Intrahepatic recurrence after percutaneous radiofrequency ablation of hepatocellular carcinoma: Analysis of the pattern and risk factors

European Journal of Radiology 59 (2006) 432–441

Intrahepatic recurrence after percutaneous radiofrequency ablation of hepatocellular carcinoma: Analysis of the pattern and risk factors Young-sun Kim a,b , Hyunchul Rhim a,b,∗ , On Koo Cho b , Byung Hee Koh b , Yongsoo Kim b a

Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, South Korea b Department of Diagnostic Radiology, Hanyang University College of Medicine, South Korea Received 1 January 2006; received in revised form 5 March 2006; accepted 6 March 2006

Abstract Purpose: To evaluate the pattern and risks for intrahepatic recurrence after percutaneous radiofrequency (RF) ablation for hepatocellular carcinoma (HCC). Materials and methods: We studied 62 patients with 72 HCCs (≤4 cm) who were treated with percutaneous RF ablation. The mean follow-up period was 19.1 months (6.0–49.1). We assessed the incidence and cumulative disease-free survival of local tumor progression (LTP) and intrahepatic distant recurrence (IDR). To analyze the risk factors, we examined the following, for the LTP: (1) tumor diameter, (2) contact with vessels, (3) degree of approximation to hepatic hilum, (4) contact with hepatic capsule, (5) presence of ablative safety margin, (6) degree of benign periablational enhancement and (7) serum alpha-fetoprotein; for the IDR: (1) severity of hepatic disease, (2) presence of HBsAg, (3) serum alpha-fetoprotein, (4) whether RF ablation was the initial treatment and (5) multiplicity of tumor for IDR. Results: The incidence of overall recurrence, LTP and IDR was 62.9%, 26.4% and 53.2%, respectively. The cumulative disease-free survival rates were 52%, 82% and 56% at 1 year, 26%, 63% and 30% at 2 years, respectively. Univariate analysis showed that the significant risk factors for LTP were: a tumor with a diameter >3 cm, contact of HCC with a vessel and an insufficient safety margin (p < 0.05). A multivariate stepwise Cox hazard model showed that the measurement of a tumor diameter >3 cm and insufficient safety margin were independent factors. Only the increased serum alpha-fetoprotein was a significant risk factor for IDR (p < 0.05). Conclusion: Intrahepatic recurrence after percutaneous RF ablation is common. Large HCC (>3 cm) with high serum alpha-fetoprotein should be treated more aggressively because of higher risk for recurrence. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Liver neoplasms; Therapeutic radiology; Radiofrequency (RF) ablation; Liver neoplasms; CT

1. Introduction Only 9–27% of the patients with hepatocellular carcinoma (HCC) are eligible for surgical resection. There are many limiting factors for successful surgical resection in patients with HCC such as severe impairment of hepatic functional reserve, bi-lobar distribution of the tumors, extra∗ Corresponding author at: Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Ilwon-dong, Kangnam-ku, Seoul 135-710, South Korea. Tel.: +82 2 3410 2507; fax: +82 2 3410 2559. E-mail address: [email protected] (H. Rhim).

0720-048X/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2006.03.007

hepatic metastasis or involvement of the portal vein [1,2]. There are many techniques developed for local treatment in patients with unresectable HCC. These include: percutaneous ethanol injection, microwave coagulation, interstitial laser coagulation and cryoablation. Among them, percutaneous ethanol injection therapy has been used most widely. However, this approach has an important drawback in that it requires many treatment sessions to achieve complete necrosis [3]. Currently, radiofrequency (RF) ablation is accepted as an established local therapeutic modality of choice for management of unresectable HCC [4]. It is also a technique used to complement surgical resection for patients with HCC [5,6].

Y.-s. Kim et al. / European Journal of Radiology 59 (2006) 432–441

However, recurrence after RF ablation remains a significant problem. Understanding the pattern and risk for recurrence is important to maximize the clinical benefits of RF ablation therapy. There are two types of intrahepatic recurrence found in patients with HCC after RF ablation, local tumor progression and intrahepatic distant recurrence. Local tumor progression (LTP) occurs along the peripheral margin of the ablative lesion and intrahepatic distant recurrence (IDR) is a new HCC tumor remote from the margin of the ablative lesion. Evaluation of overall recurrence as well as LTP and IDR may provide important information to the management of HCC patients receiving RF ablation therapy. Therefore, the purpose of this study was to determine the pattern of intrahepatic recurrence according to the type of recurrence and including the incidence and the cumulative disease-free survival rate after percutaneous RF ablation. In addition, many potential risk factors, that could be associated with each type of intrahepatic recurrence, were assessed for significant contribution to risk for recurrence.

2. Materials and methods 2.1. Study population From May 1999 to July 2004, a total of 102 patients underwent percutaneous RF ablation for the treatment of HCC. In all patients, a written informed consent was obtained. Institutional review board of our hospital does not require approval for a retrospective clinical study. Inclusion criteria for performing RF ablation in patients with HCC in our institution are as follow: the tumor or tumors should be visualized with ultrasonography (US) and accessible via the percutaneous route; a single tumor with no greater than 5 cm in the largest dimension; multiple tumors (≤3) with each tumor measuring no greater than 3 cm; no portal venous thrombosis and extrahepatic metastasis; prothrombin time ratio over 50% (prothrombin time with international normalized ratio [INR] < 1.7) and a platelet count greater than 50,000/␮L without transfusion support. In this study, we excluded patients with HCC greater than 4 cm in the largest dimension (n = 7); for these patients RF ablation was not considered to be potentially curative because of the possibility of residual tumor, which would cause sample bias. The patients who had a follow-up period less than 6 months (n = 33) were also excluded. In those 33 patients, there was neither death nor disease progression. The remaining study population was 62 patients composed of 43 men and 19 women aged 40–76 years (mean age 57.2). Fiftyseven out of 62 patients have a single HCC tumor and five have two HCC tumors at the time of initial RF ablation. Five out of 57 patients found a single IDR during the followup period. Therefore, another RF ablation for the recurrent tumor was performed; these cases were included in the study.

433

A total of 72 HCC tumors in 62 patients were evaluated. The mean value for the largest tumors diagnosed on US was 25.2 mm (range 5–40 mm). The diagnosis of HCC were confirmed by surgical (n = 4) or US-guided needle biopsy (n = 6) in 10 tumors. Twenty-two tumors were considered to be HCC on the basis of imaging findings and on an elevated serum alpha-fetoprotein (␣-FP) level (>400 ng/mL) [7]. The remaining 40 tumors were diagnosed as HCC with a satisfaction of at least two coincident radiological findings compatible with the diagnosis of HCC on US, CT, MRI, and angiography in appropriate clinical settings (chronic hepatitis or liver cirrhosis) [7]. The imaging criteria for HCC were a newly presenting, residual, or recurrent tumor at follow-up US or CT in patients with chronic liver disease or a characteristic enhancement pattern on contrast-enhanced multi-phase CT and/or MRI (hypervascularization on hepatic arterial phase and wash-out pattern on delayed phase). In 41 patients out of 62, RF ablation was the initial treatment modality for HCC. Twenty one patients had a history of treatment with transcatheter arterial chemoembolization (n = 15), surgical resection (n = 2), percutaneous ethanol injection (n = 1) and transcatheter arterial chemoembolization with surgical resection (n = 3), prior to RF ablation for the recurrent tumor. All the tumors in these 21 patients that were included in our study were newly developed lesions, and there was no evidence of viability of the tumors that had been treated previously. Mean value of the period between the latest previous treatment and RF ablation (n = 21) was 9.5 months with a range of 2.6–23.5 months. In cases of transcatheter arterial chemoembolization (n = 18), RF ablation was performed after mean 8.7 months (range 2.6–18.0). Etiologies of hepatitis or liver cirrhosis were: virus in 55 patients (type B, n = 45; type C, n = 10), alcohol in six patients and Budd–Chiari syndrome in one patient. Underlying hepatic diseases were chronic hepatitis in four patients and liver cirrhosis in 58 patients (Child-Pugh classification A, n = 36; B, n = 15; C, n = 7). Clinical features of patients and tumors are summarized in Table 1. 2.2. RF ablation We used three commercially available RF devices manufactured by two different vendors. Initially, we used only a 50-W RF generator (500 series; RITA Medical Systems, Mountain View, CA, USA) and an active multi-tined, expandable electrode with either four or seven retractable prongs for 45 tumors. Later, a 150-W generator (1500 series) and a larger and similar RF electrode with nine prongs of 5 cm from the same vendor were used for seven tumors. From May 2002, we used a 200-W RF generator and RF electrode from another company (Valleylab, Boulder, CO, USA). This RF electrode was an internally-cooled and was either of a single 17-guage straight needle type with 3 cm active tip (used for 16 tumors) or a cluster of three needles with 2.5 cm active tip mounted on a common handle (used for four tumors) (Table 1).

434

Y.-s. Kim et al. / European Journal of Radiology 59 (2006) 432–441

Table 1 Clinical and therapeutic features of the patients and the tumors Patients

62

Gender Number of males Number of females Mean age (range)

43 19 57.2 (40–76)

Etiologies of liver disease Viral hepatitis type B Viral hepatitis type C Alcoholic hepatitis Budd–Chiari syndrome

45 10 6 1

Severity of liver disease Chronic hepatitis Liver cirrhosis, Child-Pugh Class A Liver cirrhosis, Child-Pugh Class B Liver cirrhosis, Child-Pugh Class C Mean serum alpha-fetoprotein level (range) History of hepatocellular carcinoma treatment No history TACE Surgical resection Percutaneous ethanol injection TACE and surgical resection Multiplicity of tumors Patients with single tumor Single tumor only Single tumor with subsequent distant single recurrence Patients with two tumors

4 36 15 7 292.3 ␮g/L (2.1–8611.0) 41 15 2 1 3 57 52 5

2.3. Imaging protocol

5

Tumors

72

Mean diameter (range)

25.2 mm (5–40)

Diagnosis Histopathologic Clinical

10 62

Radiofrequency ablation Devices 50-W generator, multi-tined expandable electrode with four or seven prongs 150-W generator, multi-tined expandable electrode with nine prongs 200-W generator, internally-cooled electrode, single type 200-W generator, internally-cooled electrode, cluster type Mean times of radiofrequency energy application (range)

pethidine hydrochloride (Pethidine; Samsung Pharmaceutical, Seoul, Korea) for conscious sedation that was infused continuously via the intravenous route. All US procedures were performed with a 2–5-MHz convex-array transducer (HDI 5000; Advanced Technology Laboratory, Bothell, WA, USA) using a free-hand technique. Each application of RF energy was tried to be continued for at least 12 min if the patients could tolerate the pain. For large tumors, more than 3 cm in diameter, a larger (5 cm in length) multi-tined, expandable electrode or a cluster internally-cooled electrode was used. A multiple-overlapping ablation technique was adopted for complete ablation, although tumor size was less than 3 cm, in cases where a possibility of viable tumor residue was suspected based on tumor configuration and direction of electrode approach. Because, targeting and monitoring during the procedure were done exclusively using conventional US images, we estimated tissue necrosis on the basis of echogenic changes of the tumor and surrounding hepatic parenchyma. Ablation was carried out so that at least 0.5 cmthickness of surrounding apparently normal tissue remained as a safety margin. The mean value of time for RF energy application in a single treatment session was 3.1 (range 1–11 times).

45 7 16 4 3.02 (1–11)

TACE: transcatheter arterial chemoembolization.

RF ablations were performed via the percutaneous route under US-guidance in all cases. Two radiologists (YSK and HR) performed all procedures in inpatient participants after they fasted for 12 h. Laboratory examinations including complete blood count, blood coagulation test, blood typing and tumor marker for HCC (␣-FP), were performed before each procedure. Patients were treated with 2% lidocaine hydrochloride (Lidocaine; Kwang Myung Pharmaceutical, Seoul, Korea) for local anesthesia at the puncture site and with

Baseline imaging evaluation of HCC was done using a CT with intravenous contrast administration. CT scans were performed with one of two helical scanners (Somatom Plus and Somatom Plus 4; Siemens, Erlangen, Germany). A total of 120 ml of non-ionic contrast material (Iopromide, Ultravist 300; Schering-Korea, Ansung, Korea) was administrated intravenously via an antecubital vein at a rate of 3 ml/s. A triple-phase contrast-enhancement scan technique was used with a delay time of 30-, 60- and 200-s after the initiation of contrast injection. Helical CT images of pre-contrast and contrast-enhanced scans were acquired using a 7–8-mm collimation and 7–8-mm/s table speed. For post-treatment evaluation, contrast-enhanced CT scans were performed immediately or one day after the procedure in all cases. We defined a residual tumor to be present when a portion of the previous tumor was found as an enhancing lesion beyond the margin of the ablation zone during the hepatic arterial phase. In cases of non-enhancing HCC, ablation zone was carefully compared with location of the tumor before treatment. If any residual HCC was noted, we repeated the RF ablation as soon as possible (n = 5) and then, performed a second immediate follow-up CT scan. We confirmed that there was no evidence of residual tumor that went untreated in all patients. CT scans were repeated every 3–6 months during follow-up with the same protocol for all patients. Intrahepatic recurrence was divided into LTP and IDR based on standardization of terminology and reporting criteria by the international working group of image-guided tumor ablation [8]. LTP was defined by the presence of a

Y.-s. Kim et al. / European Journal of Radiology 59 (2006) 432–441

nodular lesion that was enhanced during the hepatic arterial phase and washed-out by the delayed phase and was found along the peripheral margin of the low attenuated ablative zone. IDR was defined by a lesion with similar enhancement characteristics but distant from the original ablative zone. All cases with the recurrence were diagnosed by imaging criteria instead of biopsy. Imaging analysis of intrahepatic recurrence of HCC at follow-up CT was carried out by one of five radiologists who had experiences in abdominal imaging for more than 4 years (4–20 years) (YSK, HR, OKC, BHK, YK). The readers were independent and were not blinded to the clinical history. If any intrahepatic recurrence that met inclusion criteria of this study was noted, additional percutaneous RF ablation was performed and these cases were included in the study. 2.4. Statistical analysis After review of the follow-up CT, we assessed the incidence and the cumulative disease-free survival rate of overall recurrence, LTP and IDR, respectively. For analysis of the significant risk factors for LTP, we examined (1) maximum tumor diameter (we divided the tumors into small and large group with the cutting values of 2 and 3 cm, respectively), (2) whether tumor contacted with the portal vein, hepatic vein, or inferior vena cava more than 3 mm in diameter at the contact point, (3) degree of approximation of tumor to the hepatic hilum (peripheral, center of the tumor was located within 2 cm from the outer surface of the liver; central, center within 2 cm from the hepatic hilum; intermediate, neither peripheral nor central), (4) whether tumor contacted the liver capsule, (5) whether the ablative safety margin, defined as ablated normal surrounding parenchyma more than 5 mm in thickness, at the immediate follow-up CT scan, compared to the location of tumor before RF ablation, was established in all directions or not, (6) the degree of benign periablational enhancement (grade 0, no enhancement; grade 1, disrupted, non-continuous enhancement; grade 2, continuous rim enhancement) and (7) whether the baseline serum ␣-FP level at the time of RF ablation procedure was increased (>20 ␮g/L). For the IDR, (1) the degree of underlying hepatic disease (advanced disease was defined as liver cirrhosis with Child-Pugh class B and C; early disease as chronic hepatitis and liver cirrhosis with Child-Pugh class A), (2) presence or absence of hepatitis B surface antigen (HBsAg) at the time of RF ablation, (3) whether the baseline serum ␣-FP level at the time of RF ablation procedure was increased, (4) whether RF ablation was the initial treatment for HCC, and (5) the multiplicity of HCC tumors when RF ablation was performed. These risk factors were analyzed by the consensus of two radiologists, retrospectively. Cumulative disease-free survival was estimated using the Kaplan–Meier method and the significance of the risk factors for LTP and IDR were evaluated with univariate analysis using the log-rank test. If multiple risk factors were proven to be significant by this test, we performed multivariate analysis

435

using a stepwise Cox hazards model to search for independently significant factors. All p-values were two-sided. A p-value <0.05 was considered statistically significant. Data analyses were performed using SPSS for Windows (version 11.0.1; SPSS Inc. Chicago, IL, USA).

3. Results 3.1. Pattern of intrahepatic recurrence The median follow-up period for 62 patients (72 tumors) was 19 months and ranged from 6.0 to 49.1 months (mean 19.1 months). In 39 of 62 patients (62.9%), intrahepatic recurrence (LTP and/or IDR) was found during the follow-up period. These were found 1.1–48 months after RF ablation with a median of 16.3 months (95% confidence interval [95% CI], 8.2–24.5 months). The 12- and 24-month cumulative recurrence-free survival rates were 55% (95% CI, 41.7–68.4%) and 26% (95% CI, 11.4–39.5%), respectively (Fig. 1a). LTP was found in 19 of 72 tumors (26.4%) and occurred 1.1–22.5 months after the procedure. Nineteen treated tumors that did not recur for 23 months had no LTP during the entire follow-up period up to 49.1 months. Including these patients, the mean LTP-free interval was 35.2 months. However, the mean LTP-free interval excluding them was 16.1 months. The 12- and 24-month cumulative LTP-free survival rates were 82% (95% CI, 71.9–91.4%) and 63% (95% CI, 48.1–77.6%), respectively (Fig. 1b). IDR was found in 33 of 62 patients (53.2%) during the followup period. These recurrent tumors occurred from 2.1 to 48 months after RF ablation with a median of 20.5 months (95% CI, 12.7–28.2 months). The 12- and 24-month cumulative IDR-free survival rates were 58% (95% CI, 44.8–72.0%) and 35% (95% CI, 18.9–50.1%), respectively (Fig. 1c) (Table 2). In cases with a history of previous transcatheter arterial embolization, the incidences of overall recurrence, LTP, and IDR were 66.7% (12/18), 16.7% (4/18), and 72.2% (13/18), respectively. 3.2. Risk factors analysis of intrahepatic recurrence Table 3 summarizes the results of the risk factor analysis for each type of intrahepatic recurrence associated with HCC after percutaneous RF ablation using the Kaplan–Meier method and log-rank test. 3.2.1. Local tumor progression The log rank test revealed that there were no significant findings for: approximation of tumor to the hepatic hilum, contact of tumor with the liver capsule, the degree of benign periablational enhancement as well as the baseline serum ␣-FP level. However, statistically significant findings were associated with the tumor size, contact to major blood vessels, and establishment of an ablative safe margin.

436

Y.-s. Kim et al. / European Journal of Radiology 59 (2006) 432–441

Fig. 1. Kaplan–Meier survival curves for (a) overall recurrence (n = 62), (b) local tumor progression (n = 72), and (c) intrahepatic distant recurrence (n = 62) after percutaneous radiofrequency ablation of hepatocellular carcinoma. Cross marks (+) in the curves represent censored data.

Table 2 Recurrence pattern of hepatocellular carcinoma after percutaneous radiofrequency ablation Type

Incidence

Median

Mean

12-Month

24-Month

Overall recurrence LTP IDR

39/62 (62.9%) 19/72 (26.4%) 33/62 (53.2%)

16.3 Months (8.2–24.5)

20.0 Months (15.1–24.8) 35.2 Months (30.1–40.4) 23.4 Months (18.0–28.7)

55.0% (41.7–68.4) 81.7% (71.9–91.4) 58.4% (44.8–72.0)

25.5% (11.4–39.5) 62.8% (48.1–77.6) 34.5% (18.9–50.1)

20.5 Months (12.7–28.2)

LTP: local tumor progression; IDR: intrahepatic distant recurrence; median: median of recurrence-free interval after percutaneous radiofrequency ablation of hepatocellular carcinoma, parentheses means 95% confidence interval; mean: mean of recurrence-free interval after percutaneous radiofrequency ablation of hepatocellular carcinoma, parentheses means 95% confidence interval; 12-month: 12-month cumulative recurrence-free survival rate, parentheses means 95% confidence interval; 24-month: 24-month cumulative recurrence-free survival rate, parentheses means 95% confidence interval.

Y.-s. Kim et al. / European Journal of Radiology 59 (2006) 432–441 Table 3 Potential risk factors for the intrahepatic recurrence (local tumor progression and intrahepatic distant recurrence) of hepatocellular carcinoma after percutaneous radiofrequency ablation Recura Local tumor progression (n = 72) Maximal diameter of tumor >3 cm 9 ≤3 cm 10

Not recura

p valueb

13 40

0.005c

Contact of the tumor to major hepatic vessels No contact 10 43 Contact 9 10 Central Intermediate Peripheral

5 5 9

Contact of the tumor to hepatic capsule No contact 12 Contact 7

0.022c

7 8 38

0.220

28 25

0.615

Ablative safety margin at immediate follow-up CT Insufficient 19 27 Sufficient 1 25

0.004c

Degree of benign periablational hyperemia at immediate follow-up CT Grade 0 3 14 Grade 1 9 27 Grade 2 7 12

0.480

Baseline serum alpha-fetoprotein leve >20 ␮g/L 12 ≤20 ␮g/L 7

33 20

0.865

18 11

0.737

Hepatitis surface antigen Present 24 Absent 9

21 8

0.777

Baseline serum alpha-fetoprotein >20 ␮g/L 26 ≤20 ␮g/L 7

13 16

0.019c

Intrahepatic distant recurrence (n = 62) Degree of underlying hepatic disease Advanced 22 Early 11

History of previous treatment of hepatocellular carcinoma by interventional or surgical modality Present 20 21 Absent 13 8 Multiplicity of tumor at the time of radiofrequency ablation Single 31 26 Multiple 3 2 a b c

437

CI, 17.3–71.9%) for large tumors, respectively. Small tumors had a statistically significant longer LTP-free interval compared to the larger tumors (p = 0.005) (Fig. 2a). When we set the criterion for small and large tumors as greater or less than 2 cm in diameter, we obtained similar results despite weaker statistical power (p = 0.042). For analysis of blood vessel contact with the tumor, the tumors contacting major vessels that were more than 3 mm in diameter at the contact point showed a higher incidence of LTP (9/19, 47.4%) than those more distant from the vessels (10/53, 18.9%). The Kaplan–Meier estimates of the 12- and 24-month recurrence free interval after RF ablation were 84.5% (95% CI, 73.4–95.5%) and 73.9% (95% CI, 58.9–88.9%) for the tumors more distant from blood vessels and 72.9% (95% CI, 52.1–83.7%) and 33.7% (95% CI, 4.4–63%) for the tumors in contact with the vessels, respectively. The tumors more distant from the vessels had a statistically significant, longer LTP-free interval compared to the tumors in contact with the vessels (p = 0.022) (Fig. 2b). Failure to establish a sufficient safety margin around the original tumor (>5 mm in all direction) was also a significant risk factor for LTP. The incidence of LTP in cases with a sufficient safety margin was 1/26 (3.8%); by contrast, in cases with an insufficient safety margin (but without residual tumor) LTP was 19/46 (41.3%). The Kaplan–Meier estimates of the 12- and 24-month recurrence free interval was 71.2% (95% CI, 56.9–85.5%) and 49.5% (95% CI, 31.4–67.6%) for tumors without an adequate safety margin. The only case of LTP despite a sufficient safety margin occurred 16.3 months after treatment. Tumors with a sufficient safety margin had a statistically significant, longer LTP-free interval compared to the tumors without one (p = 0.004) (Fig. 2c). To evaluate independent risk factors proven to be significant based on univariate analysis, we performed a multivariate analysis by a stepwise Cox hazards model. We found that establishment of a sufficient safety margin (risk ratio 9.3, 95% CI 1.2–69.8, p = 0.030) and the size of the tumor, in its largest dimension over 3 cm, (risk ratio 2.9, 95% CI 1.1–7.4, p = 0.026) were statistically significant risk factors for LTP after percutaneous RF ablation for HCC (Table 4).

0.418

0.333

Numbers represent the number of tumors or patients. Calculated by long-rank test. Statistically significant.

When we set the criterion for a large tumor to be more than 3 cm in the greatest dimension, the incidence of LTP was 10/50 (20%) in small tumors and 9/22 (40.9%) in large tumors. The Kaplan–Meier estimates of the 12- and 24-month recurrence free interval after RF ablation were 89.4% (95% CI, 80.4–98.4%) and 71.3% (95% CI, 55–87.6%) for small tumors and 62.4% (95% CI, 38.6–76.2%) and 44.6% (95%

3.2.2. Intrahepatic distant recurrence Among the six risk factors for IDR examined (the degree of liver cirrhosis, the presence of HBsAg, the increase of baseline serum ␣-FP level, RF ablation as a secondary treatment modality, a multiplicity of the tumor), only an increased baseline serum ␣-FP level affected the length of the IDR-free interval significantly. IDR occurred in 7 of 23 patients (30.4%) in the group with a normal ␣-FP level and 26 of 39 patients (66.7%) in the group with an increased ␣-FP level (>20 ␮g/L). The Kaplan–Meier estimates of the 12- and 24-month IDR-free interval, after RF ablation, was 73.4% (95% CI, 52.4–94.4%) for normal ␣-FP levels and 50.6% (95% CI, 33.8–67.5%) and 19.6% (95% CI, 4–35.2%) for the increased ␣-FP group, respectively. The normal ␣-FP group had a statistically sig-

438

Y.-s. Kim et al. / European Journal of Radiology 59 (2006) 432–441

Fig. 2. Significant risk factors for local tumor progression after percutaneous radiofrequency ablation of hepatocellular carcinoma. Tumor size more than 3 cm in its greatest dimension (p = 0.005) (a), contact of tumor to major hepatic vessels over 3 mm in diameter at contact point (p = 0.022) (b), and failure to establish a sufficient ablative safety margin at the immediate follow-up CT scan over 5 mm in all directions (p = 0.004) (c) are proven to be significant by Kaplan–Meier method and log-rank test. Cross marks (+) in the curves represent censored data. Table 4 Independent risk factor associated with local tumor progression after percutaneous radiofrequency ablation of hepatocellular carcinoma No. of tumors A stepwise Cox’s hazard model Safety margin Insufficient 46 Sufficient 26 Tumor size >3 cm ≤3 cm

22 50

Standard error

χ2

p value

Risk ratio

95% Confidence interval

1.029 –

8.128 –

0.030 –

9.281 1.0

1.234–69.804

0.480 –

13.958 –

0.026 –

2.899 1.0

1.132–7.423

Y.-s. Kim et al. / European Journal of Radiology 59 (2006) 432–441

Fig. 3. Significant risk factor for intrahepatic distant recurrence after percutaneous radiofrequency ablation of hepatocellular carcinoma. Increased level of baseline serum alpha-fetoprotein (>20 ␮g/L) are proven to be significant by Kaplan–Meier method and log-rank test (p = 0.019). Cross marks (+) in the curves represent censored data.

nificant, longer LTP-free interval compared to the increased ␣-FP group (p = 0.019) (Fig. 3).

4. Discussion We categorized intrahepatic recurrence of HCC after percutaneous RF ablation into local tumor progression and intrahepatic distant recurrence; each type of recurrence has a specific mechanism of pathogenesis and is thought to occur independently. Generally, LTP is considered to be related to residual tumor cells that have spread microscopically beyond the ablative margin; although there is a possibility of de novo occurrence at that site. Pathogenesis of IDR is thought to be a result of an intrahepatic metastasis of a primary HCC or due to a multicentric origin of the HCC [9]. Therefore, LTP may be associated more with a treatment methodology or result, local environment of the tumor such as a vessel contact and characteristics of the tumor itself rather than the systemic condition of the patient. By contrast, IDR may be related more to systemic factors rather than local factors. Therefore, we analyzed a variety of potential local and systemic risk factors for LTP and IDR associated with HCC, independently. There have been several studies reporting on the incidence and risk factors of LTP or IDR after RF ablation for HCC. Komorizono et al. [10] who studied LTP after a single application of RF energy for relatively small HCC reported a tumor free survival rate at 12 and 15 months of 76% and 74%, respectively. Significant risk factors for LTP were reported to be: a large tumor size over 2 cm in the greatest dimen-

439

sion and a subcapsular location. Hori et al. [11] who also studied a similar group of subjects reported that the cumulative local recurrence rates were 9.7%, 15.4% and 20.4% at 1, 2 and 3 years, respectively. Significant risk factors for LTP were tumor size and tumor location. Izumi et al. [12] reported that the IDR, after RF or microwave ablation for HCC, was found in 22 out of 84 patients (26.2%) (median follow-up period, 22 months); significant risk factors for IDR were an increased level of serum ␣-FP, a hepatitis C virus infection and multifocal HCC at the time of treatment. Harrison et al. [13] reported that the LTP rate was 39.1% and IDR rate was 30.4% in their 3-year follow-up study and that the significant risk factors for recurrence were a large tumor size, an increase of serum ␣-FP level and the presence of hepatitis. Although they investigated LTP and IDR in a similar population, they studied only the incidence of each type of recurrence not survival interval. Yamanaka et al. [14] recently reported, in their study of RF ablation for HCC in patients with hepatitis C, that the cumulative recurrence rate after 1 and 2 years was 30.8% and 86.8% for multinodular HCC and 15.4% and 29.5% for uninodular HCC; the investigators found significant risk factors to be associated with the number of HCC nodules, low serum platelets and albumin level. In our study, we evaluated the incidence of, the cumulative disease free survival for the overall recurrence, LTP and IDR. We think the results have clinical implication for advance warning and appropriate patient management for overall recurrence as well as LTP or IDR. As noted earlier, we excluded patients with HCC greater than 4 cm in the largest dimension (n = 7). In these patients, RF ablation should be performed with the multipleoverlapping technique although we use a cluster type internally cooled electrode, which theoretically can create an ablative lesion as large as 5 cm. This technique has the potential risk of incomplete tumor ablation and therefore could bias our study sample. However, this technique has been adopted and is commonly used for the ablation of smaller tumors, where incomplete ablation is unlikely. In a small portion of RF ablations included in our study (n = 5), residual tumors were found on immediate follow-up CT scans. In those cases, we repeated the procedures as soon as possible to achieve complete ablation. Then, we confirmed complete treatment on another immediate follow-up CT scan after the additional RF ablation. We included these cases in our study population and analyzed them as the same as those cases with a single complete ablation; this is because there was no reason to differentiate them from the other cases with a single session ablation. Although there have been studies on a safe tumor free margin of a surgical resection for hepatic tumors, this remains an unresolved problem [15–17]. A gross tumor free margin of 1 cm in all directions around the tumor in the resected specimen is generally accepted by most surgeons and pathologists. However, the safety margin for RF ablation used to treat HCC requires consideration independent from surgical resection because there are significant differences between

440

Y.-s. Kim et al. / European Journal of Radiology 59 (2006) 432–441

these two techniques. Because RF ablation in our study was performed percutaneously, evaluation could be done only with CT images pre- and post-procedure, rather than pathologic specimens. Securing a safety margin with RF ablation becomes more difficult exponentially as the tumor size increases. We set and evaluated a safe ablative margin as 5 mm in all directions around the tumor and two CT scans before and after the RF ablation were analyzed. We showed that the establishment of a 5 mm ablative margin was effective in suppressing LTP after RF ablation of HCC with both a univariate and a multivariate analysis. Theoretically, a tumor that is contiguous to a large vessel has more of a chance to let some tumor cells alive during local thermal therapy because there is a significant tissue cooling effect caused by blood circulation of normal body temperature [18]. We found that contiguity to a large vessel is a significant risk factor for LTP by a univariate analysis. However, this is an inconsistent finding in the literature. Lu et al. [19] who studied RF ablation for liver cancer, including metastasis, reported that a large vessel contiguous to tumors is a strong independent predictor of incomplete tumor destruction by RF ablation. However, Komorizono et al. [10] and Hori et al. [11] showed it was not a significant risk factor for LTP after RF ablation for HCC. Our results showed that an increased level of serum ␣-FP was associated with IDR after RF ablation for HCC. Although the rationale for why ␣-FP production is related to tumor recurrence is not apparent, other reports support this association. Llovet et al. [20] reported that patients with tumor recurrence had a mean ␣-FP level of 395 ␮g/L, compared to an ␣-FP of 49 ␮g/L in patients who did not have recurrence; these results were reported in a study of 32 patients with HCC treated with RF ablation. As previously noted Izumi et al. [12] and Harrison et al. [13] also reported similar results. In addition, these findings are similar to those reported for percutaneous ethanol injection therapy of HCC [21]. In our study, tumors that did not recur for about 2 years did not show any LTP during the residual follow-up period of up to 49.1 months. This time interval has great clinical significance because it can affect the follow-up CT schedule as well as patient prognosis. However, it is premature to conclude that an interval of 2 years is safe enough to neglect the possibility of LTP. Study with a larger population and a longer follow-up period is needed to ascertain a safe period from LTP. Risk factors associated with LTP after percutaneous RF ablation for HCC, include a tumor size over 3 cm in diameter and failure to establish a sufficient ablative safety margin; these are independently significant risk factors by multivariate analysis. We think that one other factor, contact of the tumor with a large vessel, may potentially be associated with tumor size; however, they were shown not to be independently significant. A possible explanation is that the larger the tumor size, the greater the chance to contact a large vessel. Our study has some important limitations. (1) Results of this study were achieved retrospectively and were not ran-

domized or controlled trials. (2) Only a small portion of the tumors in this study (10/72, 13.9%) were confirmed histologically. (3) During this study, we used five types of electrodes and three types of RF generators from two different vendors and the selection of types of electrode and RF generator was not randomized, but dependent on the time of treatment. Despite these limitations, we can conclude that, after percutaneous RF ablation for HCC, overall recurrence rate is considerable reaching two thirds of the treated patients; IDR occurs earlier and more frequently than LTP especially when the baseline serum ␣-FP level is elevated. Although less frequent, LTP tends to occur when we ablated a large HCC tumor over 3 cm in dimension. In those cases, more aggressive ablation or combination therapy with other modality such as transcatheter arterial chemoembolization or radiation therapy should be attempted and frequent follow-up protocol should be adopted to have a chance to eradicate a recurrent cancer when it is small. References [1] Primary liver cancers in Japan. Cancer 1980;45:2263–9. [2] Lai EC, Fan ST, Lo CM, Chu KM, Liu CL, Wong J. Hepatic resection for hepatocellular carcinoma. An audit of 343 patients. Ann Surg 1995;221:291–8. [3] Livraghi T, Goldberg SN, Lazzaroni S, Meloni F, Solbiati L, Gazelle GS. Small hepatocellular carcinoma: treatment with radiofrequency ablation versus ethanol injection. Radiology 1999;210:655– 61. [4] Jansen MC, van Hillegersberg R, Chamuleau RA, van Delden OM, Gouma DJ, van Gulik TM. Outcome of regional and local ablative therapies for hepatocellular carcinoma: a collective review. Eur J Surg Oncol 2005;31:331–47. [5] Choi D, Lim HK, Kim MJ, et al. Recurrent hepatocellular carcinoma: percutaneous radiofrequency ablation after hepatectomy. Radiology 2004;230:135–41. [6] Elias D, De Baere T, Smayra T, Ouellet JF, Roche A, Lasser P. Percutaneous radiofrequency thermoablation as an alternative to surgery for treatment of liver tumor recurrence after hepatectomy. Br J Surg 2002;89:752–6. [7] Bruix J, Sherman M, Llovet JM, et al. Clinical management of hepatocellular carcinoma. Conclusion of the Barcellona-2000 EASL conference. J Hepatol 2001;35:421–30. [8] Goldberg SN, Grassi CJ, Cardella JF, et al. Image-guided tumor ablation: standardization of terminology and reporting criteria. J Vasc Interv Radiol 2005;16:165–78. [9] Okuda K. Hepatocellular carcinoma: recent progress. Hepatology 1992;15:948–63. [10] Komorizono Y, Oketani M, Sako K, et al. Risk factors for local recurrence of small hepatocellular carcinoma tumors after a single session, single application of percutaneous radiofrequency ablation. Cancer 2003;97:1253–62. [11] Hori T, Nagata K, Hasuike S, et al. Risk factors for the local recurrence of hepatocellular carcinoma after a single session of percutaneous radiofrequency ablation. J Gastroenterol 2003;38: 977–81. [12] Izumi N, Asahina Y, Noguchi O, et al. Risk factors for distant recurrence of hepatocellular carcinoma in the liver after complete coagulation by microwave or radiofrequency ablation. Cancer 2001;91:949–56. [13] Harrison LE, Koneru B, Baramipour P, et al. Locoregional recurrence are frequent after radiofrequency ablation for hepatocellular carcinoma. J Am Coll Surg 2003;197:759–64.

Y.-s. Kim et al. / European Journal of Radiology 59 (2006) 432–441 [14] Yamanaka Y, Shiraki K, Miyashita K, et al. Risk factors for the recurrence of hepatocellular carcinoma after radiofrequency ablation of hepatocellular carcinoma in patient with hepatitis C. World J Gastroenterol 2005;11:2174–8. [15] Regimbeau JM, Kianmanesh R, Farges O, Dondero F, Dauvanet A, Belghiti J. Extent of liver resection influences the outcome in patients with cirrhosis and small hepatocellular carcinoma. Surgery 2002;131:311–7. [16] Poon RT, Fan ST, Ng IO, Wong J. Significance of resection margin in hepatectomy for hepatocellular carcinoma: a critical reappraisal. Ann Surg 2000;231:544–51. [17] Poon RT, Fan ST, Ng IO, Lo CM, Liu CL, Wong J. Different risk factors and prognosis for early and late intrahepatic recurrence after resection of hepatocellular carcinoma. Cancer 2000;89:500–17.

441

[18] Rhim H, Goldberg SN, Dod 3rd GD, et al. Essential techniques for successful radio-frequency thermal ablation of malignant hepatic tumors. Radiographics 2001;21:S17–35. [19] Lu DS, Raman SS, Limanond P, et al. Influence of large peritumoral vessels on outcome of radiofrequency ablation of liver tumors. J Vasc Interv Radiol 2003;14:1267–74. [20] Llovet JM, Vilana R, Bru C, et al. Increased risk of tumor seeding after percutaneous radiofrequency ablation for single hepatocellular carcinoma. Hepatology 2001;33:1124–9. [21] Pompili M, Rapaccini GL, De Luca F, et al. Risk factors for intrahepatic recurrence of hepatocellular carcinoma in cirrhotic patients treated by percutaneous ethanol injection. Cancer 1997;79: 1501–8.