Which GLP-1 Is Right for Me?

The GLP-1 medications are not interchangeable, but the differences between them are smaller and more specific than the marketing suggests. Semaglutide acts on the GLP-1 receptor. Tirzepatide acts on GLP-1 and on a second receptor, GIP. Retatrutide adds a third target and is still investigational, meaning it is not available by prescription.

That structural difference is worth understanding, because it is the reason the genetics of these drugs is not uniform, and it is the reason a variant that matters for one of them is irrelevant for another.

This is the first filter, and it is largely an approval question. The same molecule is sold under different brand names for different indications, and which one you are prescribed depends on whether you are being treated for type 2 diabetes or for weight management. Coverage usually follows the indication rather than the molecule, which is why two people on chemically identical drugs can have completely different experiences at the pharmacy counter.

Across head-to-head and comparable trials, tirzepatide has generally produced greater average weight loss than semaglutide. That is a real difference and it is usually the strongest argument for the dual agonist. It is an average, though, and individual results vary widely around it, which is worth holding on to before treating the average as a prediction about you.

Nausea, vomiting, constipation and diarrhoea are the most common reasons people stop these drugs, and they are the main practical constraint on the whole class. Most of it is dose-related and improves with slower titration, so how quickly you escalate often matters more than which drug you are on. If you have stopped a GLP-1 before because of nausea, that history should shape the next attempt more than anything else on this list.

A personal or family history of medullary thyroid carcinoma or MEN2 rules this class out. A history of pancreatitis, significant gastroparesis or active gallbladder disease all change the calculus. If you are on insulin or a sulfonylurea, adding a GLP-1 raises the risk of hypoglycaemia and usually means adjusting those. These are absolute or near-absolute considerations and they come before preference.

This is where pharmacogenetics contributes, and its contribution here is narrow but real. One variant in the GIP receptor is associated with a substantially higher chance of moderate-to-severe vomiting on tirzepatide, and shows no such association with semaglutide.^[1] Because tirzepatide acts on that receptor and semaglutide does not, the effect is specific to the drug rather than to the class.

That makes it a genuine drug-choice signal, which is unusual. It is not a reason to avoid tirzepatide outright, but it is a reason to titrate more slowly, or to consider starting with the other one if tolerability has already been a problem.

In practice this decides more cases than anything above it. These medications are expensive, coverage varies enormously, and supply has been intermittent. A drug you can obtain and afford consistently will outperform a theoretically better one you have to keep interrupting, because the weight tends to return when treatment stops.

Two genes carry the useful signal here, and they do different jobs. One is associated with how much weight people lose. The other is associated with side effects, and only on the drugs that act through it. Keeping them separate is the whole point.

These are receptor genes rather than metabolising enzymes, so the usual metabolizer language does not apply. Nobody is a poor metabolizer of semaglutide. The question is how your receptors respond, not how quickly you clear the drug.

It is also worth being blunt about the limits. Adding genetics to a model of weight loss improves how much of the variation it explains by a very small margin, so anything marketed as a genetic weight-loss prediction is overselling what these variants do.^[1] The side-effect signal is the part that changes a decision.

The realistic case for testing here is tolerability rather than efficacy. If you have already stopped tirzepatide because of nausea or vomiting, or you are choosing between the two and want to weight that risk, the GIPR result is directly relevant and points at a specific alternative. If your goal is to estimate how much weight you will lose, testing will not tell you much, and you should be sceptical of anything that claims otherwise.

One practical caveat: these genes are recent additions to pharmacogenetic reporting, so an older report may not include them.

• Retatrutide Pharmacogenetics: Triple Agonist Genetics (Investigational)
• Tirzepatide (Mounjaro, Zepbound) Not Working: GLP1R and GIPR Genetics
• Why Isn't Ozempic or Wegovy Working for Me? The GLP1R Connection
• Which Anti-Nausea Medication Is Right for Me?
• Which Anticonvulsant Is Right for Me?
• Which Antidepressant Is Right for Me?

On average, tirzepatide has produced greater weight loss in trials. That does not make it the better choice for everyone, because it is also the one affected by the GIPR variant associated with vomiting, and because cost, coverage and supply frequently decide the matter. Average superiority and individual suitability are different questions.

Not usefully. Adding genetic information to a model of weight loss improves its explanatory power by a very small amount, so any product claiming to predict your weight loss from your DNA is overstating the evidence. The genetics is far more informative about side effects than about results.

Possibly. The GIPR variant associated with moderate-to-severe vomiting acts through a receptor that tirzepatide targets and semaglutide does not, so that particular risk does not carry over. Nausea has several causes though, and much of it is dose-related rather than drug-specific, so a slower titration is often tried first. This is a conversation to have with your prescriber rather than a change to make on your own.

Retatrutide is investigational and not available by prescription. It acts on three receptors, including the GIP receptor, so the GIPR side-effect signal would be expected to apply to it as well, though that has not been established in the way it has for tirzepatide.

Not equally. The weight-loss association reached statistical significance only in participants of European ancestry, which is a common limitation in genetic studies and a reason to treat the numbers as less certain outside that group.^[1]