Diagnosis and treatment of mantle cell lymphoma

a Oncology-Haematology, Kantonsspital St Gallen, Switzerland b Oncology-Haematology, Kantonsspital Aarau, Switzerland c Department of Pathology, Kantonsspital St Gallen, Switzerland d Oncology-Haematology, Kantonsspital Baselland, Liestal, Switzerland e Oncology, Kantonsspital Münsterlingen, Switzerland f Onkozentrum Zürich, Switzerland g Oncology and Haematology, Kantonsspital Graubuenden, Chur, Switzerland


Introduction
Mantle cell lymphoma (MCL) accounts for an estimated 3%-6% of all non-Hodgkin's lymphoma (NHL) cases [1][2][3].This relatively rare lymphoma entity is characterised by an unfavourable clinical course with median overall survival (OS) of only 4-5 years [1,3,4,5].Indeed, MCL comprises the worst traits of both indolent and aggressive lymphomas, because it combines incurability with aggressive growth.Most patients present with an advanced stage of disease at diagnosis, with involvement of multiple lymph nodes, blood, spleen, bone marrow and the gastrointestinal (GI) tract.Although the disease demonstrates an initially encouraging response to treatment, its clinical course is usually marked by recurrent relapses, resulting in a dismal long-term outcome [6].Standard therapy for MCL consisted of chemotherapy regimens such as CHOP (cyclophosphamide, doxorubicin, vincristine and prednisone).The addition of the monoclonal antibody rituximab to the standard CHOP regimen improved outcome parameters such as overall response rate (ORR), complete response (CR) and OS [7][8][9].The evidence for the improvement in OS is based on a systematic review and on SEER-Medicare data [7,9].Response duration has been further improved with the application of more aggressive chemotherapy regimens including high-dose cytarabine (Ara-C) alongside high-dose chemotherapy (HDT) with autologous stem cell transplantation (ASCT).But the outcome for those who relapse after firstline treatment or ASCT is poor, and salvage therapy is usually followed by only a short duration of response.An additional obstacle to successful treatment is the fact that many MCL patients are elderly (over 65 years of age) and are not candidates for either high-dose AraC-containing regimens or HDT.This review will discuss the key factors that drive clinical practice with respect to diagnosis, treatment, and follow-up of MCL patients.

Clinical presentation
MCL patients are predominantly male (ratio of 2:1 or greater), with a median age at diagnosis of 60-65 years [10,11].Most patients initially present with stage III-IV generalised, nonbulky lymphadenopathy, often with extranodal involvement.The bone marrow, tonsils, spleen, liver and GI tract are among the most common extranodal sites.Leukaemic spread can also be detected in many cases [12].Although common, GI involvement may not be clinically obvious during presentation and is often missed unless endoscopy studies are performed [13].Data from a prospect-ive study in 13 untreated MCL patients at diagnosis revealed that 92% of this group had upper or lower GI tract infiltration by MCL [14], suggesting that the actual incidence of GI involvement may be much higher than the estimated 30% who present with obvious GI symptoms.The incidence of central nervous system (CNS) manifestation in MCL patients has been reported to range from 4%-26% [15].In most cases, CNS infiltration is a late event in the course of the disease, particularly in patients with blastoid histology [15].

Diagnosis
The diagnosis of MCL is made on biopsies of lymph nodes, bone marrow or other affected tissues, or through blood examinations [11].Classical MCL exhibits a typical morphology of small-to medium-sized monomorphic lymphoid cells with indented nuclear contours, dispersed chromatin, scant cytoplasm and inconspicuous nucleoli [16].In contrast to follicular lymphoma (FL), marginal zone B-cell lymphoma (MZL), lymphoplasmacytic lymphoma or small lymphocytic lymphoma (SLL) / chronic lymphocytic leukaemia (CLL), blastic cells such as centroblasts, immunoblasts, or paraimmunoblasts, respectively, are not found in MCL [12].Besides classical MCL, there are four cytological variants, namely the small cell variant, the marginal zone-like variant, the pleomorphic variant and the blastoid variant (fig. 1) [17].Of these, the pleomorphic and blastoid variants are usually associated with an even poorer prognosis [17,18].As these variants morphologically overlap, in particular with SLL/CLL, MZL, precursor B-cell lymphoblastic leukaemia/lymphoma (PBLL/L) and DLBCL, additional immunophenotyping is mandatory for a more comprehensive definition of MCL.Phenotypically, MCLs are positive for the B-cell markers CD20 and CD79a as well as for lineageunspecific CD5.In addition, they exhibit immunoglobulin (Ig) expression, preferably with lambda light-chain restriction, and in most cases overexpress cyclin D1 (CCND1) and SOX-11 in the nuclei.Although the latter two are highly characteristic of MCL, both may also be found in other subtypes of lymphomas, such as plasma cell myeloma (PCM), hairy cell leukaemia (HCL), Burkitt's lymphoma and lymphoblastic lymphomas [19,20].More detailed differential diagnostic immunophenotypic profiling is summarised in table 1.The ectopic and deregulated expression of cyclin D1 in MCL is due to the chromosomal translocation t(11;14)(q13;q32), which fuses the enhancer-promoter of the Ig heavy chain gene with the transcription unit of the CCND1 proto-oncogene [21].However, a few cases of MCL have been reported with alternative overexpression of cyclin D2 or D3 instead of D1, though additional con-

Staging
Initial work-up typically includes a complete blood count, chemistry tests, screening for human immunodeficiency virus (HIV) and hepatitis B and C. Staging procedures are performed with the evaluation of blood and bone marrow by flow cytometry, and computed tomography (CT) of the neck, chest, abdomen and pelvis [11].CT with 18 Ffluorodeoxyglucose positron emission tomography (FDG-PET -CT) may be useful in patients with limited-stage disease in order to confirm that the extent of their disease has not been underestimated, and there is also preliminary evidence that post-treatment PET-CT results may predict clinical outcome as demonstrated in a subset of MCL patients who received first-line treatment with high-dose chemotherapy [25].The involvement of the GI tract should be assessed with endoscopy if the patient exhibits GI symptoms or prior to starting treatment with a dose-intensive regimen [11].However, the relevance of GI tract staging to prognosis remains undefined.The cerebrospinal fluid is examined only when neurological symptoms are evident or if the patient has a blastoid variant of the disease [15], but there are no formal recommendations for CNS prophylaxis at either diagnosis or relapse.Although CNS prophylaxis is not routinely justified for all patients, Gill et al. [26] suggest that it may benefit patients with blastoid MCL or those with systemic relapse after first-line therapy.In general, however, we do not recommend CNS prophylaxis.

Prognosis
The first prognostic index for MCL was recently developed by the European MCL Network (mantle cell international prognostic index; MIPI) [27].The MIPI stratifies patients into low, intermediate and high risk groups, based on four independent prognostic factors: age, performance status, lactate dehydrogenase level and leukocyte count.A simplified version of the MIPI has also been described by the same group [27].Although it has been independently validated for use in MCL [28], use of the MIPI is limited.This index has prognostic value only for predicting OS, but cannot predict response to chemotherapy or progression-free survival (PFS) [23].
Another important prognostic factor for MCL not included in the MIPI is proliferation measured by the nuclear immunoreactivity of Ki-67 [29].The Ki-67 index is the most powerful single prognostic factor for OS, retaining its prognostic relevance in patients treated with rituximab-containing regimens [30].Although MCL may present as aggressive disease, it can take an indolent course in approximately 30% of patients [31,32].Survival is longer in this subset (up to 7-10 years), and these patients may be observed for months to years before needing chemotherapy.Observational data suggest that delayed therapy has no apparent negative impact on their overall survival [31].Although there is consensus on those patients who require immediate therapy, the identification of patients with indolent MCL who may benefit from delayed therapy remains a challenge; however, these cases are clinically leukaemic, phenotypically SOX-11 negative and genetically hypermutated with respect to the variable Ig heavy chain genes.Nevertheless, once chemotherapy is warranted in these patients, their outcomes appear to be similar to those who initially present with more aggressive disease [23].

First-line treatment
Because of the rarity of MCL, there is a paucity of large randomised studies and most of the available data are based on smaller phase II trials.In general, therapeutic approaches can be divided into two clinical contexts: the treatment of patients who are eligible for SCT, and those who are unable to undergo SCT (table 2).

Limited-stage MCL
Approximately 6%-8% of MCL patients present with limited-stage disease [33], but few data are available regarding the disease course in these patients.Leitch et al. [34] retrospectively described the clinicopathological features and outcomes of 26 patients who were identified with limited-stage MCL at the British Columbia Cancer Agency since 1984.Patients with stage IA or IIA, nonbulky (<10 cm) disease were treated according to standard protocols, namely with single-or multiple-agent chemotherapy and/ or involved-field radiation therapy (RT).Median age at diagnosis was 68 years, and the median OS for the whole group was 6.8 years.The authors noted that initial treatment with RT was the most important factor for improved PFS rates (5-year PFS 68% vs 11%; p = 0.002) with a trend towards improved OS in patients receiving RT (6-year OS

Review article
Swiss Med Wkly.2013;143:w13868 71% vs 25%; p = 0.13).Of particular importance, a plateau in the PFS curve at 60% was seen in patients receiving RT and 6 of 17 irradiated patients were free of progression after >5 years.Similar results were reported in a more recent study, which retrospectively assessed 26 patients with stage I and II MCL [35].The authors focused their analysis on the 21 patients who were treated with curative intent (chemotherapy and/or RT, followed by HDT with ASCT in two patients).An ORR of 95% as well as a local control rate of 95% was achieved in this subset of patients, leading to a median PFS and OS of 3.2 and 6.4 years, respectively.Systemic relapse was more common in patients with stage II disease and in those with the blastoid variant of MCL.
Owing to the small number of patients in all published trials [36,37], the power of any comparisons is very limited.Although limited-stage MCL is associated with better outcomes, systemic relapse is still a significant problem.
On the basis of these scant data, outside the context of clinical trials, a treatment strategy using brief multiagent (immuno)chemotherapy in combination with involved field RT is recommended for patients with limited-stage MCL.

Younger, transplant-eligible patients
In younger, fit patients with few or no comorbidities, the therapeutic approach places emphasis on more intense regimens which combine aggressive chemotherapy with rituximab, with or without HDT followed by ASCT.In this clinical context, dose-intensified high-dose Ara-C-containing induction regimens in combination with rituximab are considered as the current standard of care [38].
In a randomised trial of the German Low Grade Lymphoma Study Group (GLSG), the addition of rituximab to CHOP significantly improved ORR and CR rates [8].In an update of this trial after a median follow-up of 65 months, median time to treatment failure (TTF) was prolonged from 14 months for CHOP to 28 months for R-CHOP (p = 0.0003).However, in contrast to other lymphoma entities, no clear improvements of OS were observed, with 5-year OS rates of 59% for R-CHOP and 46% for CHOP (p = 0.27).However, with only 122 patients randomised, the trial was underpowered to detect clinically relevant differences in this secondary endpoint [39].
Owing to the failure of R-CHOP-like regimens alone to prolong OS, more aggressive treatment regimens with HDT followed by ASCT have been explored.More recent studies have shown promising survival data with doseintensified induction regimens that incorporate high-dose Ara-C [40].The combination of rituximab alongside six to eight cycles of fractionated cyclophosphamide, vincristine, doxorubicin and dexamethasone (HyperCVAD) alternating with high doses of Ara-C and methotrexate yielded an 87% CR/CRu and an impressive 3-year failure-free survival of 73% in the subgroup of patients younger than 65 years of age [40].Another rituximab and high-dose Ara-C-containing treatment schedule alternating with a dose-intensified R-CHOP regimen ("maxi-R-CHOP") followed by HDT with ASCT (MCL-2 trial) was tested by the Nordic Lymphoma Group in a large nonrandomised phase II study [41].In comparison with their previous study (MCL-1 trial), which did not contain high-dose Ara-C, results with respect to EFS, PFS and OS were significantly improved.A phase II study by the French Groupe d'Etudes des Lymphomes de l'Adulte (GELA) has also shown promising results using a rituximab and high-dose Ara-C-containing induction regimen before HDT followed by ASCT [42].
Recently, the presentation of the results from the randomised European MCL Network Younger trial has further strengthened the role of Ara-C in front-line therapy for MCL [43].In this large trial, a total of 497 patients up to the age of 65 years with previously untreated MCL were randomised to receive either six courses of CHOP plus rituximab followed by myeloablative radiochemotherapy (12 Gray total body irradiation [TBI], 2x60 mg/kg cyclophosphamide) and ASCT (control arm A) versus al- ternating courses of 3x CHOP and 3x DHAP plus rituximab followed by a high-dose Ara-C-containing myeloablative regimen (10 Gray TBI, 4x1.5 g/m 2 Ara-C, 140 mg/m 2 melphalan) and ASCT (experimental arm B).CR and CR/CRu rates were significantly higher in arm B (25% vs 36%; p = 0.012 and 40% vs 54%; p = 0.0003).After a median follow up of 51 months, a lower number of relapses occurred in the high-dose Ara-C-containing treatment arm, leading to a significantly longer TTF (46 months vs 88 months; p = 0.0382, hazard ratio [HR] 0.68).In addition, OS was improved (82 months vs median not reached; p = 0.045).The authors concluded that an induction regimen containing high dose Ara-C followed by HDT followed by ASCT should become the new standard of care of MCL patients up to 65 years of age [43].
Although HDT followed by ASCT had been shown to improve PFS after CHOP-like induction chemotherapy versus conventional chemotherapy alone [44], no prospective comparisons are available with regard to its use after the above-mentioned more intensive Ara-C-and/or rituximabcontaining induction regimens.Recently, a large retrospective analysis of 167 younger MCL patients (<65 years) from the National Comprehensive Cancer Network (NCCN) NHL database compared the outcome of patients treated with either the R-HyperCVAD regimen or the R-CHOP regimen with or without HDT/ASCT as first-line therapy [45].Interestingly, outcomes were comparable between patients treated with R-HyperCVAD, with or without HDT/ASCT, and R-CHOP followed by HDT/ ASCT -3-year PFS rates were 58% (95% confidence interval [CI] 44%-69%) for the R-HyperCVAD arm, 55% (95% CI 22%-79%) for the R-HyperCVAD + HDT/ASCT and 56% (95% CI 33%-74%) for the R-CHOP + HDT/ASCT arm, respectively) -thereby challenging the role of ASCT in patients treated with intensive Ara-C-containing induction regimens.In contrast, the R-CHOP only arm (without HDT/ASCT consolidation) demonstrated an inferior 3-year PFS of 18% (95% CI 6%-36%) compared with all the other treatment regimens.The rate of treatment-related complications was higher for patients treated with R-HyperCVAD with or without HDT/ASCT.Though there is no consensus on the optimal pretransplant induction regimen, results so far point towards intensive combinations that include high-dose Ara-C.At present, HDT followed by ASCT plays a role as part of the frontline treatment and consolidates the best possible clinical response following upfront intensive immunochemotherapy.

Elderly, fit patients and younger patients not eligible for transplant
The management of elderly patients with no significant comorbidities and younger patients who are not eligible for transplant follows similar strategies.There are two main features of the treatment of this patient population: the identification of an effective induction regimen, and the use of remission consolidating and maintenance therapy to prolong the duration of response.A recent phase III trial conducted by the European MCL Network had the two-fold objective of comparing R-CHOP with R-FC (rituximab plus fludarabine and cyclophosphamide) as induction treatment, and comparing maintenance with rituximab versus interferon-α (IFN-α) in elderly MCL patients [46].Patients were either ≥66 years of age, or 60-65 years if ineligible for HDT with ASCT, with an Eastern Cooperative Oncology Group (ECOG) status of ≤2.The key findings from this trial showed that R-CHOP was superior to R-FC in terms of efficacy and tolerability (4-year OS 62% for R-CHOP vs 47% for R-FC; p = 0.005), and that rituximab maintenance was able to prolong the duration of remission.Maintenance rituximab not only conferred a survival advantage over maintenance IFN-α, but was especially beneficial to those who had been successfully pretreated with R-CHOP (4-year OS 87% vs 63% in favour of rituximab; p = 0.005).The duration of remission and OS after a response to R-CHOP were significantly shorter amongst patients who were not assigned to any maintenance therapy, as compared with those who were randomly assigned to IFN-α (p = 0.002 and p <0.001, respectively).As an alternative to the R-CHOP induction / rituximab maintenance regimen, a consolidation strategy with radioimmunotherapy can be considered [47].Other induction regimens have been explored in this clinical setting, including R-HyperCVAD and R-bendamustine.Ten-year follow-up of a prospective phase II trial conducted by the MD Anderson Cancer Centre in 97 elderly patients treated with R-HyperCVAD alternating with rituximab plus high-dose methotrexate and Ara-C without HDT and ASCT, showed that the median OS for all patients had not been reached and the median TTF for all patients was 4.6 years [48].These promising results, however, have been tempered by the significant mucosal and haematological toxicities as well as the 8% treatment-related death rate associated with this regimen [40,49].This regimen is therefore not recommended for elderly patients with comorbidities.R-bendamustine front-line therapy has shown encouraging results in a randomised phase III trial in elderly patients with indolent and mantle cell lymphoma.At a median follow-up of 45 months, PFS was significantly prolonged with R-bendamustine compared with R-CHOP (median PFS 69.5 vs 31.2 months in favour of R-bendamustine; HR 0.58, 95% CI 0.44-0.74;p <0.0001) [50].
The superior response of R-bendamustine over R-CHOP and R-CVP was confirmed in the Bright study [51].Furthermore, patients receiving R-bendamustine reported improved quality of life (QOL) parameters, as measured by the global health status (GHS)/QOL scale and the QLQ-C30 questionnaire [52].Thus, R-bendamustine may provide an effective alternative to R-CHOP in the front-line treatment of elderly patients and those who are not eligible for transplant.The feasibility of combining this regimen with rituximab maintenance is an attractive option, and is currently under investigation.

Elderly, unfit patients
There are only limited treatment options for elderly, unfit patients with multiple comorbidities.In the case of asymptomatic patients with nonbulky disease, low Ki-67 levels and normal blood counts, initial therapy may be deferred until the onset of disease symptoms [53].For symptomatic patients, less aggressive treatment regimens may be used, such as rituximab in combination with low-dose bendamustine [54] or oral chlorambucil [10].Rituximab mono-therapy may be another option for elderly patients unable to tolerate chemotherapy.However, in contrast to the findings in FL, patients receiving prolonged treatment with single-agent rituximab did not show an improved clinical response [55].Maintaining the balance between durable remission and treatment-induced toxicity is particularly challenging, and patients are strongly encouraged to obtain treatment within the context of a clinical trial.

Treatment of relapsed disease
Despite the improved activity of first-line treatment, nearly all MCL patients experience relapse.There is a lack of randomised trials comparing the relative survival advantages of various second-line options.In general, patients with relapsed disease can be broadly classified into those who are transplant-eligible, and those for whom transplant is not an option.

Transplant-eligible patients
Regardless of the induction regimen, consolidation therapy with SCT is a key option for patients who experience relapse [56].HDT followed by ASCT is recommended for eligible patients who did not undergo this as part of their first-line treatment.Allogeneic SCT is a possibility for young, fit patients with no signs of CNS disease and who have a compatible donor [57].An analysis of the longterm outcomes (median follow-up 56 months) of 35 patients with relapsed/refractory MCL who underwent nonmyeloablative allogeneic SCT revealed a median PFS duration of 60 months, while the median OS was not yet reached.The 6-year actuarial PFS and OS rates were 46% and 53%, respectively [58].Nevertheless, mortality due to graft-versus-host disease and secondary infections are major limiting factors, and careful patient selection remains a key factor in the outcome of this treatment modality [53,59].

Patients not eligible for transplantation
Relapsing patients who are not eligible for SCT have very few options, aside from receiving treatment with new agents in a clinical trial.In the cases where the duration of remission was longer than 12 months, the patient may be rechallenged with the same induction regimen.Generally, non-cross-resistant regimens are used.Outside of clinical trials, the combination of rituximab plus chemotherapy according to the patient's performance status forms the basis of treatment, supplemented by rituximab maintenance therapy where possible [60,61].There is no regimen that has been shown to be superior, although several combinations have been tested in this patient subset, including bendamustine [62], R-bendamustine [63][64][65], R-FC [66] and R-FCM [67].Treatment with single-agent radioimmunotherapy is another option in the relapsed setting [68].The proteasome inhibitor bortezomib has been tested as a monotherapy, resulting in ORR of around 30% [69,70], including 8% CR/CRu [70].Bortezomib has also been tested in combination with other agents, including rituximab plus bendamustine [71] and rituximab plus dexamethasone [72].
A second novel class of agents are inhibitors of the mammalian target of rapamycin (mTOR), such as temsirolimus and everolimus.They act via the PI3 kinase/Akt/mTOR signalling pathway implicated in the pathogenic deregulation of cyclin D1 in MCL [73].In a randomised open-label phase III trial in 162 relapsed/refractory MCL patients, temsirolimus was evaluated in comparison with investigator's choice of therapy.The high-dose temsirolimus regimen (175 mg/75 mg) resulted in an ORR of 22% and a significantly longer PFS compared with investigator's choice therapy (median PFS 4.8 vs 1.9 months; p = 0.0009; HR = 0.44) [74].A number of ongoing trials are currently exploring the feasibility of combining temsirolimus with various chemotherapy regimens in the salvage treatment setting [73].Everolimus has been tested in a couple of small phase II trials so far and demonstrated similar response and PFS rates [75].
The immunomodulatory agent lenalidomide has also shown encouraging activity.Subgroup analysis of 15 relapsed or refractory MCL patients from an open-label phase II NHL trial resulted in a 53% ORR and a median duration of response of 13.7 months [76].Lenalidomide has also been effectively combined with rituximab, yielding 57% ORR, 36% CR and a median response duration of around 19 months with a reasonable tolerability profile

Review article
Swiss Med Wkly.2013;143:w13868 Swiss Medical Weekly • PDF of the online version • www.smw.ch[77].A multicentre phase II trial by the European MCL Network is currently underway evaluating the activity of lenalidomide with bendamustine plus rituximab in MCL patients in first relapse or with primary refractory disease [38].
Encouraging data has been obtained for ibrutinib, an orally administered inhibitor of Bruton's tyrosine kinase (BTK) [78].Preliminary results from a phase II trial in 110 evaluable patients with relapsed or refractory MCL (subdivided into bortezomib-naïve or bortezomib-exposed groups) showed ORR, CR and PR rates of 68%, 22% and 46%, respectively [78].Median duration of response, PFS and OS have not been reached.The durability of the responses, alongside the drug's tolerability profile, suggest that ibrutinib is a strong candidate for the treatment of relapsed/refractory MCL, warranting further investigation.
In addition to the above agents, other new drugs are being actively explored for the treatment of relapsed/refractory MCL.A complete overview of these novel therapies is beyond the scope of this review, but a summary of new agents under investigation is provided in table 3.

Follow-up
The primary purpose of patient follow-up in MCL is to identify relapse or progression.The majority of relapses occur within 2 years after first-line therapy.Although there is no data to support a specific follow-up strategy in MCL, patients are usually seen every 3 months for the first 2 years after treatment, every 4-6 months for an additional 3 years, and subsequently once a year.

Conclusions
The choice of strategy for approaching the treatment of MCL will remain a complex problem that still requires evidence-based guidance.Yet achieving cure for MCL remains elusive; in the absence of cure, the treatment goal is to prolong survival while maintaining the patient's quality of life.Increasing the intensity of therapeutic regimens can be a double-edged sword, prolonging a patient's survival but compromising on quality of life.Until we achieve cure, we must continue to pursue novel therapies that enable us to enhance not only the duration of the response but also the quality of the extended survival conferred to the patient.

Table 1 :
Differential diagnostic immunophenotypic profiles of mantle cell lymphoma and other B-cell lymphomas.

Table 2 :
Summary of the management of the subgroups of patients with mantle cell lymphoma.
AraC = cytarabine; ASCT = autologous stem cell transplantation; HDT = high-dose therapy; MCL = mantle cell lymphoma; R-CHOP = rituximab plus cyclophosphamide, doxorubicin, vincristine and prednisone; R-HyperCVAD = rituximab plus hyperfractionated cyclophosphamide, vincristine, doxorubicin and dexamethasone; RT = radiation therapy Levels of evidence: I -Available evidence from at least one high quality, large, randomised, controlled trial or meta-analyses of several high quality randomised trials.II -Smaller randomised studies or subanalyses of larger randomised trials with possible bias, or meta-analyses of such trials or meta-analyses of heterogeneous trials.III -Prospective cohort studies.IV-Retrospective cohort studies, case-control studies.V -Uncontrolled studies, individual case reports, expert opinions.Swiss Medical Weekly • PDF of the online version • www.smw.ch