The Heritability of Polymorphic Light Eruption

Abstract

Polymorphic light eruption is classified as an acquired idiopathic photodermatosis, yet it appears to cluster in families, suggesting a possible genetic component. In this study, we assess the heritability of polymorphic light eruption using the classical twin model. Polymorphic light eruption was investigated by a nurse-administered questionnaire in a sample of 420 pairs of adult female twins from St Thomas' Hospital UK Adult Twin Registry, including 119 monozygotic and 301 dizygotic pairs. Probandwise concordance for the presence and absence of disease was calculated and the heritability of polymorphic light eruption assessed by a quantitative genetic model fitting approach using Mx software. The prevalence of polymorphic light eruption was 21% and 18% in monozygotic and dizygotic twins, respectively. A family history of polymorphic light eruption in first-degree relatives (not including the cotwin) was present in 12% of affected twin pairs (where at least one twin had polymorphic light eruption) compared with 4% of unaffected twin pairs, providing evidence of familial clustering (p < 0.0001). The probandwise concordance for polymorphic light eruption was higher in monozygotic (0.72) than in dizygotic twin pairs (0.30), indicating a strong genetic effect. Quantitative genetic modeling found that a model comprising additive genetic (A) and unique environmental (E) factors provided the most parsimonious fit, although a dominant gene effect could also explain our data. In the AE model, 84% (95% confidence interval 65-94%) of the variance in susceptibility to polymorphic light eruption is attributed to additive genetic factors with the remaining 16% (95% confidence interval 6-35%) to unique environmental effects. These data establish a clear genetic influence in the expression of polymorphic light eruption and provide a basis for examining candidate genes that may be pathogenic in this common condition.

Study of profile of polymorphous light eruption at a tertiary referral center

“…6,17 It varied from 6.25-12% in the studies conducted by Ross and Millard. 23,24 As the disease presents with minimal symptoms, patients were not aware of similar symptoms in family members. In addition, the members of family work in different atmospheres and varying degree of sun exposure, which could be responsible for the low familial incidence.…”

“…PMLE appears to have a genetic basis, affecting patients' families one and half times more commonly than general population. 7 However, disease expression in genetically susceptible individuals is heavily dependent on environmental factors as well. 8 Typically affected sites are the bridge of the nose, malar areas of cheeks, tip of chin, sides and back of neck, upper chest, dorsa of the hands, dorsolateral aspects of arms, fronts and backs of the legs and dorsa of the feet.…”

The Heritability of Premenstrual Syndrome

“…(Millard et al, 2000). Heritability is indicated if the calculated value for MZ is higher than that of DZ twins.…”

“…The first assumption is that there is a 'continuous underlying liability to disease' and secondly, that 'a threshold of liability' dichotomizes those with PMS from those without PMS. (Millard et al, 2000). Assuming that the underlying liability approximates a continuous bivariate normal distribution, tetrachoric correlation can then be estimated by the frequencies of concordant and discordant twin pairs from the two-bytwo tables.…”

“…The elements of the genetic component are inclusive of the additive component (A) and dominant component (D) while the elements of environmental factors are comprised of shared environment (C) and unshared environment (E). In order to relate these estimates to that of the population, the overall phenotype variance was then estimated by adding the square of variances for each estimate (V p = a 2 + d 2 + c 2 + e 2 = 1) where V p is the overall phenotype variance of the population (which adds up to 1) (Millard et al, 2000). The right hand side of the formula includes items a, d, c, and e, which are the regression coefficients of observed variables with the same components as the previous formula (a: additive genetic variance, d: dominant genetic variance and so on).…”