Abstract
Obesity can affect the clinical course of a number of diseases, including breast cancer in women and mammary gland tumors in female dogs, via the secretion of various cytokines and hormones. The objective of this study was to examine the expression patterns of obesity-related molecules such as aromatase, leptin, and insulin-like growth factor 1 receptor (IGF-1 R) in canine mammary carcinomas (CMCs) on the basis of the body condition score (BCS). Comparative analyses of the expression of these molecules, together with prognostic factors for CMCs, including hormone receptors (HRs; estrogen and progesterone receptors), lymphatic invasion, central necrosis of the tumor, and histologic grade, were performed on 56 CMCs. The mean age of CMC onset was lower in the overweight or obese group (8.7 ± 1.9 years) than in the lean or ideal body weight group (10.4 ± 2.7 years). The proportion of poorly differentiated (grade III) tumors was significantly higher in the overweight or obese female dogs. Aromatase expression was significantly higher in the overweight or obese group and was correlated with the expression of HRs (P = .025). These findings suggest that overweight or obese status might affect the development and behavior of CMCs by tumor-adipocyte interactions and increased HR-related tumor growth.
Keywords
Obesity, an excessive accumulation of adipose tissue in the body, is one of the most important health problems in both veterinary and human medicine. Obesity is associated with the development of, and poor prognosis for, numerous diseases such as hypertension, cardiac disease, diabetes mellitus, respiratory disease, and neoplasms, including breast, colon, esophagus, and prostate cancers. 1,3,35,39 Among the myriad health conditions and diseases affected by obesity, mammary gland tumors are the most common neoplasms in both women 7 and female dogs. 24 Molecular interactions between adipocytes and breast cancer cells have been studied and discussed in many reports, 4–5,21 and obesity is known as a risk factor for breast cancer development in postmenopausal women. 2 Indeed, excessive deposition of adipocytes influences normal and neoplastic mammary gland epithelial cells by secretion of various cytokines and hormones, including leptin, adiponectin, insulin-like growth factor 1 (IGF-1), tumor necrosis factor–α, interleukin-6, resistin, and aromatase. 20,43
Sex steroid hormones, including estrogen and progesterone, play significant roles in the growth and invasion of both human breast cancers and canine mammary cancers through the activation of hormone receptors (HRs). 41 In particular, estrogen levels in tumor tissue can be elevated by intratumoral biosynthesis due to the secretion of cytochrome P450 aromatase by carcinoma cells and adipocytes. 25
Leptin, one of the representative hormones released from adipocytes, stimulates the proliferation of breast cancer cell lines by upregulating the transcription of aromatase. 18,21 The levels of circulating leptin correlate with adipocyte accumulation in both humans and dogs. 17,40 Recently, the role of leptin in human breast cancer was investigated. It was found that high levels of intratumoral leptin receptors and plasma leptin correlate with a poor prognosis. 26 In addition, IGF-1 exerts mitogenic and antiapoptotic effects on canine mammary carcinomas (CMCs) by binding to the IGF-1 receptor (IGF-1 R) and by increasing the activation of the estrogen receptor (ER). 42 Together, these findings suggest that obesity may affect the development of human breast cancer in postmenopausal women, and CMCs in neutered female dogs, through continued estrogen biosynthesis and stimulation.
The objective of the current study was to analyze the expression of aromatase, leptin, and IGF-1 R and to evaluate the relationships between these obesity-related molecules, clinicopathologic features, and HR status, including that of the ER and progesterone receptor (PR) in CMCs on the basis of the body condition score (BCS). Immunohistochemistry (IHC) was used to determine the expression patterns of aromatase, leptin, and IGF-1 R and to identify relationships between several clinicopathological variables and BCS.
Materials and Methods
Study Population and Tissue Samples
Primary canine mammary tumor specimens from the histopathologic database of the Department of Veterinary Pathology, Konkuk University Animal Teaching Hospital (Seoul, Korea) from 2008 to 2012 were analyzed. Among the total number of CMC samples (n = 593), 56 for which all the necessary information about the age, ovariohysterectomy (OHE) status, and BCS was available were included in this study. The cases were classified into 2 groups on the basis of their BCS: group 1, BCS = 2 or 3 (lean or optimal body weight), and group 2, BCS = 4 or 5 (overweight or obese). 22 All samples were fixed in 10% neutral buffered formalin and embedded in paraffin wax. Four-micrometer-thick sections of these samples were used for hematoxylin and eosin (HE) staining and IHC.
Histopathology
Histologic typing of CMCs was based on the Proposed Histologic Classification system by Goldschmidt et al, 11 using HE-stained tissue sections. Grade assessment was performed on the basis of the grading system proposed by Clemente et al 6 : grade I (well differentiated), grade II (moderately differentiated), and grade III (poorly differentiated). Microscopic evidence of lymphatic invasion and necrosis was also evaluated.
Immunohistochemical Staining
All CMC samples were immunostained to examine the expression of ER, PR, HER2, aromatase, leptin, and IGF-1 R. IHC analysis was performed with primary antibodies (Table 1) on formalin-fixed, paraffin-embedded tissue sections. Slides were deparaffinized in xylene, rehydrated in graded alcohol, washed 3 times with phosphate-buffered saline (PBS), and incubated in 3% hydrogen peroxide for 20 minutes at room temperature (RT) to block endogenous peroxidase activity. The slides were washed 3 further times in PBS. Heat-induced antigen retrieval (Tris-EDTA, pH 9.0; citric acid, pH 6.0) for all primary antibodies was performed using a microwave oven at high power. After antigen retrieval, the slides were washed 3 times in PBS and incubated with primary antibodies. Horseradish peroxidase secondary antibodies (DAKO REAL Envision kit; Agilent Technologies, Santa Clara, CA) were added for 20 minutes at RT after washing 3 times in PBS. The slides were then washed 4 times in PBS and development reagents applied. The development reaction was terminated by washing in distilled water when the appropriate color intensity had been achieved. All slides were counterstained with Gill’s hematoxylin, dehydrated with ethanol, and covered with coverslips. The ER- and PR-positive CMC samples (n = 2) and triple-negative CMC sample (n = 1) were used as positive and negative controls, respectively, for each ER and PR IHC. Granulosa theca cell tumor tissues (n = 2) were used as a positive control for leptin IHC. All control tissues were selected from our database.
Primary Antibodies and Immunohistochemical Staining Protocols.
ER, estrogen receptor; PR, progesterone receptor; IGF-1R, insulin-like growth factor 1 receptor; RT, room temperature.
Evaluation of Protein Expression
To evaluate protein expression and assess immunoreactivity, 2 pathologists independently evaluated CMC tissue slides without knowledge of any background features. At least 3 representative fields (200× magnification) were analyzed on each slide. Digital images were acquired using a light microscope (Olympus, Tokyo, Japan) and Leica Application Suite 2.7 digital image transfer software (Leica, Wetzlar, Germany).
To evaluate the expression of ER and PR, the percentage of cells with nuclear immunoreactivity was determined. Each tissue sample was considered positive when, in the nuclei, more than 10% of neoplastic epithelial cells expressed relevant HRs. 9,15,38 The single ER- or PR-positive CMCs and both receptor-positive CMCs are all regarded as HR positive.
Assessment of aromatase expression was performed on the basis of the scheme described previously, 9,32 whereas slides showing >10% weakly positive cells were defined as positive in this study. When neoplastic epithelial cells had moderate or strong immunoreactivity in the cytoplasm, these cases were considered positive regardless of the percentage of positive cells.
Leptin expression was defined as negative when the staining intensity of neoplastic epithelial cells was weaker than that of adipocytes on the same slide 16 or when less than 5% of the tumor cells showed strong staining intensity. When more than 5% of the tumor cells were strongly stained, these specimens were defined as positive. Only the cytoplasmic expression of leptin in cancer cells was counted. 30
Evaluation of IGF-1 R expression was based on the scheme described in a previous study, with a slight modification. 33 Both cytoplasmic and membranous expressions of antigen were assessed. Specimens with no staining, staining in <10% of neoplastic epithelial cells, or faint staining in >10% of the cells were considered negative. Those specimens with weak to strong complete membrane or cytoplasmic staining in >10% of the cells were regarded as positive.
Statistical Analyses
Pearson’s χ2 test or Fisher’s exact test was performed to evaluate the significance of any association between the expression of each molecule and the following clinicopathological parameters: OHE status, histologic type, grade, lymphatic invasion, and central necrosis. Association of mean age of the subjects according to BCS was evaluated using Student’s t-test. A probability of less than .05 was considered statistically significant. For all statistical analyses, IBM SPSS Statistics software for Windows, version 20 (SPSS, Inc, an IBM Company, Chicago, IL) was used.
Results
Clinical and Histopathological Characteristics
Tissues from a total of 34 intact and 22 neutered female dogs were studied. The mean age of the animals (n = 56) was 9.62 ± 2.53 years (range, 6–17 years). Dogs with a BCS of 2 (n = 3) or 3 (n = 27) were assigned to group 1 (n = 30), while those with a BCS of 4 (n = 20) or 5 (n = 6) were assigned to group 2 (n = 26). The breeds included Maltese (n = 17), Yorkshire Terrier (n = 9), Shih Tzu (n = 6), Poodle (n = 4), Miniature Schnauzer (n = 4), and mixed breeds (n = 4). The breeds of the remaining 12 dogs were Beagle, Cocker Spaniel, Labrador Retriever, Pekingese, Pomeranian, and Spitz. The major histologic types comprised carcinomas arising in a complex adenoma/mixed-tumor type (33.9%, n = 19) and carcinoma-complex type (14.3%, n = 8). On grade assessment, 31 CMCs were classified as grade I, 13 as grade II, and 12 as grade III. Of all the samples, 14.3% (n = 8) exhibited histologic evidence of lymphatic invasion, and 21.4% (n = 12) showed central necrosis within the tumors.
Expression of HR, Aromatase, Leptin, and IGF-1 R
Of all CMCs, 78.6% (n = 44) were HR (ER and/or PR) positive. Aromatase expression was detected predominantly in the cytoplasm of carcinoma cells and occasionally in both carcinoma and stromal cells. The results of IHC staining for aromatase were as follows: no immunoreactivity (32.1%, n = 18/56), weak immunoreactivity (19.6%, n = 11/56; Fig. 1), moderate immunoreactivity (21.4%, n = 12/56), and intense immunoreactivity (26.8%, n = 15/56; Fig. 2). On the basis of the staining intensity and percentage of stained cells, 60.7% (n = 34) of the samples were classed as aromatase positive.
Grade II ductal carcinoma, mammary gland, dog, group 1 (lean or optimal body weight). Weak staining intensity indicates low aromatase expression in the cancer cells. Immunohistochemistry (IHC).
Leptin was detected in the cytoplasm of carcinoma cells. A nuclear staining pattern was also observed in a few samples, but these were regarded as negative. Most of the leptin-positive tumors displayed homogeneous staining intensity within the tumor (Fig. 3), but some samples showed heterogeneous immunoreactivity (Fig. 4), including cells with intense leptin expression next to cells with weak staining. Leptin was stained in 48.2% (n = 27) of the specimens.
We found that 50.0% (n = 28) of the CMCs were positive for IGF-1 R. All of the positive samples showed both cytoplasmic and membranous IGF-1 R expression (Figs. 5 and 6). Intratumoral heterogeneity of staining was observed in 27.9% (n = 10) of the samples.
Correlation of IHC Results With Clinicopathological Parameters
Correlations between the results of IHC staining and clinicopathological features, including histologic grade, evidence of lymphatic invasion, necrosis, and OHE status, are presented in Table 2. The relationship between HR status and expression of aromatase (P = .025), leptin (P = .013), and IGF-1 R (P = .001) was highly significant.
Expression of Aromatase, Leptin, and Insulin-Like Growth Factor 1 Receptor (IGF-1R) According to Variations in Prognostic Factors and Ovariohysterectomy (OHE) Status.
HR, hormonal receptor; HR negative, both estrogen receptor (ER) or progesterone receptor (PR) negative; N, negative; NS, not significant; P, positive. Statistically significant, P < .05.
No associations were found between histologic type, patient age, and expression of the other molecules. However, intratumoral leptin expression had a tendency to increase with animal age. IGF-1 R expression was found much more frequently in well-differentiated (grade I) than in poorly differentiated (grade III) CMCs (P < .001). The proportion of CMCs with evidence of lymphatic invasion was greater in IGF-1R–negative than in IGF-1R–positive CMCs (P = .002).
Effects of BCS on Clinical, Histopathological, and Immunohistochemical Features
Table 3 summarizes the clinical, histopathological, and immunohistochemical features of the 56 CMCs according to BCS. The age of CMC onset was significantly lower in overweight or obese dogs (8.7 ± 1.9 years) than in lean dogs or dogs with optimal body weights (10.4 ± 2.7 years; P = .011). No significant differences in histologic classification, lymphatic invasion, central necrosis, or OHE status were found. However, the number of poorly differentiated grade III CMCs was greater in group 2 (n = 9/26, 34.6%) than in group 1 (n = 3/30, 10%; P = .035).
Pathological Features of Canine Mammary Carcinomas According to Body Condition Score (n = 56).
Values are presented as number (%) unless otherwise indicated. Group 1: lean or optimal body weight; group 2: overweight or obese. HR, hormone receptor; IGF-1R, insulin-like growth factor 1 receptor; NS, not significant (P > .05); SD, standard deviation.
aStudent’s t-test.
bFisher’s exact test.
cPearson’s χ2 test.
Aromatase-positive CMCs were observed more frequently in group 2 (n = 21/26, 80.8%) than in group 1 (n = 13/30, 43.3%; P = .004). Leptin, IGF-1 R, and HR expression was not significantly different between groups.
Discussion
Sex steroid hormones, including estrogen and progesterone, play important roles in the occurrence and progression of human breast cancer and CMCs. 24,29,41 In this regard, obesity could be an important risk factor for CMC development because adipose tissue and a high content of cholesterol can be sources of steroid hormones such as progesterone and androgens. 13,14,34 In particular, peripheral aromatization of androgens into estrogens could result in prolonged exposure of mammary tissue to estrogens. 25 Indeed, local concentrations of estrogens in breast tumors are 10 times greater than those in the circulation of postmenopausal patients due to tumor-adipocyte interactions. In addition, aromatase expression is significantly higher in canine mammary tumors than in normal mammary tissues. 23 These studies provide a basis for the hypothesis that obesity can be a risk factor for mammary gland tumors in both postmenopausal women and female dogs.
In this study, we analyzed the expression of obesity-related molecules in CMCs to find substantial evidence of a correlation between obesity and the clinicopathologic features of CMCs. First, we found that the age of CMC onset was lower in the overweight or obese group than in the lean or optimal body weight group. In addition, the proportion of grade III tumors was higher in the overweight or obese group, suggesting that the BCS might be involved in the onset and progression of CMCs. Furthermore, aromatase-positive CMCs were observed more frequently in the overweight or obese group (n = 21/26, 80.8%) than in the lean or optimal body weight group (n = 13/30, 43.3%; P = .004). This suggested that aromatase could have an important role in overweight and obese patients.
In human cancer research, breast cancers have been categorized generally into several molecular subtypes, including luminal A (ER/PR positive and HER-2 negative), luminal B (ER/PR and HER-2 positive), HER-2 overexpressing (ER/PR negative and HER-2 positive), and triple negative (all negative). 36 In accordance with these categories, tissue samples were divided into HR-positive and double-negative groups. The results showed that the incidence of aromatase-positive CMCs was higher in the HR-positive group (any ER- or PR-positive and both ER- and PR-positive CMCs) than in the HR-negative group (P = .025).
Extreme body adiposity can lead to the elevation of plasma leptin concentrations in humans, rodents, and dogs. 31 Leptin is essential for normal mammary gland development, but it can be expressed in CMCs as well. 30 Leptin is known as one of the growth factors in breast cancer because it has mitogenic effects and affects the transformation of breast cancer cells to more aggressive phenotypes. In the current study, we confirmed that the expression of leptin was not directly associated with overweight/obesity status. However, we demonstrated that leptin expression was significantly higher in HR-positive than in HR-negative CMCs (P = .013). The relationship between HR and leptin supports the idea that leptin has a possible role in CMC progression because of its involvement with obesity and sex steroid hormones.
Obesity often leads to insulin resistance in both humans and dogs. 10 In obese humans and dogs, hyperinsulinemia, and consequently increased IGF-1 levels, may lead to high mortality in breast cancer because of rapid tumor growth. 12,27 In particular, the coexpression of ER and IGF-1 R can stimulate the proliferation of human breast cancer cells synergistically. 37 In the current study, however, IGF-1 R expression was not significantly different between groups. Although the BCS was not directly correlated with the expression of IGF-1 R, the expression of HRs was higher in the IGF-1R–positive CMCs (P = .001). This finding suggested that the expression of HRs and IGF-1 R is closely related to each other. It has also been reported that the IGF-1/IGF-1 R pathway may enhance cellular responses to the mitogenic and antiapoptotic effects of IGF-1 in breast cancer cells. 19,28 In the case of dogs, some reports suggested that high IGF-1 R expression was associated with poor prognostic factors such as a high histologic grade and histologic types with a worse prognosis. 8 In the current study, however, IGF-1 R expression was significantly lower in high-grade CMCs (P < .001) and CMCs with lymphatic invasion (P = .002), which is consistent with a previous study in humans. 28
Conclusion
This study demonstrates the potential of investigating molecular links between obesity and CMCs using histopathology and IHC. Among the obesity-related molecules used in the study, only the expression of intratumoral aromatase increased in the overweight/obese group. The expression of aromatase, leptin, and IGF-1 R was associated with the HR status. Overweight or obese female dogs had an earlier onset and higher histologic grade of CMCs than did animals with lean or optimal body weights. Our findings suggest that overweight or obese status is a risk factor for the early onset and higher grade of CMCs and that increased aromatase expression in overweight or obese female dogs may affect the development of CMCs through HR signaling.
Footnotes
Acknowledgements
The authors thank Ms. Rae-Hwa Jang for her excellent technical assistance. This report represents a part of the PhD thesis of Ha-Young Lim.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Basic Science Research Program of the National Research Foundation of Korea (NRF) and was funded by the Ministry of Science, ICT, and Future Planning (2014R1A2A2A01003470).