bs_bs_banner

doi:10.1111/jgh.12470

H E PAT O L O G Y

Clinical outcomes of a cohort series of patients with hepatocellular carcinoma in a hepatitis B virus-endemic area Hee-Won Kwak,* Joong-Won Park,* Byung-Ho Nam,† Ami Yu,† Sang Myung Woo,* Tae Hyun Kim,* Seong Hoon Kim,* Young Hwan Koh,* Hyun Beom Kim,* Sang Jae Park,* Woo Jin Lee,* Eun Kyung Hong* and Chang-Min Kim* *Center for Liver Cancer, and †Cancer Biostatistics Branch, National Cancer Center, Goyang, Korea

Key words cohort, HBV-endemic area, hepatocellular carcinoma, survival outcome, temporal comparison. Accepted for publication 14 November 2013. Correspondence Dr Joong-Won Park, Center for Liver Cancer, National Cancer Center, 323 Ilsan-ro, Ilsan dong-gu, Goyang, Gyeonggi 411-769, South Korea. Email: [email protected]

Abstract Background and Aims: To evaluate the clinical outcomes of patients with hepatocellular carcinoma (HCC) and compare the findings with that of a previous cohort. Methods: Overall, 1972 HCC patients diagnosed and treated at the National Cancer Center, Korea between 2004 and 2009 were enrolled. The data of this cohort were compared with those of a previous cohort (2000–2003) from the same institution. Results: In all (mean age, 56.4 years; 1642 men), 74.6% was hepatitis B virus (HBV) positive, 81.6% were Child–Pugh (CP) class A, and 64.4% was Barcelona Clinic Liver Cancer (BCLC) stage C. The modified Union for International Cancer Control (mUICC) stage I, II, III, IVa, and IVb was found in 8.9%, 29.6%, 24.8%, 23.1%, and 13.6% patients, respectively. The most common initial treatment was transarterial chemotherapy (58.3%), followed by resection (18.6%). The 5-year survival rate of BCLC stage 0, A, B, and C were 79.6%, 67.2%, 33.9%, and 17.1%, respectively. The performance status, BCLC stage, mUICC stage, CP class, model for end-stage liver disease score, tumor characteristics, portal vein tumor invasion, and serum alpha-fetoprotein level proved to be independent prognostic variables. Overall survival in the present cohort was better than that in the previous cohort (hazard ratio, 0.829; 95% confidence interval, 0.754–0.912), especially for advanced HCC patients with HBV-positive status. Conclusions: This cohort study provides valuable insights into the characteristics of HCC in Korean patients. Our findings may help develop clinical trials, treatment strategies, and prognosis systems for HCC patients in HBV-endemic areas.

Introduction Primary liver cancer is currently ranked as the fifth most common cancer and the third leading cause of cancer-related deaths worldwide.1,2 A large proportion of the cases occur in Eastern Asia and Africa. In South Korea, primary liver cancer is the fourth and sixth most prevalent cancer in men and women, respectively, and over 15 000 new cases have developed in 2009;3 hepatocellular carcinoma (HCC) accounts for 75–90% of all primary liver cancers.3–5 HCC was the second leading cause of cancer mortality in South Korea, but the overall national 5-year survival rate (5-YSR) of patients with HCC has improved from 10.7% to 23.3% over the past 10 years.6 This improvement is assumed to be due to early detection via nationwide surveillance programs and increased use of nucleosi(ti)de analogue therapy in patients with hepatitis B virus (HBV) infections, which is a main risk factor of HCC in South Korea.5,7 Recent improvement in clinical outcomes of HCC patients may be partly due to the evidence-based treatment guidelines.8 However, data of observational outcome studies in a 820

practice-based HCC cohort are lacking. We have already reported the clinical characteristics and treatment outcomes of an HCC cohort,9,10 the data of which were used for developing a new prognosis system.11 Even after publishing the first report, we continued collecting the clinical data of patients with HCC prospectively. In this longitudinal cohort study, we analyzed the survival outcomes of patients with HCC who were diagnosed and treated according to the guidelines12,13 at a single center in Korea. The stratified descriptions of the clinical outcomes in the present cohort were compared temporally with those of the previous cohort to gain deeper insights into the characteristics of HCC and to provide useful information in an HBV-endemic area.

Patients and methods Patients. Overall, 2608 HCC patients were treated at the Center for Liver Cancer, National Cancer Center (NCC), Goyang,

Journal of Gastroenterology and Hepatology 29 (2014) 820–829 © 2013 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

H-W Kwak et al.

Outcomes and changes of the HCC cohort

South Korea between January 2004 and December 2009. Of these, 1972 patients were newly diagnosed with HCC and had not undergone prior antitumor therapy. These patients comprised the cohort of this study and were followed up until March 2012. These patients were collected prospectively, and relevant data of their clinical and tumor characteristics were extracted retrospectively from their medical records. HCC was diagnosed on the basis of the guidelines of the Korea Liver Cancer Study Group—NCC, Korea.12,13 The Modified Union for International Cancer Control (mUICC) stages were used for tumor staging,14 and the Barcelona Clinic for Liver Cancer (BCLC) staging system was used for clinical staging.15 The Child– Pugh (CP) and Model for End-stage Liver Disease (MELD) scoring systems were used for clinical diagnosis and classification of the severity of liver dysfunction, respectively.16 In addition, the data of the current cohort was compared with those of the previous cohort, which consisted of 904 HCC patients treated between November 2000 to December 2003.9 This study was approved by the institutional review board of our institute (NCC, Goyang, South Korea).

years (range, 17–89 years). The main cause of HCC was HBV infection (74.6%). Most patients (81.6%) were classified under Child–Pugh class A (CPA), and 62.1% patients had a MELD score of below 10. With respect to the BCLC stage, most patients were classified as stage C (64.4%), followed by stage A (19.8%). According to the mUICC stage, 38.5% patients were stage I or II and 61.5% patients stage III or above. Most patients (51.2%) had one tumor, and 26.1% patients had more than four tumors. In 51.3% patients, the tumor was more than 5 cm in diameter. With regard to the tumor type, more patients had well-defined HCC (67.4%) than poorly defined HCC (32.6%). Overall, 675 patients (34.2%) had PVTT, and among them, 201 patients (10.2%) had main portal vein invasion. Extrahepatic spread was found in 18.2% patients. The serum α-fetoprotein (AFP) level was normal (≤ 20 ng/mL) in 29% patients; and in 48.1% patients, it was greater (> 200 ng/mL). The proteins induced by vitamin K absence (PIVKA)-II serum level was additionally checked in 791 patients, and 71.6% of them showed an elevation (≥ 40 ng/mL).

Treatment methods. Patients received one of the following modalities: liver resection, liver transplantation, radiofrequency ablation therapy (RFA), conventional transarterial chemoembolization (cTACE) therapy, external beam radiation therapy (RT), systemic cytotoxic chemotherapy, molecular-targeted therapy (sorafenib), and conservative management. Of these treatments, liver transplantation and sorafenib therapy were newly employed during this study.9 The initial treatment modalities were applied to most patients according to the Korea guidelines.12,13 RFA was carried out by using either single or multiple cooled-tip electrodes (Cooltip; Radionics, Burlington, MA, USA). cTACE was performed according to the protocol of the NCC, Korea, by using doxorubicin hydrochloride (ADM, Dong-a Pharmaceutical Co., Seoul, Korea) in iodized oil.17 The three-dimensional conformal RT was used for cases of unresectable HCC that did not respond to cTACE or that involved portal vein tumor thrombosis (PVTT), extrahepatic metastasis of bone, the adrenal glands, the brain, and enlarged symptomatic lymph node.18 Systemic cytotoxic chemotherapy19 or sorafenib (400 mg twice daily; Nexavar, Bayer Healthcare, Leverkusen, Germany) was administered to patients with unresectable advanced HCC that was refractory to cTACE or other local treatments or involved PVTT or extrahepatic metastasis.20 Conservative management involved symptomatic treatment and management of complications of liver cirrhosis.

Initial treatment modalities. Most treatment methods, except liver transplantation and conservative management, were performed in patients with CPA or Child–Pugh class B (CPB) (Table 1). Liver transplantation was performed initially in 30 patients (1.5%), and this group showed the highest 5-YSR of 80.5%. Liver resection and RFA were conducted initially in 367 (18.6%) and 70 patients (3.5%) with mUICC stages I–III, respectively, and their 5-YSR was 68.2% and 77.9%, respectively. Nine (12.9%) patients of the RFA group and 14 (3.8%) of the resection group had liver function classified as CPB. Most patients (n = 1149, 58.3%) were initially treated with cTACE, and 81.2% of these patients had a mUICC stage of II, III, or IVa (Tables 1 and 2). Patients whose liver functions were classified as CPA or CPB but were not indicated for curative treatment or cTACE due to diffuse infiltration, extensive vascular invasion, or distant metastasis were initially treated with RT (5.0%), systemic cytotoxic chemotherapy (4.0%), or sorafenib therapy (1.7%, Table 1).

Statistical analyses. The Kaplan–Meier method, the logrank test, and the Cox proportional hazard model were used for analysis. All statistical analyses were performed using the STATA version 10.0 (Stata Corp, College Station, TX, USA), and values of P < 0.05 was considered statistically significant.

Results Clinical characteristics. The clinical characteristics and survival outcomes of the present HCC cohort of 1972 patients are summarized in Table 1. The median age of the patients was 56.0

Survival analyses. Stratified survival analysis of the present cohort is shown in Table 1. During the follow-up period (median, 21.2 months), 1336 patients (67.7%) died, and the overall median survival time was 21.2 months (Table 1, Fig. 1a). The median survival times of patients whose liver functions were classified as CPA, CPB, and Child–Pugh class C (CPC) were 26.8, 8.0, and 4.6 months, respectively. Seven (20.6%) patients in the CPC group underwent liver transplantation and survived, resulting in a better 5-YSR (22.7%) than that of the CPB group (13.9%, Fig. 1b). The 5-YSR of patients with BCLC stage 0, A, B, C, and D were 79.6%, 67.2%, 34.3%, 17.1%, and 21.4%, respectively (Table 1, Fig. 1c). Liver transplantation also improved the survival of patients classified with BCLC stage D. The 5-YSR of patients with mUICC stage I, II, III, IVa, and IVb were 71.1%, 59.8%, 25.0%, 4.6%, and 2.1%, respectively (Fig. 1d). The survival rates according to the initial treatment modalities (Table 1) were sub-analyzed according to the mUICC stage in patients with CPA (n = 1609, Table 2). In survival rate stratified by initial treatment and mUICC stage I–II in the patients with CPA,

Journal of Gastroenterology and Hepatology 29 (2014) 820–829 © 2013 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

821

Outcomes and changes of the HCC cohort

Table 1

H-W Kwak et al.

Patient characteristics and survival rate of the present cohort (n = 1972)

Characteristics Overall survival Age (years) Sex Cause

ECOG performance status

Child–Pugh class

BCLC stage

mUICC stage

MELD score

Tumor number

Tumor size (cm)

Tumor type Portal vein tumor thrombosis Extrahepatic spread AFP

PIVKA-II¶

Initial treatment

≤ 49 ≥ 50 Male Female HBV HCV Alcohol NBNCNA 0 1 2 3 A B C† 0 A B C D‡ I II III IVa IVb < 10 ≥ 10, < 20 ≥ 20, < 30 ≥ 30§ 1 2–3 ≥4 2, ≤ 5 > 5, ≤ 10 > 10 Well-defined Poorly defined None First or second branch Main Negative Positive < 20 ≥ 20, < 200 ≥ 200

Hazard ratio

0.2012 0.1418 0.0327† †

0.01 0.0221†

1 1.02 1 0.88 0.85 0.80 1 1.26 1.45

95% CI

Likelihood ratio (P-value) 0.8094

0.87

1.20

0.73 0.68 0.65

1.07 1.06 0.98

0.0614

0.0141† 1.06 1.05

1.49 1.98 0.0367†

0.7713 0.2143 0.1098 0.3367 †

0.0096 < 0.0001† < 0.0001† < 0.0001† < 0.0001

†

< 0.0001† 0.53 0.0013† 0.3203 0.2001 0.0014† †

0.0094

< 0.0001 < 0.0001† †

0.0576 0.0008 < 0.0001†

1.10 1.51 1.68 1.45 1 1.78 2.97 3.97 3.82 1 1.51 1 1.40 1 0.95 1.28 1 0.87 1.21 1.66 1 1.22 1 1.59 1.98 1 1.27 1 1.34 1.47

0.58 0.79 0.89 0.68

2.09 2.88 3.17 3.09 < 0.0001†

1.15 1.85 2.40 2.19

2.76 4.79 6.55 6.66 < 0.0001†

1.30

1.75

1.24

1.58

0.81 1.10

1.12 1.49

0.65 0.90 1.22

1.15 1.63 2.27

< 0.0001† 0.0001†

< 0.0001†

0.0094† 1.05

1.42 < 0.0001†

1.34 1.60

1.89 2.46

0.99

1.62

1.13 1.26

1.58 1.71

0.0632 < 0.0001†

PIVKA II was excluded because of missing values. † Statistically significant. ‡ ECOG 3 status excluded in univariate/multivariate analysis because of small sample size (n = 3). AFP, α-fetoprotein; BCLC, Barcelona Clinic Liver Cancer; ECOG, Eastern Cooperative Oncology Group; HBV, hepatitis B virus; HCV, hepatitis C virus; MELD, Model for End-stage Liver Disease; mUICC, modified Union for International Cancer Control; NBNCNA, non-B non-C non-alcoholic.

symptoms or mild discomfort (ECOG performance status 0 or 1) and preserved liver function (CPA, 81.6%) but were diagnosed with advanced stages of HCC (BCLC stage C, 64.4%). Such a patient population reflects the characteristics of a referral hospital (i.e. advanced HCC patients who are still feasible to treatment are referred to our institution). Therefore, this patient population may not exactly reflect the nationwide population. The one of tumor number was main (51.2%), but multinodular confluent type was considered as one of tumor number in this study.

In this cohort, cTACE was performed as an initial treatment in more than half of the patients, yielding a 5-YSR of 25.6% and a median survival time of 19.8 months (Table 1). The global observational study also reported that although HCC patients diagnosed with a BCLC stage C are predominant, cTACE is the most frequently used first recorded treatment in Asia and in North America.26 We performed cTACE for cases of HCC that are not indicated for curative treatment or that are partly associated with vascular invasion by Korean guidelines.13 The 5-YSR of patients

Journal of Gastroenterology and Hepatology 29 (2014) 820–829 © 2013 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

825

826

< 20 ≥ 20, < 200 ≥ 200

None First or second branch Main

Well-defined Poorly defined