A major clinical challenge is to accurately determine individuals with Barrett's oesophagus that are at risk for cancer development. The risk of progression is currently determined by histological assessment of the grade of dysplasia which is highly subjective. Correctly identifying a high risk group would enable optimal care for these individuals whilst reducing the burden on endoscopy units for those patients at low risk. Although biomarkers have been identified none have yet been applied to clinical practice due to lack of validation and due to the lack of a suitable assay for routine laboratories.
Useful biomarkers can be identified among different groups of molecules and with a range of approaches:
1. We are identifying novel markers using a genome wide transcriptomic and epigenomic approach in addition to investigating candidate genes identified from our in vitro studies. We have previously identified two cell cycle associated biomarkers, Mcm2 and cyclin A, whose increased expression on the epithelial surface before the development of dysplasia is associated with at least a 7 fold increase in the risk of progression. When used in a panel of biomarkers we have shown in a collaborative study that those with >30 fold increased risk can be identified (Dunn J et al., Gastroenterology, 140(5)S1:S-136 , 2011). We are also currently generating and validating a gene signature to stratify patients with Barrett’s according to their risk to develop cancer.
2. The luminal and stromal environment could play a critical role of in determining progression to cancer (e.g. Saadi et al., PNAS, 107(5):2177-82, 2010). We are using 3D in vitro model systems to evaluate the effects of these components on the epithelial cell phenotype using a range of molecular and cell biological techniques including genome wide array based approaches. This information may inform chemopreventive strategies.
Whole Organ Imaging Ex-Vivo.
3. The alteration in glycosylation on the surface of Barrett’s and dysplastic oesophagus has allowed us to use naturally occurring binding partners called lectins, as molecular imaging probes (Bird-Lieberman et al., Nat Med, 2011 in press). We will now examine this further in a clinical study involving patients with high grade dysplasia in Barrett’s oesophagus.
In addition to applying these biomarkers to biopsies taken at surveillance endoscopy we are also interested in applying these biomarkers to a non-endoscopic cell collection device called the Cytosponge. This new diagnostic device was tested in the BEST1 study (Kadri et al., BMJ 2010) and the associated biomarkers are now being developed further in the BEST2 study. We hope that this imunocytological approach will allow us to screen patients at risk for oesophageal cancer (adenocarcinoma, and in time squamous cell cancer), on a large scale and in a cost-effective way. In all of our biomarker work clinical applicability is a primary consideration (Spechler et al., Gastroenterology, 138(3):854-69, 2010).