Retatrutide phase 3 results: what TRIUMPH-1 and TRIUMPH-2 showed

What is retatrutide?

Retatrutide is an investigational, once-weekly injection from Eli Lilly that acts on three hormone receptors: GIP, GLP-1 and glucagon. Semaglutide acts on one (GLP-1). Tirzepatide acts on two (GIP and GLP-1). The glucagon piece is the new part; it is thought to raise energy use, which is one reason researchers expected larger weight loss.

It is not FDA-approved for anything. Lilly's TRIUMPH-2 release says the company "plans to submit a Biologics License Application (BLA) in Q1 2027". Outside clinical trials, it is not legally available in the United States. FDA review usually takes months after a filing, so retatrutide is not something you can be prescribed in 2026.

What did the TRIUMPH phase 3 trials show?

Two papers were published on September 29, 2026, and presented the next day at the EASD meeting in Milan:

• TRIUMPH-1 (NEJM, Jastreboff et al.): 2,339 adults with obesity and no diabetes, 80 weeks, three doses against placebo.
• TRIUMPH-2 (The Lancet, Bellido et al.): 1,152 adults with type 2 diabetes and a BMI of 27 or higher, mean starting HbA1c 7.71%, 80 weeks.

Weight loss rose with dose. At 12 mg, people without diabetes lost an average 25.0% of body weight, about 21 percentage points more than placebo. People with diabetes lost less (18.8%), which matches the pattern seen with tirzepatide and semaglutide: weight comes off more slowly with type 2 diabetes.

TRIUMPH-1 also measured two weight-related conditions. In the sleep apnea subgroup, breathing interruptions fell by 31.7 to 34.3 events an hour on the two higher doses, against 9.9 on placebo. In the knee osteoarthritis subgroup, pain scores improved more than with placebo at every dose. For diabetes, Lilly's release reports A1C falling by up to 1.6 points, with up to 40.0% of participants reaching an A1C below 5.7%, the non-diabetic range.

Why do Lilly's numbers differ from the journal numbers?

You may see 28.3% quoted for the 12 mg dose and 25.0% in the NEJM paper. Both are correct; they answer different questions.

• Treatment-regimen estimand (the journals): counts every person randomized, including people who stopped the drug or missed doses. It estimates what happens when a drug is prescribed in the real world.
• Efficacy estimand (Lilly's releases): estimates what happens if everyone keeps taking the drug as directed. It is always the larger number, because stopping treatment is set aside.

The same gap appears in TRIUMPH-2: 18.8% in The Lancet against 20.8% in Lilly's release at 12 mg. When you compare drugs, compare like with like. The tirzepatide and semaglutide figures below are treatment-regimen numbers, the same basis as the journal retatrutide figures.

What side effects did people have, and how many stopped?

As with every drug in this class, the most common side effects were gastrointestinal. The NEJM abstract states that "the most common adverse events were gastrointestinal." The Lancet paper gives the detail for people with diabetes:

Two findings are new compared with semaglutide and tirzepatide:

• Dysesthesia, an odd skin sensation such as tingling or burning, rose with dose. Trade press reporting Lilly's earlier TRIUMPH-1 topline put it at 5.1%, 12.3% and 12.5% by dose against 0.9% with placebo; we could not confirm those figures in the journal abstract.
• Low blood pressure reached 5% to 6% on the two higher doses in TRIUMPH-2. If you take blood-pressure medicine, this would matter if the drug is approved.

Stopping for side effects was 12% on 9 mg and 8% on 12 mg, against 5% on placebo, in TRIUMPH-2. For TRIUMPH-1, trade press reported 4.1%, 6.9% and 11.3% by dose against 4.9% with placebo (reported, not confirmed in the abstract). The Lancet's editorialist, Jens Juul Holst, wrote that "it will be important to continue long-term follow-up of patients on triple agonists." Read our guide to managing GLP-1 side effects for what helps with nausea and diarrhea on today's drugs.

Is retatrutide better than tirzepatide or semaglutide?

It produced among the largest average weight losses reported in a published phase 3 trial of this class, but no head-to-head trial has been published. Here are the closest comparable figures, all treatment-regimen averages in adults without diabetes:

• Retatrutide 12 mg: 25.0% at 80 weeks (TRIUMPH-1), placebo 3.9%.
• Tirzepatide 15 mg: 20.9% at 72 weeks (SURMOUNT-1), placebo 3.1%.
• Semaglutide 2.4 mg: 14.9% at 68 weeks (STEP 1), placebo 2.4%.

Subtracting placebo gives about 21 (the journal reports a 21.0-point difference), 17.8 and 12.5 percentage points. Treat this as a rough guide only. The trials ran for different lengths, enrolled different people and were done years apart. A September 2026 Annals of Internal Medicine review that reviewed 38 randomized trials found a similar order: placebo-subtracted loss of 22.1% for retatrutide, 19.0% for tirzepatide, 14.8% for injectable semaglutide and 12.4% for orforglipron. The only direct comparison between approved drugs is tirzepatide against semaglutide; see our head-to-head explainer.

Also keep in mind: tirzepatide and semaglutide have years of real-world use and FDA labels that list risks and who should not take them. Retatrutide has 80 weeks of trial data and a few new safety signals still being studied.

Is the "retatrutide" sold online the same drug?

No. Vials sold online as "research peptides" are not the drug tested in TRIUMPH, and the trial results do not apply to them. The FDA's page on unapproved GLP-1 drugs (current as of October 1, 2026) says: "Retatrutide and cagrilintide cannot be used in compounding under federal law." A JAMA news feature published October 9, 2026 adds the clinical picture:

• FDA has sent 21 warning letters to retatrutide sellers since December 2024.
• University of Queensland testing found "10 mg" vials holding from about half to about twice the labeled amount.
• An Australian patient had an esophageal rupture after a "retatrutide" product that contained semaglutide and no retatrutide; six liver-toxicity reports came from the state of Victoria.

A JAMA Network Open study of Reddit posts (October 14, 2026) found self-reported users describing fatigue, nausea, insomnia and a faster heart rate; the authors call the findings "hypothesis-generating." A Lilly spokesperson told JAMA: "What is being sold on the black market is not medicine."

What do these results mean for you?

Here is what the trial averages look like in pounds for someone who starts at 250 lb. Each figure is the trial's average percentage multiplied by 250, rounded; the last column subtracts the placebo average in the same trial.

For example, 25.0% of 250 lb is 62.5 lb, and the placebo group's 3.9% is about 10 lb, so the drug's added effect at 12 mg works out to about 53 lb on average. Some people lost much more than the average and some lost little, and people with type 2 diabetes lost less. These are averages, not a promise.

Practical steps:

1. If you want retatrutide now, the only legal route is a clinical trial. Search ClinicalTrials.gov for retatrutide studies and ask your clinician whether one near you fits.
2. If you are deciding today, approved tirzepatide and semaglutide gave about 18 and 12.5 percentage points more weight loss than placebo in SURMOUNT-1 and STEP 1. Starting one now does not stop you from switching later if retatrutide is approved.
3. If you have type 2 diabetes and take blood-pressure medicine, ask how the hypotension finding might apply to you if retatrutide reaches the market.
4. If you have used an online "retatrutide" product, tell your clinician. Seek care for yellow skin or eyes, dark urine, severe abdominal pain or a racing heart, and report reactions to FDA MedWatch.
5. Check who should not take a GLP-1 before starting any drug in this class.

What can you take now, and what does it cost without insurance?

Until retatrutide is approved, the main FDA-approved GLP-1 options for weight management include tirzepatide (Zepbound), semaglutide (Wegovy pen and pill) and orforglipron (Foundayo). Manufacturer self-pay prices in October 2026:

Compounded tirzepatide costs less. NexLife lists it at $115 a month on a 12-month plan ($129 month to month). Compounded tirzepatide is made with the drug in Zepbound, but it is not the same product, it is not FDA-approved, and the SURMOUNT results apply to Zepbound only. Compare programs in our compounded GLP-1 price index.

For more trial and label updates, see our GLP-1 news page.