Treatment patterns and outcomes of second-line rituximab and thrombopoietin receptor agonists in adult immune thrombocytopenia: A Canadian retrospective cohort study

Abstract

BackgroundThe optimal choice of second-line treatment for immune thrombocytopenia (ITP) is unclear. Guidelines recommend either rituximab, splenectomy, or thrombopoietin receptor agonists (TPO-RA). There is, however, scarce data comparing treatment patterns, outcomes and resource utilization across second-line treatments. Despite Canada's universal healthcare system, publicly funded access to second-line ITP therapies is highly variable across provinces/territories. ObjectivesTo describe treatment patterns and compare health service utilization and outcomes among recipients of second-line rituximab and TPO-RA for ITP. MethodsIn this multicentre retrospective cohort study, we included adults who received second-line ITP therapies rituximab, eltrombopag and romiplostim (2012-2020) in Alberta, Canada. Patients were identified through a provincially-funded special drug access (STEDT) program. We examined treatment patterns, predictors of second-line treatment, hospitalizations, blood product utilization, and outcomes. Kaplan-Meier survival curves were used to estimate the cumulative incidence of ITP-related hospitalizations (bleeding or infections), overall survival (OS) and relapse-free survival (RFS). Cox proportional hazards regression was used to examine the impact of second-line therapy on OS. Results223 patients received rituximab (67 %), eltrombopag (29 %), and romiplostim (4 %). TPO-RA recipients experienced significantly longer time from ITP diagnosis to second-line therapy compared with rituximab recipients (15.9 vs 6.7 months, P < 0.0001), accompanied by significantly higher platelet and IVIG utilization prior to second-line therapy. Age (adjusted odds ratio [aOR] 1.04, 95 % CI 1.02–1.07, P < 0.0001) and prior intracranial hemorrhage (aOR 12.7, 95 % CI 1.6–272.8, P = 0.03) were significant predictors of second-line TPO-RA. TPO-RA is associated with a trend towards longer median RFS (6.3 vs 3.8 years, P = 0.06) compared with rituximab, and similar rates of ITP-related hospitalizations, major bleeding, and thromboembolism. Age, time period, and Charlson comorbidity index, but not second-line ITP therapy, were significant predictors of OS. ConclusionsOur study identified older age and intracranial hemorrhage as predictors of second-line TPO-RA prescription in a real-world practice. There were no significant differences in hospitalizations and outcomes between second-line rituximab and TPO-RA, although delayed initiation of TPO-RA was associated with higher blood product utilization.