A T-cell resilience model associated with response to immunotherapy in multiple tumor types

Despite breakthroughs in cancer immunotherapy, most tumor-reactive T cells cannot persist in solid tumors due to an immunosuppressive environment. We developed Tres (Tumor-resilient T cell, https://resilience.ccr.cancer.gov), a computational model and web server that utilizes single-cell transcriptomic data to identify signatures of T cells that are resilient to immunosuppressive signals, such as TGFβ, TRAIL, and PGE2. Analyzing T-cell transcriptomic data from 72 pretreatment and 84 pre-manufacture patient samples, Tres reliably predicts the clinical effectiveness of immune checkpoint blockade or adoptive cell transfer in melanoma, lung, and B-cell malignancies. Further, Tres identified FIBP, whose functions are largely unknown without previous links to T cells, as the top negative marker of tumor-resilient T cells across many cancers. FIBP knockouts in murine and human CD8+ T cells significantly enhanced T-cell mediated cancer-killing in invitro co-cultures. Further, Fibp knockout in murine T cells potentiated the in-vivo efficacy of adoptive cell transfer by limiting cholesterol metabolism, which otherwise inhibits effector T-cell function. These results demonstrate Tres’s utility in identifying clinical biomarkers of T-cell effectiveness and potential therapeutic targets for immunotherapies in solid tumors. Examination of Fibp and Rosa26 (control) knockouts in Pmel1 TCR T cells. Each knockout condition has three samples from three independent CRISPR guide RNAs. One parental sample is also included.