Anti-CD19 chimeric antigen receptor T-cell therapy for patients with non-Hodgkin's lymphoma and concurrent autoimmune disease

Abstract

Abstract Introduction: CD19 targeted chimeric antigen receptor T cell therapy (CART) is standard of care (SOC) treatment (tx) for relapsed/refractory (r/r) B-cell non-Hodgkin's Lymphoma (NHL). Patients (pts) with r/r NHL and concurrent autoimmune diseases (AID) are often excluded from clinical trials, and the safety and efficacy of CART for these pts has not been established. Additionally, how B-cell depletion from CART affects clinical outcomes of AID has not been well described. Here we evaluated clinical outcomes related to both NHL and AID in pts who received CART for r/r NHL. Methods: We retrospectively evaluated pts treated with SOC CART for NHL at The Ohio State University (OSU) between 2017-2024. AID was defined as non-endocrine neurologic or rheumatic AID requiring FDA-indicated treatment. Fisher's exact tests and Wilcoxon rank-sum tests were used for group comparisons. Overall survival (OS) and progression free survival (PFS) were calculated via Kaplan Meier methods. Results: 290 pts were evaluated; 266 (91.7%) had no AID while 24 (8.3%) had AID. Median follow-up was 18.2 months (range: 0.4-88.1). Most baseline pt and disease (dz) characteristics did not significantly differ between non-AD and AID cohorts including age, race, dz subtype, CART product, dz bulk, prior lines, ECOG, stage, IPI, baseline lactate dehydrogenase or ferritin. More AID pts were female (70.8% vs 36.5%; p=.02). Response rates and survival outcomes did not differ. Overall response rate (ORR) in non-AID pts was 65.5% vs 66.6%. Complete response (CR) rate in non-AID pts was 49.3% vs 58.3% in the AID group (p=0.52). Median PFS from CART infusion in the non-AID vs AID cohorts was 7.6 months vs 8.9 months (p=0.95); median OS was 32.1 months vs 36.8 months (p=0.93). Peak LDH, ferritin, CRP, and peak absolute lymphocyte count post CART did not differ. There were no differences in rates of cytokine release syndrome (CRS) or immune effector cell associated neurotoxicity syndrome (ICANS). 73.2% of pts had CRS in the non-AID group vs 70.8%; the rate of Grade ≥ 3 CRS did not differ (10.3% vs 5.9%; p= 0.9). In the non-AID group, 36.5% of pts had ICANS vs 33.3%. For pts who had ICANS, the rate of Grade ≥ 2 ICANS was lower in the AID group at 37.5% (3/8 pts) vs. 78.4% (76/97) (p=.02). Severity/duration of cytopenias did not differ. AID included rheumatoid arthritis (RA) (n=9 pts), ankylosing spondylitis (1), Sjögren's (2), myasthenia gravis (1), dermatomyositis (1), psoriasis (5), and inflammatory bowel dz (IBD) (6). Of 8 pts with available data; 7/8 had seropositive AID. Of 20 pts with AID tx data, 18 required dz modifying anti-rheumatic agents (DMARDs) prior to CART. 9 required conventional DMARDs (ex. methotrexate, hydroxychloroquine (HCQ), azathioprine); 9 required biologic/targeted DMARD (ex. adalimumab, etanercept, risankizumab, infliximab, rituximab). DMARDs were weaned prior to CART however 9 pts (37.5%) remained on AID tx at the time of cell collection, including HCQ (n=4), sulfasalazine/mesalamine (n=2), and 1 each of prednisone, colestipol, sulfasalazine, and apremilast. Of 4 pts on HCQ at the time of collection, 3/4 achieved CR though 1 CR pt later relapsed. 4 pts (16.7%) remained on AID tx at CART infusion, including one each of apremilast, colestipol, prednisone, and sulfasalazine. 1 pt continued apremilast and 1 continued colestipol through CART and long-term. No pt had flares during tapering prior to CART which required re-initiation of tx. Of 20 pts off AID tx at time of CART infusion, only 2 pts restarted AID tx post-CART (both HCQ; only 1 for symptoms). The only new tx started post-CART was intermittent steroids for psoriasis and no biologics were required. Post-CART seropositivity data was unavailable at time of study. Median follow-up for AID cohort was 17.8 months [1.2-79.9].Conclusions: In this single center study of pts receiving CART for NHL we found no difference in response rates nor survival outcomes in pts with or without AID and therefore propose that AID should not preclude the use of CART. Lower severity of ICANS was seen in AID pts, though additional study with larger cohorts is required. The majority of pts were able to taper off AID tx prior to CART, while HCQ was able to be safely continued throughout CART collection with no apparent effect on CART efficacy. Our study also highlights the potential effect of CD19 directed CART on pt's AID given the majority of patients remained off AID tx for an extended time post-CART.