# The Genetics Are Real, and Small

By Nora Bissett, Pricing Editor (https://glpwatchdog.org/authors/nora-bissett) | Published: 2026-08-20 | Updated: 2026-10-27

Canonical page: https://glpwatchdog.org/research/the-genetics-are-real-and-small

> A genome-wide study found genetic predictors of GLP-1 weight loss and side effects, at overwhelming significance. The efficacy effect is about three-quarters of a kilogram per copy.

## Key points

- Four genome-wide-significant variants, in the receptor genes the drugs bind.
- The efficacy effect is ≈0.76 kg per allele, against a median total loss of 11.3 kg (11.7%) in the same study.
- Side-effect models reach AUC 65.4% and 68.0%. Chance is 50%; the usual clinical floor is 70%.
- Replicated in All of Us, did NOT replicate in UK Biobank.
- This corrects an earlier article of ours that said no such predictors had been shown.
- 23andMe self-report data, European ancestry, 82.4% women. 23andMe also sells genetic tests.

A genome-wide study of 27,885 people who had taken these drugs found genetic predictors of both how much weight they lost and how sick the drugs made them.[1] The associations are real, statistically overwhelming, and located exactly where biology says they should be. But the efficacy effect is **about three-quarters of a kilogram per copy of the variant**. That gap, between a real finding and a useful test, is the subject of this page.

> **This corrects something we published.** Our earlier piece, [the genetic test that isn’t ready](https://glpwatchdog.org/research/the-genetic-test-that-isnt-ready), concluded that nobody had shown a genetic test predicts GLP-1 response. That was written from a 27-person pilot and this study supersedes it. Associations *have* now been shown. We have corrected that article rather than leaving it standing.

## What was found

Researchers surveyed 23andMe customers who reported using a GLP-1 medication, and ran a genome-wide association study on 15,237 people of European ancestry with complete data (median age 52; 82.4% women). Four variants reached genome-wide significance.

*Genome-wide significant associations. ΔBMI% is percentage change in BMI; OR is the odds ratio per allele.[1]*

| Variant | Gene | Outcome | Effect | P |
| --- | --- | --- | --- | --- |
| rs10305420 | *GLP1R* | Weight loss | −0.641% ΔBMI per allele (≈ −0.76 kg) | 2.9 × 10^−10 |
| rs9357296 | *GLP1R* | Nausea | OR 1.36 | 2.6 × 10^−28 |
| rs11760106 | *GLP1R* | Vomiting | OR 1.57 | 2.5 × 10^−27 |
| rs1800437 | *GIPR* | Vomiting, **tirzepatide only** | OR 0.546 (protective) | 5.1 × 10^−9 |

These are not marginal findings. A P value of 10^−28 is not a fluke, and the variants sit in the receptor genes the drugs bind: *GLP1R* for both drugs, *GIPR* for the one that also targets GIP. The GIPR variant affected only tirzepatide users, which is exactly what you would predict if the biology is what it looks like. Internal coherence of that kind is strong evidence.

## Now look at the size of the efficacy effect

> Three-quarters of a kilogram, per copy of the variant.

The efficacy association is −0.641% of BMI per allele, which the authors put at roughly 0.76 kg with a confidence interval from 0.34 to 1.27 kg. Someone carrying two copies is looking at something under two kilograms of expected difference, against a median total loss in this study of 11.3 kg.[1]

So a test built on this variant can tell you, with real statistical confidence, about somewhat less than a kilogram. That is a discovery about biology. It is not information that changes which drug you take or whether you take one.

> **The authors say so themselves.** They describe the genetic contribution as “relatively modest compared with that from non-genetic factors,” and note that combined models explained 25% of the variance in weight loss. *Combined* means genetics together with everything else. That figure is not a measure of how much genetics explains.

## The side-effect predictions are closer to useful

Predicting who will be badly nauseated is arguably the more valuable application, because it bears on decisions people face: [which drug to start](https://glpwatchdog.org/learn/glp-1-medications/semaglutide-vs-tirzepatide), how fast to titrate, whether to persist.

The models reach an area under the curve of 65.4% for nausea and 68.0% for vomiting. That measure runs from 50%, which is a coin flip, to 100%. A rough convention treats 70% as the floor for acceptable discrimination in a clinical tool. **Both of these fall below it.** They are better than guessing, but not yet good enough to act on.

The GIPR finding is the most interesting of the four for a different reason: it is drug-specific. A variant that protects against vomiting only on tirzepatide is the first thing here that could in principle inform a choice *between* drugs, which is what genetic testing in this area has always been sold as doing.

## What has and has not replicated

The efficacy variant was tested in two independent cohorts. In All of Us, with 4,855 participants and electronic health record data, it held (P = 0.001), with a directionally consistent effect of −0.47%. In UK Biobank it did not, although the authors note that analysis had low statistical power.

One replication and one failure is a mixed result. The low-power caveat is a reasonable explanation rather than an excuse, because an underpowered study that fails to replicate is weak evidence against. It is still a mixed result, and reporting only the half that worked would be the easy version of this article.

## Who ran it, and who it was run in

- **The data are self-reported.** Participants told a survey how much weight they lost and whether they felt sick. The All of Us replication used medical records, which is a stronger source.
- **The GWAS was restricted to European ancestry**, as most are for statistical reasons. Whether these variants behave the same in other populations is unaddressed.
- **82.4% of participants were women.**
- **The study comes from 23andMe**, a company that sells genetic testing. That is a material disclosure in an article about whether to buy a genetic test, and it is not a reason to discount a Nature paper with these statistics behind it.

## What this means if someone offers to test you

The position has moved, and to a more interesting place than either side of the old argument. There *are* real variants in the right genes with overwhelming statistical support. And the amount of your outcome they explain is small enough that [a test would not change a decision](https://glpwatchdog.org/research/genetics-caused-it-and-the-drug-still-worked).

So the questions to put to anyone selling one are now sharper, not softer: what effect size does this variant predict, in kilograms? What is the discrimination of your side-effect model, as a number? Has the association replicated? If a seller quotes the P value and not the effect size, they are quoting the half of this paper that sounds impressive.

The best-supported predictor of your response remains your response. The first weeks on the drug carry more information than any panel currently on offer, as covered in [losing fast early](https://glpwatchdog.org/research/losing-fast-early-what-it-predicts).

## Frequently asked questions

### Do genes predict how well a GLP-1 will work?

Partly, and by a small amount. A variant in the GLP1R gene is associated with weight loss at overwhelming statistical significance, but the effect is about 0.76 kg per copy against a median total loss of 11.3 kg (11.7% of body weight) in the same study.

### Is a genetic test worth buying now?

The findings do not yet support one. The efficacy variant predicts under a kilogram per copy, and the side-effect models reach 65.4% and 68.0% on a scale where 50% is a coin flip and 70% is the usual floor for a clinically acceptable tool.

### Could genes predict who gets sick on these drugs?

That is the more promising application. Variants associated with nausea and vomiting were found at very strong significance, and one GIPR variant protected against vomiting only in tirzepatide users. That is the first result here that could in principle inform a choice between drugs.

### Did the findings replicate?

The efficacy variant replicated in the All of Us cohort (P = 0.001, directionally consistent) and did not replicate in UK Biobank, where the authors note power was low. That is a mixed result.

### Who ran the study?

23andMe, using self-reported survey data from its customers, in 15,237 people of European ancestry, 82.4% of them women. The company sells genetic testing, which readers should know, but that is not a reason to dismiss a result with these statistics.

## References

1. Su QJ, Ashenhurst JR, Xu W, et al. Genetic predictors of GLP1 receptor agonist weight loss and side effects. Nature. 2026. PMID: 41951734. https://pubmed.ncbi.nlm.nih.gov/41951734/

---

Cite as: GLP Watchdog, Nora Bissett, "The Genetics Are Real, and Small", published 2026-08-20, updated 2026-10-27. https://glpwatchdog.org/research/the-genetics-are-real-and-small

Canonical HTML page: https://glpwatchdog.org/research/the-genetics-are-real-and-small

Medical disclaimer: Nothing here is medical advice. We report on this subject; we do not treat anyone. Whether to begin a medication, supplement, or treatment, whether to adjust it, and whether to stop are questions for a clinician licensed to answer them, who knows your history.
