Genome-wide DNA methylation profiling of gastrointestinal neuroendocrine tumors to identify hypermethylation of mTOR, Wnt, and Notch pathways in GI NET pathogenesis.

Abstract

212 Background: Aberrant DNA methylation is known to play an important role in the pathogenesis of many human cancers, however little is known about its role in gastrointestinal neuroendocrine tumours (GI NET) development. We report the first unbiased genome-wide DNA methylation analysis of a large cohort of GI NETs, aiming to identify key methylation variable positions (MVPs) specific to GI NETs which may contribute to tumorigenesis and metastatic progression. Methods: Illumina Infinium Human Methylation 450 Array analysis was performed on 56 cases of GI NET DNA extracted from macrodissected tumour (n=67) and normal (n=29) specimens. Tumours were gastrointestinal primaries (n=39) or metastases (liver, mesenteric, omental or lymph node, n=28) of low (n=35), intermediate (n=17) or high grade (n=3)(unknown grade n=12). Data analysis was performed using the "ChAMP" custom pipeline and pathway analyses were performed using "GREAT," "WebGestalt," and "GSEA" web tools. A Bonferroni adjusted significance threshold value of p<0.05 was used throughout. Results: In order to identify and validate a GI NET specific methylation signature our cohort was divided into a discovery set (31 cases) and validation set (25 cases). Comparison of primary GI NET tumours with normal small bowel identified a total of 77,916 MVPs, including 1,666 sites hypermethylated by over 30% in tumour compared to normal tissue. Application of the profile to the validation set correctly identified 85% of samples. Tumours demonstrated global hypomethylation relative to normal tissue. Gene ontology analysis identified methylation of multiple cancer related pathways (including the Wnt, mTOR and Notch pathways) as a feature of hepatic metastases of GI NET primaries. Increasing RASSF1 promoter hypermethylation was associated with higher tumour grade. Conclusions: This study is the first comprehensive analysis of the epigenetic profile of GI NETs and identifies potential novel biomarkers and therapeutic targets. We are currently performing integrated analysis of epigenomic, genomic and transcriptomic data to further define the pathways involved in GI NET pathogenesis.