Venous Thromboembolism Following CAR-T Therapy in Lymphoma Patients - Disproportionality Analysis of the FDA Adverse Event Reporting System (FAERS) Database

Introduction: Venous thromboembolism (VTE) is a substantial complication of the clinical course of lymphoma patients. VTE impacts both morbidity and mortality of lymphoma patients prolonging and increasing treatment-related services and costs. Chimeric antigen receptor T-cell (CAR-T) therapy is extensively used across non-Hodgkin lymphoma (NHL) indications and has become a standard of care for relapsed/refractory settings in many types of NHL. While not frequently reported in CAR-T patients VTE can have substantial and life-threatening effects. In addition to the limited amount of data available regarding VTE events inconsistencies regarding the frequency also exist. Real-world data (RWD) is scarce in this field. Disproportionality analysis (DA) is a crucial tool in pharmacovigilance used for detecting potential safety signals in spontaneous reporting systems for adverse drug reactions (ADRs). We aimed to apply DA to assess the relative reporting rates of VTE as an adverse event in lymphoma patients receiving FDA approved CAR-T therapy.

Methods: DA of the FDA Adverse Event Reporting System (FAERS) database through Q1 2024) was performed using the LifeSphere Signal and Risk Management platform. For each US FDA-approved CAR-T therapy specifically indicated for Lymphoma (Kymriah, Yescarta, Tecartus, Breyanzi), disproportionality scores (EBGM-Empirical Bayesian Geometric Mean, and PRR-proportional reporting ratio) were generated. In the first analysis, MedDRA SMQ (Standardized MedDRA Queries): Embolic and thrombotic events, venous (narrow), was selected. In the subsequent analysis, MedDRA Preferred Term of Pulmonary embolism (PE), was chosen. The threshold criteria used was a lower bound of 90% confidence interval of EBGM (EB05)>1, and PRR>2 was used to determine statistical significance in these disproportionality criteria. In addition to this quantitative analysis, line listings were generated from the FAERS database for all the underlying cases with Embolic and thrombotic events, venous (SMQ) (Narrow), and reviewed by a Hematologist.

Results: In the extracted outputs based on the previously mentioned methodology, we identified 66 VTE events, of which 31 (47%) were PE. For Embolic and thrombotic events, venous (SMQ) (Narrow), EB05 scores were 0.775, 0.914, 0.331, and 0.513 for Kymriah, Yescarta, Tecartus, and Breyanzi, respectively. For the SMQ mentioned above, PRR scores of Kymriah, Yescarta, Tecartus, and Breyanzi were 0.836, 1.021, 0.537, and 1.189. These scores did not cross the standard statistical disproportionality threshold indicating no statistical association between embolic and thrombotic events, venous and these CAR-T therapies. For PE, EB05 scores were 0.708, 0.689, 0.112, and 0.206 for Kymriah, Yescarta, Tecartus, and Breyanzi respectively. For the abovementioned event, PRR scores of Kymriah, Yescarta, Tecartus, and Breyanzi were 0.78, 0.857, 0.276, and 0.813. These scores also did not cross the standard statistical disproportionality threshold, indicating no statistical association between PE and these CAR-T therapies.

Conclusion: DA did not demonstrate a significant association between Kymriah, Yescarta, Tecartus, Breyanzi and VTE events in lymphoma patients post CAR-T therapy. Patients who received Yescarta were close to the EB05 significance threshold for VTE events without reaching it. Besides negative results, we observed a substantial number of VTE events. 47% of observed VTE events were PE, higher than reported in data from clinical trials. Further investigation, including longer-term follow-up, is necessary to elucidate the implications of this potentially life-threatening condition and its association with CAR-T therapy.