Folate is required for DNA synthesis, repair and methylation. Low folate status has been implicated in carcinogenesis, possibly as a result of higher rate of genetic damage. The aim of this study is to compare folate status and levels of DNA damage between breast cancer and benign breast disease control patients. Fasting blood samples from 64 histologically confirmed untreated breast cancer patients (mean age 57 years) and 30 benign breast disease control patients (mean age 51 years) were obtained. Red cell folate (RCF) and plasma homocysteine were measured. Mononuclear cells (MNC) were isolated for genetic damage analysis using the basic alkaline comet assay. Results are expressed as tail moment. Data were log transformed as appropriate before analysis for normalisation purposes. The geometric mean (95% confidence interval) of RCF (ng ml(-1)) in breast cancer patients was 339.07 (333.3-404.6) vs 379.5 (335.8-505.2) in control patients (P = 0.24). Corresponding plasma homocysteine concentrations (micromol l(-1)) were 11.9 (10.6-16.4) vs 10.1 (9.3-11.9) (P = 0.073), respectively. The mean tail moment (s.d.) of DNA damage in MNC of breast cancer patients detected by the basic comet assay was 1.4 (0.66) vs -0.17 (0.79) in controls (P < 0.0001, t-test), the modified comet assay 'endonuclease III (Endo III)' was 1.7 (0.70) vs 0.86 (0.81) (P < 0.0001, t-test), and the modified comet assay 'formamidopyrimidine glycosylase (FPG)' was 1.6 (0.62) vs 0.99 (0.94) (P < 0.0001, t-test). There was a significant negative correlation between RCF levels and DNA damage detected by modified comet assay 'FPG' (Pearson Correlation Coefficient r2 = -0.26, P = 0.02) and DNA damage was found to be significantly higher in MNC of breast cancer patients compared to benign breast disease control patients. Breast cancer patients tended to have lower RCF levels and higher levels of plasma homocysteine, but these differences were not significant. The study provides preliminary evidence that reduced folate status may be implicated in the aetiology of breast cancer perhaps by increasing the in vivo level of genetic instability.