A new once-daily GLP-1 pill lowered blood sugar — with modest weight loss

Safiglipron, an oral drug that mimics the gut hormone GLP-1, lowered blood sugar more than placebo and held its own against a widely used diabetes pill in two phase 3 trials published together in Nature Medicine [s1][s2]. The results add another candidate to the fast-moving race to put GLP-1 therapy in a daily tablet rather than a weekly injection. They also come with the caveats that should travel with any early GLP-1 story: both trials were run entirely in China, both were funded by the drug's maker, and the weight loss was modest compared with the injectable drugs that have reshaped expectations.

GLP-1 receptor agonists such as semaglutide and tirzepatide are potent, but most are injected, and the oral versions approved or in late testing are either peptides needing strict fasting rules or newer small molecules. Safiglipron is a small-molecule agonist taken once daily without fasting or dietary restrictions, which is the practical selling point its developer, Jiangsu Hengrui Pharmaceuticals, is pursuing [s1].

OUTSTAND-1: against placebo

The first trial, OUTSTAND-1, enrolled 284 adults with relatively early type 2 diabetes — a median diabetes duration of 1.6 years and an average HbA1c of 7.95% — who were managing on diet and exercise alone across 46 Chinese sites [s1][s3]. Participants were randomly assigned to safiglipron 30 mg, 60 mg or 90 mg, or placebo, once daily for 32 weeks, with a 20-week extension [s1].

On the main measure, the fall in HbA1c at 32 weeks, all three doses beat placebo clearly: the placebo-adjusted reductions were 1.22%, 1.20% and 1.45% for the 30, 60 and 90 mg doses respectively, each highly statistically significant [s1]. HbA1c is a blood marker of average glucose over the preceding months, and reductions of that size are comparable to what injectable GLP-1 drugs achieve in similar patients. Weight change, by contrast, was slight: differences versus placebo of 0.65%, 2.23% and 3.56% by dose [s1]. That is far below the double-digit losses seen with the leading injectables, though this was a glucose-lowering trial in people who were not selected for obesity.

OUTSTAND-2: against dapagliflozin

The larger trial, OUTSTAND-2, tested safiglipron against dapagliflozin — an SGLT2-inhibitor pill already in routine use — in 810 adults with longer-standing diabetes (median five years, average HbA1c 8.60%) across 98 Chinese sites [s2]. All three safiglipron doses were non-inferior to dapagliflozin on HbA1c at 32 weeks, and the 90 mg dose met the pre-set bar for superiority (a further 0.25% reduction beyond dapagliflozin; one-sided p=0.0067), while the 60 mg dose did not (p=0.079), at which point the formal testing sequence stopped [s2].

Target attainment favoured safiglipron: 54.9% to 63.5% of safiglipron-treated participants reached an HbA1c below 7.0%, versus 36.5% on dapagliflozin [s2]. Weight loss ranged from 2.35% to 4.17% across safiglipron doses, against 3.60% for dapagliflozin — so on weight the two were broadly similar rather than safiglipron pulling ahead [s2].

Tolerability and the limits

As with the whole class, the side effects were mostly gastrointestinal — nausea and related symptoms — and mostly mild to moderate, but they rose with dose [s1][s2]. In OUTSTAND-1, treatment was stopped for adverse events in 1.4%, 2.9% and 6.9% of the 30, 60 and 90 mg groups versus none on placebo, so the most effective dose also drove the most dropouts [s1]. In OUTSTAND-2, discontinuation for adverse events ran around 4% across safiglipron doses versus 1.5% on dapagliflozin [s2].

Three limits deserve weight. First, both trials enrolled only Chinese adults, so how the drug performs across other populations and diets is untested [s1][s2]. Second, both were funded by the manufacturer, which does not invalidate the data but is the standard reason to wait for independent replication before treating a single programme's results as settled [s1][s2]. Third, these were 32-week glucose trials: they say nothing yet about the outcomes that matter most over years — cardiovascular events, kidney protection or durable weight control — for which GLP-1 drugs are increasingly prescribed.

What it means

Safiglipron looks like a genuine addition to the oral GLP-1 field: a once-daily tablet, no fasting rules, solid HbA1c lowering and glycaemic control at least as good as an established comparator. What it is not, on this evidence, is a weight-loss drug to rival the injectables, nor a therapy with proven long-term benefit. For patients, the practical question — whether an effective GLP-1 can be taken as a convenient pill — is edging towards yes; the harder questions of who benefits most, at what dose given the side-effect trade-off, and with what effect on hard outcomes remain open.

Sources

• [s1] Oral small-molecule GLP-1 receptor agonist safiglipron in early type 2 diabetes (OUTSTAND-1). Nature Medicine, 29 Sep 2026. https://doi.org/10.1038/s41591-026-04651-9
• [s2] Oral small-molecule GLP-1RA safiglipron versus dapagliflozin in type 2 diabetes (OUTSTAND-2). Nature Medicine, 5 Oct 2026. https://doi.org/10.1038/s41591-026-04713-y
• [s3] A Study of Safiglipron in Participants With Type 2 Diabetes (OUTSTAND-1), ClinicalTrials.gov NCT06672172. https://clinicaltrials.gov/study/NCT06672172