Relation between Ki-67 , ER , PR , Her 2 / neu , p 21 , EGFR , and TOP II-α Expression in Invasive Ductal Breast Cancer Patients and Correlations with Prognosis

Breast cancer (BC) is the most frequent malignancy in women, and the main cause of death in women aged 35-55 years old. It is a great burden to society. Despite the improvement of diagnostic methods and chemotherapeutic regimens, the 5-year overall survival (OS) rates of patients significantly depend on the stage of BC and were 56.2% in Ukraine and 88.0% in the USA from 2000-2005 (Cancer in Ukraine, 2010).Clinical characteristics such as age, tumor size, menstrual status, morphology, and lymph node status of the tumor are traditionally the most important prognostic factors. However, research on the tumor’s molecular features led to a greatly improved prognosis of the disease course. Most molecular markers, such as Ki-67, p21, EGFR, and TOP II-α, determine the ability of cells to grow malignantly (Hanahan, 2000). For BC, the estrogen and progesterone receptors (ER and PR) were the first predictive molecular markers. Patients who are determined to be positive for these steroid


Introduction
hormone receptors (HRs) generally have high sensitivity to hormone therapy. The next marker included in clinical practice was HER-2. A positive HER-2 status in patients with BC correlates with high sensitivity to targeted therapy with trastuzumab. These biomarkers play an important role in clinical practice by individualizing treatment and determining adequate chemotherapeutic schemes for patients. However, 25-50% of ER and PR positive tumors are resistant to hormone therapy (Klijin, 1991), and tumors with HER-2 overexpression do not always respond to trastuzumab therapy. In addition, about 20% of patients with BC are negative for all three markers mentioned above, and their tumors are more resistant to traditional therapy schemes.
Ki-67 is a nuclear non-histone protein, and a proliferation marker expressed in the G1, S, G2, and M-phases of the cell cycle, but not the G0 phase. Ki-67 is a large nuclear protein that is thought to be involved in cellular functions such as cell cycle regulation, ribosomal RNA processing, and organizing DNA, or to have a structural role within the nucleus (Szelachowska et al., 2012). Patients with Ki67 expression have a worse outcome in a wide variety of malignancies such as BC (Pathmanathan et al., 2013), sarcomas (Sorbye et al., 2012), lymphomas (Kanavaros et al., 2001), and gliomas (Arshad et al., 2010). Multiple published studies have observed a significant association between Ki-67 expression and clinical outcome for BC (Agboola et al., 2013;Meattini et al., 2013;Inwald et al., 2013); although, opposite findings are also reported. Cell cycle progression is regulated by cyclin-dependent kinases and their specific inhibitors (Sherr et al., 1999), includingp21WAF1/CIP1 (p21), which is a universal inhibitor of cyclin-dependent kinases (Xiong et al., 1993). It halts cell cycle progression in response to various stimuli (Bartek et al., 2007). A p21 deficiency may increase carcinogenesisin combination with other oncogenic mutations (Martín-Caballero et al., 2001;Poole et al., 2004;Carnero et al., 2004;Yang W et al., 2005). Some studies indicate that p21 can negatively modify the activity of other positive regulators of the cell cycle (CDKs), and the increasing expression of p21in cancer is consistent with the theory that abnormal cells increase p21 expression in an attempt to "brake" the process of cellular proliferation at the G1 checkpoint (Thor et al., 2000). Ashrafi found that expression of cyclinD1 and p21 was increased in BC tissues compared to normal mammary glands (Ashrafi et al., 2012). TOPisomerse IIα(Top II-α) reduces DNA supercoiling and twisting and is important for chromosome segregation and condensation in dividing cells. Its expression is highest in the G2/M phase of exponentially growing cells, and some studies found that overexpression predicts shorter disease free survival (DFS) and OS (Midura-Nowaczek et al., 2013). High Top II-α RNA levels are significantly associated with shorter metastasis-free survival in node-negative BC, but associated with a high frequency of pathological complete response in anthracycline-treated patients .
It has been confirmed that the expression of Ki-67, p21, EGFR, and TOP II were altered in BC tissues compared to normal mammary glands, and this correlated with prognosis. It is still unclear whether a correlation exits between these markers and ER, PR, and Her2/neu status. Our study aimed to investigate expression of Ki-67, ER, PR, Her2/neu, p21, EGFR, and TOP II-α in the tumor tissue of patients with invasive ductal carcinoma (IDC), as well as to analyze correlations between these markers. Our results indicated that Ki-67 expression correlates with ER, PR, HER-2, EGFR, TOP II-α, lymph node involvement, high tumor grade, lymphovascular invasion, high tumor stage, and high TNM stage in IDC. Positive Ki-67 expression was a risk factor for rapid tumor recurrence and may help tumor progression leading to a poor prognosis in IDC.

Patient selection and data collection
Specimens were obtained from 216 patients with IDC in The Third Affiliated Hospital of Southern Medical University from September 2005 to September 2010. Patients who received chemotherapy or radiotherapy before surgery and those who underwent surgery for other types of carcinomas were excluded. The clinicopathological features of these patients were assessed. This study was approved by the Institutional Review Board of the Key Lab of Breast Disease Research in Guangdong Province. All patients provided written informed consent.

Immunohistochemical and image analysis
IDC samples and paired paracancerous tissues from216 patients were analyzed by immunohistochemistry and relative optical density (OD) calculations, along with normal control tissues. Briefly, formalin-fixed and paraffin-embedded sections (4μm thick) were cut and mounted on3-aminopropyltriethoxysilane-treated slides. Slides were routinely deparaffinized with xyleneand rehydrated in ethanol washes. Non-enzymatic antigen retrieval was performed by microwave heat treatment 3 times for 7 min each in 0.01 M sodium citrate buffer (pH 6.0). Endogenous peroxidase activity and nonspecific background staining were blocked by incubating the slides in methanol containing 0.3% H 2 O 2 for 30 min. The slides were subsequently washed with PBS for 15 min and then incubated with primary antibody (anti-Ki-67, anti-p21, anti-EGFR, anti-TOP II-, In vitrogen, Camarillo, CA) for 1 h at 37°C. The working dilution of the primary antibody was 1:50. The sections were rinsed with PBS for 15 min and then incubated with ENVISION+ rabbit/horseradish peroxidase for 45 min. After washing with PBS for 15min, the final products were visualized using the 3-amino-9ethylcarbazole substrate system, and the sections were counterstained with Mayer's hematoxylin for 20 s before mounting. Both positive and negative controls were performed for each section. Normal rabbit serum IgG was used in place of the primary antibody as a negative control. All experiments were performed in duplicate.

Result analysis
The stained sections were observed under a light microscope (400×) (Olympus, Japan), and digital images were analyzed with the HPIAS-2000 multimedia colored pathology image analysis system. To determine the OD and the percentage of positive protein in the staining area, 10 complete sections without overlapping fields were randomly selected for quantitative analysis. Each section was evaluated by 3 different researchers, and the mean value of each slice was calculated.

Follow-up
After surgery, patients were regularly followed at the outpatient clinics. Every 2 to 3 months patients were monitored by abdominal and chest CT and ultrasound in the first two postoperative years, and then every5 to 6 months in the following years. In the final years, an annual abdominal CT scan was performed. CT or MR imaging of the abdomen, pelvis, and chest was performed and a bone scan was used as monitoring method to define suspicious lesions demonstrated on CT. Survival time was calculated starting with the month the surgical intervention took place to the month of final follow-up. Complete data were recorded for patients that died from IDC during this DOI:http://dx.doi.org/10.7314/APJCP.2015.16.2.823 Ki-67, ER, PR, Her2/neu, p21, EGFR, and TOP II-α Expression in Invasive Ductal Breast Cancer Correlates with Prognosis period; while, cases of survival and loss to follow-up were recorded as censored data.

Statistical analysis
The data are presented as the means ± S.D. Student's t-test and Pearson's chi-square (χ 2 )test were used for statistical analyses. The logistic regression model was used to identify independent risk factors. The survival rate of IDC patients was analyzed with Kaplan-Meier survival curves and the log-rank (Mantel-Cox) statistical method. A multivariate Cox regression model was used to identify the risk factors for DFS. Correlation analysis was used to determine the correlation between Ki-67 OD and the prognosis of IDC patients. SPSS 19.0 for Windows was used for the statistical analyses. A value of p<0.05 was considered statistically significant.

Analysis of Ki-67 and its correlation with other clinicopathological factors
The Ki-67 status could be determined in 135(62.5%) of the original 216 patients, which correlated with other clinicopathological factors (Table 3). There was an inverse correlation between Ki-67 positive expression and ER and PR positive expression (p<0.01; r=-0.263 and p<0.01; r=-0.274, respectively). As shown in Table 1, there was a direct correlation between Ki-67 and HER-2, EGFR, and TOP II-α positive expression, as well as lymph node

Comparison of Ki-67 positive rates between different groups
According to the HR and HER-2 status, the original 216 cases of invasive ductal BC were divided into four groups (Figure 2) to compare the positive rates of Ki-67 between different groups. Patients who were HR positive (ER and/or PR positive) and HER-2 negative were in Group A (n=95). Patients who were HR positive and HER-2 positive were in Group B (n=56). Patients who were HR negative (ER and PR negative) and HER-2 positive made up Group C (n=37). Patients who were HR negative and HER-2 negative made up Group D (n=28) ( Table 2).
The expression of Ki-67 was higher in the HER-2 positive group (group C) (82.9%) than the other 4 groups: 73.7% in the triple negative group (group D), 65.2% in group B (HR positive, Her-2 positive), and 43.6% in group A (HR positive) (Table 4). There were significant differences in theKi-67 positive rates in group A vs B, A vs C, A vs D, and B vs C (all p<0.05) ( Table 3). There were no significant comparisons in group B vs D and group C vs D (p>0.05). After comparing positive rates of Ki-67 between groups, only the HR positive group (group A) showed significant differences with all the groups.

Expression of Ki-67 in different groups and their relationship with other clinicopathological factors
In the HR positive group (Group A), there were direct correlations between Ki-67 positive rates and expression of EGFR and TOP II-α, lymph node involvement, high grade tumor, lymphovascular invasion, and high TNM stage (all p<0.01; r=0.264, 0.519, 0.236, 0.313, 0.356, and 0.273, respectively) ( Table 4).There was no relationship between expression of Ki-67 and p21 and high tumor stage in this group (p=0.057 and 0.098; r=0.181 and 0.164, respectively).In the HR positive and HER-2 positive group(Group B), the Ki-67 positive status directly correlated with TOP II-α,lymph node involvement, and high tumor grade (p<0.01, r=0.629; p<0.05, r=0.283; p<0.05, r=0.273, respectively) ( Table 4). There were no relationships between expression of Ki-67 and p21, EGFR, lymphovascular invasion, high tumor stage, and high TNM in this group.

Patient demographics and outcomes
To address whether HR, HER-2 status, and Ki-67 positive rates were associated with the patients' IDC progression, we analyzed the relevant clinical information and correlated the data between the four groups. As shown in Figure3A-C, there was a significant inverse correlation between HR and HER-2 status and the OS, DFS, and overall recurrence (OR) for patients in the four groups (p=0.001). After comparing OS, DFS, and OR between the four groups, we found that only the HR positive group (group A) showed significant comparisons with all the groups. Group A had a shorter OS and DFS (median, OS: 60 months; DFS: 37.1 months), and higher OR than group B (median, OS: 85 months; DFS: 59.5 months), group C (median, OS: 96 months; DFS: 68 months), and group D (median, OS: 97 months; DFS: 51 months) (all p<0.01).
We further analyzed the relationship between Ki-67 expression and IDC patients' prognoses. The results indicated there was a significant inverse correlation betweenKi-67 expression and patients' OS, DFS, and OR between the Ki-67 positive group (n=135) and Ki-67 negative group (n=81). The Ki-67 positive group had shorter OS and DFS (median, OS: 65 months; DFS: 58.5 months), and a higher OR than the Ki-67 negative group (median, OS: 81 months; DFS: 71.5 months) as shown in Figure 3D-F.

Discussion
Ki-67, a nuclear non-histone protein, is a nuclear antigen expressed in the G1, S, G2, and M-phases of the cell cycle, but not the G0 phase. Ki-67 is a large nuclear protein (395kDa) and has been hypothesized to be involved in several different cellular functions such as cell cycle regulation, ribosomal RNA processing, and organizing DNA, or to have a structural role within the nucleus (Hendrayani et al., 2013). Ki-67has been considered a proliferative marker associated with tumor invasiveness and recurrence (Salehi et al., 2009). Several studies have investigated and considered Ki-67 as a prognostic factor for BC (Tarhan et al., 2013;Inwald et al., 2013;Eom et al., 2013). Many previous studies have demonstrated Ki-67 expression is associated with poor differentiation of tumors and large tumor size in BC (Ryu et al., 2012;Loehberg et al., 2103;Synnestvedt et al., 2013).
In this study, we investigated the expression of Ki-67 and its relationship with other clinical and pathological parameters and the expression of other molecular markers in 216 female invasive ductal BC patients with a mean age of 47.21±0.75 years (range: 21 to 75 years). Patients were from a single academic tertiary care center and the clinical significance of Ki-67 as a prognostic marker of BC was evaluated. Of these cancers, 135 were Ki-67 positive using immunohistochemical techniques, which directly correlated with poor prognostic features such as lymph node involvement, high grade tumor, lymphovascular invasion, high tumor stage, and high TNM stage, and had statistically significant value in identifying high risk women with invasive ductal BC. Previous studies reported that high expression of Ki-67, detected by immunohistochemistry, is the strongest individual prognostic factor of tumor recurrence or death in high tumor stage diseases (Chan et al., 2012;Kamineni et al., 2013;Ferguson et al., 2013).
This study showed an inverse correlation between expression of the HRs (ER and PR) and Ki-67, but a direct correlation between Ki-67 and HER-2 expression. Numerous studies have shown that expression of Ki-67 had a negative correlation with ER expression (Dede et al., 2013;Galea et al., 2013;Ibrahim et al., 2013;Bjerre et al., 2013). A recent study analyzed 356 BC samples retrospectively and evaluated the association between expression of Ki-67 and some clinicopathological parameters. They found that Ki-67 expression was correlated with tumor size, grade, and lymph node metastases, but was inversely correlated with HR expression, which supports the findings from this study.
The patients were divided into four groups according to HR and HER-2 status. Patients who were HR positive (ER and/or PR positive) and HER-2 negative formed Group A. Patients who were HR positive and HER-2 positive were assigned to Group B. Patients who were HR negative (ER and PR negative) and HER-2 positive became Group C. Patients who were HR negative and HER-2 negative were in Group D. The expression levels of Ki-67 were compared between the different groups and then with the other clinical and pathological parameters, to analyze their role in the progression of invasive ductal BC. The highest expression of Ki-67 was in the HER-2 positive Group C.
Our study supports molecular subtype of BC by gene expression profiles, in which BC cases are classified into ER positive (Luminal A and B) and ER negative (HER-2 positive, triple negative) groups. Different types of tumors vary in their biology, disease outcome, and responses to therapy. In triple negative and non-triple negative BC, the ER status may play a more important role than HER-2 status to predict the prognosis. HER-2 positive tumors that are HR negative are traditionally associated with a poor prognosis. However, recent studies reported that treatment with trastuzumab may eliminate the disparity in disease outcomes (Chan et al., 2013;Lemieux et al., 2013;Perissinotti et al., 2013). In the HR positive tumor group, Ki-67 expression directly correlated with lymph node metastases, lymphovascular invasion, high tumor grade, and high TNM stage in this study. These findings are in agreement with other reports that also revealed Ki-67 was associated with lymph node metastases frequently in all types of BC (Agboola et al., 2011;Chen et al., 2013;Tawfik et al., 2013). In this study, lymph node metastases and a high tumor grade also correlated with expression of Ki-67directly in HR positive and HER-2 positive groups.
To address whether positive expression of the HRs, HER-2, and Ki-67were associated with progression in IDC patients, this study analyzed relevant clinical information and correlated the data between four groups. We found there was a significant inverse correlation between the HR and HER-2 statuses and the OS, DFS, and OR of patients in all four groups. After comparing the positive rates of Ki-67 expression between groups, only the HR positive group (group A) was significantly different than all the groups ( Table 3). The results of Kaplan-Meier survival curves also showed that only group A had significantly shorter OS and DFS, and higher OR than the other groups. These findings indicate that the Ki-67positive rates correlate with progression in IDC patients. A specific therapy for triple negative BC is lacking; although specific treatment strategies may be developed as further biological characterizations occur. In this study, 13.6% (n=29) of cases were triple negative BC; among them, 76.7% were Ki-67 positive, which directly correlated with expression of TOP II-α.
In conclusion, Ki-67 expression correlates with that of ER, PR, HER-2, EGFR, and TOP II-α, as well as lymph node involvement, high tumor grade, lymphovascular invasion, high tumor stage, and high TNM stage in IDC.
Positive Ki-67 expression is a risk factor for rapid tumor recurrence and may lead to tumor progression with a poor prognosis in IDC. Ki-67 directly correlates with EGFR and TOP II-α expression, lymph node involvement, high tumor grade, lymphovascular invasion, high tumor stage, and high TNM stage in the HR subtype of BC. In triple negative BC, Ki-67 correlates with TOP II-α. .