Breast cancer, cytomegalovirus and Epstein–Barr virus: a nested case–control study

BACKGROUND: We investigated whether elevation in serum cytomegalovirus (CMV) or Epstein–Barr virus (EBV) immunoglobulin G (IgG) antibody levels precedes the development of breast cancer. METHODS: A nested case–control study was carried out within the Janus Serum Bank cohort. Two serum samples, one taken at least 4 years before diagnosis (sample 2) and an earlier sample (sample 1) from 399 women with invasive breast cancer and from 399 controls, matched for date of blood samples and age were tested for CMV and EBV IgG antibodies. Odds ratios (ORs) with 95% confidence intervals (CIs) for CMV and EBV seroconversion between the samples and unit changes in IgG optical density (OD) examined as a continuous variable were calculated using conditional logistic regression. RESULTS: Eleven cases and three controls seroconverted for CMV IgG between the first and second blood samples, with an adjusted OR for CMV IgG seroconversion of 4.0 (95% CI=1.1–14.4). The risk of breast cancer, adjusted for parity, increased per unit difference in CMV OD between samples (OR=1.7, 95% CI=1.1–2.5). In an analysis restricted to parous cases and age-matched parous controls, the OR for CMV seroconversion for IgG between the two samples, adjusted for parity and age at first birth, was 9.7 (95% CI=1.2–77.3). The EBV seroconversion or change in EBV OD was not associated with risk of breast cancer. CONCLUSION: Our hypothesis that elevation in serum CMV IgG antibody levels precedes the development of breast cancer in some women is supported by the results of this study. Changes in EBV IgG antibody are not associated with risk of breast cancer.