Obecabtagene autoleucel (obe-cel) for Adult Relapsed/Refractory B-Cell Acute Lymphoblastic Leukemia (R/R B-ALL): Deep Molecular Remission May Predict Better Outcomes

Background: Obe-cel is an autologous 41BB-Z CD19-directed CAR T-cell therapy which uses a fast off-rate target antigen binding domain to improve persistence and reduce toxicity. The efficacy and safety of obe-cel as a treatment for adult R/R B-ALL is being investigated in the FELIX study (NCT04404660), an open-label, multi-center, global, single-arm, Phase Ib/II clinical study. In FELIX, obe-cel was associated with 12-month event-free survival (EFS) and overall survival (OS) rates of 50% and 61%, respectively, at median follow-up of 21.5 months. Forty percent of responders were in ongoing remission without subsequent stem cell transplant (SCT) or other new therapies, suggesting the potential for a long-term plateau (Jabbour E, et al. J Clin Oncol 2024; 42[Suppl 16]:6504). Here, we report on the correlation between depth of measurable residual disease (MRD)-negative remission and clinical outcomes in patients (pts) treated with obe-cel.

Methods: Bone marrow (BM) samples were collected at screening, within 1 week before lymphodepletion, and 28 days post obe-cel infusion, with optional samples collected at Months 3, 6, 12, and then every 6 months until the end of study. The BM samples were sent for central MRD evaluation by ClonoSEQ^® next-generation sequencing (NGS) assay with baseline calibration performed using BM samples taken at screening, just prior to lymphodepletion, or earlier samples depending on availability. The MRD NGS assay is sensitive to 10‾⁶ leukemic cells (i.e. 0.0001% or 1 in a million cells).

Results: In total, 96/127 (76%) pts infused with obe-cel had samples available to enable successful NGS calibration; 73/96 (76%) achieved complete remission (CR) or CR with incomplete count recovery (CRi), and 68/73 (93%) had at least one post-infusion BM sample for NGS evaluation. Among evaluable responders, based on best MRD response achieved, 4/68 (6%) pts had MRD at ≥10‾⁴ leukemic cells (i.e. MRD-positive) and 64/68 (94%) reached MRD <10^–4 leukemic cells with 7/68 (10%) having MRD between 10‾⁴ and 10‾⁶ leukemic cells, and 57/68 (84%) with MRD <10^–6 leukemic cells. At the median follow-up of 21.5 months (range: 8.6–41.4), among responders, the median duration of response (95% confidence interval [CI]) was 1.5 months (0.6, not estimable [NE]) for pts with MRD ≥10‾⁴ leukemic cells, 3.8 months (3.4, NE) for pts with MRD between 10‾⁴ and 10‾⁶ leukemic cells, and 17.1 months (11.6, NE) for pts with MRD <10‾⁶ leukemic cells. Median EFS (95% CI) was 0 months (0.03, 1.7) versus 4.5 months (0.03, NE) versus 15.1 months (11.7, NE), and median OS (95% CI) was 2.2 months (1.2, 6.9) versus 8.9 months (6.8, 12.9) versus NE, for pts with MRD ≥10‾⁴, MRD between 10‾⁴ and 10‾⁶, and MRD <10‾⁶ leukemic cells, respectively. Twelve-month EFS rates were 10%, 0%, and 61%, and 12-month OS rates were 20%, 27%, and 85%, for pts with MRD ≥10‾⁴, MRD between 10‾⁴ and 10‾⁶, and MRD <10‾⁶ leukemic cells, respectively.

Conclusions: Eighty-four percent of responders to obe-cel for whom MRD by NGS could be analyzed achieved MRD <10‾⁶ leukemic cells which was associated with more durable responses, and higher EFS and OS rates than those observed in pts with MRD ≥10‾⁴ and between 10‾⁴ and 10‾⁶ leukemic cells. Pts who did not achieve MRD <10‾⁶ had poorer outcomes.