All analyses
5 min read

Last updated:

Nutrigenomics: where genetic testing really stands in nutrition

Nutrigenomics links the genome to the nutritional response. What nutrition genetic tests establish, what they do not show, and the clinical situations where genetics and nutrition really change a decision.

7 studies citedsee the sources

Benjamin Sant

Co-founder of the School of Longevity

Professional analyzing health data
Professional analyzing health data

Nutrigenomics studies how the genome changes the body's response to a dietary intake. Commercial personalized nutrition reports rest on statistical associations, not on trials. A variant can be genuinely associated with a phenotype without explaining more than a tiny fraction of the variation between individuals. The professional's first question is therefore what the result can change in practice.

The logic of the discipline is not recent. Work linking apolipoprotein A5 polymorphisms to the lipid profile opened the way. Other work links lactase to lactose status. The reference cohorts describe selected populations, often of African American origin. Transposing these results to a person from another population is a bet, not a scientific conclusion.

The promise of the discipline lies in the possibility of one day treating monogenic conditions through targeted dietary adaptation. Lactose intolerance and celiac disease already belong to medicine. The overall level of evidence remains weak. An honest professional therefore presents nutrigenomics as a field under research.

The weak link in the chain remains transposing a statistical association observed in a group into a validated individual recommendation.

Nutrition genetic tests overinterpret weak associations

Nutrition genetic tests that analyze several hundred variants almost always produce a significant signal. That frequency is explained by chance alone, when statistical comparisons are multiplied. A large share of published associations has not been found in independent cohorts. The level of evidence required for an individual recommendation is therefore rarely reached.

The reference figure remains telling: a variant linked to weight explains barely one percent of its variation. Controlled trials show small or null differences in weight. Advice based on genotype does no better than classic advice. The strongest statistical argument therefore concerns the size of the effect, not its existence. Its existence is not enough. Review articles conclude on a level of evidence insufficient for routine practice. They invite a distinction between the analytical validity of the test and the clinical validity of the recommendation produced.

Lactose intolerance, linked to a gradual loss of lactase expression, remains the most reproducible example. Celiac disease has HLA-DQ2 and DQ8 typing that completes serology within a supervised pathway. Pharmacogenetics concerns liver enzymes that determine the efficacy of certain treatments. In these cases, the practical consequence is immediate. That urgency of decision separates these situations from consumer kits.

Levels of evidence by field of application of genetics in nutrition
FieldScientific basisCurrent clinical use
Lactose intolerance (LCT variants)Strong and reproducible gene-phenotype linkRelevant with digestive symptoms, to be confirmed clinically
Celiac disease (HLA-DQ2 and DQ8)Screening tool in addition to serologyUsable in a medical setting, not in a consumer kit
Pharmacogenetics (CYP2C19, CYP2D6)Determines the efficacy and tolerance of treatmentsRelevant in people on multiple medications
Caffeine metabolism (CYP1A2)Robust association, limited effect on behaviorContext information, with no imposed course of action
Body weight (FTO)Variance explained generally below 1%No individual recommendation
Vitamin D and iron intakeContradictory results depending on populationNo course of action changed on the basis of genotype
Response to a low-calorie dietIdentical weight loss between genotyped groupsNo additional benefit in controlled trials

Levels of evidence by field of application of genetics in nutrition

Nutrition and biomarkers reviewed in a follow-up file
The level of evidence varies from one field to another, which rules out a single position on nutrigenomics.

Personalized eating starts from real habits before the genotype profile

The most effective personalized nutrition starts from real habits, preferences and constraints. Taste, budget, culture, timing and adherence determine follow-through more reliably than genotype. Dietary analysis, clinical examination, family history and biomarkers remain the basis of the advice. Nutrition genetic tests add context to that, never the contents of the plate.

Family history remains the most accessible tool for nutritional risk analysis. A history of celiac disease, stones or early diabetes guides the advice. It does so more reliably than a commercial score. It justifies targeted screening rather than a restrictive diet. No consumer test today replaces this structured approach.

Some people look in a genotype report for an external marker to sustain a difficult dietary process. Making the limits of the report explicit soothes that search for certainty. Structured dietary change programs achieve their results through known mechanisms: repetition, feedback, follow-up and human support. A genotype report replaces none of these elements.

Genetics and nutrition: three clinical situations where the result changes the course of action

Genetics takes precedence over clinical common sense in three documented situations. Lactose intolerance is confirmed in the face of digestive symptoms, and the dietary course of action follows. HLA-DQ2 and DQ8 typing guides celiac disease screening, in addition to serology. Pharmacogenetics adjusts the dose of certain treatments in people on multiple medications. These situations are rare and fall under a medical prescription.

Lactose intolerance illustrates the difference between a robust gene-phenotype link and a weak association. Partial intolerance does not require removing all dairy products. The course of action goes through adapting quantities and the timing of consumption. The tolerance gain is verified clinically. It is observed by tracking symptoms after adaptation.

These situations fall under a medical prescription, not a wellness approach. Celiac typing is interpreted with the doctor who prescribed it, taking serology into account. Pharmacogenetics fits into a review of the prescription, especially in people on multiple medications. This requirement has a cost. It guarantees the only correct reading of the result.

Interview about metabolic nutrition with a person being supported
Detailed dietary analysis and clinical examination remain the basis of any personalization.

Commercial genetic test offers are assessed on verifiable arguments

A commercial test offer must be assessed on verifiable arguments before any use in support. The source of the data, the reference population and data protection must be known. In France, bioethics law frames the use of genetic data and requires explicit consent. The cost, regularly above one hundred and fifty euros, is measured against the absence of demonstrated benefit.

The technical reproducibility of these tests is correct. The problem does not lie there. It concerns the interpretation, the quality of the databases used and the clinical usefulness of the recommendation produced. The precision of a test does not guarantee the relevance of what it displays. A precise report on variants with no functional consequence can prove anxiety-provoking rather than useful. It then takes the place that belongs to behavioral goals.

A clear stance reassures more than dogmatic refusal. The practitioner can explain what research allows, what it does not allow, and when genetics takes precedence over clinical common sense. This transparency conditions a lasting dialogue with a motivated person who is highly exposed to commercial promises. The dietary decision remains based on clinical needs, never on a score.

The professional applies seven checks before accepting a genetic report

The first point concerns the source of the data and the publication of the reference population. The second checks the number of variants analyzed, a high figure often serving to disguise low statistical power. The third requires the identification of a responsible professional. The following ones concern data protection and the date of the result.

The seventh criterion decides everything: the existence of a course of action associated with the result. Without a validated course of action, a result has no decision value, whatever the number of variants analyzed. This grid is applied calmly, without opposing a refusal in principle. It refocuses effort on examinations with real decision value.

A report obtained several years ago is reread with caution, because knowledge of variants has evolved a great deal. The professional compares the document with current recommendations and notes the date it was obtained. The genome does not change over a lifetime. A second test therefore brings no new information to the same person. The money saved can fund behavioral support.

Documentary research and verification of scientific evidence
Seven checks are enough to decide whether a genetic report deserves to enter the file.

Sources

  1. 1.Grimaldi KA, et al. Proposed guidelines to evaluate scientific validity and evidence for genotype-based dietary advice. Genes Nutr. 2017.
  2. 2.Janssens ACJW. Validity of polygenic risk scores: are we measuring what we think we are? Hum Mol Genet. 2019.
  3. 3.Ordovas JM, et al. Personalised nutrition and health. BMJ. 2018.
  4. 4.Kirk D, et al. Precision nutrition: a systematic literature review. Comput Biol Med. 2021.
  5. 5.Voruganti VS. Precision nutrition: recent advances in obesity. Physiology (Bethesda). 2023.
  6. 6.Kiani AK, et al. Polymorphisms, diet and nutrigenomics. J Prev Med Hyg. 2022.
  7. 7.Duarte MKRN, et al. Obesity and nutrigenetics testing: new insights. Nutrients. 2024.

Reading studies is one thing, turning them into a plan is another

The Longevity Practitioner certification teaches the method that turns this kind of data into an accompaniment decision: assess, prioritise, plan over 90 days, re-measure, and know when to hand over.

See the certification

11 modules, €2,200. Enrol

Benjamin Sant

Co-founder of the School of Longevity

Cofounder of the School of Longevity with Tristan Maruyama. A former officer trained at Saint-Cyr from 2019 to 2022 and a coach since 2018, he has supported more than 4,000 people. He leads the School's editorial content with one constant principle: pass on only what can be situated, verified and applied seriously.

This article is published for information and education. It does not replace medical advice. For any personal health question, consult a healthcare professional.