Mendelian Randomization Study Supports Genetic Liability to Obsessive–Compulsive Disorder Associated With the Risk of Alzheimer's Disease

Abstract

Background

Observational studies have suggested that obsessive–compulsive disorder (OCD) may be associated with Alzheimer's disease (AD). However, whether OCD is a causal risk factor for AD remains unclear. This study aimed to assess the causal effect of OCD on AD risk by performing a two-sample Mendelian randomization (MR) analysis.

Methods

Genome-wide association summary statistics were obtained for OCD, comprising 2688 cases and 7037 controls, as well as for AD, including 21,982 cases and 41,944 controls from Kunkle et al.’s study, and 39,918 cases and 358,140 controls from Wightman et al.’s study. On the basis of two diverse thresholds, OCD-associated genetic variants were screened as instrumental variables (IVs) for subsequent MR analyses. Inverse variance weighed was the primary MR method. MR-Egger, weighted median, and weighted mode were used as supplementary MR methods. Various sensitivity tests assessed the reliability of MR results.

Results

On the basis of strict IV selecting thresholds, inverse-variance weighted (IVW) identified significant causal associations between genetic liability to OCD and increased risk of AD in two different sources ((i) Kunkle et al.: odds ratio [OR] = 1.070, 95% confidence interval [CI]: 1.015–1.127, p = 0.012; (ii) Wightman et al. 0.012; (iii) Wightman et al.: OR = 1.051, 95% CI: 1.014–1.090, p = 0.007). Three other supplementary MR methods yielded similar results to IVWs (OR > 1). Furthermore, all results were replicated in MR analyses based on lenient IV selecting thresholds. The sensitivity tests indicated that MR results were stable and not affected by significant horizontal pleiotropy.

Conclusions

This comprehensive MR study suggests that genetic liability to OCD is a causal risk factor for AD. Early intervention in patients with OCD may be beneficial in preventing future AD progression.