Animal experiments have shown that the polymeric form of fibronectin, sFN, is an effective systemic anti-metastatic agent. The sFN polymer is formed by treating soluble fibronectin with a fragment from its first type III repeat (anastellin). This proposal is based on new results showing that systemic administration of either sFN or anastellin alone, inhibits tumor growth. The likely mechanism for the anti-tumor effect is suppression angiogenesis. This application proposes optimization of the anastellin peptide for maximal efficacy. This will be accomplished by selecting for the strongest fibronectin binders from libraries of mutated anastellin peptides displayed on phage. Standardized tumor metastasis and angiogenesis assays in vivo will be used to determine which anastellin variants have optimal anti-tumor activities. In phase II, this information will be used to scale up and standardize the production of anastellin. The results could be the basis for a new, non-toxic anti-cancer compound. Moreover, these results may also be of broader significance, because the anti-angiogenic activity of anastellin may be based on a mechanism that is similar to that of certain other anti-angiogenic factors, such as angiostatin.