Mendelian Randomization and Gene-Environment Interactions in Parkinson’s Disease ; Randomisation mendélienne et interactions gène-environnement dans la maladie de Parkinson

Epidemiological studies on the role of environmental factors in Parkinson's disease (PD) have shown several associations, but it is still debated whether they are causal, in particular because of a potential reverse causality bias due to the long prodromal phase of PD. The aim of our work is to provide evidence for the causal role of different exposures (smoking, alcohol, coffee and dairy consumption, body mass index [BMI]) in PD through two-sample Mendelian randomization (MR) analyses in the international Courage-PD consortium. Our results support an inverse causal association for smoking and BMI and a positive one for dairy consumption, with a bidirectional relationship for BMI. These associations were not explained by reverse causation, survival bias or incidence-prevalence bias. No association was found for alcohol and coffee consumption, but these results should be interpreted with caution due to limited statistical power for these exposures. Finally, we showed a significant interaction between smoking and the HLA-DRB1 gene, particularly a valine at position 11, providing further evidence for the role of smoking. This work highlights the interest and difficulties of using complementary causal inference methods for the study of complex diseases. It contributes to open interesting research avenues in the field of exposures involved in PD in order to develop prevention strategies. ; Les études épidémiologiques sur le rôle de facteurs environnementaux dans la maladie de Parkinson (MP) ont mis en évidence plusieurs associations, mais leur causalité est encore discutée, en particulier en raison d’un biais potentiel de causalité inverse lié à la longue phase prodromale de la MP. Notre travail a pour objectif d’apporter des arguments en faveur du rôle causal de différentes expositions (tabagisme, consommation d’alcool, de café et de produits laitiers, indice de masse corporelle [IMC]) dans la MP au travers d’analyses de randomisation mendélienne (RM) à deux échantillons dans le consortium international Courage-PD. ...