Twin studies in multiple sclerosis. A meta-estimation of heritability and environmentality

BACKGROUND: Most twin studies of multiple sclerosis (MS) are inconclusive regarding the impact of genes and environment on disease susceptibility. In particular, high uncertainty exists about whether shared environmental factors are aetiologically relevant. OBJECTIVE: To disentangle, with a reasonable degree of confidence, the relative contributions of heritability and of shared and unique environmental components of MS susceptibility. METHODS: We performed a meta-analysis of previous twin studies. After a MEDLINE search, we selected eight twin studies in France, UK, Canada, Denmark, North America, Italy, Finland and Sweden. We conducted a biometric multi-group analysis under the liability-threshold model, by taking account of the study-specific ascertainment strategies and the population-specific prevalence rates of MS. RESULTS: The meta-analytic estimates of tetrachoric correlations were 0.71 (95% confidence interval (CI): 0.67-0.74) in monozygotic pairs and 0.46 (95% CI: 0.41-0.50) in dizygotic pairs. The biometric multi-group model provided meta-analytic estimates of 0.50 (95% CI: 0.39-0.61) for heritability, 0.21 (95% CI: 0.11-0.30) for shared environmental component and 0.29 (95% CI: 0.26-0.33) for unique environmental component. CONCLUSION: Our results support the continuing efforts to identify unknown genetic factors that fill the gap of 'missing heritability'; moreover, a 'missing environmentality' deserves future investigations into the role of non-heritable components that act as both shared and individual-specific exposures.