ABSTRACT: Proliferation, dedifferentiation, loss of cell-cell contacts and angiogenesis are among the first steps of the metastasis cascade. The complex molecular pathways associated with these events in canine mammary tumors are mostly unknown and the value of this spontaneous canine tumor as a comparative model for human breast cancer is therefore still under debate. Messenger RNA profiles of lymph-node positive canine mammary carcinomas and normal mammary glands of the same dogs were compared by microarray analysis to elucidate molecular pathways associated with metastatic progression. Differential gene expression was analyzed by gene set enrichment and pathway analysis and compared with the gene expression data of human breast cancer. Metastatic canine carcinomas had 1,312 significantly differentially expressed genes when compared to the corresponding normal mammary gland. This expression profile included a significantly increased expression of cell division and extracellular matrix invasion genes (MMP1, MMP11, MMP13, SERPINE1, TFPI2, TIMP3). In contrast, genes associated with epithelial differentiation (EGF, EGFR, KRT17, MAP2K6, STAT5), cell adhesion (CLDN5, CLDN8, CTNNAL1, MCAM, MUC1, PECAM1) and angiogenesis (ANGPT2, ANGPTL1, ANGPTL2, ANGPTL4, FGFR1, FIGF, TIE1) were mostly down-regulated. Interestingly, tumors had a significant decrease in membrane receptors and growth factor pathway gene expression (EGFR, FGFR1, GHR, LYVE1, PDGFR, PDGFR, TGFBR, TIE1), indicating a potential independence from these proliferative stimuli. Several of the identified deregulated pathways overlap with gene expression profiles of human breast cancer. Gene expression profiling of metastatic carcinomas therefore identified complex molecular pathways and functional gene families that are deregulated during malignant progression in the tumors. This study provides new insights into canine mammary tumor metastasis and suggests that canine mammary tumors may serve as a valuable model for the human disease. 13 simple mammary carcinomas with lymph node metastases at the time of tumor resection were compared with the non-neoplastic mammary gland of the same dogs. Sufficient amounts of normal mammary gland were not available for dogs no. 1 (822) and 9 (61). In addition, one normal sample without matching tumor sample was analyzed (no. 8; 2609). All patients had no radiographically detectable pulmonary metastases at the time of tumor resection. Postoperative survival was analyzed by bi-yearly telephone interviews with the owners or the attending veterinarian. Distant metastases as the cause of death were determined postoperatively by radiographic detection of metastases (nos. 1-7, 9-12) or necropsy (nos. 13 and 14). Necropsy was not authorized by the owners of dogs nos. 1-7 and 9-12. Tissue specimens of tumors and lymph node metastases were fixed in neutral-buffered 4% formalin or snap frozen in liquid nitrogen within 15 minutes after resection and stored at -80ºC until further use. Formalin-fixed tumor tissue samples were routinely embedded in paraffin and sections of 2-μm thickness were mounted on adhesive glass slides and stained with hematoxylin and eosin. Tumors and lymph nodes were evaluated histologically independently by two board-certified pathologists, following the criteria of the WHO classification of canine mammary tumors (Misdorp et al. 1999).
