Clinical and molecular insights into the hepatocellular carcinoma tumour marker des-c-carboxyprothrombin

Transcription

1 Liver International ISSN REVIEW ARTICLE Clinical and molecular insights into the hepatocellular carcinoma tumour marker des-c-carboxyprothrombin Yoshinori Inagaki 1, Wei Tang 1, Masatoshi Makuuchi 2, Kiyoshi Hasegawa 1, Yasuhiko Sugawara 1 and Norihiro Kokudo 1 1 Hepato-Biliary-Pancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, the University of Tokyo, Tokyo, Japan 2 Department of Digestive Surgery, Japanese Red Cross Medical Center, Tokyo, Japan Keywords angiogenesis cancer cell growth des-g-carboxyprothrombin diagnosis drug discovery hepatocellular carcinoma prognosis tumor marker Abbreviations AFP, a-fetoprotein; AFP-L3, Lens culinaris agglutinin-reactive fraction of a-fetoprotein; bfgf, basic fibroblast growth factor; DCP, des-g-carboxyprothrombin; ECLIA, electrochemiluminescent immunoassay; EFC, epithelial-to-fibroblastoid conversion; EGFR, epidermal growth factor receptor; EIA, enzyme immunoassay; EMT, epithelial-to-mesenchymal transition; Gla, g-carboxylated glutamic acid; Glu, glutamic acid; HCC, hepatocellular carcinoma; HUVEC, human umbilical vein endothelial cell; JAK, janus kinase; KDR, kinase insert domain receptor; LDLT, living donor liver transplantation; MAPK, mitogen-activated protein kinase; PIVKA-II, protein induced by vitamin K absence or antagonist II; PLC-g, phospholipase C-g; STAT3, signal transducers and activators of transcription3; TGF-a, transforming growth factor-a; VEGF, vascular endothelial growth factor. Abstract Des-g-carboxyprothrombin (DCP) is known as a tumour marker for hepatocellular carcinoma (HCC). Various tumour markers have been developed for serological diagnosis of cancers, including HCC, in order to increase the survival rate of cancer patients. The currently recommended combined testing of DCP and a-fetoprotein (AFP) or Lens culinaris agglutinin-reactive fraction of a-fetoprotein has been established to diagnose HCC. This combined testing using several tumour markers helps to increase the sensitivity of diagnosis of HCC, thus significantly increasing the clinical usefulness of DCP. The excessive production of DCP may be related to worse tumour behaviour, such as the presence of vascular invasion and intrahepatic metastasis of HCC cells. A high level of DCP was suggested to be useful as one of the factors in new recipient selection criteria of liver transplantation. The clinical use of DCP, therefore, might play a vital role in predicting tumour behaviour in patients with HCC. That said, the basic mechanism of DCP production has not been fully clarified. Various factors such as vitamin K 2 and g-glutamyl carboxylase may contribute to the production of DCP and have a complex relationship. Moreover, recent studies have revealed that DCP functions as a growth factor and might play significant roles in cancer progression. Thus, DCP represents a potential target of drug discovery to establish new chemotherapeutic strategy for HCC. However, various issues have to be resolved to construct a novel therapy for HCC-targeting DCP. Innovation is required to make further progress in examining DCP. Correspondence Wei Tang, MD, PhD, Hepato-Biliary-Pancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, the University of Tokyo, Hongo, Bunkyo-ku, Tokyo , Japan Tel: (ext ) Fax: tang-sur@h.u-tokyo.ac.jp Received 12 July 2010 Accepted 29 August 2010 DOI: /j x Des-g-carboxyprothrombin (DCP), also known as the protein induced by vitamin K absence or antagonist II (PIVKA-II), is known to be an effective tumour marker for hepatocellular carcinoma (HCC) (1 3). This tumour marker is currently approved in Japan, Korea and Indonesia, while the researches have been performed 22 c 2010 John Wiley & Sons A/S

2 Inagaki et al. Insights into des-g-carboxyprothrombin worldwide. Extensive scientific evidence has been assembled to indicate the clinical usage of DCP. A significant relationship between an elevated DCP level and worse tumour behaviour has been indentified, and this has helped improve prognosis for patients with HCC. In addition, recent molecular biological studies on DCP have revealed not only the usefulness of this molecule as a diagnostic marker but also its significant role in cancer progression. The present article reviews the characteristics of DCP and its clinical usefulness as a diagnostic marker and describes the biochemical functions of DCP in relation to cancer progression. The article closes with a discussion of future perspectives in the field of DCP investigation. Mechanism of production Characteristics Prothrombin, a coagulation factor, is synthesized in a vitamin K-dependent manner in liver tissues. DCP is an abnormal prothrombin that lacks the ability to interact with other coagulation factors. The difference between normal prothrombin and DCP is the composition of amino acid residues. The prothrombin molecule has some functional domain structures (4), and there are 10 g-carboxylated glutamic acid (Gla) residues in the N-terminal domain, called the Gla domain. These Gla residues are originally glutamic acid (Glu) residues in prothrombin precursor and these residues are completely synthesized by the vitamin K-dependent enzymatic reaction of g-glutamyl carboxylase as post-translational modification (Fig. 1) (5). When this reaction is insufficient in conditions like a vitamin K deficiency, DCP with Glu residues remaining in the Gla domain without g-carboxylation is expressed and secreted extracellularly (6). Thus, DCP exists as many types of molecules with different numbers of Gla residues (7). Mechanism of production The exact cause of DCP production in HCC tissues has not been understood as yet. In the past, numerous studies have been performed to clarify the mechanism of DCP production and these studies have suggested some widely accepted possibilities. One possibility is that the activity of g-glutamyl carboxylase declines in HCC tissues. Shah et al. (8) performed an analysis using rat models of the Morris hepatoma tumour to clarify the significance of g-glutamyl carboxylase activity. Their study revealed that g-glutamyl carboxylase activity was Fig. 1. Production of normal prothrombin and des-g-carboxyprothrombin (DCP) in hepatic cells. Under normal conditions, g-glutamyl carboxylase completely alters 10 glutamic acid (Glu) residues in the Gla domain of prothrombin precursor to 10 g-carboxylated glutamic acid (Gla) residues in the presence of vitamin K as a cofactor. Under abnormal conditions, this reaction is insufficient. If fewer than 10 Glu residues are altered to Gla residues, the protein molecule is abnormal prothrombin, that is, DCP. c 2010 John Wiley & Sons A/S 23

3 Insights into des-g-carboxyprothrombin Inagaki et al. lower, especially in DCP-positive HCC tissues in comparison with DCP-negative HCC tissues, although abnormal g-glutamyl carboxylase activity was not detected in normal liver tissues. Similar results were also obtained by an in vitro experiment using DCP-producing and non-dcp-producing rat hepatoma cell lines (9). Furthermore, a recent study indicated that expression of a splice variant of g-glutamyl carboxylase was detected in DCP-producing HCC cell lines. Ueda et al. (10) analysed the characteristics of g-glutamyl carboxylase mrna and protein in some HCC cell lines and compared their production of DCP. As a result, an exon 2 splice variant of g-glutamyl carboxylase was detected in DCP-positive HCC cell lines but not in DCP-negative HCC cell lines. The expression of abnormal g-glutamyl carboxylase might lead to the decrease of this enzyme activity in HCC. These studies indicated that an elevated DCP level may be a consequence of decreased g-glutamyl carboxylase activity but not of abnormal prothrombin expression. Although post-transcriptional abnormality in g-glutamyl carboxylase may cause its dysfunction, this mechanism should be analysed in further studies. Another possibility is that the availability of vitamin K declined as the result of abnormal vitamin K metabolism. Okuda et al. (11) revealed that the HCC cell line produced DCP in a time- and cell number-dependent manner in the absence of vitamin K but not in the presence of vitamin K. Another clinical study measured the serum levels of DCP in patients administered or not administered vitamin K and this study showed that the serum level of DCP declined as a result of the administration of vitamin K, and the levels of some vitamin K derivatives in patients with HCC were significantly elevated (12). These findings suggest that an elevated DCP level may be caused by a vitamin K insufficiency in HCC cells. A recent study focused on the impairment of vitamin K uptake in HCC cell as a cause of DCP production and analysed its relationship to epithelialto-fibroblastoid conversion (EFC) in vitro (13). The impairment of LDL uptake as a surrogate of vitamin K uptake was detected in EFC-induced HCC cells. An elevated DCP expression was also detected in EFCinduced HCC cells but not in untreated control cells, although the expression of E-cadherin decreased significantly. This simultaneous phenomenon of DCP expression and E-cadherin downregulation was also observed in human surgically resected HCC tissues (14). In an analysis of cytoskeletal filaments, disruption of F-actin was observed in EFC-induced HCC cells where DCP production was detected. These results suggest that the alteration of cytoskeletal filaments by EFC plays an important role in DCP production in HCC cells. Epithelial-to-mesenchymal transition (EMT) has been investigated and may play important roles in various mechanisms of cancer progression. Moreover, hypoxic stimulation of HCC cells was suggested to be an important factor for DCP expression via the induction of EFC (15). Although DCP production is not a specific event in HCC, the unavailability of vitamin K and the subsequent induction of DCP production may signify the results of cancer progression, like hypoxia and EMT. Vitamin K is significantly involved in the production of DCP, but more convincing evidence is necessary to elucidate the exact relationship among these factors. Furthermore, the overexpression of prothrombin precursor may also cause DCP production. Ono et al. (16) indicated that the level of prothrombin precursor in HCC tissues was significantly elevated in patients with an elevated serum DCP level, while there was no significant difference between HCC and non-hcc tissues in the levels of endogenous vitamin K. Excessive production of prothrombin precursor was also observed in some HCC cell lines (17). An elevated prothrombin precursor level suggests that the activity of g-glutamyl carboxylase per unit amount of endogenous prothrombin precursor decreased in HCC tissues in comparison with non-hcc liver tissues (17). These findings suggest that excessive production of prothrombin precursors may cause DCP overexpression. Production of prothrombin is affected by the expression and activity of various factors, hence indicating the complexity of the mechanism of DCP production. These various abnormal conditions or phenomena, not separately but in combination, may cause DCP production (Fig. 2). In addition, those abnormalities may vary depending on the characteristics of HCC cell lines and cancer behaviour in patients with HCC. Although further studies should be performed to clarify the principal factors, the excessive production of DCP by HCC cells is likely to be induced by the malignant transformation of HCC. Clinical usefulness Des-g-carboxyprothrombin is a well-known and widely used serological tumour marker for HCC in clinical medicine. Numerous studies have been performed to investigate the usefulness of DCP, and a variety of clinical evidence has been assembled as methods to measure the serum levels of DCP have been developed. This section reviews the development of techniques to measure DCP and the clinical usefulness of the serum DCP level. Furthermore, it describes the derivation of DCP from sera of patients with HCC and its clinical usefulness. Development of methods to measure the levels Several previous studies of DCP have provided scientific evidence on DCP and also indicated several simple and effective methods of detecting DCP derived from different kinds of materials such as tissues and sera. In 1984, Liebman et al. (18) first reported that the levels of DCP in sera of patients with HCC were significantly elevated and that this elevated level was related to the recurrence of HCC after surgery. This study was performed by means of competitive radioimmunoassay with a polyclonal antibody to detect the level of DCP. Next, a method of 24 c 2010 John Wiley & Sons A/S

4 Inagaki et al. Insights into des-g-carboxyprothrombin Fig. 2. Proposed mechanisms of production of des-g-carboxyprothrombin (DCP) in hepatocellular carcinoma (HCC) cells. Three biological abnormalities decrease of g-glutamyl carboxylase activity, insufficiency of vitamin K and overexpression of prothrombin precursor have been suggested as the contributing factors of DCP production. In recent studies, hypoxic stimulation followed by induction of epithelialto-fibroblastoid conversion (EFC)/epithelial-to-mesenchymal transition (EMT) and impairment uptake of vitamin K was suggested to cause insufficiency of vitamin K. The complex interaction among these various abnormal conditions or phenomena may play an important role in DCP production. quantifying plasma DCP levels using enzyme immunoassay (EIA) with anti-dcp monoclonal antibody was developed. Some studies reported that the levels of plasma DCP were frequently elevated in 63% of patients with HCC, with a cut-off level of 0.1 AU/ml, and that these levels normalized after surgery in some patients while they increased again in case of recurrence of the disease (19). These studies suggest that DCP could be used as a sensitive marker for HCC diagnosis and that the determination of DCP levels by EIA with monoclonal antibody is a powerful method of diagnosing HCC behaviour easily and rapidly. However, the sensitivity of this method had to be increased to screen for patients with a small HCC. Currently, an EIA kit with a higher sensitivity (Eitest PIVKA-II, Eisai, Tokyo, Japan) and an electrochemiluminescence kit (Picolumi PIVKA-II, Eisai) have been developed and used in clinical practice. They possess a 10-fold higher sensitivity in comparison with the previous EIA kit as a result of modifying the method of EIA or devising new electrochemical methods (20, 21). Using this more sensitive EIA, Mita et al. (22) analysed serum DCP levels in 91 patients with HCC and 57 with cirrhosis and indicated that 56 of 91 patients with HCC had positive DCP levels with a cut-off level of 40 mau/ml, while only three of 57 cirrhosis patients had positive DCP levels. The sensitivity and specificity of the test were estimated to be 61.5 and 94.7% respectively. In addition, the electrochemiluminescent immunoassay (ECLIA) system may have the potential to detect a small HCC (23, 24). These data on measuring the level of DCP have been assembled using the sensitive methods that are currently in use and have enabled clinicians to screen more patients with HCC. DCP is now effectively used as a tumour marker for the early detection of HCC in patients. Clinical significance of serological levels Usefulness in screening of patients with hepatocellular carcinoma Substantial evidence has been assembled by performing many clinical trials and clarifying the clinical usefulness of the serum DCP level in screening patients with HCC (1 3, 19, 22 45). Table 1 summarizes the current clinical usefulness of the serum DCP level before and after curative therapy (supporting information provides detailed information regarding the potential of the serum DCP level as a marker to distinguish patients with HCC from patients with other chronic liver diseases). The level of DCP in serum was not elevated in patients with chronic liver diseases such as hepatitis and cirrhosis, and this, therefore, represents a highly specific characteristic of DCP. While this high specificity was indicated in almost all of the studies, the sensitivity varied among c 2010 John Wiley & Sons A/S 25

5 Insights into des-g-carboxyprothrombin Inagaki et al. Table 1. Current usefulness of serum des-g-carboxyprothrombin level for diagnosis of patients with hepatocellular carcinoma in various time points Time point Usefulness of high level of DCP Before curative therapy Before liver transplantation After curative therapy Screening of patients with HCC. The combined test with AFP (or AFP-L3) is strongly recommended for good sensitivity. The current conventional cut-off level of DCP in ECLIA method is 40 mau/ml. A candidate of recipient selection criteria for liver transplantation. But the effective cut-off level of DCP is currently under consideration. Diagnosis of HCC recurrence. Serum DCP level may be elevated after curative therapy even if the level of DCP before curative therapy was not high. The combined test with AFP (or AFP-L3) is strongly recommended for good sensitivity. AFP, a-fetoprotein; AFP-L3, Lens culinaris agglutinin-reactive fraction of a-fetoprotein; DCP, des-g-carboxyprothrombin; ECLIA, electrochemiluminescent immunoassay; HCC, hepatocellular carcinoma. the studies. Although the development of methods to detect DCP has conferred increased sensitivity, low sensitivity is still a detriment of DCP and particularly so when analysing patients with small HCC. Okuda et al. (28) found that the sensitivity when analysing patients with an HCC smaller than 2 cm or smaller than 3 cm was 35.0 and 39.3%, respectively, while sensitivity in patients with HCC overall was 60.0%. Another strategy was needed to distinguish patients with small HCC with increased sensitivity. Besides DCP, there are two other serum markers for HCC diagnosis, a-fetoprotein (AFP) and Lens culinaris agglutinin-reactive fraction of AFP (AFP-L3) (22 26, 28 31, 33, 35, 38 41, 44). Many studies have been performed to compare the clinical usefulness of these three tumour markers, but the final conclusions are still under debate. A very recent study performed in cooperation with several facilities in the US highlighted the validity of these HCC markers (39). On setting the cutoff level at the recommended level for each marker, DCP (74%) displayed higher sensitivity than AFP (59%) and AFP-L3 (42%). When the cut-off level was determined by the receiver operating characteristic curve, AFP showed a higher sensitivity (70%) than DCP (66%) and AFP-L3 (50%). This study concluded that AFP was the best serum marker to diagnose early-stage HCC at 10.9 ng/ml, but suggested that a further prospective cohort study be performed in order to generalize this result to all patients with HCC. Because an elevated DCP level in sera of patients with HCC may have no relation to elevated levels of AFP and AFP-L3, a simultaneous test with a combination of these tumour markers is more effective at sensitively diagnosing HCC at present (23 26, 28 31, 40 41, 44). In the study performed by Ikoma et al. (26), on analyzing patients with an HCC smaller than 2 cm or 3 cm, the sensitivity was observed to be 38.5 and 48.6%, Table 2. Usefulness of serum des-g-carboxyprothrombin level for prediction of prognosis for patients with hepatocellular carcinoma Degree of confidence Strong Moderate Under consideration Correlated characteristics with a high level of serum DCP Higher risk of vascular invasion Higher risk of HCC recurrence Worse overall survival Higher risk of intrahepatic metastasis Higher risk of poor differentiation Higher risk of large HCC size Higher risk of pseudocapsule formation Higher risk of multiple HCC nodules DCP, des-g-carboxyprothrombin; HCC, hepatocellular carcinoma. respectively, when only DCP was used as a marker (26). However, when both DCP and AFP were used in combination, the sensitivity increased to 61.5 and 82.9% respectively. Because DCP is insufficient at distinguishing patients with small HCC, as noted earlier, AFP or AFP-L3 can compensate for its low sensitivity. The Japanese Evidence-based Clinical Practice Guidelines and Consensus-based Clinical Practice Manual has noted that periodic and simultaneous measurement of DCP and AFP (or AFP-L3) levels is strongly recommended to screen patients with HCC and particularly to detect a small HCC with a high level of sensitivity and specificity (46). Although the optimal method of diagnosing HCC using these serological tumor markers is still being debated, the combination of these makers may be essential for the early diagnosis of HCC. Usefulness in predicting prognosis for patients with hepatocellular carcinoma Many studies on the relationship between serum DCP level and various clinicopathological features of HCC have suggested that elevated DCP reflects worse tumour behaviour and prognosis for patients with HCC (47 78). Table 2 summarizes the clinical usefulness of serum DCP level to predict prognosis for patients with HCC (supporting information indicates the results of respective studies that analysed the relationship between the level of DCP and prognosis for HCC patients and the data also shows the clinicopathological significance of DCP). A number of studies showed that elevated serum DCP is significantly related to portal vein invasion and/or intrahepatic metastasis, which significantly affect prognosis for patients with HCC. In a study that simultaneously measured DCP and AFP-L3, both DCP- and AFP-L3- positive patients had higher frequencies of infiltrative growth type, vascular invasion and intrahepatic metastasis than DCP-negative patients (73). This combination may be effective both as a marker to distinguish HCC and as a diagnostic marker of HCC behaviour. Furthermore, several studies have investigated the potential use of DCP as an indicator of HCC recurrence after curative therapy. Tang et al. (76) found that a positive serum DCP level was significantly related to the presence of vascular 26 c 2010 John Wiley & Sons A/S

6 Inagaki et al. Insights into des-g-carboxyprothrombin invasion, intrahepatic metastasis, tumour size and TNM stage as well as a high frequency of tumour recurrence (76). Therefore, the overall survival rate was worse for patients with positive serum DCP levels than for those without them (76). Imamura et al. (66) indicated that the serum DCP level was significantly elevated in the patients with vascular invasion, intrahepatic metastasis and worse recurrence-free survival, but they did not designate the DCP level as an independent prognostic factor. Among the studies, the results of multivariate analysis varied; therefore, more evidence must be assembled regarding the usefulness of DCP as an independent prognostic factor. A prospective study was also performed to clarify the significance of DCP in the prediction of portal vein invasion. It revealed that portal vein invasion was significant in patients who were positive for DCP in comparison with those who were negative for DCP, and accordingly, the positivity of DCP may be the strongest predisposing factor for portal vein invasion (61). Currently, the serum DCP level can be used as a diagnostic marker to screen patients with HCC and as an indictor of a therapy s effectiveness and to predict tumour recurrence and patient prognosis. Moreover, DCP may be used as a prognostic indicator for patients with small HCC. A high level of DCP in serum was also related to a higher risk of HCC recurrence and worse overall survival of patients with an HCC smaller than 3 cm (64, 67). DCP can be used to evaluate the prognosis for patients with small HCC but is still insufficient for use in the primary screening of patients with small HCC. Although an elevated level of DCP in serum disappears after the curative treatment of primary HCC, it frequently occurs again as a result of enlargement of recurrent HCC (61). Thus, a relationship between the level of DCP in serum and proliferative potential of HCC has been suggested. An immunohistochemical study also indicated that the serum level of DCP was related to the degree of proliferation of HCC tissues (65). This significant relationship between a high level of DCP and the enlargement of HCC suggests that DCP might play a role in the activation of HCC growth. Recent molecular biological studies showed the direct effect of DCP on HCC cell growth and are noted elsewhere. Usefulness in selecting recipient of liver transplantation Des-g-carboxyprothrombin is being focused as a candidate of new recipient selection criteria for living donor liver transplantation (LDLT). Nowadays, Milan criteria (single nodule 5cm or 3 nodules, all 3 cm) are widely used as selection criteria for LDLT (79). This and other selection criteria such as the University of California at San Francisco criteria (single nodule 6.5 cm or 3 nodules, all 4.5 cm, and a total diameter 8 cm) are based on the size and number of tumour nodules (79, 80). On the other hand, the presence of vascular invasion and poor tumour grade have been clarified to be strong independent factors for poor prognosis after liver transplantation for HCC (81). Therefore, the preoperative prediction of these phenomena based on the levels of DCP and AFP was thought to contribute towards improving the prognosis of LDLT, because a high level of DCP and AFP is a good predictor of the presence of vascular invasion and poor differentiation, respectively, as noted above and in Table 2. In fact, the combination of existing criteria and levels of these tumour markers increased the accuracies for predicting the absence of vascular invasion and a well-tomoderate grade of tumour differentiation in patients with HCC (82). On the basis of these backgrounds, new criteria for selecting patients with HCC undergoing LDLT, which are combined with the levels of DCP and AFP, have been suggested. The research group in the University of Tokushima showed that disease-free survival was significantly better in patients with both a DCP level of 300 mau/ml and a tumour diameter of 5 cm than those who did not meet these criteria (57). In multicentre analysis performed by Todo et al. (70), both DCP and AFP were nominated as risk factors for patient survival and HCC recurrence, and the combination of DCP of 4100 mau/ml and AFP of 4200 ng/ml accompanied with Milan criteria was suggested to be a good predictor of worse prognosis after LDLT. However, the multivariate analysis performed by several recent studies showed that a higher DCP level represented independent risk factor for HCC recurrence after LDLT but a higher AFP level did not (58, 59). The research group in Kyushu University set the new criteria at both a DCP level of o300 mau/ml and a tumour diameter of o5 cm on the basis of the multivariate analysis (58). The research group in Kyoto University added the number of tumour nodules in their criteria and set at a DCP level of 400 mau/ml and 10 nodules, all with a diameter of o5 cm (59), and their latest research developments showed that the group of patients beyond their criteria had the worst cumulative survival rate compared with the group of patients beyond Milan criteria but within their criteria or the group of patients within Milan criteria (59). In addition, these studies clarified the significant correlation between a higher level of DCP and the presence of histological microvascular invasion in resected HCC tissues from recipients of LDLT (57 59). Thus, the results of these studies suggested that the new criteria, combined with serological DCP level, tumour size and number of nodules, can be used as the superior criteria for selecting recipient of LDLT because a high level of DCP indicates the pathological states that are important for post-operative survival such as vascular invasion. However, the newly established criteria varied among the research groups, and this might be caused by the difference of system and policy on the recipient selection of liver transplantation in each hospital. Further validation studies should be performed in order to construct the effective diagnostic system in recipient selection of liver transplantation for HCC. c 2010 John Wiley & Sons A/S 27

7 Insights into des-g-carboxyprothrombin Inagaki et al. Histological expression and its relationship to serological levels Because DCP in circulation is originally produced by HCC tissues, a reasonable assumption is that the level of DCP expression in tissues determines the serum DCP level (83). In actuality, the level of DCP expression in HCC tissues correlates with the serum DCP level and is related to the biological aggressiveness of HCC and the prognosis of small HCC (75). However, serum DCP in patients with HCC may be derived from sources other than HCC tissues (84). The overexpression of DCP is detected in HCC tissues as well as in the surrounding non-hcc liver tissues, and patients with overexpression of DCP in the surrounding non-hcc tissues have significantly elevated serum DCP levels (76). In addition, ECLIA to determine quantities of serum and tissue DCP revealed that levels of DCP expression in non-hcc tissues correlated with those in HCC tissues and serum DCP levels (85), and this result was also confirmed in a subsequent study (78). These findings suggest that an elevated serum DCP level may originate in HCC tissues as well as in surrounding non-hcc tissues. Moreover, Tang et al. (77) performed a further study to clarify the clinical usefulness of tissue and serum DCP levels. In survival analysis, a patient group with both tissue overexpressing DCP and elevated serum DCP levels had the worst outcome in comparison with the other patient groups. Furthermore, multivariate analysis revealed that DCP expression in total liver tissue was a significant prognostic factor along with intrahepatic metastasis. This study showed that DCP expression in either HCC or the surrounding liver tissue might be a powerful tool for diagnosing HCC behaviour along with the serum DCP level (Fig. 3). Moreover, investigating the biological mechanism of elevated DCP level in patients with HCC requires taking into account the overexpression of DCP in HCC tissue as well as in the surrounding noncancerous liver tissue. Biochemical significance Although numerous studies have contributed to the recognition of DCP s clinical significance as a useful diagnostic marker for HCC, reasons why DCP is overexpressed in HCC tissue and why this overexpression is related to tumour enlargement and malignancy of HCC have not been clarified thus far. However, several basic studies on the function of DCP in HCC cells may offer important clues to answer these questions. Proliferation of hepatocellular carcinoma cells The structure of DCP contains two kringle domains that are similar to those of the hepatocyte growth factor (4, 86). The kringle domains are necessary for HGF to bind to its receptor, called Met, and induce cell proliferation. DCP may have the ability to bind with Met and cause Fig. 3. Clinical usefulness of histochemical expression and serum level of des-g-carboxyprothrombin (DCP). Histochemical expression of DCP was detected in both hepatocellular carcinoma (HCC) and surrounding non-cancerous liver tissues, and this might be significantly related to worse tumour behaviour. The serum DCP level is also related to tumor malignancy and the combined serological testing of DCP and a-fetoprotein (AFP) or Lens culinaris agglutinin-reactive fraction of a-fetoprotein (AFP-L3) is a particularly effective method for diagnosing patients with HCC. Furthermore, a combination of histological and serological analyses may point to increased clinical significance. 28 c 2010 John Wiley & Sons A/S

8 Inagaki et al. Insights into des-g-carboxyprothrombin cells, and particularly HCC cells, to proliferate. In 2005, Suzuki et al. (87) investigated the biological function of DCP in HCC cell proliferation. According to their report, levels of DNA synthesis in HCC cell lines were significantly enhanced by adding purified DCP and this enhancement was more marked in non-dcp-producing cell lines than in DCP-producing cell lines. This result indicates that DCP can induce the proliferation of HCC cells. The researchers also analysed the mechanism of this phenomenon and found that DCP bound with Met and stimulated Met-Janus kinase-signal transducers and activators of the transcription 3 pathway as the signalling pathway to induce HCC cell proliferation. These results suggest that DCP can induce the proliferation of HCC cells by functioning like HGF. Ma et al. (88) investigated the effect of decreasing DCP secretion on the proliferation of HCC cells. Levels of DCP secretion decreased significantly in HCC cell lines with the addition of vitamin K 2. The researchers administered the vitamin to evaluate the effect of the decrease in DCP. As a result, doing so also affected the malignancy of HCC cell lines. Growth of those cell lines was significantly inhibited in a dose-dependent manner by vitamin K 2. In vivo analysis revealed that weight of the tumour xenograft and the serum level of DCP in nude mice decreased significantly with the addition of vitamin K 2 without any significant signs of acute or delayed toxicity. Thus, this study indicated that reducing the secreted amount of DCP may be effective at inhibiting cancer progression. DCP may relate to worse tumour behavior by enhancing the proliferation of HCC cells. Angiogenesis around hepatocellular carcinoma tissue Another biological effect of DCP in malignancy of HCC is that DCP may have the ability to enhance angiogenesis around HCC tissues. Fujikawa et al. (89) analysed the biological function of DCP in angiogenesis using human umbilical vein endothelial cells (HUVEC). Their results indicated that DCP stimulated DNA synthesis and the migrative activity of HUVEC, but normal prothrombin did not. Regarding the activation of HUVEC, they reported that DCP bound to kinase insert domain receptor (KDR), which is also known as vascular endothelial growth factor receptor, and stimulated the KDR phospholipase C-(PLC-g) mitogen-activated protein kinase (MAPK) signalling pathway and then subsequently accelerated DNA synthesis and cell migration (89). A recent in vitro study found that DCP had the ability to enhance growth and invasion of HUVEC cells (90). Furthermore, DCP induced overexpression of epidermal growth factor receptor (EGFR) and vascular endothelial growth factor (VEGF) in HUVEC cells (90). These findings indicate that DCP secreted from HCC cells can induce angiogenesis in surrounding HCC tissues and worsen the prognosis for patients with HCC. Gao et al. (91) analysed the effect of DCP on the expression of various angiogenic factors produced by HCC cells and showed that secretion of VEGF, transforming growth factor-a (TGF-a) and basic fibroblast growth factor (bfgf) by HCC cells was enhanced by DCP treatment. An immunohistochemical study to evaluate the effect of DCP in HCC cell xenografts in nude mice found that the levels of VEGF, TGF-a and bfgf expression increased in a dose-dependent manner after 2 weeks of DCP stimulation. Microvessel density around the HCC cell xenograft also increased significantly as a result of DCP stimulation. Therefore, DCP might be able to induce the migration of vascular endothelial cells as well as the production of angiogenic factors by HCC cells themselves. Angiogenesis is a particularly important phenomenon involved in the continuous growth of tumour tissues and worsening tumor behaviour, e.g. dedifferentiation and hypervascularity, and a high tumour microvessel density may cause HCC tumour recurrence (92). These studies suggest the significant roles of DCP in cancer progression via the induction of angiogenesis and may help to understand the mechanism by which DCP aggravates HCC behaviour. Future perspectives The studies cited have indicated that DCP plays several important roles in HCC progression and may explain why cancer behaviour and patient prognosis worsen in patients with DCP-positive HCC in comparison with those with DCP-negative HCC. DCP is not just an abnormal prothrombin but may be a potential cancerenhancing protein (Fig. 4). This suggests that DCP may be able to be used both as a diagnostic marker of HCC and as a therapeutic target for HCC (93). Currently, several molecular targeted anticancer agents have been developed and used in clinical medicine to treat various kinds of cancers. Nevertheless, few anticancer agents are effective at treating HCC although some treatment methods using intra-arterial infusion chemotherapy have been established (94). Recently, a novel molecular targeted agent named sorafenib, a multikinase inhibitor, has been developed as a chemotherapeutic agent for HCC, and the phase III, double-blind, randomized, placebo-controlled trial clarified that sorafenib was effective for the prolonged survival and time to progression in patients with advanced HCC (95, 96). Clinical trials using sorafenib as an adjuvant therapy after surgery are under way worldwide (97). Therefore, molecular targeted medicine is important to establish an effective method of treating HCC to improve the prognosis for patients with HCC. Investigating DCP as a target of cancer therapy may contribute to the development of novel chemotherapeutic agents for HCC. As noted earlier, DCP is related to some cancer-associated events such as self-proliferation and angiogenesis. Thus, if these events can be inhibited by some DCP inhibitor, the progression of DCP-positive HCC may be suppressed. However, constructing novel DCP inhibitors c 2010 John Wiley & Sons A/S 29

9 Insights into des-g-carboxyprothrombin Inagaki et al. Fig. 4. Proposed physiological functions of des-g-carboxyprothrombin (DCP) and the expected progression. Autocrine/paracrine secretion of DCP affects the self-proliferation of HCC cells via the Met-Janus kinase (JAK)-signal transducers and activators of the transcription3 (STAT3) signalling pathway. DCP also has the ability to enhance the production of various angiogenic factors in hepatocellular carcinoma (HCC), although the detailed mechanism is still unknown. In addition, paracrine secretion of DCP affects the proliferation and migrative activity of vascular endothelial cells via the kinase insert domain receptor (KDR) phospholipase C-g (PLC-g) mitogen-activated protein kinase (MAPK) signalling pathway. Secreted DCP can function as a growth factor and as an angiogenic factor, and hence the inhibition of DCP might help suppress cancer aggressiveness. will not be simple as the nature of DCP, such as its three-dimensional conformation, is yet to be fully understood. In addition, the mechanism that most significantly affects DCP production is still unclear, although some important biochemical events have been suggested, as noted previously. Further investigation should be conducted to increase the usefulness of DCP. As noted earlier, insufficiency of vitamin K 2 is an important factor in producing DCP in HCC cells. Previous clinical studies have found that administration of vitamin K 2 suppresses the recurrence of HCC after curative treatment and contributes to the prolonged survival of patients with HCC (98). Furthermore, molecular biological studies showed that administering vitamin K 2 or its analogue, named menatetrenone, inhibits the activity of protein with regard to intracellular signalling, regulating the growth and invasive activity of HCC cells (99 101). These studies indicate that vitamin K 2 has the potential to suppress the progression of HCC by inhibiting several biological systems, although the relationship to the reduction in DCP expression is still unclear. However, several clinical studies also noted that the effects of vitamin K 2 as a chemopreventive or chemotherapeutic agent are insufficient (102). Therefore, this strategy should be modified to make it more potent. Although further investigation on a larger scale should be performed, combining vitamin K 2 with a chemotherapeutic agent could be effective at preventing or treating HCC. A recent clinical trial showed that a combination of vitamin K 2 and an angiotensin-converting enzyme inhibitor is effective at suppressing the cumulative recurrence of HCC after curative therapy (103). A previous 30 c 2010 John Wiley & Sons A/S

10 Inagaki et al. Insights into des-g-carboxyprothrombin in vivo study found that suppression of angiogenesis may be related to the chemopreventive effect that prevents HCC recurrence (104). These studies prove that multidrug chemotherapy using agents that have no effect on the sole regimen might represent a chemotherapeutic strategy for patients with HCC. As noted previously, DCP secreted from HCC cells has the ability to act as a growth factor for HCC cells and vascular endothelial cells and is significantly associated with the progression of HCC. Developing an inhibitor of DCP and its use in multidrug chemotherapy may induce antiproliferative and anti-angiogenic action with fewer side effects. While DCP continues to be available as a diagnostic marker, DCP is also a promising target for the development of HCC chemotherapy. Conclusion Early diagnosis and proper therapeutics are the most important factors to increasing the survival rate for cancer patients in the current medical treatment. This review has focused on the clinical usefulness of DCP in diagnosing and treating HCC. DCP is now established as an effective tumour marker for HCC, and the development of highly sensitive methods to detect DCP will help in the early diagnosis of HCC. Establishment of new criteria for LDLT based on the DCP level will also contribute towards improving the prognosis of patients undergoing LDLT. Moreover, recent findings regarding the biological functions of DCP in HCC progression may provide clues to novel therapeutics for HCC. Although various issues must still be resolved, further elucidation of DCP will contribute to better medical care for patients with HCC. Acknowledgements Conflict of interest: No conflicts of interest to disclose. Funding: This work was supported by Grants-in-Aid from Japan Society for the Promotion of Science and the Ministry of Education, Culture, Sports, Science and Technology of Japan. References 1. Marrero JA, Su GL, Wei W, et al. Des-gamma carboxyprothrombin can differentiate hepatocellular carcinoma from nonmalignant chronic liver disease in American patients. Hepatology 2003; 37: Wang CS, Lin CL, Lee HC, et al. Usefulness of serum desgamma-carboxy prothrombin in detection of hepatocellular carcinoma. World J Gastroenterol 2005; 11: Kim do Y, Paik YH, Ahn SH, et al. PIVKA-II is a useful tumor marker for recurrent hepatocellular carcinoma after surgical resection. Oncology 2007; 72: Mann KG. Prothrombin. Methods Enzymol 1976; 45: Suttie JW. Vitamin K-dependent carboxylase. Annu Rev Biochem 1985; 54: Furie B, Furie BC. Molecular basis of vitamin K-dependent gamma-carboxylation. Blood 1990; 75: Naraki T, Kohno N, Saito H, et al. Gamma-carboxyglutamic acid content of hepatocellular carcinoma-associated des-gamma-carboxy prothrombin. Biochim Biophys Acta 2002; 1586: Shah DV, Engelke JA, Suttie JW. Abnormal prothrombin in the plasma of rats carrying hepatic tumors. Blood 1987; 69: Shah DV, Zhang P, Engelke JA, Bach AU, Suttie JW. Vitamin K-dependent carboxylase activity, prothrombin mrna, and prothrombin production in two cultured rat hepatoma cell lines. Thromb Res 1993; 70: Ueda N, Shiraha H, Fujikawa T, et al. Exon 2 deletion splice variant of gamma-glutamyl carboxylase causes desgamma-carboxy prothrombin production in hepatocellular carcinoma cell lines. Mol Oncol 2008; 2: Okuda H, Obata H, Nakanishi T, Furukawa R, Hashimoto E. Production of abnormal prothrombin (des-gammacarboxy prothrombin) by hepatocellular carcinoma. A clinical and experimental study. J Hepatol 1987; 4: Sakon M, Monden M, Gotoh M, et al. The effects of vitamin K on the generation of des-gamma-carboxy prothrombin (PIVKA-II) in patients with hepatocellular carcinoma. Am J Gastroenterol 1991; 86: Murata K, Sakamoto A. Impairment of clathrin-mediated endocytosis via cytoskeletal change by epithelial to fibroblastoid conversion in HepG2 cells: a possible mechanism of des-gamma-carboxy prothrombin production in hepatocellular carcinoma. Int J Oncol 2008; 33: Murata K, Suzuki H, Okano H, et al. Cytoskeletal changes during epithelial-to-fibroblastoid conversion as a crucial mechanism of des-gamma-carboxy prothrombin production in hepatocellular carcinoma. Int J Oncol 2009; 35: Murata K, Suzuki H, Okano H, et al. Hypoxia-induced des-gamma-carboxy prothrombin production in hepatocellular carcinoma. Int J Oncol 2010; 36: Ono M, Ohta H, Ohhira M, Sekiya C, Namiki M. Measurement of immunoreactive prothrombin, des-gamma-carboxy prothrombin, and vitamin K in human liver tissues: overproduction of immunoreactive prothrombin in hepatocellular carcinoma. Am J Gastroenterol 1990; 85: Yamagata H, Nakanishi T, Furukawa M, Okuda H, Obata H. Levels of vitamin K, immunoreactive prothrombin, des-gamma-carboxy prothrombin and gamma-glutamyl carboxylase activity in hepatocellular carcinoma tissue. J Gastroenterol Hepatol 1995; 10: Liebman HA, Furie BC, Tong MJ, et al. Des-gammacarboxy (abnormal) prothrombin as a serum marker of primary hepatocellular carcinoma. N Engl J Med 1984; 310: FujiyamaS,MorishitaT,HashiguchiO,SatoT.Plasma abnormal prothrombin (des-gamma-carboxy prothrombin) c 2010 John Wiley & Sons A/S 31

11 Insights into des-g-carboxyprothrombin Inagaki et al. as a marker of hepatocellular carcinoma. Cancer 1988; 61: Suzuki H, Akahane Y, Tanaka M, et al. Clinical evaluation of PIVKA-II kit (ED-036). Kan Tan Sui 1996; 33: Takatsu K, Nakanishi T, Watanabe K, et al. Development and performance of an assay kit for PIVKA-II (ED-038) by ECL technique. Jpn J Clin Exp Med 1996; 73: Mita Y, Aoyagi Y, Yanagi M, et al. The usefulness of determining des-gamma-carboxy prothrombin by sensitive enzyme immunoassay in the early diagnosis of patients with hepatocellular carcinoma. Cancer 1998; 82: Nomura F, Ishijima M, Kuwa K, et al. Serum des-gammacarboxy prothrombin levels determined by a new generation of sensitive immunoassays in patients with small-sized hepatocellular carcinoma. Am J Gastroenterol 1999; 94: Sassa T, Kumada T, Nakano S, Uematsu T. Clinical utility of simultaneous measurement of serum high-sensitivity des-gamma-carboxy prothrombin and Lens culinaris agglutinin A-reactive alpha-fetoprotein in patients with small hepatocellular carcinoma. Eur J Gastroenterol Hepatol 1999; 11: Soga K, Watanabe T, Aikawa K, et al. Serum des-gammacarboxyprothrombin level by a modified enzyme immunoassay method in hepatocellular carcinoma: clinical significance in small hepatocellular carcinoma. Hepatogastroenterology 1998; 45: Ikoma J, Kaito M, Ishihara T, et al. Early diagnosis of hepatocellular carcinoma using a sensitive assay for serum des-gamma-carboxy prothrombin: a prospective study. Hepatogastroenterology 2002; 49: Kuromatsu R, Tanaka M, Shimauchi Y, et al. Usefulness of ED036 kit for measuring serum PIVKA-II levels in small hepatocellular carcinoma. J Gastroenterol 1997; 32: Okuda H, Nakanishi T, Takatsu K, et al. Measurement of serum levels of des-gamma-carboxy prothrombin in patients with hepatocellular carcinoma by a revised enzyme immunoassay kit with increased sensitivity. Cancer 1999; 85: Tanaka Y, Kashiwagi T, Tsutsumi H, et al. Sensitive measurement of serum abnormal prothrombin (PIVKA- II) as a marker of hepatocellular carcinoma. Hepatogastroenterology 1999; 46: Cui R, He J, Zhang F, et al. Diagnostic value of protein induced by vitamin K absence (PIVKAII) and hepatoma-specific band of serum gamma-glutamyl transferase (GGTII) as hepatocellular carcinoma markers complementary to alpha-fetoprotein. Br J Cancer 2003; 88: Ishii M, Gama H, Chida N, et al. Simultaneous measurements of serum alpha-fetoprotein and protein induced by vitamin K absence for detecting hepatocellular carcinoma. South Tohoku district study group. Am J Gastroenterol 2000; 95: Durazo FA, Blatt LM, Corey WG, et al. Des-gammacarboxyprothrombin, alpha-fetoprotein and AFP-L3 in patients with chronic hepatitis, cirrhosis and hepatocellular carcinoma. J Gastroenterol Hepatol 2008; 23: Lamerz R, Runge M, Stieber P, Meissner E. Use of serum PIVKA-II (DCP) determination for differentiation between benign and malignant liver diseases. Anticancer Res 1999; 19: Tsai SL, Huang GT, Yang PM, et al. Plasma des-gammacarboxyprothrombin in the early stage of hepatocellular carcinoma. Hepatology 1990; 11: Nakao A, Suzuki Y, Isshiki K, et al. Clinical evaluation of plasma abnormal prothrombin (des-gamma-carboxy prothrombin) in hepatobiliary malignancies and other diseases. Am J Gastroenterol 1991; 86: Kasahara A, Hayashi N, Fusamoto H, et al. Clinical evaluation of plasma des-gamma-carboxy prothrombin as a marker protein of hepatocellular carcinoma in patients with tumors of various sizes. Dig Dis Sci 1993; 38: Shiraki K, Takase K, Tameda Y, et al. A clinical study of lectin-reactive alpha-fetoprotein as an early indicator of hepatocellular carcinoma in the follow-up of cirrhotic patients. Hepatology 1995; 22: Suehiro T, Sugimachi K, Matsumata T, et al. Protein induced by vitamin K absence or antagonist II as a prognostic marker in hepatocellular carcinoma. Comparison with alpha-fetoprotein. Cancer 1994; 73: Marrero JA, Feng Z, Wang Y, et al. Alpha-fetoprotein, des-gamma carboxyprothrombin, and lectin-bound alpha-fetoprotein in early hepatocellular carcinoma. Gastroenterology 2009; 137: Beale G, Chattopadhyay D, Gray J, et al. AFP, PIVKAII, GP3, SCCA-1 and follisatin as surveillance biomarkers for hepatocellular cancer in non-alcoholic and alcoholic fatty liver disease. BMC Cancer 2008; 8: Deyashiki Y, Nishioka Y, Takahashi K, Kosaka Y, Suzuki K. Evaluation of des-gamma-carboxy prothrombin as a marker protein of hepatocellular carcinoma. Cancer 1989; 64: Tanabe Y, Ohnishi K, Nomura F, Iida S. Plasma abnormal prothrombin levels in patients with small hepatocellular carcinoma. Am J Gastroenterol 1988; 83: Nomura F, Ishijima M, Horikoshi A, Nakai T, Ohnishi K. Determination of serum des-gamma-carboxy prothrombin levels in patients with small-sized hepatocellular carcinoma: comparison of the conventional enzyme immunoassay and two modified methods. Am J Gastroenterol 1996; 91: Shimauchi Y, Tanaka M, Kuromatsu R, et al. A simultaneous monitoring of Lens culinaris agglutinin A-reactive alpha-fetoprotein and des-gamma-carboxy prothrombin as an early diagnosis of hepatocellular carcinoma in the follow-up of cirrhotic patients. Oncol Rep 2000; 7: Saitoh S, Ikeda K, Koida I, et al. Serum des-gammacarboxyprothrombin concentration determined by the avidin biotin complex method in small hepatocellular carcinomas. Cancer 1994; 74: c 2010 John Wiley & Sons A/S