Chapter 100: Mantle Cell Lymphoma

Mantle cell lymphoma (MCL) was originally named centrocytic lymphoma or subsumed under the term intermediate lymphocytic lymphoma. In 1992, the term mantle cell lymphoma was adopted for this entity because of morphologic and immunophenotypic similarities of the malignant cells to lymphocytes of the mantle zone of germinal centers.¹ In 1994, the term mantle cell lymphoma was incorporated into the revised European-American classification of the International Lymphoma Study Group, and remains a distinctive lymphoma subtype in the World Health Organization classification of malignant lymphoid disorders.^2,3

Based on cytology, the classical form is characterized by small to intermediate-size cells with irregular, cleaved nuclei, dense chromatin, and indistinct nucleoli. A small cell variant, resembling chronic lymphocytic leukemia (CLL), may be associated with a more indolent course.⁴ In contrast, the blastoid cell variant, including a blastic and a pleomorphic phenotype, displays a more aggressive clinical course.³

Histologically, MCL most frequently displays a diffuse infiltration of the lymph nodes, less commonly a nodular pattern, and, rarely, a mantle zone pattern, the latter of which may represent an earlier phase of the disease.⁵ The immunophenotype of the cells resembles the lymphocytes in the mantle zone of normal germinal follicles, and is characterized by coexpression of B-cell antigens (CD19+, CD20+, CD22+, CD43+, CD79+, secretory immunoglobulin [sIg] M+, sIgD+) and the T-cell associated marker CD5+. Based on their predominantly pregerminal center origin, MCL cells stain strongly for the antiapoptotic protein BCL-2, but are negative for germinal center markers like CD10 and BCL-6.³

Because of the morphologic heterogeneity of MCL, detection of the MCL genetic hallmark, either by immunohistochemistry (cyclin D1 overexpression) or fluorescence in situ hybridization (chromosomal translocation t[11;14][q13;q32]) is crucial to confirm the diagnosis. In rare cases that are negative for cyclin D1, staining for SOX11, a transcription factor specifically expressed in more than 90 percent of MCL cases, may help to establish the diagnosis.⁶

MCL represents approximately 6 percent of all non-Hodgkin lymphomas, although a lower incidence has also been reported.^7,8 Between 1992 and 2004, the age-adjusted annual incidence more than doubled from about 0.3 to 0.7 cases per 100,000. The incidence is more than twice as high in males and increases with age. Median age at presentation is approximately 65 years. No specific etiologic agent has been associated with MCL.

The chromosomal translocation t(11;14) results in overexpression of cyclin D1, a cell-cycle protein not normally expressed in lymphoid cells. The very rare cyclin D1–negative cases usually overexpress cyclin D2 or D3.⁶ The cytogenetic abnormality t(11;14) is the primary event in the pathogenesis of MCL, facilitating the deregulation of the cell cycle at the G₁–S phase transition.⁹ However, additional genetic events are required for the clinical manifestation of MCL. Thus, a low copy-number of the t(11;14) translocation has been found in the blood cells of 1 to 2 percent of healthy individuals without evidence of clinical disease.¹⁰ Cytogenetic studies have identified frequent secondary genetic alterations that are involved in cell-cycle dysregulation and DNA repair, which may explain why MCL has one of the highest levels of genomic instability among the malignant lymphoid neoplasms. These genetic abnormalities include losses in chromosomes 1p13–p31, 2q13, 6q23–27, 8p21, 9p21, 10p14–15, 11q22–23, 13q11–13, 13q14–34, 17p13, and 22q12; gains in chromosomes 3q25, 4p12–13, 7p21–22, 8q21, 9q22, 10p11–12, 12q13, and 18q11q23; and high copy-number amplifications of certain chromosomal regions.¹¹

Figure 100–1 depicts a proposed scheme relating genetic derangements with specific MCL subtypes. Cell-cycle dysregulation is a hallmark of MCL. The overexpressed cyclin D1 complexes with CDK4, which results in phosphorylation of the retinoblastoma gene RB1 and release of elongation factor 2 (E2F) transcription factors and progression of the lymphoma cells into S-phase. In addition, mutation of the ataxia-telangiectasia mutant (ATM) gene facilitates genomic instability in lymphoma cells through impaired response to DNA damage. Phosphoinositol 3′-kinase (PI3K) and mammalian target of rapamycin (mTOR) are important downstream targets of this signaling pathway. Finally, specific gene alterations, namely p53 or p16/CDKN2, are associated with the blastoid variant and poor clinical outcome.^12,13

According to the characteristic cyclin D1 overexpression, molecular profiling has identified a cell proliferation gene signature that distinguishes patient subsets that differ by more than 5 years in median survival.¹⁴ A five-gene model to predict survival in MCL based on formalin-fixed, paraffin-embedded tissue has been devised.¹⁵ In the clinical setting, only Ki-67 expression, a cell-cycle–related protein, as determined by immunohistochemistry, has been prospectively confirmed as a reliable prognostic marker, which allows the identification of high-risk patients (Ki-67 >30 percent) who may qualify for more-aggressive therapeutic approaches.^5,16 This marker is independent of clinical features including the Mantle Cell Lymphoma International Prognostic Index (MIPI) score (see Clinical Features and Risk Factors^20,21 below).

MCL typically presents in an older male patient with lymphadenopathy in several sites (e.g., cervical, axillary, inguinal). The patient may be asymptomatic but some experience fever, night sweats, or weight loss (Table 100–1).^5,17 The spleen is enlarged in 40 percent of patients. Marrow is involved with MCL in the vast majority of patients, and 50 percent of patients present with blood involvement, sometimes with an overt leukemic phase.

In 25 percent of cases, there is symptomatic gastrointestinal involvement, typically presenting as polyposis coli.¹⁸ Gastrointestinal symptoms may include abdominal pain and diarrhea, signs of small-bowel obstruction, or hematochezia. The intestinal polyps usually appear in the ileocecal region, and histopathologic analysis and immunocytochemistry are required to confirm the diagnosis of MCL involvement. Asymptomatic gastrointestinal involvement may be detected in up to 90 percent of cases; thus, endoscopy and histologic analysis of random biopsies are recommended, especially in the few cases with localized stage.

The frequency of CNS disease is low at first diagnosis but increases with subsequent relapses and correlates with elevated lactate dehydrogenase (LDH), blastoid cytology, and cell proliferation (Ki-67).¹⁹ There is no consensus on the need for CNS prophylaxis.

Prognostic parameters include the serum level of β₂-microglobulin and LDH, blastoid cytology, age, Ann Arbor stage, extranodal presentation, and constitutional symptoms, among others.⁵ Most of these prognostic variables were determined retrospectively and are based on doxorubicin-containing regimens only. Thus, a prognostic model, the MIPI, was established, which implements four independent prognostic factors: age, performance status, LDH, and leukocyte count (Fig. 100–2).²⁰ This score has been confirmed in numerous series, including in a prospective study that corroborated the reliability of this score for various chemotherapy-based regimens.²¹

Based on frequent marrow involvement, detection of minimal residual disease with patient-specific primers allows the detection of one malignant cell in 10⁵ to 10⁶ background cells. A number of reports have confirmed a correlation between molecular residual disease and clinical recurrence.^22,23 This fact has been used in trials evaluating the value of preemptive treatment with rituximab prior to emergence of clinical evidence of recurrence.

The clinical presentation of MCL may resemble CLL or other indolent nodal lymphomas. The immunophenotype of CLL is similar with coexpression of immunoglobulin (Ig) M and IgD, the B-cell–associated antigens CD19 and CD20, and aberrant expression of the T-cell antigen CD5. In contrast to MCL cells, however, CD23 is typically highly expressed in CLL. Like follicular lymphoma, MCL is positive for CD20 and BCL-2, but in contrast to follicular lymphoma, MCL is negative for CD10 and BCL-6. However, because these expression patterns vary, analysis of cyclin D1 overexpression or t(11;14) remains crucial to confirm or exclude the diagnosis of MCL.

In general, MCL is still considered incurable and most patients follow an aggressive clinical course. However, 10 to 15 percent of patients may exhibit a more indolent evolution and may not require therapy for several years. These cases are commonly characterized by a nonnodal leukemic presentation with only marrow involvement and splenomegaly.⁴ Sox-11 negativity may also identify some of these more indolent cases, although its role is controversial, as additional p53 mutations may cause an aggressive clinical evolution (see Fig. 100–1).²⁵ Thus, the reliable diagnosis of indolent cases is difficult, and a short, watchful, waiting period under close observation seems to be appropriate in cases with low tumor burden.²⁶

LOCALIZED STAGE

Localized disease with low tumor burden is rare. A small retrospective review has suggested long-term remissions after involved-field radiotherapy (30 to 36 Gray).²⁷ In contrast, in a randomized trial all patients relapsed within a year.²⁸ Thus, a shortened chemotherapy induction followed by radiation consolidation seems to be most appropriate.

ADVANCED STAGE

Conventional Chemotherapy

Because of the typical aggressive clinical presentation, anthracycline-containing chemotherapy has usually been applied to MCL in the past, although a small randomized trial did not confirm a major clinical benefit.²⁹ Complete remission rates are only 30 to 40 percent with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP), typically with a short duration of response of 10 to 12 months.^30,31

Combined Immunochemotherapy

Rituximab monotherapy achieves low response rates of approximately 25 percent, and should be used only in medically unfit patients who are not able to tolerate cytotoxic therapy (Fig. 100–3).³² On the other hand, addition of rituximab to chemotherapy has been shown to significantly improve complete response rates, overall response rates, and overall survival in a systematic meta-analysis, making immunochemotherapy the standard of care, both in first-line and relapsed settings for patients with advanced stage MCL (Table 100–2).^33–43 In patients with an overt leukemic phase and high lymphocyte counts of greater than 50 × 10⁹/L, the first dose of rituximab should be delivered with caution because of the risk of tumor lysis syndrome or cytokine-release syndrome.

In a randomized trial, rituximab with CHOP considerably improved response rates (94 percent vs. 75 percent), and time to treatment failure in comparison to CHOP chemotherapy alone.³⁴ In contrast, rituximab with cyclophosphamide, vincristine, and prednisone (R-CVP) resulted in significantly inferior response rates and progression-free survival (PFS), and cannot be recommended in patients with MCL.⁴⁴ Similarly, fludarabine combinations result in prolonged cytopenias, and resulted in overall survival rates significantly inferior (46 percent vs. 62 percent at 4 years) to those achieved with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) in a large prospective European trial.³⁹ In contrast, a bendamustine-based combination resulted in similar response rates (93 percent vs. 91 percent) and PFS in two randomized trials, but a more favorable toxicity profile, particularly with regard to alopecia and peripheral neuropathy, making the bendamustine and rituximab (BR) regimen the most commonly administered regimen currently, especially in elderly patients.^41,44 Finally, an intensified approach combining bendamustine with cytarabine resulted in impressive response rates (100 percent), but significant thrombocytopenia, suggesting that this regimen should be used only in young, fit patients.⁴²

In conclusion, BR and R-CHOP represent the current standard approaches in older patients who represent the majority of MCL patients (see Fig. 100–3). Based on clinical presentation, BR may be preferable, especially in patients with CLL-like presentation, whereas CHOP seems to be appropriate in the more-aggressive cases with elevated LDH. Especially in blastoid variants, one might consider cytarabine-containing regimens based on the improved results in younger patients.⁴⁵ However, such an individualized approach has never been tested in a prospective fashion.

Maintenance/Consolidation

The concept of regular rituximab maintenance, previously established in follicular lymphoma, has also been investigated in MCL.³⁹ After conventional R-CHOP induction, continuous antibody maintenance results in an impressive prolongation of both PFS (58 percent vs. 29 percent at 4 years) and overall survival (79 percent vs. 67 percent at 4 years) so that rituximab maintenance is now generally recommended.

Alternatively, radioimmunotherapy consolidation has been tested after a shortened R-CHOP induction. Again, survival rates seem to be prolonged with a median time to treatment failure of 31 months, but results seem to be inferior to ongoing rituximab maintenance, possibly because of the single-application scheme used with radioimmunotherapy.⁴⁰

Intensive Chemotherapy with and without Stem Cell Transplantation

In several studies, either intensified upfront therapy or the addition of high-dose consolidation followed by autologous stem cell transplantation (ASCT) resulted in impressive survival rates (Table 100–3).^45–56 A randomized trial has proven that ASCT prolongs PFS in comparison to CHOP-like induction therapy alone (3.3 vs. 1.4 years).⁴⁶ In a subsequent meta-analysis, overall survival was also significantly prolonged by ASCT, independent of the addition of rituximab.⁵⁷ “In vivo purging” with a rituximab-containing induction regimen was shown to further improve long-term survival. Various studies have investigated the potential benefit of cytarabine-containing induction regimens (Table 100–3). Most importantly, a randomized trial demonstrated that the median PFS was almost doubled by the administration of the R-CHOP/DHAP (dexamethasone, high-dose cytarabine, and cisplatinum) regimen (7.6 vs. 3.8 years) compared to administration of R-CHOP alone prior to ASCT.⁴⁵ In contrast, a phase II study suggested that high-dose methotrexate (MTX) adds significant organ toxicity, but similar long term remissions could be also achieved with considerably reduced doses.⁴⁷

A study-to-study comparison suggests a benefit of a total-body irradiation (TBI)–containing high-dose consolidation in patients with only partial remission after induction, whereas such benefit was not observed for the addition of conventionally dosed radioimmunotherapy.^58,59 An interim analysis has suggested a benefit from subsequent rituximab maintenance, but further followup is necessary to confirm the superiority of this approach.⁵³

Alternatively, upfront dose intensification may be applied. The rituximab plus hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone (R-hyperCVAD) regimen has achieved high complete response rates and long-term remissions in various trials.^54,55,56 However, this regimen is hampered by significant therapy-associated toxicity, including secondary malignancies, and should only be considered in young, fit patients.

RECURRENT AND REFRACTORY DISEASE

Salvage Chemotherapy

The inherent resistance of MCL to conventional doses of chemotherapy becomes especially apparent in relapsed disease. Conventional immunochemotherapy options, some of them highly effective in first line treatment, achieve only short term remissions in relapsed disease. (Table 100–4).^{42,60,61,62,63,64} Thus, consolidation with ASCT deserves consideration if not already employed in the frontline setting. Unfortunately, long-term results of this approach in recurrent/refractory MCL are rather sobering.

In younger patients responding to salvage therapy, it is recommended to discuss the option of potentially curative allogeneic transplantation, based on the observed graft-versus-lymphoma activity in MCL. Reduced intensity conditioning may be applicable also in patients older than age 60 years, but is hampered by delayed toxicities, including chronic graft-versus-host disease and 20 to 25 percent treatment-related mortality.^65,66,67,68

In elderly patients, one might consider rituximab maintenance if not administered previously. Alternatively, radioimmunotherapy consolidation has achieved long-term remissions in some patients, although efficacy as a single approach is limited with a median time to progression of only 5 months.^69,70

Targeted Approaches

Chemotherapy alone has only short-term activity in relapsed disease, and therefore a number of targeted approaches have been investigated, especially in relapsed MCL, both as single agents and in combination with chemotherapy (see Tables 100–4 and 100–5).^71–97

Proteasome Inhibitors Bortezomib, a first-generation proteosome inhibitor, is thought to exert its effect in part through alterations in the nuclear factor-κB (NF-κB). In relapsed disease, this single agent has shown response rates of 30 to 40 percent with a median PFS of approximately 6 months, which led to the first Federal Drug Administration approval of a targeted drug in relapsed MCL.^71,72,73 In subsequent trials, bortezomib was evaluated in combination with different chemotherapy regimens to further improve remission rates and avoid cumulative side effects (see Table 100–4).^{88,89,90,91,92,93} Based on the encouraging data, a first-line trial has been completed. Substituting bortezomib for vincristine (within R-CHOP) led to an almost doubling of PFS.⁹⁸ In such a combined approach, bortezomib should be administered on only days 1 and 4 of each cycle to avoid cumulative thrombocytopenia.

Immunomodulatory Drugs Based on its oral formulation and its favorable tolerability, lenalidomide represents an attractive option, especially in the context of continuous treatment. After long-term efficacy was observed in multiple myeloma, various studies confirmed its benefit in relapsed MCL as well, with a response rate of 35 to 50 percent.^75,76,77,78 In a randomized phase II trial, this approach was superior to monochemotherapy (response rate 40 percent vs. 11 percent).⁷⁸ Other studies have investigated combining lenalidomide with chemotherapy.⁹⁴

Mammalian Target of Rapamycin Inhibitors Temsirolimus has been registered by the European Union based on a randomized phase III trial proving its superiority compared to monochemotherapy in a highly refractory patient population (response rate, 22 percent vs. 2 percent).⁸¹ Combining temsirolimus with chemotherapy resulted in objective responses in all evaluable patients in a phase I trial.⁹⁶

B-Cell Receptor Pathway Small molecules targeting the B-cell receptor have achieved remarkable response rates in relapsed MCL. The Bruton tyrosine kinase (BTK)-inhibitor, ibrutinib, exhibited a response rate of 68 percent in relapsed disesase.⁸⁴ Combination with rituximab led to a response in all cases with low Ki-67, whereas in highly proliferating disease, this approach was only effective in half of the patients.⁸⁵ The compound is very well tolerated with only slight immunosuppression, bleeding, and atrial fibrillation being the predominant side effects. However, early relapses with a highly aggressive course have been observed. Combinations of ibrutinib with chemotherapy are being tested in the frontline setting.

Idelalisib achieves similar high response rates of 62 percent in MCL, although many of the remissions are short-lived.⁸⁶

The prospects for patients with MCL have significantly improved over the last few years based on advances in therapy, including the addition of rituximab and optimization of chemotherapy regimens, both according to tolerability (bendamustine) as well as efficacy (cytarabine, autologous transplantation; see Fig. 100–3). Accordingly, overall survival has been observed to improve from a median of 2.7 years in the 1980s to 5.8 years in the 1990s.¹⁰⁰

Improved diagnostic tools, which allow the detection of pathognomonic cyclin D1 overexpression, show that MCL represents a spectrum of disease (see Fig. 100–1). Accordingly, a few patients may be followed with watchful waiting, whereas others present with a highly proliferative disease insensitive to conventionally dosed chemotherapy. In the majority of cases, initial responses are followed by a continuous relapse pattern, although remission durations have significantly improved as a result of improved frontline therapeutic approaches. The implementation of targeted therapies with proven efficacy in relapsed disease represents the current challenge in treatment of MCL (see Fig. 100–3). With the possible exception of ibrutinib, combined approaches are required to achieve high response rates and prolonged remission durations. In addition, predictive markers for distinct targeted therapies are urgently needed for more individualized treatment strategies which will allow patients to achieve optimal outcomes.