Writing in the American Journal of Clinical Nutrition , Kristin Guertin, PhD, and her co-authors report that the detection of 39 dietary biomarkers from 502 people for multivitamins, citrus, green vegetables, peanuts, rice, fish, shellfish, red meat, butter, coffee, beer, liquor, and total alcohol.
“The large number of correlations between self-reported diet and serum metabolites confirms that metabolomics can be applied to epidemiologic studies for identification of novel dietary biomarkers,” they wrote.
“There is a need for specific, reliable biomarkers that accurately reflect dietary intake and that can be applied to many populations. We emphasize, however, that although we appear to have uncovered objective biomarkers of diet, it should not yet be assumed that these biomarkers outperform self-report as a measure of usual dietary intake.”
A whole new level
Commenting independently on the paper, Harry Rice, PhD, VP of regulatory & scientific affairs, for the Global Organization for EPA and DHA Omega-3s (GOED), told NutraIngredients-USA that metabolomics has matured over the last couple of decades, from a fishing expedition to what he would consider to be hypothesis-driven research and it's been exciting to watch its progression.
“The validation of these metabolites as biomarkers has the potential of advancing nutritional epidemiology to a whole new level,” he said. “What remains to be determined is the quantification of the amount of different foods, but that's likely around the corner. Without knowledge of the ‘dose’, it's difficult to draw any conclusions. Even then, we wouldn't be talking about cause and effect.”
A promising tool
Andrea Wong, Ph.D., vice president, scientific and regulatory affairs for the Council for Responsible Nutrition (CRN), told us that the study shows that metabolomics may be a promising tool for more accurate measurements of dietary exposure, given the limitations associated with the commonly used self-reporting methods, such as recall bias.
“This approach also could be used to identify new biomarkers, since these are lacking for several nutrients and other dietary components. Potentially, metabolomics could be applied in epidemiology studies to investigate relationships between food/nutrients and health outcomes, and also in research conducted to assess nutrient adequacy.
“However, any novel biomarkers will need to be validated before they are used in such research.”
The NCI scientists aimed to move the subject of metabolomics forward by identifying metabolites that are biomarkers of usual dietary intake, and then narrowed that down by evaluating metabolite.
Data from 502 participants in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial was used to elucidate 12 known metabolites, which then led to the identification of 39 dietary biomarkers.
The results included both a replication of some previous findings, such as citrus intake linked to stachydrine, and novel findings, such peanuts and tryptophan betaine.
There was some variability in metabolite levels between the individuals in the study, and the researchers added that, “large, but attainable, sample sizes are required to detect associations between metabolites and disease in epidemiologic studies, further emphasizing the usefulness of metabolomics in nutritional epidemiology.”
“Ultimately, whether a biomarker is a good measure of usual diet depends on the frequency of consumption of the food or nutrient, as well as the half-life of the metabolite.
“In addition, the identification of serologic metabolites not only reflects dietary intake but also metabolic processes, including the effects of genetic variation and the gut microbiota. Nevertheless, our metabolomic approach for identifying potential dietary biomarkers showed viable biomarkers for further investigation in feeding studies,” they concluded.
Source: American Journal of Clinical Nutrition
Published online ahead of print, doi: 10.3945/ajcn.113.078758
“Metabolomics in nutritional epidemiology: identifying metabolites associated with diet and quantifying their potential to uncover diet-disease relations in populations”
Authors: K.A Guertin, S.C. Moore, J.N. Sampson, W-Y. Huang, Q. Xiao, R.Z. Stolzenberg-Solomon, R. Sinha, A.J. Cross