Background: Androgen receptor (AR) activity is modulated by the length of CAG-repeat polymorphism in exon 1. Published data suggest association between CAG-repeat alleles and PCOS, but not all studies confirmed this finding.
Aim: To test association of AR alleles with PCOS and clinical and biochemical signs of hyperandrogenism.
Study design: Casecontrol association study, approved by the institutional ethical committee, of 137 PCOS patients (Rotterdam criteria) and 130 control subjects of Romanian origin. Androgen receptor genotyping, X-chromosome methylation analysis, and phenotyping for PCOS were performed.
Results: Median CAG repeat numbers were 22 (range 1130) in PCOS and 23 (range 1330) in controls. Biallelic means of CAG repeats were significantly shorter between PCOS and control groups (22.58 vs 23.16, P=0.01, t-test). X-inactivation analysis was done in 221 heterozygous subjects (109 PCOS and 112 controls). X-inactivation status did not differ significantly between PCOS and control groups (non-random inactivation: 55.9 vs 48.2%, skewed inactivation: 11.9 vs 13.3%).
When stratified by X-inactivation status, only the non-random subgroup of PCOS patients (versus controls) had lower biallelic means (22.27 vs 23.39, P=0.0014) and X-weighted biallelic means (22.23 vs 23.39, P=0.0054). There was no evidence of preferential allele inactivation in favor of shorter alleles in PCOS.
In the control group, but not in PCOS, plasma testosterone values were significantly correlated to AR X-weighted biallelic means (r2=0.14, P=0.014). Ferriman-Gallwey hirsutism scores were not influenced by AR alleles.
Conclusion: Our results demonstrate that the CAG polymorphism of the androgen receptor is significantly associated with PCOS in Romanian women (Eastern Europe) and may play an important role in PCOS pathogenesis.
Prague, Czech Republic
24 - 28 Apr 2010
European Society of Endocrinology