Real-world evaluation of cardiac outcomes after CAR-T in patients with hematologic malignancies

Abstract

Abstract Background: Chimeric antigen receptor T-cell (CAR-T) therapy has transformed the treatment of relapsed or refractory hematologic malignancies. Retrospective studies suggest increased cardiac morbidity and major adverse cardiac events, but most are single-center, disease-specific, product-specific, and limited in follow-up. Thus, gaps remain in understanding cardiac outcomes after CAR-T in real-world settings and time-dependent patterns. Methods: TriNetX, a global health research network, provides access to real-world, HIPPA compliant, de-identified data from diverse healthcare systems. Using this platform, a retrospective cohort study was conducted to assess three major cardiac events: acute coronary syndromes (ACS), heart failure (HF), and arrhythmias. The study was not time-restricted and included adults (≥18 years) diagnosed with Non-Hodgkin Lymphoma, Multiple Myeloma, and Acute Lymphoblastic Leukemia who received FDA-approved CAR-T products during the course of their treatment . The primary outcome was incidence of the cardiac events. Secondary outcome was risk of cardiac events over time (Days 30, 100, 180, 365 and >365) . An exploratory subgroup analysis was conducted to identify potential predictors of risk. Data was analyzed using hazard ratio with 95% confidence interval. Cardiac events were assessed after excluding the first 14 days post-infusion to prevent other confounders in the acute post-infusion period. Propensity score matching was applied to ensure balanced cohorts by sex, age, ethnicity, weight, comorbidities, substance use, and prior chemotherapy. Results: A total of 4,972 patients were identified from 42 health care organizations. The cohort was predominantly male (61%), non-Hispanic (80.5%), and White (75%), with a mean age of 64 years. Diffuse large B-cell lymphoma (55%) was the most common diagnosis, followed by multiple myeloma (35%). Prior to CAR-T, 51% received ≥3 lines of chemotherapy, and 63% received targeted therapies. At baseline, 63% had hypertension, 19% had type 2 diabetes, and 29% had ischemic heart disease. The cumulative incidence of ACS, HF, and arrhythmia was 2.54%, 7.42%, and 9.77%, respectively. Risk of ACS (0.69%) and HF (6.64%) was highest between days 30–100, while arrhythmia risk peaked (3.49%) within the first 30 days. Cardiac event risk was lowest between days 100–180 (ACS 0.43%, HF 4.43%, arrythmias 1.47%). Following the first year, the incidence of cardiac events showed some increase with HF being highest at (4.84%), followed by arrythmias (2.94%) and ACS (0.53%). No significant differences in ACS or HF risk were observed when stratified by race, sex, anti-BCMA vs. anti-CD19 product, or history of cytokine release syndrome . However, HF (HR 1.50 p 0.0004) and arrythmia risk (HR 1.36 p=0.0026) was higher in patients over 65. Arrhythmia risk was also higher in males (HR 1.49 p=0.0002), and those receiving anti-CD19 therapies (HR 1.42 p=0.0015). Conclusions: These real-world data highlight time-dependent patterns of cardiac events following CAR-T therapy, underscoring the need for vigilant cardiac monitoring, particularly in older patients and those receiving anti-CD19 products. This study is limited by the inherent heterogeneity of registry data, including variability in data quality, completeness, and coding practices across sites. Additionally, findings may not be fully generalizable beyond the participating health systems. Further prospective studies are needed to confirm these findings and guide personalized management strategies.