Molecular Imaging for the Detection of High-Grade Dysplasia in Barrett's Esophagus

Dana-Farber Cancer Inst, Boston MA

Investigators

Linked publications, trials & patents

Abstract

In the U.S., the incidence of esophageal adenocarcinoma (EAC) is rising more rapidly than other cancer, increasing 350% in the last 30 years. Most cases of EAC are preceded by Barrett's esophagus (BE), observed in 2-5% of the population. Although only 0.5% of individuals with BE develop EAC annually, it is widely agreed that the rate of EAC for BE patients with high-grade dysplasia (HGD) approaches 5-8%. Thus, current guidelines recommend routine endoscopic surveillance for HGD followed by eradication of HGD lesions. A major challenge is that current white light (WL) endoscopic surveillance requires invasive and cumbersome random biopsies since HGD arising within BE is often patchy and indistinguishable from non-dysplastic mucosa. Thus, the vast majority of patients with non-dysplastic BE undergo unnecessary surveillance with biopsies. Moreover, this strategy is insensitive, with up to 40% of patients with HGD harboring additional foci of undetected EAC and does not permit real-time identification of dysplasia for immediate treatment. These inherent limitations of our current surveillance approach underscores the need for novel endosopic imaging methods which can detect, in real time and in situ, HGD with high sensitivity. Our overall goal is to advance technologies that we have developed utilizing near infrared fluorescent (NIRF) endoscopic imaging of novel optical probes activated selectively in dysplasia by cathepsin B (CTSB) proteases. In the prior funding period, we have made considerable progress in the development of this approach for more sensitive, real time detection of colorectal neoplasia. We have also demonstrated selective upregulation of CTSB in small HGD lesions within background, CTSB-negative non-dysplastic BE. Thus, we now propose to translate these innovative CTSB-selective probes for the detection of HGD in BE using mouse models which closely mimic human disease and human BE tissue samples which have been histologically and genomically characterized. We will leverage this preclinical work into a Phase l/ll clinical trial in patients with BE. If successful, this technology could fundamentally change our approach to the early detection and real time localization of HGD in BE, a critical area of unmet need for the prevention of EAC.