What is orforglipron?

A once-daily pill in the GLP-1 family of diabetes and weight-loss drugs. Unlike most drugs in that family it is not a peptide, so it can be swallowed as an ordinary tablet or capsule. It was approved in the United States for weight management in April 2026.

What did the trial find about the heart?

Cardiovascular death, heart attack, stroke or hospital admission for unstable angina occurred in 57 of 1,358 people taking orforglipron and 67 of 1,343 taking insulin glargine. That was enough to show the pill was not worse. It was not enough to show it was better.

Did fewer people die on the pill?

Yes: 1.4% against 3.2%. The trial was not designed to test deaths as a main outcome, and that comparison was not adjusted for the number of comparisons made, so it is a signal to be checked in a larger trial.

What were the side effects?

Nausea, vomiting and diarrhea were the main ones. Stomach and bowel side effects were reported by 62.1% of people on the pill, against 14.2% on insulin, and 8.9% stopped the pill because of them. Low blood sugar was less common on the pill, 6.8% against 19.2%.

What are the limits of the trial?

Participants and doctors knew which treatment was given, the comparison was with insulin and not a placebo, and the drug's maker paid for the study. Which diabetes treatment suits a particular person is assessed by their doctor. This is general information rather than medical advice.

Orforglipron pill was no worse for the heart than insulin in a 2,749-person diabetes safety trial

Major heart events occurred in 4.2% of people on the daily GLP-1 pill and 5.0% on insulin glargine over two years. The company-funded trial was built to show safety, not benefit, and six in ten on the pill had stomach or bowel side effects.

Every new diabetes drug has to answer a question before regulators will accept it: does it harm the heart? For orforglipron, a daily pill from the same family as the injectable drugs now widely used for diabetes and obesity, the answer is now in. A two-year trial published in The Lancet in September found that it was no worse for the heart than insulin.

The trial was built to show exactly that and no more. Whether the pill protects the heart, as some of the injectable drugs do, is a separate question that a larger trial is still running to answer. In the meantime the results offer a clear view of what the pill does to blood sugar and weight, and of how many people could not tolerate it.

Why a new diabetes pill needs a heart safety trial

The injectable drugs in this class copy a gut hormone and are built from chains of amino acids, called peptides, which stomach acid destroys. A trial report written last year, before any approval, described orforglipron as a small-molecule, nonpeptide glucagon-like peptide-1 (GLP-1) receptor agonist in clinical development for type 2 diabetes and weight management. Being a small molecule, it survives digestion and can be taken as an ordinary pill. Since then, the FDA approved orforglipron for chronic weight management in April 2026. It is sold as Foundayo.

The older injectable drugs have a strong record on the heart. A pooled analysis of their outcome trials begins from the settled point that GLP-1 receptor agonists reduce major adverse cardiovascular events (MACE) in patients with type 2 diabetes. One member of the class already comes as a tablet, and it has been tested against placebo. In that trial of 9,650 people, a major cardiovascular event occurred in 579 of the 4825 participants given the semaglutide tablet, or 12.0%, and in 668, or 13.8%, of the same number given placebo.

A chemically different drug cannot borrow that record. The new paper says so at the outset. While some peptide GLP-1 receptor agonists have established cardiovascular benefit, the cardiovascular safety of non-peptide GLP-1 receptor agonists has not been studied.

How the orforglipron safety trial was designed

The study, a trial called ACHIEVE-4, was conducted in 317 sites across 16 countries and territories. Its participants had type 2 diabetes that was not well controlled on up to three other medicines, and they were chosen for being at high risk. All participants had established cardiovascular or chronic kidney disease. Their average age was 63.

In all, 2749 participants were randomly assigned (1371 to orforglipron and 1378 to insulin glargine). Orforglipron was raised step by step to the highest dose each person could tolerate. Insulin glargine was titrated to a fasting glucose below 100 mg/dL, titrated meaning adjusted to reach a target. The two treatments were each administered once daily, one as a pill and one as an injection, so there was no hiding which was which. The trial was open-label.

The choice of insulin as the comparison was deliberate. Rachel Batterham, a senior executive at the drug’s maker, Eli Lilly, explained it to Healio. A placebo was not an option in this trial, Batterham said, so “we used glargine, which has been proven not to have an adverse effect, as the comparator”.

That proof comes from a large trial a decade and a half ago, in which researchers randomly assigned 12,537 people to insulin glargine or standard care and followed them for six years. Its conclusion was that insulin glargine had a neutral effect on cardiovascular outcomes and cancers.

The measure of safety was a combination of four events: death from cardiovascular causes, a heart attack, a stroke, or a hospital admission for unstable angina, which is pain in the chest that signals a threatened heart attack. Stroke is counted with the heart events, as is usual in trials of this kind. The test was one of non-inferiority. The pill would pass if the data ruled out an event rate more than 80% higher than with insulin. That is a wide margin.

Results were counted for everyone who took at least one dose. The outcomes were assessed in all participants who took at least one dose of assigned treatment, whether or not they kept taking it. That left 2,701 of the 2,749 in the count: 1,358 on the pill and 1,343 on insulin.

Heart events with the pill and with insulin

Over a median of two years	Orforglipron	Insulin glargine
Any of the four events	57 of 1,358 (4.2%)	67 of 1,343 (5.0%)
Deaths from any cause	1.4%	3.2%
Low blood sugar, clinically important or severe	6.8%	19.2%
Stomach and bowel side effects	62.1%	14.2%
Stopped treatment because of side effects	15.4%	5.4%

Over a median follow-up of 2 years, one of the four events happened to 57 people on orforglipron and 67 on insulin. The rate was 16% lower with the pill, with a plausible range from 41% lower to 20% higher.

That range easily clears the safety test, since even its worst end is nowhere near 80% higher. It does not establish a benefit, because it also includes no difference at all. The authors state the first conclusion and not the second. In their words, the cardiovascular safety of orforglipron was confirmed by demonstrating non-inferiority to insulin glargine.

A narrower measure, leaving out angina admissions, pointed the same way without settling anything. Orforglipron did not achieve superiority for 3-component events: the rate was 23% lower, with a range from 48% lower to 13% higher.

Klara Klein of the University of North Carolina, who led the trial, was explicit about the limits. “The study was not designed to demonstrate cardiovascular benefit,” Klein told Pharmacy Times, adding that the data cannot show whether the pill reduces heart events. Klein has received consulting fees and research grants from Lilly and other drug companies.

Fewer deaths on orforglipron, with a caveat

One figure in the results is striking. During the trial, 1.4% of patients in the orforglipron group and 3.2% of those in the insulin glargine group died.

The company has put that result forward. Its announcement says that in pre-planned analyses, the risk of cardiovascular death was 53% lower and the risk of death from any cause 57% lower. The same announcement carries a footnote on those analyses that reads “Not controlled for family-wise type 1 error”, which is statistical language for a result that was not protected against the chance results that arise when many outcomes are examined.

A trade report on the paper made the same point: all-cause death was not controlled for multiplicity. A difference in deaths that large, in a trial where the main heart outcome showed no clear difference, is the sort of result that needs to be seen again before it is believed. The comparison was also with insulin, which causes weight gain and low blood sugar, not with a placebo.

Blood sugar, weight and side effects of the pill

On blood sugar and weight, it outperformed insulin. Lilly reports that at one year the average blood sugar marker, HbA1c, had fallen by 1.6 percentage points with the pill and 1.0 with insulin. Body weight fell by 8.8% with the pill and rose by 1.7% with insulin.

Low blood sugar was much less common. The rates of clinically important or severe episodes were 6.8% in the orforglipron group and 19.2% of the insulin glargine group.

The cost was to the gut. Gastrointestinal adverse events were the most frequent adverse event and most common reason for orforglipron treatment discontinuation, gastrointestinal meaning of the stomach and bowel. Such events occurred in 62.1% of the orforglipron group and 14.2% of the insulin glargine group, led by nausea (31.9%), vomiting (21.0%), and diarrhea (20.4%). Most were described as mild to moderate and came early, while the dose was being raised.

A sizable minority gave up. The share who stopped treatment because of side effects of any kind was approximately 15.4% vs 5.4%. For stomach and bowel effects specifically it was 8.9% in the orforglipron group and 0.2% in the insulin glargine group.

There was also a small effect on the pulse. Pulse rate rose by a mean of about 4.0 beats per minute with orforglipron.

What the safety trial does not show about the heart

Benefit. The trial was sized to rule out harm. A definitive answer on protection is expected from a second trial, which Lilly describes as an ongoing Phase 3 study in adults with established cardiovascular or kidney disease. Klein put it this way: “We await the results of the ATTAIN-Outcomes trial, which is powered for superiority to determine whether or not there is cardiovascular superiority.”

How it compares with a placebo. Insulin was a reasonable stand-in for a neutral treatment. It is still an active drug with effects of its own on weight and blood sugar lows, both of which went against it here.

Blinding. Everyone knew who was taking a pill and who was injecting insulin. Decisions about hospital admission, one of the four events counted, can be influenced by that knowledge.

Two years. Some heart outcome trials in this class have run longer. The tablet form of semaglutide was followed for about four years.

Who paid. The trial was funded by Eli Lilly and Company, which makes the drug.

Diabetes use. In the United States the pill’s approval is for weight management. The authors describe it as a potential once-daily, oral treatment option for diabetes.

Which diabetes medicine is right for a particular person depends on their heart, kidneys, weight and tolerance of side effects, which is assessed by their doctor. Do not stop or change a prescribed medicine without talking to your doctor.

In a company-funded trial of 2,749 adults with type 2 diabetes and heart or kidney disease, the daily pill orforglipron was followed by major heart events in 4.2% of people over two years against 5.0% on insulin glargine, enough to show it was not worse for the heart but not that it was better, with stomach and bowel side effects in six of every ten people who took it.