Cardiovasc Intervent Radiol DOI 10.1007/s00270-014-1032-9
CLINICAL INVESTIGATION
Transarterial Chemoembolization for Hepatocellular Carcinomas with Central Bile Duct Invasion: Safety, Prognosis, and Predictive Factors Jin Woo Choi • Jin Wook Chung • Yun Ku Cho Yoon Jun Kim • Jung-Hwan Yoon • Hyo-Cheol Kim • Hwan Jun Jae
•
Received: 30 August 2014 / Accepted: 2 November 2014 Ó Springer Science+Business Media New York and the Cardiovascular and Interventional Radiological Society of Europe (CIRSE) 2014
Abstract Purpose To assess the safety and effectiveness of transarterial chemoembolization (TACE) of patients who have hepatocellular carcinomas (HCCs) with central bile duct invasion. Materials and Methods The institutional review board approved this retrospective study and waived informed consent. Fifty-three patients, initially treated with TACE for HCCs with central bile duct invasion from January 1999 to September 2012, were included. Clinical, laboratory, and survival data were reviewed. Complications and hospitalization length were evaluated using the v2 test, Fisher’s exact test, and logistic regression analysis. Survival was analyzed using the Kaplan–Meier method with log-rank test and Cox proportional hazard model. Results Seven patients experienced TACE-related major complications (severe post-embolization syndrome in 3, nonfatal sepsis in 3, and secondary bacterial peritonitis in 1). The overall major complication rate was 13.2 %, but there were no permanent adverse sequelae or deaths within 30 days. Serum total bilirubin C3.0 mg/dL was the only significant risk factor
Electronic supplementary material The online version of this article (doi:10.1007/s00270-014-1032-9) contains supplementary material, which is available to authorized users. J. W. Choi J. W. Chung (&) H.-C. Kim H. J. Jae Department of Radiology, Seoul National University Hospital, 101, Daehangno, Jongno-gu, Seoul 110-744, Korea e-mail: [email protected]
for long hospitalization [hazard ratio (HR) = 4.341, p = .022]. The median survival was 12.2 months. Extrahepatic metastasis (HR = 6.145, p \ .001), international normalized ratio (PT-INR) C1.20 (HR = 4.564, p \ .001), vascular invasion (HR = 3.484, p = .001), and intermediate tumor enhancement (HR = 2.417, p = .019) were significantly associated with shorter survival. Conclusion TACE can be a safe and effective treatment for patients who have HCCs with central bile duct invasion. In particular, long-term survival can be expected if patients have strongly enhancing tumors without poor prognostic factors such as extrahepatic metastasis, PT-INR prolongation, and vascular invasion. Keywords Transarterial chemoembolization Hepatocellular carcinoma Bile duct invasion Obstructive jaundice Abbreviations HCC Hepatocellular carcinoma TACE Transarterial chemoembolization CT Computed tomography AASLD American Association for the Study of Liver Disease SD Standard deviation ECOG Eastern Cooperative Oncology Group BCLC Barcelona Clinic Liver Cancer SIR Society of Interventional Radiology PT-INR International normalized ratio of prothrombin time HR Hazard ratio
Y. K. Cho Department of Radiology, VHS Medical Center, Seoul, Korea
Introduction
Y. J. Kim J.-H. Yoon Department of Internal Medicine, Seoul National University Hospital, Seoul, Korea
Central bile duct invasion of hepatocellular carcinoma (HCC) is relatively uncommon (0.5–13 %) and
123
J. W. Choi et al.: TACE for HCCs with BD Invasion
demonstrates unique clinical manifestations such as obstructive jaundice [1–6]. Thus, HCCs with central duct invasion are sometimes called ‘‘icteric types of HCC’’ or ‘‘cholestatic type of HCC’’ [1–6]. In contrast to the usual HCCs, this type of HCC may present with recognizable symptoms even when the tumor is small [3, 4]. In this light, HCCs with central bile duct invasion cause patients to seek medical care earlier in the course of their illness, and therefore there is a relatively higher chance of early detection and treatment. Conversely, some investigators have found that HCCs with bile duct invasion were associated with high rates of vascular invasion [7, 8] which suggests poor prognosis. For these reasons, HCCs with central bile duct invasion may be regarded as a special type of HCC. In regard to its management, favorable survival outcome has been reported in resectable HCCs with central bile duct invasion [9, 10]. Contrarily, there is limited knowledge regarding the treatment of unresectable HCCs with central bile duct invasion [5, 11–13]. Current practice guidelines do not make any recommendations for this particular patient population [14]. A high serum level of total bilirubin is considered as a relative contraindication for transarterial chemoembolization (TACE) because it is used as a surrogate for assessing liver function [15, 16]. However, jaundice in patients with intraductal tumor may have nothing to do with liver function; in such cases, the jaundice simply reflects bile duct obstruction. Furthermore, TACE is widely used for various conditions including HCCs with vascular invasion, bile duct invasion, etc. [13, 17–19]. Choi et al. [13] claimed that palliative TACE after biliary drainage for HCCs with bile duct invasion may improve survival outcomes and quality of life. In this context, further studies are warranted in terms of safety, prognosis, and risk factors of TACE for HCCs with central bile duct invasion. Therefore, we conducted a retrospective study to assess the safety and effectiveness of TACE for patients who have HCCs with central bile duct invasion.
Materials and Methods Patients The institutional review board approved this retrospective study and waived informed consent. This study was conducted in accordance of the declaration of Helsinki. From January 1999 to September 2012, a total of 7,051 patients were treated with TACE for HCCs at our institution. Clinicolaboratory findings, imaging features, and procedural details were accessed from the electronic database (Microsoft Access; Microsoft, Redmond, WA, USA). From
123
this database, we selected 66 patients for whom TACE was performed for newly diagnosed HCCs with bile duct invasion. Among these cases, we excluded eight patients for bile duct invasion that was peripheral to the first bifurcation level of the right or left hepatic ducts. An additional five patients were excluded for an uncertain diagnosis of HCC or bile duct invasion. Among the 53 final patients included in analysis, HCC was diagnosed by computed tomography (CT) findings in concordance with: the American Association for the Study of Liver Disease practice guidelines [20] (n = 36), equivocal CT findings with serum alpha-fetoprotein levels higher than 1,000 ng/ mL (n = 1), pathologic confirmation after surgical resection (n = 7), ultrasonography-guided biopsy (n = 6), and/ or endoscopy-guided biopsy (n = 3). In all 53 patients, bile duct invasion was diagnosed by the presence of expansile and arterially enhancing intraductal mass connected to the intrahepatic lesion [21]. Pathologic confirmation of the ductal lesions was established in 13 patients by endoscopyguided biopsy (n = 6) or surgery (n = 7). The final study population consisted of these 53 patients (39 males and 14 females) with an age range of 33–73 years [mean ± standard deviation (SD), 54.5 ± 9.6 years]. The need for, along with the method (percutaneous drainage, endoscopy-guided drainage) and timing, (before TACE, after TACE) biliary drainage was decided by hepatologists. In the presence of acute cholangitis, patients usually underwent biliary drainage and antibiotics (intravenous administration of ceftriaxone or cefotaxime) administration at first, and were treated with TACE after the resolution of acute cholangitis. Patients who received biliary drainage were also given antibiotics before and after TACE. The antibiotics were stopped after 24 h unless there were signs of infection/sepsis. Transarterial Chemoembolization All 53 patients in our study underwent conventional TACE with emulsion of 2–12 ml of iodized oil (Lipiodol; Andre Guerbe, Anulnay-sous-Bios, France) and 10–60 mg of doxorubicin hydrochloride (Adriamycin RDF; Ildong Pharmaceutical, Seoul, Korea), dependent upon the tumor size (Fig. 1). Gelatin sponge particles (Gelfoam; Upjohn, Kalamazoo, Mich; or, Cutanplast; Mascia Brunelli, Milano, Italy) are then mixed with 20 mg of doxorubicin hydrochloride or 2 mg of mitomycin-C (Mitomycin-10; Kyowa Hakko Kogyo, Tokyo, Japan). This solution was then administrated into the tumor-feeding arteries using a microcatheter until the blood flow nearly stopped. The delivery of chemoembolic agents was performed after selective catheterization of the tumor-supplying arteries as distally as possible in order to minimize unintended infusion into other vessels.
J. W. Choi et al.: TACE for HCCs with BD Invasion
Fig. 1 66-year-old woman who had unresectable hepatocellular carcinoma with central bile duct invasion. A Coronal image of portal phase computed tomography (CT) scan reveals a tumor (arrowheads) invading the common bile duct and the dilated intrahepatic ducts (arrows). B Digital subtraction celiac angiography demonstrates a hypervascular tumor (arrowheads). The patient underwent percutaneous biliary drainage and biliary stenting prior to the angiography.
C A spot image after transarterial chemoembolization shows iodized oil uptake in the tumor (arrows). Note that iodized oil is accumulated in the intraductal tumor portion (asterisk). D Axial image of postTACE unenhanced CT scan reveals compact iodized oil uptake in the intraductal tumor portion (arrowheads) as well as parenchymal tumor portion (arrows)
After the initial TACE, the patients underwent followup CT scans at a 2–3 months interval, and received additional TACE with the same regimen, if they had residual or recurred tumors.
to have portal hypertension in the presence of one of the following findings: (1) varices on upper endoscopy performed within a month prior to the TACE; (2) unequivocal varices on CT images; (3) ascites requiring diuretic treatment; or (4) splenomegaly on CT image associated with thrombocytopenia (platelet \100,000/mm3) [20].
Image Data Evaluation The patients underwent multi-phase CT scans within 2 weeks prior to TACE and unenhanced CT scans 2 weeks after the TACE to assess the distribution of the iodized oil. Two experienced radiologists (J.W.C and H.C.K) reviewed the CT images to assess the following: primary tumor size invading the bile duct (cm) by measuring the longest diameter on axial images, tumor enhancement (strong or intermediate enhancement) by visually comparing it to unaffected liver parenchymal enhancement, tumor multifocality, vascular invasion, the extent of bile duct invasion (right or left hepatic duct, common hepatic or bile duct), patterns of central bile duct invasion (peripheral to central, directly central), hemobilia on CT (intrabiliary hyperattenuation [40 Hounsfield units on pre-TACE unenhanced CT images), extrahepatic metastasis, ascites, portal hypertension, and iodized oil uptake pattern on postTACE CT (compact or partial). A patient was considered
Clinicolaboratory Data Evaluation Clinical findings and laboratory data were obtained from the electronic database and medical records of our institution, as mentioned above. We recorded clinical data including age, sex, initial symptoms, performance status score according to the Eastern Cooperative Oncology Group (ECOG), Barcelona Clinic Liver Cancer (BCLC) stage, Child-Pugh class, history of biliary drainage, hemobilia on the biliary drainage, TACE-related complications, and hospitalization length after TACE. TACErelated complications that required an additional hospital stay and special treatment for adverse events were classified as major complications [22]. Considering that most patients had symptoms such as fever and jaundice, and that clinicians considered biliary intervention before or after TACE, hospitalization length was divided into two groups
123
J. W. Choi et al.: TACE for HCCs with BD Invasion
(B7, [7 days) in our study. We also registered blood level tests examined within 3 days before TACE, including liver function tests, complete blood cell counts, international normalized ratio of prothrombin time (PT-INR), and alphafetoprotein. White blood cell counts were categorized as normal (4,000–10,000 cells/lL) and abnormal (i.e., leukocytosis or leukopenia), according to our institutional reference. Patients’ survival data were obtained from the Ministry of Public Administration and Security, which archives the death dates for all South Korean citizens. Due to this, there was no follow-up loss in our survival data. Statistical Analysis The rates of TACE-related major complications were calculated from the database. The v2 test or Fisher’s exact test was conducted to identify possible risk factors associated with the occurrence of major complications and a hospitalization length greater than 7 days. For statistical analyses, hemobilia was defined as the presence of either radiologic evidence on pre-TACE unenhanced CT images, or bloody drainage following biliary intervention. Subsequently, the variables for which p value was less than 0.10 were subjected to multivariate logistic regression analysis using the backward method. Median, 1-, 3-, and 5-year survival rates were estimated by the Kaplan–Meier method. To determine possible prognostic factors, we utilized the Kaplan–Meier method with the log-rank test. Parameters for which p values were less than 0.10 were then entered into the multivariate Cox proportional hazard model (backward method) to evaluate for associations with survival time. The proportional hazard assumption and goodness-of-fit were verified using logminus-log plot and Cox-Snell residuals. A p value of less than 0.05 was determined to indicate statistical significance. All statistical analyses were performed using software (IBM SPSS Statistics, version 19.0; SPSS Inc., IBM Company, Armonk, NY).
Results Forty-five out of 53 patients (84.9 %) experienced initial signs/symptoms which prompted each patient to seek medical care and the subsequent diagnosis: jaundice (n = 25), upper abdominal pain (n = 16), and fever (n = 4). In the remaining eight asymptomatic patients (15.1 %, 8/53), HCC was detected for seven hepatitis B virus carrier patients during routine surveillance and incidentally found in the last patient. In terms of BCLC staging, 4, 47, and 2 patients were classified as intermediate, advanced, and terminal stage, respectively. The 47 patients
123
were in advanced stage due to vascular invasion (n = 17), extrahepatic metastasis (n = 3), vascular invasion with extrahepatic metastasis (n = 3), or poor performance status (n = 24). As for Child-Pugh class, 20, 31, and 2 patients belonged to class A, B, and C, respectively. In patients with Child-Pugh class C, clinicians carefully selected the candidates for TACE by assuming their hepatic functional reserve, based on laboratory findings to the exclusion of bilirubin level and imaging features suggesting liver cirrhosis. The index tumor size ranged from 1.0 to 12.7 cm (mean ± SD, 5.4 ± 2.6 cm). Prior to TACE, 17 and 8 patients underwent percutaneous biliary drainage and endoscopy-guided biliary drainage, respectively. At the time of TACE, the mean ± SD values of total bilirubin were 3.7 ± 5.01 mg/ dL among patients without biliary intervention, and 10.0 ± 6.3 mg/dL among patients who have undergone biliary intervention (p \ .001). Among the 28 patients who had not undergone pre-TACE biliary drainage, 8 patients showed hyperbilirubinemia (range of total bilirubin, 3.4–21.6 mg/dL) at the time of TACE. However, they were managed by TACE without prior biliary drainage because they did not have acute cholangitis. After the TACE, obstructive jaundice in 3 out of the 8 patients gradually resolved without any biliary interventions. Detailed information about our study population is summarized in Table 1. Safety: Complications and Hospitalization Periods After TACE, the patients stayed in the hospital from 1 to 52 days (median, 7 days; interquartile range, 3–15 days). Twenty-two patients (41.5 %) had been hospitalized for more than 7 days either for biliary intervention after TACE (n = 15) or for TACE-related complications (n = 7). With regard to TACE-related major complications, 3 patients (5.7 %, 3/53) experienced severe post-embolization syndrome requiring hospitalization longer than 7 days, and 3 others (5.7 %, 3/53) experienced non-fatal sepsis (gram-negative bacteria from the gastrointestinal tract) in spite of prophylactic antibiotics. Among the latter group of patients with non-fatal sepsis, 2 patients had received endoscopy-guided biliary drainage prior to TACE, and the remaining patient had not received any biliary intervention beforehand. They were recovered without any permanent adverse sequelae, after dedicated antibiotic therapy according to the results of blood culture. Secondary bacterial peritonitis occurred in one patient (1.9 %, 1/53) within a week after TACE, and this patient required 22 days of hospitalization. None of the patients died within 30 days from the respective dates of TACE procedure. In addition, all of the patients were discharged without permanent adverse sequelae secondary to the TACE.
J. W. Choi et al.: TACE for HCCs with BD Invasion Table 1 Characteristics of 53 patients analyzed in our study
Table 1 continued
Item
Item
Number (%)
Age
Item
Number (%)
Primary tumor size
C55 years
26 (49.1)
\55 years
27 (50.9)
Sex
C5.0 cm
26 (49.1)
\5.0 cm
27 (50.9)
Tumor multifocality
Male
39 (73.6)
Unifocal
27 (50.9)
Female
14 (26.4)
Multifocal
26 (49.1)
Portal hypertension Absent Present
34 (64.2) 19 (35.8)
Ascites
Vascular invasion Absent Present
33 (62.3) 20 (37.7)
Tumor enhancement
Absent
40 (75.5)
Present
13 (24.5)
Albumin
Strong
21 (39.6)
Intermediate
32 (60.4)
Extent of bile duct invasion
C3.5 g/dL
29 (54.7)
\3.5 g/dL
24 (45.3)
Total bilirubin
RHD/LHD
18 (34.0)
CHD/CBD
35 (66.0)
Central bile duct invasion
C3.0 mg/dL
28 (52.8)
\3.0 mg/dL
25 (47.2)
PT-INR
Direct
32 (60.4)
From periphery
21 (39.6)
Extrahepatic metastasis
C1.20
13 (24.5)
Absent
47 (88.7)
\1.20
40 (75.5)
Present
6 (11.3)
Aspartate aminotransferase
Hemobilia
C80 IU/L
31 (58.5)
Absent
39 (73.6)
\80 IU/L
22 (41.5)
Present
14 (26.4)
Alanine aminotransferase
Iodized oil uptake after TACE
C60 IU/L
27 (50.9)
Compact
27 (50.9)
\60 IU/L
26 (49.1)
Faint
26 (49.1)
Alkaline phosphatase C200 IU/L
Performance status 22 (41.5)
ECOG 0
\200 IU/L
31 (58.5)
ECOG 1–2
Alpha-fetoprotein
8 (15.1)
Pre-TACE biliary drainage
C1,000 ng/mL
30 (56.6)
None
28 (52.8)
23 (43.4)
Percutaneous
17 (32.1)
Endoscopy
Present
8 (15.1)
BCLC stage 49 (92.5) 4 (7.5)
Etiology of HCC
Intermediate
Item
Number (%)
HBV
40 (75.5)
C
2 (3.8)
HCV
1 (1.9)
RHD/LHD right hepatic duct or left hepatic duct, CHD/CBD common hepatic duct or common bile duct, ECOG Eastern Cooperative Oncology Group
According to the results of univariate analyses, serum aspartate aminotransferase C80 IU/L (p = .033) and ascites (p = .053) were subjected to multivariate analysis. Other variables, such as total bilirubin C3.0 mg/dL (p = .426), pre-TACE biliary drainage (p = .168), hemobilia (p = .364), and performance status (p = 1.000) were unlikely to be related to major complications. In this analysis, there was no significant risk factor for major TACErelated complications. With respect to hospitalization length, serum total bilirubin C3.0 mg/dL (p = .023), ascites (p = .070), and serum alanine aminotransferase C60 IU/L (p = .087) were applied to multivariate analysis. Variables such as vascular invasion (p = .336), portal hypertension (p = .217), and bile duct invasion extent (p = .255) were unlikely to be associated with long hospitalization. From the results of the multivariate analysis, pre-TACE total bilirubin C3.0 mg/ dL was the only significant risk factor for hospitalization [7 days [hazard ratio (HR) = 4.341, 95 % confidence interval 1.236–15.247, p = .022]. Among the patients who underwent biliary drainage before TACE (n = 25), there were no significant differences in the occurrence of major complications (p = .283) and hospitalization length (p = .311), regardless of biliary intervention performed (percutaneous drainage, endoscopy-guided drainage). Effectiveness: Survival and Prognostic Factors
45 (84.9)
\1,000 ng/mL Leukocytosis or leukopenia Absent
Number (%)
4 (7.5)
Advanced
47 (88.7)
Terminal
2 (3.8)
Child-Pugh class
Alcohol
3 (5.7)
A
20 (37.7)
Cryptogenic
9 (17.0)
B
31 (58.5)
The median survival of patients who received TACE for unresectable HCCs with central bile duct invasion was 12.2 months (Fig. 2). The 1-, 3-, and 5-year survival rates were 50.9, 20.9, and 20.9 %, respectively. With regard to the BCLC stage, 88.7 % (n = 47) of our study population belonged to advanced stage, and the median survival period for these patients was 10.8 months. In terms of Child-Pugh classification, 96.2 % (n = 51) of the patients were either in class A (37.5 %, n = 20) or in class B (58.5 %, n = 31), and their median survivals were 12.6 and 10.5 months, respectively (p = .707) (Fig. E1). From the results of univariate analyses, vascular invasion (p = .002), tumor multifocality (p = .003), extrahepatic metastasis (p = .010), tumor
123
J. W. Choi et al.: TACE for HCCs with BD Invasion Table 2 Univariate analyses to determine potential prognostic factors of TACE for unresectable HCCs with central bile duct invasion Item
p value
Age C55 years
.795
.888
Present .220
.302
123
RHD/LHD
.414
Central bile duct invasion .711
\3.0 mg/dL
Direct
.107
From periphery Extraheaptic metastasis .057à
\1.20
Absent
.010à
Present
Aspartate aminotransferase C80 IU/L .196 \80 IU/L
Hemobilia Absent
.995
Present
Alanine aminotransferase .976
\60 IU/L
Performance status ECOG 0
.666
ECOG 1–2
Alkaline phosphatase C200 IU/L
.044à
CHD/CBD
PT-INR
60 IU/L
Strong Intermediate
Total bilirubin
C1.20
.002à
Extent of bile duct invasion
\3.5 g/dL
enhancement (p = .044), and PT-INR C 1.20 (p = .057) were entered into the multivariate analysis (Table 2). Other variables such as portal hypertension (p = .888), bile duct invasion extent (p = .414), total bilirubin C3.0 mg/dL (p = .711), and performance status (p = .666) were unlikely to be related to the patients’ prognosis. From the results of multivariate analysis, extrahepatic metastasis (HR = 6.145, p \ .001), PT-INR C1.20 (HR = 4.564, p \ .001), vascular invasion (HR = 3.484, p = .001), and intermediate tumor enhancement (HR = 2.417, p = .019) were significantly associated to shorter survival after TACE (Table 3). The median survival of the patients (n = 12) who did not have any of these risk factors was 78.9 months in our study (Fig. E2). For these patients, the index tumor size and total bilirubin level were 3.9 ± 1.2 cm and 5.7 ± 6.84 mg/ dL, respectively. Five out of the 12 patients underwent surgical resection during their follow-up period. In contrast, the median survival of the remaining 41 patients was 10.5 months. Among the patients who underwent biliary drainage before TACE (n = 25), there were no significant differences in survival (p = .816) regardless of the type of biliary intervention performed (percutaneous drainage, endoscopy-guided drainage). Twenty-seven out of the total 53 patients (50.9 %) demonstrated compact iodized oil uptake on post-TACE CT images, and their survival (median, 23.1 months) was significantly longer than that of patients with partial iodized oil uptake (median, 7.8 months) (p \ .001) (Fig. 3). Regarding the pre-TACE CT findings of the 27
Vascular invasion Absent Present
Present
C3.0 mg/dL
.003à
Tumor enhancement
Albumin
Fig. 2 Kaplan–Meier curve of overall cumulative survival rate after TACE (median survival, 12.2 months)
Unifocal Multifocal
Ascites
C3.5 g/dL
.585
Tumor multifocality .160
Female
Absent
C5.0 cm \5.0 cm
Sex
Portal hypertension Absent
p value
Primary tumor size
\55 years Male
Item
Leukocytosis or leukopenia .336
\200 IU/L
Absent
.957
Present
Alpha-fetoprotein C1,000 ng/mL
.430
\1,000 ng/mL RHD/LHD right hepatic duct or left hepatic duct, CHD/CBD common hepatic duct or common bile duct, ECOG Eastern Cooperative Oncology Group à Factors were included in multivariate analysis, if p value \.10 in univariate analyses
tumors, 12 and 15 HCCs demonstrated intermediate and strong enhancement, respectively (Fig. 1). A patient with compact iodized oil uptake achieved complete remission status after the third session of TACE. Additionally, 9 patients (17.0 %, 9/53), who had viable intraductal tumors after 1–5 sessions of TACE and good hepatic functional reserve, subsequently underwent surgery (hemihepatectomy in 6 patients, extended hemihepatectomy in 2, liver transplantation in 1). For the group of
J. W. Choi et al.: TACE for HCCs with BD Invasion Table 3 Multivariate analysis to determine significant prognostic factors of TACE for unresectable HCCs with central bile duct invasion Variable
Hazard ratio
95 % CI
p value
Extrahepatic metastasis
6.145
2.292–16.474 \.001 1.968–10.584 \.001
PT-INR C1.20
4.564
Vascular invasion
3.484
1.670–7.268
.001
Intermediate tumor enhancement
2.417
1.155–5.059
.019
Multifocal tumor
NS
NS not significant
patients who underwent surgery, the index tumor size and total bilirubin level before the initial TACE were 5.6 ± 1.9 cm and 11.2 ± 8.21 mg/dL, respectively. Four patients had been planned to undergo surgical resection after preoperative biliary intervention and a single session TACE. The remaining five patients decided to undergo surgery after several TACE sessions due to prolonged jaundice, in spite of their own well-controlled tumors. The median survival period for the nine patients who underwent surgical resection after TACE was 62.8 months (Fig. E3).
Discussion How to manage unresectable HCCs with central bile duct invasion remains undefined due to the lack of evidence. Therefore, this rare disease has been managed by locoregional therapy, sorafenib, or best supportive care, all dependent on local expertise. Through this study, we demonstrated acceptable complication rates (13.2 %) even after taking account of complications with uncertain causal relationship to TACE—for example, non-fatal sepsis (n = 3) and secondary bacterial peritonitis (n = 1) [23]. Furthermore, based on the earlier studies [15, 24], a considerable number of our patients had risk factors for complications such as hyperbilirubinemia or vascular invasion. These results suggest that TACE can be safely conducted for HCCs with central bile duct invasion, with selective chemoembolic agent infusion into the tumor-feeding arteries. Total bilirubin [3.0 mg/dl was a predictor of prolonged hospital stay and, not surprisingly, the main factor prolonging hospital stay in this patient group was being treated for hyperbilirubinemia. However, the patients who have undergone pre-TACE biliary drainage, including those with total bilirubin C3.0 mg/dL, did not require significantly longer hospitalization. This is because the control of cholangitis was considered more important in determining a patient’s discharge, rather than the total bilirubin level itself. These results imply that even though HCCs with
Fig. 3 Kaplan–Meier curves of cumulative survival rates after TACE by iodized oil uptake on post-TACE CT imaging
central bile duct invasion induce substantial obstructive jaundice, TACE can be performed along with appropriate biliary intervention. Concordantly, Lai et al. [4] claimed that mean survival period was remarkably longer for patients treated with biliary drainage and subsequent TACE (8.0–13.4 months) than that of patients who have received biliary drainage alone (2.5–4.5 months). These results imply that obstructive hyperbilirubinemia in patients of HCCs with central bile duct invasion is not a contraindication for TACE, and that pre-TACE biliary intervention should be considered. Our study also presented acceptable survival outcome of TACE (median survival, 12.2 months). In particular, the 12 patients without poor prognostic factors (extrahepatic metastasis, PT-INR C1.20, vascular invasion, and intermediate tumor enhancement) had 78.9 months of median survival period in our study. These results suggest that TACE can be considered as a safe and effective treatment option for HCCs with central bile duct invasion. Presently, TACE may be the only promising locoregional therapy for HCCs with central bile duct invasion, because radiofrequency ablation to these tumors is able to cause lifethreatening injury to hilar structures. In addition, irreversible electroporation is still in an evolving state. External beam radiation therapy and radioembolization may be alternatives of TACE. However, there is little or no knowledge, until now, about the efficacy and safety of these methods for HCCs with bile duct invasion.
123
J. W. Choi et al.: TACE for HCCs with BD Invasion
In terms of the prognosis factors, the PT-INR level is more useful than the total bilirubin level or performance status in these patients. Given that it reflects biosynthetic capacity of the liver [25], the PT-INR level could be a more reliable marker to assess hepatic reserve—more than the total bilirubin level or performance status in the setting of central bile duct involvement. Concordantly, PT-INR levels demonstrated a significant association to survival outcomes in our study. In addition, tumor enhancement degree on pre-TACE CT images was also a significant prognostic factor in our study. Relatively, hypovascular HCCs with intermediate enhancement on pre-TACE CT may represent poor differentiation or cholangiocarcinoma components. Hence, these tumors can be rather intractable for TACE [26–28]. On the other hand, compact iodized oil uptake after TACE, associated with vascularity and biological background of the HCC tumors, indicated better prognosis. This is consistent with previous studies which reported the importance of iodized oil uptake after TACE as a useful prognostic factor [16, 29–32]. In general, the BCLC staging system is very useful to build treatment plans for most HCC patients because this algorithm takes into account patient factors such as liver function and performance status. Additionally, the algorithm considers tumor factors such as size, number, and vascular invasion [14]. This system is largely based on the premise that patient factors reflect gradual deterioration of hepatic reserve as the HCC progresses in addition to underlying liver disease. In case this assumption is broken, the BCLC algorithm then needs to be used flexibly depending on each condition. In our study, obstructive jaundice from HCCs with central bile duct invasion appeared to have substantially over-inflated the BCLC stages and Child-Pugh classes. Specifically, 92.5 % of our patients were in advanced stage (88.7 %) or terminal stage (3.8 %) according to the BCLC staging system, and 62.3 % were in class B (58.5 %) or C (3.8 %) depending on the Child-Pugh classification. In addition, total bilirubin level and performance status did not significantly affect prognosis in our patients. These suggest that hyperbilirubinemia and compromised performance status, in cases with biliary tumor invasion, should be regarded differently from those caused by hepatic parenchymal dysfunction. There are some limitations in our study. First, the number of patients analyzed in our study was limited (n = 53). The second limitation of this study is that we did not address the efficacy of surgical intervention and the optimal timing/method of biliary intervention. Due to the retrospective nature of our study, treatment strategies varied among the patients. In conclusion, TACE can be a safe and effective treatment for patients who have unresectable HCCs with central bile duct invasion In particular, long-term survival can be
123
expected if patients have strongly enhancing tumors without poor prognostic factors such as extrahepatic metastasis, PT-INR prolongation, and vascular invasion. Conflict of interest Jin Woo Choi: no conflict of interest; Jin Wook Chung: no conflict of interest; Yun Ku Cho: no conflict of interest; Yoon Jun Kim: no conflict of interest; Jung-Hwan Yoon: no conflict of interest; Hyo-Cheol Kim: no conflict of interest; Hwan Jun Jae: no conflict of interest.
References 1. Lin TY, Chen KM, Chen YR, Lin WS, Wang TH, Sung JL (1975) Icteric type hepatoma. Med Chir Dig 4:267–270 2. Kojiro M, Kawabata K, Kawano Y, Shirai F, Takemoto N, Nakashima T (1982) Hepatocellular carcinoma presenting as intrabile duct tumor growth: a clinicopathologic study of 24 cases. Cancer 49:2144–2147 3. Lau WY, Leow CK, Leung KL, Leung TW, Chan M, Yu SC (2000) Cholangiographic features in the diagnosis and management of obstructive icteric type hepatocellular carcinoma. HPB Surg 11:299–306 4. Lai EC, Lau WY (2006) Hepatocellular carcinoma presenting with obstructive jaundice. ANZ J Surg 76:631–636 5. Huang JF, Wang LY, Lin ZY et al (2002) Incidence and clinical outcome of icteric type hepatocellular carcinoma. J Gastroenterol Hepatol 17:190–195 6. Shiomi M, Kamiya J, Nagino M et al (2001) Hepatocellular carcinoma with biliary tumor thrombi: aggressive operative approach after appropriate preoperative management. Surgery 129:692–698 7. Yeh CN, Jan YY, Lee WC, Chen MF (2004) Hepatic resection for hepatocellular carcinoma with obstructive jaundice due to biliary tumor thrombi. World J Surg 28:471–475 8. Gabata T, Terayama N, Kobayashi S, Sanada J, Kadoya M, Matsui O (2007) MR imaging of hepatocellular carcinomas with biliary tumor thrombi. Abdom Imaging 32:470–474 9. Satoh S, Ikai I, Honda G et al (2000) Clinicopathologic evaluation of hepatocellular carcinoma with bile duct thrombi. Surgery 128:779–783 10. Moon DB, Hwang S, Wang HJ et al (2013) Surgical outcomes of hepatocellular carcinoma with bile duct tumor thrombus: a Korean multicenter study. World J Surg 37:443–451 11. Kitagawa K, Yamakado K, Nakatsuka A et al (1999) Selective transcatheter hepatic arterial chemoembolization for hemobilia from hepatocellular carcinoma: report of three cases. J Vasc Interv Radiol 10:1357–1360 12. Okuda M, Miyayama S, Yamashiro M et al (2010) Sloughing of intraductal tumor thrombus of hepatocellular carcinoma after transcatheter arterial chemoembolization. Cardiovasc Interv Radiol 33:619–623 13. Choi J, Ryu JK, Lee SH et al (2013) Palliative treatment of unresectable hepatocellular carcinoma with obstructive jaundice using biliary drainage with subsequent transarterial chemoembolization. J Palliat Med 16:1026–1033 14. Llovet JM, Bru C, Bruix J (1999) Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 19:329–338 15. Chung JW, Park JH, Han JK et al (1996) Hepatic tumors: predisposing factors for complications of transcatheter oily chemoembolization. Radiology 198:33–40
J. W. Choi et al.: TACE for HCCs with BD Invasion 16. Mondazzi L, Bottelli R, Brambilla G et al (1994) Transarterial oily chemoembolization for the treatment of hepatocellular carcinoma: a multivariate analysis of prognostic factors. Hepatology 19:1115–1123 17. Pinter M, Hucke F, Graziadei I et al (2012) Advanced-stage hepatocellular carcinoma: transarterial chemoembolization versus sorafenib. Radiology 263:590–599 18. Kothary N, Weintraub JL, Susman J, Rundback JH (2007) Transarterial chemoembolization for primary hepatocellular carcinoma in patients at high risk. J Vasc Interv Radiol 18:1517–1526 19. Kim HC, Lee JH, Chung JW et al (2013) Transarterial chemoembolization with additional cisplatin infusion for hepatocellular carcinoma invading the hepatic vein. J Vasc Interv Radiol 24:274–283 20. Bruix J, Sherman M, American Association for the Study of Liver Disease (2011) Management of hepatocellular carcinoma: an update. Hepatology 53:1020–1022 21. Jung AY, Lee JM, Choi SH et al (2006) CT features of an intraductal polypoid mass: differentiation between hepatocellular carcinoma with bile duct tumor invasion and intraductal papillary cholangiocarcinoma. J Comput Assist Tomogr 30:173–181 22. Sacks D, McClenny TE, Cardella JF, Lewis CA (2003) Society of Interventional Radiology Clinical Practice guidelines. J Vasc Interv Radiol 14:S199–S202 23. Brown DB, Nikolic B, Covey AM et al (2012) Quality improvement guidelines for transhepatic arterial chemoembolization, embolization, and chemotherapeutic infusion for hepatic malignancy. J Vasc Interv Radiol 23:287–294 24. Gates J, Hartnell GG, Stuart KE, Clouse ME (1999) Chemoembolization of hepatic neoplasms: safety, complications, and when to worry. Radiographics 19:399–414
25. Thapa BR, Walia A (2007) Liver function tests and their interpretation. Indian J Pediatr 74:663–671 26. Sahani DV, Holalkere NS, Mueller PR, Zhu AX (2007) Advanced hepatocellular carcinoma: CT perfusion of liver and tumor tissue—initial experience. Radiology 243:736–743 27. Lee JH, Lee JM, Kim SJ et al (2012) Enhancement patterns of hepatocellular carcinomas on multiphasicmultidetector row CT: comparison with pathological differentiation. Br J Radiol 85:e573–e583 28. Choi BI, Han JK, Kim YI et al (1994) Combined hepatocellular and cholangiocarcinoma of the liver: sonography, CT, angiography, and iodized-oil CT with pathologic correlation. Abdom Imaging 19:43–46 29. Takayasu K, Muramatsu Y, Maeda T et al (2001) Targeted transarterial oily chemoembolization for small foci of hepatocellular carcinoma using a unified helical CT and angiography system: analysis of factors affecting local recurrence and survival rates. Am J Roentgenol 176:681–688 30. Vogl TJ, Trapp M, Schroeder H et al (2000) Transarterial chemoembolization for hepatocellular carcinoma: volumetric and morphologic CT criteria for assessment of prognosis and therapeutic success-results from a liver transplantation center. Radiology 214:349–357 31. Kawaguchi T, Ohkawa K, Imanaka K et al (2012) Lipiodol accumulation and transarterial chemoembolization efficacy for HCC patients. Hepatogastroenterology 59:219–223 32. Kim DY, Ryu HJ, Choi JY et al (2012) Radiological response predicts survival following transarterial chemoembolisation in patients with unresectable hepatocellular carcinoma. Aliment Pharmacol Ther 35:1343–1350
123
CLINICAL INVESTIGATION
Transarterial Chemoembolization for Hepatocellular Carcinomas with Central Bile Duct Invasion: Safety, Prognosis, and Predictive Factors Jin Woo Choi • Jin Wook Chung • Yun Ku Cho Yoon Jun Kim • Jung-Hwan Yoon • Hyo-Cheol Kim • Hwan Jun Jae
•
Received: 30 August 2014 / Accepted: 2 November 2014 Ó Springer Science+Business Media New York and the Cardiovascular and Interventional Radiological Society of Europe (CIRSE) 2014
Abstract Purpose To assess the safety and effectiveness of transarterial chemoembolization (TACE) of patients who have hepatocellular carcinomas (HCCs) with central bile duct invasion. Materials and Methods The institutional review board approved this retrospective study and waived informed consent. Fifty-three patients, initially treated with TACE for HCCs with central bile duct invasion from January 1999 to September 2012, were included. Clinical, laboratory, and survival data were reviewed. Complications and hospitalization length were evaluated using the v2 test, Fisher’s exact test, and logistic regression analysis. Survival was analyzed using the Kaplan–Meier method with log-rank test and Cox proportional hazard model. Results Seven patients experienced TACE-related major complications (severe post-embolization syndrome in 3, nonfatal sepsis in 3, and secondary bacterial peritonitis in 1). The overall major complication rate was 13.2 %, but there were no permanent adverse sequelae or deaths within 30 days. Serum total bilirubin C3.0 mg/dL was the only significant risk factor
Electronic supplementary material The online version of this article (doi:10.1007/s00270-014-1032-9) contains supplementary material, which is available to authorized users. J. W. Choi J. W. Chung (&) H.-C. Kim H. J. Jae Department of Radiology, Seoul National University Hospital, 101, Daehangno, Jongno-gu, Seoul 110-744, Korea e-mail: [email protected]
for long hospitalization [hazard ratio (HR) = 4.341, p = .022]. The median survival was 12.2 months. Extrahepatic metastasis (HR = 6.145, p \ .001), international normalized ratio (PT-INR) C1.20 (HR = 4.564, p \ .001), vascular invasion (HR = 3.484, p = .001), and intermediate tumor enhancement (HR = 2.417, p = .019) were significantly associated with shorter survival. Conclusion TACE can be a safe and effective treatment for patients who have HCCs with central bile duct invasion. In particular, long-term survival can be expected if patients have strongly enhancing tumors without poor prognostic factors such as extrahepatic metastasis, PT-INR prolongation, and vascular invasion. Keywords Transarterial chemoembolization Hepatocellular carcinoma Bile duct invasion Obstructive jaundice Abbreviations HCC Hepatocellular carcinoma TACE Transarterial chemoembolization CT Computed tomography AASLD American Association for the Study of Liver Disease SD Standard deviation ECOG Eastern Cooperative Oncology Group BCLC Barcelona Clinic Liver Cancer SIR Society of Interventional Radiology PT-INR International normalized ratio of prothrombin time HR Hazard ratio
Y. K. Cho Department of Radiology, VHS Medical Center, Seoul, Korea
Introduction
Y. J. Kim J.-H. Yoon Department of Internal Medicine, Seoul National University Hospital, Seoul, Korea
Central bile duct invasion of hepatocellular carcinoma (HCC) is relatively uncommon (0.5–13 %) and
123
J. W. Choi et al.: TACE for HCCs with BD Invasion
demonstrates unique clinical manifestations such as obstructive jaundice [1–6]. Thus, HCCs with central duct invasion are sometimes called ‘‘icteric types of HCC’’ or ‘‘cholestatic type of HCC’’ [1–6]. In contrast to the usual HCCs, this type of HCC may present with recognizable symptoms even when the tumor is small [3, 4]. In this light, HCCs with central bile duct invasion cause patients to seek medical care earlier in the course of their illness, and therefore there is a relatively higher chance of early detection and treatment. Conversely, some investigators have found that HCCs with bile duct invasion were associated with high rates of vascular invasion [7, 8] which suggests poor prognosis. For these reasons, HCCs with central bile duct invasion may be regarded as a special type of HCC. In regard to its management, favorable survival outcome has been reported in resectable HCCs with central bile duct invasion [9, 10]. Contrarily, there is limited knowledge regarding the treatment of unresectable HCCs with central bile duct invasion [5, 11–13]. Current practice guidelines do not make any recommendations for this particular patient population [14]. A high serum level of total bilirubin is considered as a relative contraindication for transarterial chemoembolization (TACE) because it is used as a surrogate for assessing liver function [15, 16]. However, jaundice in patients with intraductal tumor may have nothing to do with liver function; in such cases, the jaundice simply reflects bile duct obstruction. Furthermore, TACE is widely used for various conditions including HCCs with vascular invasion, bile duct invasion, etc. [13, 17–19]. Choi et al. [13] claimed that palliative TACE after biliary drainage for HCCs with bile duct invasion may improve survival outcomes and quality of life. In this context, further studies are warranted in terms of safety, prognosis, and risk factors of TACE for HCCs with central bile duct invasion. Therefore, we conducted a retrospective study to assess the safety and effectiveness of TACE for patients who have HCCs with central bile duct invasion.
Materials and Methods Patients The institutional review board approved this retrospective study and waived informed consent. This study was conducted in accordance of the declaration of Helsinki. From January 1999 to September 2012, a total of 7,051 patients were treated with TACE for HCCs at our institution. Clinicolaboratory findings, imaging features, and procedural details were accessed from the electronic database (Microsoft Access; Microsoft, Redmond, WA, USA). From
123
this database, we selected 66 patients for whom TACE was performed for newly diagnosed HCCs with bile duct invasion. Among these cases, we excluded eight patients for bile duct invasion that was peripheral to the first bifurcation level of the right or left hepatic ducts. An additional five patients were excluded for an uncertain diagnosis of HCC or bile duct invasion. Among the 53 final patients included in analysis, HCC was diagnosed by computed tomography (CT) findings in concordance with: the American Association for the Study of Liver Disease practice guidelines [20] (n = 36), equivocal CT findings with serum alpha-fetoprotein levels higher than 1,000 ng/ mL (n = 1), pathologic confirmation after surgical resection (n = 7), ultrasonography-guided biopsy (n = 6), and/ or endoscopy-guided biopsy (n = 3). In all 53 patients, bile duct invasion was diagnosed by the presence of expansile and arterially enhancing intraductal mass connected to the intrahepatic lesion [21]. Pathologic confirmation of the ductal lesions was established in 13 patients by endoscopyguided biopsy (n = 6) or surgery (n = 7). The final study population consisted of these 53 patients (39 males and 14 females) with an age range of 33–73 years [mean ± standard deviation (SD), 54.5 ± 9.6 years]. The need for, along with the method (percutaneous drainage, endoscopy-guided drainage) and timing, (before TACE, after TACE) biliary drainage was decided by hepatologists. In the presence of acute cholangitis, patients usually underwent biliary drainage and antibiotics (intravenous administration of ceftriaxone or cefotaxime) administration at first, and were treated with TACE after the resolution of acute cholangitis. Patients who received biliary drainage were also given antibiotics before and after TACE. The antibiotics were stopped after 24 h unless there were signs of infection/sepsis. Transarterial Chemoembolization All 53 patients in our study underwent conventional TACE with emulsion of 2–12 ml of iodized oil (Lipiodol; Andre Guerbe, Anulnay-sous-Bios, France) and 10–60 mg of doxorubicin hydrochloride (Adriamycin RDF; Ildong Pharmaceutical, Seoul, Korea), dependent upon the tumor size (Fig. 1). Gelatin sponge particles (Gelfoam; Upjohn, Kalamazoo, Mich; or, Cutanplast; Mascia Brunelli, Milano, Italy) are then mixed with 20 mg of doxorubicin hydrochloride or 2 mg of mitomycin-C (Mitomycin-10; Kyowa Hakko Kogyo, Tokyo, Japan). This solution was then administrated into the tumor-feeding arteries using a microcatheter until the blood flow nearly stopped. The delivery of chemoembolic agents was performed after selective catheterization of the tumor-supplying arteries as distally as possible in order to minimize unintended infusion into other vessels.
J. W. Choi et al.: TACE for HCCs with BD Invasion
Fig. 1 66-year-old woman who had unresectable hepatocellular carcinoma with central bile duct invasion. A Coronal image of portal phase computed tomography (CT) scan reveals a tumor (arrowheads) invading the common bile duct and the dilated intrahepatic ducts (arrows). B Digital subtraction celiac angiography demonstrates a hypervascular tumor (arrowheads). The patient underwent percutaneous biliary drainage and biliary stenting prior to the angiography.
C A spot image after transarterial chemoembolization shows iodized oil uptake in the tumor (arrows). Note that iodized oil is accumulated in the intraductal tumor portion (asterisk). D Axial image of postTACE unenhanced CT scan reveals compact iodized oil uptake in the intraductal tumor portion (arrowheads) as well as parenchymal tumor portion (arrows)
After the initial TACE, the patients underwent followup CT scans at a 2–3 months interval, and received additional TACE with the same regimen, if they had residual or recurred tumors.
to have portal hypertension in the presence of one of the following findings: (1) varices on upper endoscopy performed within a month prior to the TACE; (2) unequivocal varices on CT images; (3) ascites requiring diuretic treatment; or (4) splenomegaly on CT image associated with thrombocytopenia (platelet \100,000/mm3) [20].
Image Data Evaluation The patients underwent multi-phase CT scans within 2 weeks prior to TACE and unenhanced CT scans 2 weeks after the TACE to assess the distribution of the iodized oil. Two experienced radiologists (J.W.C and H.C.K) reviewed the CT images to assess the following: primary tumor size invading the bile duct (cm) by measuring the longest diameter on axial images, tumor enhancement (strong or intermediate enhancement) by visually comparing it to unaffected liver parenchymal enhancement, tumor multifocality, vascular invasion, the extent of bile duct invasion (right or left hepatic duct, common hepatic or bile duct), patterns of central bile duct invasion (peripheral to central, directly central), hemobilia on CT (intrabiliary hyperattenuation [40 Hounsfield units on pre-TACE unenhanced CT images), extrahepatic metastasis, ascites, portal hypertension, and iodized oil uptake pattern on postTACE CT (compact or partial). A patient was considered
Clinicolaboratory Data Evaluation Clinical findings and laboratory data were obtained from the electronic database and medical records of our institution, as mentioned above. We recorded clinical data including age, sex, initial symptoms, performance status score according to the Eastern Cooperative Oncology Group (ECOG), Barcelona Clinic Liver Cancer (BCLC) stage, Child-Pugh class, history of biliary drainage, hemobilia on the biliary drainage, TACE-related complications, and hospitalization length after TACE. TACErelated complications that required an additional hospital stay and special treatment for adverse events were classified as major complications [22]. Considering that most patients had symptoms such as fever and jaundice, and that clinicians considered biliary intervention before or after TACE, hospitalization length was divided into two groups
123
J. W. Choi et al.: TACE for HCCs with BD Invasion
(B7, [7 days) in our study. We also registered blood level tests examined within 3 days before TACE, including liver function tests, complete blood cell counts, international normalized ratio of prothrombin time (PT-INR), and alphafetoprotein. White blood cell counts were categorized as normal (4,000–10,000 cells/lL) and abnormal (i.e., leukocytosis or leukopenia), according to our institutional reference. Patients’ survival data were obtained from the Ministry of Public Administration and Security, which archives the death dates for all South Korean citizens. Due to this, there was no follow-up loss in our survival data. Statistical Analysis The rates of TACE-related major complications were calculated from the database. The v2 test or Fisher’s exact test was conducted to identify possible risk factors associated with the occurrence of major complications and a hospitalization length greater than 7 days. For statistical analyses, hemobilia was defined as the presence of either radiologic evidence on pre-TACE unenhanced CT images, or bloody drainage following biliary intervention. Subsequently, the variables for which p value was less than 0.10 were subjected to multivariate logistic regression analysis using the backward method. Median, 1-, 3-, and 5-year survival rates were estimated by the Kaplan–Meier method. To determine possible prognostic factors, we utilized the Kaplan–Meier method with the log-rank test. Parameters for which p values were less than 0.10 were then entered into the multivariate Cox proportional hazard model (backward method) to evaluate for associations with survival time. The proportional hazard assumption and goodness-of-fit were verified using logminus-log plot and Cox-Snell residuals. A p value of less than 0.05 was determined to indicate statistical significance. All statistical analyses were performed using software (IBM SPSS Statistics, version 19.0; SPSS Inc., IBM Company, Armonk, NY).
Results Forty-five out of 53 patients (84.9 %) experienced initial signs/symptoms which prompted each patient to seek medical care and the subsequent diagnosis: jaundice (n = 25), upper abdominal pain (n = 16), and fever (n = 4). In the remaining eight asymptomatic patients (15.1 %, 8/53), HCC was detected for seven hepatitis B virus carrier patients during routine surveillance and incidentally found in the last patient. In terms of BCLC staging, 4, 47, and 2 patients were classified as intermediate, advanced, and terminal stage, respectively. The 47 patients
123
were in advanced stage due to vascular invasion (n = 17), extrahepatic metastasis (n = 3), vascular invasion with extrahepatic metastasis (n = 3), or poor performance status (n = 24). As for Child-Pugh class, 20, 31, and 2 patients belonged to class A, B, and C, respectively. In patients with Child-Pugh class C, clinicians carefully selected the candidates for TACE by assuming their hepatic functional reserve, based on laboratory findings to the exclusion of bilirubin level and imaging features suggesting liver cirrhosis. The index tumor size ranged from 1.0 to 12.7 cm (mean ± SD, 5.4 ± 2.6 cm). Prior to TACE, 17 and 8 patients underwent percutaneous biliary drainage and endoscopy-guided biliary drainage, respectively. At the time of TACE, the mean ± SD values of total bilirubin were 3.7 ± 5.01 mg/ dL among patients without biliary intervention, and 10.0 ± 6.3 mg/dL among patients who have undergone biliary intervention (p \ .001). Among the 28 patients who had not undergone pre-TACE biliary drainage, 8 patients showed hyperbilirubinemia (range of total bilirubin, 3.4–21.6 mg/dL) at the time of TACE. However, they were managed by TACE without prior biliary drainage because they did not have acute cholangitis. After the TACE, obstructive jaundice in 3 out of the 8 patients gradually resolved without any biliary interventions. Detailed information about our study population is summarized in Table 1. Safety: Complications and Hospitalization Periods After TACE, the patients stayed in the hospital from 1 to 52 days (median, 7 days; interquartile range, 3–15 days). Twenty-two patients (41.5 %) had been hospitalized for more than 7 days either for biliary intervention after TACE (n = 15) or for TACE-related complications (n = 7). With regard to TACE-related major complications, 3 patients (5.7 %, 3/53) experienced severe post-embolization syndrome requiring hospitalization longer than 7 days, and 3 others (5.7 %, 3/53) experienced non-fatal sepsis (gram-negative bacteria from the gastrointestinal tract) in spite of prophylactic antibiotics. Among the latter group of patients with non-fatal sepsis, 2 patients had received endoscopy-guided biliary drainage prior to TACE, and the remaining patient had not received any biliary intervention beforehand. They were recovered without any permanent adverse sequelae, after dedicated antibiotic therapy according to the results of blood culture. Secondary bacterial peritonitis occurred in one patient (1.9 %, 1/53) within a week after TACE, and this patient required 22 days of hospitalization. None of the patients died within 30 days from the respective dates of TACE procedure. In addition, all of the patients were discharged without permanent adverse sequelae secondary to the TACE.
J. W. Choi et al.: TACE for HCCs with BD Invasion Table 1 Characteristics of 53 patients analyzed in our study
Table 1 continued
Item
Item
Number (%)
Age
Item
Number (%)
Primary tumor size
C55 years
26 (49.1)
\55 years
27 (50.9)
Sex
C5.0 cm
26 (49.1)
\5.0 cm
27 (50.9)
Tumor multifocality
Male
39 (73.6)
Unifocal
27 (50.9)
Female
14 (26.4)
Multifocal
26 (49.1)
Portal hypertension Absent Present
34 (64.2) 19 (35.8)
Ascites
Vascular invasion Absent Present
33 (62.3) 20 (37.7)
Tumor enhancement
Absent
40 (75.5)
Present
13 (24.5)
Albumin
Strong
21 (39.6)
Intermediate
32 (60.4)
Extent of bile duct invasion
C3.5 g/dL
29 (54.7)
\3.5 g/dL
24 (45.3)
Total bilirubin
RHD/LHD
18 (34.0)
CHD/CBD
35 (66.0)
Central bile duct invasion
C3.0 mg/dL
28 (52.8)
\3.0 mg/dL
25 (47.2)
PT-INR
Direct
32 (60.4)
From periphery
21 (39.6)
Extrahepatic metastasis
C1.20
13 (24.5)
Absent
47 (88.7)
\1.20
40 (75.5)
Present
6 (11.3)
Aspartate aminotransferase
Hemobilia
C80 IU/L
31 (58.5)
Absent
39 (73.6)
\80 IU/L
22 (41.5)
Present
14 (26.4)
Alanine aminotransferase
Iodized oil uptake after TACE
C60 IU/L
27 (50.9)
Compact
27 (50.9)
\60 IU/L
26 (49.1)
Faint
26 (49.1)
Alkaline phosphatase C200 IU/L
Performance status 22 (41.5)
ECOG 0
\200 IU/L
31 (58.5)
ECOG 1–2
Alpha-fetoprotein
8 (15.1)
Pre-TACE biliary drainage
C1,000 ng/mL
30 (56.6)
None
28 (52.8)
23 (43.4)
Percutaneous
17 (32.1)
Endoscopy
Present
8 (15.1)
BCLC stage 49 (92.5) 4 (7.5)
Etiology of HCC
Intermediate
Item
Number (%)
HBV
40 (75.5)
C
2 (3.8)
HCV
1 (1.9)
RHD/LHD right hepatic duct or left hepatic duct, CHD/CBD common hepatic duct or common bile duct, ECOG Eastern Cooperative Oncology Group
According to the results of univariate analyses, serum aspartate aminotransferase C80 IU/L (p = .033) and ascites (p = .053) were subjected to multivariate analysis. Other variables, such as total bilirubin C3.0 mg/dL (p = .426), pre-TACE biliary drainage (p = .168), hemobilia (p = .364), and performance status (p = 1.000) were unlikely to be related to major complications. In this analysis, there was no significant risk factor for major TACErelated complications. With respect to hospitalization length, serum total bilirubin C3.0 mg/dL (p = .023), ascites (p = .070), and serum alanine aminotransferase C60 IU/L (p = .087) were applied to multivariate analysis. Variables such as vascular invasion (p = .336), portal hypertension (p = .217), and bile duct invasion extent (p = .255) were unlikely to be associated with long hospitalization. From the results of the multivariate analysis, pre-TACE total bilirubin C3.0 mg/ dL was the only significant risk factor for hospitalization [7 days [hazard ratio (HR) = 4.341, 95 % confidence interval 1.236–15.247, p = .022]. Among the patients who underwent biliary drainage before TACE (n = 25), there were no significant differences in the occurrence of major complications (p = .283) and hospitalization length (p = .311), regardless of biliary intervention performed (percutaneous drainage, endoscopy-guided drainage). Effectiveness: Survival and Prognostic Factors
45 (84.9)
\1,000 ng/mL Leukocytosis or leukopenia Absent
Number (%)
4 (7.5)
Advanced
47 (88.7)
Terminal
2 (3.8)
Child-Pugh class
Alcohol
3 (5.7)
A
20 (37.7)
Cryptogenic
9 (17.0)
B
31 (58.5)
The median survival of patients who received TACE for unresectable HCCs with central bile duct invasion was 12.2 months (Fig. 2). The 1-, 3-, and 5-year survival rates were 50.9, 20.9, and 20.9 %, respectively. With regard to the BCLC stage, 88.7 % (n = 47) of our study population belonged to advanced stage, and the median survival period for these patients was 10.8 months. In terms of Child-Pugh classification, 96.2 % (n = 51) of the patients were either in class A (37.5 %, n = 20) or in class B (58.5 %, n = 31), and their median survivals were 12.6 and 10.5 months, respectively (p = .707) (Fig. E1). From the results of univariate analyses, vascular invasion (p = .002), tumor multifocality (p = .003), extrahepatic metastasis (p = .010), tumor
123
J. W. Choi et al.: TACE for HCCs with BD Invasion Table 2 Univariate analyses to determine potential prognostic factors of TACE for unresectable HCCs with central bile duct invasion Item
p value
Age C55 years
.795
.888
Present .220
.302
123
RHD/LHD
.414
Central bile duct invasion .711
\3.0 mg/dL
Direct
.107
From periphery Extraheaptic metastasis .057à
\1.20
Absent
.010à
Present
Aspartate aminotransferase C80 IU/L .196 \80 IU/L
Hemobilia Absent
.995
Present
Alanine aminotransferase .976
\60 IU/L
Performance status ECOG 0
.666
ECOG 1–2
Alkaline phosphatase C200 IU/L
.044à
CHD/CBD
PT-INR
60 IU/L
Strong Intermediate
Total bilirubin
C1.20
.002à
Extent of bile duct invasion
\3.5 g/dL
enhancement (p = .044), and PT-INR C 1.20 (p = .057) were entered into the multivariate analysis (Table 2). Other variables such as portal hypertension (p = .888), bile duct invasion extent (p = .414), total bilirubin C3.0 mg/dL (p = .711), and performance status (p = .666) were unlikely to be related to the patients’ prognosis. From the results of multivariate analysis, extrahepatic metastasis (HR = 6.145, p \ .001), PT-INR C1.20 (HR = 4.564, p \ .001), vascular invasion (HR = 3.484, p = .001), and intermediate tumor enhancement (HR = 2.417, p = .019) were significantly associated to shorter survival after TACE (Table 3). The median survival of the patients (n = 12) who did not have any of these risk factors was 78.9 months in our study (Fig. E2). For these patients, the index tumor size and total bilirubin level were 3.9 ± 1.2 cm and 5.7 ± 6.84 mg/ dL, respectively. Five out of the 12 patients underwent surgical resection during their follow-up period. In contrast, the median survival of the remaining 41 patients was 10.5 months. Among the patients who underwent biliary drainage before TACE (n = 25), there were no significant differences in survival (p = .816) regardless of the type of biliary intervention performed (percutaneous drainage, endoscopy-guided drainage). Twenty-seven out of the total 53 patients (50.9 %) demonstrated compact iodized oil uptake on post-TACE CT images, and their survival (median, 23.1 months) was significantly longer than that of patients with partial iodized oil uptake (median, 7.8 months) (p \ .001) (Fig. 3). Regarding the pre-TACE CT findings of the 27
Vascular invasion Absent Present
Present
C3.0 mg/dL
.003à
Tumor enhancement
Albumin
Fig. 2 Kaplan–Meier curve of overall cumulative survival rate after TACE (median survival, 12.2 months)
Unifocal Multifocal
Ascites
C3.5 g/dL
.585
Tumor multifocality .160
Female
Absent
C5.0 cm \5.0 cm
Sex
Portal hypertension Absent
p value
Primary tumor size
\55 years Male
Item
Leukocytosis or leukopenia .336
\200 IU/L
Absent
.957
Present
Alpha-fetoprotein C1,000 ng/mL
.430
\1,000 ng/mL RHD/LHD right hepatic duct or left hepatic duct, CHD/CBD common hepatic duct or common bile duct, ECOG Eastern Cooperative Oncology Group à Factors were included in multivariate analysis, if p value \.10 in univariate analyses
tumors, 12 and 15 HCCs demonstrated intermediate and strong enhancement, respectively (Fig. 1). A patient with compact iodized oil uptake achieved complete remission status after the third session of TACE. Additionally, 9 patients (17.0 %, 9/53), who had viable intraductal tumors after 1–5 sessions of TACE and good hepatic functional reserve, subsequently underwent surgery (hemihepatectomy in 6 patients, extended hemihepatectomy in 2, liver transplantation in 1). For the group of
J. W. Choi et al.: TACE for HCCs with BD Invasion Table 3 Multivariate analysis to determine significant prognostic factors of TACE for unresectable HCCs with central bile duct invasion Variable
Hazard ratio
95 % CI
p value
Extrahepatic metastasis
6.145
2.292–16.474 \.001 1.968–10.584 \.001
PT-INR C1.20
4.564
Vascular invasion
3.484
1.670–7.268
.001
Intermediate tumor enhancement
2.417
1.155–5.059
.019
Multifocal tumor
NS
NS not significant
patients who underwent surgery, the index tumor size and total bilirubin level before the initial TACE were 5.6 ± 1.9 cm and 11.2 ± 8.21 mg/dL, respectively. Four patients had been planned to undergo surgical resection after preoperative biliary intervention and a single session TACE. The remaining five patients decided to undergo surgery after several TACE sessions due to prolonged jaundice, in spite of their own well-controlled tumors. The median survival period for the nine patients who underwent surgical resection after TACE was 62.8 months (Fig. E3).
Discussion How to manage unresectable HCCs with central bile duct invasion remains undefined due to the lack of evidence. Therefore, this rare disease has been managed by locoregional therapy, sorafenib, or best supportive care, all dependent on local expertise. Through this study, we demonstrated acceptable complication rates (13.2 %) even after taking account of complications with uncertain causal relationship to TACE—for example, non-fatal sepsis (n = 3) and secondary bacterial peritonitis (n = 1) [23]. Furthermore, based on the earlier studies [15, 24], a considerable number of our patients had risk factors for complications such as hyperbilirubinemia or vascular invasion. These results suggest that TACE can be safely conducted for HCCs with central bile duct invasion, with selective chemoembolic agent infusion into the tumor-feeding arteries. Total bilirubin [3.0 mg/dl was a predictor of prolonged hospital stay and, not surprisingly, the main factor prolonging hospital stay in this patient group was being treated for hyperbilirubinemia. However, the patients who have undergone pre-TACE biliary drainage, including those with total bilirubin C3.0 mg/dL, did not require significantly longer hospitalization. This is because the control of cholangitis was considered more important in determining a patient’s discharge, rather than the total bilirubin level itself. These results imply that even though HCCs with
Fig. 3 Kaplan–Meier curves of cumulative survival rates after TACE by iodized oil uptake on post-TACE CT imaging
central bile duct invasion induce substantial obstructive jaundice, TACE can be performed along with appropriate biliary intervention. Concordantly, Lai et al. [4] claimed that mean survival period was remarkably longer for patients treated with biliary drainage and subsequent TACE (8.0–13.4 months) than that of patients who have received biliary drainage alone (2.5–4.5 months). These results imply that obstructive hyperbilirubinemia in patients of HCCs with central bile duct invasion is not a contraindication for TACE, and that pre-TACE biliary intervention should be considered. Our study also presented acceptable survival outcome of TACE (median survival, 12.2 months). In particular, the 12 patients without poor prognostic factors (extrahepatic metastasis, PT-INR C1.20, vascular invasion, and intermediate tumor enhancement) had 78.9 months of median survival period in our study. These results suggest that TACE can be considered as a safe and effective treatment option for HCCs with central bile duct invasion. Presently, TACE may be the only promising locoregional therapy for HCCs with central bile duct invasion, because radiofrequency ablation to these tumors is able to cause lifethreatening injury to hilar structures. In addition, irreversible electroporation is still in an evolving state. External beam radiation therapy and radioembolization may be alternatives of TACE. However, there is little or no knowledge, until now, about the efficacy and safety of these methods for HCCs with bile duct invasion.
123
J. W. Choi et al.: TACE for HCCs with BD Invasion
In terms of the prognosis factors, the PT-INR level is more useful than the total bilirubin level or performance status in these patients. Given that it reflects biosynthetic capacity of the liver [25], the PT-INR level could be a more reliable marker to assess hepatic reserve—more than the total bilirubin level or performance status in the setting of central bile duct involvement. Concordantly, PT-INR levels demonstrated a significant association to survival outcomes in our study. In addition, tumor enhancement degree on pre-TACE CT images was also a significant prognostic factor in our study. Relatively, hypovascular HCCs with intermediate enhancement on pre-TACE CT may represent poor differentiation or cholangiocarcinoma components. Hence, these tumors can be rather intractable for TACE [26–28]. On the other hand, compact iodized oil uptake after TACE, associated with vascularity and biological background of the HCC tumors, indicated better prognosis. This is consistent with previous studies which reported the importance of iodized oil uptake after TACE as a useful prognostic factor [16, 29–32]. In general, the BCLC staging system is very useful to build treatment plans for most HCC patients because this algorithm takes into account patient factors such as liver function and performance status. Additionally, the algorithm considers tumor factors such as size, number, and vascular invasion [14]. This system is largely based on the premise that patient factors reflect gradual deterioration of hepatic reserve as the HCC progresses in addition to underlying liver disease. In case this assumption is broken, the BCLC algorithm then needs to be used flexibly depending on each condition. In our study, obstructive jaundice from HCCs with central bile duct invasion appeared to have substantially over-inflated the BCLC stages and Child-Pugh classes. Specifically, 92.5 % of our patients were in advanced stage (88.7 %) or terminal stage (3.8 %) according to the BCLC staging system, and 62.3 % were in class B (58.5 %) or C (3.8 %) depending on the Child-Pugh classification. In addition, total bilirubin level and performance status did not significantly affect prognosis in our patients. These suggest that hyperbilirubinemia and compromised performance status, in cases with biliary tumor invasion, should be regarded differently from those caused by hepatic parenchymal dysfunction. There are some limitations in our study. First, the number of patients analyzed in our study was limited (n = 53). The second limitation of this study is that we did not address the efficacy of surgical intervention and the optimal timing/method of biliary intervention. Due to the retrospective nature of our study, treatment strategies varied among the patients. In conclusion, TACE can be a safe and effective treatment for patients who have unresectable HCCs with central bile duct invasion In particular, long-term survival can be
123
expected if patients have strongly enhancing tumors without poor prognostic factors such as extrahepatic metastasis, PT-INR prolongation, and vascular invasion. Conflict of interest Jin Woo Choi: no conflict of interest; Jin Wook Chung: no conflict of interest; Yun Ku Cho: no conflict of interest; Yoon Jun Kim: no conflict of interest; Jung-Hwan Yoon: no conflict of interest; Hyo-Cheol Kim: no conflict of interest; Hwan Jun Jae: no conflict of interest.
References 1. Lin TY, Chen KM, Chen YR, Lin WS, Wang TH, Sung JL (1975) Icteric type hepatoma. Med Chir Dig 4:267–270 2. Kojiro M, Kawabata K, Kawano Y, Shirai F, Takemoto N, Nakashima T (1982) Hepatocellular carcinoma presenting as intrabile duct tumor growth: a clinicopathologic study of 24 cases. Cancer 49:2144–2147 3. Lau WY, Leow CK, Leung KL, Leung TW, Chan M, Yu SC (2000) Cholangiographic features in the diagnosis and management of obstructive icteric type hepatocellular carcinoma. HPB Surg 11:299–306 4. Lai EC, Lau WY (2006) Hepatocellular carcinoma presenting with obstructive jaundice. ANZ J Surg 76:631–636 5. Huang JF, Wang LY, Lin ZY et al (2002) Incidence and clinical outcome of icteric type hepatocellular carcinoma. J Gastroenterol Hepatol 17:190–195 6. Shiomi M, Kamiya J, Nagino M et al (2001) Hepatocellular carcinoma with biliary tumor thrombi: aggressive operative approach after appropriate preoperative management. Surgery 129:692–698 7. Yeh CN, Jan YY, Lee WC, Chen MF (2004) Hepatic resection for hepatocellular carcinoma with obstructive jaundice due to biliary tumor thrombi. World J Surg 28:471–475 8. Gabata T, Terayama N, Kobayashi S, Sanada J, Kadoya M, Matsui O (2007) MR imaging of hepatocellular carcinomas with biliary tumor thrombi. Abdom Imaging 32:470–474 9. Satoh S, Ikai I, Honda G et al (2000) Clinicopathologic evaluation of hepatocellular carcinoma with bile duct thrombi. Surgery 128:779–783 10. Moon DB, Hwang S, Wang HJ et al (2013) Surgical outcomes of hepatocellular carcinoma with bile duct tumor thrombus: a Korean multicenter study. World J Surg 37:443–451 11. Kitagawa K, Yamakado K, Nakatsuka A et al (1999) Selective transcatheter hepatic arterial chemoembolization for hemobilia from hepatocellular carcinoma: report of three cases. J Vasc Interv Radiol 10:1357–1360 12. Okuda M, Miyayama S, Yamashiro M et al (2010) Sloughing of intraductal tumor thrombus of hepatocellular carcinoma after transcatheter arterial chemoembolization. Cardiovasc Interv Radiol 33:619–623 13. Choi J, Ryu JK, Lee SH et al (2013) Palliative treatment of unresectable hepatocellular carcinoma with obstructive jaundice using biliary drainage with subsequent transarterial chemoembolization. J Palliat Med 16:1026–1033 14. Llovet JM, Bru C, Bruix J (1999) Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 19:329–338 15. Chung JW, Park JH, Han JK et al (1996) Hepatic tumors: predisposing factors for complications of transcatheter oily chemoembolization. Radiology 198:33–40
J. W. Choi et al.: TACE for HCCs with BD Invasion 16. Mondazzi L, Bottelli R, Brambilla G et al (1994) Transarterial oily chemoembolization for the treatment of hepatocellular carcinoma: a multivariate analysis of prognostic factors. Hepatology 19:1115–1123 17. Pinter M, Hucke F, Graziadei I et al (2012) Advanced-stage hepatocellular carcinoma: transarterial chemoembolization versus sorafenib. Radiology 263:590–599 18. Kothary N, Weintraub JL, Susman J, Rundback JH (2007) Transarterial chemoembolization for primary hepatocellular carcinoma in patients at high risk. J Vasc Interv Radiol 18:1517–1526 19. Kim HC, Lee JH, Chung JW et al (2013) Transarterial chemoembolization with additional cisplatin infusion for hepatocellular carcinoma invading the hepatic vein. J Vasc Interv Radiol 24:274–283 20. Bruix J, Sherman M, American Association for the Study of Liver Disease (2011) Management of hepatocellular carcinoma: an update. Hepatology 53:1020–1022 21. Jung AY, Lee JM, Choi SH et al (2006) CT features of an intraductal polypoid mass: differentiation between hepatocellular carcinoma with bile duct tumor invasion and intraductal papillary cholangiocarcinoma. J Comput Assist Tomogr 30:173–181 22. Sacks D, McClenny TE, Cardella JF, Lewis CA (2003) Society of Interventional Radiology Clinical Practice guidelines. J Vasc Interv Radiol 14:S199–S202 23. Brown DB, Nikolic B, Covey AM et al (2012) Quality improvement guidelines for transhepatic arterial chemoembolization, embolization, and chemotherapeutic infusion for hepatic malignancy. J Vasc Interv Radiol 23:287–294 24. Gates J, Hartnell GG, Stuart KE, Clouse ME (1999) Chemoembolization of hepatic neoplasms: safety, complications, and when to worry. Radiographics 19:399–414
25. Thapa BR, Walia A (2007) Liver function tests and their interpretation. Indian J Pediatr 74:663–671 26. Sahani DV, Holalkere NS, Mueller PR, Zhu AX (2007) Advanced hepatocellular carcinoma: CT perfusion of liver and tumor tissue—initial experience. Radiology 243:736–743 27. Lee JH, Lee JM, Kim SJ et al (2012) Enhancement patterns of hepatocellular carcinomas on multiphasicmultidetector row CT: comparison with pathological differentiation. Br J Radiol 85:e573–e583 28. Choi BI, Han JK, Kim YI et al (1994) Combined hepatocellular and cholangiocarcinoma of the liver: sonography, CT, angiography, and iodized-oil CT with pathologic correlation. Abdom Imaging 19:43–46 29. Takayasu K, Muramatsu Y, Maeda T et al (2001) Targeted transarterial oily chemoembolization for small foci of hepatocellular carcinoma using a unified helical CT and angiography system: analysis of factors affecting local recurrence and survival rates. Am J Roentgenol 176:681–688 30. Vogl TJ, Trapp M, Schroeder H et al (2000) Transarterial chemoembolization for hepatocellular carcinoma: volumetric and morphologic CT criteria for assessment of prognosis and therapeutic success-results from a liver transplantation center. Radiology 214:349–357 31. Kawaguchi T, Ohkawa K, Imanaka K et al (2012) Lipiodol accumulation and transarterial chemoembolization efficacy for HCC patients. Hepatogastroenterology 59:219–223 32. Kim DY, Ryu HJ, Choi JY et al (2012) Radiological response predicts survival following transarterial chemoembolisation in patients with unresectable hepatocellular carcinoma. Aliment Pharmacol Ther 35:1343–1350
123
