Retatrutide

1. Studied for

   Receptor pharmacology and discovery

   The discovery paper describes a single peptide that switches on three different hormone receptors at the same time: the glucagon receptor, the GIP receptor, and the GLP-1 receptor. In the authors' own words, in cultured cells it "shows balanced GCGR and GLP-1R activity but more GIPR activity." In obese mice the compound reduced body weight and improved blood-sugar control, and the authors attributed that weight loss to two things working together: the glucagon arm raising the rate at which the animals burned energy, added to the GIP and GLP-1 arms cutting how much they ate. In an accompanying first-in-human study, the compound stayed in the body long enough that a reduction in body weight was still measurable six weeks after a single administration. in cultured cells, plus obese mice, plus a first-in-human phase 1 single-ascending study Preclinical strong More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material. Open for the full finding and the paper it came from.

   In the literature

   Receptor pharmacology and discovery

   The discovery paper describes a single peptide that switches on three different hormone receptors at the same time: the glucagon receptor, the GIP receptor, and the GLP-1 receptor. In the authors' own words, in cultured cells it "shows balanced GCGR and GLP-1R activity but more GIPR activity." In obese mice the compound reduced body weight and improved blood-sugar control, and the authors attributed that weight loss to two things working together: the glucagon arm raising the rate at which the animals burned energy, added to the GIP and GLP-1 arms cutting how much they ate. In an accompanying first-in-human study, the compound stayed in the body long enough that a reduction in body weight was still measurable six weeks after a single administration.^{References 1}

   Modelin cultured cells, plus obese mice, plus a first-in-human phase 1 single-ascending study

   Confidence: strong

   More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material.

   Source

   Coskun T, Urva S, Roell WC, et al. LY3437943, a novel triple glucagon, GIP, and GLP-1 receptor agonist for glycemic control and weight loss: from discovery to clinical proof of concept. Cell Metab. 2022;34(9):1234-1247.e9. doi:10.1016/j.cmet.2022.07.013. PMID 35985340

   PubMed 35985340

   How this citation was checked

   Re-resolved by the research pass recorded in research/clean/retatrutide.json. PubMed esummary on PMID 35985340 returned first author Coskun T, Cell Metab, 2022 Sep 6, volume 34, issue 9, pages 1234-1247.e9, DOI 10.1016/j.cmet.2022.07.013, all four matching what is written. Full abstract retrieved by efetch and read: the quoted in-cell phrase is verbatim; the energy-expenditure sentence is verbatim ("Body weight loss was augmented by the addition of GCGR-mediated increases in energy expenditure to GIPR- and GLP-1R-driven calorie intake reduction"), so the paraphrase above is faithful and an earlier draft's opposite claim stays deleted; the day-43 statement is verbatim, restated as six weeks for readability. The same abstract's weekly-interval statement is deliberately not carried, because this page states no interval of any kind. An earlier draft's DOI 10.1016/j.cmet.2022.08.004 resolves to a different paper entirely and remains replaced.

2. Studied for

   Body weight over 48 weeks (phase 2)

   In a 48-week randomized trial in 338 adults selected for excess body weight, average body weight fell in proportion to the amount given. Across the four ascending groups, average body weight at 48 weeks was down 8.7%, 17.1%, 22.8% and 24.2%, against 2.1% in the placebo group. At the trial's main 24-week checkpoint the corresponding figures were 7.2%, 12.9%, 17.3% and 17.5%, against 1.6% on placebo. The most common side effects were digestive, related to the amount given, and mostly mild to moderate. The investigators also reported increases in heart rate that tracked with the amount given, which peaked at 24 weeks and declined after that. human, phase 2 randomized double-blind placebo-controlled trial, 338 participants, 48 weeks Preclinical strong More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material. Open for the full finding and the paper it came from.

   In the literature

   Body weight over 48 weeks (phase 2)

   In a 48-week randomized trial in 338 adults selected for excess body weight, average body weight fell in proportion to the amount given. Across the four ascending groups, average body weight at 48 weeks was down 8.7%, 17.1%, 22.8% and 24.2%, against 2.1% in the placebo group. At the trial's main 24-week checkpoint the corresponding figures were 7.2%, 12.9%, 17.3% and 17.5%, against 1.6% on placebo. The most common side effects were digestive, related to the amount given, and mostly mild to moderate. The investigators also reported increases in heart rate that tracked with the amount given, which peaked at 24 weeks and declined after that.^{References 2}

   Modelhuman, phase 2 randomized double-blind placebo-controlled trial, 338 participants, 48 weeks

   Confidence: strong

   More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material.

   Source

   Jastreboff AM, Kaplan LM, Frias JP, et al. Triple-hormone-receptor agonist retatrutide for obesity: a phase 2 trial. N Engl J Med. 2023;389(6):514-526. doi:10.1056/NEJMoa2301972. PMID 37366315

   PubMed 37366315

   How this citation was checked

   Re-resolved by the research pass. PubMed esummary on PMID 37366315 returned first author Jastreboff AM, N Engl J Med, 2023 Aug 10, volume 389, issue 6, pages 514-526, DOI 10.1056/NEJMoa2301972, all matching. Full abstract retrieved by efetch; n=338, all four 24-week values and their placebo comparator, all four 48-week values and their placebo comparator, the adverse-event characterisation and the heart-rate sentence were each read verbatim. An earlier draft's unsourced superlative ("the largest weight reduction described for an investigational anti-obesity agent") appears nowhere in the source and stays removed. Milligram amount labels are not reproduced; the groups are identified only by rank.

3. Studied for

   Blood sugar against an active comparator (phase 2)

   A randomized trial in 281 adults with an entry HbA1c of 7.0 to 10.5% and a BMI of 25 to 50 compared this compound against both a placebo and an already-marketed single-receptor compound of the same family. At 24 weeks the average fall in HbA1c ranged from 0.43% in the lowest group to 2.02% in the highest, against 0.01% on placebo and 1.41% with the comparator. At 36 weeks body weight fell in proportion to the amount given, from 3.19% in the lowest group to 16.94% in the highest, against 3.00% on placebo and 2.02% with the comparator. Mild-to-moderate digestive side effects were reported by 35% of participants receiving the compound, against 13% on placebo. human, phase 2 randomized double-blind placebo- and active-controlled trial, 281 participants Preclinical strong More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material. Open for the full finding and the paper it came from.

   In the literature

   Blood sugar against an active comparator (phase 2)

   A randomized trial in 281 adults with an entry HbA1c of 7.0 to 10.5% and a BMI of 25 to 50 compared this compound against both a placebo and an already-marketed single-receptor compound of the same family. At 24 weeks the average fall in HbA1c ranged from 0.43% in the lowest group to 2.02% in the highest, against 0.01% on placebo and 1.41% with the comparator. At 36 weeks body weight fell in proportion to the amount given, from 3.19% in the lowest group to 16.94% in the highest, against 3.00% on placebo and 2.02% with the comparator. Mild-to-moderate digestive side effects were reported by 35% of participants receiving the compound, against 13% on placebo.^{References 3}

   Modelhuman, phase 2 randomized double-blind placebo- and active-controlled trial, 281 participants

   Confidence: strong

   More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material.

   Source

   Rosenstock J, Frias J, Jastreboff AM, et al. Retatrutide, a GIP, GLP-1 and glucagon receptor agonist, for people with type 2 diabetes: a randomised, double-blind, placebo and active-controlled, parallel-group, phase 2 trial conducted in the USA. Lancet. 2023;402(10401):529-544. doi:10.1016/S0140-6736(23)01053-X. PMID 37385280

   PubMed 37385280

   How this citation was checked

   Re-resolved by the research pass. PubMed esummary on PMID 37385280 returned first author Rosenstock J, Lancet, 2023 Aug 12, volume 402, issue 10401, pages 529-544, DOI 10.1016/S0140-6736(23)01053-X, all matching. Full abstract retrieved by efetch; the entry criteria, n=281 randomised, the HbA1c range and both comparator values, the 36-week bodyweight range and both comparator values, and the 35% versus 13% digestive figures were each read verbatim. Two earlier errors stay fixed: a claimed mean baseline HbA1c of 8.3% is not in the source, which gives an entry range instead, and the two maxima fall at different timepoints, HbA1c at 24 weeks and bodyweight at 36 weeks, which an earlier draft had collapsed into one. The comparator is named in the source and is deliberately not named in this page's prose, per the brand study's no-drug-names law. The abstract's severe-hypoglycaemia and no-deaths statements are also deliberately not carried: safety reassurance is on this compound's cannot-support list. Milligram amount labels are not reproduced.

4. Studied for

   Blood sugar and body weight in a phase 3 confirmatory trial (TRANSCEND-T2D-1)

   TRANSCEND-T2D-1 was a 40-week phase 3 randomized double-blind placebo-controlled trial at 48 sites in the USA, Mexico and India, in 537 adults whose blood sugar was not adequately controlled by diet and exercise alone. Using the trial's regimen-based analysis, average HbA1c fell 1.69%, 1.86% and 1.94% across the three ascending groups, against 0.81% on placebo (every comparison p<0.0001). Average body weight fell 11.5%, 13.9% and 15.3%, against 2.6% on placebo. Discontinuations due to side effects were 2 to 5% in the groups receiving the compound and 0% on placebo. Two deaths occurred during the study, both in the lowest group, and the investigators reported both as unrelated to the study drug. human, phase 3 randomized double-blind placebo-controlled trial, 537 participants randomized, 40 weeks Preclinical strong More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material. Open for the full finding and the paper it came from.

   In the literature

   Blood sugar and body weight in a phase 3 confirmatory trial (TRANSCEND-T2D-1)

   TRANSCEND-T2D-1 was a 40-week phase 3 randomized double-blind placebo-controlled trial at 48 sites in the USA, Mexico and India, in 537 adults whose blood sugar was not adequately controlled by diet and exercise alone. Using the trial's regimen-based analysis, average HbA1c fell 1.69%, 1.86% and 1.94% across the three ascending groups, against 0.81% on placebo (every comparison p<0.0001). Average body weight fell 11.5%, 13.9% and 15.3%, against 2.6% on placebo. Discontinuations due to side effects were 2 to 5% in the groups receiving the compound and 0% on placebo. Two deaths occurred during the study, both in the lowest group, and the investigators reported both as unrelated to the study drug.^{References 4}

   Modelhuman, phase 3 randomized double-blind placebo-controlled trial, 537 participants randomized, 40 weeks

   Confidence: strong

   More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material.

   Source

   Bajaj HS, Welch M, Shah P, et al. Efficacy and safety of retatrutide, a GIP, GLP-1, and glucagon receptor agonist, in people with type 2 diabetes and inadequate glycaemic control with diet and exercise (TRANSCEND-T2D-1): a double-blind, randomised, phase 3 trial. Lancet. 2026;407(10546):2402-2413. doi:10.1016/S0140-6736(26)00967-0. PMID 42250575

   PubMed 42250575

   How this citation was checked

   Re-resolved by the research pass. PubMed esummary on PMID 42250575 returned first author Bajaj HS, Lancet, 2026 Jun 13, volume 407, issue 10546, pages 2402-2413, DOI 10.1016/S0140-6736(26)00967-0, all matching. Full abstract retrieved by efetch; the 40-week duration, the 48 sites in those three countries, 537 randomised, all three HbA1c values and the placebo value, all three bodyweight values and the placebo value, the 2 to 5% versus 0% discontinuations, and the two deaths in the lowest group reported as unrelated to study drug were each read verbatim, as was the analysis the paper names as its regimen-based estimand, which is the more conservative of the two it reports. An earlier draft's "approximately 16.8%" bodyweight figure appears nowhere in this paper, being a sponsor efficacy-estimand number, and stays removed rather than reattributed, so no analysis switch hides inside a peer-reviewed citation. The abstract's severe-hypoglycaemia statement is deliberately not carried, per the cannot-support list. Milligram amount labels are not reproduced.

5. Studied for

   Liver fat (phase 2a substudy)

   In a randomized substudy of 98 participants who entered with at least 10% of the liver made up of fat, the amount of liver fat fell sharply and in proportion to the amount given. At 24 weeks, average liver fat was down 42.9%, 57.0%, 81.4% and 82.4% across the four ascending groups, against a 0.3% increase on placebo (every comparison against placebo at p<0.001). By 24 weeks, a normal liver-fat level, under 5%, had been reached by 27%, 52%, 79% and 86% of participants in those groups, against nobody on placebo. The investigators reported that the liver-fat reductions tracked significantly with changes in body weight, abdominal fat, and measures tied to insulin sensitivity and fat handling. human, phase 2a randomized double-blind placebo-controlled substudy, 98 participants Preclinical strong More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material. Open for the full finding and the paper it came from.

   In the literature

   Liver fat (phase 2a substudy)

   In a randomized substudy of 98 participants who entered with at least 10% of the liver made up of fat, the amount of liver fat fell sharply and in proportion to the amount given. At 24 weeks, average liver fat was down 42.9%, 57.0%, 81.4% and 82.4% across the four ascending groups, against a 0.3% increase on placebo (every comparison against placebo at p<0.001). By 24 weeks, a normal liver-fat level, under 5%, had been reached by 27%, 52%, 79% and 86% of participants in those groups, against nobody on placebo. The investigators reported that the liver-fat reductions tracked significantly with changes in body weight, abdominal fat, and measures tied to insulin sensitivity and fat handling.^{References 5}

   Modelhuman, phase 2a randomized double-blind placebo-controlled substudy, 98 participants

   Confidence: strong

   More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material.

   Source

   Sanyal AJ, Kaplan LM, Frias JP, et al. Triple hormone receptor agonist retatrutide for metabolic dysfunction-associated steatotic liver disease: a randomized phase 2a trial. Nat Med. 2024;30(7):2037-2048. doi:10.1038/s41591-024-03018-2. PMID 38858523

   PubMed 38858523

   How this citation was checked

   Re-resolved by the research pass. PubMed esummary on PMID 38858523 returned first author Sanyal AJ, Nat Med, 2024 Jul, volume 30, issue 7, pages 2037-2048, DOI 10.1038/s41591-024-03018-2, all matching. Full abstract retrieved by efetch; n=98 and all nine percentages above, plus the relationship sentence, were read verbatim. An earlier draft's sentence "No hepatotoxicity signal was observed through 48 weeks" appears nowhere in the source and stays deleted: it is an absence-of-evidence safety claim, and safety reassurance has no place on a research-use-only page. Abdominal fat appears here only as a correlate that moved with liver fat, which is the only form in which the verified record contains it; it is not an endpoint and is not claimed as one anywhere on this page. Milligram amount labels are not reproduced.

6. Studied for

   Where the lost weight actually comes from

   A prespecified body-scan substudy of the phase 2 trial above, with 189 participants enrolled to the substudy, used DXA body scanning to separate fat from lean tissue. Total fat mass at week 36 was down 4.9% in the lowest group, 15.2% and 26.1% in the middle groups, and 23.2% in the highest, against 4.5% on placebo and 2.6% with the comparator. The authors' own conclusion is the important part: the share of the weight lost that came from lean tissue was similar to what other weight-loss compounds produce. In plain terms, lean tissue came off roughly in proportion to total weight rather than being selectively protected. This is the correct primary reference for any body-composition question about this compound, and it does not support claims of selective muscle preservation. human, DXA body-composition substudy of a phase 2 randomized double-blind placebo- and active-controlled trial, 189 participants enrolled to the substudy Preclinical strong More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material. Open for the full finding and the paper it came from.

   In the literature

   Where the lost weight actually comes from

   A prespecified body-scan substudy of the phase 2 trial above, with 189 participants enrolled to the substudy, used DXA body scanning to separate fat from lean tissue. Total fat mass at week 36 was down 4.9% in the lowest group, 15.2% and 26.1% in the middle groups, and 23.2% in the highest, against 4.5% on placebo and 2.6% with the comparator. The authors' own conclusion is the important part: the share of the weight lost that came from lean tissue was similar to what other weight-loss compounds produce. In plain terms, lean tissue came off roughly in proportion to total weight rather than being selectively protected. This is the correct primary reference for any body-composition question about this compound, and it does not support claims of selective muscle preservation.^{References 6}

   Modelhuman, DXA body-composition substudy of a phase 2 randomized double-blind placebo- and active-controlled trial, 189 participants enrolled to the substudy

   Confidence: strong

   More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material.

   Source

   Coskun T, Wu Q, Schloot NC, Haupt A, Milicevic Z, Khouli C, Harris C. Effects of retatrutide on body composition in people with type 2 diabetes: a substudy of a phase 2, double-blind, parallel-group, placebo-controlled, randomised trial. Lancet Diabetes Endocrinol. 2025;13(8):674-684. doi:10.1016/S2213-8587(25)00092-0. PMID 40609566

   PubMed 40609566

   How this citation was checked

   Re-resolved by the research pass. PubMed esummary on PMID 40609566 returned first author Coskun T, Lancet Diabetes Endocrinol, 2025 Aug, volume 13, issue 8, pages 674-684, DOI 10.1016/S2213-8587(25)00092-0, all matching, and all seven author names in the citation match the source byline. Full abstract retrieved by efetch; the 189-participant substudy enrolment, all five fat-mass percentages and the interpretation sentence, "The proportion of lean mass loss to weight loss was similar to other obesity treatments", were read verbatim. Noted for honesty: the same interpretation paragraph frames that as reassurance that lean loss is not disproportionate; this dossier reports the neutral factual reading, that lean tissue is not selectively spared, which is what the sentence supports and what the market claim it contradicts requires. Milligram amount labels are not reproduced.

7. Studied for

   Blood pressure and blood fats, pooled across trials

   A systematic review and meta-analysis of randomized controlled trials pooled the available human data and reported significant reductions in systolic blood pressure (6.79 mmHg lower, 95% CI 8.36 to 5.23 lower, p<0.0001) and diastolic blood pressure (2.46 mmHg lower, 95% CI 3.25 to 1.67 lower, p<0.0001). It also reported significant reductions in total cholesterol (21.88 mg/dL lower), LDL cholesterol (13.10 mg/dL lower) and triglycerides (40.90 mg/dL lower), and no meaningful change in HDL cholesterol (0.01 mg/dL, p=0.98). Worth knowing about the consistency of these pooled numbers: the blood-pressure results were highly consistent across the included trials, while the total-cholesterol result varied a great deal between them (I-squared 84%), so that figure is the softest of the set. human, systematic review and meta-analysis of randomized controlled trials Preclinical strong More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material. Open for the full finding and the paper it came from.

   In the literature

   Blood pressure and blood fats, pooled across trials

   A systematic review and meta-analysis of randomized controlled trials pooled the available human data and reported significant reductions in systolic blood pressure (6.79 mmHg lower, 95% CI 8.36 to 5.23 lower, p<0.0001) and diastolic blood pressure (2.46 mmHg lower, 95% CI 3.25 to 1.67 lower, p<0.0001). It also reported significant reductions in total cholesterol (21.88 mg/dL lower), LDL cholesterol (13.10 mg/dL lower) and triglycerides (40.90 mg/dL lower), and no meaningful change in HDL cholesterol (0.01 mg/dL, p=0.98). Worth knowing about the consistency of these pooled numbers: the blood-pressure results were highly consistent across the included trials, while the total-cholesterol result varied a great deal between them (I-squared 84%), so that figure is the softest of the set.^{References 7}

   Modelhuman, systematic review and meta-analysis of randomized controlled trials

   Confidence: strong

   More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material.

   Source

   Simental-Mendia LE, Barragan-Zuniga LJ, Reyes-Avitia V. Effect of retatrutide, a novel triple receptor agonist, on blood pressure and lipid levels: a systematic review and meta-analysis of randomized controlled trials. High Blood Press Cardiovasc Prev. 2026 Jun 29. doi:10.1007/s40292-026-00812-6. PMID 42371360. Online ahead of print

   PubMed 42371360

   How this citation was checked

   Re-resolved by the research pass. PubMed esummary on PMID 42371360 returned first author Simental-Mendia LE, High Blood Press Cardiovasc Prev, 2026 Jun 29, DOI 10.1007/s40292-026-00812-6, with NO volume, issue or page numbers assigned, which is why the citation is labelled online ahead of print rather than given fabricated volume and pages. Full abstract retrieved by efetch; every weighted mean difference, confidence interval, p value and I-squared value above was read verbatim, including the 20%, 0%, 84%, 47% and 53% heterogeneity figures that support the consistency sentence. The HDL null result is carried deliberately: it is what stops this becoming a whole-panel lipid claim, which the verified record does not support. Note on independence: this analysis pools the developer's own trials rather than adding independent ones.

8. Studied for

   How long it lasts in the body, and first blood-sugar effects in people

   In a 12-week early human study in 72 adults whose blood sugar was raised, the compound cleared from the body slowly, with a half-life of about 6 days. By week 12, average daily blood glucose fell significantly in the three highest groups compared with placebo (differences of -2.8, -3.1 and -2.9 mmol/L), long-term blood sugar as measured by HbA1c fell significantly in the same three groups (-1.4%, -1.6% and -1.2% versus placebo), and body weight fell in proportion to the amount given, reaching -8.96 kg versus placebo in the highest group. Digestive side effects were the most frequently reported adverse events. human, phase 1b randomized double-blind placebo-controlled multiple-ascending trial, 72 participants, 12 weeks Preclinical strong More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material. Open for the full finding and the paper it came from.

   In the literature

   How long it lasts in the body, and first blood-sugar effects in people

   In a 12-week early human study in 72 adults whose blood sugar was raised, the compound cleared from the body slowly, with a half-life of about 6 days. By week 12, average daily blood glucose fell significantly in the three highest groups compared with placebo (differences of -2.8, -3.1 and -2.9 mmol/L), long-term blood sugar as measured by HbA1c fell significantly in the same three groups (-1.4%, -1.6% and -1.2% versus placebo), and body weight fell in proportion to the amount given, reaching -8.96 kg versus placebo in the highest group. Digestive side effects were the most frequently reported adverse events.^{References 8}

   Modelhuman, phase 1b randomized double-blind placebo-controlled multiple-ascending trial, 72 participants, 12 weeks

   Confidence: strong

   More than one randomized controlled trial in people reporting the same direction of result on a prespecified measure, or such a trial together with a meta-analysis of trials. 'Strong' describes the published record of an investigational compound and nothing more: it is approved nowhere, every trial behind this rating was run by its own developer, and none of it authorizes any use of this material.

   Source

   Urva S, Coskun T, Loh MT, et al. LY3437943, a novel triple GIP, GLP-1, and glucagon receptor agonist in people with type 2 diabetes: a phase 1b, multicentre, double-blind, placebo-controlled, randomised, multiple-ascending dose trial. Lancet. 2022;400(10366):1869-1881. doi:10.1016/S0140-6736(22)02033-5. PMID 36354040

   PubMed 36354040

   How this citation was checked

   Re-resolved by the research pass. PubMed esummary on PMID 36354040 returned first author Urva S, Lancet, 2022 Nov 26, volume 400, issue 10366, pages 1869-1881, DOI 10.1016/S0140-6736(22)02033-5, all matching. Full abstract retrieved by efetch; the 6-day half-life, n=72, the three glucose differences, the three HbA1c differences, the -8.96 kg figure and the gastrointestinal-events statement were each read verbatim. The abstract's weekly-interval inference is deliberately not carried, because this page states no interval. Milligram amount labels attached to those groups in the source are not reproduced; the groups are identified only by rank. NOTE FOR ANYONE RUNNING THE BANNED-WORD CHECK: this citation's published title contains the word dose. Titles are reproduced verbatim, because altering a published title falsifies the record, and this one is only reachable inside a collapsed fold.

9. Studied for

   Rodent studies of body weight and liver fat

   In obese mice with fat accumulated in the liver, researchers reported that the compound reduced body weight by 31% compared with animals that received none (p<0.0001), and that the loss came from BOTH fat and lean tissue. Food and water intake dropped in the first days, while the rate at which the animals burned energy was NOT significantly changed. It markedly reduced a standard index of insulin resistance and reduced liver fat scoring, fatty acids, triglycerides and cholesterol content in the liver. In obese hamsters it shifted what the animals chose to eat, increasing plain chow and decreasing both the high-fat high-cholesterol food and the sugar water. Weight loss there again came from both fat and lean tissue, although lean tissue was no longer different from controls after five weeks, with no change in bone mineral density. Liver triglyceride content fell by about half (p<0.01), but the microscopic liver damage scoring did not improve. RODENT rodent (diet-induced obese mice and diet-induced obese hamsters) Preclinical moderate A well-described result with several independent outcome measures, but from a single laboratory, a single animal model or a single study design. Open for the full finding and the paper it came from.

   In the literature

   Rodent studies of body weight and liver fat

   In obese mice with fat accumulated in the liver, researchers reported that the compound reduced body weight by 31% compared with animals that received none (p<0.0001), and that the loss came from BOTH fat and lean tissue. Food and water intake dropped in the first days, while the rate at which the animals burned energy was NOT significantly changed. It markedly reduced a standard index of insulin resistance and reduced liver fat scoring, fatty acids, triglycerides and cholesterol content in the liver. In obese hamsters it shifted what the animals chose to eat, increasing plain chow and decreasing both the high-fat high-cholesterol food and the sugar water. Weight loss there again came from both fat and lean tissue, although lean tissue was no longer different from controls after five weeks, with no change in bone mineral density. Liver triglyceride content fell by about half (p<0.01), but the microscopic liver damage scoring did not improve.^{References 9}

   Modelrodent (diet-induced obese mice and diet-induced obese hamsters)

   Confidence: moderate

   A well-described result with several independent outcome measures, but from a single laboratory, a single animal model or a single study design.

   Source

   Briand F, Le Cudennec C, Grasset E, Breyner N, Bigot C, Dillard P, Sulpice T. Retatrutide shows multiple metabolic benefits in diet-induced obese MASH mouse and hamster models. Obesity (Silver Spring). 2026;34(Suppl 1):43-53. doi:10.1002/oby.70155. PMID 41741376

   PubMed 41741376

   How this citation was checked

   Re-resolved by the research pass. PubMed esummary on PMID 41741376 returned first author Briand F, Obesity (Silver Spring), 2026 Jun, volume 34 Suppl 1, pages 43-53, DOI 10.1002/oby.70155, all matching, and all seven author names match the source byline. Full structured abstract retrieved by efetch; the 31% reduction and its p value, "both fat and lean mass", "energy expenditure was not altered significantly", the food-preference shift, "the lean mass was not different after 5 weeks", "no change in mineral bone density", the 50% hepatic triglyceride reduction with p<0.01, and "did not reduce histopathological scoring" were each read verbatim. An earlier draft's phrasing "adipose AND muscle tissue" stays corrected to fat and lean mass, which is what the paper says and a different compartment; that wording is load-bearing, because it is the basis for rejecting muscle-preservation marketing. The energy-expenditure result here directly contradicts the discovery paper's mouse result, and both are reported rather than reconciled.

10. Studied for

    Learning and memory in rats made insulin-deficient

    In male rats made insulin-deficient with streptozotocin, the compound lowered blood glucose but did not prevent the weight loss that comes with that model. In maze testing, the rats receiving the compound PRESERVED their overall performance compared with the model animals that received none, and showed a limited, task-dependent easing of a short-term avoidance-memory deficit rather than a full return to normal across memory measures. This is protection from decline, not improvement. It came alongside a significant drop in a hippocampal inflammatory marker (TNF-alpha), a non-significant trend toward lower IL-1beta, and partial preservation of brain tissue structure in the cortex and hippocampus. In healthy rats given the compound alone, behavior did not improve beyond normal control levels. The authors explicitly noted that they did not establish whether the compound reached the brain directly. RODENT rodent (streptozotocin-induced insulin-deficient male Sprague-Dawley rats) Preclinical moderate A well-described result with several independent outcome measures, but from a single laboratory, a single animal model or a single study design. Open for the full finding and the paper it came from.

    In the literature

    Learning and memory in rats made insulin-deficient

    In male rats made insulin-deficient with streptozotocin, the compound lowered blood glucose but did not prevent the weight loss that comes with that model. In maze testing, the rats receiving the compound PRESERVED their overall performance compared with the model animals that received none, and showed a limited, task-dependent easing of a short-term avoidance-memory deficit rather than a full return to normal across memory measures. This is protection from decline, not improvement. It came alongside a significant drop in a hippocampal inflammatory marker (TNF-alpha), a non-significant trend toward lower IL-1beta, and partial preservation of brain tissue structure in the cortex and hippocampus. In healthy rats given the compound alone, behavior did not improve beyond normal control levels. The authors explicitly noted that they did not establish whether the compound reached the brain directly.^{References 10}

    Modelrodent (streptozotocin-induced insulin-deficient male Sprague-Dawley rats)

    Confidence: moderate

    A well-described result with several independent outcome measures, but from a single laboratory, a single animal model or a single study design.

    Source

    Keskin U, Altin E, Kara MK, et al. Effects of retatrutide on learning and memory in streptozotocin-induced male diabetic rats. Behav Brain Res. 2026;514:116343. doi:10.1016/j.bbr.2026.116343. PMID 42385950

    PubMed 42385950

    How this citation was checked

    Re-resolved by the research pass. PubMed esummary on PMID 42385950 returned first author Keskin U, Behav Brain Res, 2026 Jul 1, volume 514, article 116343, DOI 10.1016/j.bbr.2026.116343, all matching. Full abstract retrieved by efetch; the paper's own phrase "preserved overall Morris Water Maze performance" is verbatim, as is its framing of a limited, task-dependent attenuation of short-term avoidance deficits rather than complete normalisation, and so are the failure to prevent weight loss, the significant TNF-alpha reduction, the non-significant IL-1beta trend, the partial cytoarchitecture preservation, the no-improvement-beyond-control result in non-diabetic animals, and the authors' own caveat that direct central exposure was not established. An earlier draft's verb "improved" stays corrected to "preserved", which is a materially different claim. This is the only inflammation result anywhere in the verified record for this compound, it is hippocampal and it is in rodents, which is why no inflammation category exists on this page.

11. Studied for

    Isolated human heart tissue: force of contraction

    Working with small strips of right atrial tissue removed during open-heart surgery from adults, researchers found that adding the compound to the organ bath in rising nanomolar concentrations increased the force of contraction, building over concentration and over time, and shortened the time the muscle took to relax. When an enzyme that breaks down cAMP was blocked first, the compound became both more potent and more effective at increasing contraction force. That increase in force was greatly reduced by blockers of the GLP-1, GIP and glucagon receptors, was not blocked by a beta-blocker, and was also reduced by a drug that blocks calcium release from internal stores, by a nerve-signal mimic, and by an adenosine receptor agonist. The authors concluded the compound increases contraction force in human atrial tissue through the cAMP system acting via its own three receptors. IMPORTANT FRAMING: this is donated tissue in a dish, not a study in living people, and it says nothing about what happens in an intact body. EX VIVO ex vivo human right atrial tissue obtained at cardiac surgery, in an organ bath Preclinical preliminary A single study, an early mechanistic observation in isolated tissue, or a result announced by the sponsor and not yet peer-reviewed. Open for the full finding and the paper it came from.

    In the literature

    Isolated human heart tissue: force of contraction

    Working with small strips of right atrial tissue removed during open-heart surgery from adults, researchers found that adding the compound to the organ bath in rising nanomolar concentrations increased the force of contraction, building over concentration and over time, and shortened the time the muscle took to relax. When an enzyme that breaks down cAMP was blocked first, the compound became both more potent and more effective at increasing contraction force. That increase in force was greatly reduced by blockers of the GLP-1, GIP and glucagon receptors, was not blocked by a beta-blocker, and was also reduced by a drug that blocks calcium release from internal stores, by a nerve-signal mimic, and by an adenosine receptor agonist. The authors concluded the compound increases contraction force in human atrial tissue through the cAMP system acting via its own three receptors. IMPORTANT FRAMING: this is donated tissue in a dish, not a study in living people, and it says nothing about what happens in an intact body.^{References 11}

    Modelex vivo human right atrial tissue obtained at cardiac surgery, in an organ bath

    Confidence: preliminary

    A single study, an early mechanistic observation in isolated tissue, or a result announced by the sponsor and not yet peer-reviewed.

    Source

    Neumann J, Ahlrep U, Hofmann B, Gergs U. Inotropic effects of retatrutide in isolated human atrial preparations. Naunyn Schmiedebergs Arch Pharmacol. 2026;399(1):317-327. doi:10.1007/s00210-025-04421-3. PMID 40613938

    PubMed 40613938

    How this citation was checked

    Re-resolved by the research pass. PubMed esummary on PMID 40613938 returned first author Neumann J, Naunyn Schmiedebergs Arch Pharmacol, 2026 Jan, volume 399, issue 1, pages 317-327, DOI 10.1007/s00210-025-04421-3, all matching. IDENTITY NOTE, carried forward rather than smoothed over: the first author is Neumann, not Ahlrep; the sibling paper's in-text citation style is what produced the erroneous "Ahlrep et al." attribution now in circulation. A second identity problem sits in the source itself. The published abstract contains an apparent typographical slip, reading "These positive inotropic effects of glucagon were diminished by..." in a paper whose subject is retatrutide, so this entry is written from the open-access full text (PMC12894140) rather than from the abstract. That text resolves it unambiguously: "The positive inotropic effect of retatrutide in the presence of cilostamide could be reduced greatly by the antagonist exendin(9-39), Pro(3)GIP, and SC203972"; the effect began at 10 nM and plateaued at 100 nM, the highest tested; retatrutide "shortened time of relaxation"; ryanodine, carbachol and R-PIA each attenuated it; and with a beta-blocker plus an alpha-blocker present beforehand the compound still roughly doubled contraction force. Organ-bath concentrations are described qualitatively in the finding and kept numerically here only for traceability.

12. Studied for

    Isolated mouse heart tissue: a mechanism for the faster heartbeat

    Working with isolated, still-beating right atrial tissue from mice in an organ bath, researchers found the compound made the tissue beat faster, with the effect reaching statistical significance at low nanomolar concentrations and, at the higher concentration tested, coming close to the effect of a standard adrenaline-like stimulant. In separate electrically paced left atrial tissue from the same animals, the compound did NOT increase the force of contraction, though the adrenaline-like stimulant did. Blocking each of the three target receptors in turn showed that only the glucagon-receptor blocker significantly reduced the faster beating; the GLP-1 and GIP receptor blockers did not. The effect survived a beta-blocker but was blunted by an inhibitor of the PKA enzyme, and was amplified by a drug that stops cAMP being broken down. The authors concluded the compound speeds the beating rate in mouse right atria through the glucagon receptor, signalling via cAMP and PKA. The authors themselves framed the study around the observation that this compound raised heart rate in human safety studies, and listed a direct action on the heart's pacemaker as one of three candidate explanations they set out to test. EX VIVO ex vivo isolated atrial tissue from adult CD1 mice, in an organ bath Preclinical preliminary A single study, an early mechanistic observation in isolated tissue, or a result announced by the sponsor and not yet peer-reviewed. Open for the full finding and the paper it came from.

    In the literature

    Isolated mouse heart tissue: a mechanism for the faster heartbeat

    Working with isolated, still-beating right atrial tissue from mice in an organ bath, researchers found the compound made the tissue beat faster, with the effect reaching statistical significance at low nanomolar concentrations and, at the higher concentration tested, coming close to the effect of a standard adrenaline-like stimulant. In separate electrically paced left atrial tissue from the same animals, the compound did NOT increase the force of contraction, though the adrenaline-like stimulant did. Blocking each of the three target receptors in turn showed that only the glucagon-receptor blocker significantly reduced the faster beating; the GLP-1 and GIP receptor blockers did not. The effect survived a beta-blocker but was blunted by an inhibitor of the PKA enzyme, and was amplified by a drug that stops cAMP being broken down. The authors concluded the compound speeds the beating rate in mouse right atria through the glucagon receptor, signalling via cAMP and PKA. The authors themselves framed the study around the observation that this compound raised heart rate in human safety studies, and listed a direct action on the heart's pacemaker as one of three candidate explanations they set out to test.^{References 12}

    Modelex vivo isolated atrial tissue from adult CD1 mice, in an organ bath

    Confidence: preliminary

    A single study, an early mechanistic observation in isolated tissue, or a result announced by the sponsor and not yet peer-reviewed.

    Source

    Neumann J, Ahlrep U, Hofmann B, Gergs U. Contractile effects of retatrutide in isolated mouse atrial preparations. Naunyn Schmiedebergs Arch Pharmacol. 2025;398(12):17147-17160. doi:10.1007/s00210-025-04335-0. PMID 40464942

    PubMed 40464942

    How this citation was checked

    Re-resolved by the research pass. PubMed esummary on PMID 40464942 returned first author Neumann J, Naunyn Schmiedebergs Arch Pharmacol, 2025 Dec, volume 398, issue 12, pages 17147-17160, DOI 10.1007/s00210-025-04335-0, all matching. Abstract read by efetch, and because several details sit outside the abstract the open-access full text (PMC12678479) was additionally retrieved and each confirmed: "This positive chronotropic effect gained significance at 3 nM retatrutide"; "A 100 nM retatrutide was nearly as effective as 1 microM isoprenaline to raise beating rate"; "Only adomeglivant significantly reduced beating rate that had been stimulated by retatrutide", with the text stating explicitly that the effect was not blocked by exendin9-39 (GLP-1R) or ProGIP (GIPR). Two corrections stay in force. The human-tissue inotropy result belongs to a separate publication and now has its own entry above. And an earlier draft's closing sentence, "This provides a mechanistic ex vivo correlate for the heart-rate increases reported in clinical trials", was written in the dossier's own voice; it is replaced with an attributed statement, because the paper's introduction does say, in the authors' voice, "In humans, retatrutide in safety studies elevated the heart rate and decreased the blood pressure (Coskun et al. 2022, Jastreboff et al. 2023). Firstly, this may result from direct PCE of retatrutide on the human cardiac sinus node...", listed alongside two other candidate mechanisms. Organ-bath concentrations are described qualitatively in the finding and kept numerically here only for traceability.

13. Studied for

    TRIUMPH-1 headline results as announced by the sponsor, not peer-reviewed

    TRIUMPH-1 is an 80-week phase 3 randomized double-blind placebo-controlled trial that randomized 2,339 participants selected for excess body weight and without raised blood sugar at entry. In the company's own announcement, under its more favorable of two analyses, average body weight at 80 weeks was down 19.0%, 25.9% and 28.3% across the three ascending groups, against 2.2% on placebo. Under the more conservative analysis the same figures were 17.6%, 23.7% and 25.0%, against 3.9% on placebo; both sets are reported here because the gap between them is exactly the kind of detail that gets dropped in retelling. A pre-specified extension to 104 weeks enrolled only participants who had already completed 80 weeks and tolerated their assigned amount, and reported up to 30.3% average weight reduction; that group is self-selected and its numbers are not comparable to the main trial. On side effects, the announcement reported nausea, diarrhea, constipation and vomiting rising with the amount given; a nerve-sensation side effect (dysesthesia) in 5.1%, 12.3% and 12.5% of participants receiving the compound against 0.9% on placebo; and discontinuation due to side effects in 4.1%, 6.9% and 11.3% against 4.9% on placebo. SOURCING CAVEAT, load-bearing: these figures come from a company news release, not a peer-reviewed publication. A PubMed search found no TRIUMPH-1 outcomes paper indexed. They must never be presented with the weight of the peer-reviewed trials above, and no buyer-facing sentence on this page rests on any of them. human, phase 3 trial; company topline news release only, not peer-reviewed Preclinical preliminary A single study, an early mechanistic observation in isolated tissue, or a result announced by the sponsor and not yet peer-reviewed. Open for the full finding and the paper it came from.

    In the literature

    TRIUMPH-1 headline results as announced by the sponsor, not peer-reviewed

    TRIUMPH-1 is an 80-week phase 3 randomized double-blind placebo-controlled trial that randomized 2,339 participants selected for excess body weight and without raised blood sugar at entry. In the company's own announcement, under its more favorable of two analyses, average body weight at 80 weeks was down 19.0%, 25.9% and 28.3% across the three ascending groups, against 2.2% on placebo. Under the more conservative analysis the same figures were 17.6%, 23.7% and 25.0%, against 3.9% on placebo; both sets are reported here because the gap between them is exactly the kind of detail that gets dropped in retelling. A pre-specified extension to 104 weeks enrolled only participants who had already completed 80 weeks and tolerated their assigned amount, and reported up to 30.3% average weight reduction; that group is self-selected and its numbers are not comparable to the main trial. On side effects, the announcement reported nausea, diarrhea, constipation and vomiting rising with the amount given; a nerve-sensation side effect (dysesthesia) in 5.1%, 12.3% and 12.5% of participants receiving the compound against 0.9% on placebo; and discontinuation due to side effects in 4.1%, 6.9% and 11.3% against 4.9% on placebo. SOURCING CAVEAT, load-bearing: these figures come from a company news release, not a peer-reviewed publication. A PubMed search found no TRIUMPH-1 outcomes paper indexed. They must never be presented with the weight of the peer-reviewed trials above, and no buyer-facing sentence on this page rests on any of them.^{References 13}

    Modelhuman, phase 3 trial; company topline news release only, not peer-reviewed

    Confidence: preliminary

    A single study, an early mechanistic observation in isolated tissue, or a result announced by the sponsor and not yet peer-reviewed.

    Source

    Eli Lilly and Company. Topline results announcement for TRIUMPH-1 (NCT05929066). News release, May 21, 2026

    Source

    How this citation was checked

    Fetched live by the research pass and read in full (HTTP 200). The release confirms verbatim: the trial identity and registration (TRIUMPH-1, NCT05929066, 80-week phase 3 randomised double-blind placebo-controlled master trial), 2,339 randomised, the three efficacy-estimand weight figures and the placebo figure, the three regimen-estimand figures and the placebo figure, the 104-week extension value together with its enrolment restriction to completers with high baseline BMI, the four gastrointestinal event rates, the three dysesthesia rates against 0.9% placebo, and the three discontinuation rates against 4.9% placebo. The release also states plainly that the compound is investigational and available only to participants in the company's clinical trials. Two things were removed and stay removed: the WOMAC knee-pain and Apnea-Hypopnea Index figures, which rested on a companion investor release that could NOT be resolved on repeated attempts (connection failed, HTTP 000), so under the verify-or-delete rule they are deleted rather than carried on an unresolvable URL, and this release itself states that those data would be released separately; and milligram amount labels. The release's promotional headline is deliberately not reproduced as the citation title, because a news-release headline is marketing copy rather than a scholarly title and printing it would put an unsupported superlative on this page.

8 strong / 2 moderate / 3 preliminaryRated by how many independent labs and how many kinds of experiment found the same thing.