Search for a command to run...
This study evaluated the efficacy and safety of bendamustine plus rituximab (BR) in symptomatic untreated patients with Waldenström macroglobulinaemia (WM). A total of 69 patients were enrolled in this multi-centre study between January 2013 and December 2017 in 13 French centres, 46 (67%) were males, median age at diagnosis and at initiation of therapy was 69 years (range: 45–88) and 71 years (range: 46–88) respectively. Of the patients tested for MYD88 and CXCR4 mutations, 45/51 (88%) and 11/44 (25%), respectively, were mutated. Karyotype analysis showed complex karyotype in 6/44 (14%) cases, and 1/34 (3%) had a TP53 mutation. Median β2-microglobulin and haemoglobin levels were 3·49 mg/l (range: 1·5–7·4) and 97 g/l (range: 52–170) respectively and median IgM level was 23·8 g/l (range: 0·6–12). According to the International Prognostic Scoring System for WM (Morel et al, 2009), 14 (20%) patients were low risk, 32 (46%) were intermediate and 23 (34%) were high risk. Patients received 375 mg/m2 of rituximab on day 1 and 90 mg/m2 of bendamustine on days 1 and 2, repeated every 4 weeks, for a maximum of 6 cycles. All patients except one achieved at least a minor response (MR), with overall (ORR) and major (MRR) response rates of 97%, and 96% respectively. Thirteen (19%) patients achieved a complete response (CR), 26 (37%) very good partial response (VGPR), 28 (40%) partial response (PR) and 1 (1%) MR; one patient had stable disease. The cumulative incidence of ORR was 69·6% at 3 months, 91% at 6 months, 94·5% at 9 months, 96% at 1 year and 97% at 18 months (Fig 1A). The cumulative incidence of CR was 10%, 15·9% and 17·4% at 3, 6 and 9 months and 19% at 1 year (Fig 1A). The presence of MYD88 and CXCR4 mutations did not impact on disease response. Overall, 39 (56%) patients completed the 6 cycles of BR at the full dose during the planned time and 30 had rituximab dose reduction to 70 mg/m2 and/or less than 6 or delayed cycles. The main adverse events that required total dose reduction were haematological (N = 10) and infectious (N = 8) (Table 1); 35 (51%) developed one or more long-lasting cytopenia: neutropenia (N = 26, 38%), anaemia (N = 17, 25%) and thrombocytopenia (N = 11, 16%), all grade I or II. Median duration for prolonged neutropenia, anaemia, and thrombocytopenia was 9 (3–24), 6 (3–36) and 9 (3–36) months, respectively. Haematological N = 10 Infectious N = 8 Cutaneous N = 1 After a median follow-up of 23 months (range: 3–52) after BR initiation, disease progression had occurred in 6 patients at a median time of 9 months (range: 5–19). These 6 patients were re-treated and at the last follow-up, one had reached CR, 2 VGPR and 2 PR; and one patient had progressive disease and died. Overall (OS) and progression-free (PFS) survival probabilities at 2 years were 97·1% (95% confidence interval [CI]: 81–99) and 87% (95% CI: 74–94), respectively. The reduced number of BR cycles showed no impact on PFS, therefore the respective 2-year PFS probabilities for the 39 patients who received the total 6 cycles at full dose compared to the rest of patients was 87% (95% CI: 69–95) vs. 88% (95% CI: 68–96), P = 0·793. Of note, when stratifying according the MYD88 and CXCR4 mutations, we did not find a significant impact on survival for the patients with both MYD88 and CXCR4 mutations compared to the rest of the cohort (P = 0·22). However, none of patients with both MYD88 and CXCR4 mutations have progressed whereas two of the 6 double-negative patients had disease progression (P = 0·041) (Fig 1B). At the last follow-up, none of the 6 patients with complex karyotype had progression or death events and thus had 100% PFS probability. One patient developed myelodysplastic syndrome 6 months after BR initiation, and one other patient developed breast cancer 18 months after BR initiation. This study showed that the BR combination can achieve a remarkable ORR that translated to high OS and PFS probabilities. Toxicities associated with this regimen led to reduction of BR dose in 44% of patients, however this reduction did not impact OS or PFS. Long-lasting cytopenia (grades I, II) was observed in half of the patients but was manageable in all cases. BR was compared with R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisolone) in a phase 3 trial (Rummel et al, 2013), which reported an ORR for BR of 96% compared to 94% with R-CHOP, and a remarkably longer median PFS with BR (69 months) compared to R-CHOP (28 months), and a better toxicity profile for BR. Two other retrospective studies in relapsed/refractory WM patients treated with BR demonstrated the effectiveness of this combination with an ORR of 80% and 83% respectively, and a 2-year PFS of 60% (Treon et al, 2011; Tedeschi et al, 2015). Recently, BR was compared to dexamethasone, rituximab and cyclophosphamide (DRC) in patients with WM in the treatment-naïve setting: the ORR was 93% with BR vs. 96% with DRC and the 2-year PFS was 88% with BR and 61% with DRC (Paludo et al, 2018). More interestingly, a good proportion of patients in our study had MYD88 and/or CXCR4 mutations, and 11 (16%) had both, although these mutations did not have a significant impact on disease response or survival outcome. A similar result was also reported in the recent randomized trial evaluating the combination of ibrutinib and rituximab (Dimopoulos et al, 2018). In addition, the CR rate of 19%, observed in our study at 1 year after BR initiation, is among the highest observed in this indication. Despite the limitation of the retrospective aspect of our study, our results are comparable to those from the few prospective studies of BR in WM patients, although our study had the largest number of patients. We confirmed that BR regimen can induce high and deep response rates in untreated WM patients, which translates into prolonged OS and PFS. In view of some adverse events, we recommend that the bendamustine dose should be reduced to 70 mg/m2 in elderly patients. Longer follow-up is needed to assess the long-term risks of this regimen. Author contribution: KL and VL designed the study, all authors except SP enrolled patients, SP performed NGS analysis, KL, AB de M and VL wrote the paper. All authors read and approved the final version of the manuscript for submission. The authors have no competing interests.