Risk of second primary hematological malignancy post chimeric antigen receptor T-cell therapy in diffuse large B-cell lymphoma: Propensity-matched analysis using trinetx database

Abstract Background: Chimeric antigen receptor T-cell (CAR-T) therapy has improved the outcomes for patients with relapsed or refractory diffuse large B-cell lymphoma (RR-DLBCL). However, new questions are arising over time about the long-term safety concerns of this therapy, including the risk of second primary malignancies (SPMs). Studies suggest that hematological malignancies are the most common type of SPM after CAR-T cell therapy, with MDS and AML accounting for the most common ones, and T-cell malignancies constituting a fraction of cases. Despite these findings, comprehensive real-world evidence on the incidence and distribution of hematologic SPM subtypes following CAR-T therapy remains limited. A clearer understanding of these risks is vital to inform long-term monitoring strategies and ensure the safe and effective use of CAR-T treatments. Methods: We performed a retrospective cohort study using the TriNetX database, a federated network of de-identified electronic medical records encompassing over 250 million patients. Adult patients diagnosed with diffuse large B-cell lymphoma (DLBCL) who received CAR-T cell therapy were compared to those who did not undergo CAR-T treatment. Key demographic variables, including age, sex, and race, along with comorbidities, were extracted for both groups. Comorbidity burden was categorized by organ systems. Propensity score matching (PSM) was used to reduce baseline differences between groups and mitigate confounding. The primary outcome was the incidence of second primary hematologic malignancies (SPMs). Results: A total of 1,871 patients with DLBCL received CAR-T cell therapy, while 116,603 DLBCL patients who did not receive CAR-T cell therapy were identified. Before propensity score matching, significant differences were observed in the demographic and clinical characteristics of the two cohorts. Significant disparities were found across racial and ethnic groups (p 0.001). Gender distribution also varied, with the CAR-T group comprising a higher percentage of males (60.7% vs. 54.8%, p 0.001) and a lower percentage of females (37.3% vs. 42.8%, p 0.001). Patients receiving CAR-T also had a higher burden of comorbidities across major systems, including circulatory (80.6% vs. 41.4%), endocrine/metabolic (80.1% vs. 40.6%), respiratory (62% vs. 31.7%), nervous (68% vs. 28%), and genitourinary (57.6% vs. 31%) diseases. These baseline differences were balanced after 1:1 propensity score matching, resulting in 1,864 patients in each group. During this period, patients in the CAR-T group exhibited a significantly higher incidence of myelodysplastic syndrome (3.83%), followed by acute myeloid leukemia (3.17%). No significant increase in other hematologic secondary primary malignancies subtypes was identified within the CAR-T cohort. Conclusions: In this large, real-world analysis, CAR-T cell therapy for patients with DLBCL was associated with a significantly increased risk of developing MDS and AML as second primary malignancies. These findings underscore the need for long-term hematologic surveillance in this patient population. Further prospective studies are needed to explore underlying mechanisms and to develop risk stratification strategies for post-CAR-T surveillance.