Inter-individual differences in DNA repair capacity (DRC) may play a critical role in breast cancer risk. To test whether the nucleotide excision repair (NER) pathway was deficient in breast cancer case patients, a DNA repair assay measuring the removal of in vitro-induced benzo[a]pyrene diolepoxide (BPDE)-DNA adducts in lymphoblastoid cells was used. The study took advantage of cell lines from sisters discordant for breast cancer that were available from the Metropolitan New York Registry of Breast Cancer Families. Cell lines were available from 137 families with at least one discordant set of sisters (158 case patients and 154 control subjects). Cells were treated with BPDE for 30 min and were either harvested immediately or were washed and cultured in complete medium for 4 hr to allow DNA repair. A semi-quantitative immunofluorescence assay using a polyclonal anti-BPDE-DNA primary antibody was used to quantify DNA adducts. Percent DRC was categorized into quartiles based on control values, and odds ratios and 95% confidence intervals (CIs) were calculated using conditional logistic regression models adjusted for age at blood donation, current body mass index, and smoking. Mean % DRC was significantly lower in breast cancer case patients (26.5 ± 22.4) than in control subjects (35.1 ± 24.4, P = .001). Using the quartile with the highest % DRC as the referent group, adjusted odds ratios increased from 1.23 (95% CI = 0.57 to 2.65) to 2.38 (95% CI = 1.17 to 4.86) to 2.99 (95% CI = 1.45 to 6.17) (Ptrend = .002) as DRC decreased. We have also determined genotype for a number of polymorphisms in five NER genes including Xeroderma pigmentosum complementation group D (XPD, Asp312Asn and Lys751Gln), group C (XPC, Lys939Gln and Ala499Val), group G (XPG, His1104Asp), and group A (XPA, 23G/A) and ERCC1 (8092G/T) in a total of 592 sisters. One-way ANOVA was used to compare the differences of DRC between genotype subgroups. A linear model was used to identify the main genetic determinants of DRC, adjusting for age at blood donation, current body mass index, and smoking. For all SNPs analyzed, there were no significant differences in allele frequencies between cases and controls. When we determined among controls (N=166) the relationship between genotype and mean %DRC, a significant effect was observed only with XPC Ala499Val (p=0.025). The mean %DRC for the Ala/Ala, Ala/Val and Val/Val genotypes were 30.9, 41.2 and 43.5%, respectively, suggesting a protective effect of variant allele. The model including all polymorphisms considered only explained 16% of the variance of %DRC. These findings suggest that deficient DRC may be a susceptibility factor in breast cancer risk and that polymorphisms in DNA repair genes may be associated with DNA repair capacity although only a small part of the variation is explained by the polymorphisms we have investigated to date.

[Proc Amer Assoc Cancer Res, Volume 46, 2005]