Two genetic variants linked to the effects of GLP-1 drugs for obesity

This appears to be a scientifically robust and relevant study, as it identifies genetic associations in GLP-1R and GIPR—that is, in genes encoding the biological targets of semaglutide and tirzepatide—thereby providing clear mechanistic plausibility for the findings. Moreover, the main signal associated with greater weight loss was observed in a large sample and externally replicated in a cohort based on electronic health records, which strengthens the robustness of the result. Overall, the work aligns well with existing evidence showing substantial inter-individual variability in response to these drugs (as with most interventions) and represents a step towards more refined precision medicine approaches in obesity.

However, the clinical implications should be interpreted with caution. Although the genetic signal reaches clear statistical significance—the main association in GLP1R showed a P value of 2.9 × 10⁻¹⁰ and was externally replicated in All of Us (P = 0.001)—the effect size remains modest, at around 0.76 kg of additional weight loss per allele for the lead variant. The authors demonstrate that a combined model including both clinical and genetic variables can stratify patients, but this does not imply that genetics alone is sufficient to define clinical subgroups or to justify routine genetic testing at this stage. Indeed, the study itself indicates that most of the predictive capacity still derives from non-genetic factors such as sex, age, type 2 diabetes, type of drug, dose, and treatment duration.

In this respect, the findings resemble those observed in many cardiometabolic traits—such as type 2 diabetes, glucose levels, HbA1c, BMI, or blood pressure—where genetic associations are robust and biologically informative, yet of limited individual predictive value when considered in isolation. Furthermore, a substantial proportion of the data is self-reported, the cohort is enriched for women and individuals of European ancestry, and the authors are current or former employees of 23andMe (a company specialising in direct-to-consumer genetic testing). It will therefore be important to validate these findings prospectively, independently, and in more diverse populations.

This is a remarkably elegant study that reinforces what we have long been saying. Obesity is a highly complex disease in which genetics plays a major role—although it does not explain everything, it accounts for a substantial part. It is also worth noting that this work has been published in a high-impact journal, Nature, and is based on a robust methodology, analysing around 27,000 individuals through a genetic study.

One limitation is that weight loss was self-reported by participants, meaning the data are subject to reporting bias. Nevertheless, the study provides valuable insight into a common clinical challenge: individuals respond differently to treatment. We often refer to hyper-responders and hypo-responders, and each patient may also experience different side effects. Genetics could help explain both phenomena, at least in part.

The study is both relevant and timely, as it sheds light on a current issue in clinical practice. It fits within the broader framework of what we call precision medicine—essentially, the ambition that genetic information will one day allow us to select the most appropriate drug or tailor the optimal treatment for each individual.

That said, genetics is not the whole story. As noted, obesity is a complex condition. Current evidence shows that biological factors such as sex (with women tending to lose more weight), the presence of type 2 diabetes (associated with less weight loss), age (with older individuals typically losing less weight), and fatty liver disease (also linked to reduced weight loss) all play a significant role.

Looking ahead, we are likely to integrate clinical predictors of both treatment efficacy and side effects with biological variables. We are at the beginning—indeed, the dawn—of a new era in medicine: precision medicine.

In this particular study, variants in the GLP-1R gene, located on chromosome 6, are analysed and may help explain why some individuals respond better to treatment, given the mechanism of action of these drugs on the GLP-1 receptor. Similarly, variants in the GIPR receptor gene, located on chromosome 19, may account for differences in side-effect profiles. For instance, certain variants may impair the GIPR pathway, potentially reducing the protective effect against nausea and increasing the likelihood of adverse events.

From my perspective, this points towards a highly promising future. The ability to anticipate treatment response through pharmacogenomics represents a major milestone, enabling a more precise approach not only in selecting therapies but also in optimising their use—distinguishing responders from non-responders and identifying those at higher risk of side effects.

At present, we have two highly effective drugs on the market with strong safety profiles—tirzepatide (Mounjaro) and semaglutide (Wegovy)—which are considered essential therapies. In the future, additional molecules with different mechanisms of action are expected to emerge, making genetic support even more important in guiding treatment selection, always in combination with the patient’s phenotype and clinical characteristics.

This study marks the beginning of a new era in medicine, based on the integration of genetic and clinical variables that we are already accustomed to managing in everyday practice.

Is this a high-quality study?

"This study is very interesting and provides relevant data. However, its design has some limitations, as it is not a pre-designed and registered clinical trial; rather, it uses data from participants who had previously taken part in genetic studies conducted by a genetic testing company for a different purpose. The study now analyzes individuals taking weight-loss medications and their weight loss, linking these outcomes to their genetic characteristics. What is novel is that they find a strong association between a genetic variant in the GLP-1R gene and greater weight loss when taking the drug. This finding is important. They also find some associations between genetic variants in the GLP-1R and GIPR genes and some of the adverse effects of weight-loss drugs."

Are there any limitations to consider?

“The limitation is that it needs to be replicated in other studies before being incorporated into clinical practice, as it does not have the design of a randomized, controlled clinical trial specifically designed for this purpose.”

What are the implications, and how does this fit with existing evidence?

“The implications are significant because it contributes new evidence on how genetic variants can determine the effects of the same drug, making them more effective in some people than others, and whether or not they may be associated with adverse effects. It fits very well within the existing evidence, and these results will be applicable in the new era of precision medicine.”

Genetic effects are modest compared to non-genetic factors. What does this imply for clinical practice?

“In this case, the genetic effect is not very small in magnitude. It is just one genetic variant in the context of significant weight loss. It amounts to nearly a kilo difference in weight lost with the same treatment and duration. If other genetic variants are found, the effect may be cumulative, and the important thing is that it provides new relevant data in pharmacogenetics and the genetic epidemiology of obesity. Likewise, the association of certain genetic variants with greater or lesser adverse effects, such as vomiting, etc., is relevant for better personalizing treatments, including with new weight-loss drugs.”

Is the study of high quality?

“Yes, it’s a solid piece of work. It analyzes nearly 28,000 people and uses advanced genetic tools (GWAS), in addition to validating some of the results with real clinical data. Methodologically speaking, it’s a robust and well-executed study.”

Are there any limitations to consider?

“Much of the data is self-reported, which can introduce bias. Additionally, the sample is predominantly drawn from the U.S. population of European descent, so the results cannot be automatically extrapolated to other countries like Spain. And, above all, the genetic effects detected are small.”

What are the implications, and how does this fit with existing evidence?

“It aligns well with expectations: it confirms that there is a biological and pharmacogenetic basis for the response to GLP-1, but also that clinical variability is multifactorial. It reinforces the path toward precision medicine, although still in its early stages, and is consistent with previous evidence of great heterogeneity in response to these drugs.”

Genetic effects are modest compared to non-genetic factors. What does this imply for clinical practice?

“It implies that, at present, it is much more useful and efficient to adjust treatment based on modifiable clinical variables (dose, adherence, comorbidities, age, sex) than to rely on genetic testing. Genetics may provide complementary information in the future, but its current clinical utility is limited. Furthermore, from a modern perspective on obesity, environmental factors—which also act via epigenetic mechanisms—remain the primary determinants and, therefore, the main therapeutic target.

In summary: the study provides highly relevant evidence as a starting point toward personalized medicine in obesity. However, at this time the priority should not be to generalize the use of genetic testing, but rather to develop more comprehensive approaches that integrate clinical data, environmental factors, and, likely, epigenetic markers, since the latter better reflect the interaction between the environment and each patient’s biology.”

The study appears to be well-conducted and, from a methodological standpoint, can be considered of high quality. It addresses a clinically relevant question and provides interesting evidence regarding the potential genetic contribution to variability in response to GLP-1 receptor agonists, both in terms of weight loss and adverse effects. Furthermore, it aligns reasonably well with existing evidence, which suggests that the response to this type of treatment is heterogeneous and likely multifactorial. In that sense, the study is valuable because it contributes to a better understanding of that variability and opens up a promising line of research in the field of personalized medicine.

However, I believe the practical implications of the findings are somewhat overemphasized in the final interpretation. Although genetic associations are identified, the effect sizes are modest and few, and from a clinical standpoint, their immediate utility appears limited. Non-genetic factors such as age, sex, baseline metabolic status, adherence, or tolerance to treatment likely carry greater weight and, furthermore, are more accessible and applicable variables in routine clinical decision-making. Therefore, although the study is interesting from a biological standpoint and may be useful for generating hypotheses or for future, more comprehensive predictive models, the results alone do not yet justify direct clinical application or the routine use of genetic testing to guide treatment.