Safety and preliminary efficacy of anti-CD19 CAR T cells zorpocabtagene-autoleucel in systemic autoimmune diseases: Final analysis of the castle trial

Abstract Introduction: Autoimmune diseases (AID) such as systemic lupus erythematosus (SLE), systemic sclerosis (SSc) and idiopathic inflammatory myopathies (IIM) are characterized by the activation of pathological B-cells and the formation of autoantibodies that trigger damage to cells and organs. Based on previous findings from pilot studies we found that zorpocabtagene-autoleucel (zorpo-cel, MB-CART19.1), an autologous T-cell product expressing the chimeric antigen receptor (CAR) against CD19, induce deep B-cell depletion e.g., in peripheral blood and in tissues such as lymph nodes achieving drug-free remission in these AIDs. Methods: CASTLE is a Phase I/II basket trial that tested the feasibility, safety, and efficacy of zorpo-cel in patients with severe, treatment-resistant SLE, SSc, and IIM. The primary enpoint was safety defined as the rate of cytokine-release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) within the first 4 weeks after zorpo-cel administration, the secondary clinical efficacy endpoints were DORIS remission in SLE, no progression of interstitial lung disease in SSc, and ACR Major/Moderate Response in IIM after 24 weeks. Dose-limiting toxicities were defined as CRS or ICANS of grade 3 or higher, or grade three or higher organ toxicities (including late immune effector cell-associated hematotoxicity, ICA-HT lasting beyond day 28).. Lastly, local immune effector cell associated toxicity syndrome (LICATS) was recorded according to Hagen et al 2025, Lancet Rheumatology. Patients were enrolled in 2 cohorts: 8 patients for Phase I, 16 patients for Phase II with a safety review between Phase I and Phase II. Results: All planned 24 patients with SLE, IIM or SSc were enrolled between July 2023 and February 2025. Ten SLE, nine SSc and five IIM patients had a median disease duration of 4.0 [1.3; 6.5] years and failed a median of 4 [3; 6] immunosuppressive treatments. For all patients, zorpo-cel was successfully produced and infused. CAR T cells peaked at a median of 140 [82; 360] cells/µl after 10 days. B cell depletion was achieved within 7 days and B-cell aplasia lasted a median of 83 [56; 113] days. B cells were mostly naïve in recurrence for all patients. The 8 patients of phase 1 fulfilled predefined safety and efficacy endpoints and 16 additional patients were included in Phase 2. The study met its primary endpoint of safety. No CRS > grade 2 and no ICANS occurred, there was no ICAHT persisting beyond day 28. One patient developed a CMV pneumonitis during a virus reactivation. Kidney biopsy due to reduced organ function showed thrombotic microangiopathy, possibly related to CMV. Persisting reduction of creatinine-clearance resulted in the only DLT of the trial. Twenty-two of 24 patients achieved efficacy endpoints, with 9/10 SLE patients fulfilling DORIS remission, all 9 SSc patients showing no disease progression, and 4/5 IIM patients reaching ACR/EULAR major response after 6 months. One patient with longstanding IIM did not adequately improve in muscle strength and failed ACR Moderate/Major Response Criteria. One patient with SLE did not achieve significant reduction of proteinuria (< 500 mg protein/g creatinine) and therefore did not meet DORIS remission at 6 months. Immunosuppressive therapy could be stopped in all patients. Conclusion: The CASTLE study is the first phase I/II basket trial to demonstrate feasibility, favorable safety, and high efficacy of zorpo-cel in three B-cell driven AIDs. These promising results provide strong rationale to further develop zorpo-cel in AID.