Seven-Day Vein-to-Vein Point-of-Care Manufactured CD19 CAR T Cells (GLPG5101) in Relapsed/Refractory NHL: Results from the Phase 1 Atalanta-1 Trial

Background

Chimeric antigen receptor (CAR) T-cell therapies improve survival for patients (pts) with relapsed/refractory (R/R) non-Hodgkin lymphoma (NHL). Recent studies highlighted the importance of shortening time from leukapheresis to infusion to improve clinical outcomes (Chen et al., 2022; Locke et al., 2022). A novel decentralized and automated point-of-care (PoC) manufacturing model was developed to administer fresh autologous CAR-T treatments within 7 days of apheresis. This avoids the need for cryopreservation and complex logistics, and potentially makes CAR-T therapy accessible for pts with rapidly progressive disease. Here we provide an update on the Phase (Ph) 1 Atalanta-1 trial of PoC manufactured GLPG5101 in pts with R/R NHL.

Methods

Atalanta-1 (CTIS: 2022-502661-23-00) is a Ph1/2, multicenter trial of PoC-manufactured GLPG5101 in pts with R/R diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), marginal zone lymphoma (MZL) or mantle cell lymphoma (MCL). GLPG5101 is an anti-CD19/4-1BB CAR-T therapy, administered as a fresh product after fludarabine/cyclophosphamide lymphodepleting chemotherapy. Primary Ph1 objectives are safety and establishment of a recommended Ph2 dose. The primary Ph2 objective is efficacy (objective response rate [ORR]). Secondary objectives include feasibility of PoC manufacturing, safety, additional efficacy endpoints, and pharmacokinetics.

Results

As of May 2, 2023, 14 pts were enrolled in the Ph1 study at dose level (DL) 1 (50×10⁶ CAR+ T cells, n=7) or DL2 (110×10⁶ CAR+ T cells, n=7); no screen failures occurred. Pts were diagnosed with R/R MCL (n=3), DLBCL (n=7), FL (n=3), or MZL (n=1). Median age was 65 years (range 50–77), 11/14 pts were male. Median prior lines of therapy was 4 (range 1–8). GLPG5101 was manufactured for all pts and administered as a fresh infusion; 13 pts were infused within 7 days of leukapheresis, 1 within 8 days. Three pts received DL1 instead of the intended DL2 due to a lower CAR+ T-cell yield during manufacturing. GLPG5101 showed a preserved early memory phenotype for both CD4+ and CD8+ CAR T cells in the final product, compared to apheresis starting material (Figure).

Most treatment-emergent adverse events (TEAEs) were of Grade 1–2; the majority of Grade ≥3 events were hematological (Table). Seven (50%) pts experienced cytokine release syndrome (CRS), all but 1 of Grade 1–2; 1 patient had Grade 3 CRS (DL2). Neurotoxicity was seen in 6 (43%) pts, all Grade 1. Grade 3–4 infections occurred in 2 pts. Dose-limiting toxicities (DLTs) were reported for 4 (29%) pts (DL1, n=1; DL2, n=3). Three pts had Grade 4 neutropenia, manageable with granulocyte colony-stimulating factor; these did not limit dose escalation. One Grade 5 intra-abdominal hemorrhage occurred 12 days post infusion (DLT at DL2), caused by Grade 4 disseminated intravascular coagulation, in a pt with multiple comorbidities. One Grade 5 sepsis was recorded >6 months after infusion while the pt was in ongoing complete response (CR).

In the efficacy-evaluable pts (n=13), 11 responded to treatment (best ORR 85%). CRs were observed for 9 pts (CR rate 69%). Seven (54%) pts had an ongoing response at time of analysis, with duration reported up to 12 months post-infusion (median follow-up 4.5 months; range 1–12 months). Three pts progressed after an initial response, 1 pt with a CD19−negative relapse. ORR and CR rate were 86% and 57% at DL1, and 83% and 83% at DL2, respectively.

Robust CAR T-cell expansion was observed in all pts by qPCR with a median maximum expansion of 2.8×10⁵ copies/µg DNA (C_max). Median T_max was 14 days (range 7–28 days) across doses. The median AUC_0-28d was 3.1×10⁶ copies/µg DNA × days. GLPG5101 was detected in peripheral blood up to 9 months post-infusion.

Conclusion

Data from 14 pts with NHL enrolled in the Ph1 Atalanta-1 study demonstrate that PoC CAR-T manufacturing with a short vein-to-vein time is feasible. GLPG5101 is administered as a fresh product with a median vein-to-vein time of 7 days. The safety profile of GLPG5101 is comparable to other CAR-T therapies, and high CR rates across both DLs in a heavily pre-treated pt population are demonstrated. With this novel PoC manufacturing model, an early phenotype of less differentiated CAR T cells is preserved, demonstrating rapid GLPG5101 in vivo expansion and durable persistence post-infusion. The trial is ongoing, and updated data will be presented.