Why Popular Diabetes Drugs Like Ozempic Don’t Work for Everyone: The “Genetic Glitch”

A large international study suggests that a subset of people with Type 2 diabetes carry genetic variants that blunt the effectiveness of widely used GLP-1–based drugs, despite unexpectedly elevated levels of the hormone in their bodies.

GLP-1 receptor agonists are now used by more than a quarter of people with Type 2 diabetes, and their popularity continues to rise with the widespread use of brands like Ozempic and Wegovy for weight loss.

However, new research from Stanford Medicine suggests these drugs may not work equally well for everyone. The study points to genetic differences in about 10% of the population that may quietly limit how effectively these medications control blood sugar.

The researchers identified a phenomenon called GLP-1 resistance, in which the body produces higher levels of the hormone GLP-1 (glucagon-like peptide-1) but fails to respond to it properly. This is unexpected because GLP-1 normally helps lower blood sugar, slow digestion, and reduce appetite. The findings suggest that simply having more of the hormone is not enough. What matters is how the body processes and responds to it.

Published in Genome Medicine, the study draws on a decade of global research, combining human experiments, mouse models, and clinical trial data.

“In some of the trials, we saw that individuals who had those variants were unable to lower their blood glucose levels as effectively after six months of treatment,” said Anna Gloyn, DPhil, professor of pediatrics and genetics and a senior author of the study. She noted that patients who do not respond are often switched to different medications, and identifying likely responders in advance could help guide treatment decisions and support precision medicine.

The study was co-led by Markus Stoffel, MD, PhD, of ETH Zurich. Lead authors included Mahesh Umapathysivam, MBBS, DPhil, of Adelaide University, and Elisa Araldi, PhD, of the University of Parma.

“When I treat patients in the diabetes clinic, I see a huge variation in response to these GLP-1-based medications and it is difficult to predict this response clinically,” Umapathysivam said. “This is the first step in being able to use someone’s genetic make-up to help us improve that decision-making process.”

This work represents the first detailed look at GLP-1 resistance, although its underlying cause is still unknown.

“That is the million-dollar question,” Gloyn said. “We have ticked off this enormous list of all the ways in which we thought GLP-1 resistance might come about. No matter what we’ve done, we’ve not been able to nail precisely why they are resistant.”

Unexpected resistance

The researchers focused on two variants that disrupt an enzyme called PAM (peptidyl-glycine alpha-amidating monooxygenase), which activates several hormones, including GLP-1.

“PAM is a truly fascinating enzyme because it’s the only enzyme we have that’s capable of a chemical process called amidation, which increases the half-life or the potency of biologically active peptides,” Gloyn said. “We thought, if you have a problem with this enzyme, there’s going to be multiple aspects of your biology that are not working properly.”

These variants are already known to be more common in people with diabetes and to impair insulin release. To test their effect on GLP-1, researchers studied adults with and without a specific variant, p.S539W.

Participants drank a sugary solution, and their blood was monitored every five minutes for four hours. The team initially expected lower GLP-1 levels in those with the variant.

“What we actually saw was they had increased levels of GLP-1,” Gloyn said. “This was the opposite of what we imagined we would find.”

Despite these higher levels, there was no improvement in blood sugar control.

“Despite people with the PAM variant having higher circulating levels of GLP-1, we saw no evidence of higher biological activity. They were not reducing their blood sugar levels more quickly. More GLP-1 was needed to have the same biological effect, meaning they were resistant to GLP-1.”

Seeking confirmation

The findings were so unexpected that they prompted Gloyn’s team to spend several years verifying the results.

“We couldn’t understand this, which is why we looked as many different ways as we could to see if this was a really robust observation,” she said.

To investigate further, the researchers worked with scientists in Zurich who were studying mice lacking the PAM gene. These animals showed a similar pattern, with high GLP-1 levels that failed to properly regulate blood sugar, consistent with GLP-1 resistance.

GLP-1 and the drugs that mimic it help slow the movement of food through the stomach, a process called gastric emptying, which supports both blood sugar control and weight loss. However, mice without the PAM gene had faster gastric emptying, and treatment with a GLP-1 receptor agonist did not correct this effect. The researchers also found reduced responses to GLP-1 in both the pancreas and the gut, even though the number of GLP-1 receptors in these tissues remained unchanged.

In collaboration with researchers in Copenhagen, the team showed that defects in PAM do not affect how GLP-1 binds to its receptor or how signaling occurs at the receptor level. This indicates that the resistance likely arises further along the biological pathway.

Results may vary

To assess whether this resistance affects treatment outcomes, the researchers analyzed data from several clinical trials involving people with diabetes. A meta-analysis of three studies including 1,119 participants found that individuals with PAM variants responded less well to GLP-1 receptor agonists and were less likely to lower their HbA1c, a measure of long-term blood sugar.

After six months, about 25% of participants without the variants reached target HbA1c levels, compared with 11.5% of those with the p.S539W variant and 18.5% with the p.D563G variant.

The genetic variants did not influence how patients responded to other common diabetes medications such as sulfonylureas, metformin, and DPP-4 inhibitors. “What was really striking was that we saw no effect from whether you have a variant on your response to other types of diabetes medications,” Gloyn said. “We can see very clearly that this is specific to medications that are working through GLP-1 receptor pharmacology.”

Two additional industry-funded trials, which were not included in the meta-analysis because of differences in study design, reported similar responses between those with and without the variants. These studies used longer-acting GLP-1 drugs, which Gloyn said may help overcome some aspects of the resistance.

A complex puzzle

Gloyn’s group first identified GLP-1 resistance nearly a decade ago, before these drugs became widely used for weight loss. Only two of the trials included weight data, and they showed no clear difference between groups, although the evidence remains limited.

Large datasets from clinical trials may hold further clues about how genetics affects responses to GLP-1 therapies, including weight loss. “It’s very common for pharmaceutical companies to collect genetic data on their participants,” she said. “For the newer GLP-1 medications, it would be useful to look at whether there are genetic variants, like the variants in PAM, that explain poor responders to their medications.”

The biological basis of GLP-1 resistance is still unknown and is likely complex. Gloyn compares it to insulin resistance, which remains only partly understood despite decades of research, although treatments have been developed.

“There are a whole class of medications that are insulin sensitizers, so perhaps we can develop medications that will allow people to be sensitized to GLP-1s or find formulations of GLP-1, like the longer-acting versions, that avoid the GLP-1 resistance,” she said.

Reference: “Type 2 diabetes risk alleles in peptidyl-glycine alpha-amidating monooxygenase influence GLP-1 levels and response to GLP-1 receptor agonists” by Mahesh M. Umapathysivam, Elisa Araldi, Benoit Hastoy, Adem Y. Dawed, Hasan Vatandaslar, Johanna E. Mayrhofer, Peter Lindquist, Pamuditha N. Silva, Algera Goga, Geraldine O. Trüllinger, Svenja Godbersen, Shahana Sengupta, Adrian Kaufmann, Søren Krogsgaard Thomsen, Bolette Hartmann, Yi-Chun Chen, Anna E. Jonsson, Hasan Kabakci, Swaraj Thaman, Niels Grarup, Christian T. Have, Lindsay P. Pallo, Kristine Faerch, Anette P. Gjesing, Sameena Nawaz, Jane Cheeseman, Matthew J. Neville, Oluf Pedersen, Mark Walker, Han Sun, Christopher Jennison, Andrew T. Hattersley, Jens F. Rehfeld, Rury R. Holman, Bruce C. Verchere, Torben Hansen, Fredrik Karpe, Jens J. Holst, Mette M. Rosenkilde, Angus G. Jones, Michael Ristow, Mark I. McCarthy, Ewan R. Pearson, Markus Stoffel and Anna L. Gloyn, 29 March 2026, Genome Medicine.
DOI: 10.1186/s13073-026-01630-0

The study received funding from Wellcome, Medical Research Council, European Union Horizon 2020 Programme, the National Institutes of Health (grants U01-DK105535, U01-DK085545 and UM-1DK126185), the National Institute for Health Research Oxford Biomedical Research Centre, the Canadian Institutes of Health Research, the Novo Nordisk Foundation, Boehringer Ingelheim and Diabetes Australia.

Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.