(30d) Harnessing Tangential Flow Filtration and Hydrogels to Enhance Transduction and Select Distinct Populations for CAR T Manufacturing

Autologous chimeric antigen receptor (CAR) T-cell therapies have become an important tool for the treatment of hematological cancers, with six different therapies currently approved for use by the FDA. Despite this, autologous engineered cell therapies like CAR T have faced limitations in their usage clinically, owing in part to challenges in their manufacturing process that result in high costs of the therapy (350k-450k per infusion) and variable therapeutic efficacy. Critical contributors include the viral vector used in transduction, which can constitute up to nearly 40% of the manufacturing cost, as well as the ratio of CD4/CD8 T-cells of which current manufacturing workflows do little to control. Efficient and scalable processes are needed for manufacturing of CAR T-cells, where processes that reduce the amount of virus required and allow for targeted cell isolation present significant opportunities for reducing cost and increasing therapeutic efficacy. In this work, we examined the use of a tangential flow filtration (TFF) device with a ligand functionalized hydrogel coated membrane (HCM) to evaluate the impacts of cell loading density and flow patterns to enhance and optimize transduction and selection.

To demonstrate the efficacy of our platform for lentiviral transduction, we evaluated the transduction of Jurkats with a model lentivirus in the TFF device. Modulation of the density of cells loaded into the device enhanced transduction, with a 5x increase in transduction in the device compared to static culture at low multiplicities of infection. Underlying interactions that regulate this were further probed with confocal microscopy to understand mechanism. Additionally, the scalability of this platform for transduction was examined, demonstrating the ability to decouple and tailor different operating parameters for optimal transduction.

Evaluating the efficacy of our platform for cell subtype selection, we first developed a cell-line model approach that evaluates the selection of model human T-cells from model human B-cells from a mixed population and functionalized our HCMs with anti-CD3 antibodies to enable selectivity for the model T cells. Moving to primary human cells, we then tested our system on freshly isolated T-cells from human derived peripheral mononuclear cells (hPBMCs) and sought to isolate different ratios of CD4/CD8 T-cells from a heterogeneously mixed CD4/CD8 T-cell population. Further, bringing these processing steps together, we then transduced the positively selected population in-situ.

In conclusion, the use of TFF/HCMs enables both the manipulation of operating parameters and device sizes to improve transduction in current cell therapy production workflows. Through more efficient use of lentiviral vectors, these improvements to transduction offer the potential to lower costs and increase accessibility of the therapy to patients. Furthermore, the additional functionalization of HCMs within the TFF for targeted T-cell subpopulations enables better control of cell therapy CD4/CD8 ratios for seamless integration into a CAR T manufacturing workflow.