Background: Since the increasing incidence of Hashimoto’s thyroiditis (HT) combined with papillary thyroid carcinoma (PTC), the relationship between HT and PTC has become an interesting research area. Many studies have confirmed that HT is a risk factor of PTC and can increase the incidence of PTC. On the other hand, HT has a protective effect on PTC. PTC coexisting with HT patients have several characters including that more likely to be female, tumors are more common in microcarcinomas, and multifocal, but the incidence of capsule invasion and lymph node metastasis is low[5]. At present, the pathogenesis of HT with PTC is not clear, but there are complex immunological associations and common molecular biological changes between them.

Previously, several molecular markers have been confirmed to be associated with papillary thyroid carcinoma with Hashimoto’s thyroiditis, including BRAF mutation, P63 protein, and Cytokeratin-19 (CK19), etc. BRAFV600E mutation is one of the most common mutations in PTC, with an incidence of 60.6%. BRAFV600E mutation was also found in PTC combined with HT. But compared with PTC patients without HT, the incidence of BRAFV600E mutations was lower. While the sensitivity of P63 in the diagnosis of PTC is low, CK19 is more sensitive than P63 in diagnosing PTC, but its specificity is lower. In this study, for reveal the molecular mechanism between PTC and HT we we try to find new molecular biomarkers by identifying differentially expressed genes (DEGs).

Methods: GSE138198 dataset was downloaded from the Gene Expression Omnibus (GEO) database. The common differentially expressed genes (DEGs) both highly expressed in PTC with or without HT were screened by the GEO2R tool. Gene ontology (GO) and KEGG pathway enrichment analysis was applied for DEGs. Then three hub genes were calculated by the protein-protein interaction (PPI) network which was constructed through the Retrieval of Interacting Genes (STRING) database and Cytoscape software. Finally, hub genes were validated in the Tumor Genome Atlas (TCGA) database and their relationship to the clinical characteristics of PTC patients was analyzed. Gene set enrichment analysis (GSEA) was also performed.

Results: A total of 174 common DEGs were identified. Three hub genes were filtered out that were highly expressed in PTC and PTC with HT tissues, including ITGA3, MET, and LAMB3. ROC curves showed better diagnostic values of hub genes for PTC. The AUC of ITGA3, MET, and LAMB3 were 0.914, 0.91 and 0.877, respectively, indicating their potential diagnostic values (all p<0.0001). ITGA3 was associated with the TNM stage, tumor size and lymph node metastasis (p=0.023, p=0.005, and p<0.001 respectively). However, the expression levels of MET were only related to tumor size and lymph node metastasis (p=0.005, and p<0.001). As to LAMB3, the high expression indicates large tumor, lymph node metastasis, and multifocal lesions (p=0.001, p=0.007, and p=0.025 respectively). Besides, GSEA results revealed high expression of hub genes in PTC tissues was related to cancer-related pathways and autoimmune thyroid disease pathways.

Conclusion:Our study identified three hub genes which all upregulation in PTC with or without HT. the upregulated expression of hub genes involved in the genesis and development of PTC and associated with HT.

The ITGA3 gene encodes the integrin subunit alpha 3 which a member of the integrin alpha chain family of proteins. Integrins are transmembrane receptors that mediate adhesion to several extracellular matrix(ECM) components and cell-to-cell interactions. integrins organize the cytoskeleton and regulate complex cellular behaviors, including proliferation, migration, and survival by activating intracellular signaling. A previous study has shown that ITGA3 may promote cancer progression by activating the PI3K-Akt pathway. In this study, we analyzed potential functions of ITGA3 upregulated expression in PTC. The result indicated that ITGA3 was associated with “DNA REPLICATION”, “Jak-STAT signaling pathway”, “p53 signaling pathway” pathways. The activation of Janus Kinase/Signal Transducer and Activator of Transcription (JAK/STAT) signal pathway is involved in the occurrence and development of PTC. Many studies have confirmed the expression of STAT3 in PTC tissues. ITGA3 may promote the occurrence and development of PTC through the above pathways. Besides, ITGA3 has also correlated with “AUTOIMMUNE THYROID DISEASE” pathway.

LAMB3 encodes the β3 subunit of LM-332. LMs as extracellular glycoproteins are important components of the basement membrane zone and are involved in tumor development. The expression level of LAMB3 has been identified as upregulated in PTC, meanwhile, LAMB3 regulates epithelial-mesenchymal transition(EMT)-related proteins by activating the c-MET/Akt signaling pathway, thereby promoting tumor invasion and metastasis. MET encodes the receptor tyrosine kinase which is a high-affinity receptor for hepatocyte growth factor (HGF). Dysregulation of HGF-Met signal transduction is associated with multiple cancers. In PTC, MET overexpression is associated with a higher risk of metastasis, possibly due to constitutive activation caused by MET overexpression. MET activation further leads to PI3K/Akt activation.

Besides, results of ROC curves showed that hub genes have better diagnostic values. The overexpression of hub genes related to tumor size and lymphnode metastasis of PTC patients. It can be seen that ITGA3, LAMB3, MET have reference values for the diagnosis and disease assessment of PTC. However, our study lacks HT patients as controls, and more PTC patients with HT need to be included for further verification.

In conclusion, we identified common DEGs in PTC only and PTC with HT patients by using the GEO dataset. Finally, three hub genes which may be the potential biomarkers of PTC combined with HT were explored. Data from TCGA were used to identify the diagnostic value and clinical significance of hub genes for PTC. ITGA3, LAMB3, and MET as potential biomarkers may be useful for the diagnosis, treatment and prognosis of PTC with HT. However, further molecular and biological experiments are needed to identify the function of hub genes.