||5R01CA119171-11 Interpret this number
||Fred Hutchinson Cancer Research Center
||Nutrition and Physical Activity Assessment Study (NPAAS)
DESCRIPTION (provided by applicant): This competing renewal of CA119171, "Nutrition and Physical Activity Assessment Study" (NPAAS), continues to focus the use of biomarkers and regression calibration models to understand better the relationship between diet, physical activity and chronic disease risk in postmenopausal women. Our work uses objective biomarkers in regression calibration models to correct self-report data for random and systematic measurement biases. This work has provided evidence for strong positive associations of energy intake, and strong inverse associations of activity-related energy expenditure (AREE), with chronic disease risk; these associations were absent without measurement error correction. To date, suitable quantitative biomarkers have been available for few diet components (e.g. total energy, protein, potassium, sodium) and physical activity (AREE). For this grant funding period, we will extend our work by using innovative metabolomics profiling in blood and urine specimens to characterize food and nutrient intake. Metabolomics offers the possibility to advance the field of nutritional epidemiology in a manner not been possible with individual biomarkers since metabolomics uniquely accounts for dietary exposures in an integrated manner and explains > 50% of the variation in nutrient intake. Our specific aims are: (1) To obtain serum and 24-hour urine metabolomic profiles for n=450 WHI-NPAAS participants who have stored blood and urine specimens from a previous project cycle. To use these profiles to calculate intake values for each nutritional variable for which a biomarker is established in our recently completed NPAAS feeding study. To use the resulting biomarker-based intake values to develop calibration equations for each nutritional variable using previously collected food frequency questionnaire, 4-day food record and 24-hour dietary recalls, as well as participant characteristics, from these 450 women. (2) To use the calibration equations developed in Aim 1 to estimate intake throughout WHI cohorts for each nutritional variable with suitable biomarkers. To conduct a series of analyses examining the association of metabolomics- calibrated nutrient intake with risk of major cancers, cardiovascular diseases and diabetes in WHI cohorts, where self-reported dietary intake and other key data are available at baseline for 161,000 postmenopausal women. (3) To use serum and spot urine metabolomic profiles to directly estimate intake for each nutritional variable having a suitable biomarker, in subset of about 12,000 WHI women with both stored blood and urine. Profiles will be obtained for n=708 adjudicated cancers (breast, colorectal, ovarian and endometrial) and n=708 controls. These analyses will assess novel nutrition and metabolite cancer associations without use of dietary self-report data. (4) To examine blood- and urine-based metabolomic profiles more broadly in relation to cancer risk in the Aim 3 case-control sample without regard to the role of specific metabolites in nutrient biomarker specification. Our highly productive research team is expected to make additional, innovative contributions in nutritional epidemiology and more broadly in chronic disease prevention research.
Nutritional Epidemiology Methods and Related Statistical Challenges and Opportunities.
, Huang Y.
Statistical theory and related fields, 2018; 2(1), p. 2-10.