Correlation between the Histopathological Diagnosis by AgNOR Count and AgNOR Area in Canine Mammary Tumors

• Published in: Journal of veterinary internal medicine 2008-09, Vol.22 (5), p.1174-1180
• Main Authors: Bundgaard-Andersen, K, Flagstad, A, Jensen, A.L, Hellmen, E, Trere, D
• Format: Article
• Language: English
• Subjects: Animals; Antigens, Nuclear - chemistry; Antigens, Nuclear - metabolism; Breast cancer in female dogs; diagnostic techniques; Dog Diseases - diagnosis; Dog Diseases - pathology; Dogs; Female; Image morphometry; mammary neoplasms (animal); Mammary Neoplasms, Animal - diagnosis; Mammary Neoplasms, Animal - pathology; Manual counting; Silver Staining
• ISSN: 0891-6640; 1939-1676
• DOI: 10.1111/j.1939-1676.2008.0144.x

Background: Mammary tumors are the most common type of tumor in female dogs. The histopathological diagnosis is usually made by a hematoxylin‐eosin (HE) staining of the tumor, which then requires a pathologist's judgment for assessment of malignancy. The purpose of this study was to investigate an alternative silver staining of some argyrophilic nucleolar organizer regions (AgNOR) for improving the diagnostic accuracy with mammary tumors. Hypothesis: There is a correlation between the histopathological diagnosis by AgNOR count and AgNOR area in canine mammary tumors. Animals: Seventy‐three canine mammary tumors from 33 female dogs. Materials and Methods: The AgNOR staining was evaluated retrospectively in 73 canine mammary tumors with a parallel HE staining as a “Gold Standard.” Both a quantitative manual counting method and a qualitative computerized morphometric method were tested. Result: The result from both methods indicated a clinically relevant difference in the mean values of the AgNOR in the following 4 categories: malignant, benign, hyperplastic, and normal mammary tissue. The counting method was superior, with 89% of the cases given a correct diagnosis of a malignant or a nonmalignant canine mammary tumor. The 2 methods were then compared to test their ability to classify the tumors correctly. Again, the counting method was the most reliable method, with a sensitivity of 80% and a specificity of 76% when the upper 50% of the AgNOR counts were presumed malignant. Conclusion and Clinical Importance: The results indicated that an AgNOR test could be an aid to pathologists as a prognostic indicator or to assist them in deciding between a benign or a malignant diagnosis in questionable cases.