Analysis of GWAS top hits in ADHD suggests association to two polymorphisms located in genes expressed in the cerebellum

Attention deficit/hyperactivity disorder (ADHD) is a common psychiatric disorder influenced by genetic factors. Several chromosomal regions with potential linkage and candidate genes associations have been reported, but findings are often inconsistent and non-replicated. The few genome-wide association studies (GWAS) carried out so far differ for study design and phenotypes analyzed, and did not detect any association significant at the genome-wide level. In the present study we examined the top SNPs reported in the GWAS by Neale et al. [2008] in an independent cohort. Although our sample size is smaller (415 trios vs. 909), the power was sufficient to confirm the role of candidate markers in ADHD if a true association exists. Two out of 36 top SNPs were significant at α = 0.05 in our sample, although none was still significant after correction for multiple tests. These two SNPs are both located in genes coding for as yet uncharacterized proteins expressed in the cerebellum, XKR4 in 8q12.1, and FAM190A in 4q22.1. Three other FAM190A SNPs have TDT P-values of <10^-5 in our sample, a level of significance only reached by a total of five SNPs in our genome-wide data. While these findings could be due to chance, we cannot exclude that these markers are indeed associated to disease risk. Remarkably, brain imaging studies have shown reduction of the posterior inferior cerebellar lobules volume of ADHD boys and girls compared to controls, persistent with age and not present in unaffected siblings, suggesting that the cerebellum may be directly related to pathophysiology of ADHD.