ANTICANCER RESEARCH 35: 2021-2028 (2015)

Gene Expression Analysis for Evaluation of Potential Biomarkers in Hepatocellular

HOLGER HASS1, JUERGEN JOBST1, MICHAEL SCHEURLEN3, ULRICH VOGEL4 and OLIVER NEHLS5

1Department of Oncology, Paracelsus Hospital, Scheidegg, Germany; 2Matrigene GmbH, Reutlingen, Germany; 3Department of Gastroenterology, Hematology, Oncology and Rheumatology, University of Wuerzburg, Wuerzburg, Germany; 4Department of Pathology, University of Tuebingen, Tuebingen, Germany; 5Praxis of Gastroenterology and Tumor Medicine, Stuttgart, Germany

Abstract. Background: The poor prognosis of (HCC) is the fifth most common hepatocellular carcinoma (HCC) and the lack of specific cancer and the third most common cause of death by cancer screening markers underline the need for new biomarkers for worldwide (1). In the last few decades, the incidence of HCC human hepatocarcinogenesis. Materials and Methods: We has increased, likely due to the growing worldwide prevalence investigated 10 postulated biomarkers for HCC (AFP, GPC3, of chronic hepatitis B or C. In particular, the emergence of the OPN, IGF1, HGF, SPINK1, KPNA, FUCA1, CgA, HSP90) hepatitis C virus (HCV) in developed countries can account with microarray expression analysis and real-time for approximately half of the increase in HCC (1, 2). polymerase chain reaction (RT-PCR) in HCC tissues of Currently, surgical resection and liver transplantation offer different etiologies. Results: Four candidate (FUCA1, the best potential treatments for HCC (3, 4). However, those HGF, IGF1, CgA) showed low median fold changes (fc) of treatments are mostly only available when tumors are expression compared to corresponding non-malignant liver detected early. Thus, a curable tumor resection is only a tissues (fc range=0.2-0.8; maximum 15% of samples). The realistic treatment option in a minority of patients that classic biomarker, alpha-fetoprotein (AFP), was significantly develop HCC as a late effect of chronic viral hepatitis with over-expressed (fc=2.4) in 30% of tumors. High tumor AFP advanced end-stage liver disease. These patients currently expression was associated with significantly elevated serum account for 15-25% of HCC cases. The poor prognosis of AFP concentrations (>90% of cases). Five genes (OPN, HCC and its increasing incidence underline the importance SPINK1, GPC3, HSP90, KNPA2) showed significantly of discovering novel screening biomarkers to detect tumors higher expression than AFP in 64% to 82% of samples at an earlier, curable stage. (median fc range=2.9-8.3). RT-PCR analyses gave similar Alpha-fetoprotein (AFP) is currently the most common, results. Conclusion: Unlike previous studies, our results did classical used for HCC evaluations. In 1956, not confirm FUCA1, HGF, IGF1 or CgA as potential AFP was discovered by Bergstrand and Czar (5). The markers for HCC. In contrast, OPN, SPINK1, GPC3 and earliest reports on the potential function of AFP as a KNPA2 were significantly over-expressed in HCC tissues. diagnostic marker for HCC were published in the 1960s (6). These genes may be useful in developing future biomarkers AFP is a glycoprotein of approximately 70 kDa, synthesized and therapeutic strategies for HCC. during early embryonic development and, subsequently, in fetal hepatocytes. The AFP gene has been linked to chromo - some 4, which harbors part of the albuminoid gene super - family (7). Significantly elevated serum AFP levels were associated with various, non-malignant liver diseases, including viral hepatitis or liver cirrhosis; thus, AFP had low Correspondence to: Holger G. Hass, Department of Oncology, sensitivity (20-60%) and specificity (76-96%) as a Paracelsus Hospital, Kurstrasse 5, 88175 Scheidegg, Germany. Tel: biomarker for HCC, particularly in the early stages (8, 9). +49 8381501225, Fax:+49 8381501290, e-mail: dr.holger.hass@ Other studies have shown significantly lower positive paracelsus-kliniken.de predictive values for AFP in different etiological subtypes Key Words: Hepatocellular carcinoma, biomarkers, gene expression of HCC, particularly in patients with virally-induced HCC analysis, alpha-fetoprotein, osteopontin, glypican 3. (10). Based on these data, screening with AFP is currently

0250-7005/2015 $2.00+.40 2021 ANTICANCER RESEARCH 35: 2021-2028 (2015) not recommended by the German Association for the Study Carlsbad, CA, USA) to transcribe the purified, extracted total RNA of the Liver (11). The absence of a specific biomarker for to obtain double stranded cDNA. The first strand cDNA synthesis HCC with higher sensitivity and specificity than AFP was primed with a T7-(dT24) oligonucleotide primer, after second- strand synthesis, in vitro transcription was performed in the presence underlines the need to establish other biomarkers for human of biotin-11-CTP and biotin-16-UTP (Enzo Diagnostics, city, hepatocarcinogenesis. county, country) to produce biotin labeled cRNA. Hybridization was Modern gene expression analysis allows the detection of performed by incubating (18-20 h) 200 μl of the sample with an specific genetic patterns and the molecular pathways human GeneChip (Hu133A; Affymetrix, Santa Clara, CA, USA), involved in HCC (12, 13). This approach may facilitate the which contained 22,283 probe sets for known genes or expressed detection of better screening markers for HCC. Recent sequence tags (ESTs). A Gene Array scanner G2500A (Hewlett studies that employed this technique have described several Packard, Palo Alto, CA, USA) was used to scan the chips according to procedures developed by Affymetrix. genes and gene products that were significantly overexpressed in tumor tissues and showed potential as new Statistical analysis. The statistical analyses and presentation of data biomarkers for HCC (14, 15). were performed in accordance with the MIAME criteria. The data In the present study, we analyzed gene expression levels will be published on the following web page: http://www.paracelsus- for 10 currently postulated biomarkers for HCC, including kliniken.de/scheidegg/fuer-fachkreise/forschung/gene-profiles. AFP, glypican 3 (GPC3), osteopontin (OPN), insulin-like growth factor-1 (IGF1), hepatic growth factor (HGF), Preliminary data analysis. We conducted a raw data analysis with the Affymetrix microarray suite software (MAS v5.0.1, Santa Clara, CA, specific serine protease inhibitor kazal type 1 (SPINK1), USA). Statistical analyses and post-processing were performed with karyopherin α (KPNA), fucosidase, alpha-L- 1 (FUCA1), GeneSpring (v6.1; Silicon Genetics, Redwood City, CA, USA) and chomogranin A (CgA) and heat shock -90 (HSP90) to GeneExplore software (v1.1; Applied Maths, Sint-Martens-Laten, determine their potential use in screening or as prognostic Belgium). Expression values were then log2-transformed on the basis indicators in patients with an elevated risk for HCC. of the signal:noise log ratio, which is given by comparing array results between tumor and non-malignant tissues. A p-value<0.05 Materials and Methods (t-test) and a fold-change ≥2 in 60% or more of all analyzed samples were considered significant. Acquisition of samples. Surgical specimens (both tumorous and corresponding non-malignant liver tissue) were obtained from 33 Selection of ten candidate screening markers for HCC. After searching © patients with HCC that underwent surgical treatment for the literature (Pubmed , NCBI, Bethesda, MD, USA), we identified histologically-proven HCC (10 HCV-positive, 4 hepatitis-B virus ten recently postulated screening biomarkers that we considered (HBV)-positive, 19 non-virally-induced; Table I). appropriate for individual expression analysis in the 33 HCC samples. In accordance with the Declaration of Helsinki, we obtained The expression of these genes was compared with expression in Ethics Committee approval. Parts of the resected samples were used corresponding, non-malignant liver tissues. Therefore, our screening for further genomic analyses, after obtaining written informed panel included probes for the following genes: AFP, OPN, SPINK1, consent from the patients. GPC3, HSP90, FUCA1, HGF, IGF1, CgA and KNPA2.

Comparison of AFP gene expression levels and AFP serum levels. Validation of expression data by real-time polymerase chain Serum AFP levels were measured in all patients before liver surgery. In reaction (RT-PCR). After microarray analyses, we used RT-PCR to all cases, a minimum of 5 ml serum was stored and AFP was analyzed validate the relative changes in gene expression observed for the with the chemiluminescence technique (CLIA, Immulite AFP; most consistently over-expressed genes in HCC tissues (SPINK1, Siemens Healthcare, Erlangen, Germany). Based on the fact that OPN, GPC3, KPNA). RT-PCR was performed with the chronic viral infection or inflammation may lead to elevated AFP LightCycler© system (Roche Diagnostics, Mannheim, Germany) on serum levels, an AFP elevation ≥200 ng/ml was considered sufficiently cDNA isolated from tumor tissues and corresponding non-malignant significant for suspecting hepatocellular carcinoma (HCC). liver tissues. The gene-specific primers corresponded to the coding region of each gene. These primers were designed with the OLIGO Histopathological evaluation. For each tissue sample, we obtained software, supplied by Biomers.net (Ulm, Germany) and had the clinical, virological and pathological reports of tumor typing, following sequences: staging (based on UICC criteria) and grading, in addition to patient SPINK1: 151U; GCCTTGGCCCTGTTGAGTCTA clinical data. Hematoxylin-eosin staining was performed for SPINK1: 273L; CACGCATTCATTGGGATAAGTATTT detecting features, like bile canalicular structure and Mallory GPC3: 1808U; CAGCAGGCAACTCCGAAGG hyaline bodies. Additional histological and immunohistochemical GPC3: 1929L; TGGGCACCAGGCAGTCAGT stainings with a panel of antibodies (HEP-PAR-1, AFP, CK7, CA19- KPNA: 328U; GAAAACCGCAACAACCA 9, CD10, CEA) were routinely performed for all resected tissue KPNA: 501L; GCCCAAGAAGGACACAAAT biopsies to confirm the histological diagnosis of HCC and to OPN: 696U; GGACAGCCGTGGGAAGG exclude other types of malignancies. OPN: 810L; TCAATCACATCGGAATGCTCA

Preparation of labeled cRNA and hybridization to oligonucleotide RT-PCR reactions were performed with the LC RNA Amplification arrays. We used the SuperScript Choice system (Invitrogen, Kit SYBR Green I, (Roche Diagnostics, Man nheim, Germany).

2022 Hass et al: Biomarker Expression in Hepatocellular Carcinoma

Figure 1. Two-dimensional cluster analysis of gene expression in HCC (TU; n=33) and the corresponding non-malignant liver tissue (NL; n=20). Expression of 1,085 significantly dysregulated genes/ESTs are shown (red: up-regulated, green: down-regulated genes). Gene expression was dysregulated (fc≥2) in ≥ 60% of samples (Pearson correlation).

Amplification was followed by melting curve analysis. Relative Inc., Salt Lake City, UT, USA). The relative change in gene values for the initial target concentration in each sample were expression was computed by pairwise comparisons of tumor samples determined with the LightCycler software, v3.5 (Idaho Technology to samples of adjacent normal tissue for each patient.

2023 ANTICANCER RESEARCH 35: 2021-2028 (2015)

Figure 2. Mean expression changes (fc: fold change) for different Figure 3. Comparison of serum AFP levels and corresponding tumor potential tumor markers in HCC samples. The changes in expression AFP gene expression in patients with HCC (n=33). were estimated with gene array cluster analysis (Affymetrix HU133A; light grey) and with RT-PCR (LightCycler®System; dark grey).

OPN, SPINK1, GPC3 and KNPA2 in malignant and Results corresponding non-malignant tissue samples with quantitative RT-PCR. In general, the changes in gene Gene expression profiling of HCC. Based on the 22,283 expression measured with the microarray technique reflected probe sets present on the chip, we found that, on average, the results obtained with RT-PCR. However, the dynamic 42.6% (HCC) and 39.8% (corresponding non-malignant liver range of the expression levels estimated by RT-PCR was tissue) of genes were expressed in the liver tissue samples. significantly higher (10- to 70-fold) than the ranges Out of approximately 13,000 genes expressed in the HCC and estimated in the microarray analysis (Figure 2). adjacent non-malignant tissue samples, approximately 1,200 genes were either up- or down-regulated significantly. All Gene expression and serum AFP levels. To determine statistically different genes (p<0.05) with a fold change of at whether the changes in AFP gene expression found in HCC least a 2-fold increase or decrease in at least 60% of the HCC tissues reflected changes in protein expression, we measured tumor samples were used to generate a databank of 1,085 serum AFP levels in all patients. We found that serum AFP genes/ESTs. Out of these, 445 were up-regulated and 640 were levels ranged between 5 and 170,000 U/l. Moreover, in down-regulated. With this data set, we implemented a supervised 30.3% of all patients, AFP levels were significantly elevated learning method (SLM) based on neuronal networking to obtain (≥200 U/l). In fact, 91% of serum AFP levels significantly a specific gene expression profile for HCC that enabled rapid, correlated (p=0.001) with gene expression levels estimated reproducible differentiation between malignant and non- with gene expression analysis (Figure 3). malignant liver tissues in all cases (Figure 1). Discussion Potential screening marker expression detected with microarray analysis. Four of the chosen candidate genes The poor prognosis and disappointing treatment options for (FUCA1, HGF, IGF1, CgA) were not significantly patients with non-resectable HCC underline the need for overexpressed. Their median fold change (fc) ranged from better screening parameters and therapeutic targets. To date, 0.2 to 0.8 and they were over-expressed in a maximum of AFP is the only tumor marker established for routine clinical 15% of all analyzed tumor tissues. These genes showed screening and treatment observations in patients with HCC. significantly lower expression than AFP expression; However, consistent with other published data, our results therefore, they were not considered in further analyses. demonstrated that AFP had very low specificity in the 33 AFP gene expression was increased, with a median fc of tumors analyzed in the present study. Only 30% of all tumors 2.4-fold, and it was significantly over-expressed in 30% of all showed over-expression of AFP, with a median fold change tumor samples. In addition, five genes (OPN, SPINK1, GPC3, (fc) of 2.4. Also, in contrast to other groups (10), we observed HSP90, KNPA2) showed significant over-expression (median an AFP over-expression more frequently in virally-induced fc between 2.9 and 8.3) in 64% to 82% of samples (Table I). tumors (36-40%). Among the patients that showed AFP overexpression in the microarray analysis, 91% also showed Potential screening marker expression detected with RT-PCR elevated serum AFP levels (≥200 U/l). Only two cases analysis. To validate the five most highly over-expressed showed elevated serum AFP levels without microarray candidate genes, we confirmed the expression levels of AFP, detection of significant gene over-expression. Taken together,

2024 Hass et al: Biomarker Expression in Hepatocellular Carcinoma

Table I. Expression levels of different potential tumor markers in HCC samples (n=33). Gene expression is expressed as the fold-change (fc) compared to expression in non-tumoral tissues, estimated from oligonucleotide array results.

Gene Median fc (log2) Overexpression in Overexpression in Overexpression in Overexpression in all HCV+HCC (%) HBV+HCC (%) (%) non-virally- HCC samples (%) induced HCC

AFP 2.4 36 40 26 30 SPINK1 5.7 81 80 56 66 OPN 6.2 63 60 65 64 HGF 0.5 9 20 4 7 IGF1 0.8 0 0 26 15 FUCA1 0.2 0 0 4 2 GPC3 8.3 81 80 73 76 CgA 0.2 0 0 0 0 KPNA2 3.3 100 60 78 82 HSP90 2.9 81 60 78 76 fc=Fold change (gene expression level in tumorous vs. non-tumorous liver tissue)

our results showed that AFP gene expression levels correlated Among the five genes we identified, the most very well with serum levels, but AFP was only useful as a overexpressed, independent of the etiological risk factor (i.e., screening marker for HCC in about one third of all patients. HCV, HBV or alcohol), was GPC3 (fc=8.3 in 76% of Modern diagnostic approaches, like gene expression tumors). GPC3 was first described by Filmus and colleagues profiling with DNA or oligonucleotide arrays or whole (22) (originally called OCI-5). They discovered it in an genome sequencing, are useful for discovering potential undifferentiated rat epithelial cell line. The first reports about candidate genes that may serve as biomarkers or therapeutic up-regulation of GPC3 in HCC were published in 1997 (23). targets. In the present study, we used oligonucleotide arrays In the following years, other studies confirmed those reports to analyze the expression of 10 candidate genes that were with cDNA or microarray techniques (24), which provided discussed in the literature as potential screening and the basis for performing immunohistochemical studies to prognostic markers in human hepatocarcinogenesis. To elucidate the value of GPC3 as a screening marker for HCC confirm the expression data, we used RT-PCR and melting (25). In those studies, GPC3 showed high sensitivity (>70%) curve analysis to analyze the marker genes whose over- and specificity (up to 90%) for discriminating HCC from expression was most likely linked to HCC. other tumors or non-malignant hepatopathies. Interestingly, four of the genes postulated as screening or Importantly, in addition to its usefulness as a marker for prognostic markers for HCC (FUCA1, CgA, HGF, IGF1) (16- the early diagnosis of HCC, GPC3 also functions as an 194) were found to be inferior to AFP in their association with oncogene that promotes the growth of HCC by stimulating HCC. The expression changes (fc=0.2-0.8) were significantly canonical Wnt signaling (26). Due to the high expression of lower than the change in AFP expression and expression was GPC3, particularly in malignant transformed hepatocytes, notably different from controls in only 0-15% of the analyzed compared to other molecular targets in hepatocarcinogenesis, cases. In our opinion, these genes may not be useful for current studies are developing anti-GPC3 strategies for routine clinical screening or for detecting early HCC. treating HCC (27). The other five genes (OPN, SPINK1, GPC3, KPNA2, OPN was also significantly up-regulated (fc=6.2) in over HSP90) included in our microarray analysis were found to 60% of the analyzed HCC samples. OPN (also called secreted be superior to AFP in their association with HCC. They phosphoprotein 1) is not a typical oncogene; it is mutated in showed significantly higher gene expression changes activated cancer cells and it is regulated by a variety of stimuli, (fc=2.9-8.3) than AFP and expression was notably different including the Wnt/Tcf signaling pathway, steroid receptors, from controls in 64-82% of the cases analyzed. Moreover, growth factors, tumor necrosis factor-alpha and transcription these data (except HSP90) were validated in the RT-PCR factors, such as AP-1 and the RAS proto-oncogene (28). OPN analysis. The expression levels determined with RT-PCR is a ligand for the CD44 receptor, binds to alphaV-containing correlated well with the gene expression data. However, as integrins and plays an important role in malignant cell shown previously by our group and other colleagues (20, 21), attachment and tumor invasion (29). OPN is involved in the dynamic range of the RT-PCR data was about 10-fold to several potential mechanisms in human carcinogenesis, which 70-fold higher than that of the microarray analysis. require the regulation of different signaling cascades and

2025 ANTICANCER RESEARCH 35: 2021-2028 (2015) seems to be a potent anti-apoptotic factor via inhibition of An immunohistochemical analysis of tissue samples by caspase 3 activation (30); it also induces the activation of Yoshitake et al. demonstrated (39) cancer-specific over- promatrix metalloproteinase-2 (pro-MMP2, mediated by expression of KNPA2 in up to 36% of HCCs. nuclear factor kappaB) and it induces the secretion of HSP90 was also significantly up-regulated (fc=2.9) in urokinase-type plasminogen activator (31). In a previous gene 76% of our HCC samples. HSP90 is an essential chaperone expression profiling study, our group was the first to show that that facilitates the function and integrity of a wide range of OPN was the most highly over-expressed gene in potential oncogenes (40). HSP90 expression modulates cholangiocarcinoma. In the present study, our data transcription factors (e.g., HIF-1, STAT3) and intracellular corroborated those of recent studies by showing that OPN may signaling pathways (e.g., Akt/Erk pathway), which lead to also be a useful marker for hepatocarcinogenesis. In addition, cell growth and proliferation (41). HSP90 up-regulation, in it was previously shown that OPN expression correlated with combination with cyclin-dependent kinase 4 (CDK4) early recurrence, poor prognosis and metastasis in HCC (32). activation, was predicted to contribute to HCC development. In the present study, SPINK1 (also called PSTI or TATI) Moreover, HSP90 over-expression was associated with a was significantly over-expressed (median fc=5.7), poor prognosis in patients with HCC (42). Recently, several particularly in HCV-induced HCC (up to 80%), compared to studies have shown that HSP90 may be a relevant therapeutic non-malignant liver tissues. Recent studies postulated the target in different tumors, but particularly in HCC (43). oncogenic mechanism and role of SPINK1 and it was Compared to AFP that encodes for the current “gold proposed as an early tumor marker for hepatobiliary cancer standard” for serological screening, the present study (33). This protein was first described in 1982 (34) after suggested that five other genes (OPN, SPINK1, HSP90, detection in patients with . In that study, it was KNPA2, GPC3) showed a greater potential as screening or found to inhibit trypsins and other proteases, which led to its prognostic markers for human hepatocarcinogenesis. Despite description as a tumor-associated inhibitor (TATI). At the small sample size of the current study, our data the end of the 1980s, TATI was proved to be identical to underlined the importance of these five genes in the SPINK1 and PSTI. development of human HCC. Currently, GPN3, HSP90 and As a potent proteinase inhibitor, the expression of TATI in OPN are under investigation as potential therapeutic targets malignant tissues appears to represent the proteolytic and that may contribute to our fight against HCC in the future. metastatic activity of the tumor. Stenman and colleagues identified and isolated two tumor-associated trypsins (TAT-1 Acknowledgements and -2) that were targeted by TATI. These enzymes were shown to activate matrix metalloproteases (MMPs) in vitro This study was supported by the fortüne-program of the University and in vivo in different tumor cell lines and in tumors, of Tuebingen, No. F1281305. respectively (35). Extracellular matrix re-modeling and pro - teolysis mediated by MMPs is associated with cancer Conflicts of Interest invasion and metastasis in advanced tumor stages. Consequently, TATI may play a crucial role in tumor cell The Authors declare no conflict of interest. dissemination and invasion. Other research groups postulated that TATI directly blocked apoptosis by inhibiting caspase- References independent or serine protease-dependent apoptosis, particularly in virally-induced hepatocarcinogenesis (36). 1 World Health Organization. Mortality database. 2010; Available The postulated viral induction of SPINK1 up-regulation may from: www.who.int/whosis/en. explain the up-regulation of this gene observed in the present 2 El-Serag HB and Rudolph KL: Hepatocellular carcinoma: epidemiology and molecular carcinogenesis. Gastroenterology study (37), particularly in the group of patients with HCV- 132: 2557-2576, 2007. induced HCC. This observation suggested that HCV and 3 El-Serag HB, Marrero JA, Rudolph L and Reddy KR: Diagnosis HBV might affect treatment options for both HCC and and treatment of hepatocellular carcinoma. Gastroenterology chronic viral hepatitis, one of the most important risk factors 134: 1752-1763, 2008. for human hepatocarcinogenesis. 4 Forner A, Reig ME, de Lope CR and Bruix J: Current strategy In our data, KNPA2 (also called Importin α2) was over- for staging and treatment: the BCLC update and future expressed in over 80% of all HCCs (median fc=3.3). prospects. Semin Liver Dis 30: 61-74, 2010. Currently, KNPA2 is postulated to be a new tumor marker for 5 Bergstrand CG and Czar B: Demonstration of a new protein fraction in serum from the human fetus. Scand J Clin Lab Invest different adenocarcinomas, particularly (38). 8: 174, 1956. Although its function remains under investigation, 6 Abelev GI: Production of embryonal serum alpha-globulin by particularly in human carcinogenesis, one hypothesis is that hepatomas: review of experimental and clinical data. Cancer Res KNPA2 may be involved in the nuclear transport of . 28: 1344-1350, 1968.

2026 Hass et al: Biomarker Expression in Hepatocellular Carcinoma

7 Yamashita K, Taketa K, Nishi S, Fukushima K and Ohkura T: oligonucleotide microarray and real-time PCR analysis. World J Sugar chains of human cord serum alpha-fetoprotein: characte - Gastroenterol 14: 2501-2510, 2008. ristics of N-linked sugar chains of glycoproteins produced in 21 Rajeevan MS, Vernon SD, Taysavang N and Unger ER: Valida - human liver and hepatocellular . Cancer Res 53: tion of array-based gene expression profiles by real-time 2970-1975, 1993. (kinetic) RT-PCR. J Mol Diagn 3: 26-31, 2001. 8 Kokudo N and Makuuchi M: Evidence-based clinical practice 22 Filmus J, Church JG and Buick RN: Isolation of a cDNA guidelines for hepatocellular carcinoma in Japan: the J-HCC corresponding to a developmentally regulated transcript in rat guidelines. J Gastroenterol 44: 119-121, 2009. intestine. Mol Cell Biol 8: 4243-4249, 1988. 9 Gambarin-Gelwan M, Wolf DC, Shapiro R, Schwartz ME and 23 Hsu HC, Cheng W and Lai PL: Cloning and expression of a Min AD: Sensitivity of commonly available screening tests in developmentally regulated transcript MXR7 in hepatocellular detecting hepatocellular carcinoma in cirrhotic patients undergoing carcinoma: biological significance and temporospatial liver transplantation. Am J Gastroenterol 95: 1535-1538, 2000. distribution. Cancer Res 57: 5179-5184, 1997. 10 Soresi M, Magliarisi C, Campagna P, Leto G, Bonfissuto G, Riili 24 Patil MA, Chua MS, Pan KH, Lin R, Lih CJ, Cheung ST, Ho C, A, Carroccio A, Sesti R, Tripi S and Montalto G: Usefulness of Li R, Fan ST, Cohen SN, Chen X and So S: An integrated data alpha-fetoprotein in the diagnosis of hepatocellular carcinoma. analysis approach to characterize genes highly expressed in Anticancer Res 23: 1747-1753, 2003. hepatocellular carcinoma. Oncogene 24: 3737-3747, 2005. 11 Deutschen Krebsgesellschaft e.V. Diagnostik und Therapie des 25 Ligato S, Mandich D and Cartun RW: Utility of glypican-3 in Hepatozellulären Karzinoms. Version 1.0-März 2013. AWMF- differentiating hepatocellular carcinoma from other primary and Registernummer: 032/053OL. metastatic lesions in FNA of the liver: an immunocytochemical 12 Okabe H, Satoh S, Kato T, Kitahara O, Yanagawa R, Yamaoka Y, study. Mod Pathol 21: 626-631, 2008. Tsunoda T, Furukawa Y and Nakamura Y: Genome-wide analysis 26 De Cat B, Muyldermans SY, Coomans C, Degeest G, of gene expression in human hepatocellular carcinomas using Vanderschueren B, Creemers J, Biemar F, Peers B and David G: cDNA microarray: Identification of genes involved in viral carcino- Processing by proprotein convertases is required for glypican-3 genesis and tumor progression. Cancer Res 61: 2129-2137, 2001. modulation of cell survival, Wnt signaling, and gastrulation 13 Mas VR, Maluf DG, Archer KJ, Yanek K, Kong X, Kulik L, movements. J Cell Biol 163: 625-635, 2003. Freise CE, Olthoff KM, Ghobrial RM, McIver P and Fisher R: 27 Filmus J and Capurro M: Glypican-3: a marker and a therapeutic Genes involved in viral carcinogenesis and tumor initiation in target in hepatocellular carcinoma. FEBS J 280: 2471-2476, 2013. hepatitis C virus-induced hepatocellular carcinoma. Mol Med 28 Casson AG, Wilson SM, McCart JA, O'Malley FP, Ozcelik H, 15: 85-94, 2009. Tsao MS and Chambers AF: ras mutation and expression of the 14 Gehrau RC, Archer KJ, Mas VR and Maluf DG: Molecular ras-regulated genes osteopontin and cathepsin L in human profiles of HCV cirrhotic tissues derived in a panel of markers esophageal cancer. Int J Cancer 72: 739-745, 1997. with clinical utility for hepatocellular carcinoma surveillance. 29 El-Tanani MK, Campbell FC, Kurisetty V, Jin D, McCann M and PLoS One 7: e40275. doi: 10.1371/journal. pone.0040275, 2012. Rudland PS: The regulation and role of osteopontin in malignant 15 Xiang ZL, Zeng ZC, Fan J, Tang ZY, Zeng HY and Gao DM: transformation and cancer. Cytokine Growth Factor Rev 17: 463- Gene expression profiling of fixed tissues identified hypoxia- 474, 2006. inducible factor-1α, VEGF, and matrix metalloproteinase-2 as 30 Graessmann M, Berg B, Fuchs B, Klein A and Graessmann A: biomarkers of lymph node metastasis in hepatocellular Chemotherapy resistance of mouse WAP-SVT/t carcinoma. Clin Cancer Res 17: 5463-5472, 2011. cells is mediated by osteopontin, inhibiting apoptosis down- 16 Adamek A, Kasprzak A, Mikoś H, Przybyszewska W, Seraszek- stream of caspase-3. Oncogene 26: 2840-2850, 2007. Jaros A, Czajka A, Sterzyńska K and Mozer-Lisewska I: The 31 Das R, Philip S, Mahabeleshwar GH, Bulbule A and Kundu GC: insulin-like growth factor-1 and expression of its binding Osteopontin: it's role in regulation of cell motility and nuclear protein-3 in chronic hepatitis C and hepatocellular carcinoma. factor kappa B-mediated urokinase type plasminogen activator Oncol Rep 30: 1337-1345, 2013. expression. IUBMB Life 57: 441-447, 2005. 17 Malaguarnera M, Vacante M, Fichera R, Cappellani A, Cristaldi 32 Pan HW, Ou YH, Peng SY, Liu SH, Lai PL, Lee PH, Sheu JC, E and Motta M: (CgA) serum level as a marker Chen CL and Hsu HC: Overexpression of osteopontin is of progression in hepatocellular carcinoma (HCC) of elderly associated with intrahepatic metastasis, early recurrence, and patients. Arch Gerontol Geriatr 51: 81-85, 2010. poorer prognosis of surgically resected hepatocellular carcinoma. 18 Wang K, Guo W, Li N, Shi J, Zhang C, Lau WY, Wu M and Cancer 98: 119-127, 2003. Cheng S: Alpha-1-fucosidase as a prognostic indicator for 33 Marshall A, Lukk M, Kutter C, Davies S, Alexander G and hepatocellular carcinoma following hepatectomy: a large-scale, Odom DT: Global gene expression profiling reveals SPINK1 as long-term study. Br J Cancer 110: 1811-1819, 2014. a potential hepatocellular carcinoma marker. PLoS One 8: e59459. 19 Ang CS, Sun MY, Huitzil-Melendez DF, Chou JF, Capanu M, doi: 10.1371/journal. pone.0059459, 2013. Jarnagin W, Fong Y, Dematteo RP, D'Angelica M, Allen P, Chen 34 Stenman UH, Huhtala ML, Koistinen R and Seppala M: CT, O'Reilly EM, Weiser MR and Abou-Alfa GK: c-MET and Immunochemical demonstration of an ovarian cancer-associated HGF mRNA expression in hepatocellular carcinoma: correlation urinary peptide. Int J Cancer 30: 53-57, 1982. with clinicopathological features and survival. Anticancer Res 35 Moilanen M, Sorsa T, Stenman M, Nyberg P, Lindy O, Vesterinen 33: 3241-3245, 2013. J, Paju A, Konttinen YT, Stenman UH and Salo T: Tumor- 20 Hass HG, Nehls O, Jobst J, Frilling A, Vogel U and Kaiser S: associated trypsinogen (trypsinogen-2) activates procollagenases Identification of osteopontin as the most consistently over- (MMP-1, -8, -13) and stromelysin-1 (MMP-3) and degrades type expressed gene in intrahepatic cholangiocarcinoma: detection by I collagen. Biochemistry 42: 5414-5420; 2003.

2027 ANTICANCER RESEARCH 35: 2021-2028 (2015)

36 Lamontagne J, Pinkerton M, Block TM and Lu X: Hepatitis B 41 Miyata Y, Nakamoto H and Neckers L: The therapeutic target and hepatitis C virus replication upregulates serine protease Hsp90 and cancer hallmarks. Curr Pharm Des 19: 347-365, 2013. inhibitor Kazal, resulting in cellular resistance to serine protease- 42 Sun Y, Zang Z, Xu X, Zhang Z, Zhong L, Zan W, Zhao Y and dependent apoptosis. J Virol 84: 907-917, 2010. Sun L: Differential Proteomics Identification of HSP90 as 37 Hass HG, Jobst J, Vogel U, Scheurlen M and Nehls O: Potential Serum Biomarker in Hepatocellular Carcinoma by Overexpression of tumor-associated trypsin inhibitor (SPINK1/ Two-dimensional Electrophoresis and Mass Spectrometry. Int J TATI) in hepatitis C-associated hepatocellular carcinoma: Mol Sci 11: 1423-1433, 2010. Potential implications for viral hepatocarcinogenesis. Oncol Res 43 Goyal L, Wadlow RC, Blaszkowsky LS, Wolpin BM, Abrams Treat 37: 732-738, 2014. TA, McCleary NJ, Sheehan S, Sundaram E, Karol MD, Chen J 38 Wang CI, Wang CL, Wang CW, Chen CD, Wu CC, Liang Y, Tsai and Zhu AX: A phase I and pharmacokinetic study of ganetespib YH, Chang YS, Yu JS and Yu CJ: Importin subunit alpha-2 is (STA-9090) in advanced hepatocellular carcinoma. Invest New identified as a potential biomarker for non-small cell lung cancer Drugs Sep 24, 2014. (Epub ahead of print). by integration of the cancer cell secretome and tissue transcriptome. Int J Cancer 128: 2364-2372, 2011. 39 Yoshitake K, Tanaka S, Mogushi K, Aihara A, Murakata A, Matsumura S, Mitsunori Y, Yasen M, Ban D, Noguchi N, Irie T, Kudo A, Nakamura N, Tanaka H and Arii S: Importin-α1 as a novel prognostic target for hepatocellular carcinoma. Ann Surg Oncol 18: 2093-2103, 2011. Received December 15, 2014 40 Isaacs JS, Xu W and Neckers L: Heat shock protein 90 as a mole - Revised January 10, 2015 cular target for cancer therapeutics. Cancer Cell 3: 213-217, 2003. Accepted January 16, 2015

2028