BRAF p.V600 mutation detection recently became necessary to treat metastatic melanoma patients with vemurafenib. This study compares different methods of detection of BRAF mutations. Melanoma samples from 111 patients were analyzed for BRAF mutations, and for 89 of them, results were obtained with the four following methods: Sanger sequencing, real-time PCR, immunohistochemistry, and pyrosequencing. All samples contained at least 60% of tumor cells. Directional Sanger sequencing of PCR products failed to detect 3 of 40 p.V600E-mutated cases (7.5%) (sensitivity, 92.5%; 95% CI, 78.5% to 98.0%). BRAF p.V600E-specific real-time PCR identified 39 of 40 p.V600E-mutated cases (97.6%) (sensitivity, 97.5%; 95% CI, 87.1% to 99.6%) and all 39 wild-type (WT) cases and surprisingly was also positive for 6/6 p.V600K (specificity, 87.8%; 95% CI, 75.8% to 94.3%). However, other mutations, p.V600R (n = 1), p.K601E (n = 2), and p.600_601delinsE (n = 1), were not detected. Immunohistochemistry with VE1, specific for p.V600E, identified all p.V600E and WT cases (sensitivity, 100%; 95% CI, 91.2% to 100%) but was negative for all other BRAF mutations. Pyrosequencing successfully identified all WT and mutated cases. Immunohistochemistry is highly specific for p.V600E, and could be used as a first-line method, as is currently performed for HER2 amplification detection. Pyrosequencing proved to be the most efficient method to detect BRAF mutations in melanomas and could be performed on VE1-negative or uninterpretable cases.

BRAF (V600) status could be used to stage melanoma patients with nodal deposits. Our results may also help to plan adjuvant trials in these patients, for whom the low tumor load may induce longer efficacy of BRAF-targeted therapies.

BackgroundBRAF mutations are present in 40 % of human skin melanomas. Mutated tumors with an increased percentage of BRAF mutant alleles (BRAF-M%) may have a better response to RAF/MEK inhibitors. We evaluated the BRAF-M% in melanomas, and the genetic causes of its variation.MethodsBRAF-M% was quantified by pyrosequencing, real-time PCR (rtPCR) and/or picoliter-droplet PCR (dPCR). BRAF mutant expression was detected by immunohistochemistry. Chromosomal alterations were analyzed with fluorescence in situ hybridization (FISH), and single nucleotide polymorphism (SNP) arrays.ResultsBRAF-M% quantification obtained with pyrosequencing was highly correlated (R = 0.94) with rtPCR, and with dPCR. BRAF-M% quantified from DNA and RNA were also highly correlated (R = 0.98). Among 368 samples with >80 % tumor cells, 38.6 % had a BRAFV600E mutation. Only 66.2 % cases were heterozygous (BRAF-M% 30 to 60 %). Increased BRAF-M% (>60 %) was observed in 19 % of cases. FISH showed a polysomy of chromosome 7 in 13.6 %, 35.3 % and 54.5 % of BRAF wild-type, heterozygous and non-heterozygous BRAF-mutated samples, respectively (P < 0.005). Amplification (5.6 %) and loss (3.2 %) of BRAF locus were rare. By contrast, chromosome 7 was disomic in 27/27 BRAF-mutated nevi.ConclusionsBRAF-M% is heterogeneous and frequently increased in BRAF-mutant melanomas. Aneuploidy of chromosome 7 is more frequent in BRAF mutant melanomas, specifically in those with high BRAF-M%.Electronic supplementary materialThe online version of this article (doi:10.1186/s12885-015-1515-3) contains supplementary material, which is available to authorized users.