A once-daily oral small-molecule GLP-1 receptor agonist has demonstrated greater HbA1c and weight reductions than oral semaglutide in the first head-to-head phase 3 comparison of two agents in type 2 diabetes.

In the 52-week ACHIEVE-3 trial, published in The Lancet [link here], both doses of orforglipron met non-inferiority criteria for HbA1c reduction and were statistically superior to oral semaglutide, although tolerability trade-offs and long-term outcome data remain key considerations.

For Associate Professor Sof Andrikopoulos, CEO of Australian Diabetes Society, the results are encouraging but not transformative.

Prof Sof Andrikopoulos

“I think what we need to acknowledge is that both of these agents work,” he told the limbic. “They both provide very good diabetes management and also weight loss. It’s not as straightforward as saying one’s better than the other – they both work very well.”

The trial

In ACHIEVE-3, a phase 3 trial funded and designed by Eli Lilly, 1,698 adults with inadequately controlled type 2 diabetes on metformin were randomised to receive orforglipron or oral semaglutide for 52 weeks.

The paper was led by Dr Julio Rosenstock of the University of Texas Southwestern Medical Center, with co-authors including both academic investigators and Lilly employees.

Participants had a mean baseline HbA1c of 8.3%, mean BMI of 35.1 kg/m² and mean diabetes duration of 8.6 years and were randomised to orforglipron 12 mg or 36 mg once daily, or oral semaglutide 7 mg or 14 mg.

The primary endpoint was change in HbA1c at week 52, analysed using a treatment-regimen estimand with a non-inferiority margin of 0.3%.

“Orforglipron 12 mg and 36 mg doses outperformed oral semaglutide 7 mg and 14 mg diabetes-related doses on every
key endpoint we measured, including A1C and weight loss, with improvements appearing as early as four weeks and sustained throughout the study,” Dr Rosenstock said in a statement.

Glycaemic outcomes

From baseline, mean HbA1c reductions at week 52 were:

• -1.71% with orforglipron 12 mg
  
• -1.91% with orforglipron 36 mg
  
• -1.23% with semaglutide 7 mg
  
• 1.47% with semaglutide 14 mg

Both orforglipron doses met non-inferiority and demonstrated statistical superiority over semaglutide.

The absolute difference between the higher doses was approximately 0.4-0.5%.

Whether that margin translates into meaningful long-term benefit remains uncertain, Professor Andrikopoulos cautioned.

“For me, this study shows that they both work. It’s another tool at the disposal of clinicians,” he said.

“The study is designed to show superiority, but the question is whether it’s clinically superior in day-to-day practice. That’s not entirely clear.”

He said that while the statistical difference of around 0.4-0.5% is notable, it is unclear whether such an incremental reduction – on top of an already substantial 1.5% HbA1c fall – would deliver additional long-term clinical benefit.

“But any reduction in HbA1c is a good thing,” he added.

He noted that landmark trials such as The UK Prospective Diabetes Study have shown that each 1% reduction in HbA1c is associated with meaningful reductions in microvascular and macrovascular complications.

More patients on orforglipron reached commonly used glycaemic targets, with three-quarters achieving an HbA1c below 7% and more than two-thirds reaching 6.5% or lower at the highest dose.

Weight loss and cardiometabolic markers

Mean weight loss (treatment-regimen estimand) reached:

• -6.1% with orforglipron 12 mg
  
• -8.2% with orforglipron 36 mg
  
• -3.9% with semaglutide 7 mg
• -5.3% with semaglutide 14 mg

Reductions in systolic blood pressure, triglycerides, non-HDL cholesterol and waist circumference were numerically greater with orforglipron. Percentage reductions in urine albumin-to-creatinine ratio were also larger.

The trial was not designed to assess cardiovascular outcomes.

Professor Michael Horowitz

Describing the trial as ‘important’ in an accompanying comment in The Lancet [link here], Professor Michael Horowitz, Director of the Endocrine and Metabolic Unit at the Royal Adelaide Hospital and Professor Michael Nauck, Head of Clinical Research at the Diabetes Division, St Josef-Hospital, Germany, also urged caution in interpreting the findings.

“The outcomes of the current study should not be interpreted to indicate that orforglipron is, generally speaking, more efficacious than oral semaglutide,” the authors write.

They note that oral semaglutide has previously been studied at higher doses than those used in ACHIEVE-3, with HbA1c reductions of up to 2.0% and weight loss approaching 8.5% – effect sizes similar to those observed with orforglipron in the present trial.

The comment also highlights differences in gastrointestinal tolerability and pharmacokinetics. Whereas semaglutide has a long elimination half-life, orforglipron may produce more prominent daily peak concentrations, which could partly explain variations in nausea and vomiting patterns.

The authors suggest that real-world persistence may be lower than seen in trial settings, particularly given structured dose escalation and management of gastrointestinal symptoms.

In practical terms, they conclude, “Patients in need of a substantial improvement in glycaemia or bodyweight might prefer orforglipron and those with a higher priority for tolerability might favour oral semaglutide, with the difference regarding convenience (in favour of orforglipron treatment) in mind.”

Convenience versus pharmacology

Whether the observed advantage reflects intrinsic pharmacological potency or simply more consistent real-world use remains uncertain.

Unlike oral semaglutide, which must be taken fasting with a measured volume of water and followed by a delay before eating, orforglipron does not require food or fluid timing restrictions – a practical distinction that may influence adherence outside trial settings.

“It could be exactly that,” Professor Andrikopoulos said, when asked whether easier administration might partly explain the results.

“Absorption may also be more consistent because it’s a chemical compound rather than a peptide. You may be getting more reliable bioavailability compared to an oral peptide where only a small fraction is absorbed.”

Writing in their linked comment, Professor Horowitz and Professor Nauck said “only about 1% of orally administered semaglutide is absorbed and, reflecting this very low bioavailability, daily administration is required despite the long elimination half-life (∼1 week) of semaglutide”.

Safety and discontinuation

Gastrointestinal adverse events were common and reported in 58-59% of patients receiving orforglipron compared with 37-45% in the semaglutide groups.

Discontinuation due to adverse events occurred in 8.7-9.7% of the orforglipron groups versus 4.5-4.9% with semaglutide.

“I’m not concerned about that,” Professor Andrikopoulos said. “A 10% discontinuation rate in a study like this is actually quite low. The side effects are what we expect with a GLP-1 receptor agonist – nausea, vomiting, gastrointestinal upset.”

Pulse rate increases were modestly higher with orforglipron, although rates of serious adverse events and major adverse cardiovascular events were low and similar across groups.

Four deaths occurred during the study  one in the orforglipron 12 mg group, one in the orforglipron 36 mg group and two in the semaglutide 7 mg group. The publication does not indicate that these were considered treatment-related.

Where does it fit?

If approved in Australia, positioning remains an open question Professor Andrikopoulos said.

Injectable GLP-1 receptor agonists can deliver greater weight reductions in some patients, raising the possibility that injectables may be used for more substantial weight loss, with oral agents potentially playing a role in maintenance, he suggested.

“Injectables can produce 15-20% weight loss in some cases. So you might see injectables used for more substantial weight reduction, and oral agents used for maintenance.

The biggest challenge in weight loss isn’t losing the weight – it’s maintaining it. So oral agents may have a role in maintenance therapy, perhaps even following bariatric surgery or after intensive dietary intervention”.

He also pointed to potential cost and access implications.

“Manufacturing a small-molecule chemical should theoretically be cheaper than manufacturing a peptide,” he said. “In theory, those savings should benefit the government and ideally be passed on to patients”.

And not requiring cold-chain storage could make distribution easier, particularly in regional and remote areas, he added.

However, regulatory and reimbursement pathways will ultimately determine real-world uptake. Dedicated cardiovascular outcomes data are likely to influence longer-term positioning alongside established injectable GLP-1 therapies.

In a statement accompanying publication of the trial, Eli Lilly said it plans to submit orforglipron for regulatory approval in major markets, with applications expected this year. The company has previously indicated it intends to pursue indications in both type 2 diabetes and obesity

Professor Andrikopoulos, said the broader message is one of therapeutic evolution.

“We’re in an extraordinary period in diabetes management. We have so many effective agents available.

Both agents in this study showed excellent HbA1c reductions – around 1.5% or more – which is impressive. It’s a very exciting time to be working in diabetes care”.