Dose-adjusted (DA)-EPOCH-R with high-dose methotrexate for newly diagnosed CD5-positive diffuse large b-cell lymphoma (CD5+ DLBCL): Interim results from a phase II study

Introduction: CD5 + DLBCL comprises 5-10% of DLBCL and is an immunohistochemical subgroup of DLBCL, not otherwise specified. CD5 + DLBCL shows a significantly worse survival than CD5-negative DLBCL with rituximab (R)-CHOP. Of note, central nervous system (CNS) relapse, particularly brain parenchymal relapse, frequently occurs even in the R-era (Miyazaki K, et al. Ann Oncol 2011). Most CD5 + DLBCL is classified as non-germinal center B-cell-like (non-GCB)/activated B-cell-like DLBCL. DA-EPOCH-R is reported to show promising outcomes in patients with newly diagnosed non-GCB DLBCL (Wilson WH, et al. J Clin Oncol 2008). High-dose methotrexate (HD-MTX) therapy has a greater potential than intrathecal administration of MTX to prevent brain parenchymal relapse. To explore a more effective treatment for newly diagnosed CD5 + DLBCL, we conducted a multicenter phase II study of DA-EPOCH-R combined with HD-MTX (PEARL5 study; UMIN000008507). Methods: Patients with newly diagnosed stage II-IV disease from 20 to 75 years of age with ECOG performance status (PS) of 0 to 3 were eligible for enrollment. CD5 expression in tumor cells was examined by immunohistochemistry and/or flow cytometry in participating institutes. Four cycles of DA-EPOCH-R followed by 2 cycles of HD-MTX (3.5 g/m2) and additional 4 cycles of DA-EPOCH-R were planned as the protocol treatment. The primary endpoint was 2-year progression-free survival. Secondary endpoints were complete response (CR) rate, overall response rate, 2-year overall survival, 2-year CNS recurrence, and toxicity. Response was assessed at each participating institute using the revised International Workshop Criteria. Toxicity was graded according to the Common Terminology Criteria for Adverse Events v4.0. This planned interim analysis included response and toxicity by the protocol treatment. Results: Forty-seven patients were enrolled between Aug 2012 and Nov 2015. The diagnosis of CD5+ DLBCL in all patients was confirmed by the central pathology review. Cyclin D1 and EBER were negative in all cases examined (n = 47 and n = 46, respectively). According to the Hans' criteria, 72% (33/46) of the patients had non-GCB DLBCL and 28% (13/46) had germinal center B-cell-like DLBCL. The median age was 62 years (range: 37-74 years), and 60% were > 60 years. The baseline clinical features were as follows: female sex, 62%; ECOG PS >1, 4%; stage III/IV, 53%; elevated LDH, 66%; extranodal involvement > 1 site, 34%; bone marrow involvement, 11%; and high-intermediate-risk or high-risk groups of the International Prognostic Index, 47%. Skin/subcutaneous tissue was the most frequent site of extranodal involvement (15%). One patient had primary testicular DLBCL. Forty-six (99%) patients completed the protocol treatment. In the remaining patient, the protocol treatment was discontinued due to grade 3 stomatitis during the sixth cycle of DA-EPOCH-R. The median dose level of DA-EPOCH-R was 2 (range: 1-4). There was no deviation or violation in determining the dose levels of DA-EPOCH-R. At the time of this interim analysis, 46 patients were assessed for response. The CR rate was 91% (42/46; 95% confidence interval [CI], 79-98%) and the ORR was 93% (43/46; 95% CI, 82-99%). Two patients experienced disease progression during the treatment. Toxicity was assessed in 357 cycles of DA-EPOCH-R/HD-MTX in all 47 patients. There was no treatment-related death. Grade 4 non-hematologic toxicity was observed in only one patient who experienced tumor lysis syndrome with hyperkalemia soon after the first administration of R. The most common grade 3 non-hematologic toxicity was elevated alanine transaminase, which occurred in 6% of cycles. Two patients experienced grade 3 infection (cellulitis and endocarditis). Grade 3 neuropathy occurred in only 1% of cycles. There was no grade 3 cardiac toxicity. The targeted absolute neutrophil count (> 500/mm3) occurred in 77% of cycles. Thrombocytopenia of < 25, 000/mm3 occurred in 8% of cycles. Febrile neutropenia (FN), which occurred in 23% of cycles, was manageable. Conclusion: These interim results demonstrate that DA-EPOCH-R/HD-MTX provides a high CR rate in patients with newly diagnosed stage II-IV CD5+ DLBCL. This therapeutic approach was feasible, although most patients experienced FN during the protocol treatment.