Humanized CD19 chimeric antigen receptor (CAR) T-cell therapy for high-risk and post-CAR relapse of B-cell acute lymphoblastic leukemia

Abstract

Abstract Background CD19-directed chimeric antigen receptor T-cell (CART19) therapy has transformed the treatment landscape for relapsed/refractory (r/r) B-cell acute lymphoblastic leukemia (B-ALL). Yet relapse occurs in approximately 50% of recipients. These patients and subgroups of pediatric patients in first relapse, currently ineligible for commercial CART19, have very poor outcomes with current approaches, warranting investigation of alternative strategies. We previously investigated a humanized CART19 (huCART19) in a Phase I trial with encouraging responses in both CAR-naïve and CAR-exposed cohorts (Myers JCO 2021). Here we report outcomes from a Phase II clinical trial (NCT03792633) of huCART19 for high risk r/r B-ALL, including early bone marrow (BM) relapse, a subgroup with historically poor outcomes despite intensive therapy. Methods Patients aged 0-29 years (y) with CD19+ B-ALL were eligible in 2 cohorts: CAR-naïve patients with B-ALL that is refractory, in high risk first relapse, second or greater relapse, or relapsed after or ineligible for hematopoietic stem cell transplant (HSCT); CAR-exposed patients with poor response to prior cell therapy. Patients received huCART19 at a dose of 5x106 CAR T cells/kg (maximum 2.5x108) after lymphodepletion (LD) with fludarabine and cyclophosphamide. Response was assessed on day 28 by BM and cerebrospinal fluid morphology, minimal residual disease (MRD) by flow cytometry, and biologic response by B cell aplasia (BCA). The primary endpoint was event-free survival (EFS). Secondary endpoints included overall response rate (ORR), defined as rate of complete remission (CR) or CR with incomplete hematologic recovery (CRi) with BCA, and relapse-free survival (RFS). Data cutoff was July 1, 2025. Results Of 106 patients enrolled, 100 were infused with huCART19 from 3/2019-8/2023. The median age at infusion was 12y (range 1-29), 44% were female, 20% Hispanic, 5% Asian, 7% Black, and 6% had Trisomy 21. Prior therapy included HSCT in 21%, blinatumomab in 21% and inotuzumab in 15%. The CAR-naïve cohort (n=52) included 25 with first early BM relapse within 36m of diagnosis (12 <18m). On pre-infusion BM performed post-LD, 13/52 (25%) had >25% blasts, 25/52 (48%) <0.01%. Three patients were inevaluable for response: 1 deemed ineligible due to myeloid lineage switch on pre-infusion BM; 1 died on day 2; 1 lost to follow-up at day 28. By day 28, 46/49 (94%) were in CR/CRi, 45/46 MRD-negative, 1 MRD inevaluable. With a median follow up of 51m, EFS and RFS at 2y were 65% (95% CI 52-81%) and 70% (95% CI 57-86%), and at 4y, 57% (95% CI 42-76%) and 62% (95% CI 47-81%). Relapse occurred in 14/46 (30%), of which 7 (50%) were CD19-. Ten patients pursued alternative therapy in remission, 7 for loss of BCA and 3 for MRD recurrence (2 CD19+, 1 CD19-). Overall survival (OS) at 2y and 4y was 74% (95% CI 63-87%). In a subgroup analysis of CAR-naïve patients treated for early BM relapse, ORR was 21/23 (91%), 2y EFS was 39% (95% CI 21-72%), RFS 45% (95% CI 25-81%), and OS 57% (95% CI 40-81%). The CAR-exposed cohort (n=48) included 20 with post-CART19 relapse and 28 with early (<6m) loss of BCA without relapse. On pre-infusion BM, 5/48 (10%) had >25% blasts, 34/48 (71%) <0.01%. By day 28, 43/48 patients were in CR/CRi, 43/43 MRD-negative; 6/43 did not establish BCA resulting in an ORR of 37/48 (77%). With a median follow up of 43m, EFS and RFS at 2y were 53% (95% CI 39-71%) and 72% (95% CI 57-91%), and at 4y, 50% (95% CI 36-68%) and 68% (95% CI 52-88%). Relapse occurred in 9/37 (24%), of which 4 (44%) were CD19-. Eleven patients received alternative therapy in remission, 6 for loss of BCA, 4 for CD19+ MRD recurrence, and 1 for therapy-related myeloid neoplasm. OS at 2y and 4y was 77% (95% CI 66-90%). CRS was reported in 47/52 (90%, 10 grade [Gr] 3, 5 Gr 4 on Penn scale) CAR-naïve patients and 38/48 (79%, 1 Gr 3, 2 Gr 4) CAR-exposed. There was 1 death prior to day 28, due to gastrointestinal hemorrhage on day 2 in the setting of progressive ALL and Gr 2 CRS. CAR neurotoxicity was reported in 14/52 (27%, 2 Gr 3, 2 Gr 4) CAR-naïve patients and 7/48 (15%, 1 Gr 3, 1 Gr 4) CAR-exposed, with 1 case of Gr 3 cerebral edema, fully recovered, and 1 case of ongoing myelopathy. Conclusions HuCART19 produced durable remissions in high risk r/r B-ALL and demonstrated efficacy as salvage therapy for those with poor response to prior CAR therapy, comparing favorably to historical outcomes in this extremely high risk group.