Japanese Journal of Clinical Oncology Advance Access originally published online on March 6, 2006
Japanese Journal of Clinical Oncology 2006 36(3):142-149; doi:10.1093/jjco/hyi246
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 Foundation for Promotion of Cancer Research
Cyclin E Expression in Operable Breast Cancer Quantified Using Real-Time RT–PCR: A Comparative Study with Immunostaining
1 Departments of Chemotherapy, 2 Pathology, 3 Molecular Cancerogenesis and 4 Haemostatic Disorders, Medical University of Lodz, Lodz, Poland
For reprints and all correspondence: Piotr Potemski, Department of Chemotherapy, Medical University of Lodz, 4 Paderewski Street, 93-509 Lodz, Poland. E-mail: piotrpo{at}mp.pl
Received November 7, 2005; accepted December 20, 2005
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Objective: The main purpose of this retrospective study was to compare cyclin E expression levels in operable breast cancer patients determined using real-time RT–PCR and immunostaining. The prognostic relevance of cyclin E was also investigated.
Methods: Specimens of invasive ductal breast cancer tissues obtained from 124 women during radical mastectomy were analyzed.
Results: Of the tumor samples, 40.3 and 59.7% showed high expression of cyclin E in RT–PCR and immunostaining, respectively. The overall agreement probability was 0.032 according to Scott's 
statistic. With a median follow-up of 55.5 months, cyclin E expression assessed using immunostaining was an independent negative prognostic factor in the node positive group (hazard ratio 3.1; 95% CI 1.0–9.2; P = 0.045). Cyclin E expression correlated with absence of steroid receptors and younger age. RT–PCR results did not predict survival in any group of patients.
Conclusions: Disagreement between real-time RT–PCR and immunostaining was demonstrated. Immunostaining seems to be the more reliable method for assessing cyclin E in breast cancer cells.
Key Words: breast neoplasms • cyclin E • immunochemistry • prognosis • reverse transcription PCR
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Adjuvant systemic treatment of breast cancer is based upon well-known classical prognostic factors such as age, tumor size, tumor grade, presence of lymph node metastases, steroid receptor status and HER2 status. Some patients, in spite of the presence of high-risk factors, have an excellent prognosis and are cured with regional treatment alone. On the other hand, many patients in the low- or intermediate-risk groups finally die of breast cancer. Thus, there is a common belief that classical prognostic system needs to be improved because many patients receive toxic therapy unnecessarily (1). During the past few years many studies have been conducted in order to identify molecular prognostic factors, especially among the cell cycle molecules.
Disturbances in the activity of cell cycle regulatory proteins play a key role in cancer (2,3). Progression through the cell cycle is promoted by protein complexes composed of cyclins and cyclin-dependent kinases (cdks). Cyclin D and cyclin E form active complexes with cdk2 and enable progression through the G1 phase of the cell cycle and control entry into the S phase by phosphorylation of RB (retinoblastoma) protein and subsequent activation of transcription factor E2F; cdks inhibitors exert negative control (3). In malignant cells, there is an imbalance between cyclins, cdks and cdks inhibitors, which leads to the uncontrolled cells divisions.
Cyclin E overexpression has been observed in breast cancer, gastrointestinal and hematological malignancies, lung cancer, genitourinary tract cancers, sarcomas and skin cancers (4). High levels of cyclin E in breast cancer cells have been observed more often in patients with higher disease stage, poor histological differentiation of tumor, higher proliferative status and lack of steroid receptors (5,6).
In previous studies, cyclin E expression has been assessed using protein detection methods such as immunostaining, western blot and tissue microarray analysis or using gene expression detection methods such as real-time RT–PCR (reverse transcription polymerase chain reaction) (7–10).
Investigations made by others into the prognostic value of cyclin E in breast cancer have provided conflicting data (11,12). In a retrospective study of 395 breast cancer patients, Keyomarsi et al. (8) found that a high level of cyclin E, as detected using western blot, correlated with poor survival. Cyclin E was the strongest independent prognostic factor, more important than the presence of axillary lymph node metastases. Similar observations were made by Porter et al. (12), in a group of node negative patients, and by Han et al. (13) and Lindhal et al. (14).
However, some authors did not confirm the prognostic value of cyclin E overexpression in breast cancer. Although cyclin E expression was sometimes associated with poorer outcome in univariate analyses, it was not a significant negative prognostic factor in multivariate analyses (15–17). Bukholm et al. (18), using immunostaining analysis in a group of 170 patients with invasive breast cancer, did not find any prognostic value of cyclin E.
The aim of the present study was retrospectively to compare two methods of assessing cyclin E expression: real-time RT–PCR and immunostaining. We also investigated whether the cyclin E level had an impact on survival in patients with operable breast cancer.
|
PATIENTS AND METHODS |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
TUMOR SPECIMENS AND STUDY PATIENTS
Primary tumor specimens were obtained consecutively from 128 women with operable breast cancer at the time of routine surgery at the Oncology Department of Copernicus Memorial Hospital in Lodz, Poland, between 1998 and 2001. All tumors were invasive ductal carcinomas. The surgical procedure in all cases was radical mastectomy with axillary lymph node dissection. For further mRNA analysis, fresh tumor specimens were frozen at –80°C immediately after excision. Serial sections of the tumor were obtained from archived paraffin-embedded tissue blocks. In all cases primary pathological diagnosis was confirmed using hematoxylin and eosin (H&E) staining. Subsequent slides were stained for cyclin E, steroid receptors, HER2 and Ki-67. In four cases cyclin E staining was not possible for technical reasons and these patients were excluded from the study. Thus, the study population consisted of 124 patients. All operative and pathological reports were reviewed to confirm the disease stage. The follow-up period was defined as the time from surgery to the last observation for censored cases or death for complete observations.
REAL-TIME RT–PCR ANALYSIS
Tumor samples were stored at –80°C until RNA extraction. RNA was extracted using TRIzol® reagent (Invitrogen Corporation, USA). Synthesis of cDNA was performed from 10 µg of total RNA at a total volume of 70 µL using ImProm-IITM reverse transcriptase (Promega Corporation, USA). Next, cDNA samples were diluted with sterile deionized water to a total volume of 140 µL. Volumes of 2 µL (corresponding to 0.14 µg of total RNA) were used for PCR. Real-time RT–PCR was performed using the ABI PRISM® 7000 Sequence Detection System (Applied Biosystems, USA). Sequences of primers used and annealing and detection temperatures are presented in Table 1. Primers for cyclin E were evaluated and in silico tested using the Amplify algorithm (19). All reactions were performed in triplicate. PCR products were detected with SYBR® Green I using the qPCR Core kit for SYBR® Green I (Eurogentec, Belgium). Expression levels of cyclin E were normalized using the ß2-microglobulin gene primer (20). Amplification plots for ß2-microglobulin and cyclin E genes are presented in Figure 1. Relative gene expression was calculated using the 2–
CT method with pooled cDNA from all tumor samples as a reference (21). Thus, expression level 1.0 represents an average expression in the samples examined. Assuming a cut-off level of 1.0, all tumors were divided into two groups: those with low (relative gene expression 
1.0) and those with high (relative gene expression >1.0) cyclin E expression.
|
|
IMMUNOSTAINING AND SCORING
Paraffin-embedded sections were processed routinely. Slides for immunostaining for ER (estrogen receptor), PR (progesterone receptor) and Ki-67 (all from Dako) were pre-treated with citrate buffer in a microwave oven. HER2 expression was examined using the commercially available Herceptest kit from Dako. Antibody dilutions were as follows: ER, 1:35, PR, 1:75, Ki-67, 1:200. Cyclin E expression was analyzed using anti-cyclin E monoclonal antibody (Novocastra). Endogenous peroxidase activity was inhibited using 0.5% hydrogen peroxide containing methanol, and normal horse serum was used to block the non-specific antigen binding sites. The antibody was diluted to 1:40. All subsequent procedures were performed according to standard protocols using the EnVison kit (Dako).
Cyclin E and Ki-67 labeling indexes were defined as the percentage of tumor cells displaying nuclear immunoreactivity and were calculated by counting the number of nuclear stained cells present in 1000 tumor cells. A single representative section from each sample was surveyed microscopically at x100 for at least two areas of the highest staining intensity of positive cells. Cell counts were performed at x400 in at least five fields in these areas. For cyclin E, depending on the percentage of the cells showing a nuclear staining pattern tumor, samples were judged as negative (<2%) or positive (
2%). ER and PR nuclear staining scorings were made using the method described by McCarty et al. (22). Tumors were considered positive for ER or PR if the Histo-score was above 100. HER2 staining was scaled according to the Herceptest kit manufacturer's instructions, and a score of 3+ denoted HER2 positive tumors.
STATISTICAL ANALYSIS
To analyze agreement between real-time RT–PCR and immunostaining, Scott's test and the matched-pairs Liddell exact test were used. Pearson's 
2-test or Fisher's exact test were employed to test for compatibility between dichotomized values of cyclin E expression (negative and positive) and values of other histological and clinical parameters. Two inference tests, Student's t-test and the Mann–Whitney U-test, were used to evaluate differences in age and Ki-67 expression, respectively, between the cyclin E negative and positive patients. Comparison of gene expression level between cyclin E negative and positive tumors, as measured by immunostaining, was performed using the Mann–Whitney U-test. Cancer-specific survival was calculated from the date of primary surgery to the date of death or last follow-up, and data for patients who died from causes other than breast cancer were censored at the time of death. Cancer-specific survival was estimated using the Kaplan–Meier method. Differences in survival distributions were evaluated using a log-rank test. A multivariate analysis of cancer-specific survival was performed using the Cox proportional hazard regression model with an inclusion criterion of P 0.050 and an exclusion criterion of P > 0.050. All results were considered statistically significant when P was < 0.050.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
PATIENT CHARACTERISTICS
The median follow-up period for 90 censored (surviving) patients was 61 months (range 9–68). For the whole group it was 55.5 months (range 1–68). A total of 37 deaths were observed, 14 in a node negative and 23 in a node positive group. Three of them were due to other causes (2 and 1, respectively). Patient characteristics are presented in Table 2
|
CANCER-SPECIFIC SURVIVAL
Real-Time RT–PCR
No significant differences in survival were observed between low- and high-cyclin E groups in all cases (5-year survival rate: 66.7 versus 74.0%, P = 0.374; Figure 2), in the node negative patients (76.8 versus 82.0%, P = 0.506) and in the node positive patients (55.4 versus 66.3%, P = 0.502).
|
To rule out a possible influence of the assumed cut-off level of cyclin E expression, two groups of 25 patients, with the lowest and the highest levels of cyclin E mRNA, were also compared. In the first group the relative gene expression level ranged from 0.00 to 0.10 (median 0.04) and in the second from 5.18 to 2150.61 (median 20.09). Again, no difference in survival was observed (62.0 versus 65.4%, P = 0.923).
Immunostaining
In the whole group an obvious, although not statistically significant, tendency toward better outcome in patients with tumors negative for cyclin E was observed (5-year survival rate 80.8 versus 62.6%; P = 0.072; Figure 3). No significant differences in survival were observed in the node negative (86.5 versus 73.9%; P = 0.311) and in the node positive group (75.7 versus 48.5%; P = 0.101).
|
COMPARISON OF REAL-TIME RT–PCR AND IMMUNOSTAINING
In Table 3 two methods of assessing cyclin E expression are compared. Scott's
overall agreement probability was 0.032 (95% CI 0.142–0.205). The Liddell exact test for matched pairs also revealed disagreement between methods (P = 0.003).
|
Relative gene expression did not differ between cyclin E negative and positive groups, as assessed using immunostaining (P = 0.600).
No significant differences in survival were observed between groups listed in Table 3.
RELATIONS BETWEEN CYCLIN E AND OTHER FACTORS
As shown in Table 4 no correlations between expression level of cyclin E, assessed using real-time RT–PCR, and any other prognostic factors were found. Tumors positive for cyclin E, measured using immunostaining, usually lacked steroid receptors. They also were more often observed in younger patients and had the tendency to show higher Ki-67 expression.
|
MULTIVARIATE ANALYSIS
Proportional-hazards modeling of cancer-specific survival was performed in all patients and in both the node positive and node negative groups. In these analyses, in addition to cyclin E expression measured using real-time RT–PCR and immunostaining, factors such as stage, nodal involvement, tumor grade, steroid receptor status and HER2 status were included. The results for all patients and for the patients in the node positive group are listed in Table 5. In all patients cyclin E expression measured using immunostaining almost reached statistical significance (2.2; 0.9–5.3; 0.074). In the node negative group no significant independent prognostic factors were found.
|
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This is, to our knowledge, the first study aimed directly at comparing the protein detection method and the mRNA detection method for assessing cyclin E expression in breast cancer. We demonstrated that real-time RT–PCR disagreed with immunostaining. Our observation remains to some extent in contrast with data provided by others with respect to other cancers. Tsuda et al. (23) found significant correlation of cyclin E mRNA amount and protein level in a group of 83 ovarian cancer patients. This correlation was also confirmed in a very small group of lung cancer patients (24).
There are some controversies regarding the prognostic value of cyclin E in breast cancer (8,11,12,18). The majority of authors used protein detection methods such as immunostaining, western blotting and tissue microarrays. Keyomarsi et al. (8) compared data obtained from an immunostaining analysis and a western blot analysis. They found significant discrepancies between the two methods. The survival correlated better with data obtained using the western blot technique. It can be easily explained because an immunochemistry technique for the assessment of cyclin E expression is not standardized yet and is by its nature subjective, whereas western blotting is a very reliable method for detecting proteins.
In our study, fresh frozen tumor samples were used for extracting mRNA and we were unable to extract proteins simultaneously. So, immunostaining was chosen for the detection of proteins.
We showed that cyclin E expression measured using immunostaining but not using real-time RT–PCR was an independent negative prognostic factor in the node positive group. This tendency was seen in all patients. Estimated cancer-specific survival did not differ significantly according to the cyclin E status. Nevertheless, a tendency was clearly seen toward worse survival of patients positive for cyclin E assessed using immunostaining. This might be explained by the relatively small number of patients involved in the study and the short time of follow-up. There was a higher percentage of complete observations in node positive patients than in the whole group, as node positive patients had worse prognosis overall. We believe that these findings are well supported by data obtained from multivariate analyses.
Real-time RT–PCR is considered to be one of the most precise methods of assessing the amount of mRNA. Span et al. (10) used an RT–PCR technique in a group of 277 resectable breast cancer patients and did not find any relationship of cyclin E expression to survival. We confirmed their observation, though on a group with fewer patients.
We also examined the relationship between the level of cyclin E and pathological and clinical factors. We confirmed observations made by others that high expression of cyclin E found in immunostaining was correlated with absence of steroid receptors and younger age (11,15). The relationship with higher proliferative index almost reached the statistical significance observed by Scott et al. (6). No relationship with disease stage, higher tumor grade or HER2 status was observed. This can be explained by the relatively small number of patients in our study.
These observations suggest the possibility that mRNA detection methods are of little value in assessing the prognostic significance of cyclin E. There are some data supporting this thesis.
The mRNA was used for RT–PCR could have originated from cancer cells as well as from tumor stromal cells (i.e. blood vessels or fibroblasts). In contrast, immunostaining results are specific for tumor cells. Furthermore, the amount of mRNA does not always reflect the protein level. Hyperactive, low-molecular-weight forms of cyclin E (molecular mass ranging from 34 to 49 kD) have been described in tumor cells (25). Harwell et al. (26) found that these isoforms were generated by the post-translational action of nuclear proteases present in cancer cells. Low-molecular-weight isoforms of cyclin E associated with poor prognosis were observed in breast cancer (27). These forms induced genomic instability and resistance to cdks inhibitors and predicted failure of anti-estrogen treatment.
There are some suggestions that splicing may play a substantial role in generating such variants of cyclin E (28). Deregulation of the splicing of cyclin E in breast cancer cells may contribute to creating biologically important low-molecular-weight isoforms of this protein. This may, at least in part, explain our findings. The cyclin E primer we used could possibly detect only full-length mRNA and not splice variants. Thus, expression of the cyclin E gene appeared to have no impact on survival.
Recently, in breast cancer cells mutations have been described in the hCDC4 gene coding the protein responsible for proteolysis of cyclin E (29). These mutations led to disturbances in the regulation of the cell cycle by cyclin E but were not always related to cyclin E overexpression.
In conclusion, the present study revealed the disagreement between real-time RT–PCR and immunostaining in assessing cyclin E expression in breast cancer. High expression of cyclin E measured by immunostaining was an independent negative prognostic factor in the node positive group. We also found that high expression of cyclin E determined using immunostaining correlated with lack of steroid receptors and younger age. Results obtained from RT–PCR did not predict survival. Thus, immunostaining seems to be the more reliable method for assessing cyclin E status.
|
Acknowledgments |
|---|
This study was supported by a grant from the Medical University of Lodz, Poland (No. 502-11-285), and in part by grants from the State Committee of Scientific Research (KBN, Warsaw, Poland; No. 2 P05E 099/28 and No. 3 P05A 113/23).
|
References |
|---|
|
TOPAbstractINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONReferences |
|---|
1 Senn HJ, Thurlimann B, Goldhirsch A, Wood WC, Gelber RD, Coates AS. Comments on the St. Gallen Consensus 2003 on the Primary Therapy of Early Breast Cancer. Breast 2003;12:569–82.[CrossRef][Medline]
2 Sherr CJ. Cancer cell cycle. Science 1996;274:1672–7.
3 Sherr CJ. The Pezcoller Lecture: cancer cell cycles revisited. Cancer Res 2000;60:3689–95.
4 Donnellan R, Chetty R. Cyclin E in human cancers. FASEB J 1999; 13:773–80.
5 Nielsen NH, Arnerlov C, Emdin SO, Landberg G. Cyclin E overexpression, a negative prognostic factor in breast cancer with strong correlation to oestrogen receptor status. Br J Cancer 1996;74:874–80.[ISI][Medline]
6 Scott KA, Walker RA. Lack of cyclin E immunoreactivity in non-malignant breast and association with proliferation in breast cancer. Br J Cancer 1997;76:1288–92.[ISI][Medline]
7 Foulkes WD, Brunet JS, Stefansson IM, Straume O, Chappuis PO, Begin LR, et al. The prognostic implication of the basal-like (cyclin E high/p27 low/p53+/glomeruloid-microvascular-proliferation+) phenotype of BRCA1-related breast cancer. Cancer Res 2004;64:830–5.
8 Keyomarsi K, Tucker SL, Buchholz TA, Callister M, Ding Y, Hortobagyi GN, et al. Cyclin E and survival in patients with breast cancer. N Engl J Med 2002;347:1566–75.
9 Hedberg Y, Ljungberg B, Roos G, Landberg G. Expression of cyclin D1, D3, E, and p27 in human renal cell carcinoma analyzed by tissue microarray. Br J Cancer 2003;88:1417–23.[CrossRef][Medline]
10 Span PN, Tjan-Heijnen VC, Manders P, Beex LV, Sweep CG. Cyclin-E is a strong predictor of endocrine therapy failure in human breast cancer. Oncogene 2003;22:4898–904.[CrossRef][ISI][Medline]
11 Donnellan R, Kleinschmidt I, Chetty R. Cyclin E immunoexpression in breast ductal carcinoma: pathologic correlations and prognostic implications. Hum Pathol 2001;32:89–94.[CrossRef][ISI][Medline]
12 Porter PL, Malone KE, Heagerty PJ, Alexander GM, Gatti LA, Firpo EJ, et al. Expression of cell-cycle regulators p27Kip1 and cyclin E, alone and in combination, correlate with survival in young breast cancer patients. Nat Med 1997;3:222–5.[CrossRef][ISI][Medline]
13 Han S, Park K, Bae BN, Kim KH, Kim HJ, Kim YD, et al. Prognostic implication of cyclin E expression and its relationship with cyclin D1 and p27Kip1 expression on tissue microarrays of node negative breast cancer. J Surg Oncol 2003;83:241–7.[CrossRef][ISI][Medline]
14 Lindahl T, Landberg G, Ahlgren J, Nordgren H, Norberg T, Klaar S, et al. Overexpression of cyclin E protein is associated with specific mutation types in the p53 gene and poor survival in human breast cancer. Carcinogenesis 2004;25:375–80.
15 Nielsen NH, Arnerlov C, Cajander S, Landberg G. Cyclin E expression and proliferation in breast cancer. Anal Cell Pathol 1998;17:177–88.[ISI][Medline]
16 Kuhling H, Alm P, Olsson H, Ferno M, Baldetorp B, Parwaresch R, et al. Expression of cyclins E, A, and B, and prognosis in lymph node-negative breast cancer. J Pathol 2003;199:424–31.[CrossRef][ISI][Medline]
17 Rudolph P, Kuhling H, Alm P, Ferno M, Baldetorp B, Olsson H, et al. Differential prognostic impact of the cyclins E and B in premenopausal and postmenopausal women with lymph node-negative breast cancer. Int J Cancer 2003;105:674–80.[CrossRef][ISI][Medline]
18 Bukholm IR, Bukholm G, Nesland JM. Over-expression of cyclin A is highly associated with early relapse and reduced survival in patients with primary breast carcinomas. Int J Cancer 2001;93:283–7.[CrossRef][Medline]
19 Engels WR. Contributing software to the Internet: the Amplify program. Trends Biochem Sci 1993;18:448–50.[CrossRef][ISI][Medline]
20 Mitas M, Mikhitarian K, Walters C, Baron PL, Elliott BM, Brothers TE, et al. Quantitative real-time RT-PCR detection of breast cancer micrometastasis using a multigene marker panel. Int J Cancer 2001;93:162–71.[CrossRef][ISI][Medline]
21 Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001;25:402–8.[CrossRef][ISI][Medline]
22 McCarty KS Jr, Miller LS, Cox EB, Konrath J, McCarty KS Sr. Estrogen receptor analyses. Correlation of biochemical and immunohistochemical methods using monoclonal antireceptor antibodies. Arch Pathol Lab Med 1985;109:716–21.[ISI][Medline]
23 Tsuda H, Bandera CA, Birrer MJ, Hashiguchi Y, Berkowitz RS, Mok SC. Cyclin E amplification and overexpression in clear cell adenocarcinoma of the ovary. Oncology 2004;67:291–9.[CrossRef][Medline]
24 Muller-Tidow C, Metzger R, Kugler K, Diederichs S, Idos G, Thomas M, et al. Cyclin E is the only cyclin-dependent kinase 2-associated cyclin that predicts metastasis and survival in early stage non-small cell lung cancer. Cancer Res 2001;61:647–53.
25 Porter DC, Zhang N, Danes C, McGahren MJ, Harwell RM, Faruki S, et al. Tumor-specific proteolytic processing of cyclin E generates hyperactive lower-molecular-weight forms. Mol Cell Biol 2001;21:6254–69.
26 Harwell RM, Porter DC, Danes C, Keyomarsi K. Processing of cyclin E differs between normal and tumor breast cells. Cancer Res 2000;60:481–9.
27 Akli S, Keyomarsi K. Low-molecular-weight cyclin E: the missing link between biology and clinical outcome. Breast Cancer Res 2004;6:188–91.[CrossRef][Medline]
28 Porter DC, Keyomarsi K. Novel splice variants of cyclin E with altered substrate specificity. Nucleic Acids Res 2000;28:E101–8.
29 Ekholm-Reed S, Spruck CH, Sangfelt O, van Drogen F, Mueller-Holzner E, Widschwendter M, et al. Mutation of hCDC4 leads to cell cycle deregulation of cyclin E in cancer. Cancer Res 2004;64:795–800.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||