Toxicity of Maintenance Rituximab in Older Adults with Non-Hodgkin Lymphoma

Background: The introduction of rituximab has improved the outcomes of B-cell non-Hodgkin lymphoma (BCL) across all histologies. Extended use of rituximab, or maintenance rituximab (MR), improves progression-free survival, but likely not overall survival, in follicular lymphoma patients following induction rituximab with or without chemotherapy. There is little evidence to support the use of MR in other histologic subtypes. Understanding the use and outcomes of MR is essential, particularly in older patients who may be at increased risk of adverse events. Our objective was to study the impact of MR on the risk of complications in older adults with BCL.

Methods: In the Surveillance, Epidemiology and End Results (SEER)-Medicare dataset,we identified patients age 66 or older diagnosed with BCL from 2001-2009 who had more than one claim for rituximab after diagnosis. Receipt and duration of MR were defined by the time from the first claim for rituximab after completion of induction therapy (rituximab alone or with chemotherapy) and last claim for rituximab with no gap >6 months in rituximab claims. Complications that were evaluated were hospitalization for infection, receipt of growth factors, and receipt of transfusion of blood products. We estimated the impact of MR duration (in months) on time to first complication using proportional hazards regression, treating maintenance rituximab duration as a time-dependent, continuous covariate, and controlling for demographic and disease characteristics and receipt of chemotherapy. Observations were censored at the first of end of follow-up, death, subsequent cancer diagnosis, stem cell transplant or other evidence of disease progression.

Results: There were 20,669 BCL patients across the spectrum of BCL who received rituximab at any time after diagnosis, including 3,570 who received rituximab beyond the induction period; the median age of patients receiving MR was 76 in this cohort, and only 21% of patients had a diagnosis of FL. More than 80% of all patients also received chemotherapy during induction. Duration of MR varied from 0 to 80 months. Among those who had MR, median duration was 5 months. There were 11,216 patients who experienced at least one complication after completion of induction therapy. Receipt of growth factors accounted for 56% of first complications, while infection and blood transfusion accounted for 25% and 20%, respectively. Each additional month of MR was associated with a 1% increased risk of any complication, controlling for patient demographics, disease characteristics, and receipt of chemotherapy (adjusted hazard ratio 1.013, 95% CI 1.006-1.021, p<0.0005). Results were similar when the analysis was limited to events within the first two years after the end of induction therapy, limited to patients who survived that period. Results were also similar when complications were treated as competing events; each additional month of MR increased the risk of infection by 4% (AHR 1.037, 95% CI 1.028-1.046, p<0.0001), the risk of transfusion by 2% (AHR 1.024, 95% CI 1.012-1.036, p<0.001), and the risk of growth factor receipt by almost 8% (AHR 1.079, 95% CI 1.061-1.097, p<0.001).

Conclusions: Older patients in the US receive MR for a broad range of BCL diagnoses beyond the labeled indication. However, this approach increases the risk of infections and the need for transfusional and growth factor support. Moreover, the risk of these complications increases with greater durations of rituximab exposure. These findings, from real-world practice settings, raise concerns about harms from overuse of maintenance rituximab in older patients.