A Phase I Study of CD19 Chimeric Antigen Receptor-T Cells Generated By the PiggyBac Transposon Vector for Acute Lymphoblastic Leukemia

Introduction: Chimeric antigen receptor-modified T cells targeting CD19 (CD19.CAR-T cells) have shown clinical success in patients with hematological malignancies. Despite the encouraging results obtained with this novel therapy, a major concern to its global spread, particularly in developing countries, is its high cost. We developed a method of non-viral gene transfer using piggyBac transposon to reduce the cost of CAR-T therapy. In preclinical study, the median number and transduction efficiency of CAR-T cells obtained from 2x10 ⁷ PBMC in 9 donors were 1.0x10 ⁸ (range, 0.58-1.8x10 ⁸) and 51% (range, 29-73%), respectively. The major subset of CAR-T cells was phenotypically CD8+CD45RA+CCR7+, closely related T-memory stem cells. Ex vivo, CD19.CAR-T cells showed cytotoxic effect on CD19 positive tumor cell lines. In NSG mice model, CD19.CAR-T cells successfully inhibit tumor growth. CAR gene integration sites were determined by inverse polymerase chain reaction and subsequent next-generation sequencing using MiSeq and equally distributed throughout the genome without preference for specific sites. The pre-clinical testing in mouse demonstrated safe toxicity profile at the 50 times dose of CD19.CAR-T cells. We started a human clinical trial to define feasibility, toxicity, maximum tolerated dose and clinical response of CD19.CAR-T cells (jRCTa040190099).

Methods: We report the results of cohort 1 of the study in which the safety and efficacy of autologous CD19.CAR-T in patients with relapsed or refractory B-precursor acute lymphoblastic leukemia were evaluated. We engineered autologous T cells via the piggyBac transposon system with CD19.CAR-expression transposon vector and piggyBac transposase-expression vector to express CD19.CAR incorporating CD28 costimulatory domain. We designed this phase I trial using a modified 3 + 3 design to enroll 3-12 patients with relapsed or refractory acute lymphoblastic leukemia in both children and adults. In this study, patients in cohorts 1 (16-60 years old) and 2 (1-15 years old) receive 1 × 10 ⁵ CAR-transduced T cells per kg. Patients in cohorts 3 and 4 (both 1-60 years old) receive 3 × 10 ⁵ and 1 × 10 ⁶ CAR-transduced T cells per kg, respectively. All patients receive 25mg/m ²/d of fludarabine and 250mg/m ²/d of cyclophosphamide for 3 days followed by a single infusion of CAR-T cells.

Results: Three patients were enrolled in cohort 1 and infused with 1 × 10 ⁵ CAR-transduced T cells per kg. All patients had previously undergone allogeneic hematopoietic stem cell transplantation. All patients had achieved a hematological complete response with salvage treatment before CAR-T therapy. None of the patients had dose-limiting toxicities (DLT) defined as nonhematological toxicities above grade 4 or cytokine release syndrome (CRS) above grade 4 or graft versus host disease (GVHD) above 4, or grade 3 nonhematological toxicities and GVHD not improved to grade 2 within 4 weeks after CAR-T infusion. There was no occurrence of non-hematological adverse events above grade 3. CRS was observed in one patient (grade 1) who also developed headache due to infiltration of CAR-T cells into the spinal fluid. In two patients, B cell aplasia lasted 2 and 9 months, respectively. Elevation of serum cytokine levels was observed in all patients and the peak time point was 7-21 days after CAR-T cell infusion.

Conclusions: CD19.CAR-T cell infusion produced by the piggyBac transposon gene engineering system was safe in cohort 1 of our study. As no patients had DLT in cohort 1, we are enrolling the patients in further cohorts.