Application of predictive biomarkers of sugars and animal protein intake for investigation of dietary measurement error and its effect on diet-disease associations

点击翻译按钮获取中文摘要

Project Summary Developing new approaches for obtaining accurate estimates of dietary intake is crucial for revealing true associations between dietary intakes and disease risk, with the ultimate aim of establishing evidence-based guidelines. While the evidence on dietary sugars and increased risks of dental caries and obesity have been stronger, the association between sugars and increased risk of cardiovascular disease (CVD), type 2 diabetes (T2D), and cancer remains inconclusive. The findings on total and animal protein (AP) intake in relation to T2D risk and CVD or all-cause mortality have also been inconsistent. Dietary biomarkers alleviate the problem of measurement errors (ME) associated with self-reporting dietary instruments commonly used in nutritional epidemiology, and have been used to validate and calibrate self-reported diet or to adjust for ME in self- reports, and reveal important associations. In the parent project of this renewal application, (SugarsBio study, U01-CA197902), we confirmed 24-h urinary sucrose and fructose (24uSF) as a predictive biomarker of total sugars (TS) intake, and demonstrated the transportability of the biomarker equation for estimating biomarker- based TS intake and its use across different populations. Furthermore, we have identified serum carbon isotope ratio (CIR) as a candidate predictive biomarker of the AP to total protein intake ratio (APR), a novel biomarker of protein quality, and developed a biomarker equation that can be used to generate biomarker- based APR in studies with available biological samples. The aim of this proposal is to investigate the utility and application of 24uSF and serum CIR biomarkers in diverse populations in two prospective cohorts. For this purpose, we will leverage data from two large dietary validation studies with comprehensive validation protocols (IDATA and SOLNAS) nested within cohorts, the NIH-AARP Diet and Health (AARP) Study and the Hispanic Community Health Study/Study of Latinos (HCHS/SOL). First, we will study the ME in self-reported TS, and APR, AP and plant protein (PP) intake, using 24uSF and serum CIR biomarkers. Second, we will develop regression calibration equations for self-reported intake, based on biomarkers, in IDATA and SOLNAS that we will apply in their respective cohorts. Third, we will investigate uncalibrated and calibrated (i.e., ME- corrected) self-reported intakes of TS, AP, PP and APR in relation to CVD mortality in the AARP, and T2D risk in the HCHS/SOL cohort. Our study will be the first study that applies the newly developed US population- based sugars and protein intake biomarkers and their calibration equations to race/ethnically diverse US population-based studies and evaluates ME-corrected dietary intakes in relation to chronic diseases. By informing the best practices for applying these biomarkers in future diet validation studies and studies of diet-disease associations, this proposal will significantly contribute to improving dietary assessments and enhancing scientific rigor of nutritional epidemiologic studies.