Abstract
Mantle cell lymphoma is a well-recognized distinct clinicopathologic subtype of B-cell non-Hodgkin lymphoma. The current World Health Organization (WHO) classification subdivides this entity into aggressive and other variants. The disease has a predilection for older males, and patients typically present at an advanced stage with frequent splenomegaly and extranodal involvement including bone marrow, peripheral blood, gastrointestinal, and occasional central nervous system involvement. Early studies of therapy outcomes in this disease revealed that while response rates where high, relapse was expected after a limited period of time. Prolonged survival was uncommon, with initial median survival rates typically in the 3–4-year range. Those with a high proliferative rate, blastoid morphology, and selected clinical features were recognized as having a worse prognosis. Therapeutic approaches have diverged into aggressive therapies with high response rates and promising progression free survival rates, which may be applied to younger healthy patients, and less aggressive approaches. Aggressive therapies include intensive chemotherapy alone or chemotherapy followed by autologous stem cell transplant, which has been shown to be most effective when applied in first remission. Whether these more intense therapies result in improved survival as compared with less aggressive therapies is not well established. Allogeneic transplant has also been investigated, although high treatment-related mortality and the risk of chronic graft versus host disease and the relatively advanced age of this patient population have tempered enthusiasm for this approach. A number of less aggressive therapies have been shown to produce promising results. Consolidation and maintenance strategies are an active area of investigation. A number of newer agents have shown promising activity in relapsed disease, and are being investigated in the front-line setting. Overall survival rates are improving in this disease, with current studies suggesting a median survival of 5 or more years.
Introduction
Pathology
Mantle cell lymphoma (MCL) is a distinct subtype of B-cell non-Hodgkin lymphoma that became much easier to consistently diagnose in the era of detection of cyclin D1 by immunohistochemistry, immunophenotyping of cell surface antigens, and clinical molecular testing for the t(11;14) (q13;q32) by fluorescence in situ hybridization (FISH). The tumor cells are typically characterized by a CD20+, CD5+, CD23- immunophenotype. Several subtypes of this lymphoma have been described, with the current WHO classification distinguishing between the aggressive variants (blastoid and pleomorphic) and other variants (small cell and marginal zone-like) [Swerdlow et al. 2008]. Appropriate immunophenotyping and molecular ancillary testing is required to make an accurate diagnosis. In general, the absence of demonstration of cyclin D1 overexpression or the molecular hallmark of MCL should raise doubt as to the diagnosis, although rare atypical cyclin D1 negative and/or t(11;14) negative cases otherwise indistinguishable from classic MCL have been described. Demonstration of Sox11 overexpression may be useful in these cases [Ek et al. 2008]. The incidence of MCL is widely referenced as being 2–10% of non-Hodgkin lymphomas.
Clinical presentation
The disease is most commonly seen in middle-aged to elderly adults with a median age of 60 years, and is at least twice as common in males as in females. Most patients present with advanced stage disease. Extranodal involvement is common, including bone marrow, peripheral blood, and gastrointestinal sites. A subset may present with a predominant peripheral blood and marrow involvement pattern similar to CLL. This presentation has been reported to carry a better prognosis [Fernandez et al. 2010]. Some may present with intestinal involvement only, accounting for a significant fraction of cases of lymphomatous polyposis.
Staging
Initial staging of MCL typically consists of a complete blood count with white blood cell differential, basic chemistries, lactate dehydrogenase, bone marrow aspirate and biopsy, and imaging by computed tomography (CT) and/or positron emission tomography (PET)/CT. Patients with signs or symptoms of central nervous system involvement should undergo lumbar puncture with spinal fluid analysis. Controversy exists over the need for colonoscopy in patients without gastrointestinal symptoms, with advocates pointing out the high rate of asymptomatic bowel involvement, and skeptics pointing out the lack of therapeutic implications. The role of PET is evolving. In a study of 44 untreated MCL patients, F-18-fluorodeoxyglucose PET was found to be positive at diagnosis in all cases [Bodet-Milin et al. 2010]. PET was found to be insensitive to bone marrow and gastrointestinal involvement, and so did not influence staging, but appears to be relevant to prognosis. PET was shown to identify some extranodal sites of disease not identified by CT. Other trials have shown that some cases of MCL are PET negative at diagnosis [Schaffel et al. 2009].
Prognosis
MCL demonstrates considerable variation in natural history, and the selection of therapy must take prognosis into account. The MCL international prognostic index (MIPI) was derived by applying the variables included in the international prognostic index (IPI) developed for large cell lymphoma and the variables included in the follicular lymphoma international prognostic index (FLIPI) to the analysis of MCL patients on three clinical trials [Hoster et al. 2008]. The resulting index classified patients into a low-risk group comprising 44% of patients with a median overall survival (OS) not reached, an intermediate-risk group comprising 35% of patients with median OS 51 months, and a high-risk group with 21% of the patients and median OS of 29 months (Table 1). Factors included in the index are age, performance status, lactate dehydrogenase, and leukocyte count. In the development of the MIPI, it is notable that the number of extranodal sites was not an independent prognostic factor. Tumor cell proliferation, as measured by Ki-67 assessment on paraffin-embedded tissue, was analyzed in a subset of cases with a median of 14.5% (range 1.2–91%). This value was also prognostically significant using a cutoff point of 10%, and was independent of the MIPI. Others have shown that the proliferative index may be a better indicator of progression-free survival (PFS) than the MIPI in a group of patients treated aggressively [Schaffel et al. 2010]. Interobserver variability was cited as a barrier by these authors, who used a quantitative image analysis technique to attempt to control for this. A cutoff of 30% was deemed optimal for risk assessment. The MIPI score at diagnosis has been shown to be predictive of outcomes in MCL patients treated with aggressive therapies culminating in autologous stem cell transplant (ASCT) [Geisler et al. 2010; Pan et al. 2008].
Mantle Cell International Prognostic Score (MIPI) [Hoster et al. 2008]. The sum of the points for each of the four parameters is determined. Patients with 0–3 points are classified as low risk, 4–5 points are intermediate risk, and ≥6 points are high risk.
ECOG PS, Eastern Cooperative Oncology Group performance status; LDH/ULN, ratio of lactate dehydrogenase/institutional upper limit of normal; WBC, white blood cell count.
A retrospective analysis of PET scans obtained at diagnosis demonstrated that the maximum standardized uptake value (SUVmax) of the area of most intense uptake was highly predictive of outcome. A cutpoint of SUVmax > 5 yielded a high-risk group with inferior 5-year OS (34% versus 87.7%) and median failure-free survival (FFS) (10.6 months versus 45.3 months) [Karam et al. 2009]. In contrast, initial PET SUVmax was not shown to be predictive of outcome in a group treated with sequential chemotherapy followed by ASCT, although a mid-therapy PET was able to distinguish between a good prognosis PET-negative CT partial response (PR) group versus a PET-positive CT PR group (4-year PFS 76% versus 43%) [Schaffel et al. 2009]. Others have reported PET assessment following completion of therapy to be predictive of both OS and event-free survival (EFS) [Bodet-Milin et al. 2010].
Genomic and gene expression profiling in a group of indolent MCL cases with a nonnodal presentation demonstrated hypermutated heavy chain immunoglobulin variable regions, lack of genomic complexity, and absence of SOX11 expression with a reported 5-year OS of 78% [Fernandez et al. 2010].
Minimal residual disease
Molecular remission determined by PCR has been variably identified as either not predicting outcome following rituximab and CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) [Howard et al. 2002] or being prognostically significant for response duration following rituximab and chemotherapy followed by either ACST or maintenance therapy [Pott et al. 2010]. Different therapies, assays, and patient numbers likely explain the difference.
High expression of eukaryotic initiation factor 4E, a regulatory factor influencing cyclin D1 protein levels, was shown to be associated with poorer FFS (5-year 20.6% versus 63.5%) and OS (5 year 40% versus 73.8%) when compared with low or no expression in a cohort of 70 patients uniformly treated with rituximab and chemotherapy [Inamdar et al. 2009].
Historical treatment results
Prior to the routine diagnosis of MCL, patients with this disorder were often treated on protocols designed for either indolent lymphoma or more aggressive lymphoma. The results of both approaches were suboptimal, with reasonable initial response rates but subsequent early relapse and median survival times of 3–4 years. A prospective randomized European trial in what was then termed centrocytic lymphoma failed to show an advantage in survival for an anthracycline-based regimen (CHOP) versus a non-anthracycline-based regimen (COP) [Meusers et al. 1989].
Current therapy and future directions
As investigators turned their attention to the specific entity of MCL, it became clear that the results to date were suboptimal, and new therapies were needed. Some pursued more aggressive approaches, in the hope that increased dose intensity and duration of therapy would result in prolonged survival or perhaps cure. These approaches are most applicable to younger and healthier patients. Others have taken a more measured approach in trying to identify less aggressive therapies, including an initial watch and wait strategy, which can be applied to the treatment of older or less healthy patients as well as those preferring a less intensive therapy. A number of new drugs, and one old drug new to investigators in the West, have also been shown to have significant therapeutic efficacy and are increasingly being incorporated into initial therapy of MCL. These developments appear to have had an impact on patient outcomes, with a recent analysis of the outcomes of advanced stage nonblastoid MCL from the Kiel Lymphoma Study Group from 1975–1986 to the German Low Grade Lymphoma Study Group from 1996–2004 demonstrating an increase in the 5-year survival rate from 22% to 47% [Herrmann et al. 2009]. In this review we summarize the results of upfront treatment for MCL to date, with a brief mention of therapies tested in the relapsed or refractory disease setting that are being or will be investigated in the initial therapy of MCL. Uncommon presentations including early stage disease and splenomegaly dominant disease are also discussed.
Early stage disease
While the vast majority of MCL cases are at advanced stage at presentation, early stage disease is occasionally encountered. A retrospective analysis of 26 patients with stage IA or IIA treated with involved field radiotherapy (IFR) with or without chemotherapy, chemotherapy alone, or observation yielded a 46% 5-year PFS and 70% 5 year OS [Leitch et al. 2003]. Age under 60 and receiving IFR affected PFS, with IFR overshadowing age. The 5 year PFS was 68% for those receiving IFR, and 11% for those not, suggesting a role for IFR in localized MCL.
The role of splenectomy
A subset of patients with MCL present with symptoms and/or cytopenias attributed to bulky splenomegaly. The degree of cytopenia may interfere with systemic therapy and, in some cases, splenomegaly is the dominant feature of the disease. In a series of 26 cases splenectomy has been shown to result in an improvement in hemoglobin by at least 1 g/dl in patients presenting with hemoglobin <11.0 g/dl in 69% of cases, with 90% of patients with a preoperative platelet count <100 × 109/l achieving a platelet count of ≥100 × 109/l by 1 month after surgery [Yoong et al. 2001]. For those with anemia and thrombocytopenia the response rate was 50%. Twelve patients were able to avoid chemotherapy for a minimum of 13 months after splenectomy, with three otherwise untreated patients experiencing ongoing stable disease without chemotherapy for up to 8 years postsplenectomy. A second study compared the outcome of splenectomy in MCL and chronic lymphocytic leukemia (CLL) [Ruchlemer et al. 2002]. MCL patients experienced improvement in cytopenias in 62% versus 47% of advanced CLL patients. MCL patients experienced a reduction in median leukocytosis from 60.3 × 109/l to 29.1 × 109/l, whereas CLL patients experienced an increase from 24.2 × 109/l to 44 × 109/l. Based on these studies, splenectomy should be considered for selected patients with splenomegaly dominant presentations.
Immunochemotherapy
A number of moderate intensity combination chemotherapy backbones were studied in MCL, with a general migration toward CHOP based on favorable efficacy compared with mitoxantrone, chlorambucil and prednisone (MCP) [Nickenig et al. 2006] and the observation of difficulty colecting stem cells for potential ASCT following either that regimen or fludarabine based regimens [Eve et al. 2009; Eve and Rule, 2007]. The addition of rituximab to CHOP resulted in a 96% overall response rate (ORR) with 48% in complete response (CR) and 48% in partial response (PR) [Howard et al. 2002]. Unfortunately this high response rate did not translate to prolonged remission as the median PFS was 16.6 months, with no evidence of a PFS plateau. Molecular CR did not predict outcome in this trial. A randomized trial demonstrated that the addition of rituximab to the FCM (fludarabine, cyclophosphamide and mitoxantrone) regimen for the therapy of relapsed follicular lymphoma and MCL resulted in statistically significant improvements in both ORR (58% versus 46% in MCL) and overall survival (OS) [Forstpointner et al. 2004]. A randomized trial of R-CHOP versus CHOP as initial therapy in MCL showed an improved ORR for R-CHOP (94% versus 75%), improved CR rate (34% versus 7%) and improved time to treatment failure (TTF) (21 months versus 14 months). There was no difference in PFS [Lenz et al. 2005]. Long-term results of the rituximab addition were confounded by a secondary randomization to ACST versus interferon alfa maintenance. A subsequent review and metaanalysis of R-chemotherapy versus the same chemotherapy (seven studies) in indolent lymphoma and MCL concluded that the addition of rituximab resulted in improved survival in MCL: hazard ratio for mortality 0.6 (95% confidence interval [CI] 0.37–0.98) although there was intertrial heterogeneity [Schulz et al. 2011, 2007].
Intensive immunochemotherapy without ASCT
In recognition of inadequate outcomes following milder chemotherapy regimens, a number of investigators pursued more intense chemotherapy regimens with subsequent inclusion of rituximab, but without ASCT for consolidation. Hyper-CVAD (Cyclophosphamide, Vincristine, Doxorubicin and Dexamethasone) alternating with high-dose methotrexate and cytarabine was initially given without rituximab and followed by either ASCT or allogeneic stem cell transplant. Subsequently a larger phase II trial was conducted with the addition of rituximab to this regimen, and elimination of a planned stem cell transplant [Romaguera et al. 2005]. Patients received 6–8 cycles of intensive chemotherapy, each with rituximab. Mantle zone subtype was excluded, and 100 newly diagnosed patients were enrolled with 97 being treated. The median age was 61 (range, 41-80), and the sex ratio was 3:1 male to female. The reported ORR was 97%, with 87% CR or CRu. The treatment-related mortality (TRM) was 8%, with five acute deaths during therapy and four cases of myelodysplasia or acute leukemia. Patients over 65 did particularly poorly, and this group of patients was advised not to be treated on this aggressive regimen. An update published after a median follow up of 8 years revealed a median TTF of 4.6 years for all patients with no evidence of a plateau in the curve [Romaguera et al. 2010a]. An additional case of myelodysplastic syndrome (MDS)/acute myeloid leukemia (AML) was reported. For patients 65 and younger the median TTF was 5.9 years and the median OS had not yet been reached.
A similar intensive chemotherapy regimen with alternating chemotherapy cycles and rituximab (R-MACLO-IVAM-T) was applied to 22 patients with newly diagnosed MCL [Lossos et al. 2010]. Twenty one achieved CR with three subsequent relapses after a median follow up of 37 months. There was one toxic death and one death from relapsed MCL. Three-year PFS was 78%, although it should be noted that maintenance thalidomide was given in this trial.
R-DHAP (rituximab, dexamethasone, cytarabine and cisplatinum or another platinum compound) was given for four cycles to 63 patients with newly diagnosed MCL [Le Gouill et al. 2010]. The ORR was 92% with 82.5% CR/CRu. The intent was to consolidate with stem cell transplant following this initial chemotherapy.
The National Comprehensive Cancer Network (NCCN) performed a comparison of R-CHOP, R-CHOP followed by ASCT, and R-hyper-CVAD in 156 MCL patients under the age of 65 treated at 7 participating NCCN centers [LaCasce et al. 2009]. R-CHOP alone was inferior to R-CHOP followed by ASCT or R-hyper-CVAD with regard to PFS, with a disappointing overall 3-year median PFS.
Intensive immunochemotherapy with ASCT
Improved outcomes with ASCT in first remission
Disappointing results of conventional chemotherapy in MCL led to the incorporation of ASCT into initial therapy. A randomized trial by the European MCL network demonstrated that ASCT following initial chemotherapy yielded superior 3-year PFS (median 39 months versus 17 months) when compared with alpha-interferon maintenance, but did not improve overall survival [Dreyling et al. 2005]. Multiple phase II trials reported apparently improved outcomes compared with conventional chemotherapy historical results with the incorporation of ASCT in first remission using a variety of initial chemotherapy and conditioning regimen strategies [van ’t Veer et al. 2009; Murali et al. 2008; Siddiqui et al. 2005; Thieblemont et al. 2005]. While some preferred to use conditioning regimens including total body irradiation (TBI), others used all chemotherapy conditioning regimens, such as BEAM (carmustine, etoposide, cytarabine, and melphalan). A retrospective analysis of 73 patients with MCL undergoing ASCT at a single center following a TBI-based regimen (44 patients) or chemotherapy only regimen (29 patients) demonstrated no difference in OS (1 year 89% versus 90%; 3 year 71% versus 70%) or EFS (1 year 85% versus 83%; 3 year 55% versus 58%) [Peterlin et al. 2010]. The two groups were well matched for prognostic factors, and in multivariate analysis only the MIPI score and blastoid histologic subtype were predictive of OS and EFS.
The impact of initial immunochemotherapy on outcome
As the experience with ASCT in MCL first remission grew, investigators began to analyze their results to determine the impact of different therapies given prior to the ASCT and in conjunction with ASCT. Investigators at the M.D. Anderson Cancer Center determined that the addition of rituximab to chemotherapy and transplant preparatory regimens had improved the PFS for patients transplanted in first remission, but not at relapse [Tam et al. 2009]. An analysis of the outcome of 98 MCL patients who underwent ASCT in first remission after a variety of different initial chemotherapy regimens with or without rituximab and with or without high-dose cytarabine showed a trend for improved 3-year PFS (93% versus 67%) following rituximab (hazard ratio [HR] 0.37, p = 0.07), but no suggestion of improved outcome with cytarabine. An analysis of 52 relapses in 118 patients who underwent ASCT for MCL at 3 referral centers identified not receiving rituximab prior to transplant and undergoing ASCT at relapse as predictive factors for relapse, whereas cytarabine intensification, TBI conditioning, age, and year of transplant had no impact [Dietrich et al. 2010]. A large trial of 160 patients treated by the Nordic Lymphoma Group with 6 cycles of rituximab maxi-CHOP alternating with rituximab and high-dose cytarabine followed by BEAM or BEAC (cyclophosphamide substituted for melphalan) conditioning and ASCT demonstrated 6-year OS of 70%, EFS of 56%, and PFS of 66% [Geisler et al. 2008]. The nonrelapse mortality rate was 5%. These results were substantially improved compared with a historical control group, which was treated with maxi-CHOP for four cycles without rituximab and without cytarabine. Retrospective analysis of the outcome of ASCT following (R-)hyper-CVAD/high-dose methotrexate and cytarabine versus other initial chemotherapy regimens of less intensity appeared to give more favorable results following the intensive therapy, although this analysis was not based on intent to treat, and patient drop out on (R-)hyper-CVAD due to toxicity, patient or provider choice, and stem cell collection failure would be expected to be greater than after less intensive initial therapy [Vose et al. 2006]. A second group compared 21 patients undergoing ASCT following hyper-CVAD-based therapy with 15 following CHOP-based therapy in first CR or PR. There was an insignificant trend toward lower PFS following CHOP-based therapy (HR 3.67, p = 0.11) [Till et al. 2008]. The Cancer and Leukemia Group B (CALGB) published the results of 78 patients treated with rituximab, methotrexate, augmented CHOP, high-dose cytarabine and etoposide, and ASCT following carmustine, etoposide, and cyclophosphamide conditioning [Damon et al. 2009]. An additional two doses of rituximab were given post-ASCT. The 2- and 5-year PFS were 76% and 56%, respectively, with 64% 5-year OS. A smaller trial reported the results of initial cyclophosphamide, teniposide, doxorubicin, and prednisone alternating with vincristine, high-dose methotrexate, and cytarabine for 4-6 cycles followed by allogeneic matched sibling transplant if aged ≤55 if available donor or ASCT if aged ≤65 [Evens et al. 2008]. Twenty-five patients enrolled on study, with 13 undergoing ASCT and 4 undergoing allogeneic transplant. The overall TRM was 4% with overall 5-year EFS 35% and 50% 5-year OS. Vincristine, doxorubicin, dexamethasone and chlorambucil with or without rituximab has also been studied as an initial therapy in newly diagnosed MCL prior to ASCT [Gressin et al. 2010]. After four cycles of therapy, the OS was 73% with 46% CR. Toxicity was mild, with 9% grade 3–4 thrombocytopenia. A complex regimen incorporating initial debulking chemotherapy followed rituximab and high-dose sequential chemotherapy with cyclophosphamide, cytarabine, melphalan and mitoxantrone with melphalan with ASCT has also been reported, with an initial 54-month 89% OS with 79% EFS [Cortelazzo et al. 2007; Gianni et al. 2003].
The European Mantle Cell Lymphoma Network recently presented the initial results of a complex phase III randomized prospective trial comparing the outcome of R-CHOP × 6 followed by cyclophosphamide and TBI conditioning ASCT (arm A) versus R-CHOP × 3 alternating with R-DHAP × 3 followed by TBI, cytarabine and melphalan conditioning ASCT (arm B) in newly diagnosed MCL patients age <65 [Hermine et al. 2010]. In the published analysis, the two groups were well matched with regard to prognostic factors. The primary analysis included 391 patients with median age 55 and 56 in the two arms. There were no differences in 3-year OS (79% versus 80%), ORR prior to ASCT (90% versus 94%), number going to transplant (72% versus 73%), post-transplant ORR (97%), post-transplant CR (63% versus 65%) or ASCT TRM (3% versus 4%). Hematologic toxicity was greater during induction with arm B (grade 3–4 hemoglobin, leukocytes and platelets of 28%, 75% and 74% versus 8%, 48% and 9%, respectively), as was renal toxicity (creatinine grade 1–2 of 38% versus 8%). At ASCT, grade 3–4 mucositis was more frequent in arm B (61% versus 43%). The trade off for the greater toxicity in arm B was a higher CR/CRu rate prior to transplant (60% versus 41% arm A, p = 0.0003) and longer TTF at not reached versus 49 months in arm A, p = 0.0384, HR 0.68.
Allogeneic transplant as part of the initial therapy
Given the advanced median age of patients with MCL and the relatively high-risk nature of allogeneic transplant, it is not surprising that this aggressive approach has been used sparingly as initial therapy, although the development of reduced intensity conditioning has increased the application of this therapy in the relapsed setting [Cook et al. 2010]. An analysis of allogeneic transplant versus ASCT in first remission at Johns Hopkins University revealed no difference in EFS estimated at 70% at 3 years [Kasamon et al. 2005]. A comparison of ASCT and allogeneic transplant for 97 patients with MCL at various points in the disease course at the University of Nebraska Medical Center revealed similar 5-year EFS (39% versus 44%) and 5-year OS (47% versus 49%) due to a higher TRM rate for allogeneic transplant (day 100 19% versus 0%) offsetting a lower relapse rate (21% versus 56%) [Ganti et al. 2005].
Less aggressive approaches
Chemotherapy and immunochemotherapy regimens
Given the advanced age of many patients with MCL, and the substantial toxicity associated with aggressive chemotherapy regimens and/or ASCT in first remission, numerous investigators have conducted trials using less intense therapy. The Wisconsin Oncology Network treated 22 patients with newly diagnosed MCL with a modified version of R-hyper-CVAD which contained no methotrexate or cytarabine [Kahl et al. 2006]. Rituximab was also given as maintenance for 2 years following completion of 4–6 cycles of therapy. The ORR was 77% with 64% CR. The median PFS was 37 months, and median OS had not been reached. Two patients died of TRM during immunochemotherapy. Molecular remission did not predict for improved outcome in this small study. Cladribine has been studied in MCL as either a single agent or in combination with other agents. Single-agent cladribine given as a 2-hour daily infusion × 5 days resulted in an ORR of 81% with 42% CR in untreated MCL with an ORR of 46% and CR rate of 21% in relapsed disease [Inwards et al. 2008]. A second trial of the same therapy in a variety of subtypes of lymphoma as either initial therapy or at relapse yielded an ORR of 58% in MCL [Rummel et al. 1999]. Cladribine in combination with mitoxantrone yielded an ORR of 100% with a 44% CR rate in MCL with a median duration of remission of 24 months [Rummel et al. 2002]. The combination of rituximab and cladribine yielded a CR rate of 52% (15/29) with only 3/15 relapsing at a median follow up of 21.5 months [Inwards et al. 2008]. Bortezomib was combined with traditional R-CHOP21 in a group of 36 untreated MCL patients. The ORR was 81% with 64% CR/CRu and 2-year PFS 44% [Ruan et al. 2010b, 2009]. Neuropathy occurred in 56% of patients, and was grade 2 in 8% of patients and grade 3 in 4% of patients despite decreased bortezomib dose intensity compared with single-agent therapy, likely because of concomitant vincristine therapy. A phase III randomized trial was conducted between rituximab and bendamustine (RB) and R-CHOP21 as first-line therapy in 549 patients with a variety of low-grade lymphomas, follicular lymphoma, and MCL [Rummel et al. 2009]. A total of 18% and 19% of patients on the two arms, respectively, had MCL. Published data to date do not specifically report the results in MCL, although in the cohorts as a whole RB was less toxic (grade 3–4 neutropenia 10.7% versus 46.5%) and had superior median PFS (54.8 versus 34.8 months, p = 0.0002). Proper interpretation of this trial awaits histology-specific results. Recently, the results of 20 patients with newly diagnosed low MIPI MCL treated with rituximab and oral chlorambucil followed by rituximab maintenance were published [Sachanas et al. 2010]. The ORR reported was 95% with 90% CR. Three-year PFS was 89%
Deferred initial therapy
While most patients with MCL are treated immediately at diagnosis, a report from Weill Cornell Medical Center raises doubts as to whether immediate therapy is necessary [Martin et al. 2009]. A total of 31 out of 97 patients with newly diagnosed MCL were observed for >3 months before initial systemic therapy (median time to treatment 12 months; range 4–128 months). Median follow up for the initial observation group was 55 months, and 54% were either intermediate or high risk by MIPI score. In multivariate analysis, time to treatment did not predict for overall survival, although the initially observed group had superior survival to those treated immediately with median OS not reached versus 64 months (p = 0.004).
Consolidation and maintenance approaches
Given the tendency for MCL to respond at high rates with subsequent high relapse rates, the role of consolidation and or maintenance has naturally been investigated. The most common consolidation strategy has been ASCT, as discussed above. Trials have also investigated the efficacy of radioimmunoconjugate therapy following initial chemotherapy with mature results awaited. Previous European studies demonstrated that interferon-alpha maintenance was inferior to ASCT [Dreyling et al. 2005]. Maintenance thalidomide was given in a small phase II trial with unclear impact [Lossos et al. 2010]. Rituximab has been used in the maintenance setting in upfront trials [Kahl et al. 2006], and results of randomized phase III trials are awaited with anticipation. Rituximab maintenance in the relapsed MCL setting has been shown to prolong response duration after R-FCM therapy [Forstpointner et al. 2006]. Some have suggested the use of rituximab to treat patients with molecular relapse post-ASCT in order to preempt clinical relapse [Ladetto et al. 2006].
Agents with efficacy at relapse and new regimens to be tested as part of initial therapy
Bortezomib has been extensively evaluated in the relapsed MCL setting and is increasingly being tested in addition to other agents as part of initial therapy. The single-agent ORR in relapsed MCL has been reported as being 33–46% with median TTP of 6.2 months in the largest study [Goy et al. 2009; Fisher et al. 2006]. The primary toxicity of this agent is neuropathy, with 30% (10/30) of patients in a study including both untreated and relapsed MCL patients discontinuing therapy for neurologic toxicity and myalgia [Belch et al. 2007]. Five serious adverse events including 2 deaths associated with fluid retention in the first 12 patients treated led to exclusion of patients with pleural effusions or edema. Bortezomib was combined with R-hyper-CVAD in a phase I clinical trial with a reduced frequency compared with the single-agent schedule [Romaguera et al. 2010b]. Phase II results as initial therapy are anticipated.
Temsirolimus has been tested as a single agent for the treatment of relapsed MCL in a variety of doses and schedules. The North Central Cancer Treatment Group performed 2 phase II trials at a dose of 250 mg weekly and subsequently 25 mg weekly [Ansell et al. 2008; Witzig et al. 2005]. Despite a 10-fold reduction in dose, the ORR in the low-dose trial was not different from that in the high-dose trial (41% versus 38%), nor was the TTP (6 versus 6.5 months). Hematologic toxicity was greater in the high-dose study with 71% grade 3 and 11% grade 4 toxicity, most of which was transient thrombocytopenia resolving in 1 week. The addition of rituximab to the low-dose schedule resulted in an ORR of 59% with 19% CR and median TTP 9.7 months [Ansell et al. 2011]. A phase III randomized trial in relapsed MCL tested 2 different schedules of temsirolimus versus single-agent investigator’s choice therapy [Hess et al. 2009]. The median PFS was 4.8, 3.4, and 1.9 for the 175/75-mg temsirolimus group, 175/25-mg temsirolimus group, and investigator’s choice, respectively. The high-dose temsirolimus arm was superior to investigator’s choice (p = 0.0009). This agent is being incorporated into first-line therapy trials.
Lenalidomide also appears to have significant activity in MCL, with a subset analysis of 15 patients with relapsed or refractory MCL on a phase II trial revealing a 53% ORR (3 CR, 5 PR) [Habermann et al. 2009]. The median PFS was 5.6 months. In a second larger trial, 24 of 57 MCL patients responded (21% CR/CRu, 21% PR) with median PFS 5.7 months [Witzig et al. 2011]. The combination of flavopiridol, fludarabine, and rituximab was tested in a phase I trial in a variety of B-cell disorders, including 10 patients with MCL [lin et al. 2010]. Eight of 10 MCL patients responded, with a median PFS of 26 months. A complex regimen comprising rituximab, thalidomide, prednisone, etoposide, procarbazine, and cyclophosphamide was administered to 25 patients with relapsed MCL [Ruan et al. 2010a]. The ORR was 73% (CR/CRu 32%, PR 41%) with median PFS 10 months. There were five grade 3/4 infections and two episodes of thrombosis. Grade 3/4 neutropenia occurred in 64% of patients.
Conclusions
Therapeutic outcomes in MCL have improved over the last two decades as a result of clinical investigation, and further progress is likely as newer agents, combinations, and strategies are optimized. At the present time, the first consideration in choosing initial therapy for a patient with MCL is determining whether the patient is willing and able to tolerate a relatively more aggressive approach, or whether a less intensive regimen is more appropriate or desirable for the patient. For those interested in a more aggressive approach, good results have been obtained either through intense chemoimmunotherapy approaches alone or with ASCT consolidation. Issues requiring further clarification include determination of the best induction regimens, conditioning regimens, and surveillance and maintenance strategies following initial therapy.
For patients either desiring a less intense approach or ineligible for aggressive therapy due to advanced age or comorbidities a number of other therapeutic options are available which have resulted in promising results. An initial period of observation may also be appropriate in selected patients. Areas of further investigation include the determination of the best regimens and role of maintenance therapy.
As no standard therapy for MCL has been identified, participation in well-designed clinical trials is essential if the trend of improving outcomes over time is to be sustained.
Footnotes
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Dr. Witzig has a financial interest in the use of temsirolimus in the treatment of mantle cell lymphoma (Patent on the idea of using it for MCL, this does not provide for royalties on the sale of the drug). Further details are available on request.