ACHIEVE-3: Oral Orforglipron Beat Oral Semaglutide on Blood Sugar and Weight in Head-to-Head Phase 3 Diabetes Trial

Eli Lilly and Company posted results to ClinicalTrials.gov on September 15, 2026 from ACHIEVE-3, a phase 3 trial comparing its oral GLP-1 receptor agonist orforglipron head-to-head with oral semaglutide in adults with type 2 diabetes inadequately controlled on metformin. The randomized, open-label trial (NCT06045221) is one of several studies in Lilly's ACHIEVE program evaluating orforglipron for a diabetes indication; it was completed in August 2025, and its results were first posted to the registry more than a year later.

Orforglipron is a non-peptide small molecule that activates the GLP-1 receptor and is taken as a once-daily tablet. It is FDA-approved, under the brand name Foundayo, for chronic weight management in adults with obesity or overweight with weight-related conditions, but it does not yet hold FDA approval for type 2 diabetes. Oral semaglutide is FDA-approved as Rybelsus for type 2 diabetes at doses up to 14 mg daily. ACHIEVE-3 tested orforglipron against this already-approved oral comparator rather than against a placebo.

The trial enrolled 1,698 adults across sites in the United States, Japan, China, Mexico, Puerto Rico and Argentina, with a mean baseline age in the low-to-mid 50s and roughly even numbers of men and women. Participants were randomized in four groups to reach a targeted daily dose of 12 mg or 36 mg orforglipron, or 7 mg or 14 mg semaglutide, with doses titrated upward from a 1 mg or 3 mg starting dose every four weeks over the 52-week treatment period.

The primary endpoint was change in HbA1c from baseline to week 52, tested under a prespecified statistical hierarchy for both non-inferiority (margin of 0.3 percentage points) and superiority. Mean HbA1c fell 1.91 percentage points with 12 mg orforglipron and 2.16 points with 36 mg orforglipron, compared with 1.11 points with 7 mg semaglutide and 1.45 points with 14 mg semaglutide. Every orforglipron-versus-semaglutide comparison met both the non-inferiority margin and superiority testing, with least-squares mean differences ranging from -0.46 to -1.04 percentage points and a p-value below .001 for each comparison, according to the statistical analyses posted with the results.

Body weight, a secondary endpoint, also fell more on orforglipron. Average weight loss was 6.6 kg (6.7%) with 12 mg orforglipron and 8.9 kg (9.2%) with 36 mg orforglipron, compared with 3.6 kg (3.7%) with 7 mg semaglutide and 5.0 kg (5.3%) with 14 mg semaglutide. The proportion of participants losing at least 15% of body weight was 12.0% and 23.2% on the two orforglipron doses, versus 5.0% and 6.5% on the two semaglutide doses.

Gastrointestinal adverse events were more common with orforglipron. Nausea was reported by 112 of 424 participants (26.4%) on 12 mg orforglipron and 116 of 423 (27.4%) on 36 mg orforglipron, compared with 56 of 426 (13.1%) on 7 mg semaglutide and 90 of 425 (21.2%) on 14 mg semaglutide; diarrhea and vomiting followed a similar pattern. Serious adverse events were reported in 21 of 424 participants on 12 mg orforglipron, 40 of 423 on 36 mg orforglipron, 19 of 426 on 7 mg semaglutide and 24 of 425 on 14 mg semaglutide. Deaths were reported in 1 participant on each orforglipron dose and 2 participants on 7 mg semaglutide, with none on 14 mg semaglutide; the posted results do not attribute these deaths to study treatment.

ACHIEVE-3 is sponsored by Eli Lilly, orforglipron's manufacturer, and the results were posted directly to the ClinicalTrials.gov registry rather than published or presented at a medical meeting as of this writing. The registry posting does not include a full peer-reviewed manuscript, and Lilly has not filed for an expanded diabetes indication for orforglipron based on these results as of this writing. The findings describe average outcomes in this specific trial population over 52 weeks and are not dosing or treatment guidance; orforglipron remains approved in the United States only for chronic weight management, and any decision about diabetes treatment should be made with a qualified clinician.