GLP-1 Obesity Phase 3 Weekly Digest (5 October 2026): Retatrutide Sets a New Weight-Loss Ceiling
Executive Summary
The week's main result is retatrutide's full Phase 3 publication: in TRIUMPH-1, 12 mg produced -25.0% weight loss at 80 weeks versus -3.9% for placebo, the deepest result yet in a Phase 3 obesity trial, according to a conference newsletter summary (the NEJM paper itself was blocked to me, so I could not check these figures against the paper). Retatrutide is a GIP/GLP-1/glucagon triple agonist, so it extends the GLP-1 class rather than being a pure GLP-1 drug. The data came out around the 62nd EASD meeting in Milan, with TRIUMPH-2 published in The Lancet and TRIUMPH-1 in NEJM on 29 September 2026.
The window covers about 21 September to 5 October 2026. Three other items matter. Novo's CagriSema (cagrilintide plus semaglutide) beat tirzepatide 5 mg on weight in REIMAGINE 5 in type 2 diabetes and showed 21.0% loss in REDEFINE 9. Lilly's oral orforglipron (Foundayo) cleared a cardiovascular safety bar in ACHIEVE-4. Two newer dual agonists, survodutide and olatorepatide, reported Phase 3 weight data.
Most of this week's numbers come from sponsor press releases and one secondary newsletter. Treat them as unreviewed efficacy-estimand claims until you have read the peer-reviewed papers. Several cross-agent comparisons are indirect or open-label. My bottom line: retatrutide's effect size is large and consistent across Lilly's own readouts, but its tolerability profile (dysesthesia, dose reductions for excess weight loss) and the lack of head-to-head trials against tirzepatide limit how far the evidence can be pushed today.
PICO frame: adults with obesity or overweight (with or without type 2 diabetes); GLP-1-based agents; comparator placebo, active drug, or insulin; outcomes of percent body weight change, glycaemic control, cardiovascular safety and discontinuation.
Retatrutide: deepest weight loss yet, with new tolerability signals
- TRIUMPH-1 (obesity, no diabetes): the trial enrolled 2,339 adults across 131 sites. The 12 mg dose gave -25.0% at 80 weeks versus -3.9% on placebo, deepening to -30.3% at week 104 in an extension. The extension is a subgroup of participants who stayed on treatment, so it should not be read as the randomised effect.
- Dose reductions: 22.5% of the 12 mg group needed permanent dose reductions versus 0.5% on placebo, mostly for perceived excessive weight loss or BMI of 22 or less. Discontinuation for adverse events was 4.1% (4 mg), 6.5% (9 mg) and 11.1% (12 mg) versus 4.6% on placebo. Dysesthesia reached up to 12.5%.
- TRIUMPH-2 (obesity plus type 2 diabetes): the trial randomised 1,152 participants to 4, 9 or 12 mg or placebo. At 80 weeks (efficacy estimand) weight fell 12.7%, 19.1% and 20.8% versus 4.0% on placebo, and A1C fell up to 1.6 percentage points. Nausea at 12 mg was 28.0% versus 8.0%, and AE discontinuation was 7.7% at 12 mg versus 4.9% on placebo.
- Regulatory path: Lilly plans a BLA submission in Q1 2027. That is a company plan, not a filing.
Appraisal: both trials are randomised, double-blind and placebo-controlled, which supports low risk of bias in design. The figures quoted are efficacy-estimand results (what would happen if everyone stayed on drug), which overstate real-world effect compared with treatment-regimen estimands. Lilly's table flags some secondary results (including the share reaching A1C below 5.7%) as not controlled for family-wise type 1 error, so treat those as exploratory.
KEY TAKEAWAY: Retatrutide's efficacy is large, but its numbers here are efficacy-estimand, sponsor-reported, and paired with a different safety signal than semaglutide or tirzepatide.
Other Phase 3 readouts this fortnight
| Agent (sponsor) | Trial | Key result | Source |
|---|---|---|---|
| CagriSema (Novo) | REIMAGINE 5 (T2D, week 60) | Weight loss 12.4% vs 9.1% for tirzepatide 5 mg (superior); HbA1c 1.71% vs 1.67% (non-inferior) | Novo release, 23 Sep 2026 |
| CagriSema (Novo) | REDEFINE 9 (overweight/obesity, week 68) | Weight loss 21.0% (1.0/1.0 mg) vs 2.0% placebo | Novo release, 23 Sep 2026 |
| Orforglipron / Foundayo (Lilly) | ACHIEVE-4 (T2D, 2,749 randomised, open-label) | MACE-4 HR 0.84 (95% CI 0.59 to 1.20), non-inferior vs insulin glargine; weight -8.8% vs +1.7% at 52 weeks | Lilly release, 30 Sep 2026 |
| Survodutide (BI/Zealand) | SYNCHRONIZE-2 (T2D, 755 adults, 76 weeks) | Up to -13.1% vs -3.1% placebo; vomiting up to 39%; 26% AE discontinuation in treatment arms | Biotech Readout, 2 Oct 2026 |
| Olatorepatide (Hansoh/Regeneron) | LIGHTEN (604 Chinese adults, 48 weeks) | Up to -19.3% vs -1.6% placebo | Biotech Readout, 2 Oct 2026 |
- CagriSema caveat: the REIMAGINE 5 comparator was tirzepatide at a low 5 mg dose, which the company itself notes reflects that not all patients stay on the highest available dose. The result does not show superiority over top-dose tirzepatide. Both are topline company figures. The FDA and Health Canada decisions on CagriSema are expected in Q4 2026.
- Orforglipron mortality signal: Lilly reports 53% lower cardiovascular death (HR 0.47, 95% CI 0.25 to 0.90) and 57% lower all-cause death (HR 0.43), but these were not controlled for multiplicity, the trial was open-label, and the primary MACE-4 result had a wide interval that included 1.0. Treat the death findings as hypothesis-generating. Lilly is running ATTAIN-OUTCOMES to test cardiovascular outcomes in obesity.
- Survodutide tolerability: the newsletter reads the modest HbA1c effect and heavy GI burden as favouring a liver-disease role over glycaemic control. That is the newsletter author's interpretation, not a trial conclusion.
- Olatorepatide: a China-only, 48-week trial, so generalisability is limited. Regeneron plans a global Phase 3 programme later this year.
KEY TAKEAWAY: The sponsors report positive primary or co-primary results for the trials in the table, but the comparators (low-dose tirzepatide, insulin, placebo) and open-label designs make cross-trial ranking unsafe.
Regulatory and real-world context
- Foundayo is already approved. The FDA approved orforglipron for chronic weight management (Lilly announced it on 1 April 2026), so this week's orforglipron news is post-approval evidence in type 2 diabetes, not an obesity approval. The approval release cites ATTAIN-1 (3,127 randomised, 72 weeks), where the highest dose gave 12.4% loss in those who stayed on treatment versus 0.9% on placebo.
- Indirect comparison, tirzepatide versus Wegovy HD: Lilly's analysis of SURMOUNT-1 and STEP UP data found tirzepatide 15 mg gave 4.5% greater weight loss than semaglutide 7.2 mg over 72 weeks. Under the efficacy estimand the gap shrank to 1.8% and was not statistically significant. This is a sponsor-run indirect comparison, not a head-to-head trial.
- Real-world oral switch: in Novo's OCTANE study of 194 adults switching from injectables to the Wegovy pill, mean loss was 4.1% at three months. There is no control group and follow-up is short.
- Registry check: TRIUMPH-2 is NCT05929079 and ACHIEVE-4 is NCT05803421. I did not run a full ClinicalTrials.gov sweep for newly registered trials this week, so new registrations may be missing.
Evidence certainty and what would change this view
| Finding | Design | Main concern | Provisional GRADE (my judgement) |
|---|---|---|---|
| Retatrutide 12 mg weight loss | Double-blind RCTs | Efficacy estimand, sponsor source, single programme | Moderate |
| CagriSema beats tirzepatide 5 mg | Phase 3 RCT, topline | Low-dose comparator, no full paper | Low to moderate |
| Orforglipron lower death risk | Open-label, uncorrected secondary | Imprecision, multiplicity | Low |
| Tirzepatide beats Wegovy HD | Indirect comparison | Indirectness | Low |
| Survodutide, olatorepatide weight loss | Placebo-controlled RCTs | Secondary summary only; one population China-only | Low to moderate |
GRADE labels are my own reading of the sources, not published ratings. Three things would change the picture: the full NEJM and Lancet papers showing treatment-regimen estimands that differ materially from the efficacy-estimand numbers; a head-to-head trial of retatrutide against tirzepatide; and FDA decisions on CagriSema in Q4 2026. For the next edition, watch the retatrutide BLA timing and any regulator safety commentary on dysesthesia.
Search note: I used web searches for September to October 2026 news and fetched sponsor releases and one conference newsletter. NEJM, Fierce Biotech and the FDA pages were not read in full, and PubMed and ClinicalTrials.gov were not queried, so coverage of smaller publications is incomplete.