Quantitative Immunohistochemical Expression of GP88 in Invasive Ductal Carcinoma (IDC) of Human Breast Cancer

Background: Progranulin or acrogranin is an 88-kDa glycoprotein identied by a biological screen for protein targets associated with high tumorigenicity. This work was aimed to investigate the digital expression of GP88, and HER2/neu as a predictive biomarker in human invasive ductal carcinoma (IDC) versus benign tumors and normal breast tissues, as well as, its correlation with different pathological parameters. Methods: The immunohistochemical avidin-biotin complex protocol of the paran section was used to detect the expression of GP88 and HER2/neu in IDC of 60 patients, 30 benign, and ten normal breast tissues. Results: the study showed a high expression of GP88 in IDC comparing to normal and benign breast tissues. A higher signicant statistical correlation between the expression of GP88 and large tumor size, tumor grade, and lymph node metastasis (LNM). While a negative statistical correlation was noticed between the expression of GP88 and ER, PR, and HER2/neu status. Conclusion: It could be concluded that GP88 glycoprotein may be a valuable predictive and therapeutic marker in human IDC patients.


Introduction
Breast cancer represents a major scienti c, clinical and societal problem. It is the most common malignancy and the second leading cause of cancer death in females following lung cancer [1,2] with more than million new cases and 370,000 deaths yearly worldwide [3]. In many developing countries, the incidence of breast cancer is now rising sharply due to changes in reproductive factors, lifestyle, and increased life expectancy [4]. The invasive ductal carcinoma is the most common type of breast cancer, and comprises 70-80% of all cases. It starts in a milk duct of the breast, breaks through the wall of the duct, and grows into the fatty tissue of the breast, that it may be able to spread (metastasize) to other parts of the body through the lymphatic system and blood stream [5]. The cells invasive must occur from detached and penetrate the basement membrane, adherent cells lose their normal attachment to the substrate [6]. Proliferation is highly sensitive to the extracellular matrix maintain an actively proliferating phenotype in the newly encountered matrix [7]. Ductal carcinoma is an ill-de ned mass, sometimes adherent to the skin or underlying muscle. The tumor is of varying size and may be associated with calci cation. The histomorphology of the tumor is highly variable, ranging from a low-grade tumor showing mildly pleomorphic tumor cells arranged in tubules with little mitotic activity to a high-grade tumor showing highly pleomorphic tumor morphology, with the tumor cells arranged in solid sheets and groups, brisk mitotic activity and abundant tumor necrosis [8]. The Nuclear morphology is evaluated in variation of nuclear size, regularity of the nuclear border, hyperchromasia, and prominence of nucleolus.
Also, it has been reported that histologic grade for breast cancer remains a prognostic factor despite changes in tumor size and number of positive lymph nodes [9]. Histological grade is commonly employed and has been shown to be of independent prognostic signi cance. The studies showed very strong correlation with prognosis; patients with grade I tumors have a signi cantly better survival than those with grade II and III tumors (p < .0001) [10] Progranulin is 88-kDa glycoprotein and known as GP88, PC-cell derived growth factor or acrogranin, GP88 gene is located on the 21q portion of chromosome 17, while the mouse gene was found on chromosome 11 [11]. The autocrine growth factor GP88 is abundantly expressed in epithelial cells, immune cells, neurons, and chondrocytes [11]. Also, involved in a variety of biological processes, including embryogenesis, in ammation [12], wound healing [13], host defense [14]and cartilage development and degradation [15]. The importance of the Progranulin (GP88) is no unique receptor for GP88 has been identi ed [16]. GP88 has been shown to bind the membrane proteins sortilin and tumor necrosis factor receptors (TNFR) 1 & 2 [17]. It is thought to exert its mitogenic effect through the stimulation of both the mitogen-activated protein kinase (MAPK) and the phosphatidylinositol 3-kinase (PI3K) pathways [18].
GP88 stimulates the expression of the angiogenic growth factor (VEGF) in breast cancer cell lines, the breast cancers with elevated GP88 levels have higher VEGF levels and microvascular density [19]. GP88 is an important marker in breast cancer, it can be measured in healthy and breast cancer patients [20].
Increased secretion of GP88 is associated with more aggressive forms of breast, brain, renal, cervical and other tumors [12] HER-2/neu (also called c-erbB-2) is a proto-oncogene located on chromosome 17, and encodes a 185-kd transmembrane glycoprotein receptor (p185 HER2) which has a tyrosine kinase activity and related to the epidermal growth factor receptor (EGFR). It is ampli ed or overexpressed in about 25-30% of human breast cancer, which is associated with tumor aggressiveness [11], and maybe related to variability in the interpretation of protein expression levels [21]. The predictive value of HER-2/neu status with respect to response to therapy as hormonal therapy, resistance to alkylating agent-based chemotherapy and taxanes [22,23]. So, the aim of the present work was designed to study the Immunohistochemical expression of the GP88 as a novel prognostic biomarker in human invasive breast carcinoma.

Materials And Methods
Tissue samples were obtained from patients diagnosed with breast tumors in the Department of Pathology, Medical Research Institute (MRI), Alexandria University, Egypt, Formalin-xed and para n embedded tissue specimens from 60 patients diagnosed with IDC, 30 patients diagnosed with benign breast tumor and 10 were taken from normal breast tissue adjacent to the benign tumors were included. All the cases were asked to freely volunteer to the study and informed written consents were gathered prior to their inclusion in the study according to the guide ethics of institute MRI (IORG#: IORG0008812). Samples from all studied cases were subjected to the following: 1-Clinical parameters included patients' age, tumor size, lymph node metastasis (LNM) status, 2-Histopathological examination by hematoxlin and eosin (H&E) stained slides for each patient and reviewed by two pathologists (Diagnosis of the specimens was made according to the WHO classi cation of the Tumors) [24]. 3-Biological markers of (ER, PR). 3-Immunohistochemical technique: the detection of both GP88 and HER2/neu protein used avidin biotin complex protocol (ABC) and DAB stain [25] Immunohistochemical method was utilized to study the expression of GP88 and HER2/neu. in 100 para n-embedded breast tissues. In brief; para nembedded specimens were cut into 5µm sections. The sections were depara nized using 2 changes of xylene and rehydrated. The sections were submerged in antigen retrieval (citrate buffer saline pH 6) in an oven at 95°C for 20 minutes and then left at room temperature for 20 minutes to cool. The sections were treated with 3% H 2 O 2 in PBS to quench the endogenous peroxidase activity, followed by incubation with serum blocking reagent for 30 minutes to block nonspeci c binding. The sections were incubated with primary antibody for GP88 (Biorbyt Company, London, UK) and HER2/neu (Novus Biologicals Company, USA) at 4ºC overnight. The sections were treated with conjugated 2 ry antibody (ABC-HRP reagent) for 30 minutes, stained with diaminobenzedine (DAB) and counter stained with hematoxlyen. For negative controls, antibody was replaced with PBS. Each step was followed by PBS washing. Immunohistochemical evaluation of results was arbitrarily graded as negative (0), weak (+1), moderate (+2) and strong (+3). 4-Image analysis: Integrated optical density (IOD) of GP88 and HER2/neu in normal, benign and IDC groups was analyzed using digital image analyzer. Images were viewed and recorded using Olympus microscope equipped with spot digital camera and image J software. Maximum, minimum and integrity of intensity color based on Gray-level acquisition, analysis of data were carried out by reading ten image for each case. The mean values of each reaction were based on the mean of pixel number [26]. IOD based on Gray-level transition probabilities in digitized images was graded from dark to light (0 up to 250). The average score across the whole image should be taken; IOD in digitized images was calibrated from strong to light (180 down to 70) by pixel. This calculation was proceed after subtract the pixel value from 250, the pick of lighter elimination

1-Histopathological investigation:
The para n section photomicrograph illustrated the histopathological changes of the studies groups.
The normal breast tissue showed acini lined by epithelial cells, an intact myoepithelial cell layer attached the basement membrane of acini ( Figure 1A). A mild adenosis and cyst formation with mild epithelial hyperplasia in the adjacent duct and intact myoepithelial cell layer in brocystic disease of the benign breast case ( Figure 1B). The tumor cells of invasive ductal carcinoma cases have abundant eosinophlic cytoplasm and large moderately pleomorphic round to ovoid vesicular nuclei arranged in cords and some contain small nucleoli. In ltrated tumor cells in desmoplastic stroma represented GII (Figure1C). The neoplastic growth formed of nests and tubules and hyperchromatic nuclei and largely pleomorphic vesicular nuclei were exhibiting mitotic activity in ductal epithelial cells with many pyknotic cells, as well as, many residual nuclei was seen in GIII ( Figure 1D) The histopathological diagnosis of studing cases found that the benign cases revealed as the majority (83%, 25/30) of broadenoma, while 17% (5/30) cases were represented with brocystic disease. At malignant cases, there was a majority of invasive ductal carcinoma (IDC) (98%, 59/60), while 2% (1/50) of the malignant cases were represented with invasive lobular carcinoma (ILC). In addition, according to Scarf'-Bloom-Richardson histological grading system of breast cancer, the majority of the malignant cases diagnosed as grade II (45/60, 75%), grade III cases were represented by 22% (13/60) and the minority cases were grade I (2/60, 3%) of the malignant cases as in the (Table 1).

2-Clinicopathological parameters:
The age ranges of benign and malignant groups were 15-64 and 15-74 years with a mean ± SD of 36 ± 11 and 52 ± 10 years respectively. The data reported of the benign group cases 57% (17/30) at age range > 35-45 years versus of the malignant group cases 31.5% (19/60) at age ranges > 35-45 and > 45-55 years respectively. Neither of the benign cases were above 65 years nor of malignant cases under 25 years as shown in (Table 2) According to the tumor size assessed according to TNM (Tumor, Node, Metastasis) staging system of breast cancer, there was 80% (24/30) of the benign cases were T2 (> 2-5 cm) versus 66% (27/45 grade II and 11/13 grade III) of the malignant cases were T2 (> 2-5 cm) as illustrated in the (Table 3). Whereas the percentage of positive vascular invasion (VI) was 98% (59/60) were distributed as in the (Table 4), as well as, Lymph Node Metastasis (LNM) was illustrated in the (Table 5). Also, Noticed the majority positive cases were represented (44/60, 3%) and free of lymph node metastasis cases were 16/60 (27%).
Immunostaining expression of GP88 was detected as diffuse, homogenous brown color detected in the cytoplasm and membrane of the ductal epithelial cells of the studied groups as shown in (Figure:

5-Immunostaining optical density (IOD)
A gray value detection of the binary evaluation at the 8-bit image of the stain density were quantitative digitally by pixel in the integrated region of optical density (ROD) as following: a-HER2/neu The mean and SD values of HER2/neu IOD for breast lesions illustrated the different ranges from 42. ±3.82, 63.±5.34, 143±5.49 and 169±4.46 as control, benign and IDC grade II and III were respectively.
There was statistically signi cant difference between benign group and malignant one, as well as between the malignant grade II and III of IDC (p < 0.001), as shown ( Figure:4) b-GP88 The mean and SD values of GP88 IOD for control, benign and IDC grade II and III were 35 ± 3, 41 ± 4, 140 ± 8 and 162 ± 7 respectively. A signi cant difference was noticed between GP88 IOD of grade II and III IDC and GP88 IOD of both control and benign groups (p < 0.001), but there was no statistical signi cant difference between GP88 IOD of control and benign groups (p = 0.1) as shown (Figure : 5) c-Correlation between GP88 IOD and histopathological parameters: There was no statistical signi cant correlation between the GP88 IOD and patients' age (r= -.14, P = .28), ER status (r= -.09, P = .50), PR status (r= .06, P= .65) and HER2/neu status (r.22, P= .08) of the studied breast cancer cases, while highly statistical signi cant correlation was noticed between the GP88 IOD and tumor size (r= .354 ** , P= .006), tumor grade (r= .353 ** , P= .006) and LNM status (r.493 ** , P = .000) as shown in table (10).

Discussion
Progranulin or GP88 is an autocrine growth factor and pleiotropic regulatory protein that has been shown to play role in tumorigenesis, including proliferation, survival, migration, angiogenesis invasion and matrix metallo-protease activity [27], as well as, in wound healing and in in ammation in normal tissues [28]. The study was evaluated the role of GP88 through the digital immunostaining density under the correlation with the clinical pathological parameter in the breast cancer lesions.
The study evaluated a statistical signi cant increase in the immunohistochemical expression of GP88 in IDC, versus, normal tissues and benign tumors (p < .001). This result is in alignment with previous nding reported a high level of GP88 expression in breast cancer biopsies versus benign lesions and normal mammary epithelial tissues [29,30]. The pathological studies with 203 formalin-xed para n-embedded human breast cancer tissue biopsies indicated that GP88 was preferentially expressed in ductal carcinoma with little expression in lobular carcinoma while benign lesions and normal mammary epithelial tissues were negative [31]. In addition, circulating GP88 level in the serum of breast cancer patients was increased in compared to healthy volunteers [28]. Moreover, the parental SW13 cells (cell derive from an adrenal carcinoma) poorly tumorigenic in vivo; however, by over-producing GP88, they become highly tumorigenic in mice [32,33]. This nding revealed that the SW13 cells are highly sensitive to the proliferative effects of GP88. So, the digital analysis (IOD) of the GP88 in the present work revealed the role of the GP88 expression associated with the increased tumor grade in the breast cancer and detected its role in the proliferation and invasive of the tumor cell in breast tissue. Also, the immunostaining positivity of GP88 in this work revealed the marked and intensity was present in malignant lesion and differentiated in both grade II and grade III, this results were indeed the role of the GP88 expression induced and associated with division and invasive tumor cells. This nding was con rmed by He, et al, [33] who reported that the pgrn (GP88) stimulate cell division, invasion, and survival demonstrates, other revealed the role of GP88 expression for regulates multiple processes or stages in carcinoma progression and suggested the GP88 may be a possible future therapeutic target [34].
However, there was no statistically signi cant correlation between immunohistochemial expression of GP88 and patients' age (r = -0.1, P = 0.3). This result is consistent with the results of other previous studies [28,35]. who reported the clinical and tumor characteristics (p-value risk ratio 95% ) con dence interval lower upper age in years Age ≤ 50 compared to > 50. Therefore, it could be agreed that the tissue GP88 level is predictive of recurrence in patients with hormone-responsive BC [36], while the patients with early-stage BC would be suggested that serum GP88 determination has the potential to have additional clinical utility.
The large tumor size, high tumor grade, and positive lymph nodes are all signs of poor prognosis and elevated GP88 expression have shown to be associated with tumorigenesis and poor prognosis in several cancer types including ovarian [30], renal [37], prostate [38], liver [39], esophageal [40,41] and breast cancers [42]. The current study showed a highly statistically signi cant correlation between the immunohistochemical expression of GP88 and tumor size (r = 0.354**, P = 0.006) of malignant breast cases. This result is going in accordance with Li et, al., [38] [35], but contrasted with Serrero et, al. [29] and DeMorrow, [42]. Moreover, a highly statistically signi cant correlation was observed in the current study between GP88's immunohistochemical expression and tumor histological grade (r = .353**, P = .006). This result is contrasted with Serrero et, al., [29], but consistent with the other who reported that overexpression of GP88 in 80% invasive ductal carcinoma, where it correlated with the clinical variables of poor prognosis such as tumor grade, p53 expression and Ki67index [31,35]. Immunohistochemical expression of GP88 in the present study exhibited a highly statistically signi cant correlation with LNM of the studied breast cancer cases (r = .493**, P = .000). A previous study found that high levels of GP88 expression were closely correlated with clinicopathological parameters such as lymph node metastasis [35]. The histopathological features are independent predictors of clinical outcomes such as lymph node status, tumor size and tumor grade for invasive ductal carcinoma patients were based evidence that IHC analysis of GP88 expression maybe help a predict treatment outcomes in a prognostic role. At the correlation between the GP88 and hormonal status as a routine marker and necessary for the treatment and therapy of BC, our results in IDC patients found no statistical signi cant correlation between the digital quantitative immunohistochemical expression of GP88 and ER (r = .09, P = .50) and PR (r = .06, P = .65) status. These results are con icted with Song et. al., [43] who reported that GP88 positive staining was more common in ER/PR negative samples than that in ER/PR positive tumors. The current results are going in accordance with Tkaczuk, et al., [28] who observed no statistical correlation between GP88 expression and ER and PR levels. However, screening of GP88 expression in human breast cancer cell lines indicated that GP88 was highly expressed in both ER positive and ER negative human breast carcinomas [29,44]. This lack of correlation between GP88's immunohistochemical expression and ER status indicated that the growth factor receptors promotes proliferation without requiring ER-mediated growth signaling [45]. It has been reported that level of G P88 expression was a major determinant of the intrinsic activity of the mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3'-kinase (PI3K) in cell line studies suggesting that the mitogen-activated protein kinase and phosphatidylinositol 3'-kinase signaling pathways may be involved in the promotion of tumor invasion and migration by GP88 [16,33]. However, the current study revealed no statistical signi cant correlation between the immunohistochemical expression of GP88 and HER2/neu status (r = 0.2, P = .08) of the studied breast cancer cases. This result is going in accordance with many studies reported that there was no correlation between expression of GP88 and HER2/neu, and indicated that GP88 and HER2/neu were independent biomarkers [28,31]. Other investigator found that 25% of HER2-overexpressing biopsies (3 + by IHC) were strongly positive for GP88 [46] and the treatment of HER2-overexpressing cells with Trastuzumab decreased the levels of MAPK phosphorylation [47]. The results is going to that the activation of mitogen-activated protein kinase (MAPK) signaling pathways in breast cancer cells by the over-production of growth factors [43]. So, GP88 can activate MAPK signaling pathway and phosphorylated MAPK regardless of the HER2 expression levels, thereby GP88 is not necessarily dependent on HER2 overexpression, because it has its own ability to activate MAPK pathway [48]. This may be indicated that HER2/neu and GP88 are independent biomarkers.
Finally, the study found that high progranulin expression was associated with higher breast histopathological features, which are strong independent predictors of clinical outcomes such as lymph node status, tumor size and tumor grade for patients with invasive ductal carcinoma, as well as, the digital quantitative immunostaining of GP88(IOD) in IDC cases was evident a high statistically signi cant correlation compared to the benign and malignant cases as well as between the IDC grade II and grade III (p = 0.0001). Beside of no correlation between GP88 with ER, PR, and HER2/neu expression.

Conclusion
It may be suggested that the GP88 involved in processes of cells invasive, regulate tumor cell progression and a biomarker dependent as a therapeutic target.

Declarations
Ethics consents: All the cases were asked to freely volunteer to the study and informed written consents were gathered prior to their inclusion in the study according to the guide ethics of institute MRI (IORG#: IORG0008812). Consent for publication: the co-author's consent to the publication .
Availability of supporting data: not applicable