ACS Research Highlights

Researchers Ask What People Eat—and Check Urine Samples

CPS-3 researchers evaluated diet assessments and blood and urine samples to identify 100s of metabolites linked to what people frequently eat and drink.

The Challenge

Many scientific studies use questionnaires or interviews to gather information from people about what they usually eat and drink, how often, and in what amounts. These self-reported methods are only as reliable as a person’s memory, and regardless of a study participant’s diligence and intentions, their accuracy is questionable. Likely errors in data about a person’s eating habits makes it harder to study the role of diet in cancer prevention and survival.

To overcome these problems, nutritional epidemiologists examine dietary biomarkers found in blood or urine to measure the biological effects of specific dietary components and to identify foods that influence the risk for certain metabolic diseases, like obesity.

Dietary biomarkers have great potential because they are objective and not subject to the types of errors with self-reported diet, such as relying on memory. New, robust laboratory methods allow for the investigation of biological markers of foods, beverages, and other dietary components. Dietary biomarkers include natural food constituents, like vitamins and flavonoids (nutrients in plants), food additives, as well as food contaminants. Small molecules derived from food components are called food metabolites.

Nutritional epidemiologic studies have significantly advanced the understanding of the relationships between diet and chronic diseases and have led to dietary guidelines for disease prevention in recent decades. However, the field has been limited by inconsistent findings from many studies. Confirmed, validated dietary biomarkers do not exist for most foods and dietary patterns.

The Research

Ying Wang, PhD, along with colleagues from the American Cancer Society (ACS) and Emory University, use a high-throughput technology called metabolomics to measure thousands of small molecules in blood and urine and other body tissues—to identify metabolites related to the food eaten.

Measuring a person's diet is super hard because we eat such a large variety of foods every day. In the past, when researchers for large human studies wanted to analyze the effect of diet on disease, they relied on participants to answer surveys about what they ate. Such self-reported dietary data, however, may be biased because our memory is inaccurate, and we have a tendency to report 'what I should eat' vs. 'what I actually ate.'

Ying Wang, PhD

Senior Principal Scientist, Epidemiology Research

Population Science, American Cancer Society

In comparison, Wang says, “Dietary biomarkers found in the blood and urine are considered objective measures of diet." 

The problem with reliable dietary biomarkers, though, she explains, is that they are sparse and most are nutrient-based, not food-based.

"Our study used metabolomics technology, which measures thousands of small molecules in blood and urine, to identify new food-based biomarkers for a large group of men and women.”

They used responses to questionnaires and biospecimens from almost 700 men and women in the Cancer Prevention Study-3, Diet Assessment Substudy to identify hundreds of metabolites that are related to what people frequently eat and drink, such as with oranges, broccoli, whole grains, fish, coffee, and tea.

They also used repeated measurements from blood and urine samples, collected about 6 months apart, to assess the reproducibility of the biomarkers.

They compared:

  • 101 food groups/items by a food frequency questionnaire
  • 105 food groups/items by repeated 24-hr diet recalls
  • 1,391 metabolites measured in 2 different 24-hour urine samples (urine collected in a special container for a full day, 24 hours)

They identified 513 unique metabolites correlated with 79 food groups/items. Many of these replicated metabolites found in earlier studies, adding a step towards their validity. They published their results in the journal Metabolites.

Why It Matters

Objective dietary biomarkers help validate, or add to, self-reported dietary questionnaires and are important in diet assessment in large population studies to move the field forward. Wang’s findings about identified food-related metabolic biomarkers contribute to the evidence from limited studies and will need to be confirmed and evaluated in future studies.

Identifying and learning more about dietary biomarkers may help researchers and health care professionals:

  • Shed light on physiological or pathological responses people have to certain food behaviors.
  • Help predict the risk for certain diseases.
  • Improve screening and diagnosis.
  • Monitor the progression of a disease and its responses to treatment.
  • Have information on the kinds of individual differences people have in response to diet.

In the future, knowledge about dietary biomarkers may help researchers and other experts develop personalized dietary recommendations, also called precision nutrition.