BACKGROUND: The association between androgen deprivation therapy (ADT) and dementia risk is controversial, and the dose-response relationship between them remains unclear. OBJECTIVE: We aim to further clarify the relationship between ADT and dementia risk. METHODS: PubMed, Web of Science, Embase, and Cochrane Library databases were systematically searched up to September 2024 to identify relevant studies. A meta-analysis was conducted using hazard ratios (HR) and 95% confidence intervals (CI) as pooled indicators. The robust error meta-regression (REMR) approach was performed to explore the dose-response relationship. Heterogeneity was assessed using I² statistics, and subgroup analysis, sensitivity analysis, and meta-regressions were conducted. Publication bias was evaluated with a funnel plot and Egger's test. RESULTS: Our meta-analysis of 21 studies involving 2,278,835 patients revealed that ADT significantly increased the risk of overall dementia (HR = 1.14, 95% CI: 1.06-1.20) and Alzheimer's disease (AD) (HR = 1.15, 95% CI: 1.06-1.23), but not non-AD dementia (HR = 1.01, 95% CI: 0.89-1.16). For ADT subtypes, anti-androgens increased the risk of overall dementia (HR = 1.27, 95% CI: 1.09-1.49), particularly for AD (HR = 1.53, 95% CI: 1.19-1.97), while Luteinizing hormone-releasing hormone therapy and bilateral orchiectomy were not linked to the risk of any dementia subtype. A bell-shaped non-linear relationship between ADT duration and dementia risk was observed, with the highest risk observed at 15.5 to 21.5 months (HR = 1.25), which was confirmed by subgroup analysis for AD. CONCLUSIONS: The risk of overall dementia and AD were found to be significantly associated with ADT in a bell-shaped dose-response effect.