Childhood acute lymphoblastic leukemia associated with parental alcohol consumption and polymorphisms of carcinogen-metabolizing genes.

BACKGROUND: Limited information is available on the association of parental consumption of alcohol prior to and during pregnancy with the risk of childhood leukemia, as well as for the potentially modifying role of genetic polymorphisms.
METHODS: We conducted a population-based, case-control study of 491 incident cases of acute lymphoblastic leukemia age 0-9 years and matched on age and sex to 491 healthy controls. Cases were identified at tertiary care centers in the Province of Québec between 1980 and 1993. Each parent was interviewed separately about alcohol consumption habits. We also used a case-only design with 186 cases to estimate interaction odds ratios between prenatal exposure and child DNA variants in the GSTM1 and CYP2E1 genes.
RESULTS: The adjusted odds ratio for any maternal consumption during pregnancy was 0.7 (95% confidence interval = 0.5-0.9). The interaction odds ratios for the GSTM1 null genotype during third pregnancy trimester was 2.4 (95% confidence interval = 1.1-5.4); the interaction odds ratio for CYP2E1 variant G-1295C (or allele *5) during the nursing period was 4.9 (95% confidence interval = 1.5-16.7).
CONCLUSIONS: The observed association with maternal alcohol consumption during pregnancy could be due to the potential chemopreventive effects of flavonoids found in wine and beer. These possible effects of alcohol may be at least partially genetically determined, although data are preliminary.