Effect of the m6ARNA gene on the prognosis of thyroid cancer, immune infiltration, and promising immunotherapy

Accumulating evidence suggests that N6-methyladenosine (m6A) RNA methylation plays an important role in tumor proliferation and growth. However, its effect on the clinical prognosis, immune infiltration, and immunotherapy response of thyroid cancer patients has not been investigated in detail.

Our risk model can act as an independent marker for thyroid cancer and provides promising immunotherapy targets for its treatment.

Clinical data and RNA expression profiles of thyroid cancer were extracted from the Cancer Genome Atlas-thyroid carcinoma (TCGA-THCA) and preprocessed for consensus clustering. The risk model was constructed based on differentially expressed genes (DEGs) using Least Absolute Shrinkage and Selection Operator (LASSO) and Cox regression analyses. The associations between risk score and clinical traits, immune infiltration, Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), Gene Set Enrichment Analysis (GSEA), immune infiltration, and immunotherapy were assessed. Immunohistochemistry was used to substantiate the clinical traits of our samples.

Gene expression analysis showed that 17 genes, except YHTDF2, had significant differences (vs healthy control, P<0.001). Consensus clustering yielded 2 clusters according to their clinical features and estimated a poorer prognosis for Cluster 1 (P=0.03). The heatmap between the 2 clusters showed differences in T (P<0.01), N (P<0.001) and stage (P<0.01). Based on univariate Cox and LASSO regression, a risk model consisting of three high-risk genes (KIAA1429, RBM15, FTO) was established, and the expression difference between normal and tumor tissues of three genes was confirmed by immunohistochemical results of our clinical tissues. KEGG and GSEA analyses showed that the risk DEGs were related mainly to proteolysis, immune response, and cancer pathways. The levels of immune infiltration in the high- and low-risk groups were different mainly in iDCs (P<0.05), NK cells (P<0.05), and type-INF-II (P<0.001). Immunotherapy analysis yielded 30 drugs associated with the expression of each gene and 20 drugs associated with the risk score. Conclusions: Our risk model can act as an independent marker for thyroid cancer and provides promising immunotherapy targets for its treatment.