Causal association between Parkinson's disease with multiple cancers: a Mendelian randomization study

Abstract

Abstract Background Some previous observational studies proposed a causal correlation of Parkinson's disease (PD) and cancers, but the relationship between Parkinson's disease with cancers is still controversial, and causality has not been fully established. Methods Using publicly available genome-wide association study data from databases, we identified different numbers of single nucleotide polymorphisms (SNPs) as instrumental variables to infer causality using two-sample Mendelian randomization (MR) analyses with inverse variance weighting (IVW), MR- egger method, and Weighted median method. Results The primary IVW method showed negative association between genetically predicted Parkinson's disease with malignant neoplasm of the brain (OR = 0.7077; 95%CI, 0.5586–0.8967; P = 0.004); and positive association between Parkinson's disease with breast cancer (OR = 1.0022; 95%CI, 1.0009–1.0034; P = 0.0008). However, no causal association was observed between Parkinson's disease with melanoma skin cancer, Parkinson's disease with lung cancer or bladder cancer. And these results were supported by Cochran 's Q test and MR-Egger intercept test. Conclusion This two-sample Mendelian randomization analysis found that Parkinson's disease was negatively associated with malignant neoplasm of the brain and that Parkinson's disease was positively associated with breast cancer.