Clinical Chemistry 52: 1693-1700, 2006.
First published July 27, 2006; 10.1373/clinchem.2006.071613
(Clinical Chemistry. 2006;52:1693-1700.)
© 2006 American Association for Clinical Chemistry, Inc.
Do the Survivin (BIRC5) Splice Variants Modulate or Add to the Prognostic Value of Total Survivin in Breast Cancer?
Paul N. Span1,a,
Vivianne C.G. Tjan-Heijnen2,
Joop J.T.M. Heuvel1,
Jacques B. de Kok3,
John A. Foekens4 and
Fred C.G.J. Sweep1
1 Departments of Chemical Endocrinology;
2 Medical Oncology; and
3 Clinical Chemistry; Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.
4 Department of Medical Oncology, Erasmus MC-Daniel den Hoed, Rotterdam, The Netherlands.
aAddress correspondence to this author at: 479 Department of Chemical Endocrinology, Radboud University Nijmegen Medical Centre, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. Fax 3124-354-1484; e-mail p.span{at}ace.umcn.nl.
 |
Abstract
|
|---|
Background: A total of 4 additional splice variants (survivin-
Ex3, survivin 2
, survivin-2B, and survivin-3B) have been described for survivin [baculoviral IAP repeat-containing protein (BIRC-5), approved gene symbol BIRC5], which has been implicated in both inhibition of apoptosis and regulation in mitosis in many tumor types. In this study, we assessed whether the survivin splice variants modulate or add to the prognostic value of total survivin in breast cancer.
Methods: With quantitative reverse transcription-PCR, we measured mRNA concentrations of survivin and all variants in tumor tissue from 275 patients with breast cancer and associated these with clinicopathologic characteristics and relapse-free survival.
Results: Total survivin, survivin-Ex3, and survivin 2 mRNA levels were associated with young age and ductal histology. Total survivin and survivin-Ex3 were highest in samples with advanced histological grade, whereas patients with 4–9 involved lymph nodes expressed less survivin-2B mRNA than those with 1–3 involved nodes. All variants were higher in tumors negative for steroid hormone receptors. Total survivin, survivin 2, and survivin-3B were associated with poor relapse-free survival in univariate analyses. Survivin 2 and survivin-3B added to the prognostic value of total survivin in multivariate analyses. In addition, the prognostic value of total survivin was evident only in the presence of higher expression levels of these 2 variants.
Conclusions: All variants of survivin exhibited particular associations with clinicopathologic characteristics (age, histology, grade, and steroid hormone receptor status) of breast cancer patients. Survival analyses suggest a modulating role of survivin 2 and survivin-3B on the biological function of total survivin.
|
Introduction
|
|---|
Survivin [baculoviral IAP repeat-containing protein (BIRC-5), approved gene symbol BIRC5] is a member of the inhibitors-of-apoptosis gene family that is implicated in both inhibition of apoptosis and in mitosis regulation(1). Survivin is one of the most uniformly upregulated genes in tumor tissues compared with healthy tissues(2). High survivin expression in the primary tumor, in many cancer types, is almost invariably associated with a poor prognosis for the patient. For breast cancer patients, however, the association of survivin with prognosis is ambiguous, because previous studies have reported it to be either irrelevant(3), or associated with poor(4) or good prognoses(5). We earlier reported that survivin mRNA concentrations are a strong, independent marker for poor prognosis in primary breast cancer(6). Our recent results on survivin protein concentrations in breast cancer suggest that the type of adjuvant therapy might cause these discrepancies(7). Survivin is also one of the 16 cancer-related genes represented in the Oncotype DX assay(8).
In addition to the originally described survivin form [survivin-wild type (wt)](9), Mahotka et al. later discovered 2 splice variants, survivin-Ex3 and survivin-2B(10). More recently, 2 additional splice variants, survivin-3B(11) and survivin 2(12), have been described. The survivin gene on chromosome 17q25 is now believed to consist of 4 dominant exons (1, 2, 3, and 4) and 3 additional cryptic exons located either in intron 2 (2, 2B) or intron 3 (3B). The 5 splice variants potentially produce distinct proteins with lengths of 74 (survivin 2), 120 (survivin-3B), 137 (survivin-Ex3), 142 (survivin-wt), or 165 (survivin-2B) amino acids. Evidence exists that the diverse roles of survivin might be regulated by these splice variants through heterodimerization with survivin-wt(13). Some of the distinct variants have particular functions; survivin-Ex3 retains its antiapoptotic function despite the loss of exon 3(10), whereas survivin-2B(10) and survivin 2(12) possibly counter the antiapoptotic function of survivin-wt. The survivin-3B splice variant is antiapoptotic, but lacks a carboxyl-terminal coiled-coil domain, suggesting that it is not involved in cell-cycle regulation(11). Also, the different forms of survivin have different subcellular localizations(14), which may also be relevant to its function and prognostic value(15). However, extrapolation of these in vitro observations in clinical samples is lacking.
In breast cancer, the occurrence of the mRNAs of the first 3(3)(16) or 4(17) of these survivin variants in relation to tumor characteristics has been described. So far, however, there has not been a quantitative description of all 5 variants by quantitative reverse transcription (RT)-PCR1
in human breast cancer, nor has their expression been related to disease outcome. Here, we address whether the expression levels of the splice variants add to the prognostic value of total survivin(6)(8) and/or whether these variants modulate this prognostic value.
|
Patients and Methods
|
|---|
patients
The current study, in which coded tumor tissues were used, was done in accordance with the Code of Conduct of the Federation of Medical Scientific Societies in the Netherlands (Code for Proper Secondary Use of Human Tissue in the Netherlands, http://www.fmwv.nl) and adhered to all relevant institutional and national guidelines. A series of 275 patients with unilateral, resectable breast cancer who underwent surgery of their primary tumor between November 1987 and December 1997 were selected by the availability of frozen tissue in the tumor bank of the Department of Chemical Endocrinology of the Radboud University Nijmegen Medical Centre. This bank contains frozen tumor tissue of patients with breast cancer from 5 different hospitals of the Comprehensive Cancer Centre East in the Netherlands. The patients, inclusion and exclusion criteria, and their treatment have been described earlier(6).
tissue processing
After primary surgery, a representative part of tumor was macroscopically selected by a pathologist, frozen in liquid nitrogen, and sent to our department for routine determination of estrogen and progesterone receptor status by ligand binding assay according to the dextran-charcoal method. Aliquots of tissue were pulverized using a microdismembrator (Braun) and kept in liquid nitrogen until RNA isolation. Total RNA was isolated from 
20 mg of tissue powder using the RNeasy Mini Kit (Qiagen) with on-column DNase-I treatment.
pcr
Quantitative RT-PCR consisted of either Taqman (survivin(6) and survivin-Ex3) or SybrGreen (survivin 2, survivin-2B, and survivin-3B) methodologies. See Fig. 1
for the location of the amplicons. For survivin-Ex3, the forward primer (900 nmol/L) was 5'-GAT GAC GAC CCC ATG CAA A-3', the reversed primer (600 nmol/L) was 5'-AGG CCT CAA TCC ATG GCA G-3', and the TET-labeled Taqman probe (200 nmol/L) was 5'-AGA AAG TGC GCC GTG CCA TCG A-3'. For survivin 2, the forward primer (300 nmol/L) was 5'-GCT TTG TTT TGA ACT GAG TTG TCA A-3' and the reversed primer (50 nmol/L) was 5'-GCA ATG AGG GTG GAA AGC A-3'. For survivin 2B, the forward primer (300 nmol/L) was 5'-GCA CGG TGG CTT ACG CCT G-3' and the reversed primer (300 nmol/L) was 5'-AAC CGG ACG AAT GCT TTT TAT GTT CC-3'. For survivin 3B, the forward primer (50 nmol/L) was 5'-CAG ATT CAG GGA GGG ACT GG-3' and the reversed primer (50 nmol/L) was 5'-CAA ACA TCA GGC TCT TCC TCG-3'. All data were normalized against hypoxanthine-guanine phosphoribosyl-transferase (HPRT) expression, which proved to be a proper housekeeping gene in our sample cohort(18).
View larger version (13K):
[in this window]
[in a new window]
|
Figure 1. Survivin and its splice variants.
Arrows are stop codons, and primers are shown above the locations.
|
|
statistical analyses
All statistical analyses were performed with the SPSS statistical software package 12.0.1 (SPPS Inc). Normality of value distributions were assessed by Kolomogorov-Smirnov testing. Either log-normalized data were used for analyses using parametric tests, or nonparametric testing was performed. For correlation analysis of continuous data, Spearman rank-correlation tests were used. Survivin concentrations in particular patient categories were compared by either Mann–Whitney U or Kruskall-Wallis tests. For log-normalized data, ANOVA with post hoc Tukey honestly significant difference testing was used to identify which patient group differed. Relapse-free survival (RFS) time (defined as the time from surgery until diagnosis of recurrent disease) was used as a follow-up endpoint. The Cox proportional hazards model was used to assess the prognostic value of survivin variants expression as log-transformed continuous factors, both in univariate analyses and in multivariate analyses including other clinicopathologic variables including age at primary surgery, menopausal status, tumor histology, tumor size, histological grade, estrogen and progesterone receptor status, involved axillary lymph nodes, and adjuvant systemic therapy. Multivariate analysis used a backward stepwise selection. Removal testing was based on the probability, at P <0.1, of the likelihood-ratio statistic based on the maximum partial likelihood estimates. After fitting together a base model consisting of the traditional prognostic factors with survivin, the other variants were entered separately in a second block. Survival curves were generated using the method of Kaplan and Meier, and the significance of differences in survival between dichotomized patient groups was obtained by log-rank testing. Dichotomization of the patient group was done according to the cut-point for survivin found earlier(6) or by the median level of the variants described here.
|
Results
|
|---|
correlations between variants
In addition to the data on total survivin we described earlier(6), we used quantitative RT-PCR to measure the expression of an additional 4 survivin splice variants in tumor tissue from 275 patients with primary breast cancer. In particular, the survivin-Ex3 variant was highly correlated with total survivin (rs= 0.747, P <0.001). The 3 other variants correlated less well with total survivin or with each other, with Spearman correlation coefficients ranging from 0.312 to 0.534 (all P <0.001).
association with patient characteristics
As reported earlier, the concentration of total survivin mRNA was negatively correlated with the age of the patient (P = 0.004). This was also true for survivin-Ex3 (P = 0.026) and survivin 2 (P = 0.010), but not for the other variants (Table 1
). Total survivin (P = 0.040) and survivin 2 (P = 0.050) were highest in premenopausal patients. In ductal breast cancer, survivin and most variant mRNA concentrations were higher than in lobular cancer, a difference that was found to be statistically significant for survivin (P = 0.024), survivin-Ex3 (P = 0.018), and survivin 2 (P = 0.041). Concentrations of both survivin (P = 0.008) and survivin-Ex3 variant mRNA (P = 0.001) were significantly associated with advanced histological grade. The other variants exhibited no relation with histological grade (Table 1
). Concentrations of survivin-2B variant differed significantly (P = 0.040) in relation to the number of involved lymph nodes. Post hoc testing revealed that in patients with 4–9 involved lymph nodes less survivin-2B mRNA was expressed than in patients with 1–3 involved nodes, whereas no difference was found for those patients without involved lymph nodes. Interestingly, the mRNA concentrations of all variants, most notably total survivin and the survivin-Ex3 variant, were highest in steroid hormone-receptor–negative tumors (Table 1
).
View this table:
[in this window]
[in a new window]
|
Table 1. Categorical distributions of baseline characteristics in all patients and survivin and the splice variants.
|
|
Thus survivin and some of the splice variants, particularly the survivin-Ex3 and survivin 2 variants, were associated with young age, advanced histological grade, and steroid hormone-receptor–negative tumors, all factors that denote a poor prognosis for the patient. Therefore, we further investigated the relation between the expression levels of survivin splice variants and prognosis.
univariate regression analyses
In univariate Cox regression analysis for RFS, young age (P = 0.006), premenopausal status (P = 0.029), ductal vs mixed histology (P = 0.026), pT2 or pT3/4 tumor size (P = 0.022), advanced histological grade (P = 0.021), and number of involved axillary lymph nodes (P <0.001) were all characteristic for a short time to recurrence (Table 2
). In addition, survivin (P = 0.001), survivin 2 (P = 0.008), and survivin-3B (P = 0.010), as log-transformed continuous variables, were significantly associated with poor RFS according to Cox regression analysis.
multivariate regression analyses
We performed multivariate analysis to investigate whether the variants added to the prognostic data that were obtained with survivin. A base model was constructed with classic clinicopathological variables and total survivin, and the variants were individually added. The base model that best predicted RFS in multivariate analysis consisted of age (P = 0.017), tumor size (P = 0.089), involved lymph nodes (P <0.001), adjuvant therapy (P = 0.001), and survivin (P = 0.003) (Table 3
). Adding the variants separately to this model showed that survivin 2 (hazard ratio = 1.9, 95% confidence interval = 1.1–3.2, P = 0.017) and survivin-3B (hazard ratio = 1.6, 95% confidence interval = 1.0–2.6; P = 0.048) added to the prognostic value obtained by classic clinicopathological factors and survivin in multivariate analysis (Table 3B
).
To investigate whether the variants modulated the prognostic value of survivin expression, patient groups were dichotomized based on the median expression levels of the variants, and then the prognostic value of survivin was analyzed in the resulting subgroups (Table 4
). Survivin was prognostic with similar hazard ratios of 2–2.5 in the patient groups defined by low or high survivin-Ex3 or survivin-2B values, albeit with a poorer significance in the low expressing group. In contrast, for both survivin 2 and survivin-3B, we found that survivin was prognostic only in the presence of higher expression of the variant (Fig. 2
).
View larger version (27K):
[in this window]
[in a new window]
|
Figure 2. Kaplan-Meier curves of RFS of patients categorized based on survivin (with dashed line >420 and solid line <420) in patients with low (left) or high (right) survivin 2 (A) or survivin-3B (B) mRNA concentrations in their primary tumor.
|
|
|
Discussion
|
|---|
A large number of apoptotic factors are regulated via alternative splicing, a process that allows for the production of discrete protein isoforms with often distinct functions from a common mRNA precursor(19). Here we are the first to quantify the mRNA concentrations of all 5 different forms of survivin in primary breast cancer samples, their association with clinicopathological characteristics, and their contribution to prognosis prediction in these patients. In addition to the association of survivin with particular clinicopathological characteristics we described earlier(6), we found survivin-Ex3 and survivin 2 mRNA concentrations to be associated with young age at diagnosis and ductal histology. Survivin-Ex3 was also highest in samples with advanced histological grade, whereas survivin-2B was lower in patients with more involved lymph nodes. All variants exhibited higher mRNA concentrations in steroid hormone-receptor–negative tumors. These results point to a possible relation of the splice variants with prognosis. Indeed, survivin, survivin 2, and survivin-3B were associated with poor RFS. The association with poor RFS of the latter 2 was independent of survivin expression because these variants contributed substantially in multivariate analyses including survivin. Furthermore, both these variants modulated the prognostic value of survivin, because in the presence of low concentrations of the variants the prognostic value of survivin was no longer significant. No association with prognosis was seen for the survivin-Ex3 or survivin-2B variants. Indeed, it has recently been reported that neither survivin-Ex3 nor survivin-2B acts as a survivin competitor during mitosis nor has an essential function, which is in line with our data on these variants(20).
This is the first study to use quantitative RT-PCR (Q-PCR) to quantify the amount of mRNA of all 5 variants in breast cancer. Other studies have relied on amplification over a long part of the gene and separation of the products on gel to distinguish between the variants(3)(16)(17). Q-PCR applies the amplification of short amplicons to achieve high amplification efficiencies. Therefore, it is not possible to devise a specific Q-PCR assay for survivin-wt (see Fig. 1
). Similar to O’Driscoll et al.(3) and Ryan et al.(16), we found a substantial correlation between the mRNA concentrations of the splice variants, in particular for total survivin and the survivin-Ex3 variant. In addition, survivin-wt appears to be the dominant form(3)(16)(17); the amplicon of survivin 2 would have been of approximately the same size as survivin in the study of Ryan et al.(16) and might therefore have been missed in that study. However, the amplicon of the 3B form should have been visible on the gel because it includes an additional 3B exon. A likely explanation is that Q-PCR is much more sensitive than Q-PCR. In Q-PCR, fluorescence is produced and measured with each amplification round. The amplification round (cycle) at which this fluorescence, produced by the formation of amplicons, is detected above the background (the so-called cycle threshold or Ct value) is used to quantify the amount of mRNA. The more cycles are needed to produce sufficient fluorescence, the lower the amount of starting material. The survivin-3B variant is less common than the other variants, and in our experiments, Ct values of 28–37 were established. It is likely that the 30 cycles of amplification used by both O’Driscoll et al.(3) and Ryan et al.(16) were not enough to yield enough amplicon to be visible on gel. Indeed, Vegran et al.(17) recently was the first to detect survivin-3B in breast cancer, using specific primers for this variant and amplifying for 35 cycles in a Q-PCR. With the Q-PCR technique described in this article, we were able to quantify the concentrations of all 5 variants of survivin in breast cancer.
Our results indicate that total survivin, survivin-Ex3, and survivin 2 mRNA concentrations are lower in tumors from older patients. Others(3)(17) found no significant correlation with age, and Ryan et al. did not report on this(16). The relation with age that we observed may be attributable to the greater likelihood that tumors from older, postmenopausal patients will be positive for steroid hormone receptors. The expression levels of all survivin variants were substantially lower in steroid hormone receptor positive tumors in our study. Earlier studies have not described this in breast cancer(3)(16)(17). There is particular agreement between our study and earlier report that tumors with a ductal morphology have more survivin-Ex3 mRNA than those with a lobular morphology(3)(16). We find this to be true also for total survivin and survivin 2. In the other studies a similar trend is seen for survivin-wt, whereas we are the first to measure survivin 2 in breast cancer. The positive correlation of histological grade with survivin and survivin-Ex3 seems to concur with the functions that have been suggested for these variants, because survivin-Ex3 retains it antiapoptotic function despite the loss of exon 3(10). Tumors with high concentrations of antiapoptotic proteins such as survivin-wt and survivin-Ex3, are more likely to exhibit a high-grade phenotype. Earlier studies(3)(16)(17) did not find any substantial relationship between survivin-wt, survivin-Ex3, or survivin-2B and grade, although Vegran et al.(17) found that tumors with high grade were more likely to be survivin-3B positive. We found that tumors from patients with many involved lymph nodes had lower amounts of the proapoptotic survivin-2B variant in the primary tumor than those with a limited number of involved lymph nodes. This concurs with data from Vegran et al.(17), whereas others found no substantial association with lymph node status(3)(16). This negative correlation with lymph node status would suggest a role for survivin-2B in counteracting the anti-apoptotic and/or cell cycle stimulating role of survivin-wt.
The survivin variants display a differential intracellular localization, possibly regulating their antiapoptotic potential(14)(15). Coexpression experiments indicated that survivin can heterodimerize with its splice variants. Heterodimer formation causes specific subcellular localization patterns, suggesting that high expression in tumor cells leads to formation of functionally distinct survivin complexes(13). In fact, the subcellular localization of the variants differs when cotransfected with survivin, as compared with singly transfected proteins. Coexpression of survivin-wt with survivin-Ex3 results in the recruitment of these complexes to the mitochondria, were they synergistically inhibit mitochondrial dependent apoptosis. This was not described for survivin-2B(13). On the other hand, complex formation between survivin-wt and survivin 2 leads to an attenuation of the antiapoptotic function of survivin and a sensitization to vincristine(12). As such, the data in this study confirm that some of the splice variants interact with each other to modify the prognosis of the patients. Specifically, high expression of the survivin 2 and survivin-3B variants, in addition to survivin, indicate a poorer prognosis, and the prognostic value of survivin is strongest in the presence of higher concentrations of these variants.
In conclusion, we extended our previous data on survivin expression in breast cancer(6) with data on 4 additional splice variants. All variants are present in breast cancer tissue and exhibit particular associations with clinicopathologic characteristics (age, histology, grade, lymph node status, and steroid hormone receptor status) of these tumors. Survival analyses indicate a modulating role of survivin 2 and survivin-3B, but not survivin-Ex3 and survivin-2B(20), on the function of survivin.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: RFS, relapse-free survival; RT-PCR, reverse transcription PCR; Q-PCR, quantitative RT-PCR. 
|
References
|
|---|
- Altieri DC. Validating survivin as a cancer therapeutic target. Nat Rev Cancer 2003;3:46-54.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Velculescu VE, Madden SL, Zhang L, Lash AE, Yu J, Rago C, et al. Analysis of human transcriptomes. Nat Genet 1999;23:387-388.[Web of Science][Medline]
[Order article via Infotrieve]
- O’Driscoll L, Linehan R, Kennedy M S, Cronin D, Purcell R, Glynn S, et al. Lack of prognostic significance of survivin, survivin-Ex3, survivin-2B, galectin-3, bag-1, bax-, and MRP-1 mRNAs in breast cancer. Cancer Lett 2003;201:225-236.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Tanaka K, Iwamoto S, Gon G, Nohara T, Iwamoto M, Tanigawa N. Expression of survivin and its relationship to loss of apoptosis in breast carcinomas. Clin Cancer Res 2000;6:127-134.[Abstract/Free Full Text]
- Kennedy SM, O’Driscoll L, Purcell R, Fitz-Simons N, McDermott EW, Hill AD, et al. Prognostic importance of survivin in breast cancer. Br J Cancer 2003;88:1077-1083.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Span PN, Sweep FC, Wiegerinck ET, Tjan-Heijnen VC, Manders P, Beex LV, de Kok JB. Survivin is an independent prognostic marker for risk stratification of breast cancer patients. Clin Chem 2004;50:1986-1993.[Abstract/Free Full Text]
- Span PN, Tjan-Heijnen VC, Manders P, van Tienoven D, Lehr J, Sweep FC. High survivin predicts a poor response to endocrine therapy, but a good response to chemotherapy in advanced breast cancer. Breast Cancer Res Treat 2006;98:223-230.[CrossRef][Medline]
[Order article via Infotrieve]
- Paik S, Shak S, Tang G, Kim C, Baker J, Cronin M, et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med 2004;351:2817-2826.[Abstract/Free Full Text]
- Ambrosini G, Adida C, Altieri DC. A novel anti-apoptosis gene, survivin, expressed in cancer and lymphoma. Nat Med 1997;3:917-921.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Mahotka C, Wenzel M, Springer E, Gabbert HE, Gerharz CD. Survivin-Ex3 and survivin-2B: two novel splice variants of the apoptosis inhibitor survivin with different antiapoptotic properties. Cancer Res 1999;59:6097-6102.[Abstract/Free Full Text]
- Badran A, Yoshida A, Ishikawa K, Goi T, Yamaguchi A, Ueda T, Inuzuka M. Identification of a novel splice variant of the human anti-apoptopsis gene survivin. Biochem Biophys Res Commun 2004;314:902-907.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Caldas H, Honsey LE, Altura RA. Survivin 2: a novel survivin splice variant expressed in human malignancies. Mol Cancer 2005;4:11.[CrossRef][Medline]
[Order article via Infotrieve]
- Caldas H, Jiang Y, Holloway MP, Fangusaro J, Mahotka C, Conway EM, et al. Survivin splice variants regulate the balance between proliferation and cell death. Oncogene 2005;24:1994-2007.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Mahotka C, Liebmann J, Wenzel M, Suschek CV, Schmitt M, Gabbert HE, et al. Differential subcellular localization of functionally divergent survivin splice variants. Cell Death Differ 2002;9:1334-1342.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Li F, Yang J, Ramnath N, Javle MM, Tan D. Nuclear or cytoplasmic expression of survivin: what is the significance?. Int J Cancer 2005;114:509-512.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Ryan B, O’Donovan N, Browne B, O’Shea C, Crown J, Hill AD, et al. Expression of survivin and its splice variants survivin-2B and survivin-Ex3 in breast cancer. Br J Cancer 2005;92:120-124.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Vegran F, Boidot R, Oudin C, Riedinger JM, Lizard-Nacol S. Distinct expression of survivin splice variants in breast carcinomas. Int J Oncol 2005;27:1151-1157.[Web of Science][Medline]
[Order article via Infotrieve]
- de Kok JB, Roelofs RW, Giesendorf BA, Pennings JL, Waas ET, Feuth T, et al. Normalization of gene expression measurements in tumor tissues: comparison of 13 endogenous control genes. Lab Invest 2005;85:154-159.[Web of Science][Medline]
[Order article via Infotrieve]
- Schwerk C, Schulze-Osthoff K. Regulation of apoptosis by alternative pre-mRNA splicing. Mol Cell 2005;19:1-13.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]
- Noton EA, Colnaghi R, Tate S, Starck C, Carvalho A, Ferrigno PK, Wheatley SP. Molecular analysis of survivin isoforms: evidence that alternatively spliced variants do not play a role in mitosis. J Biol Chem. 2006 Jan 13;281(2):1286-1295.
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
Y. Ma, Y. Qian, L. Wei, J. Abraham, X. Shi, V. Castranova, E. J. Harner, D. C. Flynn, and L. Guo
Population-Based Molecular Prognosis of Breast Cancer by Transcriptional Profiling
Clin. Cancer Res.,
April 1, 2007;
13(7):
2014 - 2022.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
