Parkinson disease (PD) is the second most common neurodegenerative disorder, affecting more than 1% of individuals over age 55 and 3% of those over 75 years of age (de Rijk et al. 1997). In the past decade, there has been significant progress in the identification of genes contributing to PD. Although all causative gene discoveries to date have come from studies of large kindreds with a clear Mendelian form of PD, subsequent analyses in familial and sporadic cases have demonstrated that some of these genes, notably LRRK2, play a broader role in disease susceptibility. Six Parkinson (PD) genome-wide association studies (GWAS) have been completed to date but only two loci have demonstrated repeated strong results, MAPT region and SCNA. While other gene associations have been replicated in multiple datasets, they have not reached genome-wide significance when corrected for multiple comparisons. One concern with these studies is the known genetic heterogeneity present in PD, which could lower the power of detection in these datasets. Another is the known misclassification of the clinical diagnosis. Studies on much smaller numbers of patients than used in GWAS studies suggest a 10-25% misclassification when autopsies are performed on clinically diagnosed PD patients. Thus, using neuropathic diagnoses from autopsy provides a much higher level of diagnostic accuracy for GWAS, and should provide more homogeneous groups for association studies, increasing statistical power to detect genetic effects. A second benefit to using autopsy-confirmed cases is minimizing false-positive results due to the presence of other forms of parkinsonism (with distinct genetic causes). Indeed, evaluations of the centers in this study suggest a 20-25% misclassification of clinical diagnoses, when compared with pathology. We propose to perform a GWAS analysis on PD patients with autopsy information to identify genes associated with PD risk. We propose to use the ongoing large clinical PD GWAS (~5000 cases and 5000 controls) to provide a replication dataset for significant findings in the autopsy study.
To reduce misclassification error introduced in the diagnosis of Parkinson Disease (PD) using only clinical criteria, which could mask significant results, we propose performing a genome wide association analysis using autopsy and clinically confirmed PD cases. PD is the second most common cause of neurodegeneration after Alzheimer's disease and this project will contribute to understanding the causes for PD. The preliminary written comments of the reviewers are reproduced below. These comments were prepared prior to the meeting and may not have been edited after the full committee discussion of your application.