Abstract
Background: Cyclin A is a cell-cycle regulatory gene and its overexpression promotes tumor cell growth. Y-Box-binding protein 1 (YB1) is a transcription/translation factor involved in tumor growth, invasion, and drug resistance. We investigated whether an association exists between protein products of these genes in epithelial ovarian cancer (EOC) specimens and clinicopathological parameters, patient response and EOC sensitivity to platinum-based first-line chemotherapy. Patients and Methods: Cyclin A and YB1 expression were analyzed by immunohistochemistry in 54 human primary EOC tissues. Immunolabeling of both proteins was graded according to their staining intensity (scale 0-3) and the proportion of immunostained cancer cells (scale 0-4) to obtain a staining index (SI; value=0-12). Results: Significantly higher cyclin A immunostaining (SI ≥4) in EOC specimens was discovered in patients with advanced (International Federation of Gynaecology and Obstetrics (FIGO) III and IV, p=0.003), poorly differentiated (G3, p<0.001) tumors, and tumors of those with residual disease >1 cm (p=0.001). YB1 immunostaining was significantly higher in EOCs from patients with suboptimal debulking (p=0.025). Over-expression of cyclin A (SI ≥9) in EOCs was significantly linked with poorer patient response (p=0.001) and higher resistance of tumors to platinum-based first-line chemotherapy (p=0.007), while immunolabeling of YB1 in EOCs was not significantly associated with either of these variables (p>0.05). Cyclin A expression was significantly and positively correlated with that of YB1 (R=0.588, p<0.001). Conclusion: Increased cyclin A expression in EOC is related to a more aggressive tumor behavior and predicts the response of patients to first-line platinum-based chemotherapy.
Ovarian cancer is the second most common malignant tumor of the female genital tract, with an age-adjusted incidence of approximately 11 per 100,000 women in Poland in 2011 (3,527 new cases). The incidence of ovarian cancer is rising, and despite diagnostic and therapeutic efforts, ovarian cancer-related mortality rose from 6.7 per 100,000 in 1999 (1959 deaths) to 6.9 per 100,000 women in 2011 (2558 deaths) in Poland (1). Moreover, more than 60% of women are diagnosed with disease at an advanced stage due to lack of early symptoms and effective screening tools, and for women diagnosed with advanced disease, the 5-year relative survival is merely 27.4% (2). Standard therapy includes cytoreductive surgery and combination chemotherapy regimens, however, many patients develop resistance to chemotherapy and die of disease recurrence (3, 4). Therefore, there is an urgent need to develop effective prognostic biomarkers which could accurately identify those patients who are at increased risk of a more aggressive disease progression and treatment failure, and who would require patient-tailored treatment regimens.
Cyclin A belongs to a group of regulatory proteins that are sequentially expressed and degraded at specific times during the cell cycle. Cyclins combine with cyclin-dependant kinases to form active catalytic units which regulate cell proliferation via cell-cycle transition points (5). Cyclin A regulates G1/S and G2/M phase transition and is required for DNA replication during the S phase, and it is therefore one of the most useful markers signifying proliferating cells (6, 7). Increased expression of cyclin A and dysregulation of cell-cycle checkpoints promote proliferation of cancer cells, which can be facilitated by phosphorylation of oncoproteins and tumor suppressors (7). Overexpression of cyclin A has been discovered in human cancer, including ovarian tumors (8, 9), and it was reported that cyclin A may be used as a biomarker to predict the response to chemotherapy of patients with head and neck squamous cell cancer (HNSCC) (10), soft tissue sarcoma (11), esophageal squamous cell carcinoma (ESCC) (12), acute leukemia (13), and hepatocellular carcinoma (14).
Cyclin A and YB1 immunostaining in epithelial ovarian cancer tissues related to clinicopathological variables.
Y-Box-binding protein 1 (YB1) is a transcription\translation factor involved in tumor growth, invasion, metastasis, DNA repair, and drug resistance (15-17). YB1 enhances cell proliferation by up-regulation of expression of epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor 2 (HER2), as well as of cell cycle-related genes: cyclin A, cyclin B1, cyclin D1, and cell division cycle 6 (CDC6) (15, 16). Increased YB1 expression has been found to be associated with the progression of many human malignancies, including female breast, ovarian, and cervical cancer (15, 17, 18). YB1 was reported to activate genes involved in the development of drug resistance (e.g. ATP-binding cassette sub-family B member 1 and major vault protein) (15-17). In vitro studies showed that YB1 mediates the resistance of ovarian cancer cell lines to cisplatin (19) and paclitaxel (20).
Since to our knowledge, there exist no clinical studies evaluating the usefulness of cyclin A or YB1 as biomarkers of drug resistance in ovarian cancer, the aim of our study was to evaluate cyclin A and YB1 immunostaining in primary epithelial ovarian cancer (EOC) tissues from previously untreated patients, and to investigate whether an association exists in these tumors between expression of cyclin A, YB1, clinicopathological parameters and patient response or EOC sensitivity to first-line chemotherapy.
Patients and Methods
Patients' characteristics. The study group consisted of 54 women with primary EOC, diagnosed and treated from 2004 to 2007 at the First Department of Oncologic Gynecology and Gynecology, Medical University of Lublin, and the Third Department of Gynecologic Oncology, St. John's Cancer Center Lublin. The median patient age was 56.5 years (range=23-85 years). All specimens underwent microscopic confirmation of histological type, and tumor grade by an experienced pathologist (B.J.), following established criteria (21). The optimal debulking status after initial surgery was defined as residual tumor with maximal diameter of 1 cm or less (22). Tumor specimens were collected from women during primary surgery and prior to the initiation of adjuvant therapy. After primary surgery, 41 patients (75.9%) received adjuvant chemotherapy. Paclitaxel-platinum chemotherapy (consisting of paclitaxel and carboplatin or cisplatin) was administered to 32 women (78.0%), and nine (22.0%) received cyclophosphamide-cisplatin. The patient response to chemotherapy was assessed according to RECIST criteria ver. 1.1 (23) and platinum sensitivity of tumors was determined as described previously (22). Ethical approval to carry out the study was granted by the Research Ethics Committee of Medical University of Lublin (KE-0254/137/2007).
Cyclin A and YB1 immunostaining in epithelial ovarian cancer (EOC) tissues as related to patients' response or EOCs' sensitivity to platinum-based first line chemotherapy.
Immunohistochemistry. Tissue sections (4 μm) from formalin-fixed and paraffin-embedded ovarian samples were de-paraffinized, rehydrated and subjected to heat-induced antigen retrieval in DakoCytomation Target Retrieval Solution, citrate, pH 6.0 (DAKO, Glostrup, Denmark) at 99°C for 20 min in water bath. Endo genous peroxidase was blocked using Dako REAL™ Peroxidase-Blocking Solution (DAKO) for 10 min, and then slides were incubated for 30 min in Protein Block Serum-Free (DAKO) to inhibit non-specific staining. Afterwards, the sections were incubated with the primary antibody for 30 min. The following antibodies were purchased from Santa Cruz Biotechnology (Dallas, TX, USA): clone 6E6 mouse monoclonal antibody to cyclin A (1:200) and clone 59-Q mouse monoclonal antibody to YB1 (1:100). After washing, the specifically bound antibodies were detected with the sensitive polymer-based system REAL™ EnVision™ Detection System, Peroxidase/DAB+, Rabbit/ Mouse (DAKO) according to the manufacturer's instructions. Finally, the slides were counterstained in hematoxylin and coverslipped. Breast cancer specimens were used as positive controls and a positive control section was included in each staining run. Negative controls were performed by omitting the primary antibody and no staining was detected in negative control sections.
Assessment of cyclin A and YB1 immunostaining. A pathologist (B.J.) and researcher (M.C.) independently evaluated immunohistochemically-stained slides without knowledge of the clinicopathological data of the patients, and in cases of disagreement, the slides were reviewed simultaneously, and a final agreement was made. Immunolabeling of cyclin A and YB1 was graded according to the staining intensity of cancer cells as follows: negative=0, weak=1, moderate=2, or strong=3. Moreover, the proportion of immunostained cancer cells was recorded as follows: 0%=0, 1-25%=1, 26-50%=2, 51-75%=3, or 76-100%=4. Thus, a staining index (SI) of cyclin A and YB1 (value=0-12) was obtained by multiplying the grade for staining intensity with that for the staining proportion.
Statistical analysis. The Fisher exact or Pearson's χ2 test was used for statistical analysis of the relationships between protein expressions and clinicopathological variables. Correlations between continuous variables were determined using Spearman's rank correlation test. The results with a two-sided p-value of less than 0.05 were considered statistically significant. Statistical analyses were performed using SPSS software version 14.0 (SPSS Inc., Chicago, IL, USA).
Results
Cyclin A and YB1 immunohistochemistry. Cyclin A and YB1 immunostaining was detected in all EOC specimens. Cyclin A was present predominantly in the nuclei of cancer cells (Figure 1A-C). In contrast to cyclin A, YB1 immunreactivity was detected mainly in the cytoplasm and rarely in the nuclei of cancer cells, and in the cytoplasm of stromal cells (Figure 1D-F). High cyclin A expression (SI ≥4) was found in 37 (68.5%) EOCs, whereas 20 (37.0%) tumors with SI ≥9 were considered as cyclin A overexpressing carcinomas. Forty-eight (88.9%) EOCs had high YB1 immunoreactivity (SI ≥4).
The association between expression of cyclin A, YB1 and clinicopathological characteristics. Cyclin A immunostaining in EOC specimens significantly increased with advancing International Federation of Gynaecology and Obstetrics (FIGO) stage, increasing tumor grade, and the size of residual disease (Table I). High cyclin A immunoreactivity (SI ≥4) was more common in FIGO III and IV tumors than in FIGO I and II tumors (p=0.003). Additionally, cyclin A immunolabeling was higher in poorly differentiated tumors (G3) and in EOCs of patients with residual disease >1 cm, compared to well- and moderately differentiated (G1 and G2) tumors (p<0.001), and EOCs of patients with residual disease ≤1 cm (p=0.001), respectively. YB1 immunostaining was significantly higher in EOCs from patients with suboptimal debulking (p=0.025), and tended to be higher in poorly-differentiated carcinomas (p=0.071). Among all histological subtypes, YB1 expression was the lowest (p=0.021) in mucinous tumors (Table I). Spearman correlation analysis revealed that cyclin A immunolabeling was directly and significantly correlated with YB1 expression (R=0.588, p<0.001, data not shown).
Cyclin A (A-C) and YB1 (D-F) immunostaining in epithelial ovarian cancer. A: Serous ovarian adenocarcinoma (G2) with strong expression of cyclin A in the nuclei of tumor cells (×600). B: Endometrioid ovarian adenocarcinoma (G2) with moderate expression of cyclin A in the nuclei of tumor cells (×400). C: Mucinous ovarian adenocarcinoma (G2) with weak expression of cyclin A in the nucleus of one tumor cell (×600). D: Serous ovarian adenocarcinoma (G2) with strong expression of YB1 in the cytoplasm and nuclei (upper-left part of the panel) of tumor cells (×600). E: Endometrioid ovarian adenocarcinoma (G2) with moderate expression of YB1 in the cytoplasm of tumor cells (×600). F: Mucinous ovarian adenocarcinoma (G2) with weak expression of YB1 in the cytoplasm of tumor cells. Cytoplasmic YB1 immunostaining of stromal cells (fibroblasts) is visible in the upper part of the panel (×600).
The relationship between expression of cyclin A and YB1, and patient response and EOC sensitivity to platinum-based first-line chemotherapy. Overexpression of cyclin A (SI ≥9) was significantly linked to a poorer patient response (p=0.001) and higher resistance in EOCs to platinum-based first-line chemotherapy (p=0.007). Immunolabeling of YB1 was not significantly (p>0.05) associated with either variable (Table II).
Discussion
The strategy of ovarian cancer treatment is still a challenge, and accurate assessment of treatment efficacy may help in selection of patients who will benefit from targeted therapy. We show that higher cyclin A immunostaining was significantly associated with advanced stage and poor differentiation of EOCs. Other studies have demonstrated that cyclin A protein was significantly overexpressed in malignant ovarian tumors compared with benign and borderline tumors (8, 9). Higher cyclin A immunostaining has been reported in advanced tumors: endometrial adenocarcinomas (24), HNSCCs (25, 26), gastric carcinomas (27), and ESCCs (28), and was directly correlated with histological grade in numerous tumor types, including female breast cancer (29) and endometrial adenocarcinoma (24). Importantly, higher immunolabeling of cyclin A was detected in invasive cancer: non-curative gastric adenocarcinoma (27), invasive transitional cell carcinoma of the renal pelvis and ureter (30), and invasive hepatocellular cancer (31). Accumulation of cyclin A in poorly differentiated, advanced, and invasive tumors supports the assumption that elevated expression of cyclin A provides a proliferative advantage to cancer cells, which leads to a more aggressive spread of EOCs.
Our study revealed that elevated cyclin A protein expression in EOCs was significantly associated with suboptimal debulking, similarly to the results reported by Yoon et al. (9). We are the first to demonstrate that cyclin A overexpression in EOCs was significantly associated with poorer patient response and with resistance of tumors to platinum-based first-line chemotherapy. Additionally, in our study, optimal debulking (residual disease <1 cm) was significantly associated with complete response and higher sensitivity of EOCs to platinum-based chemotherapy (p<0.001, data not shown). Our findings are in agreement with the results of Eisenhauer et al. where patients with ovarian cancer with optimal debulking had improved response to platinum-taxane chemotherapy and higher platinum sensitivity (22). It is plausible that increased pro liferation of ovarian cancer cells, as a result of cyclin A overexpression, may result in reduced effects of chemotherapy. However, this hypothesis requires further investigation. Recently, higher cyclin A immunoreactivity and gene amplification in hepatocellular carcinoma were linked to worse response to chemotherapy consisting of 5-fluouracil and leucovorin (14). On the other hand, cyclin A immunostaining did not predict chemotherapy response in women with metastatic breast cancer treated with anthracycline, methotrexate-fluorouracil or docetaxel (29). Conversely, other immunohistochemical studies showed that higher expression of cyclin A predicted a better response to therapy of patients with advanced HNSCC when treated with cisplatin, 5-fluouracil and radiotherapy (10), as well as patients with soft tissue sarcomas treated with ifosfamide, doxorubicin, dacarbazine, and mesna (11), and patients with ESCC treated with paclitaxel-cisplatin chemotherapy (12). Furthermore, flow cytometric analysis revealed higher cyclin A expression in leukemia cells of patients with acute leukemias who responded to chemotherapy (13). These results suggest that the predictive value of cyclin A as a biomarker of patient response to chemotherapy depends on the histological type of neoplasm and chemotherapy regimen.
In our study, YB1 immunostaining was significantly higher in EOCs from patients with suboptimal debulking, and tended to be higher in poorly differentiated tumors. Similarly, YB1 expression has been shown to correlate with invasion of cervical (18) and bladder cancer (32), and histological grade of breast cancer (33), HNSCC (34), and non-small-cell lung cancer (35). Experimental evidence has been provided that YB1 promotes growth, invasion, and motility of bladder cancer cells (36), migration of gastric cancer cells (37), and proliferation of osteosarcoma cells in vitro (38). Enforced expression of YB1 in non-invasive pre-malignant breast epithelial cells (MCF10AT) induced an epithelial-mesenchymal transition accompanied by enhanced invasion and metastatic potential in a mouse xenograft model (39). We discovered low YB1 expression in mucinous EOCs, which may be explained by the low percentage of G3 tumors in this group (mucinous: 22.2%, serous and endometrioid: 50.0%).
In our study, YB1 immunostaining was not associated with patient response or EOC sensitivity to platinum-based chemotherapy. Higher nuclear YB1 protein levels, detected by western blot using rabbit polyclonal antibody, were reported in a cisplatin-resistant human ovarian serous cystade nocarcinoma cell line than in sensitive parental cells (19). Silencing of YB1 with siRNA increased sensitivity of taxane-resistant EOC cells (HeyA8-MDR) to paclitaxel (20). In our study, we detected mainly cytoplasmic YB1 protein immuno staining with mouse monoclonal antibody raised against recombinant YB1 of human origin. Recently, it was reported that different antibodies to YB1 exhibit different immuno histochemical staining patterns that are determined by the accessibility of YB1 epitopes, thus YB1 immuno staining depends on the nature of the complexes containing YB1 and this may lead to different findings about the prognostic value of YB1 (33).
It was discovered that YB1 up-regulates mRNA and protein expression of cyclin A in HBL100 cells (40). However, silencing of YB1 with siRNA in human osteosarcoma cell lines (MG63 and MNNG) decreased protein expression of cyclin A, but not its mRNA level, suggesting that YB1 may post-transcriptionally regulate cyclin A expression (38). On the other hand, YB1 silencing in lung (A549) and breast (MCF-7) cancer cell lines did not affect cyclin A protein levels (41). Interestingly, ectopic expression of YB1 in breast epithelial cells (MCF10AT) led to reduced expression of cyclin A, and decreased proliferation (39). In the current work, immuno staining of YB1 and cyclin A in EOCs was significantly and directly correlated which suggests that YB1 may play a role in the control of ovarian cancer cell proliferation via the up-regulation of cyclin A expression. However, this hypothesis needs to be further investigated through in vitro analysis.
In conclusion, we demonstrate that higher cyclin A expression is related to a more aggressive behavior of EOCs, and the assessment of cyclin A immunostaining may provide important information for predicting the response of patients with EOCs to first-line platinum-based chemotherapy.
Acknowledgements
This study was supported by the grants from the Ministry of Science and Higher Education of Poland (no. N407 092 32/3452) to MC and the Medical University of Lublin, Poland to AS (no. DS 440).
- Received October 14, 2014.
- Revision received December 4, 2014.
- Accepted December 10, 2014.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
