Semaglutide
The cleanest benchmark for modern GLP-1 evidence. Approved, widely used, and supported by deep obesity, diabetes, cardiovascular, kidney, and organ-outcome data.
GLP-1 research
GLP-1 Research & Results
Here is the short version: semaglutide and tirzepatide are approved and deeply studied. Retatrutide looks exceptionally strong on weight loss, but it is still investigational and has more open questions around long-term safety and outcomes.
What seems most solid right now
- Semaglutide and tirzepatide are approved, deeply studied incretin therapies; retatrutide is still investigational as of April 21, 2026.
- The strongest established outcome domains are weight loss, glycemic improvement, lower blood pressure and triglycerides, and selected cardiovascular or kidney benefits depending on the agent.
- Retatrutide currently has the most dramatic published efficacy signal on weight loss, but it has the shallowest approval status and more unresolved long-term safety and outcomes questions.
- Body-composition claims need restraint: fat mass falls more than lean mass, but lean mass still falls, so current evidence does not justify calling these drugs proven muscle-sparing.
- Addiction and reward-behavior findings are interesting and early, not settled. The strongest current human proof-of-concept is in alcohol use disorder, not in broad social-media-style 'addiction cure' claims.
Tirzepatide
The current approved leader in average weight-loss efficacy, with strong additional evidence in obstructive sleep apnea and obesity-related HFpEF.
Retatrutide
The most advanced triple agonist. The results look unusually strong, but it is still investigational and has more unanswered long-term questions.
What the landmark obesity trials actually show
| Trial | Agent | Duration | Mean weight loss | Interpretation |
|---|---|---|---|---|
| STEP 1 | Semaglutide 2.4 mg | 68 weeks | 14.9% | Published obesity landmark trial and the cleanest semaglutide benchmark for public comparison. |
| SURMOUNT-1 | Tirzepatide 5-15 mg | 72 weeks | 15.0% to 20.9% | Published obesity landmark trial showing stronger average efficacy than semaglutide. |
| Retatrutide phase 2 obesity | Retatrutide 4-12 mg | 48 weeks | 17.1% to 24.2% | Published phase 2 signal only; directionally impressive, but not an apples-to-apples replacement for longer phase 3 comparisons. |
| TRIUMPH-4 topline | Retatrutide 12 mg | 68 weeks | 28.7% | Company-reported obesity plus knee-osteoarthritis topline, not yet a fully peer-reviewed paper as of April 21, 2026. |
Cross-trial comparisons are directional, not head-to-head proof. Trial duration, estimands, titration, and publication status all affect how confidently numbers can be compared.
Glycemia and diabetes
Established for semaglutide and tirzepatide. Retatrutide phase 2 showed HbA1c reductions up to about 2.0%, and Lilly's March 2026 topline reported 1.7% to 2.0% at 40 weeks.
Cardiovascular and kidney outcomes
Semaglutide has the deepest hard-outcome base, including SELECT and FLOW. Tirzepatide improves obesity-related organ outcomes, but its CVOT result versus dulaglutide was noninferiority, not superiority.
OSA, HFpEF, and liver disease
Tirzepatide has strong published OSA and HFpEF evidence. Semaglutide has phase 3 MASH histology data. Retatrutide's 2024 Nature Medicine MASLD substudy showed very large MRI-PDFF liver-fat reductions, and the AASLD abstract reinforced steatosis-resolution figures above 85% in the 8 mg and 12 mg groups at 48 weeks, but those retatrutide liver signals still sit below biopsy-proven outcome evidence.
Reward behavior and addiction
Alcohol-use findings are promising enough to mention, but they are still early and do not make GLP-1s established addiction treatments.
What is robust versus what is still early
- Appetite reduction and reduced energy intake are core, evidence-backed effects across the class.
- The language users recognize as 'food noise' can be useful, but the formal evidence base is still appetite, satiety, and calorie-intake oriented.
- Retatrutide may add an energy-expenditure component through glucagon signaling, but the size of that human effect remains uncertain.
- The best current human addiction signal is the small randomized semaglutide trial in alcohol use disorder, which is promising but not enough to market GLP-1s as broad addiction treatments.
The muscle question needs restraint
- Fat mass falls more than lean mass, but absolute lean mass also falls with semaglutide, tirzepatide, and retatrutide.
- Improvement in lean-to-fat ratio is not the same thing as proven preservation of strength, performance, or long-term function.
- These therapies can improve body composition meaningfully, but protein intake, resistance training, and functional outcomes still matter.
- No current agent has earned a strong 'muscle-preserving' claim.
Semaglutide, tirzepatide, and retatrutide
| Agent | Mechanism | Approved status | Strongest evidence areas | Typical weight-loss magnitude | Major caveats | Tolerability issues | Confidence |
|---|---|---|---|---|---|---|---|
| Semaglutide | GLP-1 receptor agonist | Approved for type 2 diabetes and chronic weight management; Wegovy also has U.S. cardiovascular-risk-reduction labeling in overweight or obesity with established CVD. | Weight loss, glycemic control, cardiovascular risk reduction, CKD benefit in type 2 diabetes, HFpEF symptom benefit, MASH histology improvement. | 14.9% mean loss at 68 weeks in STEP 1 | Less average weight loss than tirzepatide or retatrutide signals; substantial regain is common after withdrawal. | GI effects, gallbladder disease, pancreatitis, dehydration/AKI warnings, retinopathy warning in diabetes, boxed thyroid warning. | High |
| Tirzepatide | Dual GIP / GLP-1 agonist | Approved for type 2 diabetes and chronic weight management; U.S. approval also includes moderate-to-severe OSA in adults with obesity. | Weight loss, glycemic control, OSA, HFpEF with obesity, broad cardiometabolic improvement. | 15.0% to 20.9% mean loss at 72 weeks in SURMOUNT-1 | Stronger efficacy does not equal proven superiority on hard cardiovascular outcomes versus established GLP-1 comparators. | GI effects and the same major labeled precautions as above, including boxed thyroid warning. | High for efficacy; moderate-to-high for hard outcomes outside weight and glycemia |
| Retatrutide | Triple GIP / GLP-1 / glucagon agonist | Investigational only; no FDA or EMA approval as of April 21, 2026. | Largest current efficacy signal for weight loss, strong glycemic signal in diabetes, and early liver-fat and osteoarthritis-benefit signals. | Up to 24.2% at 48 weeks in published phase 2 obesity data; 28.7% at 68 weeks in phase 3 obesity plus knee-osteoarthritis topline. | Phase 3 evidence is still incomplete, cardiovascular outcome evidence is unfinished, and cross-trial superiority claims remain unstable. | GI effects, dose-related discontinuation, transient heart-rate increase, and dysesthesia signal in phase 3 topline. | Moderate overall; high for short-term efficacy signal, lower for long-term safety and outcomes |
What can be stated confidently
- Gastrointestinal adverse effects dominate the class: nausea, vomiting, diarrhea, constipation, and abdominal symptoms.
- Approved semaglutide and tirzepatide labels also warn about pancreatitis, gallbladder disease, kidney injury from dehydration, hypersensitivity, and diabetic-retinopathy complications in diabetes settings.
- Both U.S. labels carry a boxed warning based on rodent thyroid C-cell tumors and contraindicate use in patients with personal or family history of medullary thyroid carcinoma or MEN2.
- Retatrutide phase 2 and phase 3 toplines add dose-related discontinuation, transient heart-rate increases, and a dysesthesia signal to the discussion.
What remains unresolved
- Long-term cardiovascular safety and benefit of retatrutide in broad populations.
- True head-to-head efficacy and tolerability versus tirzepatide.
- Magnitude of regain and real-world persistence after discontinuation.
- Rare adverse-event patterns once triple agonism scales beyond trial populations.
- Whether large liver-fat reductions translate into hard hepatic clinical outcomes.
Core studies and regulatory documents
- Wilding et al., NEJM 2021 STEP 1 semaglutide obesity trial.
- Jastreboff et al., NEJM 2022 SURMOUNT-1 tirzepatide obesity trial.
- Jastreboff et al., NEJM 2023 Phase 2 retatrutide obesity trial.
- Rosenstock et al., The Lancet 2023 Phase 2 retatrutide type 2 diabetes trial.
- Lincoff et al., NEJM 2023 SELECT semaglutide cardiovascular outcomes in obesity without diabetes.
- Perkovic et al., NEJM 2024 FLOW semaglutide kidney outcomes in type 2 diabetes plus CKD.
- Kosiborod et al., NEJM 2023 and follow-up 2024 STEP-HFpEF semaglutide program.
- Sanyal et al., NEJM 2025 Phase 3 semaglutide MASH histology trial.
- Malhotra et al., NEJM 2024 SURMOUNT-OSA tirzepatide trials.
- Packer et al., NEJM 2025 SUMMIT tirzepatide in HFpEF with obesity.
- Nicholls et al., NEJM 2025 SURPASS-CVOT tirzepatide versus dulaglutide.
- Sanyal et al., Nature Medicine 2024 Retatrutide MASLD liver-fat substudy using MRI-PDFF.
- Hendershot et al., JAMA Psychiatry 2025 Semaglutide alcohol use disorder trial.
- Badve et al., Lancet Diabetes & Endocrinology 2025 GLP-1 randomized-trial meta-analysis for kidney and cardiovascular outcomes.
Keep weaker evidence in the right bucket
- Current official prescribing information and regulatory documents for Wegovy, Ozempic, Zepbound, and Mounjaro.
- 2026 Lancet review on GLP-1 receptor agonists and next-generation incretin medicines.
- FDA and EMA safety communications on suicidality review changes and current label posture.
- FDA and WHO alerts on counterfeit or unapproved semaglutide, tirzepatide, and retatrutide products.
Do not treat company toplines, animal work, observational signals, or conference abstracts like published trials or current labeling.
- Company toplines can be useful, but they are not the same as fully peer-reviewed evidence. That applies to the TRIUMPH-4 and TRANSCEND-T2D-1 retatrutide results as of April 21, 2026.
- Observational evidence can support a signal, but it should not carry headline causal claims on its own.
- Animal and preclinical data can explain plausibility, but they are not human proof.
- Conference and abstract-level findings can add direction, but they are still a weaker tier. That includes the AASLD retatrutide MASLD abstract reporting steatosis resolution above 85% in the highest-dose groups.