Chimeric antigen receptor T-cell therapy in patients with coexisting malignancy and autoimmune disease

Abstract Background: Chimeric antigen receptor (CART) therapies targeting CD19 or BCMA have shown great potential for treating autoimmune diseases (AD) in early-phase trials by depleting autoreactive B cells and plasma cells, resetting immune tolerance. While these findings suggest efficacy, current clinical evidence is limited to small, selected cohorts. We conducted a real-world analysis of patients (pts) with coexisting AD and relapsed/refractory (r/r) B-cell non-Hodgkin lymphoma (B-NHL) or multiple myeloma (MM) treated with commercial CART19 or BCMA-CART. Methods: We studied 601 pts with r/r B-NHL or MM treated with commercial CART19 or BCMA-CART from 01/2018 to 12/2024 ( data cutoff: June 2025) at our institution. Antitumor responses were evaluated by Lugano criteria (NHL) and IMWG criteria (MM), and adverse events by ASTCT guidelines. AD was defined as active in the presence of serological markers and need for therapy. AD responses were based on symptom improvement and reduced immunosuppression; flares were defined as new/worsening symptoms or increased treatment need. Serum autoantibody profiling (Chang; Nat Commun. 2021) was performed on paired samples from 90 pts (51 CART19, 39 BCMA-CART) at day 0 and month 3 (M3) post-CART. Results: Overall, 403 (67%) pts had B-NHL and 198 (33%) MM. Median follow-up was 26 months. Forty-nine pts (8.2%) had a history of AD (36 with B-NHL; 13 with MM). ADs included rheumatoid arthritis (RA, n=12), autoimmune hemolytic anemia (AIHA) or immune thrombocytopenic purpura (n=9), sarcoidosis (n=9), systemic lupus erythematosus (SLE) or undifferentiated connective tissue disease (n=7), autoimmune thyroid disease (n=4), and inflammatory bowel disease (IBD, n=4). Less common ADs included psoriasis (n=2) and one case each of type 1 diabetes, ankylosing spondylitis, Sjogren's syndrome, uveitis, polymyalgia rheumatica, and autoimmune colitis. Four pts had multiple ADs. AD was more common in females (AD: 57% vs 35%, p<0.01). Within B-NHL and MM cohorts, baseline characteristics (age, histology, CART product, prior lines, transplant history) were similar between AD and non-AD groups.We first evaluated antitumor efficacy and safety of CART19 and BCMA-CART in patients with and without AD. Response rates, progression-free survival, overall survival, and incidence and severity of cytokine release syndrome and neurotoxicity were comparable (p>0.05).We then assessed the impact of CART on AD. Among 49 pts with AD, 2 B-NHL patients (4%)—1 with RA (RA#1) and 1 with SLE (SLE#1)—had active AD at CART19 infusion. Both were symptomatic and on treatment at CART infusion. RA#1 remained in AD remission until death from lymphoma progression 2.5 years post-CART. SLE#1 had a baseline ANA titer of 1:20,000, which decreased to 1:2,500 at M3 post-CART, accompanied by reduced DNA-associated antibodies (anti-histones, KU, P70/80, H2B, H2A/4, H3, H1, nucleolin; median fold change [FC] −70%, range −39% to −84%). In contrast, anti-Smith remained stable (FC −6%) while anti-SSB increased (FC +39%), suggesting selective depletion of pathogenic B-cell clones. SLE#1 experienced SLE relapse 2.3 years post-CART with isolated cutaneous involvement, while in remission of B-NHL.We then evaluated the 47 pts with AD in remission at the time of CART infusion. While 41 (87%) pts maintained AD remission, 6 (13%) pts (3 RA, 1 psoriasis, 1 IBD, and 1 AIHA) experienced AD flares within 3 months post-CART. All resolved with appropriate therapy: corticosteroids (n=2,RA), non-steroidal anti-inflammatory (n=1, RA), topical steroids plus phototherapy (psoriasis), mesalamine (IBD), or red blood cell transfusions (AIHA). To assess the broader impact of CART19 and BCMA-CART on the pt autoreactome (individual-specific autoantibody repertoire; Bodansky; JCI 2024), independently of AD diagnosis, we profiled 90 pts using a bead-based array targeting 52 autoantigens (Chang; Nat Commun. 2021). Both therapies modulated the autoreactome, with BCMA-CART inducing a significantly greater median reduction in autoantibodies (72% vs. 31%, p<0.001), supporting its broader potential in autoantibody-driven diseases. Conclusion: CART confirms durable AD control in real-world pts with coexisting NHL/MM. However, a subset of pts with AD in remission at infusion may experience post-CART AD flares, possibly due to the post-treatment inflammatory state, warranting close monitoring. Further studies are needed to understand flares' pathogenesis and their link with CART activity.