Researchers report on 8 April 2026 that genetic differences help explain why individuals respond very differently to modern weight-loss medications. A genome-wide analysis of almost 28,000 people who used 23andMe and reported taking GLP-1–targeting drugs found variants associated with modestly greater weight loss and other variants tied to gastrointestinal side effects. The study measured outcomes over a median of eight months and linked one receptor variant to roughly 0.76 kg extra loss for carriers of a single copy and about 1.5 kg for homozygotes. Authors and outside experts emphasize that genetics is just one component of a multifactorial response to these medicines.
Key takeaways
- Sample size and source: The analysis used self-reported treatment and outcome data from nearly 28,000 23andMe customers who said they had taken prescription weight-loss drugs (median follow-up ~8 months).
- Genetic effect on weight: Carriers of one copy of a GLP-1 receptor variant lost on average 0.76 kg more than non-carriers; two copies corresponded to ~1.5 kg greater loss over the study interval.
- Side-effect risk variants: Several genetic markers were linked to higher odds of gastrointestinal adverse events, including nausea, with larger effect sizes than those seen for weight-loss magnitude.
- Drugs covered: The cohort included users of GLP-1 agonists (for example, semaglutide) and dual GLP-1/GIP agents (for example, tirzepatide), which have shown mean losses of roughly 10% body weight in randomized trials but wide individual variation.
- Clinical relevance: Investigators describe the genetic associations as modest and not yet actionable for routine clinical decision-making, though they see potential for future risk stratification research.
- Limitations: Data are self-reported, participants were drawn from a consumer genetics service with demographic skew, and causal mechanisms linking variants to outcomes remain to be proven.
Background
Next-generation anti-obesity medications mimic gut-derived hormones that control appetite and metabolism. Semaglutide acts on the glucagon-like peptide-1 (GLP-1) receptor; tirzepatide combines GLP-1 and glucose-dependent insulinotropic polypeptide (GIP) activity. Large randomized trials have reported mean weight losses of about 10% with semaglutide, but individual responses range from negligible change to losses exceeding 25%, underscoring substantial heterogeneity in efficacy.
The question of why responses vary has clinical urgency because these drugs are increasingly prescribed and considered for long-term management of obesity and metabolic disease. Genetic studies can point to biological pathways that alter drug sensitivity or side-effect susceptibility and may eventually guide safer, more effective use. However, prior pharmacogenomic findings in obesity therapeutics have been limited by small samples and inconsistent replication.
Main event
The new study analyzed hundreds of thousands of common genetic variants in participants who reported which drug they used, duration of use, weight change, and adverse effects. Researchers conducted genome-wide association scans for both efficacy (weight change) and reported side effects. The largest association for efficacy was a variant in the gene that encodes the GLP-1 receptor; carriers of one allele had on average 0.76 kg greater weight loss over a median 8 months, and homozygotes about 1.5 kg more.
For adverse events, the team identified multiple loci associated with increased risk of gastrointestinal symptoms such as nausea and vomiting. The magnitude of these associations was larger than the effect sizes seen for weight loss, suggesting genetics may be more informative for predicting tolerability than efficacy. The authors note that side effects often drive discontinuation in real-world use, making tolerability a critical outcome.
Co-authors stressed that the findings come from observational, self-reported data collected through a consumer platform, and they ran sensitivity analyses to test robustness to confounders such as age, sex, baseline BMI and medication duration. Outside experts praised the large sample while warning that measurement error and selection bias (who buys and uses at-home genetic testing) could influence results.
Analysis & implications
Biologically, a receptor variant that alters GLP-1 signalling plausibly changes both appetite suppression and gastrointestinal responses, because GLP-1 pathways operate in the brain and the gut. Even modest per-person effects can be informative: discovering relevant genes helps map the mechanisms of drug action and side effects and identifies targets for next-generation drugs with improved tolerability.
Clinically, the small average differences in weight loss (under 2 kg even for homozygotes) mean the results are not ready to guide prescribing. Physicians will likely continue to base treatment decisions on clinical factors, comorbidities and patient preference rather than genotype. Nevertheless, variants tied to substantially higher odds of nausea could inform strategies for pre-emptive side-effect management if replicated in clinical cohorts.
For regulators and drug developers, the study highlights the importance of monitoring adverse events across diverse populations. If genetic predictors of intolerance are validated, they might be used in trials to stratify participants or to design dosing and adjunctive therapies that minimize discontinuation rates. Economically, pharmacogenomic insights could affect payer decisions if genotype-informed care improves adherence and long-term outcomes.
Comparison & data
| Measure | Reported value |
|---|---|
| Cohort size | ~27,984 participants |
| Median follow-up | 8 months |
| Average weight loss in semaglutide trials | ≈10% (some >25%, some ~0%) |
| GLP-1 receptor variant effect | +0.76 kg (one copy), ~1.5 kg (two copies) |
The table summarizes key numerical findings. The 10% figure for semaglutide comes from controlled clinical trials; the genetic-effect estimates come from self-reported outcomes and reflect average differences rather than guaranteed individual results. Comparative interpretation must account for trial protocol differences and the observational nature of the 23andMe data.
Reactions & quotes
Authors and external commentators framed the study as informative but not immediately clinical. Their statements emphasize both scientific value and practical limits.
“The large sample size makes the genetic associations compelling, but the effect on weight loss is modest and not yet a clinical decision tool,”
Andrea Ganna, University of Helsinki (academic comment)
Investigators affiliated with 23andMe noted the larger genetic signal for side effects and the multifactorial nature of weight change.
“Genetics is one of many factors that influence weight loss; the association with gastrointestinal side effects was more substantial in our analysis,”
Adam Auton, 23andMe Research Institute (study co-author)
Unconfirmed
- Whether the identified variants will predict individual response well enough to guide prescribing in clinical practice is unproven and requires prospective validation.
- The causal biological mechanisms linking specific receptor variants to nausea or other gastrointestinal side effects have not been established.
- Generalisability to populations not represented among 23andMe customers—by ancestry, socioeconomic status or health status—remains uncertain.
Bottom line
This large, observational genetic analysis points to measurable but modest effects of common variants on weight loss from GLP-1–targeting drugs and stronger genetic signals for gastrointestinal side effects. The work is valuable for illuminating biology and for generating hypotheses that can be tested in clinical trials and more diverse cohorts.
For clinicians and patients today, the findings do not support routine genetic testing to choose or withhold these medications. Future research that replicates these loci in clinical trial datasets and maps mechanisms could enable genotype-informed strategies to improve tolerability and adherence, which are key to realizing long-term benefits of these therapies.