Japanese Journal of Clinical Oncology 33:522-526 (2003)
© 2003 Foundation for Promotion of Cancer Research
Prediction of Invasive Activities in Hepatocellular Carcinomas with Special Reference to 
-Fetoprotein and Des-
-carboxyprothrombin
1 Department of Clinical Chemistry, 2 Department of Physiological Testing and 3 Department of Pathology, National Cancer Center Hospital, Tokyo, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Background: Hepatocellular carcinoma (HCC) is one of the world’s major malignancies, especially in Asian countries. To improve the prognosis of HCC, it is essential to predict its invasive behavior in both intra- and extra-hepatic modalities. For this purpose, we examined the predictive values of two tumor markers,
-fetoprotein (AFP) and des-
-carboxyprothrombin (DCP). Methods: 194 HCC cases at the National Cancer Center Hospital were selected from the Pathology Records. Detailed information regarding the existence of extra-hepatic venous invasion (EHVI) at the portal vein and intra-hepatic multiple malignant tumors (IHMTs) were collected and combined with the preoperative AFP and DCP testing results. Furthermore, information about the viral infection status such as HBs antigen positive or HCV antibody positive or no viral hepatitis was obtained. The information was analyzed by the ROC (Receiver Operating Characteristic Curve) method.
Results and Conclusions: In both the EHVI group and the IHMT group, all the combinations except the HCV-positive group of IHMT revealed a tendency for DCP to offer better diagnostic accuracies, although this was statistically significant only in the HBs-positive group of IHMT. This result indicates either (1) that in a strict sense, DCP is not necessarily better than AFP at predicting the invasive characteristics or (2) that DCP is better than AFP at reflecting the invasive characteristics of HCC although not statistically significant. In either situation, from a long-term viewpoint, it is advisable to find a new marker or to re-evaluate the existing markers in order to predict the invasive characteristics more accurately.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Hepatocellular carcinoma (HCC) is one of the world’s major malignancies, especially in Asian countries (1). It is a major cause of cancer fatalities in Japan (2–5). To improve the prognosis of HCC, it is essential to predict the invasive behavior of HCC in both intra- and extra-hepatic modalities. For this purpose, we focused on two tumor markers because, although the most potent diagnostic tools for HCC are imaging studies including ultrasonography, computed tomography (CT), magnetic resonance imaging and celiac angiography, tumor markers are no less indispensable for diagnosing the disease.
-Fetoprotein (AFP) has served as a representative tumor marker for a long time (6). Des--carboxyprothrombin (DCP) is an abnormal prothrombin that lacks -carboxylation of its glutamine residue. DCP is unable to bind calcium ion that is essential for its conformational transition and functional activity. Because the -carboxylation is vitamin K dependent, the DCP protein appears when a patient is in a vitamin K-deficient state (7). In 1984, Liebman et al. (8) reported a relatively high incidence of abnormal prothrombin (DCP) in HCC patients and suggested that it might be a useful tumor marker of HCC. DCP has been further reported to have a correlation with large tumor size and proliferative activity of tumor tissue (9–13). In our Clinical Laboratory at the National Cancer Center Hospital, there exists a great deal of not only biochemical tumor marker data but also histopathological diagnosis data. Based on these routine laboratory activities, we felt that if we could analyze these biochemical and histopatholgical data statistically, we might be able to find intriguing and clinically useful information. Therefore, we reviewed HCC patients’ records carefully and analyzed them statistically. As a result, we finally revealed some novel findings.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
From January 1999 to May 2002, 33 667 tests for AFP and 17 962 for DCP were performed at the National Cancer Center Hospital. Of these, 194 HCC cases were selected from the Pathology Records because of the availability of their preoperative test results for AFP and DCP. The mean age of these 194 patients was 64.9 years with a range of from 20 to 83 years. The male:female ratio was 6.2:1. Both AFP and DCP were diagnosed at the in-hospital testing laboratory. The AFP test was performed with Lumipulse AFP-N (FujiRebio, Tokyo, Japan) and the DCP test with Eitest PIVKA-II (Sankoh Jyunyaku, Tokyo, Japan). We collected from the Pathology Records mentioned above detailed information regarding the existence of extra-hepatic venous invasion including tumor emboli at the portal vein confirmed by histopathological examination [extra-hepatic venous invasion (EHVI)] and also intra-hepatic multiple malignant tumors confirmed by histopathological examination. The latter included both intra-hepatic metastasis and intra-hepatic multiple tumors (IHMTs) of different origin. Strictly, intra-hepatic multiple tumors and tumors with parenchymal invasion are different. Theoretically, tumors with parenchymal invasion ought not to contain multiple tumors of different origins. Practically, however, it is difficult to differentiate multiple tumors of the same and different origins, although there is a report (14) that examined differentiation of multicentric origin from intra-hepatic metastasis. Therefore, in this study, we defined IHMT as intra-hepatic multiple tumors of either the same or different origin. By reviewing the records of 194 HCC cases, the following information was obtained: viral infection status such as HBs antigen (COBASCORE HBsAg II-EIA, Roche Diagnostics) or HCV antibody (COBASCORE Anti-HCV-EIA, Roche Diagnostics) or no evidence of viral hepatitis. The information obtained was analyzed statistically by the ROC (Receiver Operating Characteristic Curve) method using computer software (ROCKIT 0.9B, available at www.xray.bsd.uchicago.edu/krl/toppage11.htm) (15–17).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Of 194 patients, 46 were only HBs antigen (HBs Ag) positive, of whom 21 showed extra-hepatic venous invasion at the portal vein positive and 25 negative. Ninety-three patients were shown to be only HCV antibody (HCV Ab) positive, of whom 32 showed venous invasion at the portal vein positive and 61 negative. Fifty-one patients exhibited no evidence of viral hepatitis and, of these, 22 showed venous invasion at the portal vein positive and 29 negative (Table 1).
|
Similarly, 46 patients were shown to be only HBs Ag positive and, of these, 21 showed intra-hepatic multiple malignant tumors and 25 were negative. Ninety-three patients were shown to be only HCV Ab positive and, of these, 42 showed intra-hepatic multiple tumors and 51 were negative. Fifty-one patients exhibited no evidence of viral hepatitis and, of these, 20 showed intra-hepatic multiple tumors and 31 were negative (Table 1).
In the only-HBs Ag positive group, AFP and DCP were analyzed by the ROC with special reference to the presence of EHVI at the portal vein. Similar analyses were done for the HCV Ab positive group and also for the no viral hepatitis group. Furthermore, in the only-HBs Ag positive group, AFP and DCP were analyzed by the ROC with special reference to the existence of IHMTs (Table 2 and Fig. 1). Similar analyses were done in the only-HCV Ab positive group and in the no viral hepatitis group. Although there was no statistically significant relationship, DCP showed higher AUCs (areas under the curves) than those of AFP in EHVI in the portal vein analyses. In IHMT analyses, the only-HBs Ag positive group revealed significant diagnostic accuracy in DCP compared with AFP (P = 0.0193). The other HCV Ab positive group and the no viral hepatitis group indicated relatively moderate diagnostic accuracies in the ROC analyses (Table 2 and Fig. 1).
|
|
Regarding the cut-off values, either specificity or sensitivity was set to >60%. The actual cut-off value of AFP appears to be 40–50 ng/ml. Likewise, the cut-off value of DCP appears to be 100–210 mAU/ml. Cases above this 40–50 ng/ml of AFP and 100–210 mAU/ml of DCP consist of around 12–16% of total testings: 33 667 for AFP and 17 962 for DCP (Table 2).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Usually the invasive activities of the tumors towards the extra-hepatic direction is reflected in its EHVI at the portal vein including tumor emboli. There have been several previous studies regarding the correlation between DCP and venous invasion at the portal vein including tumor emboli. Suehiro and co-workers (12,13) repeatedly reported on the positive relationship between DCP and tumor emboli (Fig. 1A, B and C). In the current study, although not statistically significant, there was a tendency for DCP to receive better diagnostic accuracies by the ROC analyses than AFP in HCC regardless of infection status. IHMTs similarly reflect the invasive activities in the intra-hepatic direction. The analyses of this category revealed that DCP was significantly superior to AFP for the diagnosis of HCC in the HBs Ag positive group. In the no viral hepatitis group, DCP showed a better diagnostic accuracy than AFP, although not statistically significant. The last group, such as the HCC with HCV Ab positive group, showed no significant differences between DCP and AFP in diagnostic accuracy in the ROC analyses. Although they were not sub-classified according to the patients’ infection status such as HBs Ag positive, HCV Ab positive or no viral hepatitis, previous studies (10,11) indicated that there was a positive correlation between DCP and the existence of multiple HCC. Recent studies have indicated that there is a possibility that the growth factor-like action of DCP increases the invasive behavior of HCC (18–21), although the reason remains unclear.
In general, AFP and DCP tumor markers seem to be of great help in the characterization of HCC. There are reports investigating the relationship between intra-hepatic parenchymal invasive characteristics (IHMT) and extra-hepatic venous invasive characteristics (EHVI) (8,12,18,19,22). It is noteworthy that Hamamura et al. (19) and Koike et al. (18) performed intriguing studies and concluded that the number and size of tumor foci, the histological tumor grade of HCC differentiation and the serum AFP and DCP levels were significant indicators with regard to the development of portal vein invasion. They contended that among these parameters, the serum DCP level was the most significant predictor for the development of portal vein invasion (18,19). The current study revealed that DCP is significantly superior to AFP at reflecting the IHMT, intra-hepatic parenchymal invasive characteristics of HCC such as the number of tumor foci, in the HBs Ag positive group. On the other hand, regarding EHVI, there is no statistically significant difference in accuracy between DCP and AFP, although there seems to be a tendency indicating that DCP is better than AFP. The reason why there is a difference between IHMT and EHVI is not known. There could be a different mechanism in regulating the intra-hepatic or extra-hepatic direction of invasion in close relationship to the status of viral infection. Another explanation is the influence of multiple tumors of different origins. Investigation of the clonality of each tumor may reveal the actual influence of multiple tumors of different origins (14). These are subjects for future studies.
Before closing this discussion, we must emphasize that it is not our intention to conclude that AFP is not effective when diagnosing HCC. Because the cases examined were surgically resected patients, we can say that our cases of HCC were relatively advanced compared with other studies. In the future, prospective studies of each stage with histological confirmation are required. Regarding the infection status of HCC cases, such as HBs Ag positive, HCV Ab positive or no viral hepatitis, in general there were no remarkable or significant differences between AFP and DCP. This is in contrast to the earlier reports from Japan that differences existed between them (19,23–25). The elucidation of this aspect requires future study.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
We thank Mr Ken-ichi Adachi for helpful comments regarding the ROC analysis. We also thank Mr Robert Debold for English language editing.
|
FOOTNOTES |
|---|
+ Present address: Department of Pathology, Keio University School of Medicine, Tokyo 160-8582, Japan
For reprints and all correspondence: Koh Furuta, Department of Clinical Chemistry and Laboratory Medicine, National Cancer Center Hospital, 5–1–1, Tsukiji, Chuo-ku, Tokyo 104-0045, Japan. E-mail: kfuruta{at}ncc.go.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
1 Wong F, Choi TK. Primary liver cancer. Asian experience. In: Blumgart LH, editor. Surgery of the Liver and Biliary Tract. London, Churchill-Livingstone 1988.
2 Shinagawa T, Ohto M, Kimura K, Tsunetomi S, Morita M, Saisho H, et al. Diagnosis and clinical features of small hepatocellular carcinoma with emphasis on the utility of real-time ultrasonography. A study in 51 patients. Gastroenterology 1984;86:495–502.[ISI][Medline]
3 Takayasu K, Shima Y, Muramatsu Y, Goto H, Moriyama N, Yamada T, et al. Angiography of small hepatocellular carcinomas: analysis of 105 resected tumors. Am J Roentgenol 1986;147:525–9.
4 Yoshimatsu S, Inoue Y, Ibukuro K, Suzuki S. Hypovascular hepatocellular carcinoma undetected at angiography and CT with iodized oil. Radiology 1989;171:343–7.
5 Matsuda Y, Yabuuchi I. Hepatic tumors: US contrast enhancement with CO2 microbubbles. Radiology 1986;161:701–5.
6 Abelev GI. Production of embryonal serum alpha-globulin by hepatomas: review of experimental and clinical data. Cancer Res 1968;28:1344–50.
7 Weitz IC, Liebman HA. Des-gamma-carboxy (abnormal) prothrombin and hepatocellular carcinoma: a critical review. Hepatology 1993;18:990–7.[ISI][Medline]
8 Liebman HA, Furie BC, Tong MJ, Blanchard RA, Lo KJ, Lee SD, et al. Des-gamma-carboxy (abnormal) prothrombin as a serum marker of primary hepatocellular carcinoma. N Engl J Med 1984;310:1427–31.[Abstract]
9 Nakao A, Taniguchi K, Inoue S, Harada A, Nonami T, Watanabe K, et al. Usefulness of simultaneous determination of alpha-fetoprotein and des-gamma-carboxy prothrombin in hepatocellular carcinoma. Semin Surg Oncol 1996;12:160–3.[CrossRef][ISI][Medline]
10 Kasahara A, Hayashi N, Fusamoto H, Kawada Y, Imai Y, Yamamoto 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;38:2170–6.[CrossRef][ISI][Medline]
11 Izuno K, Fujiyama S, Yamasaki K, Sato M, Sato T. Early detection of hepatocellular carcinoma associated with cirrhosis by combined assay of des-gamma-carboxy prothrombin and alpha-fetoprotein: a prospective study. Hepatogastroenterology 1995;42:387–93.[Medline]
12 Suehiro T, Sugimachi K, Matsumata T, Itasaka H, Taketomi A, Maeda T. Protein induced by vitamin K absence or antagonist II as a prognostic marker in hepatocellular carcinoma. Comparison with alpha-fetoprotein. Cancer 1994;73:2464–71.[CrossRef][ISI][Medline]
13 Suehiro T, Matsumata T, Itasaka H, Taketomi A, Yamamoto K, Sugimachi K. Des-gamma-carboxy prothrombin and proliferative activity of hepatocellular carcinoma. Surgery 1995;117:682–91.[CrossRef][ISI][Medline]
14 Ng IO, Guan XY, Poon RT, Fan ST, Lee JM. Determination of the molecular relationship between multiple tumour nodules in hepatocellular carcinoma differentiates multicentric origin from intrahepatic metastasis. J Pathol 2003;199:345–53.[CrossRef][ISI][Medline]
15 Dorfman DD, Berbaum KS, Metz CE. Receiver operating characteristic rating analysis. Generalization to the population of readers and patients with the jackknife method. Invest Radiol 1992;27:723–31.[CrossRef][ISI][Medline]
16 Metz CE, Wang PL, Kronman HB. A new approach for testing the significance of differences between ROC curves measured from correlated data. In: Deconinck F, editor. Information Processing in Medical Imaging. Nijhoff: The Hague 1984;432–45.
17 Metz CE. Statistical analysis of ROC data in evaluating diagnostic performance. In: Myers R, editor. Multiple Regression Analysis: Applications in the Health Sciences. New York: American Institute of Physics 1986;365–84.
18 Koike Y, Shiratori Y, Sato S, Obi S, Teratani T, Imamura M, et al. Des-gamma-carboxy prothrombin as a useful predisposing factor for the development of portal venous invasion in patients with hepatocellular carcinoma: a prospective analysis of 227 patients. Cancer 2001;91:561–9.[CrossRef][ISI][Medline]
19 Hamamura K, Shiratori Y, Shiina S, Imamura M, Obi S, Sato S, et al. Unique clinical characteristics of patients with hepatocellular carcinoma who present with high plasma des-gamma-carboxy prothrombin and low serum alpha-fetoprotein. Cancer 2000;88:1557–64.[CrossRef][ISI][Medline]
20 Carr BI, Wang Z, Wang M, Finn FM. Prothrombin and vitamin K (VK) cycle: a novel endogenous hepatic growth inhibitory pathway. Hepatology 1994;20:272A.
21 Ni R, Nishikawa Y, Carr BI. Cell growth inhibition by a novel vitamin K is associated with induction of protein tyrosine phosphorylation. J Biol Chem 1998;273:9906–11.
22 Sakon M, Monden M, Gotoh M, Kanai T, Umeshita K, Nakano Y, et al. Relationship between pathologic prognostic factors and abnormal levels of des-gamma-carboxy prothrombin and alpha-fetoprotein in hepatocellular carcinoma. Am J Surg 1992;163:251–6.[CrossRef][ISI][Medline]
23 Ito T, Tuneo T, Sue K, Urabe K, Shirakami K, Higashi T, et al. Positivities of PIVKA-II and alpha-fetoprotein in hepatocellular carcinoma derived from hepatitis C and hepatitis B. Rinsho To Kenkyu 1998;75:701–4 (in Japanese; author’s translation).
24 Suzuki H, Akabane K, Tanaka M, Tanikawa K, Okuta H, Saitoh A, et al. Kan Tan Sui 1996;33:1069–76.
25 Takezaki E, Satoh W, Murakami S, Kawgawa K, Ohmori H. Gann No Rinshoh 1992;38:747–52.
Received May 5, 2003; accepted September 9, 2003
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Fujikawa, H. Shiraha, N. Ueda, N. Takaoka, Y. Nakanishi, N. Matsuo, S. Tanaka, S.-i. Nishina, M. Suzuki, A. Takaki, et al. Des-{gamma}-carboxyl Prothrombin-promoted Vascular Endothelial Cell Proliferation and Migration J. Biol. Chem., March 23, 2007; 282(12): 8741 - 8748. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Suzuki, H. Shiraha, T. Fujikawa, N. Takaoka, N. Ueda, Y. Nakanishi, K. Koike, A. Takaki, and Y. Shiratori Des-{gamma}-carboxy Prothrombin Is a Potential Autologous Growth Factor for Hepatocellular Carcinoma J. Biol. Chem., February 25, 2005; 280(8): 6409 - 6415. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||