Phenotypic heterogeneity in study populations may significantly confound the results of genetic association studies on alcohol dependence

BACKGROUNDThe interpretation of genetic studies on alcohol dependence may be confounded by the co-occurrence of substance dependence, psychiatric disorders and alcohol-related comorbidities, for example, cirrhosis. Significant single-marker and haplotypic associations between polymorphisms in the zinc finger gene, ZNF699, and alcohol dependence were reported in the Irish Affected Sib Pair Study of Alcohol Dependence population, one-third of whom had co-occurring substance dependence while 80% had identified psychiatric comorbidity. The aim of this study was to explore variant ZNF699 associations with alcohol dependence while exercising controls for potential confounders. METHODSThe study population was comprised of 1449 alcohol-dependent cases and 1283 population controls; all were of British or Irish ancestry. None of the cases had a history of dependence on other substances, and the frequency of comorbid depression was low. A separate, ancestry-matched cohort of 196 opioid-dependent cases was also included. Genotyping for the four previously identified SNPs of interest in ZNF699 was performed using K-Biosciences Competitive Allele Specific PCR. RESULTSNo single-marker associations were found between polymorphisms in ZNF699 and alcohol dependence per se. A significant allelic association was found between rs7254880 in ZNF699 and alcohol-related cirrhosis (n=292), using cases with no biopsy evidence of liver disease (n=314) as controls (P=0.013). Significant allelic associations were also found between rs12460279 (P=0.028), rs7252865 (P=0.012) and rs10854142 (P=0.016) in ZNF699 and opioid dependence. CONCLUSIONPhenotypic variation in study populations may contribute towards the nonreplication of genetic association studies on alcohol dependence; controls for recognised confounding variables should be exercised whenever possible.