Impact of Rituximab Maintenance Added to Ibrutinib-Containing Regimens with and without ASCT in Younger, Previously Untreated MCL Patients: An Analysis of the Triangle Data Embedded in the Multiply Project

Introduction The TRIANGLE trial (Dreyling et al. 2024) has set a novel standard in first-line treatment of mantle cell lymphoma (MCL) patients aged 18-65 years, highlighting the role of ibrutinib in the management of these patients. During the planning stage of TRIANGLE, results of the LYSA-LYMA trial showed a progression-free survival (PFS) benefit with rituximab maintenance (Rm) after a cytarabine-containing induction and autologous stem cell transplantation (ASCT) in younger patients with MCL (Le Gouill et al. 2017). Therefore, the TRIANGLE trial recommended from the beginning the additional non-randomized administration of Rm (every two months for 3 years) to all three treatment arms (arm I: IR-CHOP/ R-DHAP+ Im; arm A+I: IR-CHOP/ R-DHAP followed by ASCT + Im; arm A, i.e. pre-trial standard of care: R-CHOP/ R-DHAP followed by ASCT) according to national guidelines and/or center practice. We retrospectively analyzed the TRIANGLE data to assess the efficacy and toxicity of additional Rm to ibrutinib-containing regimens, with and without ASCT, and to independently confirm the LYSA-LYMA results in younger, untreated MCL patients. Methods All patients in remission after induction therapy (arm I) and after ASCT (arms A+I, A) were included in this analysis and assigned to Rm or noRm groups based on the as-treated principle. Every treatment arm (arm I,arm A+I, and arm A) was evaluated separately, thus, no adjustment of the alpha level was necessary. The primary endpoint PFS from end of induction (arm I)/end of ASCT (arm A+I, A) was summarized using Kaplan-Meier curves. Differences between Rm and noRm groups within each treatment arm were evaluated using a two-sided log-rank test with an alpha level of 0.05. To address baseline imbalances and to increase power, Cox regression models were calculated including MCL International Prognostic Index (MIPI) (Hoster et al. 2008), response status at the end of induction therapy (CR vs PR after induction/ASCT), Ki67,and cytology (classical vs.pleomorphic/blastoid variant). Results In the different study arms, 59% (n=159; arm I), 64% (n=151; arm A+I) and 67% (n=155; arm A) started Rm. In general, baseline characteristics were well balanced between the Rm and noRm groups in all treatment arms except for more frequent pleomorphic/blastoid cytology in the noRm group of arm I; higher Ki67 in the Rm group of arm A and A+I; and a higher Ann-Arbor stage, and more frequent CR after ASCT in the noRm group of arm A. After a median follow-up time of 4.0 years, median duration of Rm, excluding patients who relapsed, died, or withdrew, was 26 months for arm I and 30 months for arms A+I and A. The Kaplan-Meier curves for PFS and adjusted Cox regression analyses indicated a significantly longer PFS with additional Rm administration in all three treatment arms. The adjusted hazard ratio (Rm vs. noRm) for arm I was 0.50 (95% CI: 0.28 - 0.90, p-value 0.019), for arm A+I 0.26 (95% CI: 0.13 - 0.51, p-value <0.001), and for arm A 0.29 (95% CI: 0.16 - 0.52, p-value <0.001), respectively. Accordingly, the 4-year PFS rate estimated from the Kaplan-Meier curves were for arm I 86% (95% CI: 80% - 92%) in the Rm group vs. 76% (95% CI: 67% - 85%; log rank p=0.016) in the noRm group, for arm A+I 89% (95% CI: 84% - 95%) in the Rm group vs. 75% (95% CI: 65% - 87%; log rank p <0.001) in the noRm group; and for arm A 83% (95% CI: 77% - 89%) in the Rm group vs. 54% (95% CI: 42% - 68%; log-rank p <0.001) in the noRm group. Rm was associated with a modest increase in infectious complications (grade 3 or greater) in all arms (arm I: 25% in the Rm group vs. 12% in the noRm group; arm A+I: 30% in the Rm group vs. 13% in the noRm group; arm A: 19% in the Rm group vs. 1% in the noRm group). In contrast hematological toxicity (grade 3 or greater) was increased only in arm A (28% in the Rm group vs. 11% in the noRm group), whereas comparable frequencies were observed in the other study arms (arm A+I: 46% in the Rm group vs. 52% in the noRm group; arm I: 27% in the Rm group vs 23% in the noRm group). Conclusion This analysis confirms the benefit of Rm, as single maintenance (arm A) as well as in combination with ibrutinib (arms A+ I and I). No unexpected toxicity signals were observed, supporting its use also in combination with novel regimens. On behalf of the European MCL Network *: shared first/senior authorship