Damage to DNA induced by carcinogenic chemicals reflects exposure and is directly related to tumor formation, whereas modification of protein provides relatively precise dosimetry for stable adducts of proteins with a known half-life. Sophisticated methods for the detection and quantitation of DNA and protein adducts have been developed during the last approximately 25 years. For DNA adducts the most widely used methods include electrochemical detection, mass spectrometry, fluorescence and phosphorescence spectroscopy, immunoassays and immunohistochemistry and (32)P-post-labeling. Detection limits for quantitative assays are typically in the range of 1 adduct in 10(7) or 10(9) nucleotides. However, accelerator mass spectrometry, which is highly sophisticated but less accessible, has a detection limit of approximately 1 adduct in 10(12) nucleotides. Methods for the determination of protein adducts include immunoassay and a variety of elegant high-resolution mass spectrometry approaches. The detection limit of approximately 0.1 fmol for protein adducts, is based primarily on method specificity and the availability of large quantities of sample material. Using these highly sensitive methods a major achievement has been the biomonitoring of chemically exposed human populations. Validation of macromolecular adduct formation in humans has been predicated on studies in animal models. Adduct formation in humans appears to be indicative of molecular dosimetry and suggestive of increased human cancer risk. However, despite the large body of literature documenting DNA and protein adduct molecular dosimetry for many carcinogen exposures, the relationship between adduct formation and human cancer risk has been defined for only a few carcinogens. Thus, elucidation of this association remains a compelling challenge. For the future, integration of DNA and protein adduct measurements together with documentation of correlative and subsequent events, and host susceptibility factors, within the context of valid molecular epidemiologic study designs, will further our understanding of human disease mechanisms.