Heritability of Parkinson disease in Swedish twins: a longitudinal study

Abstract

Previous twin studies report no heritability of Parkinson's disease (PD) based on cross sectional information. Here, we apply a longitudinal design and re-evaluate cross sectional data in the population-based Swedish Twin Registry (STR) using clinical as well as hospital discharge and cause of death diagnoses. In the longitudinal analyses (based on 46,436 individuals), we identified 542 twins with PD and 65 twins with Parkinsonism. Concordance rates for PD were 11% for monozygotic and 4% for same-sexed dizygotic twin pairs, with a heritability estimate of 34%. Concordance rates for PD or parkinsonism were 13% for monozygotic and 5% for same-sexed dizygotic twin pairs, with a heritability estimate of 40%. In the cross sectional analyses (based on 49,814 individuals), we identified 287 twins with PD and 79 twins with parkinsonism. Concordance rates for PD were 4% for monozygotic and same-sexed dizygotic twin pairs and 0 for opposite-sexed twin pairs. Concordance rates for PD or parkinsonism were somewhat higher but the heritability estimate was nonsignificant. Our longitudinal analyses demonstrate that PD and parkinsonism are modestly heritable.

Introduction

The etiology of Parkinson's disease (PD) is considered to be complex and to involve both genetic and environmental factors. Mutations have been identified in several genes in families with autosomal dominant or recessive PD, and multiple susceptibility genes have been reported (Lesage and Brice, 2009). Familial aggregation studies have consistently shown that having a relative with PD increases the risk of PD (Thacker and Ascherio, 2008), but these studies cannot differentiate between genetic and shared familial effects, such as rearing environment. This distinction can be made in twin studies, as monozygotic twins are genetically identical and dizygotic twins share on average half of the variation of their genome, while both share rearing environment. One twin study of PD including almost 20,000 men from the National Academy of Sciences/National Research Council (NAS-NRC) World War II Veteran Twins Registry reported similar concordance rates in monozygotic and dizygotic pairs in the total sample, but in cases with age of onset before 50 years, concordance rate was significantly higher in monozygotic than in dizygotic twins (Tanner et al., 1999). These results suggested that among males with late onset PD, environmental factors are most important, while in early onset PD, genetic factors also seem to play a role.

In our previous study, we screened all twins in the Swedish Twin Registry (STR) who were 50 years or above and alive in 1998 (n = 33,780 individuals) (Wirdefeldt et al., 2004). Through a combination of telephone interview results and linking the STR to the Swedish national Inpatient Discharge Register (IDR), we identified 247 twins with self-reported or register-based PD diagnosis. Only 2 pairs were concordant, suggesting that individual-specific environmental factors are most important in the etiology of PD although low penetrant genes or gene by environment interactions could not be excluded. Shortcomings included the cross sectional design and lack of clinical evaluations. Nevertheless, both the USA and our Swedish twin data suggest that on a population basis, there is little impact of genetic variance.

The aims of the present study were: (1) to apply a longitudinal design to study heritability of PD by following the twin cohort from the 1960's onward using the national hospital discharge and cause of death registers, as well as information from clinical assessments, and (2) to re-evaluate the cross sectional data after completion of clinical assessments of twins with suspected PD.