A cancer metastasis-on-chip assay was developed to model early stages of cancer metastasis, including tumor cell invasion and intravasation, and to screen targeted anti-cancer drugs. The model used a microfluidic organ-on-chip platform with perfused microvessels, stromal fibroblasts, and a triple-negative breast cancer cell line (MDA-MB-231). High-content imaging and automated quantification were used to assess tumor cell behavior. The model showed that co-culturing MDA-MB-231 cells with fibroblasts enhanced invasion and intravasation, while MCF7 cells remained non-invasive. A high-content screening of a targeted anti-cancer drug library identified 30 compounds that reduced tumor intravasation by 60% compared to controls. Multi-parametric analysis revealed that MEK inhibitors were enriched in clusters related to cell invasion and intravasation, while drugs targeting ABL, KIT, PDGF, SRC, and VEGFR showed strong effects on intravasation with less impact on invasion or proliferation. Imatinib, a multi-kinase inhibitor, enhanced endothelial barrier stability and reduced tumor cell intravasation. The model demonstrated the potential of the metastasis-on-chip assay as a tool for studying cancer metastasis, drug discovery, and personalized therapies for triple-negative breast cancer. The study highlights the importance of modeling the tumor microenvironment to understand metastasis and develop effective therapies. The model's ability to recapitulate key aspects of the tumor microenvironment, including perfused microvessels, ECM, and stromal fibroblasts, makes it suitable for high-throughput and high-content screening. The findings suggest that targeting endothelial barrier function and tumor cell-endothelium interactions could be key strategies for preventing cancer metastasis. The study also emphasizes the need for personalized models that reflect the pathophysiological relevance of cancer to improve therapeutic outcomes.A cancer metastasis-on-chip assay was developed to model early stages of cancer metastasis, including tumor cell invasion and intravasation, and to screen targeted anti-cancer drugs. The model used a microfluidic organ-on-chip platform with perfused microvessels, stromal fibroblasts, and a triple-negative breast cancer cell line (MDA-MB-231). High-content imaging and automated quantification were used to assess tumor cell behavior. The model showed that co-culturing MDA-MB-231 cells with fibroblasts enhanced invasion and intravasation, while MCF7 cells remained non-invasive. A high-content screening of a targeted anti-cancer drug library identified 30 compounds that reduced tumor intravasation by 60% compared to controls. Multi-parametric analysis revealed that MEK inhibitors were enriched in clusters related to cell invasion and intravasation, while drugs targeting ABL, KIT, PDGF, SRC, and VEGFR showed strong effects on intravasation with less impact on invasion or proliferation. Imatinib, a multi-kinase inhibitor, enhanced endothelial barrier stability and reduced tumor cell intravasation. The model demonstrated the potential of the metastasis-on-chip assay as a tool for studying cancer metastasis, drug discovery, and personalized therapies for triple-negative breast cancer. The study highlights the importance of modeling the tumor microenvironment to understand metastasis and develop effective therapies. The model's ability to recapitulate key aspects of the tumor microenvironment, including perfused microvessels, ECM, and stromal fibroblasts, makes it suitable for high-throughput and high-content screening. The findings suggest that targeting endothelial barrier function and tumor cell-endothelium interactions could be key strategies for preventing cancer metastasis. The study also emphasizes the need for personalized models that reflect the pathophysiological relevance of cancer to improve therapeutic outcomes.