected cells with median neutrophil reconstitution occurring on day+20. Platelets did not fall <20,000/mm3 in 4/8 patients. In the other 4 patients, platelets recovered in a median of 13.5 days (range 13-18 days). Adverse events were generally mild and compatible with myeloablative conditioning; 16 Severe Adverse Events have been reported up to December 2020. Gene-marked cells appeared in bone marrow and peripheral blood in all patients at day+30 post-GT and were accompanied by restoration of early supraphysiologic IDUA activity in blood, which was maintained up to last follow-up of 24 months post-GT (median follow-up: 12 months). We observed a rapid decrease in urinary GAGs with full correction achieved between 6 and 12 months post-GT in the majority of the patients, persisting up to 24 months after GT. IDUA activity was detectable in all patients’ CSF by day+90 after GT and maintained over time, accompanied by decrease in GAG storage (1 log-fold reduction by 3-6 months post-GT). Patients with a positive anti-IDUA IgG titer cleared the antibodies by day+60 after GT. At 12 months post-GT, patients showed stable cognitive performances with improved or stable findings on brain and spine MRI, improved motor skills and joint stiffness, and growth tracking in-line with baseline percentile (mean growth velocity=9.9 cm/year).

Conclusions: HSPC-GT accomplishes extensive metabolic correction of peripheral and central compartments. Initial clinical efficacy is shown at the level of the CNS and bone which deserves further observation.

Clinical Trial Registry: NCT03488394

Disclosure: Gene therapy for MPSI was developed at San Raffaele Telethon Institute for Gene Therapy (SR-TIGET), which is a joint venture between Fondazione Telethon and Ospedale San Raffaele (OSR). The product was out-licensed to Orchard Therapeutics in May 2019. Gene therapy for MPSIH is still in development, it is not approved for use in patients outside of a clinical trial.

A. Aiuti was the PI of the trial up to May 2020, ME Bernardo is the current PI of the trial.

O056. Resolution of Serious Vaso-occlusive Pain Crises: Results From The Ongoing Phase 1/2 HGB-206 Group C Study of Lentiglobin for Sickle Cell Disease (SCD) Gene Therapy

Markus Mapara¹, Mark Walters², Alexis Thompson^3,4, Janet Kwiatkowski^5,6, Lakshmanan Krishnamurti⁷, Banu Aygun⁸, Kimberly Kasow⁹, Stacey Rifkin-Zenenberg¹⁰, Manfred Schmidt¹¹, Mauris Nnamani¹², Sara VanNest¹², Francis Pierciey Jr.¹², Alex Miller¹², Ren Chen¹², Dennis Kim¹², Sunita Goyal¹², Julie Kanter¹³, John Tisdale¹⁴

¹ Columbia University Medical Center, New York, United States, ² UCSF Benioff Children’s Hospital Oakland, Oakland, United States, ³ Northwestern University Feinberg School of Medicine, Chicago, United States, ⁴ Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, United States, ⁵ Children’s Hospital of Philadelphia, Philadelphia, United States, ⁶ University of Pennsylvania Perelman School of Medicine, Philadelphia, United States, ⁷ Children’s Healthcare of Atlanta, Emory University, Atlanta, United States, ⁸ Cohen Children’s Medical Center, Queens, United States, ⁹ University of North Carolina At Chapel Hill, Chapel Hill, United States, ¹⁰ Hackensack University Medical Center, Hackensack, United States, ¹¹ GeneWerk GmbH, GeneWerk GmbH, Germany, ¹² bluebird bio, Inc., Cambridge, United States, ¹³ University of Alabama Birmingham, Birmingham, United States, ¹⁴ National Institutes of Health, Bethesda, United States

Background: The ongoing Phase 1/2 HGB-206 Study (NCT02140554) evaluating the efficacy and safety of LentiGlobin for SCD (bb1111) gene therapy (GT) uses a modified human β-globin gene that produces GT-derived anti-sickling hemoglobin (HbA^T87Q). Group C data are presented here.

Methods: Patients (≥12 and ≤50 years) with SCD and severe vaso-occlusive events, including acute episodes of pain and acute chest syndrome (ACS), were enrolled. CD34+ cells collected by plerixafor mobilization/apheresis were transduced with BB305 LVV. LentiGlobin was infused following myeloablative busulfan conditioning. Patients were monitored for laboratory evaluations, SCD-related outcomes, and adverse events (AEs). Data are presented as median(min–max).

Results: As of 3 March 2020, 40 Group C patients (23.5[12–38] years), including 9 adolescents, initiated cell collection and underwent 2(1–4) mobilization cycles; 25/40 patients were treated with LentiGlobin. Drug product and treatment characteristics are summarized in Table 1. All patients stopped RBC transfusions by 90 days post-treatment. In 16 evaluable patients with ≥6 months of follow-up, total Hb at last visit was 11.5(9.6–16.2) g/dL, with HbA^T87Q contribution of 5.2(2.7–9.4) g/dL, HbS of 6.1(4.9–7.8) g/dL, and median HbS ≤60% of total Hb. Exploratory assays showed near pancellular expression of HbA^T87Q ≥6 months post-treatment with ~90% of RBCs containing β^A-T87Q by 18 months (n = 9). At last visit, hemolysis markers trended towards normalization (n = 25). In 14 patients with ≥6 months of follow-up and history of vaso-occlusive crisis (VOC) or ACS, the annualized VOC+ACS rate was 4(2–14) in the 2 years prior to treatment. No ACS or serious VOCs were observed after treatment. One non-serious, Grade 2 VOC occurred ~3.5 months after treatment (Fig 1). Most common non-hematalogic ≥Grade 3 AEs post-treatment were stomatitis (n = 15) and febrile neutropenia (n = 11). SAEs reported in ≥2 patients post-treatment were nausea, opioid withdrawal syndrome, and vomiting (all n = 2); no treatment-related SAEs were reported. There has been 1 death >18 months post-treatment in Group C, considered unlikely related to LentiGlobin, in a patient with significant baseline SCD-related burden. Per autopsy, cause of death was cardiovascular disease, with contributing SCD and asthma. There have been no events of graft failure, vector-mediated replication-competent lentivirus, or clonal dominance.

Table 1. Treatment and drug product characteristics

Treatment characteristics
Estimated average busulfan AUC (min*μmol/L)	4873.5 (4288-7322)
Neutrophil engraftment (days) (ANC≥500/μL x3 days)	19 (12-27)
Platelet engraftment (days) (platelets>50k/μL x3 days)	28 (19-136)
Hospitalization duration from conditioning to discharge (days)	35 (26-65)
Follow-up (months)	12.1 (2.8-24.8)
Drug product characteristics
Cell dose (x106 CD34+ cells/kg)	6.6 (3.0-16.0)
Drug product vector copy number (copies/diploid genome)	3.8 (2.3-5.7)
Transduced cells (%)	80 (63-93)

1. Values presented as median (min-max). n = 25 unless stated otherwise.
2. ANC, absolute neutrophil count; AUC, area under the curve

Conclusions: LentiGlobin for SCD GT results in near pancellular β^A-T87Q expression, reduced HbS expression, and increased total Hb. Complete resolution of VOC/ACS was observed in nearly all patients. The safety profile post-LentiGlobin remains consistent with myeloablative single-agent busulfan conditioning and underlying SCD. Additional patient data will be included in the presentation.

Clinical Trial Registry: NCT02140554

Disclosure: Markus Mapara: Nothing to declare.

Mark Walters: Editas, Consultancy; AllCells, Inc., Consultancy; Veevo Biomedicine, Consultancy.

Alexis Thompson: CRISPR/Vertex, Research Funding; Novartis, Research Funding; Bristol Myers Squibb, Consultancy, Research Funding.

Janet Kwiatkowski: bluebird bio, Consultancy, Research Funding; Agios, Consultancy; Celgene, Consultancy; Bristol Myers Squibb, Consultancy; Imara, Consultancy; Apopharma, Research Funding; Novartis, Research Funding; Terumo, Research Funding; Sangamo, Research Funding.

Lakshmanan Krishnamurti: Nothing to declare.

Banu Aygun: National Heart, Lung, Blood Institute, Research Funding; National Institute of Nursing Research, Research Funding; Patient Centered Outcomes Research Institute, Research Funding.

Kimberly Kasow: Nothing to declare.

Stacey Rifkin-Zenenberg: Nothing to declare.

Manfred Schmidt: German Cancer Research Center, Employment; GeneWerk GmbH, Equity Ownership.

Mauris Nnamani: bluebird bio, Inc., Employment, Equity Ownership.

Sara VanNest: bluebird bio, Inc., Employment.

Francis Pierciey Jr.: bluebird bio, Inc., Employment, Equity Ownership.

Alex Miller: bluebird bio, Inc., Employment, Equity Ownership.

Ren Chen: bluebird bio, Inc., Consultancy.

Dennis Kim: bluebird bio, Inc., Consultancy.

Sunita Goyal: bluebird bio, Inc., Employment, Equity Ownership.

Julie Kanter: bluebird bio, Consultancy, Honoraria; Novartis, Consultancy; Sanofi, Consultancy; Medscape, Honoraria; Guidepoint Global, Honoraria; GLG, Honoraria; Jeffries, Honoraria; Cowen, Honoraria; Wells Fargo, Honoraria; NHLBI Sickle Cell Advisory Board, Membership; SCDAA Medical and Research Advisory Board, Membership.

John Tisdale: Nothing to declare.

O057. Clinical Trial Update: EX-VIVO Autologous Stem Cell Gene Therapy in MPSIIIA

Jane Louise Kinsella¹, Simon A Jones², Adrian J Thrasher³, Claire Booth³, Karen F Buckland³, Natalia Izotova³, Stewart Rust¹, Daniel Weisberg¹, Heather J Church², Karen Tylee², Kathryn L Brammeier², Ceri Jones², Helena Lee⁴, Laura Ford⁴, Atusa Sadegholnejat⁴, Rebecca J Holley⁵, Stuart Ellison⁵, Brian Bigger⁵, Robert F Wynn¹

¹ Royal Manchester Children’s Hospital, Manchester, United Kingdom, ² Manchester Centre for Genomic Medicine, St Mary’s Hospital, Manchester, United Kingdom, ³ Great Ormond Street Institute of Child Health, University College, London, United Kingdom, ⁴ MRI, Transplantation Laboratory, Manchester, United Kingdom, ⁵ University of Manchester, Manchester, United Kingdom

Background: Autologous ex-vivo haematopoietic stem cell gene therapy (HSC-GT) to treat inborn errors of metabolism (IEM) is an exciting potential treatment modality for previously disease refractory lysosomal storage disorders (LSD). Mucopolysaccharidosis type IIIA (MPSIIIA) is a devastating LSD where mutations in the SGSH gene lead to substrate accumulation, cell inflammation and cell death. Symptomatically, these children present with developmental delay and continue to lose previously acquired skills, have significant behavioural disturbance, and die in their late teenage years. Whilst refractory to allogeneic stem cell transplant, we report the status of children with MPSIIIA treated with an ex-vivo autologous stem cell gene therapy approach as part of a phase I/II clinical trial.

Methods: This is an open label, phase I/II safety and tolerability study. Haematopoietic stem cells are collected from the peripheral blood following mobilisation with G-CSF and Plerixafor. The cells are transduced with a lentiviral vector containing the SGSH gene and a CD11b promoter. The investigational medicinal product (IMP) is infused following myeloablative conditioning.

Results: This trial has treated 3 patients to date with the IMP. Three individual IMP products have been manufactured with vector copy numbers 3.54, 3.23 and 6.26 copies/cell for patients 1,2 and 3 respectively. All three patients have successfully engrafted with the IMP. Platelet, neutrophil and haemoglobin counts demonstrated hematological engraftment - seen in figure 1a. There have been no serious adverse events related to the IMP.

All three patients demonstrate supra-physiological SGSH enzyme measured in leukocytes. At one-month post-transplant SGSH enzyme was measured to be 87, 80 and 38 times greater than the median of normal range for leukocytes for patients 1,2 and 3 respectively. At 3-months post-transplant, patients 1 and 2 had SGSH enzyme measured to be 187 and 120 times greater than the median of normal range in leukocytes, respectively (Figure 1b).

Substrate reduction measured as urinary GAG is seen to normalise in patients 1 and 2 by three months post-treatment. Further substrate analysis measuring heparan sulphate in urine and plasma is in progress and enzyme and heparan sulphate analysis in CSF is planned.

Conclusions: The delivery of ex-vivo autologous stem cell gene therapy transplants in these children with MPSIIIA was generally well tolerated and delivered supra-physiological enzyme in the blood at substantially greater levels than that observed in allogeneic transplant.

Clinical Trial Registry: NCT04201405

Disclosure: Brian Bigger - SAB member and shareholder of Orchard Therapeutics and PI on the unrestricted academic grant from Orchard Therapeutics that funds the clinical trial.

Simon Jones - SAB member and shareholder of Orchard Therapeutics.

Adrian Thrasher - Equity and consultancy for Orchard Therapeutics.

All other authors - Nothing to declare.

O058. Durable Clinical Outcomes Following Betibeglogene Autotemcel (BETI-CEL) Gene Therapy With Up to 6 Years of Follow-Up in Patients With Transfusion-Dependent Β-Thalassemia (TDT)

F. Locatelli¹, J. Kwiatkowski², M. Walters³, S. Hongeng⁴, J. Rasko⁵, M. Cavazzana⁶, M. Schmidt⁷, Y. Chen⁸, R. Colvin⁸, A. Thompson⁹ Franco Locatelli¹, Janet Kwiatkowski², Mark Walters³, Suradej Hongeng⁴, John Rasko⁵, Marina Cavazzana⁶, Manfred Schmidt⁷, Ying Chen⁸, Richard Colvin⁸, Alexis Thompson⁹

¹ IRCCS Bambino Gesù Children’s Hospital, Rome, Italy, ² Children’s Hospital of Philadelphia and Perelman School of Medicine of the University of Pennsylvania, Philadelphia, United States, ³ UCSF Benioff Children’s Hospital, Oakland, United States, ⁴ Mahidol University, Ramathibodi Hospital, Bangkok, Thailand, ⁵ The University of Sydney, Centenary Institute, and Royal Prince Alfred Hospital, Camperdown, Australia, ⁶ Necker Children’s Hospital, Assistance Publique-Hôpitaux de Paris, INSERM, and IMAGINE Institute, Université Paris Descartes, Sorbonne Paris Cité, Paris, France, ⁷ GeneWerk GmbH, Heidelberg, Germany, ⁸ bluebird bio, Inc., Cambridge, United States, ⁹ Ann & Robert H. Lurie Children’s Hospital of Chicago and Northwestern University Feinberg School of Medicine, Chicago, United States

Background: The goal of betibeglogene autotemcel (beti-cel; LentiGlobin for β-thalassemia) gene therapy in patients with TDT is lifelong, stable production of functional adult hemoglobin (Hb) sufficient for transfusion independence (TI) and improved erythropoiesis. 60 patients with TDT have been treated with beti-cel across completed phase 1/2 studies (HGB-204, HGB-205) and ongoing phase 3 studies (HGB-207, HGB-212). In patients with sufficient follow-up, 85% (23/27) of patients in phase 3 studies and 64% (14/22) of patients treated in phase 1/2 studies achieved TI (weighted average Hb ≥9g/dL without pRBC transfusions in ≥12 months). After 2 years of follow-up, all patients were invited to enroll in a 13-yr long-term follow-up study, LTF-303 (NCT02633943). Herein we report interim results of LTF-303 with up to 6 years of follow-up.

Methods: Autologous CD34+ cells transduced with BB305 lentiviral vector were infused into patients after busulfan myeloablation. The drug product manufacturing process was refined in the phase 3 studies compared to phase 1/2 studies. Iron management after beti-cel infusion through chelation or phlebotomy was at the investigator’s discretion. Data are presented as median (min–max).

Results: As of 3 March 2020, 32 patients (age at consent: 20 [12–35] years) treated with beti-cel were enrolled in LTF-303 (phase 1/2: n = 22; phase 3: n = 10) and were followed for 49.1 (23.3–71.8) months.

All patients who had achieved TI and were enrolled in LTF-303 (14/22 in phase 1/2; 9/10 in phase 3) had maintained TI in LTF-303 for an ongoing duration of 39.4 (19.4–69.4) months at last follow-up. Weighted average Hb during TI was 10.4 (9.4–13.3) g/dL (n = 14) and 12.5 (11.9–13.5) g/dL (n = 9) in patients from phase 1/2 and phase 3 studies, respectively. Of the patients who achieved TI, 74% (17/23) restarted iron chelation 12.7 (1.7–25.0) months post-infusion; 53% (9/17) of patients who restarted chelation have since discontinued chelation. Thirty percent (7/23) of patients received phlebotomy, including 3 patients who were receiving chelation. In patients who achieved TI, liver iron concentration (LIC) decreased over time following an initial increase after beti-cel infusion as a result of hypertranfusion and holding of chelation around time of myeloablation; this decrease was greater in patients with elevated baseline LIC (Figure).

No beti-cel‒related AEs were reported >2 years post-infusion. Serious AEs >2 years post-infusion included gonadotropic insufficiency, ectopic pregnancy, gall bladder wall thickening and polyp, bacteremia, neutropenia, and major depression (n = 1 each). All patients are alive at last follow-up and there have been no reports of replication competent lentivirus. Insertion site analysis as assessed every 6 months until month 60 showed unique insertions accounted for <30% of all insertions indicating polyclonal hematopoiesis and no clonal dominance. No single integration site contributed to more than 11% of all integration sites at last assessment.

Conclusions: These findings support the durability of transfusion independence and effective iron reduction after beti-cel gene therapy in patients with TDT for up to 6 years post-infusion. The favorable safety profile of beti-cel is supported by the absence of beti-cel–related AEs >2 years post-infusion and polyclonal vector integration.

Clinical Trial Registry: NCT02633943; https://clinicaltrials.gov/ct2/show/NCT02633943

Disclosure: Locatelli, Franco: honoraria, membership on board of directors or advisory committees (Amgen); consultancy, membership on board of directors or advisory committees (Bellicum Pharmaceuticals, Novarti); consultancy (bluebird bio, Inc., Sobi); honoraria (Miltenyi Biotec, Jazz Pharma, Medac). Kwiatkowski, Janet: consultancy (bluebird bio, Inc., Agios, BMS, Imara); research funding (bluebird bio, Inc., Apopharma, Novartis, Terumo Corporation, Sangamo).

Walters, Mark: consultancy (Editas Medicine, AllCells, Veevo Medicine). Hongeng, Suradej; Cavazanna, Marina: none. Rasko, John: consultancy (CRISPR Therapeutics, Rarecyte (stocks in lieu - equity ownership), Imago); honoraria (GSK, Takeda, Gilead, Cynata, Pfizer, Spark, Novartis, Celgene, bluebird bio, Inc.); Director of Pathology, stocks (Genea); committees (Gene Technology Technical Advisory, OGTR, Australian Government, Advisory Committee on Biologics, Therapeutics Goods Administration, Australian Government, NHMRC Mitochondrial Donation Expert Working Committee, Cure The Future Foundation; FSHD Global Research Foundation; Australian Cancer Research Foundation SAB). Schmidt, Manfred: Co-founder (GeneWerk GmbH). Chen, Ying; Colvin Richard: employee, ownership interest, and salary (bluebird bio, Inc.). Thompson, Alexis: consultant (bluebird bio, Inc.), consultant, research funding (Celgene, Novartis) research funding (Baxalta, Biomarin).

O059. Normalised Lysosomal Acid Lipase Activity Can Be Achieved In Wolman Fibroblasts Via Both Direct Transduction and Cross-Correction Mechanisms Using A Lentiviral-mediated Gene Therapy Approach

Jane E Potter¹, Stuart Ellison², Kathryn Brammeier³, Shaun Wood², Heather J Church³, Arunabha Ghosh³, Simon A Jones³, Robert F Wynn¹, Brian W Bigger²

¹ Royal Manchester Children’s Hospital, Manchester, United Kingdom, ² University of Manchester, Manchester, United Kingdom, ³ Manchester Centre for Genomic Medicine, St Mary’s Hospital, Manchester, United Kingdom

Background: Lentiviral vector-mediated autologous ex vivo hematopoietic stem cell (HSC) gene therapy (GT) is an emerging treatment in lysosomal storage disorders. Patient HSCs are harvested, transduced with a lentiviral vector expressing a functional copy of the defective gene, and infused back after myeloablative conditioning. Once engrafted, circulating leukocytes over express the deficient enzyme, secreting the enzyme, thereby allowing cross correction of otherwise enzyme-deficient recipient tissues.

Wolman disease arises from pathogenic variants in the LIPA gene, resulting in severe deficiency of lysosomal acid lipase (LAL) and subsequent accumulation of cholesterol esters and triglycerides. Cellular and organ dysfunction ensues, clinically observed as liver and gastrointestinal failure, with untreated Wolman patients not reaching 1 year of age. Treatment with enzyme replacement therapy (ERT) increases survival. HCT has formally been associated with high mortality and high incidence of graft failure, but we have recently reported combining ERT and HCT improves clinical outcomes compared with ERT alone, though engraftment issues persist.

Methods: A 3rd generation lentiviral construct was cloned that incorporates a codon optimised LIPA transgene for optimal expression in human cells and driven by a myeloid specific promoter, CD11b (CD11bLIPA LV). Direct transduction of Wolman fibroblasts with CD11bLIPA LV at a multiplicity of infection (MOI) of 5 (a relatively low dose), direct transduction of CHME3 (microglial like cells) with a MOI 5, and cross correction of Wolman fibroblasts (not directly exposed the lentivirus) using the transduced CHME3 MOI 5 media was evaluated. Enzyme activity (LAL/acid esterase) was used to assess if protein expression had been achieved in a validated acid esterase assay.

Results: Wolman fibroblasts have minimal enzyme activity. After transduction of Wolman fibroblasts with CD11bLIPA LV, deficient fibroblasts achieve significantly higher levels of enzyme activity (p = 0.0005) and approximately twice that of unaffected controls (p = 0.01). The CHME3 human microglial cell line, has some endogenous LAL enzyme activity and direct transduction of CHME3 cells (MOI5) results in a 3-fold increase in enzyme activity (p ≤ 0.0001). More importantly however, when considering ex vivo stem cell gene therapy these transduced cells can secrete LAL and cross correct Wolman fibroblasts. Mimicking standard HCT, we demonstrate that CHME3 media harvested from non-transduced CHME3 cells can cross correct Wolman fibroblasts, resulting in a small but not statistically significant increase in LAL. However, CHME3 media from transduced cells is able to normalise LAL activity levels to that of unaffected fibroblasts (p = 0.982).

Conclusions: We demonstrate improved cross correction of Wolman fibroblasts with myeloid cells transduced with CD11bLIPA LV compared with non-transduced cells, and propose that ex vivo autologous HSC GT will likely improve clinical outcomes compared with conventional allo-HCT, forming the future paradigm of disease management. Transduction of human HSC and in vivo models are needed to further demonstrate efficacy and to assess toxicity and safety.

Disclosure: S. A. Jones has received consultancy fees, grants, and travel support for conferences from Alexion Pharmaceuticals, Inc.

A. Ghosh has received consultancy fees/honoraria from Alexion Pharmaceuticals, Inc.

Other authors nothing to declare.

Graft-versus-host Disease – Clinical

O060. GVHD Prophylaxis With Post Transplantation Cyclophosphamide (PTCY) Versus Cyclosporine A / Methotrexate Post Allogeneic Transplantation (HSCT) From Matched Siblings For AML: From The ALWP/EBMT

Arnon Nagler¹, Myriam Labopin², Depei Wu³, Goda Choi⁴, Mahmoud Aljurf⁵, Fabio Ciceri⁶, Tobias Gedde-Dahl⁷, Ellen Meijer⁸, Riitta Niittyvuopio⁹, Ivan Moiseev¹⁰, Jean Henri Bourhis¹¹, Jan .J Cornelissen¹², Gerard Socié¹³, Yener Koc¹⁴, Jonathan Canaani¹, Bipin Savani¹⁵, Gesine Bug¹⁶, Alexandros Spyridonidis¹⁷, Sebastian Giebel¹⁸, Eolia Brissot¹⁹, Ali Bazarbachi²⁰, Jordi Esteve²¹, Mohamd Mohty²

¹ Sheba Medical Center, Ramat Gan, Israel, ² Saint Antoine Hospital, Paris, France, ³ First Affiliated Hospital of Soochow University, Suzhou, China, ⁴ University Medical Center Groningen (UMCG), Groningen, Netherlands, ⁵ King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia, ⁶ Ospedale San Raffaele s.r.l., Milano, Italy, ⁷ Oslo University Hospital, Rikshospitalet,Clinic for Cancer Medicine, Oslo, Norway, ⁸ VU University Medical Center, Amsterdam, Netherlands, ⁹ HUCH Comprehensive Cancer Center, Helsinki, Finland, ¹⁰ RM Gorbacheva Research Institute, Pavlov University, St. Petersburg, Russian Federation, ¹¹ Gustave Roussy Cancer Campus, Villejuif, France, ¹² Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, Netherlands, ¹³ Hopital St. Louis, Paris, France, ¹⁴ Medicana International Hospital Istanbul, Istanbul, Turkey, ¹⁵ Vanderbilt University Medical Center, Nashville, United States, ¹⁶ Universitätsklinikum Frankfurt, Frankfurt, Germany, ¹⁷ University Hospital of Patras, Patras, Greece, ¹⁸ Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice, Poland, ¹⁹ Hôpital Saint Antoine, Paris, France, ²⁰ American University of Beirut Medical Center, American University of Beirut Medical Center, Beirut, Lebanon, ²¹ Hospital Clínic de Barcelona, Barcelona, Spain

Background: GVHD is a major complication after HSCT. The standard regimen for prevention of GVHD is Cyclosporine A (CSA) and methotrexate (MTX). Recently post-transplant cyclophosphamide (PTCy) has proven to be highly effective in haploidentical transplants and is increasingly used in HSCT from unrelated and matched sibling donors (MSD).

Methods: Study aim was to compare PTCy to CSA / MTX as GVHD prophylaxis in adult pts (≥18 years) with AML undergoing first HSCT in CR1 from MSD in 2010‐2019.Multivariate analysis (MVA) adjusting for differences between the groups was performed using Cox’s proportional hazards regression model for main outcomes. A propensity score (PS) matched analysis was conducted to corroborate the results obtained in the overall population.

Results: 1320 pts were included: 1202 (91%) with CSA/MTX and 118 (9%) with PTCy. Median F/U was 30.5 (27-34) and 16 [13-21] months, respectively (p < 0.0001). Median age was 45 (18-73) and 52 (19-71) y, respectively (p < 0.0001). 55.3% and 44.9% were male (p = 0.63). 78.5% and 82.2% were classified as intermediate risk and 21.5% and 17.8% as adverse risk cytogenetics, respectively (p = 0.35). Conditioning was myeloablative in 81.5% and 58.5%, (p < 0.0001). Graft source was peripheral blood in 75.9% and 80.5% of transplants. Karnofsky performance score was >90 in 84.6% and 68.1% of pts (p < 0.0001).32.2% and 37.5% and 30.1% and 28.2% of the pts harbored the FLT3 and NPM1 mutation, respectively (missing in ~30%).Engraftment was achieved by 99.7% and 100% of pts. Day 180 incidence of acute (a) GVHD II-IV and III-IV was 22.9% and 25.2 (p = 0.72) and 8.1% and 5.2% (p = 0.25), respectively and 2-y total and extensive chronic (c) GVHD was 47.9% and 42 % (p = 0.52) and 29.4% and 25.5 % (p = 0. 85). The 2-y NRM was 12% and 8.7% (p = 0.39). Rate of death due to GVHD was 5.3% vs 3.5% (p = 0.28). 2y relapse rate (RI) was 33.5% and 44.6%, respectively (p = 0.06) and it was the main cause of death (49% and 62.1%). Rate of death due to infection did not differ. The 2-y LFS, OS and GRFS did not differ between the groups being 63.2% vs 51.1%, 73% vs 67.7% and 42.8 vs 35.4% (p = 0.13,0.41 and 0.64), respectively.

In MVA, RI was significantly lower while NRM, GVHD, LFS, OS and GRFS did not significantly differ between pts receiving CSA/MTX vs PTCy: HR = 1.52 (1.05-2.18, p = 0.025), HR = 0.65 (0.28-1.52, p = 0.32), HR = 1.29 (0.92-1.82, p = 0.14), HR = 1.15 (0.74-1.79, P = 0.52), and HR = 1.02(0.76-1.36, p = 0.91), respectively. Risk of aGVHD II-IV and III-IV was similar also, HR = 1.03 (0.69-1.55, p = 0.88), HR = 0.55 (0.23-1.3, p = 0.17) as was risk of total and extensive cGVHD, HR =0.8 (0.56-1.13, p = 0.2) and HR = 0.9 (0.56-1.47, p = 0.68), respectively. These results were confirmed in a matched-pair analysis of 2x110 pts well balanced for the major prognostic factors.

Conclusions: In AML pts undergoing HSCT from MSD in CR1, GVHD prophylaxis with PTCy is feasible, resulting in similar incidence of GVHD, LFS, OS and GRFS as conventional CSA/MTX. The higher RI observed with PTCy needs further evaluation and may be related to the PTCy mediated NK inhibitory and Treg cell potentiating activity observed early post HSCT.

Disclosure: Nothing to declare.

O061. Glucagon-Like-Peptide-2 (GLP-2) Analogue Teduglutide for Graft-Versus-Host Disease in Humans

Johana Norona¹, Petya Apostolova¹, Dominik Schmidt¹, Natalie Köhler¹, Tatjana Zabelina², Nicolaus Kröger², Francis Ayuk², Robert Zeiser ¹

¹ University Hospital Freiburg, Freiburg, Germany, ² University Medical Center Hamburg-Eppendorf, Hamburg, Germany

Background: The endocrine activity of the gastrointestinal tract has tissue protective and regenerative function. This function is impaired when intestinal enteroendocrine cells are damaged in the context of infection or after surgical resection. Acute graft-versus-host disease (GVHD) of the intestinal tract is a common, life-threatening complication after allogeneic hematopoietic cell transplantation (allo-HCT). Based on our promising preclinical (Norona J et al. Blood 2020), we hypothesized that treatment with the Glucagon-like-peptide-2 (GLP-2) analogue teduglutide could improve the clinical outcome of patients with acute GVHD.

Methods: We treated a cohort of six patients with steroid refractory acute GVHD having failed multiple therapies with teduglutide at a dosage of 0.05 mg/kg body weight once daily for 10 days. Human sample collection and analysis were approved by the Institutional Ethics Review Board of the Medical center, University of Freiburg, Germany. Written informed consent was obtained from each patient. All analysis of human data was carried out in compliance with relevant ethical regulations. The patient information is shown in Table 1.

Results: Intestinal biopsies taken from one patient before and during teduglutide treatment showed an increase of Paneth cells after the treatment (Figure 1A). In agreement with the increase of Paneth cells, also the clinical signs of intestinal GVHD improved in all six patients with a decline of the diarrhea frequency (Figure 1B). In addition, we analyzed the serum albumin levels in the six patients and found an increase of the serum albumin when comparing the day before treatment start with the time point 2 weeks after the treatment had been started (Figure 1C). During the teduglutide treatment period no additional intravenous albumin or fresh frozen plasma was substituted.

Table 1. Patient information

Total number 6
Variable
	median (range)
Pt. age 49 (28-74)
in years
Gender % (absolute number)
female 66.7 (4)
male 33.3 (2)
Diagnosis:
AML 50 (3)
MDS 16.6 (1)
NHL 16.6 (1)
CGD 16.6 (1)
Conditioning Regimen
MAC 50 (3)
RIC 50 (3)
Donor Type
MRD 16.7 (1)
MUD 50 (3)
MMUD 33.3 (2)
Graft Source
PBSC 100 (6)
GvHD Manifestation
Skin 66.7 (4)
Intestines 100.0 (6)
Liver 16.7 (1)
Immunosuppression, Prophylaxis
CyA 100 (6)
MMF 100 (6)
ATG 83.3 (5)
Immunosuppression, Treatment
Decortin 100 (6)
Everolimus 16.7 (1)
ECP 16.7 (1)
Ruxolitinib 83.3 (5)

1. Abbreviations: CGD: Chronic granulomatous disease, AML: acute myeloid leukemia, NHL: Non Hodgkin lymphoma, MAC: Myeloablative conditioning, RIC: reduced intensity conditioning, MRD: matched related donor, MUD: matched unrelated donor, PBSC: peripheral blood stem cells, CyA: cyclosporine A, MMF: mycophenolate mofetil, ATG: Anti Thymocyte globuli

Conclusions: These findings suggest that teduglutide, which is approved for short bowel syndrome and has a limited toxicity profile, may be effective in patients with SR-GVHD, leading to an improved absorption and thereby higher albumin levels. This is in line with previous work showing that GLP-2 can increase nutrient absorption in normal rodents and in patients with short bowel syndrome.

Clinical Trial Registry: N/A

Disclosure: RZ received speaker fees from Novartis, Incyte and Mallinckrodt.

O062. Abstract already published

O063. Ruxolitinib Vs Best Available Therapy in Patients With Steroid-Refractory/Dependent Chronic GRAFT-VS-HOST Disease (CGVHD): Subgroup Analyses of Overall Response Rate in The Phase 3 Reach3 Trial

Franco Locatelli¹, Nicola Polverelli², Ron Ram³, Shahrukh K Hashmi⁴, Ronjon Chakraverty⁵, Jan Moritz Middeke⁶, Maurizio Musso⁷, Sebastian Giebel⁸, Ant Uzay⁹, Francis Ayuketang Ayuk¹⁰, Tsila Zuckerman¹¹, Kentaro Fukushima¹², Juan Carlos Vallejo¹³, Joseph Pidala¹⁴, Nada Hamad¹⁵, Ivan Moiseev¹⁶, Francesca Bonifazi¹⁷, Shashikant Apte¹⁸, Peter Langmuir¹⁹, Norbert Hollaender²⁰, Maanasa Gowda²¹, Tommaso Stefanelli²⁰, Stephanie J. Lee²², Robert Zeiser²³, Takanori Teshima²⁴

¹ IRCCS, Ospedale Pediatrico Bambino Gesu’, Sapienza, Università di Roma, Rome, Italy, ² ASST Spedali Civili di Brescia, University of Brescia, Brescia, Italy, ³ Tel Aviv (Sourasky) Medical Center and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel, ⁴ King Faisal Specialist Hospital and Research Center, Sheikh Shakhbout Medical City, Mayo Clinic, Riyadh, Saudi Arabia, United Arab Emirates, ⁵ UCL Cancer Institute; Institute of Immunity and Transplantation, London, United Kingdom, ⁶ Medizinische Klinik und Poliklinik I, Universitätsklinikum Dresden, Dresden, Germany, ⁷ UOC di Oncoematologia e TMO, ‘la Maddalena’, Palermo, Italy, ⁸ Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice Branch, Gliwice, Poland, ⁹ Acibadem University Hospital, Istanbul, Turkey, ¹⁰ University Medical Center, Hamburg-Eppendorf, Germany, ¹¹ Hematology Institute and Bone Marrow Transplantation, Clinical Research Institute at Rambam, Rambam Health Care Campus, Haifa, Israel, ¹² Osaka University Graduate School of Medicine, Suita, Osaka, Japan, ¹³ Hospital de Donostia, Donostia, Spain, ¹⁴ Moffitt Cancer Center, Tampa, United States, ¹⁵ St Vincent’s Hospital Darlinghurst and University of NSW, Darlinghurst, NSW, Australia, ¹⁶ RM Gorbacheva Research Institute, Pavlov University, St. Petersburg, Russian Federation, ¹⁷ S. Orsola-Malpighi University Hospital, University of Bologna, Bologna, Italy, ¹⁸ Sahyadri Speciality Hospital, Pune, Maharashtra, India, ¹⁹ Incyte Corporation, Wilmington, United States, ²⁰ Novartis Pharma AG, Basel, Switzerland, ²¹ Novartis Pharmaceuticals Corporation, East Hanover, United States, ²² Fred Hutchinson Cancer Research Center, Seattle, United States, ²³ University of Freiburg, Freiburg, Germany, ²⁴ Hokkaido University, Sapporo, Japan

Background: Ruxolitinib demonstrated superior efficacy over best available therapy (BAT) in the phase 3 REACH3 study conducted in patients with steroid-refractory/dependent cGVHD. Ruxolitinib led to a significantly higher overall response rate (ORR) at week 24 vs BAT. Determining which patients derive the largest benefit is of great clinical interest; therefore, we present results from subgroup analyses of the REACH3 study population.

Methods: Patients aged ≥12 years with moderate/severe steroid-refractory/dependent cGVHD were randomized 1:1 to receive ruxolitinib 10 mg twice daily or investigator-selected BAT. Subgroup analyses comparing the odds ratios (ruxolitinib vs BAT) of ORR by demographics, cGVHD disease history, and transplant-related history were conducted. Estimation of the odds ratios and 95% CIs was performed using Mantel-Haenszel methods.

Results: A total of 329 patients were randomized to ruxolitinib (n = 165) or BAT (n = 164). Baseline characteristics were balanced between arms. Median (range) age was 49 (12-76) years. Overall, 42.9% and 56.8% of patients had moderate or severe cGVHD, respectively; 71.4% had steroid-refractory cGVHD, and 28.6% had steroid-dependent disease. At data cutoff (May 8, 2020), 204 patients (62.0%) had discontinued, mainly due to lack of efficacy (ruxolitinib, 14.5%; BAT, 42.7%).

ORR at week 24 in the overall study population was significantly higher with ruxolitinib than with BAT (49.7% vs 25.6%; odds ratio, 2.99 [95% CI, 1.86-4.80]; P < 0.0001^a). A higher ORR at week 24 and a favorable odds ratio (>1) was seen with ruxolitinib vs BAT across most patient subgroups of baseline factors and clinical GVHD characteristics (Figure); however, patient numbers in some subgroups were too small to draw definitive conclusions.

Ruxolitinib led to a higher ORR than BAT regardless of cGVHD disease history, including in patients with moderate (ruxolitinib, 59.5% vs BAT, 32.5%; odds ratio, 3.05 [95% CI, 1.59-5.84]) or severe cGVHD (40.7% vs 19.0%; odds ratio, 2.92 [95% CI, 1.46-5.84]). Response was similar between patients with steroid-refractory and those with steroid-dependent disease, and although patients were not stratified by these categories, the odds ratios favored ruxolitinib in all 3 criteria subcategories of inadequate response to steroids (see Figure footnote for details). ORR was also higher with ruxolitinib than with BAT regardless of stem cell source; ORR was higher when peripheral blood vs bone marrow cells were used (peripheral blood: ruxolitinib, 51.1% vs BAT, 29.0%; odds ratio, 2.68 [95% CI, 1.60-4.49]; bone marrow: 36.4% vs 12.9%; odds ratio, 3.69 [95% CI, 0.96-14.15]). Prior acute GVHD did not affect the response to ruxolitinib (Figure).

Odds ratios also favored ruxolitinib vs BAT regardless of transplant-related history, including conditioning regimen, T-cell depletion, stem cell source, donor type (related vs unrelated), donor HLA status, donor source/HLA match status, and donor/recipient gender match (Figure). Subgroup analyses according to organ involvement are ongoing.

Conclusions: Ruxolitinib showed clinically meaningful benefit in patients with inadequate response to steroids across different subgroups, including cGVHD severity, criteria for steroid-refractory/dependent cGVHD, prior acute GVHD, and transplant characteristics. Overall, subgroup analyses demonstrated results consistent with those of the primary analysis, with ORR at week 24 significantly higher with ruxolitinib than with BAT in most patient groups.

Clinical Trial Registry: ClinicalTrials.gov identifier, NCT03112603

Disclosure: Franco Locatelli has participated in speakers bureaus for Amgen, Jazz Pharmaceuticals, Medac, Miltenyi, Novartis, and Takeda Pharmaceutical Co Ltd and has been a member of the board of directors or advisory committee for Amgen, Bellicum Pharmaceuticals, Neovii, and Novartis. Nicola Polverelli has served as a consultant for Novartis. Ron Ram has received honoraria from Novartis, Gilead, and Janssen. Ronjon Chakraverty has participated in speakers bureaus for Mallinckrodt Pharmaceuticals (Therakos (UK) Ltd) and Neovii. Jan Moritz Middeke has been a member of an advisory board and participated in speakers bureaus for Novartis. Sebastian Giebel has been a member of an advisory board and participated in speakers bureaus for Novartis. Francis Ayuketang Ayuk has been a member of an advisory board and participated in speakers bureaus for Novartis. Joseph Pidala has served as a consultant and advisory board member for Syndax, CTI BioPharma, and Amgen and has provided clinical trial support to Novartis, Amgen, Takeda Pharmaceutical Co Ltd, Janssen, Johnson & Johnson, Pharmacyclics, AbbVie, CTI BioPharma, and Bristol Myers Squibb.

Nada Hamad has been a member of an advisory board and participated in speakers bureaus for Novartis. Ivan Moiseev has received personal fees from MSD, Pfizer, Celgene, and Takeda Pharmaceutical Co Ltd and grants and personal fees from Novartis and Bristol Myers Squibb. Peter Langmuir is employed by Incyte. Norbert Hollaender is employed by Novartis Pharma AG. Maanasa Gowda is employed by Novartis Pharmaceuticals Corporation. Tommaso Stefanelli is employed by and has equity ownership in Novartis Pharma AG. Stephanie J. Lee has received research funding from Amgen, AstraZeneca, Incyte, Kadmon, Novartis, Pfizer, Syndax, and Takeda Pharmaceutical Co Ltd and is a member of a steering committee at Incyte. Robert Zeiser has received honoraria from Incyte, Mallinckrodt, and Novartis. Takanori Teshima has received grant funding from Astellas, Chugai, Fuji Pharma, Kyowa Hakko Kirin Co Ltd, Novartis, Nippon Shinyaku, Sanofi, and Teijin Pharma; honoraria from Bristol Myers Squibb, Kyowa Hakko Kirin, MSD, Novartis, Pfizer, and Takeda; and has been a member of advisory boards for Novartis and Takeda. Shahrukh K. Hashmi, Maurizio Musso, Ant Uzay, Tsila Zuckerman, Kentaro Fukushima, Juan Carlos Vallejo, Francesca Bonifazi, and Shashikant Apte have nothing to declare.

O064. A Personalized Organ-Based Approach to The Treatment of Chronic GRAFT-VERSUS-HOST Disease

Hanaa A. Fatoum¹, Robert Zeiser², Shahrukh K. Hashmi³

¹ Alfaisal University, Riyadh, Saudi Arabia, ² University of Freiburg, Freiburg, Germany, ³ Mayo Clinic, Rochester, United States

Background: Chronic graft-versus-host-disease (cGvHD) remains one of the leading causes of morbidity and mortality amongst transplant recipients. 50-60% of patients develop steroid-refractory or resistant cGvHD and the efficacy of the second-line treatments varies widely based on many factors, including wide differences in the organ response-rates (RR).

Methods: We aimed to evaluate outcomes based on the organ response-rates to the most commonly used agents for the treatment of steroid-refractory/steroid-resistant cGVHD by conducting a systematic literature review. Clinical trials, retrospective studies, were selected if they included detailed results of various organ RR to the respective therapeutic agent.

Results: A total of 279 studies were evaluated for the organ-specific RR of 12 cGvHD treatments based on the strict selection criteria.

The highest cutaneous RR was observed in Methotrexate, Ibrutinib, and Ruxolitinib at 100, 88, and 82% respectively, while Abatacept, Rituximab, and Pentostatin yielded the lowest RR at 13, 22, and 25% respectively. Similarly, was the ocular RR as demonstrated in the graph.

With Methotrexate or Ruxolitinib was 100 percent albeit the rest of the agents ranged between 0.17 in Abatacept to 0.47 for those treated with Mesenchymal stromal cells (MSC)s.Belumosudil, Ruxolitinib, Ibrutinib, and MSCs induced a gastrointestinal RR of over 90%, and for oral GVHD a RR of 0.65, 0.99, 0.88, and 1.00. In contrast, Pomalidomide, carfilzomib, and pentostatin were only effective in less than 23 percent in oral GvHD.

Musculoskeletal-GvHD responded best to Ruxolitinib and Belomosudil at 100 and 73% respectively. However, it was only reported in 5/12 agents/treatments. For lung GvHD (BOS), 5/12 treatments reported a RR which was only 11-12% in those treated with Hydroxychloroquine, and Abatacept. Nevertheless, a lung response rate of 0.80 and 0.50 was seen in patients on ruxolitinib and carfilzomib, respectively. Lastly, genital GvHD had by far the least reported trial data with only 1 agent; Ruxolitinib, showing a 75% response in 3/4 patients. The hepatic response was observed in all 7 patients who received Ruxolitinib followed by 0.91 in MSC compared to 5% of patients on Pentostatin.

Overall, our study had a limitation of not including some key studies, due to the absence of reporting on individualized organ RRs.

Graph: Chronic GvHD organ response rates based on therapeutic agents.

Conclusions: Most of the GvHD trials conducted are focused on the overall response rate (ORR), partial, and complete response amongst transplant patients and the evidence for organ-specific response is limited and therefore our study results are striking for RR yields for some of the organs. Thus, a personalized organ-based approach to the selection of therapeutic agents in cGvHD can result in a more favorable patient outcome and achieving a better organ-specific response while minimizing toxicity and side effects.

Disclosure: Shahrukh K. Hashmi received Speakers or advisory board fees from Pfizer, Novartis, Janssen, Mallinckrodt.

Robert Zeiser received Speaker fees from Novartis, Incyte and Mallinckrodt.

O065. Discontinuation of Systemic Immunosuppressive Therapy: New Goal of Chronic GVHD Treatment?

Ivan Moiseev¹, Anna Dotsenko¹, Anna Smirnova¹, Julia Vlasova¹, Elena Morozova¹, Sergey Bondarenko¹, Alexandr Kulagin¹

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

Background: Published long-term studies of systemic therapy in chronic graft-versus-host disease patients (cGVHD) indicate that median time to complete discontinuation of immunosuppressive (IST) therapy is around 2-4 years. Complete response on the other hand is observed in the minority of patients. So continuation of systemic IST until complete response might be not the optimal therapeutic strategy. In this single-center study on the large cohort of cGVHD patients on systemic therapy we tried to analyze the factors affecting the outcomes of cGVHD treatment.

Methods: 182 patients with moderate and severe cGVHD after MRD and MUD allogeneic hematopoietic stem cell transplantation (HSCT) were included in the study. The majority of patients were allografted for malignant disease (95%), predominantly acute leukemia. Severe cGVHD was observed in 57%, moderate in 43%. As the first line of therapy 62% of patients received prednisone 1 mg/kg daily in combination with calcineurin inhibitor (CNI), 22% received CNI as the monotherapy, 16% received monotherapy with a second line treatment (pharmacological or extracorporeal photopheresis without steroids). Median follow up was 52 months.

Results: At five years the cumulative incidence of complete response (CR) was 16.9% (95% CI 10.5-24.7%), but IST discontinuation without GVHD flare was observed in 51.2% (95% CI 40.0-61.2%) of patients. At the time of analysis IST was discontinued in 55% with CR, in 28% with mild manifestations after treatment, in 26% with moderate and 8% with severe. In the multivariate analysis of NRM it was demonstrated that the initial severity of cGVHD had no impact on mortality (HR 1.70, 95%CI 0.80-3.97, p = 0.1959), while discontinuation of IST (HR 0.03, 95%CI 0.01-0.15, p = 0.0005), steroid-free starting therapy (HR 0.25, 95%CI 0.08-0.58, p = 0.0035) and use of second-line therapy (HR 0.49, 95%CI 0.25-0.96, p = 0.0322) were protective against NRM (Figure 1A). NRM in patients who discontinued IST was 2% vs 42%. The only significant factors for probability of IST discontinuation were overall initial severity of cGVHD (HR 0.45, 95%CI 0.25-0.84, p = 0.0049) and female donor for male recipient (HR 0.33, 95%CI 0.25-0.81, p = 0.0370) (Figure 1B).

Conclusions: In this study we demonstrated that successful discontinuation of IST is associated with very low NRM. IST discontinuation is feasible in patients achieving CR, mild or moderate manifestations of cGVHD. IST discontinuation as a goal of therapy in cGVHD should be evaluated in the multicenter setting.

Disclosure: Nothing to disclose.

O066. A Phase 2, Prospective, Randomised, Open-Label Study of Defibrotide For The Prevention of Acute Graft-Versus-Host Disease (AGVHD) After Allogeneic Haematopoietic Cell Transplantation (HCT)

Michelle Hudspeth¹, Shahram Mori², David Nachbaur³, Jose Antonio Perez Simon⁴, Friedrich Stölzel⁵, Marcie Riches⁶, Wendy Wu⁷, Peixin Zhang⁸, Vian Amber⁹, Ibrahim Yakoub-Agha¹⁰

¹ Medical University of South Carolina Children’s Hospital/Hollings Cancer Center, Charleston, SC, United States, ² AdventHealth, Orlando, FL, United States, ³ University Hospital for Internal Medicine V, Hematology & Oncology, Medical University, Innsbruck, Austria, ⁴ Hospital Universitario Virgen del Rocio, Instituto de Biomedicina de Sevilla (IBIS / CISC / CIBERONC), Universidad de Sevilla, Sevilla, Spain, ⁵ Universitätsklinikum Carl Gustav Carus an der TU Dresden, Dresden, Germany, ⁶ University of North Carolina at Chapel Hill, Chapel Hill, United States, ⁷ Jazz Pharmaceuticals, Palo Alto, United States, ⁸ Jazz Pharmaceuticals, Philadelphia, United States, ⁹ Jazz Pharmaceuticals, Oxford, United Kingdom, ¹⁰ CHU de Lille, INSERM U 1286, Infinite, Lille, France

Background: aGvHD is a life-threatening complication that typically occurs within the first 100 days following allogeneic HCT. Defibrotide, which has been shown to protect endothelial cells in vitro, is approved to treat patients aged >1 month with severe hepatic VOD/SOS post-HCT in the EU and VOD/SOS with renal/pulmonary dysfunction post-HCT in the US. Previous clinical trials have shown potential utility of defibrotide in preventing aGvHD. This study evaluated the efficacy and safety of defibrotide added to standard of care immunoprophylaxis (SOC) versus SOC alone for aGvHD prevention.

Methods: This phase 2 randomised study (NCT03339297) enrolled adult and paediatric patients who underwent peripheral blood stem cell or bone marrow transplantation from an unrelated donor after myeloablative or reduced intensity conditioning. Eligible patients were randomised 1:1 to receive defibrotide prophylaxis (defibrotide 25 mg/kg/day IV + SOC) or SOC alone. Randomisation was stratified by age, region, and anti-thymocyte globulin (ATG) use. The primary endpoint was cumulative incidence of Grade B-D aGvHD by Day 100 post-HCT. Secondary endpoints included cumulative incidence of Grade B-D aGvHD by Day 180 post-HCT. This study estimated incidences and was not powered to assess statistically significant differences among treatment groups.

Results: Overall, 152 patients were included: 79 in the defibrotide group (median age [range]: 57.0 [1.6, 69.0] years) and 73 in the SOC group (median age: 56.0 [1.5, 72.0] years). In the defibrotide group, 75.7% of patients received myeloablative conditioning compared to 61.4% of patients in the SOC group.

Cumulative incidence of Grade B-D aGvHD by Day 100 post-HCT was 38.4% in the defibrotide group versus 47.1% in the SOC group (Figure); this was 30.4% and 47.6%, respectively, in patients receiving ATG, and 42.0% and 46.9%, respectively, in patients who did not receive ATG. By Day 180 post-HCT, the cumulative incidence of Grade B-D aGvHD was 50.6% versus 51.6% in the defibrotide and SOC groups, respectively. The Kaplan-Meier–estimated overall survival by Day 180 post-HCT was similar in the defibrotide and SOC groups (86.0% vs 86.9%, respectively).

No defibrotide-related serious treatment-emergent adverse events (TEAEs) or deaths were reported. TEAEs were comparable between the two groups; the three most common were nausea (defibrotide group: 78.4%; SOC group: 70.0%), diarrhoea (64.9%; 75.7%, respectively), and stomatitis (56.8%; 51.4%, respectively). Serious TEAEs occurred in 31/74 (41.9%) patients in the defibrotide group and 31/70 (44.3%) in the SOC group, with gastrointestinal disorders and infections being the most common. The incidence of bleeding was 33.8% in the defibrotide group and 41.4% in the SOC group.

Conclusions: While not statistically different, topline results showed the cumulative incidence of Grade B-D aGvHD at Day 100 post-HCT was modestly lower with defibrotide prophylaxis versus SOC. Overall survival by Day 180 post-HCT was similar between the two treatment groups. Safety results were comparable between the two groups, consistent with the acceptable safety profile of defibrotide in a randomised setting. There was no increased incidence of bleeding events with defibrotide treatment versus SOC and no defibrotide-related serious TEAEs or deaths. Additional detailed analyses may help further elucidate study outcomes.

Clinical Trial Registry: ClinicalTrials.gov NCT03339297

Disclosure: Michelle Hudspeth: has served on advisory boards for Mesoblast and Jazz Pharmaceuticals. Shahram Mori: no conflicts of interest to disclose. David Nachbaur: no conflicts of interest to disclose. Jose Antonio Perez Simon: has served on advisory boards for Jazz Pharmaceuticals. Friedrich Stölzel: no conflicts of interest to disclose. Marcie Riches: has served on an advisory board for Bio-Intelect. Wendy Wu: employee of and holds stock ownership and/or stock options in Jazz Pharmaceuticals. Peixin Zhang: employee of and holds stock ownership and/or stock options in Jazz Pharmaceuticals. Vian Amber: employee of and holds stock ownership and/or stock options in Jazz Pharmaceuticals. Ibrahim Yakoub-Agha: has received honoraria from Jazz Pharmaceuticals.

O067. Ruxolitinib (RUX) Vs Best Available Therapy (BAT) in Patients With Steroid-refractory Acute GRAFT-VS-HOST Disease (SR-AGVHD): 6-Month Follow-up From The Randomized, Phase 3 Reach2 Study

Mohamad Mohty¹, Claude-Eric Bulabois², Valentin García-Gutiérrez³, David Ritchie⁴, Sung-Soo Yoon⁵, Valerie Coiteux⁶, Yok-Lam Kwong⁷, Dietger Niederwieser⁸, Reuven Or⁹, Gérard Socié¹⁰, Nikolas von Bubnoff¹¹, Robert Zeiser¹², Judith Xu¹³, Juliane Morando¹³, Bruyère Mahuzier¹⁴, Jeff Szer⁴

¹ Hôpital Saint-Antoine, Assistance Publique – Hôpitaux de Paris, Sorbonne Université, and INSERM Unité Mixte de Recherche 938, Paris, France, ² Centre Hospitalier Universitaire de Grenoble, Grenoble, France, ³ Hospital Universitario Ramón y Cajal, IRYCIS, Madrid, Spain, ⁴ Peter MacCallum Cancer Centre and The Royal Melbourne Hospital, Melbourne, VIC, Australia, ⁵ Seoul National University Hospital, Seoul, Korea, Republic of, ⁶ Hôpital Huriez, CHU de Lille, Lille, France, ⁷ Queen Mary Hospital, Hong Kong, China, ⁸ Leipzig University, Leipzig, Germany, ⁹ Cancer Immunotherapy and Immunobiology Research Center, Hadassah Medical Center, Jerusalem, Israel, ¹⁰ Assistance Publique – Hôpitaux de Paris, Hôpital Saint-Louis, Université de Paris, and INSERM Unité Mixte de Recherche 976, Paris, France, ¹¹ University Hospital Schleswig-Holstein, Campus Lübeck, Lübeck, Germany, ¹² University of Freiburg, Freiburg, Germany, ¹³ Novartis Pharmaceuticals Corporation, East Hanover, United States, ¹⁴ Novartis Pharma, Rueil-Malmaison, France

Background: RUX led to a significantly higher overall response rate (ORR) at day 28 (primary endpoint; 62.3% vs 39.4%) and durable ORR at day 56 (key secondary endpoint; 39.6% vs 21.9%) than BAT in the phase 3 REACH2 study (N = 309). We report safety and efficacy findings from a 6-month follow-up analysis of REACH2.

Methods: Patients aged ≥12 years with grade II-IV SR-aGVHD received RUX 10 mg bid (n = 154) or investigator-selected BAT (n = 155). Crossover from BAT to RUX was allowed on or after day 28. Outcomes evaluated included duration of response (DOR), event-free survival (EFS), failure-free survival (FFS), nonrelapse mortality (NRM), malignancy relapse/progression, chronic GVHD (cGVHD), patient-reported outcomes, ORR after crossover, corticosteroid dose, and safety. Overall survival will be presented at the final analysis.

Results: At data cutoff (January 6, 2020), 35 (22.7%) RUX and 21 (13.5%) BAT patients had completed the randomized period. Primary reasons for early discontinuation were lack of efficacy (RUX, 20.8%; BAT, 44.5%), death (16.2%; 13.5%), and adverse events (AEs; 17.5%; 3.9%). 49 BAT patients (31.6%) crossed over to RUX; 11 had completed the crossover period and 36 (73.5%) discontinued, with AEs (24.5%) being the most common reason. 96 RUX (62.3%) and 51 BAT patients (32.9%) entered long-term follow-up. Median exposure was 63 days for RUX and 29 days for BAT.

Median DOR was longer in the RUX arm (163 days [range, 22.0-623.0]) than in the BAT arm (101 days [range, 10.0-456.0]) (Figure). Crossover patients had responses consistent with those seen with RUX at the primary analysis: ORR at crossover day 28, 67.3%; durable ORR at crossover day 56, 42.9%.

Median FFS was significantly longer with RUX than with BAT (4.86 vs 1.02 months; P < 0.0001) (Figure). Median EFS (events: hematologic disease relapse/progression, graft failure, or death) was longer with RUX (8.18 vs 4.17 months).

Few patients experienced malignancy relapse/progression (RUX, 10.9% [16/147]; BAT, 17.0% [25/147]. Cumulative incidence of NRM indicated similar event rates over time in both arms, with NRM at 6 months being 37.7% with RUX and 42.4% with BAT. 29.2% of RUX patients and 18.7% of BAT patients developed cGVHD; fewer cases of severe cGVHD occurred in the RUX arm (RUX, n = 4; BAT, n = 7). At day 56, more RUX patients had tapered off corticosteroids (22.1% vs 14.8%). The EQ-5D-5L mean health rating at week 24 was higher with RUX (76.4 vs 63.1).

The safety profile of RUX was consistent with the primary analysis. The most common AEs (RUX vs BAT, exposure-adjusted rate per 100 patient-years) were anemia (181.7 vs 216.0), thrombocytopenia (158.6 vs 122.3), and cytomegalovirus infection (127.7 vs 169.8). AE rates in the crossover period were consistent with those in the RUX arm. There were 168 deaths on study, 82 (53.9%) in the RUX arm and 86 (57.3%) in the BAT arm.

Conclusions: RUX demonstrated sustained advantage over BAT; no new safety signals were observed with longer exposure to RUX. A lower probability of progression or addition of systemic therapy for aGVHD was observed at 6 and 12 months in the RUX arm.

Clinical Trial Registry: ClinicalTrials.gov identifier, NCT02913261

Disclosure: Mohamad Mohty, David Ritchie, Valérie Coiteux, and Yok-Lam Kwong have nothing to declare. Claude-Eric Bulabois has been a member of the board of directors or advisory committee for Astellas and Novartis and has participated in speakers bureaus for Kite/Gilead. Valentín García-Gutiérrez has served as a consultant for and has received honoraria and research funding from Novartis, Bristol Myers Squibb, Pfizer, and Incyte. Sung-Soo Yoon has served as a consultant for Amgen, Novartis, and Janssen; has received honoraria from Amgen and Novartis; and has received research funding from Kyowahako Kirin, F. Hoffmann-La Roche, Yuhan Pharmaceutical, and Genentech. Dietger Niederwieser has been a member of the board of directors or advisory committee for Cellectis, has participated in speakers bureaus for Novartis and Amgen, and has received research funding from Daiichi Sankyo. Reuven Or is employed by Hadassah Medical Center. Gérard Socié has been a member of the board of directors or advisory committee for Novartis, has participated in speakers bureaus for Novartis and Incyte, has served as a consultant for Alexion, has received honoraria from Elsalys and Alexion, and has received research funding from Alexion. Nikolas von Bubnoff has been a member of the board of directors or advisory committee, has participated in clinical biomarker research, has been a steering committee member for, and has received research support and funding from Novartis and has received honoraria from AstraZeneca. Robert Zeiser has received honoraria from Novartis, Incyte, and Mallinckrodt. Judith Xu, Juliane Morando, and Bruyère Mahuzier are employed by Novartis. Jeff Szer has been a member of the board of directors or advisory committee for, has served as a consultant for, and has received honoraria from Alexion.

O068. The Impact of GVHD on Outcomes After Adult Single Cord Blood Transplantation In European And Japanese Populations: The EUROCORD/ALWP EBMT And JSHCT/JDCHCT Collaborative Study

Junya Kanda¹, Hiromi Hayashi¹, Annalisa Ruggeri², Fumihiko Kimura³, Fernanda Volt², Satoshi Takahashi⁴, Shinichi Kako⁵, Karina Tozatto-Maio², Masamitsu Yanada⁶, Guillermo Sanz⁷, Naoyuki Uchida⁸, Emanuele Angelucci⁹, Seiko Kato⁴, Mohamad Mohty¹⁰, Edouard Forcade¹¹, Masatsugu Tanaka¹², Jorge Sierra¹³, Takanori Ohta¹⁴, Riccardo Saccardi¹⁵, Takahiro Fukuda¹⁶, Tatsuo Ichinohe¹⁷, Takafumi Kimura¹⁸, Vanderson Rocha², Shinichiro Okamoto¹⁹, Arnon Nagler²⁰, Yoshiko Atsuta²¹, Eliane Gluckman²

¹ Kyoto University, Kyoto, Japan, ² Eurocord, Paris, France, ³ National Defense Medical College, Tokorozawa, Japan, ⁴ University of Tokyo, Tokyo, Japan, ⁵ Jichi Medical University Saitama Medical Center, Saitama, Japan, ⁶ Aichi Cancer Center, Nagoya, Japan, ⁷ Hospital Universitario y Politécnico La Fe, Madrid, Spain, ⁸ Toranomon Hospital, Tokyo, Japan, ⁹ RCCS Ospedale Policlinico San Martino, Genova, Italy, ¹⁰ Hôpital Saint Antoine, Paris, France, ¹¹ CHU Bordeaux, Bordeaux, France, ¹² Kanagawa Cancer Center, Yokohama, Japan, ¹³ Hospital Santa Creu i Sant Pau, Barcelona, Spain, ¹⁴ Kitakyushu Municipal Medical Center, Kitakyushu, Japan, ¹⁵ Azienda Ospedaliera Universitaria Careggi, Firenze, Italy, ¹⁶ National Cancer Center Hospital, Tokyo, Japan, ¹⁷ Hiroshima University, Hiroshima, Japan, ¹⁸ Japanese Red Cross Kinki Block Blood Center, Ibaraki, Japan, ¹⁹ Keio University School of Medicine, Tokyo, Japan, ²⁰ Chaim Sheba Medical Center, Ramat Gan, Israel, ²¹ Japanese Data Center for Hematopoietic Cell Transplantation, Nagoya, Japan

Background: The impact of graft-versus-host disease (GVHD) and graft-versus-leukemia effect in unrelated cord blood transplantation (UCBT) is controversial.

Methods: In the present Eurocord/ALWP EBMT and the Japan Society for Hematopoietic Cell Transplantation/Japanese Data Center for Hematopoietic Cell Transplantation (JSHCT/JDCHCT) collaborative study, we evaluated the impact of GVHD on single UCBT outcomes in Japanese and European registries. Patients aged 18–75 years with acute leukemia who underwent the first allogeneic single UCBT and achieved engraftment were eligible for the study. The Cox proportional-hazards models were used to evaluate the effect of GVHD and confounding variables on overall survival (OS), leukemia-free survival (LFS), relapse and non-relapse mortality (NRM). In these regression models, the occurrence of acute and chronic GVHD was treated as a time-varying covariate.

Results: A total of 2,886 from the JSHCT/JDCHCT registry and 804 patients from the Eurocord/ALWP-EBMT registry were included. The median ages of the Japanese and European cohorts were 50 and 38 years, respectively. Three or more HLA mismatches in HLA-A, -B, and -DRB1 loci were more frequently observed in the JSHCT/JDCHCT cohort (23% vs. 3%). Median TNC counts were 2.58 and 3.51 x 107/kg in the JSHCT/JDCHCT and Eurocord/ALWP-EBMT cohorts, respectively. ATG was used in only 2% of the Japanese cohort compared with 66% of the European cohort. A multivariate analysis of OS revealed a positive impact of grade II acute GVHD compared with grade 0-I GVHD, in the Japanese cohort (HR, 0.81; P = 0.001), and an adverse impact in the European cohort (HR, 1.37; P = 0.007) (Figure 1). A negative impact of grade III-IV acute GVHD on OS was observed in both registries. In the analysis of relapse, a positive impact of grade II acute GVHD compared with grade 0-I GVHD was observed only in the Japanese cohort, regardless of disease risk. A multivariate analysis of NRM showed that an adverse impact of grade III-IV acute GVHD compared with grade 0-I GVHD was consistently observed in the Japanese cohort (HR, 2.97; P < 0.001) and in the European cohort (HR, 3.91; P < 0.001). A positive impact of limited chronic GVHD on OS and LFS was observed in the Japanese cohort (OS; HR, 0.51; P < 0.001, LFS; HR, 0.59; P < 0.001) only. An adverse impact of extensive chronic GVHD was observed in the European, but not in the Japanese cohort.

Conclusions: The deleterious impact of severe acute GVHD was observed in both registries. On the other hand, mild GVHD, i.e. grade II acute GVHD and limited chronic GVHD showed a beneficial effect in the Japanese cohort only. This could reflect ethnic differences in UCBT outcomes and might contribute to the preference of UCBT in Japan.

Disclosure: This work was supported in part by the Practical Research Project for Allergic Diseases and Immunology (Research Technology of Medical Transplantation) from the Japan Agency for Medical Research and Development, AMED (YA and JK) and JSPS KAKENHI Grant Number 18K08325 (JK).

O069. Impact of Chronic Gvhd Severity And Steroid Response on The Quality of Life In Patients Following Allogeneic Stem Cell Transplantation: Findings From A Real-World Study

Sylvie Lachance¹, Nada Hamad², Jonathan de Courcy³, Gregor Gibson³, Mike Zuurman⁴, Mohamad Mohty⁵

¹ Hôpital Maisonneuve-Rosemont and Université de Montréal, Montreal, Canada, ² St Vincent’s Hospital Darlinghurst, University of NSW, Darlinghurst, NSW, Australia, ³ Adelphi Real World, Bollington, United Kingdom, ⁴ Novartis Pharma AG, Basel, Switzerland, ⁵ Hôpital Saint-Antoine, Assistance Publique – Hôpitaux de Paris, Sorbonne Université, and INSERM Unité Mixte de Recherche 938, Paris, France

Background: Chronic GVHD (cGVHD) has a significant impact on the quality of life (QOL) of patients who undergo allogeneic stem cell transplantation. QOL should be systematically evaluated when considering optimal cGVHD treatment.

Methods: This cross-sectional survey collected data using physician-completed patient record forms and patient-completed forms. Physicians reported cGVHD severity (per NIH grading) and steroid response. Patients with cGVHD or their caregivers completed the EQ-5D-5L questionnaire, rated overall QOL on a 7-point scale ranging from very poor to very good, and reported current symptoms. Symptoms were scored from 0 (absent) to 4 (very prominent); the mean total symptom score (TSS) was calculated.

Results: Altogether, 143 patients with cGVHD from 5 countries completed the EQ-5D-5L. The median age was 54 (range, 21-76) years; 65.7% were male and 35.0% were steroid refractory/dependent. Overall, 59.4%, 29.4%, and 11.2% of patients had mild, moderate, and severe cGVHD, respectively. Most patients with severe or moderate cGVHD (81.3% and 45.2%, respectively) were steroid refractory/dependent compared with 21.2% of patients with mild disease. The median time since cGVHD diagnosis was 269 (IQR, 109.2-544.5) days.

QOL was assessed according to cGVHD severity and steroid response. While disease severity and lack of response to corticosteroids had significant negative effects on QOL, disease severity had a greater impact, with QOL decreasing progressively as cGVHD severity increased. According to EQ-5D-5L scores, patients with severe cGVHD had the worst QOL (0.58) vs patients with moderate (0.69) and mild disease (0.82) (Figure). A similar observation was seen using the EQ-5D-5L visual analog scale (VAS) (mild, 69.7; moderate, 59.2; severe, 43.9). Overall, 87.5% (14/16) of patients with severe cGVHD reported having poor (very poor, poor, somewhat poor) QOL vs patients with moderate (42.9% [18/42]) or mild disease (23.5% [20/85]).

When assessed by steroid sensitivity, the EQ-5D-5L/VAS scores were lower in steroid-refractory/dependent patients (0.69/57.1) than in steroid-responsive patients (0.79/67.3) (Figure). Poor QOL was reported in 44.0% (22/50) of steroid-refractory/dependent patients vs 32.3% (30/93) of steroid-responsive patients.

Three of the 5 EQ-5D-5L domains had the greatest impact on scores—62.0% of patients had problems performing usual activities, 79.0% experienced pain/discomfort, and 73.2% experienced anxiety/depression (Figure).

Mean TSS (symptom burden) was highest in patients with severe (30.4) and moderate cGVHD (20.6) vs those with mild disease (14.2). When assessed by steroid response, mean TSS was 21.0 in steroid-refractory/dependent patients and 16.3 in steroid-responsive patients. The mean total number of symptoms reported was 13.1, 11.0, and 9.0 in patients with severe, moderate, and mild cGVHD and 10.0 and 10.2 in steroid-responsive and steroid-refractory/dependent patients, respectively. While the mean number of symptoms was similar in both subgroup categories, patients with severe cGVHD experienced more severe symptoms (mean score, ≥2).

Conclusions: QOL in patients with cGVHD was predominantly affected by disease severity rather than by steroid response. Greater differences in EQ-5D-5L scores, overall symptom burden, and number of severe symptoms were observed when patients were categorized by disease severity. These results highlight the clinically significant impact of severe cGVHD on QOL and the need for novel preventive and curative therapies.

Disclosure: Silvy Lachance has been a consultant and an advisory board member for Novartis.

Nada Hamad has participated in advisory boards for Novartis.

Jonathan de Courcy and Gregor Gibson are employed by Adelphi Real World.

Mike Zuurman is employed by Novartis.

Mohamad Mohty has received lecturer and consulting fees from Novartis.

O070. Abstract already published

O071. Reduced Acute Graft-Versus Host Disease Incidence in Patients Receiving Anti-T Lymphocyte Globulin in Comparison to Anti-Thymocyte Globulin For The Prophylaxis of Graft-Versus-Host Disease After ALLO-HSCT

Arnaud Campidelli¹, Maud D’Aveni-Piney¹, Charline Moulin¹, Marie Detrait¹, Laura Boulangé², Gabrielle Roth-Guepin¹, Sylvie Tarillon¹, Nathalie Hottier¹, Helene Jeulin¹, Emma Verstraete¹, Anne Béatrice Notarantonio², Pierre Feugier¹, Marie T Rubio ¹

¹ CHRU Nancy, Vandoeuvre les Nancy, France, ² Lorraine University, Vandoeuvre les Nancy, France

Background: Both anti-T-lymphocyte globulin (ATLG-Grafalon) and anti-thymocyte globulin (ATG-Thymoglobulin) can prevent acute and chronic graft-versus-host-disease (GVHD) after allogeneic haematopoietic stem cell transplantation (HSCT). Because of distinct production characteristics, ATG and ATLG display distinct affinities to T and other immune cells in vitro. Whether these differences could impact post-transplant outcomes has never been explored.

Methods: In this monocentric retrospective study, we compared outcomes of 114 haemato-oncological patients who between January 2017 and September 2020 received peripheral blood stem cells transplant from matched HLA 10/10 related (MRD, n = 25) or unrelated donor (MUD, n = 89) and either ATG (5 mg/kg; n = 50) or ATLG (15 mg/kg for MRD and 30 mg/kg for MUD, n = 64) for GVHD prophylaxis. We also compared naive and memory T cell subsets, regulatory T and invariant NKT cell reconstitution after transplantation between the two groups.

Results: The comparison of patients’ characteristics at transplantation showed that the two groups were comparable in terms of disease type (64% of AML and MDS in both groups), disease risk index (intermediate in 68% vs 71% in the ATG and ATLG groups, respectively, p = 0.6), CMV status and of the donor (78% of MUD in both groups). Patients were older in the ATLG group (mean age 56 vs 51 years, p = 0.04) and received more frequently a reduced intensity/toxicity conditioning regimen (72% vs 46%, p = 0.01). Since ATLG was used more recently in our center, the median follow-up was longer in the ATG group (21 months vs 8.5 months, p < 0.001). At day 180, the cumulative incidence (CI) of grade II-IV aGVHD was 49% (95% CI 34-64%) in the ATG vs 25% (95% CI 12-33%) in ATLG group, (p = 0.005) (Figure 1). CI of CMV (p = 0.98) and EBV (p = 0.43) reactivations were comparable between the 2 groups. Incidence of chronic GVHD was similar in both groups (62% vs 72%, in the ATG vs ATLG groups at 2 years) with low incidences of moderate and severe cGVHD (29% vs 19 %, in the ATG vs ATLG groups at 2 years). At one year post-HSCT, non relapse mortality was comparable between the 2 groups (9% in ATG and 15% with ATLG, p = 0.42), while relapse incidence was higher in the ATG group (38% vs 12%, in the ATG vs ATLG groups, p = 0.016). At 2 years, extrapolated overall and GVHD-free, relapse free survival (GRFS) were not significantly different between the 2 groups (61% vs 71%, p = 0.45 and 54% vs 67% in the ATG vs ATLG groups, p = 0.74, respectively). In terms of immune reconstitution, we observed a more profound depletion of naive CD4 and CD8 T cells and reduced PD1 expression on CD8 T cells in patients receiving ATLG in comparison to those receiving ATG between day 14 and day 90 post-HSCT while levels of regulatory cells were similar.

Conclusions: Our results suggest that ATLG may better prevent the occurrence of severe aGVHD by differential effect on T cell reconstitution and activation, without increasing the incidence of relapse and of viral reactivation. These results should be confirmed in a prospective randomized trial.

Disclosure: MT Rubio has received honorarium from Neovii.

O072. Patient-Reported Outcomes in Acute Graft-Vs-Host Disease: Quality-Of-Life Findings From a Real-World Study

Nada Hamad¹, Silvy Lachance², Jonathan de Courcy³, Lou Rowaichi³, Mike Zuurman⁴, Mohamad Mohty⁵

¹ St Vincent’s Hospital Darlinghurst, University of NSW, Darlinghurst, Australia, ² Hôpital Maisonneuve-Rosemont and Université de Montréal, Montréal, Canada, ³ Adelphi Real World, Bollington, United Kingdom, ⁴ Novartis Pharma AG, Basel, Switzerland, ⁵ Hôpital Saint-Antoine, Assistance Publique – Hôpitaux de Paris, Sorbonne Université, and INSERM Unité Mixte de Recherche 938, Paris, France

Background: Quality of life (QOL) and symptom burden are important measures when studying the development of treatment for acute GVHD (aGVHD). Few studies have reported on aGVHD-associated symptom burden.

Methods: This cross-sectional survey collected data using physician-completed patient record forms and patient-completed forms. Physicians reported patient’s disease severity (per NIH grading) and steroid response. Patients with aGVHD or their caregivers completed the EQ-5D-5L questionnaire, rated overall QOL on a 7-point scale ranging from very poor to very good, and reported current symptoms. Symptoms were scored from 0 (absent) to 4 (very prominent). EQ-5D-5L scores, the mean number of symptoms, and mean total symptom score (TSS) were analyzed by grade and steroid response.

Results: Sixty-eight patients with aGVHD from 5 countries completed the EQ-5D-5L. The median age was 53 (range, 12-73) years; 66.2% were male. Overall, 39.7% of patients had grade ≥II aGVHD and 14.7% had grade III/IV aGVHD; 9.8%, 23.5%, and 70.0% of patients with grade I, II, and III/IV disease were steroid-refractory/dependent, respectively. Median time since aGVHD diagnosis was 45 (IQR, 23-96) days.

QOL was substantially lower in patients with higher aGVHD grades. The lowest EQ-5D-5L scores were seen in those with grades III/IV aGVHD (0.40) vs those with grade II (0.80) or grade I aGVHD (0.84) (Figure). Similarly, EQ-5D-5L visual analog scale (VAS) score was lowest in grade III/IV patients (grade III/IV, 48.5; grade II, 61.0; grade I, 65.5). Overall, 90.0% (9/10) of grade III/IV patients reported poor (very poor, poor, somewhat poor) QOL vs 41.2% (7/17) of grade II and 23.1% (9/39) of grade I patients.

QOL was lower in steroid-refractory/dependent patients than in steroid-responsive patients (EQ-5D-5L/VAS score, 0.53/55.0 vs 0.83/63.8, respectively) (Figure). Poor QOL was reported in 60.0% (9/15) of steroid-refractory/dependent patients vs 31.4% (16/51) of steroid-responsive patients. Furthermore, steroid-dependent patients had a much lower QOL than steroid-responsive patients (EQ-5D-5L score, 0.54 vs 0.83; poor QOL, 61.5% vs 31.4%).

Three of the 5 EQ-5D-5L domains had the greatest impact on scores—60.9% of patients had problems performing usual activities, 76.6% experienced pain/discomfort, and 64.1% experienced anxiety/depression (Figure).

The mean TSS (symptom burden) increased with grade and was 13.6, 20.9, and 33.6 in patients with grade I, II, and III/IV aGVHD, respectively. Mean TSS was also higher in steroid-refractory/dependent patients than in steroid-responsive patients (31.7 vs 14.6). The mean number of symptoms increased with grade (grade I, 8.6; grade II, 10.2; grade III/IV, 14.9) and was higher in steroid-refractory/dependent patients than in steroid-responsive patients (14.2 vs 8.7, respectively). Patients with grade ≥II aGVHD experienced more severe (mean score, ≥2) symptoms (nausea, diarrhea, indigestion, weight loss, itchy skin, and decreased libido). Steroid-dependent/refractory patients also experienced severe symptoms (diarrhea, indigestion, weight loss, and decreased libido); no steroid-responsive patients reported severe symptoms.

Conclusions: QOL is poor in patients with grade III/IV aGVHD. Higher disease severity and inadequate response to steroids had the most negative effects. Despite responding to steroids, steroid-dependent patients also had a lower QOL than steroid-responsive patients. This study highlights the need for improved first and second-line treatments.

Disclosure: Nada Hamad has participated in advisory boards for Novartis. Silvy Lachance has been a consultant and an advisory board member for Novartis. Jonathan de Courcy and Lou Rowaichi are employees at Adelphi Real World. Mike Zuurman is employed by Novartis. Mohamad Mohty has received lecturer and consulting fees from Novartis.

O073. Abstract already published

Graft-versus-host Disease – Preclinical and Animal Models

O074. Abstract already published

O075. Microbial-Derived Metabolites Induce Epithelial Recovery Via The Sting Pathway in Mice and Men and Protect From Graft-Versus-Host Disease

Erik Thiele Orberg¹, Sascha Göttert¹, Alix Pianka¹, Andreas Hiergeist², Andre Gessner², Simon Heidegger¹, Wolfgang Herr², Hendrik Poeck^1,3,2

¹ University Hospital rechts der Isar of the Technical University Munich, Munich, Germany, ² University Hospital Regensburg, Regensburg, Germany, ³ Nationales Centrum für Tumorerkrankungen, Regensburg, Germany

Background: Graft-versus-host disease (GVHD) is a dreaded complication after allogeneic hematopoietic stem-cell transplantation (allo-HSCT). Previously, we and others showed that activation of IFN-I inducing pathways such as RIG-I/MAVS or cGAS-STING can promote the integrity of the gastrointestinal (GI) barrier and limit GVHD. However, the signals that drive these protective IFN-I responses are poorly understood.

Commensal microbiota can (i) have distant effects on immune responses through modulation of IFN-I signalling and (ii) predict mortality in allo-HSCT patients. We hypothesized that microbiota-derived products such as microbial metabolites facilitate IFN-I signalling in immune and non-immune cells poising them for induction of protective responses. We established a prospective, observational clinical study in patients newly diagnosed with acute leukemia and performed longitudinal stool sampling to track targeted metabolomic profiles and microbiome. Here, we highlight a patient’s diverse microbiome and abundant metabolite profile at initial diagnosis, which is lost following allo-HSCT. To validate our clinical observations, we tested patient-detected metabolites in preclinical GVHD models and intestinal organoids.

Methods: Human stool samples from healthy controls and allo-HSCT patients (n = 20) were obtained at our Department and at the University Hospital Regensburg in accordance to IRB-approved study protocols. Patients were sampled at initial diagnosis (Dx), and weekly after allo-HSCT up to day 28. Samples were analysed by 16S, ITS sequencing and mass spectrometry for metabolomic profiling. Next, we tested metabolites which we detected in patients (desaminotyrosine [DAT], indole-3-carboxaldehyde [ICA]) as treatment in preclinical models of (i) acute chemotherapy or radiation-induced GI damage and (ii) allo-HSCT. Outcomes were assessed by clinical scoring, histology, flow cytometry (neutrophil influx, intraepithelial T cell infiltration) and organoid recovery. To obtain a mechanistic understanding of the signalling pathways involved, we stimulated WT or IFN-I-signalling-impaired mouse (incl. STING^−/−, MAVS^−/−, IFNαR^−/−) as well as human crypt-derived organoids with metabolites. Analysis was performed by microscopy, qPCR and flow cytometry.

Results: We present a 64-year-old female patient diagnosed with AML who received a 9/10 HLA-matched allo-HSCT. At day 7, i.v. antibiotics were started due to fever and Enterococcus bacteraemia. At day 15, the patient developed skin and GI GVHD (Glucksberg III). At initial diagnosis (Dx), i.e. the timepoint when we diagnosed AML but before therapy was initiated, we detected rich alpha diversity (Figure 1a). Flavonifractor plautii, a producer of the metabolite DAT, was enriched (Figure 1b). We observed high-level expression of metabolites including short-chain fatty acids, DAT, ICA and secondary bile acids (Figure 2, Dx). Following allo-HSCT, diversity and metabolite expression declined drastically (Figure 2). Next, we prophylactically administered metabolites to murine allo-HSCT recipients. Metabolite-treated mice showed significantly reduced inflammation and improved recovery of intestinal stem cells following allo-HSCT (Figure 3), an effect that was abrogated in STING^−/− recipients. Metabolite stimulation of mouse and human organoids promoted organoid numbers and size, and required intact STING signalling (Figure 4).

Conclusions: We show for the first time that microbial-derived metabolites detected in patients can engage the STING pathway in humans and mice to confer resistance from chemotherapy, radiation and immune damage. Prophylactic administration of metabolite cocktails may reduce occurrence of GVHD in allo-HSCT patients.

Disclosure: Nothing to declare.

Haematopoietic Stem Cells

O076. Outcome of Non-Myeloablative Autologous Hematopoietic Stem Cell Transplantation for Multiple Sclerosis: A Single Center Summary of 511 Patients

Richard K Burt¹, Xiaoqiang Han¹, Kathleen Quigley¹, Roumen Balabanov¹

¹ Northwestern University, Chicago, United States

Background: The MIST randomized trial of 100 patients (55 in each arm) of nonmyeloablative hematopoietic stem cell transplantation (HSCT) versus continued disease modifying therapy (DMT) for relapsing remitting multiple sclerosis (RRMS) demonstrated marked superiority for HSCT. However, no large trial of non-myeloablative HSCT for RRMS has been published to date.

Methods: Five hundred and eleven patients at Northwestern University between 7-2003 and 10-2019 underwent non-myeloablative HSCT using a non-selected graft and cyclophosphamide (Cy) / rabbit anti-thymocyte globulin (ATG) (n = 376), Cy / ATG / rituximab (n = 63), Cy / ATG / intravenous immunoglobulin (IVIG) (n = 46), or Cy / alemtuzumab (n = 26). Either an intravenous cephalosporin cefepime (maxipimeâ) or piperacillin / tozobactam (zosyn®) was started on day 0 and continued until engraftment that usually occurred on day 9 or 10.

Results: There was one treatment related death (0.19%, 1/511) due to legionella pneumonia acquired from the hospital room shower head. Only one of 511 patients developed bacteremia (Klebsiella pneumonia). Overall survival was 98.6%. Six patients had late non-treatment related deaths, a cerebrovascular accident related to medication non-compliance, a myocardial infarction, and during an elective cholecystectomy (unknown cause); colon cancer; and one died of respiratory failure from relapsed NMO (misdiagnosed as MS). The secondary autoimmune diseases (2ndADs) that occurred post HSCT were idiopathic thrombocytopenia (ITP) and hypo or hyperthyroidism. The incidence of ITP was highest with alemtuzumab (14%) and 0 to 2.8% for the non-alemtuzumab regimens. Before HSCT, hypothyroidism was present in 10.4% (n = 53) and hyperthyroidism in 1.5% (n = 8) of patients. After HSCT, 16 patients developed hypothyroidism (3.5% 16/450 at risk) and 15 developed hyperthyroidism/Grave’s disease (3.3%).

Probability of relapse free survival at 6 months, and 1, 2, 3,4, and 5 years after HSCT was 99%, 96%, 94%, 92%, and 89.5%, respectively (figure 1A). Disability measured by the Expanded Disability Status Scale (EDSS) improved for the entire group (figure 1B) from a mean before HSCT of 4.0, to 3.04, 2.95, 2.96, 2.86, 2.78, and 2.54 at 6 months and 1, 2, 3, 4, and 5 years, respectively (all p < 0.00001). Patients treated with different non-myeloablative conditioning regimens had similar post improvements in baseline EDSS (figure 1 B).

Conclusions: Autologous non-myeloablative HSCT provides drug free reversal of neurologic disability of greater than 1.0 EDSS points for greater than 5 years in 89% of patients with RRMS. The low incidence of bacteremia may be related to the use of non-myeloablative regimens that do not cause mucositis, i.e. gut endothelium bacterial barrier remains closed, placement and removal of the large bore apheresis catheter on the same day and subsequent insertion of a PICC line in the upper inner arm on day of admission resulting in less risk of cutaneous barrier breaches, a HEPA-filtered floor, and to the use of pre-emptive antibiotics when neutropenic. On the other hand, one patient developed legionella pneumonia that was subsequently cultured from the showerhead. Control over the hospital water supply is therefore of major importance. Care of these patients requires thoughtful monitoring, documentation, and prophylaxis of infections and 2ndADs.

Disclosure: Nothing to declare.

O077. The Impact of CD3 And CD34 Cell Dose on Graft Rejection And Graft Versus Host Disease in Children Undergoing Unrelated Peripheral Blood Stem Cell Transplantation

Satish Meena¹, Harika Varla¹, Rumesh Chandar¹, Venkateswaran Vellaichamy Swaminathan¹, Balasubramanium Ramakrishnan¹, Ramya Uppuluri¹, Indira Jayakumar¹, Revathi Raj¹

¹ Apollo Cancer Institute, Chennai, India

Background: There is an increasing trend in the use of peripheral blood stem cells (PBSC) from matched unrelated donor registries due to the ease of collection from the donor’s perspective. The use of PBSC results in higher risk of graft versus host disease (GVHD). In our study, we analyzed the impact of CD34 and CD3 cell doses per recipient body weight in the graft in matched unrelated donor transplantation with respect to overall survival, graft versus host disease and graft rejection.

Methods: We performed a retrospective analysis of children who underwent hematopoietic stem cell transplantation in our unit from June 2012 to June 2020. The data analyzed included the number of CD34 and CD3 cells in the PBSC product infused per kilogram of recipient body weight and its impact on graft rejection, GVHD and survival.

Results: We included 117 of the total 133 children who underwent unrelated PBSC transplantation performed at our center for whom complete follow up data is available. The male: female ratio was 1.6: 1 and the predominant indication for HSCT was benign haematological disorders at 74.4% and 25.6% for a malignancy. The donor was fully HLA matched (10/10) in 83.3% children, while a mismatched donor was used in 16.7% children. A myeloablative conditioning was used in 92.3% children and reduced intensity conditioning in 7.9% children. The median CD34 dose infused was 5.2 x 10⁶/kg and graft rejection was seen in 5.1% children. A CD34 count of less than 4.02 x 10⁶ and a CD3 dose of less than 0.71 x 10⁸ was documented in these patients. We documented GVHD in 82.1% children in which 37.5% had acute GVHD and 62.5% had chronic GVHD. Skin was the most common organ affected in GVHD in 43.75% followed by gut at 19.7%. Acute GVHD extending into chronic GVHD was present in 14.5% cases. Grade 1 to 2 GVHD was present in 53.1% cases while grade 3 to 4 GVHD was present in 46.8% cases. The dose of CD3 was higher than 1.69 x 10⁸ in children who developed GVHD with higher mortality when the dose of CD3 cells was above 1.73 x 10⁸. The overall survival was 73.5% and the most common cause of mortality was GVHD (12.8%). This was followed by infection (7.7%), relapse (4.3%) and rejection (1.7%).

Conclusions: The use of PBSC results in a higher rate of graft versus host disease and a low incidence of graft failure in unrelated donor HSCT in children. Our study highlights the importance of the dose of CD3 in the graft with a cell dose of less than 0. 71 x 10⁸ resulting in graft rejection while doses higher than 1.69 x 10⁸ showing a higher incidence of GVHD with a significant mortality due to GVHD in a dose above 1.73 x 10⁸. This data can be utilized to achieve the fine balance between graft rejection and graft versus host disease and reduce mortality whilst continuing to use PBSC as a graft source.

Disclosure: Nothing to declare.

O078. Outcome And Immune Reconstitution After TCRΑΒ/CD19 Cell Depleted Hematopoietic Stem Cell Transplantation From Haploidentical Related Donor in Adult Leukemia – A Single Centre Experience

Lucia Prezioso¹, Ilenia Manfra², Ilaria Bertaggia³, Benedetta Cambò¹, Benedetta Dalla Palma¹, Federica Falcioni¹, Elena Follini⁴, Maria Teresa Giaimo¹, Amelia Rinaldi¹, Daniele Vallisa⁴

¹ AOU Parma, Parma, Italy, ² AO Avellino, Avellino, Italy, ³ Azienda USL Toscana Nord Ovest UOC, lido di Camaiore, Italy, ⁴ AO Piacenza, Piacenza, Italy

Background: HLA-haploidentical transplant have become a strong alternative to matched donor. For years it has been burdened by a high incidence of GVHD and delayed immunological reconstitution. In the last years, thanks to the implementation of new transplant manipulation strategies and improvement of pharmacologic GVHD prophylaxis, this limitation has been partially overcome. Through ex vivo T cell depletion strategies, selective elimination of αβ+T cells retains in the graft dendritic cells, monocytes and above all NK and γδT lymphocytes, leading to a rapid and effective immunological reconstitution and to very promising outcome in terms of tumour and infection control.

Methods: Since September 2012 to august 2020, 45 adult patients, median age 55 years (19-72) affected by acute myeloid(37), lymphoblastic(7) and plasmacellular (1) leukemia were treated. Twenty-nine patients were in Complete Remission (CR:19 RC1 and 10 RC2) and 16 in advanced disease. Myeloablative conditioning.

Results: Grafts contained a median of 10.4 x 10⁶/kg(range 5-19) CD34+ cells, 5.2 x 10⁶ CD3+Tcells/kg(range 1-36), 4.7 x 10⁴/kg(range 0.4-62) αβ+T cells, 4.7 x 10⁶ γδ+Tcells/kg(range 1-34), 3.5 x 10⁴ B cells/kg(range 1.5-88) and 24 x 10⁶ CD56+NKcells/kg(range 5-440). All patients but one, who was successfully saved through a second allogenic transplant, achieved primary engraftment. Median time to reach 500 neutrophils and 20,000 platelets was of 12 and 11 days, respectively. Relapse occurred in 31% (9/16 in relapse, 5/29 in CR).OS was 60% in patients in CR1/CR2 at a median follow up of 26 months. NRM was analyzed according to age and combination of EBMT and HCT-CI score. The cumulative incidence of skin-only, grade 1-2 acute GVHD was 10%; only two patients developed fatal grade IV aGVHD, as well as chronic GVHD. CMV reactivation was documented in 16 patients and only one died due to organ disease. Three patients experienced EBV reactivation,1 Adenovirus,1 toxo, 5 HHV6 and 2 VZV, that completely resolved. Only one patient had a fungal infection. Two patients died due to Stenotrophomonas infection post endoscopic exam. T and B lymphocytes recovery was fast, and immunoglobulin serum levels normalized within 3 months. In the first months posttransplant we documented an expansion of T memory lymphocytes and of mature CD56dim/CD16+NK cells, likely of donor origin transfered to the recipient and exerting an antileukemia and antinfective function. Detection of naïve T lymphocytes as well as of the TREC (T-cell receptor excision circles) and immature CD56brightCD16dim NK cells began at 180 days post transplant. γδ T cells subpopulation showed an expansion of Vδ1 in patients with active viral infection, above all CMV, that impacted also on the expansion of “mature” phenotype NK cells, CD56dimCD16+, CD57+NKG2C+ and CD16+CD57+ NK cells. Moreover, in two patients significant expansion of pathogen-specific CD8+T cells was documented at day +30 and +54, by flow cytometry with the destramer CMV kit and it contributed to clear viral load spontaneously.

Conclusions: These data indicate that haplo-HSCT after αβT- and B-cell depletion represents a competitive option for adult patients with haematological diseases, who lack a compatible donor or are in need of urgent allograft.

Disclosure: No disclosure.

O079. Non-T Depleted Haploidentical Stem Cell Transplantation in Aml Patients Achieving First Complete Remission After One Vs Two Induction Courses: A Study From The ALWP/EBMT

Arnon Nagler¹, Myriam Labopin², Xiao-jun Huang³, Didier Blaise⁴, William Arcese⁵, Mercedes Colorado Araujo⁶, Gerard Socié⁷, Edouard Forcade⁸, Fabio Ciceri⁹, Jonathan Canaani¹, Sebastian Giebel¹⁰, Eolia Brissot¹¹, Jaime Sanz¹², Ali Bazarbachi¹³, Ibrahim Yakoub-Agha¹⁴, Mohamad Mohty¹¹

¹ Sheba Medical Center, Ramat Gan, Israel, ² Saint Antoine Hospital, Paris, France, ³ Peking University People´s Hospital, Institute of Haematology, Beijing, China, ⁴ Centre de Recherche en Cancérologie de Marseille, Institut Paoli Calmettes, Marseille, Switzerland, ⁵ ¨Tor Vergata¨ University of Rome,Policlinico Universitario Tor Vergata, Rome, Italy, ⁶ Hospital U. Marqués de Valdecilla, Santander, Spain, ⁷ Hopital St. Louis, Paris, France, ⁸ CHU Bordeaux, Hôpital Haut-Leveque, Pessac, France, ⁹ Ospedale San Raffaele s.r.l., Milano, Italy, ¹⁰ Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice, Poland, ¹¹ Hôpital Saint Antoine, Paris, France, ¹² University Hospital La Fe, Valencia, Spain, ¹³ American University of Beirut Medical Center, American University of Beirut Medical Center, Beirut, Lebanon, ¹⁴ CHU de Lille, Lille, France

Background: Achieving a first complete remission (CR1) is the primary goal in the treatment of AML and is an important prognostic factor for transplantation outcome. However, there are no data in the setting of non-T-cell-depleted haploidentical stem cell Transplantation (HaploSCT) indicating whether the number of chemotherapy courses (1 vs 2) needed to achieve CR1 is of prognostic significance.

Methods: Using the EBMT/ALWP registry, we compared transplantation outcomes of adult patients (pts) aged ≥18 year with AML that underwent HaploSCT in 2005-2019 in CR1 achieved following 1 vs 2 chemotherapy courses. Multivariate analysis (MVA) adjusting for differences between the groups were performed using Cox’s proportional- hazards regression model for main outcomes.

Results: 635 pts were included: 469 (74%) with 1 and 166 (26%) with 2 induction chemotherapies courses. Median follow-up was 28.7 (25.7-33.8) and 32.7 (29.2-37.6) months, respectively (p = 0.53). Median age was 45.3 (18.0-75.0) and 45.8 (18.3-72.2) year (p = 0.46); 58.4% of both groups were male. 429 (91.5%) and 151 (91%) pts had de novo and 40 (8.5%) and 15 (9%) secondary AML (p = 0.84). Pts with 1 and 2 induction courses, were classified by cytogenetic risk as follows: intermediate, 69.5% and 75.9%, adverse, 22.4% and 20.5%, and favorable-risk, 8.1% and 3.6% (p = 0.11). Significantly more pts were FLT3+ in the 1 induction group (24.5% and 15.9%, p = 0.046), but in 32% data was missing. Conditioning was myeloablative in 59.3% and 62.7% and reduced intensity in 40.7% and 37.3%, respectively (p = 0.45). Grafts were peripheral blood (PB) in 41.6% and 39.2%, bone marrow (BM) in 27.5% and 36.1% or PB+BM in 30.9% and 24.7% (p = 0.09). Karnofsky performance score (KPS) was >90 in 76.6% and 79.8% of pts (p = 0.47). The most frequent anti-graft-versus-host disease (GVHD) prophylaxis was post-transplant cyclophosphamide (PTCy) in 52.7% and 54.8%, anti-thymocyte globulin (ATG) in 40.5% and 38%, or both in 6.8% and 7.2% (p = 0.84). Engraftment rates were 97.2% and 97.6%. Day 180 incidence of acute GVHD II-IV and III-IV was similar in both induction groups (31.1% and 34.8%, and 10% and 10.6 %), as was 2-year total and extensive chronic GVHD (33.7% and 36.5 %, and 12.2% and 12.1%), respectively. Two -year relapse incidence (RI) was higher while leukemia free survival (LFS), overall survival (OS) and GVHD-free, relapse-free survival (GRFS) were inferior for pts achieving CR1 with 2 vs 1 course and were 29.1% vs 15.1%, (p = 0.001), 56.2% vs 66.9% (p = 0.03), 58.8% vs 72.2% (p = 0.044) and 44% vs 55.6% (p = 0.013), respectively. Non-relapse mortality (NRM) did not differ, 18% vs 14.6% (p = 0.25). These results were confirmed by MVA, hazard ratio (HR) = 1.97 (1.34-2.88, p = 0.0005), HR = 1.35 (1.01-1.81, p = 0.045), HR = 1.37 (1-1.86, p = 0.048), HR = 1.38 (1.07-1.78, p = 0.013), HR = 0.81 (0.5-1.32, p = 0.4) for RI, LFS, OS, GRFS and NRM, respectively.

Conclusions: The relapse rate was almost 2-fold higher and transplantation outcomes significantly inferior in pts with AML undergoing HaploSCT in CR1 but who received 2 lines of chemotherapy to achieve CR1. These pts may benefit from additional novel therapies in the conditioning or post-transplant in an attempt to reduce their high relapse rates and improve outcome.

Disclosure: Nothing to declare.

O080. Impact of Allogenic Transplantation In The Treatment of Secondary Acute Myeloid Leukemias. Comparison Between Myeloablative Vs Reduce Intensity Conditionings

Claudia Núñez-Torrón¹, Carlos Jiménez¹, Juan Marquet¹, Alejandro Luna¹, Adolfo Sáez¹, Lucia Rubio¹, Valentin García¹, Anabelle Chinea¹, Ana Jiménez¹, Gemma Moreno¹, Javier López¹, Pilar Herrera¹

¹ Hospital Universitario Ramón y Cajal, Madrid, Spain

Background: Patients with Secondary AML (s-AML) are a high-risk subgroup in terms of survival. Among patients candidates to intensive chemotherapy, consolidation with HSCT is postulated as de unique curative therapy.

Methods: Retrospective analysis of 90 patients transplanted in our center since 2011 to 2019 that achieved complete remission before HSCT. We divided our cohort into 2 groups: Group 1, 46 patients with de novo leukemia and Group 2, 44 patients with s-AML (including AML-MRC, t-AML and blastic phase of MPN Ph negative). The baseline characteristics of each group were compared using the Chi2 test for qualitative and T-student for quantitative variables. The survival analysis was performed through Kaplan-Meier method and the risk was calculated with Cox regression.

Results:

Variable	De novo leukemia n = 46	Secondary leukemia n = 44
Median age at HSCT, years (range)	48.5 (21-68)	58.5 (28-69)
ELN classification, n (%)
Favorable Risk	15 (32.6%)	3 (6.8%)
Intermediate Risk	16 (34.8%)	23 (52.3%)
Adverse Risk	15 (32.6%)	18 (40.9%)
State disease prior to HSCT, n (%)
CR with MRD- by flow cytometry	33 (76.7%)	30 (73.2%)
CR with MRD+ by flow cytometry	11 (23.3%)	11 (26.8%)
Conditioning intensity, n (%)
Myeloablative	34 (73.9%)	19 (43.2%)
Reduce intensity	12 (26.1%)	25 (56.8%)
Donor type, n (%)
Related donor	21 (45.6%)	10 (22.7%)
Unrelated donor	13 (28.3%)	15 (34.1%)
Haploidentical donor	12 (26.1%)	19 (43.2%)

Baseline characteristics of entire cohort are reflected in table 1. Patients with s-AML were significantly older, had less favorable risk profile at diagnosis, and the use of RIC was more common. The median follow-up was 16 months (0-77). The 2 year- EFS (2y-EFS) was 47% and the 2y-OS 57% in entire cohort. The 2y-EFS in de novo group was 61% and 33% in s-AML [HR 2.4, 95% CI 1.3-4.4)] and 2y-OS was 74% vs 41% respectively [HR 2.7, 95% CI, 1.4-5.4)].In global cohort, the 2y-EFS was 55% with the use of Myeloablative (MA) and 36% with Reduce Intensity Conditioning (RIC) [HR 1.9, 95% CI 1.2-3.4)] and the 2y-OS was 64% and 48% respectively [HR 2, 95% CI 1-3.7)]. Among patients that received MA the 2y-EFS in patients with de novo leukemia was 60% and 45% in s-AML group [HR, 1.4, 95% CI (0.6-3.4)], and the 2y-OS was 72% and 49.5% [HR 2.2, 95% CI (0.8-5.6)].Among patients that received RIC the 2y-EFS was 64% in Group 1 and 22% in Group 2 [HR 3.6, 95% CI (1.-2-10.7)] and the 2y-OS 74% in de novo vs 35% in s-AML [HR 2.7, 95% CI (0.9-8.1)] (Picture 1). In univariate analysis the characteristics with a significant HR for EFS and OS were s-AML, age>60, ELN classification, MRD+ before HSCT and the use of RIC. The variables with a significant HR in the multivariate analysis were age >60 for OS and s-AML for both EFS and OS.

Conclusions: Patients with s-AML are a high-risk subgroup compare to de novo leukemia patients with worse survival after transplantation. The use of MA conditioning could overcome part of the high risk of s-AML comparing to de novo patients. However, the results with reduced intensity conditioning in s-AML are worse than when used in de novo leukemias.

Clinical Trial Registry: No applicable.

Disclosure: Nothing to declare.

O081. The Disease Burden Prior to Allogeneic Hematopoietic Stem Cell Transplant is More Relevant in Patients With De Novo Leukemia Compare to Patients With Secondary Leukemia

Claudia Núñez-Torrón¹, Carlos Jiménez¹, Juan Marquet¹, Fernando Martín¹, Jesus Villarrubia¹, Miguel Piris¹, Adolfo Sáez¹, Alejandro Luna¹, Ana Jiménez¹, Anabelle Chinea¹, Valentín García¹, Gemma Moreno¹, Javier López¹, Pilar Herrera¹

¹ Hospital Universitario Ramón y Cajal, Madrid, Spain

Background: In AML, the persistence of disease prior to transplantation has an impact on the results. We want to explore if the disease burden with which the patient reaches the transplant is also important in S-AML.

Methods: Retrospective analysis of 110 patients with AML transplanted in our center between 2011 and 2019, all with a valuable bone marrow sample before HSCT. We divided our cohort into two groups: Group 1 n = 49 patients with de novo leukemia and Group 2 n = 61 patients with s-AML (including AML-MRC, t-AML and blastic phase of MPN Ph negative). We analyzed the differences in terms of Event-Free Survival (EFS) and Overall Survival (OS) since the day of infusion depending on the burden disease before HSCT.

Results:

Variable	De novo leukemia n = 49	Secondary leukemia n = 58
Sex, male n (%)	29 (59.2%)	37 (64.7%)
Median age at HSCT, years (range)	49 (21-68)	57 (20-69)
ELN classification, n (%)
Favorable Risk	14 (28.6%)	3 (4.9%)
Intermediate Risk	20 (40.8%)	31 (50.8%)
Adverse Risk	15 (30.6%)	27 (44.3%)
Complete remission after induction (1 or 2 cycles), n (%)	42 (89.4%)	45 (73.8%)
State disease prior to HSCT, n (%)
CR	44 (89.8%)	46 (75.4%)
Active disease	5 (10.2%)	15 (24.6%)
Conditioning intensity, n (%)
Myeloablative	33 (67.3%)	26 (42.6%)
Reduce intensity	12 (24.5%)	28 (45.9%)
Sequential reduced intensity	4 (8.2%)	7 (11.5%)

Baselines characteristics of global cohort are reflected in Table 1. Patients with s-AML were significantly older, with more adverse profile by the ELN and more refractoriness after induction and a non-significant tendency of more Active Disease (AD) before HSCT. Also in patients with s-AML RIC were more used than MAC than de novo patients. With a median follow-up of 12 months (0-77), the median EFS (mEFS) and OS (mOS) in global cohort were 13 and 24 months (m), respectively. Comparing both groups the mEFS was NR in Group 1 vs 13m in Group 2 [HR 2.6, 95% CI (1.5-4.6)] The mOS in Group 1 was NR vs 10m in Group 2 [HR 2.5, 95% CI 1.5-4.3)]. We divide global cohort into 2 groups according to disease state before HSCT by cytology: Complete Remission (CR) and AD. In global cohort, we found a mEFS 23m in CR patients and 4 months in patients with AD [HR 3.3, 95% CI (1.9-6)]. The mOS was 46m vs 8 m respectively [HR 3.2, 95% CI (1.8-5.5)]. Among de novo leukemia patients, the mEFS was NR in CR and 2m in AD patients [HR 8.2, 95% CI (2.5-27.5)], and the mOS was NR CR patients while in AD group was 2m [HR 4.8, CI 95% 1.5-14.7)]. Among s-AML patients, the mEFS was 8m in CR patients and 5m in AD [HR 2.2, 95% CI (1.2-4.1)], and the mOS was 17 months vs 9m respectively [HR 1.9, 95 CI (0.9-3.7)] (Figure 1).

Conclusions: The post-transplantation survival in patients with s-AML is worse compare to de novo AML patients. While in both groups, patients transplanted with AD had extremely poor prognosis, the benefit to achieve CR before HSCT was lower in s-AML patients.

Clinical Trial Registry: No applicable.

Disclosure: Nothing to declare.

O082. The Impact Of Platelet Recovery on Survival in Children Undergoing Hematopoietic Stem Cell Transplantation

Rumesh Chandar¹, Venkateswaran Vellaichamy Swaminathan¹, Harika Varla¹, Satish Kumar Meena¹, Ramya Uppuluri¹, Balasubramaniam Ramakrishnan¹, Indira Jayakumar¹, Revathi Raj¹

¹ Apollo Cancer Institutes, Chennai, India

Background: Platelet reconstitution on day 100 after hematopoietic stem cell transplantation (HSCT) is associated with better survival outcomes. Thrombocytopenia after platelet engraftment can be attributed to sepsis, relapse of the disease, drugs and graft versus host disease (GVHD). In this study, we aim to analyze the impact of day 100 platelet count on disease free survival in children undergoing HSCT for blood disorders.

Methods: We performed a retrospective analysis of children who underwent hematopoietic stem cell transplantation in our blood and marrow transplantation unit between November 2002 and November 2018 with a minimum follow up period of 24 months. The parameters analysed were the underlying diagnosis, conditioning regimen, graft source and overall survival. The children who died before day 100 were excluded from the analysis. Platelet recovery was defined as early (EPR) and delayed (DPR) depending on whether the day 100 platelet count was more than and less than or equal to 100,000 per cu.mm respectively.

Results: A total of 626 children were included in the study with 65% male and 35% female. The indication for HSCT was a benign haematological disorder in 497 (79%) and a malignancy in 129 (21%) children. A myeloablative conditioning regimen was used in 536 (85%) children and a reduced intensity conditioning in 90 (15%) children. The predominant source of stem cells was peripheral blood stem cells in 475 (75%) children, followed by bone marrow in 139 (23%) and cord blood in 12 (2%) children. Platelet recovery served as a good predictor of survival across different diagnoses, conditioning regimens and graft source. At a median follow up of 24 months, the overall survival of the cohort was 75%. We documented EPR in 475 (79%) children. The overall survival in the EPR group was 91% versus the DPR group at 73%. DPR was associated with poor overall survival (p < 0.001). In the group with DPR, the main causes of mortality included graft versus host disease in 13 (36%), disease relapse in 10 (27%) and sepsis in 7 (19%).

Conclusions: Our study highlights the impact of day 100 platelet count on the overall survival in children undergoing HSCT for both benign and malignant blood disorders. Platelet recovery is a simple and easily available tool to predict graft versus host disease, relapse and sepsis and serves a guide to introduce early intervention early to reduce mortality. Documentation of day 100 platelet count must be incorporated in all long term follow up records as it is a sensitive predictor of children at risk.

Disclosure: Nothing to declare.

O083. Hematopoietic Stem Cell Transplant As Treatment of Refractory Myasthenia Gravis

Claudia Lucía Sossa Melo^1,2, Angela María Peña^1,2, Luis Antonio Salazar^2,3, Manuel Rosales^1,3, Sara Ines Jimenez^1,2, Maria Luna-Gonzalez^2,3, Jose Patricio Lopez², Manuel Ardila-Baez², Daniela Vásquez², Ivan Mauricio Peña^1,2

¹ Clínica FOSCAL, Bucaramanga, Colombia, ² Universidad Autónoma de Bucaramanga, Bucaramanga, Colombia, ³ Programa para el Tratamiento de Enfermedades Hemato-Oncoloógicas de Santander, Bucaramanga, Colombia

Background: Refractory Myasthenia Gravis (RMG) is an antibody-mediated disease that affects the neuromuscular junction. The cases of patients with RMG undergoing Hematopoietic Stem Cell Transplantation (HSCT) are few, therefore the information about survival is scarce. The aim of this report is to show the survival of two patients with a history of RMG who underwent HSCT.

Methods: A case report was made of two patients diagnosed with RMG who underwent HSCT in Clinica FOSCAL, at a university hospital in Colombia, between 2010-2020.

Results: The first case is a 65-year-old male patient with a history of arterial hypertension, diagnosed at age 53 with RMG and in 2012 at age 57 he underwent HSCT. The treatment lines between diagnosis and HSCT were Pyridostigmine 180mg/day, Plasmapheresis regular use every month and in acute exacerbations. The treatment regimen at the time of the HSCT was performed with Pyridostigmine 180mg/day, Prednisolone 25mg/day, Azathioprine 100mg/day, and Mycophenolate mofetil 500mg/day. According to the mobilization and collection regimen, 2 days of collection were recorded, with a cell count of 7.0x106/kg, the mobilization regimen was Cy+G-CSF and the conditioning regimen was Cy 200mg/kg+rATG 7.5mg/kg. Febrile neutropenia and CMV viremia were presented as complications. The post-HSCT Day in reaching ANC>500/μL was +10, in reaching platelet count>50x103/μL was day +8 and he was hospitalized for 30 days.

The second case is a 53-year-old female patient with a history of arterial hypertension, hypothyroidism, fibromyalgia, convulsive status, external hemorrhoids, anemia and depression, diagnosed at age 45 with RMG, and in 2017 at age 50 she underwent HSCT. The treatment lines between diagnosis and HSCT were Pyridostigmine 60mg every 6 hours, Prednisone 100mg/day, Azathioprine 50mg every 8 hours, Cyclophosphamide 1gr IV every month for 6 months, Immunoglobulin 2014 and 2016, Rituximab 375mg/m2 IV every 4 weeks and Plasmapheresis in acute exacerbations. The treatment regimen at the time of the HSCT was performed with Pyridostigmine 60 mg every 8 hours, Prednisone 100 mg/day, Azathioprine 50 mg every 8 hours, and Plasmapheresis in case of acute exacerbations. The treatment regimen after HSCT was carried out with Pyridostigmine 60 mg/ 8 hours, Prednisone 50 mg/day and Plasmapheresis in the case of acute exacerbations. According to the mobilization and collection regimen, 1 day of collection was recorded, with a cell count of 6,205x106/kg, the mobilization regimen was Cy+G-CSF and the conditioning regimen was Cy 200mg/kg+ rATG7,5mg/kg. A convulsive status and a myasthenic crisis were presented as complications. The post-HSCT Day in reaching ANC> 500/μL was +11, in reaching platelet count>50x103/μL was day +11 and she was hospitalized for 90 days. Both cases were followed up at 7.4 and 2 years, with evidence of 100% survival. At the last follow-up, the disease was in complete remission and both patients were completely stable.

Conclusions: Although information about survival in patients with a history of RMG undergoing HSCT is scarce, in these two cases we evidence a good response of both patients and a survival of 100% for both of them at the last follow-up.

Clinical Trial Registry: No.

Disclosure: Nothing to declare.

O084. Stem Cell Transplantation From A Haploidentical Donor For Pediatric Patients With Acute Myelogenous Leukemia: A Retrospective Nationwide Study

Charlotte Nazon¹, Jean-Hugues Dalle², Gérard Michel³, Charlotte Jubert⁴, Benedicte Bruno⁵, Marie Ouachée-Chardin⁶, Fanny Rialland⁷, Anne Sirvent⁸, Nicole Raus⁹, Catherine Paillard¹

¹ Hautepierre University Hospital, Strasbourg, France, ² Hôpital Robert Debré, Paris, France, ³ Hopital de La Timone, Marseille, France, ⁴ Hôpital Pellegrin, Bordeaux, France, ⁵ Jeanne de Flandre Hospital, Lille, France, ⁶ Institute of Pediatric Hematology and Oncology, Hospices Civils, Lyon, France, ⁷ CHU de Nantes, Nantes, France, ⁸ CHU de Montpellier, Montpellier, France, ⁹ Data Manager, SFGM-TC, Lyon, France

Background: Haploidentical hematopoietic stem cell transplantation (HHSCT) can provide a quickly available chance of cure for many patients who lack an HLA-matched donor. It is particularly interesting since it is now proven that a longer time between registration and transplant have a significant negative impact on survival. HHSCT showed similar results to those reported using an HLA-matched donor in adult cohorts and notably in patients treated for an AML. In pediatrics, in the absence of an HLA-matched related or unrelated donor (MRD or MUD), the best alternative donor source remains controversial. We wanted to analyze French data regarding AML patients treated by an HHSCT.

Methods: This is a retrospective, multicenter, registry-based analysis in France. Eligibility criteria for this analysis included pediatric patients (aged <18 years) with AML who had received a hematopoietic stem cell transplantation (HSCT) from a haploidentical donor (≥ 2 antigen mismatches or more out of 8). The probability of overall survival (OS) was calculated using the Kaplan–Meier method.

Results: Between 1995 and 2020, 24 patients (25 HHSCT) under 18 years old were transplanted for AML using haploidentical donors in France. Median age at HSCT was 13,2 (1,5-17,8). It was the first HSCT for 80% of the patients, 36% were allografted on their first remission. There was 10 different types of conditioning regimen (in majority myeloablative) and different GvH prophylaxis (56% had post-graft cyclophosphamide). Eighty eight percent of them had full-donor chimerism at D+30, 68% had an acute GvHD (graft versus host disease) at a median time of 29 days (11-56) and 24% had a chronic GvHD (median time: 227 days, 91-606). Seven patients (28%) relapsed. Overall survival was 63%, 80% when considering only patients allografted in first complete remission (CR1, n = 10, Figure 1). Transplant related mortality (TRM) was 16%. Comparing patients treated before and after 2014, there was a tendency to a better OS after 2014 (Figure 1, p = 0.47).

Conclusions: Patients who underwent HHSCT are the most severe ones and it explains partially the average results of HHSCT in pediatric AML. Nevertheless, we can see an improvement of survival rates over time. Despite a high TRM, OS of CR1 patients (80%) makes HHSCT a valid alternative for patients lacking a matched donor. Finally, analysis can be confounded by the use of so many different conditioning regimens. It would be interesting to have official guidelines for HHSCT in hematologic malignancies and for conditioning regimen to harmonize our practice.

Disclosure: Nothing to declare.

O085. White Matter Development in Neonates With Infantile Krabbe Disease

Maria Escolar¹, Ashley Whited¹, Keith Werling¹, Michele Poe¹

¹ University of Pittsburgh, Pittsburgh, United States

Background: Infantile Krabbe disease (KD) is a rare neurodegenerative disorder characterized by progressive dysmyelination and demyelination. Initial symptoms can present with excessive irritability and difficulty nursing, progressing as the patient experiences seizures, vision loss, muscle spasms and cognitive impairment. Eventually most patients enter a vegetative state and death occurs—typically within 2 years of age. Previous studies have shown that diffusion measures like fractional anisotropy (FA) can accurately detect white matter degeneration in KD patients and guide clinical intervention in the form of hematopoietic stem cell transplantation (HSCT). As more advanced therapies (including gene therapy) are emerging, there is an increased need to map the development of this disease to identify and match disease progression biomarkers with treatments that provide the best outcomes for individual patients.

Methods: With the use of an automated diffusion tensor imaging (DTI) analysis pipeline, white matter structure was quantified using FA. Four major brain tracts (left and right corticospinal internal capsule (CSIC) and the splenium and genu of the corpus callosum) were analyzed. Volumetric registration of the FA maps for patients against appropriately age-matched atlases was used to align white matter structures. Successful registration was visually affirmed for quality assurance purposes. Automated fiber-tract quantification (AFQ) for FA was then performed for the four tracts listed above, and the average FA curves were compared to those of age-matched controls.

Results: Changes in white matter were recorded in 92 patients: 62 early infantile KD patients (<12 months at onset), 21 late infantile KD patients (12-36 months at onset), 5 healthy KD heterozygotes and 4 children that were identified as high risk through newborn screening. By 2 years of age, KD patients had FA values more than 50% below those of the healthy controls across the tracts with rapid decline in the first 6 months. Patients were sorted into high and low baseline FA groups and Kaplan-Meier survival curves indicated that early infantile KD patients with high FA values for the CSIC tract had a median survival 1 year longer than patients with low FA values. The baseline FA values of the splenium and the genu were not found to be associated with survival. None of the late infantile KD baseline FA values were found to be significantly associated with survival.

Conclusions: FA is a sensitive metric for assessing the progression of infantile KD and predicting therapeutic outcomes in HSCT candidates. If infantile KD is detected through newborn screening, DTI can be used to monitor changes in asymptomatic children at high risk for KD and maximize HSCT effectiveness through early diagnosis.

Disclosure: Nothing to declare.

O086. Cryopreservation of Allogeneic Hematopoietic Progenitor Cells in The Current SARS-Cov-2 Pandemic: Experience of A Single Centre

Dolores Moreno¹, Lidia Navarro², Pilar Solves², Gema Plume², Ines Gomez², Maria Dolores Picazo², Amparo Pous², Maria Isabel Peñalver², Cristina Arbona³, Guillermo Sanz², Nelly Carpio²

¹ Hospital Universitari i Politecnic la Fe, Valencia, Spain, ² Hospital Universitario La Fe, Valencia, Spain, ³ Centro Transfusión de la Comunidad Valenciana, Valencia, Spain

Background: Cryopreservation is usually performed for autologous hematopoietic stem cells (HSC), and occasionally for allogeneic hematopoietic stem cells. The current coronavirus pandemic has forced many transplant centers to change their procedures in order to adapt to the public health emergency. EBMT has recommended cryopreserving the hematopoietic progenitor cells (HPC) to avoid putting at risk the recipients due to infectious problems in donors. Our objective was to analyze the modifications in the HSC management due to the SARS-Cov-2 pandemic in the apheresis unit of a tertiary-care hospital, focusing on the cryopreservation impact on transplantation outcomes.

Methods: We reviewed the collection and management of HSC from related donors for a 6-month period. SARS-Cov-2 PCR test was performed to all donors before starting mobilization. All procedures were under-taken using Spectra Optia apheresis system, with the continuous CMN collection program software version 11.0. Data was collected retrospectively from electronic patient and donor records and laboratory records. Collected cells were sent to the local processing laboratory for cryopreservation. Once cryopreserved, clinicians proceed to perform the HSCT if the patient was fit. HSC were thawed and washed according to local procedures. Engraftment was defined as the first of 3 consecutive days with neutrophil count > 500/μl.

Results: Twenty-one donors were included in this analysis. Median patient age was 48 years (range 15-69)(25 adult and 1 child). Median pre-collection total nucleated cells (TNC)/μl and CD34+ cells/μl were 52.390 and 71.79 (range 16.6 - 179.35-), respectively. A median of 11 liters of blood (15.990-7.828) were processed per donor. Only 1 patient required two apheresis days for achieving the target dose cell (> 2 x 10⁶/Kg of recipient). The median collected TNC and CD34+ cells were 9.5 x 10⁸/Kg and 8.3 x 10⁶/Kg (range 3.2 – 14.11), respectively. The median of days between collection and infusion of progenitor cells was 8 (range 7-93). After thawing and before infusion, the median TNC was 8.3 x 10⁸/Kg and the median cell viability was 92%. All patients engrafted at a median of 17 days (15-28) after transplantation.

Conclusions: The cryopreservation of allogeneic HSC was feasible. The median of days until engraftment was similar to those HSCT performed in fresh at our hospital (Montoro J, BMT 2020).

Disclosure: Nothing to declare.

O087. Letermovir Prophylaxis For Cytomegalovirus Reactivation in Allogeneic Stem Cell Transplantation: A Multicenter Italian Real-world Experience

Annalisa Paviglianiti¹, Annalisa Pitino², Angelo Michele Carella jr^3,4, Benedetto Bruno^4,5, Alessandra Crescimanno⁶, Vincenzo Federico⁷, Alessandra Picardi^8,9, Stefania Tringali¹⁰, Claudia Ingrosso¹¹, Paola Carluccio¹², Domenico Pastore¹³, Adriana Vacca¹⁴, Bianca Serio¹⁵, Gabriella Storti¹⁶, Nicola Mordini¹⁷, Salvatore Leotta¹⁸, Michele Cimminiello¹⁹, Lucia Prezioso²⁰, Barbara Loteta¹, Anna Ferreri¹, Fabrizia Colasante⁸, Emanuela Merla³, Luisa Giaccone^4,5, Alessandro Busca⁴, Maurizio Musso⁶, Renato Scalone⁶, Nicola Di Renzo⁷, Serena Marotta⁸, Patrizio Mazza¹¹, Pellegrino Musto¹², Immacolata Attolico¹², Carmine Selleri¹⁵, Mercedes Gori², Giovanni Tripepi²¹, Gaetana Porto¹, Massimo Martino¹

¹ Grande Ospedale Metropolitano "Bianchi-Melacrino-Morelli", Reggio Calabria, Italy, ² Istituto di Fisiologia Clinica del Consiglio Nazionale dele Ricerche (CNR), Roma, Italy, ³ Ospedale I.R.C.C.S. Casa Sollievo della Sofferenza, San Giovanni Rotondo (Fg), Italy, ⁴ A.O.U. Città della Salute e della Scienza di Torino, Torino, Italy, ⁵ Università di Torino, Torino, Italy, ⁶ Istituto La Maddalena, Palermo, Italy, ⁷ P.O. “Vito Fazzi”, Lecce, Italy, ⁸ AORN Cardarelli -Napoli, Napoli, Italy, ⁹ Università di Roma Tor Vergata, Roma, Italy, ¹⁰ VIlla Sofia Cervello, Palermo, Italy, ¹¹ Ospedale San Giuseppe Moscati, Taranto, Italy, ¹² Università degli Studi “Aldo Moro” e AOUC Policlinico di Bari, Bari, Italy, ¹³ Ospedale A. Perrino, Brindisi, Italy, ¹⁴ PO Armando Businco, Cagliari, Italy, ¹⁵ University of Salerno, Salerno, Italy, ¹⁶ Azienda Ospedaliera S. G. Moscati, Avellino, Italy, ¹⁷ Azienda Ospedaliera S Croce e Carle, Cuneo, Italy, ¹⁸ Azienda Policlinico-Vittorio Emanuele, Catania, Italy, ¹⁹ Ospedale San Carlo, Potenza, Italy, ²⁰ Azienda Ospedaliero-Universitaria di Parma, Parma, Italy, ²¹ Istituto di Fisiologia Clinica del Consiglio Nazionale delle Ricerche (CNR) Reggio Calabria, Reggio Calabria, Italy

Background: Cytomegalovirus (CMV) reactivation is a significant cause of morbidity and mortality after allogeneic stem cell transplantation (HSCT). Letermovir is a novel antiviral agent recently approved for prophylaxis of CMV reactivation and disease in adult CMV-seropositive recipients.

Methods: We retrospectively analyzed the efficacy of letermovir in a multicenter Italian real-world cohort of 204 CMV-seropositive patients. Letermovir was started at dosage of 240 mg tablet once daily, no later than 28 days after HSCT, and was continued through 100 days post-transplant.The primary end-point was the proportion of patients who developed clinically significant CMV infection (defined as CMV disease or CMV viremia leading to preemptive treatment) within 100 days after HSCT.

Results: The median age was 52 (range 18-75) years and 54% were male. Disease was acute myeloid leukemia in 53% of the patients and 65% had a CMV seropositive donor. Fifty-nine patients (29%) underwent haploidentical stem cell transplantation, 66 (32%) underwent a matched related and 56 (28%) a matched unrelated donor. Graft source was peripheral blood stem cell in 86% of the cases and 66% received a myeloablative conditioning (MAC) regimen. GVHD prophylaxis consisted of cyclosporine A + short-term methotrexate in MAC and cyclosporine A ± micophenolate mofetil for a reduced intensity conditioning (RIC) regimen. Anti-thymocyte globulin was used in 98 (48%) and haploidentical stem cell recipients received post-transplant cyclophosphamide. Four patients had acute GvHD of grade ≥2 at the starting of letermovir. Patients received letermovir for a median of 93 (range: 5-100) days. One hundred thirty four (66%) patients were at high risk for CMV disease. The cumulative incidence of clinically significant CMV infection at 100 days was 5,4% (11 pts), 3,4% (7pts) with a CMV disease and 2% with a CMV positive viremia leading to preemptive treatment. The cumulative incidence of patients developing clinical significant CMV infection between 101 and 168 days (24 weeks) post-HSCT (after discontinuation of letermovir prophylaxis) was 26 (13%). Death was the primary cause of discontinuation within 100 days in 13 (6%) patients. Cause of death was: infection in 3 patients, disease relapse in 3, GvHD in 1, VOD in 1 and other reasons in 5 patients. When comparing characteristics of patients who discontinued letermovir to those who did not, MAC regimen was significantly more frequent in the group of patients who discontinued letermovir (test Chi2 p < 0.001). We did not observe a statistical significant difference in CMV infection between patients at high and low risk category who had letermovir (test Chi2 p = 0.50 at 100 days and p = 0.30 at 24 weeks).

Conclusions: Letermovir confirmed its efficacy in preventing CMV infection in the first 100 days after HSCT in a real-world population analysis. An extension of prophylaxis of up to 200 days should be considered, and clinical trials to this end are expected.

Disclosure: Nothing to declare.

O088. Impact of Fluoroquinolone Prophylaxis in Allogeneic Hematopoietic Stem Cell Transplantation; Single Center Experience

Eshrak Al-Shaibani¹, Zoe Evans¹, Kayla Madsen¹, Shiyi Chen², Carol Chen¹, Ivan Pasic¹, Wilson Lam¹, Arjun Law¹, Fotios V. Michelis¹, Armin Gerbitz¹, Dennis D. Kim¹, Rajat Kumar¹, Jeffrey H. Lipton¹, Jonas Mattson¹, Shahid Husain³, Zeyad Al-Shaibani¹, Auro Viswabandya¹

¹ Hans Messner Allogeneic Transplant Program, Princess Margaret Cancer Centre-University Health Network, Toronto, Canada, ² Princess Margaret Cancer Centre-University Health Network, Toronto, Canada, ³ University Health Network/University of Toronto, Toronto, Canada

Background: Patients undergoing allogenic hematopoietic stem cell transplantation (allo-HCT) are at high risk of infection-related mortality in the neutropenic period post-transplantation. Prophylaxis with fluoroquinolone (FQ) is recommended. However, due to an uncertain increase emergence of resistant organisms, there is no consensus in recommending FQ post-transplantation. The purpose of this study to compare infectious outcomes (primary: incidence bacteremia; secondary: febrile neutropenia (FN), C. difficile, susceptibility of bacteremia, time to discharge and 30-day mortality) for patients undergoing allo-HCT with and without FQ prophylaxis.

Methods: We retrospectively investigated 156 patients transplanted between December 2019 and May 2020. Of these patients, 52% were male and all patients had a Hickman catheter insertion. AML was the major indication for transplant in 50%. Myeloablative conditioning was administered to 42% of patients while 58% received reduced intensity conditioning. In vivo T-cell depletion was given to 99% of patients. Donors were HLA matched related in 24%, matched unrelated in 60%, and haploidentical in 16% of patients. Patients were included those who received FQ-prophylaxis (n = 88) and those who did not (n = 68). Infectious outcomes, related variables and mortality were analyzed within the first 30 days post allo-HCT.

Results: Median age 58 years (range: 18-76). Median follow-up time 13 months (range:1 - 69). Median time to neutrophil engraftment was 16 days in both groups following allo-BMT (P = 0.17).

The incidence of FN was non statistically lower in the FQ-prophylaxis group compared to those not receiving FQ (89% vs 95.6%, p = 0.18). Similar results in non-significant improvements in the blood stream infection 52.3% vs 64.7%, p = 0.12). However, Gram-negative bacteria were higher in non-FQ cohort comparing to FQ cohort (38.2% vs 19.3%, p = 0.05), while the incidence of Gram-positive bacteremia in FQ vs non-FQ cohort was 27.3% vs 19%, respectively. Streptococcus viridians was the most frequent Gram-positive bacterial species in FQ and non-FQ cohort 13.6% and 10.3%, respectively. Klebsiella pneumoniae was the main pathogen identified in non-FQ cohort (28%). Additionally, FQ-resistance was higher in the FQ cohort 14.8% vs 4.4% in non-FQ (P = 0.07). Interestingly, the incidence of C.difficile was higher in the non-FQ cohort 33.8% vs 10.2% in FQ cohort (p < 0.001).

The median duration for antibiotics used was not significantly different between those who received FQ versus those who did not (8 vs. 10 days p = 0.08). The median time to discharge from hospital was not different between groups and were 30 days each (p = 0.49). There was no difference between groups with respect to 30-days mortality and ICU admission (10.2% vs 10.3%; P = 0.99) and (4.6% vs 5.9%; P = 0.73) respectively. Acute GvHD grade III/IV was higher in FQ comparing to non-FQ and it showed a trend to be statistically significant (12.5% vs 4.4%; P = 0.036).

Conclusions: Our single center experience demonstrates a proof of concept that performing allo-HCT without FQ prophylaxis is feasible. All relevant outcome parameters were almost comparable in both groups. Considering the increase emergence of FQ resistance and lack of scientific evidence showing benefit for prophylactic use, these results warrant further investigations through a prospective study design.

Disclosure: Nothing to declare.

Haemoglobinopathy

O089. Feasibility of Premature Sirolimus Cessation Post Non-Myeloablative Transplantation in Adult Patients With Severe Sickle Cell Disease

Moussab Damlaj¹, Bader Alahmari¹, Ahmed Alaskar¹, Ayman Alhejazi¹, Husam Alsadi¹, Inaam Shehab-Eddine¹, Abdulrahman Ghamdi¹, Walid Mashaqbeh¹, Heba Alshobaki¹, Tahani Alanazi¹, Maybelle Balili¹, Muhammad Qureshi¹, Mazin Ahmed¹, Mohsen Alzahrani¹

¹ King Abdulaziz Medical City; King Abdullah International Medical Research Center; King Saud bin Abdulaziz University for Health Sciences; Saudi Blood and Marrow Transplantation Society, Riyadh, Saudi Arabia

Background: Non-myeloablative transplantation (NMA) using alemtuzumab and total body irradiation in adult patients with severe sickle cell disease (SCD) is feasible and results in favourable long term outcomes. Sirolimus is administered to induce a state of mixed chimerism and is typically continued for at least 1-year post-transplant with progressive taper initiated when lymphoid chimerism is sustained ≥ 50%. Such prolonged course of sirolimus can cause undesirable toxicity while the impact of premature cessation particularly when lymphoid chimerism is low is unknown. Our aim is to further examine patients with low CD3 chimerism that discontinued sirolimus post NMA for severe SCD at our center.

Methods: Following IRB approval, adult patients that underwent HSCT from 2015 to 2020 for severe SCD with at least 6 months of follow up were included. Conditioning regimen consisted of alemtuzumab (1 mg/kg divided over 5 days on days -7 to -3) and 300 cGy TBI on day -2 or -1 with sirolimus starting on day -1 for GVHD prophylaxis. Post-transplant, sirolimus taper began starting at 1 year if lymphoid chimerism is sustained ≥ 50%. Categorical and continuous variables were compared using chi-squared or Kruskal-Wallis, as appropriate.

Results: A total of 110 patients underwent transplant during this time frame and at data cut-off, 62 had discontinued sirolimus. Ten patients had primary or secondary graft failure while on sirolimus and were subsequently tapered. The remaining 52 patients were stratified based on chimerism status at 1-year as follows; 32 (62%) had adequate CD3 chimerism (CD3 ≥ 50%), 10 (19%) had low CD3 chimerism (CD3 < 50%) but discontinued sirolimus for various reasons including patient decision, while the remaining 10 (19%) patients had unknown CD3 chimerism but adequate and sustained whole chimerism (≥50%). Stratified by chimerism status, male gender, baseline pre-HSCT Hb, Hb S, and donor SC trait, ABO match and infused graft CD3 dose were similar across the groups. Median age was higher in the whole chimerism group (p = 0.05) while infused CD34 dose was higher in the CD3 ≥ 50% group.

Median myeloid chimerism was similar pre- and post- taper between the CD3 ≥ 50% and CD3 < 50% while median lymphoid chimerism was significantly higher pre-taper in the CD3 ≥ 50% group (65.1% vs. 41.6%, p < 0.0001) but it became similar post-taper across all three groups (p = 0.97). The median time on sirolimus post-HSCT was 536, 475 and 436 for the CD3 ≥ 50%, CD3 < 50% and whole chimerism ≥ 50% (p = 0.008). No case of engraftment failure or death was noted across all groups.

Conclusions: Albeit the small sample size herein, premature cessation of sirolimus appeared feasible in patients with < 50% lymphoid chimerism without declining chimerism or graft loss. These observations warrant further examination.

Clinical Trial Registry: Not applicable.

Disclosure: No relevant conflicts of interest.

Immunodeficiency Diseases and Macrophages

O090. International Survey And Retrospective Analysis of 28 Patients With IFN-Γ-Receptor Deficiency

Peter Olbrich^1,2, Dimana Dimitrova³, Andrew Gennery^4,5, Ansgar Schulz⁶, Sunil Bhat⁷, Mary Slatter^4,5, Jennifer Heimall⁸, Marco Zecca⁹, Mark Vander Lugt¹⁰, Paul Veys¹¹, Dagmar Berghuis¹², Agueda Molinos¹, Revathi Raj¹³, Serap Aksoylar¹⁴, Neslihan Edeer Karaca¹⁵, Lauri M Burroughs¹⁶, Troy Torgerson¹⁶, Jennifer Kanakry³, Despina Moshous¹⁷, Sheree Hazelaar¹⁸, Alexandra F Freeman¹⁹, Arjan Lankester¹², Steven M Holland²⁰, Robbert Bredius¹², Manfred Hönig⁶

¹ Hospital Infantil Virgen del Rocío, Seville, Spain, ² Instituto de Biomedicina de Sevilla, Seville, Spain, ³ National Institute of Health, National Cancer Institute, Experimental Transplantation and Immunotherapy Branch, Bethesda, United States, ⁴ Children’s Haematopoietic Stem Cell Transplant Unit, Great North Children’s Hospital, Newcastle upon Tyne, United Kingdom, ⁵ Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, United Kingdom, ⁶ University Medical Center, Ulm University, Ulm, Germany, ⁷ Pediatric Hematology, Oncology and Blood & Marrow Transplantation Mazumdar Shaw Medical Centre, Narayana Health City Bangalore, India, ⁸ Children´s Hospital of Philadelphia, Philadelphia, United States, ⁹ Policlinico San Matteo IRCCS, Pavia, Italy, ¹⁰ CS Mott Children´s Hospital Michigan Medicine, Ann Arbor, United States, ¹¹ Great Ormond Street Hospital for Children, London, United Kingdom, ¹² Leiden University Medical Center, Leiden, Netherlands, ¹³ Apollo Cancer Institutes, Chennai, India, ¹⁴ Ege University, Izmir, Turkey, ¹⁵ Ege University, Izmir, Turkey, ¹⁶ Seattle Children´s Hospital, Seattle, United States, ¹⁷ Hôpital Necker-Enfants Malades; IMAGINE - Institut des Maladies Génétiques, Paris, France, ¹⁸ Leiden University Medical Centre, Leiden, Netherlands, ¹⁹ National Institute of Health (NIAID), Bethesda, United States, ²⁰ National Institute of Allergy and Infectious Diseases National Institutes of Health, Bethesda, United States

Background: Mendelian susceptibility to mycobacterial disease (MSMD) characterizes a group of primary immunodeficiencies, which render affected individuals susceptible to weakly virulent mycobacteria. Mutations in IFNGR1 and IFNGR2, the genes encoding the components of the Interferon-γ-receptor, can – among others – cause this phenotype. Hematopoietic stem cell transplantation (HSCT) is currently the only curative option for this condition however mixed outcomes have been reported with an overall survival (OS) of 50% (Roesler et al), likely due to high rates of engraftment failures. Both supportive care and HSCT procedures have improved overtime; therefore, we sought to provide an update on HSCT outcomes.

Methods: We retrospectively analyzed data on clinical presentation and outcome of 30 HSCT procedures performed in 28 patients with the following genetic variants AR IFNGR1: n = 22, AD IFNGR1: n = 1, AR IFNGR2: n = 5 subsequently transplanted in 15 institutions between 1997 and 2020.

Results: The median age at diagnosis of the first infection was 33.6 months (IQR 12.8-100.2). 13/28 had a positive family history with at least one sibling who was affected with mycobacterial disease in early infancy. The median time between molecular diagnosis and HSCT was 347 days (IQR 108-1731). Infectious agents identified in this cohort included a broad range of mycobacteria (mostly BCG and MAC), bacterial infections such as Salmonella and Listeria, as well as viral infections (mostly herpesviruses, respiratory viruses and adenovirus). HSCT was performed at a median age of 4.5 years (0.3-58.6), with grafts donated by HLA-identical family members (n = 12), matched unrelated donors (n = 10), a mismatched unrelated donor (n = 1) and mismatched family donors (n = 7). The stem cell source was BM in 15, PBSCs in 10, CB in 4 and a combination of BM and CB in 1 transplant. Conditioning included various combinations of Busulfan, Treosulfan, Melphalan, Fludarabine, Cyclophosphamide and Thiotepa with a myeloablative intention in 23 and a non-myeloablative intention in 7 patients. Four patients experienced primary and 1 patient secondary graft failure. A second transplant was performed in two patients. Acute GvHD was reported after 9/30 transplants (Grade I: n = 3; II: n = 1; III: n = 4, IV: n = 1) and cGvHD in 4 patients. OS was 82.1% with 23/28 patients alive after a median follow up of 2.5 years (0.1-16.7).

Conclusions: In this multi-center retrospective analysis of transplant data in the largest cohort of patients with IFNGR-deficiency we have found a median time of almost 2 years between the clinical diagnosis and the curative procedure of HSCT. Although a more in-depth data analysis aiming to identify those factors related to a favorable disease/HSCT course is ongoing, overall outcomes seem to have improved considerably in the recent years.

Disclosure: P Olbrich has received funding by the Asociación Española de Pediatría (Madrid, Spain), Consejería de Salud y Familias (Andalusia, Spain) and the Instituto de Salud Carlos III Juan Rodes scholarship (Madrid, Spain).

O091. Allogeneic HSCT For Adolescents and Adults With Inborn Errors of Immunity: A Retrospective Study of The Inborn Errors Working Party (IEWP)

Michael Albert¹, Claudia Wehr², Felipe Suarez³, Maria Laura Fox⁴, Andrew Gennery⁵, Nizar Mahlaoui³, Sheree Hazelaar⁶, Katerina Bakunina⁷, Mary Slatter⁵, Matthew Collin⁸, Su Han Lum⁵, Venetia Bigley^8,9, Fabian Hauck¹, Christoph Klein¹, Siobhan Burns^10,11, Ben Carpenter¹², Ronjon Chakraverty¹³, Carsten Speckmann², Klaus Warnatz², Morgane Cheminant³, Bénédicte Neven¹⁴, Rita Beier¹⁵, Andrzej Lange¹⁶, Alexandra Laberko^5,17, Dmitry Balashov¹⁷, Ansgar Schulz¹⁸, Rob Wynn¹⁹, Arjan Lankester²⁰, Emma Morris^10,12,11

¹ Dr. von Hauner Children’s Hospital, University Hospital, LMU, Munich, Germany, ² Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany, ³ Hôpital Necker Enfants Malades, Assistance Publique-Hôpitaux de Paris, Paris, France, ⁴ Vall d’Hebron Hospital Universitari, Barcelona, Spain, ⁵ Great North Children’s Hospital, Newcastle upon Tyne, United Kingdom, ⁶ EBMT Data Office, Leiden, Netherlands, ⁷ EBMT Statistical Unit, Leiden, Netherlands, ⁸ Northern Centre for Bone Marrow Transplantation, Newcastle Hospitals NHS Foundation Trust, Freeman Hospital, Newcastle upon Tyne, United Kingdom, ⁹ Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, United Kingdom, ¹⁰ UCL Institute of Immunity & Transplantation, UCL, London, United Kingdom, ¹¹ Royal Free London Hospitals NHS Foundation Trust, London, United Kingdom, ¹² University College London Hospitals NHS Foundation Trust, London, United Kingdom, ¹³ MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford, United Kingdom, ¹⁴ Necker Enfants Malades University Hospital, Assistance Publique-Hôpitaux de Paris, Paris, France, ¹⁵ Hannover Medical School, Hannover, Germany, ¹⁶ Dolnośląskie Centrum Transplantacji Komórkowych, Wroclaw, Poland, ¹⁷ Dmitry Rogachev National Research Center of Pediatric Hematology, Oncology and Immunology, Moscow, Russian Federation, ¹⁸ Universitätsklinikum Ulm, Ulm, Germany, ¹⁹ Royal Manchester Children’s Hospital, Manchester, United Kingdom, ²⁰ Leiden University Medical Center, Leiden, Netherlands

Background: Recent single center publications from specialist centers have demonstrated that HSCT can be safe and effective for adolescents and adults with inborn errors of immunity (IEI)^1,2. There remains an urgent need, however, to identify which adult patients are most likely to benefit from HSCT and the optimal timing of transplant.

Methods: We present an EBMT IEWP retrospective multicenter study of adolescents and adults ≥15 years of age who underwent HSCT for an underlying IEI from 1.1.2000 to 31.12.2018. The Kaplan-Meier method and the log rank test were applied to assess probabilities of overall survival (OS) and the differences between risk groups. IEI diagnoses were classified as per 2019 IUIS classification³.

Results: 283 patients were included in this analysis. The most common forms of IEI were combined immune deficiency (CID) including disorders of immune dysregulation (n = 142), phagocyte disorders (n = 102) and common variable immunodeficiency (CVID; n = 35). The median age at transplant was 18 years (range: 15-62) with a median follow-up of 50.8 months (95%CI: 44.3-59.6). Stem cell sources were peripheral blood (n = 147) or bone marrow (n = 135) with one cord blood. Donors were unrelated (10/10 MUD, n = 153; 9/10 MMUD, n = 43; <9/10 MMUD, n = 9; UCB, n = 1; total unrelated n = 206), matched family donors (n = 63), haploidentical donors (n = 9), or unknown (n = 5). Conditioning intensity was predominantly reduced intensity (n = 183 reduced toxicity and n = 21 non-myeloablative), but 70 patients received myeloablative conditioning, and 9 had unknown conditioning intensity. We examined various IEI-related complications present at the time of transplant for impact on OS, including bronchiectasis, granulomatous lymphocytic inflammatory lung disease (GLILD), colitis, protracted diarrhea, body mass index (BMI), prior splenectomy, and total number of IEI-associated complications. The 2-year OS for the whole cohort was 75% (95%CI: 70-80%). Superior OS at 2 years was achieved in patients with phagocyte disorders (83%, 76-90%) and CID (74%, 67-81%) compared to those with CVID (57%, 41-74%), respectively (log-rank test p = 0.003). Neither age, donor type, conditioning intensity, BMI, nor presence of colitis, protracted diarrhea, GLILD or bronchiectasis at HSCT had a significant impact on OS, but the cumulative number of IEI-related complications at the time of HSCT did. OS was 84% (78-91%), 73% (64-83%), and 67% (55-79%), respectively for patients with no complications, only one complication, or ³ 2 complications (p = 0.006; figure). Prior splenectomy (60% vs 78%, p = 0.003) also adversely affected OS.

Conclusions: This is by far the largest study performed to date examining the outcome following HSCT for adolescents and adults with IEI. OS of this cohort was independent of donor source and conditioning intensity, and very good outcomes were observed for patients with CID and phagocyte disorders. We demonstrate that prior splenectomy and the number of IEI-associated complications at HSCT, but not patient age, adversely affect OS. Data from this study is expected to provide essential guidance for the use of HSCT in adolescent and adult IEI patients.

References:

1. Albert et al. J Allergy Clin Immunol Pract. 2018; 6: 298-301.

2. Fox et al. Blood. 2018; 131: 917-931.

3. Bousfiha et al. J Allergy Clin Immunol. 2020; 40(1): 66-81

Disclosure: Nothing to declare.

Inborn Errors

O092. T-Replete HLA-Matched Grafts Versus T-Depleted HLA-Mismatched Grafts in Inborn Errors of Immunity: Where Are We Now?

Su Han Lum¹, Sinead Greener¹, Kay Carruthers¹, Inigo Perez-Heras¹, Daniel Drozdov¹, Robert January¹, Zohreh Nademi¹, Eleri Williams¹, Stephen Owens¹, Terry Flood¹, Sophie Hambleton¹, Mario Abinun¹, Andrew Gennery¹, Mary Slatter¹

¹ Great North Children’s Hospital, Newcastle upon Tyne, United Kingdom

Background: Haematopoietic cell transplantation (HCT) has become standard of care for an increasing number of inborn errors of immunity (IEI). This study compared the transplant outcomes according to T-replete HLA-matched grafts using alemtuzumab (n = 117) and T-depleted HLA-mismatched grafts using TCR ab/CD19 depletion (n = 47) in children with IEI who underwent first HCT between 2014 and 2019 at the Great North Children’s Hospital, UK.

Methods: Main outcomes of interest were OS and EFS. An event was defined as death, graft failure or second procedures for slipping chimerism. Other endpoints were neutrophil engraftment, graft-versus-host disease (GvHD), viraemia, duration of hospitalization, immune reconstitution and donor chimerism. Subgroup differences in OS and EFS were evaluated by log-rank test. Subgroup differences in acute GvHD (aGvHD) and viraemia were evaluated by Gray’s test, with death as a competing event. A matched-pair analysis of CD3 reconstitution was carried out in T-replete HLA-matched PBSC survivors (n = 28) and T-depleted mismatched survivors (n = 28). The two-patient series were matched for age at transplant and diagnosis.

Results: All patients received treosulfan-based conditioning except patients with DNA repair disorders. There was no significant difference in number of pre-transplant risk factors between T-replete HLA-matched graft and T-depleted mismatched graft recipients. For T-replete grafts, the stem cell source was marrow in 25 (21%) patients, PBSC in 84 (72%) and CB in 7 (6%). All TCR ab/CD19 depletion was performed on PBSC from 45 haploidentical parental donors and 2 mismatched unrelated donors. The 3-year OS and EFS for the entire cohort were 85% (77-90%) and 79% (69-86%) respectively. Analysis by age at transplant revealed a comparable 3-year OS between T-replete grafts (88%, 76-94%) and T-depleted grafts (87%, 64-96%) in younger patients (<5 years of age at HCT). For older patients more than 5 years of age, the OS was significantly lower in T-depleted grafts (55%, 23-78%), compared to T-replete grafts (87%, 68-95%) (p = 0.03). Grade III-IV aGvHD was observed in 8% of T-replete marrow, 7% of T-replete PBSC, 14% of T-replete CB and 2% of T-depleted PBSC (p = 0.73). Higher cumulative incidence (CNI) of viraemia (p < 0.001) and delayed CD3 reconstitution (p = 0.003) were observed after T-depleted graft HCT.

Table 1 Transplant outcomes according to T-replete HLA-matched grafts and T-depleted HLA-mismatched grafts.

Conclusions: These data indicate that mismatched donor transplant after TCR ab and CD19 depletion is comparable to matched donor HCT for younger children with IEI in need of an allograft.

Clinical Trial Registry: Not applicable.

Disclosure: Nothing to declare.

O093. Phenotype Reversal, Excellent Overall, GVHD-free And Graft Failure-Free Survival In 55 Adolescents And Adults With Inborn Errors of Immunity Following Reduced Intensity Allogeneic HSCT

Thomas A Fox^1,2, Siobhan O Burns^1,3, Charley Lever³, Arian Laurence^2,2, Ronjon Chakraverty⁴, Sarah Grace², Filippo Oliviero², Sarita Workman³, Andrew Symes³, David M Lowe^1,3, Rachael Hough², Ben Carpenter², Emma C Morris^1,3,2

¹ UCL Institute of Immunity and Transplantation, London, United Kingdom, ² University College London NHS Foundation Trust, London, United Kingdom, ³ Royal Free London Hospitals NHS Foundation Trust, London, United Kingdom, ⁴ MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford, United Kingdom

Background: Adolescents and adults with uncorrected inborn errors of immunity (IEI) may benefit from definitive therapy to prevent progressive morbidity and premature mortality. Older patients were historically considered high-risk candidates and were not routinely offered allogeneic hematopoietic stem cell transplantation (HSCT). Recent data demonstrated that HSCT can be safe and effective in this group and led to a paradigm shift in their management. We present here detailed data on our older IEI patients (aged ≥13 years) transplanted at a single centre.

Methods: Detailed, retrospective data on pre- and post-transplant clinical and immunological status was collected from patients aged ≥13 years who were transplanted for IEI at the joint transplant unit of the Royal Free London Hospital and University College London Hospital. Probabilities of overall survival (OS) and composite GvHD-free, graft failure-free survival (GGFS) were calculated using the Kaplan-Meier and log rank methods. GvHD-free was defined as the absence of grades II-IV acute and moderate or severe chronic GvHD as per NIH classification.

Results: 55 patients underwent HSCT between 2004 and 2020. 18 patients were aged between 13-18 years (adolescent cohort) and 37 patients were ≥19 years (adult cohort) at HSCT. The underlying diagnoses were chronic granulomatous disease (CGD) (n = 19), GATA2 deficiency (n = 6), XIAP deficiency (n = 3), CVID (n = 2) and HLH (n = 2), whilst the remaining 19 had a range of combined immune deficiencies (CID). The majority (92%, 51/55) had a genetic diagnosis prior to transplant. Stem cell donors were matched unrelated, MUD (n = 30), matched related, MRD (n = 16), or mismatched unrelated, MMUD (n = 9). All patients had reduced intensity conditioning with serotherapy (either alemtuzumab or ATG). Most frequent indications for transplant were recurrent infection (42%, n = 23), autoimmunity/autoinflammation including colitis (36%, n = 20), malignancy (including myelodysplasia) (20%, n = 11) and HLH (7%, n = 4). OS and GGFS for the whole cohort at 3 years was 83% and 76% respectively, with a median follow up of 50 months (range 9m -14.5yrs) (Fig1A). Examining the influence of age at HSCT, OS and GGFS were significantly better for the adolescent group (n = 18) at 91% and 87%, respectively compared to the adult group (n = 37) at 71% and 61%, respectively (p = 0.012for OS and p = 0.025for GGFS) when all patients were examined (Fig1B).

Outcomes improved over time as evaluated by comparing transplants before and after 2010, with excellent OS and GGFS at 3 years for all patients transplanted after 2010 (n = 47) at 92% and 79% respectively (median f/u 44m). When the post 2010 cohort was analysed further, the impact of age at transplant was no longer statistically significant: p = 0.068 for OS (97% adolescent vs 83% adults) and p = 0.251 for GGFS (86% adolescents vs 65% adults) at 3 yrs (Fig. 1C).

All 14 patients with pre-transplant colitis had resolution of symptoms by last follow-up. GvHD was uncommon with only 2 patients requiring systemic immunosuppression at 6 months post-HSCT.

Conclusions: Reduced intensity HSCT is not only safe but can be highly effective at resolving clinical features across a range of IEIs in older patients. Excellent overall and GvHD-free, graft failure-free survival rates are achievable in carefully selected patients.

Disclosure: Nothing to declare.

O094. Allogeneic Hematopoietic Stem Cell Transplantation in Patients With Cerebral Adrenoleukodystrophy: Outcomes by Donor Cell Source And Match, Conditioning Regimen, And Stage of Cerebral Disease

Robert Chiesa¹, Jaap Jan Boelens², Christine Duncan³, Jörn-Sven Kühl⁴, Caroline Sevin⁵, Neena Kapoor⁶, Vinod Prasad⁷, Caroline Lindemans⁸, Simon Jones⁹, Hernan Amartino¹⁰, Mattia Algeri¹¹, Nancy Bunin¹², Cristina Díaz-de-Heredia¹³, Esther Shamir¹⁴, Alison Timm¹⁴, Elizabeth McNeil¹⁴, Andrew Dietz¹⁴, Paul Orchard¹⁵

¹ University College London, Great Ormond Street Hospital Institute of Child Health and Great Ormond Street Hospital NHS Trust, London, United Kingdom, ² Memorial Sloan Kettering Cancer Center, New York, United States, ³ Boston Children’s Hospital and Dana-Farber Cancer Institute, Boston, United States, ⁴ University Clinic Leipzig, Leipzig, Germany, ⁵ Hôpital Universitaire Hôpital Bicêtre-Hôpitaux Universitaires Paris Sud, Paris, France, ⁶ Children’s Hospital Los Angeles, Los Angeles, United States, ⁷ Duke Children’s Hospital and Health Center, Durham, United States, ⁸ Princess Maxima Center and University Medical Center Utrecht, Utrecht, Netherlands, ⁹ Saint Mary’s Hospital, Manchester, United Kingdom, ¹⁰ Hospital Universitario Austral and Medeos Medical Center, Buenos Aires, Argentina, ¹¹ Bambino Gesù Children’s Hospital, Rome, Italy, ¹² Children’s Hospital of Philadelphia, Philadelphia, United States, ¹³ Hospital Universitari Vall d’Hebron, Barcelona, Spain, ¹⁴ bluebird bio, Inc., Cambridge, United States, ¹⁵ University of Minnesota Children’s Hospital, Minneapolis, United States

Background: Early treatment with allogeneic hematopoietic stem cell transplantation (allo-HSCT) can prevent progression of cerebral adrenoleukodystrophy (CALD) and improve long-term survival but associated risks of graft-versus-host disease (GVHD) and graft rejection may increase morbidity and mortality. We report here on an observational study that assessed the impact of donor source and match, conditioning regimens, and CALD stage on allo-HSCT outcomes.

Methods: Boys with CALD with available baseline Loes score and Neurologic Function Score (NFS) (53/59 enrolled) were stratified by CALD stage into advanced disease (AD) (Loes >9 or NFS >1; n = 16) and two early disease cohorts (ED1 [Loes ≤4 and NFS ≤1; n = 24] and ED2 [Loes >4-9 and NFS ≤1; n = 13]) for efficacy analysis. Transplant-related safety outcomes, including GVHD, transplant-related mortality, and graft failure, were analyzed for all 59 boys by donor source, donor match and conditioning regimen. Data are post-first allo-HSCT only.

Results: Median follow-up was 23.0 (min 0.9 – max 49.5) months. Overall survival (95% CI) at Month 24 was 95.2% (70.7, 99.3) for ED1, 87.5% (38.7, 98.1) for ED2, and 53.0% (23.3, 75.9) for AD patients, with hazard ratios of 0.18 (95% CI 0.04, 0.91) for ED1/AD and 0.27 (95% CI 0.06, 1.38) for ED2/AD. At Month 24, major functional disabilities (MFD)-free survival (95% CI), for which events included second allo-HSCT, MFDs, or death, was 74.6% (51.9, 87.7), 74.0% (38.2, 91.0), and 41.3% (17.3, 63.9) for ED1, ED2, and AD cohorts, respectively; hazard ratios were 0.37 (95% CI 0.14, 0.99) for ED1/AD and 0.35 (95% CI 0.11, 1.16) for ED2/AD.

Median (min–max) Loes score and NFS at Month 24 were 2 (0–15) and 0 (0–1) in ED, and 14 (11–17) and 3 (1–15) in AD cohorts, respectively (Figure 1). Gadolinium enhancement (GdE+) resolution at last visit was observed in 29/30 (97%) patients with baseline GdE+ status and available follow-up.

Regarding transplant-related safety outcomes by transplant-related characteristics (Table 1), the only significant difference was fewer Grade 2-4 acute GVHD events with busulfan/fludarabine vs. busulfan/cyclophosphamide.

Conclusions: These data suggest that early CALD treatment, particularly in ED1, improves functional outcomes versus AD. Occurrence of transplant-related risks, even with matched related donors, highlights a need for alternative CALD treatments.

Table 1 Hazard ratios (95% CI) for transplant-related safety outcomes by transplant characteristics.

Clinical Trial Registry: Data described in this abstract are from the observational study ALD-103 (NCT02204904), which was sponsored by bluebird bio, Inc.

Disclosure: Chiesa, Robert; Sevin, Caroline; Lindemans, Caroline: Consulting fees from bluebird bio, Inc.

Boelens, Jaap Jan: Consulting fees from Avrobio, Advanced Clinical, Bluerock, Omeros, Takeda, and Race Oncology.

Duncan, Christine; Kapoor, Neena; Prasad, Vinod; Bunin, Nancy: Nothing to declare.

Kühl, Jörn-Sven: Honoraria from bluebird bio, Inc.

Jones, Simon: Consultancy and advisory board membership for, as well as equity ownership in and research funding from Orchard Therapeutics.

Amartino, Hernan: Research funding and honoraria fees from, and advisory board membership, for Takeda; research funding, honoraria fees from PTC Therapeutics; honoraria fees from Biomarin Pharmaceutical, Sanofi Genzyme.

Algeri, Mattia: Honoraria from bluebird bio, Inc., Miltenyi.

Díaz-de-Heredia, Cristina: Speaker’s Bureau for Novartis, Sobi; Speaker’s Bureau, travel for Jazz Pharmaceuticals; consulting fees from MSD.

Shamir, Esther; Timm, Alison; Dietz, Andrew: Employment, equity ownership, salary from bluebird bio, Inc.

McNeil, Elizabeth: Was an employee of bluebird bio, Inc. at the time of the study and reports; currently employed by Passage Bio.

Orchard, Paul: Clinical trial support received from bluebird bio, Inc., Magenta, Immusoft, Sanofi.

O095. Combining Enzyme Replacement Therapy And Hematopoietic Stem Cell Transplant for Wolman Disease: A Novel Treatment Paradigm for Optimal Long Term Disease Control

Jane E Potter¹, Gemma Petts¹, Arunabha Ghosh², Jane Kinsella¹, Fiona White², Jane Roberts², Kathryn Brammeier², Heather J Church², Stephen Hughes³, Helen Campbell¹, Denise Bonney¹, Christine Merrigan⁴, Joanne Hughes⁴, Pamela Evans⁵, Brian W Bigger⁶, Alexander Broomfield⁷, Simon A Jones⁷, Robert F Wynn¹

¹ Royal Manchester Children’s Hospital, Manchester, United Kingdom, ² Manchester Centre for Genomic Medicine, St Mary’s Hospital, Manchester, United Kingdom, ³ Royal Manchester Children’s Hospital, Manchester, United Kingdom, ⁴ Children’s Health Ireland at Temple Street, Dublin, Ireland, ⁵ Children’s Health Ireland at Crumlin, Dublin, Ireland, ⁶ University of Manchester, Manchester, United Kingdom, ⁷ St Mary’s Hospital, Manchester, United Kingdom

Background: Wolman disease is a rare lysosomal storage disease caused by pathogenic variants in the LIPA gene. Untreated patients do not survive beyond the first year of life. Hematopoietic stem cell transplant (HCT) historically carries a high mortality in WD, with rapidly progressive intestinal and liver failure around the time of HCT. Recent enzyme replacement therapy (ERT) has prolonged survival, along with dietary substrate reduction (DSR). The long term use of ERT is limited by ongoing gastrointestinal symptoms, neutralising anti-drug antibodies (ADA), the need for life long central venous access and the financial impact of treatment. We describe 5 patients who had received ERT with DSR and then underwent HCT.

Methods: A retrospective single centred analysis of all WD patients on ERT who received a HCT at Royal Manchester Children’s Hospital. All patients received allogeneic HCT. Primary endpoint is survival. Secondary endpoints include clinical and histological gastrointestinal changes, improvement in inflammatory biomarkers (Cholestane-3β, 5α,6β-triol) and peripheral blood chimerism.

Results: The most common indication for HCT in the 5 patients was worsening liver function on ERT with associated increase in ADA titre (3/5). Post HCT 4/5 patients are alive, 1/5 patients had persistent inflammation and cytopenia, and died 8 months after HCT of sepsis. In surviving patients the liver function and histology is significantly improved. Gastrointestinal improvements are most pronounced. Post HCT all patients diarrhoea gradually resolved, nutritional intake shifts from amino acids to intact proteins, minimal long chain fat to normal intake and gastrostomy feeds to oral. This is also reflected in the histology where the cholesterol clefts and foamy macrophages are reduced post HCT, and the villous structure improves. All surviving patient’s inflammatory biomarkers have improved post HCT. 3/4 patients have mixed chimerism, indicating a potential graft defect in WD. All 4 patients remain on ERT, but doses are reduced and less frequent, the aim is to stop overtime if tolerated.

Table 1 Summary of ERT, DSR + HCT patient group.

Flu-Treo-Thio – Fludarabine, Treosulfan, Thiotepa, GvHD - graft versus host disease, Treo-Cy – Treosulfan, Cyclophosphamide, PBSC – peripheral blood stem cell, MUD – matched unrelated donor, UCB - umbilical cord blood, VOD – veno-occlusive disease

Conclusions: ERT and DSR improve WD patient’s clinical condition pre HCT, reducing the risk of historically described high mortality associated with HCT. HCT addresses the limitations of ERT particularly with regards the gastrointestinal symptoms, even with the observed mixed chimerism. HCT is now a long term treatment for WD, and we propose that the paradigm of treatment for optimal long term performance might be ERT followed by HCT.

Disclosure: A Ghosh has received consultancy fees/honoraria from Alexion Pharmaceuticals, Inc.

SA Jones has received consultancy fees, grants, and travel support for conferences from Alexion Pharmaceuticals, Inc.

All others nothing to declare.

Infectious Complications

O096. Abstract already published

O097. The Association Between Cytomegalovirus And Invasive Fungal Infections in Allogeneic Hematopoietic Stem Cell Transplant Recipients: A Systematic Review And Meta-Analysis

Nipat Chuleerarux¹, Achitpol Thongkam¹, Saman Nematollahi², Veronica Dioverti-Prono², Kasama Manothummetha¹, Pattama Torvorapanit¹, Nattapong Langsiri¹, Navaporn Worasilchai¹, Rongpong Plongla¹, Ariya Chindamporn¹, Anawin Sanguankeo³, Nitipong Permpalung²

¹ Chulalongkorn University, Bangkok, Thailand, ² Johns Hopkins University School of Medicine, Baltimore, United States, ³ Siriraj Hospital, Bangkok, Thailand

Background: Cytomegalovirus (CMV) and invasive fungal infections (IFIs) are important opportunistic infections, and they are associated with poor outcomes in allogeneic hematopoietic stem cell transplant recipients (allo-HSCT). However, the inter-relationship between CMV and IFIs in allo-HSCT remains controversial. We conducted this study to understand the association between CMV infection, CMV serostatus, and IFIs post-transplant.

Methods: A systematic literature search was conducted from existence through December 4^th, 2020 using Medline, Embase, and ISI Web of Science databases. Screening, data extraction and study quality assessment were independently conducted by two reviewers. Data were analyzed with pooled effect estimates using random-effects model. IFI and CMV definitions were defined according to the standard guidelines. CMV infection was further classified into CMV reactivation (positive CMV viral load in blood without tissue invasion) and CMV disease (CMV reactivation with tissue invasion) for subgroup analyses. CMV serostatus was classified into 3 categories: high risk (D-/R+), intermediate risk (D+/R+, D+/R-), and low risk (D-/R-).

Results: A total of 20 and 13 studies were included for systematic review and meta-analysis, respectively. CMV infection significantly increased the risk of IFI with pooled hazard ratio (pHR) of 2.80 (1.96, 4.00), I² = 75% (Figure 1). The sensitivity analysis by using a leave-one-out method showed significant pHRs consistently. The source of heterogeneity was likely from studies conducted before 2010 (results not showed). Further subgroup analyses showed that CMV infection was associated with development of early IFI (≤ 40 days post allo-HSCT) with pHR of 3.92 (1.46, 10.53), I² = 85% and late IFI (> 40 days post allo-HSCT) with pHR of 2.94 (1.71, 5.03), I² = 82%. CMV disease and CMV reactivation significantly increased the risk of IFI with pHRs of 6.03 (4.04, 9.00), I² = 42%, and 4.95 (2.34, 10.48), I² = 0%, respectively. Among the 3 CMV serostatus risk groups, only the CMV high risk group significantly increased the risk of IFI with a pooled odds ratio of 1.34 (1.03, 1.74), I² = 0%.

Conclusions: CMV infection significantly increased the risk of early and late IFIs in allo-HSCT and the risk magnitude was stronger in CMV disease than CMV reactivation. The underlying mechanisms are likely that both CMV and IFIs have shared common risk factors, particularly host immunosuppressive states, and CMV infection can suppress the host immune system by its immunomodulating effects and anti-CMV treatment related leukopenia. Interestingly, only CMV D-/R+ was associated with post allo-HSCT IFIs. Whether CMV D-/R+ directly caused IFIs or IFIs occurred in CMV D-/R+ because of the increased probability of CMV reactivation in this high-risk group is unclear. Given the significant morbidity and mortality in post-allo HSCT patients with CMV disease, our findings raise the importance of CMV prevention strategies and consideration of IFI surveillance post CMV infection in allo-HSCT to reduce unfavorable outcomes.

Disclosure: Dr. Nematollahi reported receiving grant from the Fisher Center Discovery Program, Johns Hopkins University. Drs. Worasilchai, Plongla, and Chindamporn reported receiving grants from the Health Systems Research Institute (Thailand) and Rachadapiseksompotch Fund, Chulalongkorn University outside the submitted work. Dr. Permpalung reported receiving grants and salary support from the Health Systems Research Institute (Thailand), the Cystic Fibrosis Foundation, and the National of Heart, Lung, and Blood Institute outside the submitted work. Dr.Permpalung has served on the Advisory Board for Shionogi Inc. No other conflict of interest to declare.

O098. Nocardiosis Following Hematopoietic Cell Transplantation: 20 Years’ Experience. The International Study of Infectious Diseases Working Party of EBMT And Franco-Belgian Nocardia Study Group

Dina Averbuch¹, Julien De Greef², Amélie Duréault³, Lotus Wendel⁴, Gloria Tridello⁵, David Lebeaux^6,7, Malgorzata Mikulska⁸, Lidia Gil⁹, Nina Knelange⁴, Etienne Daguindau¹⁰, Christine Robin¹¹, Alienor Xhaard¹², Mahmoed Aljurf¹³, Yves Beguin¹⁴, Amandine Le Bourgeois¹⁵, Carmen Botella-Garcia¹⁶, Nina Khanna¹⁷, Jens Van Praet¹⁸, Nicolaus Kröger¹⁹, Tsila Zuckerman²⁰, Olivier Lortholary²¹, Arnaud Fontanet²², Rafael de la Camara²³, Julien Coussement²⁴, Johan Maertens²⁵, Jan Styczynski²⁶

¹ Faculty of Medicine, Hebrew University of Jerusalem, Hadassah Medical Center, Jerusalem, Israel, ² Saint-Luc University Hospital, Université catholique de Louvain (UCLouvain), Brussels, Belgium, ³ Hôpital Necker-Enfants Malades, APHP, Université Paris Descartes, Paris, France, ⁴ EBMT Data Office, Leiden, Netherlands, ⁵ Azienda Ospedaliera Universitaria Integrata Verona, Verona, Italy, ⁶ Université de Paris, Paris, France, ⁷ AP-HP, Hôpital Européen Georges Pompidou, Paris, France, ⁸ University of Genoa and Ospedale Policlinico San Martino, Genova, Italy, ⁹ University of Medical Sciences, Poznan, Poland, ¹⁰ University Hospital of Besançon, Besançon, France, ¹¹ Henri Mondor University Hospital, Creteil, France, ¹² Hospital St. Louis, Paris, France, ¹³ King Faisal Specialist Hospital & Research Center, Riyadh, Saudi Arabia, ¹⁴ CHU of Liège and University of Liège, Liège, Belgium, ¹⁵ CHU Nantes, Nantes, France, ¹⁶ CHU Bordeaux, Bordeaux, France, ¹⁷ University and University Hospital of Basel, Basel, Switzerland, ¹⁸ DAZ Sint-Jan Brugge-Oostende AV, Brugge, Belgium, ¹⁹ University Hospital Eppendorf, Hamburg, Germany, ²⁰ Rambam Medical Center, Haifa, Israel, ²¹ Université Paris Descartes, Sorbonne Paris Cité, AP-HP, Hôpital Necker Enfants Malades, Centre d’Infectiologie Necker- Pasteur and Institut Imagine, Paris, France, ²² Institut Pasteur, Unité de Recherche et d’Expertise Epidémiologie des Maladies Emergentes, Paris, France, ²³ Hospital de la Princesa, Madrid, Spain, ²⁴ Peter MacCallum Cancer Centre, Melbourne, Australia, ²⁵ Universitaire Ziekenhuizen, Leuven, Belgium, ²⁶ Nicolaus Copernicus University Torun, Bydgoszcz, Poland

Background: Nocardiosis is a rare infection after hematopoietic cell transplantation (HCT). Little is known regarding its presentation, management, and outcome in this population. We conducted an international retrospective study of nocardiosis in HCT recipients.

Methods: We reviewed nocardiosis episodes diagnosed during 1.1.2000-31.12.2018 in HCT recipients. The primary objective was to describe their clinical, microbiological, radiological, and outcome characteristics.

Results: We identified 81 nocardiosis episodes in 74 allo- and 7 auto-HCT recipients (14 countries). Nocardia infections occurred at a median of 8 (interquartile range [IQR] 4–18) months post-HCT. The most common underlying diseases were acute leukemia (25; 31%), myelodysplastic/myeloproliferative disease (18; 22%), and lymphoma (16; 20%). The median time between symptoms onset and diagnosis was 12 (IQR 5-23) days. The most common signs and symptoms were fever (66%), cough (44%), and neurological signs (30%). The most common organs involved were the lungs (70; 86%); the brain (30; 37%) and the skin/soft tissue (11; 14%).

Forty-six (57%) of nocardiosis episodes were disseminated (multiple organs involved (n = 43) and/or bacteremia (n = 13)). The most frequent combination was lungs and brain. The most common lung imaging findings were consolidations (49%) or nodules (47%; 59% of them cavitated); and brain imaging finding was multiple brain abscesses (63%).

Ten of 34 patients without neurological signs/symptoms had abnormal brain imaging. The following characteristics were significantly more frequent in patients with brain involvement: T-cell depletion, disseminated disease, neurological signs, and shorter time to diagnosis from symptom onset. The following factors were significantly more frequent in patients without brain involvement: respiratory signs/symptoms, other (non-respiratory and non-neurological) signs/symptoms, lower leukocyte count at presentation, treatment with ≥2 immune-suppressive drugs (vs. one), and bacterial infections prior or concomitant to nocardiosis.

At presentation, 47% of the patients were severely lymphocytopenic (<500 cells/mm³); 18% had acute and 44% chronic GVHD; 86% were receiving immune suppression (13% received steroids ≥1 mg/kg/day; 61% two or more agents), and 40% co-trimoxazole prophylaxis.

Species distribution is presented in Figure 1. All 45 tested isolates were susceptible to linezolid, 56/57 (98%) to amikacin, 57/63 (90%) to co-trimoxazole, and 49/57 (86%) to imipenem. The most commonly used antibiotics were co-trimoxazole (68%), imipenem (46%), amikacin (44%).

Three patients developed nocardiosis relapses within one year of therapy cessation; four remained with sequelae (mainly neurological).

Thirty-two (40%) patients died within one year from diagnosis; the median time to death in this group was 93 (IQR 38 - 187) days. In 11 patients, death was attributed to nocardiosis. The presence of bacterial infection within six months prior to nocardiosis was associated with one-year mortality (odds ratio 5.8; 95%CI 2.1-15.8, p = 0.001).

Conclusions: In HCT recipients with nocardiosis, dissemination and brain infection are particularly common. Brain imaging should systematically be performed in HCT recipients with nocardiosis, because brain involvement may occur in patients without neurological signs or symptoms; usually as a part of disseminated disease. Linezolid and amikacin are the most active antibiotics. One-year mortality is correlated with the presence of bacterial infections before or concomitant to nocardiosis.

Disclosure: There are no conflicts of interests and no funding for this study.

O099. Colonization-Guided Protocol For Empirical Antibiotic Therapy In ALLO-HSCT Recipients: First Results of A Single Center Prospective Study

Marina Popova¹, Yuliya Rogacheva¹, Aleksander Siniaev¹, Zhanna Fomina², Olga Pinegina², Anna Spiridonova², Oleg Goloshchapov¹, Yulia Vlasova¹, Maria Vladovskaya¹, Sergey Bondarenko¹, Ivan Moiseev¹, Alexander Kulagin¹

¹ Pavlov University, RM Gorbacheva Research Institute, Saint-Petersburg, Russian Federation, ² Pavlov University, Saint-Petersburg, Russian Federation

Background: Increasing resistance and the limited arsenal of new antibiotics, especially against resistant gram-negative bacteria (GNB) require carefully designed antibiotic regimens as part of antimicrobial stewardship program in allo-HSCT recipients. Prior colonization or infection by resistant GNB is the most important risk factor for infection with resistant pathogens. The role of appropriate initial antibiotic therapy in febrile neutropenia (FN) patients colonized with resistant organisms is still a subject of controversy.

Methods: We introduced a new colonization-guided protocol for empirical antibiotic therapy in allo-HSCT recipients based on a CHROMagar rapid detection of rectal colonization by resistant GNB (ESBL, carbapenemase) additionally to routine microbiological culture screening and monitoring from non-sterile sites (stool, throat, urinary), which guided the choice of initial empirical therapy regimens from May 2020. Initial empirical therapy was designed according to the colonizer’ mechanism of resistance or sensitivity at the date of FN onset. During six-month period of protocol implementation, we included 58 adult first allo-HSCT recipients. The control group was adult first allo-HSCT recipients from May to October 2019 (n = 79). Groups were comparable by main patients’ and HSCT characteristics. The median follow-up time was 142 (6-436) days after allo-HSCT.

Results: Significantly higher incidence of colonization by resistant GNB, predominantly ESBL, by the time of FN was registered in allo-HSCT recipients in was additional identified with CHROMagar screening. Overall proportion of identified colonization was 88% vs 42% (p < 0.00001), ESBL – 60% vs 23% (p = 0.00002), carbapenemase – 27.6% vs 19% (p = 0.3). Top three colonizers detected by conventional microbiological culture method were the same in two groups: Klebsiella pneumoniae – 58,5%, Escherichia coli – 19,5%, Pseudomonas aeruginosa – 14,6% in control group and 41%, 23%, 23% in study group. The etiology of microbiologically proved bloodstream infections (BSI) were represented predominantly by GNB was not different between the two groups: ESBL 35,3% vs 31,3%, carbapenemase 64,7% vs 68,7% (p > 0.1). OS at 60 days after resistant GNB BSI diagnosis was 58,8% vs 81,3% (p = 0.2) and also was not statistically differ in subgroups: ESBL – 100% vs 100% (p = 0.1), carbapenemase – 36,4% vs 70% (p = 0.2). 100-days TRM in control group was 20,3% and was significantly higher in colonized patients than non-colonized (33,3% vs 10,9%; HR 3,277 (1,137-9,440) p = 0,028). In study group 100-days TRM was 6,9% and was not differ in colonized and non-colonized patients (5,9% vs 14,3%, p = 0.4). TRM was not differ in control and study group (20,3% vs 6,9%, p = 0.069) however TRM significantly decreased in colonized patients in study group 5,9% vs 33,3% (HR 0.194 (0.054-0.698) p = 0.012).

Conclusions: New colonization-guided protocol for empirical antibiotic therapy in allo-HSCT recipients demonstrated significant improvement in detection of patients colonized by resistant gram-negative bacteria and significantly reduced the 100-days TRM in colonized patients.

Disclosure: Nothing.

O100. Colonization By Multidrug-Resistant Gram-Negative Bacteria in Early Phase of Hematopoietic Stem Cell Transplantation

Yuliya Rogacheva¹, Marina Popova¹, Aleksander Siniaev¹, Oleg Goloshchapov¹, Anna Spiridonova¹, Yulia Vlasova¹, Sergey Bondarenko¹, Ludmila Zubarovskaya¹, Alexander Kulagin¹

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

Background: Infectious complications caused by multidrug-resistant gram-negative bacteria (MDRGNB) is a global issue. Colonization by MDRGNB considered as a risk factor of infectious complications after hematopoietic stem cell transplantation (HSCT). Therefore, the question remains whether colonization by MDRGNB leads to an increased risk of subsequent bloodstream infections (BSI).

Methods: We retrospectively reviewed patients with hematological diseases who underwent allogeneic or autologous HSCT at RM Gorbacheva Research Institute (CIC 725) in 2019 year. This study included 203 adult patients with the median age of 35 years (18-73). Allogeneic HSCT was performed in 162 cases (80%), including haploidentical – 51 (25%); auto-HSCT – 41 (20%). The majority of patients had acute leukemia – 101 (49,6%), Hodgkin and non-Hodgkin lymphomas – 25 (12%), aplastic anemia – 14 (7%), myelodysplastic syndrome – 11 (5,4%), others – 52 (26%). The second allogeneic HSCT due to graft failure was performed in 13 patients (6,4%). We used a screening program with detecting bacteria by microbiological culture from non-sterile sites (stool, throat, urinary) twice a week during the HSCT. We studied colonization by MDRGNB before and after HSCT, BSI and overall survival (OS), all at D+100 after HSCT.

Results: Before HSCT 62 patients (30.5%) were colonized by at least 1 MDR bacteria, in the vast majority by ESBL (46 of 62 patients; 74%). The etiology of colonization was K. pneumoniae (n = 35, 56,5%), E. coli (n = 14, 22,6%), Pseudomonas spp. (n = 5, 8%), Enterobacter spp. (n = 1, 1,6%), Acinetobacter spp. (n = 1, 1,6%) and 6 patients (9,6%) were colonized with more than one bacteria, including K. pneumonia. After HSCT colonization rate was significantly higher (p < 0.001), 110 patients (54.1%) were colonized by at least 1 MDR bacteria, in the vast majority by ESBL+carbapenemases (72 of 110 patients; 65,4%, p < 0,001). The etiology of post-HSCT colonization was K. pneumoniae (n = 65, 59%), E. coli (n = 14, 12,7%), Pseudomonas spp. (n = 14, 12,7%), Enterobacter spp. (n = 1, 0,9%), Acinetobacter spp. (n = 1, 1,8%) and 14 patients (12,7%) were colonized with more than one bacteria, including K. pneumonia.

BSIs were diagnosed in 46 (23%) patients. The median day of BSI diagnosis was D+5 (0-81) after onset of febrile neutropenia. Gram-negative etiology was in 35 (74%) cases of BSI. In most cases BSI had been caused by K. pneumoniae – 22 (63%), E. coli – 4 (11,4%), Pseudomonas spp. – 3 (8,5%), Acinetobacter spp. – 1 (2,8%) and combination of K. pneumonia with others MDRGN – 5 (14,3%). Localized infection was registered in 51 (25%) cases. The etiology of BSI was the same bacteria that colonized non-sterile sites 2 weeks before the detection bacteria in bloodstream in 17 (48%) patients. Colonization by MDRGNB had significant association with bloodstream infection (p < 0.0001). Day 100 OS was 87,7% including 59% in patients colonized with MDRGNB and 90% in patients without colonization (p < 0.0001).

Conclusions: Colonization by multidrug-resistant gram-negative bacteria in early period after of HSCT reached 54,1% which was significantly higher than before HSCT. K. pneumonia was predominant etiology in both colonization and bloodstream infections. Colonization by multidrug-resistant gram-negative bacteria had significant association with bloodstream infection and death before D+100 after HSCT and decreased the overall survival.

Disclosure: Nothing to declare.

O101. Outcome of Covid 19 in Allogeneic Stem Cells Transplant Patients: The Single Center Experience From A Severely Affected Area (Bergamo)

Maria Caterina Mico’¹, Chiara Pavoni¹, Daniela Taurino¹, Paola Stefanoni¹, Alessandra Algarotti¹, Maria Chiara Finazzi¹, Federico Lussana¹, Anna Grassi¹, Alessandro Rambaldi^1,2

¹ ASST PG 23 Bergamo, Bergamo, Italy, ² University of Milan, Milano, Italy

Background: In Italy, Lombardia was the most affected region by COVID-19 and in the province of Bergamo 26500 cases have been reported by December 12.

Methods: We retrospectively analyzed the incidence of COVID-19 in patients undergoing allogeneic hematopoietic stem cell transplantation (alloHSCT) at our Hospital.

Results: Twenty-two allogeneic HSCT patients out of 668 in active follow-up at our hospital, proved positive for Sars CoV2 infection between February 29th and November 10th, 2020. Table 1 summarized reasons and type of transplant. When COVID-19 was diagnosed, the median age was 56 years (range 28-69), (male/female ratio, 55% /45%). Median follow-up was 212 days (range, 9–285) from the onset of COVID-19. The median time from alloHSCT to COVID-19 was 2615 days (range, 124-7741). The most recent disease status was: complete remission in 19 patients (86%), partial remission in 2 patients (9%) and progression disease in 1 patient (5%).

At COVID-19 onset, 2 or more comorbidities (hypertension, congestive heart failure, chronic obstructive pulmonary disease, diabetes mellitus, chronic kidney disease, dyslipidemia, thyroid disease) were present in 18 patients (81,8%). Neutropenia was present in 6 (27%) and lymphopenia in 7 (32%) patients. Eleven patients (50%) were under immunosuppressive treatment (tacrolimus, cyclosporine, ibrutinib, imatinib and/or ruxolitinib or ibrutinib). Ten patients (45%) had active GVHD and 7 patients were under steroids (32%). GVHD did not worsen during the COVID-19 infection. Eight patients had Covid-19 related pneumonia: six patients needed low pressure oxygen support and 2 patients required intensive oxygen support and intubation. Both these patients died (due to aspergillosis and cytomegalovirus infection). In addition to standard treatment with steroids and low molecular weight heparin, 3 patients received hyperimmune plasma due to a very prolonged (one patient more than 40weeks) nasal swab positivity to SarsCoV-2. The median time from diagnosis to swab negativization was 51days (range 8-70).

Overall survival (OS) at 30 days was 95%. The number of comorbidities and neutropenia or lymphopenia were not associated with survival.

Table 1

Characteristics	Disease numbers (%)
Disease
Acute Leukemias	ALL 1 (4.5), AML 11 (50)
Others	MDS 1 (4.5), MF 3 (13.6), CML 2 (9.1), MM 2 (9.1), NHL 2 (9.1)
Donor Type, n (%)
Sibling/MUD/Haplo/CB	9/10/2/1(41%, 45%, 9%, 4%)
Conditioning Regimen	MAC 13 (59%) RIC 9 (41%)
Cell Source of the Graft	BM 3 (13.6), PBSC 18 (81.8), Cord Blood 1 (4.5)

Conclusions: Within the limits of a single center experience, the mortality rate of HSCT patients with Covid-19 largely exceeded that observed in the general population.

Disclosure: Nothing to declare.

O102. Clinical Impact of De-escalation/discontinuation of Antibiotics Prior to Neutrophil Recovery in Patients Undergoing Hematopoietic Stem Cell Transplantation: A Single-centre Experience

Anke Verlinden¹, Sévérine De Bruijn¹, Sébastien Anguille¹, Zwi Berneman¹, Wilfried Schroyens¹, Alain Gadisseur¹

¹ University Hospital Antwerp, EDEGEM, Belgium

Background: With rising antibiotic resistance and knowledge that gut microbiome disruption has a negative impact on outcome, antibiotic policies are increasingly important. In contrast to IDSA/ESMO guidelines recommending antibiotic therapy until neutrophil recovery, ECIL-4 guidelines suggest earlier discontinuation under specific conditions. However, de-escalation/discontinuation prior to neutrophil recovery has not been widely adopted in clinical practice across Europe.

Methods: SOP for treatment of febrile neutropenia were adapted in accordance with ECIL-4 guidelines as of February 2017. Charts of patients admitted for allogeneic HSCT from November 2011 until October 2020 were reviewed for occurrence of febrile neutropenia, bacteraemia, severe sepsis, septic shock, antibiotic discontinuation, fever relapse, aGVHD, infection-related/overall mortality and duration of antibiotic therapy.

Results: 252 consecutive allogeneic HSCT admissions were included, 137 before versus 115 after introduction of stewardship SOP. Median age was 57 years (range 17-74) and 61.5% (155/252) were male. Majority of HSCT were performed for AML (46%; 116/252), followed by MDS (16.3%; 41/252) and ALL (10.3%; 26/252). Donor type was most frequently MUD (52%; 131/252), followed by sibling (36.9%; 93/252), haplo-identical (8.7%; 22/252) and mismatched MUD (2.4%; 6/252). Mean duration of hospitalisation and profound neutropenia was 30.1 days and 14.8 days respectively. Both study populations show equal distribution of these characteristics, except for a slight increase in proportion of haplo-identical transplants in the stewardship group.

Occurrence of febrile neutropenia was similar between groups [82.6% (95/115) vs 79.6% (109/137), p = 0.540]. Of the 322 recorded fever episodes, 74 (22.3%) were classified as microbiologically documented infections (MDI), 55 (16.6%) as clinically documented infections (CDI) and 203 (61.1%) as fever of unknown origin (FUO). ATG infusion was associated with 38.4% (78/203) of FUO episodes. Bacteraemia occurred more frequently in the stewardship group [47.8% (55/115) vs 16.1% (22/137); p < 0.001]. However, there was no increase in severe sepsis [7.0% (8/115) vs 3.6% (5/137); p = 0.237], septic shock [2.6% (3/115) vs 2.2% (3/137); p = 0.828], infection-related intensive care admission [3.5% (4/115) vs 2.2% (3/137); p = 0.705], overall mortality [1.7% (2/115) vs 2.2% (3/137); p = 0.798] or infection-related mortality [0.9% (1/115) vs 2.2% (3/137); p = 0.628] observed during admission.

Compliance with antibiotic discontinuation was high: 95% of MDI/CDI and 88% of FUO. Antibiotic therapy was stopped more frequently before neutrophil recovery [40.1% (65/162) vs 14.7% (25/170), p < 0.001]. Fever relapse occurred in 63.1% (41/65). Duration of antibiotic therapy was shortened significantly (median 10 days vs 14 days, p = 0.038).

Cumulative incidence of aGvHD at day 100 was significantly lower in the stewardship group [34.8% (40/115) vs 47.8% (= 65/136); p = 0.037], leaning towards less grade III/IV aGvHD [40.0% (16/40) vs 52.3% (34/65); p = 0.220]. Day 30 and day 100 overall mortality did not differ between groups [respectively 1.7% (2/115) vs 5.1% (= 7/137); p = 0.151 & 17.1% (19/111) vs 17.5% (24/137); p = 0.934] nor did infection-related mortality [respectively 1.7% (2/115) vs 2.9% (4/137); p = 0.540 & 5.4% (6/111) vs 4.4% (6/137); p = 0.708].

Conclusions: Our data confirm safety of antibiotic de-escalation/discontinuation prior to neutrophil recovery in allogeneic HSCT, leading to a significant decrease in antibiotic use. The associated decrease in aGvHD may be explained by confounding factors in transplant strategies or supportive care.

Disclosure: Nothing to declare.

O103. Influence of The SARS-COV-2 Pandemic on Bone Marrow Transplantation Centers And The Protocols Adopted In Brazil Between May And June 2020

Fernando Duarte¹, Talyta Ellen Jesus Santos Sousa², Abrahão Elias Hallack Neto³, Adriana Seber⁴, Afonso Celso Vigorito⁵, André Luís Gervatoski Lourenço⁶, Andresa Lima Melo⁷, Beatrice Araújo Duarte⁸, Beatriz Stela Gomes Souza Pitombeira Araújo⁹, Carmem Bonfim¹⁰, Celso Arrais¹¹, César Barian¹², Cilmara Kuwahara¹³, Decio Lerner¹⁴, Eduardo José de Alencar Paton¹⁵, Garles Miller Matias Vieira¹⁶, George Maurício Navarro Barros¹⁷, Gisele Loth¹⁸, Gustavo Machado Teixeira¹⁹, Jayr Schmidt Filho¹⁴, João Victor Piccolo Feliciano²⁰, Laura Maria Fogliatto²¹, Leandro Celso Grilo²², Leticia Navarro Gordan Ferreira Martins²³, Liane Esteves Daudt²⁴, Luís Fernando da Silva Bouzas²⁵, Marcio Soares Monção²⁶, Marco Aurelio Salvino²⁷, Maria Claudia Moreira²⁸, Maria Cristina M Almeida Macedo²⁹, Marina Assirati Coutinho³⁰, Nelson Hamerschlak³¹, Ricardo Chiattone³², Roberto Luiz da Silva³³, Rodolfo Soares³⁴, Rony Schaffel³⁵, Roselene Mesquita Augusto Passos³⁶, Thaisa Marjore Menezes Viana¹, Tatiana Dias Marconi Monteiro³⁷, Vaneuza Araújo Moreira Funke³⁸, Vergílio Antônio Rensi Colturato³⁹, Victor Gottardello Zecchin⁴⁰, Wellington Morais de Azevedo⁴¹, Yana Augusta S. Novis¹², Leandro de Pádua Silva⁴², Antonella Zanette⁴³, Renato Luiz Guerino Cunha³⁷, Evandro Maranhão Fagundes⁴⁴, Angelo Atalla⁴⁵, Vanderson Rocha⁴⁶

¹ Hospital Walter Cantídio, Federal University of Ceara, Fortaleza, Brazil, ² Federal University of Ceara, Fortaleza, Brazil, ³ University Hospital of Juiz de Fora, Juiz de Fora, Brazil, ⁴ Hospital Samaritano, São Paulo, Brazil, ⁵ State University of Campinas, Hemocentro TCTH, Campinas, Brazil, ⁶ Hospital dos Fornecedores de Cana de Piracicaba, Piracicaba, Brazil, ⁷ Grupo Acreditar – Oncologia DO’r, Brasília, Brazil, ⁸ Centro Universitário Christus (UNICHRISTUS), Fortaleza, Brazil, ⁹ Hospital Unimed, Fortaleza, Brazil, ¹⁰ Hospital Nossa Senhora das Graças, Curitiba, Brazil, ¹¹ Universidade Federal de São Paulo, São Paulo, Brazil, ¹² ACCG - Hospital Araújo Jorge, Goiânia, Brazil, ¹³ Hospital Pequeno Príncipe, Curitiba, Brazil, ¹⁴ Hospital A.C.Camargo Câncer Center, São Paulo, Brazil, ¹⁵ Oncobio Serviços de Saúde, Nova Lima, Brazil, ¹⁶ Hospital A.C. Camargo Câncer Center, São Paulo, Brazil, ¹⁷ Barretos Cancer Hospital, Brazilian Cooperative Group of MDS in Pediatrics, São Paulo, Brazil, ¹⁸ Federal University of Parana, Curitiba, Brazil, ¹⁹ Clinical Hospital of Federal University of Minas Gerais, Belo Horizonte, Brazil, ²⁰ Hospital de Base de São Jose do Rio Preto, São José do Rio Preto, Brazil, ²¹ Santa Casa de Misericórdia de Porto Alegre, Porto Alegre, Brazil, ²² Centro Médico de Campinas, Campinas, Brazil, ²³ Hospital Universitário Regional do Norte do Paraná - Bone Marrow Transplantation, Londrina, Brazil, ²⁴ Hospital de Clínicas de Porto Alegre, Porto Alegre, Brazil, ²⁵ Hospital Unimed Volta Redonda, Volta Redonda, Brazil, ²⁶ 26 Hospital Leforte Liberdade SA, São Paulo, Brazil, ²⁷ Hospital Português da Bahia, Salvador, Brazil, ²⁸ Complexo Hospitalar de Niterói, Niteroi, Brazil, ²⁹ Brazilian Institute of Cancer Control, Bone Marrow Transplant, São Paulo, Brazil, ³⁰ Hospital São Lucas, Rio de Janeiro, Brazil, ³¹ Hospital Israelita Albert Einstein, Bone Marrow Transplant, São Paulo, Brazil, ³² Hospital das Clínicas da Universidade Estadual de Campinas - Bone Marrow Transplantation, Campinas, Brazil, ³³ Hospital das Clínicas das Universidade Federal de Minas Gerais-Bone Marrow Transplantation, Belo Horizonte, Brazil, ³⁴ Hospital das Clínicas das Universidade Federal do Rio Grande, Petrópolis, Brazil, ³⁵ Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil, ³⁶ Instituto de Medicina Aeroespacial Brigadeiro Médico Roberto Teixeira, Rio de Janeiro, Brazil, ³⁷ Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São Paulo, Brazil, ³⁸ Federal University of Parana, Curitiba, Brazil, ³⁹ Amaral Carvalho Foundation, Jaú, Brazil, ⁴⁰ Instituto de Oncologia Pediátrica (UNIFESP), São Paulo, Brazil, ⁴¹ Santa Casa de Misericórdia de Belo Horizonte, Belo Horizonte, Brazil, ⁴² Hospital Santa Cruz, São Paulo, Brazil, ⁴³ Hospital Erasto Gaerther, Curitiba, Brazil, ⁴⁴ Hospital Luxemburgo, Belo Horizonte, Brazil, ⁴⁵ Hospital Monte Sinai, Juiz de Fora, Brazil, ⁴⁶ Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo-Bone Marrow Transplantation, São Paulo, Brazil

Background: During COVID-19 pandemic, Hematopoietic Stem Cell Transplantation Centers (HSCT) as well as other entities of onco-hematological treatment faced the challenge of continuing therapy.

Methods: This is a cross-sectional study carried out from May to June 2020, through the application of a pre-structured questionnaire of 14 questions about the possible interventions carried out in the HSCT units in the face of the COVI-19 pandemic. The project was approved by the Research Committee of the Walter Cantídio University Hospital (HUWC), in Fortaleza, Brazil, Before completing the questionnaire, those responsible for completing it signed a digital consent form, aiming to ensure the confidentiality, veracity, and security of the information. The questionnaire was published on the website of the Brazilian Society of Bone Marrow Transplantation (SBTMO) to be filled in by the technical health officials of the Brazilian HSCT units. Data were collected using the Google Forms application and analyzed using the Excel program.

Results: Table 1: Topics analyzed in the questionnaire

Topics	Positive answers from HSCT centers
	May (%)	June (%)
Use of SBTMO as protocol recommendation guideline	98	90.4
Contamination of health professionals	58	73.1
Laboratory test to confirm the SARS -CoV-2 in professionals	88.9	95
Deaths by COVID-19 in the intra or post-HSCT	26	44.2

From the 86 qualified centers in Brazil, 51 centers (59.3%) answered the questionnaire in May, which represents approximately 85% of all adult and pediatric transplants performed in Brazil. In June, 52 centers (60.4%) answered the questionnaire. In May, 4% of the centers interrupted the HSCT program and 12.2% maintained their operation without reduction. In most of them, there was a decrease in the number of HSCT, varying from 50% to 75% of the typical number in 59.2% of all centers. In June, this variation was 79.2%. The orientation for testing the donor and the asymptomatic patient in the pre-HSCT assessment was initially the main reason for discussion in the country, since the exams were not easily available. In May (88.2%) and in June (88.5%) RT-PCR exam collection was the greatest difficulty, due to the absence of tests or even an adequate place for the collection of samples. The most common symptoms reported were fever, cough, anosmia and headache. The treatment was performed with azithromycin (75%), hydroxychloroquine (55%), corticosteroids and ivermectin (both 15%). The use of immunosuppressants was maintained in 38.1% of the cases, decreased in 19% and discontinued in 14.3%.

Conclusions: The results reveal the vulnerability of patients with onco-hematological diseases to infection by COVID-19, specially during the HSCT procedures. Most of centers report they are following the coping recommendations proposed by scientific societies and are reducing the number of procedures during the pandemic. The current profile in Brazilian HSCT centers, related to the recommendations for coping with COVID-19 infection, will assist in making public policy decisions in a country such as Brazil, which suffers from increasing numbers of infection and rationalizing HSCT, so that patients who have urgency in their procedures are not harmed.

Disclosure: Nothing to declare.

O104. Immune Reconstitution During CMV Prophilaxis by Letermovir: A Single Center Experience

Ilaria Cutini¹, Antonella Gozzini¹, Chiara Nozzoli¹, Chiara Innocenti¹, Riccardo Boncompagni¹, Arianna Fani¹, Agata Suellen Guarrera¹, Giulia Raugei¹, Riccardo Saccardi¹

¹ Careggi University Hospital, Florence, Italy

Background: Cytomegalovirus (CMV) infection remains a frequent complication after allogeneic stem cell transplantation (HSCT). Letermovir (LET) was shown to be effective in preventing CMV re-activation and was EMA-approved for this indication in 2018. We speculate LET might have an impact on immune recovery after HSCT, resulting in an overall better CMV infection control.

Methods: We retrospectively evaluated 121 consecutive patients underwent HSCT for hematological malignancies during the last 5 years (from 2016 to 2020). All patients were CMV seropositive. We treated all patients from 2019 with LET and we compared it to an historical cohort which did not receive CMV prophylaxis. LET prophylaxis was administered from +2 to +100 days after HSCT. The immune reconstitution analysis was performed at +30, +60, +90, and +120 days from HSCT. Monitoring of acute or chronic GVHD, relapses and CMV reactivation after HSCT was carried out. Comparison of non-continuous variables was done by Chi-square test. Kaplan-Meyer curves were used to assess time to event analysis. Student T test for independent samples was used to assess the mean value differences.

Results: LET was administered to 31 patients. No difference was observed between the two cohorts in age at HSCT, gender, underlying disease, intensity of conditioning regimen, type of donor, graft source. LET group showed a lower incidence of overall CMV reactivation (16,1% vs 61,1%, p < 0.001). However, the difference was restricted to the LET administration time (6,5% in LET group vs 93,5%, p < 0.001), whilst beyond 100 days CMV reactivation was less frequent in the historical cohort (3.3 vs 12,9%, p = 0.049).

Regarding immune reconstitution, a higher mean of CD4⁺/CD8⁺ cells ratio at each time point was shown in LET group, (+30 days: p = 0.037, +60 days: p < 0.001, +90days: p < 0.001 and +120 days p = 0.010) as well as a higher CD3⁺CD8⁺ lymphocytes absolute mean value at +60, +90 and +120 days from HSCT (p < 0.001, p < 0.001 and p = 0.004, respectively). Moreover, a higher mean of both CD4⁺/CD8⁺ ratio and CD3⁺CD8⁺absolute lymphocytes were correlated to a lower incidence of CMV reactivation at every time point of the analysis (+30 days p = 0.002 and p = 0.003; +60 days p < 0.001 and p < 0.001; +90 days p < 0.001 and p < 0.001; +120 days p = 0.036 and p = 0.053, respectively). Although a higher CD3⁺CD4⁺ lymphocytes absolute mean value was observed in LET group only at +120 days from HSCT(p = 0.040), their protective role against CMV reactivation was reported at +30 and+60 days (p = 0.020 and p < 0.001, respectively).

The cumulative risk of CMV reactivation is lower in LET cohort at 1-year after HSCT (17,5% vs 63,7%, p < 0.001). LET did not show any impact on overall survival, acute and chronic GVHD, and incidence of relapse (p = ns).

Conclusions: A better CMV reactivation control was shown in LET group as reported in literature. LET seems to promote a better immune recovery after HSCT especially for CD3⁺CD8⁺ lymphocytes. The protective role of CD3⁺CD8⁺lymphocytes against CMV reactivation has already been described as well as the role of CD4⁺/CD8⁺ ratio and CD3⁺CD4⁺. The small sample size of LET group might prevent to assess other differences among the two groups.

Clinical Trial Registry: na.

Disclosure: Nothing yo declare.

O105. Allogeneic Stem Cell Transplantation in Acute Leukemia Patients After Covid-19 Infection

Maximilian Christopeit^1,2, Mirjam Reichard¹, Christian Niederwieser¹, Radwan Massoud¹, Evgeny Klyuchnikov¹, Nicolas Haase¹, Christine Wolschke¹, Francis Ayuk¹, Silke Heidenreich¹, Nicolaus Kröger¹

¹ University Medical Center Hamburg-Eppendorf, Hamburg, Germany, ² University Hospital Tübingen, Tübingen, Germany

Background: The risk for severe and fatal courses of COVID-19 is particularly high after allogeneic stem cell transplantation. Neither data nor recommendations exist regarding allogeneic stem cell transplantation in survivors of COVID-19.

Methods: Seven patients (n = 5 acute myeloid leukemia, AML, ELN intermediate or adverse risk, n = 2 acute lymphatic leukemia, ALL, each one standard risk with induction failure, BCRABL positive) who had survived COVID-19 were transplanted at our department. They had been diagnosed with AML/ ALL between December 2019 and April 2020 and contracted COVID-19 in March/ April 2020.

Results: COVID-19, measured as time from first positive throat and nose swab (TNS) to first of consecutive negative TNS had lasted 45 (median, range 12-70) days. ICU treatment had been necessary for 6 patients for 9 (median, range 2-22) days, 3 patients had been mechanically ventilated for 4, 4, and 12 days. Specific treatment had included plasma of reconvalescent donors, Lopinavir/ Ritonavir, and pentaglobin. ARDS, deep vein thrombosis, sepsis and catheter related blood stream infection had complicated COVID-19. Inflammation parameters were high. Anti-SARS-CoV-2 antibodies were found in 5/7 patients, 2 were antibody negative. Partially, patients had required antineoplastic treatment during COVID-19. Next to the 4 patients who needed mechanical ventilation and thus showed critical COVID-19, the remaining 3 patients who had not been in the need of mechanical ventilation were classified as having severe illness.

At the time of admission for alloSCT, all patients tested negative for SARS-CoV-2 (TNS). Thoracic computed tomography (CT) revealed small remnants of former COVID-19 infiltrations in 6 patients. Signs of pneumonia other than COVID-19 were visible in 2 patients, consistent with pulmonary mold infection. One patient was without CT signs of pneumonia. Pulmonary function testing immediately before the start of conditioning revealed moderate airway restrictions in 2 patients, with total lung capacity of 74%/ 70% and vital capacity of 67%/ 75%. Corrected diffusing capacity of lung for carbon monoxide (DLCOc) was between 45% and 87%, median 63%. No obstructive lung disease was evident. Time from onset of COVID-19 to alloSCT was 133 (median, 106-171) days, time from resolution of COVID-19 (first negative TNS) to alloSCT was 94 (64-136) days. Disease characteristics were adverse with many primary induction failures and molecular failures. Conditioning was rather intense with 6 patients receiving myeloablative conditioning, 4 including total body irradiation (3x 12 Gray, 1x 8 Gray). After a median follow-up of 77 (range 40-109) days, no reactivation of SARS-CoV-2 occurred. Leukocyte engraftment was reached in all patients after 14-31 (median 17) days. The 52 patients consecutively transplanted with peripheral blood stem cells in the same institution from April – May 2020 showed leukocyte engraftment after 7-21 days. The patients (n = 5) who had developed antibodies against SARS-CoV-2 remained antibody positive after transplant. Two patients never showed antibodies against SARS-CoV-2. All patients were discharged alive. Four patients had to be readmitted due to complications (each n = 1 sinuisoidal obstruction syndrome, relapse, infection, alloimmune phenomena).

Conclusions: AlloSCT post COVID-19 is feasible even briefly after COVID-19.

Disclosure: Nothing to declare.

O106. Colonization With Multidrug-resistant Bacteria and Disease Risk Index are Two Independent Factors That Affect The Survival of Patients After Allogeneic Haematopoietic Stem Cell Transplantation

Anna Czyz¹, Maciej Majcherek¹, Agnieszka Szeremet¹, Iwona Prajs¹, Tomasz Wrobel¹

¹ Wroclaw Medical University, Wroclaw, Poland

Background: Colonization of hematological patients with multi-drug resistant bacteria (MDRB) before allogeneic haematopoietic stem cell transplant (alloHCT) is of great concern, since a high mortality rate after transplant has been demonstrated in those patients. We aimed at determining whether MDRB colonization prior to transplant independently impacts survival after alloHCT.

Methods: A study group consisted of patients who underwent alloHCT between January 2018 and October 2020 and had been microbiologically evaluated for colonization prior to transplant. Data was collected from our institutional database.

Results: We analyzed 112 patients with various hematological diseases who received myeloablative (n = 46), reduced intensity (n = 47) or reduced toxicity conditioning (n = 19) and graft from identical siblings (n = 12), matched unrelated donor (n = 80) or family haploidentical donor (n = 20). Fifty three (47%) patients were colonized with MDRB prior to alloHCT, including 35 (31%) colonized with MDR Gram(-) bacteria, and 18 (16%) colonized with both MDR Gram(-) and Gram(+) bacteria. The most common colonizing MDRB were ESBL-producing Klebsiella pneumoniae and E.coli (31% of patients). A total of 37 patients (33%) had clinically or microbiologically documented G3-4 infections after alloHCT. The rate of G3-4 infections was significantly higher in MDRB colonized patients (47% vs 23%; p = 0.002). After the median follow-up time of 12 months (range, 2-33) overall survival (OS) for all patients was 64% at 18 months. In univariate analysis colonization with MDRB was identified as factor associated with shorter OS (58 vs. 71% at 18 months; p = 0.042; Figure). In addition, Disease Risk Index (DRI) was significantly correlated with OS rate (90%, 66%, and 35% at 18 months for low, intermediate and high DRI, respectively; p = 0.012; Figure). In multivariate analysis both factors, high DRI (HR 2.73, 95% 1.25-5.95; p = 0.011), and colonization with MDRB (HR 2.21, 95%1.03-4.74, p = 0.041) remained the only independent predictors for poor OS. Non-relapse mortality (NRM) in the whole group was 20.5% (95%CI13-31) at 18 months. MDRB colonization was associated with significantly higher overall mortality rate (36% vs 19%; p = 0.04), and infection-associated mortality rate (17% vs 3.5%; p = 0.019).

Conclusions: Our results confirm that colonization with MDRB is associated higher incidence of infections, as well as higher rates of mortality and infection-associated mortality after alloHCT. In addition, MDRB colonization, along with high DRI are two independent predictors for poor survival after transplant.

Disclosure: Authors have no conflict of interest.

O107. Prevention of Invasive Fungal Infection In Haplo Transplantation With Post-Transplantation Cyclophosphamide. Universal Fluconazole Prophylaxis And A Mould Diagnostic-directed Approach. Results in 112 Patients

Amado Karduss -Urueta¹, Angela Trujillo¹, Giovanni Ruiz¹, Rosendo Perez¹, Angelica Cardona¹

¹ Instituto de Cancerologia- Clinica Las Americas AUNA, Medellin, Colombia

Background: Patients who underwent haplotransplantation with pos-transplantation cyclophosphamide (Haplo-PTCy) are considered at high risk for Invasive fungal infection (IFI); indeed, ECIL 3 guideline suggests in this group of patients universal anti mould prophylaxis (BII level). However, there is a paucity of information about which is the best strategy for preventing IFI in this group. We inform our experience in this field.

Methods: Patients who received IFI primary prophylaxis and haplo-PTCy were included. The preventive strategy consisted of HEPA filtered rooms, fluconazole 400 mg from d- 8 to d +60, routine galactomannan determination twice a week during the first two years of the strategy, and later, only for evaluation of persistent fever or respiratory symptoms; also, chest or sinus CT imaging was done, and bronchoalveolar lavage was recommended for the evaluation of lung infiltrates. Empirical therapy with caspofungin was allowed in suspicious cases while IFI was confirmed or excluded. IFI was diagnosed according to EORTC 2008 criteria.

Results: From Jan 2016 to Aug 2020, one hundred twelve patients met the inclusion criteria. The conditioning used was a RIC regimen of fludarabine plus busulfan, or, melphalan and low TBI doses. The main characteristics of the group are presented in table 1.

The cumulative incidence of IFI at day + 100 was 6.2%; there were 4 probable and 3 possible cases. The diagnosis in probable cases was invasive aspergillosis in three and candida pneumonia in one. Mean time to IFI diagnosis was 43 days (range 12-92). Fifteen patients received empirical caspofungin; five of them had a confirmed diagnosis of IFI and in ten it was ruled out. Of the 7 patients with IFI, 3 died after 5, 18, and 30 days of treatment; however, all of them had, at the time of death, sepsis, and multi-organ failure associated with a bloodstream infection. The overall IFI associated mortality of the 112 patients was 2.6%.

Table 14 Table 1.

Conclusions: The IFI incidence in our cohort is placed in the low range of which has been informed with the use of haplo-PTCy platform which varies from 8 to 15% (1), and it is slightly superior of the threshold of 5%, above what ECIL guideline recommends for the use of universal anti-mould prophylaxis. Despite it is difficult to compare results among different studies, with the data obtained from this series of 112 cases, we can infer, that the use of fluconazole and an intensive mould diagnostic approach is a good strategy to prevent IFI in patients who undergo haplo-PTCy. However, well-structured clinical trials are needed to answer the question about which is the best alternative in this population.

1-Mediterr J Hematol Infect Dis. 2016;8(1): e2016057

Disclosure: All the authors declare do not have conflicts of interest for presenting this abstract.

O108. Strong Inflammatory Cytokine Response After Haploidentical Hematopoietic Stem Cell Transplantation With Post-Transplant Cyclophosphamide Relates to Human Herpesvirus 6 Reactivation

Kenichiro Takeda¹, Takanobu Morishita¹, Tomoki Naito¹, Kohei Ishigiwa¹, Yosuke Domon¹, Tomoe Ichiki¹, Motoki Eguchi¹, Yuka Kawaguchi¹, Rena Matsumoto¹, Marie Ohbiki¹, Tatsunori Goto¹, Yukiyasu Ozawa¹, Koichi Miyamura¹

¹ Japanese Red Cross Nagoya First Hospital, Nagoya, Japan

Background: Haploidentical hematopoietic stem cell transplantation (Haplo-SCT) has been greatly increased because of rapid and high possibility of donor availability. High dose post-transplant cyclophosphamide (PT-Cy) is being reported to be an effective GvHD prophylaxis while preserving anti leukemic allo-reactivity (GVL). Although degree of immediate post-transplant inflammatory cytokine response dramatically differs by cases, the correlation between the cytokine response and following transplant outcomes is currently under investigation.

Methods: To examine the correlation between cytokine response and transplant outcomes, we retrospectively analyzed patients (pts) receiving Haplo-SCT with PT-Cy from 2015 to 2020 at our institute. The degree of cytokine response was evaluated by max body temperature (mBT) and max serum C-reactive protein (mCRP) values between days 0 to 7 after Haplo-SCT.

Results: Twenty-two pts (female 18, male 4, median age 55) received Haplo-SCT in the period: AML 9 (de novo AML 4, AML with MRC 5), MDS 6, ALL 1, ATL 1, CML 1, MF 2, and others 2. Eight pts achieved complete remission disease status before transplantation. Myeloablative conditioning regimens were performed in 9 cases. It is noteworthy that 2 pts had HCT-CI≥3 and 4 pts received as 2nd transplantation. Median follow up period was 224 days after Haplo-SCT. One-year overall survival (OS), cumulative incidence of relapse (CIR), non-relapse mortality (NRM) was 62.0% (95% CI 34.8% - 80.5%), 19.8% (95% CI 4% - 45%), 18.2% (95% CI 5% - 36.8%), respectively. Patients with higher mBT/mCRP had significantly higher incidence rate of Human Herpesvirus 6 (HHV6) reactivation (Figure, Gray test, p = 0.01). ROS curve analysis was performed to determine the accuracy and threshold values of mBT/mCRP to predict transplant outcomes. Analysis showed that AUC of mBT/mCRP and HHV6 was 0.786 (95% CI 0.59 - 0.98), 0.854 (95% CI 0.691 - 1), respectively. Threshold value of mBT was 39.5°C (Sensitivity 68.8%, Specificity 83.3%), and mCRP was 14.8 mg/dL (Sensitivity 68.8%, Specificity 100%). In our opinion, higher inflammatory cytokine response may have obstructed rapid immune reconstitution after the transplantation and led to higher rate of HHV6 activation. Risk factors for high cytokine response was not found. Total cell count, CD34 positive cell count, or CD3 positive cell count were not related to higher cytokine response in this analysis. No correlations were seen between mBT/mCRP and other outcomes: OS, CIR, GvHD, time to engraftment, or sinusoidal obstruction syndrome/thrombic microangiopathy. Short follow up period and a low number case are limitations in this study. Because HHV6 reactivation is reported to lead to poor transplant outcome, we need to elucidate the association of inflammatory cytokine response and other outcomes. Adjusting the allo-reactive immune response to adequate degree may be the key point after Haplo-SCT with PT-Cy to successfully reconstitute immune system to avoid relapse and viral reactivations.

Conclusions: These data suggest that immediate post-transplant inflammatory cytokine response in Haplo-SCT with PT-Cy may correlates with HHV6 reactivation.

Disclosure: Nothing to declare.

O109. The Management of Clostridioides Difficile Infections In Patients After Allogeneic Hematopoietic Cell Transplantation: The Results of The Infectious Diseases Working Party EBMT Survey

Agnieszka Piekarska¹, Lidia Gil², Malgorzata Mikulska³, Patrycja Mensah-Glanowska⁴, Gulia Sbianchi⁵, Lotus Wendel⁵, Nina Knelange⁵, Dina Averbuch⁶, Rafael de la Camara⁷, Jan Styczynski⁸

¹ Medical University of Gdansk, Gdansk, Poland, ² Poznan University of Medical Sciences, Poznan, Poland, ³ University of Genoa (DISSAL) and IRCCS Ospedale Policlinico San Martino, Genova, Italy, ⁴ Collegium Medicum, Jagiellonian University, Cracow, Poland, ⁵ EBMT Data Office, Leiden, Netherlands, ⁶ Hebrew University of Jerusalem, Hadassah Medical Center, Jerusalem, Israel, ⁷ Hospital de la Princesa, Madrit, Spain, ⁸ Collegium Medicum, Nicolaus Copernicus University, Bydgoszcz, Poland

Background: The diagnostic and therapeutic approach to Clostridioides difficile infections (CDI) in allogeneic hematopoietic cell transplantation (HCT) recipients may vary according to the centre practices. Since guidelines dedicated to this specific group of patients are lacking, the Infectious Diseases Working Party undertook a survey to assess the management of CDI.

Methods: A questionnaire was mailed to the EBMT centres from July to December 2019. Responses were obtained from 144 centres (response rate 25%).

Results: Centre profile: 95 responses referred to HCT in adults, 34 in children and 15 in both groups. The CDI incidence <5% was reported by 46% of adult and paediatric centres, between 5-10% by 43% of adults and 36% of paediatric centres, while the centres treating both groups with the common approach, reported lower CDI incidence in children: <5% in 64% comparing to adults (46%).

CDI diagnosis: Paediatric centres more likely screen for CD colonization at admission in patients without CDI history (24%) and with CDI history (41%) comparing to adult centres: 11% and 30%, respectively. Diagnosis of CDI is based on stool toxins A/B detection in most centres, while glutamate dehydrogenase detection and/or verification with nucleic acid amplification tests are performed in <50% of centres. Most centres adhere to the recommended criteria for severe and fulminant CDI; however, leucocytosis >15000 cells/ml, increased creatinine, and decreased albumin were considered not applicable in HCT patients by 57%, 41%, and 41% of responders, respectively.

Treatment of CDI: Oral vancomycin or oral metronidazole is the 1st line treatment for non-severe CDI in 49% and 45% of centres, respectively, while for the 2nd line, oral vancomycin (51%) or fidaxomicin (26%) are preferred. Oral vancomycin or intravenous metronidazole is the treatment of choice in severe CDI in 60% and 31% of centres, respectively. Combined therapy with oral vancomycin and intravenous metronidazole is administered in fulminant CDI in 54% of centres, and surgical intervention is performed (34%) in case of complications. The median duration of effective CDI treatment lasts 7-10 days (in 61%). In the 1st recurrence of CDI, most centres opt for oral vancomycin, especially when metronidazole was previously administered, while in second or subsequent recurrences, vancomycin or fidaxomicin are preferred. The choice of therapy was mostly justified by high efficacy and centre experience.

Faecal microbiota transplantation is considered safe in HCT patients with recurrent CDI by 74% of responders; however, its use was declared by 30% of centres. Most responders (63%) are not convinced that available probiotics prevent CDI recurrence.

In 35% of centres, the increased prevalence of gastrointestinal-GVHD (GI-GVHD) was observed in patients suffering from CDI, but objective data are mostly lacking.

Almost 90% of responders see a need for CDI guidelines dedicated to HCT patients.

Conclusions: Definitions of severe and fulminant CDI are partially unsuitable for HCT recipients.

CDI incidence, diagnostic and therapeutic approaches vary. Still, most centres declared the significance of antimicrobial therapeutic efficacy, probably considering the risk of recurrent CDI courses after HCT and possible GI-GVHD promoting. However, newer diagnostic and treatment modalities are not frequently used in HCT patients.

Disclosure: Nothing to declare.

O110. Pattern of Fungal Infections During The Initial 100 Days In Children Undergoing Hematopoietic Stem Cell Transplantation From A Tertiary Referral Center in India

Harika Varla¹, Satish Kumar Meena¹, Rumesh Chandar Meiyalahan¹, Venkateswaran Vellaichamy Swaminathan¹, Ramya Uppuluri¹, Indira Jayakumar¹, Revathi Raj¹

¹ Apollo Cancer Institutes, Chennai, India

Background: Children undergoing hematopoietic stem cell transplantation (HSCT) are at risk of invasive fungal infections during the neutropenic period and also during the post engraftment period. Invasive fungal infection is associated with significant mortality and morbidity. We analyzed the impact of mycafungin based prophylaxis in preventing fungal infections in our unit.

Methods: We performed a retrospective analysis of the data on fungal infections in all children who underwent HSCT at our blood and marrow transplantation unit over a 3year period from January 2017 to January 2020. The children were aged from 5 months to 18 years of age. The data was analysed from day of infusion of stem cells till day 100. All children received mycafungin prophylaxis at 1mg/kg/day from day 0 till engraftment. We defined probable fungal infections as positive serum galactomannan or serum beta D glucan or fungal specific nodules in CT chest. A culture positive fungal infection was termed proven fungal infection.

Results: A total of 253 children were included with M:F ratio of 1.5:1. Among the total children, 29 (11%) had probable fungal infections while 6 (3.9%) had proven fungal infections. Among the total 204 non malignant transplants 2% (6/204) had proven fungal sepsis while 11.2% (23/204) had probable fungal sepsis. Among the total 49 children with malignant conditions, 12.2% (6/49) had probable fungal sepsis while none developed proven fungal sepsis. The incidence of fungal infections in the 88 children who underwent haploidentical transplantation was high at 21.5% (19/88) when compared to 1% (1/89) among the 89 children who underwent matched sibling donor transplantation was documented . Myeloablative conditioning (MAC) was used in 209 (82.6%) children and 44 (17.4%) received reduced intensity conditioning (RIC). Among the probable fungal infections, 10.5% (22/209) were in the MAC group while 15.9% (7/44) were in the RIC group. In proven fungal infections, 2% (5/209) received MAC while 2% (1/44) received RIC. The mortality rate in the children with fungal infection was 20.5% (6/29).

Conclusions: Mycafungin provides effective prophylaxis against fungal infections in children undergoing HSCT. The incidence of breakthrough infections is low at 13.8% in this cohort of high risk children including those with primary immune deficiency disorders, marrow failure and relapsed leukemia. The children at high risk of breakthrough infections are those undergoing haploidentical HSCT. Strategies implemented to reduce our current mortality of 20.5% due to fungal infections include augmenting the dose of mycafungin from 1 mg/kg/day to 2mg/kg/day or azole prophylaxis in selected children, meticulous follow up of serum galactomannan and beta D glucan and introduction of liposomal amphotericin B, early removal of central lines in candida sepsis and liberal use of HRCT chest to help diagnose invasive aspergillosis.

Disclosure: Nothing to declare.

Lymphoma and Chronic Lymphocytic Leukemia

O111. Abstract already published

O112. Validation of Gwas-Identified Variants For Chronic Lymphocytic Leukemia: A Study in The Context of The Crucial Consortium

Antonio José Cabrera-Serrano¹, Ana Moñiz-Díez¹, Paloma García², José Manuel Sánchez-Maldonado¹, Rob Ter Horst³, Daniele Campa⁴, Yasmeen Niazi^5,6, Stefano Landi⁴, Francisca Hernández-Mohedo⁷, Pedro González-Sierra⁷, Mónica Bernal⁷, Esther Clavero⁷, Miguel Ángel López-Nevot⁸, Lucía Moratalla⁷, Elisa López-Fernández⁷, Antonio Romero⁷, Yang Li³, Tzu Chen-Liang⁹, Irene Gámez⁹, Juan José Rodríguez-Sevilla¹⁰, Leonardo Potenza¹¹, Kari Hemminki⁵, Víctor Moreno¹², Federico Canzian⁵, Asta Försti^5,6, Rafael Marcos-Gragera¹³, María García-Álvarez¹⁴, Mihai Netea³, Yolanda Benavente¹², Javier Llorca¹⁵, Andrés Jerez⁹, Aleksandra Butrym¹⁶, Delphine Casabonne¹², Mario Luppi¹¹, Miguel Alcoceba¹⁴, Silvia de Sanjosé¹², Manuel Jurado⁷, Juan Sainz^1,8

¹ GENYO, Centre for Genomics and Oncological Research, Granada, Spain, ² Hospital Campus de la Salud, Granada, Spain, ³ Radboud University Nijmegen Medical Center, Nijmegen, Netherlands, ⁴ University of Pisa, Pisa, Italy, ⁵ German Cancer Research Center, Heidelberg, Germany, ⁶ Hopp Children’s Cancer Center (KiTZ), Heidelberg, Germany, ⁷ Virgen de las Nieves University Hospital, Granada, Spain, ⁸ University of Granada, Granada, Spain, ⁹ Morales Meseguer University Hospital, Murcia, Spain, ¹⁰ Hospital del Mar, Barcelona, Spain, ¹¹ University of Modena and Reggio Emilia, Modena, Italy, ¹² Institut Català d’Oncologia, L’Hospitalet de Llobregat, Spain, ¹³ University of Girona, Girona, Spain, ¹⁴ University Hospital of Salamanca, Salamanca, Spain, ¹⁵ University of Cantabria, Santander, Spain, ¹⁶ Medical University, Wrocław, Poland

Background: Chronic lymphocytic leukemia (CLL) is the most common leukemia among adults in western countries. Although advances in treatment options have been made, CLL remains as an incurable malignancy. Genome-wide association studies (GWAS) have identified multiple susceptibility loci for CLL. However, despite these important findings, the genetic component underlying CLL has not been completely unraveled and most of the GWAS-identified variants still require validation.

Methods: We conducted a population-based case–control study to validate associations between genetic variants identified through GWAS and CLL risk. CLL patients were diagnosed by experienced clinicians and ascertained through the Consortium for Research in Chronic lymphocytic Leukemia (CRuCIAL). The single-nucleotide polymorphisms (SNPs) were selected through an extensive literature search of relevant GWAS and meta-analyses published by 2019 using publicly available online databases. Additional criteria were potential functionality and linkage disequilibrium between the SNPs. A total of 35 SNPs in 34 loci were selected for genotyping. Hardy–Weinberg equilibrium was assessed in the control group (P > 0.01) and the association between CLL and the SNPs was tested using a multivariate unconditional logistic regression analysis. Finally, we conducted a meta-analysis of the CRuCIAL results with those from previous GWAS or GWAS meta-analysis and the I² statistic was used to assess statistical heterogeneity between the studies. The pooled OR was computed using the fixed-effect model. Statistical analyses were conducted using the STATA v12.1 software. Functional characterization of the most interesting markers was tested using data from the Human Functional Genomic Project (HFGP) cohort (cytokineQTL data after in vitro stimulation experiments, 91 annotated immune cell populations and serum hormonal and proteomic profiles; n = 520 healthy subjects).

Results: This study included 933 European CLL cases and 1,333 controls. All selected SNPs were in HWE in the control population (P < 0.01) and logistic regression analysis confirmed the association of 14 SNPs with CLL risk at P < 0.05. The strongest association was found for the GRAMD1B_rs35923643 SNP. Each copy of the GRAMD1B_rs35923643G allele increased the risk of CLL by 90% (OR = 1.90, P = 2.22∗10⁻¹⁶). Interestingly, we found an independent association with an increased risk of CLL within the GRAMD1B locus (rs2953196; OR = 1.28, P = 5.00∗10⁻⁴), which confirmed a relevant role of this gene in determining the risk of developing the disease. We also confirmed associations with CLL risk for SNPs within the IRF4, FARP3, and ODF3B loci (OR = 1.40, P = 3.76∗10⁻⁸; OR = 1.33, P = 1.25∗10⁻⁶; OR = 1.26, P = 2.00∗10⁻⁴) and more modest associations for SNPs within AC107990.1||INFE2L3P1, ACOXL, BCL2, CXXC1, FAS, GPR37, OPRM1, POT1, and QPCT (OR = 1.17-1.24). Importantly, the direction of the association of these SNPs was similar to the one reported in previous GWAS and the meta-analysis confirmed the role of these loci to modulate the risk of CLL. No significant heterogeneity was found, with the exception of the meta-analysis for the ACOXL_rs58055674 SNP (I² = 82.3%, P = 0.018). Functional studies are ongoing.

Conclusions: This study confirmed that SNPs within the GRAMD1B, IRF4, FARP3, ODF3B, AC107990.1||INFE2L3P1, ACOXL, BCL2, CXXC1, FAS, GPR37, OPRM1, POT1, and QPCT||RNU6-1116P loci are consistently associated with CLL risk and might be used to predict disease onset.

Disclosure: The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. All authors declare no conflict of interest.

O113. Evaluation of Autologous and Allogeneic Hematopoietic Cell Transplantation (HCT) in Patients With High-Risk Anaplastic Large Cell Lymphoma (ALCL)

Natalie L. Smith¹, Corinne Summers^1,2, Ted Gooley^1,2, Andrei Shustov^1,2, Rachel Salit^1,2, Rebecca Gardner^1,3, Ann Dahlberg^1,2, Mohamed L. Sorror^1,2, Leona Holmberg^1,2, Brenda M. Sandmaier^1,2, David Maloney^1,2, Monica S. Thakar^1,2

¹ University of Washington, Seattle, United States, ² Fred Hutchinson Cancer Research Center, Seattle, United States, ³ Seattle Children’s Hospital, Seattle, United States

Background: Patients with high-risk ALCL are often referred for autologous (auto) and/or allogeneic (allo) HCT for curative intent. However, there are limited data available to guide transplant-related decision making outside of multi-site or registry studies.

Methods: Single-center retrospective cohort study evaluating patients with high-risk ALCL undergoing auto and/or allo HCT at Fred Hutchinson Cancer Research Center. Kaplan-Meier method was used to estimate probabilities of progression free survival (PFS), non-relapse mortality (NRM), relapse, and overall survival (OS).

Results: Between 1997-2019, 41 patients with a median age of 44 (3-67) years underwent HCT [auto alone, n = 24; auto then allo (auto-allo), n = 6; allo alone, n = 11]. Choice of HCT strategy was physician-directed. At diagnosis, 54% had stage III-IV disease based on Ann Arbor staging and had a median of 4 (1-7) prior lines of therapy. 17% of auto and 47% of allo/auto-allo combined groups had ALK-positive disease. Median time from diagnosis to HCT was 14.7 (4.6-181.6) months. 54% were in anatomical remission (CR) prior to HCT. Auto conditioning consisted predominantly of cyclophosphamide/12 Gy total body irradiation (TBI) in the era <2009 (75%), dropping to 40% ≥2009. In the combined allo/auto-allo group, conditioning was reduced-intensity in 59%, including all of the auto-allo recipients, with fludarabine/2 Gy TBI being the predominant regimen (90%). Donors included HLA-matched sibling (n = 5), 10/10 unrelated (n = 7), and other alternative donors (n = 5). No grades 3-4 acute graft-versus-host disease (GVHD) were seen. Of the 12 patients who survived >200 days, 7 developed chronic GVHD, of whom none experienced post-HCT relapse. Overall, 18 relapses occurred at a median of 4.2 (0.9-40.3) and 2.5 (1.4-3.9) months after auto (n = 13) and allo/auto-allo (n = 5) HCT, respectively. With a median survivor follow-up of 9.0 (0.2-19.1) years, the 1-year/5-year cumulative incidence of PFS, relapse, and NRM in the auto group was 54%/44%, 46%/56%, and 0%/0%, and 70%/62%, 30%/30%, and 0%/8% in the allo/auto-allo group (A, B). Patients who entered HCT in CR (vs no CR) had numerically higher OS at 5-years: 83% (vs 27%) in auto group and 73% (vs 50%) in allo/auto-allo cohort (C, D). The 5-year OS was 86% vs 58% in ALK-negative vs positive allo/auto-allo patients. While auto HCT had similar 5-year OS across the timespan of this study (≥2009, 59% vs <2009, 56%), patients undergoing allo/auto-allo HCTs in the contemporary cohort showed numerically higher OS (≥2009, 90% vs <2009, 17%) (E, F).

Conclusions: In the largest single-center experience reported to date, we confirm that allo-HCT, including auto-allo HCT, is a reasonable and relevant treatment option conferring low rates of NRM for select high-risk ALCL patients. Given the introduction of highly active novel agents into modern ALCL treatment algorithms, further work is needed to characterize which patients would most benefit most from the additional graft-versus-lymphoma advantage that allo-HCT offers.

Disclosure: Monica Thakar is a consultant for IDRI and has received speaker honorarium/research funding from Miltenyi.

Rebecca Gardner received honorarium from Pfizer.

Mohamed L. Sorror was on advisory board committee and received honorarium from JAZZ Pharmaceuticals in June 2019.

David Maloney is an ad hoc consultant, having received honoraria from: A2 Biotherapeutics, Amgen, Bioline Rx, BMS, Juno Therapeutics, Celgene, Gilead, MorphoSys, Novartis, Genentech, Legend Biotech, and Janssen. In addition, D. Maloney has received research funding through Fred Hutchison Cancer Research Center for clinical trials on which he is a Principal Investigator from: Kite Pharma, Juno Therapeutics, and Celgene. He has patents pending (no licenses or royalties) with Juno Therapeutics and has stock options with A2 Biotherapeutics.

O114. Economic Aspects of Stem Cell Transplantation By Patients With Relapsed Diffuse B-Cell Lymphoma (DLBCL) In A German Tertiary Hospital

Bernhard Alexander Mörtl¹, Martin Dreyling¹, Eva Hoster¹, Christian Schmidt¹, Wolfgang Schoel¹, Michael von Bergwelt¹, Karin Berger¹

¹ LMU Klinikum, Munich, Germany

Background: Approximately 60% of DLBCL-patients (pts) are cured by first line of therapy: (1L) Immunochemotherapy e.g. R-CHOP. According to medical guidelines pts with relapse after 1L therapy may require a stem cell transplantation (SCT). SCT is a main cost driver in DLBCL-treatment in the pre-era of new innovative therapies. As only limited information exists on treatment patterns and costs of SCT for DLBCL-pts in Germany, we initiated an economic analysis from the third-party payers’ perspective in a German tertiary teaching hospital.

Methods: Retrospective single center observational study based on claims data and data from pts-medical records between 2007-2018. Inclusion criteria: age ≥18 years, diagnosis of relapsed DLBCL after first-line (ICD: C83.3). Exclusively inpatient stays were taken into consideration. Exclusion criteria: age <18 years, additional active malignancies. Follow-up: 2 years (in 6-months steps).

Results: We identified 84 pts with relapse after 1L. In total 34 pts received a SCT (2 pts received a second SCT in other therapy lines). Mean age at initial diagnosis: 53 years (range: 23-70), 60% were male. Number of SCTs in treatment lines (autologous-/allogenic) 2L: 1 (1/0); 2L-postremission therapy (2LPost): 24 (23/1); 3L: 5 (3/2); 3LPost: 3 (3/0); 4L: 0 (0/0); 5L: 3 (0/3). Mean costs per patient (€; range) 2L: 93.883 (-); 2LPost: 37.980 (14.974 – 113.971); 3L: 72.418 (29.908 – 148.528); 3LPost: 35.773 (30.091 – 46.468); 4L: -; 5L: 120.636 (99.358 – 153.771). In 2Lpost 24 of 31 pts (77%) received a SCT. 13 pts (55%) were in complete remission (incl. follow-up) and one death was documented during follow-up. Six patients received one further line (3L) of therapy with total additional mean costs (€) of 75.304 (3.178 – 148.528): 2 pts (8%) received one additional allogeneic-SCT and were in complete remission (incl. follow-up), 2 pts (8%) were in complete remission (incl. follow-up) and 2 pts (8%) with documented death during 2-year follow-up. After 3LPost: 2 pts (8%) were in complete remission (incl. follow-up) with total additional costs (€) of 35.688. One patient was in complete remission after 4L with total additional costs (€) of 32.820. For one patient was death during 5L follow-up documented with total additional costs (€) of 98.591.

Conclusions: The results show a huge cost range. Using the 2LPost cohort as an example, the results show that unsuccessful SCT resulted in further treatments and significant additional costs. From a clinical and economic perspective, it would be essential to identify as early as possible patients with increased risk for a relapse. The presented costs may underestimate the total hospital costs for patients with DLBCL treatment as exclusively inpatient claims data from the hematology/oncology department were taken into consideration. Costs from the radiation center and the outpatient center are not included. This is one possible explanation for the comparable low costs of 3L patients, because of a ~30% radiation-quote. Comprehensive cost and outcome analyses in DLBCL treatment are needed to put especially cost intensive innovative treatments into perspective.

Disclosure: Conflict of interest: Nothing to declare.

O115. Long Term Outcomes of Allogeneic Transplantation In Lymphoma: A Single Centre Experience

Alexander Glover¹, Nicolas Martinez-Calle¹, Christopher Fox¹, Mark Bishton¹, Jenny Byrne¹, Nigel Russell¹, Rohini Radia¹

¹ Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom

Background: Despite a growing number of targeted therapies, allogeneic stem-cell transplantation remains a curative option for patients with lymphoma, particularly due to the graft versus lymphoma (GVL) effect, which can be optimised with donor lymphocyte infusion (DLI).

Methods: Retrospective review of consecutive allografts for lymphoma at Nottingham University Hospitals (UK), between 2010-2019 (n = 70). Statistical analysis was performed using SPSS, with overall survival (OS) and progression free survival (PFS) analysed using Kaplan Meier and the Chi-square test for categorical data.

Results: Details of each lymphoma subtype are illustrated in the table. At a median follow up of 6.8 years LGNHL and HL had the highest survival rates. OS is 57% and PFS is 47% (figure 1); 10% of patients died of relapsed disease with 54% of patients in CR at last follow-up.

Conditioning included: 44% BEAM-alemtuzumab, 33% FMC, 10% RIC cord, 6% a RIC haplograft and in 6% Fludarabine/low-dose TBI RIC. Conditioning had no significant effect on PFS or OS. Overall, NRM was 31%; this was higher with FMC (48%) versus BEAM (16%). FMC was used after autograft in 82%; prior autograft was associated with higher NRM (46% vs 25%). NRM was due to infection in 50% of cases. Donor source was: Sibling 19%, MUD 50%, MMUD 16%, cord 10% and haplograft 6%. Median survival was 60 months for MMUD versus NR in other groups.

Grade I-II aGVHD was seen in 53% and III-IV in 11%. Grade III- IV aGVHD was associated with shorter OS (median 8 months versus NR, p = 0.03). Chronic GVHD was seen in 30% of patients; this was moderate or severe in 14%. Performance status was preserved - 98% of survivors had an ECOG of 0-1 at last follow up.

18 patients received DLI at a median of 197 days (range 139-1434) post-allograft, 13/18 for mixed chimerism, 2/18 for disease relapse and 3/18 for both. Of the 5 patients receiving DLI for relapse, 3 had LGNHL, 1 HGNHL and 1 MCL; 4/5 are in CR at last follow up. In 12/16 patients, DLI restored full chimerism. DLI led to grade I-II GVHD in 8/18 and III-IV in 1 patient.

	No of pts (%)	Median age (range)	Prior lines of therapy	Prior autograft (%)	No of relapses (%)	No receiving alemtuzumab (%)	Non relapse mortality- no (%)	Median OS- months	Median PFS- months
LGNHL (low grade non-Hodgkin lymphoma)	20 (28%)	56 (36-67)	3 (2-5)	9 (45%)	2 (10%)	16 (80%)	5 (25%)	NR	NR
HGNHL (high grade non-Hodgkin lymphoma)	14 (20%)	55 (31-69)	3 (2-4)	3 (21%)	4 (29%)	11 (79%)	5 (36%)	19	15
MCL (Mantle cell lymphoma)	9 (13%)	61 (51-70)	2 (2-3)	7 (78%)	2 (22%)	9 (100%)	4 (44%)	60	60
TCL (T cell lymphoma)	16 (23%)	48 (33-68)	1 (1-4)	3 (19%)	5 (31%)	10 (63%)	6 (38%)	14	9
HL (Hodgkin lymphoma)	11 (16%)	35 (23-52)	4 (3-6)	0 (0%)	1 (9%)	8 (73%)	3 (27%)	NR	NR
All Patients	70 (100%)	52 (23-70)	3 (1-6)	22 (31%)	14 (20%)	54 (77%)	23 (33%)	NR	44

Conclusions: We present a single-centre series of allogeneic SCT in lymphoma. Acknowledging histological heterogeneity, long-term OS of 57% in a high-risk cohort supports the effectiveness of allograft. T-cell depletion with alemtuzumab is associated with preserved PS and low rates of severe GVHD but high prevalence of mixed chimerism, as well as NRM due to infection. Further work to optimise conditioning, to provide equivalent GVHD prophylaxis but better immune reconstitution, is needed given that these patients are multiply treated and may require alternative donors.

Disclosure: All authors have nothing to declare in relation to this work.

Minimal Residual Disease, Tolerance, Chimerism and Immune Reconstitution

O116. Comparative Analysis of Immune Reconstitution In HSCT Patients With PT-CY OR ATG Reveals Distinct T Cell Immune Reconstitution Patterns

Saskia Leserer^1,1,2, Theresa Graf¹, Rashit Bogdanov¹, Aleksandra Pillibeit¹, Nils Leimkühler¹, Martina Franke¹, Ulrike Buttkereit¹, Katharina Fleischhauer¹, H. Christian Reinhardt¹, Dietrich W. Beelen¹, Amin T. Turki¹

¹ University Hospital Essen, University Duisburg-Essen, Essen, Germany

Background: In vivo T cell depletion with post-transplant cyclophosphamide (PT-Cy) or different formulations of anti-thymocyte globulin (ATG) can be used in graft-versus-host disease (GVHD) prophylaxis for allogeneic hematopoietic cell transplantation (HSCT) to eliminate proliferating alloreactive T cells. Previously, ATG was associated to delayed reconstitution of CD3+ and CD4+ T cells (Gooptu et al. BBMT 2018), while PT-Cy had a sparing effect on regulatory T cells (Tregs; Kanakry et al. Sci Transl Med 2013). Comparative immune reconstitution data for both regimens have only been reported from a very small cohort without differences in cell counts (Retiere et al. Oncotarget 2018).

Methods: We compared cellular immune reconstitution of 387 patients with HSCT from MUD- or haploidentical donors at the University Hospital Essen, Germany, treated between January 2017 and May 2020 with T cell depletion using either ATG (n = 305) or PT-Cy (n = 82). On a multi-color flow cytometer, 1271 peripheral blood samples obtained at months +1, +3, +6, +9 and +12 post-HSCT. Median absolute values were compared using Mann-Whitney-U-test and clinical outcomes were analyzed by Kaplan Meier analysis and Cox-regression.

Results: Early immune reconstitution differed significantly between PT-Cy and ATG cohorts with distinct cellular populations that each have been associated with an acute GVHD (aGVHD) protective effect. The PT-CY cohort had significantly higher median numbers of regulatory T cells (Tregs, p < 0.0001, Fig. 1A), while the ATG patients had significantly higher early populations of TCRγ/δ T cells (p < 0.0001, Fig 1B). We observed these differences irrespective of the donor constellation. Complete clinical outcome data was available for a major subset (72%, n = 280) with a median follow-up of > 16 months (ATG n = 231; PT-Cy: haploidentical n = 26, MUD n = 23). Here, these different cellular patterns associated with significantly higher aGVHD incidence (p = 0.016) in the ATG cohort as compared to PT-Cy. Hazard ratios for aGVHD after MUD- and haploidentical HSCT with PT-Cy were 0.6 (Confidence interval (CI) 0.37-0.97, p = 0.035) and 0.66 (CI 0.43-1.02, p = 0.062) compared to ATG patients. In addition, the median aGVHD onset was earlier in ATG patients than in haploidentical- or MUD PT-Cy patients (17 vs. 20 days). No significant differences in overall survival, non-relapse mortality or relapse were observed. Interestingly, the PT-Cy patients had higher absolute TCRα/β T cells counts compared to ATG patients (Fig. 1D), without an impact on the clinical incidence of aGVHD. Finally, we saw significantly higher median helper T cell- and conventional (Tcon) T cell counts of up to 6 months post-HSCT (p < 0.0001, Fig. 1C+E) for both MUD- and haploidentical- HSCT recipients with PT-Cy. Cytotoxic T cell, NK and B cell levels were comparable between the different cohorts.

Conclusions: Our study provides new evidence for distinct immune reconstitution patterns of different T cell depletion strategies- ATG or PTCY- in allogeneic HSCT recipients. The differences in aGVHD might be explained by distinct T cell subsets, particularly involving Tregs and TCRγ/δ T cells. Finally, the distinct GVHD prophylaxis but not the donor constellation impacted immune reconstitution patterns.

Disclosure: The authors of this abstract have potential conflicts of interest to disclose. ATT: Consultancy for MSD, JAZZ, CSL. Travel subsidies from Neovii Biotech outside the submitted work. DWB received travel subsidies from Medac, all outside the submitted work. The other authors declare no competing financial interests within the submitted work.

O117. Minimal Residual Disease and Chimerism Analysis Using Circulating Cell-Free DNA After Allogeneic Hematopoietic Cell Transplantation

Sandra Pennisi¹, Miguel Waterhouse¹, Dietmar Pfeifer¹, Justus Duyster¹, Hartmut Bertz¹, Jürgen Finke¹, Jesus Duque-Afonso¹

¹ University of Freiburg Medical Center, Freiburg, Germany

Background: Relapse of the underlying disease is a frequent complication after allogeneic hematopoietic cell transplantation (allo-HCT). We hypothesize that circulating cell-free DNA (cfDNA) analysis might result in early relapse detection compared with peripheral blood mononuclear cells (PBMCs).

Methods: A total of 285 plasma and PBMCs samples obtained from 62 patients with acute myeloid leukemia (AML) were analyzed. To assess chimerism, a panel of 12 insertion/deletion polymorphisms was used both in PBMCs and cfDNA obtained from plasma. Mutations in following genes were identified in our patient cohort and used for monitoring of minimal residual disease (MRD) in both, DNA derived from PBMCs and cfDNA: NPM1, JAK2, IDH1/2, NRAS, KRAS, DNMT3A, U2AF1, FLT3-ITD, SKF1B3.

Results: Relapse after allo-HCT was detected in 23 out of 62 patients (37.1%). The mean percentage of recipient cfDNA (chimerism) from patients in complete remission was 6.7% (range: 0-25%), while in patients with relapse, was 47.3% (range: 6-94%). A significant difference was found in the recipient cfDNA between patients with and without relapse (p < 0.001). The area under the curve (AUC) in a ROC curve was 0.968 (95% CI; 0.929-1.003 p < 0.001) and the optimal discriminating threshold between relapse and complete remission, was 18% of recipient derived cfDNA. The mean percentage of recipient DNA in PBMCs at relapse was 16.7% (range: 1-82% 95% CI; 4.5-28.9%) while in cfDNA at the same timepoint was 56.4% (range: 18-94% 95% CI; 44.2-68.5).

After performing targeted-sequencing using an AML specific gene panel in bone marrow samples at diagnosis, we were able to detect a total of 135 mutations in 57 patients. The mean mutation number per patient was 2.4 (range: 1-5). A total of 107 paired DNA from PBMCs and cfDNA samples from 24 patients were compared. The putative mutations were detected in both, in cfDNA and PBMCs, in 66 samples (61.7%), while the mutations were detected only in cfDNA in 40 samples (37.4%). Only in one patient, the mutation was detected in DNA from PBMCs but not in cfDNA (1.9%). Individual mutation comparison between cfDNA and PBMCs showed significant differences in selected mutations (FLT3-ITD, KRAS, NPM1, DNMT3A), but not in the rest of the analysed mutations (SKF1B3, U2AF1, IDH1/2, JAK2, NRAS).

The mean detection time after allo-HCT was 179 days (range: 30-272) and 282 days (range: 179-359) in either recipient cfDNA or in PBMCs, respectively. This time difference did not reach statistical significance. In 7 out of 23 patients (30%) with relapse after allo-HCT, MRD positivity was detected earlier in cfDNA (mean 397 days, range: 30-1549) than in DNA derived from PBMCs (mean 451 days, range: 100-1579). In the remaining patients relapsing after allo-HCT, MRD was detected at the same time in cfDNA and PBMCs.

Conclusions: In patients with relapse after allo-HCT, the proportion of mixed chimerism was higher in cfDNA than in PBMCs and, in several patients, MRD was detected earlier in cfDNA than in PBMCs. In summary, monitoring of chimerism and MRD in cfDNA is useful to earlier detect relapse after allo-HCT.

Disclosure: Nothing to declare.

O118. Sequential Quantification of T-Cell Receptor Excision Circles (TRECS) And K-Deleting Recombination Excision Circles (KRECS) In Allogeneic Hematopoietic Stem Cell Transplantation (ALLOHSCT) Recipients

Carlos De Miguel Jiménez¹, Ferran Briansó², Rosalía Alonso Trillo¹, Lucía Nuñez Martín-Buitrago¹, María E. Martínez-Muñoz¹, Guiomar Bautista¹, Carlos Manchado-Pergiguero², José A. García-Marco¹, Ali Sánchez-Peral², Rafael F. Duarte¹

¹ Hospital Universitario Puerta de Hierro Majadahonda, Majadahonda, Spain, ² Roche Diagnostics SL, Sant Cugat del Vallès, Spain

Background: The use of TRECs and KRECs to assess immune recovery and predict outcomes after alloHSCT has been traditionally limited by the lack of standardizable technical platforms to allow comparison and validation of results between centers.

Methods: We performed absolute TRECs/KRECs quantification using LightCycler 480 and TREC-KREC-ACTB primers (Roche Diagnostics, Barcelona, ES) on sequential samples collected prospectively (pre-transplant, 1, 3, 6 and 12-months post-transplant) in alloHSCT recipients for hematological malignancies between 1999 and 2017. Endpoint definition and statistical analyses follow EBMT statistical guidelines and were performed with R software (https://www.r-project.org/). Funding for this study comes from the Spanish CDTi (Centro de Desarrollo Tecnológico e Industrial; http://www.cdti.es/index.asp; Project IDI-20180259).

Results: 374 alloHSCT were included: 223 men (59.6%); median age 45 years (range 16-68); 69.8% acute leukemia/myelodysplastic syndromes, 22.2% chronic lymphoproliferative disorders, 8% myeloma; 52.9% matched sibling, 30.5% cord blood, 13.1% unrelated and 3.5% haploidentical donors. Pre-transplant TRECs/KRECs values and their post-transplant recovery, faster for KRECs (peak at 3 months, stable thereafter; P < 0.001) than for TRECS (reduced at 1 and 3 months, peak at 12 months; P < 0.001), associate with patient and transplant characteristics such as patient age, donor type and underlying malignancy (data not shown). TRECs levels at different timepoints have a strong association with non-relapse mortality (NRM) and overall survival (OS). In particular, patients with higher pre-transplant TRECs have a significantly reduced risk of NRM (P = 0.044) and a better OS (P = 0.0054) compared to those with lower levels, in univariate analyses (Figure panels A and B). In multivariate analyses, the impact of pre-transplant TRECs on NRM (P = 0.012) and OS (P < 0.001) remain independent from conventional clinical factors such as patient age, underlying disease and type of alloHSCT (Table). Pre-transplant KRECs levels do not associate with outcomes. As both, TRECs and KRECs levels at 6 and 12 months post-transplant had an association with OS, we performed landmark analyses to study the value of these measurements to assess long-term outcomes of early alloHSCT survivors. These analyses show that TRECs levels in survivors at 6 and 12 months and KRECs levels at 6 months after alloHSCT strongly associate with subsequent long-term survival of these patients (Table and Figure panels C and D).

Table Impact of TRECs/KRECs levels on transplant outcomes in multivariate analyses and in landmark analyses for long-term outcomes of early transplant survivors.

Conclusions: This study shows that sequential quantification of TRECs/KRECs in alloHSCT recipients for hematological malignancies using a standardized commercial platform provides novel measurable biomarker data that associate with patient and transplant characteristics and outcomes. Pre-transplant TRECs levels have an independent association with patient NRM and OS that may complement current decision-making based on clinical factors. In addition, TRECs and KRECs measurement at 6 and 12 months may provide a novel objective tool to predict long-term outcome of early survivors after alloHSCT. External, multicenter, prospective validation of these results through the standardized LightCycler 480/TREC-KREC-ACTB platform is warranted.

Disclosure: Funded by the Spanish CDTi (Centro de Desarrollo Tecnológico e Industrial; http://www.cdti.es/index.asp; Project IDI-20180259).

O119. Alemtuzumab Based-Conditioning For Non-Malignant Paediatric Diseases Depletes Donor T-Cells Allowing Exceptionally Low Rates of GVHD But Allows Persistence of Residual, Autologous, Virus-Specific T-Cells

Rubiya Nadaf¹, Prashant Hiwarkar¹, Denise Bonney¹, Helen Campbell¹, Kay Poulton¹, Helena Lee¹, Robert Wynn¹

¹ Royal Manchester Childrens Hospital, Manchester, United Kingdom

Background: We have previously reported that myeloid chimerism best reflects and therefore stem cell engraftment. We include alemtuzumab in all transplantation for non-malignant disease (NMD) since it more effectively prevents acute and chronic GVHD. Significant mixed chimerism (MC) is reported after such transplant and we sought to define the relationship between mixed T-cell chimerism and virus infection, having previously demonstrated no relationship between mixed T-cell chimerism and graft outcome.

Methods: We reviewed 130 patients aged 1-16 years, treated with allogeneic HCT using bone marrow or PBSC for non-malignant diseases at Royal Manchester Children’s Hospital. We analysed post-HSCT, lineage specific (CD3/CD15) leucocyte chimerism data to analyse the influence of virus infection on MC.

Results: Overall survival (OS) and event free survival (GVHD free) (EFS) at 6 years post-transplant were 91.1% and 81.5% respectively. No grade III-IV acute GvHD was seen and no extensive chronic GVHD requiring treatment was seen. In most of these patients with mixed chimerism, the myeloid (CD15) donor chimerism remained nearly full donor, and mixed chimerism is due to autologous T-cells persistence and expansion. At 3 and 6 months post-HCT mean T cell (CD3) and myeloid (CD15) donor chimerism values were 20.28% and 99.60% and 40.55% qne 97.34% respectively (P < 0.0001). There is no relationship between mixed T-cell chimerism and graft loss. CMV viremia was seen in 31.8% of patients, almost exclusively in CMV Seropositive (R+) patients (p < 0.0001) compared with seronegative patients (R-). CMV positive serology pre-transplant predicted for mixed T cell chimerism, at 3 and 6 months. At 3 months 64.29% (18/28) in the mixed T-cell chimerism group were CMV positive compared to only 41.77% (33/79) in the complete chimerism group. Similarly, at 6 months the group frequency was 61.7% (21/34) versus 41% (30/73) respectively (P < 0.05). In children with CMV viraemia, more children had MC at 3 and 6 months, 38.8% and 47.22%, than in those without viraemia, 19.72% and 23.9% (P = 0.03 and 0.01, respectively). CD8 T cell reconstitution is influenced by viral reactivation post-transplant, including CMV. We compared the level of CD8 T cell counts in patients who developed CMV viremia versus those who did not. CD 8 T cell count was significantly greater in CMV positive cohort. (mean 77.7 vs 548; P = 0.0001 at 3 months and mean 244 vs 780; P < 0.0001 at 6 months). (Figure 1) CD8 T cell counts were significantly higher in the mixed T-cell chimerism cohort versus complete chimerism group. (Mean 153 vs 402; P = 0.005).

Conclusions: We have previously demonstrated no relationship between T-cell chimerism and graft outcomes in non-malignant diseases. Here we confirm that this mixed T-cell chimerism is associated with positive CMV serology, with CMV viraemia and that the CD8 count of chimeric children is higher than that of non-chimeric children, confirming that mixed T-cell chimerism is due to autologous, virus-specific, predominantly CD8, T-cell expansion, and is protective and not deleterious to the recipient.

Clinical Trial Registry: N/A

Disclosure: Nothing to declare.

O120. TCR Repertoire Dynamics After Hematopoietic Stem Cell Transplantation In Myeloid Malignancies

Simona Pagliuca^1,2, Carmelo Gurnari^1,3, Sanghee Hong¹, Ran Zhao¹, Sunisa Kongkiatkamon¹, Laila Terkawi¹, Misam Zawit¹, Lisa Rybicki¹, Yihong Guan¹, Thomas Laframoboise⁴, Hassan Awada¹, Ashwin Kishtagari¹, Babal Kant Jha¹, Valeria Visconte¹, Hetty E. Carraway¹, Navneet Majhail¹, Betty K. Hamilton¹, Jaroslaw Maciejewski¹

¹ Cleveland Clinic Foundation, Cleveland, United States, ² University of Paris, Paris, France, ³ University of Rome Tor Vergata, Roma, Italy, ⁴ Case Western University, Cleveland, United States

Background: T-cell receptor (TCR) repertoire variability is a major hallmark of immune-competence. For patients undergoing HSCT, a delayed and/or impaired T-cell immune reconstitution represents a potential cause of treatment failure, associated with the occurrence of disease relapse, infections and alloreactive complications including graft-versus-host disease (GvHD) and graft failure. Although notable efforts have been made to understand the dynamics of the TCR repertoire reconstruction after transplant, critical aspects concerning the forces shaping its complexity remain still controversial.

Methods: With a deep TCRVbCDR3 sequencing we profiled the TCR repertoire of 135 serial samples at different time-points (pre-HSCT, day +30, +100, +180) derived from a cohort of 35 AML/MDS first HSCT recipients and their donors. We then built a dataset of TCR rearrangements from 130 healthy controls (HC) age-matched with patients’ donors to establish the normal range of TCR expansion, diversity and overlap.

Results: First, we established the normal patterns of clonal expansion and the quantitative measures of diversity and public overlap in HC and donors. Overall, richness (number of unique clonotypes/normalized depth) accounted for 90% of the repertoire and only 0.4% of clonotypes were pathologically expanded with a mean frequency of 0.003%. This configuration changed in patients after chemotherapy (pre-HSCT group) with a general decrease in diversity compared to donors (reduced richness and increased size of clonal expansion). This trend was conserved after transplant, when global diversity progressively diminished, irrespectively of the platform used. Specifically, across the post-transplant period, we assisted to a significant decrease in richness, with a parallel increase in proportion and size of expanded clonotypes. The richness of the repertoire according to the various timepoints was 64%, 53%, and 44%, respectively for recipients at day +30, +100 and +180, with an increase in the fraction of pathologically expanded specificities (2.1%, 2.5%, 3.7%).We then analyzed the similarities between repertoires of recipients and donors computing the proportion of the overlapping CDR3 sequences. Interestingly, only a small fraction of the recipient repertoire was shared with the respective donor across the whole post-transplant period: overlap was higher in the first month but diminished later (2.5% to 1.6%), while the rest was represented by “de novo” clonotypes, not arisen from donor’s repertoire. A consistent part of expanded shared clonotypes in recipients was also expanded in donors.

Univariable cox regression analysis showed that a higher post-transplant richness was associated with better OS [HR 0.99 (95%CI 0.998-0.99) p = 0.043] and lower risk of relapse [HR 0.99 (95%CI 0.99-1), p = 0.005]. Conversely, the presence of a higher proportion of pathological expanded clonotypes negatively impacted survival [HR: 1.04 (95%CI 1.00-1.09), p = 0.047] and the persistence of a “donor-imprinted” TCR phenotype (higher overlap) was associated with an increased risk of chronic GvHD [HR: 1.06 (95%CI 1.00-1.12), p = 0.045].

Conclusions: In sum, we performed the first systematic molecular characterization of TCR repertoire configuration after HSCT. Our findings shed light on the global architecture and the patterns of early and late reconstitution of TCR repertoires influencing transplant-related outcomes in patients with myeloid malignancies given HSCT.

Disclosure: Nothing to declare.

O121. Day +60 WT1 Assessment on Bone Marrow Might Predict Relapse And Mortality After Allogeneic Stem Cell Transplantation

Patrizia Chiusolo^1,2, Elisabetta Metafuni¹, Gessica Minnella², Sabrina Giammarco¹, Silvia Bellesi¹, Viviana Amato², Luca Laurenti², Federica Sorà², Simona Sica^1,1, Andrea Bacigalupo¹

¹ Fondazione Policlinico Universitario A. Gemelli IRCCS, Roma, Italy, ² Università Cattolica del Sacro Cuore, Roma, Italy

Background: Wilms Tumor 1 (WT1) has been widely studied recently as molecular marker for monitoring minimal residual disease in acute myeloid leukemia (AML), as well as prognostic factor for AML relapse after allogeneic stem cell transplantation (HSCT).

Methods: We enrolled 50 patients with AML submitted to HSCT in our centre between June 2018 and July 2020, who were in remission at day 60 after HSCT. A bone marrow sample collected on day +60 was tested for WT1 copies number. WT1 was assessed in selected CD34+ cells (CD34+) in 45 patients and in total mononucleated cells (MNC) in 40 patients. Two control groups of healthy bone marrow donors were included, one for WT1 in CD34+ (n = 42) and the other for WT1 in MNC (n = 18). WT1 mRNA was assessed using the validated ProfileQuant WT1 kit, European Leukemia Net, Ipsogen. WT1 mRNA expression was normalized with respect to the number of ABL transcripts and was reported as WT1 copies/104 copies of ABL.

Results: Median age was 56 years (25-69). European Leukemia Net risk group was as follows: favourable (n = 10), intermediate (n = 27), adverse (n = 13). Median copies number of WT1 in CD34+ was of 1133.6 in patients who relapsed after HSCT as compared to 387.3 in patients without relapse and 252.3 in controls (p = 0.0002). Similarly, median copies number of WT1 in MNC was 152.9 in patients who relapsed, as compared to 42.4 in patients without relapse and 43.2 in controls (p = 0.02). ROC curve allowed us to identify the best WT1 cut-off for relapse: 800 copies in CD34+ (AUC 0.842, p = 0.0006, 85.7% sensitivity and 81.6% specificity) and 100 copies in MNC (AUC 0.819, p = 0.007, 83.3% sensitivity and 88.2% specificity). Accordingly to WT1 copies number in CD34+, one year disease frees survival (DFS) was of 96.7% for patients with WT1 < 800 copies (n = 32) and 60.6% for those with WT1 ≥ 800 copies (n = 13) (p = 0.0002). One year overall survival (OS) was of 89.4% in patients with WT1 in patients with WT1 < 800 copies and 56.2% in those with WT1 ≥ 800 copies (p = 0.02). Considering WT1 detection on MNC, one year DFS was of 96.4% for patients with WT1 < 100 copies (n = 31) and 55.6% for those with WT1 ≥ 100 copies (n = 9) (p = 0.0001). Similarly, 1-year OS was of 93.1% for patients with WT1 < 100 copies and 51.8% for those with WT1 ≥ 100 copies (p = 0.01).

Conclusions: WT1 copies number assessed on bone marrow at day 60 after HSCT showed a good sensitivity and specificity in predicting relapse occurrence and survival. Here we confirmed the already used WT1 cut-off of 100 copies in total bone marrow mononucleated cells, and identified a new cut-off for WT1 in CD34+ stem cells as predictor for relapse and mortality after HSCT. Certainly, these data need to be onfirmed and validated in a larger study population.

Disclosure: Nothing to declare.

Multiple Myeloma

O122. Microenvironment Immune Reconstitution Patterns Correlate With Outcomes After Autologous Transplant in Multiple Myeloma

Harsh Parmar¹, Morie Gertz², Emilie Anderson², Shaji Kumar², Taxiarchis Kourelis²

¹ Hackensack University, Hackensack, United States, ² Mayo Clinic, Rochester, United States

Background: The immediate post-autologous stem cell transplant (ASCT) period in multiple myeloma (MM) represents a unique opportunity for long -term disease control since many patients have eradicated most of their disease but also a challenge since it is characterized by the increase of immune subsets detrimental to tumor immunosurveillance. The impact of tumor immune microenvironment (iTME) reconstitution early post-ASCT on patient outcomes is not known.

Methods: In this study, we included 58 patients undergoing upfront ASCT and evaluated their cellular and humoral iTME with CyTOF and Luminex, respectively, at day+60-100 post-ASCT. CyTOF data were analyzed using the Astrolabe cytometry platform. A two-sided false discovery rate (FDR) adjusted p value of <0.05 was considered significant when multiple comparisons were performed, otherwise a p value of <0.05 was considered significant.

Results: We identified 26 lymphoid subsets early after ASCT. Their phenotypic characteristics are shown in the figure. We clustered patients based on these immune subsets. Two major immune reconstitution patterns were noted. Group 1 was characterized by higher levels of the T cell-2 subset, a double negative, terminally differentiated subset and the T cell-4 subset, resembling NKT cells but also with markers of immune senescence/terminal differentiation and low levels of TIGIT expression. Other T cell subsets differentially increased in group 1 were T cell-14, 16, 19 and -20 subsets. Of these, the T cell-16 and -19 subsets, were CCR5 expressing subsets with features of terminal differentiation and, in the case of T cell-16, exhaustion (TIGIT+). Maximal T cell recruitment and T cell-mediated antitumor responses rely on CCR5 expression and CCR5 ligands are secreted by MM cells, suggesting that this cells are actively recruited in situ but might not be able to control tumor growth because they are exhausted or senescent. Subset -20 had a phenotype characteristic of naïve T cells. No cytokine differences were noted across the three groups. These data suggest that early immune reconstitution in a subset of patients post-ASCT is enriched in T cell subsets at the opposite ends of the spectrum of T cell differentiation, i.e. cells already terminally differentiated or naïve T cells. On multivariable analysis, this group had significantly worse hematologic responses post-ASCT, inferior overall survival and shorter time to hematologic progression independent of hematologic response, maintenance therapy and age but not of high risk FISH status. In addition, no differences in the cellular/humoral iTME were noted according to high risk FISH status, early or late relapse. Finally, males had higher levels of negative for CD16, a key receptor mediating antibody-dependent cell cytotoxicity, which a major mechanism of anti-tumor efficacy by therapeutic antibodies such as elotuzumab.

Conclusions: In summary, we define major immune reconstitution patterns early after ASCT that can help guide the rational use of plasma cell directed monoclonal antibodies and checkpoint inhibitors early after ASCT. The presence of T cells at the 2 opposite ends of the T cell differentiation spectrum, in the group with the worst outcomes, suggests that the iTME of these patients is enriched in subsets unable to mediate effective clearance of malignant cells.

Disclosure: The authors have no relevant conflicts of interest. This work was supported by an Eagles 5th District Cancer Telethon – Cancer Research Fund and the Mayo Clinic Myeloma SPORE P50 CA186781-03 NIH grant.

O123. Autologous Hematopoietic Cell Transplantation For Relapsed Multiple Myeloma Performed With Stem Cells Re-Mobilised Following A Prior AUTO-HCT – A Retrospective Study on Behalf of EBMT CMWP

Joanna Drozd-Sokołowska¹, Luuk Gras², Nienke Zinger³, John A. Snowden⁴, Mutlu Arat⁵, Grzegorz Basak¹, Anastasia Pouli⁶, Charles Crawley⁷, Keith M. O. Wilson⁸, Herve Tilly⁹, Jennifer Byrne¹⁰, Claude Eric Bulabois¹¹, Jakob Passweg¹², Zubeyde Nur Ozkurt¹³, Wilfried Schroyens¹⁴, Bruno Lioure¹⁵, Mercedes Colorado Araujo¹⁶, Xavier Poiré¹⁷, Gwendolyn Van Gorkom¹⁸, Gunhan Gurman¹⁹, Liesbeth de Wreede²⁰, Patrick Hayden²¹, Meral Beksac¹⁹, Stefan Schönland²², Ibrahim Yakoub-Agha²³

¹ Medical University of Warsaw, Warsaw, Poland, ² EBMT Statistical Unit Data Office, Leiden, Netherlands, ³ EBMT Data Office, Leiden, Netherlands, ⁴ Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, United Kingdom, ⁵ Florence Nightingale Sisli Hospital, Istanbul, Turkey, ⁶ Haematology Department, Athens, Greece, ⁷ Addenbrookes Hospital, Cambridge, United Kingdom, ⁸ Department of Haematology, Cardiff, United Kingdom, ⁹ Centre Henri Becquerel, Rouen, France, ¹⁰ Nottingham University, Nottingham, United Kingdom, ¹¹ CHU Grenoble Alpes – Université Grenoble Alpes, Grenoble, France, ¹² University Hospital, Basel, Switzerland, ¹³ Gazi University Faculty of Medicine, Ankara, Turkey, ¹⁴ Antwerp University Hospital (UZA), Antwerp_Edegem, Belgium, ¹⁵ Techniciens d`Etude Clinique suivi de Patients Greffes, Strasbourg, France, ¹⁶ Hospital U. Marqués de Valdecilla, Santander, Spain, ¹⁷ Cliniques Universitaires St. Luc, Brussels, Belgium, ¹⁸ University Hospital Maastricht, Maastricht, Netherlands, ¹⁹ Ankara University Faculty of Medicine, Ankara, Turkey, ²⁰ Leiden University Medical Center, Leiden, Netherlands, ²¹ Trinity College Dublin, St. James’s Hospital, Dublin, Ireland, ²² Medizinische Klinik u. Poliklinik V, University of Heidelberg, Heidelberg, Germany, ²³ CHU de Lille, Univ Lille, INSERM U1286, Lille, France

Background: High-dose chemotherapy and autologous hematopoietic cell transplantation (auto-HCT) constitutes the standard-of-care for transplant-eligible patients with myeloma. In selected cases, it may also be performed in patients relapsing after a previous auto-HCT. For those without stored stem cells, remobilization is necessary. We here present the results of salvage auto-HCT performed with cells procured following such remobilization.

Methods: Patients who had a salvage auto-HCT between 2000 and 2018 were selected from the EBMT database. The Kaplan-Meier estimator and log-rank test were used for overall survival (OS) and progression-free survival (PFS), and the crude cumulative incidence estimator and Gray’s test were used for competing events (relapse and non-relapse mortality (NRM); therapy-related myelodysplastic syndrome/ acute myeloid leukemia (t-MDS/t-AML), other second primary malignancy, and death before any second malignancy).

Results: 305 patients (68% male) had a salvage auto-HCT at a median age of 59 years (32-78). The median interval between the previous auto-HCT and salvage remobilization was 43.9 months (range: 7.1-152). 90% of patients underwent one remobilization attempt. Plerixafor was used in 17% of patients. The median cell dose collected was 3.39 x 106 CD34+ cells/kg (range: 0.24-16.0) and the median cell dose infused at auto-HCT was 2.92 x 106 CD34+ cells/ kg (range: 1.07-24.5).

Median follow-up after salvage auto-HCT was 31 months (95% CI, 25.9-36.4). The 2- and 4-year NRM was 5% (95% CI, 2-7%) and 9% (5-12%), respectively. Relapse incidence was 56% (95% CI, 50-62%) and 76% (95% CI, 70-81%), respectively. OS rates after 2 and 4 years were: 76% (95% CI, 71-81%) and 52% (95% CI, 45-58%) and PFS 39% (95% CI, 33-45%) and 15% (95% CI, 11-20%), respectively. An interval of more than 48 months between the prior and the salvage auto-HCT was associated with improved OS, PFS and NRM in univariate analysis.

A total of 7% of the patients developed secondary malignancies during follow-up. The cumulative incidence of t-MDS/t-AML was 1% (95% CI, 0-3%) and 3% (95% CI, 1-5%) at 2 and 4 years; the cumulative incidence of other second primary malignancies was 1% (95% CI, 0-2%) and 3% (95% CI, 1-5%), respectively; the cumulative incidence of death before any secondary malignancy was 23% (95% CI, 18-28%) and 44% (95% CI, 38-51%), respectively.

Conclusions: Salvage auto-HCT performed with stem cells remobilized after a prior auto-HCT is associated with an acceptable NRM and the expected cumulative incidence of secondary malignancies. The leading cause of failure is progression of myeloma.

Clinical Trial Registry: not applicable.

Disclosure: No conflict of interest.

O124. Autologous-Allogeneic Vs Autologous Tandem Stem Cell Transplantation And Maintenance With Thalidomide For Patients With Multiple Myeloma (MM) and Age <60 Years: A Prospective Phase II-study

Nicolaus Kröger¹, Ute Hegenbart², Matthias Stelljes³, Martin Kaufmann⁴, Lutz Müller⁵, Nico Gagelmann¹, Könecke Christian⁶, Christoph Schmid⁷, Guido Kobbe⁸, Eva-Maria Wagner⁹, Martin Bornhäuser¹⁰, Michael Kiehl¹¹, Gerald Wulf¹², Wolfgang Bethge¹³, Andreas Burchert¹⁴, Dominik Wolf¹⁵, Thomas Heinicke¹⁶, Marion Heinzelmann¹, Christine Wolschke¹, Andreas Völp¹⁷, Stefan Schönland²

¹ University Medical Center Hamburg, Hamburg, Germany, ² University Hospital Heidelberg, Heidelberg, Germany, ³ University Hospital Münster, Münster, Germany, ⁴ Robert Bosch Hospital, Stuttgart, Germany, ⁵ University Hospital Halle, Halle, Germany, ⁶ Medical School Hannover, Hannover, Germany, ⁷ University Hospital Augsburg, Augsburg, Germany, ⁸ University Hospital Düsseldorf, Düsseldorf, Germany, ⁹ University Hospital Mainz, Mainz, Germany, ¹⁰ University Hospital Dresden, Dresden, Germany, ¹¹ Hospital Frankfurt/Oder, Frankfurt/Oder, Germany, ¹² University Hospital Göttingen, Göttingen, Germany, ¹³ University Hospital Tübingen, Tübingen, Germany, ¹⁴ University Hospital Marburg, Marburg, Germany, ¹⁵ University Hospital Innsbruck, Innsbruck, Austria, ¹⁶ University Hospital Magdeburg, Magdeburg, Germany, ¹⁷ Psy Consult, Hamburg, Germany

Background: The role of autologous-allogeneic (auto-allo) tandem stem cell transplantation (TSCT) followed by maintenance therapy has not been compared to tandem autologous (auto-auto) TSCT followed by maintenance therapy in patients with newly diagnosed multiple myeloma (MM).

Methods: Between 2008 and 2014 a total of 210 MM patients up to 60 years of age were included from 23 German Centers within an open-label, parallel-group, multi-center clinical trial to investigate whether auto-allo TSCT, compared to auto-auto TSCT followed by a 2 year maintenance therapy with thalidomide (100mg/daily) in both arms, has a beneficial effect on relapse / progression-free survival (PFS) and on other clinically relevant outcomes. In auto-allo TSCT also prophylactic Donor lymphocyte infusion were allowed.

Patients received auto-allo TSCT with peripheral blood stem cells when a matched related or unrelated donor was available; otherwise, or if they declined allogeneic stem cell transplantation, they received auto-auto TSCT.

The primary endpoint was PFS at four years after TSCT and sample size was calculated on the basis of expected 48-month event rates (i. e. relapse or progression) of 50% for auto-allo TSCT and of 70% for auto-auto TSCT. A total sample size of 185 patients was estimated (n = 111 in auto-allo and n = 74 in auto-auto) to provide at least 80% power to reject the null hypothesis in a log-rank test model. A total of 178 patients underwent second SCT (allo n = 132 and auto n = 46). The median age was 51 years (range 26-61).

Results: The cumulative incidence of non-relapse mortality (NRM) at 4 years was 13% (CI: 8-20%) and 2% (0.3-2) for auto-allo and auto-auto TSCT, respectively (p = 0.04).

During the initial 48 months after the second SCT 53 (40.2%) patients in the auto-allo group and 28 (60.9%) in the auto-auto group showed progression or relapse of MM. Cumulative incidence of relapse/progression was 40% (CI: 33-50%) and 63% (CI: 50-79%) for auto-allo and auto-auto TSCT, respectively (p = 0.01).

The PFS at 48 months after the second SCT was 47% (CI: 38-55%) for auto-allo TSCT and 35% (CI: 21-49%) for auto-auto TSCT, with median survival times of 40.1 and of 29.8 months (Log-rank test: p = 0.26).

The 4-year overall survival was 66% (CI: 57-73%) for auto-allo TSCT and 66% (50-78%) for auto-auto TSCT (p = 0.9).

Any grade of acute or chronic graft versus host disease (GvHD) after allogeneic SCT or donor lymphocyte infusion (DLI) was observed in 55% or 61%.

Conclusions: We conclude, that in this prospective phase II study, auto-allo TSCT, as compared to auto-auto TSCT, reduced the rate of MM recurrence or progression by about 1/3, from 61% to 40%, during a four-year follow-up. During the same period, auto-allo TSCT was also associated with a reduction of disease-related mortality by about 1/3, from 30% to 21%. However, due to a higher NRM rate in patients receiving auto-allo TSCT the resulting 10% advantage in relapse/progression-free survival after auto-allo TSCT did not reach statistical significance.

Subgroup analyses and long term follow up data will be presented at the meeting.

The study was registered under: ClinicalTrials.gov NCT00777998

Clinical Trial Registry: ClinicalTrials.gov NCT00777998

Disclosure: NK received research funding from Neovii and Celgene.

O125. Allogeneic Hematopoietic Cell Transplantation Followed by Bortezomib in Multiple Myeloma: Final Results of A Prospective Trial Highlighting A Strong Prognostic Role of Measurable Residual Disease

Jean Roy¹, Imran Ahmad¹, Rafik Terra¹, Michael Sebag², Émilie Lemieux-Blanchard¹, Silvy Lachance¹, Nadia Bambace¹, Léa Bernard¹, Sandra Cohen¹, Jean-Sébastien Delisle¹, Thomas Kiss¹, Séverine Landais¹, Denis Claude Roy¹, Guy Sauvageau¹, Richard LeBlanc¹

¹ Université de Montréal, Montreal, Canada, ² McGill University, Montreal, Canada

Background: Prior studies have suggested that allogeneic (allo) hematopoietic cell transplantation (HCT) can overcome high-risk (HR) cytogenetics associated with poor prognosis in multiple myeloma (MM). Despite being potentially curative, the use of allo HCT remains controversial due to high rates of nonrelapse mortality (NRM), chronic graft-versus-host disease (cGVHD) and relapse. We hypothesized that bortezomib (BTZ) maintenance after allo HCT might decrease both relapse and cGVHD in HR MM and young patients (pts) who have the greatest loss in years of life. We also sought to determine prospectively the impact of bone marrow measurable residual disease (MRD) on disease progression.

Methods: This prospective trial enrolled newly diagnosed MM pts ≤ 65 years with an 8/8 sibling or unrelated donor and either HR cytogenetics, ISS 3, primary plasma cell leukemia or age ≤ 50 years regardless of biological risk. After a BTZ-based induction and autologous HCT, an outpatient nonmyeloablative allo HCT was performed, followed by BTZ maintenance initiated 120 days after HCT at 1.3 mg/m² every 2 weeks for 1 year. Responses were assessed by IMWG criteria. MRD was evaluated using Next-Generation Flow cytometry (NGF) (sensitivity < 10⁻⁵) prior to allo HCT, prior to BTZ and every 3 months thereafter for 2 years. The limit of quantification and detection (LOD) of the NGF MRD method was calculated based on identification of ≥50 and ≥30 (3 x 10⁻⁶) abnormal plasma cells (aPCs; MRD negative if < 30 aPCs).

Results: Between 11/2014 and 09/2018, 39 pts (median age 54 years) underwent allo HCT (59% with an unrelated donor). ISS 3 was found in 44% and HR cytogenetics in 65%. Thirty-six patients received BTZ (median of 26 doses; range 1-26) which was well tolerated with few BTZ-related adverse events and no death. With a median follow-up of 4.0 years (range 2.0-5.5), incidences of NRM, progression, OS and PFS (Fig. A) at 5 years are 12% (95%CI: 2-29), 48% (28-65), 80% (58-91) and 41% (21-60), respectively. Incidence of moderate/severe cGVHD was 46%, significantly better than historical controls (69%, p = 0.008). Stringent complete remission was observed in 33%, 44% and 67% and MRD negativity in 41%, 38% and 64% of patients before allo HCT, before BTZ and 3 months after initiation of BTZ, respectively. In multivariable analyses, while the small sample size did not yield significant association with age, donor type, ISS, cytogenetics, IMWG response status or cGVHD, progression risk was strongly associated with MRD status before allo HCT, before starting BTZ (Fig. B) and 3 months thereafter (adjusted HR 3.7, 11.3, 9.7, p = 0.037, 0.018, 0.001, respectively). No patient with 17p deletion (n = 4) progressed after allo HCT.

Conclusions: For HR disease and young MM patients, first line tandem autologous-allo HCT followed by BTZ maintenance is safe and yields a potential cure in approximately 40% of pts. The 17p- subgroup particularly benefited from this approach with no progression. For the first time in MM, we report that < 30 aPCs (under LOD) using NGF is predictive of a significantly better outcome after allo HCT. This threshold could be used for future studies.

Clinical Trial Registry: ClinicalTrial.gov: NCT02308280.

Disclosure: Nothing to declare.

O126. Coast (OP-501): First in Man Phase I Multicenter Study Evaluating OPD5 As A Myeloablative Conditioning Regimen in Relapsed Refractory Multiple Myeloma (RRMM)

Sergio Giralt¹, Kristine Bäck², Stefan Norin², Francesca Gay³

¹ Memorial Sloan Kettering Cancer Center, New York, United States, ² Oncopeptides AB, Stockholm, Sweden, ³ University of Torino, Torino, Italy

Background: Despite recent and continuing improvements in treatment, outcomes (such as lack of response, limited response and/or the occurrence of high-grade toxicities) in relapsed refractory multiple myeloma (RRMM) are still substandard, highlighting an urgent medical need for new therapeutic options. OPD5, an analog of melphalan flufenamide (melflufen), is a lipophilic peptide-drug conjugate (PDC). PDCs, such as OPD5, leverage aminopeptidases and release alkylating agents rapidly into tumor cells. Due to its lipophilicity, OPD5 crosses cell membranes rapidly and passively via transporter-independent inflow into cells.

Conditioning with 200 mg/m² melphalan is currently considered the standard regimen before autologous stem cell transplant (ASCT) in patients with myeloma. However, current high-dose therapy (HDT) is associated with limitations; some patients do not respond, a substantial proportion of patients have a short response, and HDT is limited by severe non-hematological toxicities, including mucositis and infections. Development and evaluation of an improved myeloablative treatment could therefore provide an alternative option to patients with MM. The COAST (OP-501) study is evaluating the safety and tolerability of OPD5 as a myeloablative conditioning regimen followed by a salvage ASCT in patients with RRMM.

Methods: An open-label, Phase 1, dose escalation study (COAST; OP-501) is to be conducted in the US and Europe. Eligible patients have previously received at least two prior lines, including an ASCT but have progressed within 24 months after the transplantation, and are refractory to a proteasome inhibitor (PI), an immunomodulatory drug (IMiD) and a CD38 antibody. The primary objective is to determine the Recommended Phase 2 Dose (RP2D) of OPD5 by evaluating the safety and tolerability by frequency and grade of adverse events (AEs) of escalating doses, as well as dose-limiting toxicities in each cohort. Secondary objectives include efficacy (overall response rate, duration of response, time to progression, time to next treatment and progression-free survival) and pharmacokinetics. Patients with RRMM will receive a single dose of OPD5 (initial dose: 30 mg/m²) after 2–9 cycles of reinduction therapy (investigator’s choice). A standard 3+3 dose escalation schedule will be applied in planned dose increments of 50% to 31%.

Results: N/A (note that this is a trial in progress [TiP]).

Conclusions: As OPD5 has similar preclinical pharmacokinetic and lipophilic properties as melflufen, it is of significant medical interest to investigate the benefit-risk profile of OPD5 as a conditioning regimen in patients where salvage ASCT is not currently considered standard of care. Findings from this first-in-man study will potentially demonstrate the value of OPD5 as a myeloablative therapy followed by stem cell support, justifying further clinical development.

Disclosure: SG has received consultant fees and honoraria related to speakers’ bureau activities from Amgen, Celgene, CSL Behring, Genzyme, Janssen, Kite Pharmaceuticals, Millennium Pharmaceuticals, Novartis, Oncopeptides, Pfizer, Sanofi, Quintiles; he also has served on advisory boards for Jazz Pharmaceuticals; FG has received consultant fees and honoraria related to speakers’ bureau activities, as well as advisory board activities from Abbvie, Adaptive, Amgen, Bristol-Myers Squibb, Celgene, Janssen, Oncopeptides, Roche, Takeda; KB and SN are employees at Oncopeptides.

O127. Characteristics And Treatment Patterns of Triple-Class-Exposed Patients With Relapsed/Refractory Multiple Myeloma Who Participated in Clinical Trials of Daratumumab

Katja C Weisel¹, Thomas Martin², Kwee Yong³, Keqin Qi⁴, Anil Londhe⁵, Rachel Kobos⁶, Ming Qi⁷, Jordan M Schecter⁶, Martin Vogel⁸, Satish Valluri⁸, Amrita Krishnan⁹, Sundar Jagannath¹⁰

¹ University Medical Center of Hamburg-Eppendorf, Hamburg, Germany, ² UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, United States, ³ University College Hospital, London, United Kingdom, ⁴ Janssen R&D, LLC, Titusville, United States, ⁵ Janssen R&D, LLC, Yardley, United States, ⁶ Janssen R&D, Raritan, United States, ⁷ Janssen R&D, LLC, Spring House, United States, ⁸ Janssen Global Services, LLC, Raritan, United States, ⁹ Judy and Bernard Briskin Center for Multiple Myeloma Research, City of Hope, Duarte, United States, ¹⁰ Mount Sinai Hospital, New York, United States

Background: Proteasome inhibitors (PIs), immunomodulatory drugs (IMiDs), and anti-CD38 monoclonal antibodies (mAbs) have improved outcomes in multiple myeloma (MM). However, many patients relapse or become refractory to these therapies, and novel agents are under investigation in patients who have received a PI, IMiD, and anti-CD38 mAb (triple-class-exposed). Since limited data are available on treatment patterns in triple-class-exposed patients with relapsed/refractory (R/R) MM, we evaluated subsequent therapies in this population using a global dataset.

Methods: Triple-class-exposed patients with R/R MM were identified from 3 trials of the anti-CD38 mAb daratumumab: POLLUX (phase 3 study of daratumumab-lenalidomide-dexamethasone vs lenalidomide-dexamethasone; NCT02076009), CASTOR (phase 3 study of daratumumab-bortezomib-dexamethasone vs bortezomib-dexamethasone; NCT02136134), and EQUULEUS (phase 1b study of daratumumab plus standard MM regimens; NCT01998971). After study treatment discontinuation, patients were followed for progression-free survival (PFS), PFS2 (PFS with therapy subsequent to trial treatment), and overall survival; subsequent treatments were per physician choice. Patients included in this analysis had Eastern Cooperative Oncology Group performance status <2; received ≥3 prior lines of therapy (LOT), including a PI, IMiD, and anti-CD38 mAb, and ≥1 subsequent therapy after becoming triple-class-exposed; and progressed ≤12 months of last LOT. Descriptive statistics for treatment patterns after triple-class-exposure were assessed overall and by LOT.

Results: In all, 329 triple-class-exposed patients with R/R MM from the European Union (69.0%), North America (15.2%), Asia/Pacific (13.4%), and other regions (2.4%) met inclusion criteria; median follow-up was 21.0 months (95% CI: 18.2-22.9). Median age was 66.5 years, 54.1% were male, and 16.7% had high-risk cytogenetics. Median number of treatments after becoming triple-class-exposed was 3 (range: 1-9). Across all subsequent LOTs after triple-class-exposure, most common therapies were pomalidomide, bortezomib, and carfilzomib, with 66.9% of patients receiving regimens containing ≥3 drugs (Table); in addition, 44.4% of patients received cyclophosphamide, 18.2% received melphalan, and 22.8% received other novel agents. The retreatment rate with the same drug after triple-class-exposure was 53.2%, including 30.0% for bortezomib, 29.5% for pomalidomide, 20.6% for thalidomide, 20.2% for lenalidomide, 11.8% for ixazomib, 11.7% for carfilzomib, and 11.6% for daratumumab.

Table PIs, IMiDs, and Anti-CD38 mAbs Received After Triple-Class-Exposure Status and Progression on Last LOT.

Conclusions: This first global analysis of real-world treatment patterns for triple-class-exposed patients with R/R MM showed a high subsequent treatment burden, with the majority receiving regimens containing at least triplets and retreatment with >1 drug received prior to becoming triple-class-exposed.

Clinical Trial Registry: NCT02076009 https://clinicaltrials.gov/ct2/show/NCT02076009?term=NCT02076009&draw=2&rank=1

NCT02136134 https://clinicaltrials.gov/ct2/show/NCT02136134?term=NCT02136134&draw=2&rank=1

NCT01998971 https://clinicaltrials.gov/ct2/show/NCT01998971?term=NCT01998971&draw=2&rank=1

Disclosure: This study was funded by Janssen Research & Development, LLC. Medical writing support was provided by Joanna Bloom, of Eloquent Scientific Solutions, and funded by Janssen Global Services, LLC.

Katja C Weisel: Honoraria (GlaxoSmithKline, Sanofi, Adaptive, Amgen, Bristol-Myers Squibb, Celgene, Janssen, and Takeda); Consultancy/Advisory (GlaxoSmithKline, Amgen, Adaptive, Bristol-Myers Squibb, Celgene, Janssen, Takeda, Sanofi, and Juno); Research Funding (Amgen, Celgene, Sanofi, and Janssen).

Thomas Martin: Research Funding (Janssen).

Kwee Yong: Honoraria (Janssen, GSK, Amgen Inc., Takeda, Sanofi); Research Funding (Janssen, Takeda, Sanofi).

Keqin Qi, Anil Londhe, Rachel Kobos, Ming Qi, Jordan M Schecter, Martin Vogel, Satish Valluri: Employee (Janssen). Amrita Krishnan: Speakers Bureau (Takeda, Amgen, BMS/Celgene); Consultancy (Sanofi, Janssen, Regeneron, BMS/Celgene); Member on an entity’s Board of Directors/Advisory Committee (Sutro, Z Predicta); Stock Ownership (BMS/Celgene). Sundar Jagannath: Consultancy (BMS, Janssen, Merck, Karyopharm, Legend Biotech, Sanofi, Takeda).

Myelodysplastic Syndromes

O128. Genomic Configuration of TP53 Mutations in Patients With Myeloid Neoplasia

Carmelo Gurnari^1,2, Vera Adema¹, Hassan Awada¹, Simona Pagliuca^1,3, Cassandra Kerr¹, Sunisa Kongkiatkamon¹, Thomas LaFramboise⁴, Valeria Visconte¹, Maria Teresa Voso², Milkkael Sekeres¹, Hetty Carraway¹, Torsten Haferlach⁵, Jaroslaw Maciejewski¹

¹ Cleveland Clinic, Cleveland, United States, ² University of Rome, Tor Vergata, Rome, Italy, ³ University of Paris, Paris, France, ⁴ Case Western Reserve University, Cleveland, United States, ⁵ Munich Leukemia Laboratory, Munich, Germany

Background: Somatic TP53 mutations are found in 10% of adult patients with MDS and de novo AML and in up to 20% of patients with therapy-related myeloid neoplasms (MNs). TP53 status is associated with complex karyotype (CK), aberrations of chromosome 5 and poor survival. Moreover, mutations in TP53 (TP53MT) may be an indication for hematopoietic cell transplantation and a predictive factor for relapse following this procedure. Here, we comprehensively characterize TP53MT MNs to better dissect the role of specific mutational configurations and identify the selective forces affecting outcomes in this poor prognostic MN category.

Methods: We analyzed results from whole genome and deep targeting sequencing from a large cohort of 4050 MNs collected at The Cleveland Clinic, other institutions and publicly available datasets.

Results: A total of 764 TP53 mutations were found in 632 TP53MT patients. Missense mutations were the most common (75%) followed by frameshifts (11%), splice sites (7%), nonsense (5%) and insertion/deletions (2%), with 20% of patients harboring more than 1 lesion. Topographical annotation revealed that missense mutations typically (98%) occurred within the DNA-binding domain and only 2% occurred outside this region (vs. 28% in case of truncating mutations, p < 0.0001). A male preponderance (1.42 vs. 1.1 M:F, p = 0.0069) and a younger age at presentation (median 68.9 vs 71, p < 0.00001) were found in wild-type (WT) vs. TP53MT. Mutant cases were also less enriched in de novo leukemia-driver genes mutations (e.g. NPM1, FLT3). When compared to WT MN, TP53MT cases were more likely to have CK (8% vs. 70%, p < 0.00001), del(5q) (4% vs. 40%, p < 0.00001), del(7q)/-7 (6% vs. 18%, p < 0.00001) and trisomy 8 (8% vs. 49%, p < 0.00001). Of note, deletion of the TP53 locus was found in 27% of mutated cases vs. only 1% of WT counterparts (p < 0.00001). With a median follow-up of 24 months (0.09-158), 1 year OS was 78% (95%CI 76.2-80.3) and 40.9% (95%CI 36-46.5%) in WT vs. mutated cases (HR 3.38, 95%CI 2.92-3.91, p ≤ 0.0001). When classifying patients according to TP53 genomic context (30% single vs. 70% double-hit, defined as a presence of biallelic, hemizigous or UPD configuration), progressive inactivation had an adverse impact on survival (p < 0.0001). Subsequent analysis revealed that double-hits patients had higher VAF (56.5% vs. 31.7%, p ≤ 0.0001) and lower number of co-mutations (median 1 vs. 0, p = 0.0007), suggesting an ancestral configuration of the TP53 mutations. On the other hand, single-hit TP53 patients with VAF < 30% showed survival outcomes similar to those of WT counterpart. Interestingly, R248Q carriers showed particular dismal outcomes when compared with patients with other non-hotspot missense mutations (p = 0.03) regardless of presence of CK, number of concurrent mutations, disease subtype, blast proportion and VAF. Finally, multivariable COX regression analysis showed that age, VAF>30%, CK and double-hits status negatively influenced survival outcomes.

Conclusions: In conclusion, TP53 genomic configuration stratifies two subgroups of patients with MNs characterized by different survival outcomes. The consideration of this parameter will possibly guide clinical decisions and lead to a better management of this particularly challenging setting of patients.

Disclosure: Nothing to declare.

O129. Retrospective Analysis Evaluating The Effect of Age on Outcomes After Allogeneic Hematopoietic Cell Transplantation In Chronic Myelomonocytic Leukemia. A Study of The EBMT-CMWP

Alicia Rovó¹, Luuk Gras², Brian Piepenbroek³, Nicolaus Kröger⁴, Christian Reinhardt⁵, Aleksandar Radujkovic⁶, Didier Blaise⁷, Guido Kobbe⁸, Riitta Niittyvuopio⁹, Uwe Platzbecker¹⁰, Sockel, Katja Sockel¹¹, Mathilde Hunault-Berger¹², J.J. Cornelissen¹³, Noel Milpied¹⁴, Jean Henri Bourhis¹⁵, Yves Chalandon¹⁶, Charles Craddock¹⁷, Véronique Leblond¹⁸, Johan Maertens¹⁹, Ahmet Elmaagacli²⁰, Nicola Mordini²¹, Patrick Hayden²², Marie Robin²³, Francesco Onida²⁴, Ibrahim Yakoub-Agha²⁵

¹ University Hospital of Bern, Bern, Switzerland, Bern, Switzerland, ² EBMT Statistical Unit, Leiden, Netherlands, ³ EBMT Leiden Study Unit, Leiden, Netherlands, ⁴ University Hospital Eppendorf, Hamburg, Germany, ⁵ University Hospital of Essen, Essen, Germany, ⁶ University of Heidelberg, Heidelberg, Germany, ⁷ Programme de Transplantation&Therapie Cellulaire, Marseille, France, ⁸ Heinrich Heine Universitaet, Duesseldorf, Germany, ⁹ HUCH Comprehensive Cancer Center, Helsinki, Finland, ¹⁰ Medical Clinic and Policinic 1, Leipzig, Germany, ¹¹ Medical Clinic and Policlinic I, University Hospital Dresden, Dresden, Germany, ¹² CHRU, Angers, France, ¹³ Erasmus MC Cancer Institute, Rotterdam, Netherlands, ¹⁴ CHU Bordeaux, Pessac, France, ¹⁵ Gustave Roussy Cancer Campus, Villejuif, France, ¹⁶ Département d`Oncologie, Service d`Hématologie, Geneva, Switzerland, ¹⁷ University Hospital Birmingham NHSTrust, Birmingham, United Kingdom, ¹⁸ Universite Paris IV, Hopital la Pitié-Salpêtrière, Paris, France, ¹⁹ University Hospital Gasthuisberg, Leuven, Belgium, ²⁰ Asklepios Klinik St. Georg, Hamburg, Germany, ²¹ Az. Ospedaliera S. Croce e Carle, Cuneo, Italy, ²² St. James’s Hospital, Dublin, Ireland, ²³ Hopital St. Louis, Paris, France, ²⁴ Fondazione IRCCS Ca’Granda Ospedale Maggiore Policlinico – University of Milan, Milan, Italy, ²⁵ CHU de Lille, Univ Lille, INSERM U 1286, Infinite, Lille, France

Background: Despite there being no consensus on the optimal therapy for CMML, it is generally recognized that allo-HCT is the only curative treatment modality. Nonetheless, CMML patient outcomes after allo-HCT are characterized by high rates of both non-relapse mortality (NRM) and relapse. In this retrospective study, we assessed for factors associated with allo-HCT outcomes, primarily focusing on age.

Methods: Patients >18 years old who underwent allo-HCT for CMML between 2010 and 2018 which were reported to the EBMT registry were included in the analysis. Overall Survival (OS) and Relapse Free Survival (RFS) rates were calculated by Kaplan-Meier methods; relapse incidence (REL) and NRM were analyzed as competing risks. Adjusted (cause specific) hazard ratios (HR) were obtained using Cox proportional hazard models. Age was modeled as a continuous linear variable and using restricted cubic splines.

Results: There were 1499 CMML patients with a median age of 60.4 years (18.9-76.3) and 69.4% were male. The WHO subtypes (available in 701 patients) were CMML-1 in 49% and CMML-2 in 51%. According to CPSS (available in 420 patients), 47% were ‘intermediate-2’ and 11.7% ‘high-risk’ at transplantation. 23% of patients had a Sorror HCT-CI ≥3. 28% had HLA-identical sibling donors, 63% were unrelated (MUD or MMUD) and 8.3% had mismatched related donors. A RIC regimen was used in 62%. Primary graft failure occurred in 3.8%. The cumulative incidence of acute GVHD at day +100 was 30% (95% CI 28-33%) whereas the incidence of chronic GVHD at 2 years was 35% (32-37%). At a median follow-up of 28.3 months, 5 years OS and RFS were 36% (95% CI, 33-40%) and 31%, (95% CI, 28-34%) respectively. REL and NRM at 5 years were 38% (95% CI, 35-41%) and 31% (95% CI, 29-34%) respectively. The HR for age as a continuous variable at transplantation (adjusted for other confounders), for each 10-year increase of age, was 1.16 (1.06-1.28) for OS and 1.13 (1.04-1.23) for RFS. The HR for REL and NRM was 1.06 (0.95-1.19) and 1.21 (1.06-1.37) for each 10-year increase of age respectively. The more flexible age model suggested a similar hazard of the endpoints death, relapse/death and death before relapse across the 55-65 age range whereas the hazard increased with age below 55 and after 65 years (Figure 1), even though no significant difference with the linear model was detected. By multivariate analysis, each 10-year increase in age and age>65 years were independently associated with significantly inferior OS, RFS and higher NRM, as were mismatched donors, Karnofsky PS ≤80, HCT-CI ≥3 and abnormal cytogenetics. The cause-specific HR for relapse comparing patients aged ≥65 years at allo-HCT with those <65 years was 1.22 (0.98-1.51).

Conclusions: Age ≥65 years negatively impacts on the survival outcomes of CMML patients undergoing allo-HCT, increasing both NRM and REL, independently of confounders including comorbidities and performance status. These findings should be borne in mind when deciding whether to transplant CMML patients older than 65 years of age.

Disclosure: Alicia Rovó has received research grants (to institution, not investigator) from Novartis, CSL Behring and AG Alexion, consulting fees from Novartis and honoraria for advisory Board from Novartis, Orpha Swiss GMBH, and BMS. All other authors have Nothing to declare. This is a study of the EBMT Chronic Malignancies Working Party.

O130. Change In IPSS-R Between Diagnosis And Transplant And Transplant Outcome in Patients With MDS. A Retrospective Analysis From The Chronic Malignancies Working Party

Christof Scheid¹, Dirk-Jan Eikema², Riitta Niittyvuopio³, Johan Maertens⁴, Jakob Passweg⁵, Didier Blaise⁶, Jennifer Byrne⁷, Nicolaus Kröger⁸, Martin Bornhäuser⁹, Patrice Chevallier¹⁰, Jean-Henri Bourhis¹¹, Jan J. Cornelissen¹², Henrik Sengeloev¹³, Jürgen Finke¹⁴, John A Snowden¹⁵, Tobias Gedde-Dahl¹⁶, Denis Guyotat¹⁷, Urs Schanz¹⁸, Amit Patel¹⁹, Linda Koster², Liesbeth C. de Wreede²⁰, Patrick J Hayden²¹, Francesco Onida²², Marie Robin²³, Ibrahim Yakoub-Agha²⁴

¹ Uniklinik Köln, Cologne, Germany, ² EBMT Data Office Leiden, Leiden University Medical Center, Leiden, Netherlands, ³ Helsinki Hospital, Helsinki, Finland, ⁴ University Hospital Gasthuisberg, Leuven, Belgium, ⁵ University Hospital Basel, Basel, Switzerland, ⁶ Institut Paoli Calmettes, Marseille, France, ⁷ Centre for Clinical Haematology, Nottingham University Hospitals Trust, Nottingham, United Kingdom, ⁸ University Hospital Eppendorf, Hamburg, Germany, ⁹ University Cancer Center (UCC), University Hospital Carl Gustav Carus, Technical University Dresden, Dresden, Germany, ¹⁰ Hematology Clinic, CHU Nantes, Nantes, France, ¹¹ Gustave Roussy, Villejuif, France, ¹² Erasmus MC Cancer Institute, Rotterdam, Netherlands, ¹³ Rigshospitalet, Copenhagen, Denmark, ¹⁴ University of Freiburg, Freiburg, Germany, ¹⁵ Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, United Kingdom, ¹⁶ Oslo University Hospital, Rikshospitalet and Institute of Clinical Medicine, University of Oslo, Oslo, Norway, ¹⁷ Lucien Neuwirth Cancer Institute, Saint-Priest-en-Jarez, France, ¹⁸ University Hospital Zurich, Zurich, Switzerland, ¹⁹ The Christie NHS Foundation Trust, Manchester, United Kingdom, ²⁰ LUMC, Leiden, Netherlands, ²¹ St. James’s Hospital, Dublin, Ireland, ²² MT Center, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico - University of Milan, Milano, Italy, ²³ Hopital Saint Louis, Paris, France, ²⁴ Univ Lille, CHU de Lille, INFINITE, INSERM U1286, Lille, France

Background: IPSS-R is a well established prognostic factor for transplant outcome in patients with MDS, irrespective whether it is assessed at diagnosis or at transplant. However it is unclear how a change in IPSS-R, e.g. by reducing bone marrow blasts through therapy, would potentially affect transplant results. In particular the decision to treat patients before transplant or perform an upfront allogeneic transplantation can so far not be based on evidence.

Methods: From the EBMT registry patients with MDS and sufficient data to calculate IPSS-R at diagnosis and before transplant were identified from the period 2005 -2018.

1482 patients were analysed. Median age was 59 (interquartile range 51-64) years, 60% were male, 40% female. Donors were related in 36% and unrelated in 64%, graft source was PBSC in 85% of cases. Conditioning was standard dose in 33% and reduced intensity in 67%.

Results: IPSS-R both at diagnosis and at transplant had a significant impact on OS and RFS. To investigate the effect of a change in IPSS-R between diagnosis and transplant we formed 3 subgroups: same IPSS-R, improved IPSS-R, worsened IPSS-R. A change in IPSS-R was noted 77.5% of patients with prior chemotherapy, 72% with prior HMA and 59.8% of untreated patients. Univariate analysis showed no significant difference in OS or RFS in patients with stable IPSS-R compared to improved or worsened IPSS-R.

To analyse the effect of a change in IPSS-R in the context of IPSS-R at diagnosis and type of therapy between diagnosis and transplant (none, chemotherapy, HMA, other) a multivariable Cox regression for OS and RFS was performed. From this model the predicted OS at 2yrs for untreated patients with high risk IPSS-R at transplant was 63%, 57% and 65% for same, improved or worsened IPSS-R, respectively. With chemotherapy predicted 2yr OS was 35%, 59% and 32%, with HMA 46%, 48% and 19%, with other treatments 2yr OS was 35%, 46% and 21% for same, improved or worsened IPSS-R. RFS at 2 years in patients with high-risk IPSS-R at diagnosis was predicted 55%, 49% and 57% with no treatment, 31%, 54% and 31% after chemotherapy, 41%, 40% and 14% after HMA, and 36%, 40% and 15% after other therapies. Similar results were obtained for the other IPSS-R risk categories.

Conclusions: In this retrospective analysis from a large cohort of patients with MDS a change in IPSS-R between diagnosis and transplant did not appear to affect OS and RFS after transplant in patients with no intermittent treatment. In treated patients worsening of IPSS-R had a negative effect on OS and RFS. Improving IPSS-R after therapy however did not show a positive effect on OS or RFS. Thus for MDS patients receiving an allogeneic transplantation our results provide no clear signal that prior therapy improves transplant outcome.

Disclosure: Scheid: Novartis: Honoraria, Research Funding; Janssen: Honoraria, Research Funding; BMS: Honoraria; Amgen: Honoraria; Takeda: Honoraria, Research Funding. Blaise: Jazz Pharmaceuticals: Honoraria. Chevallier: Incyte Corporation: Honoraria. Yakoub-Agha: Celgene: Honoraria; Novartis: Honoraria; Gilead/Kite: Honoraria, Other: travel support; Janssen: Honoraria; Jazz Pharmaceuticals: Honoraria.

Myeloproliferative Neoplasm

O131. Allogeneic Hematopoietic Cell Transplantation In Patients With Therapy-Related Myeloid Neoplasm After Breast Cancer Treatment: A Study of The Chronic Malignancies Working Party of The EBMT

Mitja Nabergoj¹, Katya Mauff², Dietrich Beelen³, Arnold Ganser⁴, Nicolaus Kröger⁵, Friedrich Stölzel⁶, Jürgen Finke⁷, Jakob Passweg⁸, Jan Cornelissen⁹, Natalie Schub¹⁰, Joan Hendrik Veelken¹¹, Yves Beguin¹², Keith Wilson¹³, Tsila Zuckerman¹⁴, Mathilde Hunault-Berger¹⁵, Bruno Lioure¹⁶, Rocio Parody Porras¹⁷, Pascal Turlure¹⁸, Tessa Kerre¹⁹, Linda Koster²⁰, Patrick Hayden²¹, Francesco Onida²², Christof Scheid²³, Yves Chalandon¹, Marie Robin²⁴, Ibrahim Yakoub-Agha²⁵

¹ Hôpitaux Universitaires de Genève, Geneva, Switzerland, ² EBMT Statistical Unit, Leiden, Netherlands, ³ Universitätsklinikum Essen, Essen, Germany, ⁴ Hannover Medical School, Hannover, Germany, ⁵ University Hospital Eppendorf, Hamburg, Germany, ⁶ Universitaetsklinikum Dresden, Dresden, Germany, ⁷ University of Freiburg, Freiburg, Germany, ⁸ University Hospital, Basel, Switzerland, ⁹ Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, Netherlands, ¹⁰ University Hospital of Schleswig-Holstein, Kiel, Germany, ¹¹ Leiden University Medical Center, Leiden, Netherlands, ¹² CHU and University of Liège, Liège, Belgium, ¹³ University Hospital of Wales, Cardiff, United Kingdom, ¹⁴ Rambam Health Care Campus, Haifa, Israel, ¹⁵ CHRU Angers, Angers, France, ¹⁶ ICANS, Strasbourg, France, ¹⁷ ICO-Hospital Duran i Reynals, Barcelona, Spain, ¹⁸ Centre Hospitalier Universitaire, Limoges, France, ¹⁹ Ghent University Hospital, Ghent, Belgium, ²⁰ EBMT Date Office, Leiden, Netherlands, ²¹ Trinity College Dublin, St. James’s Hospital, Dublin, Ireland, ²² Hematology-BMT Center, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milano, Italy, ²³ University Hospital Koeln, Cologne, Germany, ²⁴ Hôpital Saint-Louis, APHP, Université Paris 7, Paris, France, ²⁵ CHU de Lille, Univ Lille, INSERM U 1286, INFINITE, Lille, France

Background: Therapy-related myeloid neoplasms (t-MNs) represent a clinical dilemma in the transplantation field. In this retrospective registry study, we focused on t-MN following treatment for breast cancer, a relatively common cause of this disease.

Methods: We performed a retrospective EBMT registry analysis of adult patients who underwent a first allogeneic hematopoietic cell transplant (allo-HCT) for t-MDS or t-AML between 2006 and 2016, and proceeded to select those with a prior diagnosis of breast cancer which had been treated with either chemotherapy, radiotherapy or both.

Results: We identified 252 adult female patients with a median age at transplantation of 57 years, and a median time from the diagnosis of breast cancer to the diagnosis of t-MN and the subsequent allo-HCT of 3.7 and 4.6 years, respectively. t-AML was the indication for allo-HCT in 77% of cases and t-MDS in 23%. 174 (76%) patients had an abnormal karyotype. The t-MN was in complete remission (CR) at the time of transplant in 169 (67%) patients whereas 29% had either stable or progressive disease. A Karnofsky performance status of less than 90 (KPS<90) was noted in 67 patients (29%). 152 (61%) received a reduced-intensity conditioning (RIC) regimen and in vivo T-cell depletion was given in 138 (55%) cases. Mobilized peripheral blood was the stem cell source in 225 (89%). Ninety-nine (39%) patients had matched related donors. As regards the breast cancer diagnosis, eight (5%) patients presented with metastatic disease and 15 (15%) had triple-negative breast cancer; 163 (86%) received chemotherapy, 172 (88%) radiotherapy and 129 (72%) both. At transplant, 191 (98%) patients had no detectable breast cancer. The 2-year overall survival (OS), progression free-survival (PFS), relapse incidence and NRM were 50%, 45%, 33% and 22%, respectively. In multivariate analysis, the absence of t-MN CR at transplant was associated with lower OS, PFS and a higher relapse incidence, as follows: [HR 1.8 (95%CI 1.01-3.18, p = 0.045)], [HR 1.99, 95%CI (1.13 - 3.5) p = 0.018)] and [HR 2.47 (95%CI 1.21-5.05), p = 0.013)], respectively. A KPS<90 was associated with lower RFS [HR 1.85 (95%CI 1.12-3.16, p = 0.016)]. The incidence of acute grade II-IV GVHD was 26%. The two-year incidence of chronic GVHD was 41%, of which more than half (22%) was extensive. At a median follow-up of 29 months post-transplant, 17 (7%) cases of breast cancer recurrence were recorded, 13 invasive, with a curative approach adopted in seven cases. The commonest cause of death in the whole cohort was relapse of t-MN (n = 54, 36%), followed by infection (n = 42, 28%) and GVHD (n = 24, 16%). Relapse of the primary breast cancer accounted for 10 (7%) deaths.

Conclusions: This study of a large cohort of patients with t-MN following treatment for breast cancer shows encouraging transplant outcomes. Control of the t-MN prior to transplant remains crucial. The incidence of breast cancer relapse post-transplant (7%) remains a cause for concern, especially as more than half of these patients subsequently died of disease progression.

Disclosure: Nothing to declare.

O132. Impact of Prior Jak-Inhibitor Therapy With Ruxolitinib on Outcome After Allogeneic Stem Cell Transplantation For Myelofibrosis. A Large EBMT-CMWP Study

Nicolaus Kröger¹, Giulia Sbianchi², Tiarlan Sirait³, Dietrich Beelen⁴, Jakob Passweg⁵, Marie Robin⁶, Radovan Vrhovac⁷, Grzegor Helbig⁸, Katja Sockel⁹, Eibhlin Conneally¹⁰, Marie Thérèse Rubio¹¹, Yves Beguin¹², Jürgen Finke¹³, Paolo Bernasconi¹⁴, Elena Morozova¹⁵, Johannes Clausen¹⁶, Peter von dem Borne¹⁷, Nicolaas Schaap¹⁸, Wilfried Schroyens¹⁹, Francesca Patriarca²⁰, Nicola Di Renzo²¹, Zeynep Arzu Yegin²², Patrick Hayden²³, Donal McLornan^24,25, Ibrahim Yakoub-Agha²⁶

¹ University Medical Center Hamburg-Eppendorf, Hamburg, Germany, ² University of Rome ‘Tor Vergata’, Rome, Italy, ³ EBMT Data Office, Leiden, Netherlands, ⁴ Essen University Hospital, Essen, Germany, ⁵ University Hospital of Basel, Basel, Switzerland, ⁶ Hopital St. Louis, Paris, France, ⁷ University Hospital Center Rebro, Zagreb, Croatia, ⁸ Silesian Medica Academy, Katowice, Poland, ⁹ University Hospital Dresden, Dresden, Germany, ¹⁰ Hope Directorate St. James’s Hospital, Dublin, Ireland, ¹¹ Hopital d’Enfants, Vandoeuvre Nancy, France, ¹² University of Liege and CHU of Liege, Liege, Belgium, ¹³ University of Freiburg, Freiburg, Germany, ¹⁴ IRCCS Policlinico San Matteo, Pavia, Italy, ¹⁵ Pavlov First Saint Petersburg State Medical University of St. Petersburg, St. Petersburg, Russian Federation, ¹⁶ Ordensklinikum Linz - Elisabethinen, Linz, Austria, ¹⁷ Leiden University Hospital, Leiden, Netherlands, ¹⁸ Nijmegen Medical Centre, Nijmegen, Netherlands, ¹⁹ Antwerp University Hospital (UZA), Antwerp Edegem, Belgium, ²⁰ University Hospital and DAME, Udine, Italy, ²¹ Unita Operativa di Ematologia e Trapianto di Cellule Staminali, Lecce, Italy, ²² Gazi University Faculty of Medicine, Ankara, Turkey, ²³ Trinity College Dublin, St. James’s Hospital, Dublin 8, Ireland, ²⁴ Guy’s Hospital, London, United Kingdom, ²⁵ University College London Hospital, London, United Kingdom, ²⁶ CHU de Lille, Univ Lille, INSERM U 1286, Infinite, Lille, France

Background: JAK1/2 inhibitor ruxolitinib reduces spleen size and improves constitutional symptoms in patients with myelofibrosis but the impact of pretreatment with ruxolitinib on outcome after allogeneic stem cell transplantation (HSCT) remains to be determined.

Methods: We evaluated the impact of ruxolitinib on outcome in 551 myelofibrosis patients who received allogeneic HSCT between 2012 and 2016 either without (n = 274) or with (n = 277) ruxolitinib pretreatment. In the ruxolitinib pretreated group were more intermediate-2/high risk patients according to DIPSS and more patients with Karnofsky score ≤ 80. Ruxolitinib pretreated patients were separated into at transplantation responsive and no responsive/or lost response patients.

Results: Ruxolitinib rebound phenomenon after discontinuation was noted in 6%. The overall leukocyte engraftment on day 45 was 92% and significantly higher in ruxo responsive patients than those who had no or lost response to ruxolitinib (94% vs. 85%, p = 0.05). The 1 year non-relapse mortality was 22% (95% CI: 18 – 25) without significant difference between the arms. In a multivariate analysis (MVA) ruxo pretreated patients with ongoing spleen response at transplant had a significantly lower risk of relapse [8.1% vs. 19.1%; HR: 0.34 (95% CI 0.12 – 0.95, p = 0.04)] and better 2-year event-free survival [68.9% vs. 53.7%; HR: 0.61 (95% CI 0.40 – 0.91, p = 0.02)] in comparison to patients without ruxolitinib pretreatment. For overall survival the only significant factors were age > 58 years (HR: 1.42, 95% CI 1.04 – 1.95, p = 0.03) and HLA mismatch donor (HR: 2.37, 95% CI 1.40 – 4.03, p = 0.001).

Conclusions: Ruxolitinib prior to allogeneic HSCT in myelofibrosis is feasible and does not negatively impact outcome after transplantation. Ruxolitinib pretreated patients with ongoing spleen response at time of transplantation have a better leukocyte engraftment and a lower relapse rate resulting in an improved event-free survival in comparison to patients who received HSCT after failure to ruxolitinib.

Disclosure: EBMT received research grant from Novartis.

O133. Additional Chromosomal Aberrations In CML Patients Undergoing Allogeneic Hematopoietic Stem Cell Transplantation

Elena Morozova¹, Yulia Vlasova¹, Maria Barabanshikova¹, Tatjana Rudakova¹, Ksenia Jurovskaya¹, Dmitrii Zhogolev¹, Tatyana Gindina¹, Ildar Barkhatov¹, Ivan Moiseev¹, Alexander Kulagin¹

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

Background: High risk additional chromosomal aberrations (ACAs) are predictors of blast crisis evolution and are associated with poor prognosis in CML patients. Acquisition of high risk ACAs during the disease course of CML is associated with end-phase CML and allogeneic hematopoietic transplantation (alloHSCT) is indicated in this case. Here we evaluate the results of alloHSCT in patients with ACAs before and after transplantation.

Methods: A total of 42 patients were retrospectively evaluated in whom ACAs were documented before alloHSCT. High risk ACAs were registered in 37 (88%) patients, low risk ACAs – 5 (12%). After acquisition of ACAs therapy was changed in 41 (98%) patients. Thirty-eight (91%) patients carried ACAs in Ph-positive cells and 4 (9%) - in Ph-negative cells. In the majority of patients 26 (62%) complex karyotype was documented (figure 1A).

All patients received allo-HSCT with a reduced-intensity conditioning regimen with fludarabine 180 mg/m2 and busulfan 8-12 mg/kg or melphalan 140 mg/m2. AlloHSCT was performed from HLA-matched 30 (72), mismatched 11 (26) or haploidentical 1 (2) donor.

Chromosome aberrations were described according to the ISCN 2016. ACAs risk stratification was performed as previously reported by R. Hehlmann et al.

Overall survival (OS) was defined as the time from the date of alloHSCT to death, progression-free survival (PFS) - as the time between the date of alloHSCT and post-transplant relapse, death.

Results: Median follow up after alloHSCT was 31 month (1-123). Engraftment was documented in 38 (91%) patients. The cumulative incidence of non-relapse mortality at day 100 and 1 year after allo-HSCT were 10% and 20%, respectively. Grade 2-4 acute graft versus host disease (GVHD) was documented in 9 (21%), grade 3-4 acute GVHD – 7 (17%), chronic GVHD – 8 (22%) including mild, moderate and severe form in 2 (5%), 5 (14%) and 1 (3%) patients, respectively. Three-year cumulative incidence of relapse was 46% (Figure 1B). Four patients developed molecular relapse, 3 –cytogenetic, 10 – hematological. Among them in 7 patients ACAs were documented during relapse. Five patients carried the same ACAs variant as before alloHSCT. In all but one case, ACAs were associated with hematological relapse which resulted in further disease progression irrespective of therapeutic interventions such as second allo-HSCT (n = 2), TKI therapy initiation (n = 1), TKI therapy change (n = 2), donor lymphocyte infusion (n = 1). In one case cytogenetic relapse was successfully treated with DLI. ACAs were registered usually during the first year after alloHSCT (median day +228, range 40-1058). In patients without ACAs after alloHSCT molecular relapse was registered in 3 cases, cytogenetic - 2, hematological - 5. Among them 6 patients progressed, 4 – achieved stable complete molecular response. Five-year OS was 40% (95%СI 22%-57%), 5-year PFS - 28% (95%CI 13%-46%) (Figure 1C).

Conclusions: Allogeneic hematopoietic stem cell transplantation is an effective therapeutic option in considerable proportion of CML patients with high risk ACAs. Acquisition of ACAs after alloHSCT is associated with hematological relapse after alloHSCT and poor outcome. An earlier referral of patients with high risk ACAs to the transplant center may improve the outcome of alloHSCT and prognosis.

Disclosure: Nothing to declare.

O134. Donor Age, CD34 Cell Dose And Non-Blast Crisis Are Prognostic Factors For Outcome After Stem Cell Transplantation In Patients With Advance Phase CML

Christian Niederwieser¹, Elena Morozova², Ludmila Zubarovskaya², Tatjana Zabelina¹, Evgeny Klyuchnikov¹, Dietlinde Janson¹, Christine Wolschke¹, Maximilian Christopeit¹, Francis Ayuk¹, Ivan Moiseev², Boris V. Afanasyev², Nicolaus Kröger¹

¹ University Medical Center Hamburg Eppendorf, Hamburg, Germany, ² Raisa Gorbacheva Memorial Institute for Children Hematology and Transplantology, Saint Petersburg, Russian Federation

Background: The introduction of tyrosine kinase inhibitors (TKI) for BCR/ABL in patients with chronic phase (CP) chronic myeloid leukemia (CML) resulted in disappearance of the disease, restoration of normal life expectancy and even treatment free remission. In contrast, the outcome of patients with advanced disease is still dismal. HSCT remains the only curative option in advanced phase, but data on outcome and risk factors are scarce, on limited patient numbers and short follow-up. The objective of the study was to investigate the outcome in a large population of patients with advanced disease treated with different generations of TKI and transplanted in only two centers over a follow-up period of up to 15 years. The identification of risk factors in multivariate analysis should help to optimize the outcome of this high-risk patient population.

Methods: All patients transplanted in two transplant centers between 1990 and 2018 with a history of Blast Crisis (BC; n = 96) or advanced phase other than BC (n = 51) were analyzed. The predominantly male (64%) patients had a median age of 39 (7-76) years and were transplanted in ≥CP2 (n = 70), in AP (n = 40) or in BC (n = 37). Median time between diagnosis and HSCT was 1.9 (0.3-24.4) years. HSCT were performed after myeloablative (43%) or reduced intensity (57%) conditioning from peripheral blood (PB, 57.8%) or bone marrow (BM, 42.2%) using unrelated (matched 39.5%; mismatched 24.5%) or related (matched 32.0%; haploidentical 4.0%) donors with a median age of 36.0 (14-66) years.

Results: Overall survival (OS) amounted 34 (95%CI 22-46)% and progression-free survival (PFS) 26 (95% CI 16-36)% @15 years, respectively. Independent adverse risk factors for OS and PFS were low CD34+ count in the graft, donor age (>36 years) and BC. Cumulative incidence of Non-Relapse Mortality (NRM) was 28 (18-38)% and cumulative incidence of relapse (RI) 43 (33-53)% @15 years. PB-HSCT and HSCT after 2008 were favorable prognostic factors for NRM while family donor and patient age >39 years were independently associated with worse RI.

Table 1 Multivariate cox regression analysis of risk factors for overall survival (OS), progression-free survival (PFS), non-relapse mortality (NRM) and relapse incidence (RI).

Conclusions: In conclusion, HSCT resulted in long-term OS in patients with advanced phase CML and was superior in non-BC patients, with donors ≤36 years and with higher CD34+ dose in the graft.

Disclosure: Nothing to declare.

O135. Outcomes of Allogeneic Hematopoietic Stem Cell Transplantation in Patients With Myelofibrosis Treated With Ruxolitinib in The Pre-Transplant Period

Patrycja Mensah-Glanowska¹, Agnieszka Szeremet², Barbara Nasilowska-Adamska³, Maciej Majcherek², Beata Jakubas¹, Kazimierz Halaburda³, Tomasz Wrobel², Anna Czyz ²

¹ Jagiellonian University Medical College, Krakow, Poland, ² Wroclaw Medical University, Wroclaw, Poland, ³ Institute of Hematology and Blood Transfusion, Warsaw, Poland

Background: Allogeneic hematopoietic cell transplantation (alloHCT) is, at present, the only potentially curative therapy for myelofibrosis (MF). Despite improvements in transplant techniques, its’ curative efficacy is limited by morbidity and transplant-related mortality. Here, we present a cohort of MF patients treated with ruxolitinib (RUX) in pre-transplant period to enhance the published experience in this area. The goals of the study were to determine the outcomes of alloHCT, analyze the incidence of acute and chronic graft-versus-host diseasea (GvHD), as well as cytokine release syndrome (CRS).

Methods: An analysis included patients with MF who underwent alloHCT between January 2018 and November 2020 following treatment with RUX which was continued until at least start of conditioning regimen. Data was collected from our institutional databases.

Results: The study group consisted of 26 patients (median age 56 years, range 34-68) with primary (n = 12) or secondary MF (n = 14). JAK2 mutation was found in 21 patients. DIPSS Plus risk was intermediate-2 or high in all but 3 patients. The patients underwent alloHCT from identical siblings (n = 5), matched unrelated donor (n = 19) or family haploidentical donor (n = 2) following myeloablative (n = 13) or reduced toxicity (n = 13) conditioning regimen. A total of 15 patients were conditioned with busulfan and fludarabine, and 11 with thiotepa, busulfan and fludarabine (TBF). RUX was tapered gradually before alloHCT and stopped a day or two days before the start of conditioning regimen in 21 patients. In four patients RUX was continued at the doses 20 mg BID until day -4 before HCT, and one patient received RUX (5 mg per day) until day +5 post-transplant. GvHD prophylaxis consisted of calcineurin inhibitor combined with MTX +/- ATG in alloHCT from HLA matched donor, and post-transplant Cy with calcineurin inhibitor and MMF in haploHCT. A total of 4 patients (15%) developed reversible CRS with symptoms of G1 (n = 2), G2 (n = 1), and G3 (n = 1) severity at a median of 8 days (range, 4-21) after HCT. One patient died before engraftment due to diffuse alveolar hemorrhage (DAH). The rate of acute GvHD grade II-IV was 30% in evaluable patients. No patient developed moderate or severe extensive chronic GvHD. After the median follow-up time of 18 months (range, 1-33) the overall survival (OS) and progression-free survival (PFS) estimates at 24 months were 68.5% (95%CI 59-79) and 60.2% (95%CI 50-69), respectively (Figures). Non-relapse mortality (NRM) and relapse incidence at 24 months was 31.5% (95%CI13-52) and 8.4% (95%CI2-31), respectively.

Conclusions: Our results indicate that alloHCT in RUX-pretreated patients provided satisfactory PFS and OS. Interestingly, low incidence of both cGVHD and RI were observed. CRS observed early after transplant was manageable and reversible. However, NRM remains of great concern.

Disclosure: Authors have no conflict of interest.

New Drugs- and Cell-based Immune Therapies

O136. Narsoplimab (OMS721), A MASP-2 Inhibitor, For The Treatment of Adult Hematopoietic Stem Cell Transplant-Associated Thrombotic Microangiopathy (HSCT-TMA): Subgroup Analyses

Alessandro Rambaldi^1,2, Kathleen Claes³, Yeow Tee Goh⁴, Yok Lam Kwong⁵, Nelson Leung⁶, Jameela Sathar⁷, Edmund Ng⁸, Narinder Nangia⁹, Steve Whitaker⁹

¹ University of Milan, Milan, Italy, ² Azienda Socio Sanitaria Territoriale Papa Giovanni XXIII, Bergamo, Italy, ³ UZ Leuven, KU Leuven, Leuven, Belgium, ⁴ Singapore General Hospital, Singapore, Singapore, ⁵ Queen Mary Hospital, Pok Fu Lam, Hong Kong, SAR of China, ⁶ Mayo Clinic, Rochester, United States, ⁷ Ampang Hospital, Selangor, Malaysia, ⁸ Pacific Northwest Statistical Consulting, Seattle, United States, ⁹ Omeros Corporation, Seattle, United States

Background: HSCT-TMA (also known as TA-TMA) is a serious complication of hematopoietic stem cell transplantation associated with significant mortality and morbidity, with no currently approved therapy. HSCT-TMA results from endothelial injury caused by conditioning regimens, immunosuppressants, infection, and GVHD. Endothelial injury specifically activates the lectin pathway of complement. Narsoplimab, an inhibitor of mannan-binding lectin-associated serine protease-2 (MASP-2), the effector enzyme of the lectin pathway and an activator of the coagulation cascade, was studied for treatment of HSCT-TMA. A Biologics License Application has been submitted to the US FDA for the treatment of HSCT-TMA. This paper will discuss results of subgroup analyses from the pivotal trial.

Methods: This was a single-arm open-label pivotal trial in adult HSCT-TMA patients at high risk for poor outcomes (NCT02222545). Treatment consisted of IV narsoplimab once weekly for 4 or 8 weeks with a 6-week follow-up period. The FDA-agreed primary endpoint (complete response [CR]) required clinical improvements in 2 categories: 1) laboratory TMA markers (platelet count and LDH) and 2) organ function (kidney, pulmonary, gastrointestinal or neurological) or freedom from transfusion. Patients receiving at least 1 dose of narsoplimab (full analysis set [FAS]; N = 28) and patients receiving the protocol-specified narsoplimab treatment for at least 4 weeks (per-protocol [PP]; N = 23) were analyzed. Subgroup analysis was performed for patients defined by baseline demographics, clinicopathologic features, and HSCT complications.

Results: Treatment with narsoplimab showed a 61% (17/28) CR rate (CRR) in the FAS and a 74% (17/23) CRR in the PP population. One hundred-day survival post HSCT-TMA diagnosis was 68% in the FAS, 83% in the PP population, and 94% in complete responders. Subgroup analysis showed CR in 58% (14/24) of patients <65 years old vs 75% (3/4) of patients ≥65 years old, and in 65% (13/20) of men vs 50% (4/8) of women (Figure 1). Patients with GVHD, significant infection, and multiple organ TMA involvement achieved CRR of 63% (12/19), 63% (15/24), and 64% (9/14), respectively. Patients with baseline platelet counts ≤20 x 10⁹/L showed similar CR to those with platelet counts >20 x 10⁹/L (60% vs 61%). Patients with baseline organ dysfunction (kidney, pulmonary, gastrointestinal, and neurological) achieved CRR of 57% (12/21), 40% (2/5), 63% (10/16), and 100% (1/1), respectively. Lack of transfusions within 2 weeks prior to narsoplimab treatment resulted in 100% (3/3) CR. Adverse events were typical of this population and no safety signals were identified. Six deaths occurred during the core study period. Causes of death included septic shock, neutropenic sepsis, AML, GVHD, and TMA.

Conclusions: In this high-risk HSCT-TMA population with significant disease burden and high mortality risk, a high proportion of narsoplimab-treated patients showed significant improvement in laboratory markers and organ function, regardless of subgroup.

Clinical Trial Registry: ClinicalTrials.gov: NCT02222545

Disclosure:

• Alessandro Rambaldi: Amgen (Advisory board, symposia, travel support), Astellas (Advisory board), Celgene (Advisory board, symposia, travel support), Gilead (Advisory board, travel support), Italfarmaco (Advisory board, travel support), JAZZ (Advisory board), Novartis (Advisory board, symposia, travel support), Omeros Corporation (Advisory board, consultant), Pfizer (Advisory board, lecturer), Roche (Advisory board, symposia, travel support), Sanofi Aventis (Advisory board)

• Kathleen Claes: Alexion (Consultant), Astellas (Consultant), AstraZeneca (Consultant)

• Yeow Tee Goh: Nothing to declare.

• Yok Lam Kwong: Nothing to declare.

• Nelson Leung: AbbVie (Advisory board), Takeda (Advisory board), Aduro (Advisory board)

• Jameela Sathar: Nothing to declare.

• Edmund Ng: Omeros Corporation (Consultant)

• Narinder Nangia: Omeros Corporation (Employment)

• Steve Whitaker: Omeros Corporation (Employment)

O137. Improvements in Fatigue And Physical Function Evaluated Through Changes in Clinical Outcomes In Paroxysmal Nocturnal Hemoglobinuria: Post-HOC Analyses From The Pegasus Study

Alexander Röth¹, William R. Lenderking², Sujata P. Sarda³, Scott B. Baver³, Mohamed Hamdani³, Ray Hsieh², Shannon Shaffer², Michael Yeh³, David Cella ⁴

¹ University of Duisberg-Essen, Essen, Germany, ² Evidera, Waltham, United States, ³ Apellis Pharmaceuticals, Inc., Waltham, United States, ⁴ Northwestern University, Chicago, United States

Background: Paroxysmal nocturnal hemoglobinuria (PNH) is a rare, hematologic disease characterized by complement-mediated hemolysis, thrombosis, impaired bone marrow function, and anemia. In a randomized phase 3 trial (PEGASUS; NCT03500549), pegcetacoplan demonstrated a statistically significant improvement in hemoglobin levels compared to eculizumab at Week 16 in PNH patients with partial response to prior eculizumab treatment. The objective of these post-hoc analyses was to explore relationships between patient-reported measures of fatigue and physical functioning with clinical and hematological parameters, namely hemoglobin levels, absolute reticulocyte count (ARC), and indirect bilirubin levels from baseline to Week 16 in PNH patients.

Methods: Adult patients with PNH and hemoglobin levels <10.5 g/dL despite treatment with eculizumab for ≥3 months were eligible for the study. Eighty patients started a 4-week run-in period with both eculizumab and twice-weekly pegcetacoplan treatments (1080 mg, self-administered subcutaneously) before 1:1 randomization to either monotherapy for 16 weeks. Fatigue was assessed using the Functional Assessment of Chronic Illness Therapy (FACIT)–Fatigue scale and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core-30 (EORTC-QLQ-C30) questionnaire, which also evaluates physical functioning. Convergent validity was assessed using Spearman correlations. To evaluate responsiveness to change, patients were grouped by increases in hemoglobin levels (<1 g/dL, ≥1 to <2 g/dL, and ≥2 g/dL), decreases in ARC (≥70x10⁹ cells/L and <70x10⁹ cells/L) and decreases in indirect bilirubin levels (≥7.6 μmol/L and <7.6 μmol/L). Cut points were determined based on the median splits for each variable.

Results: At Week 16 (n = 80; intent-to-treat), FACIT-Fatigue scores were significantly correlated with hemoglobin (r = 0.47, p < 0.0001), ARC (r = −0.37, p < 0.01) and indirect bilirubin (r = −0.25, p < 0.05). EORTC-QLQ-C30 fatigue scores were correlated with ARC (r = 0.28, p < 0.05) and hemoglobin (r = −0.39, p < 0.001), but not with indirect bilirubin. EORTC-QLQ-C30 physical function scores were correlated with hemoglobin (r = 0.45, p < 0.0001), ARC (r = −0.28, p < 0.05), and indirect bilirubin (r = −0.26, p < 0.05). Patients with larger decreases in ARC (F = 15.2, p = 0.0002) and indirect bilirubin (F = 15.8, p = 0.0002) showed improvements in FACIT-Fatigue scores (9.3 for ARC; 9.2 for indirect bilirubin). Groups with greater improvements in hemoglobin over 16 weeks also showed the largest improvement in FACIT-Fatigue scores (F = 9.0, p < 0.0001), with the largest fatigue reduction observed in the hemoglobin group with an increase of ≥2g/dL (11.3-point improvement in FACIT-Fatigue score). Fatigue and physical functioning, as defined by the EORTC-QLQ-C30, also demonstrated improvements across the three outcomes measured. Patients with larger decreases in ARC (F = 6.1, p = 0.02) and indirect bilirubin levels (F = 7.6, p = 0.007), and greater improvements in hemoglobin levels (F = 4.1, p = 0.0093), showed the greatest improvement in fatigue. Similarly, patients in both groups showed the greatest improvement in physical function (F = 7.7, p = 0.007 for ARC; F = 8.4, p = 0.005 for indirect bilirubin levels; F = 4.1, p = 0.0103 for hemoglobin levels).

Conclusions: The results of the current study in PNH patients demonstrate that fatigue and physical functioning outcomes are correlated with improvements in clinical and hematological parameters. The FACIT-Fatigue and EORTC-QLQ-C30 scales appear to be useful and valid patient-reported measures for assessing meaningful change in PNH.

Clinical Trial Registry: NCT03500549

Disclosure: Alexander Röth, MD, reports consulting honoraria for Alexion and Apellis Pharmaceuticals, Inc., Biocryst, Novartis, Roche, and Sanofi, and research funding from Alexion and Roche. William R. Lenderking, PhD, and Ray Hsieh, MS, are current employees at Evidera, while Shannon Shaffer, MS, is a former employee. Sujata P. Sarda, PhD, Scott B. Baver, PhD, Mohamed Hamdani, and Michael Yeh, MD, MBA, MPH, are current employees and equity holders of Apellis Pharmaceuticals, Inc. David Cella, PhD reports consulting honoraria from Evidera and Apellis Pharmaceuticals, Inc., and ownership and role as President at FACIT.org.

Non-infectious Early Complications

O138. Association of Pre-Existing Comorbidities With Outcome of Allogeneic Stem Cell Transplantation. A Retrospective Analysis From The EBMT

Olaf Penack¹, Christophe Peczynski², Mohamad Mohty³, Ibrahim Yakoub-Agha⁴, Rafael de la Camara⁵, Bertram Glass⁶, Rafael Duarte⁷, Nicolaus Kröger⁸, Hélène Schoemans⁹, Christian Koenecke¹⁰, Zinaida Peric¹¹, Grzegorz Basak¹²

¹ Charité Universitätsmedizin Berlin, Berlin, Germany, ² EBMT Paris Study Office, Paris, France, ³ Hôpital Saint-Antoine, Paris, France, ⁴ Univ Lille, Inserm, CHU, Lille, France, ⁵ Hospital de la Princesa, Madrid, Spain, ⁶ Helios Klinikum Berlin-Buch, Berlin, Germany, ⁷ Hospital Universitario Puerta de Hierro Majadahonda, Madrid, Spain, ⁸ University Hospital Eppendorf, Hamburg, Germany, ⁹ University Hospitals Leuven and KU Leuven, Leuven, Belgium, ¹⁰ Medizinische Hochschule, Hannover, Germany, ¹¹ University Hospital Centre, Zagreb, Croatia, ¹² Medical University of Warsaw, Warsaw, Poland

Background: The presence of comorbidities is often used for clinical decision making on the individual risk for mortality after allogeneic stem cell transplantation (alloSCT). However, a comprehensive analysis of the impact of comorbidities on alloSCT outcome in large current datasets has not been published. To improve evidence-based assessment the EBMT started to document pre-existing comorbidities in 2010.

Methods: Here we present the first comprehensive EBMT database analysis on the association of comorbidities with outcome. We included adult alloSCT recipients with hematologic malignancies transplanted between 2010 and 2018 from matched sibling and matched unrelated donors. We performed multivariate analyses using the Cox proportional-hazards model including known risk factors for non-relapse mortality (NRM).

Results: We identified 38.759 alloSCT recipients with a full data set of pre-existing comorbidities. The prevalence of the specific comorbidities is given in Figure 1A. Pulmonary comorbidity, infections and cardiac comorbidity occurred most frequently. We found that pre-existing renal comorbidity, defined as creatinine serum values of ≥2mg/dl or previous renal transplant, had the strongest association with NRM (HR 1.96 [95% CI 1.64-2.34]. In addition, we found a significant association of multiple pre-existing comorbidities with NRM including diabetes, infections, hepatic comorbidity, cardiac comorbidity and pulmonary comorbidity. However, the HR of the association of these comorbidities with NRM was relatively low and did not exceed 1.32. A summary of the association of individual comorbidities with outcome is given in Figure 1B. Because of the rather moderate impact of these individual (non-renal) comorbidities we were interested on the performance of the hematopoietic cell transplantation comorbidity index (HCT-CI). We found that the risk of NRM was significantly but only moderately increased in patients with higher HCT-CI (HCT-CI 1-2 HR 1.14 [1.07-1.21] and HCT-CI ≥3 HR 1.35 [1.27-1.44]). Univariate NRM curves for the HCT-CI are shown in Figure 1C.

Conclusions: Under current treatment conditions severe renal comorbidity remains to be strongly associated with NRM. However, we found a weaker association of other comorbidities with NRM. When putting the present results in perspective with the previous data from North America, which was the basis for HCT-CI (Sorror, Blood. 2005;106(8):2912–2919), the impact of non-renal comorbidities on NRM has decreased sharply. As a result, the HCT-CI is not a strong predictor of NRM in the current EBMT population. Likely explanations for these favourable data are: 1) Improvements in medical practice may have diminished the negative impact of individual comorbidities on outcome; and 2) The use of HCT-CI and the improved awareness on the potentially fatal impact of comorbidities may have improved decision making on transplant indications.

Disclosure: Nothing to declare.

O139. Defibrotide Prophylaxis For Veno-occlusive Disease/Sinusoidal Obstruction Syndrome (VOD/SOS) After Haematopoietic Cell Transplantation (HCT): Analysis of A Multicentre, Multinational, Prospective, Observational Registry Study (EBMT Pass)

Mohamad Mohty¹, Jean Henri Bourhis², Maura Faraci³, Anna Paola Iori⁴, Anne Huynh⁵, Vian Amber⁶, Robert J Ryan⁷, Natalia Maximova⁸

¹ Hôpital St Antoine, Sorbonne University, INSERM UMRs 938, Paris, France, ² Institut Gustave Roussy, Villejuif, France, ³ Istituto Giannina Gaslini, Genova, Italy, ⁴ Policlinico Umberto I, Rome, Italy, ⁵ Institut Universitaire du Cancer de Toulouse-Oncopole, Toulouse, France, ⁶ Jazz Pharmaceuticals, Oxford, United Kingdom, ⁷ Jazz Pharmaceuticals, Philadelphia, PA, United States, ⁸ Institute for Maternal and Child Health – IRCCS Burlo Garofolo, Trieste, Italy

Background: VOD/SOS is a potentially life-threatening complication of HCT. Defibrotide is approved for treatment of patients aged >1 month with severe hepatic VOD/SOS post-HCT in the EU. However, several previous studies (including a phase 3 paediatric study [Corbacioglu, et al. Lancet. 2012]) have reported a benefit of defibrotide for VOD/SOS prophylaxis post-HCT.

Methods: This registry study (EBMT PASS; NCT03032016), performed by the EBMT, enrolled defibrotide-treated patients, including those receiving defibrotide for VOD/SOS prophylaxis, from April 2015 to July 2018. Defibrotide was administered prophylactically based on investigator assessment of patients’ risk for VOD/SOS post-HCT. Clinical characteristics and outcomes in patients receiving defibrotide prophylaxis and treatment were analysed. Investigators diagnosed VOD/SOS using classical/standard criteria and graded severity based on clinical expertise. This subgroup analysis evaluated outcomes in patients receiving defibrotide prophylaxis for post-HCT VOD/SOS.

Results: Defibrotide prophylaxis was received by 76 patients (42% were aged <18 years). Defibrotide was administered prophylactically a median of 7 (range: 0, 20) days on/before HCT and 19.5 (range: 0, 62) days post-HCT for prophylaxis and treatment. Nine (12%) patients receiving prophylaxis developed VOD/SOS (severe, n = 4; non-severe, n = 5); median time from defibrotide start to VOD/SOS diagnosis in these patients was 20 (range: 16, 113) days and median duration of defibrotide treatment after diagnosis was 20 days (range: 8, 55). All patients had multiple VOD/SOS risk factors. AML was the primary disease in 42% of patients. Overall, 95% of patients received allogeneic HCT (68% from unrelated/mismatched donors); 59%, 36%, and 5% of patients were undergoing their first, second, or third HCT, respectively. Myeloablative conditioning regimens were administered to 50% of patients, and 24% had prior gemtuzumab ozogamicin exposure. Busulphan-, thiotepa-, and treosulfan-containing regimens were administered to 63%, 51%, and 12% of patients, respectively. Considering patients who developed VOD/SOS, 33% were adults, and the majority (78%) had received myeloablative conditioning; 56%, 44%, and 33% of these patients had received busulphan-, thiotepa-, and treosulfan-containing regimens, respectively, and 33% had received gemtuzumab ozogamicin. All patients with VOD/SOS had received cells from mismatched/unrelated donors. The Kaplan-Meier–estimated post-HCT survival rates in patients who did and did not develop VOD/SOS are shown in the Figure; survival rates were >80% in both subgroups at Day 100. VOD/SOS resolved in all affected patients. Death occurred in 44% of patients with VOD/SOS: 1 due to disease progression and 3 due to HCT-related causes; no deaths were attributed to VOD/SOS. Serious adverse events of interest occurred in 28% of patients receiving defibrotide prophylaxis (infection: 22%; bleeding: 5%; hypotension: 1%).

Conclusions: The EBMT PASS included adult and paediatric high-risk patients who received prophylactic defibrotide. VOD/SOS resolved in all cases and no VOD/SOS-related deaths occurred. The safety of defibrotide prophylaxis in the real-world setting was consistent with the known defibrotide safety profile. These results should be interpreted in light of the previously positive phase 3 paediatric randomised study (Corbacioglu, et al. Lancet. 2012), and the ongoing phase 3 defibrotide prophylaxis study (NCT02851407), which recently stopped enrolment after meeting the protocol-defined futility criteria (awaiting analysis and publication of final results).

Clinical Trial Registry: ClinicalTrials.gov; NCT03032016

Disclosure: Mohamad Mohty: has received honoraria and research funding from Jazz Pharmaceuticals.

Jean Henri Bourhis: has no conflicts to disclose.

Maura Faraci: has no conflicts to disclose.

Anna Paola Iori: has no conflicts to disclose.

Anne Huynh: has received honoraria from Jazz Pharmaceuticals.

Vian Amber: employee of Jazz Pharmaceuticals and holds stock and/or stock options in Jazz Pharmaceuticals.

Robert J. Ryan: employee of Jazz Pharmaceuticals and holds stock and/or stock options in Jazz Pharmaceuticals.

Natalia Maximova: has no conflicts to disclose.

O140. EBMT Pass: Outcomes of Defibrotide-Treated Patients With Anicteric/Icteric Veno-Occlusive Disease/Sinusoidal Obstruction Syndrome (VOD/SOS) After Haematopoietic Cell Transplantation (HCT)

Mohamad Mohty¹, Thomas Cluzeau², Charlotte Jubert³, Sarah Lawson⁴, Robert J. Ryan⁵, Raj Hanvesakul⁶, Katie Perruccio⁷

¹ Hôpital St Antoine, Sorbonne University, INSERM UMRs 938, Paris, France, ² Hôpital de L’Archet, Nice, France, ³ CHU Bordeaux Groupe Hospitalier Pellegrin-Enfants, Bordeaux, France, ⁴ Birmingham Children’s Hospital, Birmingham, United Kingdom, ⁵ Jazz Pharmaceuticals, Biostatistics, Philadelphia, PA, United States, ⁶ Jazz Pharmaceuticals, Research and Development, Oxford, United Kingdom, ⁷ Santa Maria della Misericordia Hospital, Mother and Child Health, Perugia, Italy

Background: Hepatic VOD/SOS is a potentially life-threatening complication of HCT. Hyperbilirubinaemia (bilirubin >2 mg/dL) is a key criterion for VOD/SOS diagnosis, although recent guidelines acknowledge that VOD/SOS may occur without elevated bilirubin. In an expanded-access study, 23% of patients had anicteric (bilirubin ≤2 mg/dL) VOD/SOS at diagnosis; better outcomes were observed in these patients versus those with hyperbilirubinaemia at diagnosis. This post hoc analysis of data from a post-approval registry study evaluated outcomes in patients who received defibrotide for treatment of severe/non-severe VOD/SOS post-HCT by bilirubin level at VOD/SOS diagnosis.

Methods: This multicentre, multinational, prospective, observational study (NCT03032016; EBMT PASS), performed by the EBMT, enrolled defibrotide-treated patients from April 2015 to July 2018. Investigators diagnosed VOD/SOS using classical/standard criteria (including, but not limited to, hyperbilirubinemia, hepatomegaly, ascites, and weight gain >5%). Severity grading criteria were not predefined in the protocol; investigators graded VOD/SOS severity based on clinical expertise.

Results: Overall, 104 defibrotide-treated patients with VOD/SOS post-HCT were included in the analysis (62 with severe VOD/SOS [16% with bilirubin ≤2 mg/dL] and 42 with non-severe VOD/SOS [48% with bilirubin ≤2 mg/dL]). Demographics and baseline characteristics were similar across subgroups except that patients with bilirubin ≤2 mg/dL at diagnosis (n = 30) tended to be younger (80% and 70% aged <18 years for severe and non-severe VOD/SOS, respectively) than those with bilirubin >2 mg/dL at diagnosis (n = 74; 50% aged <18 years for both severe and non-severe VOD/SOS). Among patients with severe VOD/SOS, multi-organ dysfunction/failure was present in 63% of patients with bilirubin >2 mg/dL versus 30% of those with bilirubin ≤2 mg/dL; 9 of 10 patients with bilirubin ≤2 mg/dL at diagnosis developed VOD/SOS by Day 21.

Among patients with severe VOD/SOS, the Kaplan-Meier–estimated survival rates at Day 100 post-HCT were lower in patients with bilirubin >2 mg/dL than those with bilirubin ≤2 mg/dL at diagnosis; Day 100 post-HCT survival was also lower in those with bilirubin >2 mg/dL versus ≤2 mg/dL among patients with non-severe disease (Figure [A]). VOD/SOS resolution rates followed a similar pattern (Figure [B]).

Among patients with severe VOD/SOS, serious adverse events (SAEs) of interest occurred in 33% of patients with bilirubin >2 mg/dL and 30% of those with bilirubin ≤2 mg/dL; corresponding values in patients with non-severe disease were 23% and 20%. The most common categories of SAEs were bleeding events (severe VOD/SOS: 13% and 10% with bilirubin >2 mg/dL and ≤2 mg/dL, respectively; non-severe VOD/SOS: 9% and 10%, respectively) and infections (severe VOD/SOS: 23% and 30% with bilirubin >2 mg/dL and ≤2 mg/dL, respectively; non-severe VOD/SOS: 18% and 10%, respectively). Individual SAEs of interest were more frequent in patients with severe VOD/SOS and bilirubin >2 mg/dL at diagnosis than in other subgroups.

Conclusions: Consistent with results from a large, expanded-access study, these data suggest that treating with defibrotide before the onset of hyperbilirubinaemia may lead to better outcomes. These findings highlight the importance of prompt diagnosis using criteria that recognise VOD/SOS without hyperbilirubinaemia. The safety profile of defibrotide in the real-world setting was consistent with previous studies.

Disclosure: Mohamad Mohty: has received honoraria and research funding from Jazz Pharmaceuticals.

Thomas Cluzeau: has received honoraria from Jazz Pharmaceuticals.

Charlotte Jubert: has received funding from Jazz Pharmaceuticals.

Sarah Lawson: has received honoraria from Jazz Pharmaceuticals.

Robert J. Ryan: employee of Jazz Pharmaceuticals and holds stock and/or stock options in Jazz Pharmaceuticals.

Raj Hanvesakul: employee of Jazz Pharmaceuticals and holds stock and/or stock options in Jazz Pharmaceuticals.

Katie Perruccio: no conflicts of interest to disclose.

O141. Easix Score Predicts Overall Survival, Non-Relapse Mortality And The Risk of Intensive Care Unit Admission At Pre-Transplant Evaluation

Marta Peña Domingo¹, Maria Queralt Salas Gay¹, Alberto Mussetti¹, Gabriel Moreno Gonzalez², Anna Bosch Vilaseca¹, Beatriz Patiño Gutierrez¹, Laura Jimenez Prat¹, Meriem Kara¹, Rocío Parody Porras¹, Anna Sureda Balari¹

¹ Institut Català d’Oncologia-Hospitalet, L’Hospitalet de Llobregat, Spain, ² Bellvitge University Hospital, L’Hospitalet de Llobregat, Spain

Background: The Endothelial Activation and Stress Index (EASIX), defined as (LDH*creatinine)/platelets and evaluated prior to alloHCT (EASIX-PRE), has been proposed as a predictor of overall survival (OS) and non-relapse mortality (NRM) of the procedure. We present additional results on the predictive power of EASIX-PRE for OS and NRM in a retrospective cohort of alloHCT patients and explore the score’s ability to predict intensive care unit (ICU) admission.

Methods: Between January 2015 and June 2020, 167 adults aged between 17 and 65 years underwent alloHCT at our institution. All consecutive patients were included in the study and data was collected retrospectively. EASIX-PRE was calculated based on results obtained from blood analyses performed between day -30 and -7 prior to alloHCT. The index variable was transformed into log2 values (log2-EASIX-PRE). Patients were stratified into risk groups according to quartile values of the log2-EASIX-PRE. An optimal cut-off value based on the binary partitioning method was proposed to separate the cohort into two groups per survival curves. Main outcome variables were OS, NRM and the cumulative incidence of ICU admission. The predictive power of log2-EASIX-PRE for OS, NRM and risk for ICU admission was explored using a univariate and multivariate analysis.

Results: Baseline characteristics are summarized in Figure 1. Overall, the median age was 53 years. One hundred and twenty-one (77.2%) patients underwent reduced-intensity conditioning alloHCT and 64 (38.3%) received haploidentical donor grafts. With a median follow-up of 21.1 months among survivors, 2-year OS and NRM were 55.4% and 28.7%, respectively, while day +180 cumulative incidence of ICU admission was 18.6%.

Figure 1 shows the association between log2-EASIX-PRE and post-alloHCT outcomes. In the univariate analysis, log2-EASIX-PRE predicted OS (HR 1.2 [95% CI, 1.04-1.38], p = 0.011), NRM (HR 1.29 [95% CI, 1.08-1.53], p = 0.004) and the probability of ICU admission (HR 1.41 [95% CI, 1.2-1.66], p < 0.001). The most discriminating log2-EASIX-PRE cut-off value for OS and NRM was the median (0.102), while the 75^th percentile (1.483) was for the cumulative incidence of ICU admission. Patients with a log2-EASIX-PRE >0.102 had lower OS (2-years: 57.7% vs 68.7%, p = 0.009) and higher NRM (2-years: 38.7% vs 18.5%, p = 0.018) in comparison with patients with a log2-EASIX-PRE ≤0.102. Additionally, patients with a log2-EASIX-PRE >75^th percentile had a higher probability of ICU admission during post-transplant follow-up (day +180: 35.8% vs 12.8%, p = 0.010). In the multivariate analysis, log2-EASIX-PRE >0.102 had a negative impact in terms of OS (HR 1.70 [95% CI, 1.02-2.84], p = 0.040), NRM (HR 2.33 [95% CI, 1.21-4.48] p = 0.011) and cumulative incidence of ICU admission (HR 2.31 [95% CI, 1.23-4.36], p < 0.001).

Conclusions: The EASIX-PRE has been validated as a predictor of OS and NRM in a large, consecutive, retrospective cohort of alloHCT patients. Higher EASIX-PRE at pre-transplant evaluation is associated with a more elevated probability of ICU admission in alloHCT recipients. Therefore, EASIX-PRE may guide the design of personalized interventions for those patients at higher risk of severe events and ICU admission.

Disclosure: The authors declare no relevant conflicts of interest, financial or otherwise.

O142. The Incidence of Severe Oral Mucositis And Taste Disturbance in Patients Undergoing Different Conditioning Regimens in Hematopoietic Stem Cell Transplantation

Midori Nakagaki^1,2, Glen Kennedy^1,2, Nicole Gavin^1,3, Alexandra Clavarino², Karen Whitfield^1,2

¹ Royal Brisbane and Women’s Hospital, Brisbane, Australia, ² The University of Queensland, Brisbane, Australia, ³ Queensland University of Technology, Brisbane, Australia

Background: Oral mucositis and taste disturbance are common complications during hematopoietic stem cell transplantation (HSCT). This study aimed to review the incidence of severe mucositis and taste disturbance in patients undergoing different conditioning regimens used in matched allogeneic, haploidentical and autologous transplantation.

Methods: This was a single center retrospective study, reviewing daily oral assessment on all HSCT performed between January 2017 and June 2020. Oral care and cryotherapy with melphalan were used. Oral mucositis was routinely assessed by nursing staff and graded according to the WHO scale. In addition, taste disturbance, use of total parenteral nutrition (TPN) and patient controlled analgesia (PCA) were collected.

Results: Oral assessments for 467 patients were reviewed. The regimens, graft, patient number (N), melphalan (Mel) dose, total cyclophosphamide (Cy), fludarabine (Flu), and methotrexate (MTX) doses, and the incidence of grade 3-4 (G3-4) oral mucositis are summarized in the table. G3-4 oral mucositis was common in myeloablative TBI based regimens (CyTBI and FluTBI (12Gy) with post-transplant Cy (PTCy)) as well as reduced intensity matched allogeneic protocols (FluMel) and BEAM autologous HSCT. G3-4 oral mucositis was less commonly experienced in reduced intensity haploidentical regimens (MelFluTBI with PTCy), all non-myeloablative regimens (FluCy, FluTBI(2Gy) and FluCyTBI with PTCy) and high dose melphalan (HDM) autologous HSCT. Similarly, TPN and PCA use were common in regimens: CyTBI TPN 67%, PCA 75%; FluTBI (12Gy) with PTCy TPN 75%, PCA 69%; FluMel TPN 42%, PCA 39%; and BEAM TPN18%, PCA 41%.

Table 19

From the data with FluCy, FluCyTBI with PTCy and FluTBI(2Gy), fludarabine, high dose cyclophosphamide and low dose TBI (2Gy) are unlikely to cause severe oral mucositis. Comparing similar regimens between matched allogeneic and haploidentical regimens (CyTBI vs FluTBI (12Gy) with PTCy, FluMel vs MelFluTBI with PTCy), haploidentical regimens have less risk of severe oral mucositis due to PTCy compared to MTX. Comparing regimens with or without myeloablative TBI (CyTBI vs FluCy, FluTBI(12Gy) with PTCy vs FluCyTBI with PTCy), the myeloablative TBI appears to be the largest risk factor. Melphalan is a moderate risk factor on its own, however the risk increased when it was combined with other high-risk chemotherapy (e.g. MTX, etoposide). Taste disturbance was common regardless of the conditioning regimens, and overall incidence was 89% (range 71-95%).

Conclusions: Severe oral mucositis was associated with myeloablative TBI, MTX and melphalan in combination with MTX and in BEAM. Use of PTCy was preferable over MTX to prevent oral mucositis. Taste disturbance was common with all regimens.

Disclosure: Midori Nakagaki receives scholarship from the RBWH Foundation.

O143. Impact of Sirolimus/Tacrolimus-Based Graft-Versus-Host-Disease Prophylaxis on The Development of Early And Late Transplant-Associated Thrombotic Microangiopathy

Aldo A Acosta-Medina¹, Meera Sridharan¹, Mithun V Shah¹, Hassan B Alkhateeb¹

¹ Mayo Clinic, Rochester, United States

Background: The use of sirolimus/tacrolimus-based (SIR/TAC) graft-versus-host disease (GVHD) prophylaxis has been associated with an increased incidence of transplant-associated thrombotic microangiopathy (TMA) after allogeneic hematopoietic stem cell transplantation (alloHSCT). Impact of its use with recent recognition of early vs. late TMA phenotypes is unknown.

Methods: A retrospective review of patients undergoing alloHSCT from 2005 to 2017 was performed. Determination of the impact of SIR/TAC on the overall incidence of TMA, early TMA (<100 days post HSCT), and late TMA (≥100 days post HSCT) was carried out through the comparison of cumulative incidence functions with death as a competing risk. Analysis of impact of TMA on overall survival (OS) was performed by the construction of survival curves via the Kaplan-Meier method with differences between groups analyzed using the log-rank test.

Results: A total of 975 patients were included in the study. Baseline diagnoses included acute leukemia in 52.6% (n = 513), myelodysplastic syndrome/myeloproliferative neoplasm in 24.4% (n = 238), and lymphoma other than small lymphocytic lymphoma in 3.9% (n = 38), amongst others. In this cohort, 20 patients (2.1%) underwent GVHD prophylaxis with SIR/TAC, 2 of which received the combination after intolerance to other regimens. Most common baseline diagnosis in this subgroup was lymphoma (n = 11). There were no significant differences in terms of age, disease risk index, or conditioning intensity in patients receiving SIR/TAC but a significant increase in the use of peripheral blood as graft source (100% vs. 78.3%; p = 0.019) and matched related donors (90% vs. 46.9%; p = 0.002) were observed in this group.Throughout the study period, a total of 61 patients (6.3%) developed TMA, 29 cases (47.5%) were classified as early TMA and the remaining 32 (52.5%) as late TMA. Overall, the use of SIR/TAC was associated with an increased incidence of post-alloHSCT TMA (p = 0.0004). This association was maintained for cases developing early TMA (p = 0.0012) but was not observed for cases of late TMA (p = 0.097). Median OS for the population was 62 months (95%CI 49-84 months). The development of TMA was associated with a decrease in patient OS (p < 0.001). Within patients developing TMA, use of SIR/TAC was not associated with a worsened prognosis (p = 0.5).

Conclusions: Thrombotic microangiopathy is a rare complication in the post-HSCT setting. Nevertheless, its presence is associated with worsened survival outcomes. Concomitant use of sirolimus and tacrolimus as GVHD prophylaxis is associated with an increased incidence of TMA; however, this is primarily dependent on an increase of early TMA, rather than late TMA.

Disclosure: Nothing to declare.

O144. Risk Factors For Sinusoidal Obstruction Syndrome of The Liver After Allogeneic Hematopoietic Stem Cell Transplantation In Childhood

Bernd Gruhn¹, Jana Ernst¹, Grit Brodt¹, Jaspar Kloehn¹

¹ Jena University Hospital, Jena, Germany

Background: Sinusoidal obstruction syndrome (SOS) of the liver remains a serious complication after hematopoietic stem cell transplantation (HSCT). However, the cause of SOS is still not fully understood and the mortality remains high, especially for SOS leading to multi-organ failure with a mortality rate up to 84%. The aim of our study was to analyze several risk factors of SOS in children undergoing allogeneic HSCT. In addition, we investigated new potential risk factors.

Methods: We analyzed 105 children who underwent allogeneic HSCT for the first time and did not receive a defibrotide prophylaxis. The median age was 8.6 years and stem cell source was either bone marrow (n = 74) or peripheral blood (n = 31). All patients received a myeloablative conditioning regimen. We analyzed the transplantation-related factors graft source, donor-recipient human leukocyte antigen match, donor age, donor sex and conditioning regimen based on busulfan or total body irradiation. Furthermore, we investigated the patient-related factors patient age, patient sex, prior treatment with gemtuzumab ozogamicin as well as the following laboratory parameters: aspartate transaminase, alanine transaminase, cholinesterase, glutamyl transpeptidase, lactate dehydrogenase, alkaline phosphatase, ferritin, albumin, total bilirubin, C-reactive protein and international normalized ratio (INR). All laboratory parameters were measured before HSCT and cutoffs were determined by reference values and receiver operating characteristic (ROC) curves.

Results: SOS occurred in 15 out of 105 transplantations (14.3%). The median time of SOS onset was 12 days after HSCT. Three patients died of multi-organ failure following SOS (20%). This mortality rate was very low compared to other studies because our patients were treated with defibrotide immediately after being diagnosed with SOS. In univariate analysis, we found a significant association between patient age <1 year and SOS (Odds Ratio (OR) = 7.25, P = 0.037). Furthermore, a prior treatment with gemtuzumab ozogamicin (OR = 11.00, P = 0.020) showed a significant correlation. Patients who developed SOS had a significantly higher median ferritin level (2816.9 ng/mL vs. 1554.0 ng/mL, P = 0.026). Based on this observation, different ferritin cutoffs were selected by ROC analysis. Ferritin >1500 ng/mL (OR = 4.00, P = 0.033), ferritin >2000 ng/mL (OR = 4.69, P = 0.016) as well as ferritin >2400 ng/mL (OR = 5.29, P = 0.005) revealed significant P values. Besides these results, INR ≥1.3 (OR = 5.91, P = 0.009) was significantly associated with SOS. In multivariate analysis, the following variables showed P values less than .05: treatment with gemtuzumab ozogamicin (OR = 9.24, P = 0.048), ferritin >2400 ng/mL (OR = 5.74, P = 0.023) and INR ≥1.3 (OR = 8.02, P = 0.007).

Conclusions: Our data confirm the risk factors of young patient age (<1 year), prior treatment with gemtuzumab ozogamicin and high serum ferritin (>2400 ng/mL) in the pediatric population. Moreover, we report for the first time that there is a significant association between high INR (≥1.3) before HSCT and the occurrence of SOS. Especially this new finding could improve the risk stratification of SOS and should be evaluated in further trails.

Disclosure: Nothing to declare.

O145. Neurologic Complications After Allogeneic Stem Cell Transplantation: Analysis of Risk Factors. A Single Center Study Over 20 Years of Follow-Up

Elisa Sala¹, Verena Wais¹, Maral Saadati², Verena Gaidzik¹, Mai Thanh Baumhardt¹, Adela Neagoie¹, Stephanie von Harsdorf¹, Florian Kuchenbauer¹, Axel Benner², Hartmut Döhner¹, Donald Werner Bunjes¹

¹ University Hospital of Ulm, Ulm, Germany, ² German Cancer Research Center (DKFZ), Heidelberg, Germany

Background: The outcome of allogeneic hematopoietic Stem Cell Transplantation (allo-HSCT) is still affected by a considerable transplant-related mortality (TRM). The occurrence of neurological complications (NC), especially those of the central nervous system (CNS), is associated with a significant rate of morbidity and mortality. Calcineurin inhibitors (CNI), like Cyclosporine A (CsA) and Tacrolimus (TAC), represent the most commonly administered immunosuppression for the prophylaxis of Graft versus Host Disease (GvHD) and are associated with an increased risk of neurotoxicity. We analyzed the NC arising after allo-HSCT, in order to identify potential avoidable risk factors, with a special focus on the role of CNIs (CsA versus TAC).

Methods: We performed a retrospective analysis of CNS-complications in a cohort of 739 patients receiving allo-HSCT with a CNI-based GvHD–prophylaxis over a period of 20 years (12/1999- 04/2019) at the Bone Marrow Transplantation Unit at the University Hospital of Ulm (Germany). Patients developing NC during follow-up underwent neurological assessment, neuroimaging (CT-scan or MRI), electroencephalography and lumbar puncture, including microbiological analysis. TRM and OS were calculated using the Kaplan–Meier method. A multivariate analysis using COX regression hazard model was used in order to identify potential risk factors for the development of NC after transplant.

Results: The incidence of CNS-complications in our population was 17%, with 129 out of 739 patients developing a total of 152 NC. The most frequently detected CNS-complications were the CNI-related ones (41%), ranging from headache, confusion and hallucinations to seizures, posterior reversible encephalopathy syndrome (PRES) and transplant associated thrombotic microangiopathy (TA-TMA), followed by infections (31%), vascular events (13%) and metabolic encephalopathy (5%). The occurrence of NC after transplant proved to exert a negative impact on outcome in term of TRM (HR 7.22, p < 0.001) and OS (HR 5.37, p < 0.001). According to multivariate analysis, risk factors for the occurrence of CNS-complications were: advanced age at the time of transplant (p = 0.002), the use of TBI in the conditioning regimen (p < 0.001), the development of severe acute GvHD (grade 3-4, p < 0.001) or chronic GvHD (any severity grade, p = 0.008) and the use of TAC as GvHD-prophylaxis (p = 0.007). This higher risk of NC in the TAC cohort was almost entirely due to a higher incidence of severe TA–TMA. Comparing the outcome of patients receiving TAC versus CsA we observed that the increased risk for neurotoxcity didn’t affect the overall outcome, since in multivariate analysis the use of TAC improved TRM (HR 0.75, p = 0.053) and ultimately OS (HR 0.76, p = 0.024).

Conclusions: Our data suggest a possible increase of NC for patients undergoing GvHD-prophylaxis with TAC, especially considering the development of severe TA-TMA. Nevertheless TAC proved to exert a positive influence on OS and TRM. Although these findings need to be further validated in larger prospective and randomized studies, they could be of potential clinical relevance, suggesting the need for a balance between the risk of developing NC (e.g. avoiding TAC in high risk patients) and the potential survival benefits coming from the immunosuppressive action of TAC (e.g. considering a stricter TA-TMA monitoring in this setting).

Disclosure: Nothing to declare.

O146. Outcome After Admission to Intensive Care Unit (ICU) in Hematopoietic Transplant Recipients: Single Centre Experience

Isabel Iturrate¹, Analys Ruiz¹, Carlota Mayor¹, Ana Puchol¹, Sofía Díaz¹, Ángela Figuera¹, Adrián Alegre¹, Beatriz Aguado¹

¹ University Hospital La Princesa, Madrid, Spain

Background: Hematopoietic stem cell transplantation (HSCT) can entail significant morbidity during the procedure1. Published series report a range of ICU admission rates from 24-40% of transplant patients2 and is associated with poor prognosis3. Most frequent reasons involve septic shock, respiratory failure and veno-occlusive disease. The aim of this study was to analyze outcome of HSCT patients requiring ICU admission in our center.

Methods: We retrospectively analyzed outcome of 685 patients who underwent HSCT in our centre from January/2008 to December/2019. Data were collected from patients’ clinical histories.

Results:

Median age (range)	53 (20 - 70)
Sex: male / female	53 (55.8%) / 42 (44.2%)
Underlying disease: - AML / ALL / MDS	34 (35.8%) / 12 (12.6%) / 15 (15.8%)
- Myeloma / NHL / HL / Other	7 (7.4%) / 10 (10.5%) / 6 (6.3%) / 11 (11.6%)
Type of HSCT:
- Allogeneic:	79 (83.2%)
Sibling donor / Unrelated donor / Haploidentical	34 (35.8%) / 34 (35.8%) / 11 (11.6%)
- Autologous	16 (16.8%)
Conditioning regimen: Myeloablative / RIC	70 (73.7%) / 25 (26.3%)
GVHD: Acute / Chronic	27 (28%) / 5 (5.2%)

Ninety-five (13.8%) patients required ICU admission (baseline and HSCT characteristics on table). Median time to ICU admission was 31 days (-2-1765). Six of these patients were admitted to ICU on two occasions giving a total of 101 consecutive ICU admissions available for analysis; median time to re-admission was 70 days (19-573).

Main reason for ICU admission was respiratory distress (52%), because of pneumonia (53%), pulmonary edema (39.5%) and pulmonary haemorrhage (7.5%). Septic shock was second most common cause for ICU admission (29%) [gram-negative bacilli (66.6%), idiopathic (20%), fungal (16.6%) and gram-positive (6.6%)]. Other less frequent causes were veno-occlusive disease (9%), intracranial haemorrhage (3%), cardiorespiratory arrest (2%), hepatic failure/encephalopathy (2%), GVHD (1%), cardiogenic shock (1%), adrenal insufficiency (1%). Of the 101 ICU admissions, 32 (31.6%) required haemodialysis (26 (81.3%) died) and 69 (68.3%) required mechanical ventilation (54 (78.3%) died).

Of the 95 patients, 62 (65%) died in the ICU; of these patients, 23 (37%) received unrelated donor HSCT, 21 (33.9%) sibling donor, 8 (13%) haploidentical and 10 (16.1%) autologous. Median ICU length of stay of these patients was 12 days (range 1-76). Cause of death was same reason for ICU admission. Fifteen (15.78%) patients were discharged from ICU and died prior to hospital discharge and 18 (18.94%) survived to hospital discharge and were classified as post-discharge survivors. Of these 18 patients, 11 (61.1%) remain alive with median follow-up of 6.5 years (range 1.08-11.1) while the others (7, 38.9%) succumbed to underlying disease or complications post-HSCT.

Conclusions: In our study, 13.86% of transplant recipients required ICU admission, slightly lower than previous reports. Most common cause of admission was respiratory failure, consistent with reported. Mortality rate during ICU admission was 65%; higher death rate observed in allogeneic transplantation and those requiring aggressive ICU treatments such as mechanical ventilation or haemodialysis. ICU admission following HSCT is associated with poor prognosis, but should not be considered futile.

Disclosure: The authors declare not to have any conflicts of interest.

O147. Distribution And Impact of Acute Kidney Injury in At-Home Allogeneic Bone Marrow Transplantation Models

Marta Garcia-Recio¹, Alejandra Martinez-Roca¹, Cristina Gallego¹, Pilar Ayora¹, María Suarez-Lledó¹, Carmen Martinez¹, Montserrat Rovira¹, Gonzalo Gutierrez¹, Alvaro Urbano-Ispizua¹, Francesc Fernandez-Avilés¹

¹ Hospital Clinic, Barcelona, Spain

Background: Acute kidney injury (AKI) is a common complication for patients undergoing bone marrow transplantation (BMT). Its incidence ranges between 10-73% depending on how AKI is defined, the type of conditioning regimen used and the type of transplant (Renaghan et al. 2020). AKI has been associated with higher non-relapse mortality (Parikh et al. 2007). However, there is no data about the incidence and impact of AKI in patients undergoing at-home allogeneic BMT (allo-BMT).

Methods: Between August 2015 and October 2020, 55 consecutive adult patients underwent at-home allo-BMT. Clinical data were retrospectively collected from medical records. Renal dysfunction was assessed accordingly to the definition provided by the AKI Network (Mehta et al. 2007).

Results: Fifty-five patients were included in this study with a median follow-up of 1.66 years (0.18-5.25). Median age was 53 (23-70) and 27 (49.1%) received a myeloablative conditioning regimen. All patients received a graft versus host disease (GVHD) prophylaxis regimen that included a calcineurin inhibitor. Among patients with AKI (65.5%), 21 (78.6%) developed CMV reactivation and 11 (30.5%) required antiviral treatment. Among those who developed AKI, 12 (33.3%) presented with grade 1, 20 (55.6%) with grade 2 and 4 (11.1%) with grade 3. The majority of patients who received post-transplant cyclophosphamide (PTCy) as GVHD prophylaxis, developed AKI (77.2%). Although not statistically significant in our study (probably due to the small sample size and short follow-up), OS was inferior among those patients who developed AKI (Table 1).

Table Acute kidney injury in patients undergoing at-home allogeneic bone marrow transplantation.

Conclusions: AKI is a significant complication for patients undergoing at-home allo-BMT with impact on survival rates. Special consideration and further investigations of the increased incidence of AKI in patients receiving PTCy are needed. Patients undergoing at-home allo-BMT presented a similar incidence of AKI when compared with previous reports in hospitalized programs.

Disclosure: Authors declare no conflict of interest.

Non-infectious Late Effects, Quality of Life and Fertility

O148. Abstract already published

O149. One Thousand Voices Address Patient Engagement in Hematopoietic Stem Cell Transplantation And Cell Therapy: Results of A Survey From The EBMT

Hélène Schoemans^1,2,3,4, Bregje Verhoeven^5,4, Natacha Bolaños^6,4, Chris Lewis^7,4, Sarah Jayne Liptrott^8,4, Isabelle Scholl^9,4, Anna Barata^10,4, Paul Johnson^11,4, Deborah Buk^12,4, Dirk Pretzel^13,4, Anja Van Biezen^14,4, Rachel Phelan¹⁵, David Buchbinder¹⁶, Betty Hamilton¹⁷, Christian Koenecke^18,2, Olaf Penack^19,2, Zinaida Peric^20,2, Grzegorz Basak^21,2, Michelle Kenyon^22,4, Marianne Mol^14,4, Emmanuelle Polge^23,2,4, William Boreland^23,4, Christophe Peczynski^23,2,4, Andreu Gusi^24,4, John Snowden^25,4, Bronwen Shaw^15,4, Sabina De Geest^26,3,4, Linda Burns^27,4, Nicolaus Kroeger^28,4

¹ University Hospitals Leuven, Leuven, Belgium, ² EBMT - Transplantation Complications Working Party, Paris, France, ³ Academic Center for Nursing and Midwifery, KU Leuven, Leuven, Belgium, ⁴ EBMT - Patient Engagement Task Force, Barcelona, Spain, ⁵ EBMT - Patient Advocacy Committee, Barcelona, Spain, ⁶ Lymphoma Coalition - Global Alliance Manager - Europe, Mississauga, Canada, ⁷ Chris’s Cancer Community, Online, United Kingdom, ⁸ European Institute of Oncology IRCCS, Milan, Italy, ⁹ University Medical Center Hamburg-Eppendorf, Hamburg, Germany, ¹⁰ Moffitt Cancer Center, Tampa, FL, United States, ¹¹ Anthony Nolan, London, United Kingdom, ¹² DKMS gGmbH, Tuebingen, Germany, ¹³ Caregiver Support Group leader, Hamburg, Germany, ¹⁴ EBMT - Data Office Leiden, Leiden, Netherlands, ¹⁵ Medical College of Wisconsin & Center for International Blood and Marrow Transplant Research, Milwaukee, United States, ¹⁶ Children’s Hospital of Orange County, Orange, United States, ¹⁷ Cleveland Clinic, Cleveland, United States, ¹⁸ Hannover Medical School, Hannover, Germany, ¹⁹ Charité Universitätsmedizin, Berlin, Germany, ²⁰ University Hospital Centre Zagreb, Zagreb, Croatia, ²¹ Medical University of Warsaw, Warsaw, Poland, ²² King’s College Hospital NHS Foundation Trust, London, United Kingdom, ²³ EBMT - Study Office Paris, Paris, France, ²⁴ EBMT - Executive Office, Barcelona, Spain, ²⁵ Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, United Kingdom, ²⁶ Institute of Nursing Science, University of Basel, Basel, Switzerland, ²⁷ Center for International Blood and Marrow Transplant Research, Milwaukee, United States, ²⁸ University Medical Center Hamburg, Hamburg, Germany

Background: In line with its mission, vision and values, the EBMT aims to connect patients, the scientific community and other stakeholders. We therefore undertook a survey to understand the different perceptions regarding patient engagement.

Methods: An online multilingual survey was developed with input from patients, caregivers, donors, clinicians and EBMT staff to identify the motivational and implementation issues impacting on patient engagement, defined as Patient Reported Outcomes (PRO) and Patient Active Involvement in Research (PAIR). The survey was disseminated to a total of 14 441 stakeholders between 15/06/2020 and 19/10/2020. Responses were analyzed using descriptive statistics for all respondents providing self-reported identification (figure 1A). The two major stakeholder groups, patients and health care professionals (HCPs), were compared using Chi-square or Fisher’s test and Mann-Whitney.

Results: 1094 stakeholders responded, with all target groups represented and 813 providing self-reported identification (Figure 1A). No demographic difference were noted between patients and HCPs, except for older age (median 55 years [IQR 56-65] and 47 years [IQR 38-56] respectively) (p < 0.05).

Interest for EBMT PRO collection was documented (n = 680, 83.6% across stakeholders groups; patients n = 258, 93.1% versus HCP n = 275, 78.4% (p < 0.05)). Stakeholders agreed that PRO collection should be done ‘to understand the patient’s quality of life’ (n = 388, 49.7%) ideally using digital methods (n = 637, 78.4%). Priorities in PRO data collection differed slightly between patients and HCPs (figure 1B). More patients (n = 190, 70.4%) than HCPs (n = 178, 53.6%) wished to collect PRO ‘at home’ compared to ‘in the hospital’ (p < 0.05). Barriers were patients’ lack of ‘interest’ (n = 311, 39.9%) or ‘energy’ (n = 289, 37.1%), followed by a concern for ‘lack of time of HCPs’ reported by patients (n = 45, 29.2%) and ‘data protection issues’ reported by HCPs (n = 104, 31.4%). The main motivator for PRO collection was to provide information about the evolution of one’s results over time (n = 296, 37.9%).

Enthusiasm for PAIR was also documented (n = 702, 86.3% across stakeholders groups; patients n = 256, 92.4% versus HCPs n = 296, 85.8% (p < 0.05)). The most important reason for PAIR was ‘to improve relevance of study results’ (n = 342, 45.5%), with patient involvement identified as crucial when deciding on priorities (n = 480, 61.1%), study design (n = 358, 45.5%) and implications for care (n = 358, 45.5%). The ideal research-partners were identified as ‘randomly selected patients’ (n = 661, 83.8%). Barriers included concerns for patients’ lack of ‘knowledge about the research process’ (n = 358, 45.4%) or ‘time/energy’ (n = 304, 38.5%) and a difficulty to ‘identify representative individuals’ (n = 270, 34.2%) across stakeholder groups, followed by a concern that ‘HCPs do not reach out’ for patients (n = 78, 28.7%) or ‘language issues’ for HCPs (n = 128, 37.9%). The main facilitator was the availability of financial support for PAIR (n = 276, 34.8%) across stakeholder groups.

Conclusions: There is clear openness for patient involvement in data collection and research, particularly within patient respondents compared with HCPs. Priorities for PROs include assessing symptom experience, psychosocial and cognitive functioning. Digital tools may facilitate collection of PROs in the home setting. Stakeholders offered different perceptions, suggesting an added value to embracing patient engagement in the development of meaningful research and service design within the transplantation and cellular therapy community.

Clinical Trial Registry: not applicable.

Disclosure: Between 2017 and 2020, Helene Schoemans has participated in advisory boards for Incyte, Janssen & Novartis. She has received speaker’s fees from Jazz Pharmaceuticals, Novartis, Incyte, Jazz Pharmaceuticals & Takeda. She has received travel grants from EBMT, Celgene, Abbvie, Incyte & Gilead. Research funding was also received from Novartis for an investigator initiated study. None of these are relevant to the current study.

Bronwen Shaw reports honoraria from Mallinkrodt and consulting for Orcabio.

O150. Chronic Kidney Disease Ten Years After Pediatric Allogeneic Hematopoietic Stem Cell Transplantation

Gertjan Lugthart^1,2, Carlijn CE Jordans¹, Anne PJ de Pagter¹, Cornelia M Jol-van der Zijde¹, Dorine Bresters³, Roos WG Van Rooij-Kouwenhoven¹, Ram N Sukhai¹, Marloes Louwerens¹, Eiske M Dorresteijn², Arjan C Lankester¹

¹ Leiden University Medical Center, Leiden, Netherlands, ² Erasmus Medical Center, Rotterdam, Netherlands, ³ Prinses Maxima Center, Utrecht, Netherlands

Background: Chronic kidney disease (CKD) is an important sequela of hematopoietic stem cell transplantation (HSCT), but data regarding CKD after pediatric HSCT are limited.

Methods: In this single center cohort study, we evaluated eGFR dynamics, proteinuria and hypertension in the first decade after HSCT and assessed risk factors for chronic kidney disease in 216 pediatric long term HSCT survivors, transplanted between 2002 and 2012.

Results: The eGFR decreased from median 148 to 116 ml/min/1.73m2 between pre-HSCT and ten years after HSCT. CKD, defined as an eGFR <90 ml/min/1.73m2 and/or proteinuria (KDIGO stage ≥G2 or ≥A2) occurred in 21% of patients. In multivariate analysis, hematological malignancy as HSCT indication (HR 5.5, 95% CI 1.2-25) and cytomegalovirus reactivation (HR 2.4, 95% CI 1.1-5.4) were independent risk factors for CKD. One third of patients with CKD had both an eGFR <90 ml/min/1.73m2 as well as proteinuria, one third had isolated eGFR reduction and one third only had proteinuria. Hypertension was observed in 27% of patients with CKD compared to 4.4% of patients without CKD. Tubular proteinuria was present in 7% of the subgroup of patients (n = 71) in which β2-microglobulinuria was measured.

Conclusions: In conclusion, a significant proportion of pediatric HSCT recipients developed chronic kidney disease within ten years after HSCT. Our data stress the importance of structural monitoring of eGFR, urine and blood pressure after HSCT to identify patients with beginning CKD who could benefit most from nephroprotective interventions.

Disclosure: Nothing to declare.

Paediatric Issues

O151.Treosulfan- Versus Busulfan-Based Myeloablative Conditioning For Allogeneic Hematopoietic Stem Cell Transplantation in Pediatric Acute Lymphoblastic Leukemia. A Report From The International All-SCTPED Forum Trial

Herbert Pichler¹, Jean-Hugues Dalle², Franco Locatelli³, Ulrike Poetschger⁴, Petr Sedlacek⁵, Jochen Buechner⁶, Peter J Shaw⁷, Raquel Staciuk⁸, Marianne Ifversen⁹, Kim Vettenranta¹⁰, Peter Svec¹¹, Olga Aleinikova¹², Jerry Stein¹³, Tayfun Güngör¹⁴, Jacek Toporski¹⁵, Tony H Truong¹⁶, Cristina Diaz-de-Heredia¹⁷, Marc Bierings¹⁸, Hany Ariffin¹⁹, Mohammed Essa²⁰, Birgit Burkhardt²¹, Kirk Schultz²², Roland Meisel²³, Arjan Lankester²⁴, Marc Ansari²⁵, Martin Schrappe²⁶, Arend von Stackelberg²⁷, Adriana Balduzzi²⁸, Selim Corbacioglu²⁹, Peter Bader³⁰, Christina Peters¹

¹ St. Anna Children’s Hospital, Children’s Cancer Research Institute, University Vienna, Vienna, Austria, ² Hôpital Robert Debré, GH APHP-Nord Université de Paris, Paris, France, ³ IRCCS Ospedale Pediatrico Bambino Gesù, Sapienza University of Rome, Rome, Italy, ⁴ Children’s Cancer Research Institute, Vienna, Austria, ⁵ Hospital Motol; HSCT-Unit, Prague, Czech Republic, ⁶ Oslo University Hospital, Oslo, Norway, ⁷ The Children`s Hospital at Westmead, Sydney, Australia, ⁸ Hospital de Pediatría ‘Prof. Dr. Juan P. Garrahan’, Buenos Aires, Argentina, ⁹ Copenhagen University Hospital Rigshospitalet, Copenhagen, Denmark, ¹⁰ Children’s Hospital, University of Helsinki, Helsinki, Finland, ¹¹ National Institute of Children’s Diseases, Bratislava, Slovakia, ¹² Belarusian Research Center for Pediatric Oncology, Hematology and Immunology, Minsk, Belarus, ¹³ Schneider Children’s Medical Center of Israel, Sackler Faculty of Medicine, Tel Aviv University, Petach-Tikva, Israel, ¹⁴ Universitäts-Kinderspital, Zurich, Switzerland, ¹⁵ Skåne University Hospital, Lund, Sweden, ¹⁶ Alberta Children’s Hospital Calgary, Alberta, Canada, ¹⁷ Hospital Universitari Vall d’Hebron, Barcelona, Spain, ¹⁸ Princess Máxima Center for Pediatric Oncology, Utrecht, Netherlands, ¹⁹ University of Malaya, Kuala Lumpur, Malaysia, ²⁰ King Abdullah Specialist Children’s Hospital, King Abdullah International Medical Research Center, King Saud Bin Abdulaziz University for Health Sciences, Riyadh, Saudi Arabia, ²¹ Children’s University Hospital Münster, Munster, Germany, ²² University of British Columbia, Vancouver, Canada, ²³ Medical Faculty, Heinrich-Heine-University, Duesseldorf, Germany, ²⁴ Willem-Alexander Children’s Hospital, Leiden, Netherlands, ²⁵ Geneva University Hospital, Geneva, Switzerland, ²⁶ Universitätsklinikum Schleswig-Holstein, Kiel, Germany, ²⁷ Charité University Hospital Berlin, Berlin, Germany, ²⁸ Università degli Studi di Milano – Fondazione MBBM, Monza, Italy, ²⁹ Universitätsklinikum Regensburg, Regensburg, Germany, ³⁰ Goethe University; University Hospital Frankfurt, Frankfurt am Main, Germany

Background: Treosulfan, a comparably novel drug with a favorable toxicity profile, has been successfully used for myeloablative conditioning regimens within the last decade.

Methods: Children and adolescents with B-cell precursor (BCP-) or T-ALL treated within the prospective international randomized multicenter ALL-SCTped FORUM trial were investigated for differences in outcome after allogeneic hematopoietic stem cell transplantation (HSCT) using a Busulfan- (BU) or Treosulfan- (TREO) based myeloablative conditioning.

Results: After randomization into the chemotherapy-arm, 135 patients with BCP-ALL and 42 with T-ALL received a myeloablative conditioning consisting of either BU or TREO in combination with Thiotepa and Fludarabine.

According to the immunophenotype, no significant differences in the estimates of 2 year overall- (2yOS) and event free survival (2yEFS) were observed for patients with BCP-ALL and T-ALL respectively, with 2yOS rates of 0.79±0.04 and 0.74±0.08 (p = n.s.) and 2yEFS rates of 0.74±0.08 and 0.71±0.07 (p = n.s.) for the complete chemotherapy cohort.

With regard to the conditioning, patients with BCP-ALL after BU (n = 79) had a similar outcome compared to TREO (n = 56), with 2yOS rates of 0.79±0.05 and 0.80±0.06 (p = n.s.) and 2yEFS of 0.64±0.06 and 0.55±0.07 (p = n.s.) respectively. Comparable results were seen in T-ALL patients after conditioning with BU (n = 16) and TREO (n = 26), with 2yOS rates of 0.67±0.15 and 0.76±0.09 (p = n.s.) and 2yEFS of 0.67±0.12 and 0.73±0.09 (p = n.s.) respectively.

The 2-year cumulative incidences of relapse (2yCIR) were similar after BU and TREO for patients with BCP-ALL (2yCIR: 0.30±0.06 and 0.31±0.07; p = n.s.) and T-ALL patients (2yCIR: 0.26±0.11 and 0.23±0.09; p = n.s.).

The rates of 2-year treatment-related mortality (2yTRM) according to the conditioning regimen did neither differ in BCP-ALL patients (BU versus TREO: 0.07±0.03 vs. 0.14±0.05; p = n.s.) nor in patients with T-ALL (BU versus TREO: 0.06±0.06 vs. 0.04±0.04; p = n.s.).

Conclusions: Survival after chemoconditioning for allogeneic HSCT in children with high-risk BCP- or T-ALL is excellent despite the recently demonstrated superiority of total-body-irradiation within the ALL-SCTped FORUM trial. A Treosulfan-based conditioning results in similar outcomes as compared to a Busulfan-based regimen with low rates of TRM especially for patients with T-ALL.

Clinical Trial Registry: EudraCT: 2012-003032-22; ClinicalTrials.Gov: NCT01949129

Disclosure: CP: Grants/personal fees: Amgen, Medac, Jazz Pharmaceuticals, Riemser, Novartis, Pfizer, and Sanofi.

JHD: Grants: Fédération Enfants et Santé, Association Hubert Gouin, Sanofi-Genzyme, Keocyt; personal fees: Sanofi-Genzyme, Medac.

FL: Advisory board/speaker bureau: Amgen, Novartis; Advisory board participation: Bellicum Pharmaceutical, Neovii; Speaker bureau participation: Miltenyi, Medac, Jazz Pharmaceuticals, Takeda.

HP: Grants: Neovii Biotech; grants and personal fees from Jazz Pharmaceuticals.

JB: Grants from Norwegian Childhood Cancer Foundation.

PJS: Past direct engagement with the companies Otsuka and Orphan Pharmaceuticals more than three years ago.

MI: Personal fees from Novartis.

THT: Consultancy fees and advisory board participation with Jazz Pharmaceuticals.

CDdH: Personal fees: Novartis, Jazz Pharmaceuticals, Sobi; non-financial support: Gilead, Miltenyi, Terumo and Alexion.

ME: Personal fees: Amgen, Novartis.

BB: Advisor for Roche; Advisory board participation: Celgene, Novartis; Steering committee: Janssen, Novartis.

MS: Grants: Shire, Jazz Pharmaceuticals, Servier, Novartis, Amgen, Sigma-Tau; Speaker’s honoraria: Shire, Servier, Jazz Pharmaceuticals.

AvS: Personal fees: Amgen, Jazz Pharmaceuticals, Shire, Morphosys, Novartis, Roche.

AB: Fees: Novartis, EusaPharma, and Amgen.

PB: Grants: Deutsche Knochenmarkspender Datei, Neovii, Riemser, Medac.

RM: Personal fees: Bluebird Bio, Bellicum Pharmaceuticals, Novartis; non-financial support: Jazz Pharmaceuticals.

UP, PSedlacek, RS, KV, PSvec, OA, JS, TG, JT, MB, HA, AL, MA, SC, and KS have nothing to disclose.

O152. Serotherapy-Free Regimen Improves Non-Relapse Mortality And Early Immune Recovery Among The Recipients of ΑΒ T Cell-Depleted Haploidentical Grafts: Retrospective Study In Childhood Leukemia

Larisa Shelikhova¹, Svetlana Glushkova¹, Ruslan Nikolaev¹, Maria Dunaikina ¹, Zhanna Zhekhovtsova¹, Sergey Blagov¹, Rimma Khismatullina¹, Dmitriy Balashov¹, Elena Kurnikova¹, Dmitriy Pershin¹, Yakov Muzalevskii¹, Alexei Kazachenok¹, Elena Osipova¹, Pavel Trakhtman¹, Alexei Maschan¹, Michael Maschan¹

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

Background: Depletion of ab T cells from the graft prevents graft-versus-host disease (GVHD) and improves the outcome of HSCT from haploidentical donors. Delayed recovery of adaptive immunity remains a problem, which can be addressed by adoptive T cell transfer. In a randomized trial, we have demonstrated that low-dose memory (CD45RA-depleted) donor lymphocytes (mDLI) after HSCT with ab T cell depletion are safe but do not prevent CMV viremia. Remarkably, non-relapse mortality was low at 2%, independent of study arm. Anti-thymocyte globulin (ATG) is viewed as an essential component of preparative regimen, critical for both prevention of graft failure and GVHD. Variable pharmacokinetics of ATG may significantly affect lymphocyte subpopulations after HSCT. In order to uncover the potential of mDLI we replaced rabbit ATG with tocilizumab and abatacept. Here we compare the immune recovery and clinical outcomes, between the cohort enrolled in the prospective trial and a historic cohort, comprised of patients grafted with an ATG-based HSCT with ab T cell depletion.

Methods: A cohort of 149 children were enrolled in the prospective trial and 108 patients were selected as historic controls from a prospectively populated database. Patient population was comprised of children with high-risk hematological malignancies, over 90% represented by acute leukemia. Median age at enrollment was 8.8 years. In the prospective cohort 91% of the donors were haploidentical parents, whereas in the historical cohort 72% of the donors were haploidentical. Conditioning was based on either 12Gy TBI or treosulfan. Thiotepa, fludarabine, bortezomib and rituximab were used as additional agents. Patients in the historical cohort received rabbit ATG at 5 mg/kg total dose, while prospective cohort patients received tocilizumab at 8 mg /kg on day -1 and abatacept at 10mg/kg on days 0, 7, 14 and 28. Patients in the prospective trial cohort were randomized to receive mDLI starting on day 0, while 69% of historical cohort patients received mDLI after engraftment, as part of previous trials.

Results: Primary engraftment rate was 99% in the prospective cohort and 98% in the historical cohort. The incidence of grade II-IV aGVHD was 13% in the prospective cohort and 16% in the control arm. Chronc GVHD developed among 13% (historical) and 7% (prospective) cohorts, p 0.07. The incidence of CMV viremia was 51% in the prospective cohort arm and 54% in the historical control arm, p ns. Overall, in the prospective cohort 2-year NRM was 2%, while in the historic cohort -13%, p = 0.002. The incidence of relapse in the prospective cohort was 25%, EFS was 71% and OS was 80%, while in the historical cohort the incidence of relapse was 19%, EFS was 67% and OS was 76%, difference non-significant for all outcomes. Recovery of ab- and gd- T cells was significantly improved at days +30 and +60 after HSCT among recipients of ATG-free preparative regimens.

Conclusions: Among the recipients of ab T cell-depleted grafts, replacement of rabbit ATG with non-lymphodepleting abatacept and tocilizumab immunomodulation did not compromise engraftment and GVHD control and was associated with significantly lower NRM and better immune recovery early after HSCT.

Disclosure: nothing to declare.

O153. Excellent Transplant Survival For Children With Adenosine Deaminase (ADA) Deficiency

Elisabetta Ghimenton^1,2, Su Han Lum^1,3, Eleri Williams¹, Stephen Owens¹, Sophie Hambleton^1,3, Terry Flood¹, Zohreh Nademi¹, Aisling Flinn^1,3, Mario Abinun^1,3, Andrew Cant^1,3, Mary Slatter^1,3, Andrew Richard Gennery^1,3

¹ Great North Children’s Hospital, Newcastle-upon-Tyne, United Kingdom, ² Stellenbosch University, Cape Town, South Africa, ³ Newcastle University, Newcastle-upon-Tyne, United Kingdom

Background: ADA deficiency is a multi-system disorder resulting in combined immunodeficiency and systemic defects involving lungs, brain, skeleton, liver and kidneys, and increased cancer risk. Management options included enzyme replacement therapy (ERT), haematopoietic cell transplantation (HCT) and gene therapy.

Methods: We examined transplant outcomes of 33 children with ADA deficiency at our centre between 1989-2020. Outcomes of interest were overall survival (OS), event-free survival (EFS) (event defined as death, graft failure or second procedures), grade-versus-host disease (GvHD) and long-term disease outcomes. Log rank test was used to analyse predictors of OS. Variables included for predictor analysis were year of transplant (<2007 vs ≥2007), pre-transplant ERT, donor and conditioning.

Results: Median age at diagnosis was 1.2 months (birth-95.8), median age at HCT was 3.2 months (0.8 to 99.8). Median interval between diagnosis and HCT was 1.6 months (0.4 to 23.4). Donors were matched family donor (MFD, n = 12, 36%), matched unrelated donor (MUD, n = 17, 52%), mismatched unrelated donor (MMUD, n = 2, 6%) and haploidentical donor (HID, n = 2, 6%). Stem cell source was marrow (n = 14, 42%), PBSC (n = 7, 22%) and cord blood (n = 12, 36%); 3 had ex-vivo T-cell depletion. Sixteen (49%) received unconditioned stem cell infusion, 5 (15%) had busulfan-cyclosphosphamide (Bu-Cy), and from 2007, 12 (36%) had treosulfan-based conditioning. Three had grade II acute GvHD, 2 had grade III aGvHD, none had grade IV aGvHD or chronic GvHD.

Median duration of follow-up for surviving patients was 7.5 years (0.8-25.0 years). The 8-year OS and EFS for the entire cohort were 90% (95% CI, 72-97%) and 79% (55-91%). HCT after 2007 (n = 21) was 100% compared to 75% before 2007 (n = 12) (p = 0.02). Pre-transplant ERT was associated with 100% OS compared to patients without pre-transplant ERT (81%, 52-94%, p = 0.08). OS was 88% (58-96%) in unconditioned HCT (n = 16), 100% in in Treo-based conditioning (n = 12), 80% in Bu-Cy conditioning (p = 0.30). OS was 100% in MFD, 60% in MUD before 2007 (n = 5), 100% in MUD after 2007 (n = 12), 50% in MMUD and 100% in HID (p = 0.01). Two with second procedures (one stem cell boost for slipping chimerism; one second HCT for poor immune function) were alive. Of three total deaths, 2 had multiorgan failure and 1 encephalopathy: all 3 deceased patients received cord blood transplant (2 unconditioned; 1 Bu-Cy).

Of 20 long-term survivors (> 2-year post-transplant), median age at last follow-up was 11.2 years (range, 2.3 to 25.4) with median duration of follow-up 10.9 years (2.0 to 25 years). Median myeloid chimerism was 5% (range, 0-100%) and median T-lymphocyte chimerism was 94% (range, 46-100%). Thirteen (65%) were off immunoglobulin (median myeloid chimerism, 13%, range 3-100%; 7 unconditioned, 3 BuCy; 3 Treo-based) while 7 patients remained on immunoglobulin replacement (median myeloid chimerism, 0%, range 0-9%; 6 unconditioned; 1 had BuCy). Four patients had autoimmunity (1 transverse myelitis, 1 juvenile idiopathic arthritis, 1 hypothyroidism, 1 autoimmune haemolytic anaemia, hypothyroidism and type 1 diabetes).

Conclusions: Pre-transplant ERT and Treo-based conditioning is a promising strategy with excellent survival and long-term disease outcome. Since 2007, alternative donor HCT is comparable to MFD HCT.

Disclosure: Nothing to declare.

O154. Impact of Early Lymphocyte Immune Reconstitution on Transplant-Related Mortality Among Children After Allogeneic Hematopoietic Stem Cell Transplantation With Alpha-Beta T Cell Depletion

Svetlana Glushkova¹, Dmitriy Pershin¹, Viktoria Vedmedskaya¹, Maria Fadeeva¹, Yakov Muzalevskii¹, Alexei Kazachenok¹, Elena Kurnikova¹, Julia Starichkova¹, Elvira Sultanova¹, Svetlana Radygina¹, Maria Ilushina¹, Rimma Khismatullina¹, Larisa Shelikhova ¹, Dmitriy Balashov ¹, Alexei Maschan¹, Michael Maschan¹

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

Background: αβ T cell depletion is an established approach to graft engineering. This method allows to reduce the risk of developing acute GvHD and PTLD by removing TCRαβ+ and CD19+ cells from the graft. On the other hand, TCRαβ/CD19-depleted HSCT is associated with delayed recovery of adaptive immunity. Therefore, severe viral infections remain common, difficult to control and constitute a major cause of transplant-related mortality (TRM). The aim of this research is to investigate the correlation of early immune reconstitution of lymphocyte subpopulations and the risk of TRM among the recipients of αβ T cell-depleted grafts.

Methods: The study cohort includes 644 patients: 389 with malignant and 255 with non-malignant diseases. All patients received their first alloHSCT with TCRαβ/CD19 cell depletion from January 2012 to September 2020. The HSCT, graft preparation, laboratory control of the graft composition and after HSCT immune reconstitution analysis, was performed at one medical research center. Donors were either haploidentical related (384 patients) or matched unrelated (260 patients). Graft preparation was performed from apheresis product on Miltenyi Biotec devices (СliniMACS Prodigy™, СliniMACS™). Laboratory control of the graft composition and immune reconstitution was carried out by flow cytometry. The concentration of T cells, αβ T cells, ɣδ T cells and NK cells in peripheral blood was counted on days +30 and +60 after HSCT. Patients cohorts were divided by median absolute count of each of the lymphocyte subpopulation and TRM was calculated for each group by cumulative risk method, groups (≥median vs <median) were compared with Gray test.

Results: 53 patients died due to reasons, unrelated to primary disease relapse/progression. TRM in the whole cohort was 8,4% (12,9% non-malignant indications (NMI), 5,5% malignant indications (MI)). Among patients with MI better recovery of total T cells, αβ T cells and ɣδ T cells on day 30 was associated with significantly improved TRM, while neither day 30 NK, nor day 60 immune recovery correlated with the outcome. Among patients with NMI, there was a trend towards improved TRM with better recovery of αβ T cells on day 30, and significant improvement of TRM with better recovery of total T cells, αβ T cells and ɣδ T cells on day 60. Results are summarized on Table1 and Figure1.

Table 22

Conclusions: Better early recovery of T cells, both αβ and ɣδ subsets is associated with significantly improved non-relapse mortality among the recipients of αβ T cell-depleted grafts.

Disclosure: M.M. received speakers fee from Miltenyi Biotec.

O155. Azacitidine And Prophylactic Donor Lymphocyte Infusion After Allogeneic Stem Cell Transplantation in Pediatric Patients With High Risk Acute Myeloid Leukemia: A Retrospective Single-center Cohort Study

Natalie Booth¹, Lucia Mirea¹, Emily Huschart¹, Holly Miller¹, Dana Salzberg¹, Courtney Campbell¹, Kristen Beebe¹, Charlotte Schwalbach¹, Roberta H. Adams¹, Alexander Ngwube¹

¹ Phoenix Children’s Hospital, Phoenix, United States

Background: Pediatric patients with high risk acute myeloid leukemia (AML) who undergo allogeneic hematopoietic stem cell transplant (HSCT) continue to have poor outcomes. In the United States, 30-80% will relapse within the first year post-HSCT (PMID: 27916512). Identifying approaches for relapse prevention after HSCT in AML are warranted. The role of maintenance therapy post-HSCT has been reported in multiple small studies (PMID: 32663296). In May 2018, Phoenix Children’s Hospital (PCH) incorporated the use of post-HSCT maintenance therapy in pediatric patients with high risk AML in attempt to decrease the number of AML relapses after HSCT. Azacitidine (six cycles starting on Day +60), followed by donor lymphocyte infusion (three escalating doses beginning after day +120), was given to increase an immune response against residual leukemic cells and lead to improved post-HSCT relapse rates. Therapy was altered if patients developed cytopenias, liver function abnormalities, or graft versus host disease. We recently reported this post-HSCT maintenance therapy to be safe and well tolerated (PMID: 33150833). We compared 2 year leukemia-free survival post-HSCT between high risk AML patients who received post-HSCT maintenance therapy and historical controls who did not receive post-HCST maintenance therapy.

Methods: This retrospective historical cohort study was Institutional Review Board approved, conducted at PCH and included all high risk AML patients who underwent HSCT at PCH from January 1, 2010 to May 31, 2020. Data including demographics and pre- and post-HSCT characteristics was collected. Patients were divided into 2 groups: conventional group (historical controls, did not receive post-HSCT maintenance therapy) or intervention group (received post-HSCT maintenance therapy). We excluded all patients who died or relapsed before Day +60. The efficacy of post-HSCT maintenance therapy was studied. The primary endpoint was 2 year leukemia-free survival estimated using the Kaplan-Meier method.

Results: Forty-one patients were included in this study: 28 in the conventional group and 13 in the intervention group. Median age at transplant for the entire cohort was 9.9 years with a range of 1.2 to 23 years. The median age of the conventional group was 7 years and the median age of the intervention group was 12.6 years. Ninety percent of the total cohort received a myeloablative conditioning regimen. Of the patients in the conventional group, 29% had a matched sibling donor (MSD), 32% had a matched unrelated donor (MUD), 32% had a mismatched unrelated donor (MMUD) and 7% had a haploidentical donor transplants. In the intervention group, 15% received a MSD, 46% received a MUD, 9% received a MMUD and 30% received haploidentical donor transplants. The 2-year leukemia free survival in the intervention group was 92.3% compared to 67.9% in the conventional group. See Figure 1.

Conclusions: Our preliminary results suggest that this novel post-HSCT maintenance therapy, comprising azacitidine and prophylactic donor lymphocyte infusion may improve leukemia-free survival in pediatric patients with high risk AML.

Disclosure: Nothing to declare.

O156. Enteral Nutrition Promotes The Recovery of Gut Microbiome Homeostasis And Protects From Bloodstream Infections In Children Receiving Allogeneic Hematopoietic Stem Cell Transplantation

Daniele Zama^1,2, Riccardo Masetti^1,2, Edoardo Muratore^1,2, Davide Leardini^1,2, Federica D’amico¹, Tamara Belotti^1,2, Maura Fois^1,2, Francesca Gottardi^1,2, Martina Uberti^1,2, Silvia Turroni¹, Arcangelo Prete^1,2, Patrizia Brigidi¹, Andrea Pession^1,2

¹ Alma Mater Studiorum-University of Bologna, Bologna, Italy, ² Sant’Orsola Malpighi Hospital, Bologna, Italy

Background: The relationship between gut microbiome (GM) and outcomes in allogeneic hematopoietic stem cell transplantation (allo-HSCT) recipients has been largely described in recent years. The type of nutritional support during HSCT may play a key role in modulating the GM. Although the most recent guidelines support the use of enteral nutrition (EN), the first line nutritional approach traditionally performed is parenteral nutrition (PN). The use of PN has been linked to potential detrimental effect on the GM and clinical outcomes, but few studies have addressed this topic in allo-HSCT recipients. Particularly, the effect of EN in children receiving allo-HSCT has never been reported.

Methods: Consecutive pediatric patients undergoing allo-HSCT at our referral center between January 2016 and July 2019 were evaluated. Post-transplant and nutritional outcomes of patients receiving EN for more than 7 days (EN group) were compared with those of patients receiving EN for fewer than 7 days or receiving only PN (PN group). The GM structure was analyzed longitudinally in patients with a pre-HSCT fecal sample and at least two samples collected after HSCT. Taxonomic analysis was performed by 16S rRNA gene sequencing, using the Illumina MiSeq platform.

Results: Forty-seven patients were enrolled for the evaluation of clinical outcomes, 19 and 28 in the EN and PN groups respectively. The two groups were homogeneous regarding the main patient characteristics at admission and transplantation modalities. In the EN group, a reduced incidence of blood stream infections (BSI) was observed (14.3% vs. 53.6%; p = 0.02). The type of nutritional support was also the only variable independently associated with BSI in the multivariate analysis (p = 0.03). Body mass index (BMI) and BMI Z-score during hospitalization were not statistically significantly different between the two groups. A trend in the EN group was observed for reduced incidence of gut aGvHD Grade III-IV (0% vs 17%; p = 0.08) and steroid-resistant gut aGvHD (0% vs 21,4%; p = 0.06). The GM analysis was performed in 38 patients, 19 and 19 in the EN and PN groups respectively. We observed a prompt recovery of GM diversity post-HSCT only in EN subjects (p = 0.01). Moreover, some genera were mainly restored in the EN group, namely Faecalibacterium, Dorea, Blautia, Bacteroides, Parabacteroides, and Oscillospira, all of which are well-known health-associated GM components. In the PN group a significant decrease in short chain fatty acids levels post-HSCT was observed (p = 0.006), while in EN patients they were comparable to the baseline value.

Conclusions: The present study showed recovery of GM homeostasis after allo-HSCT only in children fed with EN. This improved gut eubiosis could be a possible explanation of the observed reduction in the incidence of BSI. Further large-scale prospective studies are warranted to determine the impact of the route of nutritional support on transplant outcomes. Metagenomic and metabolomic studies could pave the way on our understanding of the complex interplay between nutrition, intestinal epithelium and immune system.

Disclosure: The authors declare no conflict of interest.

O157. In Vivo B-Cell Depletion With Myeloablative Conditioning For Umbilical Cord Blood Transplant in Hurler Syndrome Abolishes Immune Mediated Cytopenia And Graft Rejection

Ramya Hanasoge Nataraj¹, Denise Bonney¹, Helen Campbell¹, Simon Jones², David Deambrosis³, Prashant Hiwarkar⁴, Pamela Evans⁵, Kay Poulton¹, P M v Hasselt⁶, M B Bierings⁷, Jaap-Jan Boelens⁸, Caroline Lindemans⁷, Robert Wynn¹

¹ Royal Manchester Children’s Hospital, Manchester, United Kingdom, ² Saint Mary’s Hospital, Manchester, United Kingdom, ³ University of Auckland, Auckland, New Zealand, ⁴ Bai Jerbai Wadia Hospital for Children, Mumbai, India, ⁵ Children’s Health Ireland, Crumlin, Dublin, Ireland, ⁶ Wilhelmina Kinderziekenhuis, Utrecht, Netherlands, ⁷ Princess Maxima Center for Oncology, Utrecht, Netherlands, ⁸ Memorial Sloan Kettering, Newyork, United States

Background: Hematopoietic stem cell transplantation (HCT) is the standard treatment for Hurler Syndrome (HS) and umbilical cord blood (UCB) is the preferred donor cell source. We have reported that with myeloablative, busulfan-based conditioning graft failure (GF) is aplastic and exclusively in UCB recipients.¹ In multi-institutional studies then GF is the commonest cause of death after UCB transplant for non-malignant disease.² Immune-mediated cytopenias (IMC) are common after CB transplant³ and we recently reported that IMC is a forme fruste of graft rejection, in which the residual intact immune system rejects the engrafted CB as IMC, or causes aplastic graft failure.⁴

We hypothesised that addition of rituximab in the conditioning regimen might reduce both IMC and GF, both significant limitations on the general utility of UCB as a donor cell source in pediatric HCT.

Methods: This study was a retrospective analysis of patients undergoing first UCB HCT for HS at 2 metabolic transplant centres. All received Myelo-ablative dose of busulfan along with fludarabine, and Anti-thymocyte globulin (ATG). From 2019, Rituximab 375mg/m² was added as part of conditioning on Day -10 and a 2nd dose was given post-transplant on Day + 30.

For each patient IMC, GF and Death were recorded. IMC included immune mediated haemolysis, thrombocytopenia and neutropenia. Event free survival (EFS) was defined as IMC-free GF-free survival following the first HCT.

Results: Fifty-four patients who underwent first UBC transplant for HS were enrolled. Seventeen received rituximab and 37 did not. Table 1 depicts the patient characteristics. The median time for GF and IMC was 28 days and 72 days, respectively. EFS at 1 year-post HSCT was 93.3% and 45.9% in the Rituximab group and control group respectively (p = 0.022). Overall, 6 patients died, out of which 1 was from the Rituximab group. OS at 2 years post HCT was not different among the two groups (93.3% vs 86.5%), since both IMC and GF were rescued with medical intervention and second transplant, respectively.

	Control group (n = 37)	R group (n = 17)
Male	17 (45.9%)	7 (41%)
Median age at HSCT	14 months (IQR 7-18.4)	7 months (IQR 4-18.7)
HLA mismatch	14	12
Total nucleated cell (Median)	14.53 * 107/kg	17 * 107/kg
Lymphocyte count prior to HSCT (Median)	5.8 * 109/L (3.7 -7.2)	5.4 *109/L (4.6 – 7)
Neutrophil engraftment	Day 17.1	Day 17.3
IMC	13 (35%)	0
GF	5 (13.5%)	0
Death	5 (13.5 %)	1 (5 %)

Conclusions: Both GF and IMC limit the utility of UCB transplant in Non-Malignant diseases, and are manifestations of a failure of recipient immune suppression. Here we report that the addition of rituximab to conditioning prior to transplant, results in the abolition of both IMC and GF, and we propose that this will be relevant beyond HCT in HS and influence UCB transplant in NMD.

Disclosure: Nothing to declare.

O158. Alemtuzumab Protects Children From Severe Acute And From Chronic Gvhd In PBSC As Well As Marrow Graft Recipients: A Multicentre Retrospective Cohort Analysis

Su Han Lum¹, Beki James², Giorgio Ottaviano³, Anna-Maria Ewins⁴, Salah Ali⁵, Ben Carpenter⁶, Juliana Silva⁷, Sanjay Tewari⁸, Arun Thomas⁸, Geoff Shenton¹, Katharine Patrick⁵, Denise Bonney⁹, Brenda Gibson⁴, Persis Amrolia³, Rachel Hough⁶, Kanchan Rao³, Mary Slatter¹, Robert Wynn⁹

¹ Great North Children’s Hospital, Newcastle upon Tyne, United Kingdom, ² Leeds Children’s Hospital, Leeds, United Kingdom, ³ Great Ormond Street Hospital for Children NHS Foundation Trust, London, United Kingdom, ⁴ Royal Hospital for Sick Children, Glasgow, United Kingdom, ⁵ Sheffield Children NHS Foundation Trust, Sheffield, United Kingdom, ⁶ University College London Hospitals, London, United Kingdom, ⁷ Royal Hospital for Children, Bristol, United Kingdom, ⁸ Royal Marsden Hospital, Sutton, United Kingdom, ⁹ Royal Manchester Children’s Hospital, Manchester, United Kingdom

Background: Alemtuzumab is commonly used for in vivo T cell depletion but knowledge about its impact on peripheral blood stem cells (PBSC) versus bone marrow (BM) as the stem cell source in children is lacking. This study compared transplant outcomes in 366 matched unrelated donor (MUD) recipients according to PBSC (n = 168) and BM (n = 198) after alemtuzumab based conditioning for first haematopoietic cell transplantation (HCT) between 2015 and 2019 at 8 paediatric transplant centres in the UK.

Methods: The primary end-points were overall survival (OS), event-free survival (EFS) and GvHD-relapse free survival (GRFS). EFS was defined as survival without graft failure, relapse and second procedure. GRFS was defined as survival without grade III-IV acute GvHD (aGvHD), chronic GvHD (cGVHD), graft failure, relapse and second procedure. Subgroup differences in OS, EFS and GRFS were evaluated by log-rank test. Secondary endpoints were aGvHD, cGvHD and non-relapse mortality (NRM). Competing risk methods were used for the cumulative incidence (CIN) of GvHD and NRM, with death and graft failure as competing events. Subgroup differences in GvHD and NRM were evaluated by Gray’s test. Statistical analyses were performed using STATA 14.2.

Results: Median age at HCT was 6.6 years (range 0.1-19.3 years) and median follow-up of surviving patients was 2.0 years (range, 0.2-5.4 years). The diagnoses were malignancy in 140 patients (38%; ALL, n = 87; AML, n = 23; others, n =30; BM, n = 100; PBSC, n = 40) and non-malignant disorders in 226 patients (SCID, n = 10; non-SCID PID, n = 118; aplastic anaemia, n = 35; BM failure, n = 21; HLH, n = 12; others, n = 30; BM, n = 98; PBSC, n = 128). Conditioning was MAC (n = 198, 54%), RIC (n = 113, 31%) and MIC (n = 55, 15%). GvHD prophylaxis was CNI (calcineurin inhibitor)± MTX in 195 (53%) and CSA+MMF in 171 (n = 47%). All received 0.9-1.0mg/kg alemtuzumab except 9 (2.5%) who received 0.5-0.8mg/kg. Total nucleated, CD34+, CD3+ cell doses were significantly higher in PBSC compared to BM. The neutrophil and platelet engraftment was significantly faster in PBSC compared to BM. The OS, EFS, GRFS were comparable between PBSC (2-year OS, 81%, 74-87%; EFS, 77%, 69-83%; GRFS, 67%, 59-74%) and BM (2-year OS 79%, 72-85%; EFS, 74%, 67-80%; GRFS, 65%, 57-72%). NRM was 13% (8-20%) in PBSC and 11% (7-18%) (p = 0.56). CIN of grade II aGvHD was 16% (10-24%) in BM recipients (n = 143) who had CNI monotherapy for GvHD prophylaxis and 22% (14-37%) in PBSC recipients (n = 75) who received CNI+MMF and CD3+ cell dose <5x108/kg (p = 0.35). The CIN of grade II-IV aGvHD was significantly higher in PBSC recipients who received CD3+ cell dose >5x10⁸/kg (p = 0.01). The CIN of grade III-IV aGvHD and cGvHD was comparable between BM and PBSC+CSA+MMF recipients. There was a trend of higher severe aGvHD and cGvHD rates in PBSC recipients who received CNI monotherapy.

Conclusions: In patients who received MUD transplant and alemtuzumab-based conditioning, PBSC had comparable OS, EFS and GFRS, grade III-IV aGvHD and cGvHD rates to BM recipients. Recipients of PBSC plus CNI+MMF plus CD3+ <5x108/kg had comparable grade II-IV aGvHD rate to BM plus CNI recipients.

Disclosure: Nothing to declare.

Stem Cell Donor

O159. Functional Mismatching For HLA-DP Reduces Relapse Risk After Unrelated Hematopoietic Cell Transplantation: A Study From The Cellular Therapy And Immunobiology Working Party of The EBMT

Annalisa Ruggeri^1,2, Katya Mauff³, Carlheinz Müller⁴, Pietro Crivello⁵, Liesbeth C. de Wreede⁶, Lotte Wieten^7,2, Luca Vago^1,2, Jorinde D. Hoogenboom⁸, Gérard Socié⁹, Riitta Niittyvuopio¹⁰, Ibrahim Yakoub-Agha¹¹, Tobias Gedde-Dahl¹², Jan J. Cornelissen¹³, Charles R. Crawley¹⁴, Edouard Forcade¹⁵, Claude-Eric Bulabois¹⁶, Eleni Tholouli¹⁷, Gwendolyn van Gorkom⁷, Anne Huynh¹⁸, Virginie Gandemer¹⁹, Stig Lenhoff²⁰, Xavier Poiré²¹, Jan-Erik Johansson²², Vanderson Rocha²³, Chiara Bonini^1,2, Christian Chabannon^24,2, Katharina Fleischhauer^5,2

¹ San Raffaele Scientific Institute, Milano, Italy, ² Cellular Therapy and Immunobiology Working Party, EBMT, Leiden, Netherlands, ³ EBMT Statistical Unit, Leiden, Netherlands, ⁴ Zentrales Knochenmarkspender-Register Deutschland, Ulm, Germany, ⁵ Institute for Experimental Cellular Therapy, University Hospital Essen, Essen, Germany, ⁶ Leiden University Medical Center, Leiden, Netherlands, ⁷ Maastricht University Medical Centre, Maastricht, Netherlands, ⁸ EBMT Leiden Data Office, Leiden, Netherlands, ⁹ Hôpital St. Louis, Paris, France, ¹⁰ HUCH Comprehensive Cancer Center, Helsinki, Finland, ¹¹ CHU de Lille, Lille, France, ¹² Oslo University Hospital, Rikshospitalet, Oslo, Norway, ¹³ Erasmus MC Cancer Institute, Rotterdam, Netherlands, ¹⁴ Addenbrookes Hospital, Cambridge, United Kingdom, ¹⁵ CHU Bordeaux, Pessac, France, ¹⁶ CHU Grenoble Alpes, Université Grenoble Alpes, Grenoble, France, ¹⁷ Manchester Royal Infirmary, Manchester, United Kingdom, ¹⁸ CHU - Institut Universitaire du Cancer Toulouse, Toulouse, France, ¹⁹ Centre Hospitalier Universitaire de Rennes, Rennes, France, ²⁰ Skanes University Hospital, Lund, Sweden, ²¹ Cliniques Universitaires St. Luc, Brussels, Belgium, ²² Sahlgrenska University Hospital, Goeteborg, Sweden, ²³ Hospital Sirio-Libanes, Sao Paulo, Brazil, ²⁴ Institut Paoli-Calmettes, Centre de Lutte Contre le Cancer; Centre d’Investigations Cliniques en Biothérapie, Université d’Aix-Marseille, Inserm CBT 1409, Marseille, France

Background: Relapse is the most frequent cause of death beyond day 100 in patients treated for hematologic malignancies by unrelated donor hematopoietic cell transplantation (HCT). A protective effect of donor-recipient HLA-DPB1 mismatches against relapse has been described over a decade ago (Shaw, Blood 2007), and models for functional HLA-DPB1 mismatching by T-cell epitope (TCE) groups or expression levels have been developed which predict the risk of non-relapse mortality (NRM) and acute graft-versus-host disease (GvHD), respectively (Fleischhauer, Lancet Oncol 2012; Petersdorf, N Engl J Med 2015). Here, we investigated the role of HLA-DPB1 mismatches for relapse after unrelated HCT performed over the last 15 years in Europe in adult patients with hematological malignancies.

Methods: We studied 6649 consecutive 8/8 (HLA-A,B,C,DRB1) matched unrelated HCTs reported to the EBMT in the periods 2005-2011 (26.2%) or 2012-2017 (73.8%). Only first allogeneic transplants performed for hematologic malignancies, mostly acute leukemia or myelodysplastic syndrome (69.3%), were included. Disease status at transplant was early (45.4%), intermediate (28.9%) or advanced (25.8%). Median follow-up was 36.7 months, median age was 54.4 years. Conditioning regimen was mainly reduced intensity (59.9%). GvHD prophylaxis was based on calcineurin inhibitors, mostly anti-thymocyte globulin or another form of in vivo T-cell depletion (73.5%), and did not include post-transplant cyclophosphamide. HLA data were validated using the HLAcore library and a haplotype-based probability check from the German Donor Registry. 1486 (22.3%) pairs were HLA-DPB1 allele matched and 5163 (77.7%) were mismatched. Amongst the latter, 2886 (43.4%) and 2277 (34.2%) pairs were TCE-permissive or -non-permissive, respectively, and 2753 (41.4%) had a single HLA-DPB1 mismatch in graft-versus-host direction, with the low expression-A and the high expression-G genotypes in 1467 (22.0%) and 1286 (19.3%) of cases, respectively.

Results: The 5-year cumulative incidence of relapse and probability of relapse-free survival (RFS) was 33% and 42%, respectively. HLA-DPB1 allele mismatches were significantly associated with reduced relapse risks (hazards ration [HR] 0.7, p < 0.0001). This observation was valid across both time periods analyzed, and across disease stages including advanced disease. The protective effect was maintained also when only pairs with TCE-permissive mismatches were compared to allele matches (HR 1.2, p = 0.001), and was not further increased by TCE-non-permissive compared to TCE-permissive mismatches (HR 0.9, p = 0.27). In contrast, TCE-non-permissive mismatches were associated with acute GvHD (HR 1.1, p = 0.014) and NRM (HR 1.2, p = 0.013) compared to TCE-permissive mismatches. We also confirmed the association between expression-G mismatches and acute GvHD, compared to expression-A (HR 1.2, p = 0.02). Interestingly, this translated into significantly lower relapse risks (HR 0.8, p = 0.009) without a significant increase in NRM (HR 1.1, p = 0.223), for expression-G compared to expression-A. When considering the TCE and expression models in combination, improved RFS was observed for permissive TCE mismatches with expression-G compared to the other combinations (49% vs 43%, p = 0.05).

Conclusions: Selecting an HLA-DPB1 mismatched, TCE-permissive unrelated donor reduces relapse risks after HCT regardless of disease status. Within this pool, donors that are both TCE-permissive and expression-G perform best in the combined terms of relapse, acute GvHD and NRM, resulting in improved RFS.

Clinical Trial Registry: Not applicable.

Disclosure: Nothing to declare.

O160. Post-transplant Cyclophosphamide in One-Antigen Mismatched Unrelated Donor Transplantation Versus Haploidentical Transplantation: A Retrospective Study on Behalf of The Acute Leukemia Working Party of The EBMT

Giorgia Battipaglia¹, Jacques-Emmanuel Galimard², Myriam Labopin^2,3,4, Emanuele Angelucci⁵, Didier Blaise⁶, Annalisa Ruggeri⁷, Zafer Gulbas⁸, Jean Luiz Diez-Martin⁹, Yener Koc¹⁰, Luca Castagna¹¹, Antonin Vitek¹², Patrizia Chiusolo¹³, Benedetto Bruno¹⁴, Ivan Moiseev¹⁵, Montserrat Rovira¹⁶, Massimo Martino¹⁷, Mercedes Colorado Araujo¹⁸, Mutlu Arat¹⁹, Giovanni Grillo²⁰, Hans Martin²¹, Lucia Lopez Corral²², Fabrizio Pane¹, Ali Bazarbachi²³, Fabio Ciceri⁷, Arnon Nagler^24,2, Mohamad Mohty^2,4,3

¹ Federico II University of Naples, Naples, Italy, ² EBMT Paris Study Office, Paris, France, ³ Sorbonne Universités, UPMC Univ Paris 06, INSERM, Centre de Recherche Saint-Antoine (CRSA), Paris, France, ⁴ Hôpital Saint Antoine, Service d’Hématologie et Thérapie Cellulaire, Paris, France, ⁵ UOC Ematologia e Trapianto di Midollo Osseo, IRCCS Ospedale, Genova, Italy, ⁶ Programme de Transplantation & Therapie Cellulaire, Institut Paoli-Calmettes, Aix Marseille Univ, CNRS, INSERM, CRCM, Marseille, France, ⁷ Ospedale San Raffaele s.r.l., Haematology and BMT, Milan, Italy, ⁸ Anadolu Medical Center Hospital, Kocaeli, Turkey, ⁹ Hospital General Universitario Gregorio Marañón, Instituto de Investigación Sanitaria Gregorio Marañon, Madrid, Spain, ¹⁰ Medicana International, Istanbul, Turkey, ¹¹ Humanitas Clinical and Research Center, IRCCS, Rozzano, Milan, Italy, ¹² Institute of Hematology and Blood Transfusion, Prague, Czech Republic, ¹³ Fondazione Policlinico A. Gemelli IRCCS,Istituto di Ematologia, Università Cattolica del Sacro Cuore, Rome, Italy, ¹⁴ SSCVD Trapianto di Cellule Staminali, AOU Città della Salute e della Scienza di Torino, Turin, Italy, ¹⁵ R.M. Gorbacheva Research Institute, Pavlov University, Saint Petersburg, Russian Federation, ¹⁶ Hospital Clinic, BMT Unit, IDIBAPS, Institut Josep Carreras, Barcelona, Spain, ¹⁷ Grande Ospedale Metropolitano ‘Bianchi-Melacrino-Morelli’, Reggio Calabria, Italy, ¹⁸ Hospital U. Marqués de Valdecilla, Santander, Spain, ¹⁹ Florence Nightingale Sisli Hospital, Hematopoietic SCT Unit, Istanbul, Turkey, ²⁰ ASST Grande Ospedale Metropolitano Niguarda, Milan, Italy, ²¹ Goethe-Universitaet, Medizinische Klinik II, Frankfurt, Germany, ²² Hospital Clínico, Salamanca, Spain, ²³ American University of Beirut, Beirut, Lebanon, ²⁴ Chaim Sheba Medical Center, Tel Hashomer, Israel

Background: Whether to choose Haploidentical (Haplo) or one antigen mismatched unrelated donor (MMUD) transplantation (HCT) with post-transplant cyclophosphamide (PTCY) remains an unanswered question.

Methods: This retrospective study included adults undergoing their first Haplo- or MMUD-HCT for acute myeloid leukemia in 1st or 2nd complete remission in 2010-2018 and receiving unmanipulated grafts with PTCY. Three groups were compared: MMUD-HCT with peripheral blood as stem cell source (MMUD-PB; n = 124); Haplo-HCT with bone marrow (Haplo-BM; n = 560); Haplo-HCT with PB (Haplo-PB; n = 769).

Results: Patients in Haplo-PB were older (median age of 55 versus 52 and 51 years in MMUD-PB and Haplo-BM, respectively; p < 0.01). A myeloablative conditioning regimen was used in 52%, 71% and 58% of patients in MMUD-PB, Haplo-BM and Haplo-PB, respectively (p < 0.01). A calcineurin inhibitor with mycophenolate mofetil represented the most frequent adjuvant immunosuppressive agents. Median time for neutrophil engraftment was 20 days in MMUD-PB and Haplo-BM and 19 in Haplo-PB, with a cumulative incidence (CI) of neutrophil engraftment at day 30 of 91%, 88% and 89% for MMUD-PB, Haplo-BM and Haplo-PB, respectively. The CI of grade II-IV and grade III-IV acute GVHD (aGVHD) was lower in Haplo-BM (21% and 6% versus 33% and 13% versus 34% and 13% in MMUD-PB and Haplo-PB, respectively, p < 0.01). Due to longer follow-up in Haplo-BM, outcomes were calculated after censoring at 2 years. CI of chronic GVHD (cGVHD) of all grades and of extensive cGVHD were significantly lower in Haplo-BM (26% and 9% versus 36% and 16% versus 35% and 13% in MMUD-PB and Haplo-PB, respectively, p = 0.01 and 0.03). No differences were observed in the CI of relapse (20% in both MMUD-PB and Haplo-PB and 23% in Haplo-BM; p = 0.39). The CI of non-relapse mortality (NRM) was significantly lower in MMUD-PB (11% versus 20% and 25% in Haplo-BM and Haplo-PB, respectively; p < 0.01). We observed no significant differences in the probability of either overall survival (OS, 70% versus 62% and 60% in MMUD-PB, Haplo-BM and Haplo-PB, p = 0.12) or leukemia-free survival (LFS, 69% versus 56% and 55%, p = 0.06) among the three groups. In Haplo-PB GVHD-free/relapse-free survival (GRFS) was lower (42% versus 49% in the other two groups, p < 0.01). In multivariate analysis, Haplo-HCT was associated to a lower LFS and a higher NRM as compared to MMUD-PB (reference group, hazard ratio (HR) for LFS 1.57, 95% CI 1.06-2.33 for Haplo-BM, p = 0.02; 1.52, 95% CI 1.05-2.22, p = 0.03 for Haplo-PB) (HR for NRM 2.25, 95% CI 1.16-4.33, p = 0.02 for Haplo-BM; 2.58, 95% CI 1.37-4.86, p < 0.01 for Haplo-PB). Lower risk of grade II-IV (HR 0.64, 95% CI 0.42-0.96; p = 0.03) and grade III-IV (HR 0.45, 95% CI 0.24-0.85; p = 0.01) aGVHD was found for Haplo-BM, together with lower risk of cGVHD (HR 0.52, 95% CI 0.28-0.96; p = 0.04) compared to MMUD-PB. No significant differences were found for GRFS or OS.

Conclusions: Use of MMUD is associated to higher LFS, however both MMUD- and Haplo-HCT with PTCY are valid transplant options . Despite lower risk of both aGVHD and cGVHD with Haplo-BM, use of Haplo-HCT is associated to high NRM, with need for further improvements in supportive care.

Disclosure: No conflict of interest to disclose.

O161. HLA Evolutionary Divergence (HED) Influences Outcomes of Allogeneic Hsct From Unrelated Donor (UD) In Pediatric Patients And Young Adults Affected By Malignant Disorders

Pietro Merli¹, Pietro Crivello², Luisa Strocchio¹, Rita Maria Pinto¹, Mattia Algeri¹, Francesca Del Bufalo¹, Daria Pagliara¹, Stefania Gaspari¹, Antonio Giacomo Grasso¹, Francesco Quagliarella¹, Giulia Boz¹, Emilia Boccieri¹, Maria Troiano¹, Katharina Fleischhauer², Marco Andreani¹, Franco Locatelli^1,3

¹ Bambino Gesù Children’s Hospital, Rome, Italy, ² University Hospital Essen, Essen, Germany, ³ Sapienza University, Rome, Italy

Background: The “divergent allele advantage” model hypothesizes that an increased difference in the genotype of the two alleles of a given HLA locus results into the possibility to present more diverse antigens (“immunopeptidomes”). This diversity can influence the outcomes of immunotherapies, as shown in adults patients treated with checkpoint-inhibitors (Chowel, 2019) or given allogeneic HSCT (Roerden, 2020). In this study we evaluated the influence of HED on outcomes of children/young adults (YA) undergoing allogeneic HSCT from an UD for different malignant disorders.

Methods: We retrospectively analyzed all consecutive patients undergoing allogeneic HSCT from an UD between 02/2012 and 09/2019 at Bambino Gesù Children’s Hospital for a malignant disorder. Patients and their unrelated donors were typed for HLA-A, -B, -C, -DRB1, -DRB3/4/5, -DQA1, -DQB1 and -DPB1 loci at high-resolution. Typing was performed with different molecular techniques: initially by SBT (Invitrogen, Brown Deer, WI; Atria, Abbott Park, IL) in combination with HD PCR-SSO (One Lambda Inc., Canoga Park, CA); subsequently by NGS (Thermos Fisher - One Lambda Inc., Canoga Park, CA) once this method was available in the routine activity. HED scores for each patient were obtained using an online available Perl script (Pierini, 2018), calculating the pairwise differences between the alleles of HLA-A, -B, -C or -DRB1 using Grantham Distance metric (Grantham, 1974).

Results: In the study period, 144 patients were transplanted, most of them for a diagnosis of acute leukemia. All patients received an unmanipulated graft after myeloablative conditioning regimen. In line with previous observations, HLA-DR showed the highest divergence (median 10.98, range 0-18.58, Figure 1A), followed by HLA-A and –B (median 8.10 and 8.07, respectively), while HLA–C showed the lowest diversity (median 4.82). HED-B and HED-DRB1 had an impact on disease-free survival (DFS); indeed, as reported in Figure 1 (B-C), patients with higher (i.e., above the median value) HED-B (Fig.1B) and HED-DRB1 (Fig.1C) had a DFS significantly higher than those with lower HED values. Notably, patients with higher HED-DRB1 had also an improved overall-survival (Fig.1D). The impact on DFS of HED-B (HR 0.51 (95% CI 0.27-0.93), p = 0.03) and HED-DRB1 (HR 0.53 (95% CI 0.28-0.98), p = 0.04) remained statistically significant in a multivariable model incorporating also disease status (CR versus active disease, p < 0.001), type of disease (ALL, AML, lymphomas and MDS/MPD, p < 0.01), age at transplant (0-4 years, 4-12 and 12-25, p = n.s.) and type of conditioning (TBI-based versus not, p = n.s.). Interestingly, in sub-group analyses a trend for an improved DFS was observed in patients suffering from acute lymphoblastic leukemia with high HED-B; conversely, HED-DRB1 seemed to be associated with an increased DFS in myeloproliferative neoplasms.

Conclusions: In this single-center retrospective study, high HED-B and HED-DRB1 were independently associated with improved outcome of allogeneic HSCT in pediatric/YA patients. If confirmed in larger, multicenter studies, these variables could help further refine the criteria used for donor selection [e.g., directing the choice towards donors other than unrelated (i.e., haploidentical relative or cord blood) for patients with low HED-B and HED-DRB1 scores.

Clinical Trial Registry: Not applicable.

Disclosure: Nothing to declare.

O162. Management of Donor-specific Antibodies In Haploidentical Transplant: Multicenter Experience From The Madrid Group of Hematopoietic Transplant

Rebeca Bailén^1,2, José Luis Vicario³, Laura Solán⁴, Irene Sánchez-Vadillo⁵, Pilar Herrera⁶, María Calbacho⁷, Raquel Alenda³, José Luis López-Lorenzo⁴, Karem Humala⁵, Anabelle Chinea⁶, José María Sánchez-Pina⁷, Javier Arzuaga¹, Virginia Pradillo¹, Nieves Dorado^1,2, Gillen Oarbeascoa^1,2, Javier Anguita^1,2,8, José Luis Díez-Martín^1,2,8, Mi Kwon^1,2

¹ Hospital General Universitario Gregorio Marañón, Madrid, Spain, ² Gregorio Marañón Health Research Institute, Madrid, Spain, ³ Centro de Transfusión de la Comunidad de Madrid, Madrid, Spain, ⁴ Hospital Universitario Fundación Jiménez Díaz, Madrid, Spain, ⁵ Hospital Universitario La Paz, Madrid, Spain, ⁶ Hospital Universitario Ramón y Cajal, Madrid, Spain, ⁷ Hospital Universitario 12 de Octubre, Madrid, Spain, ⁸ Universidad Complutense de Madrid, Madrid, Spain

Background: Donor specific antibodies (DSAs) are preformed IgG antibodies with specificity against HLA molecules not shared with the donor that can be responsible for graft failure (GF) in the setting of mismatched HSCT. The aim of our study is to report the experience of the Madrid Group of Hematopoietic Transplant (GMTH) in patients with DSAs undergoing haplo-HSCT.

Methods: Patients undergoing haplo-HSCT in centers from the GMTH from 2012 to 2020 were included in the study. DSAs were analyzed with a solid-phase single-antigen immunoassay (Luminex®); monitoring was performed during desensitization on days -14, -7, 0 and in a weekly basis until neutrophil engraftment was achieved. Desensitization strategies used depended on center experience, immunofluorescence intensity, complement fixation and type of antibodies.

Results: We identified a total of 20 haplo-HSCT in 19 patients performed with DSAs in 5 centers. 18 (95%) patients were female (all with prior pregnancies). One patient was diagnosed with refractory aplastic anemia (AA). All patients lacked a suitable alternative donor without DSAs. All transplants were performed using peripheral blood as stem cell source. Myeloablative conditioning was used in 53% and all patients received post-transplant cyclophosphamide based GVHD prophylaxis; the patient with AA also received ATG.

10(53%) patients presented anti-HLA class I DSAs (6 of them with >5000MFI), 4 (21%) presented anti-HLA class II (1 with >5000MFI) and 6(26%) presented both anti-HLA class I and II DSAs (5 with >5000MFI). All patients received at least two treatments as desensitization strategy (including RTX in 90%) and all experienced a decrease of MFI after desensitization (mean reduction 74%). Only one patient who developed progressive increase of MFI after infusion developed GF. Desensitization treatments used included RTX in 90% of patients, IVIG (68%), MMF (63%), therapeutic plasma exchange (TPE) (58%), incompatible platelets (31%), buffy coat (only in patients with class II DSAs, 31%), tacrolimus (26%) and steroids (5%). Seventeen (90%) patients achieved neutrophil engraftment in a median of 18 days (IQR, 15-21); one patient died before engraftment because of infection and one patient with class I DSAs and a high-risk MDS developed primary GF despite an intensive desensitization strategy including RTX, IVIG, TPE, MMF and incompatible platelets.

After a median follow-up of 10 months, OS and EFS were 60% and 58%, respectively. Cumulative incidence of relapse at 10months was 5% (2-years estimated 16%) and NRM was 32%. Cumulative incidence of grade II-IV aGVHD at day 180 was 16% and chronic GVHD at 10months was 16%. 7(37%) patients died during the study period: 1 due to GF, 3 due to bacterial infection (one in the pre-engraftment period, one the context of grade III-IV aGVHD treatment), 2 due to endothelial complications including TA-TMA and SOS, and 1 because of relapse.

Conclusions: Optimal strategy of DSAs desensitization remains unclear. The use of desensitization treatment guided by DSAs intensity kinetics constitute an effective approach with high rates of engraftment for patients with DSAs in need for an haplo-HSCT lacking an alternative suitable donor, including non-malignant disorders.

Disclosure: Nothing to declare.

O163. The Impact of Covid-19 on Unrelated And Donor Cord Provision To UK Transplant Centres During The 1st Wave of The Pandemic: UK Aligned Registry Study

Angharad Pryce¹, Eva Zoubek¹, Ann O’Leary¹, Richard Szydlo¹, Ying Li², Helen Kelly³, Christopher Harvey⁴, Katalin Balassa², Rachel Pawson², Robert Danby¹, Farheen Mir¹

¹ Anthony Nolan, London, United Kingdom, ² British Bone Marrow Registry, NHS Blood and Transplant, Bristol, United Kingdom, ³ DKMS, UK, London, United Kingdom, ⁴ Welsh Bone Marrow Donor Registry, Welsh Blood Service, Cardiff, United Kingdom

Background: The COVID-19 pandemic led to major operational changes for transplant centres and donor registries to mitigate patient and donor risk. NICE, EBMT, and BSBMT issued guidelines to maximise patient safety during the pandemic. Donor registries coordinate the search for volunteer unrelated donors and cord blood units from UK and internationally and organize collection and provision of stem cells. Registries have had to adapt to potential for delays and cancellations due to travel restrictions, increased risk of donor unavailability, unsuitability or quarantine regulations. The UK aligned registries, Anthony Nolan (AN), NHS Blood and Transplant (NHSBT), Welsh Bone Marrow Donor Registry (WBMDR) and DKMS UK have undertaken a study to understand the impact of Covid-19 on unrelated donor and cord provisions. We present data on search requests, donor work up and collections of UK and international donors for UK transplant patients during the first wave of the Covid-19 pandemic.

Methods: Retrospective data analysis from registry database including requests for searches and workups for UK patients (adults and paediatrics) and stem cell collections for unrelated or cord allogeneic haematopoietic stem cell transplant in the UK, between 1st March and 31st May 2020 encompassing UK and international donors for UK patients. To allow comparison against pre-pandemic donor provisions this data was compared to requests from 1st March-31st May 2019. SPSS used for analysis.

Results: Table 1: Impact on Searches and Work up Requests

	Mar-May 2020	Mar-May 2019
New donor search requests for patients	527	648
Repeat search requests	85	109
Workups requested	410 for 321 patients	430 for 359 patients
UK donor: UK patient	33.7%	37.2%
Source of stem cells requested
- PBSC	294 (72%)	300 (70%)
- Bone marrow	21 (5%)	56 (13%)
- Umbilical cord	48 (12%)	28 (6.5%)
- DLI	34 (8%)	36 (8%)
- Subsequent PBSC	13 (3%)	10 (2%)

107 work ups were postponed. Of these, 70 were initiated by the transplant centres with 40% of these citing Covid related reasons. 35 were initiated by donor/collection centre with 26% citing Covid related reasons. There were 129 cancellations, compared with 120 in the same time period in 2019. These were Covid related in 27 cases (21%).

Total number of collections taking place was 281, compared with 309 in the control period. There were significantly less BM harvests and significantly more cord units used. Collections were cryopreserved in 61% compared to 6% in 2019, with the majority of fresh collections happening early March.

Conclusions: Whilst activity was reduced comparatively to control period this data demonstrates a remarkable amount of activity continued despite Covid-19 disruption. Covid-19 resulted in postponement of 34.6% planned unrelated transplants in the UK and 21% of cancellations. The data shows trends in stem cell source with reduction in bone marrow harvests in keeping with the reduced access to theatre, as well as increased uptake of cord blood transplants than the previous year, suggesting favourable logistics with cord units. In line with recommendations there was widespread adoption of cryopreservation and it will be important to correlate this with engraftment and outcome data.

Disclosure: Nothing to declare.

Stem Cell mobilization, Collection and Engineering

O164. MGTA-145, in Combination With Plerixafor in A Phase 1 Clinical Study, Mobilizes Large Numbers of Hematopoietic Stem Cells And A Graft With Potent Immunosuppressive Properties

Kevin Goncalves¹, Sharon Hyzy¹, Katelyn Hammond¹, Patrick Falahee¹, Haley Howell¹, Jan Pinkas¹, Veit Schmelmer¹, Jonathan Hoggatt², David Scadden², Steven Devine³, Michael Rettig⁴, John DiPersio⁴, William Savage¹, John Davis¹

¹ Magenta Therapeutics, Cambridge, United States, ² Massachusetts General Hospital, Boston, United States, ³ Center for International Blood and Marrow Transplant Research, National Marrow Donor Program/Be The Match, Minneapolis, United States, ⁴ Washington University in St. Louis, St. Louis, United States

Background: Most hematopoietic stem cell (HSC) transplants are performed using peripheral blood mobilized by granulocyte-colony stimulating factor (G-CSF) given over 5 days, but this often results in a poor HSC yield, inconsistent engraftment, and in allogeneic HSC transplant, a significant incidence of acute and chronic graft-versus-host disease (GvHD).

Here, we show that MGTA-145, a CXCR2 agonist, when combined with plerixafor, a CXCR4 inhibitor, rapidly mobilizes large numbers of human HSCs for safe transplant after a single day of dosing and collection. Phenotypic and functional profiling demonstrate that these grafts have higher engraftment potential in NSG mice, are capable of gene modification, and lead to a marked reduction in xenogeneic GvHD compared to other graft sources.

Methods: MGTA-145 was given to healthy volunteers 2 hours after plerixafor, just prior to a 20 (13-20) L cell collection. Grafts were compared to 20 L collections of healthy volunteers mobilized with a 5-day regimen of G-CSF (commercially sourced) or a single dose of plerixafor (NCT00241358/NCT00914849). Hematopoietic cell subsets were profiled by flow cytometry. SCID-repopulating cell number was measured by limit dilution studies in primary and secondary NSG mouse recipients and GvHD was assessed in xenogeneic mouse models. Gene-editing was performed using CRISPR/Cas9 targeting β-2-microglobulin.

Results: In a Phase 1 healthy volunteer study, a peak of 40 CD34+ cells/μL were mobilized with MGTA-145+plerixafor (n = 12). 11/12 (92%) donors mobilized >20 CD34+ cells/μL compared to 8/14 (57%) achieving this cell target with plerixafor alone. A median of 4.0x10⁶ (1.5-7.0x10⁶) CD34+ cells/kg were collected after same-day mobilization/collection with MGTA-145 +plerixafor (Figure 1A, n = 8), compared to 5.3x10⁶ (5.3-7.6x10⁶) for G-CSF mobilized donors (not significant, n = 3) and 1.7x10⁶ (1.1-2.5x10⁶) for plerixafor mobilized donors (p < 0.05, n = 7). Notably, MGTA-145+plerixafor grafts contained significantly higher numbers of CD34+CD90+CD45RA- cells, a cell type enriched for HSCs (Figure 1B): 3-fold higher than that collected from G-CSF mobilized donors (p < 0.05, n = 3) and 4-fold higher than that collected from plerixafor mobilized donors (p < 0.01, n = 7). Mechanistically, MGTA-145 led to a modest, transient increase in plasma MMP- 9 with no change in neutrophil activation/degranulation. In limit dilution transplants (Figure 1C, n = 7-8 NSG mice/group), MGTA-145+plerixafor mobilized grafts led to 23-fold higher engraftment compared to G-CSF mobilized grafts (p < 0.001) and 11-fold higher engraftment versus plerixafor mobilized grafts (p < 0.001) from normal volunteers (n = 3-4 donors). MGTA-145+plerixafor CD34+ cells were edited with CRISPR/Cas9 (n = 2 donors) leading to robust long-term NSG engraftment (n = 8 mice) of gene-modified HSCs. Editing rates (~90%) were comparable between different graft sources.

B, T, and NK cell subsets were mobilized to comparable levels with MGTA-145+plerixafor or plerixafor alone, except CD8+ T cells, which at 0.2 (0.0-0.6) x10⁸/kg, was significantly lower than that mobilized by G-CSF or plerixafor. Notably, MGTA-145+plerixafor mobilized grafts resulted in significantly less GvHD than G-CSF (p < 0.01) or plerixafor (p < 0.001) grafts (Figure 1D, n = 3-6 donors/source) in a xenogeneic mouse model.

Conclusions: MGTA-145+plerixafor is a rapid, reliable, and G-CSF free method to obtain high numbers of HSCs capable of durable engraftment and gene-modification and a graft with potent immunosuppressive properties and is therefore an effective single-day mobilization/collection regimen for autologous and allogeneic HSC transplant.

Disclosure: Kevin Goncalves: employee and shareholder of Magenta Therapeutics.

Sharon Hyzy: employee and shareholder of Magenta Therapeutics.

Katelyn Hammond: employee and shareholder of Magenta Therapeutics.

Patrick Falahee: employee and shareholder of Magenta Therapeutics.

Haley Howell: employee and shareholder of Magenta Therapeutics.

Jan Pinkas: employee and shareholder of Magenta Therapeutics.

Veit Schmelmer: employee and shareholder of Magenta Therapeutics.

Jonathan Hoggatt: shareholder of Magenta Therapeutics.

David Scadden: advisory board member and shareholder of Magenta Therapeutics.

Steven Devine: Nothing to declare.

Michael Rettig: Nothing to declare.

John DiPersio: advisory board member and shareholder of Magenta Therapeutics.

William Savage: shareholder of Magenta Therapeutics.

John Davis: employee and shareholder of Magenta Therapeutics.

O165. Efficacy And Safety Of Pegfilgrastim For CD34⁺ Cell Mobilization In Healthy Volunteers

Hideki Goto¹, Koji Hayasaka², Kana Sunagoya², Rie Michimata², Mutsumi Nishida², Maki Jingu³, Yuki Ichihashi⁴, Mitsuhiko Odera⁴, Masayuki Hino⁵, Yoshinobu Maeda⁶, Masashi Sawa⁷, Takanori Teshima^8,2

¹ Hokkaido University, Graduate School of Medicine, Spporo, Japan, ² Hokkaido University Hospital, Sapporo, Japan, ³ Hokkaido University Hospital, Clinical Research and Medical Innovation Center, Sapporo, Japan, ⁴ Kyowa Kirin Co., Ltd, Tokyo, Japan, ⁵ Osaka City University Graduate School of Medicine, Osaka, Japan, ⁶ Okayama University Graduate School of Medicine, Okayama, Japan, ⁷ Anjo Kosei Hospital, Anjo, Japan, ⁸ Hokkaido University, Faculty of Medicine, Graduate School of Medicine, Spporo, Japan

Background: Generally, daily granulocyte colony-stimulating factor (G-CSF) is widely used for PBSC mobilization. On the other hand, daily G-CSF administration for a few days are big burden for healthy donors. Pegfilgrastim, a long-acting form of G-CSF, enables once-per-cycle injection after chemotherapy to reduce the chance of febrile neutropenia. PBSC mobilization with single-day injection of pegfilgrastim might have an advantage for scheduling and physical distress of healthy donor. However, it is still unclear whether pegfilgrastim can be adopted for PBSC mobilization. Here, the efficacy of pegfilgrastim on PBSC mobilization is evaluated in the first Japanese Phase 2, single-arm, single-center study, and the results are presented together with the safety results.

Methods: In pilot phase, safety and efficacy of pegfilgrastim (3.6, 7.2, and 10.8 mg in each dose escalation cohort) were evaluated in order to determine the recommended dose for evaluation phase. In evaluation phase, pegfilgrastim was subcutaneously administered on day 1. Peripheral blood CD34⁺ cells were counted daily from day 1 to 8, and day 15 using a flow cytometric analysis. The primary endpoint was the achievement of mobilization of peripheral blood CD34⁺ cells more than 20 (x10⁶/L) between day 1 and day 7 after pegfilgrastim administration.

Results: In pilot phase, 7.2 mg of pegfilgrastim was well tolerated and showed sufficient efficacy, and was determined to be the recommended dose. At the evaluation phase, 23 healthy volunteers were enrolled. Median age was 27 (range, 20 - 50) years. Median weight was 64.1 (range, 50.7 - 77.3) kg. Successful mobilization (CD34⁺ cells > 20 x10⁶/L) was achieved in all subjects (100%). In 21 of the 23 subjects (91.3%), the peak of the CD34⁺ cell counts were observed on day 5. The peak of the CD34⁺ cell of remaining 2 individuals were achieved on day 4 and day 6, respectively. Median value of CD34⁺ cells (x10⁶/L) on each day between day 4 and day 6 was 60.6 (range, 20.3 – 183.9), 84.2 (range, 22.6 – 254.2), and 49.9 (range, 20.0 – 221.5), respectively. White blood cell concentration peaked on day 4 of stimulation with a median 43.2 (range, 25.6 – 83.8) x10⁹/L. Common adverse events (AEs) were lactate dehydrogenase increased (100%), alkaline phosphatase increased (100%), back pain (87%), platelet count decreased (78%), uric acid increased (70%), headache (70%), aspartate aminotransferase increased (65%), alanine aminotransferase increased (57%) and arthralgia (17%). Severity of all adverse events was evaluated as mild. No serious AEs were observed.

[Allo PBSC mobilization outcomes in healthy volunteers.]

Conclusions: Our data indicated that a single dose of 7.2 mg pegfilgrastim on day 1 was able to mobilize CD34⁺ cells safely and effectively. Pegfilgrastim is a satisfying alternative to conventional daily G-CSF with less donor burden for PBSC mobilization.

Clinical Trial Registry: NCT03993639

Disclosure: Hideki Goto: No relationship to disclose.

Koji Hayasaka: No relationship to disclose.

Kana Sunagoya: No relationship to disclose.

Rie Michimata: No relationship to disclose.

Mutsumi Nishida: No relationship to disclose.

Maki Jingu: No relationship to disclose.

Yuki Ichihashi: Employment: Kyowa Kirin.

Mitsuhiko Odera: Employment: Kyowa Kirin.

Masayuki Hino: Corporate-sponsored research: Astellas, Daiichi Sankyo, Dainippon Sumitomo, Kyowa Kirin, Ono, Otsuka, Takeda, TEIJIN, Japan Blood Products Organization, JCR, SEKISUI, Abbott, TAIHO. Honoraria: Astellas, Chugai, Daiichi Sankyo, Dainippon Sumitomo, Eisai, Janssen, Kyowa Kirin, Novartis, Ono, Otsuka, Pfizer, Sanofi, Takeda, Japan Blood Products Organization, NIPPON SHINYAKU. Membership on an advisory board: Kyowa Kirin.

Yoshinobu Maeda: Corporate-sponsored research: Chugai, Nihon Shinyaku. Membership on an advisory board: Kyowa Kirin.

Masashi Sawa: Membership on an advisory board: Kyowa Kirin.

Takanori Teshima: Honoraria: Merck Sharp & Dohme, Kyowa Kirin, Celgene, Chugai, Novartis, Bristol-Myers Squibb, Janssen. Research funds under contract: Astellas, CIMIC, Novartis, Kyowa Kirin, Daiichi Sankyo, Chugai, NIhon servier, Takeda. Grants: Aiiku Hospital, Nippon Shinyaku, Eisai, Shionogi, Ono, Sumitomo Dainippon, The Japanese Society of Hematology, Chugai, Astellas, Kyowa Kirin, Fuji Pharma, The Japan Society for Hematopoietic Cell Transplantation. Membership on an advisory board: Kyowa Kirin.

Stem Cell Source

O166. Long Term Outcomes After Intrabone Unrelated Umbilical Cord Blood Transplant in Patients With Hematological Malignancies: A Eurocord, CTIWP-EBMT Analysis

Jacopo Peccatori¹, Fernanda Volt², Emanuele Angelucci³, Francesca Bonifazi⁴, Nicola Polverelli⁵, Nicola Mordini⁶, Fabio Giglio¹, Montserrat Rovira⁷, Franca Fagioli⁸, Fermin Sanchez- Guijo⁹, Luca Facchini¹⁰, Henrique Bittencourt¹¹, Anna Maria Raiola³, Adalberto Ibatici³, Riccardo Varaldo³, Chantal Kenzey², Vanderson Rocha², Francesco Frassoni¹², Fabio Ciceri¹, Eliane Gluckman², Annalisa Ruggeri²

¹ IRCCS San Raffaele Scientific Institute, Milano, Italy, ² Eurocord, Paris, France, ³ San Martino Hospital, Genova, Italy, ⁴ Sant’Orsola Hospital, Bologna, Italy, ⁵ Spedali Civili, Brescia, Italy, ⁶ Ospedale Civile, Cuneo, Italy, ⁷ Hospital Clinic, Barcelona, Spain, ⁸ Ospedale Infantile Regina Margherita, Torino, Italy, ⁹ IBSAL-Hospital Universitario de Salamanca, Salamanca, Spain, ¹⁰ Arcispedale Santa Maria Nuova, Reggio Emilia, Italy, ¹¹ CHU St Justine, Montreal, Canada, ¹² Gaslini Hospital, Genova, Italy

Background: Cord blood (CB) is readily available, allowing flexibility on HLA-matching, but is associated with delayed hematopoietic recovery and graft failure. To overcome the cell-dose limitation, single CB delivered by intrabone-injection (IB) directly to the bone marrow space has been reported safely and effective.

Methods: We Aim to describe IB-UCBT outcomes. We report a retrospective study of first unrelated IB-UCBT for hematological malignancies performed in 15 Eurocord/EBMT-centers.

Results: 223 patients transplanted from 2006 to 2019 were included. Median year to IB-UCBT was 2010. Median follow-up was 9 years. 30 patients had previous allogeneic-transplant. Median age was 41.4 years(94% adults). Diagnosis was acute leukemia (n = 176; 74%), lymphoproliferative disease (n = 22; 9%) or MDS/MPD (n = 28; 12%) and 61% were transplanted in disease remission (CR). Most patients received myloablative conditioning (n = 178; 76%) and the most frequent was cyclophosphamide + total-body-irradiation, followed by busulfan, thiotepa and fludarabine, or treosulfan based regimen. ATG-free protocols were preferentially applied after 2010. At cryopreservation median TNC was 3.3x107/Kg, and CD34+ was 1.43x105/Kg.

UCB units were match at 5/6 and 4/6 HLA loci in 30% and 72%, respectively (7 cases were 6/6).

Cumulative incidence function(CIF) of neutrophil recovery at 60-days was 79% (95%CI 74 - 85) in a median time of 24 days. No factors were associated with the risk of engraftment in the multivariate analysis (MVA). CIF of day-180 platelet engraftment>20x10e9/L was 63% (95%CI 57-70); median time for platelet engraftment was 37 days. Day-100 CIF of acute-GVHD was 18%. GVHD was mainly grade 2 (n = 29), grade 3 and 4 being reported in 7 and 2 cases, respectively. The 9-year CIF of chronic-GVHD was 37%, with 17 patients experiencing an extensive grade. In the multivariate analysis UCBT performed in more recent years had a lower risk of cGVHD (HR 0.85, p = 0.003). For patients transplanted for acute leukemia CIF of 9-year relapse was 34% (95%CI 27-42), being 40% (95%CI 28-57) for ALL and 31% (95%CI 24-41) for AML respectively. In the MVA patients in CR at the time of transplant had a lower rate of relapse (HR 0.44, 95%CI 0.24-0.76, p = 0.003). 9-year overall survival (OS) was 30% and disease-free-survival (DFS) was 29% (at 5-years, OS and DFS were 33% and 30%, respectively). CIF of 100-day NRM was 22%, 158 patients died (infection (n = 51) and GVHD (n = 9) were the most frequent cause of NRM). In MVA, patients undergoing IB-CBT for MDS/MPD (HR 0.40, p = 0.004) had higher OS than leukemias. Other factors associated with increased survival were female gender (HR 0.63; p = 0.006) disease remission at transplant (HR 0.679; p = 0.036) and no history of previous transplant (HR 0.4; p < 0.001). In a subgroup of patients receiving ATG free regimen and with available information on immune-reconstitution at day +90 median CD3+ was 270/mL (18-2131) and CD4+ was 240/mL (13-898), with a normal CD4+/CD8+ ratio.

Conclusions: UCBT infused via intrabone is safe, provide durable disease control in patients with high-risk hematological malignancies. The use of ATG-free-regimen and better supportive therapy may further improve results. The IB injection of UCB ex-vivo-expanded is attractive to enhance homing of expanded cord-blood derived stem cells.

Disclosure: The authors declare no COI related to this study.

O167. Impact of Covid-19 Pandemic on Release of French Cord Blood Units. on Behalf The Agency of Biomedicine, Eurocord And The SFGM-TC

Hanadi Rafii - El Ayoubi^1,2, Irina Ionescu³, Annalisa Ruggeri^4,1, Federico Garnier³, Caroline Ballot⁵, Danièle Bensoussan⁶, Christian Chabannon⁷, Bernard Dazey⁸, John De Vos⁹, Eric Gautier¹⁰, Christine Giraud¹¹, Jerome Larghero¹², Audrey Cras¹², Valérie Mialou¹³, Virginie Persoons¹⁴, Fabienne Pouthier¹⁵, Jean-Baptiste Thibert¹⁶, Jean-Hugues Dalle¹⁷, Gerard Michel¹⁸, Fernanda Volt^1,2, Chantal Kenzey^1,2, Vanderson Rocha^19,1, Jacques-Olivier Bay²⁰, Marie-Thérèse Rubio²¹, Catherine Faucher³, Evelyne Marry³, Eliane Gluckman^1,2

¹ Eurocord, Hopital Saint Louis APHP, Institut de Recherche de Saint-Louis (IRSL) EA 3518, Université de Paris, Paris, France, ² Monacord, International Observatory on Sickle Cell Disease, Centre Scientifique de Monaco, Monaco, Monaco, ³ Agency of Biomedecine, Saint Denis La Plaine, France, ⁴ IRCCS San Raffaele Scientific Institute, Haematology and Bone Marrow Transplant Unit, Milan, Italy, ⁵ Etablissement Français du Sang Hauts de France Normandie, site de LILLE – Belfort, Cell Therapy Lab, Lille, France, ⁶ Regional University Hospital, Tissue Engineering and Cell Therapy Unit, Nancy, France, ⁷ Paoli-Calmettes Institute, Inserm CBT 1409, Marseille, France, ⁸ Etablissement Français du Sang, Cell Therapy Unit, Bordeaux, France, ⁹ University Hospital, Cell Therapy Unit, Montpellier, France, ¹⁰ Cell Therapy unit, Etablissement Français du Sang, Créteil, France, ¹¹ Etablissement Français du Sang, University Hospital, Poitiers, France, ¹² AP-HP, Hôpital Saint Louis, Cell Therapy Unit and Cord Blood Bank, Paris, France, ¹³ Etablissement Français du Sang, Hopital E. Herriot, Cell Therapy Unit, Lyon, France, ¹⁴ Cell Therapy and Tissue Engineering unit, Etablissement Français du Sang, Grenoble, France, ¹⁵ Etablissement Francais du Sang, Cell and Tissue Engineering Unit, Besançon, France, ¹⁶ Etablissement Français du Sang, Cell Therapy Unit, Rennes, France, ¹⁷ Hopital Robert Debré, Assistance Publique - Hôpitaux de Paris, Paris, France, ¹⁸ Aix-Marseille University and La Timone Children’s Hospital, Marseille, France, ¹⁹ Clinics Hospital, University of São Paulo Medical School, São Paulo, Brazil, ²⁰ Clermont University, Clermont-Ferrand, France, ²¹ Regional University Hospital, Nancy, France

Background: The emergence of COVID-19 as a pandemic in early 2020 has impacted every single aspect of stem cell transplant (HCT) practice around the world. Transplant physicians were forced to face major challenges to maintain access to urgent transplants.

National lockdown, frontiers closing, flight and train circulation restrictions between countries as well as within France were established to control COVID-19 spread, making more challenging stem cells procurement from international and national donors. In addition, the transplant teams had to tackle the unpredicted situations related to donor (and harvest center staff) risk of infection/exposure which were likely to have an impact in stem cell clearance for donation.

Due to the emergence of COVID-19, many associations like WMDA and EBMT and national stem cell registries published recommendations to maintain safe access to urgent HCT. They recommended to cryopreserve all related and unrelated donor products and give preference to domestic donors /cord blood over international ones.

This study aims to retrospectively review the impact of the Covid-19 pandemic on the use of umbilical cord blood units (UCB) facilitated by the French Network.

Methods: We have reviewed the available data related to collection and utilization of unrelated UCB during the first 10 months of 2020 and compared them to data over the same period in 2019.

Results: From January to October 2020, 3023 UCB were collected by the French Network, compared to 6130 units in 2019; there was 50% decrease in the collection activity in 2020 underscoring a global reduction in UCB donation. The impact was most prominent during the 2-month lockdown period when UCB harvesting activity was totally stopped.

There was a dramatic increase in the utilization rate of UCB for recipients in France reaching 15%, with the use of 82 UCB in 2020 (vs 71 in 2019). Of these units, 34 were released from the French Network inventory to transplant 30 national recipients (compared to 22 units to transplant 20 patients in 2019) corresponding to 55% increase in utilization of French UCB and 50% increase in UCBT using French units. In addition, there was a mild impact on importing unrelated stem cells for recipients in France with 2% decrease in import activity of UCB (48 in 2020 vs 49 in 2019).

The number of French UCB released for international patients was 53 from January to October 2020, while it was 66 in 2019. International shipment of French UCB decreased 20% during the pandemic, highlighting the challenges to export stem cell products to recipients outside France due to borders closing.

Conclusions: UCB revealed to be a valuable resource during COVID-19 pandemic to transplant urgent patients, since UCB units are cryopreserved, and completely tested and ready for immediate use. Majority of the units were used to transplant national recipients due to difficulties to export stem cells across borders. All collected units had not been exposed to SARS-COV2. The possibility of skipping the adult donor work-up with UCB use is extremely important, to reduce the risk of infections and the access to hospital in the pandemic regions.

Disclosure: Nothing to declare.

O168. Inferior Clinical Outcomes in Recipients of Cryopreserved Grafts Following Reduced Intensity Allogeneic Hematopoietic Cell Transplantation: A Single Center Report

Andriyana Bankova¹, Joseph Caveney¹, Teresa Ramos¹, Jan Bögeholz¹, Kartoosh Heydari¹, Nery Diaz¹, Marin Jackson¹, Robert Lowsky¹, Janice Brown¹, Laura Johnston¹, Andrew Rezvani¹, Matthew Frank¹, Lori Muffly¹, Wen-Kai Weng¹, Surbhi Sidana¹, Robert Negrin¹, David Miklos¹, Parveen Shiraz¹, Everett Meyer¹, Judith Shizuru¹, Sally Arai¹

¹ Stanford University, Stanford, United States

Background: In 2020 cryopreservation of allogeneic donor apheresis products was implemented to overcome the challenges of donor availability and product transport related to the COVID-19 pandemic. Several studies have reported on allo-HCT outcomes with cryopreserved (cryo) grafts with conflicting findings regarding overall survival (OS), engraftment and hematopoietic recovery. Here, we report on the clinical post-HCT outcome in recipients of cryo grafts in our institution. Moreover, we compare the graft composition of cryo and fresh products by assessing the immunophenotype of hematopoietic stem cell (HSC) and lymphocyte subsets.

Methods: Data collected from 29 patients who underwent allo-HCT with cryo PBSC (cryo group) from 03 - 08/2020 were compared to a control cohort of 60 consecutive patients, who received fresh PBSC allo-HCT (fresh group) from 06/2019 to 08/2020. Primary endpoints were OS and graft failure. Secondary endpoints were hematopoietic recovery and GvHD. Flow cytometry was performed on 5 available samples in the cryo group and compared to 4 prospectively collected fresh apheresis samples from allogeneic donors.

Results: Conditioning intensity was equally distributed among both groups (cryo – 69% RIC vs 31% MAC; fresh – 65% RIC vs 35% MAC). Median post-collection (pre-freeze) dose of CD34+ cells x106/kg differed in the two groups (7.3 cryo vs 9.2 fresh) with no differences in the dose of CD3+ cells x108/kg. All donor types, except UCB were included in the analysis. The cryo group included more HLA-matched unrelated donors (72.4 vs 41.7%), whereas the fresh group included more HLA-matched sibling (33.3 vs 6.9%) and slightly more haplo/HLA-mismatched donors (25.0 vs 20.6%). 4 of 29 patients (13.8%) in the cryo group developed primary graft failure vs 1 of 60 patients (1.7%) in the fresh group (p = 0.03). All 4 cryo patients received HLA-matched unrelated grafts following RIC (3=Flu/Mel, 1=Flu/TBI), whereas the patient in the fresh cohort received a haploidentical graft, following MAC (Bu/Cy). Donor chimerism at 1 month for the RIC allo-HCT is shown in Fig. 1. We observed inferior OS in the cryo RIC group [Fig. 2; HR and 95% CI 4.20 (1.22, 14.4)], and slower recovery of neutrophils (p = 0.006) and platelets (p = 0.058). The incidence of acute GvHD was similar in the two groups. For all patients (n = 89), multivariate analysis performed for OS and graft failure after adjusting for donor type, patient age, gender, CD34 cell dose, and conditioning intensity supported the inferiority of cryo allo-HCT.

Finally, flow cytometry analysis revealed significantly lower absolute counts of NK cells (CD3-CD56+) in cryo as compared to fresh apheresis samples (p = 0.0159), with no differences in CD4+ and CD8+ T-, B- (CD19+CD20+), CD34+ cells and HSC (CD34+CD38-CD90+CD45RA-).

Conclusions: Our data show that the use of cryopreserved grafts following RIC allo-HCT led to inferior OS with increased rate of graft failure. Further studies are needed to determine a potential role of graft composition on cryo vs fresh allo-HCT outcomes.

Disclosure: Janice (Wes) Brown: Research Funding from Merck and Ansun; Lori Muffly: consultant for Amgen and Pfizer, Research Funding from Adaptive; Surbhi Sidana: consultant for Jansen, Research Funding from Magenta Therapeutics; Parveen Shiraz: Research Funding from Kite and Orca Bio; Judith Shizuru: Board of Directors and Chair of Scientific Advisory Board of Jasper Therapeutics, Inc.

O169. Clinical Results of Peripheral Blood Stem Cell Transplantation With Cryopreserved Grafts During The Covid19 Pandemic

Maximilian Christopeit^1,2, Anita Badbaran¹, Ulrike Fritsche-Friedland¹, Petra Freiberger¹, Maria Geffken¹, Sven Peine¹, Nicolaus Kröger¹

¹ University Medical Center Hamburg-Eppendorf, Hamburg, Germany, ² University Hospital Tübingen, Tübingen, Germany

Background: Cryopreservation of peripheral blood stem cell (PBSC) grafts became necessary during the ongoing COVID-19 pandemic. This followed international guidelines. Cryopreservation has been described as feasible but to be viewed with caution, e.g. when patients suffering severe aplastic anemia are transplanted.

Methods: We here describe recovery rates and engraftment kinetics from 16 PBSC grafts after cryopreservation for a median of 9 (range 6-49) days.

Results: Stem cell recovery (7 (median, range 2.7-11.9) x106 CD34+ cells before versus 0.71 (0.2-4.5), p < 0.001 after cryopreservation) and vitality of the recovered CD45+ and CD34+ cells, measured as 7AAD negative cell fraction, was drastically impaired. When compared to results of 37 patients transplanted with PBSC grafts without prior freezing – thawing, time to leukocyte and time to thrombocyte engraftment in the patients was not different. Hence, donor chimerism was statistically significant less 30 days after transplantation. This effect was lost after 100 and after 120 days when similar chimerism results could be measured. Respective values are summarized in table 1.

	Cryopreserved PBSC	PBSC without cryopreservation	p-value
Patient number	15	37
CD34+/kg BW	0.71 (0.2-4.5)	7 (2.7-11.9)	<0.001
Patients with/ without leukocyte engraftment	15/ 0	37/ 0	1
Leukocyte engraftment median (range), days	13 (9-17)	11 (7-21)	0.611
Patients with/ without thrombocyte engraftment	11/ 4	33/ 4	0.151
Thrombocyte engraftment median (range), days	15 (10-24)	13 (9-62)	0.708
Chimerism day 30 median (range), %	99.8 (90.4-99.9)	99.9 (94.6-100)	0.025
Chimerism day 100 median (range), %	99.8 (83.6-100)	99.9 (56.1-100)	0.053
Chimerism day 120 median (range), %	99.9 (85.2-99.9)	99.9 (43.1-100)	0.142

Conclusions: Cryopreservation of PBSC grafts is feasible but has to be viewed with caution and might be restricted to patients for who the graft under consideration stems from the only suitable donor available.

Disclosure: Nothing to declare.