High-throughput identification of transcription factors that enhance car t cell efficacy in solid tumors

The National Cancer Institute awarded The Broad Institute, Inc. $439,665 in a Project Grant (R21CA309357) on September 3, 2026, to conduct high-throughput identification of transcription factors that enhance CAR T cell efficacy in solid tumors under the Cancer Biology Research program (CFDA 93.396).

The project develops an in vivo screening approach to evaluate transcription factor overexpression in primary human T cells at scale. T cells are lentivirally transduced with a library of all human transcription factor open reading frames and either expanded in vitro or transferred into immunodeficient mice bearing human xenograft tumors expressing surface anti-CD3. After two weeks, T cells are isolated and sequenced to identify transcription factors that increase or reduce T cell abundance under both in vitro and in vivo conditions. This work addresses the persistent dysfunction of CAR T therapies in solid tumors caused by T cell exhaustion, a distinct transcriptional and epigenetic state that limits T cell persistence and tumor elimination. The approach targets the over 2,000 transcriptional and epigenetic modifiers in the human genome to identify candidates that could exert broad effects on T cell function, including epigenetic remodeling and acquisition of critical functional capacities.

Performance occurs in Cambridge, Massachusetts. The award period runs through August 31, 2028.

Abstract

While chimeric antigen receptor (car) t therapies can be highly effective against hematologic malignancies, their efficacy in solid tumors is less striking. A major factor limiting car t cell therapy in solid tumors is the progressive acquisition of a dysfunctional state termed "t cell exhaustion." exhaustion limits t cell persistence and the ability to eliminate target cancer cells and arises in chronic viral infections and cancer through repeated antigen stimulation. This program is hard-wired into all mature t cells to prevent autoimmune pathology but inevitably leads to loss of tumor control in cancer. Genetic approaches that delete inhibitory receptors or negative regulators have failed to reverse exhaustion, and in some cases cause unanticipated defects in t cell persistence. Because exhaustion is a distinct cell state defined by marked transcriptional and epigenetic changes, we believe that overexpression of transcription factors (TFS) could exert broad effects on t cell function, including epigenetic remodeling, differentiation, and the acquisition of critical functional capacities. The human genome encodes over 2000 transcriptional and epigenetic modifiers, making individual assessment of each impractical. To assess the impact of TF overexpression on car t cell function at scale, we developed an in vivo screening approach that evaluates all human TFS in primary human t cells. T cells are lentivirally transduced with a library of all human TF open reading frames (orfs) and either expanded in vitro or transferred into immunodeficient mice bearing human xenograft tumors that express surface anti-CD3. After two weeks, t cells are isolated and sequenced to identify TFS that increase or reduce t cell abundance under in vitro and in vivo conditions. Using this method we identified transcription factors, including GFI1B, HOXD8, HOXC8, POU2F3, and others, that increase t cell abundance in solid tumors. We validated that expression of HOXD8 and POU2F3 enhances t cell-mediated tumor control. Many of these factors have no previously characterized role in t cells, and some, such as the HOX family TFS, are not expressed in any adult lymphocytes. We hypothesize that these TFS will enhance car t therapeutic efficacy in preclinical models and drive distinct changes in the transcriptional and epigenetic states of t cells. We will determine whether these TFS increase efficacy of car-t cells in solid tumor xenograft models and profile the changes in TF binding, chromatin organization and gene expression in these car t cells. Animals are required because the trafficking, infiltration, and persistence of human car t cells within a vascularized solid tumor cannot be recapitulated in culture. We will therefore use immunodeficient NSG mice bearing human xenograft tumors, which support engraftment of both the human tumor and the human car t cells, in the minimum numbers needed for rigorous statistical comparison. These studies will not only enable the selection of the most promising candidates to pursue individually or in combination in car t cell therapies, but also provide insights into the plasticity and regulation of t cell exhaustion.