Prognostic value of DcR3 in solid tumors: A meta-analysis

Accepted Manuscript Prognostic value of DcR3 in solid tumors: A meta-analysis

Hua Ge, Chaojie Liang, Shulin Ren, Chaosen Yue, Jixiang Wu PII: DOI: Reference:

S0009-8981(18)30107-4 doi:10.1016/j.cca.2018.02.038 CCA 15090

To appear in:

Clinica Chimica Acta

Received date: Revised date: Accepted date:

20 January 2018 27 February 2018 27 February 2018

Please cite this article as: Hua Ge, Chaojie Liang, Shulin Ren, Chaosen Yue, Jixiang Wu , Prognostic value of DcR3 in solid tumors: A meta-analysis. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Cca(2017), doi:10.1016/j.cca.2018.02.038

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Title page

Prognostic value of DcR3 in solid tumors: A meta-analysis

SC

RI

PT

Short title: Prognostic value of DcR3 in solid tumors

NU

Hua Ge, Chaojie Liang, Shulin Ren, Chaosen Yue, Jixiang Wu Department of general surgery, Beijing Tongren hospital, Capital Medical University, Beijing, People’s Republic of China

MA

Corresponding author:

AC

CE

PT E

D

Dr Jixiang Wu Department of general surgery, Beijing Tongren hospital, Capital Medical University, No.2 Chongwenmennei Street, Dongcheng, Beijing 100730, People’s Republic of China. Tel +86 138 0101 5118; E-mail: [email protected]

This article includes 4 figures and 2 tables.

ACCEPTED MANUSCRIPT Background

SC

RI

PT

Cancer constitutes an enormous burden on society and is also a major cause of morbidity and mortality around the world. Based on GLOBOCAN estimates, nearly 14.1 million new cancer cases and 8.2 million deaths occurred in 2012 worldwide [1-2]. Although advances have been made in diagnostic techniques and therapeutic modalities for tumors, the overall clinical outcome in patients with solid tumors remains poor. Invasion and metastasis play very important roles in tumor progression and are difficult to deal with in clinical treatment. It is well established that numerous cancer-related biomolecules are involved in tumor pathology and can serve as biomarkers. However, only a few biomarkers have been verified for clinical usage [3]. Effective biomarkers for early diagnosis and prognosis are still lacking, limiting the treatment of malignant tumors and molecular targeting therapy. Accordingly, there is an increasingly urgent need to identify useful biomarkers that can accurately display the biological features and prognostic outcomes in patients with solid tumors.

AC

CE

PT E

D

MA

NU

The soluble decoy receptor 3 (DcR3), also known as TNF receptor superfamily member 6b (TNFRSF6B), TR6 or M68, is a member of the tumor necrosis factor receptor (TNFR) superfamily and lacks a transmembrane domain in its sequence. DcR3 functions as a kind of anti-apoptotic protein and as a death decoy. There are three known ligands of DcR3, namely the Fas ligand (FasL), lymphotoxin analogues (LIGHT), and tumor necrosis factor-like ligand 1A (TL1A), have been reported [4-6]. DcR3 is almost un-detectable in most of normal individuals. However, aberrant expression of DcR3 has been reported in a variety of human malignancies, such as gastric cancer, colon cancer, liver cancer, breast cancer and bladder cancer [7-11]. Recent studies have indicated that the expression of DcR3 is closely associated with the prognosis of cancer patients. Nevertheless, the reliability and degree of the prognostic impact of DcR3 in solid tumors has not yet been thoroughly analyzed. We thus conducted this meta-analysis to determine the correlation between overexpression of DcR3 and survival in patients with solid tumors and assess the possibility of using DcR3 as a prognostic indicator of survival in patients with solid tumor.

Materials and methods Search strategy An exhaustive literature search was conducted in electronic databases including PubMed, Web of Science, Cochrane Library, EMBASE, Chinese CNKI, and Wan Fang. The last search was conducted on January 1, 2018. The search terms used for literature retrieval were as follows: “DcR3”, “decoy receptor 3”, “TNFRSF6B”, “TR6”, “M68”, “neoplasms”, “tumor” and “prognosis”. The reference lists of the identified articles were also checked to find out other pertinent studies. Eligibility criteria

ACCEPTED MANUSCRIPT The main criteria for the inclusion of studies were: (1) The patients were diagnosed with any type of cancer and directly examined the DcR3 expression status. (2) The study was published in Chinese or English and the full text was available. (3) Expression level of DcR3 was compared to overall survival (OS) or disease-free survival/ recurrence-free survival/ progression-free survival (DFS/RFS/PFS). (4) Hazard ratios (HRs) for OS or DFS/RFS/PFS were provided or could be calculated from the published data.

RI

PT

Studies were excluded according to the following criteria: (1) Ecological studies, case reports, reviews, editorials, letters, conference abstracts, and animal trials. (2) Repeated studies based on the same dataset or patients. (3) Lacking essential information on survival, or unable to calculate HRs based on the data provided.

MA

NU

SC

Data extraction Data were extracted from the included studies by two researchers (Hua Ge and Chaojie Liang), discrepancies were resolved by independently extracting data from the original article by a third researcher (Shulin Ren), and consensus was reached through discussions. We extracted the first author’s name, publication year, sample size, country, type of cancer, detection method, duration of the follow-up, patient characteristics, HR and 95% confidence interval (CI).

PT E

D

Quality assessment The methodological quality of each candidate study was assessed according to the Newcastle-Ottawa Scale (NOS). The total score ranged from 0 to 9. A study with a NOS score of 6 or more was regarded as high quality, while a study with 5 or less score was considered as poor quality. Only studies considered to be of high quality were included in this meta-analysis.

AC

CE

Statistical analysis The meta-analysis was conducted using the RevMan 5.3 and STATA 12.0 software. HRs and 95%CIs were used to evaluate the prognostic value of DcR3 overexpression in human solid tumors. Heterogeneity among studies was evaluated by the χ2 and I2 test. When the result (I2>50% or P<0.05) indicated heterogeneity, the random effects model was used for the meta-analysis. Otherwise, a fixed effects model was used. The software Engauge Digitizer 10.0 was used to extract the survival data from a Kaplan-Meier curve in some articles, and then the Tierney’s method was used to calculate the HRs and 95%CIs [12]. If the HR>1 implied a worse prognosis for the group with positive/high DcR3 expression and would be considered to be statistically significant if the 95%CI did not overlap 1. Potential causes of statistical heterogeneity were examined by subgroup analysis. Publication bias was investigated using funnel plots with the Begg’s test. Sensitivity analysis was also tested by excluding each study individually.

Results

ACCEPTED MANUSCRIPT

PT

Study identification and characteristics According to our defined searching strategy, a total of 519 relevant articles were initially retrieved from the PubMed, Web of Science, Cochrane Library, EMBASE, Chinese CNKI, and WanFang databases. Among them, a total of 96 duplicated studies were excluded. The remaining 423 articles were further evaluated by inspecting the titles and abstracts, and 384 articles were excluded because they were non-DcR3-related, or not tested in tumor tissues. A total of 39 studies were assessed by reading their full texts, and then 23 studies were excluded due to insufficient information. Ultimately, 16 eligible articles [13-28] were included in our meta-analysis. The selection process is shown in Figure 1.

MA

NU

SC

RI

The included studies were published from 2002 to 2017. They comprised a total of 2209 cancer patients from China, Japan, Switzerland, Sweden, United States of America, and Germany who were diagnosed with a variety of cancers, including oral cavity cancer, esophageal cancer, gastric cancer, pancreatic cancer, colorectal cancer, small intestinal neuroendocrine tumor, renal cell carcinoma, bladder urothelial carcinoma, breast cancer, cervical cancer, and epithelial ovarian cancer. Studies’ sample sizes ranged from 44 to 560 patients. Twelve studies were carried out on Asians, and 4 studies on Caucasians. Regarding the quality assessment, all of the included studies had score ranging from 6 to 9, which meant they were high quality. The characteristics of the included studies are summarized in Table 1.

CE

PT E

D

Meta-analysis In this study, we evaluated the correlation between DcR3 expression and OS of patients with solid tumors. As shown in Figure 2, combined all included studies, DcR3 overexpression was significantly associated with poor survival outcome of patients with solid tumors (HR=1.63, 95%CI=1.27-2.08, p=0.0001, random). However, the expression of DcR3 was not associated with RFS (HR=1.07, 95%CI=0.71-1.63, p=0.74, fixed) (Figure 3).

AC

Subgroup analysis To explore the potential sources of heterogeneity, the subgroup analyses for OS were performed. The data presented in Table 2 reveal that DcR3 overexpression was correlated with unfavorable survival when divided by the types of tumor, sample size, sample type, detection method, NOS score and ethnicity of the patients. Subgroup analysis failed to eliminate the heterogeneity in different subgroups. However, there was no heterogeneity in the female reproductive cancer, other sample type, other detection method and Asian subgroups. Publication bias and sensitivity analysis In our meta-analysis, the Begg’s funnel plot test was used to evaluate potential publication bias. As shown in Figure 4A, the characteristic of the funnel charts is basically an inverted funnel and bilateral symmetry (p=0.092), indicating the absence of an obvious publication bias in these analyses. Sensitivity analysis revealed that no

ACCEPTED MANUSCRIPT point estimate of the omitted individual dataset lay outside the 95%CI of the combined analysis (Figure 4B). Interestingly, a study by Macher-Goeppinger S et al. [22] contributed the most to the observed heterogeneity. When this study was removed, the observed heterogeneity dramatically decreased (I2 = 45%, P=0.03 versus I2 = 0.0%, P = 0.46). However, the association between DCR3 and OS was not altered when this study was included (HR=1.63, 95%CI=1.27-2.08, p=0.0001) or excluded (HR=1.36, 95%CI=1.15-1.61, p=0.0003).

PT

Discussion

PT E

D

MA

NU

SC

RI

Apoptosis, or programmed cell death, is a key physiological function in humans. Imbalance of the strict apoptosis regulation would result in serious consequences and lead to diseases. It is well established that apoptosis inhibition is an important feature of tumor cells, which promotes their proliferation and immune escape. DcR3 was first detected in 1998 in human lung and colon tumors as a decoy receptor that binds to FasL and inhibits FasL-mediated apoptosis [4]. In addition, it was also found to neutralize LIGHT and TL1A, which leads to the antagonization of various immunoregulatory functions induced by the corresponding ligands [6,29,30]. As a soluble receptor and a secreted protein, DcR3 has not only been detected in tumor tissues, but also has been found to be elevated in the serum of cancer patients [31-33]. Additionally, it has been reported that malignant plasma cells and T lymphocytes from myeloma patients also produce DcR3, and serum DcR3 levels in myeloma patients are significantly higher compared to controls [34]. Moreover, high level of serum DcR3 has been associated with advanced tumor stage and tumor metastasis [13,16,35].

AC

CE

High levels of DcR3 have been reported in multiple human solid tumors and have been found to promote cancer progression. Many clinical studies have been conducted on the prognostic value of DcR3 overexpression. However, most of these studies involved a limited number of patients and the results remain inconclusive. From the clinical perspective, compelling evidence supporting the prognostic significance of DcR3 is still unavailable. Accordingly, we performed this meta-analysis to gain a better insight into its potential prognostic value in patients with cancer tumors. In this meta-analysis, we included sixteen studies on the prognostic value of DcR3 in patients with solid tumors. The results indicated that DcR3 overexpression was associated with poor survival outcome of patients with solid tumors. In addition, upregulation of DcR3 is associated with worse OS in gastrointestinal cancer, urinary system cancer and female reproductive cancer. However, DcR3 was not associated with RFS in cancer patients. Numerous carcinogenesis mechanisms involving alterations in the expression of DcR3 have been investigated. For instance, it has been reported DcR3 overexpression enhanced tumor cell proliferation and migration in vitro and tumorigenesis in vivo,

ACCEPTED MANUSCRIPT

NU

SC

RI

PT

and DcR3 knockdown could attenuated these malignant biological behaviors [18,36-38]. In addition, expression of DcR3 was shown to induce cytoskeleton remodeling, inhibit E-cadherin expression, and promote cancer cell migration [39]. Additionally, DcR3 was also found to be essential for TGF-β 3/SMAD signaling and thus mediated epithelial-mesenchymal transition (EMT) of CRC cells [18]. Moreover, DcR3 was shown to promote lymph node metastasis by inducing the formation of new lymphatic vessels and increasing chances for lymph node metastasis [10]. Recent studies reported that DcR3 can bind to TRAIL, and the downregulation of DcR3 with siRNA unmasked TRAIL and greatly enhanced TRAIL-induced apoptosis [40,41]. Tai SK et al. demonstrated that DcR3 modulates macrophage activation, downregulates MHC class II expression and mediates the induction of tumor-associated macrophages (TAMs), which promote tumor progression [42]. Inhibitors of DcR3 can block the malignant biological function of tumor cells. In preclinical experiments, Denbinobin and Triptolide were shown to downregulate the DcR3 protein level in pancreatic cancer cells and lead to potent anticancer activity in vitro and in vivo [43,44]. Together, these clear indications of the role of DcR3 in tumors revealed that DcR3 has a significant influence on the prognosis of cancer patients, which is consistent with our findings.

CE

PT E

D

MA

In our study, there is a certain but acceptable heterogeneity in the analysis of DcR3 and OS of the patients. Although the study by Macher-Goeppinger S et al. was identified as the main source of heterogeneity, specific elements that contributed to the heterogeneity were not fully explained. However, several factors could have contributed to the heterogeneity. First, the detection methods and cut-off values among the included studies are different, which may lead to a potential bias. Second, different type of cancers might increase the heterogeneity. Third, there were diverse sources of samples in each study. Finally, due to the variable follow-up time, publication year, and ethnic population, heterogeneity may be virtually brought in.

Limitations

AC

Our study has some limitations including: (1) Only English and Chinese language literatures were included, therefore literature published in other languages were not considered; (2) Some individual data could not be obtained from articles, we extracted these data by estimating Kaplan-Meier curves, which resulted in small statistical errors; (3) The number of included studies was limited, there is a lack of literatures on certain types of cancer, which may lead to a less powerful result in this meta-analysis.

Conclusions In summary, despite the limitations, our meta-analysis demonstrates that high DcR3 expression is associated with poor prognosis in patients with cancer, suggesting that DcR3 might serve as a potential prognostic predictor of OS for patients with solid tumors. Undoubtedly, additional studies related to specific tumor types and

ACCEPTED MANUSCRIPT perspectives are required to confirm the prognosis value of DcR3 expression in patients with solid tumors.

Acknowledgments This study was funded by the Beijing Municipal Administration of Hospital Clinical Medicine Development of Special Funding Support (ZYLX201612).

PT

Disclosure

RI

The authors report no conflicts of interest in this work.

SC

References

AC

CE

PT E

D

MA

NU

[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 (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 (2015) E359-386. [3] L. Ng, R. T. Poon, R. Pang, Biomarkers for predicting future metastasis of human gastrointestinal tumors, Cell. Mol. Life. Sci. 70 (2013) 3631-3656. [4] R.M. Pitti, S.A. Marsters, D.A. Lawrence, M. Roy, F.C. Kischkel, P. Dowd, A. Huang, C.J. Donahue, S.W. Sherwood, D. T. Baldwin, P. J. Godowski, W.I. Wood, A.L. Gurney, K.J. Hillan, R.L. Cohen, A.D. Goddard, D. Botstein, A. Ashkenazi, Genomic amplification of a decoy receptor for Fas ligand in lung and colon cancer, Nature. 396 (1998) 699-703. [5] D.N. Mauri, R. Ebner, R.I. Montgomery, K.D. Kochel, T.C. Cheung, G.L. Yu, S. Ruben, M. Murphy, R.J. Eisenberg, G.H. Cohen, P.G. Spear, C.F. Ware, LIGHT, a new member of the TNF superfamily, and lymphotoxin alpha are ligands for herpesvirus entry mediator, Immunity. 8 (1998) 21-30. [6] T.S. Migone, J. Zhang, X. Luo, L. Zhuang, C. Chen, B. Hu, J.S. Hong, J.W. Perry, S.F. Chen, J.X. Zhou, Y.H. Cho, S. Ullrich, P. Kanakaraj, J. Carrell, E. Boyd, H.S. Olsen, G. Hu, L. Pukac, D. Liu, J. Ni, S. Kim, R. Gentz, P. Feng, R.A. Moore, S.W. Ruben, P. Wei, TL1A is a TNF-like ligand for DR3 and TR6/DcR3 and functions as a T cell costimulator, Immunity. 16 (2002) 479-492. [7] Y. Wu, E. Guo, J. Yu, Q. Xie. High DcR3 expression predicts stage pN2-3 in gastric cancer, Am. J. Clin. Oncol. 31 (2008) 79–83. [8] Q.L. Liang, B.R. Wang, G.H. Li, DcR3 and survivin are highly expressed in colorectal carcinoma and closely correlated to its clinicopathologic parameters, J. Zhejiang. Univ. Sci. B. 10 (2009) 675–682. [9] C. Chen, C. Zhang, G. Zhuang, H. Luo, J. Su, P. Yin, J. Wang, Decoy receptor 3 overexpression and immunologic tolerance in hepatocellular carcinoma (HCC) development, Cancer. Invest. 26 (2008) 965-974.

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

[10] Q. Wu, Y. Zheng, D. Chen, X. Li, C. Lu, Z. Zhang, Aberrant expression of decoy receptor 3 in human breast cancer: relevance to lymphangiogenesis, J. Surg. Res. 188 (2014) 459–465. [11] K. Yamana, V. Bilim, N. Hara, T. Kasahara, T. Itoi, R. Maruyama, T. Nishiyama, K. Takahashi, Y. Tomita, Prognostic impact of FAS/CD95/APO-1 in urothelial cancers: decreased expression of Fas is associated with disease progression, Br. J. Cancer. 93 (2005) 544-551. [12] J.F. Tierney, L.A. Stewart, D. Ghersi, S. Burdett, M.R. Sydes, Practical methods for incorporating summary time-to-event data into meta-analysis, Trials. 8 (2007) 16. [13] H.F. Tu, C.J. Liu, S.Y. Liu, Y.P. Chen, E.H. Yu, S.C. Lin, K.W. Chang, Serum decoy receptor 3 level: a predictive marker for nodal metastasis and survival among oral cavity cancer patients, Head. Neck. 33 (2011) 396-402. [14] Q. Yue, M. Xiang, F. Li, H. Liu, F. Yue, J. Liu, W. Wang, C. Ji, Study on the expression level of DcR3 and MMP-2 in esophageal carcinoma and its impact on survival of the patients, CHINA. ONCOLOGY. 20 (2010) 745-750, (Chinese). [15] Y. Takahama, Y. Yamada, K. Emoto, H. Fujimoto, T. Takayama, M. Ueno, H. Uchida, S. Hirao, T. Mizuno, Y. Nakajima, The prognostic significance of overexpression of the decoy receptor for Fas ligand (DcR3) in patients with gastric carcinomas, Gastric. Cancer. 5 (2002) 61-68. [16] J. Zhou, S. Song, D. Li, S. He, B. Zhang, Z. Wang, X. Zhu, Decoy receptor 3 (DcR3) overexpression predicts the prognosis and pN2 in pancreatic head carcinoma, World. J. Surg. Oncol. 12 (2014) 52. [17] J. Zhou, S. Song, D. Zhu, H. Zhao, D. Li, Z. Zhang, Clinical Significance of Decoy Receptor 3 Expression in Pancreatic Carcinoma, Canc. Res. Prev. Treat. 40 (2013) 1143-1146, (Chinese). [18] Y.P. Liu, H.F. Zhu, D.L. Liu, Z.Y. Hu, S.N. Li, H.P. Kan, X.Y. Wang, Z.G. Li, DcR3 induces epithelial-mesenchymal transition through activation of the TGF-β 3/SMAD signaling pathway in CRC, Oncotarget. 7 (2016) 77306-77318. [19] G. Mild, F. Bachmann, J.L. Boulay, K. Glatz, U. Laffer, A. Lowy, U. Metzger, J. Reuter, L. Terracciano, R. Herrmann, C. Rochlitz, DCR3 locus is a predictive marker for 5-fluorouracil-based adjuvant chemotherapy in colorectal cancer, Int. J. Cancer. 102 (2002) 254-257. [20] L. Zong, P. Chen, D.X. Wang, Death decoy receptor overexpression and increased malignancy risk in colorectal cancer, World. J. Gastroenterol. 20 (2014) 4440-4445. [21] K. Edfeldt, K. Daskalakis, C. Bäcklin, O. Norlén, E. Tiensuu Janson, G. Westin, P .Hellman, P. Stålberg, DcR3, TFF3, and Midkine Are Novel Serum Biomarkers in Small Intestinal Neuroendocrine Tumors, Neuroendocrinology. 105 (2017) 170-181. [22] S. Macher-Goeppinger, S. Aulmann, N. Wagener, B. Funke, K.E. Tagscherer, A. Haferkamp, M. Hohenfellner, S. Kim, F. Autschbach, P. Schirmacher, W. Roth, Decoy receptor 3 is a prognostic factor in renal cell cancer, Neoplasia. 10 (2008) 1049-1056. [23] Y.Q. Jiang, T.F. Zhong, Y.W. Dang, L.S. Zou, L. Yang, X. Yang, G. Chen, Overexpression and clinicopathological contribution of DcR3 in bladder urothelial carcinoma tissues, Asian. Pac. J. Cancer. Prev. 15 (2014) 9137-9142.

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

[24] C. Kanbayashi, Y. Koyama, H. Ichikawa, E. Sakata, M. Hasegawa, C. Toshikawa, N. Manba, M. Ikarashi, T. Kobayashi, M. Minagawa, S.I. Kosugi, T. Wakai, Amplification of Genomic DNA for Decoy Receptor 3 Predicts Post-Resection Disease Recurrence in Breast Cancer Patients, World. J. Oncol. 5 (2014) 14-23. [25] Y.X. Cao, DcR3 expression and clinical signifcance of research in cervical cancer, Thesis, Central. South. University. (2011) (Chinese). [26] J.P. Connor, M. Felder, Ascites from epithelial ovarian cancer contain high levels of functional decoy receptor 3 (DcR3) and is associated with platinum resistance, Gynecol. Oncol. 111 (2008) 330-335. [27] L. Cheng, Y.H. Yang, C.R. Zhou, J.P. Yao, N. Niu, Q. Liu, Expression of decoy receptor 3 in epithelia ovarian carcinoma and its clinical signifcance, J. Mod. Oncol. 1 (2015) 116-119, (Chinese). [28] Y. Lin, X. Huang, Association between expression of decoy receptor 3 in patients with epithelial ovarian cancer and clinical pathology characteristics, Chin. J. Clin. Pharmacol. 32 (2016) 1670-1676, (Chinese). [29] M.C. Chen, M.J. Hwang, Y.C. Chou, W.H. Chen, G. Cheng, H. Nakano, T.Y. Luh, S.C. Mai, S.L. Hsieh, The role of apoptosis signal-regulating kinase 1 in lymphotoxin-beta receptor-mediated cell death, J. Biol. Chem. 278 (2003) 16073-16081. [30] C.R. Yang, S.L. Hsieh, C.M. Teng, F.M. Ho, W.L. Su, W.W. Lin, Soluble decoy receptor 3 induces angiogenesis by neutralization of TL1A, a cytokine belonging to tumor necrosis factor superfamily and exhibiting angiostatic action, Cancer. Res. 64 (2004) 1122-1129. [31] M. Yang, G. Chen, Y. Dang, D. Luo, Significance of decoy receptor 3 in sera of hepatocellular carcinoma patients, Ups. J. Med. Sci. 115 (2010) 232-237. [32] P.M. Chang, P.M. Chen, S.L. Hsieh, C.H. Tzeng, J.H. Liu, T.J. Chiou, W.S. Wang, C.C. Yen, J.P. Gau, M.H. Yang, Expression of a soluble decoy receptor 3 in patients with diffuse large B-cell lymphoma predicts clinical outcome, Int. J. Oncol. 33 (2008) 549-554. [33] I. Simon, Y. Liu, K.L. Krall, N. Urban, R.L. Wolfert, N.W. Kim, M.W. McIntosh MW, Evaluation of the novel serum markers B7-H4, Spondin 2, and DcR3 for diagnosis and early detection of ovarian cancer, Gynecol. Oncol. 106 (2007) 112-118. [34] G. Brunetti, A. Oranger, G. Mori, M. Centonze, G, Colaianni, R. Rizzi, V. Liso, A. Zallone, M. Grano, S. Colucci, The formation of osteoclasts in multiple myeloma bone disease patients involves the secretion of soluble decoy receptor 3, Ann. N. Y. Acad. Sci. 1192 (2010) 298-302. [35] D. Wang, J. Wang, G. Chen, Association of serum decoy receptor 3 protein level with the clinicopathologic features of bladder transitional cell carcinoma, Nan. Fang. Yi. Ke. Da. Xue. Xue. Bao. 33 (2013) 1831-1832, (Chinese). [36] W. Yu , Y.C. Xu, Y. Tao, P. He, Y. Li, T. Wu, Y.P. Zhu, J. Li, J.X. Wu, J. Dai, DcR3 regulates the growth and metastatic potential of SW480 colon cancer cells, Oncol. Rep. 30 (2013) 2741-2748. [37] X.N. Zhou, G.M. Li, Y.C. Xu, T.J. Zhao, J.X. Wu, Knockdown of Decoy Receptor 3 Impairs Growth and Invasiveness of Hepatocellular Carcinoma Cell Line

ACCEPTED MANUSCRIPT

CE

PT E

D

MA

NU

SC

RI

PT

of HepG2, Chin. Med. J. (Engl). 129 (2016) 2623-2629. [38] Z. Ge, A.J. Sanders, L. Ye, Y. Wang, W.G. Jiang, Expression of death decoy receptor-3 (DcR3) in human breast cancer and its functional effects on breast cancer cells in vitro, J. Exp. Ther. Oncol. 9 (2011) 109-118. [39] H. Zhang, X. Chen, D. Li, L. Cui, X. Li, X. Ye, X. Wan, DcR3 promotes hepatoma cell migration by downregulating E-cadherin expression, Oncol. Rep. 38 (2017) 377-383. [40] W. Wang, M. Zhang, W. Sun, S. Yang, Y. Su, H. Zhang, C. Liu, X. Li, L. Lin, S. Kim, P. Okunieff, Z. Zhang, L. Zhang, Reduction of decoy receptor 3 enhances TRAIL-mediated apoptosis in pancreatic cancer, PLoS. One. 8 (2013) e74272. [41] C. Liang , Y. Xu, G. Li, T. Zhao, F. Xia, G. Li, D. Zhang, J. Wu, Downregulation of DcR3 sensitizes hepatocellular carcinoma cells to TRAIL-induced apoptosis, Onco. Targets. Ther. 10 (2017) 417-428. [42] S.K. Tai, H.C. Chang, K.L. Lan, C.T. Lee, C.Y. Yang, N.J. Chen, T.Y. Chou, D.C. Tarng, S.L. Hsieh, Decoy receptor 3 enhances tumor progression via induction of tumor-associated macrophages, J. Immunol. 188 (2012) 2464-2471. [43] C.R. Yang, J.H. Guh, C.M. Teng, C.C. Chen, P.H. Chen, Combined treatment with denbinobin and Fas ligand has a synergistic cytotoxic effect in human pancreatic adenocarcinoma BxPC-3 cells, Br. J. Pharmacol. 157 (2009) 1175-1185. [44] W. Wang, X. Li, W. Sun, L. Zhang, M. Zhang, B. Hong, G. Lv, Triptolide triggers the apoptosis of pancreatic cancer cells via the downregulation of Decoy receptor 3 expression, J. Cancer. Res. Clin. Oncol. 138 (2012) 1597-1605.

Figure legends

AC

Figure 1. Flow diagram of the search strategy used in this study. Figure 2. Forest plots to assess the effect of elevated DcR3 expression on the OS in patients with solid tumors: summary of all fifteen studies with regard to OS, the estimate was 1.63 (1.27-2.08) using a random effects model. Figure 3. Forest plots to assess the effect of elevated DcR3 expression on the RFS in patients with solid tumors: summary of all two studies with regard to RFS, the estimate was 1.07 (0.71-1.63) using a fixed effects model. Figure 4. Publication bias and sensitivity analysis of DcR3 expression and the prognosis of patients with solid tumors: A. Begg’s publication bias plot for OS; B. Chart of the sensitivity analysis for OS.

ACCEPTED MANUSCRIPT Table 1. Characteristics of studies included in the meta-analysis.

DcR3 Study

Year Country

Cancer No. of

Sex

type patients (M / F)

Age

Sample Detection type

method

Follow positive/high

up

Outcomes

NOS Score

Reference

(%) 148

130/18

Yue Q

2010

China

EC

98

61/37

Takahama Y

2002

Japan

GC

84

65/19

Zhou J

2015

China

PC

50

30/20

Zhou J

2013

China

PC

50

26/24

Liu YP

2016

China

CRC

86

NR

NR

tissue

Mild G

2002 Switzerland CRC

223

NR

NR

tissue

Zong L

2014

CRC

300 177/123

2017 Sweden

SINT

96

2008 Germany

RCC

560 348/212

Jiang YQ

2014

China

BUC

166

141/25

Kanbayashi C

2014

Japan

BRC

95

F: 95

Cao YX

2011

China

CC

57

F: 57

Connor JP

2008

USA

EOC

44

F: 44

NR

Cheng L

2015

China

36-73

Lin Y

2016

China

blood

ELISA

91 (61.5)

tissue

IHC

64 (65.3)

NR

tissue

NB

22 (26)

NR

tissue

IHC

39 (78)

tissue

IHC

25 (50)

56.4 (32-76)

58.7 (41-75)

56.3 (21-85)

NR

61.6

(27-96) 56

(30-87) 54.7 (27-72)

NR

OS

7

13

OS

6

14

OS

6

15

NR

OS

6

16

NR

OS

6

17

up to 12/2009

63m

IHC

44 (51.2)

NR

OS

6

18

IHC

163 (73)

NR

OS, RFS

7

19

OS

8

20

OS

7

21

OS

7

22

tissue

IHC

blood

ELISA

48 (50)

tissue

IHC

52 (9.3)

tissue

IHC

61 (36.7)

4-40m

OS

8

23

tissue Q-PCR

48 (50.5)

4-105m

RFS

9

24

tissue

46 (80.7)

OS

7

25

MA

63±13

D

46/50

PT E

S

CE

Macher-Goeppinger

IHC

ascites ELISA

175 (58.33) 2-93m 3.7±3y 40m (median)

up to 10/2010

22 (50)

NR

OS

6

26

EOC

86

F: 86

tissue

IHC

77 (89.5)

14-69m

OS

7

27

EOC

66

F: 66 56.8±12.3 tissue

IHC

55 (83.3)

NR

OS

7

28

AC

Edfeldt K

China

NR

PT

OCC

RI

China

SC

2010

NU

Tu HF

Abbreviations: DcR3=Decoy receptor 3; ELISA=enzyme linked immunosorbent assay; NB=Northern blot; IHC=Immunohistochemistry; Q-PCR=Quantitative polymerase chain reaction; OS=Overall survival; RFS=Relapse free survival; NR=not reported; M=Male; F=Female; m=months; y=years; NOS=Newcastle-Ottawa Scale; OCC=Oral cavity cancer; EC=Esophageal cancer; GC=Gastric cancer; PC=Pancreatic cancer; CRC=Colorectal cancer; SINT=Small intestinal neuroendocrine tumors; RCC=Renal cell cancer; BUC=Bladder urothelial carcinoma; BRC=Breast cancer; CC=Cervical cancer; EOC=Epithelial ovarian cancer.

ACCEPTED MANUSCRIPT Table 2: Pooled HR for OS according to subgroup analysis No. of studies

No. of patients

Pooled HR(95% CI)

PHet

I2(%)

P value

1 8 2

148 987 726

1.68 (0.76-3.71) 1.28 (1.07-1.55) 2.57 (1.79- 3.71)

0.30 0.05

17 73

0.20 0.009 <0.0001

5

348

1.93 (1.22-3.06)

0.84

0.0

0.005

＞90

7

1591

＜90

8

523

1.57 (1.06, 2.30) 1.45(0.43, 4.84)

12 3

1826 288

1.60 (1.21-2.12) 1.87 (1.03-3.39)

11 5

1742 372

9 6

1702 412

Subgroups

11 4

CE

62

0.02

0.25

23

0.004

0.01 0.75

55 0.0

0.001 0.04

0.008 0.90

58 0.0

0.003 0.001

1.65 (1.17-2.34) 1.53 (1.07-2.18)

0.003 0.24

53 25

0.004 0.02

1.40 (1.16-1.68) 1.90 (0.95-3.81)

0.45 0.008

0.0 75

0.0004 0.07

NU

SC

RI

0.02

1.58 (1.17-2.13) 1.91 (1.14-3.17)

MA

Sample type Tissue Others Detection method Immunohistochemistry Others NOS score ＞7 ＜7 Ethnicity Asian Caucasian

D

Gastrointestinal cancer Urinary system cancer Female reproductive cancer Sample size

1191 923

PT E

Oral cavity cancer

PT

Tumor type

AC

Abbreviations: HR=hazard ratio; NOS=Newcastle-Ottawa Scale.

ACCEPTED MANUSCRIPT

Highlights

AC

CE

PT E

D

MA

NU

SC

RI

PT

1) DcR3 over-expression was significantly associated with worse OS in solid tumors. 2) Up-regulation of DcR3 is related to worse OS in gastrointestinal cancer, urinary system cancer and female reproductive cancer. 3) DcR3 expression might not be related to RFS in malignancies.

Figure 1

Figure 2

Figure 3

Figure 4