Summary: In adults with type 2 diabetes, this exploratory open-label randomised trial found that imeglimin 2000 mg/day and metformin 1000 mg/day produced a similar HbA1c reduction at 24 weeks and a comparable fall in post-challenge glucose excursion. The drugs differed mechanistically: insulin and glucose-dependent insulinotropic peptide (GIP) secretion rose only with imeglimin, whereas glucagon-like peptide-1 (GLP-1) increased with both. Glycaemic superiority was not demonstrated.
PICO Summary
| Element | Detail |
|---|---|
| Population | Adults with type 2 diabetes who were drug-naive or on a single oral hypoglycaemic agent (8-week washout before randomisation); single-centre, Japan. Exact randomised n not stated in the abstract. |
| Intervention | Imeglimin 2000 mg/day for 24 weeks, with OGTT at baseline, 12 and 24 weeks. |
| Comparison | Metformin 1000 mg/day for 24 weeks (open-label allocation). |
| Outcome | HbA1c reduction at 24 weeks was similar between arms and post-challenge glucose excursion fell comparably in both. Insulin levels increased only with imeglimin. Total and active GLP-1 increased in both arms; total and active GIP increased only with imeglimin. The imeglimin insulin rise correlated with GLP-1 at 12 weeks and with GIP at 24 weeks. The abstract reports no effect sizes, 95% confidence intervals, p-values, or ARR/NNT. |
Expert Commentary
This is an exploratory, single-centre, open-label randomised trial, and it should be read as mechanism-generating rather than practice-changing. The headline result is mechanistic, not glycaemic: HbA1c fell to a similar degree with both drugs and post-challenge glucose excursion was comparable, so no glycaemic superiority for imeglimin was shown. What distinguished imeglimin was an increase in insulin secretion accompanied by enhanced GIP secretion, alongside the GLP-1 rise seen with both agents. The earlier draft framing that imeglimin produced superior glucose control overstated the data and has been corrected. The chief limitation is the open-label, single-centre, exploratory design with no reported sample size, confidence intervals, or p-values in the abstract, which leaves the incretin findings hypothesis-level and vulnerable to expectation bias and small-sample noise. The correlation analyses are descriptive and cannot establish that GIP enhancement drives the insulin response. Can I use this with my patients? Not yet as a reason to prefer imeglimin over metformin, since glycaemic outcomes were equivalent; the incretin signal is interesting but not a treatment indication. A larger blinded trial with prespecified incretin endpoints and reported effect estimates is needed before these mechanistic differences inform drug selection.
References
Usui R, Hamamoto Y, Imura M, Omori Y, Yamazaki Y, Kuwata H, et al. Differential effects of imeglimin and metformin on insulin and incretin secretion: an exploratory randomized controlled trial. Diabetes Obes Metab. 2025;27(2):856-865. doi:10.1111/dom.16086
