Laboratory and epidemiological studies support the hypothesis that cancer incidence in human populations can be reduced by supplementing high-risk individuals with chemopreventive agents. Many candidate agents have been identified, too many to be assayed in long-term clinical trials. As an alternate approach, intermediate markers are currently being evaluated as short-term screens for the activity of chemopreventive agents in humans. These markers quantify cellular and molecular changes of biological significance to the process of carcinogenesis. One such marker is the micronucleus test on exfoliated cells. This assay has been used to quantify chromosomal breakage occurring in the human oral cavity, esophagus, cervix, lung, nasal cavity and urinary bladder. Intervention trials on high-risk populations have shown that supplementation with chemopreventive agents can modulate this breakage. This article will review the evidence in support of the use of this assay as a biological marker for the efficacy of a chemopreventive regime. Basic problem areas in the design and conduct of this assay in humans will also be discussed, as will the future potential of the assay.