GLP-1 drugs such as semaglutide (Ozempic, Wegovy) and liraglutide have changed how obesity and type 2 diabetes are treated. Yet doctors have known for a while that a share of patients barely respond. A study published in Genome Medicine in June 2026 has found part of the reason in a single gene, and it is a good example of where pharmacogenomics is heading: from antidepressants and blood thinners to metabolism itself.
What the study found
The team, led by Anna Gloyn (Stanford Medicine) and Markus Stoffel (ETH Zurich), focused on PAM, a gene coding for an enzyme that performs a finishing step called amidation on many peptide hormones, which makes them last longer and act more strongly. Two variants, p.S539W and p.D563G, are carried by roughly 10% of the population. Using data from three clinical trials with 1,119 participants with diabetes, the researchers compared how carriers and non-carriers responded to GLP-1 treatment.
After six months, about 25% of non-carriers had reached the recommended HbA1c target. Among carriers of p.S539W, only 11.5% did, and 18.5% among carriers of p.D563G. The surprising part is the mechanism: carriers actually had higher GLP-1 levels in the blood, but needed more of it to get the same effect. The authors call it GLP-1 resistance. Only two of the trials had weight-loss data, and there the evidence was too thin to conclude anything, so this is currently a finding about blood sugar, not about the scales.
What it means, and what it does not
This is not a reason to stop or change a treatment on your own. What it shows is that the same drug, at the same dose, can act very differently depending on a variant most people do not know they carry, exactly the logic behind pharmacogenomics for statins, anticoagulants or antidepressants. It also shows the limits of today's consumer tests: PAM is not part of the panels read by 23andMe-style chips, and the Fuel Your DNA report does not read it either. Anyone promising to predict your response to Ozempic from a saliva kit is ahead of the science.
What your DNA can tell you is how the biology these drugs act on is set in your body: appetite, satiety signalling, insulin response and reward eating. That is useful whether you take a GLP-1 or not.
The genes your DNA report reads
- FTO: the best-replicated appetite gene. Risk carriers feel less full after a meal and are drawn to energy-dense food, the exact feeling GLP-1 drugs dampen.
- MC4R: a key satiety switch in the brain, part of the pathway through which GLP-1 signalling reduces hunger.
- TCF7L2: the strongest common type 2 diabetes variant. Studies link it to a weaker incretin response, the hormone system these drugs mimic.
- DRD2: dopamine receptor density, which shapes how rewarding food feels (see our article on DRD2 and ultra-processed food).
What to do with this
- If a GLP-1 is not working for you, say so to your doctor. Dose, drug class and timing can all be adjusted, and genetic testing for PAM may reach clinics in the coming years.
- Protein and fibre at every meal. Both raise natural GLP-1 and stretch satiety, and they work with or without medication.
- Keep a regular eating window. Late and frequent eating works against the very hormones these drugs boost.
- Use your genes to pick your levers. An FTO or MC4R carrier gains most from structure around meals and from protein; a TCF7L2 carrier from lower-glycaemic carbohydrate placed around activity.
Important: this article is educational and is not medical advice. Never start, stop or change a prescription medication based on a genetic report. Discuss any concern about your treatment with the doctor who prescribed it.
See how your appetite and blood-sugar genes are set
The Fuel Your DNA Ultimate report reads FTO, MC4R, TCF7L2 and DRD2 alongside a pharmacogenomics section covering 11 genes that affect statins, blood thinners and antidepressants. No new test needed, just the raw file you already have from 23andMe, AncestryDNA or MyHeritage. See the reports. For the wider picture, read why you can't lose weight: what your DNA says.