Ultrasound features suspicious for malignancy predict the risk of BRAF mutation in papillary thyroid carcinoma

Purpose: This study aimed to evaluate the BRAF mutation status in papillary thyroid carcinoma (PTC) on the basis of ultrasound (US) features suspicious for malignancy. Methods: The clinical, pathologic, and radiologic characteristics of patients who were diagnosed with PTC from August 2014 to August 2015 were reviewed. A radiologist who was blinded to the BRAF status, independently reviewed all existing preoperative US examinations and recorded the sonographic characteristics for each PTC case. The US features included echogenicity, composition, height and width, margin, shape, capsule, vascularity, calcification, and number. Univariate and multivariate analyses were performed to determine sonographic predictors. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for each US characteristic that was suspicious for malignancy were calculated. The diagnostic accuracies for predictors of BRAF mutation were calculated via receiver operating characteristic analysis. Results: There was a significant association between echogenicity, composition, height, width, margin, shape, and calcification and BRAF mutation status. However, no association was observed between capsule, vascularity, and number and BRAF mutation status. Poorly defined margins and microcalcification were independent predictors of BRAF mutation via multivariate analysis. PTC with poorly defined margins predicted BRAF mutations with 86.4% sensitivity, 55.0% specificity, and 76.6% accuracy, while microcalcification predicted BRAF mutation with 84.1% sensitivity, 50.0% specificity, and 73.4% accuracy. Conclusions: Poorly defined margins and preoperative microcalcification could effectively predict the BRAF mutation status and provide a valuable guideline for aggressive treatment in PTC. Further investigations should be performed to understand the association between US features and BRAF mutation.