Research Article
Incidence of hepatocellular carcinoma after HBsAg seroclearance in chronic hepatitis B patients: A need for surveillance Gi-Ae Kim1, Han Chu Lee1,⇑, Min-Ju Kim2, Yeonjung Ha1, Eui Ju Park1, Jihyun An1, Danbi Lee1, Ju Hyun Shim1, Kang Mo Kim1, Young-Suk Lim1 1
Department of Gastroenterology, Asan Liver Center, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea; Department of Clinical Epidemiology and Biostatistics, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea
2
Background & Aims: Little is known about whether surveillance for hepatocellular carcinoma (HCC) is worthwhile in chronic hepatitis B virus (HBV)-infected patients who have achieved HBsAg seroclearance. Methods: A retrospective analysis of 829 patients (mean age: 52.3 years; 575 males; 98 with cirrhosis) achieving HBsAg seroclearance was performed at a tertiary hospital in Korea between 1997 and 2012. We evaluated incidence rates of HCC, and validated CU-HCC score based on data at the time of HBsAg seroclearance. Results: During a follow-up of 3464 patient-years, 19 patients developed HCC (annual rate: 0.55%). Liver cirrhosis (hazard ratio [HR]: 10.80; 95% confidence interval [CI]: 4.25–27.43), male gender (HR: 8.96; 95% CI: 1.17–68.80), and age P50 years at the time of HBsAg seroclearance (HR: 12.14; 95% CI: 1.61– 91.68) were independently associated with HCC. The estimated annual incidence of HCC was 2.85% and 0.29% in patients with and without cirrhosis, respectively. Among the non-cirrhotic patients, the annual rate of HCC was higher in the male patients than in the females (0.40% vs. 0%, respectively), and all the HCCs developed after age 50. The time-dependent area under the receiver operating characteristic curves for the CU-HCC score for 5 year and 10 year HCC prediction were 0.85 and 0.74, respectively. Conclusions: HCC surveillance should be considered for cirrhotic patients and non-cirrhotic male patients over age 50, even after
Keywords: HBV; Liver cancer; Survival. Received 27 August 2014; received in revised form 19 November 2014; accepted 20 November 2014; available online 28 November 2014 ⇑ Corresponding author. Address: Department of Gastroenterology, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic-ro 43-gil, Songpa-gu, Seoul 138-736, Republic of Korea. Tel.: +82 2 3010 3915; fax: +82 2 485 5782. E-mail address: [email protected] (H.C. Lee). Abbreviations: AFP, alpha-fetoprotein; ALT, alanine aminotransferase; AUC, area under the receiver operating characteristic curve; BCLC, Barcelona Clinic Liver Cancer; CHB, chronic hepatitis B; CI, confidence interval; GGT, gamma-glutamyl transferase; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HR, hazard ratio; INR, international normalized ratio; IQR, interquartile range; NUC, nucleoside analogue; ROC, receiver operating characteristics; SD, standard deviation.
HBsAg seroclearance, especially those infected with HBV genotype C. HBsAg seroclearance at age P50 years was also an independent predictor for HCC. Ó 2014 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Introduction Hepatitis B surface antigen (HBsAg) seroclearance is considered to be the most important end point of chronic hepatitis B virus (HBV) infection [1–4] because both spontaneous and therapy-induced-HBsAg seroclearance are associated with histological improvement, a reduced risk of hepatocellular carcinoma (HCC), and prolonged survival [5–11]. However, several reports have shown that clinical complications, such as hepatic decompensation or HCC, may occur even after HBsAg seroclearance, particularly in patients superinfected with other viruses or in those with liver cirrhosis [5,7,11–14]. Surveillance for HCC is cost-effective when the annual risk of HCC exceeds 0.2% in non-cirrhotic hepatitis B patients and 1.5% in cirrhotic patients [15,16]. However, little is known about whether surveillance for HCC is worthwhile in chronic HBV-infected patients who have achieved HBsAg seroclearance. Moreover, the reported rates of HCC after HBsAg seroclearance are markedly different (between 2.9% and 30.7% in cirrhotic patients and between 0% and 3.1% in non-cirrhotic patients) due to the low number of reported patients who achieve HBsAg seroclearance but still develop HCC [5,6,9,12]. There are several validated HCC risk scores that are useful tools on which to base a prognosis and decisions regarding HCC surveillance; however, it is still unclear whether these risk scores are accurate enough to predict the development of HCC in chronic HBV-infected patients with HBsAg seroclearance. Here, we examined data from 829 HBV-infected patients who achieved HBsAg seroclearance with the aim of evaluating the incidence and risk factors associated with the development of HCC. We also evaluated the accuracy and applicability of the CU-HCC score for predicting the risk of developing HCC [17].
Journal of Hepatology 2015 vol. 62 j 1092–1099
JOURNAL OF HEPATOLOGY The study was approved by the Institutional Review Board of Asan Medical Center (IRB No. 2014-0030).
1485 chronic HBV-infected patients with HBsAg seroclearance between 1997 and 2012
Follow-up evaluation and outcomes
454 were excluded • 158 did not receive further HCC surveillance during follow-up in other departments of our hospital to which they were transferred due to other illnesses • 296 refused follow-up after HBsAg seroclearance 1031 chronic HBV-infected patients with HBsAg seroclearance who underwent imaging after HBsAg seroclearance 202 were excluded • 90 had HCC before HBsAg seroclearance or within 6 months of HBsAg seroclearance • 24 had prior cirrhosis-related complications • 85 had prior immunosuppressive therapy • 3 were lost to follow-up and revisited the hospital after HCC development
829 eligible patients
Patients without cirrhosis (n = 731)
Patients with cirrhosis (n = 98)
Fig. 1. Patient flow diagram.
Patients and methods
Follow-up comprised clinical examinations and liver function tests along with the measurement of HBsAg, anti-HBs, HBeAg, anti-HBe, and HBV DNA levels every 6–12 months. Patients were screened for cirrhosis and HCC using abdominal ultrasonography every 6–12 months. A diagnosis of HCC was based on histological examination and/or imaging evidence of arterial hypervascularity and venous/delayed washout, as defined by the AASLD guidelines [15]. All images were reviewed by physicians and two radiologists (H.J.W. and Y.M.S) with 10 and 20 years of experience in imaging of the liver, respectively. Information related to a family history of HCC in first-degree relatives (parents, siblings, and children) and significant alcohol consumption (>21 drinks per week in men and >14 drinks per week in women) was also obtained by searching the clinical database [18,19]. The primary outcome measure of this study was the incidence of HCC development. The secondary outcome was incidence of death or liver transplantation. Follow-up duration for HCC development was defined as the interval between the date of HBsAg seroclearance and the time of HCC development or last imaging test (median follow-up duration: 3.2 years). Information about patient survival was obtained from the National Population Registry of Korea National Statistical Office using personal identification numbers through to March 31, 2014 (median follow-up duration: 4.1 years).
Serum assays Serum HBsAg was measured using the Architect assay (Abbott Laboratories, Chicago, IL, USA) or an immunoradiometric assay kit (DiaSorin, Vercelli, Italy). Prior to 2007, serum HBV DNA levels were measured using a hybrid capture assay (Digene Diagnostics, Gaithersburg, MD; lower limit of detection: 20,000 IU/ml); a real-time PCR assay (Abbott Laboratories, Chicago, IL; detection range: 15– 1 109 IU/ml) was used after 2007. The HBV genotype was not determined because approximately 98% of Korean patients with chronic hepatitis B (CHB) have genotype C HBV [3,20]. Serum markers of hepatitis virus infection, including HBeAg and anti-HBs, anti-HBe, anti-HCV, and anti-HIV antibodies, were measured using commercially available enzyme immunoassays (Abbott Laboratories, Chicago, IL). Routine laboratory parameters were assayed using standard analytical procedures.
Study subjects
HCC risk score
The study population comprised 1485 consecutive patients attending a tertiary referral hospital in Korea between January 1, 1997 and December 31, 2012; all were chronically infected with HBV and had achieved HBsAg seroclearance (Fig. 1). Data were collected from the hospital clinical database containing electronic medical records. The inclusion criteria were as follows: HBsAg-positive for more than 6 months; achievement of HBsAg seroclearance (defined as the loss of serum HBsAg for at least two occasions and clearance not related to liver transplantation); no evidence of co-infection with hepatitis C, hepatitis D, or human immunodeficiency viruses; and no evidence of HCC or other malignancies. We first excluded 454 patients who did not undergo imaging study after HBsAg seroclearance; patients who did not receive further HCC surveillance during the follow-up in other departments of our hospital due to other illness (n = 158); patients who refused to follow-up after HBsAg seroclearance (n = 296). Next, we excluded patients if they developed HCC before HBsAg seroclearance or within 6 months after HBsAg seroclearance (n = 90); suffered cirrhosis-related complications before HBsAg seroclearance (n = 24); received immunosuppressive therapy due to other malignancies or had an organ transplant before HBsAg seroclearance (n = 85); or were lost to follow-up after HBsAg seroclearance and revisited the hospital after HCC development (n = 3). Therefore, data from 829 eligible patients were analyzed in this study (Fig. 1). Liver cirrhosis was diagnosed from liver imaging of coarse and nodular echotexture and/or clinical evidence of portal hypertension (e.g. ascites, gastroesophageal varices, or splenomegaly with a platelet count
Incidence of hepatocellular carcinoma after HBsAg seroclearance in chronic hepatitis B patients: A need for surveillance Gi-Ae Kim1, Han Chu Lee1,⇑, Min-Ju Kim2, Yeonjung Ha1, Eui Ju Park1, Jihyun An1, Danbi Lee1, Ju Hyun Shim1, Kang Mo Kim1, Young-Suk Lim1 1
Department of Gastroenterology, Asan Liver Center, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea; Department of Clinical Epidemiology and Biostatistics, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea
2
Background & Aims: Little is known about whether surveillance for hepatocellular carcinoma (HCC) is worthwhile in chronic hepatitis B virus (HBV)-infected patients who have achieved HBsAg seroclearance. Methods: A retrospective analysis of 829 patients (mean age: 52.3 years; 575 males; 98 with cirrhosis) achieving HBsAg seroclearance was performed at a tertiary hospital in Korea between 1997 and 2012. We evaluated incidence rates of HCC, and validated CU-HCC score based on data at the time of HBsAg seroclearance. Results: During a follow-up of 3464 patient-years, 19 patients developed HCC (annual rate: 0.55%). Liver cirrhosis (hazard ratio [HR]: 10.80; 95% confidence interval [CI]: 4.25–27.43), male gender (HR: 8.96; 95% CI: 1.17–68.80), and age P50 years at the time of HBsAg seroclearance (HR: 12.14; 95% CI: 1.61– 91.68) were independently associated with HCC. The estimated annual incidence of HCC was 2.85% and 0.29% in patients with and without cirrhosis, respectively. Among the non-cirrhotic patients, the annual rate of HCC was higher in the male patients than in the females (0.40% vs. 0%, respectively), and all the HCCs developed after age 50. The time-dependent area under the receiver operating characteristic curves for the CU-HCC score for 5 year and 10 year HCC prediction were 0.85 and 0.74, respectively. Conclusions: HCC surveillance should be considered for cirrhotic patients and non-cirrhotic male patients over age 50, even after
Keywords: HBV; Liver cancer; Survival. Received 27 August 2014; received in revised form 19 November 2014; accepted 20 November 2014; available online 28 November 2014 ⇑ Corresponding author. Address: Department of Gastroenterology, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic-ro 43-gil, Songpa-gu, Seoul 138-736, Republic of Korea. Tel.: +82 2 3010 3915; fax: +82 2 485 5782. E-mail address: [email protected] (H.C. Lee). Abbreviations: AFP, alpha-fetoprotein; ALT, alanine aminotransferase; AUC, area under the receiver operating characteristic curve; BCLC, Barcelona Clinic Liver Cancer; CHB, chronic hepatitis B; CI, confidence interval; GGT, gamma-glutamyl transferase; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HR, hazard ratio; INR, international normalized ratio; IQR, interquartile range; NUC, nucleoside analogue; ROC, receiver operating characteristics; SD, standard deviation.
HBsAg seroclearance, especially those infected with HBV genotype C. HBsAg seroclearance at age P50 years was also an independent predictor for HCC. Ó 2014 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Introduction Hepatitis B surface antigen (HBsAg) seroclearance is considered to be the most important end point of chronic hepatitis B virus (HBV) infection [1–4] because both spontaneous and therapy-induced-HBsAg seroclearance are associated with histological improvement, a reduced risk of hepatocellular carcinoma (HCC), and prolonged survival [5–11]. However, several reports have shown that clinical complications, such as hepatic decompensation or HCC, may occur even after HBsAg seroclearance, particularly in patients superinfected with other viruses or in those with liver cirrhosis [5,7,11–14]. Surveillance for HCC is cost-effective when the annual risk of HCC exceeds 0.2% in non-cirrhotic hepatitis B patients and 1.5% in cirrhotic patients [15,16]. However, little is known about whether surveillance for HCC is worthwhile in chronic HBV-infected patients who have achieved HBsAg seroclearance. Moreover, the reported rates of HCC after HBsAg seroclearance are markedly different (between 2.9% and 30.7% in cirrhotic patients and between 0% and 3.1% in non-cirrhotic patients) due to the low number of reported patients who achieve HBsAg seroclearance but still develop HCC [5,6,9,12]. There are several validated HCC risk scores that are useful tools on which to base a prognosis and decisions regarding HCC surveillance; however, it is still unclear whether these risk scores are accurate enough to predict the development of HCC in chronic HBV-infected patients with HBsAg seroclearance. Here, we examined data from 829 HBV-infected patients who achieved HBsAg seroclearance with the aim of evaluating the incidence and risk factors associated with the development of HCC. We also evaluated the accuracy and applicability of the CU-HCC score for predicting the risk of developing HCC [17].
Journal of Hepatology 2015 vol. 62 j 1092–1099
JOURNAL OF HEPATOLOGY The study was approved by the Institutional Review Board of Asan Medical Center (IRB No. 2014-0030).
1485 chronic HBV-infected patients with HBsAg seroclearance between 1997 and 2012
Follow-up evaluation and outcomes
454 were excluded • 158 did not receive further HCC surveillance during follow-up in other departments of our hospital to which they were transferred due to other illnesses • 296 refused follow-up after HBsAg seroclearance 1031 chronic HBV-infected patients with HBsAg seroclearance who underwent imaging after HBsAg seroclearance 202 were excluded • 90 had HCC before HBsAg seroclearance or within 6 months of HBsAg seroclearance • 24 had prior cirrhosis-related complications • 85 had prior immunosuppressive therapy • 3 were lost to follow-up and revisited the hospital after HCC development
829 eligible patients
Patients without cirrhosis (n = 731)
Patients with cirrhosis (n = 98)
Fig. 1. Patient flow diagram.
Patients and methods
Follow-up comprised clinical examinations and liver function tests along with the measurement of HBsAg, anti-HBs, HBeAg, anti-HBe, and HBV DNA levels every 6–12 months. Patients were screened for cirrhosis and HCC using abdominal ultrasonography every 6–12 months. A diagnosis of HCC was based on histological examination and/or imaging evidence of arterial hypervascularity and venous/delayed washout, as defined by the AASLD guidelines [15]. All images were reviewed by physicians and two radiologists (H.J.W. and Y.M.S) with 10 and 20 years of experience in imaging of the liver, respectively. Information related to a family history of HCC in first-degree relatives (parents, siblings, and children) and significant alcohol consumption (>21 drinks per week in men and >14 drinks per week in women) was also obtained by searching the clinical database [18,19]. The primary outcome measure of this study was the incidence of HCC development. The secondary outcome was incidence of death or liver transplantation. Follow-up duration for HCC development was defined as the interval between the date of HBsAg seroclearance and the time of HCC development or last imaging test (median follow-up duration: 3.2 years). Information about patient survival was obtained from the National Population Registry of Korea National Statistical Office using personal identification numbers through to March 31, 2014 (median follow-up duration: 4.1 years).
Serum assays Serum HBsAg was measured using the Architect assay (Abbott Laboratories, Chicago, IL, USA) or an immunoradiometric assay kit (DiaSorin, Vercelli, Italy). Prior to 2007, serum HBV DNA levels were measured using a hybrid capture assay (Digene Diagnostics, Gaithersburg, MD; lower limit of detection: 20,000 IU/ml); a real-time PCR assay (Abbott Laboratories, Chicago, IL; detection range: 15– 1 109 IU/ml) was used after 2007. The HBV genotype was not determined because approximately 98% of Korean patients with chronic hepatitis B (CHB) have genotype C HBV [3,20]. Serum markers of hepatitis virus infection, including HBeAg and anti-HBs, anti-HBe, anti-HCV, and anti-HIV antibodies, were measured using commercially available enzyme immunoassays (Abbott Laboratories, Chicago, IL). Routine laboratory parameters were assayed using standard analytical procedures.
Study subjects
HCC risk score
The study population comprised 1485 consecutive patients attending a tertiary referral hospital in Korea between January 1, 1997 and December 31, 2012; all were chronically infected with HBV and had achieved HBsAg seroclearance (Fig. 1). Data were collected from the hospital clinical database containing electronic medical records. The inclusion criteria were as follows: HBsAg-positive for more than 6 months; achievement of HBsAg seroclearance (defined as the loss of serum HBsAg for at least two occasions and clearance not related to liver transplantation); no evidence of co-infection with hepatitis C, hepatitis D, or human immunodeficiency viruses; and no evidence of HCC or other malignancies. We first excluded 454 patients who did not undergo imaging study after HBsAg seroclearance; patients who did not receive further HCC surveillance during the follow-up in other departments of our hospital due to other illness (n = 158); patients who refused to follow-up after HBsAg seroclearance (n = 296). Next, we excluded patients if they developed HCC before HBsAg seroclearance or within 6 months after HBsAg seroclearance (n = 90); suffered cirrhosis-related complications before HBsAg seroclearance (n = 24); received immunosuppressive therapy due to other malignancies or had an organ transplant before HBsAg seroclearance (n = 85); or were lost to follow-up after HBsAg seroclearance and revisited the hospital after HCC development (n = 3). Therefore, data from 829 eligible patients were analyzed in this study (Fig. 1). Liver cirrhosis was diagnosed from liver imaging of coarse and nodular echotexture and/or clinical evidence of portal hypertension (e.g. ascites, gastroesophageal varices, or splenomegaly with a platelet count
