This study aimed at quantitatively comparing the occurrence/formation of DNA adducts with the carcinogenicity induced by a selection of DNA-reactive genotoxic carcinogens. Contrary to previous efforts, we used a very uniform set of data, limited to in vivo rat liver studies in order to investigate whether a correlation can be obtained, using a benchmark dose (BMD) approach. Dose-response data on both carcinogenicity and in vivo DNA adduct formation were available for six compounds, i.e. 2-acetylaminofluorene, aflatoxin B1, methyleugenol, safrole, 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline and tamoxifen. BMD(10) values for liver carcinogenicity were calculated using the US Environmental Protection Agency BMD software. DNA adduct levels at this dose were extrapolated assuming linearity of the DNA adduct dose response. In addition, the levels of DNA adducts at the BMD(10) were compared to available data on endogenous background DNA damage in the target organ. Although for an individual carcinogen the tumour response increases when adduct levels increase, our results demonstrate that when comparing different carcinogens, no quantitative correlation exists between the level of DNA adduct formation and carcinogenicity. These data confirm that the quantity of DNA adducts formed by a DNA-reactive compound is not a carcinogenicity predictor but that other factors such as type of adduct and mutagenic potential may be equally relevant. Moreover, comparison to background DNA damage supports the notion that the mere occurrence of DNA adducts above or below the level of endogenous DNA damage is neither correlated to development of cancer. These data strongly emphasise the need to apply the mode of action framework to understand the contribution of other biological effect markers playing a role in carcinogenicity.