Interdisciplinary Training

Sid H. Siahpush

Insulin-like Growth Factors and Esophageal Adenocarcinoma

Increased serum levels of insulin-like growth factors (IGFs), along with decreased levels of IGF binding protein-3 (IGFBP-3) have recently been reported to increase risk of several cancers. They act through interactions with several molecules, including IGF-1 receptors (IGF-1R). In addition, polymorphisms in IGF-1, IGF-1R and IGFBP-3 genes have been identified that may have functional significance with regard to tissue concentrations of IGFs. We propose to take advantage of data and specimens collected in an ongoing cohort study of predictors of neoplastic progression in persons with Barrett’s esophagus (BE) (n = 420, up to 8 years of follow up) to determine whether IGFs are related to risk of esophageal adenocarcinoma (EA), and a series of biomarkers validated as predictors of EA. The subjects donated blood and underwent endoscopic biopsies, physical examination and interview at baseline. Regular follow-up endoscopies and biopsies have been performed since baseline.

Aim 1 will determine whether increased baseline serum levels of IGF-1 and decreased levels of IGFBP-3 are associated with increased risk of neoplastic progression in BE. Neoplastic progression will be determined by the development of EA, high-grade dysplasia, increased 4N/G2 fractions, aneuploidy and/or 17pLOH. Aim 2 will determine whether polymorphisms in IGF-1, IGF-1 receptor and IGFBP-3 genes are related to risk of neoplastic progression, and whether they modify any associations observed in the first aim. Secondary aims will investigate whether observed IGF associations are modified by key risk factors for EA and BE, including obesity, waist:hip ratio, use of NSAIDs, serum selenium. Cox regression models will be used to calculate adjusted hazard ratios. Our well-established and characterized cohort, together with data and tissue that are readily available, afford a unique and cost-efficient opportunity to study these issues, with the expectation that positive results would provide i) additional methods with which to identify high risk subjects with BE for more frequent surveillance, and ii) modifiable factors for short-term intervention studies.