Prognostic Signature: Papillary Thyroid Carcinoma With Poor Prognosis Has Enhanced Immune Escape Capacity And High Tumor Heterogeneity

Abstract Background Papillary thyroid carcinoma (PTC) is considered an inflammation-driven cancer. However, a systematic investigation of the relationship between the tumor immune microenvironment and the prognosis of PTC has not been conducted. Methods A prognostic model based on differentially expressed genes (DEGs) and progression-free survival (PFS) data from The Cancer Genome Atlas (TCGA) was established by least absolute shrinkage and selection operator (LASSO) and multivariate Cox analyses. In total, 502 PTC cases were divided into low prognostic risk score (PR) (L-PR) and high PR (H-PR) groups according to the median PR. We then compared the immune characteristics between groups and verified these differences in five validation cohorts (GSE33630, GSE60542, GSE58545, GSE5364, and GSE27155). Furthermore, we explored cancer stem cells (CSCs) and the tumor mutation burden (TMB) to explain the prognostic results. Results A prognostic signature (PR) based on 13 DEGs performed well in prognostic prediction (5-year area under the curve (AUC) = 0.861). The PR was positively correlated with age, stage, T classification, metastasis, RAS mutation, and subtypes (follicular or tall cell PTC). Importantly, the H-PR group, which had poor prognostic features, exhibited four main characteristics: comprehensive weakening of the immune system that was not observed in the L-PR group, a higher ratio of tumor-promoting immune cells, more CSCs, and a higher TMB than the L-PR group. Gene set enrichment analysis (GSEA) results also showed the enrichment of immune-related pathways in the L-PR group. Conclusions Our prognostic model can effectively predict the prognosis and revealed that immune escape and tumor heterogeneity in the tumor microenvironment (TME) could be mechanisms of poor prognosis in PTC.