Hematopoietic Stem Cell Transplantation for Diffuse Large B-Cell Lymphoma Harboring a 8q24/MYC Rearrangement: A Retrospective Registry Study from the Japan Society for Hematopoietic Cell Transplantation

Background and Objectives: Approximately 10% of cases of diffuse large B cell lymphoma have a MYC rearrangement (MYC+ DLBCL) and these have been associated with an inferior outcome following standard therapy. Double-hit (DHL) and triple-hit (THL) lymphomas harbor concurrent rearrangements of MYC and BCL2 and/or BCL6 are associated with a very poor outcome. The role of hematopoietic stem cell transplantation (HSCT) for MYC+ DLBCL remains unclear. Autologous (Auto-) HSCT could not salvage chemosensitive relapse of MYC+ DLBCL in the CORAL study. However, other two retrospective studies showed the efficacy of Auto-HSCT as consolidation therapy for DHL who attained first complete response (CR), in contrast to the poor outcome of patients without HSCT. A retrospective study of HSCT for MYC+ DLBCL using Japanese registry-based data was conducted.

Materials and Methods: The data of Japanese HSCT registry were surveyed, and 54 patients with MYC+ DLBCL were identified from 4,336 adult patients with DLBCL. These pathological and cytogenetic diagnoses including G-banding and/or fluorescence in situ hybridization (FISH) were carried out at each institution. The questionnaires were sent to collect patient data of clinical presentation and therapies, as well as pathology and cytogenetic reports. This study was approved by the ethical committees of the Japan Society for Hematopoietic Cell Transplantation (JSHCT) and Shimane University.

Results: A total of 48 patient data were obtained from 36 institutions. MYC rearrangements were identified by G-banding in 44 patients and FISH in 4 patients. The median age at diagnosis was 55 (range, 21-67) years. Twenty patients were male. Six patients were transformed DLBCL from indolent lymphoma. Eleven patients had CNS involvement during clinical course. Twenty seven patients were single hit lymphoma (SHL, only MYC), 20 patients were DHL (MYC+BCL2; n = 15, MYC+BCL6; n = 5) and one patient was THL (MYC+BCL2+BCL6). Thirty-one patients (64.6%) received R-CHOP as initial therapy. The CR rate after the initial therapy was 66.7% in SHL, but was 28.6% in DHL/THL. In total, data were available for 32 Auto-HSCTs and 17 allogeneic (Allo-) HSCTs including one after relapsed of Auto-HSCT were available between 2003 and 2012 and subjected for further analysis. Median follow up time from HSCT in surviving patients was 17.1 months (range, 9.8-130.0). In Auto-HSCT in first CR, median time from diagnosis to HSCT was 6.5 months (range, 4.6-20.9). The overall survival (OS) at 2 years, the event free survival (EFS) at 2 years, the relapse rate at 1 year and the non-relapse mortality (NRM) at 1 year was 71.1% (95% CI: 51.7-83.9), 65.5% (95% CI: 46.4-79.2), 31.4% (95% CI: 16.2-47.9) and 3.1% (95% CI: 0.2-14.0), respectively. The outcome of SHL (n = 20) and DHL (n = 12) were almost similar (OS at 2 years; 73.8% vs 66.7%, P = 0.62). The disease status at HSCT was a significant adverse prognostic factor. OS of patients who received HSCT in CR (n = 23) was 82.4%, and was significantly better than that in partial response (PR, 42.9%, P = 0.04). The outcome of HSCT in first CR (n = 14) and second CR (n = 9) was similar (respective OS at 2 years; 78.6% and 88.9%, P = 0.57). In Allo-HSCT, four patients (23.5%) had poor performance status (2-4), eleven patients (64.7%) were chemoresistant prior to HSCT, and 13 patients (76.5%) received reduced intensity conditioning regimen. The median time from diagnosis to HSCT was 10.0 months (range, 2.3-54.5). Their OS at 2 years, EFS at 2 years, relapse rate at 1 year and NRM at 100 days were 29.4% (95% CI: 10.7-51.1), 17.6% (95% CI: 4.3-38.3), 41.2% (95% CI: 17.6-63.5) and 41.2% (95% CI: 17.1-64.0), respectively. The outcome of SHL (n = 8) and DHL/THL (n = 9) was almost similar (OS at 2 years; 37.5% vs 22.2%, P = 0.78).

Conclusions: Auto-HSCT as consolidation therapy for MYC+ DLBCL was effective, as reported elsewhere. Auto-HSCT for chemosensitive relapse was also effective, in contrast to the CORAL study. The efficacy of Allo-HSCT was limited, particularly for chemoresistant patients. Although the present study include several limitations of retrospective nature, HSCT is not recommended for chemoresistant MYC+ DLBCL.