Article Info

Contributors: PRTP proposed the study. CAHK performed research and wrote the first draft. CAHK, LCSC, THSC, CTT and PR analyzed the data. All authors contributed to the design and interpretation of the study and to further drafts. CAHK and PRTP are the guarantors.

BACKGROUND: Chemoembolization with doxorubucin-eluting beads (DEB) has been used to treat hepatocellular carcinoma (HCC) since 2007. This study compared the efficacy and survival between transarterial chemoembolization (TACE) with DEB and conventional approach (cTACE) in HCC treatment.

METHODS: This retrospective case-control study compared the overall survival and tumor response of HCC patients to cTACE (n=190) and DEB (n=143) by the reassessment of computed tomography and serum alpha-fetoprotein (AFP). Multivariate analysis was used to determine the factors affecting tumor response.

RESULTS: The median post-treatment to pre-treatment AFP level was 0.8 for a DEB session (n=258) and 1.0 for a cTACE session (n=452), showing a significantly greater decrease in AFP after DEB (P<0.05). More patients in the DEB group achieved objective response (complete and partial) compared with those in the cTACE group (P<0.05). Objective tumor response after DEB vs cTACE was 34.8% vs 15.4% in 0-3 months (P=0.001), 37.1% vs 20.0% in 3-6 months (P<0.05), and 50.0% vs 30.0% in 6-12 months (P=0.093). DEB predicted a 3.604 times odds of achieving at least one objective tumor response in a patient when compared to cTACE (P<0.0001). The median survival from first transcatheter therapy of patients having undergone at least once DEB was 12.53 months, while those having received cTACE only was 10.53 months (P=0.086). A tendency of improved survival appeared to maintain until >80 months after the first TACE session in the DEB group.

Patients with hepatocellular carcinoma (HCC) unsuitable for curative treatments may undergo transarterial chemoembolization (TACE).^{[1, 2]} Conventionally, a cytotoxic drug emulsified with lipiodol is injected into the hepatic artery along with gelfoam or other embolizing particles. Embolization causes tumor ischemia,^[3] while lipiodolized drug is selectively retained by HCC, exerting a loco-regional cytotoxic effect.^{[4, 5]}

However, the efficacy of conventional TACE (cTACE) may be reduced by the fact that some HCC showed no or minimal uptake of lipiodol.^[4] TACE also shares many systemic chemotherapeutic side effects as washout of the cytotoxic drug from the liver occurs.^[6] Doxorubicin-eluting beads (DEB)-TACE has been developed where DEB is an embolizing agent capable to sustain a slow release of doxorubicin to the tumor over a long period of time. Compared to the conventional approach, the peak plasma concentration of the cytotoxic drug is reduced by as much as 90%.^[7] Our group conducted one of the first trials of DEB-TACE in 2007. We have shown that DEB-TACE is a safe and effective palliative treatment for HCC, with a 70% overall tumor response rate and no evidence of doxorubicin related toxicity.^[4] DEB-TACE has improved tumor response, less side effects, and better survival in selected patients.^[8-10]

This case-control clinical study was conducted in Queen Mary Hospital, Hong Kong, China. DEB-TACE and cTACE were the two forms of TACE concerned. The primary endpoint of the study is the tumor response after one DEB or TACE session; AFP and CT were surrogates of assessment. The secondary endpoint is the survival of patients in the DEB group and the cTACE group. The DEB group consisted of patients with at least one DEB-TACE session in the year 2004-2012. The studied group was matched with the control group, cTACE, which consisted of patients who had undergone cTACE but never DEB-TACE, based on the parameters as listed in Table 1. Since TACE may be indicated in both patients with unresectable disease on diagnosis and those with recurrent tumor unsuitable for further resection, patients undergone an operation were not excluded. Also, TACE was occasionally done in adjunct with other treatments such as high-intensity focused ultrasound (HIFU) or radiofrequency ablation (RFA) in the attempt to achieve better tumor control. Other treatments did not constitute exclusion. All ranges of patients are represented in the scope of this study, as patients were included regardless of presence of poor prognostic factors such as the extent of cirrhosis, high Child-Pugh’s grade or portal venous invasion.

The 143 patients in the DEB group and 190 patients in the cTACE group had similar baseline characteristics with regard to age, gender, hepatitis virus carriage, cirrhosis, tumor size and degree of invasion (Table 1). The median of baseline AFP on diagnosis was 77.3 ng/mL (SD=72785.5; range: 1-463 900) for the DEB group and 70.0 ng/mL (SD=58239.9; range: 1-590 900) for the cTACE group. Each patient in the DEB group had received 3.72±3.01 TACE sessions, among which there were 2.19±1.65 DEB-TACE sessions. Each patient in the cTACE group had received 2.97±2.68 TACE sessions. Sixty-five (45.5%) and 12 (6.3%) patients had undergone hepatic surgery in the DEB group and cTACE group, respectively; 40 (28.0%) and 5 (2.6%) patients had received RFA, HIFU, or Yttrium radioembolization for each group, respectively.

TACE with DEB and conventional approach were administered according to standard protocols described previously.^[4] Briefly, the feeding artery to the tumor was identified and superselectively catheterized with the aid of diagnostic angiography. DEB mixed with contrast agent, of a usual dosage of 40-60 mg/m² with a maximum dosage of 150 mg, was injected in DEB-TACE, while an emulsion of maximum 60 mL, prepared by mixing lipiodol with a chemotherapeutic agent, usually cisplatin (0.5 mg/mL) was injected in cTACE.

AFP is a useful tumor marker for monitoring of HCC after treatments.^[11] With a half life of about 5-6 days, AFP is expected to show a trend of decline after a successful intervention.^[12-15] Other objective measurements of liver function such as total bilirubin and alanine aminotransferase levels after DEB-TACE had been evaluated previously, which were not useful indicators reflecting tumor response.^[16] In this study, the pre-treatment and post-treatment AFP levels were compared between the two groups after one session of procedure. Although the groups may differ in some parameters such as the number of patients having operations performed, such that the baseline AFP before treatment of each patient may be different, the main focus was the percentage change of AFP after TACE was administered. This change was more likely to be attributed to TACE and was less likely affected by previous interventions such as operations or RFA. We also analyzed the CT scans for these patients where the images and reports permitted. The emphasis was on the tumor response after a TACE session.

For survival analysis, on the other hand, it was expected that the survival of patients who had also undergone surgery, RFA, HIFU, or Yttrium radioembolization would be significantly different from those who had undergone TACE without these interventions. All these former patients were excluded from both the cTACE and DEB-TACE groups in analyzing survival. Use of sorafenib also constituted an exclusion from analyses.

Statistical significance was taken where P<0.05. For baseline characteristics, AFP and CT scan analysis, comparisons among groups were done with Mann-Whitney U test and Student’s t test for continuous variables where appropriate. Chi-square test was performed for categorical variables. For survival analysis, patients alive at the latest follow-up session were censored. Computation was performed according to the Kaplan-Meier method and difference in survival compared with the log-rank test. Statistical analyses were performed with the SPSS package (version 20).

The post-treatment AFP taken 7-90 days after a DEB-TACE or cTACE session was analyzed and was expected to reflect the change in tumor load resulting from the intervention (Fig. 1). We first analyzed the change in AFP after each chemoembolization session (Fig. 1A). 258 DEB-TACE sessions and 452 cTACE sessions were studied. For both types of TACE, the pre-treatment AFP was taken at a mean of 6.75 days beforehand. For DEB-TACE, the pre-treatment AFP ranged from 2 to 641 400 ng/mL (median=29.0, mean=16 494.8). For cTACE, the range was from 1 to 567 600 ng/mL (median=51.0, mean=11 504.1). The post-treatment levels analyzed were taken 30.8±19.4 days after a DEB-TACE session and 32.4±18.4 after a cTACE session. The median decrease in AFP level was 20% for each DEB-TACE session, and this was significantly greater than that after each session of cTACE, which is 0% (P<0.001).

We are aware that some centers recommend treatment by only two sessions of chemoembolization and we therefore proceeded to evaluate the response of AFP after the first two consecutive cTACE and DEB sessions (Fig. 1B). We compared the baseline AFP and the post-treatment AFP after the first two consecutive sessions. For cTACE, the time between the two sessions was within a median of 105 days; for DEB was within a median of 100 days. We found that AFP level decreased to 66.0% of the baseline level after two consecutive DEB sessions, while in contrast, AFP level increased to 106.0% of baseline after two consecutive cTACE (P=0.004). Of note, since each of DEB session decreased AFP by about 20.0% (see above), the decrease of AFP following two consecutive DEB sessions appeared to be due to the cumulative effects of each session.

We further analyzed the rate of decrease of AFP throughout the first two consecutive DEB sessions. We found that when the two sessions were performed within 6 months (n=43), the post-treatment AFP level decreased to 53.1% of baseline, greater than the overall decrease to 66.0% (see above). When we attempted to further stratify the time-points into a three-month interval, the analyses were constrained by an inadequate sample size. This was because only 3 patients had undergone consecutive DEB within 3 months, while the remaining patients had consecutive DEB within 3-6 months.

The reassessment CT scan was reported using the following definition: complete response (CR), no tumor was detectable; partial response (PR), tumor load was decreased; progressive disease, tumor load was increased; and stable disease for all the remainder. Objective response was defined as CR and PR. Objective tumor response after DEB-TACE vs cTACE was 34.8% vs 15.4% in 0-3 months (P=0.001), 37.1% vs 20.0% in 3-6 months (P<0.05), and 50.0% vs 30.0% in 6-12 months (P=0.093) (Fig. 2). Therefore, superiority in response after DEB-TACE maintained from 0 to 12 months, and did not appear to decrease with time at least till 12 months.

All patients undergone a hepatic surgery or other forms of interventions including HIFU, RFA, or Yttrium embolizations were excluded from analyses. Survival from first TACE session of the DEB group and the cTACE group was compared. The median survival of the DEB group was 12.53 months, while that of cTACE was 10.53 months (Fig. 3, P=0.086). Tendency of improved survival appeared to maintain until >80 months after the first TACE session in the DEB group.

We attempted to identify factors which may predict increased likelihood to achieve at least one objective response after chemoembolization. When all patients in both cTACE and DEB groups were considered, the use of DEB had 3.604 times the odds of achieving at least one objective response. Increased number of sessions of chemoembolization performed also increased the odds marginally (Table 2). For patients without hepatic surgery, the odds of achieving at least one objective response to DEB is greater (odds ratio=2.352, P=0.026) (Table 3).

DEB-TACE has emerged as a promising alternative to the conventional approach for palliative treatment of HCC. Previous studies had shown improving survival, tumor response and better adverse effects profile.^{[8, 9]} These results were applicable to stringently selected patients who had a better functional status such as Child-Pugh score, or better liver function as reflected by a relatively milder elevation of serum aminotransferases.^{[9, 17]} This was in line with the fact that survival benefits in cTACE were proven in stringently selected patients.^[18] The current study evaluated tumor response for each TACE session by studying objective data such as AFP and CT scans, and evaluated these previous results in the setting of palliative care as we reviewed our clinical data over 9 years. Our data indicated a tendency towards better tumor response and survival with DEB-TACE, considering all HCC patients regardless of their disease status.

A major limitation of this study is that besides DEB-TACE, some patients in our center received other treatments such as surgery, HIFU or RFA. As TACE is primarily a treatment of palliative intent, it is understandable that many patients may opt for different approaches in the hope of improving quality of life. TACE is also a treatment option for patients with recurrent disease despite previous tumor resections. To reflect the roles of TACE, the current study did not limit the subjects to those underwent TACE only. Two of the study endpoints, namely post-treatment to pre-treatment AFP and post-treatment reassessment CT scans, are likely valid objective assessments of tumor response attributable to the TACE session, despite of alternative treatments in the past. In addition, we recognized several limitations caused by data collection, which included the lack of standardized time-points for AFP measurement and that CT reporting of tumor response had not uniformly conformed to the modified RECIST criteria.

By analyzing AFP levels sessions by single session (Fig. 1A) and only in the first 2 consecutive sessions (Fig. 1B), we attempted to minimize the effects of the non-standardized number of chemoembolization sessions offered to each patient. However, we were aware that this was a limitation to this study and that the treatment plan on end-stage HCC was largely influenced by patients’ wish and surgeons’ decisions.

In a phase I/II clinical study which our group conducted, the concentration of plasma doxorubicin was shown to be markedly decreased compared to that known to cTACE (C_max 52.8±41.5 ng/mL [DEB] vs 900±300 ng/mL [cTACE]), with a long plasma doxorubicin half-life of 73.5±22.7 hours. This suggested a sustained loco-regional release of doxorubicin from DEB, and explained the lower systemic toxicities and better loco-regional tumor control with DEB-TACE.^[4] In this study, post-treatment AFP was found to be lower after DEB-TACE, and a greater proportion of patients achieve objective tumor response after DEB-TACE until at least 12 months.

Of note, AFP may be variably elevated in the event of acute liver injury. Liver ischemia resulting from TACE may elevate aminotransferases, and probably may have an effect on AFP.^[19] Elevation of aminotransferases may be similar or less marked after DEB-TACE when compared to the conventional approach.^[9] The lower post-treatment: pre-treatment AFP levels observed in DEB-TACE may be attributed to both improved tumor response and decreased liver toxicity. Comparing DEB and cTACE, since there was a significant decrease of AFP after every DEB session (Fig. 1A), it may be suggested that DEB could be continued even after multiple sessions.

Patients underwent DEB-TACE at least once were shown to have a tendency to improved survival compared to those received cTACE only. Although we reported that DEB-TACE may be complicated by acute pancreatitis in 4.1% of patients, these complications did not appear to lead to worsened hospital mortality.^[20] This was in line with the previous findings that selected patients treated by DEB-TACE had a 2-year survival of 48% compared to 12% for those treated by cTACE. Given this and previous evidence, the survival benefit may indeed be greater than that reported in this study if more DEB-TACE sessions were administered.

The current study provides comprehensive data of HCC patients with varying disease status who received DEB-TACE in the past 9 years. Together with a decreased systemic toxicities as previously reported, such as reduced cardiac and liver toxicities, DEB-TACE is a safe option for end-stage HCC patients and shows a tendency to improved survival and tumor response.

8 Dhanasekaran R, Kooby DA, Staley CA, Kauh JS, Khanna V, Kim HS. Comparison of conventional transarterial chemoembolization (TACE) and chemoembolization with doxorubicin drug eluting beads (DEB) for unresectable hepatocelluar carcinoma (HCC). J Surg Oncol 2010;101:476-480. PMID: 20213741

9 Lammer J, Malagari K, Vogl T, Pilleul F, Denys A, Watkinson A, et al. Prospective randomized study of doxorubicin-eluting-bead embolization in the treatment of hepatocellular carcinoma: results of the PRECISION V study. Cardiovasc Intervent Radiol 2010;33:41-52. PMID: 19908093

14 Bruix J, Sherman M, Llovet JM, Beaugrand M, Lencioni R, Burroughs AK, et al. Clinical management of hepatocellular carcinoma. Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver. J Hepatol 2001;35:421-430. PMID: 11592607

16 Song MJ, Park CH, Kim JD, Kim HY, Bae SH, Choi JY, et al. Drug-eluting bead loaded with doxorubicin versus conventional Lipiodol-based transarterial chemoembolization in the treatment of hepatocellular carcinoma: a case-control study of Asian patients. Eur J Gastroenterol Hepatol 2011;23:521-527. PMID: 21537127

18 Llovet JM, Real MI, Montaña X, Planas R, Coll S, Aponte J, et al. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial. Lancet 2002;359:1734-1739. PMID: 12049862