Introduction

Oncology Letters

1792-1074 1792-1082

D.A. Spandidos

10.3892/ol.2016.5525

OL-0-0-5525

Articles

Sorafenib therapy following resection prolongs disease-free survival in patients with advanced hepatocellular carcinoma at a high risk of recurrence

Liao

Yadi

1 2 * Zheng

Yun

1 2 * He

Wei

1 2 Li

Qijiong

1 2 Shen

Jingxian

3 Hong

Jian

1 2 Zou

Ruhai

4 Qiu

Jiliang

1 2 Li

Binkui

1 2 Yuan

Yunfei

1 2

1State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Sun Yat-Sen University Cancer Center, Guangzhou, Guangdong 510060, P.R. China 2Department of Hepatobiliary Oncology, Sun Yat-Sen University Cancer Center, Guangzhou, Guangdong 510060, P.R. China 3Department of Medical Imaging and Interventional Center, Sun Yat-Sen University Cancer Center, Guangzhou, Guangdong 510060, P.R. China 4Department of Ultrasound, Sun Yat-Sen University Cancer Center, Guangzhou, Guangdong 510060, P.R. China

Correspondence to: Dr Binkui Li or Dr Yunfei Yuan, Department of Hepatobiliary Oncology, Sun Yat-Sen University Cancer Center, 651 Dongfeng Road East, Guangzhou, Guangdong 510060, P.R. China, E-mail: libinkui@mail.sysu.edu.cn, E-mail: yuanyf@mail.sysu.edu.cn *

Contributed equally

02 2017

20 12 2016

13 2 984 992 14022015 13052016

2017

Sorafenib is the standard systemic treatment for patients with advanced hepatocellular carcinoma (HCC); however, its therapeutic value in patients with HCC following resection remains controversial. The current retrospective study was undertaken to assess the effects of sorafenib treatment following surgical resection in patients with advanced HCC disease who were at a high risk for recurrence. Between July 2010 and July 2013, a consecutive cohort of 42 patients with advanced HCC and at a high risk of recurrence (i.e., those with portal vein tumor thrombosis, adjacent organ involvement or tumor rupture) who underwent resection were analyzed. The patients were categorized into the sorafenib group (n=14) or the best supportive care (BSC) group (n=28). Although the histological grade, Barcelona Clinic Liver Cancer Stage, tumor size, nodule number and proportion of patients with high serum α-fetoprotein levels were comparable between the sorafenib and BSC groups, those receiving sorafenib following resection had significantly longer disease-free survival (DFS) of 5.2 months [95% confidence interval (CI), 1.2–9.2 months] compared with the BSC group [1.8 months (95% CI, 0.6–3.0 months)]. No differences in overall survival were noted between the groups. Furthermore, no drug-related adverse events resulted in discontinuation of sorafenib therapy. Univariate log-rank analysis revealed that sorafenib treatment (P=0.002) and treatment prior to resection (P=0.012) were significantly associated with longer DFS; however, sorafenib therapy (P=0.027) and tumor size (P=0.028) were associated with longer DFS by multivariate analysis. Furthermore, sorafenib was well-tolerated and improved DFS in patients with advanced HCC who underwent hepatic resection. Thus, tumor resection followed by sorafenib therapy may represent an effective therapeutic strategy for patients with advanced HCC. This possibility should be confirmed in larger, multicenter studies.

best supportive care disease-free survival hepatocellular carcinoma resection sorafenib

Introduction

Hepatocellular carcinoma is the fifth most prevalent type of cancer (1) and the third most common cause of cancer-related mortality worldwide (2). There are >700,000 new cases of HCC diagnosed each year worldwide (2), and the mortality rate is similar to the incidence rate due to high levels of intrahepatic recurrence and distant metastasis, and the lack of effective treatments (2). Despite advances in various treatment strategies, such as resection, transplantation, ablation, chemoembolization and targeted therapy with sorafenib, the prognosis of patients with HCC remains poor due to disease recurrence (1,3). In addition to transplantation, hepatic resection is considered the most effective treatment strategy, with a 5-year survival rate of 60–80% for selected patients (4,5).

The selection criteria for resection are as follows: i) Patients with solitary tumors and very well-preserved liver function, as normal bilirubin with either hepatic venous pressure gradient ≤10 mmHg or platelet count ≥100,000 cells; and ii) patients with multifocal tumors meeting the Milan criteria (≤3 nodules of ≤3 cm) or with mild portal hypertension who are not suitable for liver transplantation require prospective comparisons with loco-regional treatments. However, liver transplantation is considered to be the first line treatment option for patients with single tumors of <5 cm or with ≤3 nodules of ≤3 cm (Milan criteria) which are not suitable for resection (1,2).

No effective therapy to reduce the risk of HCC recurrence is currently available (6), and systemic chemotherapy, hepatic-artery chemotherapy and chemoembolization have little effect (7,8). Other treatments, such as immunotherapy and interferon therapy, may improve overall survival and recurrence-free survival (9–12). However, more studies are required before they can be recommended a part of the routine clinical practice.

HCC is further complicated by the incidence of portal vein tumor thrombosis (PVTT), tumors involving adjacent organs or those with tumor rupture. The prognosis for patients with HCC complicated by PVTT is extremely poor, with a median survival of only 2–4 months (13–17) for those receiving surgical resection, portal vein stent, transcatheter arterial chemoembolization (TACE), radiotherapy, ¹²⁵I seed implantation, percutaneous ethanol injection (PEI) or laser ablation (18).

Sorafenib is a multikinase inhibitor that targets the serine/threonine kinases, Raf-1 and B-Raf; the mitogen-activated protein kinases, mitogen-activated protein kinase kinase and extracellular signal-regulated kinase; and the receptor tyrosine kinase activity of vascular endothelial growth factor receptors 1–3; and platelet-derived growth factor receptor-β (19–22). In the phase III Sorafenib HCC Assessment Randomized Protocol (SHARP) (23) and Asian-Pacific (24) studies, sorafenib improved the overall survival of patients with advanced, unresectable HCC by 3 months. Thus, Western guidelines, including the American Association for the Study of Liver Diseases and the European Association for the Study of the Liver (3,4), recommend sorafenib as the standard systemic treatment for advanced HCC. By contrast, sorafenib is not widely available in Asian-Pacific countries; instead, surgical resection is a more common treatment for patients with advanced HCC. In a small pilot study that included 31 patients with advanced HCC, the recurrence rate was decreased in patients receiving sorafenib following resection (25). As the outcomes of patients with advanced HCC remains poor with sorafenib monotherapy, the present study aimed to determine if the survival of HCC patients with a high risk of recurrence (i.e., those with PVTT, tumors involving adjacent organs or those with tumor rupture) who underwent resection could be prolonged. The association between clinicopathological characteristics and disease-free survival (DFS) was analyzed by univariate and multivariate analyses. Sorafenib therapy following resection may represent a novel approach for treating patients with advanced HCC that are at a high risk for tumor recurrence.

Patients and methods <sec> <title>Patients

A total of 42 patients with HCC that were at high risk of disease recurrence and who underwent hepatic resection between July 2010 and July 2013 at the Sun Yat-Sen University Cancer Center (Guangzhou, China) were included in the present study. Demographic data such as Eastern Cooperative Oncology Group performance status (ECOG PS) score, Barcelona Clinic Liver Cancer Stage (BCLC) stage and Child-Pugh class were determined prior to resection. The scores were calculated on the basis of common criteria as described in previous studies (26–28). Histological grade was classified following resection according to the criteria of the World Health Organization Classification of Tumors (2000) (29). All samples were obtained prior to surgery to determine serology, α-fetoprotein (AFP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), albumin and bilirubin levels, and prothrombin time. The normal ranges are as follows: AFP, 0–25 ng/ml; ALT, 0–40 U/l; AST, 0–45 U/L; ALP, 30–110 U/l; albumin, 35–55 g/l; bilirubin, 5.0–20.5 µmol/l; and prothrombin time, 10.5–13.5 sec. The diagnosis of HCC was confirmed by pathological examination. All diagnoses were formulated according to the criteria for terminology established by the International Working Party (30). Tumor differentiation was based on the criteria of the World Health Organization Classification of Tumors (2000) (29). Tumor stage was defined according to the 2010 American Joint Committee on Cancer/International Union Against Cancer tumor-node-metastasis classification system (31). Patients were defined as being at a high risk of recurrence if they had one of the following intraoperative findings: i) Tumors involving a major branch of the portal or hepatic vein (i.e., PVTT); ii) tumors with direct invasion of adjacent organs; and iii) tumor ruptures prior to or during resection (Table I). Patients were excluded if they had one of the following: i) A history of molecularly targeted therapy or systemic chemotherapy following resection; ii) severe coagulation disorders (prothrombin activity of <40% or platelet count of <4,000 cells/µl); or iii) hepatic decompensation, including ascites, esophageal or gastric variceal bleeding, and hepatic encephalopathy. The present study was approved by the institutional review board of the Sun Yat-Sen University Cancer Center, who confirmed that patient consent was not required for the present retrospective study.

Tumor resection

The extent of resection was performed according to the standard of practice, as described by Luo et al (32). For those with PVTT, the liver tumor was completely removed such that the surgical margin was negative and the tumor thrombus was removed after opening the portal vein. For those with tumor rupture, the liver tumor was completely removed such that the surgical margin was negative. For those tumors involving adjacent organs, the liver tumor was completely removed such that the surgical margin was negative and the diaphragm or the omentum was partially removed.

Sorafenib therapy, toxicity evaluation and treatment response

Patients received either sorafenib treatment (n=14) or best supportive care (BSC; n=28) following resection. According to the American Association for the Study of Liver Diseases and the European Association for the Study of the Liver guidelines (1,5), no effective adjuvant therapy able to reduce recurrence rates following resection is available at present. Thus, the best supportive care in Sun Yat-Sen University Cancer Center included nutrition education, liver protection, pain management and psychotherapy. For patients with a high risk of recurrence, sorafenib was routinely recommended by the treating physician. Upon recommendation, patients were asked to decide whether to receive the optional sorafenib treatment. Sorafenib therapy was initiated 7 days after surgery with an initial dose of 400 mg sorafenib. Patients in this group received continuous oral treatment with 400 mg sorafenib (Bayer AG, Leverkusen, Germany) twice daily. The dose was reduced in the event of unacceptable drug-related side-effects, including hand-foot skin reaction, diarrhea, rash and hypertension. Sorafenib toxicity was evaluated according to the National Cancer Institute's Common Terminology Criteria for Adverse Events, version 3.0 (33). Tumor response and recurrence were evaluated every 4–8 weeks, according to Response Evaluation Criteria in Solid Tumors (34), by performing computed tomography or magnetic resonance imaging scans of the liver. DFS was defined as the time from resection to the date of tumor recurrence or the patient's most recent follow-up examination.

Statistical analysis

Baseline characteristics are expressed as n (%) for categorical data and mean ± standard deviation (SD) for continuous data by group. Differences between groups were compared using Pearson's χ² or Fisher's exact tests for categorical data; two-sample t-tests were used for continuous data. Data normal distribution was determined by one-sample Kolmogorov-Smirnov test with P<0.05. For continuous data that did not follow normal distribution, data are expressed as median (range) and the Mann-Whitney U test was used to compare between groups. DFS data are presented using Kaplan-Meier survival curves by group and compared using log-rank tests. Univariate and multivariate Cox regression analyses were performed to identify the prognostic factors associated with DFS. For multivariate analysis, variable selection was performed using the backward elimination method with P<0.1 set as the cutoff level. According to previous studies (35–38), factors associated with early recurrence were non-anatomical resection, microscopic vascular invasion and serum AFP level, while those contributing to late-phase recurrence were hepatitis activity, multiple tumors and histological grade. Thus, considering clinical significance and the results of previous studies (6–9), 8 factors were selected as potential confounders in the present study. The 8 factors were as follows: Sorafenib (yes vs. no), hepatitis serology (hepatitis B surface antigen-positive vs. other), histological grade (poor vs. well-moderate), tumor size (≥10 vs. <10 cm), tumor nodules (multiple versus single), AFP (≥400 vs. <400 ng/ml), previous treatment (yes vs. no) and anti-virus treatment (yes vs. no). Results are presented as hazard ratios (HRs) with 95% confidence intervals (95% CI). All statistical assessments were two-tailed and P<0.05 was considered to indicate a statistically significant difference. All statistical analyses were performed with SPSS statistical software for Windows (version 22; IBM SPSS, Chicago, IL, USA).

Results <sec> <title>Patient characteristics

Between July 2010 and July 2013, 42 consecutive HCC patients who underwent resection and were diagnosed as having a high risk of disease recurrence were enrolled in the present study, including 37 male and 5 female patients, with a mean age of 48.1 years (range, 28–69 years). Among the enrolled patients, 14 (33.3%) received postoperative sorafenib therapy (sorafenib group) for a median duration of 14.3 months (range, 2.6–24.2 months) and the other 28 received BSC. As indicated in Table I, the proportion of patients with each high-risk factor was not significantly different between the two postoperative groups.

The demographic data of all of the enrolled patients are included in Table II. The majority of patients (97.6%) had an ECOG PS score of 0–1 and 90.5% were hepatitis B virus (HBV)-positive. There was no significant difference in the proportion of patients receiving anti-viral treatment between the sorafenib and BSC groups (P=0.261). In addition, all patients were classified as Child-Pugh class A (Child-Pugh score 5–6). No differences in the histological grade or BCLC stage were noted between the groups (P=0.326 and P=1.000, respectively). Tumor size, number of nodules and proportion of patients with high serum AFP levels (≥400 ng/ml) were also similar between the two groups (P≥0.259). However, a significantly higher proportion of patients in the sorafenib group received previous treatment compared with the BSC group (P=0.003). Furthermore, the sorafenib group had higher baseline albumin levels (42.5±4.3 vs. 39.4±3.4 g/l, P=0.016) and lower median baseline levels of ALP [75.1 U/l (range, 44.4–185.5 U/l) vs. 112.9 U/l (range, 55.7–197.2 U/l); P=0.007] compared with the BSC group (Table II). Sorafenib was well-tolerated in all patients following hepatic resection, as there were no drug-related adverse events that led to discontinuity.

DFS analysis

Follow-up was performed until May 8, 2014. The median follow-up time was 13.6 months (range, 1.5–40.1 months). Of the total cohort, 9 patients (2 in the sorafenib group and 7 in the BSC group) succumbed to the disease and 38 patients (11 in the sorafenib group and 27 in the BSC group) developed recurrence during the follow-up period.

The mean overall survival time was 29.1 months (range, 1.5–40.1 months) and was comparable between the two groups (P=0.405). The overall median DFS time was 3.2 months (range, 0.3–16.6 months) and log-rank analysis revealed that patients in the sorafenib group had improved DFS of 5.2 months (95% CI, 1.2–9.2 months) compared with those in the BSC group [1.8 months (95% CI, 0.6–3.0 months); P=0.001; Fig. 1]. Furthermore, patients that had received previous treatment (TACE and radiation therapy) prior to resection had longer DFS of 9.2 months (95% CI, 0–23.7 months) compared with those that had not [2.5 months (95% CI, 1.2–3.9 months); P=0.006; Fig. 2].

Factors associated with DFS

Univariate and multivariate Cox regression analysis were performed to identify the factors associated with DFS. As indicated in Table III, there was no significant univariate association between DFS time and hepatitis serology, AFP level, tumor size, extent of tumor nodules, histological grade, Child-Pugh score or anti-virus treatment. However, sorafenib therapy (HR, 0.243, 95% CI, 0.098–0.599; P=0.002) and previous treatment prior to resection (HR, 0.204, 95% CI, 0.059–0.703; P=0.012) were associated with a decreased recurrence rate upon univariate analysis. Furthermore, multivariate Cox regression analysis revealed that sorafenib treatment (HR, 0.328; 95% CI, 0.122–0.880; P=0.027) and tumor size (HR, 0.449; 95% CI, 0.220–0.917; P=0.028) were significantly associated with DFS (Table III).

Discussion

Since the SHARP trial in 2007 (15), sorafenib has become the standard treatment for patients with advanced, unresectable HCC in the majority of countries worldwide, prolonging survival in these patients by ~3 months (23,24). However, surgical resection is also an acceptable treatment for patients with advanced HCC in Asian-Pacific countries, as sorafenib is often not readily available. Long-term survivors with advanced HCC receiving sorafenib commonly require multimodal therapies, including hepatic resection following successful sorafenib treatment, salvage therapies in the event of progressive disease, and effective post-sorafenib treatment modalities (39). Thus, the current retrospective study was undertaken to assess the therapeutic value of administering sorafenib treatment following surgical resection in patients with advanced HCC disease that were at a high risk of recurrence (i.e., those with PVTT, adjacent organ involvement or tumor rupture). Although the histological grade, BCLC stage, tumor size, nodule number and proportion of patients with high serum AFP levels were comparable between the sorafenib and BSC groups, those receiving sorafenib following resection exhibited significantly improved DFS. Although a greater proportion of patients in the sorafenib group received treatment prior to resection, multivariate analysis indicated that sorafenib and tumor size were the only variables that were associated with DFS. In univariate analysis, the error from the confounders will be reflected in the random error, which will cause the denominator to appear greater than its actual value, resulting in reduced significance. Thus, if >1 factors affect the observation value, multivariate analysis should be applied.

The overall median DFS in the current study was 3.2 months, which is comparable with previous studies (40,41). Dollinger et al observed that the time to progression of advanced HCC was 5.3 and 2.9 months for patients treated with thymostimulin and placebo, respectively (41). Nagano et al noticed that the overall median progression-free survival (PFS) of advanced HCC was 2.0 months (40). In addition, the association between sorafenib therapy and longer DFS is largely concordant with previous studies in patients with HCC (25,42–52). In a pilot study by Wang et al (25) that included 31 patients at a high risk of recurrence, sorafenib therapy following hepatic resection significantly improved the time to recurrence as well as the recurrence rate. Furthermore, univariate analysis revealed that sorafenib following resection was independently associated with recurrence-free survival (25). In a different study, a complete response was observed in a single patient with hepatitis C virus-mediated cirrhosis and recurrent HCC with PVTT following treatment with 400 mg sorafenib once daily (42). Furthermore, in a case-controlled study of 39 patients that experienced HCC recurrence following liver transplantation, the median survival time from recurrence was significantly improved with sorafenib as compared with BSC, and sorafenib was identified as the only factor associated with survival by multivariate analysis (48). The effect of sorafenib on recurrence has also been observed in animal studies. For example, Feng et al (49) demonstrated that sorafenib suppresses postsurgical recurrence and metastasis via enhancing the sensitivity of HCC cells to sorafenib in an orthotopic mouse model. In a study by Wang et al (50), sorafenib suppressed metastatic recurrence following resection in nude mice. By contrast, a recent phase III randomized, double-blind, placebo-controlled study of sorafenib therapy following resection or ablation with curative intent in 1,114 patients revealed no differences in recurrence-free survival, time to recurrence or overall survival between patients receiving sorafenib or placebo (51). It is possible that differences in disease severity (early-stage versus high-risk advanced disease) between the studies could account for the varying outcomes.

In a previous study, 44 Italian patients with HCC and established cirrhosis were treated with sorafenib for 16 weeks. In these patients, ECOG PS, extrahepatic spread, macrovascular invasion, and AFP and ALP levels at admission were identified as independent predictors of overall survival (52). In another study of 110 patients with advanced HCC that were treated with sorafenib in Germany, ascites, macrovascular invasion and BCLC stage D were independently associated with overall survival (53). Furthermore, the use of nucleoside analogues was associated with a lower risk of HCC recurrence (36). By contrast, the use of antiviral therapy was the same in both groups in the present study. Although factors associated with overall survival were not assessable for analysis in the present study due to the length of the follow-up period, histological grade, BCLC stage, tumor size, nodule number and high serum AFP levels were not associated with DFS.

Despite the fact that the present results differ from those reported by previous studies (36,52,53), the present data are still solid due to the following reasons. Firstly, in the current study, only patients with a high risk of recurrence were included. Patients were defined as being at a high risk of recurrence if they had one of the following intraoperative findings: i) Tumors involving a major branch of the portal or hepatic vein (i.e., PVTT); ii) tumors with direct invasion of adjacent organs; and iii) tumor ruptures prior to or during resection. The above risks may become the main factors that decide the prognosis. Secondly, histological grade, BCLC stage, tumor size, nodule number and high serum AFP levels were not associated with DFS in the present study.

The patients in the current study were selected on the basis of their being at a high risk for recurrence, including those with PVTT, adjacent organ involvement and tumor rupture. In addition to surgical resection, patients with PVTT are typically treated with a portal vein stents, TACE, radiotherapy, ¹²⁵I seed implantation, PEI and laser ablation (18). Although TACE appears to have potential for providing a survival benefit for HCC patients with PVTT, with a median PFS of 3.6–5.1 months (54,55), the deteriorating liver function caused by TACE limits its application (56,57). By contrast, hepatic resection may produce superior survival benefits for HCC patients with PVTT, with a 1-year overall survival rate of 34.4–42.0% (58–61), and radiotherapy may have superior local effects in these patients (62,63). The prognosis of HCC patients with PVTT is dismal. Therapy strategies for such patients, including TACE, had been compared in previous studies. Peng et al reported that hepatic resection, compared with TACE, provided survival benefits for patients with resectable HCC with PVTT, particularly for those with type I or II PVTT (58). Chen et al noticed that liver resection with thrombectomy, compared with liver resection without thrombectomy, yielded better outcomes in HCC patients with PVTT confined to the first or second branch of the main portal vein compared with those with PVTT extending into the main portal vein (59). Le Treut et al observed that liver resection in HCC patients with PVTT, compared with patients without PVTT, could provide significant survival with comparable morbidity in selected cases, that is, tumor thrombus located in the portal vein only and expected complete resection of the lesions (60). Shi et al demonstrated that liver resection with portal thrombectomy for HCC cases with PVTT, compared with resection without portal thrombectomy, is justified in selected patients with PVTT located in the segmental or sectoral branches of the portal vein (61). Zhang et al suggested that percutaneous transhepatic portal vein stenting (PTPVS)-TACE combined with three-dimensional conformal radiotherapy (3-DCRT), compared with PTPVS-TACE without 3-DCRT, is possibly an effective treatment modality for HCC with PVTT (63). Kim et al noticed that radiotherapy induced a 45.8% objective response rate for PVTT in patients with HCC, and considered that radiotherapy may be a treatment option for PVTT in patients with HCC (62).

In the present study, the median DFS in patients with PVTT who underwent resection was 5.2 months (95% CI, 2.0–8.4 months) in the sorafenib group and 1.8 months (95% CI, 0–3.7 months) in the BSC group (P=0.002). However, considering that only 5–40% of patients with HCC are receptive to resection (3), further studies should be performed to examine the impact of sorafenib treatment in combination with TACE, radiotherapy, PEI and laser ablation.

In the present study, no patient experienced an adverse event that resulted in discontinuation of sorafenib therapy. This is consistent with a previous study by Wang et al (25), in which only grades 1 and 2 drug-related adverse events were observed with sorafenib therapy, as well as other studies that observed no serious adverse events associated with sorafenib therapy in patients with advanced HCC (48,64). By contrast, grade 4 liver-related adverse events have been reported for Child-Pugh class B patients (65), with poorer outcomes observed in class B versus class A advanced HCC patients following sorafenib therapy (66). All the patients in the present study were classified as Child-Pugh class A, which may explain why sorafenib was well-tolerated.

Access to sorafenib can significantly alter the survival of patients with advanced HCC. This was demonstrated in a recent study from the UK, in which the median overall survival time in advanced HCC patients with access to sorafenib was 9.5 months compared with the 4.1 months observed in patients for whom sorafenib was unavailable due to a lack of approval by a local health funding body (67). This applies to patients in Asian-Pacific countries, where sorafenib is currently not widely available.

The major limitations of the present study include the relatively small number of patients, the short-term follow-up period and its retrospective nature. Another limitation is that only HCC patients with a high risk of recurrence were included; therefore, the role of sorafenib in patients with a low risk of recurrence remains to be elucidated. In addition, due to the short follow-up period, only 9 mortalities were observed and the median overall survival time could not be determined. Additionally, as sorafenib is not widely available to Asian-Pacific patients due to differences in clinical practice, it is likely that the group of patients receiving sorafenib financed the treatment themselves. Therefore, the sorafenib group may have had a higher socioeconomic status and may have received a more aggressive treatment protocol compared with the BSC group, representing a study bias. Furthermore, a significantly higher proportion of patients in the sorafenib group received previous treatment compared with those in the BSC group, which may have affected the results.

In conclusion, sorafenib therapy following resection may prolong DFS in patients with advanced HCC. However, further multi-center studies with larger cohorts and longer follow-up times are required to confirm the results of the current study.

Acknowledgements

The present study was supported by grants from the National Natural Science Foundation of China (nos. 81372571 and 81101863), Guangdong Department of Science and Technology Translational Medicine Center (no. 2011A080300002), and the Innovative PhD Project of Sun Yat-Sen University.

Abbreviations

AFP

α-fetoprotein

BSC

best supportive care

confidence interval

DFS

disease-free survival

HCC

hepatocellular carcinoma

PEI

percutaneous ethanol injection

PVTT

portal vein tumor thrombus

SHARP

sorafenib HCC assessment randomized protocol

TACE

transcatheter arterial chemoembolization

References 1

Raza

Sood

Hepatocellular carcinoma review: Current treatment and evidence-based medicine

World J Gastroenterol20411541272014

10.3748/wjg.v20.i15.4115

24764650

Ferlay

Shin

Bray

Forman

Mathers

Parkin

Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008

Int J Cancer127289329172010

10.1002/ijc.25516

21351269

Bruix

Sherman

American Association for the Study of Liver Diseases: Management of hepatocellular carcinoma: An update

Hepatology53102010222011

10.1002/hep.24199

21374666

European Association For The Study Of The Liver; European Organisation For Research And Treatment Of Cancer

EASL-EORTC clinical practice guidelines: Management of hepatocellular carcinoma

J Hepatol569089432012

10.1016/j.jhep.2011.12.001

22424438

Forner

Llovet

Bruix

Hepatocellular carcinoma

Lancet379124512552012

10.1016/S0140-6736(11)61347-0

22353262

Schwartz

Mandeli

Sung

Neoadjuvant and adjuvant therapy for resectable hepatocellular carcinoma: Review of the randomised clinical trials

Lancet Oncol35936032002

10.1016/S1470-2045(02)00873-2

12372721

Lai

Fan

Liu

Wong

Postoperative adjuvant chemotherapy after curative resection of hepatocellular carcinoma: A randomized controlled trial

Arch Surg1331831881998

10.1001/archsurg.133.2.183

9484732

Izumi

Shimizu

Iyobe

Yagi

Matsui

Nonomura

Miyazaki

Postoperative adjuvant hepatic arterial infusion of Lipiodol containing anticancer drugs in patients with hepatocellular carcinoma

Hepatology202953011994

10.1002/hep.1840200205

8045490

Miyake

Takaki

Iwasaki

Yamamoto

Meta-analysis: Interferon-alpha prevents the recurrence after curative treatment of hepatitis C virus-related hepatocellular carcinoma

J Viral Hepat172872922010

10.1111/j.1365-2893.2009.01181.x

19732321

Liu

Chan

Lam

Poon

Fan

Wong

A randomized, controlled trial of postoperative adjuvant interferon therapy after resection of hepatocellular carcinoma

Ann Surg2458318422007

10.1097/01.sla.0000245829.00977.45

17522506

Shen

Hsu

Chen

Cheng

Adjuvant interferon therapy after curative therapy for hepatocellular carcinoma (HCC): A meta-regression approach

J Hepatol528898942010

10.1016/j.jhep.2009.12.041

20395009

Zhong

Liu

Deng

Adjuvant systemic drug therapy and recurrence of hepatocellular carcinoma following curative resection

Drug Discov Ther71641662013

24071579

Minagawa

Makuuchi

Treatment of hepatocellular carcinoma accompanied by portal vein tumor thrombus

World J Gastroenterol12756175672006

10.3748/wjg.v12.i47.7561

17171782

Takizawa

Kakizaki

Sohara

Sato

Takagi

Arai

Katakai

Kojima

Matsuzaki

Mori

Hepatocellular carcinoma with portal vein tumor thrombosis: Clinical characteristics, prognosis and patient survival analysis

Dig Dis Sci52329032952007

10.1007/s10620-007-9808-2

17394062

Okuda

Ohtsuki

Obata

Tomimatsu

Okazaki

Hasegawa

Nakajima

Ohnishi

Natural history of hepatocellular carcinoma and prognosis in relation to treatment. Study of 850 patients

Cancer569189281985

10.1002/1097-0142(19850815)56:4<918::AID-CNCR2820560437>3.0.CO;2-E

2990661

Yeung

Liu

Wong

Fan

Wong

Natural history of untreated nonsurgical hepatocellular carcinoma

Am J Gastroenterol100199520042005

10.1111/j.1572-0241.2005.00229.x

16128944

Lee

Kim

Choi

Chung

Park

Kim

The safety and efficacy of transcatheter arterial chemoembolization in the treatment of patients with hepatocellular carcinoma and main portal vein obstruction. A prospective controlled study

Cancer79208720941997

10.1002/(SICI)1097-0142(19970601)79:11<2087::AID-CNCR5>3.0.CO;2-M

9179054

Chuan-Xing

BaoShan

Yong

PeiJian

XianYi

XiaoNing

Li-Gong

Efficacy of therapy for hepatocellular carcinoma with portal vein tumor thrombus: Chemoembolization and stent combined with iodine-125 seed

Cancer Biol Ther128658712011

10.4161/cbt.12.10.17676

22037354

Wilhelm

Carter

Tang

Wilkie

McNabola

Rong

Chen

Zhang

Vincent

McHugh

BAY 43–9006 exhibits broad spectrum oral antitumor activity and targets the RAF/MEK/ERK pathway and receptor tyrosine kinases involved in tumor progression and angiogenesis

Cancer Res64709971092004

10.1158/0008-5472.CAN-04-1443

15466206

Wilhelm

Carter

Lynch

Lowinger

Dumas

Smith

Schwartz

Simantov

Kelley

Discovery and development of sorafenib: A multikinase inhibitor for treating cancer

Nat Rev Drug Discov58358442006

10.1038/nrd2130

17016424

Wilhelm

Adnane

Newell

Villanueva

Llovet

Lynch

Preclinical overview of sorafenib, a multikinase inhibitor that targets both Raf and VEGF and PDGF receptor tyrosine kinase signaling

Mol Cancer Ther7312931402008

10.1158/1535-7163.MCT-08-0013

18852116

Chang

Adnane

Trail

Levy

Henderson

Xue

Bortolon

Ichetovkin

Chen

McNabola

Sorafenib (BAY 43–9006) inhibits tumor growth and vascularization and induces tumor apoptosis and hypoxia in RCC xenograft models

Cancer Chemother Pharmacol595615742007

10.1007/s00280-006-0393-4

17160391

Llovet

Ricci

Mazzaferro

Hilgard

Gane

Blanc

de Oliveira

Santoro

Raoul

Forner

Sorafenib in advanced hepatocellular carcinoma

N Engl J Med3593783902008

10.1056/NEJMoa0708857

18650514

Cheng

Kang

Chen

Tsao

Qin

Kim

Luo

Feng

Yang

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 Oncol1025342009

10.1016/S1470-2045(08)70285-7

19095497

Wang

Chuang

Lee

Efficacy of sorafenib as adjuvant therapy to prevent early recurrence of hepatocellular carcinoma after curative surgery: A pilot study

Hepatol Res445235312014

10.1111/hepr.12159

23672310

Oken

Creech

Tormey

Horton

Davis

McFadden

Carbone

Toxicity and response criteria of the Eastern Cooperative Oncology Group

Am J Clin Oncol56496551982

10.1097/00000421-198212000-00014

7165009

Bruix

Sherman

Llovet

Beaugrand

Lencioni

Burroughs

Christensen

Pagliaro

Colombo

Rodés

EASL Panel of Experts on HCC: Clinical management of hepatocellular carcinoma. Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver

J Hepatol354214302001

10.1016/S0168-8278(01)00130-1

11592607

Salerno

Merli

Cazzaniga

Valeriano

Rossi

Lovaria

Meregaglia

Nicolini

Lubatti

Riggio

MELD score is better than Child-Pugh score in predicting 3-month survival of patients undergoing transjugular intrahepatic portosystemic shunt

J Hepatol364945002002

10.1016/S0168-8278(01)00309-9

11943420

Kleihues

Sobin

World Health Organization classification of tumors

Cancer8828872000

10.1002/1097-0142(20000615)88:12<2887::AID-CNCR32>3.0.CO;2-F

10870076

Wanless

International Working Party: Terminology of nodular hepatocellular lesions

Hepatology229839931995

10.1002/hep.1840220341

7657307

Sobin

Fleming

TNM Classification of Malignant Tumors, fifth edition (1997). Union Internationale Contre le Cancer and the American Joint Committee on Cancer

Cancer80180318041997

10.1002/(SICI)1097-0142(19971101)80:9<1803::AID-CNCR16>3.0.CO;2-9

9351551

Luo

Peng

Guo

Zhang

Chen

Shi

Hepatic resection versus transarterial lipiodol chemoembolization as the initial treatment for large, multiple and resectable hepatocellular carcinomas: A prospective nonrandomized analysis

Radiology2592862952011

10.1148/radiol.10101072

21330557

Trotti

Colevas

Setser

Rusch

Jaques

Budach

Langer

Murphy

Cumberlin

Coleman

Rubin

CTCAE v3.0: Development of a comprehensive grading system for the adverse effects of cancer treatment

Semin Radiat Oncol131761812003

10.1016/S1053-4296(03)00031-6

12903007

Therasse

Arbuck

Eisenhauer

Wanders

Kaplan

Rubinstein

Verweij

Van Glabbeke

van Oosterom

Christian

Gwyther

New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada

J Natl Cancer Inst922052162000

10.1093/jnci/92.3.205

10655437

Imamura

Matsuyama

Tanaka

Ohkubo

Hasegawa

Miyagawa

Sugawara

Minagawa

Takayama

Kawasaki

Makuuchi

Risk factors contributing to early and late phase intrahepatic recurrence of hepatocellular carcinoma after hepatectomy

J Hepatol382002072003

10.1016/S0168-8278(02)00360-4

12547409

Chen

Hsu

Kuo

Lin

Association between nucleoside analogues and risk of hepatitis B virus-related hepatocellular carcinoma recurrence following liver resection

JAMA308190619142012

10.1001/2012.jama.11975

23162861

Zhong

Gong

Xiang

Peng

Xie

Hepatic resection associated with good survival for selected patients with intermediate and advanced-stage hepatocellular carcinoma

Ann Surg2603293402014

10.1097/SLA.0000000000000236

24096763

Shindoh

Hasegawa

Inoue

Ishizawa

Nagata

Aoki

Sakamoto

Sugawara

Makuuchi

Kokudo

Risk factors of post-operative recurrence and adequate surgical approach to improve long-term outcomes of hepatocellular carcinoma

HPB (Oxford)1531392013

10.1111/j.1477-2574.2012.00552.x

23216777

Tanaka

Shimada

Kudo

Characteristics of long-term survivors following sorafenib treatment for advanced hepatocellular carcinoma: Report of a workshop at the 50th annual meeting of the liver cancer study group of Japan

Oncology87(Suppl 1)1041092014

10.1159/000368153

25427741

Nagano

Miyamoto

Wada

Ota

Marubashi

Takeda

Dono

Umeshita

Sakon

Monden

Interferon-alpha and 5-fluorouracil combination therapy after palliative hepatic resection in patients with advanced hepatocellular carcinoma, portal venous tumor thrombus in the major trunk and multiple nodules

Cancer110249325012007

10.1002/cncr.23033

17941012

Dollinger

Lautenschlaeger

Lesske

Tannapfel

Wagner

Schoppmeyer

Nehls

Welker

Wiest

Fleig

AIO Hepatobiliary Study Group: Thymostimulin versus placebo for palliative treatment of locally advanced or metastasised hepatocellular carcinoma: A phase III clinical trial

BMC Cancer104572010

10.1186/1471-2407-10-457

20735834

Shiozawa

Watanabe

Ikehara

Matsukiyo

Kogame

Kanayama

Matsui

Kikuchi

Ishii

Igarashi

Sumino

Sustained complete response of hepatocellular carcinoma with portal vein tumor thrombus following discontinuation of sorafenib: A case report

Oncol Lett750522014

24348819

Wang

Zhou

Fan

Qiu

Huang

Tang

Effect of rapamycin alone and in combination with sorafenib in an orthotopic model of human hepatocellular carcinoma

Clin Cancer Res14512451302008

10.1158/1078-0432.CCR-07-4774

18698030

Richly

Schultheis

Adamietz

Kupsch

Grubert

Hilger

Ludwig

Brendel

Christensen

Strumberg

Combination of sorafenib and doxorubicin in patients with advanced hepatocellular carcinoma: Results from a phase I extension trial

Eur J Cancer455795872009

10.1016/j.ejca.2008.10.039

19101137

Chung

Han

Yoon

Yang

Wang

Shao

Kim

Lee

Chao

Interim analysis of START: Study in Asia of the combination of TACE (transcatheter arterial chemoembolization) with sorafenib in patients with hepatocellular carcinoma trial

Int J Cancer132244824582013

10.1002/ijc.27925

23129123

Barbier

Muscari

Le Guellec

Pariente

Otal

Suc

Liver resection after downstaging hepatocellular carcinoma with sorafenib

Int J Hepatol20117910132011

10.4061/2011/791013

22135750

Azad

Posadas

Kwitkowski

Steinberg

Jain

Annunziata

Minasian

Sarosy

Kotz

Premkumar

Combination targeted therapy with sorafenib and bevacizumab results in enhanced toxicity and antitumor activity

J Clin Oncol26370937142008

10.1200/JCO.2007.10.8332

18669456

Sposito

Mariani

Germini

Reyes

M Flores

Bongini

Grossi

Bhoori

Mazzaferro

Comparative efficacy of sorafenib versus best supportive care in recurrent hepatocellular carcinoma after liver transplantation: A case-control study

J Hepatol5959662013

10.1016/j.jhep.2013.02.026

23500153

Feng

Wang

Deng

Yang

Guan

Yao

Zhu

Qin

Wang

Sorafenib suppresses postsurgical recurrence and metastasis of hepatocellular carcinoma in an orthotopic mouse model

Hepatology534834922011

10.1002/hep.24075

21274870

Wang

Qiu

Huang

Dai

Tan

Zhou

Fan

An investigation of the effect of sorafenib on tumor growth and recurrence after liver cancer resection in nude mice independent of phosphorylated extracellular signal-regulated kinase levels

Expert Opin Investig Drugs20103910452011

10.1517/13543784.2011.588598

21671804

Bruix

Takayama

Mazzaferro

STORM: A phase III randomized, double-blind, placebo-controlled trial of adjuvant sorafenib after resection or ablation to prevent recurrence of hepatocellular carcinoma (HCC)

J Clin Oncol32(suppl 5): Abtract 40062014

Inghilesi

Gallori

Antonuzzo

Forte

Tomcikova

Arena

Colagrande

Pradella

Fani

Gianni

Predictors of survival in patients with established cirrhosis and hepatocellular carcinoma treated with sorafenib

World J Gastroenterol207867942014

10.3748/wjg.v20.i3.786

24574751

Wörns

Koch

Niederle

Marquardt

NguyenTat

Gamstätter

Schuchmann

Schulze-Bergkamen

Galle

Weinmann

The impact of patient and tumour baseline characteristics on the overall survival of patients with advanced hepatocellular carcinoma treated with sorafenib

Dig Liver Dis454084132013

10.1016/j.dld.2012.10.010

23182599

Ikeda

Okusaka

Furuse

Mitsunaga

Ueno

Yamaura

Inaba

Takeuchi

Satake

Arai

A multi-institutional phase II trial of hepatic arterial infusion chemotherapy with cisplatin for advanced hepatocellular carcinoma with portal vein tumor thrombosis

Cancer Chemother Pharmacol724634702013

10.1007/s00280-013-2222-x

23812005

Kalva

Pectasides

Liu

Rachamreddy

Surakanti

Yeddula

Ganguli

Wicky

Blaszkowsky

Zhu

Safety and effectiveness of chemoembolization with drug-eluting beads for advanced-stage hepatocellular carcinoma

Cardiovasc Intervent Radiol373813872014

10.1007/s00270-013-0654-7

23754191

Fan

Tang

Zhou

Qiu

Zhou

Multimodality treatment in hepatocellular carcinoma patients with tumor thrombi in portal vein

World J Gastroenterol728322001

10.3748/wjg.v7.i1.28

11819728

Tazawa

Maeda

Sakai

Yamane

Ohbayashi

Kakinuma

Miyasaka

Nagayama

Enomoto

Sato

Radiation therapy in combination with transcatheter arterial chemoembolization for hepatocellular carcinoma with extensive portal vein involvement

J Gastroenterol Hepatol166606652001

10.1046/j.1440-1746.2001.02496.x

11422619

Peng

Guo

Zhang

Lin

Chen

Lau

Hepatic resection versus transcatheter arterial chemoembolization for the treatment of hepatocellular carcinoma with portal vein tumor thrombus

Cancer118472547362012

10.1002/cncr.26561

22359112

Chen

Qiu

Zhang

Huang

Chen

Zhang

Effects of location and extension of portal vein tumor thrombus on long-term outcomes of surgical treatment for hepatocellular carcinoma

Ann Surg Oncol139409462006

10.1245/ASO.2006.08.007

16788755

Le Treut

Hardwigsen

Ananian

Saisse

Gregoire

Richa

Campan

Resection of hepatocellular carcinoma with tumor thrombus in the major vasculature. A European case-control series

J Gastrointest Surg108558622006

10.1016/j.gassur.2005.12.011

16769542

Shi

Lai

Guo

Xue

Lau

Cheng

Surgical treatment of hepatocellular carcinoma with portal vein tumor thrombus

Ann Surg Oncol17207320802010

10.1245/s10434-010-0940-4

20131013

Kim

Park

Lim

Lee

Yoo

Paik

Kho

Kim

Ahn

Huh

Three-dimensional conformal radiotherapy for portal vein thrombosis of hepatocellular carcinoma

Cancer103241924262005

10.1002/cncr.21043

15822130

Zhang

Wang

Yan

Qian

Zeng

Hepatocellular carcinoma with main portal vein tumor thrombus: Treatment with 3-dimensional conformal radiotherapy after portal vein stenting and transarterial chemoembolization

Cancer115124512522009

10.1002/cncr.24139

19156918

Berk

Kaplan

Tonyali

Buyukberber

Balakan

Ozkan

Demirci

Ozturk

Bilici

Tastekin

Efficiency and side effects of sorafenib therapy for advanced hepatocellular carcinoma: A retrospective study by the anatolian society of medical oncology

Asian Pac J Cancer Prev14736773692013

10.7314/APJCP.2013.14.12.7367

24460304

Zugazagoitia

Manzano

Sastre

Ladero

Puente

Díaz-Rubio

Sorafenib for non-selected patient population with advanced hepatocellular carcinoma: Efficacy and safety data according to liver function

Clin Transl Oncol151461532013

10.1007/s12094-012-0902-3

22875650

Pressiani

Boni

Rimassa

Labianca

Fagiuoli

Salvagni

Ferrari

Cortesi

Porta

Mucciarini

Sorafenib in patients with Child-Pugh class A and B advanced hepatocellular carcinoma: A prospective feasibility analysis

Ann Oncol244064112013

10.1093/annonc/mds343

23041587

Palmer

Hussain

Smith

Hargreaves

Hull

Johnson

Ross

Sorafenib for advanced hepatocellular carcinoma (HCC): Impact of rationing in the United Kingdom

Br J Cancer1098888902013

10.1038/bjc.2013.410

23880824

Figure 1.

Kaplan-Meier curves of DFS rate by postoperative treatment. DFS time was significantly different between the sorafenib and BSC groups (log-rank test, P=0.001). DFS, disease-free survival; BSC, best supportive care.

Figure 2.

Kaplan-Meier curves of DFS rate for patients with and without previous treatment prior to resection. DFS time was significantly different between the two groups (log-rank test, P=0.006). DFS, disease-free survival.

Table I.

Summary of the proportion of patients according to the risk factor selection criteria.

Risk factor	Total patients (n=42)	Sorafenib group (n=14)	BSC group (n=28)	P-value
PVTT	31 (73.8)	11 (78.6)	20 (71.4)	0.723
Tumor involving adjacent organs	6 (14.3)	2 (14.3)	4 (14.3)	1.000
Tumor rupture	5 (11.9)	1 (7.1)	4 (14.3)	0.650

Data are presented as n (%). Differences between groups were compared using the Fisher's exact test. There was no significant difference between the groups. BSC, best supportive care; PVTT, portal vein tumor thrombus.

Table II.

Demographic data of the patients by group.

Patient characteristic	All patients (n=42)	Sorafenib group (n=14)	BSC group (n=28)	P-value
Age, years	48.1±10.8	47.4±10.6	48.4±11.0	0.788
Gender				0.313
Male	37 (88.1)	11 (78.6)	26 (92.9)
Female	5 (11.9)	3 (21.4)	2 (7.1)
ECOG PS				1.000
0–1	41 (97.6)	14 (100.0)	27 (96.4)
2	1 (2.4)	0 (0.0)	1 (3.6)
Hepatitis serology				0.100
HBsAg-positive	38 (90.5)	11 (78.6)	27 (96.4)
HBsAg-negative	4 (9.5)	3 (21.4)	1 (3.6)
Child-Pugh score				1.000
5	38 (90.5)	13 (92.9)	25 (89.3)
6	4 (9.5)	1 (7.1)	3 (10.7)
Histological grade				0.326
Well/moderate	21 (50.0)	9 (64.3)	12 (42.9)
Poor	21 (50.0)	5 (35.7)	16 (57.1)
BCLC stage				1.000
A	4 (9.5)	1 (7.1)	3 (10.7)
B	1 (2.4)	0 (0.0)	1 (3.6)
C	37 (88.1)	13 (92.9)	24 (85.7)
Tumor size, cm				0.738
<10	28 (66.7)	10 (71.4)	18 (64.3)
≥10	14 (33.3)	4 (28.6)	10 (35.7)
Tumor nodules				0.259
Single	32 (76.2)	9 (64.3)	23 (82.1)
Multiple	10 (23.8)	5 (35.7)	5 (17.9)
AFP, ng/ml				1.000
<400	20 (47.6)	7 (50.0)	13 (46.4)
≥400	22 (52.4)	7 (50.0)	15 (53.6)
Previous treatment				0.003^a
Yes	7 (16.7)	6 (42.9)	1 (3.6)
No	35 (83.3)	8 (57.1)	27 (96.4)
Anti-virus treatment				0.261
Yes	16 (38.1)	7 (50.0)	9 (32.1)
No	26 (61.9)	7 (50.0)	19 (67.9)
Baseline liver function, U/l
ALT	42.7 (14.8–153.6)	43.4 (14.8–120.5)	42.5 (18.2–153.6)	0.644
AST	44.3 (18.6–412.3)	46.5 (19.7–412.3)	42.4 (18.6–295.7)	0.722
ALP	101.9 (44.4–197.2)	75.1 (44.4–185.5)	112.9 (55.7–197.2)	0.007^a
Albumin, g/l	40.4±4.0	42.5±4.3	39.4±3.4	0.016^a
Total bilirubin, µmol/l	15.6±6.2	16.2±4.6	15.2±6.9	0.622
Prothrombin time, sec	11.9 (10.6–16.4)	11.8 (10.6–16.4)	12 (10.9–15.5)	0.390
INR	1.07 (0.94–1.41)	1.1 (0.94–1.41)	1.1 (0.96–1.34)	0.334

Categorical data are expressed as n (%). Continuous data are expressed as mean ± standard deviation or median (range) by group. Differences between groups were compared using the Pearson χ² test or Fisher's exact test for categorical data and two-sample t-tests for continuous data. For continuous data that did not follow normal distribution, the Mann-Whitney U test was used. Previous treatment means TACE and radiation therapy (i.e., treatments other than sorafenib).

P<0.05 indicates significant difference between groups. BSC, best supportive care; ECOG, Eastern Cooperative Oncology Group Performance Status; HBsAg, hepatitis B virus surface antigen; BCLC, Barcelona Clinic Liver Cancer Stage; AFP, α-fetoprotein; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALP, alkaline phosphatase; INR, international normalized ratio.

Table III.

Univariate and multivariate Cox regression analysis of clinical factors associated with DFS in patients with hepatocellular cancer (n=42).

		Univariate analysis		Multivariate analysis

Clinical factor	Median DFS (95% CI)	HR (95% CI)	P-value	HR (95% CI)	P-value
Sorafenib
Yes	5.2 (1.2–9.2)	0.243 (0.098–0.599)	0.002^b	0.328 (0.122–0.880)	0.027^b
No	1.8 (0.6–3.0)	Reference		Reference
Age, years	3.0 (2.3–3.6)	0.980 (0.947–1.013)	0.232	–
Gender
Male	2.6 (1.3–3.7)	2.548 (0.761–8.525)	0.129	–
Female	9.7 (0.9–18.4)^a	Reference
ECOG PS
2	3.6 (ND-ND)	1.174 (0.159–8.671)	0.875	–
0–1	3.0 (2.3–3.6)	Reference
Hepatitis serology
HBsAg-positive	2.5 (1.8–3.3)	1.440 (0.440–4.713)	0.547	1.139 (0.328–3.957)	0.837
Other	3 (0.8–5.1)	Reference		Reference
Child-Pugh score
6	1.6 (0.0–4.3)	0.905 (0.275–2.975)	0.870	–
5	3.0 (2.3–3.6)	Reference
Histological grade
Poor	2.5 (1.1–3.8)	1.564 (0.804–3.041)	0.187	1.005 (0.440–2.293)	0.991
Well/moderate	3.4 (1.9–4.8)	Reference		Reference
BCLC stage
C	3.0 (2.3–3.6)	1.752 (0.526–5.838)	0.361	–
B	0.9 (ND-ND)	NA
A	1.6 (0.0–5.2)	Reference
Tumor size, cm
≥10	3.2 (2.7–3.6)	0.644 (0.322–1.286)	0.212	0.449 (0.220–0.917)	0.028^b
<10	1.8 (0.6–3.0)	Reference		Reference
Tumor nodules
Multiple	3.0 (1.0–4.9)	0.936 (0.451–1.944)	0.859	0.644 (0.289–1.437)	0.282
Single	2.5 (1.7–3.4)	Reference		Reference
AFP, ng/ml
≥400	2.5 (1.9–3.2)	1.180 (0.620–2.248)	0.614	1.049 (0.511–2.153)	0.896
<400	3.0 (0.0.2–5.8)	Reference		Reference
Previous treatment
Yes	9.2 (0.0–23.7)	0.204 (0.059–0.703)	0.012^b	0.299 (0.073–1.221)	0.093
No	2.5 (1.2–3.9)	Reference		Reference
Anti-virus treatment
Yes	3.1 (2.7–3.5)	0.596 (0.300–1.183)	0.139	0.851 (0.402–1.800)	0.673
No	2.4 (0.6–4.1)	Reference		Reference

Mean (95% CI).

P<0.05 indicates significant association. DFS, disease-free survival; HR, hazard ratio; 95% CI, 95% confidence interval; ECOG, Eastern Cooperative Oncology Group Performance Status; HBsAg, hepatitis B virus surface antigen; BCLC, Barcelona Clinic Liver Cancer Stage; AFP, α-fetoprotein; ND, not derived (due to small sample size); NA, not assessed (due to only one subject in BCLC stage B).