Incidence of second primary malignancies (SPMs) following bispecific antibody (BsAb) therapy for lymphoid malignancies: A descriptive analysis of clinical trials

Abstract INTRODUCTION Bispecific antibodies (BsAbs) have emerged as transformative therapies for hematologic malignancies, with several agents now approved for relapsed/refractory acute lymphoblastic leukemia (ALL), multiple myeloma (MM), and non-Hodgkin lymphoma (NHL). These therapies have demonstrated remarkable overall response rates across multiple clinical trials, with some now being evaluated in frontline settings. Recent publications have raised concerns about SPMs following CAR-T cell therapy, with the FDA requiring boxed warnings after reports indicated SPMs rates of up to 5%. However, the incidence of SPMs with BsAbs remains poorly characterized, despite shared mechanisms of T-cell activation. METHODS We conducted a systematic literature search via PubMed to identify peer-reviewed publications reporting clinical trials of FDA-approved BsAbs for hematological malignancies. Included agents were blinatumomab (CD19×CD3), glofitamab, mosunetuzumab, epcoritamab, and odronextamab (all CD20×CD3), as well as elranatamab, teclistamab, and linvoseltamab (BCMA×CD3), and talquetamab (GPRC5DxCD3). Eligible studies included phase I-III trials evaluating BsAbs either as monotherapy or combination therapy, irrespective of prior lines of treatment. Studies published only as abstracts were excluded due to insufficient adverse event reporting. Trials were screened for documentation of SPMs, both from main and supplemental publications. Trials were categorized based on SPMs reporting: “N/A” (insufficient data), “No” (explicit acknowledgment of no SPMs or comprehensive AEs reports with no malignancy entries), and “Yes” (direct reporting of SPMs). Only patients who received BsAbs were included in pooled analyses; comparator or control arms from randomized trials were excluded. For trials reporting patient numbers, median follow-up and SPM counts, we calculated crude incidence rates and cumulative incidence per 100 person-years using estimated total person-years. RESULTS We identified 19 clinical trials encompassing 5,133 patients across ALL (n=5 trials), MM (n=4), and NHL (n=10, including DLBCL and follicular lymphoma). Phase distribution included 5 phase 3 trials and 14 phase 1/2 trials. Most trials (n=16, 84%) evaluated BsAbs as monotherapy. The median number of patients enrolled per trial was 145 (range 54-718; mean 193), with a median follow-up of 14.3 months (range 4.2-43 months; mean 18.3). Four trials (21%) had insufficient adverse event data and were excluded from the analysis. The final pooled cohort analysis included 3,215 patients with available long-term safety data and a cumulative follow-up of 6,270 person-years. Across this cohort, 5 SPMs were reported, corresponding to a crude incidence of 0.16% and a cumulative incidence rate of 0.08 per 100-person years. These included 3 cases reported under “neoplasms benign, malignant and unspecified” in the TOWER trial (1 extramedullary leukemic infiltration, 1 pulmonary leukemic infiltration and 1 acute lymphocytic leukemia), along with 1 lung malignancy and 1 case of Hodgkin's lymphoma both from mosunetuzumab studies. CONCLUSION This analysis suggests a low incidence of SPMs with BsAb therapy compared to CAR-T therapy and appears consistent with expected baseline rates in heavily pretreated hematologic malignancy populations. Notably, some reported cases may represent disease progression rather than true SPMs. However, limited follow-up duration across trials warrants continued long-term surveillance, particularly as agents move to frontline settings where longer survival may reveal delayed malignancies.