The Next Generation of Metabolic Drugs: Triple Agonists in 2026
Summary
By 2026, incretin-based pharmacology is no longer evaluated only by the number on the scale. In parallel clinical programs, the same drug class is being tested for large-scale body-weight reduction, reductions in liver fat consistent with fatty liver disease biology, hard cardiometabolic outcomes, and even disease-modification in neurodegeneration—with mixed results depending on the endpoint and disease area.[1–4]
The core story is a step-change in magnitude: in a phase 3 program, the triple agonist retatrutide reached mean weight-loss levels approaching those historically associated with bariatric surgery, while the newest oral GLP-1 agents deliver double-digit losses without injections.[1, 5]
Triple agonists
Retatrutide is designed to co-activate GLP-1, GIP, and glucagon receptors, and the clinical signal to date is consistently dose-dependent weight loss across populations studied.[2, 6]
In a phase 2 randomized trial in obesity (48 weeks; ), the 24-week primary outcome showed mean percent weight change ranging from (1 mg) to (12 mg), versus with placebo.[2, 6] By 48 weeks, the mean percent change ranged from to (12 mg), versus with placebo.[2, 6] Notably, the 48-week trajectory analysis reported “no evidence of plateau” through the end of follow-up, an observation that matters when considering long-term targets for obesity management.[6]
Responder analyses underscore how far the distribution shifts at higher doses. At 48 weeks, participants receiving 12 mg achieved at least , , and weight loss at rates of 100%, 93%, and 83%, compared with 27%, 9%, and 2% in placebo, respectively.[2]
The phase 2 signal extended to type 2 diabetes (36 weeks; ), where body weight decreased dose-dependently, reaching up to with the 12 mg escalation group, compared with with placebo and with dulaglutide 1.5 mg.[2, 7]
Phase 3 data (TRIUMPH-4; 68 weeks; ; obesity/overweight with knee osteoarthritis) pushed the mean effect size even further. The co-primary efficacy estimand reported mean body-weight change with retatrutide 12 mg (and with 9 mg), versus with placebo.[1, 8] In a responder table excerpt for the same trial, the percentage achieving at least weight loss was 68% with 12 mg and 50% with 9 mg.[9]
Oral GLP-1s
A second, equally important shift is practical: oral GLP-1 therapies are now producing weight-loss magnitudes that were previously associated mainly with injectables, potentially changing how many patients can start and sustain treatment.[5]
In OASIS-4 (oral semaglutide 25 mg; 64 weeks; ), mean weight loss was 13.6%, and 63% of participants achieved at least 10% weight loss.[5] For a nonpeptide oral GLP-1 receptor agonist, ATTAIN-1 evaluated orforglipron in 3,127 adults with obesity over 72 weeks and reported mean weight loss of 11.2% and at least 10% weight loss in 54.6%.[5] In an excerpted description of the same trial, the highest dose achieved 11.2% weight loss versus 2.1% with placebo, with 54.6% reaching at least 10% weight loss (versus 12.9% on placebo), and gastrointestinal side effects contributed to discontinuation rates of 5.3–10.3% (versus 2.7% with placebo).[10]
The oral programs also come with the same broad safety theme that defines the GLP-1 class: predominantly gastrointestinal tolerability issues, which become especially important as these therapies move into larger populations and longer durations.[5, 11]
Beyond weight
The most defensible “beyond weight” claims in the supplied evidence base relate to cardiometabolic outcomes and organ fat, but the strength of evidence varies by endpoint.
Liver
For fatty liver disease biology, the strongest quantitative signal here is for reductions in liver fat measured by MRI-PDFF (a noninvasive surrogate), not biopsy-confirmed MASH resolution or fibrosis improvement.[2] In a phase 2 obesity trial that included a liver steatosis substudy (48 weeks; randomized; baseline defined as liver fat by MRI-PDFF), mean relative change in liver fat at 24 weeks was (1 mg), (4 mg), (8 mg), and (12 mg), versus with placebo.[2] At the same timepoint, normal liver fat (defined as ) was achieved by 27% (1 mg), 52% (4 mg), 79% (8 mg), and 86% (12 mg), versus 0% with placebo.[2] At 48 weeks, liver fat remained reduced (e.g., with 12 mg vs placebo), and total liver fat was reported in 89% (8 mg) and 93% (12 mg).[2]
Cardiovascular and renal
Hard cardiovascular outcomes are not yet available for retatrutide in the supplied dataset, but they are available for tirzepatide in a large outcomes trial. In SURPASS-CVOT, a primary endpoint event occurred in 12.2% of patients receiving tirzepatide and 13.1% receiving dulaglutide, corresponding to a hazard ratio of 0.92 (95.3% CI 0.83 to 1.01), meeting noninferiority but not superiority (P=0.09 for superiority).[12] All-cause mortality occurred in 8.6% versus 10.2%, respectively (hazard ratio 0.84; 95% CI 0.75 to 0.94).[12]
On kidney-related measures, the dataset includes post-hoc or marker-based results. In participants with T2D, retatrutide 12 mg reduced urine albumin-creatinine ratio (UACR) by 37%, while eGFR was unchanged.[13] A post-hoc analysis of SURPASS-4 reported that tirzepatide reduced annual eGFR decline, reduced UACR, and reduced the occurrence of composite kidney endpoints versus insulin glargine.[14]
Neurodegeneration
The neuroprotection narrative is where 2024–2026 evidence most clearly forces a reality check. The most advanced randomized, placebo-controlled trials in Alzheimer’s disease did not show clinical slowing, and a large Parkinson’s trial failed to demonstrate disease modification, even as smaller programs have produced modest signals with tolerability trade-offs.[4, 15–17]
In early Alzheimer’s disease, the phase 3 EVOKE and EVOKE+ trials did not confirm superiority of semaglutide versus placebo on the primary endpoint (change in CDR-SB).[3] In one reported analysis, oral semaglutide did not slow decline versus placebo: the treatment–placebo difference on CDR-SB to week 104 was (95% CI to ; ).[15] Secondary clinical measures similarly showed no significant differences between semaglutide and placebo groups.[18, 19] Although biomarker substudies reported approximately 10% reductions in several CSF markers (including multiple tau-related measures), this did not translate into delayed clinical progression.[19, 20]
In Parkinson’s disease, the phase 3 exenatide trial reported no meaningful difference in motor progression: at 96 weeks, OFF-medication MDS-UPDRS part III worsened by 5.7 points with exenatide versus 4.5 points with placebo (adjusted coefficient 0.92; 95% CI -1.56 to 3.39; ), and the investigators concluded there was no evidence supporting exenatide as a disease-modifying treatment.[4] By contrast, the phase 2 LIXIPARK trial of lixisenatide showed a modest 12-month signal on the primary motor endpoint (between-group difference 3.08 points; 95% CI 0.86 to 5.30; ), but gastrointestinal side effects were substantially more common, including nausea in 46% and vomiting in 13%.[16, 17]
Bottom line
The best-supported “next generation” story is not that incretins have become proven neuroprotective drugs, but that metabolic pharmacology is now delivering unusually large weight loss (retatrutide up to in TRIUMPH-4), credible oral alternatives (e.g., oral semaglutide 13.6% in OASIS-4; orforglipron 11.2% in ATTAIN-1), and clinically meaningful cardiometabolic outcomes in at least one large CVOT (SURPASS-CVOT all-cause mortality hazard ratio 0.84).[1, 5, 12] At the same time, class-consistent gastrointestinal tolerability remains a practical limiter across oral and injectable agents, and the most rigorous late-stage trials to date have not confirmed disease-modifying benefit in Alzheimer’s disease or Parkinson’s disease.[3, 4, 10]
" } }