Characteristics of Invasive Breast Ductal Carcinoma , NOS , Diagnosed in a Tertiary Institution in the East Coast of Malaysia with a Focus on Tumor Angiogenesis

Breast cancer remains the most common cancer among women in Malaysia, which accounted 29.9% of newly diagnosed cancer cases in Malaysian women registered in the national cancer registry for the year 2006 (Omar et al., 2010). Breast cancer is also the most common female cancer in Kelantan, a state located at the East Coast of Malaysia, which recorded 19.6% of newly diagnosed cancer cases in women for the year 1999-2003 in the state cancer registry (Mohd et al., 2006). Among those reported cases in Kelantan, high numbers of them were advanced cancers (stage III and stage IV), comprising 56.6% (Mohd et al., 2006). Management and prognosis of breast cancer still much depend on classic pathological factors such as histological type, tumor grade, tumor size, lymphovascular invasion, lymph node status, estrogen receptor (ER) and progesterone receptor (PR) status, and human epidermal growth factor receptor 2 (Her-2) expression status (Rakha and Ellis, 2011). Malaysian management guideline on breast cancer updated recently recommends that a pathology report should have a minimum dataset incorporating these pathological factors (Ministry of


Introduction
Breast cancer remains the most common cancer among women in Malaysia, which accounted 29.9% of newly diagnosed cancer cases in Malaysian women registered in the national cancer registry for the year 2006 (Omar et al., 2010).Breast cancer is also the most common female cancer in Kelantan, a state located at the East Coast of Malaysia, which recorded 19.6% of newly diagnosed cancer cases in women for the year 1999-2003 in the state cancer registry (Mohd et al., 2006).Among those reported cases in Kelantan, high numbers of them were advanced cancers (stage III and stage IV), comprising 56.6% (Mohd

Characteristics of Invasive Breast Ductal Carcinoma, NOS, Diagnosed in a Tertiary Institution in the East Coast of Malaysia with a Focus on Tumor Angiogenesis
Ewe Seng Ch'ng*, Sharifah Emilia Tuan Sharif, Hasnan Jaafar Health Malaysia and Academy of Medicine Malaysia, 2010), in tandem with criteria set by professional bodies in advanced countries (Lester et al., 2009;NHSBSP, 2005).
Angiogenesis -the growth and remodeling of new blood vessels -has been recognized as one of the hallmarks of cancer (Hanahan and Weinberg, 2011).It plays a crucial role in tumor progression because tumor growth, invasion and metastasis rely on tumor angiogenesis (Folkman, 2002).Weidner et al. first demonstrated that by immunostaining the blood vessels, angiogenesis quantified as microvessel density (MVD) correlated with distant metastasis in breast cancer patients (Weidner et al., 1991).Since then, it has been shown that higher microvessel densities in breast cancer predict higher risk of subsequent in situ cancers and invasive recurrence of previous in situ cancers (Teo et al., 2003), poorer response to treatment (Gasparini et al., 1996), higher occurrence of micrometastases (Fox et al., 1997;Benoy et al., 2005;Xie et al., 2009) and poorer relapse-free survival and overall survival (Uzzan et al., 2004).
Hospital Universiti Sains Malaysia is one of the two tertiary institutions in Kelantan, which has high percentage of patients presented with advanced breast cancer.This study aims to review the clinicopathological factors of invasive ductal carcinoma, not otherwise specified (NOS), the most common histological type of breast cancer, diagnosed in the Department of Pathology, School of Medical Sciences, Universiti Sains Malaysia and evaluates the relationships among the clinicopathological factors.In addition, tumor angiogenesis in these breast cancers is assessed in relation to clinicopathological factors.

Study design
This was a cross-sectional study and ethical approval was obtained from the institutional Research Ethics Committee (Human) with the reference number: USMKK/ PPP/JEPeM [226.4.(2.10)] prior to commencement of the study.Primary surgical specimens from lumpectomy, wide excision and mastectomy diagnosed as invasive ductal carcinoma, NOS, from 1 January 2006 to 31 December 2010 at the Department of Pathology, School of Medical Sciences, Universiti Sains Malaysia were included.Corresponding tissue blocks were retrieved and a single representative tumor tissue block was utilized for each case.Cases of which the corresponding paraffinized tissue blocks were not available, missing, or inadequate for sectioning, were excluded from this study.Clinicopathological data collected were age, gender, race, tumor size, tumor grade, lymphovascular invasion, lymph node involvement, estrogen and progesterone receptor status, and Her-2 expression status.In this institution, tumor grade was routinely reported using Elston-Ellis modification of the Scarff-Bloom-Richardson grading system (Elston and Ellis, 1991).Hormone receptor status was routinely considered positive for a minimum of 10% nuclear staining of any intensity in invasive tumor cells.Her-2 expression was routinely investigated by immunohistochemistry and the staining results were interpreted according to the HercepTest TM scoring guidelines (Dako, 2010).Cases with score 0 and 1+ were considered as negative, score 2+ as equivocal, and score 3+ as positive for Her-2 overexpression.

Immunohistochemistry for CD34
Standard immunoperoxidase procedures were followed.A 4um paraffin-embedded section from each case was mounted on poly -L -Lysine precoated slide (Menzel-Glasser, Germany).Sections were then deparaffinized and rehydrated.Sections were heated in Dako Target Retrieval Solution (Tris/EDTA buffer, pH 9) for 3 minutes in a pressure cooker for antigen retrieval.Dako REAL™ Peroxidase -Blocking Solution was used to quench the endogenous peroxidase activity.Sections were rinsed with wash buffer pH 7.6) and sequentially incubated with monoclonal mouse anti -human CD34 Class II primary antibody (Dako, at the dilution of 1:100 and with Dako REAL™ EnVision™ Detection System (Peroxidase/DAB+, Rabbit/Mouse) HRP-labeled polymer solution for one hour respectively at room temperature.After rinsing with wash buffer, sections were developed with 3,3'-diaminobenzidine (DAB)+substrate-chromogen solution for four minutes.
Sections were counter-stained lightly with hematoxylin, dehydrated, cleared with xylene and mounted.

Evaluation of angiogenesis using microvessel density (MVD) with anti -CD34 antibody
Using a bright-field microscope (Nikon Eclipse E600), microscopic evaluation of the slides immunostained for CD34 was performed blinded from the clinicopathological data.Microvessel density (MVD) was assessed according to the Weidner's method (Weidner et al., 1991;Fox, 2006).Each section was first scanned at low-power magnification (40x) to select three most vascularized non-overlapping areas (hotspots).The locations of these hotpots were documented as whether tumor invasive front or centre of tumor.Tumor invasive front was defined as 2mm (about field diameter of 200x field) tumor area abutting normal tissue (Giatromanolaki et al., 2004).Microvessels in sclerotic areas within tumor, where microvessels were sparse, and immediately adjacent areas of unaffected breast tissue were not considered.Individual microvessels were counted on a 200x field (i.e., 20x objective lens and 10x ocular lens).A countable vessel was judged as follows: (i) any immunostained endothelial cell or endothelial-cell cluster that was clearly separate from adjacent microvessels, tumor cells, and other connective tissue elements was considered a microvessel.(ii) vessel lumens and presence of red cells were not necessary for a structure to be defined as a microvessel.Microvessel density obtained in this manner, whether the highest value or the mean value, was used for statistical analysis.

Statistical analysis
Statistical analyses were performed using PASW® Statistics 18 software.Continuous variables were examined for their distributions and frequencies, and were expressed as means and standard deviations (SD).Frequencies and percentages of categorical variables were calculated.Univariate analyses were conducted using independent t-test or one-way ANOVA for continuous variables and Pearson's chi-square test or Fisher's exact test for categorical variables.Combination of categories was conducted when it was necessary for a valid statistical test.A p-value of <0.05 was set to denote statistical significance.

Clinicopathological factors of invasive ductal carcinoma, NOS and their inter-relationships
A total of 94 cases of invasive breast ductal carcinoma, NOS, which met the inclusion criteria, were included in this study.Pertinent clinicopathological data of the study subjects were summarized in Table 1.Inter-relationships between the clinicopathological factors were summarized in Tables 2 and 3.
Majority of the study subjects were Malays (76.6%).The mean age of the patients was 52.5(±SD 11.1) years with the age ranged from 28-89 years.Eighty-four percent of the patients underwent mastectomy with axillary clearance.Almost half of the tumors (48.9%) had tumor size larger than 50 mm.Lymph node metastasis was found  Further classification of tumors based on the hormonal receptor status and Her-2 expression was shown in Table 4.The triple negative breast cancers, which were estrogen and progesterone receptors and Her-2 negative cases, constituted 22.3% of the study subjects.

Microvessel density (MVD) in invasive breast ductal carcinoma and its relationship with clinicopathological parameters
The anti-CD34 antibody highlighted the endothelial cells of the vessels as shown in Figure 1 Relationships between the microvessel density in invasive breast ductal carcinoma, NOS and other clinicopathological parameters were assessed and the results were summarized in Table 5. Significant higher microvessel density was observed in younger patient age group, either by the average or the highest count (p<0.01 and p<0.05 respectively).The was no significant difference in microvessel density between the groups for the categories race, tumor size, tumor grade, number of lymph node metastasis, estrogen receptor, progesterone receptor, and Her-2 expression status.Limited to cases where information about lymphovascular invasion was available, significant higher microvessel density was observed in the presence of lymphovascular invasion, either assessed by the average or the highest count (both p<0.05,Table 6).For this group of patients, presence of lymphovascular invasion was also significantly associated with lymph node metastasis (p<0.01,Table 7)

Clinicopathological factors of invasive ductal carcinoma, NOS and their inter-relationships
Breast cancer remains the most common cancer among women in Malaysia.The cumulative lifetime risk was reported as 4.2% for year 2006 (Omar et al., 2010).This translates into about 1 in 24 Malaysian women will develop breast cancer in their lifetime.This risk however differs among the major ethnic groups in Malaysia: Chinese 1 in 20, Indian 1 in 23 and Malay women 1 in 31 (Omar et al., 2010).In contrast, 76.6% of the study subjects in this study were Malays, which was followed by Chinese (18.1%), other ethnics (4.3%) and Indians (1.1%).This is due to the composition bias of the local Kelantan population where 95.7% are Malays followed by 3.4% of Chinese (Department of Statistics Malaysia, 2011).However, a considerable high proportional of Chinese patients in this study as compared to the small percentage present in the local population (18.1% versus 3.4%) suggests that the incidence rate of breast cancer in Chinese women is higher as compared to that of Malay women.
Although lack of information about distant metastasis precluded proper staging of the tumors included in     doi.org/10.7314/APJCP.2012.13.9.4445 Invasive Breast Ductal Carcinoma and Tumor Angiogenesis this study, stages of the tumors could be approximated based on the tumor size and the number of lymph node metastasis.Thirty-four cases had more than three lymph node metastases.Among the patients that had 1-3 lymph node metastases, 13 cases had tumor size larger than 50 mm.These cases together would represent at least stage III disease, comprising 50% of the study subjects.The high percentage of late stage disease gauged in this study is similar to that reported in a study conducted in the East Coast of Malaysia, which had similar population composition.56.1% of the patients in that study presented with stage III and IV disease.Late stage disease in this region was attributed to the delay in diagnosis (Norsa'adah et al., 2011).Reflection of this could also be glimpsed by comparing the mean tumor size of this study with the others.Tumor size in this study had a mean of 65.5 mm, which is larger than 54.8mm and 30mm, the mean tumor sizes reported from two regional tertiary institutions in Malaysia and Singapore.Stage III and IV disease however comprised only 33% and 20% of their patients respectively (Lim et al., 2007;Mohd et al., 2008).
Majority of the tumors included in this study were grade III tumors (52.1%).Grade I tumors were merely 10.6%.This is considerably different from the published data from western studies where the proportions of grade III and grade I tumors are 29-46% and 11-38%, respectively (Rakha et al., 2010).A study from the local region of East Coast of Malaysia also reported 37.5% and 16.8% of grade III and grade I tumors respectively (Norsa'adah et al., 2011).Combined breast cancer registry from Malaysia and Singapore reported 40% and 13% of grade III and grade I tumors respectively (Bhoo et al., 2011).Higher percentage of grade III tumors in this study could be explained by the larger tumor size in this study, as larger tumor tends to have higher tumor grade (Anderson et al., 2000).Indeed, grade III tumors were significantly larger than the grade I and II tumors in this study (76.5mm versus 53.7mm, p<0.01).
In this study, estrogen receptor positive tumors were 53.2%, which is comparable to the other four studies reported in Malaysia, ranging from 48.2-57.4% (Sharifah et al., 2008;Norsa'adah et al., 2011;Teoh et al., 2011;Yip et al., 2011).However this positivity rate is lower than the rate in studies from western countries where 70-80% of invasive ductal carcinomas are estrogen receptor positive (Rhodes et al., 2000;Lal et al., 2005;Francis et al., 2007).This study also demonstrated that estrogen receptor negative tumors had significantly higher tumor grade and larger tumor size (p<0.05).It is well known that estrogen receptor negative tumors have higher tumor grade and higher tumor stage.Even with this association, estrogen receptor status remains an important prognostic factor for breast cancer after controlling the tumor grade and tumor stage (Grann et al., 2005).This study however did not show any significant association between estrogen receptor negativity and age and ethnicity as has been shown in one Malaysian study (Yip et al., 2011).
The percentage of Her-2 positive cancers in this study as defined by score 3+ by immunohistochemical methodology was 24.5%.This rate is similar to the prevalence of Her-2 positive cancers in seven Asian countries using similar method and grading system, which was 20.4±9.3% (Tan et al., 2010).Three studies from Malaysia also reported 24.2 %, 24.4% and 30.3% positive rate respectively (Sharifah et al., 2008;Tan et al., 2009a;Teoh et al., 2011).Her-2-overexpressed tumors are known to be inversely associated with the estrogen receptor expression and tends to have high tumor grade (Lal et al., 2005).Studies from this country also demonstrated this inverse relationship of Her-2 expression with estrogen receptor positivity, and positive association between Her-2 overexpression and tumor grade (Sharifah et al., 2008;Tan et al., 2010;Teoh et al., 2011).However, results of this study did not reveal these relationships.It might be due to higher percentage of triple negative breast cancers included in this study, which was 22.3% as compared to the 12-17% reported in the literature (Foulkes et al., 2010).Other Malaysian studies also reported the rate as 16.6% and 17.4% respectively (Tan et al., 2009b;Teoh et al., 2011).Triple negative breast cancers are relatively poorer outcome cancers with negative Her-2 expression and negative hormonal receptor and mostly are grade III tumors (Foulkes et al., 2010).Therefore, higher percentage of this cancer phenotype in this study would negate the general negative association between Her-2 and estrogen receptor expression and its positive association with tumor grade.
Although breast cancers in this study were limited to the invasive ductal carcinoma, NOS histological subtype, comparison with the studies mentioned above is still valid, as this histological subtype of breast cancer comprises about 90-100% of the breast cancers in most studies.

Microvessel density in invasive breast ductal carcinoma, NOS and its relationship with clinicopathological parameters
Heterogeneous distribution of microvessel density within breast cancers and vascular predominance at the tumor invasive front had been well recognized since the pioneer landmark study by Weidner et al (Weidner et al., 1991).This study also showed that most of the hotspots of microvessel density (69.9%) fell on the tumor invasive front.This phenomenon is perceived as the tumor angiogenic activity at the invasive front where proliferation of endothelial cells to form new vessels is controlled mainly by angiogenic factors.This is opposed to the inner tumor areas where anti-apoptotic ability might become more crucial for existing blood vessels to survive (Giatromanolaki et al., 2002;Giatromanolaki et al., 2004).
Studies had also revealed mixed results regarding the relationships between microvessel density and Her-2 expression or hormonal receptor status of breast cancers.It was suggested that Her-2 positive breast cancers had higher microvessel density in some studies (Koukourakis et al., 2003;Ludovini et al., 2003;Sopel et al., 2005), but not in other studies (Acenero et al., 1998;Vogl et al., 2006;Offersen et al., 2008;Thorat et al., 2009).Contradictory results were also reported in regard to hormonal receptor status.Estrogen receptor positive breast cancers were noted to have lower microvessel density in some studies (Koukourakis et al., 2003;Parentes-Vieira et al., 2007;Offersen et al., 2008).However, this observation was not substantiated in many other studies (Gasparini et al., 1996;Acenero et al., 1998;Ludovini et al., 2003;Benoy et al., 2005;Vogl et al., 2006;Thorat et al., 2009).This study revealed that the microvessel density in breast cancers was independent of the hormonal receptor or Her-2 expression status (p>0.05).
Discrepancies in the results in the literature as well as in this study could be explained by the vast heterogeneities in breast cancer characteristics included in the studies as well as the methodologies of exploration and statistical analyses.For example, a more sophisticated assessment by employing a Chalkley eyepiece graticule was claimed to be more superior in revealing significant association with other prognostic factors even though the same breast cancers were examined (Dhakal et al., 2009).
In this study, the only significant finding was the relationship with the age group; the microvessel density was significantly lower in older patients (p<0.05).Similarly, two other studies had also reported this inverse relationship between microvessel density and age (Miliaras et al., 1995;Marinho et al., 1997).It was previously reported that mean age of menopausal for women in this region was 49.4±3.4 years (Dhillon et al., 2006).Statistically, by choosing 55 as the cutoff, which is nearly the upper +2 standard deviation value of this reported age range, patients with the age more than 55 years in this study would be exclusively post-menopausal.It is therefore reasonable to extrapolate this finding that pre-menopausal breast cancers have higher angiogenic potential as compared to the post-menopausal breast cancers.Indeed, this might be a reflection of higher VEGF concentration in breast tissue in pre-menopausal women as compared to that in the post-menopausal women (Dabrosin, 2005).Higher percentage of VEGF positive tumor cells was also reported in pre-menopausal breast cancers (Fuckar et al., 2006).
In the subgroup analysis for the relationship of microvessel density to lymphovascular invasion, a significant higher microvessel density was observed in the presence of lymphovascular invasion, which was also significantly associated with lymph node metastasis (both p<0.05).However, as discussed earlier, there was no direct association between microvessel density and lymph node metastasis in this study.These results imply that tumor angiogenesis might act as prerequisite permit for vessel invasion; however, tumor cells may need to acquire additional trait to enable them establish early metastasis in the lymph nodes in line with the invasion-metastasis cascade theory (Hanahan and Weinberg, 2011).
In addition, this study demonstrated similar results when statistical analyses were conducted with either the average count or the highest count of the microvessel density.Therefore, evaluation of a single hotspot with the highest microvessel vessel density is reasonably sufficient to obtain information regarding tumor angiogenesis.
In conclusion, in summary, considerable differences exist when the findings regarding the traditional clinicopathological factors are compared against those reported from other local or western studies.Not only majority of the patients of this institution had advanced stage disease, a substantial percentage of them also had poorer prognostic factors such as higher tumor grade, lower hormonal receptor positivity and triple negative breast cancer phenotype.As compared to other studies, this study demonstrated that tumor angiogenesis was only significantly associated with age group, whereby premenopausal breast cancer might be more proangiogenic.

Figure 2 .
Figure 2. Distributions of the (A) average and (B) highest microvessel density in 94 cases of invasive breast ductal carcinoma, NOS.

Table 3 . Relationships between Tumor Size and Other Clinicopathological Factors
*One-way ANOVA.Others: Independent t-test

Table 7 . Association between Lymphovascular Invasion and Lymph Node Metastasis in 54 Cases of Invasive Breast Ductal Carcinoma, NOS
Fisher's exact test.