Sequential CD19 and CD22 CART for relapsed and refractory B cell ALL: Phase I results

Abstract Background Relapsed/refractory (R/R) B-cell acute lymphoblastic leukemia (B-ALL) is a predominantly CD19+/CD22+ hematologic malignancy that is associated with poor long-term survival in both children and adults. Chimeric antigen receptor T cells (CAR-T) targeting CD19 and/or CD22 have demonstrated impressive response rates in children and adults with R/R ALL. However, long-term durable responses are limited by poor CAR-T persistence, putting patients at risk for relapse. Here, we describe interim results of an ongoing Phase I/Ib clinical trial testing safety, feasibility and preliminary assessment of efficacy to serial infusion of commercial CD19 CAR-T followed by CD22 CAR-T. Methods In this single center study, children, adolescents, and young adults (AYA) with R/R B-ALL received fludarabine and cyclophosphamide lymphodepletion and tisagenlecleucel CD19 targeted CAR-T (tisa-cel; Kymriah) as per the standard of care (SOC) indication. Patients then received 3x105 CD22 CAR T cells/kg (humanized m791 4-1BB) within 42 days of the tisa-cel infusion without repeated lymphodepletion. Primary objectives were to assess safety of this sequential CAR-T dosing strategy, feasibility of manufacturing two CAR-T products from a single apheresis product, and feasibility of administering the CD22 CAR product within 42-days following the tisa-cel infusion. Secondary objectives were to describe the duration of B cell aplasia and disease-free survival at 6 and 12 months following tisa-cel and to describe the kinetics and durability of both CAR-T products over the first 12 months following tisa-cel infusion. Results Six patients have enrolled thus far, (50% male, 50% relapsed, 50% refractory, 7-26 years old; 1-6 prior lines of therapy); three patients had flow detectable disease (0.03-20.9%) while the remaining 3 were flow MRD negative. Four of six had detectable NGS MRD (41-21145 clonal cells/million) prior to tisa-cel infusion. All patients successfully had tisa-cel and CD22 CAR-T products manufactured from a single apheresis and all received 3x105 CD22 CAR-T cells/kg following SOC tisa-cel. CD22 CAR-T was infused 29-39 days following tisa-cel, thus meeting defined feasibility criteria. Of the first 4 patients who are through the dose limiting toxicity period, 100% achieved flow and NGS remission after tisa-cel infusion. One subject experienced Grade 1 cytokine release syndrome (CRS) and Grade 3 immune effector cell-associated neurotoxicity (ICANS) after tisa-cel. Despite receiving CD22 CAR-T while in NGS undetectable remission, 75% of patients experienced CRS (Grade 1-2) and one patient experienced Grade 2 immune-effector cell Hemophagocytic Lymphohistiocytosis-Like Syndrome (IEC-HS) following CD22 CAR-T. No ICANS was observed after CD22 CAR-T infusion. Peak CD22 expansion was observed at Day 14 (median 48.3% of CD3+ cells, range 24.5-97.6%), with predominance of CD8+ CD22 CAR+ cells (median 36%; range 20-66%). Updated data on safety, expansion, kinetics, persistence, B-cell aplasia and response will be presented. Conclusions In this Phase I portion of SOC tisa-cel CD19 CAR-T followed by CD22 CAR-T for children and AYA with relapsed or refractory B-ALL, we demonstrate acceptable safety and feasibility of this sequential infusion strategy and promising early signals of efficacy in providing prolonged B cell aplasia and disease control.