Tumor growth and metastasis are angiogenesis-dependent. The possibility of inhibiting tumor growth by interfering with the formation of new vessels has recently raised considerable interest. We previously reported that it is possible to inhibit primary tumor growth and metastasis in a transgenic model of spontaneous breast tumor, which shows many similarities to its human counterpart (including ability to metastasize) by intratumoral administration of a DNA conKeywords: gene therapy; breast adenocarcinoma; metastasis; transgenic mice; anti-angiogenesis; liposomesIt is now widely recognized that the development of new blood vessels (angiogenesis) is necessary to sustain tumor growth, invasion and metastasis. At some point during tumor growth, cancer cells acquire the ability to activate the quiescent vasculature to produce new blood vessels via a so-called 'angiogenic switch '. 1,2 Although the need for recruiting new blood vessels to the tumor has been know for more than two decades, the possibility of treating malignancies by acting on their vasculature has become popular only since the isolation of endogenous angiogenesis inhibitors, which dramatically affect cancer growth in experimental murine systems. Highly promising results on experimental tumors in rodents using these inhibitors as purified proteins as well as transduced genes, alone or in combination with traditional therapies have been reported. [3][4][5][6][7] The establishment of a dormancy status, which in some cases persists even after the suspension of therapy, has been obtained with these gene products, 5 and this, together with the absence of drug resistance, 8 raised the possibility that the anti-angiogenic effects could be useful in the treatment of human tumors.The achievement of the inhibition of local tumor growth is the usual end-point of these therapeutical attempts, but in humans it is the metastatic spread that is responsible for the majority of cancer-related deaths. Therefore, to be really useful in the human context, any new approach must be evaluated in its ability to interfere Correspondence: MG Sacco, ITBA CNR, Via F lli Cervi, 93, 20090 Segrate (MI), Italy Received 17 August 2000; accepted 11 October 2000 struct carrying the murine angiostatin cDNA driven by liposomes. Here we report that it is also possible to achieve this goal by a systemic (intraperitoneal) delivery of therapeutic DNA constructs carrying genes coding for mouse and human anti-angiogenic factors which include angiostatin, endostatin and TIMP-2. These findings may be relevant to the design of therapeutic interventions in humans. Gene Therapy (2001) 8, 67-70. with tumor cell invasion and metastasis, a therapeutical target which is rarely, if ever, investigated at the experimental level. Unfortunately, most of the results in experimental oncology have been obtained on transplantable tumors which are usually generated by inoculating highly malignant cultured cells, which rapidly grow in the site of injection and become the target for therapeutic intervention. ...