GATA-3 expression and its correlation with prognostic factors and survival in canine mammary tumors

Abstract

Introduction: The transcription factor GATA-3 plays a significant role in mammary gland development and differentiation. Recent studies on human oncology have demonstrated its association with favorable pathologic factors in breast cancer. Canine mammary tumours, proposed as comparative and translational study models, have epidemiological, clinical, biological, and genetic characteristics similar to those of human breast cancers.

Methods: Here, we evaluated the frequency of GATA-3 expression in mammary tumors of dogs and its relationship with prognostic factors and survival. Tumor samples were obtained from 40 female dogs and grouped according to histological type into benign tumors (n = 10), carcinoma in mixed tumors (CMTs) (n = 20), and aggressive tumors (n = 10). CMTs were further separated according to histological grade, and data on clinical staging and diagnosis, histopathological grading, and survival rate were collected.

Results: GATA-3 and estrogen receptor (ER) expression were higher in benign and well-differentiated carcinomas than in aggressive tumors, which showed greater Ki-67 expression. The expression rate of ER in the studied groups was equivalent to that of GATA-3. We identified a strong positive correlation between GATA-3 and ER expression frequencies and a negative correlation between those of GATA-3 and Ki-67. There were associations between GATA-3 (p < 0.001), Ki-67 (p = 0.003), tumor size (p < 0.001), clinical stage (p = 0.002), lymph node metastasis (p < 0.001), and histological grade (p < 0.001) by univariate survival analysis. The parameters ER (p = 0.015) and GATA-3 (p = 0.005) also influenced survival in a multifactorial manner.

Discussion: Kaplan-Meier analysis of survival curves validated our previous findings that dogs with GATA-3 expression in ≥79.4% of cells had significantly higher survival rates (p < 0.001). The performance analysis showed that the expression of GATA-3 in ≥79.4% of cells effectively predicted survival or death in dogs with mammary tumors. Collectively, these results suggest that GATA-3 can be a relevant marker in the study of mammary tumor progression and has potential as a prognosis marker for predicting outcomes in canine mammary tumors.

Keywords: breast cancer; canine mammary tumors; prognostic biomarker; survival; transcription factor; tumor progression.

Figure 1

Immunohistochemical expression of GATA-3, ER, and Ki-67 markers in canine mammary cancer samples (benign tumor, CMT grade I, CMT grade II and aggressive tumor) in objective of 40×. (A) Tubular adenoma in objective of 40× (HE). (B) Tubular adenoma in objective of 40× (GATA-3). (C) Tubular adenoma in objective of 40× (ER). (D) Tubular adenoma in objective of 40× (Ki-67). (E) CMT grade I in objective of 40× (HE). (F) CMT grade I in objective of 40× (GATA-3). (G) CMT grade I in objective of 40× (ER). (H) CMT grade I in objective of 40× (Ki-67). (I) CMT grade II in objective of 40× (HE). (J) CMT grade II in objective of 40× (GATA-3). (K) CMT grade II in objective of 40× (ER). (L) CMT grade II in objective of 40× (Ki-67). (M) Solid carcinoma in objective of 40× (HE). (N) Solid carcinoma in objective of 40× (GATA-3). (O) Solid carcinoma in objective of 40× (ER). (P) Solid carcinoma in objective of 40× (Ki-67).

Figure 2

Graphic representation of the expression of biomarkers in samples of canine mammary tumors. Brackets with asterisks indicate significant differences between the corresponding groups. Data were analyzed using the Kruskal–Wallis test, followed by the Dunn’s test. Statistical significance: (^*) p < 0.05, (^**) p < 0.01, and (^***) p < 0.0001. G-BT, group of benign tumors; G-CMT-I, group carcinomas in a mixed tumor (CMT) grade I; G-CMT-II, group of CMT grade II; G-AT/CMT-III, group of aggressive tumors and CMT grade III.

Figure 3

Kaplan–Meier survival curve according to the group studied (G-CMT-I, G-CMT-II and G-AT/CMT-III), regardless of GATA-3 expression. (^***) express a significant statistical difference among the curves by the log-rank test with p < 0.001. G-CMT-I, group carcinomas in a mixed tumor (CMT) grade I; G-CMT-II, group of CMT grade II; G-AT/CMT-III, group of aggressive tumors and CMT grade III.

Figure 4

Relationship between the survival rate of canine mammary tumors and the expression of the biomarker GATA-3. (A) Kaplan–Meier survival curve for all animals with malign mammary tumors (G-CMT-I, G-CMT-II and G-AT/CMT-III) characterized according to the intensity of GATA-3 expression (high ≥79.4% and low <79.4%) of GATA-3-stained cells. (^***) express a significant statistical difference among the curves by the log-rank test with p < 0.001. (B) Graphs represent the relationship of GATA-3 categorized according to the frequency of cases with high (≥79.4%) and low (<79.4%) expression intensity in the different types of canine mammary tumors (BT, CMT I, CMT II, and AT). (C) Correlation graph between GATA-3 expression and survival of animals with malignant mammary tumors. The correlation coefficient was positive, proving to be a directly proportional relationship. Data were obtained by Pearson’s correlation test. Degree of significance: p < 0.0001. G-BT, group of benign tumors; G-CMT-I, group carcinomas in a mixed tumor (CMT) grade I; G-CMT-II, group of CMT grade II; G-AT/CMT-III, group of aggressive tumors and CMT grade III.

Figure 5

Graphical representation of cut-off performance indices of GATA-3 expression. (A) Scatter plot between GATA-3 expression and disease outcome in 365 days after mastectomy in animals with malignant mammary tumors. (B) ROC (Receiver Operating Characteristic) curve indices, using a specific cut-off point, including: area under curve/overall accuracy (AUC), Co-positivity (Co-pos), Co-negativity (Co-neg), negative and positive predictive values (NPV and PPV) and relative risk/likelihood ratio (LR⁻ and LR⁺).