Elevated expression of DKK1 is associated with cytoplasmic/nuclear β-catenin accumulation and poor prognosis in hepatocellular carcinomas
Bin Yu,Xinrong Yang,Yang Xu,Genfu Yao,Huiqun Shu,Biaoyang Lin,Leroy Hood,Hongyang Wang,Shengli Yang,Jianren Gu,Jia Fan,Wenxin Qin
DOI: https://doi.org/10.1016/j.jhep.2008.11.020
IF: 25.7
2009-01-01
Journal of Hepatology
Abstract:Results The expression level of DKK1 was associated with the staining pattern of β-catenin in HCC cell lines, and DKK1 overexpression correlated with β-catenin cytoplasmic/nuclear accumulation in clinical HCC samples ( P = 0.011, correlation coefficient = 0.144). High DKK1 expression predicted unfavorable prognosis in HCC patients, especially in early stage patients and those with normal AFP levels. In multivariate analyses, DKK1 was an independent predictor for overall survival (OS) ( P = 0.002) and disease-free survival (DFS) ( P = 0.002) of HCC patients. Furthermore, the HCC patients with high DKK1 expression and cytoplasmic/nuclear β-catenin accumulation had very poor prognosis. Conclusions Elevated expression of DKK1 is a critical event in patients with HCC that indicates poor clinical outcome. DKK1, alone or combined with β-catenin, is a novel prognostic predictor for HCC patients. Abbreviations DKK1 Dickkopf-1 HCC hepatocellular carcinoma OS overall survival DFS disease-free survival qRT-PCR quantitative real-time RT-PCR Keywords DKK1 HCC Predictor β-Catenin Tumor tissue microarray 1 Introduction Hepatocellular carcinoma (HCC) is one of the most prevalent tumor types, and both the incidence and mortality rates of HCC have increased in recent years [1] . Although survival of patients with HCC has improved due to advances in surgical techniques and perioperative management, long-term survival after surgical resection remains unsatisfactory because of the high rate of recurrence and metastasis [2,3] . Biomarkers that can define the risk of recurrence and the metastatic potential of HCC may allow appropriate therapeutic regimes to be applied earlier in the disease course, but suitable predictors that can be widely used in clinical settings are not currently available [4–7] . Dickkopf-1 (DKK1), the prototype of a family of secreted proteins, plays a crucial role in head formation in vertebrate development and is a potent antagonist of Wnt signaling [8,9] . DKK1 is downregulated in human colon tumors [10] , indicating that DKK1 might act as a tumor suppressor and that its role as an antagonist of Wnt signaling is lost in colon cancer. However, overexpression of DKK1 was found in 26 out of 32 human hepatoblastomas and 5 out of 6 cases of Wilms’ tumor [11] . Forget et al. reported expression of DKK1 in 21 out of 73 cases of breast cancer, in particular hormone-resistant breast tumors [12] , and preliminary experiments using gene microarrays, including studies in our institute, have shown high expression of DKK1 in HCCs [13,14] . These studies suggest that DKK1 performs an oncogenic, rather than tumor-suppressing, function in these tumors. However, except for the expression status of DKK1 described in these preliminary reports, the role of DKK1 expression and activation in the progression and prognosis of HCC has not been investigated. In this study we show that DKK1 expression can be induced by an active Wnt/β-catenin signaling pathway and that its overexpression preferentially exists in metastatic HCC cell lines and HCC patients with recurrence of disease. Furthermore, tissue microarray data suggest that DKK1 is a novel prognostic predictor for patients with HCC, especially in early stages of disease and patients with normal AFP levels. 2 Materials and methods 2.1 Cell lines HCC cell lines HuH-7, SMMC-7721, HepG2 and metastatic HCC cell lines MHCC97-L and HCCLM3 [2] were maintained as described previously [15] . 2.2 Patients and specimens Fifty cases used in quantitative real-time RT-PCR (qRT-PCR) were randomly collected from patients after operations performed between 2000 and 2002. Fourteen frozen tissues used in Western blot analysis were randomly chosen from these 50 cases. Tumor specimens used in tissue microarray analysis were obtained from 314 patients with HCC who underwent surgical resection in the Liver Cancer Institute, Zhong Shan Hospital, Fudan University between 1997 and 2000. Ethical approval for human subjects was obtained from the research ethics committee of Zhong Shan Hospital, informed consent was obtained from each patient. The entrance criteria for all patients in this study were patients with: (1) distinctive HCC diagnosis by pathology based on the WHO criteria; (2) no prior anticancer treatment (3) indication for surgical resection (4) suitable formalin-fixed, paraffin-embedded tissues and frozen tissues (5) clinicopathologic and follow-up data. Surgical resection which was defined as: the complete resection of all tumor nodules and the cut surface being free of cancer by histological examination [16] . Tumor differentiation was defined according to the Edmondson grading system [17] . Liver function was assessed by Child-Pugh classification. Tumor staging was defined according to the sixth edition of TNM classification of UICC. The clinicopathologic characteristics of 314 patients are summarized in Table S1. 2.3 Western blotting Cell and tumor tissue lysates and serum-free conditioned media were clarified by centrifugation at 10,000 g . Cytoplasmic and nuclear extracts from cultured cells were prepared using NE-PER Nuclear and Cytoplasmic Extraction Reagents (Pierce Chemical Company). Immunoblotting experiments were performed according to standard procedures [15] . 2.4 Tissue microarray and immunohistochemistry Tissue microarrays were constructed and immunohistochemistry performed using a two-step protocol as previously described [18] . 2.5 Evaluation of immunohistochemical variables Immunohistochemical staining was assessed by two independent pathologists with no knowledge of patient characteristics and any discrepancies were resolved by consensus. The mean percentage value of two cores was considered representative of one tumor. The concordance between scores from different cores of the same tumor was greater than 90%. The intensity of DKK1 staining was evaluated using the following criteria: strong staining (scored as 2+), dark brown staining in >50% of tumor cells completely obscuring cytoplasm; weak staining (1+), any lesser degree of brown staining appreciable in tumor cell cytoplasm; and absent (scored as 0), no appreciable staining in tumor cells [19] . Only strong staining was considered a positive result. For β-catenin protein immunoreaction, we regarded more than 10% of cells exhibiting cytoplasmic or nuclear staining as a positive result [20] . 2.6 Statistical analysis Statistical analyses were performed by SPSS 12.0 for windows (SPSS). Cumulative survival time was calculated by the Kaplan–Meier method and analyzed by the log-rank test. Univariate and multivariate analyses were based on the Cox proportional hazards regression model. The correlation significance was analyzed by Kendall’s tau-b rank correlation analysis. The chi-square test, Fisher’s exact probability and student’s t -test were used for comparison between groups. P < 0.05 was considered statistically significant. 3 Results 3.1 Expression of DKK1 in HCCs from patients with or without disease recurrence qRT-PCR analysis showed high expression of DKK1 mRNA in MHCC-97L and HCCLM3 cells, but low expression in SMMC-7721 and HuH-7 ( Fig. 1 A). Consistent with these findings, DKK1 protein expression was not detected in HuH-7 and weakly observed in SMMC-7721, but strongly detected in metastatic HCC cell lines MHCC97-L and HCCLM3 ( Fig. 1 B). In tissue samples, DKK1 was expressed in four of 7 patients (T1, T2, T3 and T5) in the group with disease recurrence, but in only one of 7 patients (T12) without recurrence ( Fig. 1 C). Patients suffering HCC recurrence had higher levels of DKK1 mRNA expression than those without recurrence ( P = 0.005) ( Fig. 1 D). 3.2 DKK1 expression in HCC cells correlated with the staining patterns of β-catenin Five cell lines were screened for larger mRNA deletions by RT-PCR using primers flanking exon 3 and exons 2–4 of the β-catenin gene. PCR products of reduced size were observed in MHCC97-L, HCCLM3 and HepG2 ( Fig. S1A ). We found deletion of entire exon 3 in MHCC97-L and HCCLM3, and partial deletion of exons 3–4 in HepG2. The deletion of exon 3 resulted in altered β-catenin transcripts and proteins lacking a functional NH 2 -terminal degradation box. No larger deletion in β-catenin mRNA was found in HuH-7 or SMMC-7721. We next examined β-catenin expression and phosphorylation in HCC cell lines. Western blot analysis showed that the phosphorylation level of β-catenin in the cytoplasmic fraction of HuH-7 and SMMC-7721 cells was higher than that in HepG2 and MHCC-97L, and the expression level of total β-catenin in the nuclear fraction of HuH-7 and SMMC-7721 cells was lower than that in HepG2 and MHCC-97L ( Fig. S1B ). We further analyzed the cellular localization of β-catenin in HCC cells by immunostaining. Membranous staining of β-catenin was observed in HuH-7 and SMMC-7721 cells, and nuclear staining was seen in HepG2 and MHCC-97L cells ( Fig. S1C ). Secretion of DKK1 into the culture medium of HCC cells was assessed by ELISA. High concentrations of DKK1 were detected in the culture medium of HepG2 and MHCC-97L cells, and low concentrations in the medium of HuH-7 and SMMC-7721 ( Fig. S1D ). Thus, the level of DKK1 secretion in HCC cells closely correlated with the staining patterns of β-catenin. To further investigate the effect of Wnt/β-catenin signaling on the expression of DKK1 in HCC cells, β-catenin mutants (Δ−β-catenin) were transfected into SMMC-7721 cells, resulting in increased expression of DKK1 ( Fig. S1E ). In addition, reduction of β-catenin protein levels in HepG2 cells using short interfering RNA (siRNA) directed against two separate regions of β-catenin resulted in decreased expression of DKK1 ( Fig. S1E ). 3.3 Immunohistochemical analysis of HCC tumor samples We next investigated the relationship between β-catenin and DKK1 expression level in clinical HCC samples by immunohistochemical staining of tumor tissue microarrays. Representative images are presented for DKK1 and β-catenin in Fig. 2 . Cytoplasmic/nuclear staining of β-catenin was observed in 126 of the 314 HCC samples (40.1%), membranous staining in 105 cases (33.5%), and no staining in 83 cases (26.4%). DKK1 staining with polyclonal antibody specific to DKK1 was mainly observed in the cytoplasm of tumor cells. DKK1 was strongly stained (score 2+) in 99 cases of the 314 samples (31.5%), weakly stained (score 1+) in 141 cases (44.9%) and not stained (score 0) in 74 cases (23.6%). We regarded strong staining as a positive result for DKK1 and cytoplasmic/nuclear staining as a positive result for β-catenin (Table S1). 3.4 Prognostic factors For the whole study population, the overall survival (OS) and disease-free survival (DFS) rates were 68.1% and 59.6% at 3 years, 54.3% and 48.8% at 5 years, 43.5% and 43.4% at 7 years and 39.1% and 39.4% at 9 years, respectively. Univariate analysis showed that GGT, tumor number, vascular invasion and TNM stage were unfavorable predictors for OS and DFS. In addition, tumor differentiation, tumor encapsulation and tumor size were associated with OS ( Table 1 ). DKK1 was found to be prognostic for OS ( P = 0.001) and DFS ( P < 0.001) in HCC. The 3-, 5-, 7-, 9-year OS and DFS rates in the DKK1 − group were significantly higher than those of the DKK1 + group (73.9% and 66.2% vs. 55.6% and 44.4%, 59.3% and 55.2% vs. 43.4% and 34.2%, 49.3% and 49.9% vs. 31.0% and 31.3%, 44.5% and 47.0% vs. 27.6% and 21.9%, respectively) ( Fig. 3 A and B and Table 1 ). Most of the DKK1 + patients with normal pre-operative levels of AFP (29/47) suffered disease recurrence within 5 years. For patients with normal pre-operative AFP, the 5-year OS and DFS rates for DKK1 − patients were significantly higher than those of DKK1 + patients (OS 64.4% vs. 42.6%, P = 0.014; DFS 54.5% vs. 31.7%, P = 0.038) ( Fig. 3 C and D). When we further stratified patients by TNM stage, we found that the majority of the DKK1 + patients (75 of 99) were in TNM stage I, which may have been caused by the heterogeneous TNM stage distribution of the patient population (stage I, n = 228; stages II–III, n = 86). Most of the DKK1 + patients in TNM stage I (43/75) suffered recurrence within 5 years. The 5-year OS and DFS rates of these patients were 46.7% and 39.5%, significantly lower than those of DKK1 − patients in TNM stage I (66.7% and 62.4%, P = 0.0007 and P < 0.0001, respectively) ( Fig. 3 E and F). Multivariate analysis further demonstrated that DKK1 was an independent prognostic factor for both OS ( P = 0.002) and DFS ( P = 0.002) of HCC patients ( Table 2 ). β-Catenin was also found to be prognostic for OS ( P = 0.040) and DFS ( P < 0.0001). The 3-, 5-, 7-, 9-year OS and DFS rates in the β-catenin − group were significantly higher than those of the β-catenin + group (73.3% and 70.4% vs. 60.3% and 43.3%, 59.4% and 57.7% vs. 46.7% and 35.5%, 48.1% and 52.1% vs. 36.7% and 30.4%, 42.1% and 46.8% vs. 34.7% and 28.1%, respectively) ( Fig. 4 A and B and Table 1 ). In multivariate analysis, β-catenin was an independent prognostic factor for both OS ( P = 0.031) and DFS ( P < 0.0001) of HCC patients ( Table 2 ). The combined influence of DKK1 and β-catenin on prognosis of patients with HCC was also evaluated. We divided patients into three groups, DKK1 − and β-catenin − ( n = 139), DKK1 + or β-catenin + ( n = 125), and DKK1 + and β-catenin + ( n = 50), and found significant differences in both OS ( P = 0.003) and DFS ( P < 0.0001) between the three groups. The 5-year OS and DFS rates were 66.0% and 59.8% for the DKK1 − and β-catenin − group, 44.6% and 49.1% for the DKK1 + or β-catenin + group, and 46.0% and 18.0% for the DKK1 + and β-catenin + group ( Fig. 4 C and D and Table 1 ). In addition, the co-index of DKK1/β-catenin was an independent prognosticator for both OS ( P = 0.0002) and DFS ( P < 0.0001) in the multivariate analysis ( Table 3 ). 3.5 Relationship between immunomarkers and clinicopathologic features Kendall’s tau-b rank correlation analysis demonstrated a significant correlation between high level of DKK1 expression and cytoplasmic/nuclear staining of β-catenin ( P = 0.011, correlation coefficient = 0.144) (Table S2). Neither DKK1 nor DKK1/β-catenin positive staining was associated with liver cirrhosis, AFP, tumor encapsulation, tumor differentiation, TNM stage, or vascular invasion. Sites of disease recurrence were classified into three groups: type I was local recurrence in remnant liver with a single or double lesion ( n = 119); type II was multinodular (⩾3) or diffuse pattern consisting of many nodules scattered throughout the remaining liver ( n = 23); and type III was extrahepatic metastasis ( n = 27). In our study, DKK1 + /β-catenin + patients had high rates of type II and III recurrence compared with the DKK1 − /β-catenin − group (38.5% vs. 18.6%, P = 0.030, Table S2). Using 24 months as the cut-off, recurrence was classified as early recurrence ( n = 95), mainly resulting from intrahepatic metastasis, and late recurrence ( n = 74), usually the result of a multicentric new tumor [21] . DKK1 + /β-catenin + patients were inclined to develop early recurrence than patients in the DKK1 − /β-catenin − group ( P = 0.090, Table S2). 3.6 Activation of cellular invasion by DKK1 To examine the functional role of DKK1 in HCC cells, metastatic HCCLM3 cells were transfected with three siRNA duplexes against DKK1. As shown in Fig. 5 A and B, DKK1 expression in HCCLM3 cells was effectively knocked down by siRNA-1 and siRNA-2. The effect of DKK1 reduction on the invasive behavior of HCC cells was tested using a Boyden chamber assay. At 48 h, there was a 49% reduction in cell invasion in the siRNA-1-transfected cells and a 73% reduction in the invasion of siRNA-2-knockdown cells compared with cells transfected with the control nonspecific siRNA ( Fig. 5 C). 4 Discussion DKK1 was reported to be a target gene of β-catenin/TCF by computational screening and experimental analysis [10,22] . In this study, we found that DKK1 expression was associated with cytoplasmic/nuclear accumulation of β-catenin in HCC cells and clinical HCC samples, suggesting that DKK1 expression is regulated by the Wnt/β-catenin signaling pathway in some HCCs. Our in vitro assays demonstrated that DKK1 performs its inhibitory action on upstream of the Wnt/β-catenin pathway and is thus unable to decrease Wnt signaling in the cells that already exhibit cytoplasmic/nuclear β-catenin accumulation ( Fig. S2 ). So, the biological capability of DKK1 to inhibit Wnt signaling through a negative feedback loop seems to be abrogated in HCC, most likely because genetic alterations disrupt the central multiprotein complex that controls the stability of β-catenin. Overexpression of DKK1 has been reported in hepatoblastoma, Wilms’ tumor, and breast cancer [11,12] , implying that DKK1 functions as a potential oncogenic factor in these tumors, and not as a tumor suppressor through inhibition of Wnt signaling. A recent research article showed that overexpression of DKK1 enhanced the invasive activity of NIH3T3 and COS-7 cells and that DKK1 was a potential prognostic marker for primary non-small cell lung cancer and esophageal squamous cell carcinoma [19] . We found that DKK1 was preferentially expressed in metastatic HCC cell lines and HCC patients with disease recurrence, and we fully evaluated the prognostic value of DKK1 in patients with HCC using tissue microarrays consisting of tumor tissues from 314 HCC patients. Our data showed that high level of DKK1 expression in HCC tissues correlated with poor prognosis of HCC patients, and that DKK1 was an independent prognosticator for OS and DFS in HCC. Despite advances in surveillance and clinical treatment strategies, the high rate of recurrence or metastasis after surgical resection has hindered improvements in HCC survival rates [2,3] . It is critical to identify early those patients with a high probability of recurrence and metastasis in order to initiate a timely intervention; however, specific predictive markers for metastasis and recurrence are lacking, and current clinicopathologic factors such as TNM stage and tumor differentiation cannot accurately predict the outcome of all HCC patients [6] . At present, AFP is the best marker for monitoring recurrence and metastasis in AFP-positive HCC patients after surgery [23] , but it is hard to predict prognosis in AFP-normal and early stage HCC patients [6,7] . When we stratified patients according to AFP, we found that most DKK1 + patients in the AFP-normal group suffered recurrence within 5 years, suggesting that DKK1 might be a promising marker for poor prognosis in AFP-normal HCC patients. Interestingly, unlike previously described predictive markers for poor prognosis, which are mainly expressed in advanced HCC [6,24,25] , we found that the 5-year DFS rate of DKK1 + patients in TNM stage I was significantly reduced compared with DKK1 − patients in the same stage (39.5% vs. 62.4%, P < 0.0001), suggesting that DKK1 might be a very good marker for poor prognosis in early stage HCC patients. In the clinical context, it has been noticed that even after surgical resection, the recurrence rate of early stage HCC is still very high (50.00% in our series). The predictive value of DKK1 has great potential to help clinicians to distinguish patients at high risk of recurrence in early stages. Therefore, DKK1 is a predictive marker of HCC invasiveness and unfavorable prognosis in AFP-normal and early stage HCC patients, in whom prediction is very difficult using conventional clinical indexes. Reduced membranous expression of β-catenin is associated with short survival in gastric cancer and primary colorectal cancer patients [26,27] . Although Inagawa et al. reported that nuclear β-catenin accumulation was associated with poor prognosis in poorly differentiated HCC patients [28] , Schmitt-Gräff et al. did not find a correlation between β-catenin expression and prognosis in HCCs [29] . We investigated the prognostic significance of the expression pattern of β-catenin in a larger cohort of HCC patients with a long-term follow-up study by tissue microarray and found that cytoplasmic/nuclear accumulation of β-catenin was correlated with poor prognosis. DKK1 + patients and β-catenin + patients did not perfectly overlap in our tissue microarray data (Table S2), and it is possible that DKK1 expression is also regulated by other molecules or signaling pathways [30–32] . As the level of DKK1 was associated with the staining pattern of β-catenin, we combined high expression of DKK1 and cytoplasmic/nuclear accumulation of β-catenin, and found that this correlated with very poor prognosis of HCC patients. Furthermore, the co-index of DKK1/β-catenin was an independent prognosticator for both OS and DFS of HCC patients and patients with DKK1 + /β-catenin + had higher rates of type II and III recurrence than the DKK1 − /β-catenin − group. As type II and III recurrences were mainly caused by dissociation of tumor cells from the primary tumor and dissemination into other sites before surgery, these data indicate greater malignancy and higher metastatic potential of tumors in DKK1 + /β-catenin + patients. We further showed that knockdown of DKK1 expression inhibited invasion of the HCC cell line HCCLM3 in vitro , implying that DKK1 expression might promote dissemination of tumors by stimulating cell invasion. Concordantly, tissue microarray analyses revealed a correlation between strong DKK1 staining in HCC tissues and a higher recurrence rate of the patients. Additional functional assays in HCC cell lines with different β-catenin mutations will contribute towards a better understanding of the signaling pathway mediated by DKK1 overexpression. We also detected high levels of DKK1 protein in some serologic samples from 128 HCC patients (data not shown), and the significance of DKK1 for diagnosis and prognosis of HCC patients with normal or absent AFP expression is currently being evaluated in a separate experiment using sera from a large cohort of HCC patients ( n > 1000). To the best of our knowledge, this is the first demonstration of DKK1 as a predictive marker of HCC invasiveness and unfavorable prognosis in HCC patients after surgery. The availability of a prognostic predictor in the early stages of disease may allow initiation of adjuvant treatment for DKK1 + or DKK1 + /β-catenin + HCC patients after surgical procedures and may reduce disease recurrence and prolong survival. Acknowledgements This work was supported by National Key Basic Research Program of China Grant 2009CB521803 and National Key Sci-Tech Special Project of China Grant 2008ZX10002-019 and Shanghai Municipal Program of International Cooperation in Science and Technology Grant 08410700900 (to W. Qin), Shanghai Science and Technology Development Fund 06QA14012 and the Program for Excellent Disciplinary Leaders of Shanghai Health Bureau Grant LJ06004 (to J. Fan). Support was also provided by NSFC/RGC 30731160004. W. Qin is a Visiting Scientist from the Leroy Hood Laboratory, the Institute for Systems Biology, Seattle, WA 98103, USA. We are grateful to Xin Hu, Yingjun Zhao, Shenglin Huang, Lin Wei and Hua Jiang for skillful technical support. Appendix A Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jhep.2008.11.020 . Appendix A Supplementary data Supplementary Fig. S1 Supplementary Fig. S2 Supplementary data References [1] A. Jemal R. Siegel E. Ward T. Murray J. Xu M.J. Thun Cancer statistics, 2007 CA Cancer J Clin 57 2007 43 66 [2] Z.Y. Tang S.L. Ye Y.K. Liu L.X. Qin H.C. Sun Q.H. Ye A decade’s studies on metastasis of hepatocellular carcinoma J Cancer Res Clin Oncol 130 2004 187 196 [3] Y. Yang H. Nagano H. Ota O. Morimoto M. Nakamura H. Wada Patterns and clinicopathologic features of extrahepatic recurrence of hepatocellular carcinoma after curative resection Surgery 141 2007 196 202 [4] C.D. Mann C.P. Neal G. Garcea M.M. Manson A.R. Dennison D.P. Berry Prognostic molecular markers in hepatocellular carcinoma: a systematic review Eur J Cancer 43 2007 979 992 [5] J.M. Llovet J. Bruix Novel advancements in the management of hepatocellular carcinoma in 2008 J Hepatol 48 2008 S20 S37 [6] L.X. Qin Z.Y. Tang Recent progress in predictive biomarkers for metastatic recurrence of human hepatocellular carcinoma: a review of the literature J Cancer Res Clin Oncol 130 2004 497 513 [7] J.M. Llovet A.M. Di Bisceglie J. Bruix B.S. Kramer R. Lencioni A.X. Zhu Design and endpoints of clinical trials in hepatocellular carcinoma J Natl Cancer Inst 100 2008 698 711 [8] A. Glinka W. Wu H. Delius A.P. Monaghan C. Blumenstock C. Niehrs Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction Nature 391 1998 357 362 [9] P. Fedi A. Bafico A. Nieto Soria W.H. Burgess T. Miki D.P. Bottaro Isolation and biochemical characterization of the human DKKK1 homologue, a novel inhibitor of mammalian Wnt signaling J Biol Chem 274 1999 19465 19472 [10] J.M. González-Sancho O. Aguilera J.M. Garcı´a N. Pendás-Franco C. Peña S. Cal The Wnt antagonist DICKKOPF-1 gene is a downstream target of beta-catenin/TCF and is downregulated in human colon cancer Oncogene 24 2005 1098 1103 [11] O. Wirths A. Waha S. Weggen P. Schirmacher T. Kühne C.G. Goodyer Overexpression of human Dickkopf-1, an antagonist of wingless/WNT signaling, in human hepatoblastomas and Wilms’ tumors Lab Invest 83 2003 429 434 [12] M.A. Forget S. Turcotte D. Beauseigle J. Godin-Ethier S. Pelletier J. Martin The Wnt pathway regulator DKK1 is preferentially expressed in hormone-resistant breast tumours and in some common cancer types Br J Cancer 96 2007 646 653 [13] M.A. Patil M.S. Chua K.H. Pan R. Lin C.J. Lih S.T. Cheung An integrated data analysis approach to characterize genes highly expressed in hepatocellular carcinoma Oncogene 24 2005 3737 3747 [14] T. Yamashita M. Forgues W. Wang J.W. Kim Q. Ye H. Jia EpCAM and alpha-fetoprotein expression defines novel prognostic subtypes of hepatocellular carcinoma Cancer Res 68 2008 1451 1461 [15] L. Xie W. Qin J. Li X. He H. Zhang G. Yao BNIPL-2 promotes the invasion and metastasis of human hepatocellular carcinoma cells Oncol Rep 17 2007 605 610 [16] R.T. Poon I.O. Ng C. Lau W.C. Yu Z.F. Yang S.T. Fan Tumor microvessel density as a predictor of recurrence after resection of hepatocellular carcinoma: a prospective study J Clin Oncol 20 2002 1775 1785 [17] C. Wittekind Pitfalls in the classification of liver tumors Pathology 27 2006 289 293 [18] Q. Gao S.J. Qiu J. Fan J. Zhou X.Y. Wang Y.S. Xiao Intratumoral balance of regulatory and cytotoxic T cells is associated with prognosis of hepatocellular carcinoma after resection J Clin Oncol 25 2007 2586 2593 [19] T. Yamabuki A. Takano S. Hayama N. Ishikawa T. Kato M. Miyamoto Dikkopf-1 as a novel serologic and prognostic biomarker for lung and esophageal carcinomas Cancer Res 67 2007 2517 2525 [20] B. Terris P. Pineau L. Bregeaud D. Valla J. Belghiti P. Tiollais Close correlation between beta-catenin gene alterations and nuclear accumulation of the protein in human hepatocellular carcinomas Oncogene 18 1999 6583 6588 [21] M. Shimada T. Hamatsu Y. Yamashita T. Rikimaru K. Taguchi T. Utsunomiya Characteristics of multicentric hepatocellular carcinomas: comparison with intrahepatic metastasis World J Surg 25 2001 991 995 [22] A. Niida T. Hiroko M. Kasai Y. Furukawa Y. Nakamura Y. Suzuki DKK1, a negative regulator of Wnt signaling, is a target of the beta-catenin/TCF pathway Oncogene 23 2004 8520 8526 [23] S.Y. Peng W.J. Chen P.L. Lai Y.M. Jeng J.C. Sheu H.C. Hsu High alpha-fetoprotein level correlates with high stage, early recurrence and poor prognosis of hepatocellular carcinoma: significance of hepatitis virus infection, age, p53 and beta-catenin mutations Int J Cancer 112 2004 44 50 [24] Y.C. Hsu H.H. Fu Y.M. Jeng P.H. Lee S.D. Yang Proline directed protein kinase FA is a powerful and independent prognostic predictor for progression and patient survival of hepatocellular carcinoma J Clin Oncol 24 2006 3780 3788 [25] X.D. Zhu J.B. Zhang P.Y. Zhuang H.G. Zhu W. Zhang Y.Q. Xiong High expression of macrophage colony-stimulating factor in peritumoral liver tissue is associated with poor survival after curative resection of hepatocellular carcinoma J Clin Oncol 26 2008 2707 2716 [26] S. Ramesh J. Nash P.G. McCulloch Reduction in membranous expression of beta-catenin and increased cytoplasmic E-cadherin expression predict poor survival in gastric cancer Br J Cancer 81 1999 1392 1397 [27] T.J. Hugh S.A. Dillon B.A. Taylor M. Pignatelli G.J. Poston A.R. Kinsella Cadherin–catenin expression in primary colorectal cancer: a survival analysis Br J Cancer 80 1999 1046 1051 [28] S. Inagawa M. Itabashi S. Adachi T. Kawamoto M. Hori J. Shimazaki Expression and prognostic roles of beta-catenin in hepatocellular carcinoma: correlation with tumor progression and postoperative survival Clin Cancer Res 8 2002 450 456 [29] A. Schmitt-Gräff V. Ertelt H.P. Allgaier K. Koelble M. Olschewski R. Nitschke Cellular retinol-binding protein-1 in hepatocellular carcinoma correlates with beta-catenin, Ki-67 index, and patient survival Hepatology 38 2003 470 480 [30] S. Tulac M.T. Overgaard A.E. Hamilton N.L. Jumbe E. Suchanek L.C. Giudice Dickkopf-1, an inhibitor of Wnt signaling, is regulated by progesterone in human endometrial stromal cells J Clin Endocrinol Metab 91 2006 1453 1461 [31] O. Aguilera C. Peña J.M. Garcı´a M.J. Larriba P. Ordóñez-Morán D. Navarro The Wnt antagonist DICKKOPF-1 gene is induced by 1alpha,25-dihydroxyvitamin D3 associated to the differentiation of human colon cancer cells Carcinogenesis 28 2007 1877 1884 [32] N. Kane M. Jones J.J. Brosens P.T. Saunders R.W. Kelly H.O. Critchley Transforming growth factor-beta1 attenuates expression of both the progesterone receptor and Dickkopf in differentiated human endometrial stromal cells Mol Endocrinol 22 2008 716 728