The study protocol conformed to the ethical guidelines of the World Medical Association Declaration of Helsinki and was approved by the Institutional Review Board (IRB no. 2020-L09-01). The requirement for informed consent from patients was waived due to the retrospective nature of this study.

This is a pilot non-randomized trial through a retrospective chart review in a single center. Consecutive patients who underwent at least one session of DEB-TACE or RFA for first treatment of single small HCC from October 2010 to March 2017 and who met the following eligibility criteria were included. Eligibility criteria were as follows: 1) newly diagnosed, treatment-naïve HCC, 2) Child-Pugh class A or B, 3) Eastern Cooperative Oncology Group (ECOG) performance status 0,1 or 2, and 4) absence of main portal vein thrombosis and extrahepatic metastases.7 Patients at high risk for surgery or those who were unwilling to receive operation in spite of the clinician’s advice were included. We excluded patients with previous histories of intra- or extrahepatic malignancies (n=3), poor hepatic reservoir function of Child-Pugh class C (n=11), medically uncontrolled or recurrent hepatic encephalopathy (n=4), and poor performance status of ECOG 3 or higher (n=7). As a result, a total of 40 patients were finally included in this study. The Strengthening the Reporting of Observational studies in Epidemiology (STROBE) reporting guidelines were followed (Supplementary Table 1).

Diagnosis of a small (≤3 cm in greatest dimension) solitary HCC was made based upon the typical radiological findings (arterial hypervascularity and venous/late phase washout) on dynamic computed tomography (CT). In all patients, a radiological diagnosis of HCC was ascertained by gadoxetic acid (Primovist^®, Bayer HealthCare, Leverkusen, Germany)-enhanced magnetic resonance imaging (MRI) following dynamic CT.8,9 On gadoxetic acid-enhanced MRI, lesions satisfying at least two findings within the following criteria were considered as HCC nodules: 1) arterial enhancement with or without portal/equilibrium phase washout, 2) hypointensity on hepatobiliary phase, 3) hyperintensity on T2-weighted image, and 4) hyperintensity on diffusion-weighted image.10

One radiologist with 25 years of experience in intervention performed all procedures. After hepatic and superior mesenteric artery angiography to check for arteriovenous shunts, and identification of arterial tumor supply, feeding arteries were selectively catheterized using a microcatheter. After positioning the microcatheter tip close to the tumor-feeding branch, a mixture of 1 vial of DEB (70–150 or 100–300 μm diameter; DC Bead^®, Biocompatibles, Farnham, UK) loaded with doxorubicin (50 or 75 mg) and contrast agent was infused until the flow through the tumor-feeding artery was slowed down.

Under conscious sedation and local anesthesia, all RFA procedures were performed percutaneously by the aforementioned radiologist. LeVeen needle electrode^® (Boston Scientific, Marlborough, MA, USA) or Well-Point RF electrode^® (STARmed, Goyang, Korea) were used according to the manufacturers’ recommended protocols. Real-time sonographic guidance and monitoring was made for adequate ablation during procedure. In case of invisible HCC at sonographic targeting, CT was applied for adequate needle positioning prior to and post-ablation.

Primary outcomes were tumor response and local and distant intrahepatic recurrences of HCC. Local recurrence of HCC was defined when enhancing tumor reemerged at the target lesion and distant intrahepatic recurrence was defined when the new HCC was developed at a separate intrahepatic site.11 These were evaluated by regular follow-up with CT scan and/or MRI at 1, 3, and 5–6 months after receiving DEB-TACE or RFA, using the modified Response Evaluation Criteria in Solid Tumors.12 Complete response was defined as disappearance of intratumoral enhancement; partial response was defined as a ≥30% decrease in the sum of the diameters of viable lesion; progressive disease was defined as an increase of ≥20% in the sum of diameters of viable tumor; and stable disease was defined as any cases that did not qualify as either partial response of progressive disease. Patients who were diagnosed with HCC recurrence after initial treatment were assigned to sequential therapeutic modalities.

The secondary endpoint was treatment-related complication. Laboratory tests were performed 1, 3–7, 14 days and every 4–12 weeks after treatment. Adverse events were assessed using the Common Terminology Criteria for Adverse Events (version 4.0).13

Results are summarized either as number (%), median (ranges) or mean (95% confidence interval [CI]) as appropriate. Continuous data were analyzed using the Mann-Whitney U test or student’s t-test. Categorical data were evaluated using the Fisher exact test or chi-square test. Insufficient response to first treatment was censored at the time of first evaluation of therapeutic response. Follow-up loss or death without tumor recurrence was censored at the time of the last follow-up. Recurrence rates and curves were obtained using the Kaplan-Meier method and compared using the log-rank test. Multivariate Cox regression was performed to identify independent factors associated with tumor recurrence using variables with P-values less than 0.1 in univariate analysis. A P-value of less than 0.05 (2-tailed) was considered statistically significant. Statistical analyses were performed using MedCalc for Windows, version 19.4 (MedCalc Software, Ostend, Belgium).

The baseline characteristics of the patients are listed in Table 1. Twenty-one patients (52.5%) underwent DEB-TACE and 19 patients underwent RFA (11 ultrasound-guided and 8 CT-guided). The patients included 27 men and 13 women with median age of 67.5 years (range, 43–84). About 87% of the patients had Child-Pugh grade A liver function, and 60% of patients had ECOG performance grade of 0 while the remaining were either grade 1 or 2. The median tumor diameter was 2.0 cm (range, 1.0–3.0). There was no difference between the groups with regard to tumor stage (P =0.936). Although Child-Pugh scores of the two groups were not significantly different, the serum albumin of the RFA group was lower than that of the DEB-TACE group (4.2 vs. 3.5 g/dL, P =0.028) and prothrombin time international normalized ratio of the RFA group was higher than that of DEB-TACE groups (1.10 vs. 1.20, P =0.011).

The target tumor responses evaluated after first treatment are summarized in Table 2. Complete response rates in DEB-TACE and RFA groups were 90.5% and 94.7%, respectively (P =1.000). In DEB-TACE group, partial response and stable disease rates were equally 4.8%. One patient (5.3%) in the RFA group showed progressive disease.

Fig. 1. describes the case of a patient with 2.0 cm-sized solitary enhancing nodule who experienced a complete necrosis on follow-up CT after a single session of DEB-TACE and underlines the absence of local tumor recurrence for 28.6 months.

Overall mean follow-up duration for local recurrence was 87.6 months (95% CI, 74.4–102) and 65.6 months (95% CI, 48.6–82.6) for distant intrahepatic recurrence. Of the 21 patients who underwent DEB-TACE, one patient had partial response and one patient showed stable disease on first follow-up imaging. Local recurrences were noted in another four patients. Among 19 patients who received RFA, one patient showed progressive disease at first response evaluation and another three patients developed local recurrences. The 6- and 12-month cumulative local recurrence rate was 5.0% and 21.8% in DEB-TACE vs. 11.1% and 17.0% in RFA group (Fig. 2A, P =0.877) and reached a plateau in 12 months. Distant intrahepatic recurrences developed in 14 cases during follow-up (7 and 7, respective groups). The 12- and 24-month cumulative distant intrahepatic recurrence rate was 20.6% and 42.7% in DEB-TACE vs. 17.2% and 36.3% in RFA group (Fig. 2B, P=0.844).

On Cox regression analysis with multivariable adjustment, there was no associated factor regarding local recurrence. Distant intrahepatic recurrence was significantly associated with albumin level and tumor size (Table 3). Treatment modality was not an independent risk factor for local or distant intrahepatic recurrences.

As treatment-related adverse events, there were no grade 3–4 laboratory toxicities in either group within 4 weeks and no mortality directly related to the procedures. Development of post-embolization or post-ablation syndrome under control with analgesic or antipyretic drugs was shown in 6 patients (28.6%) in DEB-TACE and 4 patients (21.1%) among RFA group. In RFA group, two patients experienced small pneumothorax (<2 cm rim present between the lung edge and chest wall) which was absorbed with supplemental oxygen within 1–2 days.

Two patients (9.5%) experienced gangrenous cholecystitis requiring cholecystectomy in DEB-TACE group. Fig. 3. illustrates one of these patients who underwent emergent operation and the biopsy resulted in ischemia due to deposition of embolic particles in the vessel. Another patient experienced liver abscess which required percutaneous catheter drainage.