I take Zepbound, I have lost about 170 pounds on it, and the drug everyone now wants to measure it against comes from the same company that makes mine. Retatrutide is Lilly’s next obesity molecule, it is not approved anywhere, and going by its trial results it is probably going to be the most effective obesity drug on the market when it gets here.
Disclosure, since it colors all of this: I inject tirzepatide and I own Eli Lilly stock. This is a company racing itself.
Which is the problem. A drug this good that you cannot legally buy has thousands of people injecting something they ordered off a website. Almost all of them would do great on the drug I am already on.
| FDA approved | No |
| Development stage | Phase 3. Lilly said in July 2026 the data package for registration is complete; other Phase 3 trials are still running |
| FDA application | Planned Q1 2027, as a biologic (a fight of its own), for obesity, sleep apnea and knee osteoarthritis pain |
| Brand name | Not announced |
| Last checked | September 1, 2026 |
How much weight does retatrutide take off compared to Zepbound?
More, and it gets there without needing its top dose.
In TRIUMPH-1, 2,339 adults took 4, 9 or 12 mg of retatrutide or placebo for 80 weeks. The top dose came in at 28.3 percent, an average of 70 pounds. SURMOUNT-1 reported 20.9 percent on 15 mg of tirzepatide. You cannot just subtract one from the other; the two trials report weight loss under different statistical rules, and lined up the same way the gap is four to six points.
What impressed me is further down the dose ladder. Retatrutide’s middle dose, 9 mg, lost 25.9 percent, beating tirzepatide’s maximum dose under either rule. The lowest dose, 4 mg, lost 19.0 percent, and fewer people quit it over side effects than quit placebo. Nobody has to climb to 12 mg to beat the drug I am on.
A McGill review in Annals of Internal Medicine, out the morning I finished this, pools 38 trials and gets 22.1 percent for retatrutide against 19.0 for tirzepatide, while saying outright the trials are too different to rank.
Lilly is running the trial that settles it. TRIUMPH-5 put about 800 adults on retatrutide or tirzepatide, double-blind, and finishes collecting data in November 2026.
Get the next one in your inbox
I cover the latest GLP-1 trials and access changes weekly. Free, no spam, unsubscribe whenever.
Does retatrutide cause less muscle loss than Zepbound?
No study has shown that, and the trial that could have settled it does not measure body composition at all.
The paper people reach for is Coskun et al., Lancet Diabetes & Endocrinology (2025), a DXA substudy of retatrutide’s Phase 2 diabetes trial. It found real fat loss, but its interpretation says “the proportion of lean mass loss to weight loss was similar to other obesity treatments.” That is reassurance retatrutide does not cost you more muscle, not a claim it costs you less. Tirzepatide’s own substudy (Look et al., 2025) reported roughly 75 percent fat and 25 percent lean “for both tirzepatide and placebo.”
TRIUMPH-5 will not close this either. It registers no DXA scan and no lean-mass measure of any kind. The levers for holding onto muscle have not changed: enough protein, enough resistance training, whichever molecule you take.
What are retatrutide’s side effects compared to Zepbound?
Both are gut drugs first: nausea, diarrhea, constipation, vomiting, worst while you climb the dose. Where they differ is dysesthesia, an odd skin sensation usually described as tingling or burning. In TRIUMPH-1 it hit 12.5 percent at the top dose against 0.9 percent on placebo; TRIUMPH-2 and TRIUMPH-3 came in lower. Not retatrutide’s private problem, though. Semaglutide at 7.2 mg produced it in 22.9 percent in Novo’s STEP UP trial.
Heart rate is the one I have a personal stake in, because I have an ascending aortic aneurysm. The 5 to 7 beats per minute figure circulating for retatrutide came from a transient peak at 24 weeks on the top dose in the Phase 2 trial (Jastreboff et al., NEJM, 2023) and declined after. Zhang et al. (2026), pooling 12 trials, put it at 3.46 beats over placebo against 2.05 for tirzepatide. About a beat and a half above the drug I inject.
On heart outcomes the gap is enormous and it favors Zepbound. SURPASS-CVOT returned a hazard ratio of 0.92 for tirzepatide against dulaglutide, and FDA approved a cardiovascular indication for Mounjaro on that basis in August. Retatrutide has nothing like it. TRIUMPH-3 counted 44 major cardiovascular events on drug against 52 on placebo, too few to rule out either benefit or harm, and TRIUMPH-Outcomes does not finish until February 2029.
How much will retatrutide cost compared to Zepbound?
Nothing yet, because there is nothing to price. Zepbound is the closest model. On LillyDirect with the savings card, on commercial insurance that denied the drug, the KwikPen runs $299 a month at 2.5 mg, $399 at 5 mg and $449 from 7.5 mg up, that last one only if you refill within 45 days. The card counts a month as 28 days and caps you at 11 fills a year, so it runs dry before December and the last fills revert to $699.
Where retatrutide lands depends partly on a fight over what it legally is. Lilly is in the Seventh Circuit arguing it should be licensed as a biologic rather than approved as a drug, twelve years of exclusivity instead of five, and it turns on a single amino acid. I wrote that up in my Zepbound will go generic, and Lilly is in court to make sure retatrutide doesn’t. Either way there is no cheap version of this molecule in the 2020s.
Can you get retatrutide now?
Not legally, unless you are in one of Lilly’s trials or clear its expanded access program, and almost nobody reading this will. That program is NCT07629401, and STAT reported its gate, attributed to a source familiar rather than to Lilly on the record: refractory obesity plus at least two serious obesity-related complications. Lilly said it was making “authentic” retatrutide available, a quiet admission about everything else sold under the name.
That has not slowed anyone down. A preprint searching the records of twenty-nine million Americans found 652 people who had taken retatrutide, and of those whose source was documented, 71 percent got it from an online vendor, a compounding pharmacy or a wellness clinic. Lilly has sued six sellers and flagged over 14,000 websites, which I covered in there’s no legal retatrutide to buy.
I understand the appeal and it is still a bad trade. The distance between retatrutide and the drug you can get with a prescription is a few percentage points, measured in people who got the real molecule in a verified vial. Almost everyone chasing grey-market retatrutide would lose a great deal of weight on Zepbound.
What I am doing about it
Staying on Zepbound. With the aneurysm I am the wrong guy to volunteer for the higher heart-rate number on a hunch, and two of retatrutide’s three planned indications are already on the label of the drug in my refrigerator.
TRIUMPH-5 finishes in November and sometime after that there will be a number worth arguing about. Retatrutide is going to be a very good drug. It is not going to be so much better than Zepbound that it is worth buying from a stranger to get it a year early.
More on this class on the GLP-1 pipeline tracker, and the full TRIUMPH-1 readout is in what TRIUMPH-1 just showed.