Abstract
OBJECTIVE: To investigate whether galactose-1-phosphate uridyl transferase (GALT) variant genotypes were associated with epithelial ovarian cancer risk, and to determine if this association was modified by lactose intake. DESIGN: Two prospective cohort studies and a case-control study. SETTING: Academic institution. PATIENT(S): A total of 992 cases and 1,050 population-based control samples from a New England case-control study and 240 cases and 900 control samples from the Nurses' Health Studies. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Genotyping of the N314D variant and the 4-bp deletion (-119delGTCA) of GALT with the use of the Taqman 5' nuclease assay. Duarte1 (D1) genotype individuals have a missense mutation (N314D) associated with normal GALT activity unless it occurs together with an associated 4-bp deletion leading to reduced GALT activity (Duarte2 or D2). RESULT(S): Logistic regression analysis identified no association between D1/D2 genotypes and ovarian cancer risk (pooled risk ratio 1.1 [95% confidence interval (CI) 0.8-1.5] for D1 and 1.0 [95% CI 0.7-1.4] for D2). We did not observe a significant interaction between D1 and D2 genotypes in analyses stratified by level of lactose intake. CONCLUSION(S): D1 and D2 genotypes do not appear to play a role in the association between galactose intake, possible ovarian dysfunction, and the link with ovarian cancer.