Several groups have shown that quantitation of tumor angiogenesis by counting blood vessels in primary breast cancer gives an independent assessment of prognosis. Poor prognosis is associated with high blood vessel counts. We have shown that the rate of cell division in endothelial cells is much higher in breast tumours than in normal breast. Breast cancer cell lines and primary human breast tumours express a wide range of vascular growth factors, including VEGF, placenta growth factor, pleiotrophin, TGF beta 1, acidic and basic FGF, and platelet-derived endothelial cell growth factor. Inhibiting angiogenesis by blocking vascular growth factors would be difficult with highly specific agents, but drugs with a broader spectrum of antagonism may be effective. We have developed several suramin analogues which are less toxic than suramin in vivo but more potent in inhibiting angiogenesis, and these have been developed for Phase I. A combination of anti-angiogenesis agents with drugs activated by hypoxia may also be useful, because anti-angiogenesis alone may not kill cells, whereas activation of hypoxic drugs could synergize. New endpoints may be necessary because inhibition of new blood vessel formation may not cause tumour regression. Thus, the endpoint of stable disease and biochemical assessment of inhibition of angiogenesis may be much more important in therapeutic studies and for drug development in the future. The prognostic importance of angiogenesis suggests that this should be a major new therapeutic target.