Risk of second primary malignancy following chimeric antigen receptor T-cell therapy: A real-world cohort study

Abstract Background: Chimeric Antigen Receptor T-cell (CAR-T) therapy has significantly improved outcomes in relapsed or refractory diffuse large B-cell lymphoma (DLBCL). However, the long-term risk of second primary malignancies (SPMs) following CAR-T treatment remains insufficiently studied, particularly in real-world populations. Methods: We performed a retrospective cohort study using the TriNetX research network to identify adult patients (aged ≥18 years) diagnosed with DLBCL who received CAR-T therapy. Patients were divided into two cohorts based on CAR-T cell therapy exposure: those who received CAR-T therapy (CAR-T group) and those who did not receive CAR-T therapy (non-CAR-T group). To reduce confounding, propensity score matching was performed to balance baseline characteristics, including age at index, sex, race, prior chemotherapy and line of therapy. As cancer stage data were unavailable in the database, disease site distribution was balanced between groups and used as a proxy to estimate disease burden. The primary outcome was the incidence of SPMs occurring 3 years after CAR-T therapy, which included hematologic malignancies and solid tumors, within 3 years following CAR-T therapy. Result: Baseline characteristics were well balanced following propensity score matching, resulting in 1,480 patients in each cohort. The mean follow-up duration was 494.9 days in the CAR-T group and 632.2 days in the non-CAR-T group. The mean age at index was comparable between groups (62.3 ± 13.2 years vs. 62.1 ± 15.5 years). Males comprised 62.30% of the CAR-T group and 62.57% of the non-CAR-T group, while females accounted for 36.42% in both groups. The majority of patients were White (76.69% in the CAR-T group and 78.04% in the non-CAR-T group), followed by African American patients (6.96% vs. 7.23%) and Asian patients (3.78% vs. 2.57%). The CAR-T group showed a higher incidence of acute myelogenous leukemia (AML) (3.23% vs. 1.18%, p = 0.0002) and myelodysplastic syndrome (MDS) (2.93% vs. 1.11%, p = 0.0005). There were no significant differences between groups in rates of follicular lymphoma (5.13% vs. 4.69%, p = 0.6173), mantle cell lymphoma (0.94% vs. 0.69%, p = 0.4598), chronic lymphocytic leukemia (1.37% vs. 1.79%, p = 0.3773), acute lymphoblastic leukemia (0.90% vs. 1.56%, p = 0.1102), multiple myeloma (2.25% vs. 1.68%, p = 0.2709), and sézary syndrome (0.66% vs. 0.66%, p = 0.9905). However, the CAR-T group showed a lower incidence of Hodgkin lymphoma (2.01% vs. 3.30%, p = 0.0337) and mature T/NK cell lymphoma (0.90% vs. 2.65%, p = 0.0004). Among solid tumors, there were no significant differences in the incidence of basal cell carcinoma (0.68% vs. 0.69%, p = 0.9926), squamous cell carcinoma (0.68% vs. 1.03%, p = 0.3005), lung cancer (1.02% vs.1.4%, p = 0.3783), and prostate cancer (1.04% vs. 1.07%, p = 0.9446) between the two groups. Conclusion: We found that DLBCL patients treated with CAR-T therapy were associated with a significantly high risk of therapy-related myeloid malignancies, particularly AML and MDS. Alternatively, patients who received CAR-T therapy were associated with a reduced risk of Hodgkin lymphoma and mature T/NK-cell lymphoma. The findings from this study demonstrate the importance of improved long-term hematologic follow-up for all patients receiving CAR-T therapy to allow for monitoring for the earliest signs of secondary myeloid neoplasms. Future studies should continue to develop our understanding in this area and explore the mechanisms for both the increased and decreased risks of secondary malignancy in patients treated with CAR-T. In this way, we will be able to develop strategies for risk mitigation both pre- and post-CAR-T and best practices for patient selection with the hopes of expanding the potential indications for CAR-T therapy.