Myoepithelial cells and extracellular matrix in the cytologic differentiation of canine mammary tumors

BACKGROUND Mammary neoplasms are common tumors in intact female dogs. Fine-needle aspiration cytology (FNAC) is a valuable diagnostic tool and has gained some credibility in the diagnosis of mammary tumors in dogs. Prompt classification of canine mammary tumors using cytology would enhance feasibility as a prognostic tool and guide clinical and surgical management. OBJECTIVES We aimed to examine background elements to differentiate mammary tumors using FNAC. We proposed to distinguish simple from complex and mixed tumors by identifying myoepithelial (ME) cells and different types of extracellular matrix. Additionally, we determined the accuracy of FNAC to differentiate benign from malignant tumors. METHODS One hundred and one mammary tumors from female dogs were included in this study. We compared FNAC using histopathology as the gold standard. Cellular and background components were evaluated and identified. The cytologic accuracy, sensitivity (Se), specificity (Sp), positive predictive value (PPV), and negative predictive value (NPV) for diagnosing malignancy were determined, excluding inadequate samples. RESULTS The cytologic-histologic agreement was 92.5% for simple carcinomas, 57.9% for complex-type carcinomas, 57.1% for mixed-type carcinomas, 27.3% for carcinosarcomas, and 100% for osteosarcomas. Myoepithelial cells were successfully identified using FNAC. Myxoid and chondroid/osteoid matrix were satisfactorily recognized. Cytologic accuracy, Se, Sp, PPV, and NPV for diagnosing malignancy were 99%, 100%, 83%, 99%, and 100%, respectively. CONCLUSIONS Chondroid/osteoid matrix was noted in mixed tumors but not in complex tumors. Myxoid matrix, often associated with ME cells, was noted in complex and mixed tumors. Mesenchymal cells were differentiated from ME cells, allowing the distinction of simple carcinomas with scirrhous reaction from complex and mixed tumors.