No trial has compared eloralintide or EloraTZP with retatrutide, and none of the 22 eloralintide trials on ClinicalTrials.gov includes retatrutide. Every comparison so far sets numbers from different trials beside each other.
Those trials differ in ways that move the numbers: who took part (with or without type 2 diabetes), how long they ran (28 to 80 weeks), and which estimand they report (assuming everyone stayed on treatment, or counting everyone randomised). Lined up by those three things, the data show three effective drugs and no winner.
The closest design to the EloraTZP trial is retatrutide's own phase 2 in type 2 diabetes, which ran 36 weeks instead of 48. Even that pair cannot rank the two.
Three different drugs
The three are related but not interchangeable. Eloralintide is an amylin drug, EloraTZP adds it to tirzepatide, and retatrutide is a single molecule that reaches three incretin and glucagon receptors. All three are Lilly's, and none is approved anywhere as of October 2, 2026.
| Eloralintide | EloraTZP | Retatrutide | |
|---|---|---|---|
| What it activates | Amylin receptors (designed to favour them over the calcitonin receptor) | Amylin, GIP and GLP-1 receptors (two drugs) | GIP, GLP-1 and glucagon receptors (one molecule) |
| Given as | One weekly injection | Two weekly injections in phase 2b; one planned for phase 3 | One weekly injection |
| Developer | Eli Lilly | Eli Lilly | Eli Lilly |
| Furthest stage | Phase 3 (ENLIGHTEN, recruiting) | Phase 2b complete; phase 3 planned by end of 2026 | Phase 3 complete in several trials; filing planned for early 2027 |
Retatrutide's filing plan comes from Lilly's second-quarter results release of August 5, 2026, as covered in our post on the filing plan. Eloralintide's and EloraTZP's stages come from ClinicalTrials.gov and Lilly's September 30, 2026 announcement.
Like for like: the type 2 diabetes trials
EloraTZP has only been tested in people with type 2 diabetes, so the fair comparisons are other trials in type 2 diabetes:
| Trial | Weeks | People | Estimand | Top dose: weight | Placebo | HbA1c (top dose) |
|---|---|---|---|---|---|---|
| EloraTZP phase 2b (2026) | 48 | 367 | Efficacy | -23.3% (eloralintide 9 mg + tirzepatide 15 mg) | -3.0% | -2.9 points |
| Retatrutide phase 2 (2023) | 36 | 281 | Not named in the abstract | -16.9% (12 mg) | -3.0% | -2.02 points at 24 weeks |
| TRIUMPH-2, retatrutide (2026) | 80 | 1,152 | Treatment-regimen | -18.8% (12 mg) | -5.1% | Not in the abstract |
| REDEFINE 2, CagriSema (2025) | 68 | 1,206 | Treatment-policy | -13.7% | -3.4% | 73.5% reached 6.5% or lower |
The retatrutide phase 2 trial is the closest match in design: it enrolled people with type 2 diabetes in the same HbA1c window, 7.0% to 10.5%, on diet and exercise or metformin, and it was double-blind with a placebo group. But it ran for 36 weeks rather than 48, it did not allow SGLT2 inhibitors, its HbA1c figures are the 24-week primary endpoint, and its abstract does not say which estimand its least-squares means use. At its top dose retatrutide gave 16.9% weight loss and a 2.02-point HbA1c fall; EloraTZP's top combination gave 23.3% and 2.9 points twelve weeks later. Part of that gap may be time: in retatrutide's obesity trial, weight loss on 12 mg grew from 17.5% at 24 weeks to 24.2% at 48.
TRIUMPH-2 is the largest and longest trial in the table, and the hardest test: 80 weeks, 1,152 people, and a treatment-regimen estimand that keeps people who stopped the drug in the result. It reported 18.8% on 12 mg. CagriSema's REDEFINE 2, another amylin and incretin pairing, reported 13.7% over 68 weeks under a similar estimand. Neither can be ranked against EloraTZP's 23.3%, which assumes nobody stopped, from a 48-week trial with about 37 people per group.
Without diabetes: the obesity trials
Eloralintide on its own has a trial in people without diabetes, and so do retatrutide, CagriSema and petrelintide:
| Trial | Weeks | People | Estimand | Top dose: weight | Placebo |
|---|---|---|---|---|---|
| Eloralintide phase 2 (2025) | 48 | 263 | Efficacy | -20.1% (9 mg) | -0.4% |
| Retatrutide phase 2 (2023) | 48 | 338 | Not named in the abstract | -24.2% (12 mg) | -2.1% |
| TRIUMPH-1, retatrutide (2026) | 80 | 2,339 | Treatment-regimen | -25.0% (12 mg) | -3.9% |
| REDEFINE 1, CagriSema (2025) | 68 | 3,417 | Treatment-policy | -20.4% | -3.0% |
| ZUPREME 1, petrelintide (2026) | 28 (primary) | 493 | Efficacy | -9.8% (5 mg) | -1.7% |
The two phase 2 trials, eloralintide's and retatrutide's, share a length (48 weeks) and a broadly similar population, and retatrutide's top dose came out higher, 24.2% against 20.1%. That is the most like-for-like monotherapy comparison available, and it still crosses two separate trials, with different dose schedules and placebo responses (2.1% against 0.4%), run years apart. TRIUMPH-1 and REDEFINE 1 are larger, longer and count everyone randomised; petrelintide's primary result is at 28 weeks and is not comparable on time at all.
Why cross-trial comparisons mislead
Each of these differences can move a headline by several points on its own:
- Estimand. An efficacy estimand reports what would have happened had everyone stayed on treatment; a treatment-regimen or treatment-policy estimand counts everyone randomised, including people who stopped. When many people stop, as in the EloraTZP combination groups, the first runs higher than the second.
- Population. People with type 2 diabetes tend to lose less weight on the same drug. Retatrutide 12 mg gave 25.0% in TRIUMPH-1 and 18.8% in TRIUMPH-2, both at 80 weeks and under the same estimand; eloralintide 6 mg gave 17.6% without diabetes and 12.3% with it.
- Duration. Weight loss keeps growing for many months on these drugs, so a 36-week result and an 80-week result measure different points on the curve.
- Placebo. Placebo groups lost from 0.4% to 5.1% across these trials. Subtracting placebo helps, but does not fix the other differences.
- Size. A group of about 37 people, as in the EloraTZP trial, carries far more uncertainty than a group of several hundred, and Lilly has not yet published confidence intervals for EloraTZP.
- Source. The EloraTZP figures come from a company announcement; the others come from peer-reviewed papers. The papers carry the detail that settles questions like these; announcements do not.
Side effects and stopping, as reported
| Trial | Gastrointestinal events or nausea | Stopped for adverse events |
|---|---|---|
| Eloralintide phase 2 | Nausea 11% to 64% by group (placebo 14%); fatigue up to 46% (placebo 12%) | 0 to 2 people per group left the study for an adverse event (registry) |
| EloraTZP phase 2b | Mostly gastrointestinal, more on the combinations; rates by group not reported | 10.8% to 27.0% on combinations; 2.9% on tirzepatide; 16.7% on placebo |
| Retatrutide phase 2, diabetes | Gastrointestinal events 13% to 50% by group (placebo 13%) | Not in the abstract |
| TRIUMPH-2 | Diarrhoea 27% to 34%, nausea 14% to 28% (placebo 13% and 8%) | 4% to 12% stopped for adverse events or death (placebo 5%) |
| REDEFINE 2 | Gastrointestinal events 72.5% (placebo 34.4%) | Not in the abstract |
The trials report side effects in different ways, which makes this the weakest comparison of all. What can be said is narrower. Eloralintide alone had high rates of nausea when started at a high dose and notably high fatigue at 9 mg. EloraTZP lost more people to adverse events than tirzepatide alone. Retatrutide brings its own signals, including higher heart rate in phase 2, and low blood pressure and dysaesthesia, an altered skin sensation, in TRIUMPH-2.
What a head-to-head trial would need
Ranking these drugs needs one randomised trial that gives them to the same kind of people at the same time. To be decisive it would need:
- the same population, ideally run separately in people with and without type 2 diabetes;
- the doses and escalation schedules each drug will actually be approved with;
- the same length, long enough to reach the plateau, 64 to 80 weeks in these programmes;
- a treatment-regimen estimand as the primary result, so that stopping counts;
- enough people to tell a few points apart, and side effects and discontinuations reported the same way for both arms.
No such trial is registered among the 22 eloralintide studies. Until one is, the honest answer to "which is better" is that the data do not say.
The amylin landscape around them
Eloralintide is one of several long-acting amylin drugs. Cagrilintide, in Novo Nordisk's CagriSema with semaglutide, also activates the calcitonin receptor; CagriSema has published phase 3 results in people with and without type 2 diabetes. Petrelintide, from Zealand Pharma, is in phase 2: its ZUPREME 1 trial reported 7.9% to 9.8% at 28 weeks, with no gain from doses above 5 mg and nausea in 20% of people against 6% on placebo. Lilly's bet with eloralintide is selectivity for amylin receptors, and with EloraTZP, adding an amylin drug to its existing GIP and GLP-1 agonist. Whether either beats the alternatives is a question for the trials still running.
Frequently asked questions
Is EloraTZP better than retatrutide?
Nobody knows. No trial has compared them, and the published figures come from trials with different populations, lengths and estimands. EloraTZP's 23.3% is a 48-week efficacy-estimand figure in type 2 diabetes; retatrutide's 18.8% in type 2 diabetes is an 80-week treatment-regimen figure.
Which trial is closest to EloraTZP's?
Retatrutide's phase 2 trial in type 2 diabetes, with the same HbA1c entry range and background treatment. It ran 36 weeks rather than 48, and retatrutide 12 mg gave 16.9% weight loss there.
Is eloralintide stronger than retatrutide on its own?
In the two 48-week phase 2 trials in people without diabetes, retatrutide's top dose gave 24.2% and eloralintide's 20.1%. Those are separate trials with different designs, so the gap is suggestive at most.
Why do people compare them?
All three are Lilly obesity drugs reporting large weight losses within weeks of each other, and EloraTZP's headline number is higher than retatrutide's figure in type 2 diabetes. The numbers were measured differently.
Are any of them approved?
No. As of October 2, 2026 none is approved anywhere. Lilly plans to file retatrutide in early 2027; eloralintide is in phase 3, and EloraTZP's phase 3 is planned to start by the end of 2026.
Related on Grey Peptides
Sources
- Eli Lilly and Company. EloraTZP (combination of eloralintide and tirzepatide) phase 2b results. News release, September 30, 2026. Read October 2, 2026.
- Billings LK, et al. "Eloralintide, a selective amylin receptor agonist for the treatment of obesity: a 48-week phase 2, multicentre, double-blind, randomised, placebo-controlled trial." Lancet. 2025;406(10520):2631-2643.
- Jastreboff AM, et al. "Triple-Hormone-Receptor Agonist Retatrutide for Obesity: A Phase 2 Trial." N Engl J Med. 2023;389(6):514-526.
- Rosenstock J, 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.
- Jastreboff AM, et al. "Retatrutide, a Triple Hormone Receptor Agonist, for Treatment of Obesity." N Engl J Med. 2026 (TRIUMPH-1).
- Bellido V, et al. "Retatrutide in adults with obesity and type 2 diabetes (TRIUMPH-2): a double-blind, parallel-group, randomised, placebo-controlled, phase 3 trial." Lancet. 2026.
- Garvey WT, et al. "Coadministered Cagrilintide and Semaglutide in Adults with Overweight or Obesity." N Engl J Med. 2025;393(7):635-647 (REDEFINE 1).
- Davies MJ, et al. "Cagrilintide-Semaglutide in Adults with Overweight or Obesity and Type 2 Diabetes." N Engl J Med. 2025;393(7):648-659 (REDEFINE 2).
- Garvey WT, et al. "Petrelintide, a human amylin analogue for the treatment of obesity (ZUPREME 1): a randomised, double-blind, placebo-controlled, phase 2 trial." Lancet Diabetes Endocrinol. 2026.
- ClinicalTrials.gov: NCT06230523 (eloralintide phase 2 results) and the 22 studies naming eloralintide or LY3841136. Read October 2, 2026.
Medical disclaimer: This page is for education only and is not medical advice. EloraTZP and eloralintide are investigational and not approved for use. Do not start, stop, or self-administer any drug based on this page; consult a licensed clinician.