Risk of Parkinson's disease and depression severity in different populations: A two-sample Mendelian randomization analysis

Abstract

Background: Depression is widely recognized as a common non-motor symptom of Parkinson's disease (PD). Across different studies, the reported prevalence of depression in PD varies widely, ranging from 2.7% to 90%, but it is unclear whether this association is due to genetic or acquired factors. Whether there is a causal relationship remains unknown. The aim of this study was to use a two-sample Mendelian randomization (MR) approach to investigate the causal effect of PD on depression.

Methods: Analyses were conducted separately for individuals of European and East Asian ancestry using publicly available summary data from genome-wide association studies. Depression was divided into two categories: ever depressed for a whole week and major depressive disorder (MDD). PD data were used as the exposure and were obtained from the International Parkinson's Disease Genomics Consortium and the BioBank Japan PheWeb, while depression data were used as the outcome and were obtained from the ntegrative Epidemiology Unit (IEU) Open GWAS Project(A public GWAS database） and the Psychiatric Genomics Consortium. The influence of PD on depression was assessed using inverse variance weighted (IVW), weighted median, MR-Egger, and weighted mode methods. Heterogeneity and pleiotropy were tested, and the results were validated using FinnGen GWAS data from version R9.

Results: In individuals of European ancestry, there was a causal relationship between PD and ever depressed for a whole week (IVW method, odds ratio [OR] = 0.990; 95% CI, 0.984-0.996; p = .002), but no causal relationship was observed between PD and MDD (IVW method, OR = 0.974; 95% CI, 0.942-1.009; p = .141). In individuals of East Asian ancestry, no causal relationship was observed between PD and ever depressed for a whole week (IVW method, OR = 1.001; 95% CI, 0.829-1.209; p = .990) and between PD and MDD (IVW method, OR = 1.017; 95% CI, 0.982-1.052; p = .342). The results of the three additional analysis methods were similar to those of the IVW method, and there was no heterogeneity according to Cochran's Q-test. There was no evidence of pleiotropy based on MR-Egger intercept test and MR-PRESSO. The FinnGen validation dataset supported these findings. The results are stable and reliable.

Conclusion: The observed increase in depression among PD patients could potentially be attributed to modifiable acquired factors. Consequently, there is an urgent need to strengthen the management of PD patients in order to prevent the development of depression in the future.

Keywords: Mendelian randomization; Parkinson's disease; ever depressed for a whole week; major depressive disorder.

FIGURE 1

Forest plots showed the causal associations between depression severity and Parkinson's disease (PD) in different populations by inverse variance weighted (IVW). MDD, major depressive disorder; OR, odds ratio

FIGURE 2

The causal relationship between Parkinson's disease (PD) and ever depressed for a whole week in European ancestry. (a) Mendelian randomization (MR) leave‐one‐out sensitivity analysis of PD's causal effect on having experienced depression for a whole week. The figure shows that excluding a particular SNP does not lead to a significant change in the overall result. (b) Scatter plot: The slope of the line corresponds to the causal estimate from each method. (c) The forest plot shows the estimate of the effect of genetically increased PD risk on ever depressed for a whole week risk. Each black dot represents the log odds ratio (OR) for ever depressed for a whole week per standard deviation (SD) increase in log OR for PD. Horizontal lines represent 95% confidence intervals (95% CIs). (d) Funnel plot showing the relationship between the causal effect of PD on ever depressed for a whole week estimated using each individual SNP as a separate instrument against the inverse of the standard error of the causal estimate. There is relatively symmetry in the plot.

FIGURE 3

The causal relationship between Parkinson's disease (PD) and major depressive disorder (MDD) in European ancestry. (a) Mendelian randomization (MR) leave‐one‐out sensitivity analysis for PD on MDD. The figure shows that excluding a particular SNP does not lead to a significant change in the overall result. (b) Scatter plot: The slope of the line corresponds to the causal estimate from each method. (c) The forest plot shows the estimate of the effect of genetically increased PD risk on MDD risk. Each black dot represents the log odds ratio (OR) for MDD per standard deviation (SD) increase in log OR for PD. Horizontal lines represent 95% confidence intervals (95% CIs). (d) Funnel plot showing the relationship between the causal effect of PD on MDD estimated using each individual SNP as a separate instrument against the inverse of the standard error of the causal estimate. There is relatively symmetry in the plot.

FIGURE 4

The causal relationship between Parkinson's disease (PD) and ever depressed for a whole week in East Asian ancestry. (a) Mendelian randomization (MR) leave‐one‐out sensitivity analysis of PD's causal effect on having experienced depression for a whole week. The figure shows that excluding a particular SNP does not lead to a significant change in the overall result. (b) Scatter plot: The slope of the line corresponds to the causal estimate from each method. (c) The forest plot shows the estimate of the effect of genetically increased PD risk on ever depressed for a whole week risk. Each black dot represents the log odds ratio (OR) for ever depressed for a whole week per standard deviation (SD) increase in log OR for PD. Horizontal lines represent 95% confidence intervals (95% CIs). (d) Funnel plot showing the relationship between the causal effect of PD on ever depressed for a whole week estimated using each individual SNP as a separate instrument against the inverse of the standard error of the causal estimate. There is relatively symmetry in the plot.

FIGURE 5

The causal relationship between Parkinson's disease (PD) and major depressive disorder (MDD) in East Asian ancestry. (a) Mendelian randomization (MR) leave‐one‐out sensitivity analysis for PD on MDD. The figure shows that excluding a particular SNP does not lead to a significant change in the overall result. (b) Scatter plot: The slope of the line corresponds to the causal estimate from each method. (c) The forest plot shows the estimate of the effect of genetically increased PD risk on MDD risk. Each black dot represents the log odds ratio (OR) for MDD per standard deviation (SD) increase in log OR for PD. Horizontal lines represent 95% confidence intervals (95% CIs). (d) Funnel plot showing the relationship between the causal effect of PD on MDD estimated using each individual SNP as a separate instrument against the inverse of the standard error of the causal estimate. There is relatively symmetry in the plot.