Genome-wide CRISPRa screens nominate modulators of CAR T cell survival within distinct tumor cytokine milieus

Abstract

Chimeric Antigen Receptor (CAR) T cell therapy has revolutionized the treatment of B cell malignancies and translating this success to other cancers remains an ongoing clinical objective. Next-generation T cell products in development aim to genetically modulate many facets of cell behavior, for which gene-nominating platforms provide a useful framework for prioritization. Among competing screening approaches, CRISPR activation (CRISPRa) technology permits gain-of-function (GoF) gene surveys at genome-wide scale, but routine implementation in primary T cells has been stymied by high cell requirements ([~]107 - 108) and abbreviated activity. Here, we describe a novel cell manufacturing schema using an all-in-one transposon-based gene delivery system coupled with CAR-restricted cell expansion to generate yields (109) of primary T cells bearing CAR and CRISPRa transgenes that are well above the threshold needed for genome-scale screening. CRISPRa activity is sustained via the inclusion of divergent, duplicate Elongation Factor 1 core/human T-cell leukemia virus (EF1-HTLV) hybrid promoters; while guide RNA representation is preserved through late lentiviral transduction, thus preventing bottlenecking and premature candidate pruning. CRISPRa-CAR T cells manufactured via this pipeline retain potent on-target gene-overexpression (>85% target+) across varied cell subsets (e.g. Tim-3+Lag3+ or serial-challenge) and timescales (>14 days). When deployed to survival-based genome-wide selection landscapes, CRISPRa-CAR pools nominate known and novel endogenous genes capable of enhancing CD8+ CAR T survival in cytokine-rich (e.g. MYC, FUT6, IRF4, GSE1) and cytokine-depleted (e.g. CSF2RB, STAT6, IRF4, GSE1) settings of tumor challenge. This system will have broad utility for therapy-enhancing gene discovery.