Identification of a Lactate Accumulation Model to Explain the Heterogeneity in Prognosis, Immune Landscape, and Tumor Environment for HNSCC patients.

Abstract

Head and neck squamous cell carcinoma (HNSCC) is one of the most frequent cancers with a high mortality rate. Lactate accumulation, a hallmark of cancer, has received extensive attention, but its role in HNSCC remains underexplored. Therefore, we identified 33 prognostic genes related to lactate accumulation. By consensus clustering, we separated all HNSCC samples into cluster_A or cluster_B and explored the difference of clinicopathological characteristics and genomics landscape. Next, we performed LASSO analysis and RSF to calculate the lactate-related gene score (LRGS) and constructed a risk model with high accuracy for predicting survival, as estimated by ROC, nomogram, and calibration curve. Then, through OncoPredict algorithm and TCIA, we filter the suitable drugs, especially immunology with diverse LRGS. GSEA analysis showed that the DEGs of LRGS were enriched in activation of immune response and positive regulation of immune response. Moreover, we developed a tumor-infiltrating immune-related lncRNA signature (TILSig) through a combination of 115 immune cell lines from 16 GEO datasets and DealGPL570. Subsequently, we identified the 9 tumor-infiltrating immune-related lncRNAs and calculated the TIL_score. The correlations among these tumor-infiltrating immune-related lncRNAs, hub lactate-related genes and LRGS levels were visualized. According to validation using multiple datasets including TCGA, GSE65858, GSE41613, GSE27020, and the IMvigor 210 database, CARS2, NFU1, and SYNJ1 were identified as hub genes. In light of a comprehensive pan-cancer study, we analyzed these genes to detect the potential clinical value. In conclusion, the constructed LRGS provides important insights for subsequent mechanistic research and can guide clinicians in proposing therapeutic strategies for HNSCC patients.