BRAF and KRAS Mutations in hyperplastic polyps and serrated adenomas of the colorectum: relationship to histology and CpG island methylation status.

The aim of this study was to test the hypothesis that mutations of the oncogenes BRAF or KRAS are early events in the putative serrated polyp neoplasia pathway and more advanced pathology is associated with acquired mutator and suppressor gene inactivation by CpG island methylation of promoter regions. We assayed 79 sporadic hyperplastic polyps (HPs) classified according to the schema of Torlakovic et al and 25 serrated adenomas (SAs) for BRAF and KRAS mutations and related the findings to histologic characteristics and CpG island methylation phenotype (CIMP). Mutations at exon 15, codon 599, of BRAF were assayed using an allele-specific PCR (AS-PCR) technique and confirmed in a sample of AS-PCR- positive cases by direct sequencing of exon 15. AS-PCR-negative HPs and SAs were also sequenced on exon 15 and exon 11 to detect additional mutations. PCR-RFLP was used to assay KRAS codon 12 and 13 mutations, and these mutations were further validated by direct sequencing of the KRAS gene. BRAF599 mutations were identified in a total of 55 HPs (69.6%) and KRAS mutations in a total of 13 (16.5%). BRAF599 mutations occurred with similar frequencies among microvesicular serrated polyp (76.3%) and serrated polyp with abnormal proliferation (82.1%) subtypes but less frequently in goblet cell serrated polyps (23.1%). Conversely, KRAS mutations were most frequent in goblet cell serrated polyp (46.2%) and less frequent in microvesicular serrated polyp (13.2%) and serrated polyp with abnormal proliferation (7.1%). BRAF599 and KRAS mutations were present in 15 (60.0%) and 7 (28.0%) of SAs, respectively. BRAF 599 mutation and KRAS were mutually exclusive findings in the polyps studied and one or the other occurred in 68 of 79 (86.1%) HPs and 22 of 25 (88.0%) SAs. CpG island methylation involving 2 or more genes (CIMP-H) was present in 80.0% of SAs, 75% serrated polyp with abnormal proliferations, 47.4% of microvesicular serrated polyps, and 15.4% of goblet cell serrated polyps. SAs were significantly more likely to be CIMP-H than HPs (odds ratio 3.7; 95% confidence interval, 1.27-10.86; P = 0.017). A similar high frequency of KRAS or BRAF mutations across the histologic spectrum of the serrated polyps assayed suggests that these are early events in the serrated polyp neoplasia pathway. In contrast, the association of higher levels of CpG island methylation with more advanced histologic changes suggests that CpG island methylation plays a role in serrated polyp progression toward colorectal carcinoma.