CD19 CAR T-cell therapy (GLPG5101) for relapsed or refractory B-cell non-Hodgkin lymphoma (ATALANTA-1): phase 1 results from a single-arm, multicentre, phase 1/2 study.

Chimeric antigen receptor (CAR) T-cell therapies have transformed the treatment of B-cell non-Hodgkin lymphoma, but centralised manufacturing-with its complex logistics and long vein-to-vein times-drives up costs and restricts access. We aimed to evaluate the safety of GLPG5101, a fresh, CD19 CAR T-cell product manufactured through a decentralised process, and determine the recommended phase 2 dose. This phase 1 dose-escalation part of the ATALANTA-1 phase 1/2, single-arm study, was executed in five hospitals in the Netherlands and Belgium. Patients aged 18 years or older, with histologically confirmed diffuse large B-cell lymphoma (DLBCL), follicular lymphoma, marginal zone lymphoma (MZL), or mantle cell lymphoma (MCL) after two or more lines of therapy, measurable disease according to the Lugano classification, Eastern Cooperative Oncology Group Performance Status 0-2, and adequate organ function were enrolled. Patients were treated with a single intravenous infusion at one of three dose levels of GLPG5101 (dose level 1 [35-50 × 106 viable CAR+ T-cells], dose level 2 [85-110 × 106 viable CAR+ T-cells], and dose level 3 [200-250 × 106 viable CAR+ T-cells]). The phase 1 part of the study applied a Bayesian Optimal Interval design and the primary endpoints were safety (incidence of adverse events and serious adverse events until end of treatment [week 14], and dose-limiting toxicities [DLTs] until day 28) in patients who received fresh GLPG5101 at any dose (excluding recipients of non-conforming product [not meeting prespecified release criteria other than dose]) and determination of the recommended phase 2 dose. The study was registered with ClinicalTrials.gov (NCT06561425) and is closed for recruitment. From March 15, 2022, to Sept 10, 2024, 27 patients were screened for eligibility, 24 of whom were enrolled, underwent leukapheresis and lymphodepleting chemotherapy and received GLPG5101. Data cutoff was June 1, 2025. In the intention-to-treat population (n=24), the median age was 66·5 years (IQR 59·0-71·5), 15 (63%) patients were male, nine (38%) were female, and 21 (88%) were White. One of the 24 patients received a non-conforming product and was excluded from the safety analysis population. Median follow-up was 24·0 months (IQR 20·7-24·9). Five DLTs were reported: grade 3 thrombocytopenia (n=1, dose level 1), death from intra-abdominal haemorrhage (n=1, dose level 2), and grade 4 prolonged neutropenia not resolving to grade 2 or lower within 28 days (n=1 at dose level 1 and n=2 at dose level 2). All 23 patients in the safety analysis population had grade 3 or higher treatment-emergent adverse events; the most common were neutropenia (22 [96%]), leukopenia (nine [39%]), lymphopenia (seven [30%]), anaemia (six [26%]), and thrombocytopenia (five [22%]). Treatment-related deaths occurred in four patients: one during the 14-week treatment period due to intra-abdominal haemorrhage, and three after the treatment period due to Escherichia coli sepsis (n=1), immune-effector cell-associated haemophagocytic lymphohistiocytic syndrome (n=1), and COVID-19 (n=1). The safety review committee selected 110 × 106 (range 50-110 × 106) viable CAR T-cells as the recommended phase 2 dose. The results of this study showed the feasibility of rapid multicentre decentralised manufacturing and delivery of fresh CAR T-cell therapy in heavily pretreated B-cell non-Hodgkin lymphoma. The phase 2 part of ATALANTA-1 will provide further information on clinical activity and safety. CellPoint, Lakefront Biotherapeutics.