Pilot of BRAF mutation analysis in indeterminate, suspicious and malignant thyroid FNA cytology

Abstract

BRAF V600E mutation has been reported to show a high specificity for papillary thyroid carcinoma (PTC). Using this marker to upgrade 'indeterminate' or 'suspicious' thyroid fine needle aspiration (FNA) cytology to 'malignant' could potentially allow one-stage therapeutic total thyroidectomy. For a 14-month period, FNA cytology specimens in the Thy3-5 categories, which are the UK equivalents of indeterminate (Thy3a, atypical; Thy3f, follicular), suspicious for malignancy (Thy4) and malignant (Thy5) in the Bethesda System, underwent BRAF mutation testing by melt curve analysis. The results were correlated with histology. We tested 123 cytology specimens of which 12 (9.8%) failed. The BRAF mutation rate in the remainder was 16.2% (18/111), with 93 showing the wild-type. Seventeen mutations were V600E and one was non-V600E. The rate of mutation increased significantly (P<0.0001 if Thy3a and Thy3f were combined) with the cytology category: 1/42 Thy3a (2.4%), 1/36 Thy3f (2.8%), 4/15 Thy4 (26.7%), 12/18 Thy5 (66.7%). All BRAF mutations correlated with PTC on histology, except for one recurrent PTC without histology. One mutation-positive case with Thy3a cytology showed the target lesion to be a 10-mm follicular adenoma on histology with an immediately adjacent 4-mm micro-PTC, in a patient who did not require total thyroidectomy. BRAF mutational analysis by melt curve analysis is feasible in routine thyroid cytology, and in our series had a 100% specificity for PTC in subsequent histology. The application of BRAF analysis could be useful for indeterminate cytology, but we suggest that it would be most appropriate and cost-effective for Thy4/suspicious cases, for which it could enable one-stage therapeutic surgery in the context of multidisciplinary discussion. In contrast, the sensitivity is low and there is no role for avoiding diagnostic thyroid surgery if wild-type BRAF is found.