There is considerable evidence that insulin and/or insulin-like growth factors (IGFs) can increase risk of colorectal neoplasia. Epidemiologic risk factors for colorectal neoplasia are similar to those for insulin resistance syndromes, and prospective studies have shown both diabetes and higher levels of IGF-I to be associated with colorectal cancer risk. No previous studies have included direct measures of insulin resistance, nor have any included complete ascertainment of colorectal neoplasia by direct examination of the entire colorectum. This study will assess the relationship between insulin resistance and colorectal neoplasia by taking advantage of a unique opportunity to examine a multi-ethnic cohort on whom prior measures of insulin sensitivity have been made. The Insulin Resistance and Atherosclerosis Study (IRAS) is a cohort study supported by the National Heart Lung and Blood Institute. IRAS examined 1628 people of average age 55 in 1991-1994 for atherosclerosis risk factors. The cohort, assembled in four clinical centers (Alamosa, Co., Los Angeles, Oakland, and San Antonio) was established to be multi-ethnic (34 percent Hispanic, 28 percent African American, and 38 percent non-Hispanic white), bi- gender, and varied in diabetes risk. In 1998-1999 over 85 percent of the surviving cohort was re-examined. Both of the examinations have included measures of self-reported risk factors for atherosclerosis (diet, physical activity, tobacco use, family history) as well as anthropometry and, most importantly, oral glucose tolerance testing and frequently- sampled intravenous glucose tolerance tests (FSIGT). The FSIGT is a sensitive and specific measure of insulin resistance. All surviving cohort members (estimated 1518) will be invited to have a screening colonoscopy. Feasibility data indicate that 1000 will agree to have a colonoscopic exam, among whom we estimate 240 (range 206-274) will have adenomas. Mucosal biopsies will be taken from the cecum and rectum of all subjects, and all adenomas will be removed and examined for histologic features, Ki-ras mutations, proliferation, and apoptosis. Serum samples will be assayed for insulin, IGF-I, IGFBPI, and IGFBP3 levels for all cohort members at both the time of colonoscopy, as well as at the time of two earlier examinations (1991-4 and 1998-9) using stored serum samples. This study offers the advantage of the availability of prospective measures of glucose tolerance, insulin resistance, measurements of most colorectal neoplasia risk factors, and the availability of stored blood samples from a multi-ethnic and bi-gender cohort. Complete colorectal visualization of this entire cohort will enable unbiased estimates of colorectal neoplasia risk related to these factors. This study therefore offers a time-efficient and a cost-efficient method to test the hypothesis that colorectal neoplasia risk is increased substantially by factors related to insulin resistance, and to examine the biologic mechanisms whereby that risk is increased.
