Mantle cell lymphoma: The promise of new treatment options

Transcription

1 Critical Reviews in Oncology/Hematology 80 (2011) Mantle cell lymphoma: The promise of new treatment options Andre Goy a,, Brad Kahl b a Lymphoma Division, John Theurer Cancer Center, Hackensack Medical University Cancer, NJ Center, 20 Prospect Avenue, Hackensack, NJ 07601, USA b University of Wisconsin Carbone Cancer Center, Madison, WI, USA Accepted 15 September 2010 Contents 1. Introduction Search strategy and selection criteria Diagnosis Prognostic markers Treatment Previously untreated MCL Chemotherapy plus rituximab Dose intensive or high-dose chemotherapy approaches Unmet need for patients not eligible for high-dose chemotherapy Consolidation and maintenance therapy strategies Relapsed or refractory MCL Bortezomib new class of proteasome inhibitors New class of mtor inhibitors Flavopiridol new class of CDK inhibitors Lenalidomide new class of IMiDs compounds Vorinostat new class HDAC inhibitors Other inhibitors Conclusions Reviewers Conflict of interest Acknowledgments References Biographies Abstract Though the expected overall survival (OS) for mantle cell lymphoma (MCL) has doubled in the last 30 years it is still in the range of only 4 5 years. Despite high response rates with current first-line treatments, most patients eventually relapse and become typically chemoresistant, leading to very poor outcome in the relapsed setting. Here, we summarize the clinical characteristics of MCL and frontline strategies used in MCL, and review a number of novel options that are currently being investigated in an effort to extend survival outcomes for this difficultto-treat patient population. Among these novel options figure cytotoxics (bendamustine, cladribine), new biologicals/small molecules such as proteasome inhibitors (bortezomib 1st drug approved in the USA for MCL), mtor inhibitors with temsirolimus (1st drug approved in EU for MCL), CDK inhibitors (flavopiridol); IMiDs (thalidomide, lenalidomide); HDAC inhibitors, Bcl-2 inhibitors and second or third Corresponding author. Tel.: ; fax: address: agoy@humed.com (A. Goy) /$ see front matter 2010 Published by Elsevier Ireland Ltd. doi: /j.critrevonc

2 70 A. Goy, B. Kahl / Critical Reviews in Oncology/Hematology 80 (2011) generation monoclonal antibodies or immunotoxins. The panel of novel drugs approved or being tested offers new opportunities in the management of MCL from combination in the frontline setting (e.g. bortezomib-r-chemo) to post-induction strategies such as consolidation (e.g. radioimmunotherapy, bortezomib) or maintenance therapy (e.g. rituximab, lenalidomide) Published by Elsevier Ireland Ltd. Keywords: Mantle cell lymphoma; Non-Hodgkin s lymphoma; Rituximab; Bortezomib; Flavopiridol; Bendamustine; Temsirolimus; Lenalidomide 1. Introduction Mantle cell lymphoma (MCL) is a lymphoma subtype that accounts for 5 7% of non-hodgkin s lymphomas (NHL). Although the expected overall survival (OS) for MCL has doubled over the last 3 decades [1], it currently remains in the range of only 4 5 years. Though MCL patients respond to initial therapy, a continuous pattern of relapse follows with common subsequent chemoresistance and very poor outcome. With the exception of rare patients who enjoy long term disease-free survival after non-myeloablative allogeneic stem cell transplantation (SCT), there is no curative therapy defined for MCL, which carries one of the worst prognoses among B-cell lymphomas [2,3], especially in the relapse setting. Modern treatment options consisting of anthracyclinebased, rituximab-containing chemotherapy combinations and high dose chemotherapy (HDT)/myeloablative consolidation with autologous stem cell transplantation (ASCT) regimens have improved overall response rates (ORR), complete response (CR) rates, progression-free survival (PFS) as well as time to treatment failure (TTF) [4,5]. When added to first-line, anthracycline-based chemotherapy, rituximab improved ORR and TTF, but provided no benefit in either PFS or overall survival (OS). Only about onequarter of patients remained progression-free after 2 years following R-CHOP ([rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone]) [6]. Although chemotherapy regimens containing rituximab followed by ASCT, as well as dose-intense regimens (R-HyperCVAD) ([rituximab plus hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone]), have improved clinical outcomes (extending PFS) relative to standard chemotherapy, patients continue to relapse and whether such approaches have a beneficial impact on OS is still debated [7,8]. Following relapse, median life expectancy for a patient with MCL declines to 1 2 years [9], and there are few effective treatment options available. Bortezomib and temsirolimus are currently the only drugs approved by the United States Food and Drug Administration (USFDA) and European Medicines Agency (EMEA), respectively, for the treatment of relapsed or refractory MCL. This paper will briefly summarize the clinical characteristics of MCL and review the available data on new treatment options as well as strategies that are currently being explored in an effort to extend survival outcomes for this difficult-to-treat patient population. 2. Search strategy and selection criteria Data for this review were identified by searches of EMBASE, PubMed, and references from relevant articles using the search term mantle cell lymphoma. Abstracts and reports from meetings were included only when they related directly to previously published work. Only papers published in English between 1995 and 2009 were included. 3. Diagnosis Patients with MCL usually present with advanced disease [3,9], though less than one-third have B symptoms at presentation. Typical clinical characteristics of MCL patients include a median age at diagnosis of >60 years, a male predominance, advanced disease ( 70% Ann Arbor stage IV), and extranodal involvement, including bone marrow, blood, spleen, liver, and gastrointestinal tract [10 12,3,9]. Gastrointestinal tract involvement can vary from regular polyps to extensive polyposis coli or normal mucosa, but random biopsies have been shown to be positive in approximately 90% of patients at baseline (hence not indicated as part of initial work but to be considered for restaging on therapy) [13]. Histologically, MCL is characterized by neoplastic expansion of the mantle zone surrounding lymph node germinal centers (antigen naïve cell) and a homogeneous population of small- to medium-sized lymphocytes with irregular nuclei, condensed chromatin, inconspicuous nucleoli, and scant cytoplasm (Fig. 1) [10 12,14,15]. The histological pattern may be diffuse, nodular, or a combination of the two mixed histologies, while the mantle-zone variant is rare and more indolent [11]. At the cytological level, variants include classic small cells subtype, lymphoblastoid and pleiomorphic subtypes with clearly some overlap, illustrating the fact that MCL may encompass a spectrum of diseases [16], which is also suggested by molecular profiling studies [16]. The classical immunophenotype of MCL reflects a mature B cell (CD19+, CD20+, CD22+, CD79a+) and is usually CD5+, CD43+, while negative for CD10, CD23 and Bcl- 6 [10 12,17]. In some rare cases, however, MCL may be CD5 or CD23+ [17]. The t(11;14)(q13;q32) translocation is the hallmark cytogenetic abnormality of MCL. This translocation places the cyclin D1 gene, which regulates progression through the G1 checkpoint of the cell cycle, under the transcriptional control of the immunoglobulin (Ig) heavy chain gene enhancer region on chromosome 14. This results in a significant overexpression of cyclin D1, which is not nor-

3 A. Goy, B. Kahl / Critical Reviews in Oncology/Hematology 80 (2011) Fig. 1. Morphology and cytogenetic features of MCL. (A) Normal morphological structure of the spleen. The germinal center, mantle, and marginal zones can be seen. Reprinted by permission of Macmillan Publishers Ltd: Mateo et al. [14]. (B) Classical MCL is characterized by a proliferation of small- to medium-sized lymphocytes with irregular nuclei and inconspicuous nucleoli. Reprinted by permission of Macmillan Publishers Ltd: Jares et al. [15]. (C and D) Examples of interphase fluorescence in situ hybridization (FISH) patterns in normal nuclei (C; 2 reds, 2 green signals) and in t(11;14) + (D; 1 red, 1 green and 2 yellow fusion signals; lymph node section, 1000 ). Reprinted by permission of Macmillan Publishers Ltd: Belaud-Rotureau et al. [119]. mally expressed on resting cells of the lymphoid lineage, thus promoting cellular proliferation [9]. In the absence of the classical nuclear staining of cyclin D1 (sometimes technically difficult), the diagnosis can be confirmed by fluorescence in situ hybridization (FISH) for t(11;14) translocation (Fig. 1) [17]. A small subset of truly cyclin D1-negative MCL (5 10%) will show the expression of cyclin D2 or D3, in some cases through alternative translocations with other Ig loci [18]. These rare cases of cyclin D1-negative MCL show otherwise similar characteristics and outcomes to cyclin D1-positive MCL. 4. Prognostic markers Features frequently associated with favorable prognosis in patients with MCL include younger age (<65 years), normal levels of serum lactate dehydrogenase (LDH), and low levels of 2 -microglobulin. On the other hand, advanced-stage disease, occurrence of B symptoms, poor Eastern Cooperative Oncology Group (ECOG) performance status, and high mitotic index Ki-67 (MIB-1 >30%) are often associated with poor prognosis [10,19]. Other factors associated with shorter survival rates include blastic morphology, high white blood cell count, bulky disease, and liver and bone marrow involvement, as well as elevated levels of beta-2 microglobulin [10,11]. Data on minimal residual disease (MRD) initially showed that in the setting of R-CHOP (rituximab plus CHOP therapy), achievement of a molecular CR did not impact PFS [20]. However, more recent data from Europe (with much larger series) suggest that an MRD negative status following HDT-ASCT or even conventional chemotherapy (FCR vs R-CHOP in patients aged 60y; ongoing phase III study in EU) may be its own important prognostic factor [21,22]. A new combined clinical and biological score, the MCL International Prognostic Index (MIPI), has been proposed based on data from >450 MCL patients who received uniform treatment in prospective clinical trials [23]. The MIPI incorporates age, ECOG performance status, LDH level, and leukocyte count into one calculated index, which provides a consistent and reliable estimation of each patient s clinical course. A consistent use of such an index also helps when comparing outcomes based on patient populations between clinical studies (though 2/4 factors PS and age might be difficult to implement for new strategies using high dose therapy). It should be noted that the MIPI, which is the first prognostic index in MCL, was established based on a population of patients treated with different regimens, only approximately one-third of whom had received rituximab. However, in a retrospective analysis by Smith et al., this index has now also been validated in the rituximab era. MIPI maintained its prognostic capacity, and the use of rituximab containing regimens was not associated with improved OS at any time compared to none [24]. In contrast, the MIPI index still requires a prospective validation in respect to combination therapy with R-hyperCVAD; as 2 trials have shown that the MIPI risk model loses its significance when associated with the OS and time to treatment failure [25,26] similarly to our experience [27]. Other prognostic factors include deletions or mutations in the p53 gene [28], proliferation signature [29], other cytogenetic abnormalities (comparative genomic hybridization (CGH)), single-nucleotide polymorphisms (SNPs), cyclin D1 transcript length (secondary 3 UTR deletion which leads to shorter transcript with longer half-life) [30], monoallelic deletions of p27 and p16, micro RNAs (e.g. mir16), methylation pattern [31], as well as the inactivation of RB1 or SOX11 [32], while ZAP-70 and somatic mutations status are not as clearly prognostic as in chronic lymphocytic leukemia (CLL) [33]. Clinically a small subset of MCL behaves more indolently and presents more like CLL with frequent nonnodal presentation (high white count, splenomegaly and less lymphadenopathy), they carry more often somatic mutations and are SOX11 negative by opposition to classic MCL [34].

4 72 A. Goy, B. Kahl / Critical Reviews in Oncology/Hematology 80 (2011) Table 1 Outcomes following chemotherapy, with or without rituximab, in patients with newly diagnosed MCL. Treatment Reference n ORR (%) CR (%) Median OS Median PFS Median response duration (months) CHOP Nickenig et al. [35] months NR 21 Lenz et al. [6] months NR NR HyperCVAD Romaguera et al. [13] NR 15 months a NR CHOP + DHAP Lefrère et al. [36,8] months 51 months a,b NR CHOP + R Howard et al. [20] c NR 16.6 months NR Lenz et al. [6]; Hoster % at 5 years NR 29 et al. [37] HyperCVAD + R Romaguera et al. [38]; c 65% at 5 years 48% at 5 years a NR Fayad et al. [39] Kahl et al. [40] months 37 months NR DHAP + R De Guibert et al. [41] c 69% at 3 years 65% at 3 years b NR Delarue et al. [42] Not reached >67 months NR NR CR, complete response; NR, not reported; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; R, rituximab. a Failure-free survival. b Event-free survival. c Includes complete response unconfirmed. 5. Treatment 5.1. Previously untreated MCL Chemotherapy plus rituximab Various chemotherapy regimens have been studied in patients with MCL and have been shown to produce relatively high response rates in previously untreated patients (Table 1) [13,20,35 42,6,8]. In a randomized trial conducted by the German Low Grade Lymphoma Study Group, patients with advanced MCL and follicular lymphoma were treated with mitoxantrone, chlorambucil, and prednisone (MCP) or CHOP [35]. A subset analysis of the patients with MCL indicated that there was no statistical difference between the regimens with regard to response rate, duration of response, TTF, or OS. There is some evidence to suggest that sequential chemotherapy potentially increases complete remission (CR) rates by further reducing the disease burden through intensified consolidation regimens that use high doses of cytarabine. Treatment of MCL patients in a small trial with 4 cycles of CHOP followed by cisplatin, high dose cytarabine, and dexamethasone (DHAP) resulted in an 84% CR rate in patients who completed the treatment regimen [36,8]. Although the addition of rituximab to anthracycline-based chemotherapy has been shown to improve ORR and TTF in the first-line setting, these have not translated into significant improvement in PFS and OS. In an initial phase II study of 40 previously untreated patients with MCL, 36% of patients achieved molecular remission [20]. However, at 16.6 months, the median PFS remained relatively short, and molecular remission (from blood and/or BM) did not appear to significantly affect the PFS compared to patients without molecular remissions (16.5 vs months). In a randomized trial comparing CHOP with R-CHOP, the addition of rituximab to first-line anthracycline-based chemotherapy improved both ORR and CR rates [6] (Table 1). In each treatment group, the 2-year PFS and OS rate was only 25% and 77%, respectively. Adverse events were similar with CHOP and R-CHOP, with the exception of infusion reactions related to the first infusion (none with CHOP vs. 7% with R- CHOP), and grade 3 or 4 granulocytopenia (53% vs. 63%, respectively) Dose intensive or high-dose chemotherapy approaches Treatment with fractionated cyclophosphamide, doxorubicin, vincristine, and dexamethasone (hypercvad), alternating every 3 weeks with high doses of methotrexate and cytarabine, resulted in a median failure-free survival (FFS) of 15 months, which indicated no significant improvement relative to historical data [13]. High response rates were also achieved when rituximab was added to the hypercvad regimen (R-hyperCVAD) alternating with rituximab plus methotrexate and cytarabine (Table 1). After a median follow-up of 7 years, patients who were 65 years of age achieved a FFS and OS of 52% and 68%, respectively [43]. However, improved efficacy came at the expense of increased toxicity. The main adverse event was myelosuppression, and 5% of patients died from acute toxicity. Moreover, 6 patients developed secondary hematological malignancies including myelodysplastic syndromes (MDS) (n = 5) and acute myelogenous leukemia (AML) (n = 1). Thus, while the results represent some of the most favorable outcomes achieved in the first-line setting, the significant toxicity of this regimen suggests that it may not be suitable for all patients, particularly those with advanced age (median age at diagnosis mid 60s).

5 A. Goy, B. Kahl / Critical Reviews in Oncology/Hematology 80 (2011) A pilot study by Kahl et al. [40] investigated rituximab and a modified R-hyperCVAD regimen in 22 patients with previously untreated MCL. The modification consisted of no methotrexate, no cytarabine, no vincristine administered on day 11 nor steroids on days Doxorubicin was administered by intravenous continuous infusion and cycles were repeated every 28 days. In an effort to improve response duration, the initial regimen was followed by rituximab maintenance (4 weekly doses rituximab 375 mg/m 2 every 6 months for 2 years). The main adverse events were myelosuppression, with rates of grade 3 or 4 neutropenia 57%, thrombocytopenia 23%, and anemia 12%. Based on these results (Table 1), the National Comprehensive Cancer Network (NCCN) guidelines were edited to note that modified hypercvad plus rituximab maintenance is an option in firstline therapy of MCL patients who are over 65 years of age [17]. This also served as a useful backbone to combine with bortezomib; as shown thorough a small study initially, and a recently completed larger study with VcR-HyperCVAD (ECOG1405) [44,45]. In a retrospective analysis of the NCCN NHL Outcomes Database, the efficacy and safety profile of R-CHOP alone, R-hyperCVAD, R-CHOP followed by HDT and ASCT was investigated as a first-line treatment approach in patients with MCL [46]. At a median follow-up of 30 months, no significant differences in OS were observed between the 3 groups. However, R-CHOP without consolidation had a significantly poorer PFS than both R-hyperCVAD and R- CHOP followed by HDT/ASCT. No significant difference in PFS was observed between R-hyperCVAD and R-CHOP plus HDT/ASCT. A recently published study by Feldman and colleagues confirm these findings [47]. The Nordic Lymphoma Group evaluated the use of intensive first-line immunochemotherapy and rituximab in vivo purged ASCT in 160 previously untreated patients aged 65 years [48]. Induction immunochemotherapy consisted of augmented CHOP ( maxi-chop ) therapy plus rituximab alternating with high-dose cytarabine (6 cycles) followed by carmustine, etoposide, cytarabine, and melphalan (BEAM) or carmustine, etoposide, cytarabine, and cyclophosphamide (BEAC) (see Table 1 in Geisler et al. [48] for full dosing description). Almost 40% of patients discontinued the study, mostly due to non-response, relapse, or progression (30%) or non-relapse events (8%). Of the 160 MCL patients enrolled on study, 74 patients achieved a CR and 4 patients had progressive disease following ASCT [49]. Patients in clinical response, without signs of clinical relapse, who converted to a polymerase chain reaction (PCR)-positive result for PCR-detectable MRD (quantitative PCR assay for Bcl- 1/Immunoglobulin heavy chain gene (IgH) and clonal IgH rearrangements) in bone marrow or blood, were offered preemptive therapy with rituximab 375 mg/m 2 weekly for 4 weeks, to prevent clinical relapse [49]. Conversion to PCRdetectable MRD occurred in some patients 2 5 years post transplantation, and it should be noted that these patients were not considered failures in this study [48]. Preemptive treatment was administered to 26/36 patients, that underwent a molecular relapse up to 6 years (mean 18.5 months) after ASCT, leading to reinduction of molecular remission in 92% [49]. Out of the 26 patients that underwent preemptive therapy, 62% remain in CR while 38% have relapsed (38%), suggesting that PCR-based preemptive treatment using rituximab is feasible, may reinduce molecular remission, and may prevent clinical relapse. Ultimately, the 6-year OS, EFS, and PFS rates were 70%, 56%, and 66%, respectively, and no clinical relapses had occurred within 5 years [48,49]; though a recent update suggest late relapse beyond 6 years, illustrating the need for improvement and potential other maintenance strategies post induction Unmet need for patients not eligible for high-dose chemotherapy Although recent studies have demonstrated promising outcomes with aggressive induction therapies and dose-intense chemotherapies, such approaches may not be suitable for all patients [50]. Intensive treatment strategies tend to be poorly tolerated by patients who are elderly or have comorbidities [51]. Furthermore, there is some evidence that intensive treatment strategies may not provide better outcomes than standard therapies. A retrospective analysis was conducted of 181 patients with MCL, of whom 111 had a date of diagnosis and 75 had adequate information on therapy. Seventy percent of patients had received CHOP-like or R-CHOP-like treatment and only 5% received hypercvad and/or ASCT as part of first-line therapy [51]. Three years from diagnosis, the OS was 86% and the median OS was 7.1 years, which compares favorably to outcomes achieved with more aggressive therapies such as hypercvad. Perhaps surprisingly, promising outcomes have also been reported for a strategy in which initial therapy is deferred, or a watch and wait approach. A retrospective analysis investigated outcomes from 97 patients, 31 (32%) of whom were observed for >3 months before initiating treatment (early treatment group) [52]. Patients in the watch and wait category tended to have a better performance status and a lower-risk International Prognostic Index (IPI) index score. Despite these differences, it was interesting to note that the median OS for the watch and wait group was not reached, compared with a median OS of 64 months in patients who were initiated on early therapy (p = 0.004) [52]. These provocative data certainly are consistent with the diversity of clinical presentation of MCL, with younger and potentially less symptomatic patients in the observation group. Of notice also, the cut-off for watch and wait was from 3 months up to 5 years and 2/3 patients were in the 0 3 month s group. Although these results may be skewed by a few patients with positive outcomes, these results do indicate that much like with other histological subtypes of NHL, MCL itself is quite heterogeneous. Given the aggressive clinical course and the poor survival outcomes attributed to most patients with MCL, it is clear that further research is required to better char-

6 74 A. Goy, B. Kahl / Critical Reviews in Oncology/Hematology 80 (2011) acterize individual patients in an effort to determine optimal treatment strategies. While new treatment strategies such as consolidation and maintenance are also being explored for this difficult to treat patient population, an impressive number of promising novel agents are being developed in MCL including proteasome inhibitors, mtor inhibitors and CDK inhibitors, as well as new biologicals and monoclonal antibodies, immunotoxins and immunomodulatory agents, both alone and as part of novel combination strategies that target multiple pathogenic mechanisms Bendamustine. Bendamustine is a unique, cellcycle nonspecific, cytostatic chemotherapeutic agent which contains an alkylating group (mechlorethamine group/nitrogen mustard), a benzimidazole ring (thought to exhibit antipurine-like properties). Bendamustine has shown impressive activity in indolent lymphomas, and a few of these studies have included patients with MCL. Three phase II studies [53 55] evaluated bendamustine in combination with rituximab with or without mitoxantrone in patients with relapsed or refractory indolent NHL or MCL. In patients with MCL, these studies reported ORR ranging from 75 to 92%, CR rates ranging from 33 to 50%, and a median PFS of months. The most common adverse event was reversible myelosuppression, including grade 3 or 4 leukopenia (16 78%) and neutropenia (36 46%). These data need to be confirmed in larger studies, however, as the number of MCL patients enrolled was small (12 18 patients). Recently, in a study by Rummel et al. the role of bendamustine and rituximab first-line combination therapy was investigated in 260 patients with follicular, indolent and MCL [56]. In this study, an ORR for all patients was reported of 93.8%, with a 40.1% CR rate and a PFS of 54.8 months. Hematological adverse events included grade 3 or 4 neutropenia (10.7%) and leukopenia (12.1%) [56]. The results seen with bendamustine, offer a new backbone for MCL that is being looked at in a variety of combinations with mitoxantrone (BMR) or bortezomib and rituximab (BBR) [57 60] Cladribine. The purine analog cladribine (2-CdA) is of the same family of drugs as pentostatin and fludarabine, and has shown high activity in patients with indolent lymphoid malignancies [61,62]. Following emergence of the anti-cd20 monoclonal antibody rituximab, several trials were initiated to assess the safety profile and clinical benefit of rituximab and 2-CdA combination therapy in patients with lymphoid malignancies, some trials including patients with MCL. Based on some success in the treatment of other B- cell malignancies such as B-CLL, Robak et al. evaluated rituximab and 2-CdA combination therapy (RC) with or without the addition of cyclophosphamide (RCC) in 54 patients, including 9 patients with MCL [63]. For the 9 MCL patients, an ORR of 67% was observed (RC 44% vs RCC 22%), with 11% CR in both groups. The most common AEs for the total population were hypersensitivity to rituximab (35%), resulting in fever, chills, rush, and hypotonia. Grade 3 and 4 hematological adverse events included neutropenia (11%), thrombocytopenia (7%), and anemia (7%). These results compare favorably with studies investigating fludarabine monotherapy in MCL patients, resulting in the achievement of an ORR of 33 41% [64,65]. Such ORR were improved with the addition of cyclophosphamide to therapy, and resulted in a 100% ORR with a 70% CR rate [66]. Myelosuppression and infection were the main AEs associated with fludarabine treatment [66], as well as the risk for future stem cells collection in that setting. More recently, in a study by Inwards et al. the role 2-CdA alone or in combination with rituximab was investigated in 80 patients with MCL [67]. In previously untreated patients (n = 26, median age 68 years), 2-CdA monotherapy resulted in an ORR of 81% with a 42% CR rate. The median PFS for these patients was 13.6 months, and 81% of patients remained alive at 2 years. For patients with relapsed disease (n = 25, median age 68 years), an ORR of 46% was achieved with a 21% CR rate. Median PFS was 5.4 months and 36% remain alive at 2 years. The addition of rituximab to therapy resulted in an ORR of 66% for previously untreated patients (n = 29, median age 70 years), with a 52% CR rate. The PFS was 12.1 months and 78% of patients remained alive at 2 years. Grade 3 or 4 hematological adverse events were most common for the three treatment groups. Thrombocytopenia and neutropenia occurred in 17% and 50%, respectively, of previously untreated patients receiving single agent 2-CdA. Adverse events for patients with relapsed MCL treated with 2-CdA included thrombocytopenia (8%) and neutropenia (53%). The addition of rituximab to therapy for previously untreated patient was associated with thrombocytopenia (17%) and neutropenia (31%) [67]. Ongoing new studies are looking at 2-CdA combination with rituximab and temsirolimus or clofarabine plus rituximab Consolidation and maintenance therapy strategies In general, significant strides have been achieved with MCL therapy in terms of improving response rates, CR rate and PFS. However, these increased rates of response have not necessarily translated into significant improvement in OS. In an effort to prolong the duration of response achieved with first-line therapy, several post-induction approaches, such as maintenance or consolidation therapy, have been evaluated after initial chemotherapy. In the previously mentioned study by Kahl and colleagues [40], the use of rituximab maintenance may have contributed to the observed prolongation of PFS to a median of 37 months. Also, use of intensifying consolidation approaches, to deepen responses to a level of MRD negativity, may also translate into improved PFS and duration of response. Several studies have shown that consolidation with HDT and ASCT can improve outcomes in eligible patients who respond to first-line induction therapy [48,68,69,7,70 72]

7 A. Goy, B. Kahl / Critical Reviews in Oncology/Hematology 80 (2011) Table 2 Outcomes following consolidation therapy with ASCT for newly diagnosed MCL. Treatment Reference n ORR (%) CR (%) Median OS Median PFS ASCT after CHOP-like induction therapy Dreyling et al. [68] % at 3 years a 39 months R-in vivo purged ASCT after maxi-chop Geisler et al. [48] b 70% at 6 years 66% at 6 years R-HDS with R-in vivo purged ASCT after Gianni et al. [72] % at 4.5 years 79% at 4.5 years a standard chemotherapy ASCT after HDT Vigouroux et al. [7] months 37 months ASCT after hyper-cvad Khouri et al. [69] % at 5 years c 43% at 5 years d ASCT after hypercvad (±R) Vose et al. [70] 32 NR NR 97% at 3 years 78% at 3 years ASCT after R-hyperCVAD Ritchie et al. [71] % at 3 years 92% at 3 years CR, complete response; HDS, high-dose sequential chemotherapy; NR, not reported; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; R, rituximab. a Event-free survival. b Includes complete response unconfirmed. c Estimated value. d Disease-free survival. (Table 2). The type of induction prior to ASCT is still debated though data suggest high dose cytarabine is critical while the use of R-hyperCVAD as induction prior to ACST suggests better outcome with the caveats of small retrospective phase II studies with potential selection bias [73,74]. In a trial with MCL patients who achieved a CR or partial response (PR) using CHOP-like induction therapy, further treatment consisted of ASCT or maintenance therapy with interferon-alpha (IFN ) [68]. In this study, the ASCT arm showed a significantly longer PFS (median, 39 months vs. 17 months; p = ), which translated into a significantly longer TTF (median, 29 vs. 15 months; p = ). However, the estimated 3-year OS did not differ significantly between groups (83% with ASCT vs. 77% with IFN ). An early small study involving 16 MCL patients, of whom only 3 had received prior rituximab, showed that rituximab plus thalidomide induction followed by thalidomide maintenance achieved marked anti-tumor activity, with an ORR of 81%, including 31% CR [75]. A subsequent single-arm phase II study was conducted in 22 subjects with newly diagnosed MCL to assess the utility of thalidomide maintenance therapy after a novel, intensive, and very complex induction regimen [76]. The R-MACLO-IVAM-T regimen consisted of R-MACLO (rituximab, doxorubicin, cyclophosphamide, vincristine, methotrexate) in cycles 1 and 3, and R-IVAM (rituximab, cytarabine, etoposide, ifosfamide, mesna) in cycles 2 and 4. Patients achieving CR at the end of therapy received thalidomide maintenance (200 mg/day) until relapse or intolerable toxicity. In 21 evaluable patients, the ORR was 100%, including 95% CR, and estimated 2-year PFS was 94%. Although median followup is short (15 months), median PFS and OS had not been reached at time of reporting. Most common grade 3 4 toxicities were neutropenia, thrombocytopenia, and anemia, which were observed in 33%, 19%, and 17% of R-MACLO cycles and in 50%, 88%, and 68% of R-IVAM cycles, respectively. The thalidomide maintenance dose was reduced or stopped in 8 patients due to grade 3 4 peripheral neuropathy and was reduced in 6 patients due to grade 3 4 neutropenia. In the previously described Nordic Lymphoma group trial, which evaluated the use of intensive first-line immunochemotherapy and rituximab-in vivo-purged ASCT, responders who did not achieve a CR pre-transplant had significantly poorer outcomes [48]. Consequently, therapy was intensified in the MCL-3 study by adding 90 Y-ibritumomab to the high-dose BEAC/BEAM to responders not in CR [77]. The response rates in this trial were 50% CR, 18% complete response unconfirmed (CRu), and 28% PR. There was no indication of any added toxicity for 90 Y-ibritumomab tiuxetan-treated patients in CRu or PR. Molecular remission was achieved in 36% of patients with a positive PET scan pre-transplant compared to 92% of patients with a negative PET scan (p < 0.001). Lastly, a retrospective analysis of registry data on 195 MCL patients who underwent consolidation with ASCT revealed a median survival of 59 months, which was longer than the approximately 36 months achieved with conventional chemotherapy in historical series of patients [78]. Although with a median age of 50 years, the patients in this study were very young, the data suggest that consolidation therapy might improve outcomes in MCL. These data also show that patients do benefit from post-induction approaches that continue treatment, as they can help prevent relapse and deepen the responses achieved with induction. These data overall show that ASCT in MCL patients in first remission may positively influence survival outcomes compared with conventional therapy alone, although data from large randomized trials in the rituximab era are still lacking. In a study spanning 17 years, the longterm outcome of a risk-adapted transplantation strategy for MCL was analyzed in 121 patients from a single institution, who were enrolled in sequential transplantation protocols [79]. Rituximab-containing autologous transplantation in first remission resulted in long-term disease control, with only 1 relapse occurring in 11 patients followed between 2 and 8 years. In contrast, 29 patients who received ASCT without rituximab experienced relapse in a continuous fashion. This trend of improved long-term disease control with rituximab was not observed in patients who received ASCT in second

8 76 A. Goy, B. Kahl / Critical Reviews in Oncology/Hematology 80 (2011) Table 3 Response to bortezomib and conventional combination therapies in patients with relapsed or refractory MCL. Treatment Reference n ORR (%) CR (%) Median OS (months) Median PFS Median response duration (months) Bortezomib Fisher et al. [85]; Goyet a 23.5 NR 9.2 al. [86] Goy et al. [87] NR 42% at 6 NR months b O Connor et al. [88] a NR NR NR Fludarabine + cyclophosphamide Cohen et al. [66] 6; 14 c 83; 29 c 0; 14 c 17.5; 15.9 c 4.8 months; NR 2.5 months c,d R-FCM Forstpointner et al. [89] months NR R maintenance after FCM or R-FCM Forstpointner et al. [83] NR NR 14 EPOCH Wilder et al. [90] a months e NR R-EPOCH Jermann et al. [91] NR 15 months f NR Gemcitabine + dexamethasone ± cisplatin Morschhauser et al. [92] NR 8.5 months NR HyperCVAD + R Romaguera et al. [93] a NR 18 months d NR CR, complete response; EPOCH, etoposide, doxorubicin, and vincristine with cyclophosphamide and prednisone; FCM: fludarabine, cyclophosphamide, and mitoxantrone; NR, not reported; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; R, rituximab. a Includes complete response unconfirmed. b Estimated value. c Patients with 1 prior therapy and patients with multiple prior therapies, respectively. d Failure-free survival. e Event-free survival. f Patients treated with 175 mg 3 weekly followed by 75 mg or 25 mg weekly, respectively. or later remission [79]. Additionally, in the recently published CALGB study, 78 patients aged up to 69 years with newly diagnosed MCL were treated with rituximabcontaining high-dose chemotherapy followed by ASCT and rituximab to eliminate remaining lymphoma cells [80]. With a median follow-up of 4.7 years, the 5-year PFS was 56% (range 43 68%). The 5-year OS was 64% (range 50 75%). These data suggest that the incorporation of rituximab with aggressive chemotherapy and ASCT may be beneficial for patients with newly diagnosed MCL. The NCCN recommends consolidation with ASCT in eligible patients who achieve initial remission [17]. Overall, dose-intense or high-dose chemoimmunotherapy clearly prolongs PFS but patients still relapse over time, and the optimal treatment of patients who are ineligible for ASCT due to their physical condition, age, or lack of initial remission remains undefined Relapsed or refractory MCL Although the treatment of patients with relapsed or refractory MCL requires the use of aggressive therapies, there is currently no consensus regarding a standardized approach for treating this population. Two major issues associated with the treatment of relapsed or refractory MCL are that (1) response durations are even shorter with subsequent lines of therapy than with initial rituximab-containing chemotherapy [53,81 85], and (2) high-dose chemotherapy regimens are associated with significant toxicity and also shorter remissions than in the front-line setting [79]. Chemoimmunotherapy regimens in the relapsed/refractory setting have improved ORR, but CR rates remain low, and the best response durations are in the range of 8 9 months (Table 3) [66,83,85 93]. Currently, the only treatment for patients with relapsed or refractory MCL that has been approved by the USFDA is the proteasome inhibitor bortezomib. Its specific indication is for the treatment of patients with MCL who have received at least 1 prior therapy [94]. In Europe, only the mtor (mammalian target of rapamycin) inhibitor temsirolimus has been approved by the EMEA for the treatment of relapsed or refractory MCL patients. However, future treatments are promising as new investigational therapies are being investigated. These new biologicals include amongst others the CDK inhibitor flavopiridol and the immunomodulatory agent lenalidomide. With respect to treatment regimens, these agents have shown promising results as single agent in the relapse or refractory setting, and combination therapies with chemotherapy regimens and/or other biologicals (e.g. rituximab) are being investigated to improve upon clinical outcomes Bortezomib new class of proteasome inhibitors Regulatory approval of bortezomib was based on the results of the phase II multicenter PINNACLE study [85,95] and additional supporting data from 2 smaller single- and multicenter phase II studies [87,88] (Table 3). In the smaller trials, which evaluated the safety and efficacy of bortezomib in 10 and 29 patients, respectively, the ORR was 50% and 41%, respectively, and the CR rate was 10% and 21%, respectively. The PINNACLE study included 155 patients with MCL who had received 1 3 prior regimens [85,95]. Bortezomib treatment resulted in an ORR of 33% (8% CR or CRu), a median response duration of 9.2 months, a median time

9 A. Goy, B. Kahl / Critical Reviews in Oncology/Hematology 80 (2011) to progression (TTP) of 6.7 months, and a median OS of 23.5 months [85,95]. For patients who achieved CR or CRu, the median response duration and TTP were not reached after a median follow-up of 26.4 months, and the median OS was 36.0 months [95]. The most common grade 3 or higher adverse events were peripheral neuropathy, fatigue, and thrombocytopenia. Treatment-associated adverse events caused treatment discontinuation in 26% of patients [85]. Similar activity was reported in 2 other phase II studies from Canada and the UK [96,97]; with remarkably the same ORR as single agent in untreated vs. relapsed/refractory MCL in the NCIC study. Early phase I/II clinical trials investigated the combination of bortezomib with other agents based on extensive preclinical rationale, which suggest additive or synergistic effects with a variety of compounds from cytotoxics to biologicals. A phase I dose-escalation study of bortezomib in combination with fludarabine and rituximab in 23 patients, including 2 with MCL, reported that the inclusion of rituximab may worsen the toxicity (one grade 3 neuropathy) [98]. Gerecitano and colleagues [99] found that weeklyand twice-weekly schedules of bortezomib in combination with cyclophosphamide, prednisone, and rituximab (CBoRP) were well tolerated in 39 patients with relapsed/refractory lymphoma, including 9 patients with MCL. In another small study of 8 heavily pretreated patients who received bortezomib, high-dose cytarabine, dexamethasone, and rituximab, the ORR was 50% and CR 25%. Patients achieved a median PFS of 5 months, and a median OS of 15.5 months [100]. However, all patients in this study experienced grade 3 or 4 hematological toxicity. Based on preclinical murine models using bortezomib and rituximab [101], a phase II trial has been initiated to determine the clinical benefit and tolerability profile of bortezomib and rituximab combination therapy for patients with relapsed/refractory MCL (NCT ). Another small dose-escalation trial investigating the combination of bortezomib with the pan bcl-2 inhibitor obatoclax mesylate (GX15-070) found that this combination had acceptable tolerability, with the main grade 3/4 adverse event being thrombocytopenia (22%). There was also some evidence of efficacy as 3 patients experienced CR/CRu; 2 of these 3 patients had received prior HDT with SCT and the third had received prior bortezomib [102]. Multiple trials are looking at combination with R-CHOP [103], modified hyper-cvad [45], classic hyper-cvad, consolidation or maintenance post R-CHOP induction (SWOG) or post HDT (CALGB), BTZ + R-EPOCH (NCI) with a single agent window of opportunity initially, and also maintenance post induction among others as well as a large ongoing phase III study of R-CHOP vs R-CHP-BTZ (replacing bortezomib with vincristine) in the frontline setting. In a small study investigating bortezomib and ibritumomab combination therapy, 9 out of 12 patients with NHL completed therapy which resulted in an ORR of 56% (33% CR). At 1 year, the relapse rate was 25%, OS was 80%, and the PFS was 60% [104]. Other ongoing studies are also evaluating bortezomib in combination with gemcitabine (ClinicalTrials.gov NCT ), everolimus (NCT ), vorinostat, 17-AGG (HSP90 inhibitor) (NCT ), obatoclax mesylate (Bcl-2 family inhibitor) (NCT ), several chemotherapy or chemoimmunotherapy regimens (NCT and NCT ), and radioimmunotherapy, 131I-tositumomab (NCT ). Of notice, 65 studies are currently registered in ClinicalTrials.gov website using BTZ in MCL. Second generation proteasome inhibitors: carfilzomib (also an inhibitor of the chymotrypsin-like site but irreversible) or NPI-0052 (inhibits the 3 enzymatic sites within the proteasome unit) are also being looked at while immunoproteasome inhibitors might be very appealing due to their preferential inhibition of the proteasome within lymphoid cells New class of mtor inhibitors Temsirolimus. The mtor inhibitor temsirolimus has been shown to be clinically active in relapsed or refractory MCL patients during phase II and III studies, but CR rates were low. In the phase II study, temsirolimus resulted in a 38% ORR (3% CR), median TTP of 6.5 months, and response duration of 6.9 months [82] (Table 4). Grade 3 or 4 adverse events were observed in >90% of patients, including thrombocytopenia, neutropenia, and anemia. Thrombocytopenia was the most frequent cause of dose reduction but typically resolved within 1 week. In the phase III study, 2 dosing schedules (high and low dose) of temsirolimus were compared [105] (Table 4). The most common adverse events in this trial included thrombocytopenia, anemia, neutropenia, and asthenia. Compared to low-dose temsirolimus, high-dose temsirolimus resulted in a significant improvement in PFS and ORR. Based on these studies, temsirolimus has been approved by the EMEA for the treatment of relapsed or refractory MCL. As rituximab has improved response rates and overall outcome of lymphoma patients, the efficacy and safety profile of temsirolimus and rituximab combination therapy has also been investigated as second-line therapy for patients with relapsed or refractory MCL. In a phase II trial, Ansell et al. [106] showed that temsirolimus in combination with rituximab was well tolerated and had clinical activity in relapsed or refractory MCL. Combination therapy resulted in an ORR of 48% with 20% CR and 28% PR, and a median response duration of 9.5 months (95% confidence interval [CI], months). For rituximab-sensitive and refractory patients, the median response duration was 9.5 months (95% CI, months) and 7.15 months (95% CI, months), respectively. Hematological adverse events were most common and included thrombocytopenia (7%) and neutropenia (4%) Everolimus. Everolimus is an oral derivative of rapamycin, an inhibitor of the mammalian target of rapamycin (mtor). mtor is a serine/threonine kinase, and

10 78 A. Goy, B. Kahl / Critical Reviews in Oncology/Hematology 80 (2011) Table 4 Investigational therapies evaluated in patients with relapsed or refractory mantle cell lymphoma. Treatment Reference n ORR (%) CR (%) Median OS Median PFS (months) Flavopiridol a Kouroukis et al. [81] NR NR 3.3 Bendamustine + R Rummel et al. [54] NR 18 NR Bendamustine + Weide et al. [55] % at 2 years 21 NR mitoxantrone + R Robinson et al. [53] NR NR 19 Temsirolimus Witzig et al. [82] NR NR 6.9 Hess et al. [105] 54; 54 b 22; 6 b NR 12.8 months; 4.8; 3.4 b NR 10.0 months b Ansell et al. [113] months NR 6 Lenalidomide Habermann et al. [109] a NR Zinzani et al. [110] a NR NR NR Lenalidomide + R Wang et al. [112] 10 c NR NR NR CR, complete response; NR, not reported; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; R, rituximab. a Includes complete response unconfirmed. b Patients treated with 175 mg 3 weekly followed by 75 mg or 25 mg weekly, respectively. c Number of patients enrolled at the MTD. Median response duration (months) a downstream target of the PI3K and Akt/PKB pathway, and thus regulates many aspects of cell proliferation and survival. Single agent everolimus treatment was shown to be effective in inhibiting growth of MCL cells in vitro, and has shown synergistic activity in combination with other agents such as vincristine or rituximab [107]. A recently published phase I trial investigated the safety profile and clinical benefit of everolimus in a small group of patients with relapsed or refractory NHL (n = 13, DLBCL: 2, MCL: 2; FL: 5; PTCL: 1; CTCL: 1; ALCL: 2) [108]. Responses were observed among patients with DLBCL (1 CR, 1 PR) and FL (1 CR, 1 PR). Everolimus was well tolerated in this patient population, hematological grade 3 or 4 AEs included reversible thrombocytopenia (n = 2), lymphopenia (n = 4), and anemia (n = 1). However, as the number of enrolled patients was small (n = 13), further studies are warranted to assess both the clinical benefit and the tolerability profile of everolimus for patients with NHL [108]. A large international trial (PILLAR-1) with RAD001 (everolimus) in relapsed/refractory MCL is currently ongoing Flavopiridol new class of CDK inhibitors Flavopiridol is a semisynthetic flavone derivative of the alkaloid rohitukine. Flavopiridol inhibits the activity of several cyclin-dependent kinases (CDK), including the CDK4-cyclin D1 complex, thus inducing growth arrest in the G1 or G2 phase of the cell cycle. Given its effect on cyclin D1, MCL was one of the 1st tested NHL, though the initial schedules showed modest activity of flavopiridol; with both previously untreated and relapsed or refractory MCL patients showing an ORR of 11% (only PR) and a median response duration of 3.3 months [81] (Table 4; [105, ,53 55,81,82]). Pharmacodynamics based changes led to new schedules with impressive acitivity in CLL (with tumor lysis syndromes), but still modest in MCL [114]. Hematological adverse events were mild with flavopiridol, but 47% of patients developed grade 3 or 4 non-hematological adverse events including diarrhea, fatigue, and nausea [81]. Other CDK inhibitors (PD , P1446A-05, P2760) are under evaluation as well in the clinic Lenalidomide new class of IMiDs compounds Lenalidomide is an oral, anti-angiogenic, antiproliferative immunomodulatory agent that is active in several hematological malignancies. Lenalidomide has direct anti-proliferative effects in MCL cells, correlating with baseline levels of cyclin D1 and appears to be mediated by increased expression of the tumor suppressor genes p21 cip1 and secreted protein acidic and rich in cysteine (SPARC) [115]. In addition, lenalidomide enhances activation of T- and NK-cells through improved tumor B-cell/T-cell immunological synapses formation, thus leading to an enhancement in immune-mediated cell killing [ ]. Data from 2 phase II studies suggest that lenalidomide monotherapy is active in patients with relapsed or refractory MCL [109,110] (Table 4). In a subset analysis of the multicenter NHL-002 trial that focused on the 15 patients with relapsed or refractory MCL, lenalidomide produced an ORR of 53%, including a CR rate of 20% in these heavily pretreated patients (median of 4 prior therapies; range, 2 7) [109]. Responses were observed in patients who had received prior ASCT and those who received prior bortezomib. Notably, the Kaplan-Meier estimate for the median duration of response was 13.7 months, and the median PFS was 5.6 months. Adverse events were predominantly hematological, manageable, and consistent with that observed with lenalidomide therapy in patients with other hematological malignancies. Dose reductions were required in 53% of patients, mainly due to neutropenia, although only 1 patient