Oral GLP-1 Pill Safiglipron Beats Standard Diabetes Drug on Blood Sugar Control
A phase 3 trial shows safiglipron, a no-fasting oral GLP-1 pill, outperforms dapagliflozin on HbA1c reduction in type 2 diabetes.
Summary
Safiglipron is a once-daily GLP-1 pill that can be taken without fasting or food restrictions. In a 32-week trial, 810 adults with type 2 diabetes who were not controlled on metformin received one of three safiglipron doses or the SGLT2 inhibitor dapagliflozin. All three doses lowered HbA1c at least as well as dapagliflozin, and the highest dose lowered it more (about 0.25 percentage points more). More people reached an HbA1c under 7% on safiglipron (55-64%) than on dapagliflozin (37%). Weight loss was similar, around 3.5-4% at the two higher doses. The trade-off was digestive side effects: nearly half of people on the top dose had nausea, and about 4% stopped treatment because of side effects, versus 1.5% on dapagliflozin. Participants were all in China and mostly men.
Detailed Summary
Type 2 diabetes affects an estimated 589 million adults worldwide, and many need a second drug on top of metformin as beta cell function declines. GLP-1 receptor agonists and SGLT2 inhibitors are the preferred add-ons, but injectable GLP-1s can limit adherence, and oral semaglutide requires fasting dosing with restricted water. Safiglipron (HRS-7535) is an oral, non-peptide, small-molecule GLP-1 receptor agonist taken without fasting or dietary restrictions. OUTSTAND-2 is described as the first double-blind phase 3 trial comparing such a drug against an SGLT2 inhibitor.
The trial enrolled 810 adults aged 18-75 at 98 sites in China, all with HbA1c of 7.5-11.0% on stable metformin. Participants were randomized in a double-dummy design to safiglipron 30 mg (n=202), 60 mg (n=203), 90 mg (n=203) or dapagliflozin 10 mg (n=202) for 32 weeks, followed by a 20-week extension. Mean age was 52.2 years, 35.7% were women, median diabetes duration was 5.0 years, mean HbA1c was 8.60% and mean BMI was 27.1 kg/m2.
The primary endpoint was the change in HbA1c at week 32. Least-squares mean reductions were 1.58%, 1.50% and 1.68% with safiglipron 30, 60 and 90 mg versus 1.28% with dapagliflozin. Differences were -0.30%, -0.22% and -0.40%, establishing non-inferiority for all doses against a 0.4% margin. Under the treatment policy estimand, the 90 mg dose was superior (difference -0.25%; 95% CI -0.45 to -0.05; one-sided P=0.0067). The 60 mg dose was not (-0.14%; P=0.0789), so fixed-sequence testing stopped.
Secondary results favored safiglipron on glycemic targets. HbA1c below 7.0% was reached by 54.9-63.5% versus 36.5% with dapagliflozin, and 6.5% or lower by 38.9-48.6% versus 16.1%. Fasting glucose fell 2.23-2.59 mmol/L versus 2.00. Weight loss was -2.35%, -3.55% and -4.17% versus -3.60% for dapagliflozin. HOMA-beta rose 77-98% versus 11.5%, while HOMA-IR fell less than with dapagliflozin (16-22% versus 38.5%). Exploratory results showed 90 mg lowered LDL cholesterol 2.7%, whereas dapagliflozin raised it 2.9%. Both drugs reduced systolic blood pressure and albuminuria.
Gastrointestinal events were the main cost. Nausea occurred in 22.8%, 38.1% and 46.8% of participants on safiglipron versus 6.9% on dapagliflozin, and vomiting in 13.4%, 29.7% and 34.0% versus 2.0%. Adverse events led to discontinuation in about 4% of safiglipron participants versus 1.5%. Level 2 hypoglycemia occurred in 1.5-2.0% versus none, with no severe episodes. Lipase elevations were more common (15.8-18.3% versus 4.5%), and one acute pancreatitis case occurred at 60 mg. One death from myocardial infarction at 90 mg was judged unrelated. No Hy's law cases or genital mycotic infections were reported.
Several caveats apply. The population was entirely from China, about 64% male and relatively lean, which limits generalizability. The primary endpoint was at 32 weeks, and no cardiovascular or kidney outcomes were assessed. Because fixed-sequence testing stopped at the 60 mg superiority comparison, later secondary findings are not confirmatory. Completion was lower with safiglipron (82-88%) than dapagliflozin (93%). Safiglipron was compared only with dapagliflozin, not with other GLP-1 drugs.
Key Findings
- HbA1c fell 1.58%, 1.50% and 1.68% with safiglipron 30/60/90 mg versus 1.28% with dapagliflozin; all three doses were non-inferior (margin 0.4%, all P<0.0001)
- The 90 mg dose was superior under the treatment policy estimand (difference -0.25%; 95% CI -0.45 to -0.05; one-sided P=0.0067); 60 mg was not (P=0.0789)
- HbA1c <7.0% was reached by 54.9-63.5% on safiglipron versus 36.5% on dapagliflozin; HbA1c ≤6.5% by 38.9-48.6% versus 16.1%
- Body weight fell 2.35%, 3.55% and 4.17% with safiglipron versus 3.60% with dapagliflozin; the difference at 90 mg was not significant (-0.56%; 95% CI -1.37 to 0.24)
- HOMA-beta rose 76.9-98.5% with safiglipron versus 11.5% with dapagliflozin, but HOMA-IR fell less (16-22% versus 38.5%)
- Nausea occurred in 22.8%, 38.1% and 46.8% on safiglipron versus 6.9% on dapagliflozin; vomiting in 13.4-34.0% versus 2.0%
- Treatment discontinuation due to adverse events was 4.0%, 4.0% and 3.9% versus 1.5%, with gastrointestinal events driving most dropouts (3.0-3.5% versus 0%)
Methodology
OUTSTAND-2 was a multicenter, randomized, double-blind, double-dummy, active-comparator phase 3 trial at 98 sites in China, enrolling 810 adults on stable metformin with HbA1c 7.5-11.0%. It included a 3-week run-in, a 32-week core period, a 20-week extension and a 2-week safety follow-up. The primary endpoint (week 32 HbA1c change) was analyzed with a non-inferiority estimand (0.4% margin) and a treatment policy estimand for superiority, using mixed models and a fixed-sequence multiplicity strategy. Sensitivity analyses used multiple imputation and tipping-point methods.
Study Limitations
All participants were enrolled in China and about 64% were men, with a mean BMI of 27 kg/m2, which limits generalizability. Superiority testing stopped at the 60 mg comparison, so later key secondary findings are not confirmatory, and the primary endpoint covered only 32 weeks with no cardiovascular or renal outcomes. Dropout was higher with safiglipron than dapagliflozin. Funding and conflict-of-interest disclosures were not included in the supplied text and should be checked in the full publication.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe:
