Comparative efficacy and safety of antidiabetic agents in Alzheimer's disease: A network meta-analysis of randomized controlled trials

Alzheimer's disease (AD) is a progressive neurodegenerative disorder with limited treatment options. Emerging evidence suggests that antidiabetic agents may offer neuroprotective effects by targeting shared pathophysiological mechanisms such as insulin resistance and neuroinflammation. However, the comparative efficacy, and safety of these agents in the treatment of AD remain unclear. This study aimed to systematically evaluate and compare the efficacy and safety of antidiabetic agents for improving cognitive outcomes, reducing amyloid-β (Aβ) deposition, and managing adverse effects in patients with AD, using a network meta-analysis of randomized controlled trials (RCTs). A comprehensive literature search was conducted across multiple databases to identify RCTs examining the effects of antidiabetic agents in patients with AD. The primary outcomes included cognitive performance (e.g., MMSE scores), Aβ deposition (measured via CSF biomarkers), and safety/adverse effects. A network meta-analysis was performed to integrate direct and indirect evidence, ranking interventions using Surface Under the Cumulative Ranking (SUCRA) probabilities. Risk of bias was assessed using the Cochrane risk-of-bias tool. A total of 26 studies, involving 7,361 participants, were included in the analysis. The interventions evaluated included insulin detemir (both low-dose and high-dose), liraglutide, exenatide, metformin, and pioglitazone. Both low-dose insulin detemir (mean difference: 2.10, 95 % CI: 1.04 to 3.15), high-dose insulin detemir (mean difference: 1.40, 95 % CI: -0.07 to 2.88), exenatide (mean difference: 1.19, 95 % CI: 0.06 to 2.32), and metformin combined with exenatide (mean difference: 1.06, 95 % CI: -1.68 to 3.80) showed cognitive improvements compared to placebo. Among these, low-dose insulin detemir demonstrated the most significant improvement. In terms of reducing Aβ deposition, metformin ranked highest in effectiveness, with the highest SUCRA score (84.6), followed by high-dose insulin detemir (SUCRA: 54.1). Low-dose insulin detemir (SUCRA: 51.1) also demonstrated moderate efficacy. Low-dose insulin detemir showed some reduction in Aβ deposition (mean difference: -0.31, 95 % CI: -2.82 to 2.20), although statistical significance was limited. Liraglutide exhibited the highest rate of study treatment withdrawal (mean difference: 1.97, 95 % CI: -0.07 to 4.00), while pioglitazone demonstrated the lowest withdrawal rates (mean difference: 0.07, 95 % CI: -0.03 to 0.17). This network meta-analysis provides valuable insights into the comparative efficacy and safety of antidiabetic agents in AD. Low-dose insulin detemir demonstrated the most significant cognitive improvement and a moderate effect on reducing Aβ deposition. Metformin emerged as the most effective agent for reducing Aβ levels, though its effects on cognitive function were less pronounced. Safety profiles varied, with liraglutide associated with the highest rate of treatment withdrawals, while pioglitazone demonstrated the lowest incidence of treatment-related discontinuations. These findings support the potential use of antidiabetic agents, particularly insulin detemir, as a therapeutic option for AD, although further studies are needed to confirm their long-term benefits and safety.