Biomarkers Screening For Early Diagnosis Of Thyroid Carcinoma And Primary Cytological Validation

Background and PurposeThyroid gland is a butterfly-shaped gland located in the neck.It secretes a variety of hormones and exerts physiological functions such as regulating blood pressure,body temperature,heart rate,metabolism,and body weight.Thyroid carcinoma（TC）is one of the most common endocrine malignancies.In recent years,the incidence of thyroid carcinoma has increased significantly in China and worldwide.At present,fine needle aspiration cytology（FNAC）is a standardized clinical procedure recommended by the American Thyroid Association,and it is the most accurate and economical method for clinical evaluation of thyroid nodules.However,in fine-needle aspiration biopsies,cytological results for approximate 30%of cases are uncertain,which require partial or complete thyroidectomy and further histopathological examination.It is still challenge to establish accurate early diagnosis methods in clinic to determine the benign and malignant nodules.It is of great significance to discover and screen new biomarkers of thyroid carcinoma with potential application value to improve the accuracy of early diagnosis of thyroid carcinoma patients,especially for the preoperative evaluation of patients with uncertain cytology results.MethodIn this study,the WGCNA method was used to analyze the gene expression profile data of 265 samples with cytology indeterminate from the NCBI expression profile database,so as to obtain co-expression modules significantly related to malignant histological characteristics.The results were further validated in 3 independent data sets derived from the cBioPortal Cancer Case Genomics Database.At the same time,this research used gene ontology（GO）functional annotation and KEGG pathway enrichment analysis methods to analyze the tumorigenesis signal pathways involved in related genes.The genes found in early stage of thyroid cancer analysis were verified by qRT-PCR and Western Blot method.The risk factors derived from clinical information of 84 thyroid patients were analyzed.A model for screening high-risk patients from these risk factors was established.Result1)The blue module with the highest correlation（cor=0.77）and significance(p=6.8e^-6)with histopathological malignant features identified by WGCNA.2)The 24 genes were verified in original samples and 3 independent data sets with larger sample sizes.It was found that 15 genes（CC2D2B,CFH,CITED1,FN1,G0S2,GABRB2,KRT19,TENM1,PPP2R2B,PROS1,RXRG,SCEL,SERGEF,SLC34A2 and STK32A）were significantly overexpressed in thyroid tumors.Among them,5 genes with specific point mutations,specifically FN1（R534P）,PROS1（K200I,Q571K）,SCEL（T320S）,SLC34A2（T688M）and TENM1（S1131F）,were highlighted as potential biomarkers for early diagnosis of thyroid carcinoma.3)GO analysis and KEGG pathway analysis was performed on genes in the blue module.Genclip analysis revealed that these genes are mainly involved in pathways including the physiological responses of hormones and steroid hormones,regulation of cell migration and adhesion,and cell connection interaction.These pathways are mainly focused on the occurrence of cancers such as thyroid cancer,small/non-small cell lung cancer,signal transduction,extracellular regions,and transporter activity.4)The transcription levels of 10 genes（FN1,PROS1,TENM1,SCEL,SLC34A2,CITED1,GABRB2,KRT19,RXRG and STK32A）were verified by qRT-PCR.Compared with normal thyroid cells,the transcription levels of 10 genes in thyroid cancer cells were increased.5)Western Blot results showed that SLC34A2 was significantly overexpressed in thyroid cancer cell lines,compared with normal thyroid cells.6)Analysis of clinical data with small sample sizes showed that women in the group of 50⁵4 age are susceptible to thyroid carcinoma,but there is no significant differences in age and maximum tumor diameters among low-risk and high-risk thyroid patients.