Significance of Hepatectomy for AJCC/UICC T3 Hepatocellular Carcinoma

Transcription

1 Significance of Hepatectomy for AJCC/UICC T3 Hepatocellular Carcinoma TOSHIYA OCHIAI 1,2, HIROMICHI ISHII 1, YUSUKE YAMAMOTO 2, RYO MORIMURA 2, HISASHI IKOMA 2 and EIGO OTSUJI 2 1 Department of Surgery, North Medical Center, Kyoto Prefectural University of Medicine, Kyoto, Japan; 2 Division of Digestive Surgery, Department of Surgery, Kyoto Prefectural University of Medicine, Kyoto, Japan Abstract. Background: It is difficult to treat the American Joint Committee on Cancer (AJCC)/International Union against Cancer (UICC) T3 hepatocellular carcinoma (HCC), curatively. Patients and Methods: We compared the clinicopathology of T3 group (n=44: T3a 25, T3b 19) with that of the T1 (n=257) or T2 group (n=120) and evaluated favorable conditions of hepatectomy for T3 HCC patients. Results: The T3 group had significantly higher hepatitis B surface antigen (HBsAg)-positive rates and better liver function. Infiltrative large tumors located beyond one subsegment with intrahepatic metastasis were significantly more common. Significantly, more non-curative large hepatectomies with transarterial embolization were performed. There was no significant difference between T3 and 2 groups in 5-year disease-free survival (DFS) and survival (S). Tumor size more than 55 mm and serum albumin less than 3.5 g/dl were risk factors of hepatectomy for T3 HCC patients by multivariate analysis. Conclusion: Surgeons should resect AJCC/UICC T3 HCC lesions if the patient is able to tolerate surgery. The American Joint Committee on Cancer (AJCC)/ International Union against Cancer (UICC) Tumor-Node- Metastasis (TNM) classification (1, 2) is one of the most popular staging systems for hepatocellular carcinoma (HCC) worldwide. This system was developed based on the results of a survival analysis of patients who underwent hepatectomy (3); therefore, most liver surgeons select treatments considering this classification. In particular, the T factor is important for surgeons as lymph node dissection and/or metastatectomy of distant organs are not performed in most patients with HCC and the stage of operable HCC is Correspondence to: Toshiya Ochiai, MD, Department of Surgery, North Medical Center, Kyoto Prefectural University of Medicine, 481 Otokoyama, Yosano-cho, Yosa-gun, Kyoto, , Japan. Tel/Fax: , tochiai@koto.kpu-m.ac.jp Key Words: Hepatocellular carcinoma, UICC, stage, surgery. determined based only on the T factor. While the AJCC/UICC TNM classification itself does not suggest feasible treatments, the modified Barcelona Clinic Liver Cancer (BCLC) guidelines (4) indicate which treatments should be selected based on the stage. These guidelines consider the tumor stage, as well as the patient s liver function and physical status. The definition of T3 HCC in the AJCC/UICC TNM classification, 7 th edition (2), consists of the T3a (multiple tumors with a maximum diameter above 5 cm) and T3b (single or multiple tumors of any size involving a major branch of the portal or hepatic vein) categories. According to the modified BCLC guidelines, most T3 HCC cases are classified as being in the intermediate or advanced stage, which carry no indications for surgical resection and are associated with a predictive 3- year survival of % (5). However, in Japan, most cases of AJCC/UICC T3 HCC are resected because there currently exist no effective nonsurgical therapies for lesions of these stages and the prognosis of surgical versus non-surgical cases is relatively good. The results of hepatectomy for AJCC/UICC T3 in Japan are not as poor as those indicated in the modified BCLC guidelines. For these reasons, AJCC/UICC T3 HCC patients are considered to have operative indications in Japan. Currently, there are no special rules regarding treatment of macroscopic tumor thrombi, which meet the criteria for AJCC/UICC T3b, in the Japanese guidelines (6). In this report, we analyzed the clinicopathological features of AJCC/UICC T3 HCC patients and sought to clarify the significance of hepatectomy for AJCC/UICC T3 HCC. Patients and Methods Between April 1980 and August 2011, 452 patients with HCC underwent primary hepatic resection at the Division of Digestive Surgery, Department of Surgery, Kyoto Prefectural University of Medicine. Among these subjects, clinicopathological data were available for 421 HCC patients who underwent curative resection for HCC. We compared the clinicopathological findings in the T3 group (n=44: T3a 25, T3b 19) with those observed in the T1 (n=257) and T /2015 $

2 (n=120) groups and identified favorable conditions for hepatectomy in T3 HCC patients. In the T3 group, one patient had a macroscopic tumor thrombus at the common bile duct, which is not defined in the AJCC/UICC TMN classification. We classified this patient into the T3b group. We also studied risk factors for T3-HCC using univariate and multivariate analyses. Curative resection was defined as the complete removal of macroscopic tumors not exposed on the cut surface. Basically, resection of primary HCC at our institution is carried out via anatomic hepatectomy (7). In patients with concomitant liver cirrhosis or an impaired liver function, however, limited (nonanatomical) resection is performed in some cases as a second-choice procedure. The indications for hepatic resection and the operative method were selected based on the degree of tumor progression and Makuuchi s criteria, i.e. the presence or absence of ascites, the serum total bilirubin level and the indocyanine green (ICG) retention rate at 15 minutes (ICG R15) (8). HCC patients with a serum total bilirubin level of >1.0 mg/dl or ICG R15 of >29% received limited (non-anatomical) resection. All resected liver specimens were cut at a thickness of approximately 5 mm and microscopic sections were viewed after staining with hematoxylin and eosin. The pathologic diagnosis and classification of the resected HCC tissues were determined according to the General Rules for the Clinical and Pathological Study of Primary Liver Cancer (9). All tumors were staged based on the definitions of TNM stages proposed by the International Hepato-Pancreato-Biliary Association and International Union against Cancer (2). Follow-up. Patients were followed-up using hepatic ultrasonography, computed tomography and assessments of the serum alpha-fetoprotein and protein induced by vitamin K absence II levels every three to six months. Disease-free survival (DFS) was defined as the interval between surgery and the date of diagnosis of the first recurrence or last follow-up. Survival (S) was also defined as the interval between surgery and the date of death due to HCC recurrence or the last follow-up. We treated the date of death due to liver dysfunction or variceal rupture the same as the date of the last follow-up. The median follow-up duration was 36 months (range=1-269 months). Statistics. Statistical analysis was carried out with the Mann- Whitney U-test for unpaired observations and the chi-square test for the frequencies of various attributes among the groups. The S and DFS rates were estimated according to the Kaplan-Meier method and assessed for statistical significance using the Wilcoxon and logrank tests. Risk factors for T3 HCC were estimated using a Cox proportional hazards analysis to identify factors associated with the postoperative prognosis. Differences with a p-value less than 0.05 were considered to be statistically significant. Results Clinicopathological features. Regarding host-related factors, the T3 group had significantly lower serum total bilirubin levels and rates of hepatitis B surface antigen (HBsAg) negativity, associated liver cirrhosis and ICG retention rates than the T1 and T2 groups. With respect to tumor-related factors, large infiltrative tumors located beyond one subsegment with intrahepatic metastasis were significantly more common in the T3 group than in the T1 and T2 groups. For treatment-related factors, significantly more non-curative lobectomy and hepatectomy procedures with preoperative transarterial embolization (TAE) were performed in the T3 group than in the T1 and T2 groups (Table I). Perioperative data. Regarding the perioperative data in cases of hepatic resection performed since 2000, the amount of blood loss, operative time, morbidity and length of postoperative hospital stay were significantly greater in the T3 group than in the T1 or T2 groups. There were no significant differences in mortality between the three groups (Table II). Prognosis. In terms of the prognosis, the cumulative 5-year S and DFS rates were 31.6/20.3%, 50.1/20.4% and 79.0/44.4% in the T3, T2 and T1 groups, respectively (Figure 1-a, b). While there were no significant differences between the T3 and T2 groups, the DFS and S rates in the T3 group were significantly worse than those observed in the T1 group. In the T2 and T3 groups, significant differences in the S rates were noted between the T3a and T3b subgroups (Figure 2-a, b); however, there were no significant differences between the T2 and T3a sub-groups. With respect to the recurrence pattern, the rates of postoperative early diffuse and extrahepatic recurrence were significantly higher in the T3 group than in the T1 group (Table III). Risk factors for a poor prognosis. A Cox proportional hazards analysis revealed the tumor size and a serum albumin level less than 3.5 g/dl to be prognostic risk factors for hepatectomy in the T3 HCC patients according to a multivariate analysis taking into consideration clinical factors that could be clarified either pre- or intraoperatively (Tables IV and V). As for macroscopic tumor thrombi, there were significant differences in survival between the patients with and without these features in the T3 group based on the Wilcoxon test (Figure 2-a). We analyzed the feasible size of the primary tumor for resection using a receiver operating characteristic (ROC) curve and the Youden index. An area under the ROC curve of was obtained taking into consideration the sensitivity and specificity of the primary tumor size. A tumor size of >55 mm was identified as the best cut-off value for predicting a poor prognosis. In this series, HCC patients with tumors greater than 55 mm in size were predicted to have a poor prognosis (specificity: 0.455, sensitivity: 0.818) (Figure 3). Discussion For surgeons, the stage of malignancy is one of the most important factors for determining whether to perform surgery. The efficacy of surgery is greatest in the early stages of the disease. Vauthey s simplified staging (10), which was 2922

3 Ochiai et al: Hepatectomy for AJCC/UICC T3 HCC Table I. Clinicopathologic features in each group. T3 T2 T1 p-value n=44 n=120 n=257 Host-related factors Age 59.0± ± ±9.3 vst1<0.05* Gender (male/female) 36/8 100/20 188/69 ns HBsAg (positive rate %) All <0.05* HCV (positive rate %) ns ICG15 % 14.9± ± ±12.6 All<0.05* T-Bil mg/dl 0.87± ± ±0.39 All<0.05* Alb g/dl 3.86± ± ±0.50 ns PLT( 10 4 ) 22.0± ± ±6.4 All<0.01* Cirrhosis (Cirrhosis/Fibrosis, Normal) 14/30 69/51 144/113 All<0.01* Child-Pugh (A/B/C) 44/0/0 117/3/0 250/7/0 ns Tumor-related factors Solitary/Multiple 9./35 50/70 254/3 All<0.05* Size mm 73.7± ± ±25.5 All<0.01* AFP 100 ng/dl (%) All<0.01* Capsule (present/absent) vpvvb (negative/positive) $ 28/16 100/20 208/49 vst1<0.01* im (negative/positive) $$ 11/33 47/73 256/1 All<0.05* Gross type (Massive, Infiltrative 12/32 69/51 254/3 vst1<0.01* Confluent nodular, Multinodular(multiple)/Simple nodular 20/24 25/95 2/250 All<0.01* Simple nodular with extragrowth Hepatic involvement (Hs/H1,H2,H3) 7/37 50/70 181/76 All<0.01* Treatment-related factors Preoperative TAE (present/absent) 25/19 67/53 82/168 vst1<0.01* Operation (small hepatectomy/large hepatectomy.) # 14/30 86/34 215/42 All<0.01* Blood transfusion (%) vst1<0.01* Surgical margin (>10mm/ 10mm) 18/26 43/77 109/143 ns Curability## (A1, A2/B) 2/42 15/ /29 vst1<0.01* # Small hepatectomy: Resection of less than one subsegment, one subsegment or one segment; large hepatectomy: resection of two or three segments ## Curability A1: Stage I and the absence of cancer invasion at the surgical margin (SM-). Curability A2: Stage II and SMCurability. B: No residual cancer; however, the criteria for A1 and A2 are not met. Curability C: Definite residual cancer. HBsAg: Hepatitis B surface antigen, HCV: hepatitis C virus, ICG: indocyanin green, PLT: platelet, Alb: albumin, AFF: alpha feto-protein, TAE: transarterial embolization. $ vpvvb: Ductal invasion (portal venous and/or hepatic venous and/or biliary involvement). $$ im: Intrahepatic metastasis. subsequently adopted as the TNM staging system for AJCC/UICC of HCC, is based on the stratification of the prognosis in each stage. Although parameters of the liver function are not included, this system covers all patients with HCC and was designed for use in operable cases. The indications for resection, including tumor stage and liver function, remain controversial. Therefore, surgeons consider the AJCC/UICC TMN staging system in addition to various parameters (8, 11, 12) of the liver function in order to determine whether to perform hepatectomy. A poor prognostic stage, i.e. stage IV, is not always optimal for surgery, even if the patient has enough liver function for hepatectomy. Performing curative resection in such cases is difficult and causes damage that is far worse than that associated with other non-surgical treatments based on the predicted postoperative prognosis. In cases of HCC, stage III is the most difficult stage for selecting the best treatment. According to the BCLC staging system, an AJCC/UICC TMN classification of T3 is not feasible for hepatectomy; such cases of HCC are classified as being in the intermediate or advanced stage and treatment with transcatheter arterial chemoembolization (TACE) or sorafenib is recommended in these patients. On the other hand, Torzilli et al. (13) maintained that patients with intermediate or advanced disease can tolerate hepatectomy with low mortality, acceptable morbidity and survival benefits if resection is performed under strict intraoperative ultrasonograhic guidance. In addition, Yang T. et al. (14) reported 1-, 3- and 5-year overall/disease-free survival rates of 69.9/48.2%, 41.2/30.3% and 30.5/24.0%, respectively, among 511 Chinese patients with BCLC advanced-stage HCC who underwent surgical resection. These results are almost the same as in our series. Recently, Zhong J.H. et al. (15) also reported that, for selected patients 2923

4 Table II. Perioperative data for each group since T1 T2 T3 p-value n=11 n=47 n=115 Operating time (min) 487.3± ± ±119.0 All<0.01* Blood loss (ml) 1859.± ± ±1314 vst1<0.05* Post operative hospital stay (day) 55.0± ± ±21.8 All<0.01* Mortality 0 (0.0%) 1 (0.8%) 0 (0.0%) ns Morbidity 6 (54.5%) 8 (17.0%) 19 (16.5%) All<0.01* Biliary tract complications 2 (18.2%) 2 (4.2%) 7 (6.1%) ns Liver dysfunction 2 (18.2%) 3 (6.4%) 2 (1.7%) vst1<0.01* Pleural effusion 1 (9.1%) 1 (2.1%) 3 (2.6%) ns Intraperitoneal abscess 0 (0.0%) 0 (0.0%) 1 (0.1%) NA Others 1 (9.1%) 2 (4.2%) 6 (5.3%) NA *Significant; NA: not analyzed. Table III. Recurrence patterns in each group. T3 T2 T1 p-value n=44 n=120 n=257 No recurrence 10 (22.7%) 35 (29.2%) 135 (52.5%) Recurrence 34 (77.3%) 85 (70.8%) 122 (47.5%) vst1<0.01* Solitary 6 (17.6%) 26 (30.6%) 53 (43.4%) vst1<0.01* Multiple 16 (47.1%) 40 (47.1%) 61 (50.0%) ns Diffuse 4 (11.8%) 9 (10.6%) 0 (0.0%) vst1<0.01* Other organ 8 (23.5%) 10 (11.8%) 8 (6.6%) vst1<0.01* * % (/whole recurrence). Average follow-up duration: 49.9 months. with intermediate and advanced-stage HCC, hepatectomy is associated with good survival, superior to that of transarterial chemoembolization. Moreover, various studies (16, 17) have reported good short- and long-term outcomes in patients with intermediate-advanced HCC treated with liver resection. Based on these reports, although the BCLC guidelines are thought to be one of the most reliable staging systems for HCC, revision of the BCLC recommendations is required. The size or number of tumors and presence of macroscopic tumor thrombi are prognostic risk factors for HCC and included in the AJCC/UICC T3 definition. Among these factors, vascular invasion has the strongest impact on survival (3). Shimada et al. (18) reported that even if curative hepatectomy cannot be achieved, hepatectomy may provide better survival in large HCC patients without macroscopic tumor thrombi than in those with macroscopic tumor thrombi. In the current series, there were statistically significant differences in S between the T3a and T3b groups. Postoperative early diffuse and extrahepatic recurrence patterns are related with macroscopic tumor thrombi. Resecting HCCs with macroscopic tumor thrombi is technically challenging with poor prognosis. There are various patterns of macroscopic tumor thrombus progression (19), i.e. extension within the ipsilateral portal vein, extension to or beyond the portal vein bifurcation, which can be treated with en bloc resection associated with portal vein reconstruction, as well as extension to or beyond the portal vein bifurcation, which can be treated with thrombectomy; no significant differences have been reported in the prognosis among these three groups. Moreover, Inoue et al. (20) showed that the peeling off technique achieved a similar prognosis to en bloc resection in 49 patients who underwent curative hepatectomy for HCC with macroscopic portal venous tumor thrombi. However, Shi J. et al. (21) maintained that the degree of extension of macroscopic tumor thrombi was able to better stratify and predict the prognosis in 441 patients with HCC with such thrombi. With regard to size and number of tumors, it is true that as the tumor grows, so do prognostic risk factors, i.e. vascular invasion, intrahepatic metastasis and the serum alphafetoprotein level. Tsoulfas et al. (22) reported that identifying risk factors that affect the outcomes of patients undergoing surgery for large HCC is critical as hepatectomy can obtain excellent outcomes in carefully selected patients. In the present series, there were no significant differences in the S 2924

5 Ochiai et al: Hepatectomy for AJCC/UICC T3 HCC Figure 1. Postoperative cumulative survival (a) and disease-free survival (b) curves for each group. While there were no significant differences between the T3 and T2 groups, the DFS and S rates in the T3 group were significantly worse than those in the T1 group. The five-year cumulative survival/disease-free survival rate of T1, T2 and T3 group was 79.0/44.4, 50.1/20.4 and 31.6/20.3%, respectively. Figure 2. In the T2 and T3 groups, a significant difference in the S rate was noted between the T3a and T3b groups; however, there were no significant differences between the T2 and T3a groups (a). There were no significant differences in disease-free survival between the three groups (b). between the T2 and T3a HCC groups as a result of patient selection. As for size, numerous factors and liver function most surgeons, including us, consider them in determining the operative indication correctly. Compared to size and numerous factors, the presence of macroscopic tumor thrombi has a far strong prognostic impact as described above. AJCC/UICC T3a and T3b lesions should be considered to involve different stages. The presence of macroscopic tumor thrombi in the main branch of the portal or hepatic vein and bile duct should be treated as a factor indicative of a more advanced stage. T3 HCC group may include many non-cirrhotic patients with HBs carrier. Due to good liver function and tumor local invasion, many patients of T3 HCC had received large hepatectomies. As the results, the perioperative data for T3 HCC were worse than those for the T1 and T2 stages in this study, but they were acceptable. Priority, however, is given to the postoperative prognosis of HCC patients. The prognosis of T3 HCC is also worse than that of T1 but far better than that obtained with non-surgical therapy and/or best supportive care. In the current report, we identified prognostic risk factors in patients with T3 HCC. When performing hepatectomy of T3 HCC, it is possible to achieve a better prognosis in patients with a serum protein level above 3.5 g/dl and a tumor size of less than 55 mm. 2925

6 Table IV. Risk factors for disease-free survival in the T3 group. Univariate analysis p-value Multivariate analysis p-value RR (95% lower RR (95% lower -upper limit) -upper limit) Host-related factors HBsAg (positive vs. negative) 0.67 ( ) 0.30 HCV (positive vs. negative) 1.71 ( ) 0.26 T-Bil>0.9 mg/dl 0.62 ( ) 0.34 Alb>3.5 g/dl 0.40 ( ) 0.04* 0.55 ( ) 0.005* Cirrhosis (no vs. yes) 1.02 ( ) 0.96 Age>60y.o 0.97 ( ) 0.92 Treatment-related factors Anatomic hepatectomy (no vs. yes) 0.95 ( ) 0.93 Blood transfusion (yes vs. no) 1.55 ( ) 0.32 Subseg. or limited resection (no vs. yes) 0.88 ( ) 0.81 Tumor-related factors Size mm 1.01 ( ) * 1.01 ( ) 0.001* AFP 100 ng/dl (no vs. yes) 1.33 ( ) 0.47 Gross type (Simple nodular, Simple nodular with extragrowth 1.68 ( ) 0.14 vs. Massive,Infiltrative, Confluent nodular, Multinodular (multiple)) Ductal invasion (absent vs. present) 1.01( ) 0.97 Hepatic involvement (Hs vs. H1-3) 1.98 ( ) 0.26 *Significant risk factors by logistic regression model. RR: Relative risk, HBsAg: hepatitis B surface antigen, HCV: hepatitis C virus, Alb: albumin, AFF: alpha feto-protein, H1: cancer limited to one segment, H2: cancer limited to two segments, H3: cancer limited to three. segments, Hs: cancer limited to a sub-segment. The limitations of this study include its retrospective design and patient selection bias. However, surgery should not be avoided as one of recommended treatments for T3 group due to its invasiveness. In conclusion, AJCC/UICC T3 HCC is operable. Revision of the BCLC treatment recommendations for intermediate or advanced stage is required. Surgeons should resect T3 HCC lesions if the patient is able to tolerate surgery. However, the presence of macroscopic tumor thrombi in a major branch of the vein or bile duct should be considered a risk factor for a poor prognosis. Conflicts of Interest There exist no conflicts of interest to disclose. References 1 AJCC: AJCC Cancer Staging Manual, 7th ed. New York Springer Sobin LH, Wittekind CH (eds): TNM Classification of malignant tumours, 7th edn In: Liver. New York, Wiley-Liss., pp , Minagawa M, Ikai I, Matsuyama Y, Yamaoka Y and Makuuchi M: Staging of hepatocellular carcinoma Assessment of the Japanese TNM and AJCC/UICC TNM Systems in Cohort of 13,772 patients in Japan Ann Surg 245: , Figure 3. Analysis of the size of the primary tumor for resection using a receiver operating characteristic (ROC) curve and the Youden index. An area under the ROC curve of was obtained taking into consideration the sensitivity and specificity of the primary tumor size. A tumor size of >55 mm (gray dot) was identified to be the best cut-off value for predicting a poor prognosis. In this series, the HCC patients with tumors greater 55 mm in size were predicted to have a poor prognosis (specificity: 0.455, sensitivity: 0.818). 2926

7 Ochiai et al: Hepatectomy for AJCC/UICC T3 HCC Table V. Risk factors for survival in the T3 group. Univariate analysis p-value Multivariate analysis p-value RR (95% lower RR (95% lower -upper limit) -upper limit) Host-related factors HBsAg (positive vs. negative) 0.49 ( ) 0.09 HCV (positive vs. negative) 3.54 ( ) 0.034* 0.86 ( ) 0.55 T-Bil>0.9 mg/dl 0.54 ( ) 0.25 Alb>3.5 g/dl 0.90 ( ) 0.84 Cirrhosis (no vs. yes) 1.04 ( ) 0.93 Age>60y.o 0.41 ( ) 0.028* 0.93 ( ) 0.65 Treatment-related factors Anatomic hepatectomy (no vs. yes) 1.17 ( ) 0.83 Blood transfusion (yes vs. no) 1.10 ( ) 0.86 Subseg. or limited resection (no vs. yes) 1.02 ( ) 0.97 Tumor-related factors Size mm 1.01 ( ) * 1.01 ( ) 0.016* AFP 100 ng/dl (no vs. yes) 1.52 ( ) 0.36 Gross type (Simple nodular, Simple nodular with extragrowth 1.34 ( ) 0.46 vs. Massive, Infiltrative, Confluent nodular, Multinodular(multiple)) Ductal invasion (absent vs. present) 0.84 ( ) 0.69 Hepatic involvement (Hs vs. H1-3) 1.02 ( ) 0.98 *Significant risk factors by logistic regression model. RR: Relative risk, HBsAg: hepatitis B surface antigen, HCV: hepatitis C virus, Alb: albumin, AFF: alpha feto-protein, H1: Cancer limited to one segment, H2: Cancer limited to two segments, H3: Cancer limited to three. segments, Hs: Cancer limited to a subsegment. 4 Llovet JM, Fuster J and Bruix J: The Barcelona Approach: Diagnosis, Staging, and Treatment of hepatocellular carcinoma. Liver Transpl 10 Suppl: S115-S120, Pons F, Varela M and Llovet JM: Staging system in hepatocellular carcinoma HPB 7: 35-37, Makuuchi M, Kokudo N, Arii S, Futagawa S, Kaneko S, Kawasaki S, Matsuyama Y, Okazaki M, Okita K, Omata M, Saida Y, Takayama T and Yamaoka Y: Development of evidence-based clinical guidelines for diagnosis and treatment of hepatocellular carcinoma in Japan. Hepatol Res 238: 37-51, Makuuchi M, Hasegawa H and Yamazaki S: Ultrasonically guided subsegmentectomy. Sure Gynecol Obstet 161: , Makuuchi M, Kosuge T, Takayama T, Yamazaki S, Kakazu T, Miyagawa S and Kawasaki S: Surgery for small liver cancers. Semin Surg Oncol 9: , Liver Cancer study Group of Japan: General rules for the clinical and pathological study of primary liver cancer, 2nd English edn. In: Tokyo, Kanehara., pp , Vauthey JN, Lauwers GY, Esnaola NF, Do KA, Belghiti J, Mirza N, Curley SA, Ellis LM, Regimbeau JM, Rashid A, Cleary KR and Nagorney DM: Simplified staging for hepatocellular carcinoma. J Clin Oncol 20: , Pugh RNH, Murray-Lyon IM, Dawson JL, Pietroni MC and Williams R: Transection of the oesophagus for bleeding oesophaeal varices. Br J Surg 60: , Kudo M, Chung H and Osaki Y: Prognostic staging system for hepatocellular carcinoma (CLIP score): its value and limitations, and a proposal for a new staging system, the Japan Integrated Staging Score (JIS score). J Gastroenterology 38: , Torzilli G, Donadon M, Marconi M, Palmisano A, Del Fabbro D, Spinelli A, Botea F and Montorsi M: Hepatectomy for stage B and stage C hepatocellular carcinoma in Barcelona Clinic Liver Cancer Classification: Results of prospective analysis. Arch Surg 143: , Yang T, Lin C, Zhai J, Shi S, Zhu M, Zhu N, Lu JH, Yang GS and Wu MC: Surgical resection for advanced hepatocellular carcinoma according to Barcelona Clinic Liver Cancer (BCLC) staging. J Cancer Res Clin Oncol 138: , Zhong JH, Xiang BD, Gong WF, Ke Y, Mo QG, Ma L, Liu X and Li LQ: Comparison of long-term survival of patients with BCLC stage B hepatocellular carcinoma after liver resection or transarterial chemoembolization. PLoS One 8: e68193, Guglielmi A, Ruzzenente A, Conci S, Valdegamberi A, Vitali M, Bertuzzo F, De Angelis M, Mantovani G and Iacono C: Hepatocellular carcinoma:surgical perspectives beyond the Barcelona clinic liver cancer recommendations. World J Gastroentrrtol 20: , Zhong JH, Ke Y, Gong WF, Xiang BD, Ma L, Ye XP, Peng T, Xie GS and Li LQ: Hepatic resection associated with good survival for selected patients with intermediate and advancedstage hepatocellular carcinoma. Ann Surg 260: , Shimada K, Sakamoto Y, Esaki M and Kosuge T: Role of a hepatectomy for the treatment of large hepatocellular carcinomas measuring 10 cm or larger in diameter. Langenbeck s Arch Surg 393: , Chok KS, Cheung TT, Chan SC, Poon RT, Fan ST and Lo CM: Surgical outcomes in hepatocellular carcinoma patients with portal vein tumor thrombosis. World J Surg 38: ,

8 20 Inoue Y, Hasegawa K, Ishizawa T, Aoki T, Sano K, Beck Y, Imamura H, Sugawara Y, Kokudo N and Makuuchi M: Is there any difference in survival according to the portal tumor thrombectomy method in patients with hepatocellular carcinoma? Surgery 145: 9-19, Shi J, Lai EC, Li N, Guo WX, Xue J, Lau WY, Wu MC and Cheng SQ: A new classification for hepatocellular carcinoma with portal vein tumor thrombus. J Hepatobiliary Pancreat Sci 18: 74-80, Tsoulfas G, Mekras A, Agorastou P and Kiskinis D: Surgical treatment for large hepatocellular carcinoma: does size matter? ANZ J Surg 82: , Received February 8, 2015 Revised February 21, 2015 Accepted February 24,