Safety of bendamustine for the treatment of indolent non-Hodgkin lymphoma: a UK real-world experience

Abstract

Introduction: Bendamustine is among the most effective chemotherapeutics for indolent B-cell non-Hodgkin lymphomas (iNHL), but trial reports of significant toxicity, including opportunistic infections and excess deaths, led to prescriber warnings. We conducted a multicentre observational study evaluating bendamustine toxicity in real-world practice. Methods: Patients receiving at least one dose of bendamustine (B) +/- rituximab (R) for iNHL were included. Demographics, lymphoma and treatment details and grade 3-5 adverse events (AEs) were analysed. Results: 323 patients were enrolled from 9 NHS hospitals. Most patients (96%) received BR and 46% R maintenance. 21.7% experienced serious AEs (SAE) related to treatment, including infections in 12%, with absolute risk highest during induction (63%), maintenance (20%), and follow-up (17%), and the relative risk highest during maintenance (54%), induction (34%) and follow-up (28%). Toxicity led to permanent treatment discontinuation in 13% of patients, and 2.8% died of bendamustine-related infections (n=5), myelodysplastic syndrome (n=3), and cardiac disease (n=1). More SAEs per patient were reported in patients with mantle cell lymphoma, poor pre-induction PS, poor pre-maintenance PS, abnormal pre-induction total globulins and in those receiving growth factors. Use of antimicrobial prophylaxis was variable, and 3/10 opportunistic infections occurred despite prophylaxis. Conclusion: In this real-world analysis, bendamustine-related deaths and treatment discontinuation were similar to trial populations of younger, fitter patients. Poor PS, mantle cell histology and maintenance rituximab were potential risk factors. Infections, including late onset events, were the most common treatment-related SAE and cause of death warranting extended antimicrobial prophylaxis and infectious surveillance, especially in maintenance-treated patients.