Breast cancer is the most frequent cause of cancer death in women in developing regions including Egypt (Jemal et al., 2011; Ibrahim et al., 2014). The variability in the clinical course of patients with breast cancer poses a challenge for clinicians regarding the choice of optimal treatment based on outcome prediction. Prognostic and predictive biomarkers could play a vital role in accomplishing this task (Farr et al., 2013).
Many studies have identified that the interaction between stromal components and tumor cells in breast carcinomas is a critical step in tumor progression. A better understanding of stromal components and its contributions to cancer progression will identify specific signals that promote invasion and metastasis. Eventually, this will result in the identification of new therapeutic targets for future treatment (Wever and Mareel, 2003; Troester et al., 2009; Puri et al., 2011; Salama et al., 2015).
Matrix metalloproteinases (MMPs) are a large family of calcium-dependent, zinc-containing endopeptidases. They are responsible for tissue remodeling and cleavage of extracellular matrix proteins. The activities of MMPs are usually tightly controlled, and they are minimally expressed in normal physiological conditions (Rakashanda et al., 2012).
In tumors, MMPs promote tumor progression through their role in formation of active transforming growth factor-β, a cytokine produced by carcinoma-associated fibroblasts, which stimulates tumorigenesis and angiogenesis (Curran and Keely, 2013). In addition, some growth factors and matrix components cleaved by MMPs (e.g. collagen, laminin) promote tumor cell migration through their chemotactic activity for tumor cells. Thus, expressions of MMPs are increased in aggressive tumors, and this mostly correlates with decreased overall survival (Kass et al., 2007).
CD10, also known as neprilysin and common acute lymphoblastic leukemia/lymphoma antigen, is the prototype of the MMP family. It is a cell-surface, zinc-dependent metalloproteinase produced by myofibroblasts. The main function of CD10 is its central role in tissue remodeling (Maguer-Satta et al., 2011).
Physiologically, CD10 is widely expressed in various tissues (Iezzi et al., 2008). The pathological role of CD10 was first discovered in acute lymphoblastic leukemia. Moreover, it was found to be expressed in B-lymphoblastic leukemia/lymphoma and in certain mature B-cell lymphomas and very rarely in T-cell lymphomas (Greaves et al., 1975; Maguer-Satta et al., 2011).
In breast tissue, CD10 is considered a specific marker of the myoepithelial cells, and is constantly expressed during development and after maturation by the myoepithelial cells. However, other components of adult breast including luminal epithelial cells and surrounding fibroblasts do not express CD10 (Sanchez-Cespedes et al., 2013).
Recently, a few studies have investigated the prognostic role of stromal CD10 expression in breast carcinomas (Iwaya et al., 2002; Makretsov et al., 2007; Kim et al., 2010; Puri et al., 2011; Hosni et al., 2012; Mohammadizadeh et al., 2012; Jana et al., 2014; Taghizadeh-Kermani et al., 2014; Vo et al., 2015; Anuradha Devi et al., 2016; Sadaka et al., 2016). The present study was designed to evaluate the immunohistochemical stromal expression of CD10 in breast carcinoma and to correlate its expression with clinicopathological parameters and patients’ clinical outcome.
Patients and methods
This study comprised 60 Egyptian female patients who presented with primary invasive carcinoma of no special type (IC-NST) during the period between January 2004 and July 2011.
Exclusion criteria included the following: patients who received preoperative neoadjuvant chemotherapy, studies with follow-up period less than 2 years, cases of equivocal HER2/neu immunostaining (score 2+), which were not evaluated by fluorescence in-situ hybridization test, and insufficient tissue for immunostaining.
Follow-up data including documentation of the treatment modality and disease relapse of patients were collected from the files of the Clinical Oncology and Nuclear Medicine Department, Faculty of Medicine, Alexandria University.
Data on gross examination were collected from pathology reports from the files of the Pathology Department, Faculty of Medicine, Alexandria University. Formalin-fixed, paraffin-embedded tissue specimens were cut into 5-μm-thick sections, and were stained with hematoxylin and eosin for light microscopic examination. Grading and TNM staging were carried out according to established criteria (Edge et al., 2010; Lakhani et al., 2012).
An avidin–biotinylated immunoperoxidase methodology was used for both CD10 and HER2/neu immunostaining. CD10 was evaluated using the ready-to-use mouse monoclonal primary antibody CD10 (clone 56C6) (antihuman CD10/common acute lymphoblastic leukemia/lymphoma antigen; Thermo Scientific, Fremont, California, USA). HER2/neu was evaluated using a mouse monoclonal primary HER/neu antibody (clone e2-4001+3B5; Thermo Scientific) at a dilution of 1 : 200.
Negative and positive controls (reactive lymph node: for CD10 and a known case of invasive breast carcinoma positive for HER2/neu score +3; for HER2/neu) were included in all runs.
Assessment and evaluation of the intensity and extent of CD10-positive immunostaining were carried out in the stroma. Stromal CD10 expression was evaluated as three categories: negative (no staining), weak positive staining defined as focal weak staining or diffuse weak staining or strong focal staining in less than 30% of stromal cells, and strong positive staining defined as strong staining of greater than or equal to 30% of stromal cells (Makretsov et al., 2007).
HER2/neu staining was assessed according to the guidelines of https://www.ncbi.nlm.nih.gov/pubmed/19548375 (Wolff et al., 2007). Expression of HER2/neu was scored from 0 to 3. Scores were interpreted as follows: negative (0 and 1+), equivocal (2+), and positive (3+).
Information on hormonal receptor status was obtained from the pathological records and reviewed according to published guidelines (Hammond et al., 2010). Both average intensity and extent of staining were reported, and the Allred scoring system was used (Collins et al., 2005). Molecular subtypes were determined according to the hormonal receptors and the HER2/neu status (Nishimura et al., 2010).
Data were collected, coded, and analyzed using IBM SPSS (version 21, 2012; International Business Machines Corp., Boston, Massachusetts, USA) for Windows XP. The following statistical tests were used: the Student t-test and the Mann–Whitney test for abnormally distributed data; one-way analysis of variance test or the Kruskal–Wallis H-test for continuous variables; and post-hoc tests (Hochberg or Games-Howell) for pairwise comparison of significant results. The effect of different categorical variables of CD10 expression on median time till recurrence was tested by survival analysis using log-rank test with χ2 distribution. The median survival time was summarized by the Kaplan–Meier curve. The cutoff point to detect significance (P) was 0.05 or less.
The present study included 60 female patients. Their ages ranged from 30 to 68 years, with a mean value of 49.167 and SD of 8.31. The clinicopathological characteristics are summarized in Table 1.
Out of the studied 60 IC-NST cases, 18.3% showed negative stromal expression (Fig. 1a). Positive stromal CD10 expression was detected in 81.7%, of which 65% showed weak positive staining (Fig. 1b and c), and 16.7% showed strong positive staining (Fig. 1d). The nearby fibrocystic areas and the adjacent normal breast tissue showed negative CD10 stromal expression. The strong positive expression of CD10 in adjacent non-neoplastic myoepithelial cells was used as a built-in internal control. In addition, there was no expression of CD10 in normal ductal cells (Fig. 1e and f).
Correlations between different CD10 expression categories and demographic and clinicopathological variables (Table 1) showed that there was a statistically significant association between different CD10 expression categories and different tumor grades (P=0.000). Pairwise comparison showed that grade 3 tumors showed a significantly higher stromal CD10 expression compared with grade 1 tumors.
In addition, a statistically significant, positive correlation was found between stromal CD10 expression and presence of necrosis (P=0.021) (Table 1).
In contrast, the correlation between CD10 expression categories and other clinicopathological features was statistically insignificant (Table 1) (F=1.973, P=0.148).
Stromal CD10 expression showed a statistically significant inverse correlation with estrogen receptor (ER) and progesterone receptor (PR) expressions (P=0.029 and 0.006, respectively). In addition, a statistically significant inverse correlation between CD10 expression categories and Allred scores of ER and PR was found (P=0.020 and 0.005, respectively). Correlation with molecular subtypes showed that CD10 was significantly overexpressed in the triple-negative group (P=0.015) (Table 2).
On the other hand, there was no statistically significant association between different CD10 expression categories and HER2/neu immunostaining (Table 2).
The follow-up period ranged from 36 to 120 months with a mean of 61and SD of 24.432 (median=48). CD10 upregulation was significantly associated with a shorter median disease-free survival time, development of local and distant recurrence, and occurrence of a new primary in the contralateral breast (P=0.016, 0.013, and 0.011, respectively) (Table 3).
According to the log-rank test, there was a statistically significant difference in median survival time between cases showing stromal CD10 expression (median=29 months) and those showing negative CD10 expression (median=60 months) (Fig. 2 and Table 3).
Breast tumor progression is a multistep process starting with epithelial hyperplasia that progresses to in situ, invasive, and metastatic carcinomas. Despite the malignant epithelial nature of invasive breast carcinoma, the central role of the tumor microenvironment in cancer progression is currently elucidated (Wever and Mareel, 2003; Schedin and Elias, 2004; Albini and Sporn, 2007).
Stromal markers are now emerging as novel markers in assessing the prognosis of invasive breast cancer, such as CD10 (Makretsov et al., 2007). The present study was carried out to evaluate the stromal expression of CD10 in breast carcinoma, and correlate its expression with different clinicopathological parameters.
In the present study, 60 cases of IC-NST were immunohistochemically studied for stromal CD10 expression. It was found that the nearby fibrocystic areas and the adjacent normal breast tissue showed strong positive expression of CD10 in myoepithelial cells with lack of expression of CD10 in normal ductal cells and stroma. Similar results were documented by other investigators (Iwaya et al., 2002; Makretsov et al., 2007; Kim et al., 2010; Mohammadizadeh et al., 2012; Taghizadeh-Kermani et al., 2014; Vo et al., 2015; Anuradha Devi et al., 2016).
In the present study, 81.7% of cases showed positivity for CD10 immunostaining in the stroma surrounding the tumor cells. Among the positive cases, 65% showed weak positive staining and 16.7% showed strong positive staining.
These findings are in concordance with some studies that showed stromal CD10 expression in 71.4, 79, 80, 81, and 81.6% of IC-NST cases, respectively (Makretsov et al., 2007; Puri et al., 2011; Hosni et al., 2012; Mohammadizadeh et al., 2012; Anuradha Devi et al., 2016).
On the other hand, lower percentages of CD10 positivity were documented in other studies including 18, 36.1, 48.6, 49.5, and 53.4% of IC-NST cases, respectively (Iwaya et al., 2002; Kim et al., 2010; Jana et al., 2014; Vo et al., 2015; Sadaka et al., 2016).
This discrepancy in the results might be due to the different cutoff points in assessing the positivity used in these studies. Our study along with some of these previous studies (Makretsov et al., 2007; Hosni et al., 2012; Mohammadizadeh et al., 2012; Taghizadeh-Kermani et al., 2014) did not use a cutoff point for negativity, whereas others have used different cutoff points to judge CD10 expression in stromal cells. These cutoff points ranged from 10 (Iwaya et al., 2002; Kim et al., 2010; Puri et al., 2011; Jana et al., 2014; Anuradha Devi et al., 2016; Sadaka et al., 2016) to 20% (Vo et al., 2015).
The relatively high expression of stromal CD10 in the studied cases of IC-NST and its total lack of expression in the adjacent non-neoplastic breast tissue suggested that CD10 stromal expression may be implicated in breast cancer tumorigenesis. This was also explained by the model proposed by Maguer-Satta et al. (2011).
In the present study, a statistically significant association was found between the different CD10 expression categories and tumor grades (P=0.000). Strong positive CD10 expression was significantly associated with a higher tumor grade. All included cases of grade 3 tumors (100%) showed CD10 expression with a different intensity in comparison with those of grade 1 tumors, which showed weak CD10 expression in only one (20%) case.
This result is in agreement with those reported by other authors (Makretsov et al., 2007; Kim et al., 2010; Hosni et al., 2012; Mohammadizadeh et al., 2012; Jana et al., 2014; Taghizadeh-Kermani et al., 2014; Anuradha Devi et al., 2016; Sadaka et al., 2016). Thus, a stronger CD10 expression in a higher tumor grade may suggest a role of CD10 in tumor dedifferentiation and aggressiveness.
In contrast to our findings, Iwaya et al. (2002), Puri et al. (2011), and Vo et al. (2015) reported that there was no statistically significant correlation between CD10 expression and different tumor grades. The lack of standardized methodology for measuring stromal CD10 expression and the use of different cutoff points might explain these different findings.
In addition, a statistically significant, positive correlation was found between stromal CD10 expression and presence of necrosis (P=0.021). However, no statistically significant correlation between CD10 expression categories and demographic and other clinicopathological parameters was found.
Some researchers (Iwaya et al., 2002; Kim et al., 2010; Hosni et al., 2012; Mohammadizadeh et al., 2012; Jana et al., 2014; Vo et al., 2015) reported that there was no statistical significant correlation between stromal CD10 expression and patient’s age. In addition, similar results were reported by other authors (Taghizadeh-Kermani et al., 2014; Anuradha Devi et al., 2016; Sadaka et al., 2016) regarding patient age as well as menstrual status.
Considering lymph node status and tumor size, our results are in agreement with other investigators (Makretsov et al., 2007; Hosni et al., 2012; Jana et al., 2014) who reported that there was no correlation between stromal CD10 expression categories and lymph node status and tumor size.
In contrast to our finding, lymphovascular invasion was reported to be significantly associated with stromal CD10 expression by Sadaka et al. (2016).
Considering the pathological TNM staging, Jana et al. (2014) and Iwaya et al. (2002) reported that the pathological TNM staging did not have any correlation with positive CD10 expression. These findings are also in agreement with the results of the present study.
On the other hand, Kim et al. (2010) reported overexpression of CD10 in advanced tumor stages.
This discrepancy in the findings may be explained by the great variability in the number of studied cases in different studies. In addition, the used methodology for immunostaining was different regarding antigen retrieval and different antibody concentrations.
In the 60 studied IC-NST cases, a statistically significant inverse correlation was found between different CD10 expression categories and ER and PR expressions (P=0.029 and 0.006, respectively).
This is in accordance with the findings reported by Kim et al. (2010) and Sadaka et al. (2016) who reported a strong inverse relationship between CD10 expression and ER and PR expressions.
In the present study, an additional statistical test was performed to correlate the different CD10 expression categories with the Allred scores of ER and PR, and a statistically significant inverse correlation was found between different CD10 expression categories and ER and PR expressions (P=0.020 and 0.005, respectively).
Regarding HER2/neu staining, no statistically significant association was found between the different CD10 expression categories and the HER2/neu status (P=0.432).
These findings are in accordance with those reported by Makretsov et al. (2007), Mohammadizadeh et al. (2012), and Vo et al. (2015), who reported that there was no statistically significant correlation between the different CD10 expression categories and the HER2/neu status.
In contrast, Puri et al. (2011), Jana et al. (2014), and Sadaka et al. (2016) detected a statistically positive correlation between stromal CD10 expression and HER2/neu status.
In the present study, a statistically significant correlation was reported between the different CD10 expression categories and the molecular subtypes (P=0.015), and it was overexpressed in triple-negative tumors and less expressed in luminal A tumors.
A single previous study by Jana et al. (2014) investigated the relationship between CD10 expression and different molecular subtypes and reported similar results of low-level CD10 expression in the luminal A subtype.
These findings could be explained by the fact that triple-negative tumors are mostly associated with phosphatase and tensin homolog loss. However, this event is absent in luminal tumors (Maguer-Satta et al., 2011).
In the present study, a stronger CD10 expression was associated with increased incidence of recurrence and development of a new primary in the contralateral breast (P=0.013 and 0.011, respectively). Thus, positive stromal CD10 expression could be considered as a potential poor prognostic factor in breast cancer and might help identify a group of patients with a higher propensity to tumor recurrence, which might aid in patient management.
In line with our results, Vo et al. (2015) reported a strong positive correlation between CD10 expression and poor clinical outcome. They suggested that CD10 expression could be applied as a prognostic factor, independent of the regional lymph node and ER status in IC-NST.
In the present study, the log-rank test was performed, and it showed a statistically significant correlation between short-term disease-free survival and stromal CD10 expression (χ2log-rank=5.77, P=0.016).
This result is in concordance with the results reported by Iwaya et al. (2002), Makretsov et al. (2007), and Sadaka et al. (2016), who reported a decreased, long-term, disease-specific overall survival with positive stromal CD10 expression.
From the present study, it can be suggested that stromal CD10 expression may be implicated in breast cancer tumorigenesis. It was associated with poor prognostic indicators – namely, higher tumor grade, presence of necrosis, negative hormonal receptor status, and triple-negative molecular subtype, which also suggests that CD10 may contribute to tumor aggressiveness and progression.
This can be supported by the documented role of CD10 in the protein kinase B pathway (or Akt pathway). During tumor progression, enzymatic activity of CD10 induces cleavage of growth factors such as fibroblast growth factor 2 as well as blockage of phosphatase and tensin homolog functions. These events lead to Akt signaling, which favors apoptosis inhibition, cell proliferation, endothelial cell growth, and angiogenesis (Maguer-Satta et al., 2011; Cancer Genome Atlas Network, 2012).
However, these results need to be supported by further studies with larger sample sizes to further elucidate the impact of CD10 on breast cancer patients’ outcome.
Conflicts of interest
There are no conflicts of interest.
Albini A, Sporn MB (2007). The tumor microenvironment as a target for chemoprevention. Nat Rev Cancer 7:139–147.
Anuradha Devi BS, Sekhar SC, Saritha C, Anil SS, Rani HS (2016). A study on stromal CD10 expression in invasive breast carcinoma. IAIM 3:142–147.
Cancer Genome Atlas Network (2012). Comprehensive molecular portraits of human breast tumours. Nature 490:61–70.
Collins LC, Botero ML, Schnitt SJ (2005). Bimodal frequency distribution of estrogen receptor immunohistochemical staining results in breast cancer. Am J Clin Pathol 123:16–20.
Curran CS, Keely PJ (2013). Breast tumor and stromal cell responses to TGF-β and hypoxia in matrix deposition. Matrix Biol 32:95–105.
Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A (2010). AJCC cancer staging manual, 7th ed. New York: Springer.
Farr A, Wuerstlein R, Heiduschka A, Singer CF, Harbeck N (2013). Modern risk assessment for individualizing treatment concepts in early-stage breast cancer. Rev Obstet Gynecol
Greaves MF, Brown G, Rapson NT, Lister TA (1975). Antisera to acute lymphoblastic leukemia cells. Clin Immunol Immunopathol 4:67–84.
Hammond MEH, Hayes DF, Dowsett M, Allred DC, Hagerty KL, Badve S, et al (2010). American Society of Clinical Oncology/College of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer. J Clin Oncol 28:2784–2795.
Hosni HN, Abd ELAziz A, Tabak SA, ELSayed M (2012). Immunohistochemical study of stromal CD10 expression in mammary duct carcinoma. Med J Cairo Univ 80:37–44.
Ibrahim AS, Khaled HM, Mikhai NN, Baraka H, Kamel H (2014). Cancer incidence in Egypt: results of the national population-based cancer registry program. J Cancer Epidemiol 2014:1–18.
Iezzi G, Piattelli A, Rubini C, Artese L, Goteri G, Fioroni M, Carinci F (2008). CD10 expression in stromal cells of ameloblastoma variants. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 105:206–209.
Iwaya K, Ogawa H, Izumi M (2002). Stromal expression of CD10 in invasive breast carcinoma: a new predictor of clinical outcome. Virchows Arch 440:589–593.
Jana SH, Jha BM, Patel C, Jana D, Agarwal A (2014). CD10-A new prognostic stromal marker in breast carcinoma, its utility, limitations and role in breast cancer pathogenesis. Indian J Pathol Microbiol 57:530–536.
Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D (2011). Global cancer statistics. CA Cancer J Clin 61:69–90.
Kass L, Erler JT, Dembo M, Weaver VM (2007). Mammary epithelial cell: influence of extracellular matrix composition and organization during development and tumorigenesis. Int J Biochem Cell Biol 39:1987–1994.
Kim HS, Kim GY, Kim YW, Park YK, Song JY, Lim SJ (2010). Stromal CD10 expression and relationship to the E cadherin/beta-catenin complex in breast carcinoma. Histopathology 56:708–719.
Lakhani SR, Ellis IO, Schnitt SJ, Tan PH, van de Vijver MJ (2012). WHO classification of tumours of the breast
Vol 4, 4th ed. Lyon, France: IARC Press.
Maguer-Satta V, Besançon R, Bachelard-Cascales E (2011). Concise review: neutral endopeptidase (CD10): a multifaceted environment actor in stem cells, physiological mechanisms, and cancer. Stem Cells 29:389–396.
Makretsov N, Hayes M, Carter B (2007). Stromal CD10 expression in invasive breast carcinoma correlates with poor prognosis, estrogen receptor negativity, and high grade. Mod Pathol 20:84–89.
Mohammadizadeh F, Salavati M, Moghaddam NA (2012). CD10 expression in stromal component of invasive breast carcinoma: a potential prognostic determinant. J Res Med Sci 2:194–199.
Nishimura R, Osako T, Okumura Y, Hayashi M, Toyozumi Y, Arima N (2010). Ki-67 as a prognostic marker according to breast cancer subtype and a predictor of recurrence time in primary breast cancer. Exp Ther Med 1:747–754.
Puri V, Jain M, Thomas S (2011). Stromal expression of CD10 in invasive breast carcinoma and its correlation with ER, PR, HER2-neu, and Ki67. Int J Breast Cancer 2011:1–4.
Rakashanda S, Rana F, Rafiq S, Masood A, Amin S (2012). Role of proteases in cancer. Biotechnol Mol Biol Rev 7:90–101.
Sadaka E, Almorsy W, Elsaka A (2016). CD10 expression as a prognostic factor in female patients with invasive ductal carcinoma of the breast. J Am Sci 12:71–77.
Salama ME, Ibrahim WS, Khairy DA (2015). Comparison of CD10 expression in stroma of epithelial and mesenchymal tumors of the breast. Global Adv Res J Med Med Sci 4:51–56.
Sanchez-Cespedes R, Suarez-Bonnet A, Millan Y, Guil-Luna S, Reymundo C, Herráez P, et al (2013). Use of CD10 as a marker of canine mammary myoepithelial cells. Vet J 195:192–199.
Schedin P, Elias A (2004). Review: multistep tumorigenesis and the microenvironment. Breast Cancer Res 6:93–101.
Taghizadeh-Kermani A, Jafarian AH, Ashabyamin R, Seilanian-Toosi M, Pourali L, Asadi M, et al (2014). The stromal overexpression of CD10 in invasive breast cancer and its association with clincophathologic factors. Iran J Cancer Prev 7:17–21.
Troester MA, Lee MH, Carter M, Fan C, Cowan DW, Perez ER, et al (2009). Activation of host wound responses in breast cancer microenvironment. Clin Cancer Res 15:7020–7028.
Vo TD, Mekata E, Umeda T, Abe H, Kawai Y, Mori T, et al (2015). Prognostic impact of CD10 expression in clinical outcome of invasive breast carcinoma. Breast Cancer 22:117–128.
Wever OD, Mareel M (2003). Role of tissue stroma in cancer cell invasion. J Pathol 200:429–447.
Wolff AC, Hammond MEH, Schwartz JN, Hagerty KL, Allred DC, Cote RJ, et al (2007). American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol 25:118–145.