• Asthma genetics

Two recent reports in Thorax failed to show an association between polymorphisms within PHF11 and asthma phenotypes.1 2 Against this, there is evidence for an association between PHF11 and one or more phenotypes of total immunoglobulin E (IgE), atopic eczema and asthma in six independent cohorts.3–5 Here, we argue in support of the continued investigation into the biological role of PHF11 in the unravelling of the pathogenesis of atopic disease and asthma.

In the report by McClenaghan et al, positive associations were identified but lost upon correction for multiple testing; this approach is arguably overconservative given that it is a replication analysis. The study was powered to detect an association with an OR of at least 1.46, an OR the authors consider a requirement for ‘a major locus modifying asthma risk’. However, results emerging from genome-wide association studies of complex diseases show that apart from the human leucocyte antigen (HLA) locus it is rare to find an association with an OR of this level.

Both studies report on asthma phenotypes in non-selected cohorts drawn from the general population. Elevated total serum IgE is a marker for atopy; the mean total IgE in the twin study of McClenaghan et al was 22 IU/ml and this contrasts with 399 IU/ml in the childhood asthma study of Hersh et al and 337 IU/ml in our study of young children, all of whom developed atopic eczema by age 3. Both our study and that of Hersh et al reported an association with PHF11.

Early childhood eczema and atopic sensitisation contribute to the risk of later atopic respiratory disease. Since the initial association with PHF11 was with atopy/total IgE, we suggest that highly selected cohorts, such as young children with severe allergic disease, are better suited to detect positive single-gene associations with PHF11. In cohorts of less severely affected individuals genetic effects might only be detected by assessing gene–gene interactions, as suggested by Blakey et al. Overall, we believe an association, albeit with a modest OR, remains likely between PHF11 and IgE production/atopy in individuals afflicted by atopic asthma and eczema. Associations with low ORs may still provide important insights into the pathogenesis of complex genetic diseases; it is this principle that continues to drive the international effort in gene discovery.