Akt Phosphorylation Is a Risk Factor for Early Disease Recurrence and Poor Prognosis in Hepatocellular Carcinoma

Transcription

1 307 Akt Phosphorylation Is a Risk Factor for Early Disease Recurrence and Poor Prognosis in Hepatocellular Carcinoma Kazuaki Nakanishi, M.D., Ph.D. 1,2 Michiie Sakamoto, M.D., Ph.D. 1,3 Susumu Yamasaki, M.D., Ph.D. 4 Satoru Todo, M.D., Ph.D. 2 Setsuo Hirohashi, M.D., Ph.D. 1 1 Division of Pathology, National Cancer Center Research Institute, Tokyo, Japan. 2 Department of General Surgery, Hokkaido University Graduate School of Medicine, Sapporo, Japan. 3 Department of Pathology, Keio University School of Medicine, Tokyo, Japan. 4 Division of Hepatobiliary and Pancreatic Surgery, National Cancer Center Hospital, Tokyo, Japan. BACKGROUND. Patients with hepatocellular carcinoma (HCC) who showed early massive disease recurrence due to hematogenous intrahepatic metastasis after curative resection had a poor prognosis. The authors previously reported that Akt phosphorylation was correlated with hematogenous intrahepatic metastasis, using HCC cell lines. METHODS. The authors analyzed clinicopathologic features and the status of selected biologic markers, including phosphorylated Akt, to identify risk factors for early disease recurrence and poor prognosis in HCC. In the current series, 49 postoperative patients developed intrahepatic disease recurrence within 6 months () and 86 patients remained disease recurrence free 3 years after resection (). was further divided into 2 subgroups: 19 patients who died of disease recurrence within a year after resection (A) and 27 patients who survived 1 year (B). RESULTS. Using univariate analysis, the risk factors for early disease recurrence were tumor size, macroscopic classification, tumor differentiation, microscopic capsule infiltration, microscopic portal vein (MPV) invasion, microscopic intrahepatic metastasis (MIM), and positive immunostaining for phosphorylated Akt, Ki-67, and p53 (P 0.05). The risk factors for poor prognosis were the number of intrahepatic metastases, tumor differentiation, and positive immunostaining for phosphorylated Akt and Ki-67 (P 0.05). Multivariate analysis revealed that the risk factors for early disease recurrence were MPV invasion, MIM, and positive immunostaining for phosphorylated Akt, and that the risk factors for poor prognosis were positive immunostaining for phosphorylated Akt and Ki-67 (P 0.05). CONCLUSIONS. The current clinical study showed the critical involvement of Akt phosphorylation in the aggressiveness of HCC. The potential benefits of surgery should be assessed carefully in patients with any of these risk factors. Cancer 2005; 103: American Cancer Society. Supported by a Grant-in-Aid for the Second Term Comprehensive 10-Year Strategy for Cancer Control from the Ministry of Health, Labor, and Welfare of Japan. Dr. Nakanishi is the recipient of a Research Resident Fellowship from the Foundation for Promotion of Cancer Research in Japan. Address for reprints: Setsuo Hirohashi, M.D., Ph.D., National Cancer Center Research Institute, Tsukiji, Chuo-ku, Tokyo , Japan; Fax: (011) ; shirohas@ncc.go.jp Received April 27, 2004; revision received September 21, 2004; accepted September 28, KEYWORDS: hepatocellular carcinoma, early disease recurrence, poor prognosis, immunostaining, phosphorylated Akt. Hepatectomy for hepatocellular carcinoma (HCC) has become a safe surgical procedure with a very low operative mortality rate as a result of progress in evaluation of hepatic function reserve, surgical technique, and perioperative management. However, prognosis after hepatectomy is still unsatisfactory because of a high incidence of disease recurrence. 1,2 Pathologic and genetic studies have identified two types of recurrence of HCC, i.e., intrahepatic metastasis of the primary HCC and multicentric development of new tumors, although the two types cannot always be distinguished. It is generally believed that disease 2004 American Cancer Society DOI /cncr Published online 8 December 2004 in Wiley InterScience (

2 308 CANCER January 15, 2005 / Volume 103 / Number 2 recurrence due to intrahepatic metastasis is likely to appear soon after surgical resection and that its risk decreases with prolongation of the postoperative period. Conversely, disease recurrence due to multicentric liver carcinogenesis occurs throughout the observation period at a constant rate that varies with the severity of liver disease and the type of hepatitis virus infection. 3 Recent studies also have shown that early disease recurrence is associated with a worse prognosis compared with late disease recurrence. 4 8 A considerable number of patients develop early postoperative massive intrahepatic disease recurrence with poor prognosis, even after curative resection of relatively small HCC tumors. The generation of intrahepatic metastases requires spreading of tumor cells from the primary HCC tumor into the circulation and the survival of the circulating cells without cell-substratum interaction, which is called anchorage-independent growth. Our previous report showed that anchorage-independent growth correlates with phosphorylated (activated) Akt (p-akt) and plays a critical role in hematogenous intrahepatic metastasis in an orthotopic implantation model of HCC. 9 Therefore, it is possible that p-akt status is correlated with early disease recurrence and poor prognosis. In the current study, we retrospectively analyzed clinicopathologic features and the status of p-akt and other biologic markers in patients with clinical HCC to identify the risk factors for early disease recurrence and poor prognosis. MATERIALS AND METHODS Patients and Follow-Up From January 1990 to December 2000, 394 patients underwent curative resection of solitary HCC at the National Cancer Center Hospital (Tokyo, Japan). Patients who had intrahepatic metastasis within the whole liver and/or extrahepatic metastasis were excluded from the study. Of these 394 patients, 49 developed intrahepatic disease recurrence within 6 months after resection (). Of these 49 patients, 19 died due to disease recurrence within a year after resection (A) and 27 survived 1 year ( 1B). Eighty-six patients remained disease recurrence free 3 years after surgical resection (). These 135 patients (1 [n 49] and [n 86]) were included in the current study. Disease recurrence after surgical resection was diagnosed by serial measurement of tumor markers, ultrasonography, and computed tomography scans every 3 6 months. Clinical Parameters and Tumor Characteristics We compared the backgrounds of the patients, including age, gender, hepatitis virus infection status, and history of preoperative treatment. We also compared the macroscopic features of the resected specimens with regard to tumor size and macroscopic classification using criteria described previously. 10 Evaluation of microscopic features included portal vein invasion, tumor capsule formation, capsule infiltration, intrahepatic metastasis, and the histology of the noncancerous liver tissue specimens. We also evaluated the histologic grade of the tumor specimens using the World Health Organization classification. Immunohistochemistry of Biologic Markers Immunohistochemistry (IHC) was performed to evaluate three biologic markers: Ki-67, p53, and p-akt. The antibodies used were MIB-1 anti Ki-67 antibody (1: 200 dilution), DO-7 anti-p53 antibody (1:50 dilution), and phospho-akt (Ser 473, IHC-specific) anti p-akt antibody (1: 20 dilution). Both the MIB-1 anti Ki-67 antibody and the DO-7 anti-p53 antibody were purchased from Dako (Glostrup, Denmark) and phospho- Akt (Ser 473, IHC-specific) anti p-akt antibody was purchased from Cell Signaling Technology (Beverly, MA). Sections (5 m thick) of formalin-fixed, paraffinembedded tissue were deparaffinized with xylene, treated with 0.3% hydrogen peroxide in methanol, immersed in 10 mm citrate buffer (ph 6.0), heated at 121 C in an autoclave for 10 minutes, and allowed to cool at room temperature for 30 minutes. The sections were preincubated in 2% normal swine serum in phosphate-buffered saline (PBS) and then incubated with each antibody at 4 C overnight, followed by washing 3 times with PBS. The sections with MIB-1 and DO-7 were incubated for 30 minutes with biotinylated secondary antibody at room temperature. For detection, a Vectastain ABC kit (Vector Laboratories, Burlingame, CA) was used. A PolyMICA detection kit (The Binding Site Ltd., Birmingham, UK) 11 was used according to the manufacturer s instructions for detection of the anti p-akt antibody. Staining of Ki-67, p53, and p-akt was estimated by counting the percentage of positive nuclei per 1000 tumor cells in the region of the tumor specimen with the greatest density of staining. The percentage of cells expressing nuclear staining of Ki-67, p53, and p-akt was categorized as follows: cells with positive staining in 50% (Ki-67), 10% (p53), and 10%(p-Akt) of tumor cells were graded as negative and those with 50%, 10%, and 10%, respectively, were graded as positive.

3 Risk Factors for Early HCC Recurrence/Nakanishi et al. 309 FIGURE 1. Immunostaining of phosphorylated Akt. (A) Inhibition of Akt activation by LY Immunoblot analysis of total (Akt) and activated (p-akt) Akt protein in Li7 cells after incubation with various concentrations of LY for 12 hours. (B) Formalin-fixed, paraffin-embedded Li7 cell pellets treated with (right panel) and without (left panel) LY were immunostained with phospho-akt antibody. Cells treated without LY were stained more strongly in the nucleus compared with those treated with LY (C) Positive staining of poorly differentiated HCC (left panel) and negative staining of moderately differentiated HCC (right panel). Original magnification 200 (C); 400 (B). Statistical Analysis For univariate analysis, the chi-square test and/or the Fisher exact test was used to compare categorized variables and the t test was used to compare continuous variables. Continuous variables were expressed as the mean the standard deviation. Logistic regression analysis was used for multivariate analysis. All statistical analyses were performed with Stat View (Version 5.0) software (Abacus Concepts, Berkeley, CA). P 0.05 was considered to be statistically significant. RESULTS Immunostaining of Phosphorylated Akt As previously reported, 9 Western blotting analysis showed that the Akt protein was constitutively phosphorylated in the human HCC cell line, Li7 (Fig. 1A). Moreover, a specific inhibitor of phosphatidylinositol 3-kinase (PI3K), LY294002, dose dependently suppressed phosphorylation of Akt. Phosphorylation of Akt in Li7 cells was suppressed completely by 50 M LY294002, and the effect was sustained for 24 hours. IHC of p-akt was evaluated with formalin-fixed, paraffin-embedded Li7 cell pellets that had been pretreated with or without 20 M LY In parallel with phosphorylation status by Western blotting analysis, Li7 cells treated without LY stained more strongly in the nucleus than did those treated with LY (Fig. 1B). Thus, the nuclear staining was specific to the phosphorylation status of Akt, and could be evaluated as positive in surgical tissue specimens (Fig. 1C). Staining of p-akt was observed in cancer cells and in some lymphocytes of surgical tissue specimens. In positive cases, p-akt staining was distributed diffusely, although the staining was stronger in the periphery than in the center of the lesions. Risk Factors for Early Disease Recurrence Table 1 shows the univariate analysis of clinical parameters comparing the early () and no () disease recurrence groups. None of the clinical parameters were significantly different. Tables 2 and 3 show the univariate analyses of macroscopic and microscopic characteristics, respectively. Two macroscopic and four microscopic factors were associated with early disease recurrence: tumor size, macroscopic classification, tumor differentiation, microscopic capsule infiltration, microscopic portal vein

4 310 CANCER January 15, 2005 / Volume 103 / Number 2 TABLE 1 Clinical Parameters Associated with Early Disease Recurrence and Poor Prognosis a Clinical parameters A B (n 86) Gender Male 38 (77.6) 15 (78.9) 21 (77.8) 65 (75.6) Female 11 (22.4) 4 (21.1) 6 (22.2) 21 (24.4) Age (yrs) Virus infection HBsAg( ) HCVAb( ) 6 (12.2) 4 (21.1) 2 (7.4) 17 (19.8) HBsAg( ) HCVAb( ) 32 (65.4) 12 (63.1) 18 (66.7) 43 (50.0) HBsAg( ) HCVAb( ) 3 (6.1) 0 3 (11.1) 4 (4.7) HBsAg( ) HCVAb( ) 6 (12.2) 2 (10.5) 3 (11.1) 15 (17.4) Unknown 2 (4.1) 1 (5.3) 1 (3.7) 7 (8.1) History of preoperative treatment Presence 33 (67.3) 14 (73.7) 17 (63.0) 49 (57.0) Absence 16 (32.7) 5 (26.3) 10 (37.0) 37 (43.0) Disease recurrence-free interval (days) HBsAg ( ): hepatitis B surface antigen positive; HBsAg ( ): hepatitis B surface antigen negative; HCVAb ( ): hepatitis C virus antibody positive; HCVAb ( ): hepatitis C virus antibody negative. a Continuous data are expressed as mean standard deviation. TABLE 2 Macroscopic Characteristics Associated with Early Disease Recurrence and Poor Prognosis a Macroscopic characteristics 1A 1B (n 86) Tumor size (cm) b Macroscopic classification b Early (12.8) Type 1 10 (20.4) 1 (5.3) 8 (29.6) 31 (36.0) Type 2 25 (51.0) 13 (68.4) 11 (40.7) 27 (31.4) Type 3 14 (28.6) 5 (26.3) 8 (29.6) 17 (19.8) No. of intrahepatic metastases c 3 37 (75.5) 18 (94.7) 18 (66.7) 0 1 or 2 12 (24.5) 1 (5.3) 9 (33.3) (100) a Continuous data are expressed as mean standard deviation. b Significant difference (P 0.05) between s 1 and 2. c Significant difference (P 0.05) between s 1A and 1B. (MPV) invasion, and microscopic intrahepatic metastasis (MIM; P 0.05). All 3 biologic markers studied by IHC (i.e., Ki-67, p53, and p-akt) were significantly different (P 0.05; Table 4). Of the clinical parameters and tumor characteristics evaluated by univariate analysis, MPV invasion and MIM were associated significantly with early disease recurrence as demonstrated by multivariate analysis. Positive immunostaining for p-akt was an independent risk factor for early disease recurrence as revealed by multivariate analysis of all three biologic markers (Table 5). TABLE 3 Microscopic Characteristics Associated with Early Disease Recurrence and Poor Prognosis Microscopic characteristics 1A 1B (n 86) Tumor differentiation a,b Well (27.9) Moderately 24 (49.0) 5 (26.3) 17 (63.0) 43 (50.0) Poorly 23 (46.9) 12 (63.1) 10 (37.0) 16 (18.6) Unknown 2 (4.1) 2 (10.5) 0 3 (3.5) Tumor capsule formation Negative 3 (6.1) 1 (5.3) 1 (3.7) 16 (18.6) Positive 46 (93.9) 18 (94.7) 26 (96.3) 70 (81.4) Capsule infiltration a Negative 16 (34.8) 5 (27.8) 10 (38.5) 40 (57.1) Positive 30 (65.2) 13 (72.2) 16 (61.5) 30 (42.9) Portal vein invasion a Negative 9 (18.4) 3 (15.8) 6 (22.2) 65 (75.6) Positive 40 (81.6) 16 (84.2) 21 (77.8) 21 (24.4) Intrahepatic metastasis a Negative 12 (24.5) 3 (15.8) 8 (29.6) 81 (94.2) Positive 37 (75.5) 16 (84.2) 19 (70.4) 5 (5.8) Noncancerous liver tissue specimen No cirrhosis 36 (73.5) 13 (72.2) 19 (70.4) 67 (77.9) Nomal 4 (8.2) 2 (11.1) 2 (7.4) 7 (8.1) CH 32 (65.3) 11 (61.1) 14 (63.0) 53 (69.8) Cirrhosis 13 (26.5) 5 (27.8) 8 (29.6) 19 (22.1) CH: chronic hepatitis; PC: precirrhosis. a Significant difference (P 0.05) between s 1 and 2. b Significant difference (P 0.05) between s 1A and 1B.

5 Risk Factors for Early HCC Recurrence/Nakanishi et al. 311 TABLE 4 Biologic Markers Associated with Early Disease Recurrence and Poor Prognosis Biologic markers A B (n 81) p-akt a,b Negative 36 (73.5) 10 (52.6) 23 (85.2) 79 (97.5) Positive 13 (26.5) 9 (47.4) 4 (14.8) 2 (2.5) Ki-67 a,b Low 34 (69.4) 10 (52.6) 23 (85.2) 72 (88.9) High 15 (30.6) 9 (47.4) 4 (14.8) 9 (11.1) p53 a Negative 33 (67.3) 12 (63.2) 20 (74.1) 73 (90.1) Positive 16 (32.7) 7 (36.8) 7 (25.9) 8 (9.9) p-akt: phosphorylated Akt. a Significant difference (P 0.05) between s 1 and 2. b Significant difference (P 0.05) between s 1A and 1B. TABLE 5 Multivariate Analysis of the Risk Factors Associated with Early Disease Recurrence Variables Risk ratio 95% CI P value Macroscopic and microscopic characteristics Portal vein invasion (positive) Intrahepatic metastasis (positive) Biologic markers p-akt (positive) CI: confidence interval; p-akt: phosphorylated Akt. Risk Factors for Poor Prognosis in Patients with Early Disease Recurrence To investigate the clinical impact of early disease recurrence, we subdivided into A (which included patients who died within 1 year) and B (which included patients who survived 1 year). In general, prognosis after disease recurrence is believed to depend on the treatment received, the effectiveness of the treatment, and the length of the disease recurrence-free interval after surgical resection. In the current study, the use of transarterial chemoembolization after disease recurrence (data not shown) and the disease recurrence-free interval after surgical resection (Table 1) were similar between s 1A and 1B. The number of intrahepatic metastases was associated significantly with a poor prognosis (Table 2). Moreover, in the univariate analysis, one pathologic characteristic and two biologic markers showed significant differences between s 1A 1B: tumor differentiation and positive immunostaining for Ki-67 and p-akt (P 0.05; Tables 3 and 4). Clinical parameters were not significantly different TABLE 6 Multivariate Analysis of the Risk Factors Associated with Poor Prognosis in Patients with Early Disease Recurrence Variables Risk ratio 95% CI P value p-akt (positive) Ki-67 (positive) CI: confidence interval; p-akt: phosphorylated Akt. (Table 1). When the multivariate analysis was conducted, including the significant variables obtained on univariate analysis, positive immunostaining for Ki-67 and p-akt was an independent risk factor associated with poor prognosis (Table 6). DISCUSSION The disease recurrence rate after curative resection for HCC is still high. Surgery might be of little benefit in patients with early massive disease recurrence and very poor prognosis. Therefore, it is necessary to identify this high-risk group of patients with early disease recurrence and poor prognosis. In the current study, p-akt was recognized as a risk factor for early disease recurrence and poor prognosis. The serine/threonine kinase Akt was identified originally as an oncogene in mouse thymoma. 12 Akt is activated fully by phosphorylation after activation of PI3K. 13 p-akt regulates multiple processes, such as apoptosis, cell proliferation, and glucose utilization. 14 Overexpression of wild-type Akt may cause oncogenesis, as Akt has been found to be amplified with high frequency in gastric, pancreatic, breast, and ovarian tumors Moreover, we previously showed that p- Akt plays a critical role in hematogenous intrahepatic metastasis in an orthotopic implantation model of HCC. We also demonstrated a significant correlation between Akt activation and early disease recurrence, in the form of intrahepatic metastasis, in clinical patients. Currently, the molecular mechanism of early disease recurrence remains unclear. Intrahepatic metastasis is a more likely mechanism for the generation of early disease recurrence than multicentric development of new tumors. The first step of intrahepatic metastasis is dissemination of cancer cells into the circulation. Indeed, dissemination of cancer cells from primary HCC into the portal vein during surgical resection has been observed. 20 Nevertheless, disseminated cells do not always survive in the circulation and form metastatic sites, although anchorage-independent growth and rapid growth of cancer cells after surgical resection may be important factors for early

6 312 CANCER January 15, 2005 / Volume 103 / Number 2 disease recurrence. The current findings suggest that Akt phosphorylation is involved in such malignant behavior. Previous reports have suggested risk factors for early disease recurrence, such as tumor volume, histopathologic dedifferentiation, microscopic invasiveness, high cellular proliferation, and p53 mutation. A high serum concentration of -fetoprotein, which reflects tumor volume, was reported to be a predictor of early disease recurrence, and a previous report showed that p53 mutation was associated with the progression of HCC as a late event in hepatocarcinogenesis. 27 These observations are compatible with our current results. In conclusion, our clinical study has demonstrated the critical involvement of Akt phosphorylation in the aggressiveness of HCC and has identified Akt phosphorylation as a significant risk factor for early disease recurrence and poor prognosis. In patients with this risk factor, surgery may be of little benefit. Therefore, careful assessment of the indications for surgery is necessary in these patients. REFERENCES 1. Nagao T, Inoue S, Goto S, et al. Hepatic resection for hepatocellular carcinoma. Clinical features and long-term prognosis. Ann Surg. 1987;205: Nagao T, Inoue S, Yoshimi F, et al. Postoperative recurrence of hepatocellular carcinoma. Ann Surg. 1990;211: Sakon M, Umeshita K, Nagano H, et al. Clinical significance of hepatic resection in hepatocellular carcinoma: analysis by disease-free survival curves. Arch Surg. 2000;135: Ikeda Y, Kajiyama K, Adachi E, Yamagata M, Shimada M, Yanaga K. Early recurrence after surgery of hepatocellular carcinoma. Hepatogastroenterology. 1995;42: Matsumata T, Kanematsu T, Takenaka K, Yoshida Y, Nishizaki T, Sugimachi K. Patterns of intrahepatic recurrence after curative resection of hepatocellular carcinoma. Hepatology. 1989;9: Poon RT, Fan ST, Ng IO, Lo CM, Liu CL, Wong J. Different risk factors and prognosis for early and late intrahepatic recurrence after resection of hepatocellular carcinoma. Cancer. 2000;89: Shirabe K, Kanematsu T, Matsumata T, Adachi E, Akazawa K, Sugimachi K. Factors linked to early recurrence of small hepatocellular carcinoma after hepatectomy: univariate and multivariate analyses. Hepatology. 1991;14: Yamanaka J, Yamanaka N, Nakasho K, et al. Clinicopathologic analysis of stage II-III hepatocellular carcinoma showing early massive recurrence after liver resection. J Gastroenterol Hepatol. 2000;15: Nakanishi K, Sakamoto M, Yasuda J, et al. Critical involvement of the phosphatidylinositol 3-kinase/Akt pathway in anchorage-independent growth and hematogeneous intrahepatic metastasis of liver cancer. Cancer Res. 2002;62: Kanai T, Hirohashi S, Upton MP, et al. Pathology of small hepatocellular carcinoma. A proposal for a new gross classification. Cancer. 1987;60: Mangham DC, Isaacson PG. A novel immunohistochemical detection system using mirror image complementary antibodies (MICA). Histopathology. 1999;35: Bellacosa A, Testa JR, Staal SP, Tsichlis PN. A retroviral oncogene, akt, encoding a serine-threonine kinase containing an SH2-like region. Science. 1991;254: Kandel ES, Hay N. The regulation and activities of the multifunctional serine/threonine kinase Akt/PKB. Exp Cell Res. 1999;253: Datta SR, Brunet A, Greenberg ME. Cellular survival: a play in three Akts. Genes Dev. 1999;13: Aoki M, Batista O, Bellacosa A, Tsichlis P, Vogt PK. The akt kinase: molecular determinants of oncogenicity. Proc Natl Acad Sci USA. 1998;95: Staal SP. Molecular cloning of the akt oncogene and its human homologues AKT1 and AKT2: amplification of AKT1 in a primary human gastric adenocarcinoma. Proc Natl Acad Sci USA. 1987;84: Bellacosa A, de Feo D, Godwin AK, et al. Molecular alterations of the AKT2 oncogene in ovarian and breast carcinomas. Int J Cancer. 1995;64: Cheng JQ, Ruggeri B, Klein WM, et al. Amplification of AKT2 in human pancreatic cells and inhibition of AKT2 expression and tumorigenicity by antisense RNA. Proc Natl Acad Sci USA. 1996;93: Ruggeri BA, Huang L, Wood M, Cheng JQ, Testa JR. Amplification and overexpression of the AKT2 oncogene in a subset of human pancreatic ductal adenocarcinomas. Mol Carcinog. 1998;21: Yamanaka N, Okamoto E, Fujihara S, et al. Do the tumor cells of hepatocellular carcinomas dislodge into the portal venous stream during hepatic resection? Cancer. 1992;70: Yamanaka N, Okamoto E, Toyosaka A, et al. Prognostic factors after hepatectomy for hepatocellular carcinomas. A univariate and multivariate analysis. Cancer. 1990;65: Hanazaki K, Wakabayashi M, Sodeyama H, Kajikawa S, Amano J. Hepatic function immediately after hepatectomy as a significant risk factor for early recurrence in hepatocellular carcinoma. Hepatogastroenterology. 1999;46: Poon RT, Fan ST, Lo CM, Liu CL, Wong J. Intrahepatic recurrence after curative resection of hepatocellular carcinoma: long-term results of treatment and prognostic factors. Ann Surg. 1999;229: Shimada M, Takenaka K, Gion T, et al. Prognosis of recurrent hepatocellular carcinoma: a 10-year surgical experience in Japan. Gastroenterology. 1996;111: Takata M, Yamanaka N, Tanaka T, et al. What patients can survive disease free after complete resection for hepatocellular carcinoma? A multivariate analysis. Jpn J Clin Oncol. 2000;30: Hayashi H, Sugio K, Matsumata T, Adachi E, Takenaka K, Sugimachi K. The clinical significance of p53 gene mutation in hepatocellular carcinomas from Japan. Hepatology. 1995; 22: Oda T, Tsuda H, Sakamoto M, Hirohashi S. Different mutations of the p53 gene in nodule-in-nodule hepatocellular carcinoma as a evidence [sic] for multistage progression. Cancer Lett. 1994;83: