Abstract
We conducted a case-control study to evaluate the association between paternal smoking and childhood leukemia and to evaluate potential modification by polymorphisms in CYP1A1. Histologically confirmed childhood leukemia cases (n=164) and non-cancer controls (n=164) were recruited from three teaching hospitals in Seoul, Korea. Five single nucleotide polymorphisms in CYP1A1 (-17961T>C, -9893G>A, I462V, 1188C>T (*2A), and 11599C>G) were genotyped and haplotypes were estimated by the expectation-maximization method. We also conducted a meta-analysis of 12 studies that have reported the association between paternal smoking and childhood leukemia risk. Paternal smoking at home was associated with all leukemias (OR=1.8, 95% CI=1.1-2.8) and acute lymphoblastic leukemia (ALL) (2.0, 1.2-3.4). An increasing trend in risk was observed for pack-years smoked after birth (P(trend)=0.06 and 0.02, respectively) and the number of smokers in the home during the child's life (P(trend)=0.05 and 0.03, respectively). Among those without the CGACC haplotype, ALL risk was significantly increased by the father's smoking at home (2.8, 1.5-5.3) and the presence of at least one smoker in the home (2.3, 1.2-4.4), and the test for interaction was significant (P(interaction)=0.03 and 0.02, respectively). The meta-analysis showed that overall paternal smoking (1.13, 1.04-1.24) and smoking before the pregnancy of the child (1.12, 1.04-1.21) were significantly associated with childhood leukemia risk. Our results suggest that paternal smoking is a risk factor for childhood leukemia and the effect may be modified by CYP1A1 genotype.