Cervical cancer is the third most common malignancy in women worldwide and is often diagnosed at an advanced clinical stage.1 Despite recent advances in cancer treatments, radiotherapy remains the most common therapeutic intervention for cervical cancer as a preoperative adjuvant therapy or a primary treatment. The National Comprehensive Cancer Network (NCCN) 2011 principles of radiation for cervical cancer noted that brachytherapy is a critical component of therapy for all patients with intact cervical cancer. In China, preoperative adjuvant brachytherapy followed by radical hysterectomy and pelvic lymph node dissection after approximately 2 weeks is frequently used for locally bulky cervical cancer (International Federation of Gynecology and Obstetrics [FIGO] stage IB2-IIA2) patients.
Over the last 2 decades, the translation initiation complex EIF4F (eukaryotic initiation factor 4F), which up-regulated protein synthesis, has been shown to play important roles in oncogenesis.2–4 The EIF4F complex is a trimeric complex composed of the adenosine triphosphate–dependent RNA helicase EIF4A1, 5′ cap messenger RNA (mRNA)–binding protein EIF4E, and scaffolding protein EIF4G1. EIF4A1, EIF4E, and EIF4G1 are overexpressed in various cancers, including primary hepatocellular carcinomas,5 melanoma,6 nasopharyngeal cancer (NPC), lymphomas,7 and cancers of the breast,8 lung,9 head and neck,10 esophagus,11 skin,12 bladder,13 colon,14 and cervix.15 Recently, targeting EIF4A1 to reverse lymphoma cancer cell resistance to chemotherapy16 and the activity of EIF4E for enhancing the radiosensitivity of cancer cell lines17 has been shown. Significant correlations between the expression of EIF4E and EIF4G1 and cancer patient prognosis have been reported.14,17,18 It has been demonstrated that EIF4A1, EIF4E, and EIF4G1 play important roles in cancer pathogenesis, progress, and treatment, but there are few studies focused on these proteins in cervical cancer. In the present study, we determined the expression levels and roles of these proteins in cervical cancer, particularly in the prognostic assessment for patients with preoperative brachytherapy.
MATERIALS AND METHODS
Patients, Samples, and Follow-up
Archival tissues of the 137 patients with invasive cervical cancer and 35 patients with benign gynecologic disease diagnosed between January 2002 and December 2005 were obtained from the Fudan University Shanghai Cancer Center. The patients were staged according to the FIGO staging criteria. All of the cervical cancer patients underwent radical hysterectomy and bilateral pelvic lymphadenectomy, and 50 patients with FIGO IB2 to ΠA2 disease received high-dose-rate brachytherapy 2 to 3 weeks before the operation, with a total dose of 1500 cGy to point A in 3 fractions. For these 50 patients, paired specimens before and after the brachytherapy were assembled. The protocol was approved by the institutional review board of the Fudan University Shanghai Cancer Center, and all of the patients gave informed consent. The disease-specific survival was recorded from the date of diagnosis to the time of the last follow-up or cancer-related death. The patients were examined every 3 months for 2 years and every 6 months thereafter.
Tissue microarrays (TMAs) were constructed with 35 normal cervix samples obtained from patients with benign gynecologic disease, 87 cervical cancers treated without preoperative therapy, and as 50 pairs of cervical cancer samples collected before and after preoperative brachytherapy. The procedure for the construction of TMAs was as previously described.19 Briefly, hematoxylin and eosin stained slides were reviewed under a microscope, and the cores were punched using a tissue-arraying instrument (Beecher Instruments, Silver Spring, MD) from selected paraffin blocks. A 1.0-mm-diameter cylinder of tissue was removed and reembedded into a predetermined position in a recipient paraffin block. For every sample, 2 separate cores were selected. Multiple sections (4-μm-thick) were cut from the TMA block and mounted on microscope slides.
A rabbit polyclonal antibody against EIF4A1 was obtained from Epitomics, Inc (Burlingame, CA). Rabbit antibodies against EIF4E and EIF4G1 were purchased from Cell Signaling Technology (Danvers, MA). An EnVision + System peroxidase kit (Gene Tech Biotechnology Company Ltd, Shanghai, China) was used to evaluate the expression of the target proteins. The TMAs were dewaxed and dehydrated before being boiled in an autoclave with citrate buffer (pH 6.0) for 10 minutes. Nonspecific binding was blocked with 10% normal bovine serum for 30 minutes. The TMA slides were incubated with the first antibody overnight at 4°C. EIF4A1, EIF4E, and EIF4G1 were diluted to 1:50, 1:100, 1:40, respectively, according to the manufacturer’s protocol and our results of titrated concentrations. A goat anti–rabbit antibody was applied as a secondary antibody for 60 minutes at 37°C. Finally, the TMAs were treated with 3,3′-diaminobenzidine. The negative control sections were subjected to the same procedure except the first antibody was replaced with a phosphate buffer. No significant staining was observed in the negative controls.
All of the staining results were evaluated by 2 pathologists, without any knowledge of the diagnosis, prognosis, and the other observer’s scores. For each tissue sample, the intensity and the proportion of stained tumor cells were recorded. The staining intensity was estimated on a 4-step scale (negative, 0; weak, 1; moderate, 2; or strong, 3 scores). The tumors were categorized into 4 groups according to the proportion of positive stained cells; these are as follows: score 0 (no staining or <5%), score 1 (>5% or <25%), score 2 (>25% or <50%), and score 3 (>50%). The sum of the staining intensity and proportion scores (0–6) was used as the final staining score. For the final score of each patient, the average score was used if the scores of the 2 duplicated cores agreed with each other; the higher score was used if there was a discrepancy. For the statistical analysis, final staining scores of 0, 1–2, 3–4, and 5–6 were considered to be negative, low, medium, and high expression levels, respectively.
Tumor Response to Brachytherapy
The brachytherapeutic response was assessed 2 weeks after the last course of treatment according to the World Health Organization criteria20 as follows: complete response (CR), complete resolution of the tumor; partial response (PR), more than 50% decreased in the tumor volume; stable disease (SD), less than 50% decreased or a less than 25% increase in the tumor volume; and progressive disease (PD), more than 25% increase in the tumor volume. The patients with a CR or PR were defined as the response group, and the patients with SD or PD were deemed as the nonresponse group.
SPSS software for PC (version 19.0 for Windows) was used for the statistical analysis. The χ2 test or Fisher exact test were applied to assess the statistical significance of the association between the expression of EIF4A1, EIF4E, and EIF4G1 and clinicopathologic variables. The Kaplan-Meier method was used to evaluate the disease-specific survival. A multivariate Cox regression analysis with a backward logistic regression method was applied for all parameters of significance in the univariate analysis. All of the tests were 2-sided, and the threshold for statistical significance was P < 0.05.
The median follow-up time for the 137 invasive cervical cancer patients was 59 months (range, 2–95 months), and cancer-related deaths were identified in 35 patients. The patterns of EIF4A1, EIF4E, and EIF4G1 staining were predominantly cytoplasmic (Fig. 1). These proteins were positively expressed in 131of 137, 131 of 137, and 132 of 137 cervical cancer tissues, respectively. Low expression levels of EIF4A1, EIF4E, and EIF4G1 were observed in 8, 7, and 10 of the 35 normal cervical tissues, respectively. We redesignated the staining index of 0 to 2 as normal expression, 3 to 4 as weakly positive, and 5 to 6 as strongly positive expression. According to this designation, overexpression of EIF4A1, EIF4E, and EIF4G1 was observed in 115 (83.9%) of 137, 116 (84.7%) of 137, and 110 (80.3%) of 137 cervical cancer tissues, respectively, and in none of the normal cervical tissues (P < 0.001).
Expression Levels of EIFs and Clinicopathological Variables
The correlations between the expression of EIF4A1, EIF4E, and EIF4G1 and clinicopathological characteristics are shown in Table 1. EIF4A1, EIF4E, and EIF4G1 were increasingly expressed with increasing FIGO stages (P < 0.001, 0.001, and 0.001, respectively). For squamous carcinomas, these were expressed at strongly positive levels in 64.2%, 62.1%, and 65.3% of patients, respectively, whereas they were expressed as strongly positive in only 16.7%, 14.3%, and 16.7% of nonsquamous carcinomas (P < 0.001). They were well associated with lymph node metastasis status (P = 0.004, 0.003, and 0.006, respectively). For patients with positive pelvic lymph nodes, EIF4A1, EIF4E, and EIF4G1 showed strongly positive expression at a ratio of 73.1%, 73.1%, and 65.4%, respectively. In patients with negative pelvic lymph nodes, the proteins showed strongly positive expression in 44.1%, 41.4%, and 46.8%.
For the 87 patients without preoperative brachytherapy, the expression levels of EIF4A1, EIF4E, and EIF4G1 were closely related to parametrial invasion (P = 0.03, 0.001, and 0.001, respectively) and the depths of stromal invasion (P = 0.008 and 0.036, respectively). No obvious differences were observed about tumor size or histopathologic grades (P > 0.05). In regard to preoperative brachytherapy, the expression of EIF4A1, EIF4E, and EIF4G1 after brachytherapy presented no significance with the abovementioned clinicopathological characteristics (P > 0.05, data not shown), except that the EIF4G1 after brachytherapy was statistically higher in the patients with positive pelvic lymph nodes (P = 0.009).
In our center, if patients were evaluated to have FIGO IB2 to IIA stages with imaging data and gynecological examinations, preoperative brachytherapy was recommended. We found significant differences in the clinicopathological variables between the patients with and without brachytherapy in this study, including the FIGO stages, pelvic lymph node metastasis status, histological type, and the depth of stromal invasion (Table 2, P < 0.05).
Expression of EIFs and Cervical Cancer Radiosensitivity
In total, 31 patients with detailed tumor volume data before and after preoperative brachytherapy were classified as the response (0 CR + 12 PR) and nonresponse groups (17 SD + 2 PD). Compared with the expression patterns of EIF4A1, EIF4E, and EIF4G1 before brachytherapy, different expression profiles were found after brachytherapy. Statistically, the decreased expression of EIF4A1 and EIF4E after brachytherapy predicted good tumor responses to the brachytherapy (P = 0.029, 0.012, respectively, Table 3).
Correlations Between EIFs and Prognosis
The expression levels of EIF4A, EIF4E, and EIF4G1 were examined with respect to the tumor-specific survival by the Kaplan-Meier and log-rank tests, and there was no statistical significance (P > 0.05). We noticed that the constitutions of the EIF4F complex staining scores after brachytherapy were different from before brachytherapy. The percentage of strongly positive expression dramatically declined after brachytherapy, whereas the numbers of patients with normal and weakly positive expression increased greatly (Fig. 2). We attempted to determine the relationships between these alterations with patient survival. We found that the patients with decreased EIF4A1 had significantly longer survival time than the patients without decrease EIF4A1 (P = 0.020; Tables 4, 5; Fig. 3). The alteration of EIF4A1 was confirmed as 1 independent predictor for tumor-specific survival (P = 0.047; hazards ratio, 0.272; 95% confidence interval, 0.076–0.982; Table 6) using the multivariate Cox regression mode to examine the parameters significantly correlated with patient prognosis in the univariate analysis. No significant relevance between the alterations of EIF4E and EIF4G1 and prognosis were seen.
In this study, the overexpression of EIF4A1, EIF4E, and EIF4G1 was frequently seen in cervical cancer, and the frequencies of overexpression increased with respect to the FIGO stage, the histotype, the status of lymph node metastasis, parametrial invasion, and deep stromal invasion (P < 0.05). The expression levels of the 3 proteins changed after preoperative brachytherapy, but only the altered expression of EIF4A1 showed significant relevance to the prognosis of cervical cancer.
EIF4A1 is a canonical DEAD-box helicase that exhibits adenosine triphosphate–dependent RNA helicase activity in mRNA translation, particularly for the overexpression of mRNAs that contain highly structured 5′ UTR.21 Two thirds of these mRNAs containing a highly structured 5′ UTR encode for oncoproteins or proteins that are implicated in cell growth, cell death, or cell proliferation.22 A growing number of studies have confirmed that EIF4A1 is an important cancer-promoting protein. Elevated expression of EIF4A1 has been reported in primary hepatocellular carcinomas, melanoma cell lines, and congenital melanocytic nevi.5,6 EIF4A1 has been reported to be used as an early diagnostic marker of endometrioid endometrial cancer.23 Ji et al24 found that patients with very high EIF4A1 (top 20%) had a very poor prognosis, whereas EIF4A1 expression slightly above the median did not affect the patient survival. Bordeleau et al16 reported that the depletion of EIF4A1 from the EIF4F complex increased chemosensitivity to doxorubicin. Similar to these results, we found that the EIF4A1 expression significantly correlated with the progress and prognosis of cervical cancer, including the FIGO stage, histotype, and lymph node metastasis, indicating the great clinical importance of this protein. The present study showed that the alteration of EIF4A1 was an independent predictor for the prognosis and response to radiotherapy in cervical cancer treated with preoperative brachytherapy and radical hysterectomy.
We concluded that the elevated expression of EIF4A1 was positively correlated with cancer stages, tumor size, and lymph node metastasis, but the underlined mechanisms for this remain unclear, suggesting that the overexpression of EIF4A1 could enable cancer cells or cancer cells after treatment to have the capacity of regeneration, invasion, or metastasis. If the activity or expression of EIF4A1 was not decreased after treatments, including preoperative adjuvant therapy, patients would likely experience an unsatisfactory response to brachytherapy and worse survival. In this study, a poor response to brachytherapy and prognosis in patients with nondecreased expression of EIF4A1 after brachytherapy could be interpreted by this assumption.
Elevated levels of EIF4E have been reported in some types of cancers and are correlated with disease progression and survival, including in 3 studies on cervical neoplasia. Compared with normal cervical tissues, Van et al25 found that EIF4E mRNA was overexpressed 7-fold in cervical cancer. Another study showed a progressive elevation in EIF4E staining intensity with increasing cervical pathology (P < 0.001), whereas no difference was seen in EIF4E staining intensity with tumor type or tumor grade.15 The findings of Lee et al26 agreed with those results, as they found that EIF4E staining intensity increased from CIN to ISCC (P < 0.001). Our findings in this study provide further evidence that EIF4E expression is significantly correlated with cervical cancer about the clinicopathological characteristics. Decreased expression of EIF4E suggested a favorable response to preoperative brachytherapy. These results showed that EIF4E might play an important role in cervical cancer diagnosis and therapy.
EIF4G1 functions as a scaffold containing translation initiation factor–binding sites, including sites for EIF4E and EIF4A1. Increasing evidence indicates that EIF4G1 could be an oncogenic protein. Significant overexpression of EIF4G1 was observed in 29 of 37 inflammatory breast cancer (IBC) and 25 of 34 stage III (T3, ≥5 cm) non-IBC patients. It facilitated the formation of IBC tumor emboli, which contributes to tumor cell survival.27 Frequent amplification of EIF4G1 on chromosome 3q27.1 and overexpression of EI4G1 mRNA have been reported in lung cancer.28,29 EIF4G1 expression was found to be positively correlated with clinicopathological factors, including tumor T classification, lymph node involvement, and clinical stages in NPC patients,18 which is consistent with our results. Others have identified EIF4G1 as an independent prognostic indicator of the overall survival for NPC patients.
The current report has some limitations. First, unexpected bias might exist because this was a retrospective study with a limited number of patients. Second, the assessments of the cervical cancer response to radiosensitivity were not always available because some patients had incomplete clinical data. To our knowledge, this is a pioneering work investigating eukaryotic initiation factors as relevant prognostic and predictive markers in cervical cancer patients, and encouraging findings were observed.
The present study assessed, for the first time, the expression of EIF4A1, EIF4E, and EIF4G1 in normal and malignant cervical tissues (with or without preoperative brachytherapy) and demonstrated that overexpression of EIF4A1, EIF4E, and EIF4G1 were acquired malignant phenotypic features of cervical cancer, which may provide useful information for cervical cancer diagnosis, progression, and prognosis. This study has introduced us to the idea that decreased expression of EIF4A1 after preoperative brachytherapy is an independent prognostic marker for better tumor-specific survival in cervical cancer.
The authors thank all of the people and patients who participated in this study.
1. Jemal A, Bray F, Center MM, et al. Global cancer statistics. Ca Cancer J Clin. 2011; 61: 69–90.
2. Hagner PR, Schneider A, Gartenhaus RB. Targeting the translational machinery as a novel treatment strategy for hematologic malignancies. Blood. 2010; 18: 2127–2135.
3. Mamane Y, Petroulakis E, Rong L, et al. eIF4E—from translation to transformation. Oncogene. 2004; 19: 3172–3179.
4. Fukuchi-Shimogori T, Ishii I, Kashiwagi K, et al. Malignant transformation by overproduction of translation initiation factor eIF4G. Cancer Res. 1997; 57: 5041–5044.
5. Shuda M, Kondoh N, Tanaka K, et al. Enhanced expression of translation factor mRNAs in hepatocellular carcinoma. Anticancer Res. 2000; 20: 2489–2494.
6. Eberle J, Krasagakis K, Orfanos CE. Translation initiation factor eIF-4A1 mRNA is consistently overexpressed in human melanoma cells in vitro. Int J Cancer. 1997; 71: 396–401.
7. Wang S, Rosenwald IB, Hutzler MJ, et al. Expressionof the eukaryotic translation initiation factors 4E and 2alpha in non-Hodgkin’s lymphomas. Am J Pathol. 1999; 155: 247–255.
8. Kerekatte V, Smiley K, Hu B, et al. The proto-oncogene/translation factor eIF4E: a survey of its expression in breast carcinomas. Int J Cancer. 1995; 64: 27–31.
9. Rosenwald IB, Hutzler MJ, Wang S, et al. Expression of eukaryotic translation initiation factors 4E and 2alpha is increased frequently in bronchioloalveolar but not in squamous cell carcinomas of the lung. Cancer. 2001; 92: 2164–2171.
10. Nathan CO, Amirghahari N, Abreo F, et al. Overexpressed eIF4E is functionally active in surgical margins of head and neck cancer patients via activation of the Akt/mammalian target of rapamycin pathway. Clin Cancer Res. 2004; 10: 5820–5827.
11. Salehi Z, Mashayekhi F. Expression of the eukaryotic translation initiation factor 4E (eIF4E) and 4E-BP1 in esophageal cancer. Clin Biochem. 2006; 39: 404–409.
12. Salehi Z, Mashayekhi F, Shahosseini F. Significance of eIF4E expression in skin squamous cell carcinoma. Cell Biol Int. 2007; 31: 1400–1404.
13. Crew JP, Fuggle S, Bicknell R, et al. Eukaryotic initiation factor–4E in superficial and muscle invasive bladder cancer and its correlation with vascular endothelial growth factor expression and tumour progression. Br J Cancer. 2000; 82: 161–166.
14. Rosenwald IB, Chen JJ, Wang S, et al. Upregulation of protein synthesis initiation factor eIF-4E is an early event during colon carcinogenesis. Oncogene. 1999; 18: 2507–2517.
15. Matthews-Greer J, Caldito G, De Benedetti A, et al. eIF4E as a marker for cervical neoplasia. Appl Immunohistochem Mol Morphol. 2005; 13: 367–370.
16. Bordeleau ME, Robert F, Gerard B, et al. Therapeutic suppression of translation initiation modulates chemosensitivity in a mouse lymphoma model. J Clin Invest. 2008; 118: 2651–2660.
17. Hayman TJ, Williams ES, Jamal M, et al. Translation initiation factor eIF4E is a target for tumor cell radiosensitization. Cancer Res. 2012; 72: 2362–2372.
18. Tu L, Liu Z, He X, et al. Over-expression of eukaryotic translation initiation factor 4 gamma 1 correlates with tumor progression and poor prognosis
in nasopharyngeal carcinoma. Mol Cancer. 2010; 9: 78.
19. Kononen J, Bubendorf L, Kallioniemi A, et al. Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nat Med. 1998; 4: 844–847.
20. WHO. Handbook for Reporting Results of Cancer Treatment. Geneva: WHO Offset Publication; Geneva, Switzerland. 1979: 48.
21. Svitkin YV, Pause A, Haghighat A, et al. The requirement for eukaryotic initiation factor 4A (elF4A) in translation is in direct proportion to the degree of mRNA 5′ secondary structure. RNA. 2001; 7: 382–394.
22. Kozak M. An analysis of 5′-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 1987; 15: 8125–8148.
23. Lomnytska MI, Becker S, Gemoll T, et al. Impact of genomic stability on protein expression in endometrioid endometrial cancer. Br J Cancer. 2012; 27: 1297–1305.
24. Ji P, Diederichs S, Wang W, et al. MALAT-1, a novel noncoding RNA, and thymosin beta 4 predict metastasis and survival in early-stage non-small cell lung cancer. Oncogene. 2003; 22: 8031–8041.
25. Van Trappen PO, Ryan A, Carroll M, et al. A model for co-expression pattern analysis of genes implicated in angiogenesis and tumour cell invasion in cervical cancer
. Br J Cancer. 2002; 87: 537–544.
26. Lee JW, Choi JJ, Lee KM, et al. eIF-4E expression is associated with histopathologic grades in cervical neoplasia. Hum Pathol. 2005; 36: 1197–1203.
27. Silvera D, Arju R, Darvishian F, et al. Essential role for eIF4GI overexpression in the pathogenesis of inflammatory breast cancer. Nat Cell Biol. 2009; 11: 903–908.
28. Bauer C, Brass N, Diesinger I, et al. Overexpression of the eukaryotic translation initiation factor 4G (eIF4G-1) in squamous cell lung carcinoma. Int J Cancer. 2002; 98: 181–185.
29. Bauer C, Diesinger I, Brass N, et al. Translation initiation factor eIF-4G is immunogenic, overexpressed, and amplified in patients with squamous cell lung carcinoma. Cancer. 2001; 92: 822–829.
Keywords:© 2014 by the International Gynecologic Cancer Society and the European Society of Gynaecological Oncology.
Cervical cancer; Brachytherapy; Prognosis; EIF4A1; Translational control; EIF4F