2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Sunday, December 13, 2026
09:30 a.m. - 11:00 a.m.
MCCNO - 217-219
Treatment Considerations, Biological Therapies, Chimeric Antigen Receptor, (CAR)-T Cell Therapies, Non-Biological Therapies, Anticoagulant Drugs, Chemotherapy, Adverse Events, Biological Processes, CHIP, Diseases, Bleeding and Clotting, Thromboembolism, Thrombotic Disorders, Lymphoid Malignancies, Lymphoid Leukemias, Lymphomas, ALL, Research, Clinical Research, Clinical Trials, Epidemiology, Health Outcomes Research, Real-World Evidence, Registries, Study Population, Human, Adult, Pediatric
PAPER

476 Risk of venous thromboembolism and bleeding following CAR-T cell therapy: A multicenter analysis using epic cosmos

Katia Luciani, MD, Omid Jafari, Senior Software Engineering and Programming Associate2, Danielle Guffey3, Zihan Yang, MS4, Avi Leader, MD5, Marc Carrier Jr, MD6, Natasha Kekre, MD7, Ang Li, MD, MS8
1 University of Ottawa, Ottawa, Ontario, Canada
2 Baylor College of Medicine, Houston, Texas, United States
3 Baylor College of Medicine, Houston, Texas, United States
4 University of Texas Health Science Center at Houston, Houston, Texas, United States
5 Memorial Sloan Kettering Cancer Center, New York, New York, United States
6 Ottawa Hospital Research Institute, Ottawa, Ontario, Canada, University of Ottawa, Ottawa, Ontario, Canada
7 Ottawa Hospital Research Institute, Ottawa, Ontario, Canada, University of Ottawa, Ottawa, Ontario, Canada
8 Baylor College of Medicine, Houston, Texas, United States

Background

Venous thromboembolism (VTE) and clinically relevant bleeding (CRB) are recognized complications after chimeric antigen receptor T-cell (CAR-T) therapy. While single-center studies suggest high risk of VTE and CRB after CAR-T, multicenter estimates accounting for prior VTE and cytokine release syndrome (CRS) remain limited.

Methods

Data were obtained from Epic Cosmos, a collaborative dataset of Epic health systems including more than 300 million patients from all 50 states, D.C., and non-U.S. sites (cosmos.epic.com). Adults receiving first CD19- or BCMA-directed CAR-T therapy for hematologic malignancies from 2021-2024 in the U.S. were identified. After excluding therapeutic anticoagulation before CAR-T (n=976) and acute VTE diagnosis in the last 30 days, the final analytic cohort had 3,624 patients.

VTE and CRB after CAR-T infusion date were defined using previously validated ICD-10 diagnosis codes, and the diagnosis dates were anchored by the POA flag and CPT codes. Cumulative incidence of VTE and CRB was estimated with death as a competing risk. Multivariable cause-specific Cox models were built to evaluate factors associated with early VTE or CRB onset. CRS onset/severity was treated as a time-varying covariate. Baseline enoxaparin prophylaxis (PPX) was defined using a -2 to +1-day window from CAR-T infusion.

Results

Among 3,624 patients, the median age was 62 years; 38.4% were female and 72.8% were white. There were 1,804 diffuse large B-cell lymphoma (DLBCL), 274 mantle cell lymphoma (MCL), 274 follicular lymphoma, 1,064 multiple myeloma (MM), and 208 acute lymphoblastic leukemia (ALL). Axicabtagene was the most used CAR-T product (36.6%); cyclophosphamide/fludarabine lymphodepletion and inpatient administration were each observed in 76.6% of patients. Prior VTE (from -31d to -365d) occurred in 3.7% and recent myocardial infarction (MI) occurred in 1.5% of patients. Enoxaparin PPX was given in 1,195 patients (33%). Median time to CRS was 4 days (IQR 2-6), with 807 Grade 2 and 152 Grade 3-4.

At 30- and 90-days following CAR-T infusion, the cumulative incidence of VTE was 2.5% (2.1-3.9) and 4.2% (3.5-4.9), while the incidence of CRB was 2.6% (2.1-3.2) and 4.8% (4.1-5.5), respectively. Over half of VTE events (54.8%) were upper extremity thrombosis and 31.7% of CRB were gastrointestinal bleed. VTE incidence at 90-day was highest in MCL (6.7%; 4.1-10.2) and DLBCL (4.8%; 3.9-5.9), while CRB was highest in ALL (10.4%; 6.7-15.1%) and MCL (7.8%; 5.0-11.5). No meaningful differences in VTE were observed across CAR-T products after cancer type stratification. Among DLBCL, lisocabtagene (2.9%; 1.7-4.6) and tisagenlecleucel (2.6%; 0.9-6.2) had lower 90-day CRB than axicabtagene (6.6%; 5.2-8.2).

In multivariable analysis, CRS onset/severity was associated with impaired hemostasis. The corresponding HR for VTE was 1.35 (0.92-1.98) for grade 2 and 2.14 (1.09-4.19) for grade 3-4 vs grade 0-1, and for CRB was 1.77 (1.25-2.50) for grade 2 and 2.80 (1.63-4.83) for grade 3-4 vs grade 0-1. Other factors significantly associated with VTE included cancer type (HR 0.58; 0.38-0.90 for MM vs DLBCL), VTE history (HR 3.38; 1.97-5.83), MI history (HR 2.78; 1.18-6.58), and albumin <3.5 g/dL (HR 1.57; 1.03-2.38). Enoxaparin PPX was marginally protective for VTE (HR 0.69; 0.47-1.01). Other factors significantly associated with CRB included cancer type (HR 0.50; 0.33-0.75 for MM vs DLBCL), MI history (2.84; 1.27-6.37), inpatient CAR-T (HR 1.60; 1.00-2.56), hemoglobin <10 mg/dL (2.32; 1.63-3.31), platelet <50 (2.97; 1.98-4.44), and albumin <3.5 g/dL (1.83; 1.28-2.60). Enoxaparin PPX was not associated with increased CRB (0.84; 0.57-1.22).

Conclusion

In this multicenter cohort, VTE and CRB occurred in 4.2% and 4.8% of patients, respectively, within 90 days after CAR-T therapy. Notably, 23% of patients were excluded from the analysis for existing anticoagulation or recent VTE diagnosis, highlighting the high burden of VTE related to refractory hematologic malignancy independent of the CAR-T therapy itself. Severe CRS was associated with both thrombotic and bleeding complications, supporting the central role of CAR-T-associated inflammation in vascular toxicity. Prior VTE identified patients at increased risk of recurrent VTE, whereas thrombocytopenia was the strongest predictor of CRB. Enoxaparin PPX was marginally protective for VTE without increased risk of CRB. These findings support risk-stratified PPX and monitoring strategies after CAR-T.

 

Katia Luciani, MD: No relevant conflicts of interest to declare.