Semaglutide

GLP-1 is released by the gut after a meal. It helps the pancreas release insulin when blood sugar is high, slows the emptying of the stomach and acts on appetite centers in the brain. Semaglutide is built to resist the enzymes that clear natural GLP-1 within minutes, so a single injection lasts about a week [2].

The injections are taken once a week. Wegovy injection starts at 0.25 mg and steps up every 4 weeks, through 0.5 mg, 1 mg and 1.7 mg, to a usual maintenance dose of 2.4 mg; adults who tolerate 2.4 mg for at least 4 weeks can go up to 7.2 mg if more weight loss is needed. The Wegovy tablet is taken once a day, starting at 1.5 mg and building to 25 mg. The slow start is there to limit stomach and bowel side effects, and the label tells prescribers to consider delaying the next step by 4 weeks if a dose is not tolerated [2]. Ozempic follows its own diabetes schedule [3].

In the STEP 1 trial, adults with obesity or overweight who took 2.4 mg a week for 68 weeks lost 14.9% of their body weight on average, against 2.4% with placebo [4]. Individual results vary widely. Nausea, diarrhea, vomiting and constipation are the most common side effects, and they are most likely while the dose is going up [2].

GLP1R carries the instructions for the GLP-1 receptor, the protein semaglutide binds. Gene2Rx reads one position in it, rs10305420. The common form, C, codes for the amino acid proline at position 7 of the receptor's signal peptide, the short leading section that guides the new protein to the cell surface. The other form, T, codes for leucine (Pro7Leu). Everyone has two copies, so a result is C/C, C/T or T/T.

The study behind the guideline

The evidence comes from a genome-wide association study by the 23andMe research team, published in Nature in 2026 [1]. In a survey, 27,885 research participants described their experience with Ozempic, Wegovy, compounded semaglutide, Mounjaro, Zepbound or compounded tirzepatide: weight before and during treatment, dose, time on the drug and side effects. The researchers then scanned the genome for variants linked to weight loss and to moderate or severe side effects.

One signal stood out for weight loss: rs10305420 in GLP1R. Each T allele was associated with an extra 0.64% fall in body mass index, about 0.76 kg of weight. The effect was additive, so two copies meant roughly twice the effect of one. It held after allowing for type 2 diabetes and smoking, and it was repeated in an independent group of 4,855 people in the All of Us program, whose weights came from medical records [1].

The same region of GLP1R was linked to nausea and vomiting. The strongest markers there were associated with about 1.4 times the odds of moderate or severe nausea and 1.6 times the odds of moderate or severe vomiting per copy, and statistical tests suggested that the weight-loss and side-effect signals come from the same underlying variant. Put simply, the version of the receptor linked to more weight loss was also linked to more stomach trouble. The effect on side effects did not differ between semaglutide and tirzepatide [1].

How the change works is not settled. Leucine is more water-repelling than proline, and the authors suggest it may help the receptor reach the cell surface, so cells carry more receptor without the receptor itself working differently. Two earlier, much smaller studies, of liraglutide in women with polycystic ovary syndrome and of exenatide in people with type 2 diabetes, reported the opposite direction; the 2026 authors note those studies had little power to detect an effect of this size [1].

How big the effect is

The effect is real but small. For someone who loses 15 kg on semaglutide, one copy of the T allele is associated with less than one extra kilogram. Within the study, the effect looked smaller with semaglutide (0.51% of body mass index per copy) than with tirzepatide (0.95%), although the two estimates overlapped. A model built from age, sex, diabetes, dose and other details explained about a quarter of the differences in weight loss between people; adding the genetic variants raised that share by about a tenth of one percentage point [1].

How common it is

The allele frequencies come from the gnomAD database, as reported in the study [1]; the last column is calculated from them, assuming the two copies are inherited independently. The weight-loss association was statistically significant only in people of European ancestry. Estimates in the Latino, African American, East Asian, South Asian and Middle Eastern groups pointed the same way, but those groups were smaller [1].

The same study found a second variant, rs1800437 in GIPR, the gene for the receptor of a second gut hormone, GIP. Its Gln354 form was associated with more vomiting and nausea in people taking tirzepatide, which acts on both receptors. Semaglutide does not act on the GIP receptor, and in the study the GIPR variant made no difference to side effects with semaglutide [1]:

These are observed rates among survey participants, from the study's Supplementary Table 14, and are not adjusted for dose or time on treatment [1]. They show that the GIPR result is relevant to tirzepatide; it does not change anything for semaglutide, which is why the Gene2Rx semaglutide guideline uses GLP1R alone.

Gene2Rx wrote this guideline itself: CPIC, the FDA and the Dutch DPWG have published no pharmacogenetic guidance for semaglutide. The guideline adjusts expectations and points to the dose schedule; it never says to avoid semaglutide. The table summarizes what a report shows.

"Enhanced Response" is the report's label for the group linked to more weight loss. It describes an average difference of about a kilogram per copy, not a guarantee of a better result.

Most pharmacogenetic guidance concerns enzymes that break a drug down. This finding is different: the variant sits in the drug's own target, the receptor semaglutide acts on, which gives it a clear biological reason to matter. It was found by scanning the whole genome in a large group of real users, it reached genome-wide significance, it was repeated in an independent group with medically recorded weights, and the side-effect signal in the same gene fits the drug's biology, since GLP-1 medicines cause nausea through this receptor [1].

The effect is small, so Gene2Rx does not use it to say who should take semaglutide or how much weight anyone will lose. It uses it where a small shift is useful to know about: for people likely to feel more nausea, the dose schedule that already exists to manage side effects is the practical lever. That is also why every row of the guideline keeps standard dosing.

The guideline will change if better evidence arrives, for example studies that start people on the drug and follow them prospectively, or data on the Wegovy tablet, which was not part of this study.

If you carry one or two T alleles, tell the clinician who prescribes semaglutide, especially if you are about to start or have struggled with nausea. The label already allows the next dose step to wait 4 weeks when a dose is not tolerated [2]; your result is one more reason to raise that early rather than stop treatment. Do not change your dose or schedule on your own.

If you have no copies, your chance of side effects is not zero. Nausea is common with every GLP-1 medicine, and in the study about a quarter of people taking semaglutide reported moderate or severe nausea whatever their GIPR result [1].

Other factors matter far more than this variant for weight loss. In the study, weight loss was greater in women, in people taking higher doses for longer and with tirzepatide, and smaller in people with type 2 diabetes [1]. The labels also list practical points that are not genetic: semaglutide slows stomach emptying, which can affect how some tablets taken by mouth are absorbed, and it can cause low blood sugar when combined with insulin or a sulfonylurea [2].

This is an association found across thousands of people. It cannot tell you how much weight you will lose or whether you will feel sick. The study relied on what participants reported in a survey. Among the 195 people who also shared medical records, self-reported weight loss averaged 14.1% against 8.4% in their records. Most participants were women of European ancestry, and the drugs were mainly injections, including compounded versions [1].

Gene2Rx reads rs10305420 from your DNA data, measured directly or inferred by imputation from nearby markers. Consumer arrays (23andMe, AncestryDNA, MyHeritage) are not clinical-grade tests, and their raw data is labeled not for medical use. A typical result means the T allele was not found in the data that covers this position.

Many other genes, and much more than genes, shape how people respond to GLP-1 medicines. A typical result is not a reason to expect an easy course, and an enhanced-response result is not a reason to push the dose faster.

Semaglutide isn't working well for me. Is it my genes?

This variant is unlikely to be the main reason. Its effect is about a kilogram per copy, while dose, time on treatment, diabetes and other factors make much larger differences. The semaglutide not working guide covers the common causes.

Should I take tirzepatide instead?

The GLP1R variant points the same way with both drugs, so it does not favor one over the other. GIPR matters only for tirzepatide's side effects (see above). The which GLP-1 is right for me guide walks through the choice.

Does this apply to the Wegovy pill or Rybelsus?

Probably, since the tablets act on the same receptor, but the study did not include them, so the evidence is indirect.

1. Su QJ, Ashenhurst JR, Xu W et al. Genetic predictors of GLP1 receptor agonist weight loss and side effects. Nature 653:770-775 (2026). The study behind the guideline, including its Supplementary Tables 14 (side effects by GIPR genotype) and 22 (model performance).
2. Wegovy (semaglutide) injection and tablets: prescribing information. DailyMed. Uses, dose escalation, side effects and interactions.
3. Ozempic (semaglutide) injection: prescribing information. DailyMed. Uses and dosing in type 2 diabetes.
4. Wilding JPH et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1). N Engl J Med (2021). Average weight loss with semaglutide 2.4 mg.