Enzymatic saturation of metabolic pathways is one factor that potentially contributes to the nonlinear exposure-response relations that are frequently reported in occupational epidemiologic studies. The authors propose an approach to explore the contribution of saturable metabolism to previously reported exposure-response relations by integrating predictive models of relevant biomarkers of exposure into the epidemiologic analysis. The approach is demonstrated with 2 studies of leukemia in benzene-exposed workers, one conducted in the Australian petroleum industry (1981-1999) and one conducted in a US rubber hydrochloride production factory in Ohio (1940-1996). The studies differed greatly in their magnitudes and durations of exposure. Substitution of biomarker levels for external estimates of benzene exposure reduced the fold difference of the log relative risk of leukemia per unit of cumulative exposure between the 2 studies by 11%-44%. Nevertheless, a considerable difference in the log relative risk per unit of cumulative exposure remained between the 2 studies, suggesting that exposure misclassification, differences in study design, and potential confounding factors also contributed to the heterogeneity in risk estimates.