|
|
||||||||
, PR, HER-2/neu, and EGFR in Different Cell Origin Subtypes of High Grade and Non-High Grade Ductal Carcinoma In Situ
Address correspondence to Ping Tang, M.D., Ph.D., Department of Pathology and Laboratory Medicine, University of Rochester Medical Center, 601 Elmwood Ave, Box 626, Rochester, New York 14642 USA; tel 585 275 6640; fax 585 273 3637; e-mail: ping_tang{at}urmc.rochester.edu.
| Abstract |
|---|
|
|
|---|
), progesterone receptor (PR), HER-2/neu, and epidermal growth factor receptor (ERFR) in 53 cases of non-high grade and 46 cases of high nuclear grade DCIS. Using a panel of antibodies to ER-
, PR, HER-2/neu, and EGFR, along with cytokeratin (CK) markers (CK5/6, CK8, CK14, CK17, and CK18), we found that all 3 cell origin subtypes can express ER-
and PR, and their expression is higher in non-high grade DCIS than in high grade DCIS; the expression of HER-2/neu is associated with luminal subtype only in non-high grade DCIS, but can be seen in all 3 subtypes in high grade DCIS; the expression of EGFR is low and is present only in luminal cell subtypes in both high and non-high grade DCIS. Basal/ stem cell and null cell subtypes occur in younger patients in non-high grade DCIS compared to high grade DCIS. In conclusion, the expression patterns of ER-
, PR, HER-2/neu, and EGFR are markedly different in different cell origin subtypes of both high grade and non-high grade DCIS, suggesting that cell origin subtypes as well as nuclear grade contribute to the biological and molecular heterogeneity of DCIS.
Keywords: breast cancer, ductal carcinoma in situ, cell origin markers, ER-
, PR, HER-2/neu, EGFR, cytokeratins, nuclear grade
| Introduction |
|---|
|
|
|---|
Numerous studies have attempted to identify morphologic and molecular factors that influence the prognosis of DCIS. Nuclear grade, one of the most important factors, is the main criterion used to classify DCIS as high, intermediate, and low grade [3,4], and is also one of the important pathologic features used in the Van Nuys Prognostic Index for DCIS [5]. However, nuclear grade is not the only factor determining the prognosis of DCIS. Low grade DCIS is frequently associated with invasive carcinoma and metastasis, suggesting that other factors are also important. Cornfield et al [6] reported that many molecular markers, including ER, PR, p53, Ki-67, HER-2/neu, bcl-2, and p21, have no significant independent prognostic value for DCIS. We have reported that both high grade and non-high grade DCIS can be further subdivided into 3 subgroups according to the expression patterns of cytokeratin cell origin markers (CK5/6 for stem cells, CK8 and 18 for luminal cells, and CK14 and 17 for basal cells) and that their distributions are strongly associated with nuclear grade [7]. High grade DCIS has significantly more basal and stem cell subtypes compared to non-high grade DCIS, suggesting that at least in a subpopulation, cell origins of DCIS may play a key role in determining its biological behavior. Here we extend our study of relationships between cytokeratin cell origin markers and the expression of ER-
, PR, HER-2/neu, and EGFR.
| Materials and Methods |
|---|
|
|
|---|
, PR, HER-2/neu, and EGFR, as well as antibodies to CKs 5/6, 8, 14, 17, and 18. Positive staining was defined as
10% of the tumor cells with nuclear staining for ER-
and PR, with 3+ complete membrane staining for HER-2/neu and EGFR, and with strong cytoplasmic staining for CKs. The sources, dilutions, and pretreatments for each antibody are listed in Table 1
|
, PR, and HER-2/neu. For example, CMH was used to test the relationship between nuclear grade and ER positivity while the confounding factor, cell type, was controlled. Fishers exact test was used to calculate p values for ER-
, PR, HER-2, and EGFR in luminal cell types. All statistical calculations were carried out with SAS software (Statistical Analysis System, Inc., Cary, NC). | Results |
|---|
|
|
|---|
, PR, HER-2/neu, and EGFR in high grade and non-high grade DCIS are illustrated in Fig. 1
and PR, and with no expression of HER-2/neu or EGFR, were identified on the same sections in the majority of the cases tested. CK5/6 antibody stained cells in basal layers, CK8 and 18 stained cells in luminal cell layers, and CK14 and 17 stained cells in basal layers. The DCIS was subclassified as luminal cell, basal/stem cell, or null cell subtype based on
10% of the tumor cells being positive for luminal cell markers (CK8 and CK18 luminal), basal or stem cell markers (CK14, CK17, and CK5/6 basal/stem), or negative for all of these markers (null).
|
and PR in all 3 subtypes.
The expression of ER-
and PR was observed in all 3 subtypes of high grade and non-high grade DCIS, and had higher frequency in non-high grade DCIS (92% and 77%, respectively) than in high grade DCIS (30% and 23%, respectively) (Tables 2
, a CMH Chi-square value of 34.74 with 1 degree of freedom (df) indicated a p value <0.0001. For PR, the CMH Chi-square value of 20.98 (df = 1) indicated a p value <0.0001. The expression of ER-
and PR was most often associated with luminal cell subtypes of both high grade (35% and 29%) and non-high grade DCIS (93% and 80%), although expression of both receptors was also noted in basal/stem cell subtype in both high grade DCIS (14% and 7%) and non-high grade DCIS (75% and 25%) with a p value of 5.476 x 108 for ER-
and 1.182 x 105 for PR.
|
|
EGFR expressed only in luminal cell type.
The expression of EGFR was a rare event. It was seen in only 3 of 46 (7%) high grade DCIS and 1 of 53 (2%) non-high grade DCIS (Table 2
). Furthermore, all 4 EGFR positive cases were of the luminal subtype. There was no apparent correlation between the expression of EGFR and the expression of ER-
, PR, or HER-2/neu. Among the 3 EGFR-positive high grade DCIS, one was positive for ER-
, PR, and HER-2/neu, one was positive for ER-
, and negative for PR and HER-2/neu, and one was negative for ER-
, PR, and HER-2/neu. The one EGFR-positive non-high grade DCIS was positive for HER-2/neu and negative for ER and PR.
Younger patients.
Although there was no significant age difference between high grade (mean 58.5 yr) and non-high grade DCIS (54.9 yr) as a whole, the patients were younger in non-high grade DCIS (53.8 and 46.6 yr) compared to high grade DCIS (61.6 and 73.0 yr) in basal/stem cells and null cell subtypes, respectively (Table 2
).
| Discussion |
|---|
|
|
|---|
Nuclear grade has been a key morphological factor used not only for classification, but also as a prognostic factor for DCIS [35]. We have reported that both high grade and non-high grade DCIS are associated with all 3 cell origin subtypes, although a significantly higher proportion of high grade DCIS belongs to stem/basal subtype compared to non-high grade DCIS, suggesting that within the same nuclear grade there are genetically distinct subtypes for DCIS, which may ultimately determine the biological course of individual tumors [7]. Furthermore, a triple expression pattern of the 3 cell origin markers is associated with high grade DCIS, indicating that a more complicated carcinogenesis pathway (trans-differentiation) exists in high grade lesions. These findings not only provide insight into early carcinogenesis in the breast, but also permit subdivision of breast carcinoma into 3 subgroups with possibly different prognosis and potential to progress to invasive carcinoma.
Estrogen plays a critical role in the development of breast cancer. Many breast carcinomas express ER-
and PR, and their expressions have been used routinely to guide clinical management and predict response to therapy [16]. Recently, ER-
status has been shown to be associated with a distinct gene expression pattern [17]. A second type of estrogen receptor (ER-ß), identified in 1996, that differs from ER-
in tissue distribution and function, is shown to be a counterpart of ER-
in breast carcinogenesis [18,19]. In the present study, we have shown that the expressions of ER-
and PR are not only associated with better-differentiated (non-high grade) DCIS, but are also more likely associated with DCIS from luminal cell lineage. These findings support the concepts that distinct pathways exist in breast carcinogenesis and that different progenitor cells may give rise to genetically different subtypes of DCIS.
HER-2/neu over-expression is often associated with aggressive histological subtypes and poor prognosis for breast carcinoma [20]. Over-expression or amplification of the gene or its product has been routinely tested clinically by immunohistochemical staining or FISH to help guide clinical therapy with Herceptin and other anticancer agents [21]. Our results show that over-expression of HER-2/neu is strongly associated with high grade DCIS. More important, HER-2/neu is overexpressed only in the luminal subtype in low grade DCIS, but in all three subtypes in high grade DCIS, suggesting that a tight regulatory mechanism for HER-2/neu expression is disrupted in high grade DCIS. These results indicate that although HER-2/neu has an important role in carcinogenesis in high grade DCIS, its role in non-high grade DCIS is limited. These findings support the concept that different progenitor cells give rise to genetically different subtypes of DCIS. Phosphorated HER-2/neu has been shown to be a better predictor for chemotherapy response than overexpression of "total" HER-2/neu [22,23], and it is also a better prognostic marker for breast cancer [24]. Thus, further studies of phosphorated HER-2/neu may have significant impact on the clinical management of breast carcinomas.
The expression of EGFR has not been routinely tested clinically in breast carcinomas. However, EGFR expression is associated with high histological grade, high Nottingham Prognostic Index (NPI) score, negative ER status, larger tumor size, distant metastasis, and death [21]. An inhibitor of EGFR has been shown to decrease epithelial proliferation in ER-negative and EGFR positive DCIS [25]. Furthermore, co-expression of growth factor receptor family members, including EGFR and HER-2/neu, is associated with poor clinical outcome [26,27]. In the present study, we found that EGFR over-expression is rare and that its over-expression is seen only in luminal cell subtypes in both high grade and non-high grade DCIS, differs from the expression pattern seen for HER-2/neu, and does not correlate with the expressions of ER-
, PR, or HER-2/neu. These findings suggest that regulation of EGFR expression is restricted to the terminally differentiated luminal cell subtypes, and that its regulatory pathways and role in breast carcinogenesis are different from those of HER-2/neu.
Age has long been recognized as an important factor that affects the prognosis for breast carcinoma and age has recently been added to the revised Van Nuys Prognostic Index [5]. Age affects not only the incidence, but also the biology of breast cancer [28]. Breast carcinoma arising in older patients has a slower growth rate, is more likely to be ER-positive, and is less likely to be positive for p53, HER-2/neu, and EGFR [29]. We have observed variation in the expression of cytokeratin cell origin markers according to age. The basal/stem cell and null cell subtypes in non-high grade DCIS are more frequently seen in younger patients. Since the number of cases in our study was limited, the precise relationships between age and different cell origin subtypes remain to be determined.
| Acknowledgment |
|---|
| Footnotes |
|---|
b Steven Hajdu, M.D., Professor of Pathology, New York University School of Medicine, New York, NY. ![]()
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. Antuofermo, M. A. Miller, S. Pirino, J. Xie, S. Badve, and S. I. Mohammed Spontaneous Mammary Intraepithelial Lesions in Dogs A Model of Breast Cancer Cancer Epidemiol. Biomarkers Prev., November 1, 2007; 16(11): 2247 - 2256. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Mimeault and S. K. Batra Interplay of distinct growth factors during epithelial mesenchymal transition of cancer progenitor cells and molecular targeting as novel cancer therapies Ann. Onc., October 1, 2007; 18(10): 1605 - 1619. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Khilko, P. Bourne, Qi Yang, and Ping Tang Mismatch Repair Genes hMLH1 and hMSH2 May Not Play an Essential Role in Breast Carcinogenesis International Journal of Surgical Pathology, July 1, 2007; 15(3): 233 - 241. [Abstract] [PDF] |
||||
![]() |
S. Steinman, J. Wang, P. Bourne, Q. Yang, and P. Tang Expression of Cytokeratin Markers, ER-alpha, PR, HER-2/neu, and EGFR in Pure Ductal Carcinoma In Situ (DCIS) and DCIS with Co-existing Invasive Ductal Carcinoma (IDC) of the Breast Ann. Clin. Lab. Sci., January 1, 2007; 37(2): 127 - 134. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |