Over a century ago, Theodor Boveri paved the way to mechanistic studies linking chromosomal abnormalities to cancer pathogenesis. Since then, theoretical and empirical evidence has been accumulated, supporting a causal role of these events in the aetiology of human cancer. A powerful tool for measurement of chromosomal abnormalities is the cytokinesis-block micronucleus cytome (CBMNcyt) assay. The validation of the micronucleus (MN) as marker of phenotypic susceptibility to cancer has received decisive support from mutagens sensitivity studies, particularly from a recent case-control study on lung cancer, which showed increased frequency of tobacco carcinogen-induced MN, nuclear buds and especially nucleoplasmic bridges in cancer patients (odds ratios of 2.3, 10.0 and 45.5, respectively). Recently, a large international cohort study showed a significant association between MN frequency in healthy subjects and cancer risk. The study assembled data on 6718 individuals from 10 countries (62,980 person-years). Cancers incidence was significantly higher in groups with medium (RR=1.84; 95% confidence interval: 1.28-2.66) and high MN frequency (RR=1.53; 95% CI: 1.04-2.25). This study provided preliminary evidence that MN frequency in peripheral blood lymphocytes is predictive of cancer risk, suggesting that increased MN formation is associated with early events in carcinogenesis. These results, in combination with mechanistic evidence, prospected the use of MN frequency in cancer screening programmes. However, issues such as interindividual variability and preventive strategies in high-risk groups need to be further addressed to consolidate these achievements.