Your genes may partially determine how many kilograms you will lose when taking the popular drugs from the GLP‑1 agonist group - as well as whether you will be tormented by nausea as a side effect. This is the conclusion reached by the authors of a large-scale study published this week in the journal "Nature". The results suggest that patients' hereditary characteristics affect both the effectiveness and tolerability of therapies such as semaglutide ("Wegovy") and tirzepatide ("Mounjaro", "Zepbound").
The study was conducted by scientists from the "23andMe" Research Institute and the University of Copenhagen. The team analyzed genetic data and self-reports from nearly 28,000 people taking GLP‑1 receptor agonists, including semaglutide and tirzepatide, to track the relationship between specific DNA variants, weight loss and side effects.
What the researchers found
Within a genome-wide association study, a missense variant rs10305420 in the GLP1R gene was identified, associated with a more pronounced weight reduction in people treated with GLP‑1 agonists. According to the publication, each copy of the so-called "effective" allele is associated with an average additional weight loss of 0.76 kg. No signs of a dominant effect were found, which speaks of a rather additive, dose-dependent contribution of the variant to the effectiveness of the therapy.
A second key observation links the variant rs1800437 in the GIPR gene to a higher risk of nausea and vomiting in patients taking tirzepatide. However, this variant is not associated with the amount of weight loss. Variants in GLP1R have also shown a link to side effects in the wider group of GLP‑1 agonist users, suggesting that genetic differences in the receptor may affect both the benefits and the adverse reactions.
"Our study identifies a sustained genetic association with both the effectiveness of weight reduction with GLP‑1 drugs and the associated side effects," the authors write, but emphasize that the observed effect is "moderate" and in itself is not decisive for the outcome.
Why experts are calling for caution
Independent experts welcomed the results as an important step towards understanding individual differences in treatment response, but warned of their limited clinical significance at this stage. Dr. Marie Spreckley, head of a research program at the University of Cambridge, commented that the work "provides biologically plausible evidence that variations in the drugs' target molecule itself (GLP1R) and in related pathways (GIPR) contribute to inter-human variability in response."
At the same time, she warned against making hasty conclusions. According to Spreckley, "non-genetic factors – such as gender, specific medication, dose and duration of therapy – obviously explain a much larger share of this variability." She adds that for now "there is not enough evidence to use genetic information as a basis for therapeutic decisions in routine clinical practice", and that these results are more research-oriented than directly applicable.
A step towards personalized medicine in the treatment of obesity
The authors of the study point out that they have included genetic data in a broader model for predicting the response to GLP‑1 therapies, which allows for the stratification of patients according to expected effectiveness and the risk of side effects. In their words, this work "lays the foundation for personalized medicine approaches in the treatment of obesity" – for example, in future clinical trials or in a more precise selection of candidates for different types of drug therapy.
The study builds on a genetic program launched by "23andMe" in August 2024, which aims to clarify why GLP‑1 receptor agonists work much better in some people than in others. Against the backdrop of the fact that today tens of millions of patients around the world are taking such drugs, even moderate genetic effects can in perspective help doctors personalize therapy – but for now, the strongest predictors of success remain factors such as lifestyle, concomitant diseases and drug regimens, and not the DNA profile itself.