P06.15 Highly multiplexed, single-cell functional proteomics of CAR-T products enables more predictive product characterization, cell manufacturing optimization, and cellular biomarkers across product types

Chimeric antigen receptor (CAR) T cell therapy has already paved the way for successful immunotherapies to fight against liquid tumors and is quickly expanding to solid tumors. Nevertheless, the biggest challenges are how to evaluate the quality of CAR-T cells and how to predict their in vivo behaviors once reinfused into a patient. In this report, we review single-cell polyfunctional profiling results obtained from several different sets of pre-infusion CAR-T samples, including CD19 CAR-T products from Novartis and Kite Pharma (Gilead), GoCAR-T cell products targeting Prostate Stem Cell Antigen from Bellicum, bispecific CD19/22 CAR-T cells from the NIH, trimeric APRIL-based CAR-T cells targeting both BCMA and TACI from MGH and CAR-T cells targeting glypican 3 in hepatocellular carcinoma from NIH. In each case, CD4+ and CD8+ CAR-T cells were stimulated and subsequently analyzed at a single-cell level using IsoPlexis’ IsoCode proteomic chips. Our single-cell data revealed highly polyfunctional and heterogeneous responses across each cohorts. The polyfunctional strength index (PSI) of the pre-infused CAR-T products is significantly associated with the clinical outcome of the patients after receiving the treatment, as well as post-infusion grade 3+ CRS. The CAR-T cells secreted a wide range of cytokines/chemokines in response to antigen specific stimulation and a significant portion of the CAR-T cells were polyfunctional (2+cytokines/cell). These results highlight the potential benefits of single-cell proteomics to comprehensively understand how CAR-T products behave in response to antigen-specific stimulation. Analyzing the single-cell polyfunctionality of CAR-T profiles also provides a valuable quality check for optimizing the manufacturing process and a powerful tool for next generation biomarker developments. Disclosure Information D. Liu: None. P. Paczkowski: None. S. MacKay: None. J. Zhou: None.