Venous and arterial thromboembolic events associated with osimertinib in patients with lung cancer: a study by the thrombosis and cancer group of the spanish society of medical oncology (SEOM)

Abstract

PURPOSE: To determine the real-world clinical practice incidence and predictive factors of venous and arterial thromboembolic events (VTE/ATE) in lung cancer patients treated with osimertinib. METHODS/PATIENTS: A multicenter retrospective study including 546 lung cancer patients who initiated treatment with osimertinib before or during 2024, with a minimum follow-up of 6 months. The primary objective was to evaluate the incidence of VTE/ATE. Secondary objectives included identifying predictive factors for thrombosis and assessing its impact on overall survival (OS). Competing risk regression (Fine-Gray), multivariable Cox models, and Kaplan-Meier analyses (including landmark analysis at 6 months) adjusted for time-dependent covariates were performed to prevent immortal time bias. Missing data were handled using a complete-case strategy (n = 544) given <0.5% missingness in model covariates. RESULTS: Overall VTE/ATE incidence was 15.0% (82/546; 10.5 events per 100 person-years)[cite: 3, 4]. The cumulative incidence of first VTE/ATE (accounting for non-thrombotic death as a competing risk) was 8.34% (95% CI 6.08-10.74%) at 6 months and 10.54% (95% CI 7.96-13.21%) at 12 months. In the multivariable Fine-Gray model, higher body mass index (BMI) (sHR 1.07; 95% CI 1.02-1.12; p = 0.009) and central nervous system (CNS) metastases (sHR 2.30; 95% CI 1.30-4.07; p = 0.004) were independently associated with VTE/ATE. Metastatic recurrence versus de novo stage IV showed no association with thrombosis (sHR 0.85; 95% CI 0.48-1.50; p = 0.573). Median OS was significantly shorter in patients with VTE/ATE (22.0 vs. 31.0 months; log-rank p = 0.049). Modeled as a time-dependent covariate, thrombosis significantly increased mortality hazard (HR 2.43; 95% CI 1.78-3.33; p < 0.001). Sensitivity analysis excluding events at tumor progression confirmed these associations (Cox HR 1.78; p = 0.002). Neoplastic progression accounted for 77.5% of total deaths and 64.9% of deaths in patients with thrombosis. CONCLUSIONS: The incidence of thromboembolic events during osimertinib treatment in routine practice was 15.0% and was associated with poorer overall survival. Elevated BMI and CNS metastases identify higher-risk patients, providing a rationale for evaluating individualized risk stratification in future prospective clinical trials.