Table 1.
Scoring system established for cellularity, presence of clusters and background details for evaluation of cytological samples of CMTs in the present study.
Fig 1.
Cytological and histopathological samples of benign CMTs.
Cytological smears were stained with Giemsa, while histopathological samples with hematoxylin and eosin method. Lobular hyperplasia (A, B–cytology; C–histopathology), arrows indicate cholesterol clefts; Adenoma complex (D, E–cytology; F–histopathology); Benign mixed tumor (G, H–cytology; I–histopathology), an arrow indicates the extracellular matrix, an insert in the bottom-left corner of Fig 1H shows octeoclast. Original magnification: (A, C, D, F, G, I) 100x; (B, E, H) 400x.
Fig 2.
Cytological and histopathological samples of malignant CMTs.
Cytological smears were stained with Giemsa, while histopathological samples with hematoxylin and eosin method. Simple carcinoma (A, B–cytology; C–histopathology); Complex carcinoma (D, E–cytology; F–histopathology), an arrow indicates the extracellular matrix; Carcinoma arising in benign mixed tumor (G, H–cytology; I–histopathology). Original magnification: (A, C, D, F, G, I) 100x; (B, E, H) 400x.
Fig 3.
Robinson’s grading system adopted in canine mammary tumors.
Robinson’s grade 1. Uniform neoplastic cells in a cluster (A), with a smooth nuclear membrane, vesicular chromatin, and indistinct or nucleoli (D). Robinson’s grade 2. Mildly pleomorphic cells (B), with visible nucleoli, a slightly irregular nuclear margin and granular chromatin (E). Robinson’s grade 3. Pleomorphic cells in a loose cluster (C), with prominent nucleoli, an irregular nuclear membrane and chromatin clearing (F). Giemsa stain. Original magnification: (A, B, C) 400x; (D, E, F) 1000x.
Table 2.
Robinson’s grading system described by Robinson et al. [15].
Table 3.
The numbers and percentages of each CMT based on tumor behavior (benign/malignant) and tumor type determined by cytology (CP) and histopathology (HP, definitive diagnosis) without showing variations in diagnoses between CP and HP (n = 73).
Table 4.
The comparison of the cytopathological results according to the histopathological diagnosis for the differentiation of benign tumors, simple carcinomas and complex carcinoma/carcinoma arising in BMT/other malignant tumors (n = 73).
Table 5.
Benign tumors and malignancy grade classifications (1, 2, 3) of total of 73 CMTs based on cytological vs. histopathological diagnoses.
Fig 4.
A receiver operating characteristic (ROC) curve of cytopathological Robinson’s grading used for detecting malignant tumors (grade 2 and 3 by HP).
Only malignant tumors were included (in cytopathology, n = 61). Number of points on a scale is shown. A broken line is a line of non-discrimination.
Table 6.
Cytological features in cytological samples from benign (n = 18) and malignant (n = 55) mammary tumors ultimately diagnosed by histopathology.
Table 7.
Univariable analysis of cytological features in association with metastases 2 years after the mastectomy.
Fig 5.
A Kaplan-Meier survival plot of dogs with mammary tumors diagnosed by cytopathology and evaluated with the Mantel-Cox log rank test (α = 0.05).
The blue line indicates dogs with tumors of grade 1 (n = 24). The red line indicates dogs with tumors of grade 2 or 3 (n = 26). Short vertical lines signify censored observations. The median overall survival for dogs with mammary tumor graded 2 or 3 by CP was 28 months (IQR 1.5 to 46 months).
Table 8.
Univariable overall survival analysis of dogs with CMTs in relation to cytological examination.