Optimal frontline management of mantle cell lymphoma: can we agree?

KEYWORDS:

• Mantle cell lymphoma
• autologous stem cell transplant
• lymphoma
• chemotherapy
• minimal residual disease

1. Introduction

Mantle cell lymphoma (MCL) has historically borne a poor prognosis, though recent developments such as intensified induction with autologous stem cell transplant (ASCT) consolidation for younger patients have significantly improved outcomes in this disease. Nevertheless, MCL remains incurable. Given the marked clinical heterogeneity of MCL, a ‘one size fits all’ approach to frontline management is clearly inadequate; rather we should focus on risk-adapted therapy that maximizes disease control while minimizing toxicity in appropriate patient groups. Observation has been demonstrated to be a potential management strategy, but baseline factors do not adequately identify this patient population. The challenge lies in delineating which groups could benefit from intensification versus those that could safely receive less toxic therapy, potentially through the use of minimal residual disease (MRD) testing and/or incorporation targeted therapies at an earlier stage of treatment.

2. Watch-and-wait approach

A small proportion of MCL patients can be safely watched prior to commencing treatment. These include asymptomatic patients with low-bulk disease, and those recognized in the updated WHO classification as ‘Leukemic non-nodal MCL’ negative for SOX11 with an indolent disease course [1]. Despite increasing recognition, identifying such patients at outset remains challenging, thereby warranting close follow-up if adopting this approach.

3. Frontline therapy for young patients with standard MCL

Upfront ASCT consolidation of transplant-eligible MCL patients was established as standard of care through the European MCL Network’s randomized trial of CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) followed by myeloablative ASCT versus IFN-a maintenance [2]. This trial demonstrated that ASCT consolidation improves progression-free survival (PFS); however, persistence of MRD was subsequently recognized as a key driver of MCL relapse despite ASCT [3].

Induction intensification through the use of high-dose cytarabine prior to ASCT has proven integral to improving survival in younger MCL patients. The pivotal MCL Younger study, alternating R-CHOP/R-DHAP (rituximab, dexamethasone, high-dose cytarabine, and cisplatin) with ASCT consolidation set a new benchmark, with median time to treatment failure of 9.1 years [4]. Much of this benefit was driven by the incorporation of cytarabine, leading to high rates of MRD negativity in both blood and marrow at end of induction, further translating to prolonged PFS irrespective of the quality of clinical remission prior to ASCT. Similar results over a long period of follow-up (median PFS of 11 years) were achieved by the Nordic MCL2 study utilizing high-dose cytarabine alternating with maxi-CHOP and rituximab induction prior to ASCT, though extended late relapses continue to plague patients over a decade beyond initial induction [5]. An extended failure-free survival of 15 years in 30% of younger patients was also achieved via HyperCVAD without ASCT consolidation (rituximab, hyper-fractionated cyclophosphamide, vincristine, doxorubicin, dexamethasone, alternating with cytarabine and methotrexate) [6], but this could not be replicated in multicenter settings [7,8].

Maintenance therapy with rituximab has been applied either in an MRD-driven or ‘blanket’ manner post-ASCT in an attempt to reduce relapse post-ASCT. An extension of Nordic MCL2 used preemptive rituximab in patients with molecular relapse post-ASCT. Rituximab could convert patients with molecular relapse back to MRD-negativity, with median time from molecular to clinical relapse extending out to 55 months [9]. However, this should be interpreted with caution in the absence of randomized data. The LyMa group’s phase III data demonstrated blanket use of rituximab maintenance following four cycles of R-DHAP and R-BEAM (rituximab, carmustine, etoposide, cytarabine, and melphalan) improved PFS and overall survival (OS) at 4 years [10]. Hence, currently available evidence supports routine use of rituximab maintenance post-ASCT, though the potential for an MRD-adapted maintenance strategy should be further investigated.

While effective, the toxicities of high-dose cytarabine containing induction and ASCT still warrant consideration. Drop-out rates in intensive induction trials range from 10% to 30% [11], with significant infectious and myelotoxic sequelae. Secondary myeloid malignancies are reported in 2.4–3.1% [4,5], with rates of solid tumors (excluding non-melanomatous skin cancers) reaching 12.5% over 15 years [5]. Hence it is relevant to reexamine the true benefit of ASCT consolidation in the cytarabine and rituximab era. The ECOG Intergroup’s EA4151 trial will evaluate the necessity of ASCT consolidation in patients who are MRD-negative post-induction [12]. Efforts are also underway to integrate novel agents with established efficacy in the relapsed setting (such as Ibrutinib) to the frontline, with view to minimizing chemotherapy toxicities. Of particular interest is the MCL Triangle trial incorporating Ibrutinib to R-CHOP/R-DHAP induction and maintenance, as an alternative to the paradigm of R-CHOP/R-DHAP followed by myeloablative ASCT [13]. MRD analysis in this trial will also address whether maintenance with novel agents such as ibrutinib can address the dismal prognosis for patients with early reemergence of MRD within 6 months of ASCT (median PFS of 20 vs. 142 months in MRD-negative) [9]. Indeed, the ideal scenario would be the use of effective novel agents able to attain high rates of MRD-negativity upfront, followed by MRD-driven maintenance in order to maximize durable remission while avoiding significant long-term treatment toxicity.

4. Treating high-risk patients – blastoid variant and TP53 alterations

Blastoid morphology has been reported as a predictor of aggressive behavior [14]; however, more recent studies have found that its impact on survival may largely be driven by Ki-67 index rather than histology [15]. Limited prospective data address treatment in this population; however, recent retrospective series suggest that ASCT consolidation could improve median OS in younger blastoid variant patients [16].

TP53 mutations confer poor prognosis in hematological malignancies [17], yet optimal treatment of TP53-mutated MCL remains undefined. Its prevalence in unselected MCL cohorts is reported at 14–20%, and is associated with median OS of 13-14 months, i.e. less than a third that of wild-type TP53 patients [18,19]. TP53 mutations were seen in 11% of patients included in the Nordic MCL studies, with median OS of only 1.8 years despite intensive induction/ASCT consolidation, in stark contrast to the 12.7 years achieved by their unmutated counterparts [20]. MCL Younger also demonstrated an additive effect of TP53 deletion to CDKN2A deletion leading to poor median OS [21]. Hence, the largest trials to date of younger MCL patients present a clear signal that risk from TP53 alterations remains inadequately addressed by intensified induction and ASCT consolidation.

Unfortunately, there is as yet no defined frontline escalation strategy in TP53-mutated MCL. Reduced intensity conditioning allograft as part of frontline consolidation has demonstrated no survival benefit in unselected MCL cohorts [22], though further prospective trials examining TP53-mutated patients are warranted. In contrast, preliminary data incorporating novel agents report (phase II PHILEMON trial combining ibrutinib, lenalidomide, and rituximab in relapsed/refractory patients) a similar PFS for TP53-mutated and unmutated patients [23]. Combination ibrutinib and venetoclax in a small phase II study of heavily pretreated young MCL patients achieved CR in 50% of TP53-mutated patients included [24], supporting the potential efficacy of novel agent combinations in this poor-performing group of patients.

5. Older and transplant-ineligible patients

The median age at diagnosis in MCL is 68 years [25], thus limiting the population eligible for dose-intensive therapy to a smaller subset. Where high-intensity induction and ASCT cannot be applied, management of this population should thus focus on less toxic treatment strategies.

Both R-CHOP and bendamustine-rituximab (BR) have been studied in patients >65 years of age [26,27]. The Study Group of Indolent Lymphoma (StiL) trial demonstrated that BR achieved superior PFS versus R-CHOP in all indolent subtypes, including 94 (18%) MCL patients achieving PFS 35 vs. 22 months with reduced neutropenia and infective sequelae [26]. The BRIGHT trial demonstrated a similar toxicity benefit, although its design limited the interpretation of the efficacy data [28]. With favorable toxicity and no difference in OS with 9-year follow-up [29], BR has increasingly become a standard of care approach in non-transplant eligible MCL patients.

Attempts to build on R-CHOP or BR have also evaluated cytarabine and targeted agents such as bortezomib. Modification of R-CHOP through replacement of vincristine for bortezomib (VR-CAP) in a large phase III trial demonstrated significant improvement in PFS (30.7 vs 16.1 months), though increasing hematologic toxicity without OS benefit [30]. Cytarabine has been incorporated in a phase II trial of R-BAC (lower bendamustine dose 70mg/m2 and cytarabine 500mg/m2) which reported favorable efficacy (CR rate 91%; 2-year PFS and OS of 81%) albeit with significant grade 3–4 hematologic toxicity [31], once again highlight the challenging dynamic between efficacy and toxicity in treating MCL patients. Results of a phase III trial (SHINE) where ibrutinib is being used to augment both BR induction and rituximab maintenance are also eagerly awaited [32].

Rituximab maintenance as a strategy to delay relapse has been evaluated in this population, and the benefit of this may vary on the induction regimen employed. Patients >65 years treated with R-CHOP and rituximab maintenance until progression achieved PFS and OS (57% and 87% at 4 years) approaching that of their intensively treated younger counterparts [27]. The benefit of maintenance rituximab may be more modest in patients receiving BR (from limited subgroup analysis in the StiL MAINTAIN trial) [33] and as yet unknown in other regimens such as R-BAC or VR-CAP.

For ‘frail’ elderly who cannot tolerate the above regimens, ‘chemotherapy-free’ approaches are increasingly attractive. Again, novel agents will have a pivotal role in this space. Recently, lenalidomide plus rituximab for induction and maintenance until progression has demonstrated promising results in the elderly (5-year PFS and OS of 63.9% and 77.4%, respectively) [34]. Similarly, ibrutinib/rituximab may also be a viable frontline option subject to further results from an ongoing phase II trial.

6. Conclusion and future directions

While there is clear evidence for intensive ASCT-based frontline treatment for younger fit patients with MCL, the optimal chemotherapy in the elderly/unfit remains undefined although maintenance rituximab post-R-CHOP appears to be a clear standard. MRD-driven approaches and the incorporation of targeted therapy in the frontline setting require prospective assessment. In particular, there is an urgent need to understand the biology driving high-risk patients in order to address this with novel strategies.

Declaration of interest

J Kuruvilla declares research support from Leukemia and Lymphoma Society Canada, Princess Margaret Cancer Foundation, Roche, and Janssen; consultancy activity with AbbVie, Bristol-Myers Squibb, Gilead, Janssen, Merck, Roche, and Seattle Genetics; has received honoraria from Amgen, Bristol-Myers Squibb, Celgene, Gilead, Janssen, Karyopharm, Lundbeck, Merck, Roche, and Seattle Genetics; and has participated in scientific advisory board activity for Lymphoma Canada. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.