Sage Journals: Discover world-class research

Abstract

Cyclin E is a critical regulator in cell cycle and promotes the initiation of DNA replication and centrosome duplication in late G1. The overexpression of cyclin E is common in cancers of the digestive system. However, whether cyclin E represents a prognostic biomarker in gastrointestinal cancer remains controversial. We reviewed the published literatures to clarify the association between cyclin E determined by immunohistochemistry (IHC) and survival in gastrointestinal cancer. Literatures were searched in PubMed and Cochrane Library published up to December 1, 2014. A total of 282 articles were initially identified, and 14 articles were included in this study. Meta-analysis was performed for 10 studies with a total of 1300 patients. Combined hazard risk (HR) and corresponding 95% confidence interval (CI) were calculated by random-effect model due to the heterogeneity. The quality of included studies was assessed by the Newcastle-Ottawa Scale and the Methodological Index for Non-Randomized Studies (MINORS). We found that high level of cyclin E was a predicator of poor prognosis among patients with gastrointestinal cancer (HR = 1.67, 95% CI = 1.06-2.63, P = .028). In summary, overexpression of cyclin E is associated with poor prognosis in gastrointestinal cancer and expression of cyclin E determined by IHC might be a prognostic marker for gastrointestinal cancer in clinical practice.

Keywords

cyclin E gastric cancer colorectal cancer prognosis meta-analysis

Introduction

Progression between phases in cell cycle (G0, G1, S, G2, M) is an orderly process tightly regulated by complex yet redundant mechanisms. It includes multiple checkpoints used to assess growth signals, cell size, and DNA integrity.¹ Cyclins are prime cell cycle regulators, and they play a critical role in the control of cell proliferation by forming a complex with different cyclin-dependent kinases (CDKs). Cyclin D-CDK4, cyclin D-CDK6, and cyclin E-CDK2 drive G1 progression through the restriction point, which commits the cell to complete the cycle.² Abnormalities in cell cycle regulators and subsequent deregulation of the G1/S transition may be one of the important biologic events in the development and progression of cancer. Overexpression of cyclins and CDKs and loss of expression of CKIs (CDK inhibitors) are frequently reported in human neoplasia. The overexpression of cyclin E is common in various types of cancer, including breast cancer,^3,4 nonsmall cell lung cancer (NSCLC),^5,6 colon cancer,⁷ bladder cancer,⁸ and ovarian cancer.⁹

Cancers of the digestive system are common worldwide, while the pattern and incidence vary widely between different parts of the world. According to the data of World Health Organization, in 2005 to 2009, the highest overall rates (per 100 000) among men were in the Republic of Korea (25.3). In women, the highest rates (per 100 000) were observed in Ecuador (10.5).¹⁰ It is estimated that there are more than 1.2 million men and women living in the United States with a previous colorectal cancer diagnosis up to January 1, 2014, and the 1-year and 5-year relative survival rates for patients with colorectal cancer are 83.4% and 64.9%, respectively.¹¹ Mortality rates for both gastric and colorectal cancer decline in the past decade attributed to the prevention, early detection, and treatment,¹² but gastric cancer and colorectal cancer still rank the second and fourth leading cause of cancer deaths worldwide.^13,14

A number of studies have investigated the association between expression of cyclin E and survival in patients with gastric and colorectal cancer. However, whether cyclin E represents a prognostic biomarker remains controversial and no consensus has been reached. Tenderenda¹⁵ reported that increased expression of cyclin E in gastric cancer tissue correlated with poorer prognosis. In contrast, Takano et al¹⁶ and Li et al¹⁷ reported independently that patients with gastric cancer and colorectal cancer whose tumors were cyclin E positive showed a favorable prognosis. However, So et al¹⁸ and Ioachim¹⁹ reported that expression of cyclin E was not related to survival of patients with gastric or colorectal cancer. Hence, we performed a systematic review of published studies with meta-analysis to assess the effects of overexpression of cyclin E in tumor tissue on survival in patients with gastrointestinal cancer.

Methods

Search Strategy and Study Selection

The electronic databases PubMed and Cochrane Library were searched for studies dealing with the prognostic implications of different expression levels of cyclin E protein to include in the present meta-analysis. The references reported in the identified studies and conference papers were also used to complete the search. The deadline of the included articles was December 1, 2014. Searches included the terms “gastric,” “intestinal or colorectal,” “cancer or carcinoma or tumor,” and “Cyclin E.”

Studies were considered eligible if they met the following criteria: (1) peer-reviewed original articles, (2) patients with gastric and intestinal cancer were studied, (3) the expression of cyclin E in cancer tissues was measured with immunohistochemistry (IHC), and (4) the primary outcomes of interest were overall survival (OS) and/or relapse-free survival (RFS), hazard ratios (HRs), and 95% confidence interval (95% CI) according to cyclin E level either had to be reported or had to be calculated from the data presented. The exclusion criteria for the studies were (1) HRs of OS or RFS could not be extracted based on the given information and (2) animal or laboratory studies, non-English articles, reviews or letters, case reports, articles published in a book.

The titles and abstracts of relevant articles were read by 2 reviewers (Lanshan Huang and Fanghui Ren), respectively. The articles which could not be categorized in accordance with title and abstract were retrieved for full-text review. Disagreements were resolved through consensus with a third reviewer (Gang Chen).

Data Extraction

The final articles included were assessed independently by 2 reviewers (Lanshan Huang and Fanghui Ren) using a standard protocol. The following characteristics were extracted: first author, publication year, origin of population, type of cancer, tumor stage, test method, cutoff value, follow-up, and survival data. Differences in the data extraction were resolved together by 2 reviewers.

Statistical Analysis

Included studies were tested by performing chi-square Q-tests for heterogeneity. A P value greater than .05 for the Q-test indicated lack of heterogeneity among studies, so the fixed-effect model was used for meta-analysis. If not, DerSimonian-Laird random-effect model²⁰ was used.

Hazard ratio with 95% CI was used to combine the data. When described in original articles, these values were obtained directly. When these statistical variables were not provided, they were calculated from available numerical data or Kaplan-Meier survival curve. The total values were combined to obtain the overall HR and 95% CI²¹ consisted of a DerSimonian-Laird random-effect model.

Sensitivity analyses were performed by removing each study in turn to establish the extent to which they contributed to heterogeneity and to the overall result. Studies with clear survival values provided were included in a sensitivity analysis, excluding the studies in which data could only be obtained from calculation or Kaplan-Meier survival curves.

The quality of included studies was assessed using the Newcastle-Ottawa Scale (NOS) for cohort studies.²² The risk of bias for included studies was assessed using the NOS and the Methodological Index for Non-Randomized Studies (MINORS). The potential publication bias was assessed using Beggs²³ and Eggers tests.²⁴ The statistical analyses were performed by Stata version 11.0 and Rev Man 5.3. Again, disagreements were resolved by discussion of 2 reviewers (Lanshan Huang and Fanghui Ren) and consensus with a third reviewer (Gang Chen).

Results

Study Characteristics

A total of 282 potentially relevant articles were found after initial searching of PubMed and Cochrane Library. Finally, the remaining 15 potential eligible studies were obtained for full-text evaluation.^{15

-19,25

-34} One study²⁵ was excluded due to the duplicate for the same population. Fourteen studies^{15

-19,26

-34} were included in the systematic review, of which 10 studies^{15
-17,26

-29,31,33,34} with sufficient data or with data that could be calculated indirectly were evaluated in the meta-analysis and 4 studies^18,19,30,32 were with lack of eligible HR data. Figure 1 showed the flowchart of study inclusion. The main characteristics of the included studies are summarized in Table 1.

Figure 1.

Study inclusion flowchart.

Table 1.

Characteristics of Patients in Different Studies.

First Author, year	Origin of Population	Type	Stage	Cyclin E (+)	Method	Cutoff	Follow-up, months	Survival Analysis	HR Statistics	HR (95% CI)	P
Jang et al,²⁶ 1999	Korea	Gastric cancer	I-IV	34/84	IHC	50%	38^a	OS	Reported	2.23 (1.13-4.39)	.021
Muller et al,²⁸ 1999	Germany	Gastric cancer	–	36/413	IHC	5%	27.6^a	OS	SC	1.175 (0.715-1.925)	.4
Kwon,²⁷ 1999	Korea	Gastric cancer	I-III	24/115	IHC	50%	58^a	OS	Reported	0.86 (0.48-1.55)	.626
Aoyagi et al,²⁹ 2000	Japan	Gastric cancer	I-IV	40/76	IHC	10%	84^b	OS	SC	2.13 (1.05-4.34)	<.01
Takano et al,¹⁶ 2000	Japan	Gastric cancer	–	34/116	IHC	5%	65.3^a	OS	SC	0.91 (0.465-1.77)	.07
Li et al,¹⁷ 2001	Japan	Colorectal carcinoma	0-IV	83/105	IHC	5%	36^b	OS	SC	0.11 (0.02-0.54)	.0168
Bani-Hani et al,³¹ 2005	Jordan	Gastric cancer	I-IV	35/89	IHC	25%	33.7^c	OS	Reported	5.32 (2.60-10.87)	<.0001
Tenderenda,¹⁵ 2005	Poland	Gastric cancer	–	8/80	IHC	50%	NA	OS	Reported	7.245 (3.021-17.38)	NA
								DFS	Reported	7.309 (3.031-17.62)	NA
Choi et al,³³ 2009	Korea	Gastric cancer	I-IV	77/166	IHC	50%	35.7^a	OS	SC	1.23 (0.805-1.415)	.54
Kouraklis et al,³⁴ 2009	Greece	Gastric cancer	–	23/56	IHC	5%	22^a	OS	SC	2.67 (1.18-6.075)	.26
Lim,³⁰ 2004^d	Korea	Colorectal carcinoma	II	3/41	IHC	1%	62^a	RFS	SC	32.74 (0.28-3890.27)	.442
So et al,¹⁸ 2000^d	Singapore	Gastric cancer	NA	20/71	IHC	10%	NA	NA	NA	NA	NA
Bondi et al,³² 2005^d	Norway	Colorectal carcinoma	A-D	55/217	IHC	5%	60^b	NA	NA	NA	.07
Ioachim,¹⁹ 2008^d	Greece	Colorectal carcinoma	B-C	18/60	IHC	5%	56.4^c	NA	NA	NA	NA

Abbreviations: NA, not available; A/B/C/D Dukes classification; I/II/III/IV TNM stage; IHC, immunohistochemistry; OS, overall survival; DFS, disease-free survival; RFS, recurrence-free survival; HR hazard ratio; SC survival curve; CI, confidence interval.

^aMedian follow-up.

^bTotal follow-up.

^cMean follow-up.

^dThe 4 studies lack of eligible HR data.

In the meta-analysis, we collected data from the 10 studies between 1999 and 2009 including a total of 1300 patients from Japan, Korea, Germany, and Poland. All of them were retrospective in design. All of the studies reported the prognostic value of cyclin E status for survival in patients with gastric cancer or colorectal carcinomas. Immunohistochemistry was applied for cyclin E expression assessment in tissue samples in all the studies included. Of the 10 studies, 4 studies directly reported HRs while the other 6 studies provided survival curves. Seven studies identified overexpression of cyclin E as an indicator of poor prognosis and the other 3 studies showed inverse results of which 2 studies showed no statistically significant impact of overexpression of cyclin E on prognosis.

Since the HRs of OS could not be obtained based on the data presented in the articles of So et al, Bondi et al, and Ioachim ^18,19,32 and HR with 95% CI calculated from Kaplan-Meier survival curve in Lim article³⁰ was not eligible which may be due to the small number of the study and low-positive rate, these articles were excluded in the meta-analysis and remained in the systematic review.

Meta-Analysis and Sensitivity Analysis

The 10 studies including 1300 cases were evaluated for the effect of overexpression of cyclin E on OS. A random-effect model was used to combine HRs because of the heterogeneity present among the studies by chi-square Q-tests (χ² = 46.73, P < .001, I² = 80.7%). The combined HR was 1.67 (95% CI 1.06-2.63; P = .028), suggesting that high level of cyclin E was a predicator of poor prognosis among patients with gastrointestinal cancer. Forest plot is shown in Figure 2. In addition, subgroup meta-analysis was not conducted because of insufficient data.

Figure 2.

Forest plot of meta-analysis of the effect of expression of cyclin E on survival in patients with gastrointestinal cancer. Random effects model was used when combining data. The combined hazards ratio (HRs) displayed in this figure suggest that high level of expression of cyclin E might be an independent prognostic factor that predicts the adverse outcome in patients with gastrointestinal cancer.

Sensitivity analyses showed that the study by Tenderenda,¹⁵ Bani-Hani et al,³¹ and Li et al¹⁷ were the top 3 with heterogeneity, and removal of these studies changed the effect into a significant one with lowest heterogeneity (HR 1.365, 95% CI 1.03-1.81, P = .031; χ² = 10.56, I² = 43.2%). In the studies with clear survival data,^15,26,27,31 the effect reaches significance with heterogeneity (HR 2.85, 95% CI 1.08-7.49, P = .034; χ² = 22.7, I² = 86.8%).

Quality and Bias Assessment

The quality assessment for a total of 14 studies are summarized in Table A1 of the appendix. For quality, scores ranged from 0 (lowest) to 9 (highest), and studies with scores of 6 or more were rated as high quality. Scores of the 14 studies ranged from 4 to 7, with a mean of 6.21 ± 1.19. The risk of bias was shown in Figure A1 of the appendix. Baseline equivalence of groups and incomplete outcome data were considered high risk of bias in the 14 studies which indicated information bias and attrition bias.

The Begg funnel plot and the Egger test were applied for assessing publication bias in the meta-analysis. Begg funnel plot revealed evidence of asymmetry in this study (Figure 3). However, the P value of the Egger test (P = .547) suggested no evidence of publication bias. These results suggest that there is no obvious publication bias in this analysis in terms of the OS of patients with gastrointestinal cancer.

Figure 3.

Begg’s funnel plot for the assessment of potential publication bias. Studies are distributed asymmetrically and suggest the potential meta-analysis publication bias.

Systematic Review

As the detailed results of the 4 studies excluded from meta-analysis were not shown in Table 1, we summarize the information here. In the study of So et al,¹⁸ expression of cyclin E was not related to survival of patients with gastric cancer. In the studies of Lim et al,³⁰ Bondi et al,³² and Ioachim,¹⁹ the correlation of cyclin E with prognosis of patients with colorectal carcinoma was reported. According to Lim et al,³⁰ in patients with stage II colorectal cancer, the overexpression of cyclin E could not predict poor survival. According to Bondi et al,³² when analyzing the expression of the different cyclin proteins individually (cyclins A, B1, C, D1, D3, E, and H) with regard to survival, only low expression of cyclin A showed a significant association with an unfavorable prognosis. Ioachim¹⁹ reported that the prognostic significance of cyclin E in determining the risk of recurrence and OS showed no statistically significant differences.

In total, among the 14 studies included, 7 studies identified overexpression of cyclin E as an indicator of poor prognosis, 3 studies showed inverse results, and 4 studies showed no association of overexpression of cyclin E with prognosis. Of 14 studies, there were 10 on gastric cancer and 4 on colorectal carcinoma. In the 10 studies on the association of cyclin E with gastric cancer prognosis, 7 studies identified cyclin E as a poor prognosis marker, 2 studies showed inverse results, and 1 study found no relation. In the 4 studies on colorectal carcinoma, 1 study showed that low level of cyclin E associated with poor survival and 3 studies found no correlation between cyclin E and survival.

Discussion

In the previous published literatures, several systematic reviews with meta-analyses on overexpression of cyclin E for tumors have been performed; high level of cyclin E was found to be a poor prognostic factor for survival in breast cancer and NSCLC.^35
-37 As far as we know, no meta-analysis had been undertaken to evaluate cyclin E as a prognostic marker in case of digestive system. In the present meta-analysis, we identified 10 studies concerning the expression of cyclin E on survival of gastrointestinal cancer. The expression of cyclin E was determined by IHC with surgical specimens. Our results showed that overexpression of cyclin E was associated with poor survival for patients with gastrointestinal cancer.

Cyclins E and A and the activator of the cyclin E(A)/CDK2 kinase—the CDC25A phosphatase—are induced in late G1.³⁸ Cyclin E accumulates at G1/S boundary and forms complex with CDK2 and subsequently promotes initiation of DNA replication and centrosome duplication. The abundant cyclin E eventually becomes phosphorylated and degraded that allows normal cell cycle progression.^39,40 However, the level of cyclin E and activity of cyclin E–CDK2 can be aberrantly regulated and this excessive activity of the cyclinE–CDK2 complex, in turn, drives cells to replicate their DNA prematurely, resulting in tumorigenesis.⁴¹

Alterations of cyclin E have been shown in gastrointestinal cancer including amplification of cyclin E gene and overexpression of cyclin E protein.^4,8 Like other regulators of cell cycle such as cyclin D1,^42,43 the association of cyclin E with prognosis of these cancers also cannot reach a consensus hither to. Inconsistent results were shown in previous studies that were performed with methods of polymerase chain reaction (PCR), Western blot (WB), or IHC. Expression of Cyclin E is easy to be detected in human tissues by standard IHC techniques and IHC is cost effective and relatively easy to perform. In this study, we assessed the prognostic role of cyclin E protein in patients with gastrointestinal cancer. Evaluating the prognosis of patients with gastrointestinal cancer by detecting cyclin E level with IHC techniques might be an alternative scheme.

As a meta-analysis, this study was restricted to some limitations that should be discussed for further consideration. First, there might be biases if negative studies were less frequently published or contained less detailed results. Second, we conducted a literature search in PubMed and Cochrane Library. As studies included were published in English only, selection and language bias might exist. Third, due to the limited published studies on association of cyclin E protein with prognosis of colorectal carcinoma and lack of subgroup meta-analysis, this meta-analysis cannot offer strong evidence on the overall of gastrointestinal cancer. Fourth, extrapolating the survival rates from the survival curves may add inaccuracy in the extracted survival rates and lead to potential bias. Finally, the cutoff used to assess overexpression of cyclin E, the technique used for IHC, and the analysis of univariate or multivariate were variable between studies might contribute to the heterogeneity. Therefore, our results should be interpreted with caution.

In conclusion, despite the limitations described earlier, this study suggests that overexpression of cyclin E is associated with poor prognosis in gastrointestinal cancer. The expression of cyclin E determined by IHC might be a prognostic marker for gastrointestinal cancer in clinical practice. To strengthen our findings, well-designed prospective studies with better standardized assessment of prognostic markers are necessary to help to define the precise prognostic significance of cyclin E in patients with gastrointestinal cancer.

Footnotes

Abbreviations

Appendix A

Table A1.

The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of 14 studies.

Studies	Selection				Comparability	Outcome
First author, year	Representativeness of the Exposed Cohort	Selection of the Nonexposed Cohort	Ascertainment of Exposure	Demonstration that Outcome of Interest was not Present at Start of Study	Comparability of Cohorts on the Basis of Metastasis or Age, Stage, Grade, etc	Assessment of Outcome (Death or Recurrence)	Adequacy of Follow-up Length for Outcome (2 years)	Lost to Follow-up Acceptable (less than 25% or reported)	Total Quality Score
Jang et al,²⁶ 1999	*	*	*	*		*	*	*	7
Muller et al,²⁸ 1999	*	*	*	*		*	*	*	7
Kwon,²⁷ 1999	*	*	*	*		*	*	*	7
Aoyagi et al,²⁹ 2000	*	*	*	*		*	*	*	7
Takano et al,¹⁶ 2000	*	*	*	*		*	*	*	7
Li et al,¹⁷ 2001	*	*	*	*		*	*	*	7
Bani-Hani et al³¹, 2005	*	*	*			*	*	*	6
Tenderenda,¹⁵ 2005	*	*	*			*			4
Choi et al,³³ 2009	*	*	*	*		*	*	*	7
Kouraklis et al,³⁴ 2009	*	*	*	*		*	*	*	7
Lim,³⁰ 2004*	*	*	*	*		*	*	*	7
So et al,¹⁸ 2000*	*	*	*			*			4
Bondi et al,³² 2005*	*	*	*			*	*		5
Ioachim,¹⁹ 2008*	*	*	*			*	*		5

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Zhu

Shih

Ling

Tsao

. Immunohistochemical markers of prognosis in non-small cell lung cancer: a review and proposal for a multiphase approach to marker evaluation. J Clin Pathol. 2006;59(8):790–800. doi:10.1136/jcp.2005.031351.

Planas-Silva

Weinberg

. The restriction point and control of cell proliferation. Curr Opin Cell Biol. 1997;9(6):768–772.

Berglund

Landberg

. Cyclin e overexpression reduces infiltrative growth in breast cancer: yet another link between proliferation control and tumor invasion. Cell Cycle. 2006;5(6):606–609.

Shaye

Sahin

Hao

Hunt

Keyomarsi

Bedrosian

. Cyclin E deregulation is an early event in the development of breast cancer. Breast Cancer Research Treat. 2009;115(3):651–659. doi:10.1007/s10549-008-0266-0.

Dosaka-Akita

Hommura

Mishina

. A risk-stratification model of non-small cell lung cancers using cyclin E, Ki-67, and ras p21: different roles of G1 cyclins in cell proliferation and prognosis. Cancer Res. 2001;61(6):2500–2504.

Kosacka

Piesiak

Porebska

Korzeniewska

Dyla

Jankowska

. Cyclin A and Cyclin E expression in resected non-small cell lung cancer stage I-IIIA. In Vivo (Athens, Greece). 2009;23(4):519–525.

Corin

Di Giacomo

Lastella

Bagnulo

Guanti

Simone

. Tumor-specific hyperactive low-molecular-weight cyclin E isoforms detection and characterization in non-metastatic colorectal tumors. Cancer Biol Ther. 2006;5(2):198–203.

Richter

Wagner

Kononen

. High-throughput tissue microarray analysis of cyclin E gene amplification and overexpression in urinary bladder cancer. Am J Pathol. 2000;157(3):787–794.

Lee

Heo

Kim

. Significance of cell cycle regulatory proteins as malignant and prognostic biomarkers in ovarian epithelial tumors. Int J Gynecol Pathol. 2011;30(3):205–217. doi:10.1097/PGP.0b013e3182063e71.

10.

Ferro

Peleteiro

Malvezzi

. Worldwide trends in gastric cancer mortality (1980-2011), with predictions to 2015, and incidence by subtype. Eur J Cancer (Oxford, England: 1990). 2014;50(7):1330–1344. doi:10.1016/j.ejca.2014.01.029.

11.

DeSantis

Lin

Mariotto

. Cancer treatment and survivorship statistics, 2014. CA Cancer J Clin. 2014;64(4):252–271. doi:10.3322/caac.21235.

12.

Siegel

Zou

Jemal

. Cancer statistics, 2014. CA Cancer J Clin. 2014;64(1):9–29. doi:10.3322/caac.21208.

13.

Elimova

Shiozaki

Wadhwa

. Medical management of gastric cancer: a 2014 update. World J Gastroenterol. 2014;20(38):13637–13647. doi:10.3748/wjg.v20.i38.13637.

14.

Ahmed

Johnson

Ahmed

Iqbal

. Advances in the management of colorectal cancer: from biology to treatment. Int J Colorectal Dis. 2014;29(9):1031–1042. doi:10.1007/s00384-014-1928-5.

15.

Tenderenda

. A study on the prognostic value of cyclins D1 and E expression levels in resectable gastric cancer and on some correlations between cyclins expression, histoclinical parameters and selected protein products of cell-cycle regulatory genes. J Exp Clin Cancer Res. 2005;24(3):405–414.

16.

Takano

Kato

van Diest

Masuda

Mitomi

Okayasu

. Cyclin D2 overexpression and lack of p27 correlate positively and cyclin E inversely with a poor prognosis in gastric cancer cases. Am J Pathol. 2000;156(2):585–594. doi:10.1016/s0002-9440(10)64763-3.

17.

Miki

Ohmori

Funamoto

. Expression of cyclin E and cyclin-dependent kinase 2 correlates with metastasis and prognosis in colorectal carcinoma. Hum Pathol. 2001;32(9):945–953. doi:10.1053/hupa.2001.27116.

18.

Samarasinge

Raju

Moochhala

. Expression of cell-cycle regulators p27 and cyclin E correlates with survival in gastric carcinoma patients. J Surg Res. 2000;94(1):56–60. doi:10.1006/jsre.2000.5998.

19.

Ioachim

. Expression patterns of cyclins D1, E and cyclin-dependent kinase inhibitors p21waf1/cip1, p27kip1 in colorectal carcinoma: correlation with other cell cycle regulators (pRb, p53 and Ki-67 and PCNA) and clinicopathological features. Int J Clin Pract. 2008;62(11):1736–1743. doi:10.1111/j.1742-1241.2006.01105.x.

20.

DerSimonian

Laird

. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177–188.

21.

Parmar

Torri

Stewart

. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med. 1998;17(24):2815–2834.

22.

Wells

Shea

O’Connell

. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Web site. www.ohri.ca/programs/clinical_epidemiology/oxford.asp. 2011. Published 2011. Accessed November 25, 2012.

23.

Begg

Mazumdar

. Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994;50(4):1088–1101.

24.

Egger

Davey Smith

Schneider

Minder

. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629–634.

25.

Ahn

Kim

Jang

. Expression of cyclin D1 and cyclin E in human gastric carcinoma and its clinicopathologic significance. J Korean Med Sci. 1998;13(5):513–518.

26.

Jang

Park

. Expression of cell-cycle regulators, cyclin E and p21WAF1/CIP1, potential prognostic markers for gastric cancer. Eur J Surg Oncol. 1999;25(2):157–163. doi:10.1053/ejso.1998.0619.

27.

Kwon

Kang

Suh

Chang

Jang

Chung

. The loss of p27 protein has an independent prognostic significance in gastric cancer. Anticancer Res. 1999;19(5B):4215–4220.

28.

Müller

Noguchi

Wirtz

Hommel

Gabbert

. Expression of cell-cycle regulatory proteins cyclin D1, cyclin E, and their inhibitor p21 WAF1/CIP1 in gastric cancer. J Pathol. 1999;189(2):186–193.

29.

Aoyagi

Koufuji

Yano

. Immunohistochemical study on the expression of cyclin D1 and E in gastric cancer. Kurume Med J. 2000;47(3):199–203.

30.

Lim

Kim

Ahn

. Cyclin E, p27 and mutant p53 do not predict the Prognosis in AJCC Stage II Colorectal Carcinomas. Korean J Gastroenterol. 2004;44(6):314–320.

31.

Bani-Hani

Almasri

Khader

Sheyab

Karam

. Combined evaluation of expressions of cyclin E and p53 proteins as prognostic factors for patients with gastric cancer. Clin Cancer Res. 2005;11(4):1447–1453. doi:10.1158/1078-0432.CCR-04-1730.

32.

Bondi

Husdal

Bukholm

Nesland

Bakka

Bukholm

. Expression and gene amplification of primary (A, B1, D1, D3, and E) and secondary (C and H) cyclins in colon adenocarcinomas and correlation with patient outcome. J Clin Pathol. 2005;58(5):509–514. doi:10.1136/jcp.2004.020347.

33.

Choi

Noh

. Expression levels of cyclin G2, but not cyclin E, correlate with gastric cancer progression. J Surg Res. 2009;157(2):168–174. doi:10.1016/j.jss.2008.06.020.

34.

Kouraklis

Katsoulis

Theocharis

. Does the expression of cyclin E, pRb, and p21 correlate with prognosis in gastric adenocarcinoma? Dig Dis Sci. 2009;54(5):1015–1020. doi:10.1007/s10620-008-0464-y.

35.

Wang

Shao

. Cyclin e expression and prognosis in breast cancer patients: a meta-analysis of published studies. Cancer Invest. 2006;24(6):581–587. doi:10.1080/07357900600894799.

36.

Gao

. Prognostic value of cyclin E expression in breast cancer: a meta-analysis. Tumour Biol. 2013;34(6):3423–3430. doi:10.1007/s13277-013-0915-8.

37.

Huang

Wang

Chen

. Meta-analysis for cyclin E in lung cancer survival. Clin Chim Acta. 2012;413(7-8):663–668. doi:10.1016/j.cca.2011.12.020.

38.

Kastan

Bartek

. Cell-cycle checkpoints and cancer. Nature. 2004;432(7015):316–323. doi:10.1038/nature03097.

39.

Clurman

Sheaff

Thress

Groudine

Roberts

. Turnover of cyclin E by the ubiquitin-proteasome pathway is regulated by cdk2 binding and cyclin phosphorylation. Genes Dev. 1996;10(16):1979–1990.

40.

Moberg

Bell

Wahrer

Haber

Hariharan

. Archipelago regulates Cyclin E levels in Drosophila and is mutated in human cancer cell lines. Nature. 2001;413(6853):311–316. doi:10.1038/35095068.

41.

Bortner

Rosenberg

. Induction of mammary gland hyperplasia and carcinomas in transgenic mice expressing human cyclin E. Mol Cell Biol. 1997;17(1):453–459.

42.

Zhang

Jiang

. The role of cyclin D1 expression and patient's survival in non-small-cell lung cancer: a systematic review with meta-analysis. Clin Lung Cancer. 2012;13(3):188–195. doi:10.1016/j.cllc.2011.10.003.

43.

Zhao

Wei

. Clinicopathological and prognostic role of cyclin D1 in esophageal squamous cell carcinoma: a meta-analysis. Dis Esophagus. 2012;25(6):520–526. doi:10.1111/j.1442-2050.2011.01278.x.

Prognostic Value of Expression of Cyclin E in Gastrointestinal Cancer

Abstract

Keywords

Introduction

Methods

Search Strategy and Study Selection

Data Extraction

Statistical Analysis

Results

Study Characteristics

Meta-Analysis and Sensitivity Analysis

Quality and Bias Assessment

Systematic Review

Discussion

Footnotes

Abbreviations

Appendix A

Declaration of Conflicting Interests

Funding

References