A gene expression profile (GEP) based on IFN-γ-related mRNA levels predicts clinical response to PD-1 blockade. Researchers analyzed RNA from tumor samples of patients treated with pembrolizumab and identified immune-related gene signatures associated with clinical benefit. A 10-gene "preliminary IFN-γ" signature was developed, which could separate responders and non-responders in melanoma patients. This was validated in a larger cohort of 62 melanoma patients, leading to the development of a 28-gene "preliminary expanded immune" signature. Further refinement of these signatures across multiple cancer types resulted in an 18-gene T cell-inflamed GEP that predicted response to PD-1 blockade in 9 different cancer types. This GEP was compared with PD-L1 immunohistochemistry (IHC) in 96 patients with head and neck squamous cell carcinoma (HNSCC) and showed comparable performance. The GEP was found to be robust and predictive of clinical outcomes, with high discriminatory ability. The T cell-inflamed GEP was developed into a clinical-grade assay for use in ongoing pembrolizumab trials. The study highlights the importance of IFN-γ signaling and T cell biology in predicting response to PD-1 blockade, and suggests that a T cell-inflamed microenvironment is necessary but not sufficient for clinical benefit. The results indicate that IFN-γ-related gene signatures may offer a more robust predictive biomarker than PD-L1 IHC in PD-L1-unselected populations. The study also suggests that distinct resistance mechanisms may be at play in tumors lacking T cell inflammation, and that further research is needed to understand the biology of resistance to PD-1 checkpoint blockade. The findings support the use of multigene signatures to predict clinical response to PD-1 blockade and highlight the potential of these signatures as a diagnostic tool in clinical trials.A gene expression profile (GEP) based on IFN-γ-related mRNA levels predicts clinical response to PD-1 blockade. Researchers analyzed RNA from tumor samples of patients treated with pembrolizumab and identified immune-related gene signatures associated with clinical benefit. A 10-gene "preliminary IFN-γ" signature was developed, which could separate responders and non-responders in melanoma patients. This was validated in a larger cohort of 62 melanoma patients, leading to the development of a 28-gene "preliminary expanded immune" signature. Further refinement of these signatures across multiple cancer types resulted in an 18-gene T cell-inflamed GEP that predicted response to PD-1 blockade in 9 different cancer types. This GEP was compared with PD-L1 immunohistochemistry (IHC) in 96 patients with head and neck squamous cell carcinoma (HNSCC) and showed comparable performance. The GEP was found to be robust and predictive of clinical outcomes, with high discriminatory ability. The T cell-inflamed GEP was developed into a clinical-grade assay for use in ongoing pembrolizumab trials. The study highlights the importance of IFN-γ signaling and T cell biology in predicting response to PD-1 blockade, and suggests that a T cell-inflamed microenvironment is necessary but not sufficient for clinical benefit. The results indicate that IFN-γ-related gene signatures may offer a more robust predictive biomarker than PD-L1 IHC in PD-L1-unselected populations. The study also suggests that distinct resistance mechanisms may be at play in tumors lacking T cell inflammation, and that further research is needed to understand the biology of resistance to PD-1 checkpoint blockade. The findings support the use of multigene signatures to predict clinical response to PD-1 blockade and highlight the potential of these signatures as a diagnostic tool in clinical trials.