DNA methylation biomarkers for esophageal adenocarcinoma and precursor disease
Esophageal adenocarcinoma has one of the poorest outcomes of all solid tumors, attributable, at least in part, to lack of an early stage diagnostic test. Aberrant methylation is an early and frequent event in carcinogenesis providing an opportunity for early cancer detection. The overall aim of this thesis is to identify and validate regions of aberrant methylation as a biomarker for early detection of esophageal adenocarcinoma and the dysplastic stages of its precursor disease, Barrett’s esophagus. By using well-classified patient data and stringent, quality controlled biospecimen selection for training and validation cohorts, I found regions of diseaseassociated aberrant methylation that are novel for esophageal carcinogenesis. With comprehensive technical and independent validation by targeted amplicon sequencing and whole genome methylation profiling of a large external validation cohort, I demonstrated potential utility of these target regions for identification of intervention requiring disease. For subsequent blood investigation, all target regions are unmethylated in peripheral blood from healthy patients and amplification assays for targeted sequencing are suitable for degraded, shorter fragments of cell-free circulating DNA in blood. I proposed a panel of three methylation biomarkers (TUBA3FP, VANGL2, ARL10) for identification of intervention requiring disease, reporting 100% sensitivity and 84.6% specificity and demonstrated biomarker application for prediction of disease progression as well as utility for monitoring disease status with treatment. I was also able to show utility for predicting the necessity of treatment for low-grade dysplasia, which is controversial in guidelines worldwide. By performing genome-wide methylation and expression profiling, as well as cancer-associated mutation screening on single tissue biopsies from all stages of the metaplasia-dysplasiaadenocarcinoma sequence, I was able to gain a more complete understanding of the genetic and epigenetic changes occurring in esophageal adenocarcinogenesis. The research presented in this thesis demonstrates that I have been able to propose potentially clinically valuable methylation biomarkers for the detection of intervention-requiring disease, with potential application for non-invasive, high-risk population screening for identification of esophageal adenocarcinoma at an early, treatable stage.