Effectiveness of GLP-1 RAs and SGLT2 Inhibitors in preventing T2DM in high-risk patients: An updated Systematic Review and Meta-Analysis

Georgios I Tsironikos¹Vasiliki Tsolaki^2*George E Zakynthinos³Despoina Kyprianidou⁴Vasiliki Rammou⁵Thomas Antonogiannis⁵Theodoros Mprotsis⁶Epaminondas Zakynthinos²Alexandra Bargiota²

• ¹Panepistemio Ioanninon, Ioannina, Greece
• ²Panepistemiako Geniko Nosokomeio Larisas, Larissa, Greece
• ³Geniko Nosokomeio Nosematon Thorakos Athenon E Soteria, Athens, Greece
• ⁴Ethniko kai Kapodistriako Panepistemio Athenon, Athens, Greece
• ⁵Panepistemio Thessalias Tmema Iatrikes, Larissa, Greece
• ⁶Panepistemio Thessalias Ergasterio Biomathematikon, Larissa, Greece

Introduction: There are conflicting results for Glucagon-like peptide 1 receptor agonists' (GLP-1 RAs') and limited data for Sodium-glucose cotransporter 2 (SGLT2) inhibitors' effectiveness in preventing Type-2 Diabetes Mellitus (T2DM) in high-risk adults. An updated investigation is warranted. We aimed to explore their effectiveness in preventing T2DM in high-risk patients, and assess changes in body weight/body mass index (BMI), glycemic parameters, and safety. Materials and Methods: Search in PubMed, Cochrane Library Central Register of Controlled Trials and SCOPUS for eligible Randomized Controlled Trials (RCTs). A Systematic Review (SR) and Meta-Analysis (MA). GRADE assessment of overall evidence. Results: All 24,157 participants in 10 GLP-1 RAs RCTs were overweight/obese. Compared to placebo, GLP-1 RAs reduced T2DM incidence (OR 0.51, 95%CI 0.28, 0.94; P-value 0.03). 2.4 mg of semaglutide were overall effective (OR 0.38, 95%CI 0.16, 0.94; P-value < 0.0001). Subgroup analysis indicated effectiveness in patients more than 50 years across the world, in cardiovascular disease, after 100 weeks, and during post-intervention. Liraglutide was not overall effective. However, subgroup analyses demonstrated effectiveness for studies that were performed worldwide, for women more than 40 years, at 3.0 mg daily, after 55 weeks of administration, and only during intervention. Exenatide was not effective. Heterogeneity was large (Q 54.56, P-value < 0.0001; I² 84%, 95%CI 74, 89%) and the MA was performed by random effects model. Heterogeneity was explained by countries' performance in semaglutide-and liraglutide-based RCTs and participants' mean age, dosage, duration and post-intervention evaluation in liraglutide-based RCTs. Sensitivity analyses considering studies with post-intervention assessment and studies with the largest sample size and drop-out more than 5% in semaglutide-based RCTs explain further heterogeneity. The Quality of evidence was low. Compared to placebo, GLP-1 RAs reduced weight (kg) and BMI (kg/m²) (mean difference ‒ 6.35; P-value < 0.00001 and ‒2.46; P-value < 0.00001, respectively). Finally, GLP-1 RAs were safe (OR for adverse events 1.01; P-value 0.95). Conclusions: GLP-1 RAs may prevent diabetes in high-risk adults and ameliorate body and glycemic factors. Their effectiveness should be considered carefully due to the low quality of evidence. No safety issues were identified. Future investigation is necessary to provide consistency of estimations.