ORIGINAL ARTICLE. Abstract. Introduction

Transcription

1 ORIGINAL ARTICLE Prognosis of Patients with Hepatocellular Carcinoma Treated Solely with Transcatheter Arterial Chemoembolization: Risk Factors for One-year Recurrence and Two-year Mortality (Preliminary Data) Michitaka Matsuda 1,FumioOmata 1, Sokun Fuwa 2, Yukihisa Saida 2, Shoko Suzuki 1, Masayo Uemura 1, Naoki Ishii 1, Yusuke Iizuka 1, Katsuyuki Fukuda 1 and Yoshiyuki Fujita 1 Abstract Objective Transcatheter arterial chemoembolization (TACE) is an essential therapy for patients with hepatocellular carcinoma (HCC) in whom administering other treatments such as liver transplantation, resection or local therapy is not feasible. The purpose of our study was to determine the independent risk factors for oneyear recurrence and two-year mortality in patients treated solely with TACE. Methods We conducted a retrospective cohort study of 34 consecutive patients (Group 1) with incident HCC who were treated solely with epirubicin-based TACE between April 2004 and March A subgroup analysis was performed among 24 patients (Group 2) who underwent complete TACE confirmed with abdominal computed tomography (CT) one month later. Tumor recurrence was evaluated using contrast CT every three months after the initial TACE. We calculated Kaplan-Meier estimates and performed a multiple regression analysis using a Cox-proportional hazard model. Results The patients in Group 1 (men, 59%), all of whom had liver cirrhosis, underwent TACE as the sole therapy for HCC. Kaplan-Meier estimates revealed a two-year survival rate [95% CI] of 70% [48-84%]. For the non-child A patients, the adjusted hazard ratio (HR) [95% CI] for two-year survival was 7.1 [ ]. In Group 2, the Kaplan-Meier estimate of the one-year recurrence rate [95% CI] was 61% [42-81%]. The adjusted HRs [95% CIs] for one-year recurrence for age and indocyanine green (ICG) 15-min >30% were 1.1 [ ] and 7.87 [ ], respectively. Conclusion Non-Child A cirrhosis is an independent risk factor for two-year mortality in patients treated solely with TACE. For ICG 15-min >30%, careful monitoring for HCC recurrence at one year, even after complete TACE, is warranted. Key words: hepatocellular carcinoma, transcatheter arterial chemoembolization, prognosis, mortality, recurrence (Intern Med 52: , 2013) () Introduction Hepatocellular carcinoma (HCC) is the sixth most common cancer and the third most common cause of cancer mortality worldwide (1). There were an estimated 700,000 deaths from HCC in 2008 worldwide, with an age-adjusted mortality rate of 10 per 100,000 inhabitants (1). In Japan, the prevalence of HCC is declining (2) but remains substantial, with an age-adjusted mortality rate of 26.7 per 100,000 observed in 2006 (3). The prognosis of HCC remains unsatisfactory (4), primarily because administering curative treatments such as surgical resection, radiofrequency ablation (RFA) and liver transplantation is feasible only in a relatively small percentage of patients and because most patients with HCC have concomitant liver cirrhosis. Department of Internal Medicine, St. Luke s International Hospital, Japan and Department of Radiology, St. Luke s International Hospital, Japan Received for publication November 17, 2012; Accepted for publication December 25, 2012 Correspondence to Dr. Michitaka Matsuda, michimtd@me.com 847

2 Figure 1. Study flow diagram. A total of 269 patients were treated with epirubicin-based transarterial chemoembolization (TACE) for unresectable hepatocellular carcinoma (HCC) between April 2004 and March Two hundred and twenty-six patients with a history of treatment for HCC and nine patients treated with combined TACE-radiofrequency ablation were excluded. The final 34 patients were treated with TACE only (Group 1). In 10 patients, TACE was classified as incomplete per the findings of CT or MRI obtained one month after the initial therapy. For the remaining 24 patients, TACE was categorized as complete (Group 2). Transcatheter arterial chemoembolization (TACE) is an essential therapy for improving the prognoses of patients with HCC in whom performing liver transplantation, surgical resection and percutaneous ablation is not feasible (5). It is therefore important to determine the prognosis, as well as risk factors for recurrence and mortality, among HCC patients treated solely with TACE. TACE has been reported to be less effective than other local therapies such as RFA (6), with many patients requiring retreatment for recurrence. In addition, local HCC recurrence after TACE remains a critical problem for patients. However, few studies have investigated the local HCC recurrence rates after TACE. This data gap may stem from protocols in the United States and European countries, which commonly add TACE as intermittent, palliative therapy (for example, every three months) without confirming HCC recurrence. Japanese protocols, in contrast, typically repeat TACE only when surveillance CT or MRI reveal local disease recurrence. To repeat TACE on demand, it is useful to determine the recurrence rate after TACE in order to estimate the appropriate follow-up period for cross-sectional imaging studies. In order to quantify mortality and recurrence in HCC patients treated solely with TACE, we first examined the twoyear survival rate of patients with incident HCC treated with on-demand TACE. Additionally, we investigated the HCC recurrence rate in a population whose success of initial TACE was confirmed on CT or MRI one month after the procedure. Materials and Methods A total of 34 consecutive patients with unresectable HCC were treated solely with epirubicin-based TACE at a tertiary referral center in Tokyo, Japan between April 2004 and March First, we conducted a retrospective cohort study of these 34 patients (Group 1) focusing on two-year survival. Twenty patients with Barcelona Clinic Liver Cancer (BCLC) (7) stage A HCC were included due to patients being unsuitable for local therapy or surgical resection and because performing liver transplantation was not feasible. Second, we performed a subgroup analysis of disease recurrence among 24 patients (Group 2) who underwent complete TACE. We defined complete TACE as a lack of residual lesions and recurrence on multiphasic CT scans performed one month after TACE. All patients with a history of prior treatment for HCC, local therapy added in the follow-up period, hepatic vein or portal vein tumor invasion or extrahepatic metastasis to lymph nodes and/or other organs were excluded from this study (Fig. 1). The etiology of cirrhosis, the platelet count, the levels of serum albumin and serum alpha-fetoprotein (AFP), multiphasic CT results (tumor size, tumor number, lymph node metastasis), one-year recurrence and the followup time to two-year survival were investigated retrospectively. Diagnosis of HCC The diagnosis of HCC was made based on ultrasonography, multiphasic CT, MRI with or without abdominal angiography and/or peripheral biopsies in accordance with the 2005 American Association for the Study of Liver Diseases practice guidelines (8). Abnormal elevations in the levels of tumor markers such as serum AFP and protein induced by vitamin K absence or antagonist-2 (PIVKA-2) were also taken into account. Typical HCC exhibited arterial enhancement and portal venous washout on multiphasic CT. The lack of extrahepatic metastasis was confirmed using ultrasonography, CT and chest X-ray. TACE A 2.2/2.9-Fr catheter tip was advanced as far as possible into the feeding artery of the HCC. 5-8 ml of emulsion consisting of 50 mg of epirubicin and 3-6 ml of lipiodol followed by gelatin sponge particles (Spongel ) dissolved in saline was injected under fluoroscopy. The doses of the emulsion and embolic materials were determined based on the tumor size. The disappearance of the arterial flow was confirmed after embolization using digital subtraction angiography before the end of the procedure. The accumulation of lipiodol in the tumor region was confirmed on CT a few days after the procedure. The completeness of TACE and the presence of tumor recurrence were confirmed on multiphasic CT or MRI one month after TACE. Subsequently, tumor recurrence was assessed using multiphasic CT or MRI every three to four months. TACE was repeated when local recurrence, intrahepatic metastasis or a second primary HCC was found. 848

3 Table 1. Basic Characteristics (Group 1 and Group 2) Group 1 (n=34) Group 2 (n=24) Age, mean (SD) 70 (7.1) 69 (6.8) Gender Male, n (%) 20 (59) 15 (62) Etiology Alcohol, n (%) 3 (9) 2 (8) Hepatitis C, n (%) 25 (73) 17 (70) Hepatitis B, n (%) 4 (12) 4 (16) Non-B, Non-C, n (%) 3 (9) 3 (12) Platlet count / L, mean (SD) ( ) ( ) Serum albumin g/dl, mean (SD) 3.5 (0.5) 3.5 (0.5) Number of lesions Single, n (%) 17 (50) 14 (58) 2-3, n (%) 12 (35) 9 (38) 4 <, n (%) 5 (15) 1 (4) Maximum tumor size in cm, median (range) 2.4 (0.5-19) 2.3 (0.5-19) Child A, n (%) 26 (76) 20 (83) Child B, n (%) 8 (24) 4 (17) TNM stage I (T1N0M0), n (%) 10 (30) 9 (38) II (T2N0M0), n (%) 11 (32) 6 (25) III (T3N0M0), n (%) 13 (38) 9 (38) IV (T4N0M0), n (%) 0 (0) 0 (0) BCLC stage BCLC-A, n (%) 22 (65) 19 (79) BCLC-B, n (%) 12 (35) 5 (21) AFP ng/ml, mean (SD), 857 (2,499) 930 (2,618) AFP > 400ng/mL, n (%) 8 (24) 5 (21) AFP < or = 400ng/mL, n (%) 26 (76) 19 (79) ICG-15min %, mean (SD) 32.5 (20.5) 30.7 (20.6) SD: standard deviation, TNM: Tumor-Node-Metastasis staging system, BCLC: Bacelona Clinic Liver Cancer, AFP: Alpha-fetoproein, ICG-15min: indocyanine green retention rate at 15 minutes Bivariate analyses of continuous variables were performed using Student s t-test or the Wilcoxon rank-sum test depending on the distribution. Fisher s exact test was used for bivariate analyses of dichotomous variables. We calculated Kaplan-Meier estimates of mortality and tumor recurrence and performed a multiple regression analysis using a Coxproportional hazard model. All p values were two-sided, and a p value of <.05 was considered to be statistically significant. All statistical analyses were performed using the JMP version 8 software package. Results Figure 2. Two-year survival curve in Group 1. The solid line represents the Kaplan-Meier estimates of two-year survival. The dotted line represents the 95% confidence interval. The two-year survival rate [95% CI] was 70% [48-84%]. Statistical analysis Table 1 shows the basic characteristics of the patients stratified by group. The median follow-up duration [range] in Group 1 and Group 2 was 585 [19-1,867] and 726 [19-1,624] days, respectively. Group 1 consisted of 20 men (59%) and 14 women (41%) with a mean age (standard deviation (SD)) of 70 (7.1) years. Twenty-six patients (76%) were classified as having Child A disease and eight (24%) patients were classified as having Child B disease. The most common etiology was hepatitis C virus infection, present in 25 cases (73%). Seventeen patients (50%) had solitary lesions, while 12 patients (35%) had two to three lesions and five patients (15%) had more than four lesions. Regarding the maximum tumor size, the median [range] maximum 849

4 Table 2. Bivariate and Multivariate Analysis for Two-year Suvival Crude Hazard Ratio 95% CI p value Adjusted Hazard Ratio 95% CI p value Age Male gender Maximum size of HCC in cm Single (vs. Multiple) T stage Platlet count < PT-INR > Total bilirubin > 2.0mg/ Albumin < 3.0g/ AFP > 400ng/m ICG-15min > 30% Child-Pugh score Non Child A Incomplete TACE HCC: hepatocellular carconoma, PT-INR: prothorombin time-international normalized ratio, AFP: Alpha-fetoproein, ICG-15min: indocyanine green retention rate at 15 minuntes Figure 3. One-year local recurrence curve in Group 2. The solid line represents the Kaplan-Meier estimates of one-year recurrence. The dotted line represents the 95% confidence interval. One-fourth of the patients who underwent complete TACE experienced local recurrence within three months. The one-year local recurrence rate [95% CI] was 61% [42-81]. HCC diameter was 2.4 [0.5-19] cm. In Group 1, eight patients died during the follow-up period. The causes of death were hepatic failure (three patients), uncontrolled HCC (two patients), variceal rupture (one patient), sepsis (one patient) and cerebral hemorrhage (one patient). Group 2 consisted of 15 men (62%) and nine women (38%) with a mean age (SD) of 69 (6.8) years. Twenty patients (83%) were classified as having Child A disease and four patients (17%) were classified as having Child B disease. Seventeen patients (70%) were infected with hepatitis C, the most common etiology. Fourteen patients (58%) had solitary lesions, nine patients (38%) had two to three lesions and one patient (4%) had more than four lesions. The median [range] maximum HCC diameter was 2.6 [0.5-7] cm. According to the tumor-node-metastasis (TNM) staging revised by the Liver Cancer Study Group of Japan in 2000 (9), which differs from the TNM staging system of the American Joint Committee on Cancer (AJCC)/International Union Against Cancer (UICC) (10), the T category consists of three factors: the presence of a single lesion, a tumor size 2 cm in diameter and a lack of vascular or bile duct invasion. The tumor stage between I and IV-A is determined solely by the T factor when both the N and M factors are zero. In Group 1, 10 patients (30%), 11 patients (32%) and 13 patients (38%) were categorized as having stage I, II and III disease, respectively. In Group 2, nine patients (38%), six patients (25%) and nine patients (38%) were categorized as having stage I, II and III disease, respectively. Per the exclusion criteria, there were no Stage IV-A or IV-B patients in either group. According to the BCLC classification, Group 1 contained 20 patients (65%) with early stage (stage 0-A) disease and 12 patients (35%) with intermediate stage (stage B) disease, while Group 2 contained 19 patients (79%) with early stage disease and five patients (21%) with intermediate stage disease. The Kaplan-Meier estimates demonstrated a two-year survival rate [95% CI] in Group 1 of 70% [48-84%] (Fig. 2). The crude hazard ratios (HRs) [95%CIs] associated with larger maximum HCC size, prothrombin time (PT-INR) >1, total bilirubin >2 mg/dl, serum albumin <3 g/dl, serum AFP >400 ng/ml, non-child A cirrhosis and incomplete TACE were 1.2 [1-1.47], 2.9 [ ], 2.8 [ ], 3.5 [ ], 3 [ ], 4.2 [ ] and 2.68 [ ], respectively. The adjusted HRs [95% CIs] for older age and non-child A associated with two-year survival were 1.1 [ ] and 7.1 [1.1-52], respectively (Table 2). The Kaplan-Meier estimate of one-year recurrence [95% CI] in Group 2 was 61% [42-81%] (Fig. 3). The crude HRs [95%CIs] associated with older age, larger maximum HCC size, PT-INR >1 and indocyanine green (ICG) 15-min >30% were 1 [ ], 1.4 [ ], 1.9 [ ] and 4.6 [1.3-21], respectively. The adjusted HRs [95% CIs] for oneyear recurrence associated with age and ICG 15-min >30% 850

5 Table 3. Bivariate and Multivariate Analysis for One-year Recurrence Crude Hazard Ratio 95% CI p value Adjusted Hazard Ratio 95% CI p value Age Male gender Maximum size of HCC in cm Single (vs. multiple) T stage Platlet count < / L PT-INR > Total bilirubin > 2.0mg/dL Albumin < 3.0g/dL AFP > 400ng/dL ICG-15min > 30% Child-Pugh score Non-Child A HCC: hepatocellular carconoma, PT-INR: prothorombin time-international normalized ratio, AFP: Alpha-fetoproein, ICG-15min: indocyanine green retention rate at 15 minutes were 1.1 [ ] and 7.87 [ ], respectively (Table 3). Discussion Our study showed that the two-year survival rate of patients treated solely with TACE remains largely unsatisfactory. Interestingly, incomplete TACE therapy was not found to be associated with diminished two-year survival. Moreover, the rate of one-year recurrence of HCC was substantial, even after complete TACE. The risk factors for twoyear mortality included a non-child A status, while age and ICG 15-min >30% were found to be independent risk factors for one-year recurrence. Interestingly, tumor multiplicity was found to be associated with neither two-year mortality nor one-year recurrence, while the maximum HCC size was marginally associated with one-year recurrence. In this cohort, the major cause of death was related to an impaired hepatic function, compatible with the finding that a non-child A status is a significant risk factor for two-year mortality, while tumor control itself is not. When characterizing the prognosis of patients with liver cirrhosis and HCC, not only tumor extension, but also the residual liver function must be evaluated. The BCLC system is an established HCC staging system for selecting therapy and is especially utilized in Europe and the United States (11). However, it does not necessarily reflect the liver function. For example, TACE is recommended for patients with BCLC stage B disease, including both Child A and Child B patients. Heterogeneous patients with different prognoses are thus included in a single stage, per the BCLC system. Our results suggest that the prognosis should be estimated using a staging system that better reflects the liver function. Moreover, individual BCLC stages do not consistently correspond to specific therapies. Our study population included not only BCLC stage B patients, but also those with stage A disease in whom performing neither resection, local therapy nor liver transplantation was feasible. Local therapies such as RFA may be the most frequently used modalities but have inherent limitations depending on both tumor size and tumor location (namely, proximity to the diaphragm, Inferior vena cava (IVC) and main portal trunk) (12, 13). As such, TACE is indicated not only for BCLC stage B patients, but also for those with BCLC stage A disease. Although many reports provide consistent evidence that TACE can benefit patients with unresectable HCC and thus propose the use of TACE as standard therapy in this population, extant protocols lack uniformity, particularly regarding intervals of treatment. Llovet et al. (5) conducted a randomized controlled trial comparing embolization or TACE with symptomatic treatment in patients with unresectable HCC. They reported that TACE improved survival, with two-year survival rates for TACE and the controls of 63% versus 27%. The authors performed repeat TACE two months after the initial TACE and then every six months thereafter. Two meta-analyses (14, 15) also support the survival benefits of TACE. Most studies included in these metaanalyses also performed TACE repeatedly at scheduled intervals without conducting intermittent imaging surveillance. However, Ernst et al. (16) suggested that the efficacy of TACE is enhanced when repeated only as necessary based on follow-up imaging rather than at scheduled intervals. Such on-demand TACE may be a reasonable strategy to preserve the liver function, which was shown to be a critical factor for two-year survival in the present study. Takayasu et al. (17) reported that the two-year survival rate of patients treated with TACE is 63%. The authors repeated TACE only when local recurrence, intrahepatic metastases and/or second primary HCC were found on quarterly surveillance multiphasic CT or MRI, similar to our protocol. However, their study did not include the use of multiphasic CT one month after index TACE to confirm completion. Our study suggests that no less than 50% of patients exhibit local recurrence of HCC within one year, de- 851

6 spite undergoing complete TACE, and that providing close follow-up within three months is mandatory, as one-fourth of post-tace patients develop local recurrence within this period. Even if the benefits of closer follow-up are considered, our two-year survival rate was far better than those reported in the aforementioned studies, as our study population included more cases of early BCLC stage disease. In addition, in four cases, TACE was repeated within four months when the initial procedure was found to be incomplete. Another possible factor is the existence of variations in TACE technique between countries and/or radiologists. The participants in our study underwent embolization as selectively as possible after the catheter tip was advanced, i.e. segmental or subsegmental arterial embolization, which provides a tradeoff between achieving maximum effectiveness of HCC and minimum injury of the background liver. Takayasu et al. (17) suggested that the degree of liver damage (assessed based on the Liver Cancer Study Group of Japan criteria), a maximum tumor size >2.1 cm and the presence of tumor multiplicity are independent risk factors for eight-year mortality. In contrast, our study did not suggest that maximum HCC size or tumor multiplicity are independent risk factors for mortality at two years, possibly due to our much shorter follow-up period. We assume that the degree of liver damage is a more important risk factor, especially in earlier phases of the disease. Our study is the first report to suggest that an ICG 15- min >30% is a significant marker of one-year HCC recurrence after complete TACE. Wu et al. (18) reported that a higher ICG 15-min is a significant risk factor for late recurrence of hepatitis type B virus (HBV)-related HCC at two years after partial hepatectomy. The ICG 15-min has previously been used to evaluate the liver function and also provides a good estimate of the hepatic blood flow (19). An insufficient blood supply, as reflected by an elevated ICG 15- min, may preclude sufficient TACE, leading to higher recurrence rates. Kim et al. (20), in their retrospective observational study, suggested that liver transplantation was associated with better survival rates during a five-year follow-up in patients with unresectable HCC, beyond the Milan criteria (21). Liver transplantation can be more effective in these patients, as it not only removes the HCC itself, but also replaces a debilitated residual liver function, a critical factor determining the survival rate in several previous studies. Welldesigned randomized controlled trials comparing TACE with liver transplantation in BCLC stage B patients are needed in the future. Our study is associated with several limitations. First, this was a single-center study with a limited number of patients. Therefore, the statistical power was low. Second, there may be unexpected factors that influence the probability of survival and/or recurrence. Third, we cannot deny the possibility that occult portosystemic shunts influenced the values of ICG 15-min. Fourth, an external validation study is warranted before generalizing our results. In conclusion, TACE is not a satisfactory stand-alone treatment for patients with intermediate or early-stage HCC in terms of two-year survival. As expected, the residual liver function is a key factor for two-year survival. Additionally, providing close follow-up using cross-sectional imaging is warranted, even after complete TACE, especially in patients with an older age or an elevated ICG 15-min. The authors state that they have no Conflict of Interest (COI). Acknowledgement We thank Drs. Gautam A. Deshpande and Takeshi Kimura for providing comments and editing our manuscript. References 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN Int J Cancer 127: , Qiu D, Katanoda K, Marugame T, Sobue T. A Joinpoint regression analysis of long-term trends in cancer mortality in Japan ( ). Int J Cancer 124: , Umemura T, Ichijo T, Yoshizawa K, Tanaka E, Kiyosawa K. Epidemiology of hepatocellular carcinoma in Japan. J Gastroenterol 44 Suppl 19: , Parkin DM. Global cancer statistics in the year Lancet Oncol 2: , Llovet JM, Real MI, Montana X, et al. Arterial embolisation or chemoembolization versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomized controlled trial. Lancet 359: , Lencioni R. Loco-regional treatment of hepatocellular carcinoma. Hepatology 52: , Llovet JM, Bru C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 19: , Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology 42: , The Liver Cancer Study Group of Japan. The general rules for the clinical and pathological study of primary liver cancer. Version 4. Kanehara & Co., Ltd, Tokyo, Greene FL, Page DL, Fleming ID. AJCC cancer staging manual. 6th ed. Springer, Chicago, 2002: Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet 379: , Mulier S, Ni Y, Jamart J, Ruers T, Marchal G, Michel L. Local recurrence after hepatic radiofrequency coagulation: multivariate meta-analysis and review of contributing factors. Ann Surg 242: , Hu RH, Lee PH, Chang YC, Ho MC, Yu SC. Treatment of centrally located hepatocellular carcinoma with central hepatectomy. Surgery 133: , Camma C, Schepis F, Orlando A, Albanese M, Shahied L, Trevisani F. Transarterial chemoembolization for unresectable hepatocellular carcinoma: meta-analysis of randomized controlled trials. Radiology 224: 47-54, Llovet JM, Bruix J. Systemic review of randomized trials for unresectable hepatocellular carcinoma: chemoembolization improves survival. Hepatology 37: , Ernst O, Serqent G, Mizrahi D, Delemazure O, Paris JC, L Hermine C. Treatment of hepatocellular carcinoma by transcatheter arterial chemoembolization: comparison of planned periodic chemoembolization and chemoembolization based on tumor re- 852

7 sponse. Am J Roentgenol 172: 59-64, Takayasu K, Arii S, Ikai I, Omata M, Okita K, Ichida T. Prospective cohort study of transarterial chemoembolization for unresectable hepatocellular carcinoma in 8510 patients. Gastroenterology 131: , Wu JC, Huang YH, Chau GY, Su CW, Lai CR, Lee PC. Risk factors for early and late recurrence in hepatitis B-related hepatocellular carcinoma. J Hepatol 51: , Bircher J. Quantitative assessment of deranged hepatic function: a missed opportunity? Semin Liver Dis 3: 275, Kim JM, Kwon CH, Joh JW, Kim EY, Moon JI, Jung GO. Patients with unresectable hepatocellular carcinoma beyond Milan criteria: should we perform transarterial chemoembolization or liver transplantation? Tansplant Proc 42: , Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Eng J Med 334: , The Japanese Society of Internal Medicine 853