Replication of Chinese PCOS Susceptibility Loci in Patients diagnosed with PCOS from Caucasian Descent

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: MON 548-560-Hyperandrogenic Disorders
Basic/Clinical
Monday, June 17, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board MON-549
Yvonne V. Louwers*, Lisette Stolk and André G Uitterlinden
Erasmus MC University Medical Center, Rotterdam, Netherlands
BackgroundChinese Genome Wide Association Studies (GWAS) have revealed new susceptibility loci for PCOS. It has been well established that ethnic background adds to the phenotypic differences in PCOS patients. Therefore, it seems plausible that genetic variants associated with PCOS act differently in diverse ethnic populations.

AimThe aim of this study was to investigate the association of the genetic variants identified in Han Chinese women also demonstrate association with PCOS in a population of Caucasian decent.

Methods Association with PCOS of 17 genetic variants mapping to the 11 Chinese PCOS loci, i.e., LHCGR, THADA, DENND1A, FSHR, c9orf3, YAP1, RAB5B/SUOX, HMGA2, TOX3, INSR, SUMO1P1, was studied in a population from Caucasian descent. Study population consisted of 703 PCOS patients and 2164 controls from the general population. PCOS was diagnosed according the Rotterdam criteria and Caucasian ancestry was determined by Principle Component Analysis. Since we only had moderate power (0.40-0.75) to detect effects of similar size as observed in the Chinese cohort we also performed a meta-analysis for the variants in LHCGR, THADA, DENND1A which have been replicated in an Caucasian population by others (1).  Taken into account multiple testing, a p-value of > 0.002 was considered statistically significant (0.05/17 SNPs =0.002).

Results Risk variants mapping THADA (rs12468395 and rs12478601), DENND1A (rs10986105), FSHR (rs2349415) and c9orf3 (rs4385527) were nominally significantly associated with PCOS susceptibility in our case control set of Caucasian descent. The strongest PCOS association was with rs705702 mapping RAB5B/SUOX and rs1894116 mapping YAP1 (OR 1.21, p-value= 0.003; OR 1.37, p-value = 0.002, respectively). However, most likely due to lack of power, these signals did not reach significance. Meta-analysis with the published data by Welt et al (1) resulted in a statistically significant replication of DENND1A(rs10986105, p-value 0.0003).

Conclusion Meta-analysis with the published data by Welt et al (1) resulted in a statistically significant replication of DENND1A (rs10986105) in a large Caucasian case control sample. The direction and the magnitude of effects of majority of remaining SNPs tested were similar to those observed in the GWAS in the Han Chinese. However, since allele frequencies as well as phenotypic characteristics differ between PCOS patients from different ancestry, large consortia as we are conducting now, are needed to confirm these findings.

(1) Welt et al., J Clin Endocrinol Metab 2012; 97:1342

Nothing to Disclose: YVL, LS, AGU

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm