Investigation of variants of the aromatase gene (CYP19A1) in female pattern hair loss

Funding sources S.R. was a recipient of a BONFOR Gerok fellowship from the Medical Faculty of the University of Bonn. M.M.N. is recipient of a grant from the Alfried Krupp von Bohlen und Halbach‐Stiftung. R.C.B. is recipient of a Heisenberg Professorship from the German Research Foundation (DFG). This study was also funded by the British Skin Foundation (A.G.M. and R.T.‐A., 2005). Conflicts of interest: none declared. A.G.M. and R.C.B. have contributed equally to this work. Madam, Female pattern hair loss (FPHL) is a common disorder which affects around 12% of women by the age of 30 years, and 30–40% of women by the age of 70 years.1 2–3 Although its precise aetiopathogenesis is unknown, reports of familial occurrence strongly implicate genetic factors.4 5–6 Yip et al.7 recently performed a gene‐wide study of association between FPHL and the aromatase gene encoding CYP19A1, which is known to convert androgens to oestrogens in scalp hair follicles.8 They examined 61 tagging single nucleotide polymorphisms (SNPs) of CYP19A1 in an Australian sample of 484 caucasian women with grade 3–5 FPHL (according to the Sinclair scale9) and 471 controls.7 While no significant association was obtained at the allelic level, the CC genotype of rs4646 was significantly more frequent in cases (58·1%) than controls (48·9%) when compared with collapsed AC and AA groups (P =0·006), suggesting a recessive genetic effect. Although this result did not withstand experiment‐wide correction for multiple testing, its validity was supported by the following lines of evidence: (i) the highest frequency of the CC genotype (68·2%) occurred in women < 40 years, which is consistent with the hypothesis that genetic predisposition is more evident among early‐onset cases; and (ii) the rs4646 CC genotype has been associated with higher oestrogen levels than the AA genotype, which is consistent with reports that oestrogen has inhibitory effects on hair growth. Given the lack of experiment‐wide significance, the authors were cautious in their interpretation, and stressed the need for replication in independent samples. To our knowledge, the present study is the first to attempt replication of this finding.