Researchers unlock precision nutrition research with biomarkers of food intake

15 Aug 2024 --- Biomarkers of food intake (BFIs) can increase the objectivity and accuracy of dietary assessments, which is crucial to enable precision nutrition. In a review article, scientists from the University of Copenhagen, Denmark, provide a comprehensive overview of BFIs measured in blood and urine and present a roadmap to develop and implement these tools to enable precision in nutrition research. 

A biomarker of food intake measures the consumption of specific food groups, foods or components to estimate recent or average intake. BFIs are circulating or excreted food constituents’ metabolites, usually non-nutrients.

The researchers note that BFIs can support precision nutrition, which requires precise tools. At the same time, current dietary assessment instruments are subjective and limit the understanding of causal relationships between diet and health.

Nutrition Insight discusses BFI best practices, challenges and developments with the study’s lead author, Dr. Catalina Cuparencu, an assistant professor at the Department of Nutrition, Exercise and Sports at the University of Copenhagen.

“While diet is an important modifiable risk factor for many chronic diseases, well-designed dietary assessment is often disregarded in the era of deep-phenotypic cohorts aimed to deliver precision medicine,” says Cuparencu.

She highlights that BFIs have now reached the potential to “improve confidence in nutrition research” by enhancing the assessment of adherence to dietary interventions and even by determining individual intakes of certain foods and dietary patterns.

“We hope that by broadening the implementation of biomarkers as dietary assessment tools in nutrition and health research, we highlight the need to include multi-sampling for objective dietary assessment in the new generation of emerging cohorts and make precision nutrition a core part of the efforts to deliver precision medicine.”

BFI best practices 

Cuparencu highlights that a significant bottleneck in biomarker development is the misconception that “each food in the human diet will be measurable by one biomarker, which will tell if, when and how much of the food was consumed. That is unrealistic.”

She explains that the review, published in Nature Metabolism, includes strategies to disentangle this false expectation. “For example, some biomarkers are well-validated to indicate if a certain food, like meat or lentils, has been consumed at all. These can be used in dietary intervention studies to monitor compliance with the intervention conducted — was the food consumed at all, yes or no, or was any meat consumed during a one-month vegan intervention?”

BFIs can enhance dietary intervention assessment and adherence and determine individual intakes of some foods and dietary patterns.The study highlights that the purpose of using BFIs in dietary intervention studies is often to identify non-compliant participants or adjust for misreporting and non-adherence.

“Although counter-intuitive, even in the most well-designed dietary intervention studies, researchers still rely on study volunteers to consume the foods provided or to not consume certain foods that would, for example, confound the results,” says Cuparencu.

Additionally, she emphasizes that sampling schemes must match the period within which the biomarker is measurable following food intake for the tools to work.

“Most current BFIs reflect intake only up to one to two days. Therefore, biomarker implementation will require multiple samples collected throughout the intervention, rather than before and after samples usually collected in feeding trials.”

“Indicating amounts of foods consumed is a bit more difficult with the current evidence, not because it is not possible, but because efforts targeting quantitation have only addressed a few foods,” she continues.

“Therefore, a combination of biomarkers and 24-hour dietary records, ideally collected during the same time, will still be needed for a while to detect dietary intake and will further contribute to validating candidate biomarkers and speed up the development of the field.”

Validating biomarkers

To be implemented in nutrition studies, BFIs need to be specific, unaffected by a varied dietary matrix and comparable with other intake measurements. Therefore, the review article stresses the importance of validating BFIs according to their plausibility, robustness and reliability. Time response is also important, as it provides guidance to determine the appropriate window for sampling.

In the review, the researchers rank BFIs according to their utility level, from level one (well-validated BFIs) to level four — missing BFIs because studies have not revealed any candidate BFI or because BFIs for these foods have not been investigated. The authors reviewed several urine and blood BFIs based on systematic literature searches.

Validated urinary BFIs exist for total meat and fish, chicken, some processed meats, fatty fish, total fruit, citrus fruit (orange), banana, whole-grain wheat and rye, alcohol, beer, wine and coffee. The researchers also identified promising candidate BFIs (utility level two) that require additional validation for total plant foods, vegetables, legumes, dairy and several specific fruits, tubers and vegetables.

The researchers present a comprehensive roadmap to unlock biomarkers as dietary assessment tools further.At the same time, the authors note that biological factors affecting intra- and interindividual variability of a BFI are not currently documented for most BFIs. Such factors include genetics, sex, age, dietary matrix, microbiota, disease condition, medication and ADME — absorption, distribution, metabolism and elimination rates.

Additionally, BFI detection in a blood or urine sample is affected by a person’s typical diet, the frequency of consuming a particular food and the sampling time.

To improve the specificity for single BFIs, the authors note that combining two or more biomarkers to indicate consumption of a food or food group is “well known as a solution.”

Roadmap for development

The researchers present a comprehensive roadmap to unlock biomarkers as dietary assessment tools further. It recommends BFI development focus on discovering and validating BFIs, implementing them in nutrition research for proof-of-concept studies, and developing routine methods to eventually “harvest benefits” for public health nutrition.

These benefits include improved reliability of dietary assessment, increased trust in associations between dietary components and specific health outcomes and the ability to enable precision nutrition research.

Cuparencu explains that different angles of biomarker development and implementation can be tackled simultaneously.

“On the one hand, discovering and validating current single and combined BFI candidates for single foods and food groups, as well as for contrasting foods and processing methods, should be prioritized, in line with current dietary recommendations tackling human and planetary health.”

For BFI development, Cuparencu underscores the need for data sharing platforms to search metabolomics across studies and store data.“On the other hand, proof of concept studies utilizing existing biomarkers of food intake as objective dietary assessment tools should be conducted. A few such studies already exist in the literature, yielding promising results.”

Lastly, she calls for the development of analytical and data-sharing platforms, which are needed to accelerate the routine use of BFIs in nutrition and health research. 

Overcoming limitations

Although the field is developing rapidly and Cuparencu observes solutions for many initial problems, she points to two prevailing challenges.

“First, the biomarker discovery part continues to be a bottleneck despite the available databases and fragmentation software, and thereby, many metabolites remain unidentified. A data sharing platform where untargeted metabolomics data can be queried across studies (to understand the specificity of the unknown metabolite) and where data can be stored easily and fast would be a great help.”

“Second, analytical standards do not exist for many metabolites, hampering their definite identification. Without standards, only semi-quantitative analytical methods can be developed to measure the biomarkers, which may affect how precisely we detect the amounts of food consumed. This might turn crucial, particularly for the foods consumed in small amounts.”

Regarding new research, Cuparencu comments that current initiatives work on discovering and validating BFIs and developing analytical methods targeting multiple BFIs.

“These will enable an increase in proof-of-concept studies and will enlarge the panel of foods that biomarkers can cover. Through data sharing and aligned efforts across laboratories, biomarkers of food intake may become the truth detector in nutrition research faster than anticipated,” she concludes.

By Jolanda van Hal