The Book of PeptidesThe Book of Peptidesthe "triple-G"
1923 (glucagon reported as a contaminant of pancreatic extracts); 2013 (Imperial College: GLP-1 cancels glucagon’s glucose rise); 2015 (first monomeric triagonist); 2019 (LY3437943 first-in-human)
Glucagon entered medicine in 1923 as an impurity — the hyperglycemic junk fouling early insulin extracts, named for driving blood sugar up. A century later, chemists built it into a diabetes drug on purpose.
Glucagon arrived in the medical literature in 1923 as a nuisance. C.P. Kimball and John R. Murlin at the University of Rochester described a hyperglycemic factor riding along inside pancreatic extracts — the same extracts then being purified into the first commercial insulin — and named it for what it did: the substance that pushes glucose up. For the century that followed, that stayed its whole reputation: glucagon is the emergency-kit hormone, the injection that pulls someone out of severe hypoglycemia. So a drug for obesity and type 2 diabetes that deliberately switches the glucagon receptor on reads, at first glance, like a category error — an accelerator bolted onto a brake.
The counter-clues had been in print almost as long. By 1957, Schulman and colleagues were already studying glucagon's effect on food intake and body weight in humans, in the Journal of Applied Physiology. Clinicians knew that patients with glucagonoma — a rare tumor that floods the blood with glucagon — tend to waste away; a 2025 review of 86 reported cases recorded weight loss in about 62% of those whose clinical features were tabulated. And glucagon does one thing GLP-1 does not: a 2016 study in Diabetes, Obesity and Metabolism reported that infusing it raised human energy expenditure roughly 15% — comparable, within that same experiment, to cold exposure. The problem was never that glucagon did nothing. The problem was the blood sugar.
The fix was demonstrated at Imperial College London, in a small crossover study published in Diabetes in 2013. Tan and colleagues infused ten overweight volunteers with glucagon alone, then glucagon plus GLP-1: alone, glucagon raised plasma glucose exactly as the textbooks promised; with GLP-1 added, that excursion rapidly flattened while the extra energy expenditure survived. Nature had already sketched the trick — oxyntomodulin, a gut hormone that activates both the GLP-1 and glucagon receptors, had cut body weight against placebo in a 2005 Imperial trial and raised total energy expenditure 9.4% in a 2006 follow-up. Glucagon's liability could be cancelled by the very drug class it was being bolted to.
The core holds: glucagon really was first reported in 1923 as a hyperglycemic contaminant of pancreatic extracts, and it really was switched on deliberately. The lone-gamble framing is the embellishment — glucagon's effect on food intake and body weight was under human study by 1957, oxyntomodulin (nature's own dual agonist) went through human trials in 2005–2006, and the triagonist concept came from academia, not the sponsor. Nor is it an even three-way split: the discovery paper reports more GIP activity.
Retatrutide is not approved by any regulator, anywhere. Lilly's phase 3 TRIUMPH program has so far reported through company announcements — TRIUMPH-1's 28.3% mean weight loss at 80 weeks on 12 mg, versus 2.2% on placebo, among them — and as of August 2026 none of those phase 3 results appear in the peer-reviewed literature. The announcements carried a tolerability cost, equally unpublished: roughly 18% of participants on the highest dose in TRIUMPH-4 stopped for adverse events, and dysesthesia — abnormal skin sensation, not a familiar GLP-1 class effect — was reported in up to 12.5% versus 0.9% on placebo, rising with dose. What is published is the phase 2 record: 24.2% mean weight loss at 48 weeks on 12 mg, dose-related gastrointestinal effects, and a dose-dependent heart-rate rise. Meanwhile the compound moves briskly through research-chemical channels labeled not for human use.
Turning that into one molecule ran largely through Richard DiMarchi — the chemist credited with inventing the insulin analog Humalog at Eli Lilly, inducted into the National Inventors Hall of Fame in 2014 — who carried the problem into academia at Indiana University. His collaboration with Matthias Tschöp produced a dual glucagon/GLP-1 agonist in rodents in 2009, then the first monomeric triagonist in Nature Medicine in 2015 — adding GIP to strengthen the incretin effect and, in the authors' own framing, moderate glucagon's diabetogenic effect. Lilly's LY3437943 first entered a human in March 2019; its 2022 Cell Metabolism discovery paper describes balanced glucagon and GLP-1 activity but more GIP — not an even split. In the phase 2 diabetes trial, the top dose cut HbA1c by 2.02% versus 0.01% on placebo and 1.41% on the GLP-1 agonist dulaglutide.