Radiofrequency ablation (RFA) and microwave ablation (MWA) were found to be effective in treating hepatocellular carcinoma (HCC) smaller than 3 cm; however, to the authors' knowledge, the usefulness of thermal ablation in treating larger HCC, especially those >5 cm, has not been well documented. The present study evaluated the therapeutic efficacy of percutaneous thermal ablation with curative intention for HCC measuring between 3.0 cm and 7.0 cm.
Percutaneous RFA or MWA were used to treat 109 HCC patients with at least 1 tumor measuring between 3.0 cm and 7.0 cm. Fifty‒eight patients received thermal ablation as the first treatment, and the remaining 51 were treated for posthepatectomy recurrent HCC. A total of 89 patients had a main tumor measuring 3.0 cm to 5.0 cm, and 20 patients had main tumors measuring 5.0 cm to 7.0 cm. Local therapeutic efficacy, long-term outcome, and prognostic factors were analyzed.
There were no treatment-related deaths, and the major complication rate was 9.2%. Complete ablation rate was 92.6%. Local recurrence (LR) occurred in 22% patients, with a median time to LR of 4.6 months. Distant recurrences developed in 53.2% patients. The 1-year, 3-year, and 5-year survival rates were 75.8%, 30.9%, and 15.4%, respectively. Univariate analysis indicated that incomplete tumor ablation, posthepatectomy recurrence, and preablation α-fetoprotein (AFP) ≥200 ng/mL were 3 unfavorable prognostic factors for long-term survival (P = .000, .015, and .008, respectively). Cox regression analysis confirmed that incomplete tumor ablation, recurrent tumors, and preablation AFP ≥200 ng/mL were independent unfavorable prognostic factors, with an exp(B) of 4.158 (P = .001), 1.568 (P = .082), and 1.593 (P = .082), respectively.
In the past 2 decades, percutaneous local ablative therapies, including percutaneous ethanol injection (PEI), radiofrequency ablation (RFA), and microwave ablation (MWA), have emerged as effective treatments for small hepatocellular carcinoma (HCC) measuring ≤3 cm.1-5 Retrospective studies suggested that they produced long-term survival similar to surgical resection.6-8 At present, these techniques have been used as standard treatment for small HCC in patients who have high surgical risks.
Randomized control trials demonstrated that RFA and MWA have greater tumoricidal efficacy and need fewer treatment sessions to achieve complete necrosis of tumors compared with PEI in treating small HCC.4, 5, 9, 10 Shibata et al compared the therapeutic efficacy of RFA and MWA in a randomized control trial, and found no significant differences in local tumor control and complication rates between RFA and MWA.11 Our previous study demonstrated that both modalities produced not only similar complete ablation rates, complication rates, and local recurrence rates, but also long-term survival.12
With improvements in devices and techniques, RFA and MWA have displayed potential capability for treating HCC measuring larger than 3 cm.12-16 Lin et al16 treated 50 HCC patients with a tumor dimension of up to 4 cm using RFA, and achieved a complete ablation rate of 96% and a 3-year survival rate of 74% in all patients and 62% in patients with tumors measuring 3.1 to 4 cm. Lencioni et al17 reported a complete ablation rate of 98.6% and a 2-year survival rate of 98% in 52 HCC patients with a tumor dimension up to 5 cm using RFA. Recently, encouraging results were reported from prospective randomized control trials, in which RFA or MWA were shown to be as effective as surgical resection in treating HCC measuring ≤5 cm.18, 19 Usefulness of thermal ablation in treating large HCC, especially those measuring >5 cm, however, has not been well documented. The present study evaluated local therapeutic efficacy, long-term outcome, and prognostic factors of percutaneous thermal ablation with curative intention for HCC measuring between 3 cm and 7 cm.
MATERIALS AND METHODS
A total of 109 HCC patients with at least 1 tumor measuring between 3 cm and 7 cm in dimension who were treated at our hospital by ultrasound-guided percutaneous thermal ablation (RFA in 59 patients and MWA in 50 patients) from October 1997 through May 2007 were included in the current study. HCC was diagnosed by 1 of following criteria20: 1) histologic evidence, or 2) detection of hypervascular nodule(s) in a cirrhotic liver by 2 imaging techniques with an elevated α-fetoprotein (AFP) >400 ng/mL.
The inclusion criteria were as follows: 1) aged 18 to 75 years; 2) no more than 3 tumors with at least 1 tumor measuring between 3 cm and 7 cm in maximum dimension; 3) absence of vascular invasion, lymph node involvement, and distant metastases; 4) liver function status at Child-Pugh A or B; 5) platelet count>50 × 109/L; and 6) unsuitable or refused partial hepatectomy. Patients with intractable ascites and/or uncorrectable coagulopathy were excluded from the study.
The demographic data of these patients are shown in Table 1. There were 95 males and 14 females, with an average age of 53 ± 12 years (range, 25‒75 years). Eighty-nine patients (81.7%) had concomitant cirrhosis. With regard to liver function status, 86 had Child-Pugh classification grade A and 23 grade B. A total of 70 patients had a solitary tumor, and 39 had multiple tumors. Of all the patients, 89 had a main tumor measuring 3.0 to 5.0 cm (medium), and 20 had a main tumor measuring 5.0 to 7.0 cm (large). There were no significant differences in demographic features between patients with medium and large tumors (Table 2). With respect to type of tumor (ie, previous history of treatment), 58 of 109 patients received thermal ablation as the first treatment (first-treated tumor), and 51 for recurrent tumor after hepatectomy (recurrent tumor). The median interval between partial hepatectomy and tumor recurrence was 6.9 months (range, 1.0‒117.6 months). Of 109 patients, 59 were treated by RFA and 50 by MWA.
Table 1. Demographic Data
No. of Cases
HBsAg indicates hepatitis B surface antigen; HCV, hepatitis C virus; +, positive; -, negative.
Using the Child-Pugh classification.
Maximum dimension of 1 tumor in patients with a solitary tumor or the largest tumor in patients with 2 or 3 tumors.
The study was approved by the ethics committee of our hospital and written informed consent was obtained from each patient before treatment.
RFA and MWA were performed percutaneously under real-time ultrasound guidance using a Toshiba 240 ultrasound scanner (Toshiba, Tokyo, Japan) or SSD 2000 ultrasound scanner (Aloka Inc, Tokyo, Japan) and a 3.5-megahertz (MHz) puncture probe. Both ablation procedures were performed under local anesthesia with 1% lidocaine plus intravenous fentanyl (Yiyao Groups Ltd., Hubei, China) and Droperidol (Yongkang Pharmaceuticals, Beijing, China), or propofol (Diprivan; Zeneca Pharmaceuticals, Wilmington, DE) and ketamine (the First Pharmaceuticals of Shanghai, Shanghai, China). The aim of the treatments was to destroy the entire tumor with 0.5 to 1 cm of surrounding nontumoral tissue.
RFA was performed using a WE-7568 RF delivery system (Welfare Electronic Co., Beijing, China) or cool-tip RF ablation system (Valleylab, Boulder, Colo). WE-7568 RF delivery system consisted of a RF generator with a frequency of 290 kHz and a maximum power of 200 W, a 14-gauge multi-tined electrode, and 2 dispersive pads. The electrode contained 10 expandable tines inside its shaft with a thermocouple placed at the tip to continuously monitor the local tissue temperature during ablation. At full deployment, the 10 expandable tines gave rise to an umbrella-like structure with a maximum diameter of 4.5 cm. The cool-tip RFA system consisted of an RF generator with a maximum power of 200 watts, a 17-gauge internally cooled electrode, and 2 dispersive pads. Grounding was obtained by attaching 1 dispersive pad to each thigh of the patient.
MWA was undertaken by using the UMC-I MW delivery system (Institute 207 of Aerospace Industry Company and PLA General Hospital, Beijing, China) or FORSEA MW delivery system (Qinghai Microwave Electronic Institute, Nanjing, China). The UMC-I MW delivery system consisted of a MW generator with a frequency of 2450 MHz and a power output of 10 to 80 watts, and a 16-gauge monopolar electrode with 2.7-cm-long antenna, which was used to deliver MW energy into liver tissue. The FORSEA MW delivery system consisted of a MTC-3 MW generator with frequency of 2450 MHz and power output of 10 to 150 watts, and a 14-gauge cooled-shaft electrode. The cooled-shaft electrode contains a 16.5‒cm double-lumen shaft and a 1.5-cm antenna. The chilled saline solution(4°C) is circulated continuously through the double-lumen shaft at 50 to 60 mL/minute to maintain the shaft temperature at 10 ± 2°C by using a BT01-100 steady-flow pump(LanGe Steady Flow Pump Corporation, Baoding, China).
An overlapping ablative technique with multiple electrode insertions was applied to treat tumors. Each tumor usually required 2 to 7 electrode insertions (depending on its size) and multiple energy applications. Each RF energy delivery was set for 10 minutes with tissue temperature achieving 90°C for at least 6 minutes in WE-7568 RF ablation and 12 minutes in cool-tip ablation. Each MW energy application was set at 60 watts for 5 minutes in UMC-I MW ablation and 70 watts for 20 minutes in FORSEA MW ablation. The treatment of all tumors in the patient at 1 time was referred as to 1 treatment cycle. Reapplication of ablation when follow-up computed tomography (CT) demonstrated incomplete ablation was considered as an additional treatment cycle.
Assessment of Therapeutic Efficacy
Local therapeutic efficacy was evaluated by contrast-enhanced CT scanning (Toshiba Xpress/SX, Tokyo, Japan) at 1 month after treatment. Complete ablation was defined as uniform hypoattenuation without enhancement in the ablated zone. For the incompletely ablated nodules, an additional session of RFA or MWA was given with the aim of complete ablation. If complete ablation failed to be achieved after additional treatment, incomplete ablation was considered.
All patients were followed with color Doppler ultrasound examination of the liver and serum AFP monthly for the first 6 to 12 months, and once every 3 months thereafter. Contrast-enhanced CT scanning was ordered if local recurrence (LR) and/or distant recurrence (DR) were detected on ultrasonography. Local recurrence and distant recurrence were determined by 2 radiologists who reviewed contrast-enhanced CT scanning in consensus.
Regrowth of a successfully ablated tumor at or adjacent to the original site was defined as LR. The presence of intrahepatic or extrahepatic new tumor nodules was defined as DR. Both LR and DR were treated with RFA or MWA if applicable.
The results were given as the mean ± standard deviation, or median + range if the data were not normally distributed. Statistical analysis was undertaken using the commercially available SPSS software package (version 10.0; SPSS Inc, Chicago, Ill). Intergroup differences in categoric variables were analyzed by the chi-square test or Fisher exact probability, and intergroup differences in continuous variables were compared by the Student t test for unpaired data. The Kaplan-Meier method was used to calculate survival rate, and the log-rank test was used to compare intergroup differences. Cox regression multivariate analysis was used to identify the independent survival prognostic factors. Statistical significance was considered at P < .05.
Tumor Response to Treatment
A total of 137 treatment cycles of RFA and MWA were performed on 109 patients (1.3 ± 0.6 cycles per patient). Assessment of local therapeutic efficacy was available in 108 of 109 patients. Complete ablation was achieved in 84 patients after initial treatment and in 16 patients after additional treatment, with a complete ablation rate of 92.6% (100 of 108 patients). Of the 59 patients treated by RFA, complete ablation was achieved in 41 after initial treatment and 12 after additional treatment. In comparison, of 49 patients treated by MWA, complete ablation was achieved in 43 after initial treatment and 4 after additional treatment (P = .062). There was no significant difference in the complete ablation rate between RFA and MWA (89.8% [53 of 59 patients] vs 95.9% [47 of 49 patients]; P = .288). Conversely, of 88 patients with tumors measuring 3.0 cm to 5.0 cm, complete ablation was obtained in 71 after initial treatment and 13 after additional treatment. In comparison, of 20 patients with tumors measuring 5.0 cm to 7.0 cm, complete ablation was obtained in 13 after initial treatment and 3 after additional treatment (P = .717). The complete ablation rate was greater in patients with tumors measuring 3.0 cm to 5.0 cm than that in patients with tumors measuring 5.0 cm to 7.0 cm, being 95.4% (84 of 88) versus 80% (16 of 20) (P = .037).
Major Complications and Side Effects
There were no ablation-related deaths reported. The major complication rate was 9.2% (10 of 109 patients), including a hepatic subcapsular hematoma in 1 patient, an intra-abdominal hemorrhage in 1 patient, a second-degree skin burn in 1 patient, wound infection in 2 patients, massive hydrothorax in 4 patients, and acute cholecystitis in 1 patient. All were cured with conservative treatment.
Side effects mainly included mild to moderate pain in 81.7% of patients (89 of 109) and fever in 71.6% of patients (78 of 109). These patients were managed symptomatically with nonsteroidal anti-inflammatory analgesics. The symptoms usually subsided within 1 to 3 days.
This study was censored on August 31, 2007. Of 109 patients, 3 (2.8%) were lost to follow-up at 2.0 months, 11.0 months, and 16.6 months after ablation, respectively. The mean follow-up time for the remaining 106 patients was 22.0 ± 18.5 months (range, 2.2‒93.5 months).
A total of 22 of 100 patients (22%) developed LR after complete ablation during a mean follow-up of 23.0 ± 18.9 months. The 2-year local recurrence-free survival rate was 38.3%, with a median time to local recurrence of 4.6 months (range,2.0‒27.8 months). LR was 20.2% in tumors measuring 3.0 cm to 5.0 cm (17 of 84), and 31.2% in tumors measuring 5.0 cm to 7.0 cm (5 of 16) (P = .336). There were no significant differences in LR between RFA-treated and MWA-treated medium HCC tumors (P = .274), and between RFA-treated and MWA-treated large HCC tumors (P = 1.000). Univariate analysis was performed to assess the influence of factors including age of patients, preablation liver function status, type of tumor treated (first-treated vs recurrent tumors), tumor number, tumor size, and treatment modality (RFA vs MWA) on LR. The results demonstrated that LR was not associated with any of these factors (Table 3).
Table 3. Univariate Analysis of Factors Associated With Local Recurrence
LR indicates local recurrence; RFA, radiofrequency ablation; MWA, microwave ablation.
Determined using the chi-square test.
Age(≤50 y vs >50 y)
13.6 vs 28.6
Type of tumor
First treatment vs recurrent tumor
20.0 vs 24.4
Preablation Child-Pugh grade
A vs B
23.8 vs 15.0
No. of tumors
Solitary vs multiple
21.0 vs 23.7
Maximum dimension of main tumor, cm
3-5 vs 5-7
20.2 vs 31.2
RFA vs MWA
26.4 vs 17.0
There were 58 of 106 patients (54.7%) developed DR after ablation, including intrahepatic new lesions in 54 patients, lung metastases in 3 patients, and retroperitoneal lymph node metastases in 1 patient. The median interval time to DR was 6.7 months (range, 1‒36 months). Patients who received thermal ablation as the first treatment had a markedly lower cumulative DR rate and a longer interval time to DR in comparison with patients treated for recurrent tumors. The 3-year DR rates were 61.2% and 80.8%, respectively, and the interval times to DR were 21.0 ± 10.5 months and 6.7 ± 2.1 months, respectively (P = .046) (Fig. 1).
Of 106 patients, 38 were still alive at the time of censor of this study. They survived from 2.5 months to 93.5 months after ablation (23.9 ± 22.6 months), including survival of <1 year in 16 patients, 1 to 2 years in 7 patients, 2 to 3 years in 7 patients, 3 to 4 years in 5 patients, and >5 years in 3 patients. The remaining 68 patients died of various causes, including tumor progression in 52 patients, liver failure in 13 patients, and upper gastrointestinal bleeding in 3 patients. The survival was <1 year in 23 patients, 1 to 2 years in 24 patients, 2 to 3 years in 11 patients, 3 to 4 years in 5 patients, 4 to 5 years in 2 patients, and >5 years in 3 patients, with a mean survival of 21.2 ± 16.1 months (range, 2.2‒70.6 months).
The 1-year, 2-year, 3-year, and 5-year postablation overall survival rates were 75.8%, 47.2%, 30.9%, and 15.4%, respectively, with a median survival of 23.4 months (95% confidence interval [95% CI], 20.3-26.5 months) (Fig. 2). Univariate analysis was undertaken to evaluate the impact of the various factors, including age of patients, type of tumor treated (first-treated vs recurrent tumors), preablation AFP level, liver function status, tumor number, tumor size, treatment modality (RFA vs MWA), and completeness of ablation (completely vs incompletely ablated) on long-term survival. The results demonstrated that completeness of ablation, type of tumor treated, and preablation AFP level were significant prognostic factors for long-term survival, whereas other factors had insignificant impact (Table 4). The patients with completely ablated tumors had much better survival than those with incompletely ablated tumors, with 1-year, 2-year, 3-year, and 5-year postablation survival rates of 80.0% versus 25.0%, 51.0% versus 0%, 34.3% versus 0%, and 17.1% versus 0%, respectively (P = .0000) (Fig. 3). Patients who received ablation as the first treatment had markedly longer survival when compared with patients treated for recurrent tumors, with 1-year, 2-year, 3-year, and 5-year postablation survival rates of 78.9% versus 72.5%, 65.5% versus 30.8%, 45.5% versus 19.0%, and 25.3% versus 8.5%, respectively (P = .015) (Fig. 4). The 1-year, 2-year, 3-year, and 5-year postablation survival rates were significantly higher in patients with a preablation AFP level<200 ng/mL than those with a preablation AFP level≥200 ng/mL, being 83.1% versus 60.9%, 54.6% versus 32.6%, 38.4% versus 16.9%, and 18.4% versus 12.7%, respectively (P = .008) (Fig. 5). Conversely, patients with tumors measuring 3.0 cm to 5.0 cm had similar 1-year, 2-year, 3-year, and 5-year postablation survival rates compared with those with tumors measuring 5.0 cm to 7.0 cm, being 77.1% versus 70.0%, 48.7% versus 41.5%, 30.4% versus 34.6%, and 14.9% versus 17.3%, respectively (P = .84) (Fig. 6).
Table 4. Univariate Analysis of Prognostic Factors for Long-term Survival
Cox regression multivariate analysis, which included age of the patients, type of tumor treated (first-treated vs recurrent), preablation AFP level, liver function status, tumor number, tumor size, treatment modality (RFA vs MWA), and completeness of ablation, was then performed to identify the independent prognostic factors. It confirmed that incomplete ablation of tumors, posthepatectomy recurrent tumors, and preablation AFP level ≥200 ng/mL were 3 independent unfavorable prognostic factors, with an exp(B) of 4.158 (95% CI, 1.801-9.601; P = .001), 1.568 (95% CI, 0.945-2.692; P = .082), and 1.593 (95% CI, 0.945-2.692; P = .082), respectively.
The management of medium and large HCC (>3 cm) in patients who are not candidates for surgical resection remains a major challenge. With the advantages of minimal invasiveness and predictability of therapeutic efficacy, thermal ablation has quickly gained a great deal of attention in the management of HCC.21, 22 A rapidly growing body of medical literature has produced convincing evidence that thermal ablation is very effective for small HCC.2-5, 9, 10 With further refinement in the devices and techniques, thermal ablation has exhibited promising capability in treating larger HCC measuring up to 5 cm.
The question now is whether a tumor larger than 5 cm can be treated effectively by thermal ablation. Livraghi et al23 reported complete ablation rates of 23% to 25% in tumors measuring 5.1 to 9.5 cm using RFA. The unsatisfactory response to treatment in their series was due to large tumors being included in their study. In our study using RFA or MWA, we found that although the complete ablation rate in tumors measuring 3 to 5 cm was significantly higher than that in tumors measuring 5 to 7 cm, the complete response to treatment in tumors measuring 5 to 7 cm was as high as 80%, which we considered acceptable. In our center during the same period of time as the current study, we also treated 7 HCC patients with tumors measuring >7 cm (range, 7.2‒13 cm) with RFA or MWA. However, complete ablation was achieved in none of them (unpublished data). These results suggested that 7 cm might be the upper limit of tumor size suitable for thermal ablation. A further increase in tumor size appeared to be beyond the ability of the currently available percutaneous ablative technique and would markedly lower its therapeutic efficacy.
With respect to LR, it has been reported to occur in 24% for patients with tumors measuring 3 to 4 cm after RFA.16 The current study data indicated that the incidence of LR was 20% in tumors measuring 3 to 5 cm, and 31% in tumors measuring 5 to 7 cm. There was no significant difference noted between these 2 groups of patients. These results further supported that thermal ablation could treat tumors measuring up to 7 cm in greatest dimension.
For the long-term survival of patients treated by RFA, Lencioni et al17 reported a 2-year survival rate of 98% for 52 patients with HCC measuring ≤5 cm. Lin et al16 achieved a 3-year survival rate of 74% for 50 patients with HCC measuring <4 cm. Recently, Chen et al19 treated 71 patients with solitary HCC measuring ≤5 cm using RFA, and obtained a 3-year survival rate of 71%. All these series included patients with tumors measuring <3 cm. In our study using RFA or MWA to treat patients with HCC measuring 3 to 7 cm, the 1-year, 2-year, 3-year, and 5-year overall survival rates were 76%, 47%, 31%, and 15%, respectively. Univariate analysis and Cox regression multivariate analysis demonstrated that incomplete ablation of tumors, posthepatectomy recurrence, and baseline level of AFP ≥200 ng/mL were 3 independent unfavorable prognostic factors for long-term survival. Among them, completeness of ablation was the most important prognostic factor. Sala et al24 treated 282 cirrhotic patients with early HCC using PEI, RFA, or transcatheter arterial embolization-PEI and discovered that the initial complete tumor ablation with treatment was associated with improved long-term survival. Because tumor size affects the incidence of complete tumor ablation,23, 25 large tumors have a high possibility of incomplete ablation and consequently negatively influence the long-term outcome. Our results demonstrated that there was no significant difference in the long-term survival between patients with tumors measuring 3 to 5 cm and 5 to 7 cm. A similar conclusion has been drawn from a study by Guglielmi et al,26 who treated 65 HCC (tumors measuring 1‒7 cm) in 53 patients with RFA. These authors demonstrated that the 3-year survival rate was not affected by tumor size and number. Thus, with the currently available thermal ablation devices and techniques, HCC measuring up to 7 cm can be treated with no remarkable compromise in patient survival.
Patients who received thermal ablation as the first treatment had a markedly longer survival than the patients with posthepatectomy tumor recurrence, with a 5-year survival rate of 25% versus 8%, and a median survival of 35 months versus 18 months. Similar results have been shown in a study reported by Tateishi et al, who demonstrated that patients who received thermal ablation as the first treatment had a markedly longer 5-year survival than the patients with recurrent tumors after partial hepatectomy (54% vs 38%).27 The reason for the poorer prognosis of the patients with recurrent tumors is that the recurrent HCC re-recurred within a shorter time interval. The results of the current study revealed that patients who received thermal ablation as the first treatment had a markedly lower DR rate and a longer interval time to DR than patients who had posthepatectomy tumor recurrence, with a 3-year DR rate of 61% versus 81% and a mean interval time to DR of 21 months versus 7 months.
Patients with a preablation AFP<200 ng/mL had significantly better survival than those with a preablation AFP≥200 ng/mL, with a 5-year survival rate of 18% versus 13%, and a median survival time of 29 months versus 18 months. The unfavorable influence of high AFP level on long-term survival has been shown in HCC patients who underwent partial hepatectomy.28, 29 To our knowledge, the exact mechanism for unfavorable influence of AFP on prognosis still remained unclear.
The thermal ablative treatment was well tolerated in the current series, with a morbidity rate of 9.2% and a mortality rate of 0%. Our morbidity rate was close to that reported in the medical literature. An Italian multicenter study of percutaneous RFA of malignant liver tumors in 2320 patients reported a morbidity rate of 7.1% and a mortality rate of 0.3%.30 Another review reported a morbidity rate of 8.9% and a mortality rate of 0.5% after percutaneous, laparoscopic, or open RFA of hepatic tumors in 3670 patients.31
In conclusion, percutaneous thermal ablation using RF or MW energy was found to be effective and safe in treating medium and large HCC tumors measuring up to 7 cm, with an acceptable local tumor control and long-term survival.