Screening of six novel candidate genes for association with Graves' disease
Ruth Tisdall, Matthew Simmonds, Paul Newby, Jayne Franklyn, Stephen Gough & Oliver Brand
Genome wide association screening (GWAS) has proved invaluable in determining otherwise undetected genetic effects for several common endocrine diseases. The largest GWAS performed in Graves disease (GD), to date, has not only confirmed association of several known gene regions, including the HLA region, TSHR and FCRL3, but has also identified several other possible regions of association with GD. As GD shares several susceptibility loci with other endocrine autoimmune diseases, including type 1 diabetes (T1D), several of the novel gene regions detected through GWAS in T1D, including PTPN2, CD226 and IFIH1, have also been associated with GD. A recent meta-analysis conducted on three of the largest T1D GWAS performed, investigating over 8207 T1D subjects and controls, identified six novel T1D susceptibility regions including, the intergenic rs165738, gene desert located rs6887079, PRKCQ rs947474, CTSH rs3825932, TNP2 rs416603 and C1QTNF6 rs229541 SNPs. To determine whether these six newly detected T1D susceptibility loci could also be associated with GD, these SNPs were screened in a large UK Caucasian GD cohort consisting of 2504 GD patients and 2688 controls subjects. All subjects gave informed written consent and the project was approved by the local ethics committee. None of the six SNPs studied showed any association with GD (P=0.980.05). Clinical correlations were preformed to determine if these SNPs were associated with a specific clinical GD sub-phenotype including age of onset, severity of Graves ophthalmopathy (NOSPECS 26), presence of goitre and autoantibody status, however no associations were detected (P=0.990.07). This study can confidently rule out the possibility of any major genetic contributions of these six novel T1D regions to GD (OR≥1.15). However we can not rule out the possibility of these SNPs contributing smaller effects to GD susceptibility (OR<1.15) without screening these SNPs in additional GD datasets of equal/greater size and the use of meta analysis.