Va s c u l a r a n d I n t e r ve n t i o n a l R a d i o l o g y • O r i g i n a l R e s e a r c h

Kim et al.
RFA to Treat Intrahepatic Cholangiocarcinoma

Vascular and Interventional Radiology

Original Research

Radiofrequency Ablation for the

Treatment of Primary Intrahepatic
Cholangiocarcinoma
Jin Hyoung Kim1 OBJECTIVE. We present the results of percutaneous radiofrequency ablation (RFA) in
Hyung Jin Won patients with unresectable primary intrahepatic cholangiocarcinoma.
Yong Moon Shin MATERIALS AND METHODS. From 2000 to 2009, 13 patients with 17 primary in-
Kyung-Ah Kim trahepatic cholangiocarcinomas underwent RFA at our institution. Intrahepatic cholangio-
Pyo Nyun Kim carcinoma was unresectable because of poor hepatic reserve due to liver cirrhosis in nine pa-
tients, extrahepatic extension in two, atrophy of the left hepatic lobe in one, and underlying
Kim JH, Won HJ, Shin YM, Kim KA, Kim PN comorbidities in one. Ten tumors had a diameter of less than 3 cm, five were between 3 and 5
cm, and two were larger than 5 cm. Technical effectiveness was defined as the complete abla-
tion of the tumor, shown by imaging follow-up 1 month later. Local progression-free survival,
overall survival periods, and complications after RFA were also evaluated.
RESULTS. Technical effectiveness of RFA was achieved for 15 of the 17 tumors (88%),
all smaller than 5 cm in diameter. Treatment failure occurred in two patients with large tu-
mors (7 and 8 cm). After the 17 RFA sessions, one major complication (6%), a liver abscess,
occurred 1 month later. During follow-up (median, 19.5 months; range, 3.3–82.1 months),
nine patients died and four remain alive. Median local progression-free survival and overall
survival periods were 32.2 and 38.5 months, respectively. The 1-, 3-, and 5-year survival rates
were 85%, 51%, and 15%, respectively.
CONCLUSION. RFA may provide successful local tumor control in patients with pri-
mary intrahepatic cholangiocarcinomas of intermediate (3–5 cm) or small (< 3 cm) diameter.
RFA for unresectable primary intrahepatic cholangiocarcinoma resulted in a median overall
survival period of 38.5 months.

T
he prognosis for patients with un- diation therapy, these options afford little or
Keywords: CT, hepatectomy, intrahepatic cholangiocar-
treated unresectable cholangio- no improvement in survival compared with
cinoma, radiofrequency ablation carcinoma is dismal, with a me- supportive therapy alone because intrahe-
dian survival time of 3.9 months patic cholangiocarcinomas respond poor-
DOI:10.2214/AJR.10.4937 [1]. The major cause of death is liver failure or ly to such therapies [2, 5, 6]. Transarterial
cholangitis and sepsis resulting from progres- chemoembolization (TACE), which can in-
Received May 6, 2010; accepted after revision
September 20, 2010. sive, refractory biliary obstruction [2]. Intra- crease the local concentration of chemother-
hepatic cholangiocarcinoma usually presents apeutic agents thus killing cancer cells and
Supported by grant 05-382 from the Asan Institute for as advanced disease at the time of diagnosis reducing systemic side effects, has shown
Life Sciences. because of the lack of symptoms until late in promising results when used as a palliative
1
All authors: Department of Radiology and Research
disease progression, and the overall prognosis treatment of patients with liver-dominant he-
Institute of Radiology, Asan Medical Center, University of is far worse than that of extrahepatic cholang- patic malignancies [7]. However, such palli-
Ulsan College of Medicine, 388-1, Poongnap-2dong, iocarcinoma [1–3]. Although hepatic resec- ative therapy may not be effective in patients
Songpa-gu, Seoul, Republic of Korea. Address tion may be curative, most patients with intra- with hypovascular tumors because it is not
correspondence to H. J. Won (hjwon@amc.seoul.kr).
hepatic cholangiocarcinoma are not candidates possible to deliver a chemotherapeutic agent
WEB for curative resection because of advanced or embolic material more effectively and se-
This is a Web exclusive article. cancer at the time of initial presentation, like- lectively than in patients with hypervascu-
ly insufficient function of the remaining liver, lar tumors [8, 9]. Radioembolization has re-
AJR 2011; 196:W205–W209 or underlying patient comorbidities [2–4]. cently shown promising results for palliative
0361–803X/11/1962–W205
Although most patients with intrahepatic treatment of patients with unresectable intra-
cholangiocarcinoma receive palliative thera- hepatic cholangiocarcinoma, with a reported
© American Roentgen Ray Society py including systemic chemotherapy and ra- median survival period after radioemboliza-

AJR:196, February 2011 W205

 Kim et al.

TABLE 1:  Patient Characteristics and Clinical Outcomes

Tumor Recurrence Follow-Up
Patient Location LTP-Free Overall
No. Age (y) Sex Cause of Unresectability Stage No. (Segment) Size (cm) TS TE Pattern Second Tx SP (mo) SP (mo) Survival
1 45 M Bone metastasis IV 1 IV 8 No No TF RFA 0.0 13.7 Death
2 61 F Underlying comorbidities I 1 V 7 No No TF None 0.0 3.3 Death
(DM, pulmonary Tbc)
3 61 M Poor hepatic reserve (LC) I 1 VIII 3.3 Yes Yes LTP RFA 31.0 38.5 Death
4 47 M Poor hepatic reserve (LC) I 1 VIII 2.8 Yes Yes NL TACE, RTx 13.2 13.2 Death
5 55 M Poor hepatic reserve (LC) I 1 VIII 2.5 Yes Yes NL RFA 50.1 50.1 Death
6 51 F Atrophy of left hepatic lobe I 1 VI 4.5 Yes Yes LTP STx 7.0 14.6 Death
7 61 M Poor hepatic reserve (LC) I 1 VIII 3.6 Yes Yes LTP, NL STx 34.8 42.5 Death
8 49 F Poor hepatic reserve (LC) II 2 VI, VI 2.4, 0.9 Yes Yes None None 82.1 82.1 Alive
9 66 M Poor hepatic reserve (LC) I 1 VI 3.6 Yes Yes None None 19.5 19.5 Death
10 63 M Tumor extension into IIIB 1 IV 3.3 Yes Yes None None 6.4 6.4 Death
common hepatic duct
11 61 M Poor hepatic reserve (LC) II 3 IV, IV, VIII 0.8, 2.1, 1.4 Yes Yes LTP, NL Ctx 32.2 49.5 Alive
12 66 M Poor hepatic reserve (LC) I 1 V 0.9 Yes Yes None None 22.0 22.0 Alive
13 71 M Poor hepatic reserve (LC) II 2 VII, VIII 2, 2.1 Yes Yes None None 10.0 10.0 Alive
Note—TS = technical success, TE = technical effectiveness, Tx = treatment, LTP = local tumor progression, SP = survival period, TF = technical failure, RFA = radiofre-
quency ablation, DM = diabetes mellitus, Tbc = tuberculosis, LC = liver cirrhosis, NL = new lesions, TACE = transarterial chemoembolization, RTx = radiation treatment,
STx = supportive treatment, Ctx = chemotherapy.

tion of 9.3 and 14.9 months in two small se- of grade 2–4) [18] and the presence of more than der conscious sedation and local anesthesia. A sin-
ries [10, 11]. However, further prospective or three intrahepatic cholangiocarcinomas, vascular gle needle or a needle cluster with an internally
large study is still required to determine the invasion, progressive extrahepatic metastases, or cooled electrode was used depending on the size
value of radioembolization [10, 11]. coagulopathy (platelet count < 50 × 103/μL; inter- of the tumor. For all tumors 2 cm or less in diam-
Regardless of tumor vascularity, percu- national normalized ratio > 1.5). eter, a single electrode with a 3-cm exposed tip
taneous radiofrequency ablation (RFA) has Early in our experience with RFA for the treat- was used. For tumors more than 2 cm in diame-
been reported to be safe and effective in the ment of intrahepatic cholangiocarcinoma, we per- ter, a cluster electrode or multiple overlapping in-
local control of hepatic malignancies in pa- formed RFA on two patients with large (7 and 8 sertions of a single electrode were used. Radiofre-
tients considered unsuitable for surgical re- cm in diameter) liver tumors. However, after treat- quency current was emitted for 12 or 15 minutes
section [12–15]. One case report [16] and a ment failed for both patients, we decided to per- using a 200-W generator set to deliver maximum
study of 10 patients [17] described the use form RFA only in patients with intrahepatic cho- power under automatic impedance control. Each
of RFA in patients with primary intrahe- langiocarcinomas smaller than 5 cm in diameter. tumor received 1–7 ablations (median, 2 ablations)
patic cholangiocarcinoma, but neither pro- All study patients underwent contrast-enhanced per session depending on its size and shape. The
vided survival data after RFA. We therefore CT, with or without MRI, to evaluate the charac- end point of the ablation was to achieve complete
assessed the outcomes—including survival teristics of the tumor and determine whether ex- ablation of both the visible tumor and an ablation
results—of percutaneous RFA performed in trahepatic metastases were present. margin in the normal liver parenchyma surround-
13 patients with unresectable primary intra- We included 13 patients with 17 primary intra- ing the tumor of at least a 0.5–1.0 cm.
hepatic cholangiocarcinoma. hepatic cholangiocarcinomas who underwent RFA
between February 2000 and June 2009. All tumors Follow-Up
Materials and Methods were diagnosed as intrahepatic cholangiocarcino- Immediately after RFA, all patients underwent
Patient Population mas on the basis of the histologic results of imag- contrast-enhanced CT to evaluate for possible
Our institutional review board approved this ret- ing-guided percutaneous needle biopsies. Baseline complications such as bleeding or fluid collection.
rospective review of patient medical and imaging patient and tumor characteristics are summarized The efficacy of RFA was evaluated by contrast-
records. All included patients had undergone RFA in Table 1. Tumors were staged according to the enhanced CT 1 month after the procedure. If a re-
for the treatment of three or fewer histologically American Joint Committee on Cancer staging sys- sidual tumor was present in the ablated area, an
proven primary intrahepatic cholangiocarcinomas tem, also known as TNM staging [4]. Ten tumors additional session of RFA was performed to treat
not amenable to curative surgery, showed no imag- had a diameter of less than 3 cm, five were between the lesion further.
ing evidence of vascular invasion by the tumor, and 3 and 5 cm, and two were larger than 5 cm. In cases of complete ablation of the tumor with
had no evidence of extrahepatic disease in the face no appearance of a new tumor in other liver sites
of stable extrahepatic metastases. Exclusion crite- Radiofrequency Ablation Technique on 1-month follow-up CT, subsequent follow-up
ria were poor performance status (Eastern Coop- RFA was performed percutaneously using contrast-enhanced CT examinations were repeat-
erative Oncology Group performance status rating sonographic guidance while the patient was un- ed every 2–3 months. All new tumors in the ablat-

W206 AJR:196, February 2011

 RFA to Treat Intrahepatic Cholangiocarcinoma

ed lesion or in other liver sites that emerged during gression [20], and the overall survival period was Three patients experienced postablation
follow-up were treated with RFA if the patient still defined as the time interval, in months, between syndrome that resolved within 10 days in all
met the requirements for RFA. the initial RFA and patient death. patients without any special treatment. A small
amount of pleural effusion and a small degree
Definition and Evaluation of Data Results of hematoma around the ablated area occurred
We adopted the reporting standards of the Soci- Technical Success and Technical Effectiveness in five and two patients, respectively. All of
ety of Interventional Radiology with respect to ter- Clinical outcomes after RFA are summa- these problems disappeared after 1 month.
minology and reporting criteria [19]. Technical suc- rized in Table 1. Technical success after one
cess was achieved when a tumor that was treated session of RFA was achieved for 15 of the 17 Local Tumor Progression-Free Survival Period
according to protocol was completely covered at the tumors (88%) in 11 of the 13 patients (85%). In addition to the two patients (patients 1
time of the procedure. Technical effectiveness was In two patients (patients 1 and 2) with large tu- and 2) showing initial treatment failure, two
defined as the complete ablation of the tumor shown mors (8 and 7 cm in diameter, respectively), the patients (patients 3 and 6) showed local tu-
on imaging follow-up 1 month after RFA. Any ir- tumors were not completely ablated. In these mor progression 31 and 7 months, respective-
regular or nodular peripheral enhancement was con- patients, contrast-enhanced CT scans obtained ly, after the initial procedure. Two patients
sidered to reflect residual tumor at the ablation mar- 1 month after RFA showed residual, irregu- (patients 7 and 11) had local tumor progres-
gin and a treatment failure. Local tumor progression lar, peripherally enhanced areas. An addition- sion and new lesions in the liver or distant ar-
was defined as nodular or irregular enhancement at al RFA session, delivered 4 months after the eas 34.8 and 32.2 months, respectively, after
any follow-up examination performed more than initial RFA session, also did not result in com- the initial procedure. Two patients (patients 4
1 month after RFA. A major complication was de- plete ablation of tumor in patient 1. Contrast- and 5) showed new lesions in the liver or dis-
fined as any event that resulted in additional treat- enhanced CT performed 1 month after RFA of tant areas without local tumor progression at
ment including an increased level of care; a hospital the remaining 15 tumors of 11 patients showed the ablated area after 7.1 and 16 months, re-
stay beyond observation status, including readmis- no evidence of residual unablated tumor; these spectively. The treatments used for patients
sion after initial discharge; and permanent adverse findings indicate that RFA was technically ef- with local tumor progression and new lesions
sequelae including substantial morbidity or disabil- fective as well as technically successful for the are summarized in Table 1. The median local
ity and death. All other complications were classi- treatment of 15 of 17 tumors (88%) in 11 of the tumor progression-free survival period was
fied as minor. Postablation syndrome was defined as 13 patients (85%) (Fig. 1). 32.2 months (Fig. 2).
a transient self-limiting complex of low-grade fever,
pain, and general malaise [17, 19]. Complications Overall Survival Period
Local tumor progression-free survival and A liver abscess developed in the ablated During the median follow-up period of
overall survival were calculated using the Kaplan- area 1 month after the first RFA session (6%, 19.5 months (range, 3.3–82.1 months), nine
Meier method. The local tumor progression-free 1/17) in one patient (patient 2). This patient patients died and four remain alive. Of the
survival period was defined as the time interval, died of sepsis 3.3 months after the procedure nine patients who died, six died from disease
in months, between the initial RFA treatment and despite percutaneous drainage of the liver progression, one (patient 2) died of a liver ab-
any follow-up imaging showing local tumor pro- abscess and antibiotic therapy. scess related to RFA, one (patient 9) died of

A B
Fig. 1—Contrast-enhanced axial CT images of 61-year-old man with intrahepatic cholangiocarcinoma (patient 3 in Table 1).
A, Image in portal phase obtained 4 days before radiofrequency ablation (RFA) shows heterogeneously increasing mass (arrowheads), 3.3 cm in largest diameter, in
segment VIII.
B, Image obtained 28 months after RFA shows good local tumor control (arrowheads).

AJR:196, February 2011 W207

 Kim et al.

may be a risk factor for major complications

100
after RFA [24].
100
The principal limitation of this study was
80
the small number of study patients and the
80
lack of a control group. Nevertheless, we be-

% of Patients
% of Tumors

60 lieve that our results suggest that RFA may

60
play a role in the treatment of patients with
40
40
unresectable primary intrahepatic cholang-
iocarcinoma and that our results provide sup-
20
20
port for future prospective investigations.
In conclusion, RFA may provide success-
0 0 ful local tumor control in patients with pri-
0 20 40 60 80 0 20 40 60 80
mary intrahepatic cholangiocarcinomas of
Local Tumor Progression-Free intermediate (3–5 cm) or small (< 3 cm) di-
Overall Survival Period (mo)
Survival Period (mo) ameter. RFA to treat unresectable primary
intrahepatic cholangiocarcinoma resulted
Fig. 2—Graph shows local tumor progression-free Fig. 3—Graph shows overall survival period after in a median overall survival period of 38.5
survival period after each session of radiofrequency radiofrequency ablation in 13 patients. months in our present patient series.
ablation.
variceal bleeding, and one (patient 10) died tween tumor size or multiplicity and the lo- References
of an infection related to a refractory biliary cal tumor progression-free survival period. 1. Park J, Kim MH, Kim KP, et al. Natural history
obstruction. The median overall survival pe- The median survival time of patients with and prognostic factors of advanced cholangiocar-
riod after RFA was 38.5 months (Fig. 3). The untreated, unresectable intrahepatic cholang- cinoma without surgery, chemotherapy, or radio-
1-, 3-, and 5-year survival rates were 85%, iocarcinoma has been reported to be 3 months therapy: a large-scale observational study. Gut
51%, and 15%, respectively. [1]. TACE has recently shown promising re- Liver 2009; 3:298–305
sults for palliative treatment of such patients, 2. Burger I, Hong K, Schulick R, et al. Transcatheter
Discussion with a reported median survival period after arterial chemoembolization in unresectable cho-
We found that RFA was both technically TACE of 9.1–10 months [5, 8]. There have langiocarcinoma: initial experience in a single in-
successful and technically effective in 15 of been no reports to date on overall survival af- stitution. J Vasc Interv Radiol 2005; 16:353–361
17 primary intrahepatic cholangiocarcino- ter RFA in patients with primary intrahepatic 3. Nakeeb A, Tran KQ, Black MJ, et al. Improved
mas (88%), all of which were smaller than cholangiocarcinoma. We found that the me- survival in resected biliary malignancies. Surgery
5 cm in diameter. Treatment failure occurred dian overall survival period after RFA was 2002; 132:555–563
in two patients, both of whom had large tu- 38.5 months in 13 patients with primary in- 4. Aljiffry M, Walsh MJ, Molinari M. Advances in
mors (7 and 8 cm, respectively). In a previ- trahepatic cholangiocarcinoma during a me- diagnosis, treatment and palliation of cholangio-
ous study [17], technical effectiveness was dian follow-up period of 19.5 months (range, carcinoma: 1990–2009. World J Gastroenterol
achieved in eight of 10 intrahepatic cholan- 3.3–82.1 months). The 1-, 3-, and 5-year pa- 2009; 15:4240–4262
giocarcinomas (80%) after one or two ses- tient survival rates were 85%, 51%, and 15%, 5. Gusani NJ, Balaa FK, Steel JL, et al. Treatment of
sions of RFA, with the two patients who ex- respectively. Even after curative resection for unresectable cholangiocarcinoma with gemcitabi-
perienced treatment failure having larger intrahepatic cholangiocarcinoma, patient sur- ne-based transcatheter arterial chemoemboliza-
tumors (4.6 and 6.8 cm in diameter) [17]. To- vival rates remain low, with 5-year survival tion (TACE): a single-institution experience. J
gether, these findings indicate that RFA may rates ranging between 15% and 36% [21–23]. Gastrointest Surg 2008; 12:129–137
provide successful local tumor control in pa- Thus, the 5-year survival rate we observed after 6. Mazhar D, Stebbing J, Bower M. Chemotherapy
tients with intermediate (3–5 cm) or small RFA ablation was comparable to that seen after for advanced cholangiocarcinoma: what is stan-
(< 3 cm) intrahepatic cholangiocarcinomas. curative resection. dard treatment? Future Oncol 2006; 2:509–514
To our knowledge, there have been no pre- We found that a major complication oc- 7. Liapi E, Geschwind JE. Transcatheter and abla-
vious reports on local tumor progression- curred after only one of 17 (6%) RFA ses- tive therapeutic approaches for solid malignan-
free survival periods after RFA in patients sions. This problem, a liver abscess, oc- cies. J Clin Oncol 2007; 25:978–986
with primary intrahepatic cholangiocarcino- curred 1 month after RFA in a patient with 8. Kim JH, Yoon HK, Sung KB, et al. Transcatheter
ma. We found that the median local tumor a large tumor (7 cm) and eventually result- arterial chemoembolization of chemoinfusion for
progression-free period after the initial RFA ed in the patient’s death despite percutane- unresectable intrahepatic cholangiocarcinoma:
session for the treatment of intrahepatic cho- ous drainage of the abscess and the use of clinical efficacy and factors influencing outcomes.
langiocarcinoma was 32.2 months. Tumor antibiotic therapy. The patient’s comorbid Cancer 2008; 113:1614–1622
size and tumor multiplicity have been shown conditions—diabetes mellitus and pulmo- 9. Kim JH, Yoon HK, Ko GY, et al. Nonresectable
to be significantly associated with local tu- nary tuberculosis—also may have contribut- combined hepatocellular carcinoma and cholang-
mor progression-free survival after RFA for ed to death. Because patients with large tu- iocarcinoma: analysis of the response and prog-
treatment of hepatic malignancies [14]. Be- mors may require a greater number of RFA nostic factors after transcatheter arterial chemoem-
cause of the small sample size of the current sessions and because these sessions carry a bolization. Radiology 2010; 255:270–277
study, we did not evaluate relationships be- risk of major complications, large tumor size 10. Ibrahim SM, Mulcahy MF, Lewandowski RJ, et

W208 AJR:196, February 2011

 RFA to Treat Intrahepatic Cholangiocarcinoma

al. Treatment of unresectable cholangiocarcinoma Sironi S, Lee FT Jr. Breast cancer liver metasta- terminology and reporting criteria. J Vasc Interv
using yttrium-90 microspheres. Cancer 2008; ses: US-guided percutaneous radiofrequency ab- Radiol 2009; 20[suppl 7]:S377–S390
113:2119–2128 lation—intermediate and long-term survival rates. 20. Sofocleous CT, Nascimento RG, Gonen M, et al.
11. Saxena A, Bester L, Chua TC, Chu FC, Morris Radiology 2009; 253:861–869 Radiofrequency ablation in the management of
DL. Yttrium-90 radiotherapy for unresectable in- 16. Zgodzinski W, Espat NJ. Radiofrequency ablation liver metastases from breast cancer. AJR 2007;
trahepatic cholangiocarcinoma: a preliminary as- for incidentally identified primary intrahepatic 189:883–889
sessment of this novel treatment option. Ann Surg cholangiocarcinoma. World J Gastroenterol 21. Jan YY, Yeh CN, Yeh TS, Hwang TL, Chen MF.
Oncol 2010; 17:484–491 2005; 11:5239–5240 Clinicopathological factors predicting long-term
12. de Baere T, Deschamps F, Briggs P, et al. Hepatic 17. Chiou YY, Hwang JI, Chou YH, Wang HK, overall survival after hepatectomy for peripheral cho-
malignancies: percutaneous radiofrequency abla- Chiang JH, Chang CY. Percutaneous ultrasound- langiocarcinoma. World J Surg 2005; 29:894–898
tion during percutaneous portal or hepatic vein guided radiofrequency ablation of intrahepatic 22. Ohtsuka M, Ito H, Kimura F, et al. Extended he-
occlusion. Radiology 2008; 248:1056–1066 cholangiocarcinoma. Kaohsiung J Med Sci 2005; patic resection and outcomes in intrahepatic cho-
13. Cho YK, Kim JK, Kim MY, Rhim H, Han JK. 21:304–309 langiocarcinoma. J Hepatobiliary Pancreat Surg
Systematic review of randomized trials for hepa- 18. Falkson G, MacIntyre JM, Moertel CG, Johnson 2003; 10:259–264
tocellular carcinoma treated with percutaneous LA, Scherman RC. Primary liver cancer: an East- 23. Inoue K, Makuuchi M, Takayama T, et al. Long-
ablation therapies. Hepatology 2009; 49:453–459 ern Cooperative Oncology Group trial. Cancer term survival and prognostic factors in the surgi-
14. Stang A, Fischbach R, Teichmann W, Bokemeyer 1984; 54:970–977 cal treatment of mass-forming type cholangiocar-
C, Braumann D. A systematic review on the clini- 19. Goldberg SN, Grassi CJ, Cardella JF, et al.; Soci- cinoma. Surgery 2000; 127:498–505
cal benefit and role of radiofrequency ablation as ety of Interventional Radiology Technology As- 24. Takaki H, Yamakado K, Uraki J, et al. Radiofrequen-
treatment of colorectal liver metastases. Eur J sessment Committee and the International Work- cy ablation combined with chemoembolization for
Cancer 2009; 45:1748–1756 ing Group on Image-Guided Tumor Ablation. the treatment of hepatocellular carcinomas larger
15. Meloni MF, Andreano A, Laeseke PF, Livraghi T, Image-guided tumor ablation: standardization of than 5 cm. J Vasc Interv Radiol 2009; 20:217–224

AJR:196, February 2011 W209