This proposal seeks to extend our prospective evaluation of predictors of Type 2 diabetes mellitus (DM) in the Nurses' Health Study, a large prospective cohort with archived blood specimens. Diabetes mellitus is a major and increasing public health problem, affecting at least 16 million Americans (of whom 15 million have Type 2 DM). A novel hypothesis implicates inflammation and endothelial dysfunction in the pathogenesis of Type 2 DM. We propose to study the role of several novel and promising biomarkers of inflammation (including tumor necrosis factor alpha receptor 2, interleukin-6, C-reactive protein, and ferritin) and endothelial dysfunction (including E-selectin, intercellular adhesion molecule-1 [VCAM-1]) as predictors of risk of Type 2 DM. We also propose to study the pathogenic roles of specific genetic markers associated with inflammation and endothelial dysfunction, including the peroxisome proliferator-activated receptor gamma [PAR-g2} Pro12A1a, tumor necrosis factor alpha G308A, and E-selectin Ser128Arg polymorphisms. Elucidation of interrelationships between these biomarkers and development of Type 2 DM may suggest new treatment and/or prevention strategies. Previous work in this cohort has contributed to clarifying the major roles of obesity, body fat distribution, physical activity level, and several dietary factors as determinants of Type 2 DM. We seek to build on this body of work by utilizing an existing database which includes 121,700 U.S. female nurses currently aged 54-79, with follow-up and documentation of DM endpoints through the 1998-90, with 774 incident cases of type 2 DM by the year 2000. We will use a nested case control design for the biomarker analyses. Several unique features of this established cohort stud, including its prospective design, large size, long duration, high follow-up rates (exceeding 90 percent over 20 years), availability of stored blood specimens, and cost-efficiency, make this database an unparalleled resource in the etiologic study of Type 2 DM in women.
Showing the most recent 10 out of 333 publications
