Three hexitol derivatives, dibromomannitol (DBM), dibromodulcitol (DBD), and dianhydrogalactitol (DAG), originally investigated in Hungary, have been evaluated as anticancer agents in the United States. Their principal mechanism of action is attributed to alkylation via actual or derived epoxide groups. Their preclinical spectrum includes activity against murine leukemias and against the murine ependymoblastoma, which is particularly noteworthy for DAG. Dibromomannitol trials were targeted to chronic myelogenous leukemia but no advantage over busulfan therapy was demonstrable. Dibromodulcitol and DAG were sequentially evaluated for their usefulness against a wide variety of tumors. The activity of DBD against breast cancer has stimulated several continuing trials in this disease. On the other hand, DAG was disappointing in breast cancer and in several other malignancies, but some activity has been noted against lung cancer. Both DBD and DAG are being investigated for possible usefulness in the management of patients with intracranial neoplasms. The present clinical experience does not allow firm judgment on the advantage of one analogue over another. Such comparative analysis does point out the desirable direction of future studies as well as the limitations of current preclinical systems for the selection of analogues.