Orforglipron heart safety results are reassuring for adults with type 2 diabetes and existing heart or kidney disease. ACHIEVE-4 met its planned safety standard against insulin glargine, a long-acting insulin injection. It did not establish fewer major heart events. The lower death rates are encouraging signals. They need confirmation before anyone says Foundayo helps people live longer.
What changed in orforglipron heart safety today?
As of September 30, 2026, the day this article was written, Lilly has released detailed ACHIEVE-4 results. The paper by Klein and colleagues in The Lancet (2026) appeared online the same day.
Lilly's April 16 first readout already reported the main safety estimate. Today's development is the detailed report and journal publication.
Foundayo is orforglipron's brand name. It acts on GLP-1, a gut hormone signal involved in fullness and blood sugar. Our Foundayo approval explainer covers the tablet's place among the different medication options.
The ACHIEVE-4 registry records 2,749 adults assigned by chance to orforglipron or insulin glargine. Everyone had type 2 diabetes, overweight or obesity, and established heart, blood-vessel or long-term kidney disease. The planned observation period ran up to roughly 104 weeks. The paper reports a median two-year follow-up: half were followed longer, half less.
The study was open-label, meaning people knew which medicine they received. Lilly funded it. These were people already facing substantial health risks, rather than otherwise healthy adults seeking weight loss.
What Foundayo versus insulin glargine showed
About 42 in every 1,000 people had a counted heart event on orforglipron, versus 50 in every 1,000 on glargine: 4.2% versus 5.0%. The four-part measure, called MACE-4, combined heart-related death, heart attack, stroke and hospitalization for unstable chest pain.
The hazard ratio, a comparison of how quickly events occurred during follow-up, was 0.84. Its 95% confidence interval, the statistical range around that estimate, was 0.59–1.20. That range includes no difference between treatments.
Non-inferiority means ruling out an increase in risk beyond a limit chosen beforehand. Here, the range's upper end had to stay below 1.8. It was 1.20. The safety test passed, but better heart outcomes were not established.
The three-event measure, MACE-3, left out hospitalization for chest pain. Its hazard ratio was 0.77 (95% confidence interval 0.52–1.13), again including no difference.
For someone comparing Foundayo with tirzepatide, the comparator matters: ACHIEVE-4 tested insulin, not tirzepatide. The SURPASS-CVOT results cover tirzepatide's separate heart study.
Why the death headlines need care
Lilly reported a 53% lower estimate for heart-related death: hazard ratio 0.47 (95% confidence interval 0.25–0.90). For death from any cause, the estimate was 57% lower: 0.43 (0.25–0.75).
Those analyses were planned beforehand. But the release's footnote says they were “not controlled for family-wise type 1 error.” In plain words, they lacked the statistical adjustment for checking many outcomes at once.
Checking more outcomes creates more opportunities for a chance finding. Even a striking result can therefore remain a clue to test further. These death findings suggest a possible benefit; they do not settle whether orforglipron extends life.
Blood sugar, weight and stopping treatment
A1C is a blood test reflecting average blood sugar over roughly three months. A change from 8% to 7% is one percentage point. Lilly reported these changes at 52 weeks:
| Measure | Orforglipron | Insulin glargine |
|---|---|---|
| A1C change, percentage points | −1.6 | −1.0 |
| Body-weight change | −8.8% | +1.7% |
These used an efficacy estimand, an estimate assuming people stayed on assigned treatment without adding blood-sugar medicines for more than 14 days. It allowed treatment interruptions or dose changes. It isn't the same as counting outcomes regardless of whether people continued treatment.
About one person in ten, 10.6%, stopped orforglipron because of a reported health problem during the minimum 52-week treatment period. Common problems included nausea, vomiting, diarrhea, decreased appetite and constipation. Passing a heart-safety test doesn't mean a medicine is easy for everyone to tolerate.
Foundayo versus oral semaglutide 25 mg
As of September 30, 2026, the day this article was written, Lilly's September 29 indirect comparison adds another comparison. An indirect comparison combines separate studies rather than assigning people to both treatments within one trial.
Its adjusted estimate favored orforglipron 17.2 mg by 1.5 percentage points of body weight (95% confidence interval −2.7 to −0.2). A1C fell 0.3 percentage points more (−0.6 to −0.1) at 52 weeks. These estimates assumed continued treatment without additional blood-sugar medication.
The analysis drew on ACHIEVE-3 and PIONEER PLUS. The direct ACHIEVE-3 trial, published in The Lancet in 2026, compared orforglipron with oral semaglutide 7 mg and 14 mg, not 25 mg.
Statistical adjustments cannot make different study groups one head-to-head trial. This analysis concerns blood sugar and weight, not heart protection. It also cannot rank compounded semaglutide preparations against Foundayo.
What this does not change for a prescription
As of September 30, 2026, the day this article was written, Foundayo's FDA label reads “Revised: 07/2026.” It covers long-term weight management in eligible adults alongside reduced-calorie eating and physical activity. It does not list treating type 2 diabetes or preventing major heart events as authorized uses.
The label still carries a boxed warning about thyroid C-cell tumors and cautions about pancreas inflammation and dehydration-related kidney injury. A reassuring heart result does not remove those risks.
Other medicines have their own evidence. SELECT tested branded semaglutide in people with existing heart or blood-vessel disease and no diabetes (Lincoff and colleagues, NEJM, 2023). The SELECT trial results explain that separate question.
Through Promise, semaglutide and tirzepatide are dispensed as compounded medications, which are different from FDA-approved products: the formulations offered here are not FDA-approved. FDA does not verify compounded drugs' safety, effectiveness or quality before marketing. These trials do not establish identical heart outcomes for a compounded formulation.
A licensed provider may still prescribe a compounded formulation; that decision is between the patient and the doctor.
What to watch next
As of September 30, 2026, the day this article was written, Lilly's EASD schedule lists the ACHIEVE-4 kidney presentation for October 1. That presentation is still upcoming. It is a separate question from whether the main heart-event comparison established a benefit.
Lilly is also conducting ATTAIN-OUTCOMES in adults with obesity or overweight and established heart or kidney disease, with or without type 2 diabetes. That broader trial addresses questions ACHIEVE-4 cannot settle for people without diabetes.
For a prescribing conversation, the useful questions are which product was studied, whether the person's history resembles the trial group, and what treatment goal matters. Current medicines and tolerability belong in that discussion too. A headline alone cannot decide whether a different prescription fits.
At Promise, a licensed U.S. provider reviews every request, and not everyone qualifies. The prescriber sets the actual dose.
