Enhancing prognostic accuracy in head and neck squamous cell carcinoma chemotherapy via a lipid metabolism-related clustered polygenic model

Abstract

ObjectiveSystemic chemotherapy is the first-line therapeutic option for head and neck squamous cell carcinoma (HNSCC), but it often fails. This study aimed to develop an effective prognostic model for evaluating the therapeutic effects of systemic chemotherapy.MethodsThis study utilized CRISPR/cas9 whole gene loss-of-function library screening and data from The Cancer Genome Atlas (TCGA) HNSCC patients who have undergone systemic therapy to examine differentially expressed genes (DEGs). A lipid metabolism-related clustered polygenic model called the lipid metabolism related score (LMRS) model was established based on the identified functionally enriched DEGs. The prediction efficiency of the model for survival outcome, chemotherapy, and immunotherapy response was evaluated using HNSCC datasets, the GEO database and clinical samples.ResultsScreening results from the study demonstrated that genes those were differentially expressed were highly associated with lipid metabolism-related pathways, and patients receiving systemic therapy had significantly different prognoses based on lipid metabolism gene characteristics. The LMRS model, consisting of eight lipid metabolism-related genes, outperformed each lipid metabolism gene-based model in predicting outcome and drug response. Further validation of the LMRS model in HNSCCs confirmed its prognostic value.ConclusionIn conclusion, the LMRS polygenic prognostic model is helpful to assess outcome and drug response for HNSCCs and could assist in the timely selection of the appropriate treatment for HNSCC patients. This study provides important insights for improving systemic chemotherapy and enhancing patient outcomes.