Donor-Derived CAR T Cells Engineered with Sleeping Beauty in Pediatric and Adult Patients with Acute Lymphoblastic Leukemia Relapsed Post-HSCT

Introduction Allogeneic Chimeric Antigen Receptor (CAR) T cells engineered with non-viral methods offer a modality to reduce costs and logistical complexity of the viral process and allow lymphodepleted patients to access CAR T cell treatment. We recently proposed the use of Sleeping Beauty (SB) transposon to engineer donor-derived T cells differentiated according to the cytokine-induced killer (CIK) cell protocol (Magnani CF et al. J Clin Invest. 2021). We report here outcomes on B-cell acute lymphoblastic leukemia (B-ALL) patients, relapsing after transplantation, treated with donor-derived anti-CD19 CAR T cells (CARCIK-CD19). Methods We conducted an academic, multi-center, phase I/II dose-escalation trial in patients relapsed after allogeneic hematopoietic stem cell transplantation (HSCT). The infusion product was manufactured in-house starting from 50 mL of peripheral blood from the HSCT donor by electroporation with GMP-grade plasmids. All patients underwent lymphodepletion with Fludarabine (30 mg/m 2/day x 4 days) and Cyclophosphamide (500 mg/m 2/day x 2 days), before proceeding to CARCIK-CD19 infusion. We used the Bayesian Optimal Interval (BOIN) design to define a four-dose escalation scheme. Primary objectives were to define the Maximum Tolerated Dose (MTD), safety, and feasibility. Secondary objectives included the assessment of complete hematologic response (CR), duration of response (DOR), progression-free (PFS), event-free (EFS), and overall survival (OS). This study was registered at ClinicalTrials.gov, NCT03389035. Results From January 2018 to June 2021, a total of 32 patients were screened, 26 enrolled (6 children and 20 adults) and 21 infused (4 children and 17 adults). Reasons for not receiving infusion included consent withdrawal (N=1), disease progression not controlled by bridging therapy (N=3), acquisition of myeloid phenotype (N=1). The median number of prior therapies was 4 (range, 1-7) with a median time interval from HSCT to relapse of 9 months. The median BM blasts was 60% (range, 5-100%) at enrollment and 7% (range, 0-96%) post lymphodepletion. Of the 21 patients infused, CARCIK-CD19 were obtained by HLA-identical sibling (n=6, 29%), matched unrelated (n= 7, 33%), and haploidentical donors (n=8, 38%). Three patients (14%) received the first dose level of 1x10 6 CARCIK-CD19 cells/Kg, three (14%) the second of 3x10 6, and three (14%) the third of 7.5x10 6 whereas 12 patients (57%) received the fourth and last planned dose level of