IJC International Journal of Cancer

Transcription

1 IJC International Journal of Cancer Adjuvant chemotherapy for patients with primary hepatocellular carcinoma: A meta-analysis Zhouying Zheng 1 *, Wenhua Liang 2 *, Dongping Wang 1 *, Paul M. Schroder 3, Weiqiang Ju 1, Linwei Wu 1, Zheng Zheng 1, Yushu Shang 1, Zhiyong Guo 1 and Xiaoshun He 1 1 Organ Transplant Center, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China 2 Department of Cardiothoracic Surgery, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China 3 University of Toledo College of Medicine, Toledo, OH Numerous studies have investigated the effects of adjuvant chemotherapy for primary hepatocellular carcinoma (HCC) patients. We conducted this analysis to evaluate the efficacy of adjuvant chemotherapy in HCC patients after hepatectomy. PubMed/ MEDLINE, EMBASE, Cochrane, and other databases were searched for eligible studies. The major endpoints were overall survival (OS) and disease-free survival (DFS). The pooled odds ratio (OR) was calculated using a random-effects model to summarize the results. In the meta-analysis of 13 randomized control trials (RCTs) and 35 observational studies with 4747 patients, hepatectomy plus adjuvant chemotherapy showed superiority over hepatectomy alone in 1-year DFS (OR , , p < 0.001), 3-year DFS (OR , , p < 0.001) and 5-year DFS (OR , , p < 0.001), as well as 1-year OS (OR , 95% confidence interval , p < 0.001), 3-year OS (OR , , p < 0.001) and 5-year OS (OR , , p < 0.001). Subgroup and sensitivity analysis revealed that only adjuvant TACE had significant survival benefits. The meta-analysis of studies involving patients with portal vein tumor thrombus (PVTT), but not other factors related to recurrence risk, revealed favorable outcomes of the Treatment arm over the Control arm. The present study shows that adjuvant chemotherapy can improve outcomes for HCC patients. The benefits of adjuvant TACE have been confirmed whereas the effects of other adjuvant chemotherapy modalities remain uncertain. Adjuvant chemotherapy is likely to be more applicable to certain patient populations for instance those with PVTT, but further research in identifying these patient factors is of importance for tailoring adjuvant therapies to individual patients in the future. Hepatocellular carcinoma (HCC) is the most common liver cancer and fifth leading cause of cancer death around the world, with 85% of the cases occuring in developing countries. 1 Surgical resection, a potential curative treatment, has been considered as the first-line treatment for operable HCC for the past few decades. 2,3 However, the recurrence rates Key words: hepatocellular carcinoma, chemotherapy, hepatectomy, recurrence, meta-analysis Abbreviations: CI: confidence interval; DFS: disease-free survival; HAIC: hepatic arterial infusion chemotherapy; HCC: hepatocellular carcinoma; NCCN: national comprehensive cancer network; OR: odds ratio; OS: overall survival; PVC: portal vein chemotherapy; PVCE: portal vein chemoembolization; PVTT: portal vein tumor thrombus; RCT: randomized controlled trial; SC: systemic chemotherapy; TACE: transcatheter hepatic arterial chemoembolization Additional Supporting Information may be found in the online version of this article. *Z. Zheng, W. Liang, and D. Wang contributed equally to this work. Grant sponsor: National High Technology Research and Development Program of China (863 Program); Grant number: 2012AA021008; Grant sponsor: Key Clinical Project from the Ministry of Health; Grant number: ; Grant sponsor: National Natural Science Foundation of China; Grant numbers: , , , ; Grant sponsor: Special Fund for Science Research by Ministry of Health; Grant number: ; Grant sponsor: Research Fund for the Doctoral Program of Higher Education of China by the Ministry of Education; Grant numbers: , , ; Grant sponsors: Science and Technology Planning Key Clinical Project of Guangdong Province; Grant number: 2011A ; Grant sponsor: Major Sci-tech Special Program of Science and Technology Planning Project of Guangzhou City; Grant number: 2012Y ; Grant sponsors: Science and Technology Project of Guangdong Province; Grant number: 2010B ; Grant sponsor: State Key Clinical Specialty Construction Project; Project by Division of Medical Service Management of the Ministry of Health (2010) DOI: /ijc History: Received 18 May 2014; Accepted 29 July 2014; Online 10 Sep 2014 Correspondence to: Xiaoshun He, Organ Transplant Center, the First Affiliated Hospital, Sun Yat-sen University, No. 58 Zhongshan Er Road, Guangzhou , China, gdtrc@126.com and Zhiyong Guo, Organ Transplant Center, the First Affiliated Hospital, Sun Yat-sen University, No. 58 Zhongshan Er Road, Guangzhou , China, rockyucsf1981@126.com Int. J. Cancer: 136, E751 E759 (2015) VC 2014 UICC

2 E752 Adjuvant chemotherapy for HCC: A meta-analysis What s new? There are numerous randomized control trials and observational studies investigating the effects of adjuvant chemotherapy for hepatocellular carcinoma (HCC) patients. The results remain controversial, however. Here, the authors conduct a metaanalysis of all available studies to evaluate the efficacy of adjuvant chemotherapy in patients with primary HCC. They show that adjuvant chemotherapy can improve outcomes for HCC patients. They confirm the benefits of adjuvant transcatheter hepatic arterial chemoembolization (TACE), whereas the effects of other adjuvant chemotherapy modalities remain uncertain. Notably, adjuvant chemotherapy appeared particularly beneficial for patients with portal vein tumor thrombus as a complication of their disease. after hepatic resection remains about 80% at 5 years. 4 Microscopic vascular invasion, intrahepatic metastasis, multicentric occurrence, and non-anatomical resection are predisposing factors to tumor recurrence. 5,6 Adjuvant chemotherapy has been one of the major modalities in treating cancers in recent decades. 7 In general, the majority of HCC tumors can be removed after hepatectomy, and the remnant part can be eliminated by chemotherapy. 8 The benefit of adjuvant chemotherapy in controlling tumor growth and decreasing recurrence has been showed in many other solid tumors like breast cancer and non-small cell lung cancer. 9 There are numerous randomized control trials (RCTs) and observational studies investigating the effects of adjuvant chemotherapy for HCC patients. However, the results of these studies have been controversial. Many individual studies have reported conflicting results and systematic reviews or metaanalyses have a limited number of included studies to evaluate the effects of adjuvant chemotherapy Furthermore, there are no statements regarding adjuvant chemotherapy for HCC patients after hepatectomy in the National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines Version for Hepatobiliary Cancers. 13 In addition, no metaanalyses, reviews, or systematic reviews have evaluated the benefits of all the major modalities of adjuvant chemotherapy. To address these issues, we conducted this meta-analysis of all available studies to evaluate the efficacy of adjuvant chemotherapy in patients with primary HCC. Methods Definitions of outcomes The primary outcome was disease-free survival (DFS), which was defined as the time from randomization (RCTs) or diagnosis (observational studies) until first recurrence or death by any cause. Living patients without disease were censored on the date of last follow-up. The secondary outcome was overall survival (OS), which was defined as the time from randomization (RCTs) or diagnosis (observational studies) until death by any cause. Patients who were still alive at the endpoints of the study were censored on the date of last follow-up. Study searching Sources such as MEDLINE, EMBASE, the Cochrane Library database, ISI Web of Science, and the China Journal Full-text Database were searched. We used a combination of the terms hepatocellular carcinoma, or hepatic cancer, or hepatic carcinoma, or hepatoma, or hepatocarcinoma, or hepatic neoplasm, or liver cancer, or liver carcinoma, or liver neoplasm and adjuvant chemotherapy, or postoperative chemotherapy. An additional search through Google Scholar and a manual search through published literature were performed. Two authors carried out the search independently. The search was not restricted by language or study date. Inclusion and exclusion criteria Trials were included in this meta-analysis if they met the following criteria: (i) The study subjects were primary HCC patients without metastases and other cancerous diseases; (ii) An integrated description of methods and baseline characteristics were provided; (iii) Data contained at least one item of 1-, 3-, 5-year DFS and OS, or DFS and OS curves that cover these time points; (iv) The studies were prospective RCTs and observational studies or retrospective observational studies comparing hepatectomy plus adjuvant chemotherapy (Treatment arm) with hepatectomy alone (Control arm). Trials were excluded if they met any of the following criteria: (i) case series, case reports, reviews and conference reports; (ii) duplicate publications; (iii) studies based on overlapping cohorts from the same institution; (iv) studies with patients undergoing pre-operative treatments, such as chemotherapy, embolization, radiotherapy, radiofrequency and immunotherapy; (v) Studies with statistically significant differences in clinicopathologic and baseline characteristics between the Treatment and Control arms. Note express software (Beijing Aegean Software Company, Beijing, China) was used to manage the information of all included studies. Duplicate references were removed. Two authors independently selected the related references, assessed their eligibility according to the inclusion and exclusion criteria, and any disagreement was solved by discussion. Data extraction and quality assessment We extracted data on author details, year of publication, study methods, participants characteristics, hepatic resection pattern, chemotherapy interventions, patient DFS and OS rate by using a data extraction form. Attempts were made to

3 Zheng et al. E753 Figure 1. Flow chart of study selection. contact the authors of the original articles for the missing data. We evaluated the quality of each study with the modified Jadad scale or modified Newcastle Ottawa Scale. Statistical analysis The data was analyzed by Review Manager 5.1 (The Cochrane Collaboration, 2011, The Nordic Cochrane Centre, Denmark). Treatment effects were expressed as odds ratios (OR) with 95% confidence intervals (CI) for dichotomous outcomes. The Mantel Haenszel analysis was used to assess the efficacy of adjuvant chemotherapy. Random-effects models were used to summarize the results. Heterogeneity between trials was analyzed by v 2 -test with significance set at p < The extent of inconsistency was assessed by I 2 statistic. If heterogeneity existed, an attempt to identify the causes was made using subgroup and sensitivity analyses when appropriate. The point estimate of OR was considered statistically significant at p < 0.05 levels or if the 95% CI did not include the value of 1. Graphical funnel plots were generated to visually assess a publication bias. The statistical methods to detect funnel plot asymmetry were the rank correlation test of Begg and Mazumdar and the regression asymmetry test of Egger. Results Characteristics of included studies Figure 1 illustrates the flow chart of study selection. There were 48 eligible studies, including 13 RCTs and 35 observational studies, with 4,747 participants included in this analysis. Among them, there were 2,235 patients from 19 studies included in the transcatheter hepatic arterial chemoembolization (TACE) group, 615 patients from 10 studies in the hepatic arterial infusion chemotherapy (HAIC) group, 288 patients from four studies in the portal vein chemotherapy (PVC) group, 481 patients from six studies in the systemic chemotherapy (SC) group, and 582 patients from eight studies in the Combined (combination of two or more adjuvant chemotherapy modalities) group. Five observational studies were identified concerning locoregional chemotherapy with unclear classification, and one evaluated portal vein chemotherapy with lipiodol embolism (PVCE). These studies fell into none of the groups mentioned above. Therefore, we included these six studies with 666 patients in the ELSE group. More detailed information and the clinical characteristics of the included studies are summarized in Supporting Information Table 1. According to the pre-established criteria, we performed a quality assessment of each study with modified Jadad Scale or modified Newcastle Ottawa Scale (Supporting Information Tables 2 4). Seven RCTs, the scores of which were less than 4 in the modified Jadad Scale, were considered to be of low quality. Other RCTs and all of the observational studies were of high quality. Considering of the above, we conducted a subgroup analysis according to the quality of each study. Meta-analysis of all included studies Thirteen RCTs and 35 observational studies were included in this analysis, which contained 53 comparisons between the Treatment arm and Control arm. The pooled results showed significant improvement in 1-year DFS (OR , 95% CI , p < 0.001) (Fig. 2), 3-year DFS (OR , , p < 0.001) (Fig. 3), and 5-year DFS (OR , , p < 0.001) (Fig. 4) without statistically significant heterogeneity identified in the analysis of DFS. In the assessment of OS, the pooled results revealed statistically significant improvement in 1-year OS (OR , , p < 0.001) (Fig. 5), 3-year OS (OR , , p < 0.001) (Fig. 6), and 5-year OS (OR , , p < 0.001) (Fig. 7) without statistically significant heterogeneity identified in the analysis of 1-year OS (p , I %) or 3-year OS (p , I %), but there was heterogeneity identified in the analysis of 5-year OS (p , I %). Subgroup analysis and sensitivity analysis Although no significant heterogeneity was observed in most of the comparisons, we sought to minimize the potential

4 E754 Adjuvant chemotherapy for HCC: A meta-analysis Figure 2. Meta-analysis of 1-year DFS in all included studies comparing hepatectomy plus adjuvant chemotherapy (Treatment) with hepatectomy heterogeneity and probe into detailed results in the subpopulation by performing a subgroup analysis and a sensitivity analysis. We stratified the analyses according to several important confounding factors, which were believed to have significant effects on the results of a study including study design (RCTs or non-rcts), modalities of adjuvant chemotherapy (TACE, HAIC, PVC, SC, Combined and ELSE group), complication of portal vein tumor thrombus (PVTT; all, some or none of the patients), multiple tumor (more than 50% of the patients complicated or not), mean tumor size (larger than 5cm or not), complication of cirrhosis (more than 50% of the patients complicated or not), TNM stage (more than 50% of the patients in stage I and II or III and IVA), and the quality assessment of each study (low or high quality). Supporting Information Table 5 summarizes all of the subgroup results and overall results. The results of RCTs or non-rcts were similar to the overall results, while the outcomes of 1-year DFS and 5-year OS from RCTs revealed no statistically significant difference between the treatment arm and control arm. As to the subgroup results of adjuvant chemotherapy modalities, TACE showed significant improvement in 1-year DFS (OR , , p ) and OS (OR , , p < 0.001), 3-year DFS (OR , , p < 0.001) and OS (OR , , p < 0.001), 5-year DFS (OR , , p ) and OS (OR , , p ) with significant heterogeneity only identified in 5-year OS (p , I %) across studies. The results of HAIC treatment showed favorable outcomes of 1-, 3-, 5-year DFS and 3-, 5-year OS compared to the Control arm. However, most of these results were derived from observational studies. In addition, the pooled results of other adjuvant chemotherapy modalities remained undetermined because there were a limited number of RCTs and observational studies reporting results for these therapies that met the inclusion criteria. In terms of the subgroup results of those studies involving patients with PVTT (either all or some of the patients in the study) as a complication of their disease course (PVTT group), the analysis revealed favorable outcomes in the Treatment arm compared to the Control arm, while the results of other studies (non-pvtt group) did not. Importantly, the pooled OR values of the analysis of the PVTT group were higher than those of the non-pvtt group and the overall results in most of the outcomes. Moreover, we carried out a meta-analysis of

5 Zheng et al. E755 Figure 3. Meta-analysis of 3-year DFS in all included studies comparing hepatectomy plus adjuvant chemotherapy (Treatment) with hepatectomy Figure 4. Meta-analysis of 5-year DFS in all included studies comparing hepatectomy plus adjuvant chemotherapy (Treatment) with hepatectomy C 2014 UICC Int. J. Cancer: 136, E751 E759 (2015) V

6 Adjuvant chemotherapy for HCC: A meta-analysis E756 Figure 5. Meta-analysis of 1-year OS in all included studies comparing hepatectomy plus adjuvant chemotherapy (Treatment) with hepatectomy those studies in which all the included patients had a disease course complicated with PVTT. It showed an obviously significant improvement of DFS (OR , , p ) (Supporting Information Fig. 1) and OS (OR , , p < 0.001; Supporting Information Fig. 2) without statistically significant heterogeneity. According to the quality assessment of each study, we performed a subgroup analysis, most of the results are consistent with the overall results except the subgroup result of low quality in 5-year DFS. Also, we observed that most studies reported adjuvant chemotherapy tended to improve the patient s prognosis after hepatectomy. This was true for all studies except the studies reported by Lai et al. and Hasegawa et al. When we excluded the Lai et al. study, heterogeneity was significantly reduced in 3-year DFS (p C 2014 UICC Int. J. Cancer: 136, E751 E759 (2015) V

7 Figure 6. Meta-analysis of 3-year OS in all included studies comparing hepatectomy plus adjuvant chemotherapy (Treatment) with hepatectomy value increased from 0.01 to 0.32 and I2 dropped from 43 to 11%), and the pooled OR increased from 2.37 ( ) to 2.45 ( ). In addition, when we left the Hasegawa et al. study out, homogeneity was improved in 5-year OS (p value increased from to 0.02 and I2 dropped from 53 to 43%), and the pooled OR increased from 1.92 ( ) to 2.04 ( ) in 5-year OS. Publication bias The meta-analysis of 1-, 3-year DFS and 3-, 5-year OS showed potential publication biases with the p values lower than 0.05 given by Begg s or Egger s test. There was no evidence of pubc 2014 UICC Int. J. Cancer: 136, E751 E759 (2015) V lication bias for other results, as the funnel plot analysis showed a symmetrical appearance and p value ranging from 0.26 to 0.88 given by Begg s test and p value ranging from 0.46 to 0.69 by Egger s test (Supporting Information Fig. 3). Discussion In the meta-analysis of all included studies, we found that adjuvant chemotherapy was significantly superior to Control arm with higher 1-, 3-, 5-year DFS and OS. The pooled results of RCTs revealed statistically significant improvement of patients in the Treatment arm compared to the Control arm in all DFS and OS except 1-year DFS and 5-year OS. Our results E757 Zheng et al.

8 E758 Adjuvant chemotherapy for HCC: A meta-analysis Figure 7. Meta-analysis of 5-year OS in all included studies comparing hepatectomy plus adjuvant chemotherapy (Treatment) with hepatectomy provide evidence that adjuvant chemotherapy after hepatectomy provides benefits for operable HCC patients. However, distinct modalities of adjuvant chemotherapy may yield different patient outcomes. The meta-analyses of all included studies and only RCTs concerning TACE show improved 1-,3-, 5-year DFS and OS. The subgroup analysis results of adjuvant HAIC are similar to those of TACE. However, the meta-analysis of only RCTs concerning HAIC is not applicable with no more than one study included. The pooled OR values of HAIC in 1-, 5- yeardfsand3-,5-yearosarehigherthanthatoftace,but withlowerorvaluesin3-yeardfsand1-yearos.somestudies have compared the efficacy between TACE and HAIC in patients not receiving hepatectomy, indicating that chemotherapy drugs and patient selection may affect the outcomes of patients undergoing TACE and HAIC therapies. 14,15 Therefore, it is difficult to determine which one is better based on the current studies, and it is of interest to carry out studies directly comparing adjuvant TACE with HAIC in the future. In the subgroup analysis of adjuvant SC, the superior results of the Treatment arm were revealed in 3-, 5-year DFS and 3- year OS in comparison to the Control arm. Considering the small number of included studies, the efficacy of adjuvant SC is not yet clear and needs to be confirmed in future studies. On the other hand, with the growing knowledge of the molecular pathway of carcinogenesis, oral molecular targeted agent therapies have become the blue ocean of cancer treatment. 16 Most notably, sorafenib has been approved as a standard treatment for patients with advanced and unresectable HCC in most countries worldwide. 17,18 Recently, a pilot study showed that setting sorafenib as adjuvant therapy for HCC to prevent early recurrence after hepatic resection could be a potential indication. 19 However, because of the small sample size of the study, the therapeutic potential of sorafenib has not yet been proven in HCC patients after resection. Similarly, the efficacy of combined chemotherapy is unclear based on the current analysis. Considering the anticancer agents of adjuvant chemotherapy in each study, the most frequently used anticancer agents for adjuvant chemotherapy are the different combinations of ADM, MMC, DDP, 5-Fu and their derivatives. Besides these traditional anticancer agents, the application of some new drugs such as oxaliplatin (OXA), hydroxycamptothecine (HCPT) and capecitabine have been investigated by researchers. It has been revealed that OXA-based chemotherapy is effective in advanced HCC and represented a viable option in these patients. 20 HCPT with a low concentration could induce hepatocyte-specific differentiation of cancer stem-like cells from HCC cell line. 21

9 Zheng et al. E759 Notably, metronomic capecitabine can be well tolerated by patients with advanced HCC and appears to have activity both in treatment-naive patients and in those previously treated with sorafenib. 22 These novel drugs provide alternatives to the traditional anticancer agents for HCC adjuvant chemotherapy. In terms of patient selection for adjuvant chemotherapy, we have conducted a subgroup analysis of those studies involving patients with and without PVTT as a complication of their disease. Only in patients with (but not without) PVTT, the benefit of adjuvant chemotherapy was confirmed. HCC patients with PVTT would be considered to be advanced stage of liver cancer, 23 which often leads to intrahepatic metastasis or recurrence. In agreement with this result, some studies have demonstrated that adjuvant TACE is thought to be more applicable to patients at high risk for recurrence. 8,24 We have conducted a further subgroup analysis of studies with distinct tumor characteristics including multiple tumors, mean tumor size, cirrhosis status, and TNM stage. However, we failed to document any of these factors affecting the pooled analysis results. Notably, the risks of adjuvant chemotherapy should be weighed against its benefits. Adjuvant chemotherapy, especially adjuvant systemic chemotherapy, may not be sensitive to the majority of HCC patients owing to the drug tolerance of cancerous cells. In addition, excessive adjuvant chemotherapy may suppress the immune system and promote the transfer of the liver cancer. Finally, adjuvant chemotherapy, especially chemoembolization, may cause hepatic parenchymal damage and decline the compensatory function of cirrhotic liver. All these factors would inevitably affect the longterm survival of HCC patients. 8 Although this meta-analysis summarizes all available eligible studies concerning the use of adjuvant chemotherapy for HCC patients, there are still some drawbacks of this study. First, the number of included RCTs was relatively small and not all studies described the method of randomization. Second, there was some significant heterogeneity that may come from different clinical baseline characteristics and intervention protocols between the included studies. Third, some studies only reported DFS rates while some only reported OS rates. Finally, the quality of life was not assessed in those that received chemotherapy. In consideration of the limitations above, the results and conclusions in this study should be interpreted with cautions. In conclusion, we demonstrate that adjuvant chemotherapy is an effective treatment to improve the prognosis of primary HCC patients who underwent hepatectomy. Among all the modalities of adjuvant chemotherapy, only the efficacy of TACE has been verified by both observational studies and RCTs. Notably, adjuvant chemotherapy may be particularly beneficial for patients with PVTT as a complication of their disease. Collectively, we recommend adjuvant chemotherapy for selected HCC patients after resection with a suitable chemotherapy modality and chemotherapeutic agents. References 1. El-Serag HB. Epidemiology of viral hepatitis and hepatocellular carcinoma. Gastroenterology 2012; 142: Arii S, Yamaoka Y, Futagawa S, et al. Results of surgical and nonsurgical treatment for small-sized hepatocellular carcinomas: a retrospective and nationwide survey in Japan. The Liver Cancer Study Group of Japan. Hepatology 2000;32: Makuuchi M, Donadon M, Torzilli G. Hepatic resection for hepatocellular carcinoma in cirrhosis. Ann Ital Chir 2008;79: Ueno M, Uchiyama K, Ozawa S, et al. Adjuvant chemolipiodolization reduces early recurrence derived from intrahepatic metastasis of hepatocellular carcinoma after hepatectomy. Ann Surg Oncol 2011;18: Imamura H, Matsuyama Y, Tanaka E, et al. Risk factors contributing to early and late phase intrahepatic recurrence of hepatocellular carcinoma after hepatectomy. J Hepatol 2003;38: Wu ZQ. Evaluation of postoperative adjuvant chemotherapy for hepatocellular carcinoma. J. Abdominal Surg 2001;14: Huang JD, Li YM, Lei ZM. Evaluation of preoperative or post-operative transarterial chemotherapy for primary liver carcinoma. J Evidence- Based Med 2007;7: Ren ZG. Postoperative adjuvant chemotherapy for hepatocellular carcinoma after resection. Oncol Prog 2005;3: Fukui T, Mitsudomi T. Postoperative adjuvant chemotherapy for lung, gastric, colorectal and breast cancer. Nippon Rinsho 2010;68: Chan ES, Chow PK, Tai B, et al. Neoadjuvant and adjuvant therapy for operable hepatocellular carcinoma. Cochrane Database Syst Rev 2000;2: CD Mathurin P, Raynard B, Dharancy S, et al. Metaanalysis: evaluation of adjuvant therapy after curative liver resection for hepatocellular carcinoma. Aliment Pharmacol Ther 2003;17: Zhong JH, Li LQ. Postoperative adjuvant transarterial chemoembolization for participants with hepatocellular carcinoma: a meta-analysis. Hepatol Res 2010;40: National Comprehensive Cancer Network (NCCN). Clinical Practice Guidelines in Oncology: Hepatobiliary Cancers Version Okusaka T, Kasugai H, Shioyama Y, et al. Transarterial chemotherapy alone versus transarterial chemoembolization for hepatocellular carcinoma: a randomized phase III trial. J Hepatol 2009;51: Kim HY, Kim JD, Bae SH, et al. A comparative study of high-dose hepatic arterial infusion chemotherapy and transarterial chemoembolization using doxorubicin for intractable, advanced hepatocellular carcinoma. Korean J Hepatol 2010,16: Woo HY, Heo J. Sorafenib in liver cancer. Expert Opin Pharmacother 2012;13: Llovet JM, Ricci S, Mazzaferro V, et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 2008;359: Cheng AL, Kang YK, Chen Z, et al. Efficacy and safety of sorafenib in patients in the Asia- Pacific region with advanced hepatocellular carcinoma: a phase III randomised, double-blind, placebo-controlled trial. Lancet Oncol 2009;10: Wang SN, Chuang SC, Lee KT. Efficacy of sorafenib as adjuvant therapy to prevent early recurrence of hepatocellular carcinoma after curative surgery: a pilot study. Hepatol Res / hepr Petrelli F, Coinu A, Borgonovo K, et al. Oxaliplatin-based chemotherapy: a new option in advanced hepatocellular carcinoma. A systematic review and pooled analysis. Clin Oncol (R Coll Radiol) 2014;26: Zhang Y, Song WJ, Zhang FQ, et al. Differentiation-inducing activity of hydroxycamptothecin on cancer stem-like cells derived from hepatocellular carcinoma. Dig Dis Sci 2011;56: Brandi G, de Rosa F, Agostini V, et al. Metronomic capecitabine in advanced hepatocellular carcinoma patients: a phase II study. Oncologist 2013;18: Mitsunobu M, Toyosaka A, Oriyama T, et al. Intrahepatic metastases in hepatocellular carcinoma: the role of the portal vein as an efferent vessel. Clin Exp Metastasis 1996;14: Li JQ, Zhang CQ, Zhang YQ, et al. Value of adjuvant transcatheter hepatic arterial chemoembolization after hepatectomy for primary liver carcinoma. Chin J Bases Clin Gen Surg 2006;13: