Efficacy of SCRI-CAR19x22 T cell product in B-ALL and persistence of anti-CD22 activity.

Abstract

3035 Background: Loss of CD19 expression is a major cause of limited durable B-ALL remission following CD19 CAR T cells, which might be overcome by utilization of dual CD19xCD22 CAR T cell targeting. Methods: A Phase I trial (NCT03330691) of SCRI-CAR19x22 was developed using dual transduction of lentiviral vectors encoding for either a CD19- or CD22-specific CAR T cell construct, both with 4-1BB co-stimulation. Manufacturing was performed in a closed G-Rex system with IL-7, IL-15 and IL-21. After lymphodepletion, CAR T cells were infused at 1 or 3 X 106 CAR T cells/kg dose levels. Leukemic response and CAR T cell persistence were evaluated by flow cytometry. Results: Products were successfully manufactured in all 28 enrolled subjects with 7.92 average days in culture (range of 7-11 days) and consisted of an average CD8:CD4 ratio of 3.09 (range 0.19 to 8.9). The cellular product CAR composition was 29% CD19, 31% CD22 and 39% CD19 and CD22 targeting. 13 subjects had prior exposure to CD19 or CD22 targeting therapies with diverse expression of CD19 and CD22 on the leukemic blasts. No dose limiting toxicities occurred in the 27 infused subjects. The recommended phase 2 dose is 3 x 106 CAR+ cells/kg. CRS was present in 80% of subjects, with 85% of CRS being grade 2 or less, and a peak grade of 3 (n = 3). Mild neurotoxicity occurred in 38%, with a single grade 3 event. 84.6% obtained a CR, of which 95% were MRD negative. Of the 4 subjects who did not achieved a CR, 2 had a pre-existing CD19 negative population and one had previously received CAR T cells and rejected SCRI-CAR19x22. There have been 4 relapses with varying CD19 and CD22 expression as follows: 1 CD19-CD22-, 1 CD19+CD22+, and 2 CD19-CD22+. The in vivo engraftment of CAR T cells peaked most frequently between day +7 and +14 and was predominated by the CD19 CAR+ T cells. Conclusions: We demonstrate manufacturing feasibility and safety of SCRI-CAR19x22. While initial efficacy is demonstrated, CD22 activity is poor due to limited expansion of the CD22 CAR-containing components and subjects with pre-existing CD19 negative leukemia fared poorly. Development of a revised CD22 CAR that exhibits a reduction tonic signaling is underway, with plans to explore the new construct in the context of a dual-targeting CD19xCD22 CAR T cell product. Clinical trial information: NCT03330691 .