TSTScope Unifies Single-Cell Multi-Omics to Identify Functional T Cell States Predictive of Immunotherapy Response.

Immune checkpoint blockade (ICB) can produce durable responses in cancer, but reliable predictors of benefit are still lacking. CD8+ tumor-specific T cells (TSTs) are essential for ICB efficacy, yet it remains unclear which functional states of these cells are associated with therapeutic benefit. To address this, we developed TSTScope, an interpretable deep learning framework that integrates single-cell transcriptomic and T-cell receptor sequencing data to generate unified representations of CD8+ T-cell identity. By applying TSTScope to non-small cell lung cancer (NSCLC) datasets, we characterized the gene programs defining tumor specificity and computationally inferred a population of potential TSTs (pTSTs). Our analyses show that clinical response is associated with the functional state of these cells rather than their abundance alone. We derived the major pathological response (MPR) score, a metric capturing this functional potential. In an independent validation cohort, the MPR score was associated with pathological response and recurrence-free survival and provided complementary information to selected response-associated biomarkers. Collectively, TSTScope identifies a distinct functional state of tumor-specific T cells linked to ICB response, providing an interpretable framework for studying receptor-linked T-cell function in immunotherapy cohorts.