A landmark decade-long study led by Stanford Medicine has uncovered a startling finding that could reshape how doctors prescribe one of the most popular classes of diabetes medications. Published in Genome Medicine, the research reveals that roughly one in ten people carry genetic variants that make GLP-1 receptor agonist drugs, including semaglutide sold under the brand name Ozempic, substantially less effective at controlling blood sugar levels. The discovery highlights a critical gap in the current one-size-fits-all approach to diabetes treatment.
The research team identified two specific variants in the PAM gene, known as p.S539W and p.D563G, that are classified as type 2 diabetes risk alleles. The PAM gene encodes peptidyl-glycine alpha-amidating monooxygenase, which is the only enzyme in the human body capable of performing amidation. This biochemical process is essential because it increases both the potency and the half-life of GLP-1 hormones, making them more effective at regulating insulin secretion and blood sugar levels. When the PAM enzyme functions poorly due to these genetic variants, the entire GLP-1 signaling pathway is compromised.
One of the most paradoxical findings of the study is that carriers of these genetic variants actually circulate higher levels of GLP-1 hormone in their bloodstream compared to non-carriers. Despite this abundance, their bodies remain significantly less responsive to the hormone and to drugs that mimic its action. Researchers describe this as a form of pathway-specific resistance, where the problem lies not in hormone production but in downstream processing and receptor activation. This paradox had previously confused clinicians who observed high GLP-1 levels yet poor treatment outcomes in certain patients.
The study drew on data from three large diabetes clinical trials encompassing 1,119 participants tracked over multiple years. The results were striking in their consistency across all three trials. Among patients carrying the p.D563G variant, only 18.5 percent reached their blood sugar management goals after six months of GLP-1 drug therapy. For those with the p.S539W variant, the success rate dropped even further to just 11.5 percent. These figures stand in stark contrast to the roughly 55 to 60 percent success rate observed in patients without either variant.
Critically, the resistance appears to be specific to the GLP-1 pathway rather than a general defect in glucose metabolism. Patients carrying these PAM gene variants responded normally to other classes of diabetes medications, including metformin, sulfonylureas, and DPP-4 inhibitors. This specificity suggests that genetic screening could help physicians identify patients who are unlikely to benefit from GLP-1 drugs before prescribing them, potentially saving months of ineffective treatment and allowing earlier use of alternative therapies.
The implications extend well beyond diabetes management. GLP-1 drugs have surged in popularity for weight loss, cardiovascular protection, and even emerging applications in addiction and neurodegenerative disease. If one in ten people carries variants that blunt the effectiveness of these drugs, millions of patients worldwide may be receiving suboptimal treatment without knowing why their results fall short of expectations. Stanford researchers are now calling for pharmacogenomic testing to be integrated into clinical practice before initiating GLP-1 therapy.
Public health experts have welcomed the findings as a step toward precision medicine in metabolic disease. The study underscores that genetic diversity plays a far larger role in drug response than previously appreciated. Researchers are already working on next-generation GLP-1 formulations designed to bypass the amidation bottleneck, which could potentially restore drug effectiveness even in patients with compromised PAM function. Clinical trials for these modified compounds are expected to begin within two years.
Comments