Tobacco-constituent biomarkers are metabolites of specific compounds present in tobacco or tobacco smoke. Highly reliable analytic methods, based mainly on mass spectrometry, have been developed for quantitation of these biomarkers in both urine and blood specimens. There is substantial interindividual variation in smokingrelated lung cancer risk that is determined in part by individual variability in the uptake and metabolism of tobacco smoke carcinogens. Thus, by incorporating these biomarkers in epidemiologic studies, we can potentially obtain a more valid and precise measure of in vivo carcinogen dose than by using self-reported smoking history, ultimately improving the estimation of smoking-related lung cancer risk. Indeed, we have demonstrated this by using a prospective study design comparing biomarker levels in urine samples collected from smokers many years before their development of cancer versus those in their smoking counterparts without a cancer diagnosis. The following urinary metabolites were associated with lung cancer risk, independent of smoking intensity and duration: cotinine plus its glucuronide, a biomarker of nicotine uptake; 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol and its glucuronides (total NNAL), a biomarker of the tobacco carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK); and r-1-,t-2,3,c-4- tetrahydroxy-1,2,3,4-tetrahydrophenanthrene (PheT), a biomarker of polycyclic aromatic hydrocarbons (PAH). These results provide several possible new directions for using tobacco smoke-constituent biomarkers in lung cancer prevention, including improved lung cancer risk assessment, intermediate outcome determination in prevention trials, and regulation of tobacco products.