Production of Anti-CD19 CAR T Cells for ZUMA-3 and -4: Phase 1/2 Multicenter Studies Evaluating KTE-C19 in Patients With Relapsed/Refractory B-Precursor Acute Lymphoblastic Leukemia (R/R ALL)

Introduction: ZUMA-3 and -4 (NCT02614066 and NCT02625480) are ongoing phase 1-2 multicenter trials evaluating KTE-C19, a CD28/CD3ζ anti-CD19 chimeric antigen receptor (CAR) T cell therapy, in adult and pediatric patients respectively with R/R ALL. The manufacturing process developed to support the clinical trials aimed to accommodate variability in the incoming apheresis material, maximize manufacturing success rate, and minimize timing between leukapheresis collection and KTE-C19 administration at bedside.

Methods: Once eligibility criteria were satisfied, patients underwent leukapheresis to process 12-15 L of blood (2-5 blood volumes for pediatric patients) to target collection of 10 × 10⁹ mononuclear cells (MNCs). After collection, leukapheresis material was shipped to the central manufacturing site and processed to enrich T cells using an automated and closed system. The T cell enrichment step was implemented to remove cellular impurities such as leukemic blasts and myeloid cells that may negatively impact T cell activation and expansion. Enriched T cells were activated using an antigen presenting cell (APC)-free and bead-free method and cultured in serum-free medium containing 300 IU/mL of interleukin-2. Activated T cells were transduced with a gamma retroviral vector that encodes the CAR gene and further expanded to achieve target dose of 2 × 10⁶ CAR-positive T cells/kg body weight (minimum of 1 × 10⁶). Overall target duration of the manufacturing process was 8 days with an allowed range for harvest set between process days 6 and 10. Final KTE-C19 product was washed, formulated, cryopreserved and tested for identity, potency, and adventitious agents. After acceptance criteria were met, KTE-C19 was shipped to the clinical site using a validated cryoshipper.

Results: As of July 2016, 5 of the 6 patients enrolled in the clinical trials received the target dose of 2 × 10⁶ CAR-positive cells/kg body weight. The assigned dose could not be generated for one patient enrolled in the ZUMA-4 trial. In this patient, analysis performed on the leukapheresis material revealed >99% leukemic blasts. For the other 5 patients, the target dose was achieved within 6 - 7 process days notwithstanding a wide range in T cell composition of the incoming leukapheresis material (range, 4.7% to 81%), and wide variation in percent leukemic blasts (range, 11% to 92%). Median frequency of CAR expression in the KTE-C19 final product was consistent across product lots. The manufacturing process led to final products highly enriched for T cells with a median CD3+ cell content of >99%. Expansion of T cells was consistent among the 5 product lots with a mean 7-fold expansion from transduction to harvest. Median time from leukapheresis collection to final product release was 14 days.

Conclusions: Reproducible manufacturing of high-quality, clinical-grade autologous CAR T cell products may be challenged by the inherent variability of starting material in patients with high content of leukemic blasts in peripheral blood. The bead- and serum-free manufacturing process developed to support the ZUMA-3 and -4 clinical trials was able to generate products within 6 -7 days with a low failure rate (including from patients with high percentages of leukemic blasts), and most importantly, a short duration from leukapheresis collection to shipment of KTE-C19 final product back to the clinical center for patient administration.