Identification of key pathways and biomarkers in anaplastic thyroid cancer using an integrated analysis

Thyroid carcinoma (THCA) is one of the most common endocrine tumours with high morbidity worldwide. Anaplastic thyroid cancer (ATC) is the most fatal and has the poorest prognosis of the four THCA types, as it lacks effective treatments. Early screening of ATC is problematic and so identifying ATC biomarkers is increasingly crucial.

We performed a systematic search of the thyroid transcriptome in the Gene Expression Omnibus (GEO) database and an integrative analysis of gene expression profiles. Moreover, we conducted a pathway enrichment analysis in ATC using the WEB-based GEne SeT AnaLysis Toolkit. We identified the intersections of all the differentially expressed genes (DEGs) between ATC and normal samples and DEGs between ATC and non-ATC samples in the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING). Finally, we used Cytoscape software to visualize the protein-protein interaction (PPI) network.

Six gene expression datasets containing 131 thyroid cancer samples and 98 normal control samples were collected to identify the significant DEGs. A total of 1489 DEGs were identified between ATC and normal samples, and 522 DEGs between ATC and non-ATC samples. ATC showed a greater association with the cell cycle. The Principal component analysis (PCA) results revealed 222 genes with substantial contributions to the identification of ATC.

Cell cycle plays a decisive role in the high mortality rate of ATC. TOP2A, NUSAP1, PBK, KIF15, CENPF, CEP55, CDK1, CCNB2, CDCA8 and CDC20 were identified as hub genes.