Next-Generation Sequencing Identifies a Highly Accurate miRNA Panel That Distinguishes Well-Differentiated Thyroid Cancer from Benign Thyroid Nodules.

Mazeh H#1, Deutch T#2, Karas A1, Bogardus KA3, Mizrahi I1, Gur-Wahnon D2, Ben-Dov IZ4.

Author information

Abstract

Background: Fine needle aspiration biopsy (FNAB) is the gold-standard procedure for diagnosing malignant thyroid nodules. Indeterminate cytology is identified in 10% to 40% of cases, and molecular testing may guide management in this setting. Current commercial options are expensive, and are either sensitive or specific. The aim of this study was to utilize next-generation sequencing (NGS) technology to identify informative diversities in the miRNA expression profile of benign versus malignant thyroid nodules.Methods:Ex vivo FNAB samples were obtained from thyroid specimens of patients who underwent thyroidectomy at a referral center. miRNA levels were determined using NGS and multiplexing technologies. Statistical analyses identified differences between normal and malignant samples and miRNA expression profiles that associate with malignancy were established. The accuracy of the miRNA signature in predicting histologic malignancy was validated using a group of patient specimens with indeterminate cytology results.Results: A total of 274 samples were obtained from 102 patients undergoing thyroidectomy. Of these samples, 71% were benign and 29% were malignant. Nineteen miRNAs were identified as statistically different between benign and malignant samples and were used to classify 35 additional nodules with indeterminate cytology (validation). The miRNA panel's sensitivity, specificity, negative and positive predictive values, and overall accuracy were 91%, 100%, 87%, 100%, and 94%, respectively.Conclusions: Using NGS technology, we identified a panel of 19 miRNAs that may be utilized to distinguish benign from malignant thyroid nodules with indeterminate cytology.Impact: Our panel may classify indeterminate thyroid nodules at higher accuracy than commercially available molecular tests. Cancer Epidemiol Biomarkers Prev; 27(8); 858-63. ©2018 AACR.