- Email: [email protected]
Matricellular protein CCN1 (CYR61) expression is associated with high-grade ductal carcinoma in situ
Human Pathology (2014) xx, xxx–xxx
www.elsevier.com/locate/humpath
Original contribution
Matricellular protein CCN1 (CYR61) expression is associated with high-grade ductal carcinoma in situ☆ Ozlen Saglam MD a,⁎, Feng Dai PhD b , Seema Husain PhD c , Yilei Zhan BS b , Gokce Toruner MD, PhD c , G. Kenneth Haines III MD a a
Yale University School of Medicine, New Haven, CT 06512, USA Yale Center for Analytic Sciences, New Haven, CT 06511, USA c Rutgers New Jersey Medical School, Newark, NJ 07103, USA b
Received 15 December 2013; revised 25 January 2014; accepted 7 February 2014
Keywords: DCIS; CCN1; Cyclin D1; β-catenin
Summary Cysteine-rich protein 61, connective tissue growth factor, and nephroblastoma overexpressed gene (CCN) comprise a family of matricellular proteins that have multiple physiologic functions including development, tissue repair, cell adhesion, migration, and proliferation. The expression of CCN1, cyclin D1, β-catenin, and p53 was explored by immunohistochemistry in different grades of ductal carcinoma in situ (DCIS) cases. These cases did not contain any infiltrating carcinoma components. In addition, all cysteine-rich protein 61 gene exons (encoding the CCN1 protein) were sequenced in 30 samples. Allred and H-scores were calculated for expression in both DCIS and the surrounding benign breast tissue. All cases of DCIS showed degrees of cytoplasmic CCN1 staining with median H-scores of 170, 160, and 60 in grades 3, 2, and 1, respectively (P = .043). Twelve of 28 DCIS 3, 1 of 15 DCIS 2, and 0 of 18 DCIS 1 also showed nuclear staining for CCN1. The cytoplasmic staining difference was preserved when the cases were divided into estrogen receptor (ER)+/DCIS grade 1, ER+/DCIS 2 and 3, and ER−/DCIS 2 and 3 by the H-score (P = .037). Cyclin D1 expression was positively correlated with the CCN1 cytoplasmic H-score in all DCIS samples (P = .038). Membranous β-catenin expression correlated with the grade of intraepithelial carcinoma by both H-score (P = .047) and Allred score (P = .026). Our results suggest that CCN1 has a role in the development of intraepithelial carcinoma. CCN1 expression correlates with grade of DCIS independent of ER status. It can induce cell cycle progression through cyclin D1. It is warranted to study high expression of CCN1 in DCIS as an independent risk factor in a larger cohort. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The risk for progression to cancer among different breast lesions was described by Page and Dupont [1] in 1988. ☆ Disclosures: Authors do not have any conflict of interest. ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (O. Saglam).
0046-8177/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.humpath.2014.02.007
Ductal carcinoma in situ (DCIS) carries the highest risk for progression to infiltrating duct carcinoma [2]. Most authors consider DCIS a heterogeneous group of diseases rather than a single entity that progresses from a low- to high-grade lesion before invasion [3]. Low-grade DCIS has a much more indolent course for development of an infiltrating carcinoma compared with a high-grade lesion [4]. Younger patients with DCIS are more likely to have multicentric disease and loco-regional recurrence compared with older
2 women [5]. Based on differences in the clinical behavior and age of occurrence, it is reasonable to hypothesize that distinct molecular pathways are involved in the pathogenesis and progression of different grades of DCIS. The role of extracellular matrix proteins has been studied in the development and progression of the breast carcinoma [6]. Cysteine-rich protein 61 (CYR61), connective tissue growth factor, and nephroblastoma overexpressed gene, together termed “CCN,” comprise a family of matricellular proteins that have multiple physiologic functions including development, tissue repair, cellular adhesion, migration, and proliferation [7,8]. There are 6 known members of the family named in order of discovery [9]. CCN family members have different roles in a variety of tissues. CCN proteins bind to cell surface integrins and heparin sulfate proteoglycans [10]. Depending on the subtype of CCN family member, type of receptors, and target tissue, they may promote or inhibit cancer formation. CCN1, CCN2, and CCN3 induce intracellular events including kinase activation and gene transcription [7]. CCN3 promotes prostate cancer bone metastasis [11], whereas CCN2 appears to suppress lung cancer metastasis [12]. The role of the CCN family in infiltrating breast carcinoma has been described [13-15]. CCN1 expression is associated with higher-grade breast cancer with lymph node metastasis, whereas CCN6 inhibits breast cancer progression [16]. Multiple signaling pathways including PI3K, AKT, and ILK, transcription factors such as β-catenin, and regulatory molecules such as p53 and cyclin D1 are activated or upregulated through CCN1 signaling [17,18]. Overexpression of CCN1 has been found to promote β-catenin nuclear translocation in some cancers such as glioma and non–small cell lung cancer [18,19]. This nuclear translocation in turn activates many growth-related genes including CCN1 in a positive feedback loop. Although CCN1 has been reported to play a role in infiltrating breast carcinoma, there is no detailed study about CCN1’s role in the pathologic precursor, DCIS. Immunostains for CCN1, p53, β-catenin, and cyclin D1 were used to evaluate their expression and possible interaction in different grades of breast ductal intraepithelial neoplasia.
2. Materials and methods After approval from the human investigation committee, DCIS cases (n = 61) were identified from the Yale Pathology Department’s archives diagnosed between 2007 and 2010. Most cases (n = 49) were from core needle biopsies. An additional 12 cases were obtained from excision specimens. Hematoxylin and eosin–stained sections, estrogen receptor (ER), and progesterone receptor (PgR) immunostained slides were reviewed. Diagnosis and the grade of disease were confirmed. DCIS grade was determined by using guidelines published by College of American Pathologists. These include nuclear size, pleomorphism, chromatin pattern, orientation, presence of nucleoli, and mitosis. DCIS archi-
O. Saglam et al. tectural patterns and presence or absence of necrosis were reviewed. Selected blocks from 61 cases were used for CCN1 immunostain, and 42 of the cases were also immunstained for cyclin D1, p53, and β-catenin. All CYR61 gene exons (encoding CCN1 protein) were sequenced in 30 samples. All DCIS cases studied by immunohistochemistry and polymerase chain reaction (PCR) were pure in situ carcinomas. They did not show any invasive components.
2.1. Immunohistochemistry Five-micron sections were cut from paraffin blocks and mounted on charged slides. They were deparaffinized in xylene and rehydrated through graded concentrations of ethanol. Slides for cyclin D1 immunostain were immersed in EDTA, whereas those for CCN1, p53, and β-catenin were immersed in citrate buffer and kept between 95°C to 101°C for 30 minutes. They were allowed to cool to room temperature, rinsed, and transferred to Tris-buffered saline. Tris-buffered saline was used as a wash between each subsequent step. Endogenous peroxidase was quenched with hydrogen peroxide. Primary antibodies were applied in the following dilutions: CCN1 (ab82040; Abcam, Cambridge, MA), prediluted antibody required no further dilution; p53 (MS187; Thermo, Waltham, MA), diluted to 1:100; cyclin D1 (CME432; Biocare Medical, Concord, CA), diluted to 1:200; and β-catenin (CM406; Biocare Medical), diluted to 1:100.
2.2. Evaluation of immunostains DCIS and the surrounding benign-appearing breast tissue were evaluated separately for each immunostain. The intensity (0-3 scale) and percentage of positive DCIS cells and benign epithelium were separately recorded. Initially, 2 pathologists (O. S. and K. H.) read each case independently. If there was a disagreement, the case was reviewed and consensus agreement obtained. Both H-score and Allred score were calculated on each case for DCIS and the surrounding benign-appearing breast tissue when present. Hscores were calculated by multiplying the percentage of cells demonstrating each intensity scored from 0 to 3+ and adding the results. They ranged from 0 to 300. Allred scores were calculated by adding together proportion scores obtained from the percentage of positively stained cells and intensity scores. They ranged from 0 to 8. Immunostains for CCN1 were evaluated for both cytoplasmic and nuclear staining. Only nuclear staining was considered positive for cyclin D1 and p53. β-catenin immunostain was read for nuclear versus cytoplasmic and membranous staining.
2.3. DNA extraction and mutation analysis Tumor DNA was extracted from the marked dissected formalin-fixed, paraffin-embedded tissues. The formalinfixed, paraffin-embedded tissues were deparaffinized in
CCN1 expression in in situ breast carcinoma Hemo-De (Scientific Safety Solvents, Keller, TX) and washed with 100% ethanol. Genomic DNA was extracted using QIAamp DNA Micro kit (Qiagen, Valencia, CA). PCR and Sanger sequencing analysis of CYR61 gene were performed. All 5 exons were sequenced using the following primers: Exon 1 forward: ACCAGCTTGTTGGCGTCTT; exon 1 reverse: GGGAATAGGTGGCAAAAGGA Exon 2 forward: CCGAGTCTCACGCGTATCTT; exon 2 reverse: TTTTAAAGGGGCCAAACCAC Exon 3 forward: TCCCCTTCTACCTTTCTCTTTTG; exon 3 reverse: GACACAAACAGGCTCAAGCA Exon 4 forward: TGTCGGTATGCCTCTGAGAA; exon 4 reverse: CTGCCAGGCAACAGTCTACA Exon 5 forward: CCCCTAACTTTCCTTCTCTCC; exon 5 reverse: GTCTAGGTGTGCCCTGGAAA The PCR conditions were as follows: (i) 95°C for 5 minutes; (ii) 94°C for 30 seconds, 60°C for 30 seconds (for exons 1, 2, 3, and 4), and −58°C for 30 seconds (for exon 5); 72°C for 1 minute (40 cycles); and (iii) 72°C for 10 minutes. The PCR products were identified on a 1% agarose gel. Sequencing was carried out on the ABI 3130 XL Genetic Analyzer (Applied Biosystems, Grand Island, NY). Mutation Surveyor software package Version 4.0.5 (SoftGenetics, State College, PA) was used for data analysis.
2.4. Statistical analysis All continuous data were assessed for normality of distribution using the Shapiro-Wilk test of normality. As they were not normally distributed, our data were summarized as a median, min-max, and interquartile range (IQR). The Spearman rank correlation coefficient was computed to help quantify the association between expression levels (measured by Allred and H-scores) of 4 different biomarkers or association of age and biomarker expression levels, from using all tumor samples or subsamples stratified by DCIS grade and/or ER status. The differences of expression levels of 4 biomarkers among the subsamples with different DCIS grade (or DCIS grade + ER status) were compared using the nonparametric Kruskal-Wallis test or Wilcoxon rank sum test as appropriate. The differences in the biomarker expression levels between normal breast tissue and DCIS were compared using the Wilcoxon signed rank test. All statistical analyses were 2-sided and were performed with the use of statistical software SAS version 9.2 (SAS, Cary, NC). A P b .05 was considered to indicate statistical significance, unless otherwise specified.
3. Results The patients’ ages ranged from 34 to 82 years (mean: 58 years). The mean ages for patients with DCIS grades 1 to 3 were 60, 61, and 55, respectively. There was no association between age and marker expression. Architectural types
3 include solid, cribriform, papillary, and micropapillary. Ten of 28 cases of DCIS grade 3 were ER−/PgR−, 13 cases ER+/ PgR+, 4 cases ER+/PgR−, and 1 case ER−/PgR+. Necrosis was identified in 27 of 28 cases. All 15 cases of DCIS grade 2 showed necrosis. There were 2 ER−/PgR−, 1 ER+/PgR−, and 12 ER+/PgR+ DCIS grade 2 cases. The DCIS grade 1 (n = 18) cases were all ER+/PgR+. All DCIS cases showed varying degrees of cytoplasmic staining for CCN1. The percentage of positively stained cells in DCIS 3 cases ranged from 5% to 95% and staining intensity from 1+ to 3+ (Fig.). H-scores ranged from 5 to 285 (median: 170) and Allred score from 3 to 8 (median: 7) (Table 1). Twelve cases of 28 DCIS grade 3 (48%) also showed nuclear staining ranging from 5% to 80% in quantity. The H-score for DCIS grade 2 cases varied from 20 to 285 (median: 160) and Allred scores between 4 and 8 (median: 7). Nuclear staining for immunostain CCN1 was observed only in 1 DCIS grade 2 case (20%, 2+). There was no nuclear CCN1 staining in any DCIS grade 1 cases. Only 4 of 15 DCIS grade 1 cases (27%) showed cytoplasmic staining reaching 3+ intensity. H-scores of DCIS grade 1 cases ranged from 1 to 270 (median: 60) and Allred scores from 2 to 7 (median: 5). There was a difference in CCN1 cytoplasmic H-scores (P = .043) but not Allred scores among DCIS grades. When cases were divided into ER+/DCIS 1, ER+/DCIS 2 and 3, and ER−/DCIS 2 and 3, the difference in cytoplasmic CCN1 staining was still significant (P = .037) (Table 2). The highest median was observed in ER+/DCIS grades 2 and 3 (H-score median: 160), followed by ER−/DCIS grades 2 and 3 (median: 100), and ER+/DCIS grade 1 (median: 60). Sequencing of the CYR61 gene encoding CCN1 revealed no pathogenic somatic mutations in any analyzed specimens. Twenty-one DCIS grade 3, 10 DCIS grade 2, and 11 DCIS grade 1 were available for β-catenin, p53, and cyclin D1 immunostaining. No nuclear β-catenin staining was seen in any case. Membranous and cytoplasmic β-catenin staining was observed in all 21 DCIS grade 3 cases, with H-scores ranging from 1 to 285 (median: 150) and Allred scores from 4 to 8 (median: 7). The median H-scores of βcatenin cytoplasmic/membranous expression in DCIS grades 2 and 1 were 150 and 60, respectively. There was a significant difference in β-catenin expression among different DCIS grades by both H-score (P = .047) and Allred score (P = .026). This difference remained when the cases were divided into ER+/DCIS grade 1, ER−/DCIS 2 and 3, and ER+/DCIS 2 and 3 groups for H-score (P = .045) and Allred score (P = .026). Cyclin D1 was positive in 19 of 21 DCIS grade 3 cases. The H-scores ranged from 0 to 270 (median: 120) and Allred score was between 0 and 8 (median: 7). In DCIS cases with focal cyclin D1 staining, the positive cells were located at the periphery of the duct adjacent to the residual myoepithelial layer. DCIS grades 2 and 1 had the median H-scores of 180 and 60, respectively. The H-score and Allred score for cyclin D1 showed positive correlations with that for CCN1 cytoplasmic expression in all DCIS samples (P = .038 and
4
O. Saglam et al.
Fig. Photomicrograph shows examples of marker expression. A, CCN1 immunostain: the cytoplasmic marker expression is diffuse and strong in DCIS200 Inset: CCN1 nuclear staining is shown. B, Cyclin D1. C, β-catenin. D, p53. ×200 (A and B), ×400 (inset), ×100 (C and D).
Table 1
Summary of marker expression by DCIS grade
Marker
H-score CCN-cytoplasmic Cyclin D1 p53 β-catenin Allred CCN-cytoplasmic Cyclin D1 p53 β-catenin
DCIS grade 1
DCIS grade 2
DCIS grade 3
n
Median (IQR)
n
Median (IQR)
n
Median (IQR)
18 11 11 11
60.0 (10.0-160.0) 60.0 (40.0-180.0) 30.0 (5.0-80.0) 60.0 (20.0-120.0)
15 10 10 10
160.0 (120.0-240.0) 180.0 (60.0-240.0) 50.0 (20.0-80.0) 150.0 (40.0-210.0)
28 21 21 21
170.0 (95.0-195.0) 120.0 (60.0-210.0) 90.0 (20.0-140.0) 150.0 (100.0-180.0)
18 11 11 11
5.0 (3.0-7.0) 6.0 (5.0 - 7.0) 4.0 (3.0-6.0) 5.0 (4.0-6.0)
15 10 10 10
7.0 (6.0-8.0) 7.5 (6.0-8.0) 5.0 (4.0-6.0) 7.0 (5.0-8.0)
28 21 21 21
NOTE. Data are summarized by median (interquartile range). The Kruskal-Wallis test was used for statistical analysis.
7.0 (6.0-7.5) 7.0 (6.0-8.0) 6.0 (4.0-7.0) 7.0 (6.0-7.0)
P
.043 .302 .214 .047 .089 .249 .180 .026
CCN1 expression in in situ breast carcinoma Table 2
5
Summary of marker expression by DCIS grade and ER status
Marker
H-score CCN-cytoplasmic Cyclin D1 p53 β-catenin Allred CCN-cytoplasmic Cyclin D1 p53 β-catenin
DCIS grade 1, ER+
DCIS grade (2 + 3), ER−
DCIS grade (2 + 3), ER+
n
Median (IQR)
n
Median (IQR)
n
Median (IQR)
18 11 11 11
60.0 60.0 30.0 60.0
(15.0-160.0) (40.0-180.0) (5.0-80.0) (20.0-120.0)
13 8 8 8
100.0 (90.0-180.0) 120.0 (20.0-210.0) 105.0 (30.5-160.0) 170.0 (80.0-195.0)
30 23 23 23
160.0 (140.0-210.0) 150.0 (60.0-240.0) 60.0 (20.0-140.0) 140.0 (40.0-210.0)
18 11 11 11
5.0 6.0 4.0 5.0
(3.0-7.0) (5.0-7.0) (3.0-6.0) (4.0-6.0)
13 8 8 8
6.0 (6.0-7.0) 6.5 (2.5-8.0) 6.0 (3.5-7.0) 7.0 (5.5-7.5)
30 23 23 23
7.0 (7.0-8.0) 7.0 (6.0-8.0) 5.0 (4.0-7.0) 7.0 (5.0-8.0)
P
.037 .259 .243 .045 .060 .220 .252 .026
NOTE. Data are summarized by median (interquartile range). The Kruskal-Wallis test was used for statistical analysis. There were no DCIS grade 1 and ERnegative patients.
P = .006) (Supplementary Tables 1 and 2). This association was lost when the cases were divided into ER−/+ groups. Immunostain for p53 was positive in 20 of 28 DCIS grade 3 cases with a median H-score of 90. For intermediate- and low-grade DCIS, median H-scores were 50 and 30, respectively. There was no statistical difference in p53 expression among the grades of DCIS. There was no nuclear expression of CCN1 in the benign breast tissue surrounding any grade of DCIS. The H-score of cytoplasmic CCN1 varied from 1 to 160 in all grades. The median H-scores of cytoplasmic CCN1, cyclin D1, p53, and β-catenin were 10, 10, 20, and 35, respectively. Wilcoxon signed rank test showed significant differences in expression of all markers between DCIS and the adjacent benign breast tissue (P b .001).
3.1. Clinical follow-up and additional findings on excision specimens Two patients did not have any further treatment after initial diagnosis in our institution. On excision specimens, most patients had only DCIS with matching grade to the core biopsy diagnosis. There were 7 invasive carcinomas (including 2 with microinvasion) on the excision materials after DCIS diagnosis on core needle biopsies. Two patients with DCIS grade 3 had microinvasive duct carcinoma. Their H-scores for nuclear CCN1 were 240 and 150. Cytoplasmic H-scores were 180 and 285. The remaining cases consisted of 3 T1a, 1 T1b, and 1 T1c tumors. Two of these were associated with DCIS grade 3, 1 with DCIS grade 2, and 2 with DCIS 1. Both tumors with DCIS grade 1 were welldifferentiated invasive carcinomas. The cytoplasmic Hscores were 210 and 10 on the core needle biopsies. Clinical follow-up ranged from 11 to 68 months (median: 36 months). One patient with DCIS 1 developed contralateral DCIS grade 1 with microinvasive carcinoma 2 years after her initial diagnosis. The cytoplasmic H-score for CCN1 was 210 for this case. Another patient with DCIS 1 developed recurrent
disease within 3 years. The H-score for cytoplasmic CCN1 was 160 in this case. A patient with DCIS 3 had metastatic lung cancer to the axilla. There were no other documented recurrent or contralateral diseases in our records.
4. Discussion Extracellular matrix components play a significant role in regulating epithelial cell growth and survival in a variety of settings [20]. The interaction between DCIS and its surrounding stroma produces changes in the microenvironment before tumor cells actually invading into the stroma [21]. CCN1 as a matricellular protein is up-regulated in wound healing and induces several genes such as matrix metalloproteinase (MMP)-1, MMP-3, vascular endothelial growth factors A and C, interleukins, and others [22]. MMP1 activates the breast cancer invasion–associated molecule protease activator receptor 1. Protease activator receptor 1 can induce CCN1 expression in breast cancer cells, which promotes invasion by further augmenting MMP-1 expression in adjacent stromal fibroblasts [23]. Leu et al [24] developed an anti-CCN1 antibody that suppresses Rac1mediated tumor cell migration and invasion. We observed a positive correlation between CCN1 expression and grades of DCIS independent of ER status. Showing higher expression of CCN1 in DCIS 3 is compatible with higher CCN1 expression in the advanced-stage infiltrating duct carcinoma as previously reported in the literature [25]. Recent genetic expression profiling demonstrated similarities between low-grade DCIS and its low-grade invasive carcinoma component and between high-grade DCIS and high-grade infiltrating duct carcinoma [26]. One of 2 DCIS grade 1 cases with associated invasive carcinoma on the excision specimen has higher cytoplasmic expression of CCN1 compared with median CCN1 expression in this group. On the other hand, both recurrence and contralateral disease were observed in patients with DCIS 1
6 expressing high cytoplasmic levels CCN1. These results are limited observations. Whether high expression of CCN1 in low-grade DCIS is a risk factor for recurrence or progression to invasive disease would require long-term follow-up in a larger cohort. CCN1 expression has been correlated with estrogen levels. Estradiol was shown to increase CCN1 expression in the uterus of oophorectomized mice [27]. Estrogen induces CCN1 in breast cancer cell lines in a dose-dependent manner and is blocked by neutralizing antibodies [15]. In our cohort, ER+ intermediate- and high-grade DCIS have higher median CCN1 scores compared with ER− equivalents. This finding is compatible with previous studies. On the other hand, Tsai et al [28] have demonstrated that CCN1 is sufficient to demonstrate estrogen independence and resistance to antiestrogens in vivo. ER+ low-grade DCIS cases had the lowest expression of CCN1 among all 3 groups. These results indicate that the grade of disease also has an impact in the cyctoplasmic expression of CCN1 regardless of hormone receptor status. Nuclear expression of CCN1 was observed in 12 DCIS grade 3 and 1 DCIS grade 2 cases. Truncated forms of CCN1 are known to translocate to the nucleus, where they are involved in transcriptional regulation [29]. For example, CCN1 was found to localize both in cytoplasm and nucleus of cultured bladder smooth muscle cells [30]. Nuclear expression of CCN1 in invasive breast carcinoma has not been previously reported. The significance of this finding in highgrade DCIS is not certain and requires further investigation. CCN1 and β-catenin interact closely in the Wnt signaling pathway. When Wnt binds to its receptor, β-catenin is released from the cell membrane and translocates to the nucleus, where it associates with ternary complex factor. Transcriptional targets of the Wnt pathway include the cellular oncogenes cyclin D1 and c-myc [31]. CCN1 reportedly leads to β-catenin translocation through degradation of E-cadherin [8]. CCN1 induces β-catenin translocation in squamous cells carcinoma of esophagus [32] and is involved in the progression of hepatocellular carcinoma via Wnt/β-catenin signaling [33]. There was no nuclear β-catenin staining in our DCIS cases. Membranous/ cytoplasmic β-catenin expression varied significantly by both H-score and Allred scores in different DCIS grades regardless of ER status. The biological significance of membranous staining in our DCIS cases is not clear. Cell cycle progression requires coordinated interactions between cyclins and cyclin-dependent kinases (CDK). The binding of cyclin D1 to CDK4 and CDK6 is a rate-limiting step during cell cycle progression [34]. Cylin D1 is overexpressed in a number of cancers including the breast, liver, lung, and brain [35-37]. Assoian and Klein [38] showed cyclin D1 induction requires coordinated signaling from the extracellular matrix and soluble growth factors. Supporting this finding, we observed a positive correlation between CCN1 and cyclin D1 expression in all DCIS grades. In summary, our results indicate that the role of CCN1 in the breast cancer development begins in intraepithelial carcinoma
O. Saglam et al. stage. Although no mutation in the exons of CCN1 gene was detected, the CCN1 protein expression has a positive correlation with cyclin D1 and possibly with β-catenin. In addition, cytoplasmic expression of CCN1 was associated with DCIS grade independent of ER status. We found no definite interaction between CCN1 and p53 in intraductal carcinoma. Further observational and functional studies are warranted to determine the exact role of CCN1 in pathogenesis of DCIS.
Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.humpath.2014.02.007.
References [1] Page DL, Dupont WD. Histopathologic risk factors for breast cancer in women with benign breast disease. Semin Surg Oncol 1988;4:213-7. [2] Betsill Jr WL, Rosen PP, Lieberman PH, Robbins GF. Intraductal carcinoma. Long-term follow-up after treatment by biopsy alone. JAMA 1978;239:1863-7. [3] Punglia RS, Schnitt SJ, Weeks JC. Treatment of ductal carcinoma in situ after excision: would a prophylactic paradigm be more appropriate? J Natl Cancer Inst 2013;105:1527-33. [4] Sanders ME, Schuyler PA, Dupont WD, Page DL. The natural history of low-grade ductal carcinoma in situ of the breast in women treated by biopsy only revealed over 30 years of long-term follow-up. Cancer 2005;103:2481-4. [5] Alvarado R, Lari SA, Roses RE, et al. Biology, treatment, and outcome in very young and older women with DCIS. Ann Surg Oncol 2012;19:3777-84. [6] Vargas AC, Reed AE, Waddell N, et al. Gene expression profiling of tumour epithelial and stromal compartments during breast cancer progression. Breast Cancer Res Treat 2012;135:153-65. [7] Chen CC, Lau LF. Functions and mechanisms of action of CCN matricellular proteins. Int J Biochem Cell Biol 2009;41:771-83. [8] Chong HC, Tan CK, Huang RL, Tan NS. Matricellular proteins: a sticky affair with cancers. J Oncol 2012;2012:351089. [9] Brigstock DR. The CCN family: a new stimulus package. J Endocrinol 2003;178:169-75. [10] Kireeva ML, Mo FE, Yang GP, Lau LF. Cyr61, a product of a growth factor-inducible immediate-early gene, promotes cell proliferation, migration, and adhesion. Mol Cell Biol 1996;16:1326-34. [11] Chen PC, Cheng HC, Tang CH. CCN3 promotes prostate cancer bone metastasis by modulating the tumor-bone microenvironment through RANKL-dependent pathway. Carcinogenesis 2013;34:1669-79. [12] Chang CC, Yang MH, Lin BR, et al. CCN2 inhibits lung cancer metastasis through promoting DAPK-dependent anoikis and inducing EGFR degradation. Cell Death Differ 2013;20:443-55. [13] Lin J, Huo R, Wang L, et al. A novel anti-Cyr61 antibody inhibits breast cancer growth and metastasis in vivo. Cancer Immunol Immunother 2012;61:677-87. [14] Tsai MS, Hornby AE, Lakins J, Lupu R. Expression and function of CYR61, an angiogenic factor, in breast cancer cell lines and tumor biopsies. Cancer Res 2000;60:5603-7. [15] Sampath D, Winneker RC, Zhang Z. Cyr61, a member of the CCN family, is required for MCF-7 cell proliferation: regulation by 17betaestradiol and overexpression in human breast cancer. Endocrinology 2001;142:2540-8.
CCN1 expression in in situ breast carcinoma [16] Jiang WG, Watkins G, Fodstad O, Douglas-Jones A, Mokbel K, Mansel RE. Differential expression of the CCN family members Cyr61, CTGF and Nov in human breast cancer. Endocr Relat Cancer 2004;11:781-91. [17] Menendez JA, Vellon L, Mehmi I, Teng PK, Griggs DW, Lupu R. A novel CYR61-triggered “CYR61-alphavbeta3 integrin loop” regulates breast cancer cell survival and chemosensitivity through activation of ERK1/ERK2 MAPK signaling pathway. Oncogene 2005;24:761-79. [18] Tong X, O'Kelly J, Xie D, et al. Cyr61 suppresses the growth of nonsmall-cell lung cancer cells via the beta-catenin-c-myc-p53 pathway. Oncogene 2004;23:4847-55. [19] Xie D, Yin D, Tong X, et al. Cyr61 is overexpressed in gliomas and involved in integrin-linked kinase-mediated Akt and beta-cateninTCF/Lef signaling pathways. Cancer Res 2004;64:1987-96. [20] Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144:646-74. [21] Ma XJ, Dahiya S, Richardson E, Erlander M, Sgroi DC. Gene expression profiling of the tumor microenvironment during breast cancer progression. Breast Cancer Res 2009;11:R7. [22] Chen CC, Chen N, Lau LF. The angiogenic factors Cyr61 and connective tissue growth factor induce adhesive signaling in primary human skin fibroblasts. J Biol Chem 2001;276:10443-52. [23] Nguyen N, Kuliopulos A, Graham RA, Covic L. Tumor-derived Cyr61(CCN1) promotes stromal matrix metalloproteinase-1 production and protease-activated receptor 1-dependent migration of breast cancer cells. Cancer Res 2006;66:2658-65. [24] Jim Leu SJ, Sung JS, Huang ML, Chen MY, Tsai TW. A novel antiCCN1 monoclonal antibody suppresses Rac-dependent cytoskeletal reorganization and migratory activities in breast cancer cells. Biochem Biophys Res Commun 2013;434:885-91. [25] Xie D, Nakachi K, Wang H, Elashoff R, Koeffler HP. Elevated levels of connective tissue growth factor, WISP-1, and CYR61 in primary breast cancers associated with more advanced features. Cancer Res 2001;61:8917-23.
7 [26] Ma XJ, Salunga R, Tuggle JT, et al. Gene expression profiles of human breast cancer progression. Proc Natl Acad Sci U S A 2003;100:5974-9. [27] Rivera-Gonzalez R, Petersen DN, Tkalcevic G, Thompson DD, Brown TA. Estrogen-induced genes in the uterus of ovariectomized rats and their regulation by droloxifene and tamoxifen. J Steroid Biochem Mol Biol 1998;64:13-24. [28] Tsai MS, Bogart DF, Castaneda JM, Li P, Lupu R. Cyr61 promotes breast tumorigenesis and cancer progression. Oncogene 2002;21:8178-85. [29] Yeger H, Perbal B. The CCN family of genes: a perspective on CCN biology and therapeutic potential. J Cell Commun Signal 2007;1:159-64. [30] Tamura I, Rosenbloom J, Macarak E, Chaqour B. Regulation of Cyr61 gene expression by mechanical stretch through multiple signaling pathways. Am J Physiol Cell Physiol 2001;281:C1524-32. [31] Smalley MJ, Dale TC. Wnt signalling in mammalian development and cancer. Cancer Metastasis Rev 1999;18:215-30. [32] Chai J, Modak C, Ouyang Y, Wu SY, Jamal MM. CCN1 induces betacatenin translocation in esophageal squamous cell carcinoma through integrin alpha11. ISRN Gastroenterol 2012;2012:207235. [33] Li ZQ, Ding W, Sun SJ, et al. Cyr61/CCN1 is regulated by Wnt/betacatenin signaling and plays an important role in the progression of hepatocellular carcinoma. PLoS One 2012;7:e35754. [34] Fu M, Wang C, Li Z, Sakamaki T, Pestell RG. Minireview: cyclin D1: normal and abnormal functions. Endocrinology 2004;145:5439-47. [35] Gillett C, Smith P, Gregory W, et al. Cyclin D1 and prognosis in human breast cancer. Int J Cancer 1996;69:92-9. [36] Hall M, Peters G. Genetic alterations of cyclins, cyclin-dependent kinases, and Cdk inhibitors in human cancer. Adv Cancer Res 1996;68:67-108. [37] Molenaar JJ, Ebus ME, Koster J, et al. Cyclin D1 and CDK4 activity contribute to the undifferentiated phenotype in neuroblastoma. Cancer Res 2008;68:2599-609. [38] Assoian RK, Klein EA. Growth control by intracellular tension and extracellular stiffness. Trends Cell Biol 2008;18:347-52.
www.elsevier.com/locate/humpath
Original contribution
Matricellular protein CCN1 (CYR61) expression is associated with high-grade ductal carcinoma in situ☆ Ozlen Saglam MD a,⁎, Feng Dai PhD b , Seema Husain PhD c , Yilei Zhan BS b , Gokce Toruner MD, PhD c , G. Kenneth Haines III MD a a
Yale University School of Medicine, New Haven, CT 06512, USA Yale Center for Analytic Sciences, New Haven, CT 06511, USA c Rutgers New Jersey Medical School, Newark, NJ 07103, USA b
Received 15 December 2013; revised 25 January 2014; accepted 7 February 2014
Keywords: DCIS; CCN1; Cyclin D1; β-catenin
Summary Cysteine-rich protein 61, connective tissue growth factor, and nephroblastoma overexpressed gene (CCN) comprise a family of matricellular proteins that have multiple physiologic functions including development, tissue repair, cell adhesion, migration, and proliferation. The expression of CCN1, cyclin D1, β-catenin, and p53 was explored by immunohistochemistry in different grades of ductal carcinoma in situ (DCIS) cases. These cases did not contain any infiltrating carcinoma components. In addition, all cysteine-rich protein 61 gene exons (encoding the CCN1 protein) were sequenced in 30 samples. Allred and H-scores were calculated for expression in both DCIS and the surrounding benign breast tissue. All cases of DCIS showed degrees of cytoplasmic CCN1 staining with median H-scores of 170, 160, and 60 in grades 3, 2, and 1, respectively (P = .043). Twelve of 28 DCIS 3, 1 of 15 DCIS 2, and 0 of 18 DCIS 1 also showed nuclear staining for CCN1. The cytoplasmic staining difference was preserved when the cases were divided into estrogen receptor (ER)+/DCIS grade 1, ER+/DCIS 2 and 3, and ER−/DCIS 2 and 3 by the H-score (P = .037). Cyclin D1 expression was positively correlated with the CCN1 cytoplasmic H-score in all DCIS samples (P = .038). Membranous β-catenin expression correlated with the grade of intraepithelial carcinoma by both H-score (P = .047) and Allred score (P = .026). Our results suggest that CCN1 has a role in the development of intraepithelial carcinoma. CCN1 expression correlates with grade of DCIS independent of ER status. It can induce cell cycle progression through cyclin D1. It is warranted to study high expression of CCN1 in DCIS as an independent risk factor in a larger cohort. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The risk for progression to cancer among different breast lesions was described by Page and Dupont [1] in 1988. ☆ Disclosures: Authors do not have any conflict of interest. ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (O. Saglam).
0046-8177/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.humpath.2014.02.007
Ductal carcinoma in situ (DCIS) carries the highest risk for progression to infiltrating duct carcinoma [2]. Most authors consider DCIS a heterogeneous group of diseases rather than a single entity that progresses from a low- to high-grade lesion before invasion [3]. Low-grade DCIS has a much more indolent course for development of an infiltrating carcinoma compared with a high-grade lesion [4]. Younger patients with DCIS are more likely to have multicentric disease and loco-regional recurrence compared with older
2 women [5]. Based on differences in the clinical behavior and age of occurrence, it is reasonable to hypothesize that distinct molecular pathways are involved in the pathogenesis and progression of different grades of DCIS. The role of extracellular matrix proteins has been studied in the development and progression of the breast carcinoma [6]. Cysteine-rich protein 61 (CYR61), connective tissue growth factor, and nephroblastoma overexpressed gene, together termed “CCN,” comprise a family of matricellular proteins that have multiple physiologic functions including development, tissue repair, cellular adhesion, migration, and proliferation [7,8]. There are 6 known members of the family named in order of discovery [9]. CCN family members have different roles in a variety of tissues. CCN proteins bind to cell surface integrins and heparin sulfate proteoglycans [10]. Depending on the subtype of CCN family member, type of receptors, and target tissue, they may promote or inhibit cancer formation. CCN1, CCN2, and CCN3 induce intracellular events including kinase activation and gene transcription [7]. CCN3 promotes prostate cancer bone metastasis [11], whereas CCN2 appears to suppress lung cancer metastasis [12]. The role of the CCN family in infiltrating breast carcinoma has been described [13-15]. CCN1 expression is associated with higher-grade breast cancer with lymph node metastasis, whereas CCN6 inhibits breast cancer progression [16]. Multiple signaling pathways including PI3K, AKT, and ILK, transcription factors such as β-catenin, and regulatory molecules such as p53 and cyclin D1 are activated or upregulated through CCN1 signaling [17,18]. Overexpression of CCN1 has been found to promote β-catenin nuclear translocation in some cancers such as glioma and non–small cell lung cancer [18,19]. This nuclear translocation in turn activates many growth-related genes including CCN1 in a positive feedback loop. Although CCN1 has been reported to play a role in infiltrating breast carcinoma, there is no detailed study about CCN1’s role in the pathologic precursor, DCIS. Immunostains for CCN1, p53, β-catenin, and cyclin D1 were used to evaluate their expression and possible interaction in different grades of breast ductal intraepithelial neoplasia.
2. Materials and methods After approval from the human investigation committee, DCIS cases (n = 61) were identified from the Yale Pathology Department’s archives diagnosed between 2007 and 2010. Most cases (n = 49) were from core needle biopsies. An additional 12 cases were obtained from excision specimens. Hematoxylin and eosin–stained sections, estrogen receptor (ER), and progesterone receptor (PgR) immunostained slides were reviewed. Diagnosis and the grade of disease were confirmed. DCIS grade was determined by using guidelines published by College of American Pathologists. These include nuclear size, pleomorphism, chromatin pattern, orientation, presence of nucleoli, and mitosis. DCIS archi-
O. Saglam et al. tectural patterns and presence or absence of necrosis were reviewed. Selected blocks from 61 cases were used for CCN1 immunostain, and 42 of the cases were also immunstained for cyclin D1, p53, and β-catenin. All CYR61 gene exons (encoding CCN1 protein) were sequenced in 30 samples. All DCIS cases studied by immunohistochemistry and polymerase chain reaction (PCR) were pure in situ carcinomas. They did not show any invasive components.
2.1. Immunohistochemistry Five-micron sections were cut from paraffin blocks and mounted on charged slides. They were deparaffinized in xylene and rehydrated through graded concentrations of ethanol. Slides for cyclin D1 immunostain were immersed in EDTA, whereas those for CCN1, p53, and β-catenin were immersed in citrate buffer and kept between 95°C to 101°C for 30 minutes. They were allowed to cool to room temperature, rinsed, and transferred to Tris-buffered saline. Tris-buffered saline was used as a wash between each subsequent step. Endogenous peroxidase was quenched with hydrogen peroxide. Primary antibodies were applied in the following dilutions: CCN1 (ab82040; Abcam, Cambridge, MA), prediluted antibody required no further dilution; p53 (MS187; Thermo, Waltham, MA), diluted to 1:100; cyclin D1 (CME432; Biocare Medical, Concord, CA), diluted to 1:200; and β-catenin (CM406; Biocare Medical), diluted to 1:100.
2.2. Evaluation of immunostains DCIS and the surrounding benign-appearing breast tissue were evaluated separately for each immunostain. The intensity (0-3 scale) and percentage of positive DCIS cells and benign epithelium were separately recorded. Initially, 2 pathologists (O. S. and K. H.) read each case independently. If there was a disagreement, the case was reviewed and consensus agreement obtained. Both H-score and Allred score were calculated on each case for DCIS and the surrounding benign-appearing breast tissue when present. Hscores were calculated by multiplying the percentage of cells demonstrating each intensity scored from 0 to 3+ and adding the results. They ranged from 0 to 300. Allred scores were calculated by adding together proportion scores obtained from the percentage of positively stained cells and intensity scores. They ranged from 0 to 8. Immunostains for CCN1 were evaluated for both cytoplasmic and nuclear staining. Only nuclear staining was considered positive for cyclin D1 and p53. β-catenin immunostain was read for nuclear versus cytoplasmic and membranous staining.
2.3. DNA extraction and mutation analysis Tumor DNA was extracted from the marked dissected formalin-fixed, paraffin-embedded tissues. The formalinfixed, paraffin-embedded tissues were deparaffinized in
CCN1 expression in in situ breast carcinoma Hemo-De (Scientific Safety Solvents, Keller, TX) and washed with 100% ethanol. Genomic DNA was extracted using QIAamp DNA Micro kit (Qiagen, Valencia, CA). PCR and Sanger sequencing analysis of CYR61 gene were performed. All 5 exons were sequenced using the following primers: Exon 1 forward: ACCAGCTTGTTGGCGTCTT; exon 1 reverse: GGGAATAGGTGGCAAAAGGA Exon 2 forward: CCGAGTCTCACGCGTATCTT; exon 2 reverse: TTTTAAAGGGGCCAAACCAC Exon 3 forward: TCCCCTTCTACCTTTCTCTTTTG; exon 3 reverse: GACACAAACAGGCTCAAGCA Exon 4 forward: TGTCGGTATGCCTCTGAGAA; exon 4 reverse: CTGCCAGGCAACAGTCTACA Exon 5 forward: CCCCTAACTTTCCTTCTCTCC; exon 5 reverse: GTCTAGGTGTGCCCTGGAAA The PCR conditions were as follows: (i) 95°C for 5 minutes; (ii) 94°C for 30 seconds, 60°C for 30 seconds (for exons 1, 2, 3, and 4), and −58°C for 30 seconds (for exon 5); 72°C for 1 minute (40 cycles); and (iii) 72°C for 10 minutes. The PCR products were identified on a 1% agarose gel. Sequencing was carried out on the ABI 3130 XL Genetic Analyzer (Applied Biosystems, Grand Island, NY). Mutation Surveyor software package Version 4.0.5 (SoftGenetics, State College, PA) was used for data analysis.
2.4. Statistical analysis All continuous data were assessed for normality of distribution using the Shapiro-Wilk test of normality. As they were not normally distributed, our data were summarized as a median, min-max, and interquartile range (IQR). The Spearman rank correlation coefficient was computed to help quantify the association between expression levels (measured by Allred and H-scores) of 4 different biomarkers or association of age and biomarker expression levels, from using all tumor samples or subsamples stratified by DCIS grade and/or ER status. The differences of expression levels of 4 biomarkers among the subsamples with different DCIS grade (or DCIS grade + ER status) were compared using the nonparametric Kruskal-Wallis test or Wilcoxon rank sum test as appropriate. The differences in the biomarker expression levels between normal breast tissue and DCIS were compared using the Wilcoxon signed rank test. All statistical analyses were 2-sided and were performed with the use of statistical software SAS version 9.2 (SAS, Cary, NC). A P b .05 was considered to indicate statistical significance, unless otherwise specified.
3. Results The patients’ ages ranged from 34 to 82 years (mean: 58 years). The mean ages for patients with DCIS grades 1 to 3 were 60, 61, and 55, respectively. There was no association between age and marker expression. Architectural types
3 include solid, cribriform, papillary, and micropapillary. Ten of 28 cases of DCIS grade 3 were ER−/PgR−, 13 cases ER+/ PgR+, 4 cases ER+/PgR−, and 1 case ER−/PgR+. Necrosis was identified in 27 of 28 cases. All 15 cases of DCIS grade 2 showed necrosis. There were 2 ER−/PgR−, 1 ER+/PgR−, and 12 ER+/PgR+ DCIS grade 2 cases. The DCIS grade 1 (n = 18) cases were all ER+/PgR+. All DCIS cases showed varying degrees of cytoplasmic staining for CCN1. The percentage of positively stained cells in DCIS 3 cases ranged from 5% to 95% and staining intensity from 1+ to 3+ (Fig.). H-scores ranged from 5 to 285 (median: 170) and Allred score from 3 to 8 (median: 7) (Table 1). Twelve cases of 28 DCIS grade 3 (48%) also showed nuclear staining ranging from 5% to 80% in quantity. The H-score for DCIS grade 2 cases varied from 20 to 285 (median: 160) and Allred scores between 4 and 8 (median: 7). Nuclear staining for immunostain CCN1 was observed only in 1 DCIS grade 2 case (20%, 2+). There was no nuclear CCN1 staining in any DCIS grade 1 cases. Only 4 of 15 DCIS grade 1 cases (27%) showed cytoplasmic staining reaching 3+ intensity. H-scores of DCIS grade 1 cases ranged from 1 to 270 (median: 60) and Allred scores from 2 to 7 (median: 5). There was a difference in CCN1 cytoplasmic H-scores (P = .043) but not Allred scores among DCIS grades. When cases were divided into ER+/DCIS 1, ER+/DCIS 2 and 3, and ER−/DCIS 2 and 3, the difference in cytoplasmic CCN1 staining was still significant (P = .037) (Table 2). The highest median was observed in ER+/DCIS grades 2 and 3 (H-score median: 160), followed by ER−/DCIS grades 2 and 3 (median: 100), and ER+/DCIS grade 1 (median: 60). Sequencing of the CYR61 gene encoding CCN1 revealed no pathogenic somatic mutations in any analyzed specimens. Twenty-one DCIS grade 3, 10 DCIS grade 2, and 11 DCIS grade 1 were available for β-catenin, p53, and cyclin D1 immunostaining. No nuclear β-catenin staining was seen in any case. Membranous and cytoplasmic β-catenin staining was observed in all 21 DCIS grade 3 cases, with H-scores ranging from 1 to 285 (median: 150) and Allred scores from 4 to 8 (median: 7). The median H-scores of βcatenin cytoplasmic/membranous expression in DCIS grades 2 and 1 were 150 and 60, respectively. There was a significant difference in β-catenin expression among different DCIS grades by both H-score (P = .047) and Allred score (P = .026). This difference remained when the cases were divided into ER+/DCIS grade 1, ER−/DCIS 2 and 3, and ER+/DCIS 2 and 3 groups for H-score (P = .045) and Allred score (P = .026). Cyclin D1 was positive in 19 of 21 DCIS grade 3 cases. The H-scores ranged from 0 to 270 (median: 120) and Allred score was between 0 and 8 (median: 7). In DCIS cases with focal cyclin D1 staining, the positive cells were located at the periphery of the duct adjacent to the residual myoepithelial layer. DCIS grades 2 and 1 had the median H-scores of 180 and 60, respectively. The H-score and Allred score for cyclin D1 showed positive correlations with that for CCN1 cytoplasmic expression in all DCIS samples (P = .038 and
4
O. Saglam et al.
Fig. Photomicrograph shows examples of marker expression. A, CCN1 immunostain: the cytoplasmic marker expression is diffuse and strong in DCIS200 Inset: CCN1 nuclear staining is shown. B, Cyclin D1. C, β-catenin. D, p53. ×200 (A and B), ×400 (inset), ×100 (C and D).
Table 1
Summary of marker expression by DCIS grade
Marker
H-score CCN-cytoplasmic Cyclin D1 p53 β-catenin Allred CCN-cytoplasmic Cyclin D1 p53 β-catenin
DCIS grade 1
DCIS grade 2
DCIS grade 3
n
Median (IQR)
n
Median (IQR)
n
Median (IQR)
18 11 11 11
60.0 (10.0-160.0) 60.0 (40.0-180.0) 30.0 (5.0-80.0) 60.0 (20.0-120.0)
15 10 10 10
160.0 (120.0-240.0) 180.0 (60.0-240.0) 50.0 (20.0-80.0) 150.0 (40.0-210.0)
28 21 21 21
170.0 (95.0-195.0) 120.0 (60.0-210.0) 90.0 (20.0-140.0) 150.0 (100.0-180.0)
18 11 11 11
5.0 (3.0-7.0) 6.0 (5.0 - 7.0) 4.0 (3.0-6.0) 5.0 (4.0-6.0)
15 10 10 10
7.0 (6.0-8.0) 7.5 (6.0-8.0) 5.0 (4.0-6.0) 7.0 (5.0-8.0)
28 21 21 21
NOTE. Data are summarized by median (interquartile range). The Kruskal-Wallis test was used for statistical analysis.
7.0 (6.0-7.5) 7.0 (6.0-8.0) 6.0 (4.0-7.0) 7.0 (6.0-7.0)
P
.043 .302 .214 .047 .089 .249 .180 .026
CCN1 expression in in situ breast carcinoma Table 2
5
Summary of marker expression by DCIS grade and ER status
Marker
H-score CCN-cytoplasmic Cyclin D1 p53 β-catenin Allred CCN-cytoplasmic Cyclin D1 p53 β-catenin
DCIS grade 1, ER+
DCIS grade (2 + 3), ER−
DCIS grade (2 + 3), ER+
n
Median (IQR)
n
Median (IQR)
n
Median (IQR)
18 11 11 11
60.0 60.0 30.0 60.0
(15.0-160.0) (40.0-180.0) (5.0-80.0) (20.0-120.0)
13 8 8 8
100.0 (90.0-180.0) 120.0 (20.0-210.0) 105.0 (30.5-160.0) 170.0 (80.0-195.0)
30 23 23 23
160.0 (140.0-210.0) 150.0 (60.0-240.0) 60.0 (20.0-140.0) 140.0 (40.0-210.0)
18 11 11 11
5.0 6.0 4.0 5.0
(3.0-7.0) (5.0-7.0) (3.0-6.0) (4.0-6.0)
13 8 8 8
6.0 (6.0-7.0) 6.5 (2.5-8.0) 6.0 (3.5-7.0) 7.0 (5.5-7.5)
30 23 23 23
7.0 (7.0-8.0) 7.0 (6.0-8.0) 5.0 (4.0-7.0) 7.0 (5.0-8.0)
P
.037 .259 .243 .045 .060 .220 .252 .026
NOTE. Data are summarized by median (interquartile range). The Kruskal-Wallis test was used for statistical analysis. There were no DCIS grade 1 and ERnegative patients.
P = .006) (Supplementary Tables 1 and 2). This association was lost when the cases were divided into ER−/+ groups. Immunostain for p53 was positive in 20 of 28 DCIS grade 3 cases with a median H-score of 90. For intermediate- and low-grade DCIS, median H-scores were 50 and 30, respectively. There was no statistical difference in p53 expression among the grades of DCIS. There was no nuclear expression of CCN1 in the benign breast tissue surrounding any grade of DCIS. The H-score of cytoplasmic CCN1 varied from 1 to 160 in all grades. The median H-scores of cytoplasmic CCN1, cyclin D1, p53, and β-catenin were 10, 10, 20, and 35, respectively. Wilcoxon signed rank test showed significant differences in expression of all markers between DCIS and the adjacent benign breast tissue (P b .001).
3.1. Clinical follow-up and additional findings on excision specimens Two patients did not have any further treatment after initial diagnosis in our institution. On excision specimens, most patients had only DCIS with matching grade to the core biopsy diagnosis. There were 7 invasive carcinomas (including 2 with microinvasion) on the excision materials after DCIS diagnosis on core needle biopsies. Two patients with DCIS grade 3 had microinvasive duct carcinoma. Their H-scores for nuclear CCN1 were 240 and 150. Cytoplasmic H-scores were 180 and 285. The remaining cases consisted of 3 T1a, 1 T1b, and 1 T1c tumors. Two of these were associated with DCIS grade 3, 1 with DCIS grade 2, and 2 with DCIS 1. Both tumors with DCIS grade 1 were welldifferentiated invasive carcinomas. The cytoplasmic Hscores were 210 and 10 on the core needle biopsies. Clinical follow-up ranged from 11 to 68 months (median: 36 months). One patient with DCIS 1 developed contralateral DCIS grade 1 with microinvasive carcinoma 2 years after her initial diagnosis. The cytoplasmic H-score for CCN1 was 210 for this case. Another patient with DCIS 1 developed recurrent
disease within 3 years. The H-score for cytoplasmic CCN1 was 160 in this case. A patient with DCIS 3 had metastatic lung cancer to the axilla. There were no other documented recurrent or contralateral diseases in our records.
4. Discussion Extracellular matrix components play a significant role in regulating epithelial cell growth and survival in a variety of settings [20]. The interaction between DCIS and its surrounding stroma produces changes in the microenvironment before tumor cells actually invading into the stroma [21]. CCN1 as a matricellular protein is up-regulated in wound healing and induces several genes such as matrix metalloproteinase (MMP)-1, MMP-3, vascular endothelial growth factors A and C, interleukins, and others [22]. MMP1 activates the breast cancer invasion–associated molecule protease activator receptor 1. Protease activator receptor 1 can induce CCN1 expression in breast cancer cells, which promotes invasion by further augmenting MMP-1 expression in adjacent stromal fibroblasts [23]. Leu et al [24] developed an anti-CCN1 antibody that suppresses Rac1mediated tumor cell migration and invasion. We observed a positive correlation between CCN1 expression and grades of DCIS independent of ER status. Showing higher expression of CCN1 in DCIS 3 is compatible with higher CCN1 expression in the advanced-stage infiltrating duct carcinoma as previously reported in the literature [25]. Recent genetic expression profiling demonstrated similarities between low-grade DCIS and its low-grade invasive carcinoma component and between high-grade DCIS and high-grade infiltrating duct carcinoma [26]. One of 2 DCIS grade 1 cases with associated invasive carcinoma on the excision specimen has higher cytoplasmic expression of CCN1 compared with median CCN1 expression in this group. On the other hand, both recurrence and contralateral disease were observed in patients with DCIS 1
6 expressing high cytoplasmic levels CCN1. These results are limited observations. Whether high expression of CCN1 in low-grade DCIS is a risk factor for recurrence or progression to invasive disease would require long-term follow-up in a larger cohort. CCN1 expression has been correlated with estrogen levels. Estradiol was shown to increase CCN1 expression in the uterus of oophorectomized mice [27]. Estrogen induces CCN1 in breast cancer cell lines in a dose-dependent manner and is blocked by neutralizing antibodies [15]. In our cohort, ER+ intermediate- and high-grade DCIS have higher median CCN1 scores compared with ER− equivalents. This finding is compatible with previous studies. On the other hand, Tsai et al [28] have demonstrated that CCN1 is sufficient to demonstrate estrogen independence and resistance to antiestrogens in vivo. ER+ low-grade DCIS cases had the lowest expression of CCN1 among all 3 groups. These results indicate that the grade of disease also has an impact in the cyctoplasmic expression of CCN1 regardless of hormone receptor status. Nuclear expression of CCN1 was observed in 12 DCIS grade 3 and 1 DCIS grade 2 cases. Truncated forms of CCN1 are known to translocate to the nucleus, where they are involved in transcriptional regulation [29]. For example, CCN1 was found to localize both in cytoplasm and nucleus of cultured bladder smooth muscle cells [30]. Nuclear expression of CCN1 in invasive breast carcinoma has not been previously reported. The significance of this finding in highgrade DCIS is not certain and requires further investigation. CCN1 and β-catenin interact closely in the Wnt signaling pathway. When Wnt binds to its receptor, β-catenin is released from the cell membrane and translocates to the nucleus, where it associates with ternary complex factor. Transcriptional targets of the Wnt pathway include the cellular oncogenes cyclin D1 and c-myc [31]. CCN1 reportedly leads to β-catenin translocation through degradation of E-cadherin [8]. CCN1 induces β-catenin translocation in squamous cells carcinoma of esophagus [32] and is involved in the progression of hepatocellular carcinoma via Wnt/β-catenin signaling [33]. There was no nuclear β-catenin staining in our DCIS cases. Membranous/ cytoplasmic β-catenin expression varied significantly by both H-score and Allred scores in different DCIS grades regardless of ER status. The biological significance of membranous staining in our DCIS cases is not clear. Cell cycle progression requires coordinated interactions between cyclins and cyclin-dependent kinases (CDK). The binding of cyclin D1 to CDK4 and CDK6 is a rate-limiting step during cell cycle progression [34]. Cylin D1 is overexpressed in a number of cancers including the breast, liver, lung, and brain [35-37]. Assoian and Klein [38] showed cyclin D1 induction requires coordinated signaling from the extracellular matrix and soluble growth factors. Supporting this finding, we observed a positive correlation between CCN1 and cyclin D1 expression in all DCIS grades. In summary, our results indicate that the role of CCN1 in the breast cancer development begins in intraepithelial carcinoma
O. Saglam et al. stage. Although no mutation in the exons of CCN1 gene was detected, the CCN1 protein expression has a positive correlation with cyclin D1 and possibly with β-catenin. In addition, cytoplasmic expression of CCN1 was associated with DCIS grade independent of ER status. We found no definite interaction between CCN1 and p53 in intraductal carcinoma. Further observational and functional studies are warranted to determine the exact role of CCN1 in pathogenesis of DCIS.
Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.humpath.2014.02.007.
References [1] Page DL, Dupont WD. Histopathologic risk factors for breast cancer in women with benign breast disease. Semin Surg Oncol 1988;4:213-7. [2] Betsill Jr WL, Rosen PP, Lieberman PH, Robbins GF. Intraductal carcinoma. Long-term follow-up after treatment by biopsy alone. JAMA 1978;239:1863-7. [3] Punglia RS, Schnitt SJ, Weeks JC. Treatment of ductal carcinoma in situ after excision: would a prophylactic paradigm be more appropriate? J Natl Cancer Inst 2013;105:1527-33. [4] Sanders ME, Schuyler PA, Dupont WD, Page DL. The natural history of low-grade ductal carcinoma in situ of the breast in women treated by biopsy only revealed over 30 years of long-term follow-up. Cancer 2005;103:2481-4. [5] Alvarado R, Lari SA, Roses RE, et al. Biology, treatment, and outcome in very young and older women with DCIS. Ann Surg Oncol 2012;19:3777-84. [6] Vargas AC, Reed AE, Waddell N, et al. Gene expression profiling of tumour epithelial and stromal compartments during breast cancer progression. Breast Cancer Res Treat 2012;135:153-65. [7] Chen CC, Lau LF. Functions and mechanisms of action of CCN matricellular proteins. Int J Biochem Cell Biol 2009;41:771-83. [8] Chong HC, Tan CK, Huang RL, Tan NS. Matricellular proteins: a sticky affair with cancers. J Oncol 2012;2012:351089. [9] Brigstock DR. The CCN family: a new stimulus package. J Endocrinol 2003;178:169-75. [10] Kireeva ML, Mo FE, Yang GP, Lau LF. Cyr61, a product of a growth factor-inducible immediate-early gene, promotes cell proliferation, migration, and adhesion. Mol Cell Biol 1996;16:1326-34. [11] Chen PC, Cheng HC, Tang CH. CCN3 promotes prostate cancer bone metastasis by modulating the tumor-bone microenvironment through RANKL-dependent pathway. Carcinogenesis 2013;34:1669-79. [12] Chang CC, Yang MH, Lin BR, et al. CCN2 inhibits lung cancer metastasis through promoting DAPK-dependent anoikis and inducing EGFR degradation. Cell Death Differ 2013;20:443-55. [13] Lin J, Huo R, Wang L, et al. A novel anti-Cyr61 antibody inhibits breast cancer growth and metastasis in vivo. Cancer Immunol Immunother 2012;61:677-87. [14] Tsai MS, Hornby AE, Lakins J, Lupu R. Expression and function of CYR61, an angiogenic factor, in breast cancer cell lines and tumor biopsies. Cancer Res 2000;60:5603-7. [15] Sampath D, Winneker RC, Zhang Z. Cyr61, a member of the CCN family, is required for MCF-7 cell proliferation: regulation by 17betaestradiol and overexpression in human breast cancer. Endocrinology 2001;142:2540-8.
CCN1 expression in in situ breast carcinoma [16] Jiang WG, Watkins G, Fodstad O, Douglas-Jones A, Mokbel K, Mansel RE. Differential expression of the CCN family members Cyr61, CTGF and Nov in human breast cancer. Endocr Relat Cancer 2004;11:781-91. [17] Menendez JA, Vellon L, Mehmi I, Teng PK, Griggs DW, Lupu R. A novel CYR61-triggered “CYR61-alphavbeta3 integrin loop” regulates breast cancer cell survival and chemosensitivity through activation of ERK1/ERK2 MAPK signaling pathway. Oncogene 2005;24:761-79. [18] Tong X, O'Kelly J, Xie D, et al. Cyr61 suppresses the growth of nonsmall-cell lung cancer cells via the beta-catenin-c-myc-p53 pathway. Oncogene 2004;23:4847-55. [19] Xie D, Yin D, Tong X, et al. Cyr61 is overexpressed in gliomas and involved in integrin-linked kinase-mediated Akt and beta-cateninTCF/Lef signaling pathways. Cancer Res 2004;64:1987-96. [20] Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144:646-74. [21] Ma XJ, Dahiya S, Richardson E, Erlander M, Sgroi DC. Gene expression profiling of the tumor microenvironment during breast cancer progression. Breast Cancer Res 2009;11:R7. [22] Chen CC, Chen N, Lau LF. The angiogenic factors Cyr61 and connective tissue growth factor induce adhesive signaling in primary human skin fibroblasts. J Biol Chem 2001;276:10443-52. [23] Nguyen N, Kuliopulos A, Graham RA, Covic L. Tumor-derived Cyr61(CCN1) promotes stromal matrix metalloproteinase-1 production and protease-activated receptor 1-dependent migration of breast cancer cells. Cancer Res 2006;66:2658-65. [24] Jim Leu SJ, Sung JS, Huang ML, Chen MY, Tsai TW. A novel antiCCN1 monoclonal antibody suppresses Rac-dependent cytoskeletal reorganization and migratory activities in breast cancer cells. Biochem Biophys Res Commun 2013;434:885-91. [25] Xie D, Nakachi K, Wang H, Elashoff R, Koeffler HP. Elevated levels of connective tissue growth factor, WISP-1, and CYR61 in primary breast cancers associated with more advanced features. Cancer Res 2001;61:8917-23.
7 [26] Ma XJ, Salunga R, Tuggle JT, et al. Gene expression profiles of human breast cancer progression. Proc Natl Acad Sci U S A 2003;100:5974-9. [27] Rivera-Gonzalez R, Petersen DN, Tkalcevic G, Thompson DD, Brown TA. Estrogen-induced genes in the uterus of ovariectomized rats and their regulation by droloxifene and tamoxifen. J Steroid Biochem Mol Biol 1998;64:13-24. [28] Tsai MS, Bogart DF, Castaneda JM, Li P, Lupu R. Cyr61 promotes breast tumorigenesis and cancer progression. Oncogene 2002;21:8178-85. [29] Yeger H, Perbal B. The CCN family of genes: a perspective on CCN biology and therapeutic potential. J Cell Commun Signal 2007;1:159-64. [30] Tamura I, Rosenbloom J, Macarak E, Chaqour B. Regulation of Cyr61 gene expression by mechanical stretch through multiple signaling pathways. Am J Physiol Cell Physiol 2001;281:C1524-32. [31] Smalley MJ, Dale TC. Wnt signalling in mammalian development and cancer. Cancer Metastasis Rev 1999;18:215-30. [32] Chai J, Modak C, Ouyang Y, Wu SY, Jamal MM. CCN1 induces betacatenin translocation in esophageal squamous cell carcinoma through integrin alpha11. ISRN Gastroenterol 2012;2012:207235. [33] Li ZQ, Ding W, Sun SJ, et al. Cyr61/CCN1 is regulated by Wnt/betacatenin signaling and plays an important role in the progression of hepatocellular carcinoma. PLoS One 2012;7:e35754. [34] Fu M, Wang C, Li Z, Sakamaki T, Pestell RG. Minireview: cyclin D1: normal and abnormal functions. Endocrinology 2004;145:5439-47. [35] Gillett C, Smith P, Gregory W, et al. Cyclin D1 and prognosis in human breast cancer. Int J Cancer 1996;69:92-9. [36] Hall M, Peters G. Genetic alterations of cyclins, cyclin-dependent kinases, and Cdk inhibitors in human cancer. Adv Cancer Res 1996;68:67-108. [37] Molenaar JJ, Ebus ME, Koster J, et al. Cyclin D1 and CDK4 activity contribute to the undifferentiated phenotype in neuroblastoma. Cancer Res 2008;68:2599-609. [38] Assoian RK, Klein EA. Growth control by intracellular tension and extracellular stiffness. Trends Cell Biol 2008;18:347-52.