Background <p>In women with prediabetes after prior gestational diabetes mellitus (GDM), whether intermittent metformin treatment as needed is as effective as continuous use in preventing type 2 diabetes mellitus (T2DM) is unknown. We aimed to test whether an intermittent metformin treatment is non-inferior to continuous metformin strategy for prevention of T2DM in women with prediabetes after GDM.</p> Methods <p>A randomized, open-label, non-inferiority trial was conducted at three centers in China. Women aged 18–45 years with prior GDM and prediabetes at 4–12 weeks postpartum were randomly assigned to receive lifestyle intervention plus intermittent or continuous metformin treatment. In the continuous group, participants were initially treated with lifestyle intervention and metformin was initiated when lifestyle intervention alone failed to maintain euglycemia and continued without interruption throughout the trial. In the intermittent group, participants were initially treated with lifestyle intervention and metformin was added when they failed to maintain euglycemia and was discontinued if the glycemic status returned to normal. The primary outcome was the difference in incidences of newly diagnosed diabetes between the two groups during the 3-year follow-up, compared against a non-inferiority margin of 10-percentage-point by following the intention-to-treat principle. Secondary outcomes included the percentage of prediabetes remission, changes from baseline in body weight, body mass index, waist circumference, fasting plasma glucose, 2-h plasma glucose, and hemoglobin A1c.</p> Results <p>Among the 376 women randomized, 11.5% of women in the intermittent group and 11.1% in the continuous group developed newly diagnosed T2DM after three years (absolute risk difference 0.4% points [95% CI −6.5 to 7.4], meeting criteria for non-inferiority). None of the secondary outcomes were significantly different between the two groups. Women in the continuous group experienced higher rate of metformin-related vitamin B₁₂ deficiency (13.5% vs. 3.7%, <i>P</i> = 0.001).</p> Conclusions <p>In postpartum women with prediabetes after GDM, a strategy of intermittent metformin as needed plus lifestyle intervention met the pre-specified 10-percentage-point non-inferiority margin for 3-year T2DM incidence compared with continuous metformin plus lifestyle intervention. The equivalence of the two metformin strategies needs to be confirmed in further trials with a narrower margin and a larger event count.</p> Trial registration <p>The study was registered in the Chinese Clinical Trial Registry (ChiCTR-IOR-17012770).</p>