Bias in effect size of systemic lupus erythematosus susceptibility loci across Europe: a case-control study

Introduction: We aimed to investigate whether the effect size of the systemic lupus erythematosus (SLE) risk alleles varies across European subpopulations. Methods: European SLE patients (n = 1,742) and ethnically matched healthy controls (n = 2,101) were recruited at 17 centres from 10 different countries. Only individuals with self-reported ancestry from the country of origin were included. In addition, participants were genotyped for top ancestry informative markers and for 25 SLE associated SNPs. The results were used to compare effect sizes between the Central Eureopan and Southern European subgroups. Results: Twenty of the 25 SNPs showed independent association with SLE, These SNPs showed a significant bias to larger effect sizes in the Southern subgroup, with 15/20 showing this trend (P = 0.019) and a larger mean odds ratio of the 20 SNPs (1.46 vs. 1.34, P = 0.02) as well as a larger difference in the number of risk alleles (2.06 vs. 1.63, P = 0.027) between SLE patients and controls than for Central Europeans. This bias was reflected in a very significant difference in the cumulative genetic risk score (4.31 vs. 3.48, P = 1.8 x 10(-32)). Effect size bias was accompanied by a lower number of SLE risk alleles in the Southern subjects, both patients and controls, the difference being more marked between the controls (P = 1.1 x 10(-8)) than between the Southern and Central European patients (P = 0.016). Seven of these SNPs showed significant allele frequency clines. Conclusion: Our findings showed a bias to larger effect sizes of SLE loci in the Southern Europeans relative to the Central Europeans together with clines of SLE risk allele frequencies. These results indicate the need to study risk allele clines and the implications of the polygenic model of inheritance in SLE.