Paramount therapy for young and fit patients with mantle cell lymphoma: strategies for front-line therapy

Romaguera et al. reported a Phase II study of hyper-CVAD/MA regimen at MDACC, which included 65 younger patients (≤ 65 years old) with newly diagnosed MCL in a total of 97 cases. A recent 15-year follow-up report of this regimen (median follow-up 13.4 years) shows a mOS of 13.4 years for younger patients, with no recurrence in nearly one-third of the patients. This regimen also achieved a complete response rate (CR) of 89% in younger patients [27]. However, the toxicity treatment-related mortality (TRM) was 6.5% in patients less than 60 years old including 2 infections and 1 from an unknown cause. Twenty-five patients (26.0%) experienced secondary malignancies. The secondary malignancies were myelodysplastic syndrome/acute myeloid leukemia (MDS/AML, N = 13), prostate cancer (N = 4), lung cancer (N = 2), esophageal cancer (N = 1), renal cell carcinoma (N = 3), marginal zone lymphoma (N = 1), Burkitt lymphoma (N = 1), multiple myeloma (N = 1), pancreatic cancer (N = 1), colon cancer (N = 1) and thyroid cancer (N = 1). Six patients experienced MDS/AML in first remission while three patients experienced multiple malignancies. The incidence of a 10-year cumulative MDS/AML and all type of secondary malignancy for young patients in remission following R-HCVAD/MA was 6.2% (95% CI: 2.0–13.8%) and 15.4% (95% CI: 7.9–25.1%), respectively [7].

Based on the data from a single center trial at MDACC, clinical effectiveness and feasibility for RHyper-CVAD / RMA regimen requires more multicentered clinical trials. From 2005 to 2010, a prospective multicenter study from Gruppo Italiano Studio Linfomi (GISL) tested this regimen in 60 patients with a median age of 57 years old (29–66). They received 4 courses of R- hyper-CVAD/MA. If CR was not achieved after induction chemotherapy, ASCT followed. The RHyper-CVAD/ RMA regimen showed a good overall response rate (ORR) (83%) and CR (72%). At the median follow-up of 46 months, the estimated 5-year OS and PFS were 73 and 61%, respectively. The prognosis of low-risk MIPI score patients (estimated 5-year OS was 89 and 80%) was significantly better than the high-risk patients (24%). As expected, toxicity was evident, with 5% of TRM and 2% incidence of secondary AML [24]. However, a higher rate of treatment failure was observed in comparison to the MDACC trial. Only 37% of the patients (22 cases) completed the four courses of treatment in this study and 61% of patients completed the planned course of treatment in the SWOG multicenter, while the rate was 70% for the MDACC study [25].

Furthermore, a multicenter phase II study from SWOG showed that RHyper-CVAD/RMA was an effective regimen. The SWOG 0213 study treated 49 cases of newly diagnosed MCL patients with progressive stage with a median age of 57.4 years (35–69.8). The results also showed good ORR (86%). However, in comparison to the MDACC GISL trial, they had a significantly lower CR rate (47%). After a median follow-up of 4.8 years, the mPFS was 4.8 years (5.5 years for those ≤65 years) and mOS was 6.8 years. The low to intermediate-risk MIPI score accounted for 86% of patients. TRM for one and secondary MDS for two patients were observed [25]. Recently, another clinical trial from SWOG regarding RHyper-CVAD/RMA versus R-bendamustine followed by ASCT has been terminated due to insufficient collection of stem cells in the Hyper-CVAD/MA group. To improve the shortcomings of this regimen, which includes treatment-related complications (i.e. neutropenia), more intensive therapies are needed.

To summarize, the R-hyperVCAD/MTX/cytarabine regimen indeed demonstrates remarkable efficacy for MCL patients ≤65 years of age in the trials performed in MDACC, SWOG and GISL. Despite of the difference in patient characteristics in these three studies, considerably high ORR (> 80%) and CR/CRu (> 70%) were achieved in younger patients except for in the SWOG study, which had a CR/CRu of only 55%. Furthermore, the estimated 5-year OS and PFS were beyond 50% for younger patients. Meanwhile, encouraging results were shown by MDACC with the longest median survival of 13.4 years due to the longest median follow-up time of 13.4 years. Of note, the fact that at least half of the patients in these three studies were at low risk by MIPI indicates low risk patients may benefit most from this intensive treatment. Alternatively, the overall outcome may be driven in part by the favorable outcome of this low-risk group. However, toxicity was also very obvious, with incompletion of the planned courses for 29% in MDACC, 39% in SWOG, and 63% in GISL. The higher rate of incompletion of courses in GISL was possibly influenced by more patients being aged > 60 years (35%). Thus, this regimen may be too toxic in community hospital settings.

The European Group for Blood and Marrow Transplantation (EMBT) and the European Mantle Cell Lymphoma Network (EMCL) recommends that the induction therapy for MCL to include HD-Ara-c plus rituximab [28].

The addition of HD-Ara-c as a single agent to intensive therapy followed by supportive stem cell regimen was studied by the Nordic group in a non-randomized phase II multicenter trial [22]. The Nordic Lymphoma Group (NLG) conducted its first MCL phase II protocol (NLG MCL-1) in 1996 to 2000 with an induction treatment of 4 cycles of dose-intensified CHOP without rituximab, followed by carmustine, etoposide, cytarabine, and melphalan (BEAM) or the same regimen with cyclophosphamide instead of melphalan (BEAC), high-dose chemotherapy with unpurged or ex vivo–purged ASCT. Forty-one newly diagnosed patients below 66 years were enrolled and given three series of an augmented CHOP regimen. Responders underwent stem cell mobilization with a fourth course of CHOP, stem cell harvest, and ASCT. However, the results were disappointing. 85% of the patients failed the therapy and most of the evaluable patients had a demonstrable amount of minimal residual disease after transplantation, as did most of the evaluable stem cell products [16].

Compared with Nordic MCL1 trial, Nordic MCL2 was added with the intensification of the induction therapy to include high-dose cytarabine (3 g/m² every 12 h for a total of 4 doses; patients older than 60 years, cytarabine 2 g/m²) and rituximab [22]. In this 2nd Nordic MCL trial, 160 patients with previously untreated MCL less than 66 years of age received intensive therapy including R plus maxi-CHOP alternating with R plus HD-Ara-c. Responders received high-dose chemotherapy with BEAM or BEAC followed by salvage with R-in vivo purged ASCT. When compared with a historical control without HD-Ara-c and R [16], the 2nd Nordic MCL trial had significantly higher ORR (96% vs. 76%, p < 0.001) and CR rates (54% vs. 27%, p < 0.001). After a median follow-up of 11.4 years, the mOS and mPFS were 12.7 years and 8.5 years, respectively. Although half of the patients were still alive and 40% went under first remission after more than 12 years, an excess disease-related mortality was observed. Additionally, 20 cases (12.6%) of secondary malignancies were reported [9].

Compared with Nordic MCL2 trial, Nordic MCL3 trial had similar treatment with the MCL2 trial except for the addition of 90Y-Ibritumomab Tiuxetan (Zevalin) before transplant. In the Nordic MCL3 trial [29], 160 untreated MCL patients < 66 years received R-maxi-CHOP alternating with R-high-dose cytarabine (6 cycles total), followed by high-dose BEAM/BEAC and ASCT. Zevalin was given to responders not in CR before transplantation. The outcome did not differ from that of the MCL2 trial: OS, EFS, and PFS at year 4 were 78, 62, and 71%, respectively. For responding non-CR patients receiving Zevalin, the duration of response was shorter than for the CR group. Inferior PFS, EFS, and OS were predicted by positron emission tomography (PET) scan positivity prior to transplant and detectable minimal residual disease (MRD) post-transplant. Thus, intensification with Zevalin may be too late to improve the outcome of patients not in CR before the transplant.

Therefore, the Nordic MCL1 trial indicated an important strategy to improve the outcome of younger MCL patients – to intensify the induction chemotherapy. The Nordic MCL2 trial provided strong evidence of intensive induction chemotherapy including R plus HD-Ara-c for the dramatic improvement of clinical outcome with acceptable toxicities. Additionally, the Nordic MCL3 trial suggested the need for an early intensified induction therapy as the front-line treatment for younger MCL patients.

Another form of HD-Ara-c in the induction therapy is DHAP (Dexamethasone, HD-Aara-c, Platinum salt). Nevertheless, the efficacy of R-DHAP regimen is not only due to HD AraC but also the platinum salt. A prospective phase III LyMa Trial assessed R-DHAP as the induction therapy for younger MCL patients. In this study, R-DHA-Cisplatin combined with ASCT has been used as the first-line for the treatment of 299 MCL patients (median age: 57y, range: 27–65), with maintenance treatment by either rituximab or the wait and watch strategy. CR/Cru rates were 81.4 and 92% before and after the transplant, respectively. At a median follow-up of 35.8 months, mPFS and the mOS were not reached. The estimated 3y-PFS and OS were 73.7 and 82.6%, respectively [30]. R-DHA-Cisplatin was also used alternating with R-CHOP for induction therapy before transplantation in patients who did not reach at least a PR after 4 courses of R-DHA-Cisplatin. In the MCL Younger Trial (up to the age of 65 years) of the European Mantle Cell Lymphoma (EMCL) study, alternating courses of CHOP (3 cycles) and DHA-Cisplatin (3 cycles) plus R followed by a H-D Ara-C myeloablative regimen and ASCT (group B) increased CR rates and time to treatment failure (TTF) when compared to 6 courses of CHOP plus R followed by myeloablative chemo radiotherapy and ASCT (group A) in MCL. Conditioning regimen of group A was total body irradiation (TBI) + CTX and group B was TAM (TBI + HD-AraC + melphalan). The ORR for two groups before transplantation was similar (90% vs. 94%; p = 0.19), and CR rate of group B was significantly higher than group A (26% vs. 39%, P = 0.012). This indicated that the CR rate could be significantly improved when DHA-Cisplatin (combination form of HD-Ara-c) was added to induction therapy with R-CHOP [31]. After a median follow-up of 6.1 years [32], Group B had a significantly longer TTF (median 9.1 years [95% CI 6.3–not reached], 5 year rate 65% [95% CI 57–71]) compared to group A (3.9 years [3.2–4.4], 40% [33‐46]; hazard ratio 0.56; p = 0.038). During induction immunochemotherapy, group B had increased grade 3 or 4 hematological toxicity and grade 1 or 2 renal toxicity.

In another multicenter Phase II study from France Groupe d’Etude desLymphomes de l’Adulte, HD-Ara-c was added to the induction chemotherapy prior to ASCT in 60 cases of young MCL patients (median age 57.5, range 40–66) with advanced stage. Serial induction therapy consisted of 3 courses of CHOP, followed by 3 courses of DHA-Cisplatin regimen. The addition of rituximab was started from the third CHOP regimen. The conditioning regimen TAM was used as the early consolidation prior to ASCT. A total of 93% of patients responded to R-CHOP regimen, but only 12% of patients achieved CR. After R-DHA-Cisplatin chemotherapy, CR rate increased to 57%. Seven patients did not complete the planned treatment, including three due to disease progression and four due to toxicity, which was mainly renal insufficiency. Finally, 49 patients underwent ASCT, while only one failure in stem cell collection was found. After a median follow-up of 5.6 years, the 5-year EFS and OS was 64 and 75%, respectively. No treatment-related deaths and MDS or AML events occurred. Surprisingly, high incidence of other secondary tumors was observed in 11 patients, of which 5 were kidney cancer [33].

In a subgroup analysis of the LyMa Trial, a comparison was made between the three platinum salt in R-DHAP regimesn for young previously untreated MCL patients (n = 298) [34]. The results showed that R-DHA-Oxaliplatin is a better induction regimen before ASCT than R-DHA-Carboplatin and R-DHA-Cisplatin in MCL patients. MCL patients treated with R-DHA-Oxaliplatin had longer PFS and OS. Compared with carboplatin and oxaliplatin, Cisplatin showed more toxicity particularly concerning renal failure. All cases of renal failure during induction (n = 8) occurred in Cisplatin group. Twenty-seven (15%) and 38 (21%) out of 184 patients in cisplatin group switched to carboplatin and oxaliplatin, respectively.

EBMT/ EMCL had come up with a consensus for ASCT as the standard first-line consolidation therapy in MCL, giving it a big role in regards to intense therapy regimens [28]. ASCT had high response rates and significantly prolonged survival, especially PFS, with acceptable toxicity. In a randomized controlled trial, the European MCL Network compared two groups with or without ASCT for consolidation. The study revealed that the PFS was significantly prolonged in patients aged younger than 65 years with advanced stage, who had consolidation with ASCT compared to the interferon group (IFN) (39 months vs. 17 months, P = .0108) [21]. On the other hand, 85% of patients in the ASCT group had infectious complications due to cytopenias, while this was the case for only 22% of the IFN group. The mortality was 5% in the ASCT group, while 0% in the IFN group. In addition, there was no significant difference in OS between the two groups [21]. This may also be explained by the fact that some patients in the IFN group who experienced relapse subsequently received transplantation. A longer follow-up is needed to determine the effect on OS.

EBMT/ EMCL had also reached the consensus that CR or PR must be achieved before ASCT [28]. The vast majority of clinical trials showed that the outcome of SCT in relapsed MCL patient was often poor, with the expected PFS of about 20 to 40% [35]. Therefore, more recent studies evaluated ASCT as a consolidation therapy in patients sensitive to chemotherapy after the first remission. As shown in Table 1, the median PFS was prolonged to greater than 5 years for the MCL patients receiving ASCT in the first remission, with acceptable toxicity [8, 9, 22, 36, 37]. Meanwhile, for young MCL patients who do not achieve CR after induction therapy, ASCT can further improve the CR rate by 20–40% based on the prior treatments according to MCL net data [31].

Thus far, there are no prospective studies on the choice of conditioning regimen. As demonstrated in the large retrospective study SFGM-TC [38], it was difficult to prove which conditioning regimen was optimal. Additionally, different induction therapies prior to preparative treatment increases the difficulty of comparing between preparative regimens. According to the NCCN NHL Database study [26], commonly used conditioning regimens includes BEAM, TAM (TBI-aracytine- melphalan), CBV (cyclophosphamide, BCNU, VP-16) and TBI/Cy. The MCL network Phase III trial showed that alternative R-CHOP/R-DHAP induction followed by TAM had a longer PFS when compared to R-CHOP followed by TBI/Cy [31]. MCL is sensitive to TBI. However, if the conditioning and induction regimen consist of TBI without HD-Aar-c or other aggressive chemotherapy before transplantation, it seems to be difficult to benefit from the OS. Another comparative retrospective analysis of the European MCL (with TBI) and MCL Nordic group (no TBI), which used a similar induction regimen containing HD Ara-C, showed that TBI seems to improve PFS only in the group of patients who are in PR before ASCT [39]. Because the goal of most front-line induction regimens is CR, TBI is no longer commonly used in Europe and instead, the BEAM regimen has become the new standard [40].

A trial conducted by National Cancer Institute, Bethesda enrolled a total of 26 patients with median age of 57 years (range: 22–73 years) with previously untreated MCL who were treated with 6 courses of Dose-adjusted -EPOCH-R (etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin combined with rituximab) followed by 5 courses of immunotherapy plus autologous tumor-derived idiotype (Id) -vaccine. The study showed a good response rate, including CR rate of 92% and PR rate of 8%. After a median follow-up of 11 years, the mPFS and mOS were 2 years and 8.7 years, respectively [41]. Another phase II clinical trial reported by the same group used Dose-adjusted-EPOCH-R combined with bortezomib and achieved a considerably high CR rate (up to 92%) in patients with newly diagnosed MCL. Forty-three cases, aged 41–75 years, were included in the trial [42]. However, no detailed age related subgroup analysis was conducted for the response or outcome of the treatment in both these studies.

A phase II single-arm clinical trial from two centers (Dana-Farber Cancer Institute and Beth Israel Deaconess Medical Center) tested the induction regimen of RB (rituximab/bendamustine) followed by RC (rituximab/high-dose cytarabine) for 23 cases (median age: 57, range, 42–69), who were transplant-eligible with newly diagnosed MCL. Patients received three cycles of RB followed by three cycles of RC. This induction regimen had a considerable high CR/unconfirmed rate (96%) and MRD-negative rate (93% in 15 evaluable patients). Twenty-one patients received consolidation with ASCT. After a median follow-up of 13 months, the median PFS was found to be 96%. No TRM was observed. However, the high remission rates found in this study may relate to a higher proportion of low-risk patients (70%). In addition, the follow-up time for the study was short, which warrants further observation [43].

CdM was used for the treatment of 62 cases of MCL and low grade NHL with a median age of 59 (31–76) years. CdM displayed high activity for MCL with an ORR of 100%. However, the CR rate was only 44%. In 9 patients with previously untreated MCL, the ORR also reached 100%, while the CR rate was found to be 33%. Among all patients, myelosuppression was the major toxicity with 23% grade 3 and 50% grade 4 granulocytopenia [44]. However, the results of this study are controversial since the study included only a few patients with previously untreated MCL.

A phase II trial of R-CHOP followed by Zevalin in untreated MCL was performed by the Eastern Cooperative Oncology Group Study E1499 [45]. Fifty-six patients with 91% of stage III/IV and 78% of marrow-positive were enrolled between November 2003 and February 2005. Median age was 61 (range: 33–83) years while low to intermediate MIPI score patients accounted for 77%. After four cycles of R-CHOP, responding (CR/PR) and stable patients received Zevalin. Fifty-one (91%) patients received all treatment, with 42% CR/Cru, 32% PR, 12% stable and 4% unevaluable, and an improvement in response in 16 patients after Zevalin. At a median follow-up of 9.8 years, median OS for those patients under 65 years has not been reached at 10 years. No therapy related myeloid neoplasia was observed in this trial. The second malignancies include non-small cell lung cancer (N = 2), bladder cancer (N = 1), ampullary adenocarcinoma (N = 1) and resected localized non-melanoma skin cancers (N = 2) [46].

Howard, O. M. et al. conducted a phase II study with six cycles of R-CHOP regimen as the induction therapy in 40 patients (median age: 55 years, range: 31–69) with newly diagnosed MCL. The addition of monoclonal anti-CD20 antibody R to CHOP regimen significantly improved the ORR to 96% and CR / unconfirmed CR (CRu) rate to 48% [20], which were in accordance with the results from a prospective randomized trial of the German Low Grade Lymphoma Study Group (ORR: 94% in RCHOP vs 75% in CHOP, P = 0.0054; CR: 34% in RCHOP vs 7% in CHOP; P = 0.00024) [47]. However, the duration of remission was short, with a median PFS time of only 16.5 months at a median follow-up of 25 months [19, 20]. Also, no differences were observed for PFS in the two studies [20, 47]. Compared with CHOP regimen, R-CHOP could not translate favorable clinical and molecular response rates into prolonged PFS. Nevertheless, R-CHOP may transiently clear peripheral-blood (PB) or bone marrow (BM) of detectable tumor cells [20]. Another study by LaCasce and colleagues [26] showed poor survival advantage for the newly diagnosed younger MCL patients receiving R-CHOP alone. R-CHOP (n = 29) demonstrated inferior PFS compared with other aggressive regimens (n = 138) (P < .004).

A phase I/Ib study from 2012 to 2014 conducted in Ohio State University demonstrated that ibrutinib in combination with bendamustine and rituximab in 48 patients with B-cell lymphoma, which included 17 patients with MCL (5 untreated, ages 62–72), produced an ORR and CR rate for MCL of 94 and 76%, respectively. The mPFS and mOS that were not reached along with an acceptable toxicity profile. However, the sample size of untreated young patients with MCL in this study was too small to make any significant conclusions [48].

In a phase 1b, open-label, non-randomized study from 2012 to 2013 at six centers in the USA and France involving 32 patients (ages 18 years or older) with CD20 positive B-cell NHL, five patients with MCL received ibrutinib combined with standard doses of R-CHOP regimen. While the ORR and CR rate for all patients was 94 and 72%, respectively, and the combination was found to be safe, the sample size of MCL was small to make any significant conclusions [49]. Larger sample sizes are needed to test this regimen in phase III trials.

A pivotal phase III trial called the LYM-3002 study from 2008 to 2011 was conducted at 128 sites in 28 countries across Europe, Asia, North America, and South America. This study compared 244 patients who received R-CHOP with 243 patients who received VR-CAP (vincristine in R-CHOP regimen was replaced by bortezomib) and all were previously untreated with MCL. Approximately one third of patients were less than 60 years of age. CR rate in the VR-CAP group was 53%, which was slightly higher than 42% of R-CHOP group. After median follow-up of 40 months, median duration of CR in VR-CAP group was significantly longer (42.1 months vs. 18.0 months). Meanwhile, the PFS was increased by 96%, which almost doubled (16.1 months vs. 30.7 months, p < 0.001). There was no significant difference in mOS between the two groups. However, outcome analysis according to age subgroups was lacking [50].