To our knowledge, this is the largest retrospective analysis on MCL patients in China. The median age at diagnosis was 59 years, younger than that reported in Western cohorts (68 years)(3). The incidence of bone marrow involvement in this cohort (31%) were much lower than that in Western patients (82%)(25). Other clinical features of MCL patients in our cohort were comparable to those reported in the Western cohorts. The PFS and OS of the patients who have received rituximab-containing induction therapy was analogous to those recently reported in several Western cohorts(26, 27). Younger patient cohort also had similar survival outcomes(25, 28). However, survival outcome of the older patients was significantly inferior compared with that reported in the Western countries(25). Conventional prognostic factors including age ≥ 65 years, high-risk MIPI, blastoid/pleomorphic histology, Ki-67 > 30% and bone marrow involvement were associated with worse PFS or OS on MVA in this cohort.
Most younger and older patients received frontline immunochemotherapy with R-CHOP given the time period of data capture. Unsurprisingly, younger patients with R-HD-AraC regimens as first-line treatment had the lowest rate of POD24, and novel agents-containing regimens also demonstrated favorable outcomes. The most commonly utilized R-HD-AraC regimen was R-CHOP/R-DHAP, and R-HyperCVAD was associated with unsatisfactory disease control, which may be caused by the intolerable toxicity of the R-HyperCVAD regimen. Although previous study has demonstrated a high CR rate with R-HyperCVAD, almost 40% of patients were unable to complete the planned treatment(29). Therefore, R-HyperCVAD is not recommended as induction regimen by the British Society of Hematology (BSH)(30).
For the older patients, first-line regimens used were varied except for R-CHOP in this cohort. Patients who received BR and VR-CAP have achieved the lowest rate of POD24. A recent study has reported that exposure to frontline HD-AraC in older patients appeared to mitigate differences in clinical outcome between younger and older patients(25). In contrast, our study failed to demonstrate the benefit of frontline HD-AraC in older patients. Substantial hematological toxicity of HD-AraC may prevent the completion of therapy and thus reduction of cytarabine dose is considered as a solution. The combination of a lower dose of cytarabine and BR backbone (RBAC500) has shown a manageable hematological toxicity that allowed 95% of patients to complete at least four RBAC500 cycles with a median age of 71 years, and achieved a 2-year PFS of 81%(31). Overall, less aggressive frontline regimens, especially BR and VR-CAP, are expected to change the adverse outcome of older patients in China. Adjunctive low dose cytarabine and targeted agents could also be considered.
AHCT was uncommon even in patients younger than 65 years in this cohort, suggesting the limited availability and affordability of AHCT in real-world practice in China. Only AHCT followed by MR was associated with better PFS, and AHCT alone was only associated with improved OS in the rituximab-exposed cohort. However, this improvement in OS with AHCT alone did not persist for the CR and novel-agents-exposed cohort. Nevertheless, the value of HD-AraC-based induction therapy followed by consolidated AHCT in the frontline setting is becoming uncertain in the era of targeted therapy. Large retrospective studies in recent years showed that AHCT was not clearly associated with improved OS(27, 28) and the use of novel agents at relapse may help explain the lack of improvement. Also, there is an increasing desire for more convenient and tolerable treatment options that could be widely utilized. Our exploratory analysis demonstrated that the efficacy of BTKi-containing treatment was comparable to that of the standard first-line treatment with R-HD-AraC regimens followed by consolidated AHCT. While the use of PSM has minimized the selection bias, small sample size and heterogeneous chemotherapy regimens that was combined with BTKi may limit the interpretability of the result. Currently, randomized trials are evaluating the necessity of HD-AraC (ECOG 4181 trial; NCT04115631) and consolidated AHCT (TRIANGLE trial; NCT02858258) when BTKi is incorporated in first-line therapy, and results from these trials are expected to change the therapeutic pattern for younger patients. Such effort is also being made by researchers in China, with an ongoing phase II trial (NCT04624958) assessing zanubrutinib induction and maintenance as an alternative of consolidated AHCT.
Similarly, maintenance rituximab was uncommon in our cohort, which may be resulted from limited reimbursement for rituximab as maintenance. Consistent with the phase III Mantle Cell Lymphoma Efficacy of Rituximab Maintenance (LyMA) study and European MCL Elderly Trials(17, 18), our results support the use of maintenance rituximab irrespective of younger or older patients. However, several problems remained in the practice of maintenance rituximab therapy. First, the optimal schedule and duration of rituximab maintenance was difficult to discern for this cohort given the retrospective nature of the study. Maintenance rituximab is typically offered every 2 months and the duration ranges from 2 years to continuous course until disease progression(17, 18). Secondly, rituximab-maintenance therapy was reported to offer survival benefit when used after R-CHOP or R-cytarabine-based therapy in the frontline but the benefit following other induction therapies such as bendamustine- or fludarabine-containing treatment remains unknown(32). Thus, the value of maintenance rituximab following BR and VR-CAP induction is not convincing. Lastly, novel maintenance strategies incorporating other targeted agents are expected as more effective alternatives to maintenance rituximab, especially for older patients. Single-agent lenalidomide maintenance has been shown to improve PFS after R-CHOP + HD-AraC induction and AHCT versus observation, at the cost of an increased rate of adverse events(33). Currently, the ECOG-ACRIN E1411 trial (NCT01415752) is not only evaluating the use of bortezomib (BR ± bortezomib) in induction therapy, but also comparing maintenance strategies of lenalidomide with rituximab (LR) versus rituximab monotherapy in patients < 65 years. The MCL-R2 Elderly trial (NCT01865110) is also comparing LR versus rituximab maintenance after R-CHOP ± cytarabine induction in patients > 60 years. The SHINE study (NCT01776840) is evaluating ibrutinib-rituximab maintenance after BR ± ibrutinib induction in patients > 65 years.
In this study, patients with consolidated AHCT followed by MR had better PFS compared with MR alone. However, results may have been skewed by the small sample size and inherent limitations of retrospective study; thus, it remains uncertain whether combined AHCT and MR provides advantage over MR alone. The ongoing phase III EA4151 trial (NCT03267433) is comparing these two consolidated modalities in younger patients with negative minimal residual disease in first complete remission.
For the r/r MCL, the OS-2 of our cohort was similar to that reported in the Western cohorts(26, 34). Time to POD confirmed its crucial importance to predict the overall survival. 30% of the r/r MCL received BTK inhibitors during the salvage treatment and demonstrated a good outcome, in accordance with the well-known efficacy of BTKi in relapsed or refractory setting. Since the primary objective of our study was to characterize treatment patterns in the front-line setting, in-depth analysis of salvage treatment pattern in this cohort would remain for further study.
Given the endemic HBV infection, HBV reactivation related to immunosuppressive chemotherapies remains a common but serious complication for Chinese MCL patients. For HBsAg-positive patients, anti-HBV prophylaxis should be initiated at least 7 days before anticancer therapy and drugs with a high resistance barrier (entecavir or tenofovir) are preferred. For HBsAg-negative, anti-HBc–positive patients, anti-HBV prophylaxis is recommended for all the patients receiving rituximab-containing treatment or stem cell transplantation(24, 35). However, preventive measures are not always successfully implemented in these patients, possibly resulted from unawareness of the risk of HBV reactivation and consideration of cost effectiveness of anti-HBV prophylaxis. In the current study, most of the HBsAg-negative, anti-HBc–positive patients did not receive anti-HBV prophylaxis and 5.3% of these untreated patients developed HBV reactivation, including 3 patients with HBV-related hepatitis. This result supported the necessity of anti-HBV prophylaxis in this patient cohort. Except for traditional cytotoxic drugs and anti-CD20 antibody that have been identified as risk factors of HBV reactivation, the association between other targeted agents and HBV reactivation is uncertain. Our preliminary analysis suggested BTKi may not be a risk factor and more solid evidence is needed to decide the prophylaxis recommendation for patients receiving BTKi.
This study has several limitations. First, as a multi-center study, there was a lack of central pathological and response review. Second, as a major consideration of treatment selection, safety profile of the treatment modalities in this patient cohort was not collected and analyzed, thus it is unable to assess whether ethnic characteristics could affect treatment-related side effect. Third, biologic features (i.e., TP53 mutations or deletions) were not included in the analysis because these features were not routinely tested in clinical practice and may not be available for most of the included patients. Lastly, inherent time bias existed in the analysis of AHCT and maintenance rituximab therapy. To reduce the biases, we conducted secondary analyses that excluded patients with stable or progressive disease during the induction therapy.