Pyrosequencing cut-off value identifying BRAFV600E mutation in fine needle aspiration samples of thyroid nodules

Context: Recently, tremendous efforts have been made towards the development of sensitive techniques to detect the BRAF(V600E) mutation in fine needle aspiration biopsy (FNAB) samples. However, newly developed quantitative and semi-quantitative methods, such as dual-priming oligonucleotide (DPO)-based multiplex polymerase chain reaction (PCR), have the potential to generate false-positive (FP) results.

Objectives: To eliminate the possibility of FP results, we generated a receiver operating characteristic (ROC) curve to investigate the diagnostic accuracy of pyrosequencing using quantitative data.

Design: Cytological diagnoses of 983 thyroid nodules were made according to the Bethesda System 2007. The BRAF(V600E) mutation was analysed by pyrosequencing, and statistical analyses were performed.

Results: Of the 983 nodules, 902 were adopted to evaluate the diagnostic value of pyrosequencing. The number of pathologically confirmed malignancies was 192, of which 182 were papillary thyroid cancer (PTC). By generating an ROC curve, we defined the optimal cut-off value of the mutant allele peak as 5·95% (area under the curve, 0·849; sensitivity, 0·55; 1-specificity, 0). When we applied this selective cut-off value, the number of PTCs positive for BRAF(V600E) was 99 (54·4% of the total number of PTCs). With cytology alone, the diagnostic sensitivity and specificity of detecting malignancy were 71·2% and 100%, respectively. Pyrosequencing improved the diagnostic sensitivity from 71·2% to 78·5% (McNemar's test, P < 0·001), without any change in the diagnostic specificity. When 'suspicious for malignancy' was considered a positive cytological outcome, pyrosequencing increased the diagnostic sensitivity of cytology from 95·8% to 96·9%; however, this improvement did not show statistical significance (McNemar's test, P > 0·05).

Conclusions: Pyrosequencing is an effective method for detecting the BRAF(V600E) mutation in FNAB samples. By allowing the optimal cut-off value to be determined, pyrosequencing improves the diagnostic sensitivity while eliminating the possibility of FP results.