A Phase I Study of CD19 CAR T-Cells with Escalating Doses of Lymphodepletion with or without Rituximab in Relapsed/Refractory Diffuse Large B-Cell Lymphoma

Introduction: CD19 chimeric antigen receptor (CAR) T-cells have transformed the landscape of treatment for relapsed/refractory (R/R) diffuse large b-cell lymphoma (DLBCL); however, only ~40% of patients have durable responses (Neelapu 2023). Essential to a favorable environment for CAR T-cells to expand, exert anti-tumor activity, and persist is lymphodepleting chemotherapy. Cyclophosphamide and fludarabine is a frequently used combination of lymphodepletion; however, the dosing of these agents varies and there is limited data prospectively comparing these doses with efficacy and safety. Rituximab has also demonstrated synergism with CAR T-cells in pre-clinical models and is also believed to mediate peripheral B-cell depletion, facilitating better tumor targeting by CARs, but it’s role in lymphodepletion is not well described. In addition, timely access to this potentially curable therapy remains a challenge. As such, point of care manufacturing with rapid vein-to-vein time is critical to expanding access to CAR T-cell therapy.

Methods: This is a phase I, single-center 3+3 dose escalation study of locally manufactured CD19 CAR T-cells at our university’s GMP facility. The primary objectives include feasibility of manufacturing CD19 CAR T-cells, safety of escalating doses of lymphodepleting chemotherapy with or without rituximab, and safety of CD19 CAR T-cells in subjects with R/R DLBCL. Overall response rate (ORR), complete response (CR), partial response (PR), progression free survival (PFS), and overall survival (OS) are secondary objectives. This is an autologous CD19-directed CAR created with a lentiviral vector (Miltenyi Biotec), a 4-1BB costimulatory domain, and a CD3 zeta intracellular domain. Patients with R/R DLBCL after > 2 prior lines of therapy are eligible. Subjects receive escalating doses of lymphodepleting chemotherapy with cyclophosphamide and fludarabine with or without rituximab in 6 dose levels (DL). The fludarabine dose remains constant at 25 mg/m² Days -5 to -3, and the cyclophosphamide dose escalates from 500 mg/m2 x 3 doses in DL 1 and 2, to 60 mg/kg x 1 dose in DL 3 and 4, and to 60 mg/kg x 2 doses in DL 5 and 6. Rituximab is given in DL 2, 4, and 6. A CAR T-cell dose of 1x10⁶ cells/kg is given in each DL. The CAR T-cells were initially harvested at Day 12 then thawed from a cryopreserved product; they are currently given as a fresh infusion after an 8-day manufacturing period.

Results: At time of data cutoff, 20 subjects enrolled and 15 are evaluable: 3 on DL 1, 4 on DL 2, 5 on DL 3, and 3 on DL 4. 6 patients have received fresh cell infusion since the manufacturing time was shortened. 11 (73%) subjects are female; median age is 63 (range 28-77). 9 subjects (60%) had primary refractory disease; 5 (33.3%) relapsed > 12 months from initial therapy and 1 (6.7%) relapsed < 12 months from initial therapy. 2 subjects had double hit lymphoma, 4 were double expressors, and 3 transformed from follicular lymphoma. 5 subjects have GCB immunophenotype, 8 have non-GCB, and 2 are unknown. Stage of disease includes 3 subjects with stage 2, 2 with stage 3, and 10 with stage IV. The median number of prior lines of therapy is 3 (range 2-8). 5 subjects had prior autologous stem cell transplantation (SCT) and 1 had prior allogeneic SCT.

Day +30 PET shows ORR 86% (13/15), CR 46.7% (7/15), and PR 40% (6/15). On Day +90 PET, 3 additional subjects converted to CR (10/15; 67.7%); 2 subjects received no additional therapy and 1 started lenalidomide between Day +30 and +90. The median follow-up time is 9 months. At time of data cutoff (30 months; July 21, 2024), PFS is 67.7% (10/15) and OS 73.3% (11/15). 4 patients have expired: 3 from disease progression and 1 from infection. 6 subjects (40%) experienced CRS, with 4 (26.7%) Grade 1 events in DL 2, 3, and 4, and 2 (13.3%) Grade 2 events in DL 2 and 3. 1 subject (6.7%) had Grade 2 ICANS in DL 3. There have been no Grade 3 or 4 CRS or ICANS events. There are no dose limiting toxicities to date.

Conclusions: These results demonstrate the safety and efficacy of CD19 CAR T-cells with escalating doses of lymphodepleting cyclophosphamide and fludarabine with and without rituximab. There have been no dose limiting toxicities and no Grade 3-4 CRS or ICANS. The feasibility of point of care manufacturing with vein-to-vein time of 8 days is demonstrated in this phase I study. The efficacy of the CARs is maintained and comparable to commercial CD19 products without any severe CRS or ICANS. Enrollment continues on DL 5.