Detection of the BRAF V600E mutation is required for use of the BRAF inhibitor, vemurafenib, in patients with metastatic melanoma. Although the Roche Cobas 4800 BRAF V600 Mutation Test is approved, it detects primarily the single-nucleotide V600E mutation and could miss other potentially relevant V600 mutations. To assess the detection rate of the cobas assay for V600 mutations in clinical specimens, we compared the results of this assay with Sanger sequencing in 295 melanoma FFPE samples. Twenty samples were excluded because of invalid results on the cobas (n = 3), sequencing (n = 15), or both (n = 2). V600 mutations were detected by the cobas test in 96 (34.9%) of 275 samples and by Sanger sequencing in 118 (42.9%) of 275 samples. Thus, relative to Sanger sequencing, the cobas test exhibited 80.5% sensitivity (95% CI, 72.4% to 86.6%) and 99.4% specificity (95% CI, 96.5% to 99.9%). Of 23 samples with positive sequencing results but negative cobas results, 21 harbored dinucleotide mutations (V600E in 6, V600K in 10, and V600R in 5); the other two involved single-nucleotide mutations (V600E and V600G). These findings indicate that the cobas assay may miss many V600 mutations in clinical specimens. In our study, the addition of Sanger sequencing for samples with negative cobas results increased the detection rate to 42.9%. This approach could help maximize the number of patients who benefit from BRAF inhibitor treatment.
Copyright © 2013 American Society for Investigative Pathology and the Association for Molecular Pathology. Published by Elsevier Inc. All rights reserved.