BRAF-mutated carcinomas among thyroid nodules with prior indeterminate FNA report: a systematic review and meta-analysis.

Background: Several molecular analyses have been investigated for risk stratification of thyroid nodules, with a particular focus on the V600E mutation of the BRAF gene [BRAF(V600E)]. To date, there is no high-level evidence supporting or refuting a role for BRAF analysis in thyroid nodules with prior indeterminate cytology. To obtain more robust evidence, we reviewed and meta-analysed data from published articles.

Research Design And Methods: A comprehensive literature search of the PubMed/MEDLINE, Embase and Scopus databases was conducted using the terms 'BRAF', 'thyroid' and 'indeterminate'. The search was updated until March 2015, and references of the retrieved articles were also screened. Only original articles reporting BRAF mutation testing within nodules with indeterminate FNA were eligible for inclusion.

Results: The literature search revealed 82 articles, of which 8 were eligible for the study. Five studies were prospective and three retrospective. The majority of authors analysed BRAF mutations in FNA samples which were classified by the British or Bethesda system. Of the initial series of studies, a pooled number of 1361 cases were achieved of which 43 were BRAF mutated. Overall, the BRAF mutation rate was 4·6% (95% CI: 1-10·8%), ranging from 0 to 22·9%. When we included only histological series, 978 thyroid nodules were found. Of these, 245 were cancers.

Conclusions: A very low rate of lesions with indeterminate cytology are BRAF mutated. Thus, the role of this biomarker to detect or exclude cancers in patients with such FNA reports is marginal and should be reconsidered in guidelines.