Defining Optimal T Cell Characteristics for Pediatric Chimeric Antigen Receptor (CAR) T Cell Trials

Abstract

Background: Engineered chimeric antigen receptor (CAR) T cells have emerged as a powerful, highly personalized immunotherapy in pediatric cancer. Early phase clinical trials using CAR T cells targeting CD19 have resulted in complete response (CR) rates as high as 93% in children with relapsed and refractory acute lymphoblastic leukemia (ALL). Despite this success, there are many challenges that must be overcome before CAR T cell therapy can be used routinely in pediatric ALL or other malignancies. Objective: To develop novel biomarkers that will identify patients at high risk for poor product expansion, treatment failure and/or toxicity with current immunotherapy protocols and that may ultimately help improve manufacturing methods to produce safer and more effective CAR T cells. Design/Methods: Using flow cytometry, we evaluated T cell characteristics (memory phenotype, cytokine production, presence of markers of activation/exhaustion) of starting products in 43 patients enrolled and treated in a Phase I clinical trial, PLAT-02. Potential predictors of poor product expansion were assessed using the Wilcoxon Rank Sum Test. Results: A higher percentage of cells producing cytokines and expressing PD-1 in CD4 starting products was associated with poor expansion. Poor expansion was not associated with patient toxicities or outcomes. Discussion: Increased cytokine production and PD-1 expression suggest a more differentiated, effector-like phenotype of starting products that subsequently experience poor expansion, consistent with preclinical data, although numbers are limited. Conclusion: Ongoing correlative biology studies will be important in future immunotherapy trials as we seek to identify biomarkers that predict product expansion, toxicities and outcomes.