Why We Need Biomarkers

Dietary assessment is the bedrock of nutrition science. Accurate and reliable methods for assessing what people eat are essential to understanding the relationship between diet and health. However, despite decades of research, dietary assessment remains difficult and error-prone.


Challenges in Dietary Assessment

Reliance on Self-Reporting

Most dietary studies depend on methods like food frequency questionnaires, 24-hour recalls, or food diaries. These self-reported approaches are fraught with inaccuracies due to:

  • Recall Bias: Individuals may forget or misremember what they ate.
  • Underreporting: Foods perceived as less healthy are often omitted or underestimated.
  • Overestimation: Portion sizes, especially for healthier foods, are often exaggerated.

But these are only some of the shortcomings of self-reported methods. A crucial – yet often overlooked – aspect is the variability in food composition. Foods vary in their nutritional content, even between two items of the same food such as two apples. Indeed: even apples harvested at the same time from the self-same tree differ in their nutrient composition.

The following figure shows the range in (-)-epicatechin, flavan-3-ols and nitrate in a range of commonly consumed foods. These compounds are commonly referred to as bioactives, non-essential compounds found in foods that have health-promoting properties. They are not required for survival, but their presence in the diet can reduce the risk of chronic diseases.

Variability in flavan-3-ol, (–)-epicatechin (Rothwell et al., 2013), and nitrate (Blekkenhorst et al., 2017) content of foods commonly eaten.
Data show the range of food content (black) and mean (red).

This is of course not restricted to plant foods – most processed foods vary in their composition, because they are not standardised for composition but for taste.


Does this matter?

The variability in food composition matters: it means it is virtually impossible to estimate the actual intake of certain compounds – bioactives and nutrients, but also food additives like emulsifiers – from dietary records alone. But it also affects research because estimated associations between intake and disease risks are affected.

We can estimate the impact of bioactives on health outcomes by combining detailed dietary intake data (e.g., from 24-hour dietary recalls) with available information on the range of bioactive concentrations in foods. For some bioactives, this enables us to estimate the potential intake range for each study participant and examine its association with health outcomes.

In our analysis, we applied this approach to three different bioactives. The figure below illustrates the results of 10,000 simulations (represented as dots). These simulations highlight how the variability in food composition can lead to markedly different associations with health outcomes. For reference, the true association is indicated by a solid line, providing context for the variability observed.

Association between estimated bioactive intake (flavan-3-ols, (–)-epicatechin, and nitrate, based on the 24 hr dietary recall and food composition data [DD-FCT method]) and systolic blood pressure at baseline (estimated difference between low [p10] and high [p90] intake and p-value for Wald-test (as -log10(p)) in men [purple], women [green], and all participants [red]). Data are based on 10,000 simulations and adjusted for age, body mass index (BMI), plasma vitamin C, smoking status, physical activity, and self-reported health at baseline; additionally for menopausal status for women and sex for nitrate. Results based on the intake estimated by simulating food content within minimum and maximum food content reported in databases (circle), intake based on mean food content as reported in databases (diamond), and intake based on biomarker data (|). 95% CI is shown for biomarker only.

What about food-food interactions?

Food-food interactions can affect the uptake of individual compounds – such as the effect of phytates on the uptake of zinc. Another example is the effect of phenol-oxidase, found for example on bananas, on flavanols. In both instances, relying on dietary intake alone would overestimate actual intake and yield misleading results.


What can be done?

Intake assessments based on nutritional biomarkers can provide an advanced alternative, as they reflect the systemic presence of nutrients and bioactives. They are ideal to investigate and assess the intake of individual compounds – or classes of compounds – but are have limited use to estimate intake of individual foods or dietary patterns.

As molecular tools to assess dietary intake, nutritional biomarkers are crucial for nutrition research. They are the only reliable instrument to estimate intake of bioactive food constituents. The main application of nutritional biomarkers are observational cohort studies, where obtaining reliable estimates of intake at population-relevant scales is crucial. However, applications can go far beyond this: nutritional biomarkers can be used to assess compliance in clinical studies or to stratify participants in clinical trials according to intake. They are also crucial for personalised nutrition, as accurate dietary advice can only be provided with a known baseline and adequate and objective feedback loops.

However, there are also a number of pitfalls – which we have outline in more detail previously. In brief, they are:

Candidate biomarkers with very short half-life or high inter-individual variabilityPharmacokinetic parameters of biomarkers are important to allow the selection of suitable metabolites. They can help identify metabolites with particularly long half-life or lower inter-individual variability.
Avoiding colonic metabolites because of variability of gut microbiomeColonic metabolites can be very useful as biomarkers as they often have a much longer plasma half-life, but they require careful evaluation regarding their inter-individual variability
Unreliable validation dataBiomarkers have to be evaluated and validated against actual intake – not estimated intake based on food composition data.
Dose-dependent metabolism and inter-individual variabilityMetabolism depends on a range of factors, including the amount consumed and differences in genotype. A combination of mutilple metabolites can therefore improve outcomes over a larger range of intake
Lack of specificityMany candidate-biomarkers have more than one pre-cursor, not just colonic metabolites. It is therefore important to establish specificity of any new biomarker.
Selection of appropriate specimen and adjusting for dilutionWhile 24h urine is ideal, spot urine samples are often more feasible. Adjusting spot urine samples for dilution by creatinine can introduce considerable bias that might attenuate (or exaggerate) associations.
Unreliable analytical methodsAnalytical methods should be based on the use of LC-MS with authentic standards. Further development should seek translating methods into more accessible analytical platforms

Looking Forward

Nutritional biomarkers are indispensable for advancing the field of dietary assessment. They provide the only reliable method to quantify the intake of bioactive compounds, addressing a critical gap in nutrition research. While challenges remain, continued innovation in biomarker discovery, validation, and application will enable the development of more precise and impactful dietary recommendations. By integrating biomarkers with traditional dietary assessment methods, we can unlock a deeper understanding of the complex relationship between diet and health, paving the way for improved public health outcomes and personalised nutrition strategies.

In summary, biomarkers are not just tools but foundational elements in the next era of nutrition science, enabling researchers to go beyond estimation and achieve true accuracy and reliability