In the CORAL study, 255 chemosensitive relapses with diffuse large B-cell lymphoma (DLBCL) were consolidated with autologous stem cell transplantation (ASCT), and 75 of them relapsed thereafter. The median time between ASCT and progression was 7.1 months. The median age was 56.1 years; tertiary International Prognosis Index (tIPI) observed at relapse was 0–2 in 71.6% of the patients and >2 in 28.4%. The overall response rate to third-line chemotherapy was 44%. The median overall survival (OS) was 10.0 months (median follow-up: 32.8 months). Thirteen patients received an allogeneic SCT, and three a second ASCT. The median OS was shorter among patients who relapsed <6 months (5.7 months) compared with those relapsing ⩾12 months after ASCT (12.6 months, P=0.0221). The median OS in patients achieving CR, PR or no response after the third-line regimen was 37.7 (P<0.0001), 10.0 (P=0.03) and 6.3 months, respectively. The median OS varied according to tIPI: 0–2: 12.6 months and >2: 5.3 months (P=0.0007). In multivariate analysis, tIPI >2, achievement of response and remission lasting <6 months predicted the OS. This report identifies the prognostic factors for DLBCL relapsing after ASCT and thus helps to select patients for experimental therapy.
Salvage second-line chemotherapy followed by high-dose therapy and autologous stem cell transplantation (ASCT) is the standard of care for relapsed/refractory diffuse large B-cell lymphoma (DLBCL). In the CORAL study, two salvage regimens (R-DHAP (rituximab, dexamethasone, cytarabine, cisplatinum) or R-ICE (rituximab, ifosfamide, carboplatinum, etoposide)) were compared in 477 patients, and no difference was found. Only 50% of the patients could proceed to per protocol ASCT, afterwards they were randomly assigned to rituximab or observation. The 4-year event-free survival post ASCT was 52% and 53% for the rituximab and observation groups, respectively (P=0.7). International Prognosis Index (IPI) at relapse, that is, secondary IPI (sIPI), independently predicted event-free survival, PFS and OS. Patients with DLBCL relapsing <12 months after rituximab-containing first-line therapy were also identified as a subgroup with dismal outcome.1, 2
In DLBCL patients who fail second-line therapies, third-line strategies are not well established. The CORAL investigators reported on the outcome of patients who could not proceed to CORAL-scheduled ASCT, mainly because of treatment failure.3 The results indicated that in ~30% of these patients, particularly those with a lower tertiary IPI (tIPI), response to third-line therapy and the possibility to consolidate with ASCT, long-term disease control could be achieved. In the remaining patients, the outcome was dismal, and innovative strategies are awaited. There are few registry data in the rituximab era for relapse post ASCT, with a reported median overall survival (OS) of 9.9 months in rituximab-pretreated patients.4 The CORAL study provided a unique opportunity to look at 75 patients who relapsed after BEAM/ASCT and to shed more light on the outcome and prognostic factors in this population who had an initial response to first salvage treatment.
Patients and methods
Study design and patient selection
The CORAL study was a phase III, multicentre, randomised trial that compared the efficacy of three R-ICE or R-DHAP cycles, followed by ASCT with or without rituximab maintenance in patients aged 18–65 years with relapsed DLBCL. Details of the inclusion criteria, treatment, assessment of response and monitoring have been reported previously.1, 2 In total, between July 2003 and June 2008, 477 patients were randomly assigned to R-ICE (n=243) or R-DHAP (n=234). A total of 255 patients who achieved CR, PR or stable disease after the third cycle of salvage treatment with adequate stem cell collection received consolidation with BEAM (carmustine, etoposide, cytarabine and melphalan), followed by ASCT. Among them, 244 patients were further randomised between rituximab maintenance (n=122) or observation (n=120). The present report focuses on the 75 patients included in the CORAL study who relapsed after ASCT either before (n=4) or after randomisation between rituximab (n=37) and observation (n=34; Figure 1). The CORAL study was registered under EudraCT (no. 2004-002103-32) and ClinicalTrials.gov (no. NCT 00137995). The Lymphoma Academic Research Organization (LYSARC) monitored the data centrally. All patients had given written informed consent both to participate and to provide tissue material for biologic studies.
The investigators were asked to update their patients’ status. The patient characteristics at relapse, time between ASCT and failure, type of third-line regimen, response to third-line treatment, response by the CORAL maintenance arm (rituximab or observation) and OS were retrospectively collected after an amendment. IPI was calculated based on the characteristics at relapse post ASCT (tIPI). Third-line treatments were grouped into major categories (see Van Den Neste et al.3 for a full description). All patients had been previously exposed to rituximab in the first- and/or second-line treatment and data on use of rituximab in the third-line was not recorded.
Evaluations and analyses
Patients’ response was assessed by the investigators using conventional methods that included CT scans according to the International Working Group criteria.5 PFS and event-free survival after the third-line treatment were not considered because it was anticipated that these results would be less reliable; indeed, the initial CORAL report showed that over 90% of deaths were lymphoma related.1 Thus, only the OS is given, defined as the time from the date of relapse after ASCT until death. The median follow-up was 32.8 months (range 24.3–45.8 months).
Material for histology was only available in a subset of patients to assess tumour biology. Immunohistochemistry and fluorescent in situ hybridisation were performed as previously reported and were centrally reviewed.6 The cell of origin was defined according to the algorithm published by Hans et al.7
The Kaplan–Meier method was used to estimate the OS. The Wilcoxon's signed-rank test or χ2-test was used to compare the patient characteristics. Cox regression analysis was used to calculate the hazard ratio (HR) between different patient categories. All reported P-values are two-sided, and P-values <0.05 were considered significant. All statistical analyses were carried out using the SAS 9.2 software (SAS Institute, Cary, NC, USA).
In total, 75 patients relapsed after CORAL-scheduled ASCT (Table 1). The median time between ASCT scheduled in CORAL and relapse was 7.1 months (range 3.2–61.9), with 32.9% of patients relapsing ⩾12 months. The median age at relapse was 56.1 years (range 20.9–67.7), and the sex ratio was 51 males/24 females. For the patients in the second relapse, tIPI could be determined and was 0–2 in 71.6% and >2 in 28.4%. Of these patients, 49.3% were in the rituximab arm, and 45.3% were in the observation arm of the CORAL study; 5.3% relapsed before the second randomisation (Figure 1). All patients had previously received rituximab. Third-line therapy consisted of ICE-type (17.3%), DHAP-type (24%), gemcitabine-containing (28%), CHOP-like (13.3%) and miscellaneous regimens (17.3%).
The immunohistochemical expression of BCL2 and c-MYC in tumour cells was observed in 9/46 (19.6%) and 2/15 (13.3%) of the evaluable patients, respectively. Among the tumour samples displaying interpretable fluorescent in situ hybridisation signals, BCL2/18q21 and c-MYC/8q24 gene rearrangements were found in 11/26 (42.3%) and 3/27 (11.1%), respectively, with only one double hit. Using the algorithm of Hans,7 36.4% (16/44) of the patients were classified as germinal centre B (GCB)-cell like, and 63.6% (28/44) were classified as non-germinal centre B (not shown).
Response to third-line chemotherapy
The overall response rate to third-line chemotherapy, based on local investigator assessment, was 44%, with 32% CR/CR unconfirmed (CRu) and 12% PR. No significant difference among the various salvage regimens was observed. For patients previously treated into the R-ICE arm of the CORAL study, the following chemotherapies (+/− rituximab) were used as the third-line: ICE (n=2), DHAP (n=18), gemcitabine-containing (n=9), CHOP-like (n=6) and other (n=9). For patients treated in the R-DHAP arm of the CORAL study, regimens (+/− rituximab) used as the third-line were: ICE (n=11), gemcitabine-containing (n=12), CHOP-like (n=4) and other (n=4). Chemotherapy switching from ICE to DHAP, and oppositely from DHAP to ICE-yielded CR/PR rates of 35.3%/17.6% and 36.4%/18.2%, respectively.
Among the 75 patients, 16 (21.6%) could eventually be transplanted again, three patients received an autologous SCT and 13 an allogeneic SCT, with various conditioning regimens, which are detailed in Table 1. The intensity of conditioning regimens for allogeneic SCT was reduced in in nine patients and myeloabative in four of them. Among the transplanted patients, 10 were in CR/CRu, one was in PR and 4 were in stable or progressive disease at the time of transplantation; the remaining patient was not evaluable.
The median OS was 10.0 months for the entire population with an estimated 1-year OS of 39.1% (Figure 2). The median OS was significantly different according to IPI at CORAL failure: tIPI 0–2: 12.6 months (1-year OS 51.3%) compared with tIPI >2: 5.3 months (1-year OS 21.1%, hazard ratio (HR) 2.805, 95% confidence interval (CI) 1.510–5.210, P=0.0007; Figure 3).
The median OS was particularly dismal among patients who relapsed <6 months after CORAL-scheduled ASCT (5.7 months, n=28) compared with those relapsing either ⩾6 or <12 months (median OS: 11.3 months, n=21, HR 0.407, 95% CI 0.206–0.806, P=0.0099) or ⩾12 months after ASCT (median OS: 12.6 months, n=24, HR 0.467, 95% CI 0.244–0.896, P=0.0221; Figure 4).
The median OS in patients achieving CR/CRu or PR after third-line regimen was 37.7 months (HR 0.132, 95% CI 0.059–0.294, P<0.0001) or 10.0 months (HR 0.375, 95% CI 0.152–0.929, P=0.034), respectively, compared with 6.3 months in those with no response (Figure 5). When patients were analysed separately according to the type of third-line salvage regimen, no significant difference was found (not shown).
The median OS of patients who could eventually be transplanted (after a median follow-up of 32.7 months, range: 24.3–55.2 months) was 17.4 months (1-year OS 68.2%) compared with a median OS of 8.0 months in those who were not transplanted (1-year OS 31.2%), with a HR of 0.575, but this difference was not statistically significant (P=0.1106; Figure 6). There was no difference (HR 1.035, P=0.9654) according to the type of transplantation performed (ASCT or allogeneic SCT, not shown). Importantly, the characteristics of patients who could or could not be transplanted were compared. The only difference was the response to third-line regimen with 21/58 (36.2%) non-transplanted patients in CR/CRu/PR compared with 11/16 (68.7, P=0.024) in the transplanted group. Other characteristics (age, sex, arm of second randomisation, disease-free survival from ASCT, tIPI and type of salvage) were similar in both groups.
The OS according to prognostic factors is depicted in Table 2. Of note, the tumour biology was included in neither the univariate nor multivariate analysis because of missing data. There was one double hit Bcl2 Myc rearrangement and two double expressors. Transplantation was not included in the model because of the small number of patients and closely related to CR/PR. It was moreover not significantly different. In multivariate Cox analysis (for age, sex, tIPI, response to third-line regimen and time between CORAL ASCT and relapse), tIPI >2 (HR 2.464, P=0.0139), achievement of CR/CRu (HR 0.104, P<0.0001) or PR (HR 0.242, P=0.0186), and post-ASCT disease-free interval (DFI) lasting <6 months (HR 2.270, P=0.0497) independently predicted for OS. This analysis was performed on 57 patients for whom all data were available.
In this report, we retrospectively analysed the outcome of DLBCL patients in the second relapse after ASCT. All patients had been prospectively included in the CORAL study at the time of their first relapse and uniformly treated with DHAP or ICE combined with rituximab. Patients who could not proceed to CORAL-planned ASCT were excluded; these patients have been previously reported.3 The present analysis thus focuses on the 75 patients who were in CR/PR with adequate stem cell collection after R-DHAP/ICE, underwent ASCT and eventually relapsed. The objective was to define the efficacy of the third-line regimen and prognostic factors for a better outcome in this population, for which a standard strategy is lacking and the prognosis is notoriously dismal. One of the goals was to set an historical standard comparator for the development of new approaches.
The ORR to third-line salvage was 44%, including 32% CRs. We previously reported an ORR of 38% (CR 24%) in CORAL-included patients who could not proceed to the transplantation because of refractoriness to the first salvage chemotherapy. We showed that this ORR could be achieved, at least partially, by shifting from DHAP-type to ICE-type and vice versa and similar results were observed here.3 The nature of our study does enable a true comparison between different salvage regimens. However, our results show that even if this proportion is reduced, a significant proportion of patients with DLBCL in second relapse after ASCT remain chemosensitive.
The OS of the entire population was 10 months, similar to that reported by Nagle et al.4 (9.9 months) in a retrospective analysis of 56 patients with disease progression after ASCT. Although limited, this OS appears to be better than initially reported in the pre-rituximab era (3 months).8 This observation matches the recent finding that for patients who respond to salvage chemotherapy and manage to proceed to ASCT after a failure of first-line immunochemotherapy, ASCT is at least as effective as in patients who had failed chemotherapy in the preceding decade.9, 10 A subgroup analysis indicated three independent factors that were predictive of a significantly improved outcome: low or low-intermediate tIPI at relapse, progression occurring ⩾6 months after ASCT and chemosensitivity to third-line salvage. Patients who relapsed<6 months after CORAL transplantation had a median outcome (OS 5.7 months) in the range of those refractory to CORAL salvage (OS 3.6 months), and should thus be considered similarly to the latter. The time to progression following ASCT has been identified as a factor impacting OS in previous reports in similar patients.4, 11 Our results thus indicate that in selected DLBCL patients relapsing after ASCT, potentially on the basis of these latter characteristics, implementing a classical strategy with third-line salvage chemotherapy and attempting to achieve a response could remain a valid option, especially when an immune consolidation with allogeneic transplantation is possible. Conversely, patients with post-ASCT relapse bearing high/intermediate–high tIPI and/or short post-ASCT DFI could be eligible for experimental strategies.
In our study, the actual role of consolidative allogeneic SCT is difficult to define because most (69%) of the transplanted patients were in CR/CRu/PR at transplantation. The disease status has indeed been reported to be a major parameter affecting the outcome at the time of allogeneic SCT as second transplant.11, 12, 13 Overall, allogeneic SCT in DLBCL has been shown to be associated with durable disease control and graft-versus-lymphoma effect, as demonstrated by the role of immunotherapeutic intervention post allo-SCT.14, 15, 16 Although prior ASCT may adversely affect PFS, encouraging OS has been reported for allogeneic SCT as second transplant procedure in DLBCL after various conditioning regimens with various donor types.12, 17, 18, 19 Recently, the CIBMTR reviewed 503 patients with allogeneic SCT (25% myeloablative conditioning) after experiencing disease progression or relapse following prior ASCT. A prognostic model using 3 factors was proposed: (i) Karnofsky score <80 (2 points), (ii) interval between ASCT and allogeneic SCT <1 year (1 point), and (iii) chemoresistant disease at allogeneic SCT (2 points). Only patients with low risk (0-1 point) benefitted from acceptable PFS and OS (38% and 43% at 3 years, respectively).20 In the present report, we observed long-term disease control and a potential plateau in ~30% of patients who could receive allogeneic SCT, similar to what was observed in registry data (9,11). There was no specific recommendation in the CORAL protocol for the management of failures, and the decision to perform allogeneic SCT was thus left to the investigators‘ choice.
When analysed critically, our data show that the majority (~60%) of patients with DLBCL relapsing after ASCT will not respond to salvage therapy, thus profoundly jeopardising survival and emphasising the need for better salvage strategies. It is unlikely that improvement may come from chemotherapy-only strategies, even by switching to drugs with non-overlapping resistance mechanisms. Hopefully, novel strategies, including agents targeting oncogenic drivers in DLBCL or chimeric T cells, will allow more tailored and effective strategies (as reviewed in references 21, 22, 23). In our study, when analysed for 44/75 patients, we found a majority of non-GC subtypes, which suggests that the pharmacological inhibition of BCR signalling might be a promising therapeutic option. The ongoing LYSA trial (NCT02055924) combining DHAP with ibrutinib in relapsed/refractory DLBCL is illustrating this strategy.
In conclusion, we analysed the outcome of DLBCL patients who relapsed after ASCT. All of the patients had been prospectively included in the CORAL study at the second relapse and uniformly treated prior to ASCT. All patients had received prior rituximab. We identified two prognostic factors (post-ASCT disease-free interval and tIPI) that could help clinicians to select an adequate strategy. We also confirm the importance of achieving a response before proceeding to further treatment including transplantation. Future progress will hopefully come from a better understanding of the molecular characteristics of relapsing patients, This series could be used as a comparator to evaluate more active salvage and consolidation modalities.
Gisselbrecht C, Glass B, Mounier N, Singh Gill D, Linch DC, Trneny M et al. Salvage regimens with autologous transplantation for relapsed large B-cell lymphoma in the rituximab era. J Clin Oncol 2010; 28: 4184–4190.
Gisselbrecht C, Schmitz N, Mounier N, Singh Gill D, Linch DC, Trneny M et al. Rituximab maintenance therapy after autologous stem-cell transplantation in patients with relapsed CD20(+) diffuse large B-cell lymphoma: final analysis of the collaborative trial in relapsed aggressive lymphoma. J Clin Oncol 2012; 30: 4462–4469.
Van Den Neste E, Schmitz N, Mounier N, Gill D, Linch D, Trneny M et al. Outcome of patients with relapsed diffuse large B-cell lymphoma who fail second-line salvage regimens in the International CORAL study. Bone Marrow Transplant 2015; 51: 51–57.
Nagle SJ, Woo K, Schuster SJ, Nasta SD, Stadtmauer E, Mick R et al. Outcomes of patients with relapsed/refractory diffuse large B-cell lymphoma with progression of lymphoma after autologous stem cell transplantation in the rituximab era. Am J Hematol 2013; 88: 890–894.
Cheson BD, Horning SJ, Coiffier B, Shipp MA, Fisher RI, Connors JM et al. Report of an international workshop to standardize response criteria for non-Hodgkin's lymphomas. NCI Sponsored International Working Group. J Clin Oncol 1999; 17: 1244.
Thieblemont C, Briere J, Mounier N, Voelker HU, Cuccuini W, Hirchaud E et al. The germinal center/activated B-cell subclassification has a prognostic impact for response to salvage therapy in relapsed/refractory diffuse large B-cell lymphoma: a bio-CORAL study. J Clin Oncol 2011; 29: 4079–4087.
Hans CP, Weisenburger DD, Greiner TC, Gascoyne RD, Delabie J, Ott G et al. Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray. Blood 2004; 103: 275–282.
Vose JM, Bierman PJ, Anderson JR, Kessinger A, Pierson J, Nelson J et al. Progressive disease after high-dose therapy and autologous transplantation for lymphoid malignancy: clinical course and patient follow-up. Blood 1992; 80: 2142–2148.
Robinson SP, Boumendil A, Finel H, Blaise D, Poire X, Nicolas-Virelizier E et al. Autologous stem cell transplantation for relapsed/refractory diffuse large B-cell lymphoma: efficacy in the rituximab era and comparison to first allogeneic transplants. A report from the EBMT Lymphoma Working Party. Bone Marrow Transplant 2015; 51: 365–371.
Redondo AM, Pomares H, Vidal MJ, Pascual MJ, Quereda B, Sancho JM et al. Impact of prior rituximab on outcomes of autologous stem-cell transplantation in patients with relapsed or refractory aggressive B-cell lymphoma: a multicentre retrospective Spanish group of lymphoma/autologous bone marrow transplant study. Br J Haematol 2014; 164: 668–674.
van Kampen RJ, Canals C, Schouten HC, Nagler A, Thomson KJ, Vernant JP et al. Allogeneic stem-cell transplantation as salvage therapy for patients with diffuse large B-cell non-Hodgkin's lymphoma relapsing after an autologous stem-cell transplantation: an analysis of the European Group for Blood and Marrow Transplantation Registry. J Clin Oncol 2011; 29: 1342–1348.
Rigacci L, Puccini B, Dodero A, Iacopino P, Castagna L, Bramanti S et al. Allogeneic hematopoietic stem cell transplantation in patients with diffuse large B cell lymphoma relapsed after autologous stem cell transplantation: a GITMO study. Ann Hematol 2012; 91: 931–939.
Sirvent A, Dhedin N, Michallet M, Mounier N, Faucher C, Yakoub-Agha I et al. Low nonrelapse mortality and prolonged long-term survival after reduced-intensity allogeneic stem cell transplantation for relapsed or refractory diffuse large B cell lymphoma: report of the Societe Francaise de Greffe de Moelle et de Therapie Cellulaire. Biol Blood Marrow Transplant 2010; 16: 78–85.
Bacher U, Klyuchnikov E, Le-Rademacher J, Carreras J, Armand P, Bishop MR et al. Conditioning regimens for allotransplants for diffuse large B-cell lymphoma: myeloablative or reduced intensity? Blood 2012; 120: 4256–4262.
Bishop MR, Dean RM, Steinberg SM, Odom J, Pavletic SZ, Chow C et al. Clinical evidence of a graft-versus-lymphoma effect against relapsed diffuse large B-cell lymphoma after allogeneic hematopoietic stem-cell transplantation. Ann Oncol 2008; 19: 1935–1940.
Glass B, Hasenkamp J, Wulf G, Dreger P, Pfreundschuh M, Gramatzki M et al. Rituximab after lymphoma-directed conditioning and allogeneic stem-cell transplantation for relapsed and refractory aggressive non-Hodgkin lymphoma (DSHNHL R3): an open-label, randomised, phase 2 trial. Lancet Oncol 2014; 15: 757–766.
Rezvani AR, Kanate AS, Efron B, Chhabra S, Kohrt HE, Shizuru JA et al. Allogeneic hematopoietic cell transplantation after failed autologous transplant for lymphoma using TLI and anti-thymocyte globulin conditioning. Bone Marrow Transplant 2015; 50: 1286–1292.
Kim JW, Kim SW, Tada K, Fukuda T, Lee JH, Lee JJ et al. Allogeneic stem cell transplantation in patients with de novo diffuse large B-cell lymphoma who experienced relapse or progression after autologous stem cell transplantation: a Korea-Japan collaborative study. Ann Hematol 2014; 93: 1345–1351.
Avivi I, Canals C, Vernant JP, Wulf G, Nagler A, Hermine O et al. Matched unrelated donor allogeneic transplantation provides comparable long-term outcome to HLA-identical sibling transplantation in relapsed diffuse large B-cell lymphoma. Bone Marrow Transplant 2014; 49: 671–678.
Fenske TS, Ahn KW, Graff TM, DiGilio A, Bashir Q, Kamble RT et al. Allogeneic transplantation provides durable remission in a subset of DLBCL patients relapsing after autologous transplantation. Br J Haematol 2016; 174: 235–248.
Mondello P, Younes A . Emerging drugs for diffuse large B-cell lymphoma. Expert Rev Anticancer Ther 2015; 15: 439–451.
Nowakowski GS, Czuczman MS . ABC, GCB, and double-hit diffuse large B-cell lymphoma: does subtype make a difference in therapy selection? Am Soc Clin Oncol Educ Book 2015; doi: 10.14694/EdBook_AM.2015.35.e449 e449–e457.
Camicia R, Winkler HC, Hassa PO . Novel drug targets for personalized precision medicine in relapsed/refractory diffuse large B-cell lymphoma: a comprehensive review. Mol Cancer 2015; 14: 207.
We thank the LYSARC for coordinating the study; Fabienne Morand, Sami Boussetta, Marion Fournier, Laurence Girard, Clemence Capron and the project leaders from the different countries; the patients and their families; AJE for reviewing the English language manuscript; Catherine Druon for preparing the manuscript; and all investigators and pathologists. Preliminary results presented at the 57th ASH 2015 annual meeting (abstract #731). Written on behalf of CORAL (Collaborative trial in Relapsed Aggressive Lymphoma).
EVDN: Roche: Travel accommodation. CG: Roche: Research funding. MT: Roche: Honoraria, research funding. JR: Roche: Consultancy. UD: Roche Pharma: Honoraria, research funding. CHM: Genentech: Consultancy, research funding. BG: Roche: Honoraria, research funding.
About this article
Long-term safety and activity of axicabtagene ciloleucel in refractory large B-cell lymphoma (ZUMA-1): a single-arm, multicentre, phase 1–2 trial
The Lancet Oncology (2019)
Impact of the soluble interleukin-2 receptor level in the relapsed or refractory phase on the clinical outcome of patients with diffuse large B-cell lymphoma
Leukemia & Lymphoma (2019)
Treatment of aggressive B-cell non-Hodgkin lymphoma beyond frontline therapy in patients not eligible for stem cell transplantation: a structured review
Leukemia & Lymphoma (2019)
Outcomes of patients with large B‐cell lymphomas and progressive disease following CD19‐specific CAR T‐cell therapy
American Journal of Hematology (2019)
Postrelapse survival in diffuse large B-cell lymphoma after therapy failure following autologous transplantation
Blood Advances (2019)