Novel T cell exhaustion gene signature to predict prognosis and immunotherapy response in thyroid carcinoma from integrated RNA-sequencing analysis.

Exhausted CD8 T lymphocytes and tumor-associated macrophages play critical roles in determining cancer prognosis and the efficacy of immunotherapy. Our study revealed a negative correlation between exhausted CD8 T lymphocytes and prognosis in thyroid carcinoma (THCA). Consensus clustering divided patients into two subgroups of exhaustion with different prognoses, as defined by marker genes of exhausted CD8 T cells. Subsequently, we constructed an eight-gene prognostic signature, and developed a risk score named the exhaustion-related gene score (ERGS) to forecast both prognosis and immunotherapy response in THCA. Bulk RNA sequencing analysis revealed a higher prevalence of M2 macrophages, indicative of an immunosuppressive tumor microenvironment (TME), in the high-ERGS group. Single-cell RNA sequencing showed that SPP1 macrophages and CD14 monocytes infiltrations were positively associated with higher ERGS. Functionally, it was determined that SPP1 macrophages exert an immunosuppressive role, while CD14 monocytes were implicated in promoting tumor progression and angiogenesis. Analysis of cell-cell interactions between SPP1 macrophages and T cells highlighted the activation of the SPP1-CD44 and MIF-CD74 axes, both of which could foster an immunosuppressive TME. Therapeutic strategies that target SPP1 macrophages, CD14 monocytes, and the SPP1-CD44 and MIF-CD74 axes may potentially improve the prognosis and amplify the immunotherapy response in THCA patients.