No abstract

Keywords: eating disorders; catechol-O-methyltransferase (COMT); polymorphism; haplotype relative risk (HRR); transmission disequilibrium test (TDT)Anorexia nervosa (AN) is a common, severe and disabling psychiatric disorder, characterized by profound weight loss and body image disturbance. 1 Family and twin studies indicate a significant genetic contribution 2,3 and pharmacological data suggest possible dysfunction of the serotonergic 4,5 and dopaminergic 6-9 pathways. Catechol-O-methyltransferase (COMT) is a candidate gene for mediating susceptibility to AN since it is involved in the dopamine catabolism 10 and because its functional polymorphism (Val/Met 158) determines high (H) and low (L) enzymatic activity alleles. 11 Fifty-one Israeli AN patients and their parents were genotyped with the COMT polymorphism. Using the haplotype relative risk (HRR) method it was found that the frequency of the H allele among alleles transmitted to AN patients from their parents was significantly higher than in those not transmitted (68% vs 51% 2 = 5.20, df = 1, P = 0.023, odds ratio: 2.01). Transmission disequilibrium test (TDT) revealed that out of 49 heterozygote parents the H allele was transmitted to AN patients 33 times while the L allele was transmitted only 16 (McNemar's 2 = 5.90, df = 1, P = 0.015). Our study suggests that the COMT gene is associated with genetic susceptibility to AN, and that individuals homozygous for the high activity allele (HH) have a two-fold increased risk for development of the disorder. Molecular Psychiatry (2001) 6, 243-245.Fifty-one DSM-IV anorexia nervosa female patients and both their parents were genotyped for the COMT gene (valine to methionine at codon 158) polymorphism. In the light of inherent problems facing association studies, we used the haplotype relative risk (HRR) 12 and the transmission disequilibrium test (TDT) 13 methods that are robust to artifacts caused by population stratification. The genotypes in each family were examined and analyzed so as to determine which of the parents' alleles were transmitted to the affected daughters. The non-transmitted alleles were used to construct the nontransmitted genotypes. According to the HRR method, genotypes and alleles in the patient population are compared with the non-transmitted alleles and genotypes that form the optimal ethnically matched control population. The distribution of the transmitted and non-transmitted genotypes and alleles in the fifty-one trios is presented in Table 1.Homozygotes for the high activity allele (HH) were significantly more frequent among AN patients than in the non-transmitted controls (51% vs 29%, 2 = 4.01, df = 1, P = 0.043; odds ratio: 2.50, 95% CI 1.10-5.64). When allelic distribution was compared, it was found that the high activity allele is significantly more common among patients than in the non-transmitted alleles (68% vs 51%, 2 = 5.20, df = 1, P = 0.023, odds ratio: 2.01, 95% CI 1.39-3.55). The parent population did not differ from the 172 healthy Israeli controls in terms of genotypic and...

No abstract