Abstract
Cell therapy manufacturing of primary T cells often results in heterogeneous cell populations within a final product, with many cells lacking desired of receptor expression or those that have exhausted or other dysfunctional phenotypes. Here, we design a novel cell-intrinsic strategy to genetically reprogram primary human T cells to autonomously detect and eliminate dysfunctional cells. This integrated detection and elimination process, known as directed fratricide, is programmed via nonviral CRISPR genome-editing to eliminate the T cell receptor (TCR) alpha chain ( TRAC gene knockout) and integrate a chimeric antigen receptor (CAR) against the urokinase-type plasminogen activator receptor (uPAR), also known as CD87. Within these cell products, strong T cell stimulation or activation during manufacturing causes a small subset of cells to express uPAR, which subsequently triggers CAR-mediated killing by a separate subset of cells within the product. This fratricide induces proliferation in the desired cells and destroys undesired cells, a process that could be modeled computationally and controlled robustly via supplements to the culture media. The strategy enabled enrichment of anti-uPAR and anti-GD2 CAR T cell products up to ≥99% CAR+/TCR-, favoring a stem cell memory-like phenotype (CD45RA (high) /CD62L (high) ). Understanding growth dynamics among T cell subsets and reprogramming them via CRISPR could accelerate the biomanufacturing of potent cell products without extensive selection methods.