This retrospective, single-center study was conducted in a TACE-naïve consecutive cohort treated between January 2011 and August 2016. The study protocol was approved by our institutional review board and was conducted in accordance with local laws and the Declaration of Helsinki. All patients provided written informed consent for the treatment procedures.

Criteria to perform TACE were as follows: i) HCC diagnosis with histopathological confirmation and/or radiological examination based on the European Association for the Study of the Liver imaging criteria (3); ii) tumor location and extent not amenable to elective curative approach (resection and ablation); iii) no previous TACE or systemic chemotherapy; iv) absence of macrovascular invasion and of any suspicion of extrahepatic tumor spread; v) Child-Pugh classes A or B; vi) Eastern Cooperative Oncology Group (ECOG) performance status scores of 0 to 1; vii) adequate renal function (serum creatinine <1.5 times the upper limit of the normal range).

We administered C-TACE to consecutive patients between January 2011 and April 2014; after SAP was approved, we administered SAP-TACE to consecutive patients between May 2014 and August 2016, and we did not change the embolics choice, for the repeated use, in any case.

Angiography was performed using the Seldinger technique via the femoral artery. The femoral artery was catheterized under local anesthesia, and a 4-F catheter (Selecon; Terumo) was inserted into the hepatic artery. Then, a 1.9-F microcatheter (Progreat Σ; Terumo) was advanced into the feeder arteries of each tumor.

In the C-TACE group, a mixture of 5 ml of iodized oil (Lipiodol; Guerbet Japan) and 50 mg of epirubicin hydrochloride (Farmorubicin; Pfizer Japan Inc.) were injected to a maximum administered dose of 50 mg, followed by embolizing with absorbable gelatin sponge particles (Gelpart; Nippon Kayaku).

In the SAP-TACE group, 50-100 µm HepaSphere microspheres were prepared as previously reported (12); 25-mg vials of embolics were preloaded with 25-30 mg of epirubicin dissolved in 5 ml of non-ionic contrast medium and were left unperturbed for 20 min to allow the embolics to expand and absorb the epirubicin after injecting the solution into the vacuum-sealed vials containing the embolics. The SAP injection was administered until near stasis (the contrast column cleared within 2-5 heartbeats) (23).

Liver function was evaluated at just before first TACE, after 3 cycles of TACE, and after 5 cycles of TACE by Child-Pugh class and albumin-bilirubin (ALBI) scores (24). Quantitative differences between the groups were analyzed using the Mann-Whitney U test, and categorical differences were analyzed using Fisher's exact or χ² test. Survival time was calculated from the moment of the initial TACE. Survival curves were created using the Kaplan-Meier method and were compared using the log-rank test. Prognostic factors related to overall survival were identified by univariate and multivariate analyses. Multivariate analyses were performed using the Cox proportional hazards model to identify the independent prognostic factors. Bonferroni adjustment was used to correct for multiple comparisons. The level of significance was set up to P<0.05. Statistical analyses were performed using SPSS version 25 software for Windows (IBM; SPSS, Inc.).

Baseline characteristics of the patients are shown in Table I. During 2011-2016, a total of 155 consecutive patients underwent TACE, including 71 patients treated with C-TACE and 84 patients treated with SAP-TACE. The follow-up period, patient age, gender, background liver disease, number of tumors, maximum tumor size, BCLC stage, Child-Pugh scores, and ALBI scores were not statistically different between the two groups. The proportion of patients that received any prior treatment before TACE was significantly different between the groups (P<0.05). There were no significant differences between the two groups in terms of laboratory data, the levels of aspartate aminotransferase, albumin, total bilirubin, prothrombin time, alpha-fetoprotein (AFP), and des-gamma-carboxy prothrombin (DCP).

We evaluated overall survival rates during follow-up for a median of 25 months. At the time of the analysis, 110 patients had died (59 in the C-TACE cohort, and 51 in the SAP-TACE cohort). Treatment-related death was not experienced. Median survival of the C-TACE cohort was 26 months, compared with 28 months for the SAP-TACE cohort (Fig. 1A). The overall survival rates were not significantly different between the two groups (P=0.289). The 1-, 2-, and 3-year overall survival rates were 74, 50 and 35% in the C-TACE cohort and 75, 60 and 39% in the SAP-TACE cohort, respectively. Unilobar tumor-distribution, BCLC stage A, Child-Pugh class A, AFP <400 ng/ml and DCP <1,000 mAU/ml were significant favorable prognostic factors in the univariate analysis. Age <70, Child-Pugh class A, AFP <400 ng/ml and DCP <1,000 mAU/ml were significant favorable prognostic factors in the multivariate Cox proportional hazard model (Table II). Treatment modality, i.e., C-TACE or SAP-TACE, did not affect long-term survival. In a subgroup analysis of patients with Child-Pugh score of 5, the SAP-TACE cohort showed longer survival compared to the C-TACE cohort (P=0.043) (Fig. 1B). Subgroup analysis did not show significant survival differences in patients with Child-Pugh scores of 6, 7 and 8, and with BCLC stages A and B (Fig. 1C-F).

The repeated TACE cycles were three (median) in the C-TACE cohort. In patients receiving C-TACE, median Child-Pugh scores were 5 before C-TACE, 6 after 3 cycles of C-TACE, and 7 after 5 cycles of C-TACE. The Child-Pugh scores after 3 and 5 cycle of C-TACE significantly worsened compared to those before C-TACE (before vs. 3-cycle, P=0.006; before vs. 5-cycle, P=0.012) (left side of Fig. 2). The median ALBI scores were-2.40 before C-TACE, -2.25 after 3 cycles of C-TACE, and -1.94 after 5 cycles of C-TACE. The ALBI scores after 3 and 5 cycle C-TACE significantly worsened compared to those before C-TACE (before vs. 3-cycle, P=0.009; before vs. 5-cycle, P=0.045) (right side of Fig. 2).

The repeated TACE cycles were three (median) in the SAP-TACE cohort. In patients receiving SAP-TACE, Child-Pugh scores were 6 before SAP-TACE, after 3 cycles of SAP-TACE, and after 5 cycles of SAP-TACE. The Child-Pugh scores after 3 and 5 cycles of SAP-TACE did not change compared to those before SAP-TACE (left side of Fig. 3). The median ALBI scores were-2.31 before SAP-TACE, -2.45 after 3 cycles of SAP-TACE, and -2.64 after 5 cycles of SAP-TACE. The ALBI scores after 3 and 5 cycle of SAP-TACE were not statistically different compared to those before SAP-TACE (right side of Fig. 3).

Subsequent treatments are showed in Table III. After cessation of C-TACE, 41 patients (57.7%) received secondary treatment with radiofrequency ablation (RFA) (n=14; 19.7%), surgical resection (n=1; 1.4%), radiotherapy (n=2; 2.8%), hepatic arterial infusion chemotherapy (n=15; 21.1%), systemic chemotherapies (n=14; 19.7%). Thirty patients (42.3%) received no secondary treatment. After cessation of SAP-TACE, 49 patients (58.3%) received secondary treatment with RFA (n=22; 26.2%), surgical resection (n=2; 2.4%), radiotherapy (n=5; 6.0%), hepatic arterial infusion chemotherapy (n=5; 6.0%), systemic chemotherapies (n=29; 34.5%). Thirty-five patients (41.7%) received no secondary treatment. The proportions of patients who received hepatic arterial infusion chemotherapy and systemic chemotherapies after repeated TACE were different between the C-TACE cohort and the SAP-TACE cohort (P=0.005 and 0.004, respectively), although those who received RFA, surgical resection, and radiotherapy were not different between the C-TACE cohort and the SAP-cohort (P=0.341, 0.661 and 0.349, respectively).