Risk of venous and arterial thromboembolic events following CAR-T therapy in relapsed or refractory non-Hodgkin lymphoma: A propensity-matched multicenter analysis.

Abstract Background: Chimeric antigen receptor T-cell (CAR-T) therapy has transformed outcomes in relapsed or refractory non-Hodgkin lymphoma (NHL), but it may increase the risk of thromboembolic events (TEEs), including venous thromboembolism (VTE), myocardial infarction (MI), and stroke, through mechanisms such as cytokine-driven inflammation and endothelial injury. However, comparative data on the short-term incidence of these events versus similar NHL patients not receiving CAR-T remain limited. This study aimed to quantify and compare the 90-day risk of major arterial and venous TEEs in relapsed or refractory NHL patients treated with CAR-T versus matched non–CAR-T controls. Methods We used the TriNetX U.S. Collaborative Network (69 healthcare organizations) to identify adults (≥18 years) diagnosed between January 2017 and December 2024 with relapsed or refractory diffuse large B-cell lymphoma (DLBCL) or follicular lymphoma (FL). The CAR-T cohort included patients who received any FDA-approved CAR-T product (axi-cel, tisa-cel, liso-cel). The comparison cohort consisted of relapsed or refractory NHL patients without CAR-T exposure. Outcomes included: MI, VTE, ischemic stroke, and acute limb ischemia (ALI), were assessed within 90 days. Propensity score matching (1:1, nearest neighbor, caliper 0.1) was performed to balance baseline characteristics, including age, sex, race, hypertension, diabetes, hyperlipidemia, chronic ischemic heart disease, chronic kidney disease (all stages), atrial fibrillation, prior ischemic heart disease, prior thromboembolic events, and use of aspirin, warfarin, apixaban, clopidogrel, ticagrelor, or rivaroxaban before CAR-T cell therapy. Post-matching balance was evaluated using standardized differences (<0.1 considered acceptable). Risk differences (RD), risk ratios (RR), and hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated, and Kaplan–Meier curves with log-rank tests were used for time-to-event analysis. Results After propensity score matching, 2,401 patients were included in each cohort, achieving balance across all demographics, comorbidities, and baseline antiplatelet/anticoagulant therapies (standardized differences <0.1). Within 90 days of the index event, CAR-T recipients experienced significantly higher rates of several thromboembolic and ischemic events compared to matched non–CAR-T patients: Venous thromboembolism: composite of acute DVT of lower extremity, upper extremity thrombosis, and pulmonary embolism: 372/2,401 (15.5%) vs 251/2,401 (10.5%); RD 5.0% (95% CI 3.1–6.9); RR 1.48 (95% CI 1.28–1.72); OR 1.57 (95% CI 1.32–1.86); HR 1.44 (95% CI 1.23–1.69); p<0.001. VTE-free survival at 90 days: 84.2% vs 89.2% (log-rank p<0.001).Acute myocardial infarction: including initial and subsequent STEMI/NSTEMI: 51/2,401 (2.1%) vs 41/2,401 (1.7%); RD 0.4% (95% CI –0.4–1.2); RR 1.24 (95% CI 0.83–1.87); OR 1.25 (95% CI 0.83–1.89); HR 1.21 (95% CI 0.80–1.83); p=0.36.Ischemic stroke: cerebral infarction: 51/2,401 (2.1%) vs 33/2,401 (1.4%); RD 0.7% (95% CI –0.1–1.5); RR 1.52 (95% CI 0.98–2.36); OR 1.53 (95% CI 0.97–2.41); HR 1.46 (95% CI 0.94–2.26); p=0.09.Acute limb ischemia: 24/2,401 (1.0%) vs 10/2,401 (0.4%); RD 0.6% (95% CI 0.1–1.0); RR 2.55 (95% CI 1.29–5.05); OR 2.54 (95% CI 1.26–5.11); HR 2.42 (95% CI 1.22–4.81); p=0.01. ALI-free survival at 90 days: 98.9% vs 99.6% (log-rank p=0.01). Conclusion In this large propensity score–matched analysis of patients with relapsed or refractory NHL, CAR-T therapy was associated with a substantially higher 90-day risk of venous thromboembolism and acute limb ischemia in the early post–CAR-T period. These findings highlight the need for close clinical monitoring and the potential role of preventive strategies, such as the use of prophylactic anticoagulation in patients with a low risk of bleeding.