Design and Characterisation of a Randomized Food Intervention That Mimics Exposure to a Typical UK Diet to Provide Urine Samples for Identification and Validation of Metabolite Biomarkers of Food Intake

Challenges Associated With the Design and Deployment of Food Intake Urine Biomarker Technology for Assessment of Habitual Diet in Free-Living Individuals and Populations—A Perspective

“…Preliminary investigation of the performance of any specific biomarker in the context of a complex biomarker panel requires urine samples from complex food intervention studies designed specifically to emulate habitual eating patterns. The MAIN Study at Newcastle was designed with this specific objective in mind (1,2,22). By validating biomarkers in urine samples from studies researching eating behaviour in free-living populations (62,67) it is anticipated that BFI technology will mature rapidly over the next few years.…”

“…There is a rich history of nutrition research spanning many decades, much of which has had at its core a need for accurate information on dietary intake for investigation of the links between exposure to individual food/food groups and specific health outcomes. Food intervention projects commonly rely on participants collecting pre-prepared foods from research centres for consumption at home and then confirming compliance at a later date (1,2). On the other hand, large-scale nutritional epidemiological projects and nutrition surveys involving freeliving individuals consuming their habitual diet rely almost totally on self-reporting of dietary exposure.…”

“…Using data-driven approaches, we have shown that the potential utility of a biomarker is dependent on the type, portion size, and frequency of consumption of individual foods (20). Data concerning nationally-representative estimates of intakes of foods by the UK population are collected by the UK National Diet and Nutrition Survey (NDNS) (21) and this database can be explored to identify foods and food groups for which dietary exposure biomarker discovery might be feasible and relevant (1,2,22).…”

“…Over the past decade, our collaborative research projects and those of other teams (see Supporting Data 1 for a comprehensive list) have contributed to the discovery of putative dietary intake urinary biomarkers of specific foods including poultry and red meat (23)(24)(25)(26)(27)(28) citrus fruits (29,30), crucifers (31,32), oily fish (26,27,32), red berries/strawberries (2,(32)(33)(34), wholegrain/rye (35)(36)(37), sugary drinks (38,39), artificial sweeteners (2,40), peas/beans/legumes (2,41,42), grapes (41,(43)(44)(45), apples (41,46,47), and potatoes (48). In addition, consensus guidelines for the critical assessment of candidate BFIs has been established (49).…”

“…Optimal sampling requirements for urine biomarker analysis will be dependent largely on the study design and objectives ( Table 1). For example, a food intervention study with free-living participants lasting several weeks investigating links between a health outcome and a specific food/food group will require appropriate samples on multiple days taken at random to assess compliance with dietary intake targets (1,2). In contrast, assessing the general eating habits of a large population in an epidemiological survey may only require sampling of a large number of people on a single random day or multiple days (60).…”