Abstract
BACKGROUND: Molecular testing is recommended to refine risk stratification in indeterminate thyroid nodules (Bethesda III-IV), but data on dual-gene (BRAF and RAS) testing using fresh FNA washout specimens are limited. We aimed to evaluate the performance of BRAF and RAS mutation analysis from fresh thyroid FNA washout material, with a focus on indeterminate cytology. METHODS: We retrospectively analyzed 1139 patients who underwent washout-based molecular testing between May 2022 and October 2024 at a tertiary endocrine center. Of these, 307 had available histopathologic results after surgery. Primary outcomes were sample adequacy, mutation spectrum, and diagnostic metrics (sensitivity, specificity, PPV, NPV, and accuracy). Analyses were repeated under two assumptions that classified borderline/low-risk neoplasms as benign vs. malignant, and within the Bethesda III-IV subset. RESULTS: Adequate material for molecular analysis was obtained in 1037/1139 samples (90.9%). In the operated cohort (n = 307), malignant lesions comprised 31.9% and low-risk neoplasms 8.5%. When borderline lesions were considered benign, mutation positivity yielded a sensitivity of 48.0%, a specificity of 89.6%, a PPV of 75.9%, an NPV of 71.9%, and an accuracy of 72.9%. In Bethesda III-IV nodules (n = 153), sensitivity, specificity, and accuracy were 41.0%, 85.2%, and 66.0% (malignant assumption). Isolated BRAF positivity showed high specificity (~96.7%) with modest sensitivity. CONCLUSIONS: Our findings extend current diagnostic approaches by showing that dual-gene (BRAF and RAS) testing from fresh FNA washouts is technically feasible (≥90% adequacy) and provides high specificity with modest sensitivity for malignancy in indeterminate nodules. In settings lacking comprehensive commercial panels, this low-complexity approach offers a practical adjunct to cytology and imaging for preoperative decision-making.