MicroRNA profile for diagnostic and prognostic biomarkers in thyroid cancer

Abstract

The challenge in managing thyroid nodules is to accurately diagnose the minority of those with malignancy. We aimed to identify diagnostic and prognostic miRNA markers for thyroid nodules. In a discovery cohort, we identified 20 candidate miRNAs to differentiate between noninvasive follicular thyroid neoplasms with papillary-like nuclear features (NIFTP) and papillary thyroid carcinomas (PTC) by using the high-throughput small RNA sequencing method. We then selected three miRNAs (miR-136, miR-21, and miR-127) that were differentially expressed between the PTC follicular variant and other variants in The Cancer Genome Atlas data. High expression of three miRNAs differentiated thyroid cancer from nonmalignant tumors, with an area under curve (AUC) of 0.76-0.81 in an independent cohort. In patients with differentiated thyroid cancer, the high-level expression of the three miRNAs was an independent indicator for both distant metastases and recurrent or persistent disease. In patients with PTC, a high expression of miRNAs was associated with an aggressive histologic variant, extrathyroidal extension, distant metastasis, or recurrent or persistent disease. Three miRNAs may be used as diagnostic markers for differentiating thyroid cancers from benign tumors and tumors with extremely low malignant potential (NIFTP), as well as prognostic markers for predicting the risk of recurrent/persistent disease for differentiated thyroid cancer.