Nutrigenomics and Nutrigenomic Testing: How Genes and Nutrients Interact

Genetic testing has become cheap and accessible, and a growing industry offers diet plans, supplement regimens and lifestyle advice based on a person’s DNA. The idea is compelling: if people respond differently to foods because of their genes, why not eat according to them? The science of gene nutrient interaction is real and advancing, and a few applications are firmly established. But most of what is currently sold goes well beyond the evidence. This article maps what is known, what is promising and what is marketing.

3 related fields

In practice, consumer nutrigenomic tests are mostly nutrigenetic: they genotype selected variants and generate recommendations.

Established gene nutrient interactions

A handful of interactions have clear clinical importance.

• Lactase persistence. Variants near the lactase gene determine whether lactase production continues into adulthood. Most of the world’s adults have reduced lactase, and genetic testing can support a diagnosis of lactose malabsorption, though symptoms and tolerance vary.
• Coeliac disease risk genes. Almost all people with coeliac disease carry HLA DQ2 or DQ8. These genes are common in the general population, so a positive result does not diagnose coeliac disease, but a negative result makes it very unlikely, which can be useful when diagnosis is uncertain.
• Hereditary haemochromatosis. Homozygosity for the C282Y variant of the HFE gene causes iron overload in some people; identifying it guides treatment and avoidance of iron supplements.
• Phenylketonuria and other inherited metabolic disorders. Detected through newborn screening and managed with precise dietary therapy.
• Familial hypercholesterolaemia. Genetic diagnosis guides treatment and family screening, although management is primarily medical.

These examples share features: large effects, well understood mechanisms and clear clinical actions.

Plausible but modest interactions

In the DIETFITS trial, genotype patterns proposed to predict response to low fat or low carbohydrate diets did not predict weight loss. The same trial found no interaction with baseline insulin secretion.

Does DNA based advice change behaviour?

A key question is whether giving people genetic information about their diet leads to better behaviour or outcomes. In the European Food4Me trial, personalised nutrition advice improved dietary quality more than generic advice, but adding genetic information to personalised advice did not produce further improvement. A systematic review of trials communicating genetic risk found little or no effect on risk reducing behaviours such as diet, physical activity and smoking. Personalisation helps; genetics, so far, adds little to it for most people.

Polygenic scores

Polygenic scores combine the effects of many variants to estimate genetic predisposition to conditions such as obesity, type 2 diabetes or coronary disease. They are useful research tools and may eventually refine risk prediction, but for most conditions they add modestly to established clinical risk factors, perform less well in populations underrepresented in the genetic studies from which they were derived, and do not yet have clear dietary applications.

Nutritional epigenetics

Epigenetic mechanisms, such as DNA methylation, regulate gene expression without changing the DNA sequence and can be influenced by nutrition. Studies of people exposed to famine before birth during the Dutch Hunger Winter of 1944 to 1945 found differences in DNA methylation decades later, alongside higher rates of some metabolic conditions. Maternal nutrition, including folate and B12 status, influences offspring epigenetics. These findings underscore the importance of nutrition in early life, but commercial epigenetic age tests and epigenetic diet plans lack clinical validation.

Concerns about consumer testing

• Weak evidence for most reported variants and recommendations.
• Inconsistent results between companies interpreting the same variants.
• Incidental findings, such as APOE status, with implications for disease risk that people may not be prepared for.
• Privacy of genetic data, which concerns relatives as well as the person tested.
• Commercial bundling of tests with supplement sales.
• Fatalism or false reassurance, in which people conclude that their genes determine their health or that they are protected.

A practical approach

1. Use genetic testing for established indications, such as suspected haemochromatosis, familial hypercholesterolaemia, uncertain coeliac diagnosis or inherited metabolic disease, ideally with genetic counselling where appropriate.
2. Prefer direct measures of nutrient status, such as vitamin D, B12 or homocysteine, over genotypes that predict them indirectly.
3. Personalise through history and response, which usually offers more than genotype.
4. Help patients interpret consumer reports, distinguishing robust findings from speculative ones.
5. Discuss privacy and incidental findings before testing.

An illustrative case

A 42 year old man brings a consumer nutrigenomic report recommending a low fat diet, extra B vitamins, avoidance of coffee and several supplements based on variants in FTO, MTHFR, CYP1A2 and others. It also reports that he carries one APOE e4 allele, which has caused him considerable anxiety about dementia.

The practitioner explains the modest effects of the reported variants and that the dietary advice for him would be similar regardless: a healthy dietary pattern he can sustain. His homocysteine and B12 are measured directly and are normal, so extra B vitamins are unnecessary. His moderate coffee intake is fine. The APOE result is discussed sensitively, including that most people with one e4 allele do not develop Alzheimer disease, and that the modifiable risk factors for dementia, blood pressure, activity, hearing, social engagement and others, apply to him as to everyone. He is offered referral to genetic counselling. His plan focuses on those modifiable factors.

Genes load the gun, but the evidence decides

Gene nutrient interactions are real, and a few are clinically important. For most people, however, genetic testing does not yet improve on careful personalised nutrition built from history, preferences and measured response. The field is promising; its commercial applications should be judged by the same evidence standards as any other intervention.