Tumour hypoxia: the picture has changed in the 1990s.

Since the 1950s, the presence of hypoxic cells in human tumours has been widely regarded as a problem, and a variety of strategies have been developed and tested, both in experimental and clinical studies, to overcome this perceived problem. One of these strategies was the development of bioreductive cytotoxins--drugs which in themselves were relatively innocuous, but when metabolized under hypoxic conditions, became highly cytotoxic, thereby preferentially killing the hypoxic cells. Modelling studies and experimental data with newly developed hypoxic cytotoxins, such as SR 4233 (tirapazamine) and RSU 1069, have led to the realization not only that it is better to kill hypoxic cells in tumours than to radiosensitize or oxygenate them, but also that with these bioreductive cytotoxins hypoxic cells in tumours can be an advantage in cancer therapy. However, to realize the advantage of adding the drug with each radiation dose, the tumour must undergo a process analogous to reoxygenation, which we have termed 'rehypoxiation', by which hypoxic cells are regenerated after each dose of the hypoxic cytotoxin. In addition, we also discuss the fact that hypoxia is a cellular stress which activates many new genes. The activation of these genes will be a major focus for research in coming years and will undoubtedly lead to new approaches in cancer detection and treatment. In summary, the 1990s are bringing a fundamental change in our perception of tumour hypoxia, from a position of being a problem to that of being a solution in cancer treatment.