Evaluate fibroadenoma and phyllodes tumor sizes

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


Aims and background. Fibroadenoma (FA) and phyllodes tumor (PT) are a group of fibroepithelial lesion of the breast comprising a heterogenous neoplasm of the terminal ductal unit. The main aim of the study was to evaluate the expression status of ERα, ERβ, Bcl2, p53, and MIB-1 protein in fibroepithelial lesion and correlate with clinicopathological factors.

Methods and study design. A total of 114 fibroadenomas and 67 PTs (43 benign, 12 borderline and 12 malignant) were examined using immunohistochemistry (IHC) on tissue microarray (Alphelys 0.6mm punch) to determine the expression of ERα, ERβ, Bcl2, p53 and MIB-1 in epithelia and stroma.

Results: The mean of tumor size was 3.4±1.97 cm, 6.6±4.48, 14.1±8.4 and 9.9±8.3 cm for fibroadenoma, benign PTs, borderline PTs and malignant PTs, respectively. The expression of ERβ (p<0.001), p53 (p=0.012) and MIB-1 (p=0.238) in the stroma component were greater in a larger tumor size. The p53 was significant in stroma component of malignant PTs (P<0.001). In addition, p53 and MIB-1 expression showed a significant association in both epithelia and stroma (p=0.015, p<0.001) as well as in ER and Bcl2 (p<0.001).

Conclusions: The association between MIB-1 and p53 with histological grade, may indicate their immunopositivity in benign breast tumor increase risk for malignancies and demonstrate to be a valuable marker in PTs, while a positive association between ER and Bcl2 defines that ER may be the regulator of Bcl2 protein.

Key words: phyllodes tumor, fibroadenoma, MIB-1 protein, p53 antigen, estrogen, receptor, immunohistochemistry


Fibroepithelial tumors of the breast illustrated as a proliferation of epithelia and stroma component. This tumor group ranges from benign to malignant; includes fibroadenomas and phyllodes tumors (PTs)1. Fibroadenomas (FAs) are the most common fibroepithelial tumor among young women, but it can be also diagnosed in post-menopausal women. It is composed of both glandular and connective tissue, characterized by hyperplasia and abnormal lobular breast units2,3,4. The phyllodes tumors are rare neoplasms and accounting for less than 1% of all primary breast tumors 5,6,7. Phyllodes tumors show predominantly of fibroadenoma criteria, with a leaf-like projection cells and more cellular connective tissue stroma component.

Phyllodes tumors may be classified as benign, borderline and malignant subtypes according to gross and microscopic features including margin appearance, cellular pleomorphism, stromal cellularity, mitotic activity and stromal distribution1,6,7,8. Apoptosis is one of the factors that play a key role in the development and growth regulation of normal and neoplastic mammary tissues and its dysregulation is suggested to stimulate the mutation that causes breast tumor 9,10 . Bcl2 is an oncoprotein that plays a role in apoptosis; by blocking programmed cell death, discourages the promotion of cellular proliferation and induces tumor development 10,11.

The expression of Bcl2 is associated with cells that are protected from apoptosis such as stem cells or epithelia that undergo hyperplasia; including breast and prostate (Moore & Lee 2001) and its expression is related to steroid hormone receptor 10,12.

The estrogen receptor (ER), is ligand activated transcription factors that normally present in normal and benign lesions of the breast 10,13 . Generally, ER has been implicated in both normal and neoplastic mammary tissue as a DNA binding transcription factor and induces cell proliferation 4,14. It was first cloned as ER located on the long arm of chromosome 6q25 and is now known as ERα. The second type of ER is ERβ, located on chromosome 14q22-24 13. ERα was reported to express mainly by epithelial cells in fibroepithelial tumors, while was commonly difficult to define the presence of ERα in the stroma component 15 .

Ki-67 antigen is a cell proliferation-related protein that can be detected with monoclonal antibody MIB-1 and to evaluate proliferative activity in different types of tumor 16,17. Ki-67 is linked to cell cycle (Jacklin et al 2006), specially expresses during late G1, S, M and G2 phase of the cell cycle, is undetected in cells of the G0 phase 6,18. It is observed as a marker of cell proliferation and the best indicator in determining the growth fraction of cells 6.

Expression of the proliferation marker Ki-67 has been reported to vary among histological categories of phyllodes tumor as well as between phyllodes tumors and fibroadenomas and was also reported in several studies to show a correlation between MIB-1 positivity and the histological grade 7,19,20.

Besides, p53 immunohistochemical expression, commonly used as an identification for tumor-suppressor gene mutation has been also correlated with tumor grade 7,21. It is located on the short arm of chromosome 17p13, is a well-known tumor suppressor gene, plays a role in the regulation of normal cell growth and division, DNA repair and apoptosis 22 .

The aim of this study was to conduct immunohistochemical analysis to determine the expression status of ERα, ERβ, Bcl2, p53, and MIB-1 protein in fibroepithelial lesion by tissue microarray. We determined the relationship between the immunostaining expression with clinical criteria such as age, tumor size and tumor types and to assess the usefulness of these markers in distinguishing benign from malignant tumors.

Methods and study design


Prior ethical approval was taken to conduct the study. The study population consisted of 181 patients recruited during 1990 to 2008. All samples and patient's data were collected from Department of Pathology, Universiti Kebangsaan Malaysia Medical Centre (UKMMC). A total of 114 fibroadenomas and 67 phyllodes tumors (benign PT, borderline PT and malignant PT) histology slides were examined through microscope and evaluation by pathologists to determine the spot area; consisting of both epithelia and stroma components of fibroepithelial lesions. The phyllodes tumors were graded into benign, borderline and malignant tumor, according to WHO classification23.

Tissue Microarray

Tissue microarray (TMA) was conducted to all selected cases using 0.6mm diameter punch kit MTA Booster (Alphelys, France). Approximately 50-70 cases were punched into a single TMA block and quadruplicate, giving a total of 16 TMA blocks24.

Immunohistochemistry (IHC)

Immunohistochemical staining was performed manually to determine the expression of selected biomarkers in epithelia and stroma using DAB horseradish chromogen. 3µm sections were cut onto poly-L-lysine coated slides and baked at 60°C for 30 minutes. Deparaffinisation in xylene and rehydration through graded alcohol followed by treating with antigen retrieval for 40 minutes at 98°C. The sections then were cooled down in room temperature and incubated with blockage agent, hydrogen peroxide for 10 minutes followed by primary antibody incubation for 30-60 minutes in room temperature. The primary antibody and their dilution were shown in Table 1.

Incubation with Dako EnVision was followed by DAB chromogen detection and then counterstained in haematoxylin solution. Sections of tonsil, colon adenocarcinoma, breast carcinoma and endometrium shown strong staining (3+) were used as a positive control for MIB-1 and Bcl2, p53, ERα and ERβ, repectively. Negative controls were performed by omitting the primary antibody. IHC staining of p53 was interpreted as positive when more than 10% staining of the tumor nuclei were detected and Ki67 positive nuclei divided by the total number of stroma or epithelial cells expressed as a percentage, modified from previous studies 7,17,25. The ER and Bcl2 were required as positive reactivity when ≥10% of nuclear or cytoplasm staining of at least 2+ intensity of expression detected 26 .

Statistical analysis

Kruskal-Wallis test was used to examine differences of biomarkers expression between the four groups, and were correlated between different variables using chi-square test. Differences were considered to be statistically significant at p<0.05. All analyses were carried out using the software Statistical Packages for the Social Sciences, SPSS v12.0 (SPSS Inc., Chicago, IL, USA). All human samples were collected with ethic approval from ethic committees of Faculty of Medicine, Universiti Kebangsaan Malaysia.


A total of 114 fibroadenoma and 67 phyllodes tumor patients showed an increase in age and tumor size between benign to malignant tumors, except for malignant phyllodes tumor which was slightly lower than in borderline patients (Table 2). The mean age of patients with fibroadenoma was 28.5 years (range, 10-59). The mean age for benign, borderline and malignant phyllodes tumors were 39.8 (range, 16-74), 49.8 (38-78), and 44.8 (29-67) years, respectively (shown in Table 2). The mean of tumor sizes were 3.4±1.97 cm, 6.6±4.48, 14.1±8.4 and 9.9±8.3 cm for fibroadenoma, benign PTs, borderline PTs and malignant PTs, respectively.

This study showed a significant association between p53 (p=0.012) expression in the stroma with tumor size, which 10.7% (6/56) were less than 3cm, 4.2% (3/72) ranges 3 to 5 cm and 21.2% (11/52) were more than 5 cm size of tumor, as shown in the Figure 1. Our data showed positive expression of ERβ (p<0.001) and MIB-1 (p=0.238) in the stroma component that were greater in the larger tumor size (>5cm).

The expression of p53 in the epithelia component (Figure 2) was primarily seen in small tumor size, but was not statistically significant compared to Bcl2 expression in the epithelia showed a significant correlation to the small size of tumor (p=0.018).

Figure 3 showed a tissue microarray slide. A positive expression of study markers for fibroadenoma and three types of phyllodes tumors (benign, borderline and malignant) were shown in Figure 4; fibroadenoma (Figure 4A), benign PT (Figure 4B), borderline PT (Figure 4C), and malignant PT (Figure 4D). ERα showed a strong staining in nucleus of epithelia component, while ERβ immunoreactivity was detected in both epithelia and stroma components. Bcl2 staining was detected in the nuclei of epithelia cells and also expressed in stroma component of malignant phyllodes tumor. Besides, p53 was strongly expressed in malignant phyllodes tumor and weak staining was detected in benign tumors.

The results of overexpression of biomarkers are listed in Figure 5. The p53 expression was low in the stroma of benign tumor and increased in malignant tumor (>10% expression). Both types of ER expression were greater in epithelia of benign than malignant tumors. Surprisingly, both types of ER expression were common in stroma of malignant phyllodes tumors compared to benign tumors. However, ERβ expressed higher percentage than ERα especially in the stroma component, while ERα was undetected in fibroadenoma to borderline phyllodes tumors. Contrary to stroma component, expression of p53, ER and Bcl2 were greater in most of epithelial of benign tumors than in borderline PTs and malignant PTs, but were not statistically significant for p53 and ER. Significant association between ER and Bcl2 expression was seen in the stroma component (P<0.001).

The present study showed a different pattern of p53 expression which had greater percentage of staining in benign tumors than malignant tumors in the epithelia cells, contrary to MIB-1 which strongly expressed in malignant tumors in both epithelia (Table 3) and stroma component (Table 4). However, fibroadenoma and benign PTs showed weak staining of p53 in the majority of cases. The monoclonal antibodies p53 showed a significant value in the stromal component, of malignant PTs only (P<0.001). In addition, p53 and MIB-1 expression showed a significant association in both epithelia and stroma components (p=0.015, p<0.001).


Tissue development is the result of a balance between cell proliferation, differentiation and apoptosis 10 . Proliferation of fibroepithelial tumor is mainly in the stroma component, followed by proliferation of epithelia cells 15. The stromal elements were considered as the neoplastic component and therefore believed as the determinant of biological activity 27.

The Bcl2 protein are usually expressed in normal tissue and benign proliferative lesions, as they discourage the promotion of cell proliferation by inhibit apoptosis 10 . There was a significant difference of the Bcl2 expression between the benign (60.5%), borderline (75%) and malignant PT (25%) in epithelia, and 12.5%, 8.3% and 25% in benign, borderline and malignant PT, respectively of the stroma component, but was not statistical significant in the stroma. Bcl2 expression can be detected in many hormonally regulated cells, such as epithelia cells of normal breast and ER/PR-positive breast carcinomas10,28,29. Besides, Bcl2 has been shown to be important downstream mediates of estrogen action survival of cancer cells and other study (Yang et al 2006) showed relation of Bcl2 protein expression to the expression of steroid hormone receptors such as ER, similar to this study (p=0.001) 4,10 .

Estrogen has a significant role in promoting malignancy in breast and potentially to be a mediator of the mammary gland since it is necessary for development and metastasis13. The results of the present study showed a significant differences of ER expression in stroma component, which increased with malignancy (p<0.05). In other studies, ER regularly present in normal and benign lesions of breast and metastasis tumors were usually negative for ER 10. As in this study, ERα expression was common in benign tumor, 63.2% and 60.5% of fibroadenoma and benign PT, respectively, but low expression was detected in epithelia cells of malignant PT (50%). However, the expression of ER was slightly increased in borderline cases (66.7%), suggesting a different cellular activity between tumors. On the other hand, ERβ showed an excellent pattern in epithelia component with high expression in benign cases and reduced with malignancy; 81.4% of, 74.4% in, 66.7% and 33.3% in fibroadenoma, benign PT, borderline PT and malignant PT; respectively, but was not statistically significant.

Besides, other studies also reported that ERα was mainly expressed by epithelia cells while its expression by stromal cell was controversial with only 8.3% expression in malignant PT and absent in borderline and benign PT, in the present study 15. Previous studies demonstrated that ERβ were found in stromal cells in the fibroadenoma and PT 13,15.We opine that ERβ showed better expression than ERα which was highly detectable in the stromal component. The greater expression of ERβ suggests that it may be the main ER in the breast. So that, Speirs and colleagues (2002) believed that cells initially considered ERα negative may actually be expressing ERβ 13 .

Recent study demonstrated the expression of ER was related to the proliferative index of the tumors measured by immunolabelling with antibodies against Ki-67 and indicating a good prognosis with endocrine therapy 10. Another study suggested that cells that were immunopositive for ERα very rarely undergo proliferation, as determined by the lack of expression of the cell cycle associated antigen Ki-67 and proliferative cell nuclear antigen which was similar to the present study but was not statistically significant 13. Expression of the proliferation marker, Ki-67 has been reported to be different between histological categories 7,21.

Other authors and our study have shown a correlation between MIB-1 positivity and the histological grades 7,19,30 . Our results showed a significant difference of MIB-1 positivity in both epithelia and stroma components. The expression of MIB-1 in epithelium was 14% (6/42), 42% (5/12) and 33% (1/3), respectively in benign, borderline and malignant PTs and 12%, 33% and 17% in the stromal component, similar to an earlier report by Tse et al (2010); 5-25%, 15-100% in benign and malignant cases, respectively 30. MIB-1 negativity or low expression indicated a very low proliferation rate, apparently noticeable in fibroadenoma and benign PT. Benign phyllodes tumors with more than 10% of cells were positive for MIB-1 were at risk of malignant change, as described by Chan et al (2004) 17,31.

Like Ki-67, p53 status may be correlated to histological grade of tumors. Published studies reported that Ki-67 and p53 expressions correlated well with the morphologic grading of phyllodes tumor 7,17,19. P53 is useful as independent criteria for evaluating the malignancy of phyllodes tumor and its expression tends to be greater in the phyllodes tumor with a higher malignant potential 32. In the present study, high level p53 expression was confined to stromal cells, with only low levels of expression found in the epithelium, 26.8% (30/112), 9.8% (4/41), 8.3% (1/12) of fibroadenoma, benign PT and borderline PTs respectively, but was absence in the epithelium of malignant PT. In the stroma component, there was significant association of Ki-67 with histological grade with 7% (8/114), 4.8% (2/42), 33.3% (4/12) and 50% (6/12) of fibroadenoma benign PT, borderline PT and malignant PTs respectively, supporting the assertion that the stroma rather than the epithelia component is neoplastic 27 . Even so, interactions between both stroma and epithelia, may influence the pathogenesis of PTs 27. Although p53 expression correlated well with the histological grades, however immunopositive of p53 may not detect an increased risk for developing breast cancer among patient with benign tumor 33. In addition, the negativity of p53 in the stromal cells confirmed the benign of the phyllodes tumor 21,30,34. In this study, only 4.8% of weak staining was detected in the stroma of benign PTs while others remained negative.

Neither Ki-67 nor p53 expression may distinguish benign from malignant phyllodes tumor in diagnostically difficult cases such as borderline or low proliferative malignant cases. Ki67 antigen presents as the true proliferation activity and overexpression of p53 is caused by detection of mutation in a case that progressed from benign to malignant tumor 17. The association between Ki67 and p53 with histological grade, indicates that their immunoreactivity in benign breast lesion may increase risk for malignancies and demonstrate to be possible indicator in PTs while ER expression status plays the role in determining the treatment in breast tumor 35. A positive association between ER and Bcl-2 suggests that ER may be the regulator of Bcl2 protein. Besides, high detecting of ERβ subtype compared to ERα especially in the stroma component may suggest that ERβ is a dominant ER in the breast tissues (Speirs et al 2002).