Background
Open-label, non-inferiority phase III German StiL trial of 514 patients with indolent NHL (follicular grade 1/2, lymphoplasmacytic, small lymphocytic, mantle cell, marginal zone) requiring therapy. Eligibility: untreated stage III/IV with ≥1 of cytopenias (Hb <10 g/dL, ANC <1.5, PLT <100K), B-symptoms, bulky disease, organ compression, rapid (>50% in <6 mo) growth, or hyperviscosity.
Interventions and follow up
Arm A (B-R): Bendamustine 90 mg/m2 D1-2 + rituximab 375 mg/m2 D1, q28d, up to 6 cycles (n=274).
Arm B (R-CHOP): Cyclophosphamide 750, doxorubicin 50, vincristine 1.4 mg/m2 D1, prednisone 100 mg D1-5 + rituximab 375 mg/m2 D1, up to 6 cycles (n=275).
Primary endpoint: PFS (non-inferiority).
mFollow up: 45 months.
Arm B (R-CHOP): Cyclophosphamide 750, doxorubicin 50, vincristine 1.4 mg/m2 D1, prednisone 100 mg D1-5 + rituximab 375 mg/m2 D1, up to 6 cycles (n=275).
Primary endpoint: PFS (non-inferiority).
mFollow up: 45 months.
Results
mPFS: 69.5 vs 31.2 mo (B-R vs R-CHOP); HR 0.58; 95% CI 0.44-0.74; P<.0001.
CR: 40% vs 30%; P=.021.
FL subgroup PFS: HR 0.61; 95% CI 0.42-0.87; P=.0072.
MCL subgroup PFS: HR 0.49; 95% CI 0.28-0.79; P=.0044.
MZL subgroup PFS: HR 0.70; 95% CI 0.34-1.43; P=.3249.
OS: no significant difference — HR 0.82; 95% CI 0.58-1.15; P=.249.
Second-line therapy required: 34% vs 52%.
CR: 40% vs 30%; P=.021.
FL subgroup PFS: HR 0.61; 95% CI 0.42-0.87; P=.0072.
MCL subgroup PFS: HR 0.49; 95% CI 0.28-0.79; P=.0044.
MZL subgroup PFS: HR 0.70; 95% CI 0.34-1.43; P=.3249.
OS: no significant difference — HR 0.82; 95% CI 0.58-1.15; P=.249.
Second-line therapy required: 34% vs 52%.
Adverse events
Hematologic (grade 3-4, B-R vs R-CHOP): neutropenia 29% vs 69%.
Infections: any-grade 37% vs 50%; sepsis <1% vs 3%.
Nonhematologic (any grade): alopecia 0% vs 100%; paresthesia 7% vs 29%; allergic skin reaction 1% vs 6%.
Second malignancy: 7% vs 8.3%.
Infections: any-grade 37% vs 50%; sepsis <1% vs 3%.
Nonhematologic (any grade): alopecia 0% vs 100%; paresthesia 7% vs 29%; allergic skin reaction 1% vs 6%.
Second malignancy: 7% vs 8.3%.
Conclusions
Bendamustine-rituximab provides superior PFS and a more favorable toxicity profile (less neutropenia, infection, alopecia, neuropathy) versus R-CHOP for first-line indolent and mantle cell NHL, without an OS difference.
Key Limitations
Open-label design; PFS benefit did not translate into OS difference. The mixed histology (including MCL) complicates interpretation across subtypes, and no rituximab-maintenance arm was prespecified (maintenance was censored). MZL subgroup was underpowered.
Clinical Context
StiL, with the corroborating BRIGHT trial, established B-R as a preferred first-line option for indolent and mantle cell NHL per ASCO/ESMO guidance, owing to its favorable PFS and tolerability versus R-CHOP.
Study update: Rummel MJ et al, 2017, ASCO abstract — 10-yr OS 71% vs 66%, P=.249.
Study update: Rummel MJ et al, 2017, ASCO abstract — 10-yr OS 71% vs 66%, P=.249.