This is a retrospective observational study of 108 patients in a prospectively collected cohort with HBV-related HCC who underwent TACE at the Catholic university of Korea Incheon St. Mary’s hospital between January 2007 and January 2013 (Fig. 1). Primary HCC was diagnosed in 561 patients in our institution. Patients who met the following criteria were entered into the study: age ≥ 18 years; hepatitis B surface antigen positive for at least 6 months; Child-Pugh Class A or B status; Barcelona Clinic Liver Cancer stage 0, A, or B HCC; treated with TACE; and antiviral therapy using entecavir 0.5 mg or no antiviral therapy. Patients were excluded for the following reasons: the presence of other liver diseases, including chronic hepatitis C, alcoholic liver disease, or autoimmune hepatitis (n=199); Child-Pugh class C status (n=43); BCLC stage C or D (n=137); other treatment for HCC including surgery, radiofrequency ablation, percutaneous ethanol injection, liver transplantation, or radiotherapy (n=50); mutation of HBV (n=3); or other antiviral therapy (n=21). Therefore, the study cohort consisted of 108 patients treated with TACE for HBV-related BCLC stage 0-B HCC, of whom 30 were given entecavir 0.5 mg prior to TACE. The diagnosis of HCC was done on the basis of the AASLD practice guidelines [3]. The initiation of antiviral therapy was generally based on the Korean Association for the Study of the Liver guidelines [15] and the Korean national insurance policy: HBV DNA≥2,000 IU/mL and serum aminotransferase level above upper limit normal. We regarded at least 6 months as a significant duration of preemptive antiviral treatment before diagnosis of HCC. This study was approved by the local ethical committee and Institutional Review Board of the Catholic University of Korea.

TACE was performed by injection of doxorubicin hydrochloride (50 mg/session) and iodized oil (Lipiodol Ultra Fluid; Guerbet, France). After the first TACE session, follow-up TACE sessions were performed every 2 months until CT scans suggested complete tumor remission and TACE showed no more hypervascular staining of the tumor. After TACE, liver and renal function tests, including albumin, total bilirubin, aspartate transaminase, alanine transaminase, prothrombin time-international normalized ratio, and creatinine levels were monitored at least every 3 days. CT scans were performed every 2 months to evaluate treatment response and progression of liver cirrhosis.

The primary endpoint was development of treatment-related deterioration of hepatic function after TACE for HBV-related HCC. Secondary endpoints included development of HBV reactivation, hepatitis, severe hepatitis, or ≥grade 3 common toxicity criteria of hepatic measures, including alanine aminotransferase, aspartate aminotransferase, and blood bilirubin. Acute deterioration of hepatic function was defined as newly developed encephalopathy, ascites, variceal bleeding, hepatorenal syndrome, a bilirubin level more than 2.5 times the upper normal limit, prolongation of prothrombin time by more than 3 seconds, or an elevation of the Child-Pugh score (≥2) within 2 weeks following TACE [10,16,17]. HBV reactivation was defined as a greater than 10-fold increase in serum HBV DNA from nadir during antiviral therapy. Hepatitis and severe hepatitis were defined as AST or ALT scores that were more than 2 times and 5 times the upper normal limits, respectively. Hepatic common toxicity criteria were based upon the common terminology criteria (CTC) for adverse events version 4.0 [18].

Categorical variables are described as frequencies and percentages. Continuous variables are presented as means ± standard deviations (SD) and medians with interquartile ranges for parametric and non-parametric variables, respectively. Baseline characteristics were analyzed using the χ² test for categorical variables and Student’s t test for continuous variables. Cox proportional hazards models were used to assess the risk factors for treatment-related decompensation after TACE. The variables used for multivariate analysis were selected on the basis of statistical significance in the univariate analysis (P<0.10). Multivariate analysis was performed using a forward conditional stepwise procedure to exclude confounding variables. Kaplan-Meier curves and a log-rank test were used to estimate the cumulative incidence of treatment-related decompensation. A P-value less than 0.05 on a two-tailed test was regarded as statistically significant. Statistical analyses were carried out using the Statistical Package for the Social Sciences version 16.0 for Windows (SPSS, Inc., Chicago, IL, USA).

The clinical characteristics of the patients at the time of diagnosis of HCC are shown in Table 1. The study population was comprised of 108 patients who met the eligibility criteria, 30 of whom received antiviral therapy (preemptive group; entecavir 0.5mg) and 78 of whom did not receive antiviral therapy (non-preemptive group) prior to TACE. The mean age of the 108 patients was 57.6 years (SD: 9.6 years), and 84 (77.8%) were male. In baseline liver function, 84 patients (77.8%) were Child-Pugh class A and 24 patients (22.2%) were class B. Regarding tumor characteristics, 23 patients (21.3%) were BCLC stage 0, 45 patients (41.7%) were stage A, and 40 patients (37.0%) were stage B. There were no significant differences in age, gender, HBeAg positivity, HBV DNA levels, total bilirubin, ALT, creatinine, α-fetoprotein, total tumor size, Child-Pugh class, and BCLC stage of HCC between the two groups. However, patients in the preemptive group had significantly lower serum albumin levels (3.6 vs. 3.8 mg/dL, P=0.036) and platelet counts (102 vs. 136 ×10³/mm³, P=0.008), and higher levels of PT INR (1.23 vs. 1.13, P=0.008).

Among all 108 patients, 25 (23.1%) suffered from treatment-related decompensation. The rate of decompensation was significantly lower in the preemptive group than in the non-preemptive group (6.7% vs. 29.5%, P=0.008). HBV reactivation occurred more frequently in the non-preemptive group than in the preemptive group (60.3% vs. 0.0%, P<0.001). Hepatic CTC for adverse events grade 3 or 4, hepatitis, and severe-grade hepatitis following TACE developed in 35 (32.4%), 59 (54.6%), and 27 (25.0%) patients, respectively. However, there were no significant differences in the hepatic CTC for adverse events grade 3 or 4 (P=0.245), hepatitis (P=0.317), and severe-grade hepatitis (P=0.485) following TACE between the preemptive group and the non-preemptive group. Although the results revealed no significant statistical differences, the incidences of hepatic adverse events, hepatitis, and severe-grade hepatitis were lower in the preemptive group than in the non-preemptive group (Fig. 2).

Treatment-related decompensation was observed in 25 patients (23.1%); 3 patients developed ascites, 1 developed hepatic encephalopathy, 7 demonstrated an increase in bilirubin levels, and 21 demonstrated an increase in Child-Pugh score. Univariate analysis was performed to screen the risk factors for treatment-related decompensation (Table 2). The predictive factors included in the univariate analysis were gender, age, HBeAg positivity, HBV DNA levels, Child-Pugh class, total bilirubin, ALT, albumin, creatinine, platelet counts, PT INR, alpha-fetoprotein (AFP) levels, total tumor size, BCLC stage of HCC, and preemptive antiviral therapy. Univariate analysis revealed that treatment-related decompensation was significantly associated with higher levels of total bilirubin (≥0.8 mg/dL, P=0.003), lower levels of albumin (≤3.8 g/dL, P=0.003), and non-antiviral therapy (P=0.040). In multivariate analysis, higher levels of total bilirubin (≥0.8 mg/dL; hazard ratio (HR), 3.425; 95% CI, 1.294-9.066; P=0.013), lower levels of albumin (≤3.8 g/dL; HR, 3.990; 95% CI, 1.307-12.182; P=0.015), and non-antiviral therapy (HR 7.597; 95% CI, 1.768-32.645; P=0.006) were the statistically significant factors predictive of treatment-related decompensation (Table 3). The cumulative incidence of treatment-related decompensation was significantly higher for patients with higher levels of total bilirubin (≥0.8 mg/dL, P=0.001) or lower levels of albumin (≤3.8 g/dL, P=0.001) (Fig. 3A, 3B). Patients who did not receive preemptive antiviral therapy had a higher risk of developing treatment-related decompensation than those who were given preemptive antiviral therapy (P=0.024) (Fig. 3C).