Obesity and diabetes medicine has moved a long way from the single-hormone era. Dual and triple agonists are engineered molecules that switch on two or three hormone receptors at once. They now dominate the late-stage pipeline, and the highest doses tested have produced mean weight reductions approaching 30 per cent in randomised trials. Making sense of them means understanding three hormones: GLP-1, GIP and glucagon.
What dual and triple agonists actually are
Dual and triple agonists are named for how many receptors they hit. An agonist binds to a cell surface receptor and switches it on, mimicking the hormone that normally does the job. A dual agonist activates two receptors, a triple agonist three. The engineering challenge is substantial, because a single peptide must fit more than one receptor at useful potency. That is not the same as combining two drugs in one syringe.
Tirzepatide, sold as Mounjaro and Zepbound, remains the best known example. It is a 39 amino acid synthetic peptide built on the GIP sequence, with a fatty di-acid chain that extends its circulating half-life and allows weekly dosing. The molecule activates both the GIP and GLP-1 receptors, though not evenly: pharmacology studies indicate greater relative engagement of the GIP receptor, an imbalance that appears deliberate.
Retatrutide adds a third receptor. The investigational once-weekly peptide from Eli Lilly activates GIP, GLP-1 and glucagon receptors together. Published receptor pharmacology shows it is more potent at the human GIP receptor than native GIP, and less potent at the glucagon and GLP-1 receptors than the natural ligands. Balancing those three signals is the central design problem for triple agonists.
Reading the numbers: a note on estimands
Most efficacy figures in this field are reported in two versions, and the difference is material. The efficacy estimand models what would have happened had every randomised participant remained on treatment. The treatment-regimen estimand reflects the average effect across everyone randomised, whether or not they stayed on the drug. The efficacy estimand always gives the larger number, and is what companies lead with in topline announcements. Both are given below wherever disclosed.
In the phase 3 survodutide trial discussed below, participants on placebo lost 5.4 per cent of body weight on the treatment-regimen estimand, well above the 2 to 3 per cent the investigators had anticipated. The published paper records that 16.5 per cent of the placebo group reported taking another medication with GLP-1 receptor agonist activity while still enrolled. In that trial at least, the placebo arm was not an untreated arm in any straightforward sense, which complicates comparison with programmes where such use was lower or was not reported.
GLP-1: the pathway that started it all
Glucagon-like peptide-1 is released from L cells in the intestine within minutes of eating. It stimulates glucose-dependent insulin secretion, suppresses glucagon release, slows gastric emptying and acts on hypothalamic circuits to increase satiety. Semaglutide, the GLP-1 receptor agonist behind Ozempic and Wegovy, established this pathway as the foundation of modern weight management.
Developers have nonetheless moved beyond single-pathway GLP-1 agonism, on the working assumption that it reaches a practical ceiling. That assumption has shaped every dual and triple agonist now in late-stage testing, though it remains an inference from comparative trial results rather than a settled finding.
GIP: the receptor that divides the field
Glucose-dependent insulinotropic polypeptide is secreted by K cells in the upper small intestine and is the more powerful of the two incretins for stimulating insulin release. Its role in body weight, however, has been contested for decades. GIP raises glucagon secretion where GLP-1 suppresses it, and beta cell responsiveness to GIP is blunted in type 2 diabetes.
Clinical evidence nonetheless favours adding GIP agonism, and most dual and triple agonists in late-stage testing include it. SURMOUNT-5 (NCT05822830) was the first published head-to-head comparison of the two mechanisms in obesity without diabetes, following an earlier comparison in type 2 diabetes in the SURPASS-2 trial. It randomised 751 adults with obesity, or overweight with at least one weight-related complication, and without type 2 diabetes, to tirzepatide or semaglutide at maximum tolerated dose for 72 weeks. The New England Journal of Medicine published the results online in May 2025, alongside presentation at the European Congress on Obesity. Least-squares mean weight reduction was 20.2 per cent with the dual agonist against 13.7 per cent with semaglutide. Waist circumference fell by 18.4 cm and 13.0 cm respectively.
The trial was open-label and funded by the manufacturer of the better-performing drug, a limitation worth holding in mind. The direction of the result has nonetheless held up across subsequent analyses.
Glucagon: burning energy rather than blunting appetite
Glucagon is best known for raising blood glucose, which makes it a counterintuitive target in metabolic disease. Its appeal lies elsewhere: glucagon receptor activation increases energy expenditure and drives hepatic fat oxidation, and pairing it with GLP-1 offsets the glucose-raising effect while adding a second route to weight loss.
Survodutide, an investigational agent developed by Boehringer Ingelheim under licence from Zealand Pharma, leads this approach in Western markets. The phase 3 SYNCHRONIZE-1 trial (NCT06066515) randomised 726 adults with obesity, or overweight with at least one obesity-related complication, and without type 2 diabetes, of whom 725 received treatment, to weekly injections at 3.6 mg or 6.0 mg, or placebo.
The trial’s primary analysis used the treatment-regimen estimand. On that measure, weight fell by 12.2 per cent at 3.6 mg and 13.0 per cent at 6.0 mg after 76 weeks, against 5.4 per cent on placebo. On the efficacy estimand the same doses reached 15.3 and 16.6 per cent, against 3.2 per cent. Up to 85.1 per cent of treated participants lost at least 5 per cent of body weight on the efficacy estimand, against 38.8 per cent on placebo.
Full results were presented at the American Diabetes Association Scientific Sessions in June 2026 and published simultaneously in the New England Journal of Medicine.
Tolerability shaped the trial. Gastrointestinal adverse events led to discontinuation of the trial regimen in 17.8 per cent of the 3.6 mg group and 20.2 per cent of the 6.0 mg group, against 2.9 per cent on placebo.
The market read those figures harshly. Zealand Pharma shares closed almost 23 per cent lower on 8 June 2026, the worst performer on the Stoxx 600 that day, with analysts at Barclays and Citi arguing that a pooled discontinuation rate near 19 per cent sat above commercially viable levels for this class. Investigators pointed to the rigid dose-escalation schedule.
A pre-specified sub-study reported visceral fat reductions of up to 34 per cent and liver fat reductions of up to 63.1 per cent at the highest dose, with lean mass loss accounting for no more than 10.8 per cent of the change in total tissue mass at that dose. Those figures come from magnetic resonance imaging in 25 participants per group, a small sample. The profile is nonetheless why glucagon-containing dual and triple agonists are being positioned for liver disease as well as obesity, a case strengthened by the SYNCHRONIZE-MASLD results at the same meeting.
China reached this market first. The National Medical Products Administration approved mazdutide for chronic weight management in June 2025, in adults with a body mass index of 28 or above, or 24 or above with a weight-related comorbidity. Approval for glycaemic control in type 2 diabetes followed that September. Innovent Biologics developed the oxyntomodulin analogue under licence from Lilly. Innovent describes it as the first glucagon and GLP-1 dual agonist approved anywhere in the world, a characterisation carried through to the Adis drug profile published in Drugs in September 2025.
What the trial data show for dual and triple agonists
Among the dual and triple agonists tested in humans, the strongest efficacy figures belong to retatrutide, although that comparison rests on separate trials with different durations and populations rather than head-to-head data. Lilly now describes the molecule as having succeeded across five phase 3 studies. TRIUMPH-1 (NCT05929066) randomised 2,339 adults with obesity or overweight, at least one weight-related comorbidity and no diabetes, to 4 mg, 9 mg or 12 mg of retatrutide, or placebo. At 80 weeks, mean weight loss on the efficacy estimand was 19.0, 25.9 and 28.3 per cent respectively, against 2.2 per cent on placebo. On the treatment-regimen estimand the same doses delivered 17.6, 23.7 and 25.0 per cent, against 3.9 per cent.
On the efficacy estimand, 45.3 per cent of those on the highest dose lost at least 30 per cent of body weight, against 0.5 per cent on placebo.
A pre-specified extension enrolled 532 participants with a baseline body mass index of 35 or above who had completed the main study on treatment. At 104 weeks, those continuing on 12 mg to maximum tolerated dose lost 30.3 per cent, the 4 mg and 9 mg arms escalated to maximum tolerated dose reached 27.9 and 29.5 per cent, and participants switched from placebo reached 19.2 per cent.
Ania Jastreboff is professor of medicine and paediatrics at Yale School of Medicine and lead investigator on the trial. In the sponsor’s announcement she said that every dose of retatrutide produced clinically meaningful weight reduction for nearly all participants.
Two further phase 3 readouts followed in July 2026. Lilly reported that TRIUMPH-2 and TRIUMPH-3, in adults with type 2 diabetes and in adults with severe obesity and established cardiovascular disease respectively, both met their primary endpoints, with weight loss of up to 22.6 per cent at 80 weeks in TRIUMPH-3. The company has said it intends to submit a Biologics License Application for retatrutide to the FDA in the first quarter of 2027.
None of the three trials has yet been published in a peer-reviewed journal. Further TRIUMPH-1 results were presented at the American Diabetes Association Scientific Sessions in June 2026, but the TRIUMPH-2 and TRIUMPH-3 data remain sponsor topline releases. Readers should weigh them against the published SURMOUNT-5 and SYNCHRONIZE-1 datasets.
Tolerability, and what the data do not yet settle
Tolerability remains the constraint on the most potent dual and triple agonists. In TRIUMPH-1, nausea affected 42.4 per cent of participants on 12 mg against 14.8 per cent on placebo, and vomiting 25.3 per cent against 4.8 per cent. Discontinuation due to adverse events reached 11.3 per cent on the highest dose, against 4.9 per cent on placebo, though at 4 mg the rate was 4.1 per cent, below placebo.
Dysaesthesia, an altered or unpleasant sensation typically affecting the skin, is an unusual signal for this class and is thought to relate to glucagon receptor activity. It occurred in 12.5 per cent of the 12 mg group, against 0.9 per cent on placebo. The sponsor reported these events as generally mild to moderate, mostly resolving during treatment, with most affected participants continuing. Independent characterisation is not yet available.
Two label-level risks apply across the approved agents in this class. Tirzepatide and semaglutide both carry a boxed warning for thyroid C-cell tumours based on rodent data, and both are contraindicated where there is a personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia syndrome type 2. Pancreatitis is a labelled warning for each. Retatrutide and survodutide, being investigational, carry no label yet.
One earlier concern has resolved in the other direction. The European Medicines Agency’s Pharmacovigilance Risk Assessment Committee concluded in April 2024 that the evidence did not support a causal association between GLP-1 receptor agonists and suicidal or self-harming thoughts. In January 2026 the FDA asked manufacturers to remove the suicidal ideation and behaviour warning from GLP-1 labelling, following a meta-analysis of 91 trials and more than 100,000 participants that found no increased risk against placebo.
Three questions remain genuinely open. No dedicated cardiovascular outcomes trial has yet reported for any dual or triple agonist other than tirzepatide, so the assumption that greater weight loss translates into fewer cardiovascular events is untested for this group. TRIUMPH-3 enrolled adults with established cardiovascular disease and took weight as its primary endpoint, and its pre-specified cardiovascular analyses were inconclusive because events occurred less frequently than anticipated in every arm, with hazard ratios of 0.82 (95 per cent CI 0.55 to 1.22) for five-component MACE and 1.12 (0.64 to 1.96) for three-component MACE. SYNCHRONIZE-CVOT remains under way. Body composition reporting is inconsistent between programmes, which makes lean mass preservation hard to compare across trials. And durability after discontinuation is largely unstudied at these levels of weight loss.
Why some dual and triple agonists block a receptor instead
Not every developer accepts that GIP should be switched on. Maridebart cafraglutide from Amgen, known as MariTide, is an investigational peptide-antibody conjugate that activates the GLP-1 receptor while blocking the GIP receptor, the opposite of the tirzepatide approach. Human genetics work linking reduced GIP receptor signalling to lower body weight underpinned the design.
Phase 2 data published in the New England Journal of Medicine in June 2025 showed weight loss of 16.3 to 19.9 per cent at 52 weeks on the efficacy estimand, against 2.6 per cent on placebo, in participants without type 2 diabetes. Dosing was monthly or less frequent, and no clear plateau appeared. The phase 3 MARITIME programme is now under way across obesity, type 2 diabetes and several related conditions.
Two opposing GIP strategies therefore deliver broadly comparable weight loss, albeit in separate trials, which limits the comparison.
Where dual and triple agonists go next
Receptor stacking is not stopping at three. Boehringer Ingelheim began a phase 2 trial of BI 3034701 in July 2026 (NCT07662122), a triple agonist targeting GLP-1, GIP and the NPY2 receptor rather than glucagon. NPY2 agonism acts on central hunger signalling and is the least established of the three mechanisms. No efficacy data have been released. Others are pairing incretins with amylin analogues rather than adding further receptors to a single peptide.
Funding is the constraint that receptor engineering cannot solve. List prices for the current generation already strain health system budgets, and the most potent agents are unlikely to arrive cheaper. Access is uneven, as GLP-1 uptake by country shows, and more effective agents may widen rather than narrow that gap. The wider field of late-stage obesity drugs now runs to at least seven phase 3 candidates, and the GLP-1 drug pipeline continues to broaden beyond the incretins alone.
For now, only one dual agonist is approved in the United States and Europe: tirzepatide. Mazdutide is approved in China. Novo Nordisk’s CagriSema, a fixed-dose combination rather than a single multi-receptor peptide, remains under FDA review, with a decision expected in late 2026 and no action date publicly confirmed. Whether the triple agonists join them depends on regulatory submissions beginning in 2027. On current evidence, dual and triple agonists have shifted the ceiling of medical weight management well beyond the first generation of GLP-1 drugs. Whether that shift changes long-term health outcomes, and for whom, is the question the next round of trials has to answer.
