Tumor cell proliferation is frequently associated with genetic or epigenetic alterations in key regulators of the cell cycle. Most known oncogenes and tumor suppressors target entry into the cell cycle and control the G(1)/S transition. However, tumor-associated alterations in spindle formation or chromosome segregation are also frequent and may result in chromosomal instability. In fact, a few centrosomal or mitotic proteins such as aurora A, polo-like kinase 1 and PTTG1 (securin) have been reported to act as oncogenes. Some spindle checkpoint regulators such as the BUB kinases or MAD2 protect cells from aberrant chromosome segregation and may therefore function as suppressors of malignant transformation. However, few cancer-associated mutations in these or other mitotic regulators have been described thus far and many of these molecules do not fit into the classical definition of 'oncogenes' or 'tumor suppressor genes'. In some cases, both over-expression and decreased expression of these genes result in mitotic arrest. Moreover, some mitotic regulators such as MAD2 are either up- or down-regulated depending on the tumor types and, in both cases, these alterations result in chromosomal imbalances and tumor development. Minor changes in protein levels that do not compromise cell viability might therefore be sufficient to dysregulate the mitotic cycle and induce genomic instability. Despite the limited knowledge on the molecular basis of these processes, the clinical success of mitotic poisons such as taxanes reinforces the interest in these molecules, their involvement in human cancer and the therapeutic opportunities to modulate their function in cancer treatment.