Retatrutide's new Phase 3 data: diabetes, heart disease, and a catch

Two more trials, and a fifth win for Lilly

Retatrutide has been posting strong obesity numbers since 2023, but obesity-only trial populations tend to be healthier than the people who actually show up in a doctor's office asking about weight loss. On July 23, 2026, Lilly announced topline results from TRIUMPH-2 and TRIUMPH-3, the two Phase 3 trials built to test whether retatrutide still works once you add real comorbidities: type 2 diabetes in one trial, established cardiovascular disease in the other. Both trials met their primary weight-loss endpoint. That makes five positive Phase 3 readouts for retatrutide, following TRIUMPH-1 and TRIUMPH-4, and it clears a real hurdle: sicker patients don't always respond as well to a drug as trial-healthy volunteers do.

TRIUMPH-2: retatrutide in people who already have diabetes

TRIUMPH-2 enrolled 1,152 adults with obesity or overweight who also have type 2 diabetes, randomized to 4mg, 9mg, or 12mg of retatrutide, or placebo, over 80 weeks. Weight loss came in at 12.7% on the low dose, 19.1% on the middle dose, and 20.8% on the top dose, against 4.0% on placebo. A1C, which had a baseline around 7.7%, dropped by up to 1.6 percentage points more than placebo. Those numbers sit close to what tirzepatide's own diabetes trials have shown, which is notable mainly because it means the diabetes diagnosis itself didn't blunt retatrutide's effect the way some clinicians worried it might.

TRIUMPH-3: retatrutide in people with existing heart disease

TRIUMPH-3 is the bigger swing. It enrolled 1,949 adults with severe obesity (BMI 35 or higher) and established cardiovascular disease, with or without diabetes, randomized to 9mg or 12mg of retatrutide or placebo. Weight loss reached up to 22.6%, or about 55.8 pounds, at 80 weeks, against 3.2% on placebo. Lilly also reported meaningful drops in triglycerides (down 37.0% at the top dose), non-HDL cholesterol (down 16.5%), systolic blood pressure (down 9.3 mmHg), and waist circumference (down 7.5 inches). On paper, that is a strong cardiometabolic risk-factor profile.

The number that didn't move the way anyone hoped

Here is the part most headlines glossed over. TRIUMPH-3 tracked actual cardiovascular events, not just risk-factor markers, and the results were mixed. For the broader MACE-5 composite (all-cause death, heart attack, stroke, heart failure event, or coronary revascularization), retatrutide had 44 events against 52 on placebo, a hazard ratio of 0.82. That favors the drug, but the confidence interval (0.55 to 1.22) crosses 1.0, so it is not statistically significant. For the narrower MACE-3 composite (cardiovascular death, heart attack, stroke), retatrutide actually had more events than placebo, 27 versus 23, a hazard ratio of 1.12 with a confidence interval of 0.64 to 1.96. Neither number proves harm. Neither proves benefit either.

Compare that to semaglutide's SOUL trial, which was purpose-built with enough participants and enough time to detect a cardiovascular signal, and it did: a 14% cut in major cardiac events. TRIUMPH-3 was not designed or powered as a dedicated cardiovascular outcomes trial in that sense. It measured MACE as a safety endpoint alongside a weight-loss primary endpoint, with fewer events and less statistical power to detect a real effect either way. Lilly is running a separate, larger dedicated cardiovascular outcomes trial, TRIUMPH-Outcomes, that should eventually give a cleaner answer. Until that reads out, anyone claiming retatrutide protects the heart the way semaglutide does is getting ahead of the data.

Side effects looked familiar

Nothing new showed up on safety. In TRIUMPH-2, diarrhea hit 27.4% to 33.6% of participants across doses versus 13.2% on placebo, nausea ranged from 13.7% to 28.0% versus 8.0%, and constipation ran 14.0% to 16.8% versus 9.4%. Discontinuations from side effects ranged from 3.8% to 11.6% on the drug versus 4.9% on placebo. TRIUMPH-3's sicker, older population saw somewhat higher discontinuation rates, 9.8% to 13.5% versus 4.8% on placebo, which tracks with a population that has more going on medically and less tolerance for GI side effects while their doctors are also managing heart disease. None of this is a new signal. It is the same GI-side-effect profile every drug in this class has shown, just in a group of patients more likely to feel it.

Why the filing timeline moved

Lilly is now targeting a Q1 2027 filing, submitted as a Biologics License Application covering the combined obesity, diabetes, and cardiovascular data. That is later than the Q4 2026 New Drug Application timeline the company had floated earlier this year around the TRANSCEND-T2D-1 results. The switch from an NDA to a BLA reflects the broader package Lilly is now assembling, five trials instead of one or two, which takes longer to compile but gives the FDA a fuller picture of how the drug performs across different patient types. Add the standard 10-to-12-month review clock on top of a Q1 2027 filing, and an approval before late 2028 would be on the fast side of normal.

Where this leaves retatrutide today

Five positive Phase 3 trials is a genuinely strong package, and the weight-loss numbers in diabetes and cardiovascular populations answer a real question: does the drug still work once patients are sicker and more medicated? Yes, apparently. What it hasn't answered is whether retatrutide protects the heart beyond what weight loss alone would predict. That distinction matters for anyone weighing retatrutide against tirzepatide or wondering what the triple-agonist mechanism actually buys you. It is not approved. It is not available outside a trial. And on cardiovascular protection specifically, the honest answer right now is: not proven, come back after TRIUMPH-Outcomes.