Improving risk stratification in Barrett's esophagus

Barrett’s esophagus is the precursor lesion of esophageal adenocarcinoma, a cancer with a poor 5-year survival rate of 10-20% if diagnosed late. In contrast, early detection of Barrett’s cancer allows for endoscopic resection with an excellent prognosis and low morbidity. It is Malignant transformation from non-dysplastic Barrett’s to cancer is thought to be a stepwise progression from non-dysplastic Barrett’s to low-grade dysplasia, followed by high-grade dysplasia and finally cancer. This allows for endoscopic surveillance every 3-5 years with the aim to detect cancer early and amenable to endoscopic treatment. However, the current surveillance strategy fails as the majority of BE patients is non-dysplastic and do never progress to cancer, and the majority of cancers are diagnosed in advanced stages and outside of the surveillance program. Improved risk stratification would allow to identify and stop surveillance in low-risk patients, whereas surveillance or even preventive therapy could be offered to high-risk patients. This thesis summarizes current methods and limitations of risk stratification in BE. We developed a nested case-control cohort using the most stringent inclusion criteria to optimize exploratory studies on biomarkers predicting progression in non-dysplastic BE. The second part of the thesis focuses on the evaluation of histological features (degree of crypt atypia) to improve risk stratification. The third part of the thesis includes several studies investigating an objective tissue systems pathology test to improve risk stratification in BE irrespective of the presence of dysplasia.