Determining the occurrence of disease(s) in families is a traditional strategy to begin the study of genetic causation. Typically, as diseases cluster in families, the likelihood of a genetic basis for the disease is thought to increase, assuming, of course, no known common, causative environmental exposure. Family studies have formed the basis for many analytic genetic methods, including association, segregation, and linkage analyses; such studies focus on description, characterization, and identification of potential disease-causing genes. In this issue of the ARCHIVES, Levy et al1 use a basic family history design to describe the occurrence of Alzheimer disease (AD) among the relatives of persons with Parkinson disease (PD) as compared with AD occurrence in relatives of similar persons without PD. Because AD and PD are both neurodegenerative diseases, occurrence of one could possibly predispose family members to occurrence of the other, if similar etiologic disease genes were in play. Specifically, Levy and colleagues used the family history design to test the hypothesis that PD and AD may have "major shared genetic contributions to the etiology," which would result in an excess of AD among probands with PD. Their study includes information about AD occurrence in more than 4800 first-degree relatives of probands with PD and control probands; it is reportedly the largest study yet to address this hypothesis. Levy and colleagues found that the probands with PD showed no significant increased risk of AD compared with the AD risk among control relatives. Further, the relative risk (hazard) estimates were consistently null when stratifying by age at PD onset, by presentation (tremor-dominant or postural instability gait disorder), or by type of first-degree relative (parent or sibling); all risk estimates were also adjusted for sex, education, and ethnicity, and proportional hazards modeling accounted for age at AD onset or censoring. Despite the detailed and complete analytic methods, no significant differences emerged. All right, then; the evidence appears quite convincing! Should we as investigators conclude that genetic, etiologic overlap between AD and PD is nonsignificant and inconsequential?
Kukull WA. Perspectives on Shared Genetic Contributions for Parkinson Disease and Alzheimer Disease. Arch Neurol. 2004;61(7):1007-1008. doi:10.1001/archneur.61.7.1007