Chimeric antigen receptor (CAR) T-cell therapy has revolutionized the therapeutic landscape for relapsed or refractory lymphoid malignancies, achieving remarkable rates of durable remission. Despite this success, significant variability among patients in clinical responses and treatment-related toxicities remains a critical challenge, highlighting an urgent need for robust predictive biomarkers. Key intrinsic CAR T-cell attributes predictive of therapeutic efficacy and safety include the composition of memory T cell subsets, particularly central memory (Tcm) and stem cell memory (Tscm) populations, CAR density and transduction efficiency, cytokine production profiles with emphasis on polyfunctionality, and metabolic fitness. Additionally, the systemic immune contexture significantly modulates outcomes, including baseline systemic inflammatory cytokines, presence of regulatory immune cell populations, and the pre-treatment immunosuppressive tumor microenvironment. Recent advances in single-cell transcriptomics, comprehensive proteomic profiling, and cytokine polyfunctionality assays have provided greater resolution for identifying predictive biomarkers and optimizing therapeutic strategies. High-dimensional immunophenotyping combined with advanced machine learning methods enables automated CAR T-cell manufacturing quality control and precise immunological synapse quantification. Furthermore, tumor antigen (epitope) spreading following CAR T-cell therapy has risen as a provisional biomarker indicating broadened antitumor immunity and potentially sustained remission. Integrating these emerging biomarkers and advanced multi-omics approaches into clinical practice can refine patient stratification, enhance CAR T-cell manufacturing processes, and improve therapeutic outcomes in patients with lymphoid malignancies.
Yoo et al. (2025) studied this question.