Prognostic Protein Biomarker Screening for Thyroid Carcinoma Based on Cancer Proteomics Profiles

Abstract

Thyroid carcinoma (THCA) ranks among the most prevalent cancers globally. Integrating advanced genomic and proteomic analyses to construct a protein-based prognostic model promises to identify effective biomarkers and explore new therapeutic avenues. In this study, proteomic data from The Cancer Proteomics Atlas (TCPA) and clinical data from The Cancer Genome Atlas (TCGA) were utilized. Using Kaplan–Meier, Cox regression, and LASSO penalized Cox analyses, we developed a prognostic risk model comprising 13 proteins (S100A4, PAI1, IGFBP2, RICTOR, B7-H3, COLLAGENVI, PAR, SNAIL, FAK, Connexin-43, Rheb, EVI1, and P90RSK_pT359S363). The protein prognostic model was validated as an independent predictor of survival time in THCA patients, based on risk curves, survival analysis, receiver operating characteristic curves and independent prognostic analysis. Additionally, we explored the immune cell infiltration and tumor mutational burden (TMB) related to these features. Notably, our study proved a novel approach for predicting treatment responses in THCA patients, including those undergoing chemotherapy and targeted therapy.