Risk Factors for Severe Infection in Patients Receiving Bispecific Antibody Therapies for Lymphoma

Introduction: Several CD20/ CD3 bispecific antibodies (BsAb) have been approved for relapsed/refractory DLBCL and FL. As research is ongoing to move these effective therapies earlier in the treatment landscape, it is critical to understand the real-world infection risks. Our study sought to examine risk factors for infection in patients with B-cell non-Hodgkin's lymphoma treated with commercial or investigational BsAb. Methods: We conducted a retrospective, single center cohort study of adult patients with B-cell NHL treated at the University of Michigan between September 2018 and April 2024. Patients received BsAb as single agent or combination therapy. The primary endpoint was risk factors for severe infection defined as any infection grade 3 or higher using CTCAE v5. We evaluated characteristics of infection including time to infection, time from last treatment of bendamustine, autologous SCT, or CAR-T to infection, and outcomes after infection. A univariate analysis and multivariate logistic regression analysis was performed to evaluate risk factors for infection. Results: 84 patients were included in this study who were treated with BsAb therapy. The median age was 69 (range 62-78). Patients had DLBCL (n=48), FL (n=17), MCL (n=15), and Richter's transformation-DLBCL (n=4). 86% of patients included had an ECOG of 0-1. 74 patients received BsAb for R/R disease with 3.5 (range 3-5) median lines of therapy. 57 R/R patients received BsAb as monotherapy. 17 R/R patients received BsAb as combination therapy with Polatuzumab-CHP (n=5), R-DHAOx (n=3), GemOx (n=3), lenalidomide (n=5), or RCHOP (n=1). 10 patients received BsAb as part of frontline therapy, of which 9 patients received BsAb in combination with RCHOP (n=5), mini-RCHOP (n=2), BR (n=1), or Polatuzumab-CHP (n=1). The median duration of BsAb treatment was 84 days (range 31-177). Patients were treated with epcoritamab (n=39), glofitamab (n=11), mosunetuzumab (n=16), and plamotamab (n=18). The incidence of grade 3 or higher infection was 23%, 45%, 44% and 44% in patients receiving epcoritamab, glofitamab, mosunetuzumab, and plamotamab respectively. 37 (44%) patients received prophylaxis for pneumocystis jiroveci pneumonia (PJP), and 62 (74%) patients received viral prophylaxis. 47 (56%) patients developed any grade infection, of which 29 (62%) patients had severe infection. Most common types of infection were bacterial (45%) and viral (28%). Most common infections were pneumonia (23%), upper respiratory tract infection (21%), or urinary tract infection (21%). 30 (64%) patents had an identifiable organism. 3 patients had PJP infection of which 1 patient was on prophylaxis. 2 patients had varicella zoster virus of which 1 patient was on prophylaxis. Of the 29 patients with severe infection, 19 (66%) patients had prior CRS of which 42% received steroids beyond step-up dosing in cycle 1. 24 (51%) patients with any infection required hospitalization and 22 (47%) patients had recurrent infections. 22 (47%) patients required treatment hold due to infection, however only 7 (15) patients required treatment discontinuation. 4 (9%) patients died from infection due to COVID 19 (n=1), bacteremia (n=1), and bacterial pneumonia (n=2). Time from treatment to infection was 36 days (range 12-84). There was no significant difference in grade 3 or higher infection when evaluating for any variable including histology, prophylaxis, prior treatment, or patients receiving BsAb therapy on a clinical trial. The median duration of treatment in patients with and without any grade infection was 105 days (range 42-287) and 70 days (range 20-108) respectively (P=0.027). A univariate analysis showed a trend toward increased risk of severe infection in patients receiving plamotamab and patients with ECOG of 2, however there was no statistical significance in a multivariate analysis. Conclusion: Our study is largest real-world cohort to report on infection rates in patients treated with CD20/CD3 BsAb for B-cell NHL. Our study confirms that patients receiving BsAb have a high risk of severe infections, even in patients treated in the frontline setting or while on prophylaxis. Multivariate analysis showed no difference in risk of severe infection with regards to combination therapy, histology, product, or previous treatment and further research with larger cohorts is warranted to fully understand risk factors for infection in this high risk patient population.