Some patients lose 30 percent of their body weight on drugs like Wegovy and Mounjaro. Others lose nothing at all. A study of nearly 28,000 people published today in Nature has identified a genetic explanation for the gap — and points toward a future where doctors could predict whether these drugs will work before writing the prescription.

Researchers at the 23andMe Research Institute conducted a genome-wide association study using data from 27,885 people who had used GLP-1 receptor agonists, the class of medications that includes semaglutide (Ozempic, Wegovy) and tirzepatide (Mounjaro, Zepbound). Participants lost a median of 11.7 percent of their body weight over roughly eight months of treatment — but the range was enormous. Previous research on semaglutide, cited in the study, found that while the average patient lost 10.2 percent of body weight, 4.9 percent of patients lost more than 25 percent, and 32.2 percent lost less than 5 percent or even gained weight.

The variant that matters

The study identified a missense variant — a single-letter DNA change that alters a protein — in the GLP1R gene, which encodes the receptor these drugs activate to suppress appetite. People carrying one copy lost an additional 0.76 kg (about 1.7 pounds) on average. Those with two copies roughly doubled that benefit.

“The genetic variant we found lands right in this gene [for] the GLP-1 receptor, which happens to be the target for these medications,” Adam Auton, study co-author and vice president of human genetics at the 23andMe Research Institute, told Scientific American.

The variant is unevenly distributed across populations. The variant’s effect allele has a frequency of roughly 40 percent in European and 38 percent in Middle Eastern populations, compared to about 7 percent in African populations, according to the study. That disparity raises questions about whether clinical trial results — drawn predominantly from white populations — overstate typical outcomes for other groups.

Side effects, written in genes

The researchers also identified genetic markers linked to nausea and vomiting, the most common side effects of GLP-1 drugs. Variants in both GLP1R and GIPR were associated with worse gastrointestinal symptoms. The GIPR variant appeared only in people taking tirzepatide, which targets both the GLP-1 and GIP pathways — a drug-specific genetic interaction not previously documented.

People carrying two copies each of both risk variants were approximately 15 times more likely to experience severe vomiting on tirzepatide than those without either variant, the study found.

Beyond DNA

Genetics alone explains only part of the variation. The study found that women lost more weight than men, younger patients responded better than older ones, and people with type 2 diabetes saw significantly less benefit — roughly a 2.9 percentage-point reduction in BMI loss compared to non-diabetics. Drug type, dose, and treatment duration all contributed. Together, these non-genetic factors accounted for about 21 percent of the variance in outcomes.

“Genetics is only one part of a much more complex picture,” Dr Marie Spreckley of the University of Cambridge told the BBC. Behavioral factors — diet, exercise, clinical support — remain central to how patients respond.

A test before the prescription

The study’s broader significance lies in the predictive model the researchers built from these findings. By combining genetic and clinical data, they demonstrated the ability to stratify patients by likely efficacy and side-effect risk. That model was validated against a separate electronic health record dataset of 4,855 patients from the All of Us research program.

23andMe has already released a report for its Total Health subscribers that provides personalized estimates of weight-loss and nausea risk based on the study’s genetic markers, alongside clinical factors. According to the company’s announcement, predicted nausea risk among research participants ranged from 5 to 78 percent depending on the combination of genetics and other variables — a spread wide enough to matter clinically.

Whether the findings constitute genuine clinical utility is contested. Professor Naveed Sattar, a metabolic health expert at the University of Glasgow, told the BBC: “Overall, these findings are scientifically interesting, but they are a long way from changing clinical practice.” Ruth Loos, an obesity geneticist at the University of Copenhagen who peer-reviewed the study, characterized the genetic effects as “not trivial.”

The current prescribing model for GLP-1 drugs is essentially trial and error — expensive, slow, and for roughly a quarter of patients, ineffective. A genetic test that reduced even some of that uncertainty would reshape the economics and experience of obesity treatment. Whether this study’s findings are robust enough to build that test on remains, as Sattar suggested, a question for future trials.

As Giles Yeo, an obesity geneticist at the University of Cambridge not involved in the study, told Scientific American: “Your genes do matter, but it’s not only your genes.”

Sources