ASH 2014 Abstract 145. Bendamustine Plus Rituximab Versus Fludarabine Plus Rituximab in Patients with Relapsed Follicular, Indolent, or Mantle Cell Lymphomas – 8-Year Follow-up Results of the Randomized Phase III Study NHL 2-2003 on Behalf of the StiL (Study Group Indolent Lymphomas, Germany)
Mathias J. Rummel, MD et al.
Introduction: Fludarabine plus rituximab (F-R) is an established treatment option for patients (pts) with relapsed/refractory follicular lymphoma (FL), other indolent lymphoma, or mantle cell lymphoma (MCL). To further improve the treatment in this setting we initiated in 2003 a multicenter, randomized phase III study to compare the efficacy and safety of bendamustine plus rituximab (B-R) versus F-R for pts with relapsed FL, other indolent lymphomas or MCL.
Results: A total of 219 pts were evaluable for the analysis (114 B-R; 105 F-R). There were no significant differences between arms for patient characteristics, including age, stage, LDH, IPI, FLIPI, bone marrow infiltration, and extranodal involvement. Most pts had stage IV (71.6% B-R; 60.6% F-R) or stage III disease (21.1% B-R; 25.3% F-R). Median patient age was 68 yrs (range 38–87). Patients had received a median of 1 prior therapy (range 1–7). Histological subtypes were distributed equally between the B-R and F-R arms: follicular, 45.9% and 47.5%, respectively; Waldenström’s Macroglobulinemia, 11.9% and 11.1%; MCL, 20.2% and 21.2%; other indolent lymphomas, 23% and 20.2%. A median of 6 cycles were given in both treatment arms, with 75.2% and 53.4% of B-R and F-R pts receiving 6 cycles, respectively. At the time of this analysis (June 2014), the median observation time was 96 months. The ORR was significantly higher with B-R than with F-R (83.5% vs. 52.5%, respectively; p< 0.0001). The CR rate with B-R was also significantly higher than that with F-R (38.5% vs. 16.2%; p=0.0004). Median PFS was significantly prolonged with B-R compared with F-R (34 vs. 12 months; hazard ratio [HR] 0.54, 95% confidence interval [CI] 0.38–0.72; p<0.0001). The longer PFS translated into a survival benefit with a significantly longer median overall survival in the B-R group than in the F-R group (110 vs. 49 months; HR 0.64, 95% CI 0.45–0.91; p=0.0125) comprising 55 and 71 deaths in the B-R and F-R groups, respectively.
There were no significant differences in the rates of alopecia, stomatitis, erythema, allergic reactions, peripheral neuropathy, or infectious episodes between groups. Hematologic toxicities were also similar between arms: 8.9% grade 3/4 neutropenia with B-R vs. 9.1% with F-R; 11.8% grade 3/4 leukocytopenia with B-R vs. 12.4% with F-R. The overall incidence of serious adverse events was similar for the B-R and F-R groups (17.4% and 22.2%, respectively). 17 pts (14.9%) developed a secondary neoplasia after B-R compared with 16 pts (15.2%) after F-R. Of these, 5 pts in the B-R group, and 3 pts in the F-R group developed a secondary hematological neoplasia (2 AML [1 AML M4], 1 CML, 1 DLBCL, and 1 HD after B-R; and 2 AML M4, and 1 MDS after F-R).
An unplanned subanalysis showed that rituximab maintenance therapy significantly prolonged overall survival (HR 0.38, 95% CI 0.32-0.71; p=0.0003) and PFS (HR 0.35, 95% CI 0.31-0.62; p< 0.0001) in the small group of 40 pts who received this treatment (23 B-R, 17 F-R) compared with those who did not. Although the numbers are too small in this non-randomized comparison to draw validated conclusions, these results appear to confirm the favorable role of rituximab maintenance.
Conclusions: B-R was more effective than F-R in this setting of relapsed FL, other indolent lymphomas and MCL due to higher overall and complete response rates, a longer PFS, and an improved OS. These data confirm the high anti-lymphoma activity of B-R.
CARE Faculty Canadian Perspective:
For several years, fludarabine with or without rituximab has been a common option in relapsed/refractory iNHL after failure of first-line regimens such as CVP-R. The STiL-2 study suggests that B-R results in improved response rates, PFS, and OS compared to F-R, without significant incremental toxicity. Therefore, the StiL-2 trial places B-R as an attractive option for this patient population. However, many centers in Canada have now shifted practice to B-R as the fist-line regimen of choice for iNHL based on the STiL-1 and BRIGHT trials. It may be difficult to apply the results of STiL-2 to those patients initially treated with B-R who then experience relapse or progression in terms of the efficacy of B-R re-treatment. In STiL-2, 16% patients had received bendamustine with or without rituximab as a prior line of therapy, but response rates and outcomes with B-R re-treatment were not mentioned in the abstract or oral presentation. The final publication may provide additional information.