| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular Diagnostics and Genetics |
Departments of1 Chemical Endocrinology, 2 Clinical Chemistry, and 3 Medical Oncology, University Medical Center Nijmegen, Nijmegen, The Netherlands.
aAddress correspondence to this author at: 530 Department of Chemical Endocrinology, University Medical Center Nijmegen, PO Box 9101, 6500 HB Nijmegen, The Netherlands. Fax 31-24-3541484; e-mail p.span{at}ace.umcn.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Methods: Survivin mRNA was measured by quantitative TaqMan reverse transcription-PCR in 275 breast cancer tissues from patients with operable tumors and was correlated with established clinicopathologic factors, relapse-free survival [(RFS); 102 events], and overall survival [(OS); 81 events].
Results: High survivin mRNA concentrations were found mainly in tissues from younger patients and in high-grade cancer tissues. High survivin concentrations were most strongly associated with estrogen receptor- or progesterone receptor-negative tumors. In univariate Cox regression analysis for RFS, survivin concentrations were significantly associated with poor prognosis with a hazard ratio (HR) of 1.99 (95% confidence interval, 1.31–3.02; P = 0.001) for every 10-fold increase in expression. For OS, a significant contribution of survivin to poor prognosis was found with a HR of 2.76 (1.67–4.55; P <0.001). Multivariate analyses were performed including established clinicopathologic factors. For RFS, age (P = 0.027), nodal category (P <0.001), and survivin [HR = 1.78 (1.18–2.68); P = 0.006] contributed significantly to the model. For OS, only nodal category (P <0.001) and survivin [HR = 3.05 (1.83–5.10); P <0.001] were significant.
Conclusion: Survivin demonstrates a strong, independent, association with poor prognosis. Survivin might be used as a new marker to stratify breast cancer patients for more optimal treatment modalities, or it could be a promising new target for therapy.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
200 000 new cases and 70 000 deaths each year in the European Community, breast cancer is the leading malignancy in women and a leading cause of death. The basic understanding of breast cancer initiation and progression is still incomplete. In addition, there is a need to develop improved methods to stratify breast cancer patients into different risk groups more accurately than can be achieved with current clinicopathologic classification methods. Hence, low-risk patients can be spared unnecessary treatment, avoiding side effects and reducing the cost of treatment. Moreover, high-risk patients could be rapidly identified and offered treatment modalities customized (more aggressive) to individual patients. Newly designed biological therapies aimed at specific tumor cell-associated target molecules could also be devised. Survivin [baculoviral inhibitor of apoptosis (IAP)1 repeat-containing protein-5 (Birc-5)] is a member of the IAP gene family, which has been implicated in both inhibition of apoptosis and mitosis regulation [for a review, see Ref. (1)]. Survivin is one of the most uniformly up-regulated genes in tumor tissues compared with healthy tissues (2). Uncontrolled growth of cancerous cells requires antiapoptotic strategies to extend an otherwise limited lifespan and to counter the customary apoptotic triggers. In addition, cells that are unresponsive to apoptotic triggers will also be more resistant to radiation and chemotherapy. Indeed, high survivin expression in the primary tumor is almost invariably associated with poor patient prognosis in many cancer types (3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14).
The association of survivin with prognosis in breast cancer patients is, however, ambiguous; previous studies have reported it to be either irrelevant (15) or associated with poor (16) or with good prognosis (17). However, qualitative reverse transcription-PCR (RT-PCR) or immunohistochemistry using antibodies with different sensitivities toward the known survivin variants were used in these studies. Measuring survivin mRNA by real-time fluorescence RT-PCR has the advantages of being more quantitative then classical RT-PCR and, in general, more specific and sensitive then antibody-based assays. Because survivin concentrations are largely controlled at the level of gene transcription (1)(18), quantitative RT-PCR should yield representative data on survivin protein concentrations. In the present study, we measured survivin mRNA by quantitative TaqMan RT-PCR in 275 breast cancer tissues and correlated the copy numbers with established clinicopathologic factors, relapse-free survival (RFS), and overall survival (OS).
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
tissue processing
After primary surgery, a representative part of the tumor was macroscopically selected by a pathologist, frozen in liquid nitrogen, and sent to our department for routine determination of ER and PgR status by ligand-binding assay according to the dextran-charcoal method. Aliquots of tissue were pulverized with a microdismembrator (Braun) and kept in liquid nitrogen until RNA isolation. The samples were coded, and clinical information was unavailable to the technicians performing the mRNA quantification. Total RNA was isolated from 20 mg of tissue powder by use of the RNeasy mini reagent set (Qiagen) with on-column DNase-I treatment. The quality of the RNA was checked by examining ribosomal RNA bands after agarose gel electrophoresis and by amplifying hypoxanthine phosphoribosyltransferase (HPRT) as a control (see below). RNA concentrations were determined spectrophotometrically based on absorbance at 260 nm (Genequant; Amersham). No association of RNA degradation or concentration with length of storage was found. The selected part of the tumor tissue was considered representative for the whole tumor because mammacarcinoma is a heterogeneous tumor in which both malignant cells and "healthy" tissue (components) interact. Additionally, in our experience only limited amounts of RNA can be extracted from healthy breast tissue because this contains mostly fat and fibrogenous materials. Measuring survivin mRNA and correcting for HPRT concentration thus partly compensates for "dilution" by healthy tissue.
rt-pcr
Purified total RNA (1.0 µg) was denatured for 10 min at 70 °C and immediately cooled on ice. Reverse transcription was performed with the Reverse Transcription System (Promega Benelux BV) according to the manufacturer’s protocol. After annealing of random hexamers for 10 min at 20 °C, cDNA synthesis was performed for 60 min at 42 °C, followed by an enzyme inactivation step for 5 min at 95 °C. Quantitative PCR was performed as reported previously (12), with both survivin and HPRT mRNA concentrations expressed in absolute copy numbers. Four plates were run for each amplicon in the present study. Survivin and HPRT mRNA copy numbers were quantified by constructing plasmids containing either of the amplicons. A triplicate 5-log-range calibration curve containing 10 to 106 copies of either survivin or HPRT was included in each real-time PCR assay plate. The characteristics of the survivin calibration curves were as follows [mean (SD) of four curves]: slope = –3.464 (0.089); y-intercept = 38.323 (0.426); and correlation coefficient = 0.996 (0.004). For HPRT this amounted to the following: slope = –3.468 (0.057); y-intercept = 37.006 (0.231); and correlation coefficient = 0.998 (0.001).
statistical analyses
Statistical analyses were carried out with SPSS 10.0.5 software (SPSS Benelux BV). The normality of the distribution was tested by the method of Kolmogorov–Smirnov. Differences in expression in samples from patients categorized by clinicopathologic characteristics, used as grouping variables, were assessed with Student’s t-test or ANOVA after normalization by log-transformation where appropriate. Nonparametric correlations (rs) were established using Spearman rank correlation testing. RFS time (defined as the time from surgery until diagnosis of recurrent disease) and OS time (defined as the time between date of surgery and death by any cause) were used as follow-up endpoints. The Cox proportional hazards model was used to assess the prognostic value of survivin expression as a log-transformed continuous factor and in addition to other clinicopathologic factors. Kaplan-Meier survival curves were generated after we established an optimum cutoff value in the total group of patients to visually inspect the proportionality assumption. Equality of survival distributions was tested by log-rank testing. Two-sided P values <0.05 were considered to be statistically significant. Cases with >96 months of follow-up were censored at 96 months because of the rapidly decreasing number of patients still surviving after that length of time, although data on some patients were available for up to 169 months after primary surgery. After a certain period of observation, patients were frequently redirected to their general practitioners for checkups and mammography and ceased to belong to the outpatients collective of our breast cancer clinic. Further inclusion of the small remaining groups in statistical analyses would be noninformative.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
). We next investigated the correlation of survivin mRNA concentrations in the primary tumor with other clinicopathologic factors (age, menopausal status, nodal category, tumor histology, tumor size, histologic grade, ER and PgR status) and with type of surgery and adjuvant therapy (Table 1
). Patients <50 years of age had significantly higher survivin mRNA concentrations (P = 0.021). We found a negative correlation between continuous survivin concentrations and age (rs = –0.173; P = 0.004). Survivin mRNA concentrations were higher in ductal than in lobular tumors (P = 0.015), and grade III tumors had significantly higher survivin mRNA concentrations than did grade I or grade II tumors (P = 0.008). Most pronounced was the association of high survivin concentrations with ER- and PgR-negative tumors (P <0.001 for both). This was not the result of a relationship between survivin and absolute ER concentrations; we found no correlation within the ER-positive group (rs = –0.028; P = 0.710). Thus, survivin mRNA concentrations were highest in tumors that exhibited a generally worse prognosis according to present established clinicopathologic factors.
|
|
association with prognosis
Whether survivin is indeed associated with a poor prognosis, as suggested by its association with other patient and tumor characteristics, was subsequently investigated in univariate survival analyses. Survivin concentrations were normalized by log-transformation and entered in univariate Cox regression analysis for RFS and OS. For RFS, survivin concentrations were significantly associated with poor prognosis with a hazard ratio (HR) of 1.99 [95% confidence interval (CI), 1.31–3.02; P = 0.001; Table 2
]. Because survivin is entered as a log-transformed continuous factor, this HR stands for the increase in risk for every 10-fold increase in survivin concentration. For OS, a significant contribution of survivin to poor prognosis was found with a HR of 2.76 (95% CI, 1.67–4.55; P <0.001; Table 3
).
|
|
dichotomization into risk groups
To allow analysis of survivin expression as a categorized variable, and for visualization in Kaplan–Meier survival curves to classify tumors as high vs low risk for relapse, tumor survivin concentrations were dichotomized by an optimal cutoff value. We found the most significant difference in RFS (P <0.0001, log-rank; Fig. 2A
) after dichotomizing at a survivin/HPRT ratio of 420. When we used this ratio to dichotomize results, tumors from 207 (75.3%) patients were considered having low and 68 (24.7%) as having high concentrations of survivin. The patients with high survivin concentrations had a HR of 2.34 (95% CI, 1.54–3.56; P <0.001). For optimal differentiation of patients for their OS time, 206 (75.7%) patients had low and 66 (24.3%) had high survivin concentrations in their tumors (P = 0.0004, log-rank; Fig. 2B
). For OS, patients with high survivin concentrations had a HR of 2.39 (95% CI, 1.45–3.94; P = 0.001).
|
multivariate regression analysis
The independent relationship of survivin with RFS and OS was studied with Cox multivariate regression analysis. We thus could establish whether the prognostic value of survivin as found in univariate analyses was attributable to its relationship with other clinicopathologic factors or whether survivin itself contributes independently to prognosis. A multivariate analysis was performed including age, menopausal status, nodal category, tumor size, tumor grade, and ER/PgR status. For RFS, age (P = 0.027), nodal category (P <0.001), and survivin concentration (HR = 1.78; 95% CI, 1.18–2.68; P = 0.006) contributed significantly to the model (Table 2
). For OS, only nodal category (P <0.001) and survivin concentration (HR = 3.05; 95% CI, 1.83–5.10; P <0.001) were significant (Table 3
).
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Our results concur with studies in other tumor types, which have shown that survivin relates to poor disease outcome in a variety of tumors, i.e., neuroblastoma (3), colorectal cancer (4), non-small-cell lung cancers (5), B-cell lymphoma (6), T-cell leukemia (7), hepatocellular carcinoma (8), esophageal carcinoma (9), rectal cancer (10), glioma (11), bladder cancer (12), soft tissue sarcoma (13), and astrocytic tumors (14). Results obtained in previous studies into the association of survivin with prognosis of breast cancer patients were inconclusive (15) or contradictory (16)(17). The prediction of a poor prognosis by survivin mRNA concentrations, as we show here, is to be anticipated from the function of survivin as an inhibitor of effector caspases, thus inhibiting both intrinsic and extrinsic apoptosis pathways (1). Cells that are unresponsive to apoptotic triggers will also be more resistant to cytotoxic treatments, as are cells that overexpress survivin (24). As such, the recent report on survivin as playing a pivotal role in vascular endothelial growth factor (VEGF)-mediated chemoresistance in endothelial cells (25) is of importance because we recently found VEGF to be an excellent predictor of poor disease outcome in breast cancer patients treated with either radiotherapy (26) or endocrine therapy (27). Possibly, the prediction of poor prognosis by survivin we report here is attributable to its association with (VEGF-related) therapy resistance. Indeed, the fact that absolute, log-transformed survivin concentrations are more strongly related to OS than to RFS would suggest an association with success of therapy given after relapse. It should be noted that the stronger prognostic value of survivin for OS than for RFS was not appreciable after the patient group was dichotomized on the basis of an optimal cutoff value for survivin mRNA. Survivin might also function as a mitosis regulator, and the overexpression of survivin in tumor cells irrespective of cell cycle progression might relate to independent proliferation irrespective of therapy (1). Whether survivin is related to disease progression (i.e., prognostic) or treatment resistance (predictive) should be addressed in larger patient groups to allow for proper subgroup analyses.
Measuring survivin mRNA by real-time RT-PCR has the advantages of being more quantitative than classical RT-PCR. Because survivin concentrations are largely controlled at the level of gene transcription (1)(18), quantitative RT-PCR should yield representative data on survivin protein concentrations. One previous study also used RT-PCR to detect survivin mRNA in breast cancer tissues, but in contrast to our findings, they failed to find a correlation with disease outcome (15). Importantly, a qualitative RT-PCR was used in that study, in a more limited number (n = 106) of tissues than we report on here. We show in the present study that the quantity of survivin mRNA is related to prognosis. Furthermore, the same group reported that nuclear survivin protein, as measured by immunohistochemistry in 293 samples, was associated with a favorable disease outcome in breast cancer (17). This could not be replicated in the 106 samples used for the RT-PCR study (15). These results are at variance with another immunohistochemical study that reported that survivin protein concentrations were correlated with inhibition of apoptosis and, indirectly, with poor prognosis (16). Of importance, different antibodies have been used in the immunohistochemical analyses with differing sensitivities toward the splice variants of survivin (16)(17). Only the antibody used by Kennedy et al. (17) was capable of detecting the survivin 
Ex3 splice variant. It could be speculated that nucleus-localized survivin is represented by the survivin Ex3 splice variant. This splice variant lacks the nuclear export signal and, because of a frame shift induced by exon 3 skipping, possesses a nuclear localization signal (1). Indeed, similar to in breast cancer (17), nucleus-localized survivin is associated with favorable prognosis in gastric cancer (28) and osteosarcoma (29). However, nuclear staining for survivin can also be related to poor prognosis in esophageal squamous cell carcinoma (30) and Mantle cell lymphoma (31). Thus, the prognostic value of nucleus-localized survivin is not necessarily different from that of cytosolic survivin.
In conclusion, survivin mRNA, as measured by quantitative RT-PCR in the primary tumor, has strong and independent prognostic value in human breast cancer. Survivin mRNA concentrations in the tumor can be used to classify breast cancer patients into different risk groups. Thus, low-risk patients can be spared unnecessary treatment and high-risk patients can be offered more aggressive treatment modalities. These results also support survivin as a promising target for therapy in breast cancer, e.g., by vaccination or administration of antisense oligonucleotides or mutant survivin adenoviruses (1).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
and MRP-1 mRNAs in breast cancer. Cancer Lett 2003;201:225-236.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. Ansems, S. Hontelez, M. W. G. Looman, N. Karthaus, P. Bult, J. J. Bonenkamp, J. H. Jansen, F. C. G. J. Sweep, P. N. Span, and G. J. Adema DC-SCRIPT: Nuclear Receptor Modulation and Prognostic Significance in Primary Breast Cancer J Natl Cancer Inst, January 6, 2010; 102(1): 54 - 68. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Giaccone, P. Zatloukal, J. Roubec, K. Floor, J. Musil, M. Kuta, R. J. van Klaveren, S. Chaudhary, A. Gunther, and S. Shamsili Multicenter Phase II Trial of YM155, a Small-Molecule Suppressor of Survivin, in Patients With Advanced, Refractory, Non-Small-Cell Lung Cancer J. Clin. Oncol., September 20, 2009; 27(27): 4481 - 4486. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. W. Tolcher, A. Mita, L. D. Lewis, C. R. Garrett, E. Till, A. I. Daud, A. Patnaik, K. Papadopoulos, C. Takimoto, P. Bartels, et al. Phase I and Pharmacokinetic Study of YM155, a Small-Molecule Inhibitor of Survivin J. Clin. Oncol., November 10, 2008; 26(32): 5198 - 5203. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Shen, L. Hu, L. Xia, and Y. Li Quantitative Real-time RT-PCR Detection for Survivin, CK20 and CEA in Peripheral Blood of Colorectal Cancer Patients Jpn. J. Clin. Oncol., November 1, 2008; 38(11): 770 - 776. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Brennan, E. Rexhepaj, S. L. O'Brien, E. McSherry, D. P. O'Connor, A. Fagan, A. C. Culhane, D. G. Higgins, K. Jirstrom, R. C. Millikan, et al. Altered Cytoplasmic-to-Nuclear Ratio of Survivin Is a Prognostic Indicator in Breast Cancer Clin. Cancer Res., May 1, 2008; 14(9): 2681 - 2689. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. N. Span, V. C.G. Tjan-Heijnen, J. J.T.M. Heuvel, J. B. de Kok, J. A. Foekens, and F. C.G.J. Sweep Do the Survivin (BIRC5) Splice Variants Modulate or Add to the Prognostic Value of Total Survivin in Breast Cancer? Clin. Chem., September 1, 2006; 52(9): 1693 - 1700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lacroix, R.-A. Toillon, and G. Leclercq p53 and breast cancer, an update. Endocr. Relat. Cancer, June 1, 2006; 13(2): 293 - 325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. W. Rahman, Y. Li, Z. Wang, S. H. Sarkar, and F. H. Sarkar Gene Expression Profiling Revealed Survivin as a Target of 3,3'-Diindolylmethane-Induced Cell Growth Inhibition and Apoptosis in Breast Cancer Cells. Cancer Res., May 1, 2006; 66(9): 4952 - 4960. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Fukuda and L. M. Pelus Survivin, a cancer target with an emerging role in normal adult tissues Mol. Cancer Ther., May 1, 2006; 5(5): 1087 - 1098. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Ryan, G. E. Konecny, S. Kahlert, H.-J. Wang, M. Untch, G. Meng, M. D. Pegram, K. C. Podratz, J. Crown, D. J. Slamon, et al. Survivin expression in breast cancer predicts clinical outcome and is associated with HER2, VEGF, urokinase plasminogen activator and PAI-1 Ann. Onc., April 1, 2006; 17(4): 597 - 604. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Asanuma, T. Torigoe, K. Kamiguchi, Y. Hirohashi, T. Ohmura, K. Hirata, M. Sato, and N. Sato Survivin Expression Is Regulated by Coexpression of Human Epidermal Growth Factor Receptor 2 and Epidermal Growth Factor Receptor via Phosphatidylinositol 3-Kinase/AKT Signaling Pathway in Breast Cancer Cells Cancer Res., December 1, 2005; 65(23): 11018 - 11025. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
