- Email: [email protected]
Ten-year outcomes of percutaneous radiofrequency ablation as first-line therapy of early hepatocellular carcinoma: Analysis of prognostic factors
Research Article
Ten-year outcomes of percutaneous radiofrequency ablation as first-line therapy of early hepatocellular carcinoma: Analysis of prognostic factors Young-sun Kim1, Hyo Keun Lim1,⇑, , Hyunchul Rhim1,⇑, , Min Woo Lee1, Dongil Choi1, Won Jae Lee1, Seung Woon Paik2, Kwang Cheol Koh2, Joon Hyeok Lee2, Moon Seok Choi2, Geum-Youn Gwak2, Byung Chul Yoo2 1
Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea; 2Department of Medicine and Digestive Disease Research Center, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
Background & Aims: The aim was to assess 10-year outcomes of radiofrequency ablation as a first-line therapy of early-stage hepatocellular carcinoma with an analysis of prognostic factors. Methods: From April 1999 to April 2011, 1305 patients (male: female = 993:312; mean age, 58.4 years) with 1502 early-stage hepatocellular carcinomas (mean size, 2.2 cm) were treated with percutaneous radiofrequency ablation as a first-line option. Follow-up period ranged from 0.4 to 146.6 months (median, 33.4 months). We assessed the 10-year follow-up results of recurrences and survival with the analyses of prognostic factors. Results: Recurrences occurred in 795 patients (1–17 times), which were managed with various therapeutic modalities. The cumulative local tumor progression rates were 27.0% and 36.9% at 5 and 10 years, respectively, for which the only significant risk factor was large tumor size (B = 0.584, p = 0.001). Cumulative intrahepatic distant and extrahepatic recurrence rates were 73.1% and 88.5%, and 19.1% and 38.2% at 5 and 10 years, respectively. Corresponding overall survival rates were 59.7% and 32.3%, respectively. Poor survival was associated with old age (B = 0.043, p = 0.010), Child-Pugh class B (B = 1.054, p <0.001), absence of antiviral therapy during follow-up (B = 0.699, p = 0.034), and presence of extrahepatic recurrence (B = 0.971, p = 0.007).
Keywords: Hepatocellular carcinoma; Radiofrequency ablation; Outcome analysis; Ten-year survival; Prognostic factor. Received 30 January 2012; received in revised form 7 September 2012; accepted 13 September 2012; available online 27 September 2012 ⇑ Corresponding authors. Address: Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, #50, Irwon-dong, Gangnam-gu, Seoul 138-225, Republic of Korea. Tel: +82 2 3410 2518; fax: +82 2 3410 6368. E-mail addresses: [email protected] (H.K. Lim), [email protected] (H. Rhim). These authors contributed equally to this work. Abbreviations: RFA, radiofrequency ablation; HCC, hepatocellular carcinoma; LTP, local tumor progression; IDR, intrahepatic distant recurrence; ER, extrahepatic recurrence; OS, overall survival; US, ultrasonography; CT, computed tomography; MRI, magnetic resonance imaging; AFP, alpha-fetoprotein; BCLC, Barcelona Clinic Liver Cancer.
Conclusions: Ten-year survival outcomes after percutaneous radiofrequency ablation as a first-line therapy of hepatocellular carcinoma were excellent despite frequent tumor recurrences. Overall survival was influenced by age, Child-Pugh class, antiviral therapy, or extrahepatic recurrence. Ó 2012 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Introduction Since the first clinical study in 1995 [1], clinical implications of percutaneous radiofrequency ablation (RFA) in the treatment of early-stage hepatocellular carcinoma (HCC) have expanded. Currently, RFA is recognized as a curative modality for early-stage HCC [2] whose outcomes are comparable to those of surgery, as evidenced by many studies [3–7]. Clinical roles of percutaneous RFA are actually more meaningful than those of surgery from the viewpoint that RFA could be applied to patients whose hepatic functional reserve is insufficient to endure surgery, and can be conducted repeatedly with the same patient [8]. Despite such a large potential, long-term (i.e., more than 10 years) follow-up results are rare [9] because of its relatively short history of clinical applications although there has been plenty of reports regarding 5-year outcomes [8,10–12]. During the last 12 years, our institution has performed percutaneous RFA of HCC in more than 3000 patients, and approximately more than 40% were performed as a first-line therapy. Through these clinical experiences, the best way to perform RFA has evolved toward developing an effective as well as safe treatment, for example, the use of artificial ascites techniques, multiple overlapping ablations with ‘‘switch box’’, or fusion image guidance. In addition, antiviral therapy has become more and more readily available, all of which seemingly affect therapeutic outcomes regarding not only recurrence but also overall survival (OS). These long-term changes raise the necessity for a new analysis of the longer-term outcomes. Therefore, the purposes of our study were to evaluate 10-year follow-up results with regard to recurrences and OS after
Journal of Hepatology 2013 vol. 58 j 89–97
Research Article percutaneous RFA as a first-line therapeutic option for the treatment of early-stage HCC, and to assess factors significantly influencing survival.
Patients and methods This study was approved by the institutional review board of our hospital, and informed consent from patients was waived because this was a retrospective clinical study. However, written informed consent for RFA procedures and the use of data for research purposes were obtained from patients prior to every treatment. Study population We reviewed the departmental database of RFA procedures and retrieved the following data. Between April 1999 and April 2011, a total of 10,334 planning ultrasonography (US) examinations were conducted by one of six radiologists (Y.S.K., H.K.L., H.R., M.W.L., D.C., and W.J.L.) on a referral basis, and among them, USguided percutaneous RFAs were performed for the treatment of a total of 5097 HCC tumors in 3084 patients (male:female = 2402:682 mean age, 58.5 years; age range, 24–89). Out of 5097 HCC tumors, 1502 tumors (mean size, 2.2 cm; size range, 0.5–4.9 cm) in 1305 consecutive patients (male:female = 993:312; mean age, 58.4 years; age range, 28–86 years), which were initially diagnosed and treated with US-guided percutaneous RFA as a first-line option, were analyzed in this study (Fig. 1). Two hundred six tumors (206/1502, 13.7%) were confirmed as HCC by means of a US-guided core needle biopsy, and the remaining 1296 tumors (1296/1502, 86.3%) were considered to be HCC, based on one of two clinical criteria from the American Association for the Study of Liver Diseases according to the time of RFA procedures [2,13].
Our institutional inclusion criteria for percutaneous RFA procedures were as follows: (1) presence of a single nodular HCC <5 cm in maximum diameter; (2) presence of multinodular HCCs (63 in number, each <3 cm in maximum diameter); (3) absence of portal venous thrombosis; (4) Child-Pugh class A or B; and (5) prothrombin time ratio >50% (prothrombin time with an international normalized ratio <1.7) and platelet count >50,000 cells/mm3 (50 cells 109/L). The decision to perform RFA was made by consensus of an interdisciplinary discussion. In cases of a single HCC in a Child-Pugh class A patient with neither portal hypertension nor hyperbilirubinemia, surgical resection was primarily considered. However, patients’ age, co-morbidity and preference were also taken into account. The technical feasibility of RFA was assessed with planning US exam [14] by the radiologist who would perform the procedure. The presence of ascites was not either a relative or an absolute contraindication. Clinical characteristics of the patients and tumors analyzed in this study are summarized in Table 1. RFA procedure All RFA procedures were performed percutaneously under US guidance. Procedures were performed on an inpatient basis by one of six radiologists (Y.S.K., H.K.L., H.R., M.W.L., D.C., and W.J.L.), who had at least seven years of experience with this procedure, by the end of the study period. Details about RFA procedures are reported as Supplementary data. Follow-up after RFA For early evaluation of the therapeutic response or possible complications, either contrast-enhanced US (n = 199, 15.2%), CT (n = 1091, 83.6%), or MRI (n = 15, 1.5%) was performed within 24 h. Between July 2002 and December 2003, contrastenhanced US using a microbubble contrast agent (SH U508A, Levovist; Bayer
Planning US for percutaneous RFA of HCC n = 10,334
Feasible n = 4460
Not feasible n = 5874
Choice of other modalities or follow-up loss, n = 59
Percutaneous RFA of HCC 3084 patients with 5097 tumors (4401 procedures)
Recommendation of follow-up TACE Surgical resection Percutaneous RFA with TACE Radiation therapy Liver transplantation Intraoperative RFA Percutaneous ethanol injection Follow-up loss
Percutaneous RFA as a first-line option
Percutaneous RFA for recurrence
1305 patients with 1502 tumors (1305 procedures)
1779 patients with 3595 tumors (3096 procedures)
n = 783 n = 3054 n = 841 n = 142 n = 78 n = 49 n = 41 n = 12 n = 874
Final inclusion Fig. 1. Flow of study inclusion. A total of 10,334 patients were examined with planning US, and 1305 patients with 1502 HCC tumors, treated with percutaneous RFA as a first-line option, were finally included.
90
Journal of Hepatology 2013 vol. 58 j 89–97
JOURNAL OF HEPATOLOGY Table 1. Baseline features of the study population.
Clinical Features
Numbers
Patients
n = 1305
Male:female
993:312
Age (yr)
mean 58.4, range 28-86
Liver disease
A follow-up study was performed 1 month later. If RFA was considered to be technically effective, a follow-up was repeated every 3 months until two years after RFA. After then, follow-ups were conducted every 4–6 months according to the risks for recurrences. Chest radiographs were also checked at each follow-up. Patients who showed specific clinical or radiological features such as headache for brain metastasis or abnormality in chest radiograph for pulmonary metastasis underwent appropriate imaging examinations for the detection of extrahepatic recurrence (ER). Treatment strategy after initial RFA
Liver cirrhosis
1077
Acute or chronic hepatitis
103
Healthy carrier
75
None
50
When an unablated residual tumor was observed at the immediate follow-up, additional RFA was primarily chosen. Unless percutaneous RFA was technically feasible, other therapeutic modalities were carried out. If the index tumor was completely covered by the ablation zone, the treatment was regarded as a technical success regardless of the sufficiency of ablative margin. If LTP or intrahepatic distant recurrence (IDR) was observed during subsequent follow-up visits, an additional treatment was immediately performed, preferably using percutaneous RFA.
Etiology of liver disease Hepatitis B virus
912
Hepatitis C virus
233
Hepatitis non-B, non-C virus
33
Hepatitis B and C viruses
7
Alcohol
59
Autoimmune hepatitis
1
Cryptogenic
10
No liver disease
50
Analysis of therapeutic efficacy and survival Local therapeutic efficacy in terms of technique effectiveness and LTP was assessed on a tumor basis. IDR, ER, event-free survival, and OS were evaluated on a per-patient basis. If the patient underwent liver transplantation, we regarded the operation date as the end of follow-up for LTP and IDR. The OS time was defined as the interval between the first RFA and either death or last follow-up visit before August 31, 2011. If death resulted from unrelated causes such as a car accident, that data was disregarded. Event-free survival was defined as the time from the first RFA to either the earliest event (i.e., LTP, IDR, ER or death) or the last follow-up date without an event.
Child-Pugh classification Class A
1021
Class B
284
Definitions of terminology
Baseline serum AFP level* Range
1-13,563.1 ng/ml
Mean ± standard deviation
207.8 ± 796.2 ng/ml
≤20 ng/ml
625
>20 ng/ml, ≤200 ng/ml
429
>200 ng/ml
213
Number of tumors One
1122
Two
169
Three
14
BCLC stage Stage 0
427
Stage A
878
Tumors
n = 1502
Size Range
0.5-4.9 cm
Mean ± standard deviation
2.2 ± 0.8 cm
Diagnosis of hepatocellular carcinoma Histopathology
89
Edmondson-Steiner grade II
100
Edmondson-Steiner grade III
8
Edmondson-Steiner grade IV
9
Clinical criteria ⁄
206
Edmondson-Steiner grade I
1296
Available in 1267 patients.
Schering Pharma, Berlin, Germany) was predominantly conducted for research purposes. In other periods, triple-phase contrast-enhanced CT was the preferred method; however, if the patient had a contraindication to CT, MRI was performed.
Definitions are based on the standardization by the International Working Group on Image-Guided Tumor Ablation [20]. Technical success was defined when the tumor was treated according to the protocol and was completely replaced by RFA zones at the immediate follow-up. Achievement of technique effectiveness was defined when complete ablation of macroscopic tumors was evident on the 1-month follow-up exams for those patients whose HCC tumors were completely removed by the initial RFA treatment. LTP was diagnosed when a follow-up exam demonstrated findings of interval development/growth of the tumor along the margin of the ablation zone where the RFA had been considered to be technically effective. IDR was defined by a lesion with similar characteristics but not contacting the original ablation zone in the liver. Cancer seeding was regarded as extrahepatic recurrence. Major complication was defined as an event that leads to substantial morbidity and disability, increasing the level of care, or results in hospital admission or substantially lengthens the hospital stay. All other complications were regarded as minor. Statistical analysis The sizes of RFA-treated tumors in the first and last years were compared using the Mann–Whitney test. The cumulative rates for each type of recurrence and survival were estimated with the Kaplan–Meier method. Prognostic factors for LTP were assessed using the multivariate Cox proportional hazard model. Potential prognostic factors for LTP included sex, age, etiology of liver disease, Child-Pugh class, tumor size, Edmondson-Steiner grade (if histopathologically confirmed), baseline serum a-fetoprotein level (AFP), and AFP changes within 30 days in response to RFA (group 1, initially >20 ng/ml and decrease to <2/3 of the baseline after RFA; group 2, initially >20 ng/ml and increase, no change or decrease to P2/3 of the baseline after RFA; group 3, consistently 620 ng/ml regardless of RFA). OS curves of different BCLC (Barcelona Clinic Liver Cancer) stages as well as each potential prognostic factor were compared with the log rank test as univariate analysis. Prognostic factors for OS were also assessed using the multivariate Cox proportional hazard model. Potential prognostic factors for OS included sex, age, etiology of liver disease, Child-Pugh class, number and the largest diameter of the RFA-treated tumors, Edmondson-Steiner grade (if histopathologically confirmed; higher grade if multiple), baseline AFP, AFP changes within 30 days in response to RFA, antiviral therapy during follow-up period, and presence of LTP, IDR, and ER after RFA.
Journal of Hepatology 2013 vol. 58 j 89–97
91
Research Article A p-value <0.05 was considered to be statistically significant. Data analyses were performed using the commercially available software (SPSS, ver. 19.0; SPSS Inc., Chicago, IL).
Results The average size of the index tumors decreased each year from 2.6 cm in 1999 to 1.9 cm in 2011 (Supplementary Fig. 1), which was significantly different (1999 vs. 2011, p = 0.001). The median follow-up period for 1305 patients was 33.4 months (range, 0.4– 146.6 months; mean, 38.7 months). Two patients died within 30 days after RFA treatment because of hypovolemic shock caused by variceal bleeding and unusually rapid progression of HCC, respectively. Fifty-two out of 1305 patients (4.0%) underwent liver transplantation 26.4 ± 18.1 months (range, 3.4– 76.4 months) after percutaneous RFA. Technical success Technical success was achieved in one session in 94.8% of cases (1237/1305). Sixty-seven cases (67/1305, 5.1%) with 68 tumors (68/1502, 4.5%) turned out to be technical failures due to either residual unablated tumors (n = 54) or mistargeting (n = 14) at immediate follow-up examinations. Procedures were stopped for four cases (4/1305, 0.3%) because of unstable vital signs, which in turn left residual unablated tumors at immediate follow-up. We repeated percutaneous RFA in 51 cases with 52 tumors and finally succeeded in completely ablating the index tumors in 47 cases (47/51, 92.2%) with 48 tumors. The remaining 20 cases with 20 untreated or residual tumors were managed with chemoembolization (n = 18) or surgical resection (n = 2). Recurrences and managements Of 1283 patients who were completely treated with RFA and survived more than 30 days, recurrences occurred in 795 patients (62.0%, 795/1283). Among them, 41 cases were lost to followup or refused treatment, thus 754 patients with 1st recurrence (LTP, n = 154; IDR, n = 535; LTP and IDR, n = 55; LTP and ER, n = 2; IDR and ER, n = 8) were managed with the various therapeutic modalities. Among the 754 patients, 245 (32.5%, 245/ 754) showed no recurrence, and 509 (67.5%, 509/754) experienced 2nd recurrence. Recurrences were observed up to 17 times (n = 2) as of August 31, 2011, and were also managed with the various therapeutic modalities as shown in Fig. 2. Local therapeutic efficacy Among 1482 tumors which were regarded as to be completely ablated by the initial RFA treatments, 20 showed viable residues at 1 month follow-up; therefore, the technique effectiveness rate was 98.7% (1462/1482). Of 1462 tumors with complete technique effectiveness, LTP was identified in 284 (284/1462, 19.4%) from 2.4 months to 113.2 months. The mean estimate of LTP-free survival was 105.3 months (95% confidence interval (95% CI) 100.7– 110.0 months). Cumulative rates of LTP were estimated as 9.7% (95% CI 8.9–10.5%), 21.4% (20.1–22.7%), 27.0% (25.4–28.6%), 30.2% (29.2–33.6%), and 36.9% (32.5–41.3%) at 1, 3, 5, 8, and 10 years, respectively. Multivariate analysis revealed that large
92
tumor size was the only independently significant risk factor for LTP (B = 0.584, p = 0.001) (Fig. 3A) (Table 2). Initial LTPs were treated with chemoembolization (n = 165), percutaneous RFA (n = 103), combination therapy of percutaneous RFA with chemoembolization (n = 7), surgical resection (n = 5), liver transplantation (n = 1), intraoperative RFA (n = 1), percutaneous ethanol injection (n = 1), or radiation therapy (n = 1) with a mean delay of 14.2 days after diagnosis. Remote recurrences IDRs were observed in 715 of 1305 (54.8%) patients. The first IDRs occurred 0.7–99.7 months after RFA. Mean and median estimates of the IDR-free survival time were 43.6 months (95% CI 40.0– 47.1 months) and 27.4 months (24.8–30.1 months), respectively. Cumulative rates of IDR were estimated as 24.4% (95% CI, 23.2– 25.6%), 59.5% (57.9–61.1%), 73.1% (71.3–74.9%), 87.0% (84.7– 89.3%), and 88.5% (86.1–90.9%) at 1, 3, 5, 8, and 10 years, respectively (Fig. 3B). The initial IDRs were managed with conventional chemoembolization (n = 372), percutaneous RFA (n = 243), combination therapy of percutaneous RFA and chemoembolization (n = 38), surgical resection (n = 15), systemic chemotherapy (n = 9), liver transplantation (n = 8), radiation therapy (n = 3), intraoperative RFA (n = 2), or percutaneous ethanol injection therapy (n = 2). Twenty patients did not receive any specific treatment prior to their death. The remaining three patients were lost to follow-up. ERs including cancer seeding were found in 165 of 1305 (12.6%) patients during follow-up. ERs occurred 1.0– 130.6 months after treatment, and the mean estimate of ER-free survival time was 113.3 months (95% CI 107.6–119.1 months). Estimated cumulative rates were 2.7% (95% CI 2.2–3.2%), 10.2% (9.2–11.2%), 19.1% (17.5–20.7%), 28.5% (25.8–31.2%), and 38.2% (32.9–43.5%) for 1, 3, 5, 8, and 10 years, respectively (Fig. 3B). Cancer seedings were detected in 5 cases with a mean delay of 19.6 months (range 7.1–38.2 months). Long-term survival Of a total of 1305 patients, 407 (31.2%) died of HCC progression (n = 333), hepatic failure or complications of liver cirrhosis, such as variceal bleeding (n = 39) or systemic or other organ complications, such as sepsis or cerebral hemorrhage (n = 35). Time of death after RFA ranged between 0.4 and 132.0 months with a mean survival time of 82.9 months (95% CI 78.4–87.4 months) and a median survival time of 75.0 months (66.8–83.1 months). Thirty-seven patients (37/1305, 2.8%) were lost during followup, and 861 (66.0%) were alive as of August 31, 2011. The number of 5- and 10-year survivors was 219 and 13, respectively. Cumulative OS rates were estimated as 95.5% (95% CI 94.9–96.1%), 77.9% (76.6–79.2%), 59.7% (57.9–61.5%), 43.2% (40.8–45.6%), and 32.3% (28.5–36.1%) at 1, 3, 5, 8, and 10 years, respectively (Fig. 4A). Estimates of OS of the subgroups according to each potential prognostic factor and the results of univariate comparisons are summarized in Supplementary Table 2. By multivariate analysis, old age (B = 0.043, p = 0.010), Child-Pugh class B (B = 1.054, p <0.001), absence of antiviral therapy during follow-up (B = 0.699, p = 0.034), and presence of ER (B = 0.971, p = 0.007) were independently significant poor prognostic factors for OS (Table 3). Patients with BCLC stage 0 showed a significantly better OS than patients with BCLC stage A (p = 0.002) (Fig. 4B).
Journal of Hepatology 2013 vol. 58 j 89–97
JOURNAL OF HEPATOLOGY Percutaneous RFA as a first-line option n = 1305 Mortality within 30 days (n = 2) Untreated or residual tumor after RFA (n = 20)
1st recurrence, n = 795
No recurrence, n = 488
No treatment (n = 41) Treatment of 1st recurrence, n = 754 n = 154 IDR
LTP
RFA n = 79 TACE n = 66 RFA after TACE n = 4 SR n=1 PEI n=1
n = 535
TACE RFA RFA after TACE LT SR Others
LTP + IDR n = 55
n = 270 n = 220 n = 26 n=7 n=5 n=8
TACE RFA SR Chemo.
(LTP or IDR) + ER n = 10 TACE+Chemo. RT+Chemo. TACE+RT Chemo.
n = 42 n = 10 n=2 n=1
2nd recurrence, n = 509
n=6 n=2 n=1 n=1
No recurrence, n = 245
No treatment (n = 22) Treatment of 2nd recurrence, n = 487 LTP
n = 74
TACE RFA RT SR RFA after TACE
n = 43 n = 20 n=6 n=4 n=1
IDR
n = 405
TACE RFA RFA after TACE LT SR Others
n = 264 n = 109 n = 13 n=6 n=4 n=9
LTP + IDR TACE
ER
n=3
n=2
Chemo.
n=3
IDR + ER
n=3
RT Chemo.
n=2 n=1
n=2
No recurrence, n = 143
3rd recurrence, n = 344 No treatment (n = 10) Treatment of 3rd recurrence, n = 334 LTP TACE RFA SR RFA after TACE
n = 28 n = 17 n=8 n=2 n=1
IDR TACE RFA RFA after TACE LT RT SR
n = 302 n = 233 n = 53 n=5 n=4 n=4 n=3
ER Chemo. RT
n=4 n=2 n=2
4th recurrence, n = 242
5th recurrence, n = 188
6th recurrence, n = 131
9th recurrence, n = 42
8th recurrence, n = 60
7th recurrence, n = 97
10th recurrence, n = 27
11th recurrence, n = 19
12th recurrence, n = 10
15th recurrence, n = 4
14th recurrence, n = 5
13th recurrence, n = 5
16th recurrence, n = 3
17th recurrence, n = 2
Fig. 2. Recurrences and management after percutaneous RFA of HCC. After percutaneous RFA in 1305 patients, recurrences occurred in 795 patients up to 17 times, which were managed with various therapeutic modalities. RFA, radiofrequency ablation; TACE, transcatheter arterial chemoembolization; PEI, percutaneous ethanol injection; SR, surgical resection; LT, liver transplantation; Chemo., Systemic chemotherapy; RT, radiation therapy; LTP, local tumor progression; IDR, intrahepatic distant recurrence; ER, extrahepatic recurrence.
Journal of Hepatology 2013 vol. 58 j 89–97
93
Research Article Cumulative event-free survival rates were estimated as 66.8% (95% CI 65.5–68.1%), 29.0% (27.5–30.5%), 17.5% (16.1–18.9%), 7.0% (5.5–8.5%), and 3.8% (2.2–5.4%) at 1, 3, 5, 8, and 10 years, respectively (Fig. 4C).
1.0 Cumulative local tumor progression rate
A
0.8 0.6
Complications
0.4
Following 1305 RFA treatments, a total of 108 (8.3%) complications were encountered during follow-up. Of these, 26 (2.0%) major complications were identified (Table 4). There were no procedure-related mortalities.
0.2 0.0 0 20 40 60 80 100120140 Duration (mo)
Cumulative recurrence rate
B 1.0 0.8 0.6 0.4 0.2 0.0
0 20 40 60 80 100120140 Duration (mo) Fig. 3. Recurrence curves. (A) Local tumor progression of 1462 HCC tumors after percutaneous RFA as a first-line therapy. Cumulative rates of local tumor progression were estimated as 9.7%, 21.4%, 27.0%, 30.2%, and 36.9% at 1, 3, 5, 8, and 10 years, respectively. Local tumor progressions according to tumor size were also demonstrated. (B) Intrahepatic distant recurrence and extrahepatic recurrence in 1305 patients with HCC, treated with percutaneous RFA as a first-line therapy. Cumulative rates of intrahepatic distant recurrence were estimated as 24.4%, 59.5%, 73.1%, 87.0%, and 88.5% at 1, 3, 5, 8, and 10 years, respectively. The estimated cumulative rates of extrahepatic recurrence were 2.7%, 10.2%, 19.1%, 28.5%, and 38.2% for 1, 3, 5, 8, and 10 years, respectively. +, censored data.
Discussion Although controversial, it is generally accepted that 5-year OS outcomes after percutaneous RFA of HCC are comparable to those of surgical resection in cases of small tumors [3–7,11]. However, because of a relatively short history of clinical application of RFA, most publications have dealt with at most 5-year results. To compare the results of our study with those of surgical resection, we investigated the surgical literature with 10-year follow-up, and we confined our search to those publications fulfilling the Milan criteria. Studies by Cucchetti et al. [15] and Yamamoto et al. [16] reported that 5-/10-year OS rates were 61.5%/28.7% and 56%/20%, respectively. However, both populations had a higher proportion of Child-Pugh class A than our study (95.7 and 85.3% vs. 78.2%), which might play a role as a favorable confounder for OS. Taking into account this discrepancy in demographics, our long-term survival outcome after percutaneous RFA of HCC, 59.7% at 5 years and 32.3% at 10 years, appears to be comparable to those of surgical resection. There have been several studies with regard to the effects of residual or recurrent HCC or treatment on survival. Sala et al. [17] reported that the initial complete response (i.e., absence of residual tumor) after percutaneous ablation therapy significantly improved survival. Ng et al. [10] demonstrated that different seg-
Table 2. Results of univariate and multivariate analysis on potential prognostic factors for LTP after percutaneous RFA of HCC, as first-line therapy.
Univariate analysis*
Multivariate analysis**
p value
Regression coefficient
Standard error
p value
Age†
0.393
-0.29
0.020
0.147
Gender
0.930
0.542
0.426
0.204
Etiology of liver disease
0.099
0.520
Child-Pugh class
0.203
-0.245
0.440
0.577
Tumor size‡
<0.001§§
0.584
0.180
0.001§§
Edmondson-Steiner grade
0.027§§
Baseline AFP§
0.897
0.128
0.387
AFP change in response to RFA
0.806
0.060
⁄
Log rank test. ⁄⁄ Cox proportional hazard model. Grouped based on median age (i.e.,658;>58) for univariate analysis. à Grouped into 61 cm, >1 and 62 cm, >2 and 63 cm, >3 and 64 cm, or >4 and 65 cm for univariate analysis. § Grouped into 620 ng/ml, >20 and 6200 ng/ml, or >200 ng/ml for univariate analysis. §§ Statistically significant.
94
Journal of Hepatology 2013 vol. 58 j 89–97
0.741 0.777
JOURNAL OF HEPATOLOGY A
B
0.8 0.6 0.4 0.2
Cumulative event-free survial rate
Cumulative overall survial rate
Cumulative overall survial rate
C 1.0
1.0
0.8 0.6 0.4 0.2 0.0
0.0 0
20
40 60 80 100 120 140 Duration (mo)
1.0 0.8 0.6 0.4 0.2 0.0
0
20
40 60 80 100 120 140 Duration (mo)
0
25
50 75 100 Duration (mo)
125
Fig. 4. Survival curves. (A) Overall survival curve of 1305 patients with HCC treated with percutaneous RFA. Cumulative overall survival rates were estimated as 95.5%, 77.9%, 59.7%, 43.2%, and 32.3% at 1, 3, 5, 8, and 10 years, respectively. (B) Patients with BCLC stage 0 demonstrated a significantly better overall survival than those with BCLC stage A. (C) Event-free survival curve of 1305 patients. The estimated cumulative rates for event-free survival were 66.8%, 29.0%, 17.5%, 7.0%, and 3.8% for 1, 3, 5, 8, and 10 years, respectively. +, censored data.
Table 3. Results of univariate and multivariate analysis on potential prognostic factors for overall survival after percutaneous RFA of HCC as a first-line therapy.
Univariate analysis*
Age†
⁄
Multivariate analysis** Regression coefficient Standard error
p value
<0.001§§
0.043
0.017
0.010§§
-0.059
0.310
0.849
-1.054
0.299
<0.001§§
Gender
0.089
Etiology of liver disease
<0.001§§
Child-Pugh class
<0.001§§
0.926
Number of tumors treated
0.579
0.052
0.315
0.868
Maximal tumor size treated‡
<0.001§§
0.235
0.165
0.155
Edmondson-Steiner grade
0.872
Baseline AFP§
0.205
-0.365
0.327
0.265
AFP change in response to RFA
<0.001§§
Antiviral therapy during follow-up
<0.001§§
0.699
0.329
0.034§§
0.369 0.220
Occurrence of local tumor progression
0.108
0.296
0.305
0.333
Occurrence of intrahepatic distant recurrence
0.361
-0.589
0.330
0.074
Occurrence of extrahepatic recurrence
<0.001§§
0.971
0.358
0.007§§
Log rank test; post hoc analyses are available in Supplementary Table 2.
⁄⁄
Cox proportional hazard model. Grouped based on median age (i.e., 658; >58) for univariate analysis. à Grouped into 61 cm, >1 and 62 cm, >2 and 63 cm, >3 and 64 cm, or >4 and 65 cm for univariate analysis. § Grouped into 620 ng/ml, >20 and 6200 ng/ml, or >200 ng/ml for univariate analysis. §§ Statistically significant.
ment intrahepatic recurrence and distant metastasis after RFA carried significant poor prognostic influence on OS outcome. In their study, LTP failed to show statistical significance by multivariate analysis. Lam et al. [18] demonstrated that aggressive treatment of LTP after RFA improved OS outcome. Our study showed that either LTP or IDR had no influence, and ER had a negative influence on survival based on both univariate and multivariate analyses. Our results with regards to LTP and ER were compatible with the results from the study by Ng et al. [10]. We could compare our results to those from another recent study by Shiina et al. in which 10-year follow-up results of RFA
are available [9]. The study population consisted of 1170 patients with a median follow-up of 38.2 months. Their cumulative LTP rates were much lower than ours (3.2% at 5 years and 3.2% at 10 years vs. 27.0% at 5 years and 36.9% at 10 years); however the differences in OS were too small to be regarded as different (60.2% at 5 years and 27.3% at 10 years vs. 59.7% at 5 years and 32.3% at 10 years). Considering the difference in ratios of hepatitis B (10.9% vs. 69.9%) and hepatitis C (74.4% vs. 17.9%) viral infections, and favorable prognosis of hepatitis B virus-induced liver cirrhosis, which was also evidenced by our univariate analysis, OS survival in our study may be worse. However, it does
Journal of Hepatology 2013 vol. 58 j 89–97
95
Research Article Table 4. Complications after percutaneous RFA of HCC as a first-line therapy.
Number of complications
Number of major complications
Hepatic infarction
25 (1.9%)
3 (0.2%)
Biloma
18 (1.4%)
2 (0.2%)
Hepatic abscess
15 (1.1%)
7 (0.5%)
Intraperitoneal hemorrhage
13 (1.0%)
1 (0.1%)
Bile duct stricture
10 (0.8%)
0 (0%)
Diaphragm injury
5 (0.4%)
1 (0.1%)
Cancer seeding
5 (0.4%)
5 (0.4%)
Pneumothorax
4 (0.3%)
0 (0%)
Hemobilia
3 (0.2%)
0 (0%)
Hematoma in the abdominal wall
3 (0.2%)
1 (0.1%)
Thermal injury of the hepatic flexure colon
2 (0.2%)
2 (0.2%)
Hepatic arterial pseudoaneurysm
2 (0.2%)
1 (0.1%)
Hemothorax
1 (0.1%)
1 (0.1%)
Acute cholecystitis
1 (0.1%)
1 (0.1%)
Perihepatic abscess due to biloma rupture
1 (0.1%)
1 (0.1%)
Total (n = 1305)
108 (8.3%)
26 (2.0%)
not seem appropriate to draw affirmative conclusion because other important factors, for instance, antiviral therapy, were not standardized between the two studies. Much lower LTP rates might be related to a difference in the natural course of HCC caused by different etiologies or the strategy coping with insufficient ablative margin after RFA immediately after treatment. The annual average size of the treated tumor decreased over time from 2.6 cm in 1999 to 1.9 cm in 2011, as shown in Supplementary Fig. 1. We guess that this slow decrease should be attributed to the technological developments of the diagnostic modalities including a contrast agent, e.g., gadoxetic acid-enhanced 3.0-tesla MRI of the liver, which enabled to diagnose smaller HCCs [19]. Our study had several limitations. First, immediate follow-up examinations were performed with contrast-enhanced US in 15.2% of our study population. Therefore, some of the immediate minor complications may not have been detected. Second, we could not include insufficiency of ablative margin as a risk factor for LTP because objective evaluation was not feasible in this huge population. Instead, we included tumor size, which was likely to be closely correlated with ablative margin. Third, we diagnosed LTP according to the definition by the standardized criteria [20]. However, this definition is limited in differentiating true LTP from new HCC occurrences abutting the margin of the RFA zone. This might have contributed to higher rates of LTP in our study. Fourth, the evaluation of ER was performed when specific clinical or radiological features were noticed. This procedure might underestimate the occurrence of ER. Finally, the majority of HCC tumors were not histopathologically confirmed, but were considered to be HCC on the basis of clinical criteria [2,13]. However, we think this was an insurmountable limitation in real clinical practices of HCC management. In conclusion, 10-year survival outcomes after percutaneous radiofrequency ablation as a first-line therapeutic option of hepatocellular carcinoma are excellent and comparable to those of
96
surgery, even though tumor recurrences are relatively frequent. Poor overall survival turned out to be associated with old age, Child-Pugh class B, absence of antiviral therapy during the follow-up period, or occurrence of extrahepatic recurrence. Aggressive antiviral therapy after radiofrequency ablation appears to prolong survival of the patients if the underlying liver disease is related to hepatitis viral infection.
Financial support This study was funded by Samsung Biomedical Research Institute Grant (C-B0-231-1).
Conflict of interest The Authors who have taken part in this study do not have a relationship with the manufacturers of the drugs involved either in the past or present and did not receive funding from the manufacturers to carry out their research. The Authors received support from Samsung Biomedical Research Institute Grant (C-B0-231-1). Acknowledgments The authors wish to thank Jung-eun Kim, R.N. for her assistance with this work.
Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jhep.2012. 09.020.
Journal of Hepatology 2013 vol. 58 j 89–97
JOURNAL OF HEPATOLOGY References [1] Rossi S, Di Stasi M, Buscarini E, Cavanna L, Quaretti P, Squassante E, et al. Percutaneous radiofrequency interstitial thermal ablation in the treatment of small hepatocellular carcinoma. Cancer J Sci Am 1995;1:73–81. [2] Bruix J, Sherman M. Management of hepatocellular carcinoma: an update. Hepatology 2011;53:1020–1022. [3] Cho YK, Kim JK, Kim WT, Chung JW. Hepatic resection versus radiofrequency ablation for very early stage hepatocellular carcinoma: a Markov model analysis. Hepatology 2010;51:1284–1290. [4] Liu Z, Zhou Y, Zhang P, Qin H. Meta-analysis of the therapeutic effect of hepatectomy versus radiofrequency ablation for the treatment of hepatocellular carcinoma. Surg Laparosc Endosc Percutan Tech 2010;20:130–140. [5] Molinari M, Helton S. Hepatic resection versus radiofrequency ablation for hepatocellular carcinoma in cirrhotic individuals not candidates for liver transplantation: a Markov model decision analysis. Am J Surg 2009;198:396–406. [6] Wang JH, Wang CC, Hung CH, Chen CL, Lu SN. Survival comparison between surgical resection and radiofrequency ablation for patients in BCLC very early/early stage hepatocellular carcinoma. J Hepatol 2012;56:412–418. [7] Zhou Y, Zhao Y, Li B, Xu D, Yin Z, Xie F, et al. Meta-analysis of radiofrequency ablation versus hepatic resection for small hepatocellular carcinoma. BMC Gastroenterol 2010;10:78. [8] Lencioni R, Crocetti L. Local-regional treatment of hepatocellular carcinoma. Radiology 2012;262:43–58. [9] Shiina S, Tateishi R, Arano T, Uchino K, Enooku K, Nakagawa H, et al. Radiofrequency ablation for hepatocellular carcinoma: 10-year outcome and prognostic factors. Am J Gastroenterol 2012;107:569–577. [10] Ng KK, Poon RT, Lo CM, Yuen J, Tso WK, Fan ST. Analysis of recurrence pattern and its influence on survival outcome after radiofrequency ablation of hepatocellular carcinoma. J Gastrointest Surg 2008;12:183–191. [11] Livraghi T, Meloni F, Di Stasi M, Rolle E, Solbiati L, Tinelli C, et al. Sustained complete response and complications rates after radiofrequency ablation of
[12]
[13] [14]
[15]
[16]
[17]
[18]
[19]
[20]
very early hepatocellular carcinoma in cirrhosis: Is resection still the treatment of choice? Hepatology 2008;47:82–89. Choi D, Lim HK, Rhim H, Kim YS, Lee WJ, Paik SW, et al. Percutaneous radiofrequency ablation for early-stage hepatocellular carcinoma as a firstline treatment: long-term results and prognostic factors in a large singleinstitution series. Eur Radiol 2007;17:684–692. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology 2005;42:1208–1236. Rhim H, Lee MH, Kim YS, Choi D, Lee WJ, Lim HK. Planning sonography to assess the feasibility of percutaneous radiofrequency ablation of hepatocellular carcinomas. AJR 2008;190:1324–1330. Cucchetti A, Zanello M, Cescon M, Ercolani G, Del Gaudio M, Ravaioli M, et al. Improved diagnostic imaging and interventional therapies prolong survival after resection for hepatocellular carcinoma in cirrhosis: the university of bologna experience over 10 years. Ann Surg Oncol 2011;18:1630–1637. Yamamoto J, Kosuge T, Saiura A, Sakamoto Y, Shimada K, Sano T, et al. Effectiveness of hepatic resection for early-stage hepatocellular carcinoma in cirrhotic patients: subgroup analysis according to Milan criteria. Jpn J Clin Oncol 2007;37:287–295. Sala M, Llovet JM, Vilana R, Bianchi L, Sole M, Ayuso C, et al. Initial response to percutaneous ablation predicts survival in patients with hepatocellular carcinoma. Hepatology 2004;40:1352–1360. Lam VW, Ng KK, Chok KS, Cheung TT, Yuen J, Tung H, et al. Risk factors and prognostic factors of local recurrence after radiofrequency ablation of hepatocellular carcinoma. J Am Coll Surg 2008;207:20–29. Kim JE, Kim SH, Lee SJ, Rhim H. Hypervascular hepatocellular carcinoma 1 cm or smaller in patients with chronic liver disease: characterization with gadoxetic acid-enhanced MRI that includes diffusion-weighted imaging. AJR Am J Roentgenol 2011;196:W758–765. Goldberg SN, Grassi CJ, Cardella JF, Charboneau JW, Dodd 3rd GD, Dupuy DE, et al. Image-guided tumor ablation: standardization of terminology and reporting criteria. Radiology 2005;235:728–739.
Journal of Hepatology 2013 vol. 58 j 89–97
97
Ten-year outcomes of percutaneous radiofrequency ablation as first-line therapy of early hepatocellular carcinoma: Analysis of prognostic factors Young-sun Kim1, Hyo Keun Lim1,⇑, , Hyunchul Rhim1,⇑, , Min Woo Lee1, Dongil Choi1, Won Jae Lee1, Seung Woon Paik2, Kwang Cheol Koh2, Joon Hyeok Lee2, Moon Seok Choi2, Geum-Youn Gwak2, Byung Chul Yoo2 1
Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea; 2Department of Medicine and Digestive Disease Research Center, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
Background & Aims: The aim was to assess 10-year outcomes of radiofrequency ablation as a first-line therapy of early-stage hepatocellular carcinoma with an analysis of prognostic factors. Methods: From April 1999 to April 2011, 1305 patients (male: female = 993:312; mean age, 58.4 years) with 1502 early-stage hepatocellular carcinomas (mean size, 2.2 cm) were treated with percutaneous radiofrequency ablation as a first-line option. Follow-up period ranged from 0.4 to 146.6 months (median, 33.4 months). We assessed the 10-year follow-up results of recurrences and survival with the analyses of prognostic factors. Results: Recurrences occurred in 795 patients (1–17 times), which were managed with various therapeutic modalities. The cumulative local tumor progression rates were 27.0% and 36.9% at 5 and 10 years, respectively, for which the only significant risk factor was large tumor size (B = 0.584, p = 0.001). Cumulative intrahepatic distant and extrahepatic recurrence rates were 73.1% and 88.5%, and 19.1% and 38.2% at 5 and 10 years, respectively. Corresponding overall survival rates were 59.7% and 32.3%, respectively. Poor survival was associated with old age (B = 0.043, p = 0.010), Child-Pugh class B (B = 1.054, p <0.001), absence of antiviral therapy during follow-up (B = 0.699, p = 0.034), and presence of extrahepatic recurrence (B = 0.971, p = 0.007).
Keywords: Hepatocellular carcinoma; Radiofrequency ablation; Outcome analysis; Ten-year survival; Prognostic factor. Received 30 January 2012; received in revised form 7 September 2012; accepted 13 September 2012; available online 27 September 2012 ⇑ Corresponding authors. Address: Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, #50, Irwon-dong, Gangnam-gu, Seoul 138-225, Republic of Korea. Tel: +82 2 3410 2518; fax: +82 2 3410 6368. E-mail addresses: [email protected] (H.K. Lim), [email protected] (H. Rhim). These authors contributed equally to this work. Abbreviations: RFA, radiofrequency ablation; HCC, hepatocellular carcinoma; LTP, local tumor progression; IDR, intrahepatic distant recurrence; ER, extrahepatic recurrence; OS, overall survival; US, ultrasonography; CT, computed tomography; MRI, magnetic resonance imaging; AFP, alpha-fetoprotein; BCLC, Barcelona Clinic Liver Cancer.
Conclusions: Ten-year survival outcomes after percutaneous radiofrequency ablation as a first-line therapy of hepatocellular carcinoma were excellent despite frequent tumor recurrences. Overall survival was influenced by age, Child-Pugh class, antiviral therapy, or extrahepatic recurrence. Ó 2012 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Introduction Since the first clinical study in 1995 [1], clinical implications of percutaneous radiofrequency ablation (RFA) in the treatment of early-stage hepatocellular carcinoma (HCC) have expanded. Currently, RFA is recognized as a curative modality for early-stage HCC [2] whose outcomes are comparable to those of surgery, as evidenced by many studies [3–7]. Clinical roles of percutaneous RFA are actually more meaningful than those of surgery from the viewpoint that RFA could be applied to patients whose hepatic functional reserve is insufficient to endure surgery, and can be conducted repeatedly with the same patient [8]. Despite such a large potential, long-term (i.e., more than 10 years) follow-up results are rare [9] because of its relatively short history of clinical applications although there has been plenty of reports regarding 5-year outcomes [8,10–12]. During the last 12 years, our institution has performed percutaneous RFA of HCC in more than 3000 patients, and approximately more than 40% were performed as a first-line therapy. Through these clinical experiences, the best way to perform RFA has evolved toward developing an effective as well as safe treatment, for example, the use of artificial ascites techniques, multiple overlapping ablations with ‘‘switch box’’, or fusion image guidance. In addition, antiviral therapy has become more and more readily available, all of which seemingly affect therapeutic outcomes regarding not only recurrence but also overall survival (OS). These long-term changes raise the necessity for a new analysis of the longer-term outcomes. Therefore, the purposes of our study were to evaluate 10-year follow-up results with regard to recurrences and OS after
Journal of Hepatology 2013 vol. 58 j 89–97
Research Article percutaneous RFA as a first-line therapeutic option for the treatment of early-stage HCC, and to assess factors significantly influencing survival.
Patients and methods This study was approved by the institutional review board of our hospital, and informed consent from patients was waived because this was a retrospective clinical study. However, written informed consent for RFA procedures and the use of data for research purposes were obtained from patients prior to every treatment. Study population We reviewed the departmental database of RFA procedures and retrieved the following data. Between April 1999 and April 2011, a total of 10,334 planning ultrasonography (US) examinations were conducted by one of six radiologists (Y.S.K., H.K.L., H.R., M.W.L., D.C., and W.J.L.) on a referral basis, and among them, USguided percutaneous RFAs were performed for the treatment of a total of 5097 HCC tumors in 3084 patients (male:female = 2402:682 mean age, 58.5 years; age range, 24–89). Out of 5097 HCC tumors, 1502 tumors (mean size, 2.2 cm; size range, 0.5–4.9 cm) in 1305 consecutive patients (male:female = 993:312; mean age, 58.4 years; age range, 28–86 years), which were initially diagnosed and treated with US-guided percutaneous RFA as a first-line option, were analyzed in this study (Fig. 1). Two hundred six tumors (206/1502, 13.7%) were confirmed as HCC by means of a US-guided core needle biopsy, and the remaining 1296 tumors (1296/1502, 86.3%) were considered to be HCC, based on one of two clinical criteria from the American Association for the Study of Liver Diseases according to the time of RFA procedures [2,13].
Our institutional inclusion criteria for percutaneous RFA procedures were as follows: (1) presence of a single nodular HCC <5 cm in maximum diameter; (2) presence of multinodular HCCs (63 in number, each <3 cm in maximum diameter); (3) absence of portal venous thrombosis; (4) Child-Pugh class A or B; and (5) prothrombin time ratio >50% (prothrombin time with an international normalized ratio <1.7) and platelet count >50,000 cells/mm3 (50 cells 109/L). The decision to perform RFA was made by consensus of an interdisciplinary discussion. In cases of a single HCC in a Child-Pugh class A patient with neither portal hypertension nor hyperbilirubinemia, surgical resection was primarily considered. However, patients’ age, co-morbidity and preference were also taken into account. The technical feasibility of RFA was assessed with planning US exam [14] by the radiologist who would perform the procedure. The presence of ascites was not either a relative or an absolute contraindication. Clinical characteristics of the patients and tumors analyzed in this study are summarized in Table 1. RFA procedure All RFA procedures were performed percutaneously under US guidance. Procedures were performed on an inpatient basis by one of six radiologists (Y.S.K., H.K.L., H.R., M.W.L., D.C., and W.J.L.), who had at least seven years of experience with this procedure, by the end of the study period. Details about RFA procedures are reported as Supplementary data. Follow-up after RFA For early evaluation of the therapeutic response or possible complications, either contrast-enhanced US (n = 199, 15.2%), CT (n = 1091, 83.6%), or MRI (n = 15, 1.5%) was performed within 24 h. Between July 2002 and December 2003, contrastenhanced US using a microbubble contrast agent (SH U508A, Levovist; Bayer
Planning US for percutaneous RFA of HCC n = 10,334
Feasible n = 4460
Not feasible n = 5874
Choice of other modalities or follow-up loss, n = 59
Percutaneous RFA of HCC 3084 patients with 5097 tumors (4401 procedures)
Recommendation of follow-up TACE Surgical resection Percutaneous RFA with TACE Radiation therapy Liver transplantation Intraoperative RFA Percutaneous ethanol injection Follow-up loss
Percutaneous RFA as a first-line option
Percutaneous RFA for recurrence
1305 patients with 1502 tumors (1305 procedures)
1779 patients with 3595 tumors (3096 procedures)
n = 783 n = 3054 n = 841 n = 142 n = 78 n = 49 n = 41 n = 12 n = 874
Final inclusion Fig. 1. Flow of study inclusion. A total of 10,334 patients were examined with planning US, and 1305 patients with 1502 HCC tumors, treated with percutaneous RFA as a first-line option, were finally included.
90
Journal of Hepatology 2013 vol. 58 j 89–97
JOURNAL OF HEPATOLOGY Table 1. Baseline features of the study population.
Clinical Features
Numbers
Patients
n = 1305
Male:female
993:312
Age (yr)
mean 58.4, range 28-86
Liver disease
A follow-up study was performed 1 month later. If RFA was considered to be technically effective, a follow-up was repeated every 3 months until two years after RFA. After then, follow-ups were conducted every 4–6 months according to the risks for recurrences. Chest radiographs were also checked at each follow-up. Patients who showed specific clinical or radiological features such as headache for brain metastasis or abnormality in chest radiograph for pulmonary metastasis underwent appropriate imaging examinations for the detection of extrahepatic recurrence (ER). Treatment strategy after initial RFA
Liver cirrhosis
1077
Acute or chronic hepatitis
103
Healthy carrier
75
None
50
When an unablated residual tumor was observed at the immediate follow-up, additional RFA was primarily chosen. Unless percutaneous RFA was technically feasible, other therapeutic modalities were carried out. If the index tumor was completely covered by the ablation zone, the treatment was regarded as a technical success regardless of the sufficiency of ablative margin. If LTP or intrahepatic distant recurrence (IDR) was observed during subsequent follow-up visits, an additional treatment was immediately performed, preferably using percutaneous RFA.
Etiology of liver disease Hepatitis B virus
912
Hepatitis C virus
233
Hepatitis non-B, non-C virus
33
Hepatitis B and C viruses
7
Alcohol
59
Autoimmune hepatitis
1
Cryptogenic
10
No liver disease
50
Analysis of therapeutic efficacy and survival Local therapeutic efficacy in terms of technique effectiveness and LTP was assessed on a tumor basis. IDR, ER, event-free survival, and OS were evaluated on a per-patient basis. If the patient underwent liver transplantation, we regarded the operation date as the end of follow-up for LTP and IDR. The OS time was defined as the interval between the first RFA and either death or last follow-up visit before August 31, 2011. If death resulted from unrelated causes such as a car accident, that data was disregarded. Event-free survival was defined as the time from the first RFA to either the earliest event (i.e., LTP, IDR, ER or death) or the last follow-up date without an event.
Child-Pugh classification Class A
1021
Class B
284
Definitions of terminology
Baseline serum AFP level* Range
1-13,563.1 ng/ml
Mean ± standard deviation
207.8 ± 796.2 ng/ml
≤20 ng/ml
625
>20 ng/ml, ≤200 ng/ml
429
>200 ng/ml
213
Number of tumors One
1122
Two
169
Three
14
BCLC stage Stage 0
427
Stage A
878
Tumors
n = 1502
Size Range
0.5-4.9 cm
Mean ± standard deviation
2.2 ± 0.8 cm
Diagnosis of hepatocellular carcinoma Histopathology
89
Edmondson-Steiner grade II
100
Edmondson-Steiner grade III
8
Edmondson-Steiner grade IV
9
Clinical criteria ⁄
206
Edmondson-Steiner grade I
1296
Available in 1267 patients.
Schering Pharma, Berlin, Germany) was predominantly conducted for research purposes. In other periods, triple-phase contrast-enhanced CT was the preferred method; however, if the patient had a contraindication to CT, MRI was performed.
Definitions are based on the standardization by the International Working Group on Image-Guided Tumor Ablation [20]. Technical success was defined when the tumor was treated according to the protocol and was completely replaced by RFA zones at the immediate follow-up. Achievement of technique effectiveness was defined when complete ablation of macroscopic tumors was evident on the 1-month follow-up exams for those patients whose HCC tumors were completely removed by the initial RFA treatment. LTP was diagnosed when a follow-up exam demonstrated findings of interval development/growth of the tumor along the margin of the ablation zone where the RFA had been considered to be technically effective. IDR was defined by a lesion with similar characteristics but not contacting the original ablation zone in the liver. Cancer seeding was regarded as extrahepatic recurrence. Major complication was defined as an event that leads to substantial morbidity and disability, increasing the level of care, or results in hospital admission or substantially lengthens the hospital stay. All other complications were regarded as minor. Statistical analysis The sizes of RFA-treated tumors in the first and last years were compared using the Mann–Whitney test. The cumulative rates for each type of recurrence and survival were estimated with the Kaplan–Meier method. Prognostic factors for LTP were assessed using the multivariate Cox proportional hazard model. Potential prognostic factors for LTP included sex, age, etiology of liver disease, Child-Pugh class, tumor size, Edmondson-Steiner grade (if histopathologically confirmed), baseline serum a-fetoprotein level (AFP), and AFP changes within 30 days in response to RFA (group 1, initially >20 ng/ml and decrease to <2/3 of the baseline after RFA; group 2, initially >20 ng/ml and increase, no change or decrease to P2/3 of the baseline after RFA; group 3, consistently 620 ng/ml regardless of RFA). OS curves of different BCLC (Barcelona Clinic Liver Cancer) stages as well as each potential prognostic factor were compared with the log rank test as univariate analysis. Prognostic factors for OS were also assessed using the multivariate Cox proportional hazard model. Potential prognostic factors for OS included sex, age, etiology of liver disease, Child-Pugh class, number and the largest diameter of the RFA-treated tumors, Edmondson-Steiner grade (if histopathologically confirmed; higher grade if multiple), baseline AFP, AFP changes within 30 days in response to RFA, antiviral therapy during follow-up period, and presence of LTP, IDR, and ER after RFA.
Journal of Hepatology 2013 vol. 58 j 89–97
91
Research Article A p-value <0.05 was considered to be statistically significant. Data analyses were performed using the commercially available software (SPSS, ver. 19.0; SPSS Inc., Chicago, IL).
Results The average size of the index tumors decreased each year from 2.6 cm in 1999 to 1.9 cm in 2011 (Supplementary Fig. 1), which was significantly different (1999 vs. 2011, p = 0.001). The median follow-up period for 1305 patients was 33.4 months (range, 0.4– 146.6 months; mean, 38.7 months). Two patients died within 30 days after RFA treatment because of hypovolemic shock caused by variceal bleeding and unusually rapid progression of HCC, respectively. Fifty-two out of 1305 patients (4.0%) underwent liver transplantation 26.4 ± 18.1 months (range, 3.4– 76.4 months) after percutaneous RFA. Technical success Technical success was achieved in one session in 94.8% of cases (1237/1305). Sixty-seven cases (67/1305, 5.1%) with 68 tumors (68/1502, 4.5%) turned out to be technical failures due to either residual unablated tumors (n = 54) or mistargeting (n = 14) at immediate follow-up examinations. Procedures were stopped for four cases (4/1305, 0.3%) because of unstable vital signs, which in turn left residual unablated tumors at immediate follow-up. We repeated percutaneous RFA in 51 cases with 52 tumors and finally succeeded in completely ablating the index tumors in 47 cases (47/51, 92.2%) with 48 tumors. The remaining 20 cases with 20 untreated or residual tumors were managed with chemoembolization (n = 18) or surgical resection (n = 2). Recurrences and managements Of 1283 patients who were completely treated with RFA and survived more than 30 days, recurrences occurred in 795 patients (62.0%, 795/1283). Among them, 41 cases were lost to followup or refused treatment, thus 754 patients with 1st recurrence (LTP, n = 154; IDR, n = 535; LTP and IDR, n = 55; LTP and ER, n = 2; IDR and ER, n = 8) were managed with the various therapeutic modalities. Among the 754 patients, 245 (32.5%, 245/ 754) showed no recurrence, and 509 (67.5%, 509/754) experienced 2nd recurrence. Recurrences were observed up to 17 times (n = 2) as of August 31, 2011, and were also managed with the various therapeutic modalities as shown in Fig. 2. Local therapeutic efficacy Among 1482 tumors which were regarded as to be completely ablated by the initial RFA treatments, 20 showed viable residues at 1 month follow-up; therefore, the technique effectiveness rate was 98.7% (1462/1482). Of 1462 tumors with complete technique effectiveness, LTP was identified in 284 (284/1462, 19.4%) from 2.4 months to 113.2 months. The mean estimate of LTP-free survival was 105.3 months (95% confidence interval (95% CI) 100.7– 110.0 months). Cumulative rates of LTP were estimated as 9.7% (95% CI 8.9–10.5%), 21.4% (20.1–22.7%), 27.0% (25.4–28.6%), 30.2% (29.2–33.6%), and 36.9% (32.5–41.3%) at 1, 3, 5, 8, and 10 years, respectively. Multivariate analysis revealed that large
92
tumor size was the only independently significant risk factor for LTP (B = 0.584, p = 0.001) (Fig. 3A) (Table 2). Initial LTPs were treated with chemoembolization (n = 165), percutaneous RFA (n = 103), combination therapy of percutaneous RFA with chemoembolization (n = 7), surgical resection (n = 5), liver transplantation (n = 1), intraoperative RFA (n = 1), percutaneous ethanol injection (n = 1), or radiation therapy (n = 1) with a mean delay of 14.2 days after diagnosis. Remote recurrences IDRs were observed in 715 of 1305 (54.8%) patients. The first IDRs occurred 0.7–99.7 months after RFA. Mean and median estimates of the IDR-free survival time were 43.6 months (95% CI 40.0– 47.1 months) and 27.4 months (24.8–30.1 months), respectively. Cumulative rates of IDR were estimated as 24.4% (95% CI, 23.2– 25.6%), 59.5% (57.9–61.1%), 73.1% (71.3–74.9%), 87.0% (84.7– 89.3%), and 88.5% (86.1–90.9%) at 1, 3, 5, 8, and 10 years, respectively (Fig. 3B). The initial IDRs were managed with conventional chemoembolization (n = 372), percutaneous RFA (n = 243), combination therapy of percutaneous RFA and chemoembolization (n = 38), surgical resection (n = 15), systemic chemotherapy (n = 9), liver transplantation (n = 8), radiation therapy (n = 3), intraoperative RFA (n = 2), or percutaneous ethanol injection therapy (n = 2). Twenty patients did not receive any specific treatment prior to their death. The remaining three patients were lost to follow-up. ERs including cancer seeding were found in 165 of 1305 (12.6%) patients during follow-up. ERs occurred 1.0– 130.6 months after treatment, and the mean estimate of ER-free survival time was 113.3 months (95% CI 107.6–119.1 months). Estimated cumulative rates were 2.7% (95% CI 2.2–3.2%), 10.2% (9.2–11.2%), 19.1% (17.5–20.7%), 28.5% (25.8–31.2%), and 38.2% (32.9–43.5%) for 1, 3, 5, 8, and 10 years, respectively (Fig. 3B). Cancer seedings were detected in 5 cases with a mean delay of 19.6 months (range 7.1–38.2 months). Long-term survival Of a total of 1305 patients, 407 (31.2%) died of HCC progression (n = 333), hepatic failure or complications of liver cirrhosis, such as variceal bleeding (n = 39) or systemic or other organ complications, such as sepsis or cerebral hemorrhage (n = 35). Time of death after RFA ranged between 0.4 and 132.0 months with a mean survival time of 82.9 months (95% CI 78.4–87.4 months) and a median survival time of 75.0 months (66.8–83.1 months). Thirty-seven patients (37/1305, 2.8%) were lost during followup, and 861 (66.0%) were alive as of August 31, 2011. The number of 5- and 10-year survivors was 219 and 13, respectively. Cumulative OS rates were estimated as 95.5% (95% CI 94.9–96.1%), 77.9% (76.6–79.2%), 59.7% (57.9–61.5%), 43.2% (40.8–45.6%), and 32.3% (28.5–36.1%) at 1, 3, 5, 8, and 10 years, respectively (Fig. 4A). Estimates of OS of the subgroups according to each potential prognostic factor and the results of univariate comparisons are summarized in Supplementary Table 2. By multivariate analysis, old age (B = 0.043, p = 0.010), Child-Pugh class B (B = 1.054, p <0.001), absence of antiviral therapy during follow-up (B = 0.699, p = 0.034), and presence of ER (B = 0.971, p = 0.007) were independently significant poor prognostic factors for OS (Table 3). Patients with BCLC stage 0 showed a significantly better OS than patients with BCLC stage A (p = 0.002) (Fig. 4B).
Journal of Hepatology 2013 vol. 58 j 89–97
JOURNAL OF HEPATOLOGY Percutaneous RFA as a first-line option n = 1305 Mortality within 30 days (n = 2) Untreated or residual tumor after RFA (n = 20)
1st recurrence, n = 795
No recurrence, n = 488
No treatment (n = 41) Treatment of 1st recurrence, n = 754 n = 154 IDR
LTP
RFA n = 79 TACE n = 66 RFA after TACE n = 4 SR n=1 PEI n=1
n = 535
TACE RFA RFA after TACE LT SR Others
LTP + IDR n = 55
n = 270 n = 220 n = 26 n=7 n=5 n=8
TACE RFA SR Chemo.
(LTP or IDR) + ER n = 10 TACE+Chemo. RT+Chemo. TACE+RT Chemo.
n = 42 n = 10 n=2 n=1
2nd recurrence, n = 509
n=6 n=2 n=1 n=1
No recurrence, n = 245
No treatment (n = 22) Treatment of 2nd recurrence, n = 487 LTP
n = 74
TACE RFA RT SR RFA after TACE
n = 43 n = 20 n=6 n=4 n=1
IDR
n = 405
TACE RFA RFA after TACE LT SR Others
n = 264 n = 109 n = 13 n=6 n=4 n=9
LTP + IDR TACE
ER
n=3
n=2
Chemo.
n=3
IDR + ER
n=3
RT Chemo.
n=2 n=1
n=2
No recurrence, n = 143
3rd recurrence, n = 344 No treatment (n = 10) Treatment of 3rd recurrence, n = 334 LTP TACE RFA SR RFA after TACE
n = 28 n = 17 n=8 n=2 n=1
IDR TACE RFA RFA after TACE LT RT SR
n = 302 n = 233 n = 53 n=5 n=4 n=4 n=3
ER Chemo. RT
n=4 n=2 n=2
4th recurrence, n = 242
5th recurrence, n = 188
6th recurrence, n = 131
9th recurrence, n = 42
8th recurrence, n = 60
7th recurrence, n = 97
10th recurrence, n = 27
11th recurrence, n = 19
12th recurrence, n = 10
15th recurrence, n = 4
14th recurrence, n = 5
13th recurrence, n = 5
16th recurrence, n = 3
17th recurrence, n = 2
Fig. 2. Recurrences and management after percutaneous RFA of HCC. After percutaneous RFA in 1305 patients, recurrences occurred in 795 patients up to 17 times, which were managed with various therapeutic modalities. RFA, radiofrequency ablation; TACE, transcatheter arterial chemoembolization; PEI, percutaneous ethanol injection; SR, surgical resection; LT, liver transplantation; Chemo., Systemic chemotherapy; RT, radiation therapy; LTP, local tumor progression; IDR, intrahepatic distant recurrence; ER, extrahepatic recurrence.
Journal of Hepatology 2013 vol. 58 j 89–97
93
Research Article Cumulative event-free survival rates were estimated as 66.8% (95% CI 65.5–68.1%), 29.0% (27.5–30.5%), 17.5% (16.1–18.9%), 7.0% (5.5–8.5%), and 3.8% (2.2–5.4%) at 1, 3, 5, 8, and 10 years, respectively (Fig. 4C).
1.0 Cumulative local tumor progression rate
A
0.8 0.6
Complications
0.4
Following 1305 RFA treatments, a total of 108 (8.3%) complications were encountered during follow-up. Of these, 26 (2.0%) major complications were identified (Table 4). There were no procedure-related mortalities.
0.2 0.0 0 20 40 60 80 100120140 Duration (mo)
Cumulative recurrence rate
B 1.0 0.8 0.6 0.4 0.2 0.0
0 20 40 60 80 100120140 Duration (mo) Fig. 3. Recurrence curves. (A) Local tumor progression of 1462 HCC tumors after percutaneous RFA as a first-line therapy. Cumulative rates of local tumor progression were estimated as 9.7%, 21.4%, 27.0%, 30.2%, and 36.9% at 1, 3, 5, 8, and 10 years, respectively. Local tumor progressions according to tumor size were also demonstrated. (B) Intrahepatic distant recurrence and extrahepatic recurrence in 1305 patients with HCC, treated with percutaneous RFA as a first-line therapy. Cumulative rates of intrahepatic distant recurrence were estimated as 24.4%, 59.5%, 73.1%, 87.0%, and 88.5% at 1, 3, 5, 8, and 10 years, respectively. The estimated cumulative rates of extrahepatic recurrence were 2.7%, 10.2%, 19.1%, 28.5%, and 38.2% for 1, 3, 5, 8, and 10 years, respectively. +, censored data.
Discussion Although controversial, it is generally accepted that 5-year OS outcomes after percutaneous RFA of HCC are comparable to those of surgical resection in cases of small tumors [3–7,11]. However, because of a relatively short history of clinical application of RFA, most publications have dealt with at most 5-year results. To compare the results of our study with those of surgical resection, we investigated the surgical literature with 10-year follow-up, and we confined our search to those publications fulfilling the Milan criteria. Studies by Cucchetti et al. [15] and Yamamoto et al. [16] reported that 5-/10-year OS rates were 61.5%/28.7% and 56%/20%, respectively. However, both populations had a higher proportion of Child-Pugh class A than our study (95.7 and 85.3% vs. 78.2%), which might play a role as a favorable confounder for OS. Taking into account this discrepancy in demographics, our long-term survival outcome after percutaneous RFA of HCC, 59.7% at 5 years and 32.3% at 10 years, appears to be comparable to those of surgical resection. There have been several studies with regard to the effects of residual or recurrent HCC or treatment on survival. Sala et al. [17] reported that the initial complete response (i.e., absence of residual tumor) after percutaneous ablation therapy significantly improved survival. Ng et al. [10] demonstrated that different seg-
Table 2. Results of univariate and multivariate analysis on potential prognostic factors for LTP after percutaneous RFA of HCC, as first-line therapy.
Univariate analysis*
Multivariate analysis**
p value
Regression coefficient
Standard error
p value
Age†
0.393
-0.29
0.020
0.147
Gender
0.930
0.542
0.426
0.204
Etiology of liver disease
0.099
0.520
Child-Pugh class
0.203
-0.245
0.440
0.577
Tumor size‡
<0.001§§
0.584
0.180
0.001§§
Edmondson-Steiner grade
0.027§§
Baseline AFP§
0.897
0.128
0.387
AFP change in response to RFA
0.806
0.060
⁄
Log rank test. ⁄⁄ Cox proportional hazard model. Grouped based on median age (i.e.,658;>58) for univariate analysis. à Grouped into 61 cm, >1 and 62 cm, >2 and 63 cm, >3 and 64 cm, or >4 and 65 cm for univariate analysis. § Grouped into 620 ng/ml, >20 and 6200 ng/ml, or >200 ng/ml for univariate analysis. §§ Statistically significant.
94
Journal of Hepatology 2013 vol. 58 j 89–97
0.741 0.777
JOURNAL OF HEPATOLOGY A
B
0.8 0.6 0.4 0.2
Cumulative event-free survial rate
Cumulative overall survial rate
Cumulative overall survial rate
C 1.0
1.0
0.8 0.6 0.4 0.2 0.0
0.0 0
20
40 60 80 100 120 140 Duration (mo)
1.0 0.8 0.6 0.4 0.2 0.0
0
20
40 60 80 100 120 140 Duration (mo)
0
25
50 75 100 Duration (mo)
125
Fig. 4. Survival curves. (A) Overall survival curve of 1305 patients with HCC treated with percutaneous RFA. Cumulative overall survival rates were estimated as 95.5%, 77.9%, 59.7%, 43.2%, and 32.3% at 1, 3, 5, 8, and 10 years, respectively. (B) Patients with BCLC stage 0 demonstrated a significantly better overall survival than those with BCLC stage A. (C) Event-free survival curve of 1305 patients. The estimated cumulative rates for event-free survival were 66.8%, 29.0%, 17.5%, 7.0%, and 3.8% for 1, 3, 5, 8, and 10 years, respectively. +, censored data.
Table 3. Results of univariate and multivariate analysis on potential prognostic factors for overall survival after percutaneous RFA of HCC as a first-line therapy.
Univariate analysis*
Age†
⁄
Multivariate analysis** Regression coefficient Standard error
p value
<0.001§§
0.043
0.017
0.010§§
-0.059
0.310
0.849
-1.054
0.299
<0.001§§
Gender
0.089
Etiology of liver disease
<0.001§§
Child-Pugh class
<0.001§§
0.926
Number of tumors treated
0.579
0.052
0.315
0.868
Maximal tumor size treated‡
<0.001§§
0.235
0.165
0.155
Edmondson-Steiner grade
0.872
Baseline AFP§
0.205
-0.365
0.327
0.265
AFP change in response to RFA
<0.001§§
Antiviral therapy during follow-up
<0.001§§
0.699
0.329
0.034§§
0.369 0.220
Occurrence of local tumor progression
0.108
0.296
0.305
0.333
Occurrence of intrahepatic distant recurrence
0.361
-0.589
0.330
0.074
Occurrence of extrahepatic recurrence
<0.001§§
0.971
0.358
0.007§§
Log rank test; post hoc analyses are available in Supplementary Table 2.
⁄⁄
Cox proportional hazard model. Grouped based on median age (i.e., 658; >58) for univariate analysis. à Grouped into 61 cm, >1 and 62 cm, >2 and 63 cm, >3 and 64 cm, or >4 and 65 cm for univariate analysis. § Grouped into 620 ng/ml, >20 and 6200 ng/ml, or >200 ng/ml for univariate analysis. §§ Statistically significant.
ment intrahepatic recurrence and distant metastasis after RFA carried significant poor prognostic influence on OS outcome. In their study, LTP failed to show statistical significance by multivariate analysis. Lam et al. [18] demonstrated that aggressive treatment of LTP after RFA improved OS outcome. Our study showed that either LTP or IDR had no influence, and ER had a negative influence on survival based on both univariate and multivariate analyses. Our results with regards to LTP and ER were compatible with the results from the study by Ng et al. [10]. We could compare our results to those from another recent study by Shiina et al. in which 10-year follow-up results of RFA
are available [9]. The study population consisted of 1170 patients with a median follow-up of 38.2 months. Their cumulative LTP rates were much lower than ours (3.2% at 5 years and 3.2% at 10 years vs. 27.0% at 5 years and 36.9% at 10 years); however the differences in OS were too small to be regarded as different (60.2% at 5 years and 27.3% at 10 years vs. 59.7% at 5 years and 32.3% at 10 years). Considering the difference in ratios of hepatitis B (10.9% vs. 69.9%) and hepatitis C (74.4% vs. 17.9%) viral infections, and favorable prognosis of hepatitis B virus-induced liver cirrhosis, which was also evidenced by our univariate analysis, OS survival in our study may be worse. However, it does
Journal of Hepatology 2013 vol. 58 j 89–97
95
Research Article Table 4. Complications after percutaneous RFA of HCC as a first-line therapy.
Number of complications
Number of major complications
Hepatic infarction
25 (1.9%)
3 (0.2%)
Biloma
18 (1.4%)
2 (0.2%)
Hepatic abscess
15 (1.1%)
7 (0.5%)
Intraperitoneal hemorrhage
13 (1.0%)
1 (0.1%)
Bile duct stricture
10 (0.8%)
0 (0%)
Diaphragm injury
5 (0.4%)
1 (0.1%)
Cancer seeding
5 (0.4%)
5 (0.4%)
Pneumothorax
4 (0.3%)
0 (0%)
Hemobilia
3 (0.2%)
0 (0%)
Hematoma in the abdominal wall
3 (0.2%)
1 (0.1%)
Thermal injury of the hepatic flexure colon
2 (0.2%)
2 (0.2%)
Hepatic arterial pseudoaneurysm
2 (0.2%)
1 (0.1%)
Hemothorax
1 (0.1%)
1 (0.1%)
Acute cholecystitis
1 (0.1%)
1 (0.1%)
Perihepatic abscess due to biloma rupture
1 (0.1%)
1 (0.1%)
Total (n = 1305)
108 (8.3%)
26 (2.0%)
not seem appropriate to draw affirmative conclusion because other important factors, for instance, antiviral therapy, were not standardized between the two studies. Much lower LTP rates might be related to a difference in the natural course of HCC caused by different etiologies or the strategy coping with insufficient ablative margin after RFA immediately after treatment. The annual average size of the treated tumor decreased over time from 2.6 cm in 1999 to 1.9 cm in 2011, as shown in Supplementary Fig. 1. We guess that this slow decrease should be attributed to the technological developments of the diagnostic modalities including a contrast agent, e.g., gadoxetic acid-enhanced 3.0-tesla MRI of the liver, which enabled to diagnose smaller HCCs [19]. Our study had several limitations. First, immediate follow-up examinations were performed with contrast-enhanced US in 15.2% of our study population. Therefore, some of the immediate minor complications may not have been detected. Second, we could not include insufficiency of ablative margin as a risk factor for LTP because objective evaluation was not feasible in this huge population. Instead, we included tumor size, which was likely to be closely correlated with ablative margin. Third, we diagnosed LTP according to the definition by the standardized criteria [20]. However, this definition is limited in differentiating true LTP from new HCC occurrences abutting the margin of the RFA zone. This might have contributed to higher rates of LTP in our study. Fourth, the evaluation of ER was performed when specific clinical or radiological features were noticed. This procedure might underestimate the occurrence of ER. Finally, the majority of HCC tumors were not histopathologically confirmed, but were considered to be HCC on the basis of clinical criteria [2,13]. However, we think this was an insurmountable limitation in real clinical practices of HCC management. In conclusion, 10-year survival outcomes after percutaneous radiofrequency ablation as a first-line therapeutic option of hepatocellular carcinoma are excellent and comparable to those of
96
surgery, even though tumor recurrences are relatively frequent. Poor overall survival turned out to be associated with old age, Child-Pugh class B, absence of antiviral therapy during the follow-up period, or occurrence of extrahepatic recurrence. Aggressive antiviral therapy after radiofrequency ablation appears to prolong survival of the patients if the underlying liver disease is related to hepatitis viral infection.
Financial support This study was funded by Samsung Biomedical Research Institute Grant (C-B0-231-1).
Conflict of interest The Authors who have taken part in this study do not have a relationship with the manufacturers of the drugs involved either in the past or present and did not receive funding from the manufacturers to carry out their research. The Authors received support from Samsung Biomedical Research Institute Grant (C-B0-231-1). Acknowledgments The authors wish to thank Jung-eun Kim, R.N. for her assistance with this work.
Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jhep.2012. 09.020.
Journal of Hepatology 2013 vol. 58 j 89–97
JOURNAL OF HEPATOLOGY References [1] Rossi S, Di Stasi M, Buscarini E, Cavanna L, Quaretti P, Squassante E, et al. Percutaneous radiofrequency interstitial thermal ablation in the treatment of small hepatocellular carcinoma. Cancer J Sci Am 1995;1:73–81. [2] Bruix J, Sherman M. Management of hepatocellular carcinoma: an update. Hepatology 2011;53:1020–1022. [3] Cho YK, Kim JK, Kim WT, Chung JW. Hepatic resection versus radiofrequency ablation for very early stage hepatocellular carcinoma: a Markov model analysis. Hepatology 2010;51:1284–1290. [4] Liu Z, Zhou Y, Zhang P, Qin H. Meta-analysis of the therapeutic effect of hepatectomy versus radiofrequency ablation for the treatment of hepatocellular carcinoma. Surg Laparosc Endosc Percutan Tech 2010;20:130–140. [5] Molinari M, Helton S. Hepatic resection versus radiofrequency ablation for hepatocellular carcinoma in cirrhotic individuals not candidates for liver transplantation: a Markov model decision analysis. Am J Surg 2009;198:396–406. [6] Wang JH, Wang CC, Hung CH, Chen CL, Lu SN. Survival comparison between surgical resection and radiofrequency ablation for patients in BCLC very early/early stage hepatocellular carcinoma. J Hepatol 2012;56:412–418. [7] Zhou Y, Zhao Y, Li B, Xu D, Yin Z, Xie F, et al. Meta-analysis of radiofrequency ablation versus hepatic resection for small hepatocellular carcinoma. BMC Gastroenterol 2010;10:78. [8] Lencioni R, Crocetti L. Local-regional treatment of hepatocellular carcinoma. Radiology 2012;262:43–58. [9] Shiina S, Tateishi R, Arano T, Uchino K, Enooku K, Nakagawa H, et al. Radiofrequency ablation for hepatocellular carcinoma: 10-year outcome and prognostic factors. Am J Gastroenterol 2012;107:569–577. [10] Ng KK, Poon RT, Lo CM, Yuen J, Tso WK, Fan ST. Analysis of recurrence pattern and its influence on survival outcome after radiofrequency ablation of hepatocellular carcinoma. J Gastrointest Surg 2008;12:183–191. [11] Livraghi T, Meloni F, Di Stasi M, Rolle E, Solbiati L, Tinelli C, et al. Sustained complete response and complications rates after radiofrequency ablation of
[12]
[13] [14]
[15]
[16]
[17]
[18]
[19]
[20]
very early hepatocellular carcinoma in cirrhosis: Is resection still the treatment of choice? Hepatology 2008;47:82–89. Choi D, Lim HK, Rhim H, Kim YS, Lee WJ, Paik SW, et al. Percutaneous radiofrequency ablation for early-stage hepatocellular carcinoma as a firstline treatment: long-term results and prognostic factors in a large singleinstitution series. Eur Radiol 2007;17:684–692. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology 2005;42:1208–1236. Rhim H, Lee MH, Kim YS, Choi D, Lee WJ, Lim HK. Planning sonography to assess the feasibility of percutaneous radiofrequency ablation of hepatocellular carcinomas. AJR 2008;190:1324–1330. Cucchetti A, Zanello M, Cescon M, Ercolani G, Del Gaudio M, Ravaioli M, et al. Improved diagnostic imaging and interventional therapies prolong survival after resection for hepatocellular carcinoma in cirrhosis: the university of bologna experience over 10 years. Ann Surg Oncol 2011;18:1630–1637. Yamamoto J, Kosuge T, Saiura A, Sakamoto Y, Shimada K, Sano T, et al. Effectiveness of hepatic resection for early-stage hepatocellular carcinoma in cirrhotic patients: subgroup analysis according to Milan criteria. Jpn J Clin Oncol 2007;37:287–295. Sala M, Llovet JM, Vilana R, Bianchi L, Sole M, Ayuso C, et al. Initial response to percutaneous ablation predicts survival in patients with hepatocellular carcinoma. Hepatology 2004;40:1352–1360. Lam VW, Ng KK, Chok KS, Cheung TT, Yuen J, Tung H, et al. Risk factors and prognostic factors of local recurrence after radiofrequency ablation of hepatocellular carcinoma. J Am Coll Surg 2008;207:20–29. Kim JE, Kim SH, Lee SJ, Rhim H. Hypervascular hepatocellular carcinoma 1 cm or smaller in patients with chronic liver disease: characterization with gadoxetic acid-enhanced MRI that includes diffusion-weighted imaging. AJR Am J Roentgenol 2011;196:W758–765. Goldberg SN, Grassi CJ, Cardella JF, Charboneau JW, Dodd 3rd GD, Dupuy DE, et al. Image-guided tumor ablation: standardization of terminology and reporting criteria. Radiology 2005;235:728–739.
Journal of Hepatology 2013 vol. 58 j 89–97
97