Genetic susceptibility modifies the association of long-term air pollution exposure on Parkinson's disease

Abstract

Inconsistent findings exist regarding the potential association between polluted air and Parkinson's disease (PD), with unclear insights into the role of inherited sensitivity. This study sought to explore the potential link between various air pollutants and PD risk, investigating whether genetic susceptibility modulates these associations. The population-based study involved 312,009 initially PD-free participants with complete genotyping data. Annual mean concentrations of PM_2.5, PM₁₀, NO₂, and NO_x were estimated, and a polygenic risk score (PRS) was computed to assess individual genetic risks for PD. Cox proportional risk models were employed to calculate hazard ratios (HR) and 95% confidence intervals (CI) for the associations between ambient air pollutants, genetic risk, and incident PD. Over a median 12.07-year follow-up, 2356 PD cases (0.76%) were observed. Compared to the lowest quartile of air pollution, the highest quartiles of NO₂ and PM₁₀ pollution showed HRs and 95% CIs of 1.247 (1.089-1.427) and 1.201 (1.052-1.373) for PD incidence, respectively. Each 10 μg/m³ increase in NO₂ and PM₁₀ yielded elevated HRs and 95% CIs for PD of 1.089 (1.026-1.155) and 1.363 (1.043-1.782), respectively. Individuals with significant genetic and PM₁₀ exposure risks had the highest PD development risk (HR: 2.748, 95% CI: 2.145-3.520). Similarly, those with substantial genetic and NO₂ exposure risks were over twice as likely to develop PD compared to minimal-risk counterparts (HR: 2.414, 95% CI: 1.912-3.048). Findings suggest that exposure to air contaminants heightens PD risk, particularly in individuals genetically predisposed to high susceptibility.

Fig. 1. Associations of air pollution and genetic risk with incident PD.

The joint associations of long-term NO₂ (a), NO_x (b), PM_2.5 (c), and PM₁₀ (d) exposure and PRS with the risk of incident PD. Cox regression models adjusted for age, sex, BMI, alcohol status, smoking status, education level, employment status, household income, Townsend deprivation index, genotyping array, the number of long-term morbidities, and the first 10 principal components of ancestry. Error bars represent HR with 95% confidence intervals (95% CI).