Eli Lilly's once-daily oral small-molecule GLP-1 agonist orforglipron beat oral semaglutide on HbA1c reduction and weight loss in a 52-week phase 3 trial of 1,698 adults with type 2 diabetes, while producing more gastrointestinal side effects and double the discontinuation rate. The head-to-head,…
Eli Lilly's once-daily oral small-molecule drug orforglipron beat oral semaglutide on both blood sugar control and weight loss in a phase 3 trial in adults with type 2 diabetes, while producing higher rates of gastrointestinal side effects. The results were reported in The Lancet and were the subject of a 10 July 2026 dispatch from the Qazinform News Agency written by correspondent Yerzhan Zhanibekov. The trial is the first major head-to-head comparison to position a non-peptide GLP-1 receptor agonist against the established oral peptide in the class, and the small molecule came out ahead on both efficacy endpoints.
Orforglipron is an oral tablet with a simpler storage and dosing profile than existing GLP-1 medicines, and its small-molecule format carries lower manufacturing costs. Those properties are the basis of the argument that a non-peptide GLP-1 agonist could widen access to a class of drugs that has been limited by biology, logistics, and price. The trial tests that argument in a direct contest with oral semaglutide, the leading oral peptide GLP-1 therapy.
The win is not unconditional. Gastrointestinal side effects were more frequent with orforglipron, and more patients stopped the drug because of adverse effects. The balance between superior efficacy and inferior gastrointestinal tolerability is now the central question for regulators, prescribers, and anyone developing the next generation of incretin medicines.
The trial tracked four core outcomes. Change in HbA1c, the standard measure of long-term glucose control; change in body weight; gastrointestinal adverse events including nausea, vomiting, diarrhea, and constipation; and discontinuation due to adverse effects.
Participants entered with an average baseline HbA1c of 8.3%, a starting point well above most treatment targets. With orforglipron, average HbA1c fell by 1.71% to 1.91% over the course of the trial, versus 1.47% with oral semaglutide. The gap, 0.24 to 0.44 percentage points, is clinically meaningful in a disease where the goal of therapy is often to bring patients below 7%. Average weight loss was 6.1 kg to 8.2 kg with orforglipron and 5.3 kg with oral semaglutide, a difference of 0.8 kg to 2.9 kg.
Tolerability is where the comparison inverts. About 59% of orforglipron participants reported gastrointestinal side effects, compared with 37% to 45% in the oral semaglutide group. About 10% of orforglipron participants discontinued treatment because of adverse effects, roughly double the comparator's rate. Those two numbers are large enough to offset some of the efficacy advantage, and they will shape how the drug is titrated, prescribed, and evaluated by regulators.
The trial enrolled 1,698 adults with type 2 diabetes and ran for 52 weeks across six countries. The indexed record on file at Peptide Atlas identifies the study as ACHIEVE-3, described in The Lancet on 21 March 2026 as a multinational, multicentre, non-inferiority, open-label, randomised, phase 3 trial PMID 41765029 . The news-agency dispatch that carried the results did not describe randomization, blinding, or dose arms, so the open-label design is an important detail for reading the data.
Open-label means participants and investigators knew which drug each person was taking. That is common in trials comparing different formulations, but it introduces the risk that expectations shape adherence and the reporting of subjective symptoms such as nausea. Randomization protects the comparison against confounding by baseline differences. What the design cannot provide is a blinded safety assessment, and the dispatch contains no confidence intervals or p-values, so the statistical consistency of the advantage across doses cannot be judged from it.
The non-inferiority framing matters as well. ACHIEVE-3 was designed to test whether orforglipron was not worse than oral semaglutide, not primarily to prove superiority. The reported results place orforglipron numerically ahead on both main endpoints, but the full dose-level analysis will determine how much of that advantage is consistent and generalizable. The six-country recruitment supports generalizability, but the dispatch does not report how participants were distributed across countries, ethnicities, or background therapies. Type 2 diabetes care varies by health system, and the size of the treatment difference could shift with the standard of care in each region.
GLP-1 is an incretin hormone released from the gut after meals. It binds the GLP-1 receptor on pancreatic beta cells, potentiating glucose-dependent insulin secretion, suppressing glucagon release, slowing gastric emptying, and acting on appetite circuits in the brain. Those coordinated effects are what produce the HbA1c reductions and weight losses seen across this drug class.
Semaglutide is a peptide GLP-1 receptor agonist. In its oral form it must be taken on an empty stomach and followed by a 30-minute wait before eating or drinking, a constraint that reflects the fragility of peptide drugs in the gastrointestinal tract. Its injectable presentations, Ozempic and Wegovy, avoid the gut but require injection and refrigeration. Orforglipron is a small molecule that activates the same receptor and is taken as a once-daily tablet with no fasting requirement and no cold chain. For patients, those differences translate directly into the daily logistics of taking the medicine.
Manufacturing is the second structural difference. Peptide drugs are assembled, purified, and formulated products that often need cold-chain distribution. Small molecules are produced by chemical synthesis using standard pharmaceutical infrastructure. The claim that orforglipron is less expensive and simpler to manufacture than peptide-based GLP-1 medicines, and therefore more viable for low- and middle-income countries, rests on that distinction.
The gastrointestinal side effects are not off-target toxicity. They are a direct result of GLP-1 receptor activation in the gut, where slowed motility produces nausea, vomiting, diarrhea, and constipation. The higher event rate with orforglipron and the higher discontinuation rate point to a steeper dose-response for tolerability. Whether the difference comes from higher plasma exposure, dose selection, or the kinetics of receptor activation is not established by the public summary.
The Peptide Atlas trial registry holds 668 registered clinical trials for semaglutide. Of those, one is Phase 3, four are Phase 2, and four are Phase 4, and ten are currently recruiting. Semaglutide's registered program reaches well beyond diabetes into psychiatry, oncology, cardiology, and pediatrics, as six notable trials show:
The registry is backed by 197 indexed PubMed papers on semaglutide. Recent additions include the STRIDE trial of semaglutide in peripheral artery disease and diabetes by baseline severity and age European Heart Journal, July 2026 , a randomized trial of semaglutide and effort-based decision-making in major depressive disorder JAMA Psychiatry, July 2026 , a systematic review and meta-analysis of long-term safety and renal outcomes in non-diabetic obesity with chronic kidney disease or hypertension Clin Ter, July 2026 , and a scoping review of retinal vascular events in semaglutide users Ophthalmologica, June 2026 . Peptide Atlas also holds eight third-party lab purity tests for semaglutide, with a highest observed purity of 99.979%.
Orforglipron's registered footprint is far smaller. The registry currently lists zero clinical trials for the drug, though it has 8 indexed PubMed papers. The zero count reflects the limits of registry coverage rather than an absence of clinical activity; the ACHIEVE program alone includes multiple multinational trials. The indexed literature includes the ACHIEVE-3 report itself The Lancet, 21 March 2026 , a meta-analysis of orforglipron's effects on cardiometabolic outcomes Cardiovascular Diabetology and Endocrinology Reports, 20 February 2026 , mechanistic work on G protein-biased agonism at the GLP-1 receptor Molecular Metabolism, March 2026 , a humanised GLP-1 receptor mouse model for translational drug development EBioMedicine, February 2026 , and a review of incretin modulation in diabetes Diabetes Therapy, March 2026 . Peptide Atlas maintains reference pages for both molecules at https://peptideatlas.co/peptides/semaglutide and https://peptideatlas.co/peptides/orforglipron.
In the days before its orforglipron dispatch, Qazinform also reported promising preclinical results for a separate vitamin B12-based compound that crosses the blood-brain barrier and targets glioblastoma tumors, a reminder that non-peptide drug development extends beyond metabolic disease.
For clinicians, the attractions of orforglipron are immediately legible. A once-daily oral tablet that needs no refrigeration and no empty-stomach routine removes two barriers that shape real-world adherence to GLP-1 therapy. The counterweight is that people who stop the drug because of nausea or vomiting cannot benefit from the better HbA1c and weight results. The higher discontinuation rate in the trial is the most important fact for estimating net clinical benefit in routine care, where titration schedules and patient counseling will determine whether the efficacy advantage survives contact with daily life. Clinical management will likely lean on gradual dose escalation, already the standard approach for GLP-1 drugs, but a daily oral tablet raises a question that injections do not: whether patients will keep taking a medicine that makes them feel ill when the consequence of stopping is a missed pill rather than a missed injection. Real-world cohorts, not registration trials, will answer that.
For researchers, the comparison sets a new benchmark. Efficacy is no longer the only axis on which small-molecule candidates are judged against peptide GLP-1 drugs; tolerability now carries equal weight. The open-label design of ACHIEVE-3 means the true difference in gastrointestinal symptoms is probably smaller than the reported gap, since patients and investigators knew the assignment. A blinded comparison with standardized adverse-event capture would be needed to quantify the real tolerability difference, and the humanised GLP-1 receptor mouse model already indexed for orforglipron research is one tool for studying the pharmacology behind it.
For the peptide supply chain, the result is a substitution risk at the margins. Oral semaglutide is the established oral peptide option, and its logistics are built around peptide production and stable formulation. If orforglipron reaches the market with a price that reflects small-molecule synthesis, it could draw patients from the oral peptide…
Peptides referenced: Semaglutide, Orforglipron, Glucagon, GLP-1.
Related reading: After Novo Unit Closure, AI Biotech Hires Expert for Parkinson's Stem Cell Review, Democratic Lawmakers Question If Trump Got Unapproved Obesity Drug, Assessing Retatrutide Supplier Trustworthiness in 2026, Amylin Makes a Comeback: The Forgotten Hormone Returns to Focus.