Background The clinical response to adoptive T cell therapies is strongly associated with transcriptional and epigenetic state. Thus, technologies to discover regulators of T cell gene networks and their corresponding phenotypes have great potential to improve the efficacy of T cell therapies.

Methods We developed pooled CRISPR screening approaches with compact epigenome editors to systematically profile the effects of activation and repression of 120 transcription factors and epigenetic modifiers on human CD8 T cell state.

Results CRISPR interference and activation screens nominated known and novel regulators of T cell phenotypes with BATF3 emerging as a high confidence gene in both screens. We found that BATF3 overexpression promoted specific features of memory T cells such as increased IL7R expression and glycolytic capacity, while attenuating gene programs associated with cytotoxicity, regulatory T cell function, and T cell exhaustion. In the context of chronic antigen stimulation, BATF3 overexpression countered phenotypic and epigenetic signatures of T cell exhaustion. For example, only 13% of BATF3 engineered T cells co-expressed canonical exhaustion markers (LAG3, TIM3, TIGIT), whereas 65% of wild type T cell co-expressed all three markers after multiple rounds of antigen stimulation. CAR T cells overexpressing BATF3 significantly outperformed control CAR T cells in both in vitro and in vivo tumor models. Moreover, we found that BATF3 programmed a transcriptional profile that correlated with positive clinical response to adoptive T cell therapy. Finally, we performed CRISPR knockout screens with and without BATF3 overexpression to define co-factors and downstream factors of BATF3, as well as other therapeutic targets.

Conclusions BATF3 overexpression markedly enhanced the therapeutic potential of CD8 T cells in both in vitro and in vivo tumor models. The compact size of BATF3 could seamlessly integrate into current manufacturing processes of FDA-approved adoptive T cell therapies, which all use lentivirus to deliver the CAR construct to donor T cells. To our knowledge, this work is the first example that combines overexpression of a specific transcription factor with a transcription factor wide knockout screen to dissect co-factors and downstream factors and highlights the power of this approach. These screens pointed to a model where BATF3 interacts with JUNB and IRF4 to regulate gene expression and illuminated several other novel targets for further investigation.