Browse the corpus

To the Editor: Recent reports of T-cell lymphomas/leukemias after chimeric antigen receptor T-cell (CAR-T) therapy for relapsed/refractory B-cell lymphomas/leukemias have garnered attention as a potential CAR-T therapy-related adverse effect.1,2 However, it has long been known that B-cell lymphoma/leukemia patients have increased risk of subsequent malignancies, including T-cell lymphomas/leukemias, related to treatments, immunosuppression, and other risk factors.3–5 Whether T-cell lymphomas/leukemias following CAR-T therapy for B-cell lymphomas/leukemias occur more often than expected is an unanswered but critical question because of disease- and prior treatment-related risk factors and the rarity of T-cell lymphomas/leukemias. To better understand baseline T-cell lymphoma/leukemia risks after B-cell lymphoma/leukemia in the pre-CAR-T era, we quantified incidence rates per 100,000 persons (IRs), standardized incidence ratios (SIRs), and excess absolute risks per 100,000 person-years (EARs) of T-cell lymphomas/leukemias in B-cell lymphoma/leukemia patients and an age-, sex-, and race and ethnicity-matched general population sample from 17 US population-based cancer registry data (2000–2017). IRs were calculated as the observed (B-cell lymphoma/leukemia survivors) or expected (general population) number of cases divided by the total person-years at risk.

leukemia patients and an age-, sex-, and race and ethnicity-matched general population sample from 17 US population-based cancer registry data (2000–2017). IRs were calculated as the observed (B-cell lymphoma/leukemia survivors) or expected (general population) number of cases divided by the total person-years at risk. Among 235,665 B-cell lymphoma/leukemia patients treated with initial chemo/immunotherapy, we identified 298 subsequent T-cell lymphomas/leukemias (IR=22.7). In the general population, 62.4 cases (IR=4.7) were expected, representing an SIR=4.8 and EAR=18 cases. IRs, SIRs, and EARs were highly variable by time since B-cell lymphoma/leukemia diagnosis, age at diagnosis, and T-cell lymphoma/leukemia subtype (Figure 1). Despite including 26.5% of the US population over an 18-year period, the rarity of T-cell lymphomas/leukemias precluded IR and SIR estimation when stratifying by B- and T-cell lymphoma/leukemia subtypes simultaneously; however, overall T-cell lymphoma/leukemia risks also varied substantially by B-cell lymphoma/leukemia subtype. These population-based data emphasize the difficulty of accurately quantifying an “expected” T-cell lymphomas/leukemias rate for patients treated with CAR-T therapy. First, general population T-cell lymphoma/leukemia rates are not appropriate for B-cell lymphoma/leukemia survivors because they have elevated risk in the pre-CAR-T era. Second, T-cell lymphomas/leukemias are too rare to reliably estimate expected rates after B-cell lymphoma/leukemia.

tients treated with CAR-T therapy. First, general population T-cell lymphoma/leukemia rates are not appropriate for B-cell lymphoma/leukemia survivors because they have elevated risk in the pre-CAR-T era. Second, T-cell lymphomas/leukemias are too rare to reliably estimate expected rates after B-cell lymphoma/leukemia. Patients treated with CAR-T therapy should continue to be encouraged to participate in registries collecting detailed treatment information and patient outcomes so that future studies can identify factors contributing to T-cell lymphoma/leukemia risk following B-cell lymphoma/leukemia. Biologic specimen collection will enable detailed molecular studies to potentially detect the CAR transgene, quantify allele frequency, and assess viral integration to further assess causality. The importance of clarifying CAR-T therapy-related adverse effects is further emphasized as CAR-T therapies move towards earlier lines of treatment and indications expand to non-malignant disorders, possibly altering the risk-benefit balance.