GLP‑1 receptor agonists are among the success stories of modern diabetes care. They lower HbA1c, support weight loss, and—depending on the specific drug and the patient’s risk profile—offer additional cardiometabolic benefits. Nevertheless, a familiar phenomenon persists in practice: Some patients respond less well than expected, even though the administration, dose, and duration of therapy are correct.
A recent study in Genome Medicine provides a possible genetic explanation for this. The focus is on the PAM gene, which encodes peptidyl-glycine alpha-amidating monooxygenase.
This enzyme is involved in the amidation of numerous bioactive peptides—including GLP‑1. Two hypomorphic PAM variants studied, p.S539W and p.D563G, are associated with an increased risk of type 2 diabetes; combined, they are carried by about one in ten people.
More GLP‑1 does not mean increased effectiveness
In the study, PAM activity was significantly reduced in carriers of these variants: by about 52% for p.S539W and by about 20% for p.D563G.
Interestingly, carriers also had elevated circulating GLP‑1 levels. While this may initially sound like an advantage, it does not translate into a stronger effect. Reduced endogenous GLP‑1 sensitivity was particularly evident with p.S539W—consistent with the concept of GLP‑1 resistance.
The mouse data also fit this picture: Pam knockout mice exhibited accelerated gastric emptying, responded less well to exendin‑4, and showed attenuated signal transduction via the GLP‑1 receptor in the pylorus.
So what does this mean for treatment?
Clinically, the effect was measurable: In a meta-analysis of three cohorts with 1,119 participants, the reduction in HbA1c under GLP-1RA therapy was lower in p.S539W carriers than in non-carriers. This difference was not observed for metformin, sulfonylureas, and DPP‑4 inhibitors.
This does not yet change any guidelines. But PAM could help identify GLP-1RA non-responders earlier —and make treatment decisions less based on the principle of “let’s see if it works.”