Predictors of discontinuation, efficacy, and safety of memantine treatment for Alzheimer's disease: meta-analysis and meta-regression of 18 randomized clinical trials involving 5004 patients

Background: The risk-benefit relationship of memantine treatment for Alzheimer's disease (AD) remains unclear. In addition, variability between the results of clinical trials has been observed. The aim of this study was to investigate the risk-benefit relationship of memantine treatment in patients with AD and to determine the predictor effect of patient, intervention, and study design related covariates. Methods: A systematic review and meta-analysis of double-blind, placebo controlled clinical trials was performed. Primary outcomes were all-cause discontinuation, discontinuation due to adverse events (AE) and efficacy on cognitive function. Odds ratio (OR) and standard mean difference (SMD) with 95% confidence intervals were calculated. Meta-regression was conducted to identify related covariates. Cochrane Collaboration tool was used to evaluate the risk of bias of included trials. Results: Eighteen studies involving 5004 patients were included. No differences between memantine and placebo were found for all-cause treatment discontinuation (OR=0.97 [0.82, 1.14]) and discontinuation due to AE (OR=1.18 [0.91, 1.53]). Memantine showed small improvement on cognitive function (SMD=0.15 [0.08, 0.22]). Baseline functional ability was positively associated with all-cause treatment discontinuation and discontinuation due to AE. Conclusions: Our study suggests that memantine has a very small efficacy on AD symptomatology and its safety profile is similar to that of placebo. No evidence of treatment discontinuation improvement with memantine is found, indicating a dubious risk-benefit relationship. No intervention characteristic or subgroup of patients clearly shows a significantly better risk-benefit relationship.