International Journal of Surgery 48 (2017) 38–44
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.elsevier.com/locate/ijsu
Review
Clinicopathological and prognostic significance of FoxM1 in gastric cancer: A meta-analysis
MARK
Dongdong Jianga,b,1, Lu Jianga,b,1, Baiying Liua,b, He Huanga,b, Wenbin Lia, Taotao Zhanga, Guo Zua,∗, Xiangwen Zhanga,∗∗ a b
Department of Gastroenterology Surgery, The Dalian Municipal Central Hospital Affiliated of Dalian Medical University, Dalian 116033, China Dalian Medical University, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Prognosis Clinical parameters FoxM1 Gastric cancer Meta-analysis
Background: The relationship between expression of FoxM1 and clinical parameters of patients with gastric cancer (GC) has yet to be fully established. Methods: A systematic search was performed. Odds ratio (OR) and confidence interval (CI) were used to assess association between expression of FoxM1 and clinical parameters and the prognostic value of patients with GC. Results: Eight studies involving 529 patients with GC were identified. Overall, the pooled results showed that expression of FoxM1 was associated with TNM stage (OR: 0.482, 95%CI: 0.275–0.845, P = 0.011), depth of invasion (OR: 0.617, 95%CI: 0.382–0.998, P = 0.049) and lymph node metastasis (OR: 2.084, 95%CI: 1.305–3.328, P = 0.002) in the patients with GC. Whereas, expression of FoxM1 was not associated with gender (OR: 1.143, 95%CI: 0.726–1.799, P = 0.564) and tumors' differentiation (OR: 0.991, 95%CI: 0.624–1.575, P = 0.971) of GC. Expression of FoxM1 was also associated with poor prognosis of overall survival (OS) in the patients with GC (one year OS: OR: 0.218, 95%CI: 0.103–0.459, P = 0.000; three years OS: OR: 0.178, 95%CI: 0.093–0.340, P = 0.000; five years OS: OR: 0.180, 95%CI: 0.095–0.341, P = 0.000). Conclusion: Expression of FoxM1 is associated with TNM stage, depth of invasion, lymph node metastasis and poor prognosis of the patients with GC.
1. Introduction Gastric cancer (GC) is the fifth most common malignant cancer in the world. In spite of the advances in both diagnosis and therapy it is the third leading cause of cancer death worldwide [1]. Of recorded cases, 50% of the world's total occur in Asian countries, including China [2]. Most patients with GC are diagnosed in the advanced stages of disease with lymph node or distant metastases [3]. Identifying suitable biomarkers for the early detection and diagnosis of patients with GC is the subject of much research. Forkhead box M1 (FoxM1) plays multiple roles as a member of the forkhead family of transcription factors in cell proliferation, DNA repair and cell cycle progression at the G1-S and G2-M phases with controlling expression of essential genes and the progression of mitosis [4,5]. FoxM1 has been linked to tumorigenesis and progression of several kinds of malignancies. Expression of FoxM1 has been detected in
various cancers including liver, lung and breast [6]. Recently, the expression of FoxM1 in GC and the relationship between expression of this gene and clinical parameters, such as gender, tumors' differentiation, depth of invasion, lymph node metastasis and TNM stage have been studied, with inconsistent findings. Hu et al. reported that overexpression of FoxM1 was positively correlated with advanced TNM stage and lymphatic metastasis [7]. However, Okada et al. indicated that expression of FoxM1 was not correlated with any clinical parameters [8]. Considering the inconsistent results of current findings, we have undertaken a meta-analysis with all eligible studies to explore the relationship between expression of FoxM1 and both clinical and pathological parameters of patients with GC.
∗ Corresponding author. Gastrointestinal Surgery, The Dalian Municipal Center Hospital Affiliated of Dalian Medical University, No. 826 Southwest Road, Shahekou District of Dalian, 116027, China. ∗∗ Corresponding author. Gastrointestinal Surgery, The Dalian Municipal Center Hospital Affiliated of Dalian Medical University, No. 826 Southwest Road, Shahekou District of Dalian, 116027, China. E-mail addresses: [email protected] (G. Zu), [email protected] (X. Zhang). 1 Co-first author.
http://dx.doi.org/10.1016/j.ijsu.2017.09.076 Received 2 August 2017; Received in revised form 30 August 2017; Accepted 12 September 2017 Available online 07 October 2017 1743-9191/ © 2017 Published by Elsevier Ltd on behalf of IJS Publishing Group Ltd.
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Fig. 1. Flow diagram showing selection of articles.
Table 1 Characteristics of included studies. No.
Author
Year
Journal
Country
Study period
Cases no.
M/F
NOS
1 2 3 4 5 6 7 8
Hu [7] Okada [8] Li [9] Li [10] Zheng [11] Zhang [12] Yang [13] Feng [14]
2015 2013 2013 2009 2013 2015 2009 2012
Biochimica et Biophysica Acta Ann SurgOncol ClinExp Med Cancer Res Chin J Gastroenterol Hepatol Chin J Curr Adv Gen Surg Journal of Clinical and Experimental Medicine Chinese Journal of Clinical oncology
China Japan China China China China China China
2010–2011 2001–2008 2007–2008 n.r. n.r. 2012–2013 2014–2015 2010–2011.
40 77 103 62 98 39 40 70
26/14 53/24 81/22 43/19 74/24 23/16 32/8 55/15
7 8 7 7 6 7 6 7
NOS, Newcastle-Ottawa Scale. n.r., not reported.
Fig. 2. Forest plots showed that the association between expression of FoxM1 and gender (a: forest plots, b: funnel plots).
cancer” or “stomach neoplasms” or “gastric cancer”) and (“survival” or “prognosis”) and (“FoxM1”, “FoxM1a” or “FoxM1b” or “FoxM1c”). Manual reference checks of accepted papers in recent reviews and included papers were performed to supply the electronic searches.
2. Methods 2.1. Search strategy Searching from PubMed, EMBASE, Elsevier, Web of Science, the Cochrane Library, Chinese electronic databases (CNKI database, WanFang database, and VIP database) and the Cochrane Library, the ultimate search date was July 1, 2017. NO language was restricted with the following terms and keywords: (“gastric carcinoma” or “stomach
2.2. Inclusion criteria and exclusion criteria The following inclusion criteria of all articles identified in the literature search ware: (1) the patients were histologically diagnosed with 39
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Fig. 3. Forest plots showed that the association between expression of FoxM1 and tumors' differentiation (a: forest plots, b: funnel plots).
Fig. 4. Forest plots showed that the association between FoxM1 expression and depth of invasion. (a: forest plots, b: funnel plots).
Fig. 5. Forest plots showed that the association between FoxM1 expression and lymph node metastasis (a: forest plots, b: funnel plots).
patients received adjuvant chemoradiation.
GC; (2) the relationship between expression of FoxM1 and clinical parameters of patients with GC was performed; (3) only studies assessed identical target factors were included to control between studies variability. The exclusion criteria were as follows: (1) animal research; (2) nonoriginal research (reviews, comments, letters and case reports); (3) repeated studies were based on the same database or patients; (4)
2.3. Data extraction Two authors identified and screened the search findings. The titles and abstracts were screened for potentially eligible studies. Full-text articles were obtained for detailed evaluation. When studies were 40
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Fig. 6. Forest plots showed that the association between FoxM1 expression and TNM stage (a: forest plots, b: funnel plots).
Table 2 Patient OS in relationship with Expression of FoxM1. Author(year)
Li (2013) Okada (2013) Hu (2015) Li (2009) Zhang (2015) Zheng (2014) Overall
1 Year
3 Years
5 Years
OR(95%CI)
Weight
OR(95%CI)
Weight
OR(95%CI)
Weight
0.170(0.009–3.030) 0.180(0.010–3.389) 0.917(0.146–5.757) 0.093(0.019–0.454) 0.240(0.025–2.286) 0.235(0.064–0.859) 0.218(0.103–0.459)
10.47% 8.77% 6.17% 33.84% 9.89% 30.86% 100%
0.287(0.097–0.852) 0.210(0.044–1.002) 0.347(0.090–1.338) 0.025(0.005–0.116) – – 0.178(0.093–0.340)
27.35% 18.47% 14.55% 39.63% – – 100%
0.302(0.113–0.806) 0.139(0.029–0.652) 0.247(0.062–0.976) 0.044(0.008–0.236) – – 0.180(0.095–0.341)
30.18% 25.14% 16.97% 27.71% – – 100%
Relative risks refer to no relationship in survival for patients with expression of FoxM1. One year: heterogeneity χ2 = 3.52; P = 0.620; I2 = 0.0%; z-test = 4.01; P = 0.000; Three years: heterogeneity χ2 = 8.00; P = 0.046; I2 = 62.5%; z-test = 5.23; P = 0.000; Five years: heterogeneity χ2 = 4.09; P = 0.252; I2 = 26.6%; z-test = 5.27; P = 0.000.
included in this meta-analysis.
conducted in the same institution, we included either the study of better quality or the more recent publication. Two reviewers independently extracted the following data from each identified study: first author, year of publication, details of where the studies were conducted, study period, sample size, gender, tumors' differentiation, depth of invasion, lymph node metastasis and TNM stage and overall survival (OS). Newcastle-Ottawa Scale (NOS) scores ranged from 0 to 9, and a score equal to or higher than 6 indicates high quality.
The pooled analysis enrolled 427 patients from six studies, including 283 males and 144 females. Meta-analysis of available covariates showed no difference in expression between the sexes. (OR:1.143, 95%CI: 0.726–1.799, P = 0.564) (Fig. 2).
2.4. Statistical analysis
3.3. Expression of FoxM1 and tumors' differentiation
This meta-analysis was conducted using STATA statistics software (Version 10.0). The odds ratio (OR) and hazard ratios (HR) with 95% confidence intervals were computed using fixed- or random-effects models to evaluate relevant clinical outcomes. Statistical heterogeneity between trials was evaluated by the χ2 test (P < 0.100 was considered to be significant) and I2 values. An I2 value of 50% or greater indicated the presence of heterogeneity. P < 0.05 was considered to be statistically significant.
Tumors' differentiation (well-moderate group or poor-undifferentiated group) was reported in 428 gastric patients from five studies was available. No significant association with FoxM1 expression was found (OR: 0.991, 95%CI: 0.624–1.575, P = 0.971) (Fig. 3).
3.2. Expression of FoxM1 and gender
3.4. Expression of FoxM1 and depth of invasion A total of 387 patients in five studies were included to detect the relationship between depth of invasion (T1-T2 group and T3-T4 group) and expression of FoxM1 in this meta-analysis. We found an association between depth of invasion and expression of FoxM1 (OR: 0.617, 95%CI: 0.382–0.998, P = 0.049) (Fig. 4).
3. Results 3.1. Characteristics of studies included in meta-analysis
3.5. Expression of FoxM1 and lymph node metastasis
Initially, we identified 83 potential articles. 4 records data were repeated from the same population and were eliminated. Among the 79 articles, 71 articles were excluded, including other cancers (N = 10), no clinical parameters (N = 32), review only (N = 4) and cell experiment (N = 25). Eight articles were proved to be within the scope of the study and all followed an observational design, scoring 6 or more on the NOS (Fig. 1 and Table 1) [7–14]. A total of 529 patients were
Five studies with a total of 387 patients report the association between lymph node involvement and expression of FoxM1 (OR: 2.084, 95%CI: 1.305–3.328, P = 0.002) (Fig. 5). The meta-analysis indicates that expression of FoxM1 is related to the presence of lymph node metastases. 41
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Fig. 7. Forest plots showed that the association between FoxM1 expression and one year survival rate (a: forest plots, b: funnel plots), three years survival rate (c: forest plots, d: funnel plots), five years survival rate (e: forest plots, f: funnel plots).
with a poor prognosis in GC patients (one year OS: OR: 0.218, 95%CI: 0.103–0.459, P = 0.000; three years OS: OR: 0.178, 95%CI: 0.093–0.340, P = 0.000; five years OS: OR: 0.180, 95%CI:0.095–0.341, P = 0.000) (Fig. 7).
3.6. Expression of FoxM1 and TNM stage We pooled five studies with 292 patients and found an association between expression of FoxM1 and the tumor stage of GC (I-II group and III-IV group) (OR: 0.482, 95%CI: 0.275–0.845, P = 0.011) in Fig. 6.
4. Discussion 3.7. Expression of FoxM1 and OS GC is a leading cause of death and has a poor prognosis compared to many other cancers. In China, patients with advanced GC have only a median OS of less than 12 months [15,16]. Although diagnosis and treatment has improved, many patients are diagnosed with GC at an
The combined analysis of six studies with 419 patients were assessed for the correlation between expression of FoxM1 and OS of GC. The results (Table 2) indicated that expression of FoxM1 was associated 42
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Funding
advanced stage and have a poor prognosis. Recently, suitable biomarkers for early detection and diagnosis have attracted much attention. In this meta-analysis we found that expression of FoxM1 was associated with lymph node metastasis, depth of invasion, TNM stage and poor OS. However, we did not observe significant differences of FoxM1 expression with gender and tumors' differentiation. As a biomarker FoxM1 may have a role in predicting patient prognosis. FoxM1 plays an essential role in cell proliferation and progression of the cell cycle, is an oncofetal transcription factor. It involves in the processes of cell migration, invasion, angiogenesis and cellular senescence. Defects in these processes can affect tumor initiation, progression and metastasis [17,18]. Several studies reported the relationship between expression of FoxM1 and TNM stage, depth of invasion and lymph node metastasis, but the findings were inconsistent. Okada et al. had indicated that the expression of FoxM1 had no significant association with clinical pathological features, including TNM stage, depth of invasion and lymph node metastasis [8]. On the contrary, Hu and colleagues reported that expression of FoxM1 was positive correlated with both TNM stage and lymph node metastasis [7]. Chen et al. also found that expression of FoxM1b in primary gastric tumors significantly enhanced the incidence of lymph node metastasis [17]. In our metaanalysis the expression of FoxM1 was associated with TNM stage, depth of invasion and lymph node involvement of GC patients. Recent studies have looked at the relationship between expression of FoxM1 and gender and tumors' differentiation. Li et al. reported a total of 103 patients including 81 males and 22 females finding that expression of FoxM1 was not associated with either gender and tumors' differentiation [9]. However, Zheng et al. combining analysis of 98 patients indicated that expression of FoxM1 was positively correlated with tumors' differentiation [11]. Our meta-analysis reports that expression of FoxM1 is not associated with tumors' differentiation or gender in the patients with GC. FoxM1 is important in cell cycle at the G1-S and G2-M phases with controlling expression of essential genes and the progression of mitosis [5]. It impacts tumor initiation, progression, metastasis, angiogenesis and drug resistance [16]. A comprehensive genomics-based study included 18,000 cases with OS data for patients with various cancers reported that expression of FoxM1 was most frequently associated with an adverse outcome [19]. A meta-analysis also found a significant relationship between high expression of FoxM1 and poor outcome of human solid tumors [20]. Available evidence supports the relationship between high expression of FoxM1 and poor OS in patients with GC [8–10]. In this study we observed that high expression of FoxM1 was associated with reduced survival at one, three and five years. This meta-analysis clearly has limitations. The studies included were small both in number and overall patients. The populations studied came from China and Japan and no reported studies from outside these areas were found. Furthermore, some data was gained from assessing survival curves rather than individual patient data, and the heterogeneity should not be neglected. The findings however support further evaluation as FoxM1 as a biomarker for gastric cancer and a possible prognostic indicator. Larger studies re clearly needed to validate the association of FoxM1 expression with the GC-related clinicopathological characteristics. In conclusion, this meta-analysis reports that expression of FoxM1 was not associated with gender or tumors' differentiation. However, expression of FoxM1 is associated with TNM stage, depth of invasion, lymph node metastasis and an overall poor prognosis in patients with GC.
This work was supported by grants from the National Natural Science Foundation of China (No. 81700465). Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors. Research registration unique identifying number (UIN) reviewregistry296. Author contribution G. Zu and X. Zhang designed the research; D. Jiang and L. Jiang conducted the research; B. Liu, H. Huang, T. Zhang and W. Li analyzed data; D. Jiang and L. Jiang wrote the draft; all authors read, reviewed and approved the final manuscript. G. Zu and X. Zhang had primary responsibility for final content. Guarantor D. Jiang Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx. doi.org/10.1016/j.ijsu.2017.09.076. References [1] L.A. Torre, F. Bray, R.L. Siegel, J. Ferlay, J. Lortet-Tieulent, A. Jemal, Global cancer statistics, 2012, CA Cancer J. Clin. 65 (2) (2015) 87–108. [2] J. Ferlay, I. Soerjomataram, R. Dikshit, S. Eser, C. Mathers, M. Rebelo, D.M. Parkin, D. Forman, F. Bray, Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012, Int. J. Cancer 136 (5) (2015) E359–E386. [3] E. Van Cutsem, X. Sagaert, B. Topal, K. Haustermans, H. Prenen, Gastric cancer, Lancet 388 (10060) (2016) 2654–2664. [4] S.S. Myatt, E.W. Lam, Targeting FOXM1, Nat. Rev. Cancer 8 (3) (2008) 242. [5] M. Alvarez-Fernández, R.H. Medema, Novel functions of FoxM1: from molecular mechanisms to cancer therapy, Front. Oncol. 3 (2013) 30. [6] Christian Pilarsky, Michael Wenzig, Thomas Specht, Hans Detlev Saeger, Robert Grützmann, Identification and validation of commonly overexpressed genes in solid tumors by comparison of microarray data, Neoplasia 6 (6) (2004) 744–750. [7] C.J. Hu, B. Wang, B. Tang, B.J. Chen, Y.F. Xiao, Y. Qin, X. Yong, G. Luo, J.W. Zhang, D. Zhang, S. Li, F. He, S.M. Yang, The FOXM1-induced resistance to oxaliplatin is partially mediated by its novel target gene Mcl-1 in gastric cancer cells, Biochim. Biophys. Acta 1849 (3) (2015) 290–299. [8] K. Okada, Y. Fujiwara, T. Takahashi, Y. Nakamura, S. Takiguchi, K. Nakajima, H. Miyata, M. Yamasaki, Y. Kurokawa, M. Mori, Y. Doki, Overexpression of forkhead box M1 transcription factor (FOXM1) is a potential prognostic marker and enhances chemoresistance for docetaxel in gastric cancer, Ann. Surg. Oncol. 20 (3) (2013) 1035–1043. [9] X. Li, D. Tang, Y. Yao, W. Qi, J. Liang, Clinical significance and positive correlation of FoxM1 and Her-2 expression in gastric cancer, Clin. Exp. Med. 14 (4) (2014) 447–455. [10] Q. Li, N. Zhang, Z. Jia, X. Le, B. Dai, D. Wei, S. Huang, D. Tan, K. Xie, Critical role and regulation of transcription factor FoxM1 in human gastric cancer angiogenesis and progression, Cancer Res. 69 (8) (2009) 3501–3509. [11] J. Zheng, L. Zhang, L. Li, Effects of Gli1 and Foxm1 on the metastasis and prognosis of gastric cancer, Chin. J. Gastroenterol. Hepatol. 23 (4) (2014) 1394–1404. [12] Y. Zhang, Y.X. Chai, M. Han, X.L. Qu, Expression of FOXM1 in gastric carcinoma and its clinical significance, Chin. J. Curr. Adv. Gen. Surg. 18 (10) (2015) 827–829. [13] S.F. Yang, N. Sheng, Expression changes and clinical significance of serum D-D, Foxm1, Gli1 and CD147 in patients with gastric cancer, 16 (6) (2017) 543–546. [14] Y.G. Feng, Z.Q. Wei, S.B. Zeng, Y. Pan, Expression of Gli1 and Foxm1 in gastric carcer and its clinical significance, Chin. J. Clin. Oncol. 39 (7) (2012) 377–381. [15] J. Ferlay, I. Soerjomataram, R. Dikshit, S. Eser, C. Mathers, M. Rebelo, D.M. Parkin, D. Forman, F. Bray, Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012, Int. J. Cancer 136 (5) (2015) E359–E386. [16] D. Cunningham, N. Starling, S. Rao, T. Iveson, M. Nicolson, F. Coxon, G. Middleton, F. Daniel, J. Oates, A.R. Norman, Capecitabine and oxaliplatin for advanced esophagogastric cancer, N. Engl. J. Med. 362 (9) (2010) 858–859.
Conflict of interest The authors declare no conflict of interest.
43
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Y. Xu, A. Khuong, C.D. Hoang, M. Diehn, R.B. West, S.K. Plevritis, A.A. Alizadeh, The prognostic landscape of genes and infiltrating immune cells across human cancers, Nat. Med. 21 (8) (2015) 938–945. [20] L. Li, D. Wu, Q. Yu, L. Li, P. Wu, Prognostic value of FOXM1 in solid tumors: a systematic review and meta-analysis, Oncotarget 8 (19) (2017) 32298–32308.
[17] H. Chen, Y. Zou, H. Yang, J. Wang, H. Pan, Downregulation of FoxM1 inhibits proliferation, invasion and angiogenesis of HeLa cells in vitro and in vivo, Int. J. Oncol. 45 (6) (2014) 2355–2364. [18] L. Bella, S. Zona, G. Nestal de Moraes, E.W. Lam, FOXM1: a key oncofoetal transcription factor in health and disease, Sem. Cancer Biol. 29 (2014) 32–39. [19] A.J. Gentles, A.M. Newman, C.L. Liu, S.V. Bratman, W. Feng, D. Kim, V.S. Nair,
44
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.elsevier.com/locate/ijsu
Review
Clinicopathological and prognostic significance of FoxM1 in gastric cancer: A meta-analysis
MARK
Dongdong Jianga,b,1, Lu Jianga,b,1, Baiying Liua,b, He Huanga,b, Wenbin Lia, Taotao Zhanga, Guo Zua,∗, Xiangwen Zhanga,∗∗ a b
Department of Gastroenterology Surgery, The Dalian Municipal Central Hospital Affiliated of Dalian Medical University, Dalian 116033, China Dalian Medical University, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Prognosis Clinical parameters FoxM1 Gastric cancer Meta-analysis
Background: The relationship between expression of FoxM1 and clinical parameters of patients with gastric cancer (GC) has yet to be fully established. Methods: A systematic search was performed. Odds ratio (OR) and confidence interval (CI) were used to assess association between expression of FoxM1 and clinical parameters and the prognostic value of patients with GC. Results: Eight studies involving 529 patients with GC were identified. Overall, the pooled results showed that expression of FoxM1 was associated with TNM stage (OR: 0.482, 95%CI: 0.275–0.845, P = 0.011), depth of invasion (OR: 0.617, 95%CI: 0.382–0.998, P = 0.049) and lymph node metastasis (OR: 2.084, 95%CI: 1.305–3.328, P = 0.002) in the patients with GC. Whereas, expression of FoxM1 was not associated with gender (OR: 1.143, 95%CI: 0.726–1.799, P = 0.564) and tumors' differentiation (OR: 0.991, 95%CI: 0.624–1.575, P = 0.971) of GC. Expression of FoxM1 was also associated with poor prognosis of overall survival (OS) in the patients with GC (one year OS: OR: 0.218, 95%CI: 0.103–0.459, P = 0.000; three years OS: OR: 0.178, 95%CI: 0.093–0.340, P = 0.000; five years OS: OR: 0.180, 95%CI: 0.095–0.341, P = 0.000). Conclusion: Expression of FoxM1 is associated with TNM stage, depth of invasion, lymph node metastasis and poor prognosis of the patients with GC.
1. Introduction Gastric cancer (GC) is the fifth most common malignant cancer in the world. In spite of the advances in both diagnosis and therapy it is the third leading cause of cancer death worldwide [1]. Of recorded cases, 50% of the world's total occur in Asian countries, including China [2]. Most patients with GC are diagnosed in the advanced stages of disease with lymph node or distant metastases [3]. Identifying suitable biomarkers for the early detection and diagnosis of patients with GC is the subject of much research. Forkhead box M1 (FoxM1) plays multiple roles as a member of the forkhead family of transcription factors in cell proliferation, DNA repair and cell cycle progression at the G1-S and G2-M phases with controlling expression of essential genes and the progression of mitosis [4,5]. FoxM1 has been linked to tumorigenesis and progression of several kinds of malignancies. Expression of FoxM1 has been detected in
various cancers including liver, lung and breast [6]. Recently, the expression of FoxM1 in GC and the relationship between expression of this gene and clinical parameters, such as gender, tumors' differentiation, depth of invasion, lymph node metastasis and TNM stage have been studied, with inconsistent findings. Hu et al. reported that overexpression of FoxM1 was positively correlated with advanced TNM stage and lymphatic metastasis [7]. However, Okada et al. indicated that expression of FoxM1 was not correlated with any clinical parameters [8]. Considering the inconsistent results of current findings, we have undertaken a meta-analysis with all eligible studies to explore the relationship between expression of FoxM1 and both clinical and pathological parameters of patients with GC.
∗ Corresponding author. Gastrointestinal Surgery, The Dalian Municipal Center Hospital Affiliated of Dalian Medical University, No. 826 Southwest Road, Shahekou District of Dalian, 116027, China. ∗∗ Corresponding author. Gastrointestinal Surgery, The Dalian Municipal Center Hospital Affiliated of Dalian Medical University, No. 826 Southwest Road, Shahekou District of Dalian, 116027, China. E-mail addresses: [email protected] (G. Zu), [email protected] (X. Zhang). 1 Co-first author.
http://dx.doi.org/10.1016/j.ijsu.2017.09.076 Received 2 August 2017; Received in revised form 30 August 2017; Accepted 12 September 2017 Available online 07 October 2017 1743-9191/ © 2017 Published by Elsevier Ltd on behalf of IJS Publishing Group Ltd.
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Fig. 1. Flow diagram showing selection of articles.
Table 1 Characteristics of included studies. No.
Author
Year
Journal
Country
Study period
Cases no.
M/F
NOS
1 2 3 4 5 6 7 8
Hu [7] Okada [8] Li [9] Li [10] Zheng [11] Zhang [12] Yang [13] Feng [14]
2015 2013 2013 2009 2013 2015 2009 2012
Biochimica et Biophysica Acta Ann SurgOncol ClinExp Med Cancer Res Chin J Gastroenterol Hepatol Chin J Curr Adv Gen Surg Journal of Clinical and Experimental Medicine Chinese Journal of Clinical oncology
China Japan China China China China China China
2010–2011 2001–2008 2007–2008 n.r. n.r. 2012–2013 2014–2015 2010–2011.
40 77 103 62 98 39 40 70
26/14 53/24 81/22 43/19 74/24 23/16 32/8 55/15
7 8 7 7 6 7 6 7
NOS, Newcastle-Ottawa Scale. n.r., not reported.
Fig. 2. Forest plots showed that the association between expression of FoxM1 and gender (a: forest plots, b: funnel plots).
cancer” or “stomach neoplasms” or “gastric cancer”) and (“survival” or “prognosis”) and (“FoxM1”, “FoxM1a” or “FoxM1b” or “FoxM1c”). Manual reference checks of accepted papers in recent reviews and included papers were performed to supply the electronic searches.
2. Methods 2.1. Search strategy Searching from PubMed, EMBASE, Elsevier, Web of Science, the Cochrane Library, Chinese electronic databases (CNKI database, WanFang database, and VIP database) and the Cochrane Library, the ultimate search date was July 1, 2017. NO language was restricted with the following terms and keywords: (“gastric carcinoma” or “stomach
2.2. Inclusion criteria and exclusion criteria The following inclusion criteria of all articles identified in the literature search ware: (1) the patients were histologically diagnosed with 39
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Fig. 3. Forest plots showed that the association between expression of FoxM1 and tumors' differentiation (a: forest plots, b: funnel plots).
Fig. 4. Forest plots showed that the association between FoxM1 expression and depth of invasion. (a: forest plots, b: funnel plots).
Fig. 5. Forest plots showed that the association between FoxM1 expression and lymph node metastasis (a: forest plots, b: funnel plots).
patients received adjuvant chemoradiation.
GC; (2) the relationship between expression of FoxM1 and clinical parameters of patients with GC was performed; (3) only studies assessed identical target factors were included to control between studies variability. The exclusion criteria were as follows: (1) animal research; (2) nonoriginal research (reviews, comments, letters and case reports); (3) repeated studies were based on the same database or patients; (4)
2.3. Data extraction Two authors identified and screened the search findings. The titles and abstracts were screened for potentially eligible studies. Full-text articles were obtained for detailed evaluation. When studies were 40
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Fig. 6. Forest plots showed that the association between FoxM1 expression and TNM stage (a: forest plots, b: funnel plots).
Table 2 Patient OS in relationship with Expression of FoxM1. Author(year)
Li (2013) Okada (2013) Hu (2015) Li (2009) Zhang (2015) Zheng (2014) Overall
1 Year
3 Years
5 Years
OR(95%CI)
Weight
OR(95%CI)
Weight
OR(95%CI)
Weight
0.170(0.009–3.030) 0.180(0.010–3.389) 0.917(0.146–5.757) 0.093(0.019–0.454) 0.240(0.025–2.286) 0.235(0.064–0.859) 0.218(0.103–0.459)
10.47% 8.77% 6.17% 33.84% 9.89% 30.86% 100%
0.287(0.097–0.852) 0.210(0.044–1.002) 0.347(0.090–1.338) 0.025(0.005–0.116) – – 0.178(0.093–0.340)
27.35% 18.47% 14.55% 39.63% – – 100%
0.302(0.113–0.806) 0.139(0.029–0.652) 0.247(0.062–0.976) 0.044(0.008–0.236) – – 0.180(0.095–0.341)
30.18% 25.14% 16.97% 27.71% – – 100%
Relative risks refer to no relationship in survival for patients with expression of FoxM1. One year: heterogeneity χ2 = 3.52; P = 0.620; I2 = 0.0%; z-test = 4.01; P = 0.000; Three years: heterogeneity χ2 = 8.00; P = 0.046; I2 = 62.5%; z-test = 5.23; P = 0.000; Five years: heterogeneity χ2 = 4.09; P = 0.252; I2 = 26.6%; z-test = 5.27; P = 0.000.
included in this meta-analysis.
conducted in the same institution, we included either the study of better quality or the more recent publication. Two reviewers independently extracted the following data from each identified study: first author, year of publication, details of where the studies were conducted, study period, sample size, gender, tumors' differentiation, depth of invasion, lymph node metastasis and TNM stage and overall survival (OS). Newcastle-Ottawa Scale (NOS) scores ranged from 0 to 9, and a score equal to or higher than 6 indicates high quality.
The pooled analysis enrolled 427 patients from six studies, including 283 males and 144 females. Meta-analysis of available covariates showed no difference in expression between the sexes. (OR:1.143, 95%CI: 0.726–1.799, P = 0.564) (Fig. 2).
2.4. Statistical analysis
3.3. Expression of FoxM1 and tumors' differentiation
This meta-analysis was conducted using STATA statistics software (Version 10.0). The odds ratio (OR) and hazard ratios (HR) with 95% confidence intervals were computed using fixed- or random-effects models to evaluate relevant clinical outcomes. Statistical heterogeneity between trials was evaluated by the χ2 test (P < 0.100 was considered to be significant) and I2 values. An I2 value of 50% or greater indicated the presence of heterogeneity. P < 0.05 was considered to be statistically significant.
Tumors' differentiation (well-moderate group or poor-undifferentiated group) was reported in 428 gastric patients from five studies was available. No significant association with FoxM1 expression was found (OR: 0.991, 95%CI: 0.624–1.575, P = 0.971) (Fig. 3).
3.2. Expression of FoxM1 and gender
3.4. Expression of FoxM1 and depth of invasion A total of 387 patients in five studies were included to detect the relationship between depth of invasion (T1-T2 group and T3-T4 group) and expression of FoxM1 in this meta-analysis. We found an association between depth of invasion and expression of FoxM1 (OR: 0.617, 95%CI: 0.382–0.998, P = 0.049) (Fig. 4).
3. Results 3.1. Characteristics of studies included in meta-analysis
3.5. Expression of FoxM1 and lymph node metastasis
Initially, we identified 83 potential articles. 4 records data were repeated from the same population and were eliminated. Among the 79 articles, 71 articles were excluded, including other cancers (N = 10), no clinical parameters (N = 32), review only (N = 4) and cell experiment (N = 25). Eight articles were proved to be within the scope of the study and all followed an observational design, scoring 6 or more on the NOS (Fig. 1 and Table 1) [7–14]. A total of 529 patients were
Five studies with a total of 387 patients report the association between lymph node involvement and expression of FoxM1 (OR: 2.084, 95%CI: 1.305–3.328, P = 0.002) (Fig. 5). The meta-analysis indicates that expression of FoxM1 is related to the presence of lymph node metastases. 41
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Fig. 7. Forest plots showed that the association between FoxM1 expression and one year survival rate (a: forest plots, b: funnel plots), three years survival rate (c: forest plots, d: funnel plots), five years survival rate (e: forest plots, f: funnel plots).
with a poor prognosis in GC patients (one year OS: OR: 0.218, 95%CI: 0.103–0.459, P = 0.000; three years OS: OR: 0.178, 95%CI: 0.093–0.340, P = 0.000; five years OS: OR: 0.180, 95%CI:0.095–0.341, P = 0.000) (Fig. 7).
3.6. Expression of FoxM1 and TNM stage We pooled five studies with 292 patients and found an association between expression of FoxM1 and the tumor stage of GC (I-II group and III-IV group) (OR: 0.482, 95%CI: 0.275–0.845, P = 0.011) in Fig. 6.
4. Discussion 3.7. Expression of FoxM1 and OS GC is a leading cause of death and has a poor prognosis compared to many other cancers. In China, patients with advanced GC have only a median OS of less than 12 months [15,16]. Although diagnosis and treatment has improved, many patients are diagnosed with GC at an
The combined analysis of six studies with 419 patients were assessed for the correlation between expression of FoxM1 and OS of GC. The results (Table 2) indicated that expression of FoxM1 was associated 42
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Funding
advanced stage and have a poor prognosis. Recently, suitable biomarkers for early detection and diagnosis have attracted much attention. In this meta-analysis we found that expression of FoxM1 was associated with lymph node metastasis, depth of invasion, TNM stage and poor OS. However, we did not observe significant differences of FoxM1 expression with gender and tumors' differentiation. As a biomarker FoxM1 may have a role in predicting patient prognosis. FoxM1 plays an essential role in cell proliferation and progression of the cell cycle, is an oncofetal transcription factor. It involves in the processes of cell migration, invasion, angiogenesis and cellular senescence. Defects in these processes can affect tumor initiation, progression and metastasis [17,18]. Several studies reported the relationship between expression of FoxM1 and TNM stage, depth of invasion and lymph node metastasis, but the findings were inconsistent. Okada et al. had indicated that the expression of FoxM1 had no significant association with clinical pathological features, including TNM stage, depth of invasion and lymph node metastasis [8]. On the contrary, Hu and colleagues reported that expression of FoxM1 was positive correlated with both TNM stage and lymph node metastasis [7]. Chen et al. also found that expression of FoxM1b in primary gastric tumors significantly enhanced the incidence of lymph node metastasis [17]. In our metaanalysis the expression of FoxM1 was associated with TNM stage, depth of invasion and lymph node involvement of GC patients. Recent studies have looked at the relationship between expression of FoxM1 and gender and tumors' differentiation. Li et al. reported a total of 103 patients including 81 males and 22 females finding that expression of FoxM1 was not associated with either gender and tumors' differentiation [9]. However, Zheng et al. combining analysis of 98 patients indicated that expression of FoxM1 was positively correlated with tumors' differentiation [11]. Our meta-analysis reports that expression of FoxM1 is not associated with tumors' differentiation or gender in the patients with GC. FoxM1 is important in cell cycle at the G1-S and G2-M phases with controlling expression of essential genes and the progression of mitosis [5]. It impacts tumor initiation, progression, metastasis, angiogenesis and drug resistance [16]. A comprehensive genomics-based study included 18,000 cases with OS data for patients with various cancers reported that expression of FoxM1 was most frequently associated with an adverse outcome [19]. A meta-analysis also found a significant relationship between high expression of FoxM1 and poor outcome of human solid tumors [20]. Available evidence supports the relationship between high expression of FoxM1 and poor OS in patients with GC [8–10]. In this study we observed that high expression of FoxM1 was associated with reduced survival at one, three and five years. This meta-analysis clearly has limitations. The studies included were small both in number and overall patients. The populations studied came from China and Japan and no reported studies from outside these areas were found. Furthermore, some data was gained from assessing survival curves rather than individual patient data, and the heterogeneity should not be neglected. The findings however support further evaluation as FoxM1 as a biomarker for gastric cancer and a possible prognostic indicator. Larger studies re clearly needed to validate the association of FoxM1 expression with the GC-related clinicopathological characteristics. In conclusion, this meta-analysis reports that expression of FoxM1 was not associated with gender or tumors' differentiation. However, expression of FoxM1 is associated with TNM stage, depth of invasion, lymph node metastasis and an overall poor prognosis in patients with GC.
This work was supported by grants from the National Natural Science Foundation of China (No. 81700465). Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors. Research registration unique identifying number (UIN) reviewregistry296. Author contribution G. Zu and X. Zhang designed the research; D. Jiang and L. Jiang conducted the research; B. Liu, H. Huang, T. Zhang and W. Li analyzed data; D. Jiang and L. Jiang wrote the draft; all authors read, reviewed and approved the final manuscript. G. Zu and X. Zhang had primary responsibility for final content. Guarantor D. Jiang Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx. doi.org/10.1016/j.ijsu.2017.09.076. References [1] L.A. Torre, F. Bray, R.L. Siegel, J. Ferlay, J. Lortet-Tieulent, A. Jemal, Global cancer statistics, 2012, CA Cancer J. Clin. 65 (2) (2015) 87–108. [2] J. Ferlay, I. Soerjomataram, R. Dikshit, S. Eser, C. Mathers, M. Rebelo, D.M. Parkin, D. Forman, F. Bray, Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012, Int. J. Cancer 136 (5) (2015) E359–E386. [3] E. Van Cutsem, X. Sagaert, B. Topal, K. Haustermans, H. Prenen, Gastric cancer, Lancet 388 (10060) (2016) 2654–2664. [4] S.S. Myatt, E.W. Lam, Targeting FOXM1, Nat. Rev. Cancer 8 (3) (2008) 242. [5] M. Alvarez-Fernández, R.H. Medema, Novel functions of FoxM1: from molecular mechanisms to cancer therapy, Front. Oncol. 3 (2013) 30. [6] Christian Pilarsky, Michael Wenzig, Thomas Specht, Hans Detlev Saeger, Robert Grützmann, Identification and validation of commonly overexpressed genes in solid tumors by comparison of microarray data, Neoplasia 6 (6) (2004) 744–750. [7] C.J. Hu, B. Wang, B. Tang, B.J. Chen, Y.F. Xiao, Y. Qin, X. Yong, G. Luo, J.W. Zhang, D. Zhang, S. Li, F. He, S.M. Yang, The FOXM1-induced resistance to oxaliplatin is partially mediated by its novel target gene Mcl-1 in gastric cancer cells, Biochim. Biophys. Acta 1849 (3) (2015) 290–299. [8] K. Okada, Y. Fujiwara, T. Takahashi, Y. Nakamura, S. Takiguchi, K. Nakajima, H. Miyata, M. Yamasaki, Y. Kurokawa, M. Mori, Y. Doki, Overexpression of forkhead box M1 transcription factor (FOXM1) is a potential prognostic marker and enhances chemoresistance for docetaxel in gastric cancer, Ann. Surg. Oncol. 20 (3) (2013) 1035–1043. [9] X. Li, D. Tang, Y. Yao, W. Qi, J. Liang, Clinical significance and positive correlation of FoxM1 and Her-2 expression in gastric cancer, Clin. Exp. Med. 14 (4) (2014) 447–455. [10] Q. Li, N. Zhang, Z. Jia, X. Le, B. Dai, D. Wei, S. Huang, D. Tan, K. Xie, Critical role and regulation of transcription factor FoxM1 in human gastric cancer angiogenesis and progression, Cancer Res. 69 (8) (2009) 3501–3509. [11] J. Zheng, L. Zhang, L. Li, Effects of Gli1 and Foxm1 on the metastasis and prognosis of gastric cancer, Chin. J. Gastroenterol. Hepatol. 23 (4) (2014) 1394–1404. [12] Y. Zhang, Y.X. Chai, M. Han, X.L. Qu, Expression of FOXM1 in gastric carcinoma and its clinical significance, Chin. J. Curr. Adv. Gen. Surg. 18 (10) (2015) 827–829. [13] S.F. Yang, N. Sheng, Expression changes and clinical significance of serum D-D, Foxm1, Gli1 and CD147 in patients with gastric cancer, 16 (6) (2017) 543–546. [14] Y.G. Feng, Z.Q. Wei, S.B. Zeng, Y. Pan, Expression of Gli1 and Foxm1 in gastric carcer and its clinical significance, Chin. J. Clin. Oncol. 39 (7) (2012) 377–381. [15] J. Ferlay, I. Soerjomataram, R. Dikshit, S. Eser, C. Mathers, M. Rebelo, D.M. Parkin, D. Forman, F. Bray, Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012, Int. J. Cancer 136 (5) (2015) E359–E386. [16] D. Cunningham, N. Starling, S. Rao, T. Iveson, M. Nicolson, F. Coxon, G. Middleton, F. Daniel, J. Oates, A.R. Norman, Capecitabine and oxaliplatin for advanced esophagogastric cancer, N. Engl. J. Med. 362 (9) (2010) 858–859.
Conflict of interest The authors declare no conflict of interest.
43
International Journal of Surgery 48 (2017) 38–44
D. Jiang et al.
Y. Xu, A. Khuong, C.D. Hoang, M. Diehn, R.B. West, S.K. Plevritis, A.A. Alizadeh, The prognostic landscape of genes and infiltrating immune cells across human cancers, Nat. Med. 21 (8) (2015) 938–945. [20] L. Li, D. Wu, Q. Yu, L. Li, P. Wu, Prognostic value of FOXM1 in solid tumors: a systematic review and meta-analysis, Oncotarget 8 (19) (2017) 32298–32308.
[17] H. Chen, Y. Zou, H. Yang, J. Wang, H. Pan, Downregulation of FoxM1 inhibits proliferation, invasion and angiogenesis of HeLa cells in vitro and in vivo, Int. J. Oncol. 45 (6) (2014) 2355–2364. [18] L. Bella, S. Zona, G. Nestal de Moraes, E.W. Lam, FOXM1: a key oncofoetal transcription factor in health and disease, Sem. Cancer Biol. 29 (2014) 32–39. [19] A.J. Gentles, A.M. Newman, C.L. Liu, S.V. Bratman, W. Feng, D. Kim, V.S. Nair,
44
