Survivin, encoded by
Breast cancer (BC) is one of the most common malignancies among women worldwide. Despite generally good prognosis for BC patients, there is a wide variation in survival [
This study included BC patients who had their regular check-up at the Eljuga Polyclinic and agreed to participate. Upon signing the consent forms their blood was drawn. 26 patients, diagnosed between 2000 and 2012 and whose archival paraffin embedded tumour tissue was available from archive at the University Hospital for Tumours, University Hospital Centre “Sestre Milosrdnice,” were included in the study. One patient had a bilateral BC and both tumours were available for testing. Demographic and clinicopathological data on patients was collected from medical records. Demographic characteristics included age of onset and time since operation, and clinicopathological characteristics included histological grade, tumour size, oestrogen and progesterone receptor status, Her2 status, lymph node status, type, and Ki67 status (determined routinely as part of the diagnostic procedure in the clinic). Two 5-
Survivin protein was analysed as follows: slides were deparaffinized, prewarmed in Epitope Retrieval Solution (Dako, Glostrup, Denmark) and kept at 95-99°C for 20 min, left to cool for additional 20 min, and then washed in PBS 3x5 min. Endogenous peroxidase was blocked with 3 % H2O2/methanol for 10 min and then washed 3x5 min in PBS. The sample was circled by PAP-PEN (Kiyota, Baltimore, MD). Protein block serum-free (DAKO) was added for 10 min.
Immunohistochemical staining was performed using monoclonal rabbit anti-survivin antibody (1:200, 71G4B7, Cell Signalling Technology, Danvers, MA). Following an overnight incubation, the immunodetection was completed using the LSAB Visualization System (DAKO) utilizing 3, 3-diaminobenzidine (DAB) chromogen as substrate, according to the manufacturer’s instructions. All sections were counterstained with haematoxylin (DAKO). Negative controls were obtained by omitting the primary antibody.
Stained slides were examined by two experienced pathologists. Samples were scored between 1 and 3 based on survivin expression, taking into account the percentage of positive cells and the staining intensity, as previously described by Xu et al. [
Whole coding region of the
Genotype and allele frequencies of patients and controls were compared using
A total of 26 patients with BC were recruited in this study. The main clinical and histopathological features are summarized in Table
Patient clinicopathological characteristics.
Characteristic | No. of patients (%) |
---|---|
Age at diagnosis (years) | |
| |
< 40 | 1 (3.8) |
40-50 | 9 (34.6) |
50-60 | 8 (30.8) |
> 60 | 8 (30.8) |
| |
Years since operation | |
| |
5 | 8 (30.8) |
10 | 15 (57.7) |
15 | 3 (11.5) |
| |
Histological grade | |
| |
1 | 4 (14.8) |
2 | 12 (44.4) |
3 | 10 (37.0) |
N.A | 1 (3.7) |
| |
Tumour size | |
| |
1 (≤ 2 cm) | 20 (74.1) |
2 (> 2 ≤ 5 cm) | 6 (22.2) |
3 (> 5 cm) | 1 (3.7) |
| |
Oestrogen receptor status | |
| |
Positive | 21 (77.8) |
Negative | 6 (22.2) |
| |
Progesterone receptor status | |
| |
Positive | 16 (59.3) |
Negative | 11 (40.7 |
| |
Her2 status | |
| |
Positive | 2 (7.4) |
Negative | 22 (81.5) |
N.A. | 3 (11.1) |
| |
Lymph node status | |
| |
Positive | 11 (40.7) |
Negative | 16 (59.3) |
| |
Pathologic Characteristics | |
| |
Invasive ductal carcinoma | 22 (81.5) |
Medullary carcinoma | 2 (7.4) |
Invasive lobular carcinoma | 1 (3.7) |
Mixed invasive ductal/lobular carcinoma | 1 (3.7) |
Papillary carcinoma | 1 (3.7) |
| |
Ki67 status | |
| |
Positive | 12 (44.4) |
Negative | 12 (44.4) |
N.A. | 3 (11.1) |
Tumour size was associated with positive lymph nodes (p=0.004). Histological grade was associated with Ki67 expression (p=0.0009). All samples with histological grade 1 were negative for Ki67, while only one sample with histological grade 3 was negative for Ki67. Progesterone receptor (PR) and oestrogen receptor (ER) status were weakly positively associated (0.053). These results are in agreement with previously published international data [
Survivin expression was analysed in 27 BC samples from 26 patients using immunohistochemical staining. Samples were scored between 1 and 3 based on survivin expression, taking into account the percentage of positive cells and the staining intensity (Figure
Levels and localization of survivin expression.
Characteristic | No. of samples (%) |
---|---|
Survivin immunoreactivity score | |
| |
weak | 9 (33.3) |
moderate | 11 (40.7) |
strong | 7 (25.9) |
| |
Survivin cellular localization | |
| |
cytoplasmic | 1 (3.7) |
nuclear | 25 (92.6) |
cytoplasmic+nuclear | 1 (3.7) |
Examples of scored staining intensities: (a) negative control, (b) score 1, (c) score 2, and (d) score 3.
The high nuclear staining of survivin is different than most published data, where the nuclear staining is mostly lower than cytoplasmic [
Clinicopathological findings (age at diagnosis, time since operation, histological grade, tumour size, oestrogen and progesterone receptors, Her2 status, lymph node status, and Ki67 status) were compared with the immunohistochemical characterization of survivin expression. Since the majority of samples (22 of 27 samples, 84.6%) were ductal BC all samples were grouped together. The statistical analyses revealed a significant association between survivin expression and ER status. High survivin expression was significantly associated with negative ER status (p=0.007). No samples with low survivin expression (score 1), 18.2% samples with medium expression (score 2), and 57.1% of samples with high survivin expression (score 3) had negative ER status (Figure
Association between survivin expression and ER status. Survivin expression was ranked weak, moderate, and strong, and percentage of ER negative and ER positive samples per group is shown.
There is controversial data reported about association between high survivin expression and negative ER and/or PR status in BC. While some studies [
Fifteen different polymorphisms were found in the constitutional DNA of 26 BC patients and 74 healthy controls.
The whole
Comparison of minor allele frequencies of
gene region | SNP ID number | nucleotide change | minor allele frequency controls (n/N, %) | minor allele frequency BC samples (n/N, %) | p (for allele frequencies) |
---|---|---|---|---|---|
promoter | rs3764383 | c.-1547C>T | 37/148 (25.0) | 16/52 (30.8) | 0.466 |
| |||||
promoter | rs8073903 | c.-644T>C | 49/148 (33.1) | 18/52 (34.6) | 0.865 |
| |||||
promoter | rs8073069 | c.-625G>C | 33/148 (22.3) | 13/52 (25.0) | 0.704 |
| |||||
promoter | rs17878731 | c.-267G>A | 1/148 (0.7) | 1/52 (1.9) | 0.453 |
| |||||
promoter | rs17878467 | c.-241C>T | 16/148 (10.8) | 6/52 (11.5) | 1.000 |
| |||||
promoter | rs17887126 | c.-235G>A | 2/148 (1.3) | 0/52 (0) | |
| |||||
5'UTR | rs9904341 | c.-31G>C | 55/148 (37.2) | 18/52 (34.6) | 0.867 |
| |||||
intron 2 | rs4789551 | c.221+209T>C | 7/148 (4.7) | 1/52 (1.9) | 0.683 |
| |||||
exon 4 | rs2071214 | c.9194G>A | 5/148 (3.4) | 1/52 (1.9) | 1.000 |
| |||||
3'UTR | rs17885521 | c.9288G>C | 3/148 (2.0) | 1/52 (1.9) | 1.000 |
| |||||
3'UTR | rs17882627 | c.9342G>A | 2/148 (1.3) | 2/52 (3.8) | 0.278 |
| |||||
3'UTR | rs2239680 | c.9386T>C | 34/148 (23.0) | 14/52 (26.9) | 0.575 |
| |||||
3'UTR | rs17882139 | c.9387_9388insAA | 3/148 (2.0) | 2/52 (3.8) | 0.606 |
| |||||
3'UTR | rs1042489 | c.9809T>C | 53/148 (35.8) | 18/52 (34.6) | 1.000 |
| |||||
3'UTR | rs2661694 | c.10611C>A | 38/148 (25.7) | 11/52 (21.1) | 0.578 |
According to the TRANSFAC database (
Structure of
Fourteen different polymorphisms were found in BC samples while in controls one more polymorphism was present (c.-235G>A). The majority of detected polymorphisms were located either upstream or downstream of the coding region of the gene, and only one in the coding region. In addition to ten polymorphisms selected for this study, five more rare polymorphisms were found: one in the promoter (c.-267G>A), one in intron 2 (c.221+209T>C), and 3 in the 3'UTR (c. 9288G>C, c.9342G>A, and c.9387_9388insAA). There was no significant difference in genotype or allele frequencies between BC and control samples.
The observed allele frequencies for most of the polymorphisms are in accordance with previously observed frequencies for either global or European populations [
Analysis showed there is a difference in number of polymorphisms in linkage disequilibrium (LD) between BC samples and controls. BC cases showed increased nonrandom association of alleles at multiple loci. In case of c.-644C>T and c.9809T>C, these two polymorphisms always appear in the same combination in all tested BC samples showing complete LD (r2=0.99) (Figure
Pairwise linkage disequilibrium (LD) of eight
LD between various
Association of genotype and allele frequencies of polymorphisms with age of onset.
polymorphism | mean±SD (years) | p-value | ||
---|---|---|---|---|
c.-1547C>T | genotype | CC | 62.3±6.5 | 0.054 |
CT | 55.8±8.1 | |||
TT | 49.0±10.1 | |||
allele | C | 58.3±7.9 | | |
T | 50.9±9.8 | |||
| ||||
c.-644T>C | genotype | TT | 59.4±8.2 | |
TC | 50.7±8.6 | |||
CC | 39.0±2.8 | |||
allele | T | 55.8±9.2 | | |
C | 48.1±9.1 | |||
| ||||
c.-241C>T | genotype | CC | 54.7±10.0 | 0.143 |
CT | 49.0±4.5 | |||
TT | 37.0±0.0 | |||
allele | C | 54.2±9.6 | | |
T | 45.0±7.1 | |||
| ||||
c.9386T>C | genotype | TT | 49.1±9.3 | |
TC | 57.4±8.4 | |||
CC | 62.3±6.5 | |||
allele | T | 50.8±9.6 | | |
C | 59.5±7.6 | |||
| ||||
c.9809T>C | genotype | TT | 59.4±8.2 | |
TC | 50.7±8.6 | |||
CC | 39.0±2.8 | |||
allele | T | 55.8±9.2 | | |
C | 48.1±9.1 |
Association of age of onset with the significantly associated polymorphisms. (a) c.-1547C>T. (b) c.-644T>C. (c) c.-241C>T. (d) c.9386T>C. (e) c.9809T>C.
For the c.-1547C>T polymorphism, the minor allele T was associated with earlier age of onset. The same trend is visible for the c.-644C>T allele and the c.9089C>T allele, where in both cases the minor C allele is associated with earlier age of onset. For the c.9386T>C allele, the major allele T is associated with earlier age of onset. For the final, c.-241C>T polymorphism, where the difference was significant only on allele but not on genotype frequencies, the minor allele T also seems to be associated with earlier age of onset but the number of samples in the TT group is too low to make this comparison.
Until now, only two studies showed association between
This was the first study investigating the possible role of
This is not applicable, since there are no big datasets connected with this article.
Results published in this paper were presented as poster at HDIR-5: The Fifth Meeting of the Croatian Association for Cancer Research with International Participation, Zagreb in November, 2018.
The authors declare that there are no conflicts of interest regarding the publication of this paper.
Ilona Sušac and Petar Ozretić are equally contributing authors.
This study was funded and supported by Terry Fox Foundation (Terry Fox Run Croatia 2012 and 2016) and Foundation of the Croatian Academy of Sciences and Arts (10-102/168-3-2015). The authors are grateful to Croatian League Against Cancer and the Terry Fox Foundation for their administrative and financial support. They are also extremely grateful to all participating patients, without whose help and willingness to participate this study would not be possible.
Supplementary Table