Development of a microfluidic cell transfection device into gene-edited CAR T cell manufacturing workflow

Abstract

Genetic reprogramming of immune cells to recognize and target tumor cells offers a possibility of long-term cure. Cell therapies, however, lack simple and affordable manufacturing workflows, especially to genetically edit immune cells to more effectively target cancer cells and avoid immune suppression mechanisms. Microfluidics is a pathway to improve the manufacturing precision of gene modified cells. However, to date, it remains to be demonstrated that microfluidic treatment preserves the functionality of T cell products in a complete workflow. In this study, we used microfluidics to perform CRISPR/Cas9 gene editing of CD5, a negative T-cell regulator, followed by the insertion of a chimeric antigen receptor (CAR) transgene via lentiviral vector transduction to generate CAR T cells targeted against the B cell antigen CD19. As part of the workflow, we have optimized a microfluidic device that relies on convective volume exchange between cells and surrounding fluid to deliver guide RNA and Cas9 ribonucleoprotein to primary T cells. We comprehensively tested critical design features of the device to improve the gene-edited product yield. By combining high-speed video and cell mechanics measurements using the atomic force microscope, we validate a model that relates the device design features to cell properties. Our findings showed enhanced performance was obtained by focusing the cells to counteract the flow resistance caused by the ridge constrictions, providing a ridge layout that allows sufficient cycles of compression and time for volume recovery, and including a gutter to clear aggregates that could reduce cell viability. The optimized device was used in a workflow to generate CD5-knockout CD19 CAR T cells. The microfluidics approach resulted in >60% CD5 editing efficiency, ≥80% cell viability, similar memory phenotype composition as unprocessed cells, and superior cell growth. The microfluidics workflow yielded 4-fold increase of edited T cells compared to an electroporation workflow post-expansion. The transduced CAR T cells showed similar transduction efficiency and cytotoxicity against CD19-positive leukemia cells. Moreover, patient-derived T cells showed the ability to be similarly edited, though their distinct biomechanics resulted in slightly lower outcomes. Microfluidics-based manufacturing is a promising path towards more productive clinical manufacturing of gene edited CAR T cells.