Integrating Bulk RNA-Seq and Spatial Transcriptomics Data to Reveal Distinct Immune Escape Mechanism and...

Post-translational modification (PTM) plays a crucial role in head and neck squamous cell carcinoma (HNSCC) progression, and their specific prognostic implications in HNSCC have not been thoroughly investigated. TCGA-HNSCC, GSE41613, GSE42743, and GSE65858 were merged into a meta-cohort, and 21 types of PTM were generated consensus cluster. Then, WGCNA was utilized to identify module genes. Finally, a machine learning approach was used to create the PTM.score. This analysis revealed 2 distinct subtypes of PTMs, each characterized by unique molecular signatures. By integrating different categories of genes, including DEGs, prognosis-related DEGs, module genes, and PTM-related genes, 13 hub genes were identified, and a PTM.score was developed. PTM.score was rigorously validated across 4 independent external cohorts and an in-house cohort, demonstrating its reliability and potential applicability. The PTM.score serves a dual purpose in its application, as it encapsulates the essential clinical context and offers valuable insights regarding the efficacy of immunotherapy treatments. In particular, patients categorized with a high PTM.score displayed a TME that was more actively engaged, which corresponded with a poor prognosis. Furthermore, these patients demonstrated a low level of responsiveness to immunotherapy interventions. In addition, an analysis utilizing spatial transcriptomics revealed that the PTM.score was markedly expressed within the tumor cells. This novel PTM-related prognostic signature could effectively assess the prognosis and therapeutic responses of HNSCC patients, providing new perspectives for individualized treatment for the patient population.