Get the next issue in your inbox — clear, evidence-based GLP-1 analysis, every week.

For three years, we ranked these drugs like a leaderboard. Semaglutide put up 15%. Tirzepatide answered with 21%. Then a triple agonist crossed 28%, and every headline treated that number like a finish line sprinting away from us.

I want to argue the opposite. The most important thing that happened to obesity pharmacology in the last 18 months is not that the top number got bigger. It is that the field fragmented. Into pills and injections. Into appetite drugs and metabolism drugs. Into weight drugs and organ drugs. The single ladder became a menu, and a menu is exactly what a chronic, biologically heterogeneous disease always needed.

Start with the ground truth, because the pipeline noise makes people forget what is already sitting in the pharmacy.

Semaglutide (Wegovy, Ozempic) is the workhorse single agonist. It hits the GLP-1 receptor and nothing else, and at the standard 2.4 mg dose it delivers roughly 15% weight loss over 68 weeks. Tirzepatide (Zepbound, Mounjaro) added the GIP receptor and pushed the class into the low-to-mid 20s, around 21 to 23% in SURMOUNT-1. Those two drugs defined the first era, and that era was a ladder. One receptor, then two, more weight with each rung.

Then 2026 broke the ladder in three places at once.

First, the incumbent leveled up. In March 2026 the FDA cleared Wegovy HD, a 7.2 mg dose of semaglutide that lifted the old workhorse to 20.7% at 72 weeks. Second, the pill era arrived. Novo's oral semaglutide was approved in December 2025, and on April 1, 2026, the FDA cleared orforglipron, brand name Foundayo. This one matters more than its weight-loss figure suggests. It is the first non-peptide oral GLP-1 ever approved: a small molecule that survives stomach acid, with no fasting window, no refrigeration, and no needle. Third, the ceiling itself kept climbing, which is where the triple agonist comes in.

Orforglipron produces only about 11 to 12% weight loss, less than half of what retatrutide does. That is the point. It is not competing on the leaderboard. It is competing on the front door.

If you only track the weight-loss percentage, the last two years look like a single line going up. If you track mechanism, you see the real structure. The field has organized itself around four hormonal levers, and every drug in development is some combination of them.

Read that table again and the whole pipeline decodes itself. Tirzepatide is GLP-1 plus GIP. Survodutide, mazdutide, and pemvidutide are GLP-1 plus glucagon. CagriSema and amycretin are GLP-1 plus amylin. And retatrutide is the only molecule that stacks three at once: GLP-1, GIP, and glucagon.

Here is the conceptual shift worth internalizing. Semaglutide and tirzepatide make you eat less. The glucagon component makes you burn more and clear fat out of your liver. That is a different physiological attack on the same disease. When you add glucagon, you stop treating obesity purely as a problem of intake and start treating it as a problem of energy balance and organ health. That distinction is the entire second half of this article.

The whole board on one screen. Weight-loss figures are peak reported values and are not directly comparable across trials of different length, dose, and population.

Retatrutide is the drug everyone is watching, and 2026 is the year it stopped being a promise. In the Phase 2 obesity trial published in the New England Journal of Medicine, the 12 mg dose produced 24.2% weight loss at 48 weeks, a record at the time. Then the pivotal readout landed, and it was bigger.

On May 21, 2026, Eli Lilly reported topline results from TRIUMPH-1, the first registrational Phase 3 obesity trial. At the 12 mg dose, participants lost 28.3% of body weight at 80 weeks. In the study extension, patients with a baseline BMI of 35 or higher who stayed on treatment reached 30.3% at 104 weeks, and 45.3% of all participants crossed 30% weight loss, a threshold long associated with bariatric surgery. An earlier trial, TRIUMPH-4 in obesity plus knee osteoarthritis, had already shown 28.7% at 68 weeks in December 2025, alongside meaningful osteoarthritis pain relief.

One caveat deserves emphasis, and I say this as someone who reads these releases for a living. These are topline press releases, not yet full peer-reviewed papers, with detailed data slated for the American Diabetes Association meeting and journal publication. The broader TRIUMPH program spans multiple large Phase 3 trials, with several more readouts expected through 2026 and an FDA filing anticipated late 2026 into 2027. Treat the numbers as strong signals, not settled facts.

Still, the trajectory is unmistakable. Each generation has added roughly 5 to 10 points. Single agonist to dual added about 6. Dual to triple added 6 to 8 more. We are now looking at surgical-magnitude weight loss, sustained to 2 years, from a once-weekly injection. That reframes what is possible for patients who were told surgery was their only option.

If retatrutide is the story of how high we can go, orforglipron and oral amycretin are the story of how many people we can reach. And reach is where population health actually lives.

Injectable peptides carry hidden friction: cold-chain logistics, manufacturing bottlenecks, needle aversion, and a specialty-pharmacy gauntlet that quietly filters out patients before they ever start. A shelf-stable pill that a primary-care physician can prescribe on a Tuesday afternoon collapses most of that friction. In the head-to-head ACHIEVE-3 trial, orforglipron beat oral semaglutide on both blood sugar and weight, and it does not require the 30-minute fasting ritual that makes the oral semaglutide protocol so hard to sustain in real life.

Novo's answer is amycretin (zenagamtide), and it is the more scientifically interesting pill. It fuses GLP-1 and amylin into a single molecule, and its early data are startling: 13.1% weight loss in just 12 weeks as an oral tablet, with the curve still descending at the end of the study. The injectable version reached up to 24% at 36 weeks. Most incretin drugs take 36 to 72 weeks to plateau. Amycretin was still accelerating at week 12. Phase 3 began in early 2026.

The lesson for practice is not that pills will replace injections. The market is splitting into a convenience tier and a maximum-efficacy tier, and most patients belong in the first. The person who needs 10% to reverse prediabetes and the person who needs 25% to get off a CPAP machine are no longer competing for the same prescription.

This is the part of the landscape I think clinicians are underrating. The glucagon receptor does something none of the pure appetite drugs do. It tells the liver to burn its own fat. That makes the GLP-1/glucagon duals less like weight-loss drugs and more like steatohepatitis drugs that happen to cause weight loss.

The Phase 2 data are hard to ignore. Survodutide (Boehringer Ingelheim and Zealand) hit its primary endpoint in biopsy-confirmed MASH: in the published trial, up to 62% of patients on the 4.8 mg dose achieved improvement in MASH without worsening of fibrosis at 48 weeks, versus 14% on placebo. You will also see a headline figure of 83% attached to survodutide. That was the topline responder number from the initial press release; the more conservative 62% is the published, prespecified analysis, and it is the one to quote.

Pemvidutide (Altimmune), a balanced 1:1 GLP-1/glucagon agonist, delivered a 54.7% reduction in liver fat at its top dose in the IMPACT trial and earned FDA Breakthrough Therapy Designation for MASH. And mazdutide (Innovent) became one of the first GLP-1/glucagon dual agonists approved anywhere, cleared in China as Xinermei. Even the triple agonist plays a role here: retatrutide showed dramatic liver fat clearance in a dedicated Phase 2a MASLD trial, which is almost certainly the glucagon component doing its work.

Why does this reframe the field? Because MASH is on track to become a leading cause of liver transplant, and until recently we had almost nothing for it. If the Phase 3 programs confirm these signals, a single weekly injection could treat obesity and liver disease at the same time. The question shifts from "how much weight did the patient lose" to "did we resolve the steatohepatitis?" That is a fundamentally different clinical conversation, and it is happening now.

Not every next-generation bet has paid off cleanly, and honesty about that is what separates analysis from hype. CagriSema, Novo's fixed-dose combination of semaglutide plus the amylin analog cagrilintide, was built to be the answer to tirzepatide. In REDEFINE 1 it delivered about 22.7% weight loss, clearly additive over each component alone. The market had hoped for closer to 25%, and the stock fell hard on the miss. Novo filed with the FDA in December 2025 anyway, with a decision expected late 2026.

Then came the head-to-head. In REDEFINE 4, CagriSema was tested directly against tirzepatide and failed to meet non-inferiority: 23.0% versus 25.5% under the efficacy estimand, and 20.2% versus 23.6% when real-world adherence was factored in. A two-drug amylin combination lost to a single-molecule dual agonist across 84 weeks. That result does not sink the program, and CagriSema will likely still reach the market as the first amylin-based product. But it punctures the assumption that stacking pathways automatically wins. Amylin is a real, additive lever. It is not a trump card.

The strategic tell is what Novo did next. It pushed amycretin, the single-molecule GLP-1/amylin drug, hard into Phase 3. The future of the amylin pathway is probably unimolecular and oral, not a two-drug co-injection.

For a decade, obesity medicine had one tool and one question: how much weight can we take off? In 2026, we have a mechanism map, two oral options, a triple agonist near-filing, and drugs that resolve liver disease as a byproduct of treating obesity. The horse race is over, not because someone won, but because we no longer need a single winner.

The work now is matching a heterogeneous disease to a differentiated toolkit. The needle-averse patient with early metabolic dysfunction and the patient with biopsy-proven MASH and a BMI of 45 no longer get the same prescription, because they never should have. That is not a complication. That is obesity medicine finally growing into the chronic-disease framework it always deserved.

The best news in the field is that it got complicated. Complexity, here, is just another word for options.

REFERENCES