Bayesian network meta-analysis ranking anti-prediabetes drugs (metformin, acarbose, pioglitazone, GLP-1 RAs including semaglutide, SGLT2 inhibitors) for preventing T2DM progression in prediabetes. Synthesizes RCT evidence on diabetes incidence prevention, weight reduction, and adverse event profiles across all available agents. Provides an evidence-based drug ranking for prediabetes pharmacological intervention—informing the emerging indication for semaglutide in prediabetes prevention as a high-value early intervention point in the cardiometabolic disease continuum.
Abstract
BACKGROUND: Prediabetes refers to the transitional stage from normal glucose metabolism to diabetes. The International Diabetes Federation guidelines reported that, as of 2024, approximately 1.12 billion people globally were in the prediabetes stage. Without intervention, individuals with prediabetes are highly likely to progress to type 2 diabetes mellitus. It can be seen that prediabetes is posing a threat to human health and life and leads to a significant global public health concern.
METHODS: PubMed, Embase, Cochrane Library, Web of Science, and ClinicalTrials.gov were searched before March 29, 2025. Eligible randomized controlled trials (RCTs) enrolled adults with prediabetes, compared the efficacy and safety of placebo and anti-prediabetic drugs (e.g., metformin, sodium-glucose cotransporter 2 inhibitors, glucagon-like peptide-1 receptor agonists, and thiazolidinedione) with a follow-up duration of at least 12 weeks. Bayesian network meta-analysis was employed in statistical analysis.
RESULTS: Fifty-five eligible RCTs involving 37 interventions with 16,610 participants were included in this study. Compared with placebo, most anti-prediabetic drugs significantly reduced levels of hemoglobin A1c (HbA1c) (mean difference (MD), - 0.94 ~ - 0.27%), fasting plasma glucose (FPG) (MD, - 26.42 ~ - 0.15 mg/dL), weight loss (WL) (MD, - 13.59 ~ - 5.99 kg) and body mass index (BMI) (MD, - 4.50 ~ - 0.08 kg/m). Specifically, 2.4 mg of semaglutide SC demonstrated the most optimal efficacy in WL (MD - 13.59 kg; 95% confidence interval (CI) - 17.30 to - 9.91) and favorable efficacy in lowering HbA1c (MD - 0.39%; 95% CI - 0.55 to - 0.25); 15 mg of tirzepatide showed significant efficacy in lowering FPG (MD - 9.58 mg/dL; 95% CI - 12.00 to - 7.15), and potent efficacy in lowering BMI. Thirty milligrams of pioglitazone showed excellent efficacy in lowering lipid and FPG. Among the interventions, there was no significant difference in the incidence of adverse events (AEs), while 100 mg of sitagliptin demonstrated higher incidence of serious adverse events (SAEs).
CONCLUSIONS: Among all the included interventions, GLP-1RAs, GIP/GLP-1RAs, and TZDs demonstrated favorable anti-prediabetic efficacy and acceptable safety. 2.4 mg of semaglutide SC and 15 mg of tirzepatide were the best option among the included interventions considering favorable glucose and BMI control.
SYSTEMATIC REVIEW REGISTRATION: PROSPERO CRD42025636991.