One of the major problems in angiogenesis research remains the lack of suitable methods for quantifying the angiogenic response in vivo. We describe the development and application of the directed in vivo angiogenesis assay (DIVAA) and demonstrated that it is reproducible and quantitative. This assay consists of subcutaneous implantation of semiclosed silicone cylinders (angioreactors) into nude mice. Angioreactors are filled with only 18 μl of extracellular matrix premixed with or without angiogenic factors. Vascularization within angioreactors is quantified by the intravenous injection of fluorescein isothiocyanate (FITC)-dextran before their recovery, followed by spectrofluorimetry. Angioreactors examined by immunofluorescence show cells and invading angiogenic vessels at different developmental stages. The minimally detectable angiogenic response requires 9 days after implantation and ≥50 ng/ml (P < 0.01) of either fibroblast growth factor-2 or vascular endothelial growth factor. Characterization of this assay system demonstrates that the FITC-labeled dextran quantitation is highly reproducible and that levels of FITC-dextran are not significantly influenced by vascular permeability. DIVAA allows accurate dose-response analysis and identification of effective doses of angiogenesis-modulating factors in vivo. TNP-470 potently inhibits angiogenesis (EC50 = 88 pmol/L) induced by 500 ng/ml of fibroblast growth factor-2. This inhibition correlates with decreased endothelial cell invasion. DIVAA efficiently detects differences in anti-angiogenic potencies of thrombospondin-1 peptides (25 μmol/L) and demonstrates a partial inhibition of angiogenesis (∼40%) in a matrix metalloprotease (MMP)-2-deficient mouse compared with that in wild-type animals. Zymography of angioreactors from MMP-deficient and control animals reveals quantitative changes in MMP expression. These results support DIVAA as an assay to compare potencies of angiogenic factors or inhibitors, and for profiling molecular markers of angiogenesis in vivo.