Thrombotic events in patients receiving immune checkpoint inhibitors: Incidence, risk factors, and prognostic outcomes from a retrospective cohort study.

Abstract

e24155 Background: Immune checkpoint inhibitors (ICIs) have transformed cancer outcomes across multiple malignancies. However, emerging evidence suggests that thrombotic events, including venous thromboembolism (VTE) and arterial thromboembolism (ATE) are serious complications that may affect survival. The true incidence, timing, risk factors, and impact of these events in ICI patients remain unknown. This study aims to determine their incidence, identify independent risk factors, and evaluate effects on outcomes and survival. Methods: We conducted a retrospective cohort study of adult patients treated with ICIs at a tertiary academic center. Thrombotic events were classified as VTE or ATE occurring from ICI initiation through follow-up. Clinical, laboratory, and treatment-related variables were collected, including use of anticoagulants or antiplatelet agents, prior thrombotic history, age, sex, tumor type, stage, ICI regimen, and performance status. Outcomes assessed included overall survival (OS) . Kaplan–Meier methods and multivariable Cox proportional hazards models were used to evaluate survival impact and identify independent predictors. Results: Among 746 patients treated with immune checkpoint inhibitors (median age at initiation 65.1 years; 64.7% male; 94.1% ECOG 0–1), 125 thrombotic events (TE) occurred over a median follow-up of 23 months, corresponding to a cumulative incidence of 16.3%. Of these events, 105 (84.0%) were venous thromboembolism (VTE) and 20 (16.0%) were arterial thromboembolism (ATE). Of all TE, 20.8% occurred within 0–6 months, 8.0% at 6–12 months, 5.6% at 13–18 months, and 4.0% at 18–24 months (38.4% within 24 months). VTE accounted for 21.9% of all events in the first 0–6 months and declined in proportion among subsequent events, while ATE increased progressively (11.5%, 50.0%, 85.7%, and 100% across successive intervals). TE was associated with worse overall survival (log-rank p<0.001) and remained significant on multivariable analysis (HR 2.63, 95% CI 1.90–3.58, p<0.001). Independent predictors of TE included prior VTE (HR 1.92, p=0.035), lower baseline hemoglobin (HR 0.87, p=0.01), and older age at immunotherapy initiation (HR 1.02 per year, p<0.001). Conclusions: Thrombotic events are common in patients receiving immune checkpoint inhibitors, with a higher incidence than reported in clinical trials and consistent with real-world data and are independently associated with worse survival. A temporal shift from early venous to late arterial events highlights the need for time-dependent risk stratification.