What happened
In Genome Medicine, researchers at Stanford Medicine and ETH Zurich identified variants in the PAM gene, carried by roughly 1 in 10 people, that create a kind of "GLP-1 resistance." Oddly, carriers have higher natural GLP-1 levels, yet their bodies respond to it more weakly. Across three trials with 1,119 participants, after six months about 25 percent of non-carriers hit their blood-sugar targets, versus 11.5 percent for one variant (p.S539W) and 18.5 percent for another (p.D563G). Crucially, the variants did not affect metformin or other diabetes drugs, only the GLP-1 pathway.
Why this matters: this is precision medicine getting concrete. Instead of cycling patients through an expensive drug to see whether it sticks, a genetic marker could flag in advance who is unlikely to respond and route them to something that works. Professors Anna Gloyn at Stanford and Markus Stoffel at ETH Zurich built this over roughly a decade.
One gene is not the whole story of why these drugs vary person to person, and this is not yet a clinic-ready test. But matching the right metabolic drug to the right biology is exactly where weight and diabetes care should go. If your genes could tell you which treatment to skip, would you want to know first?
Source
Reported by Scientists discover why Ozempic may not work for some people via sciencedaily.com, published June 5, 2026.