Journal of Medical Virology
Serum Hepatitis B Surface Antigen Correlates With Tissue Covalently Closed Circular DNA in Patients With Hepatitis B-Associated Hepatocellular Carcinoma Qin Wang,1 Wei Luan,1 Leslie Warren,1 M. Isabel Fiel,2 Sima Blank,1 Hena Kadri,1 Daniel Tuvin,1 and Spiros P. Hiotis1* 1 2
Department of Surgery, The Icahn School of Medicine at Mount Sinai, New York City, New York Department of Pathology, The Icahn School of Medicine at Mount Sinai, New York City, New York
Quantitation of hepatitis B surface antigen (HBsAg) in hepatitis B-associated hepatocellular carcinoma (HBV-HCC) remains to be clearly defined. This study aims to determine the association of HBsAg quantity with intrahepatic HBV viral load and activity in both tumor and non-neoplastic liver of HBV-HCC patients. Data were obtained from 89 prospectively enrolled patients treated with primary liver resection for HBV-HCC at a single Western institution (2008–2013). Circulating HBsAg was quantitated using ELISA. HBV DNA, covalently closed circular (cccDNA) and precore-pregenomic RNA (preC-pgRNA) in both tumor and non-neoplastic liver were quantitated by real-time PCR from fresh liver resection specimens. Circulating HBsAg was detectable in all 89 patients. HBsAg negatively correlated with age, and positively correlated with pre-operative serum AFP and ALT levels. HBsAg correlated with HBV cccDNA copy number in tumor or non-neoplastic liver tissue. It also correlated with preC-pgRNA copy number in non-neoplastic liver tissue. HBsAg did not correlate with serum HBV DNA, total intrahepatic HBV DNA, viral replicative activity or transcriptional activity. In HBV-HCC patients, HBsAg levels correlated with cccDNA copy number in tumor or non-neoplastic liver tissue, suggesting that a greater pool of cccDNA is associated with a higher rate of HBsAg production. J. Med. Virol. 9999: XX–XX, 2015.
hepatocellular carcinoma (HCC) [Fattovich et al., 2004; Bonilla Guerrero and Roberts, 2005; Liu et al., 2006; Liu et al., 2009; Lok and McMahon, 2007; El-Serag, 2011; Yang et al., 2011]. HBV viral products as well as HBV viral integration into the host genome may have direct oncogenic potential that contributes to hepatocarcinogenesis [Sung et al., 2012]. Besides HBV-specific mechanisms, chronic HBV infection also induces inflammation, liver cell injury and regeneration, and fibrosis, thereby creating an environment that may favor hepatocarcinogenesis [Arzumanyan et al., 2013]. Hepatitis B is a DNA virus that is surrounded by a host cell-derived envelope. HBsAg is embedded within the lipid bilayer of the viral envelope where it is present in three forms (small, medium, and large HBsAg), and is important for viral structure as well as viral entry into the host cell [Chan et al., 2011; Lee and Ahn, 2011; Yan et al., 2012]. The HBV genome is a relaxed circular DNA (rcDNA) that is partially double-stranded and about 3.2 kb in length [Seeger and Mason, 2000]. Once inside the host cell, the rcDNA is transported into the host cell nuclei, where the rcDNA is converted into covalently closed circular DNA (cccDNA) [Levrero et al., 2009]. The HBV genome contains four overlapping open reading frames (ORF) that encode all viral proteins, including HBsAg. Four mRNAs, each under the control of a separate promoter, are transcribed. Two mRNA species (PreS1 and PreS2/S mRNA) give rise to the
# 2015 Wiley Periodicals, Inc.
Conflict of interest: None. Correspondence to: Spiros P. Hiotis, MD, PhD, Division of Surgical Oncology, Department of Surgery, The Icahn School of Medicine at Mount Sinai, 5 East 98th Street, 12th Floor, Box 1259, New York City, NY 10029. E-mail: [email protected] Accepted 4 July 2015
KEY WORDS:
HBsAg; cccDNA; HCC; AFP
INTRODUCTION Hepatitis B surface antigen (HBsAg) is an important diagnostic marker for Hepatitis B (HBV) infection, one of the greatest risk factors for the development of C 2015 WILEY PERIODICALS, INC.
DOI 10.1002/jmv.24326 Published online in Wiley Online Library (wileyonlinelibrary.com).
2
Wang et al.
small, medium and large HBsAg, while the preCorepregenomic RNA (preC-pgRNA) gives rises to hepatitis B E antigen (HBeAg), core protein, polymerase and serves as the template for reverse transcription into rcDNA of a newly formed HBV viron. The viral DNA also integrates into host genome, causing instability of host cell genome and increased synthesis of viral proteins including HBsAg. Although antiviral therapy often leads to undetectable serum HBV DNA levels, HBV DNA is often detectable in the liver tissue of chronic HBV patients even in those with undetectable serum HBV DNA [Wang et al., 2013b]. However, liver biopsies in order to assess intrahepatic HBV-DNA levels are invasive and expensive; therefore, it is useful to determine serological markers for intrahepatic viral activity. In treatment-naiive patients, HBsAg changes during the natural history of HBV infection [Nguyen et al., 2010]. In addition, HBsAg correlates with cccDNA copy number, and the reduction of circulating HBsAg is thought to predict responses to interferon anti-viral therapy [Wursthorn et al., 2006; Lee et al., 2011; Matthews et al., 2013; Wang et al., 2013a]. Quantitative measurement of HBsAg is not well understood for patients with HBV-HCC. The goal of this study is to quantitatively measure HBsAg levels in HBV-HCC patients and to determine whether HBsAg correlates with intrahepatic viral DNA in tumor and non-neoplastic liver tissue.
116–9.89 108 copies/ml [Faria et al, 2008]. Information on other available biochemical markers including preoperative alpha-fetoprotein (AFP), platelet count, albumin, alanine aminotransferase (ALT), bilirubin, creatinine and INR was obtained from medical records. Following liver resection, all specimens were reviewed by a single dedicated liver pathologist who was blinded to all patient demographics and clinical outcomes. During dedicated research pathology analysis of specimens, the degree of hepatic fibrosis in non-neoplastic liver tissue and necro-inflammatory grade according histologic activity index (HAI) were scored according to the modified Ishak method [Goodman, 2007]. Vascular invasion and histologic differentiation were also assessed independently of the standard clinical pathology report, which was used only to obtain information on the number of tumors and the largest tumor diameter.
PATIENTS AND METHODS
Quantitative Intrahepatic HBV, DNA, and cccDNA Measurements
Patients, Histopathologic Assessment, and Follow-Up Data were obtained from selected HBV-HCC patients who underwent primary surgical resection at the Mount Sinai Medical Center in New York, NY, USA between 2008 and 2013. The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki and all aspects of research were approved by the Institutional review board prior to study initiation. Informed consent was obtained from every patient enrolled in the study. HBV-HCC patients were assessed pre-operatively by axial imaging (3-phase computerized tomography (CT) with intravenous contrast, or multi-phase magnetic resonance imaging (MRI) with intravenous contrast). Liver resection was performed on patients who had technicallyresectable disease and normal synthetic liver function (according to pre-operative bilirubin, albumin, and INR). Patients with portal hypertension (platelet count 200 ng/ml) is more likely to be detected in HCC patients with positive serum HBsAg [Peng et al., 2004]. HBV transcription, including the transcription of HBsAg, requires host cell transcription factors; therefore, it is possible that HBsAg and AFP share common transcription factors. Both the preS1 and preS2/S promoter and the AFP promoter have binding sites for transcription factors including hepatocyte nuclear factor 1 (HNF-1), nuclear factor 1 (NF-1), and CAAT/enhancer binding protein (C/EBP) [Bois-Joyeux and Danan, 1994; Crowe et al., 1999; Quasdorff and Protzer, 2010]. Moreover, enhanced HBV DNA integration events into host genome correlates with higher AFP expression [Sung et al., 2012], and may result in increased HBsAg production. Another viral protein, HBx, has also been shown to alleviate p53-mediated repression of AFP expression [Ogden et al., 2000], adding another layer of complexity to the possible interaction of HBV viral proteins and AFP expression. Because a role for AFP in HCC prognosis has been demonstrated in the literature [Personeni et al., 2012; Blank et al., 2013], future studies to elucidate the interaction of HBV and AFP will be important to advance further understanding of the prognostic capacity of AFP in HBVassociated HCC. Our previous studies have demonstrated that intrahepatic HBV DNA and viral replicative activity was associated with necro-inflammation and ALT, while
TABLE II. Correlations Between Circulating HBsAg (log10 IU/ml) and Intrahepatic HBV Viral Load and Activity in Tumor or Non-Neoplastic Liver Tissue (n ¼ 89) Correlation with HBsAg (log10 IU/ml)
Tumor Total HBV DNA (log10 copies/cell) cccDNA (log10 copies/cell) preC-pgRNA (log10 copies/cell) cccDNA/HBV DNA ratio pgRNA/cccDNA ratio Non-neoplastic liver tissue Total HBV DNA (log10 copies/cell) cccDNA (log10 copies/cell) preC-pgRNA (log10 copies/cell) cccDNA/HBV DNA ratio pgRNA/cccDNA ratio
N
Range
76 86 55 76 54
4.48–3.65 6.35–0.43 3.70–2.10 12.49–17.03 2.61–5.61
85 82 58 80 53
4.00–1.24 5.59 to 0.80 4.00–1.80 13.07–31.18 1.80–7.63
Pearson R 0.11 0.28 0.15 0.08 0.10 0.15 0.46 0.34 0.10 0.17
P-value 0.35 0.01 0.28 0.58 0.40 0.17
Serum Hepatitis B Surface Antigen Correlates With Tissue Covalently Closed Circular DNA in Patients With Hepatitis B-Associated Hepatocellular Carcinoma Qin Wang,1 Wei Luan,1 Leslie Warren,1 M. Isabel Fiel,2 Sima Blank,1 Hena Kadri,1 Daniel Tuvin,1 and Spiros P. Hiotis1* 1 2
Department of Surgery, The Icahn School of Medicine at Mount Sinai, New York City, New York Department of Pathology, The Icahn School of Medicine at Mount Sinai, New York City, New York
Quantitation of hepatitis B surface antigen (HBsAg) in hepatitis B-associated hepatocellular carcinoma (HBV-HCC) remains to be clearly defined. This study aims to determine the association of HBsAg quantity with intrahepatic HBV viral load and activity in both tumor and non-neoplastic liver of HBV-HCC patients. Data were obtained from 89 prospectively enrolled patients treated with primary liver resection for HBV-HCC at a single Western institution (2008–2013). Circulating HBsAg was quantitated using ELISA. HBV DNA, covalently closed circular (cccDNA) and precore-pregenomic RNA (preC-pgRNA) in both tumor and non-neoplastic liver were quantitated by real-time PCR from fresh liver resection specimens. Circulating HBsAg was detectable in all 89 patients. HBsAg negatively correlated with age, and positively correlated with pre-operative serum AFP and ALT levels. HBsAg correlated with HBV cccDNA copy number in tumor or non-neoplastic liver tissue. It also correlated with preC-pgRNA copy number in non-neoplastic liver tissue. HBsAg did not correlate with serum HBV DNA, total intrahepatic HBV DNA, viral replicative activity or transcriptional activity. In HBV-HCC patients, HBsAg levels correlated with cccDNA copy number in tumor or non-neoplastic liver tissue, suggesting that a greater pool of cccDNA is associated with a higher rate of HBsAg production. J. Med. Virol. 9999: XX–XX, 2015.
hepatocellular carcinoma (HCC) [Fattovich et al., 2004; Bonilla Guerrero and Roberts, 2005; Liu et al., 2006; Liu et al., 2009; Lok and McMahon, 2007; El-Serag, 2011; Yang et al., 2011]. HBV viral products as well as HBV viral integration into the host genome may have direct oncogenic potential that contributes to hepatocarcinogenesis [Sung et al., 2012]. Besides HBV-specific mechanisms, chronic HBV infection also induces inflammation, liver cell injury and regeneration, and fibrosis, thereby creating an environment that may favor hepatocarcinogenesis [Arzumanyan et al., 2013]. Hepatitis B is a DNA virus that is surrounded by a host cell-derived envelope. HBsAg is embedded within the lipid bilayer of the viral envelope where it is present in three forms (small, medium, and large HBsAg), and is important for viral structure as well as viral entry into the host cell [Chan et al., 2011; Lee and Ahn, 2011; Yan et al., 2012]. The HBV genome is a relaxed circular DNA (rcDNA) that is partially double-stranded and about 3.2 kb in length [Seeger and Mason, 2000]. Once inside the host cell, the rcDNA is transported into the host cell nuclei, where the rcDNA is converted into covalently closed circular DNA (cccDNA) [Levrero et al., 2009]. The HBV genome contains four overlapping open reading frames (ORF) that encode all viral proteins, including HBsAg. Four mRNAs, each under the control of a separate promoter, are transcribed. Two mRNA species (PreS1 and PreS2/S mRNA) give rise to the
# 2015 Wiley Periodicals, Inc.
Conflict of interest: None. Correspondence to: Spiros P. Hiotis, MD, PhD, Division of Surgical Oncology, Department of Surgery, The Icahn School of Medicine at Mount Sinai, 5 East 98th Street, 12th Floor, Box 1259, New York City, NY 10029. E-mail: [email protected] Accepted 4 July 2015
KEY WORDS:
HBsAg; cccDNA; HCC; AFP
INTRODUCTION Hepatitis B surface antigen (HBsAg) is an important diagnostic marker for Hepatitis B (HBV) infection, one of the greatest risk factors for the development of C 2015 WILEY PERIODICALS, INC.
DOI 10.1002/jmv.24326 Published online in Wiley Online Library (wileyonlinelibrary.com).
2
Wang et al.
small, medium and large HBsAg, while the preCorepregenomic RNA (preC-pgRNA) gives rises to hepatitis B E antigen (HBeAg), core protein, polymerase and serves as the template for reverse transcription into rcDNA of a newly formed HBV viron. The viral DNA also integrates into host genome, causing instability of host cell genome and increased synthesis of viral proteins including HBsAg. Although antiviral therapy often leads to undetectable serum HBV DNA levels, HBV DNA is often detectable in the liver tissue of chronic HBV patients even in those with undetectable serum HBV DNA [Wang et al., 2013b]. However, liver biopsies in order to assess intrahepatic HBV-DNA levels are invasive and expensive; therefore, it is useful to determine serological markers for intrahepatic viral activity. In treatment-naiive patients, HBsAg changes during the natural history of HBV infection [Nguyen et al., 2010]. In addition, HBsAg correlates with cccDNA copy number, and the reduction of circulating HBsAg is thought to predict responses to interferon anti-viral therapy [Wursthorn et al., 2006; Lee et al., 2011; Matthews et al., 2013; Wang et al., 2013a]. Quantitative measurement of HBsAg is not well understood for patients with HBV-HCC. The goal of this study is to quantitatively measure HBsAg levels in HBV-HCC patients and to determine whether HBsAg correlates with intrahepatic viral DNA in tumor and non-neoplastic liver tissue.
116–9.89 108 copies/ml [Faria et al, 2008]. Information on other available biochemical markers including preoperative alpha-fetoprotein (AFP), platelet count, albumin, alanine aminotransferase (ALT), bilirubin, creatinine and INR was obtained from medical records. Following liver resection, all specimens were reviewed by a single dedicated liver pathologist who was blinded to all patient demographics and clinical outcomes. During dedicated research pathology analysis of specimens, the degree of hepatic fibrosis in non-neoplastic liver tissue and necro-inflammatory grade according histologic activity index (HAI) were scored according to the modified Ishak method [Goodman, 2007]. Vascular invasion and histologic differentiation were also assessed independently of the standard clinical pathology report, which was used only to obtain information on the number of tumors and the largest tumor diameter.
PATIENTS AND METHODS
Quantitative Intrahepatic HBV, DNA, and cccDNA Measurements
Patients, Histopathologic Assessment, and Follow-Up Data were obtained from selected HBV-HCC patients who underwent primary surgical resection at the Mount Sinai Medical Center in New York, NY, USA between 2008 and 2013. The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki and all aspects of research were approved by the Institutional review board prior to study initiation. Informed consent was obtained from every patient enrolled in the study. HBV-HCC patients were assessed pre-operatively by axial imaging (3-phase computerized tomography (CT) with intravenous contrast, or multi-phase magnetic resonance imaging (MRI) with intravenous contrast). Liver resection was performed on patients who had technicallyresectable disease and normal synthetic liver function (according to pre-operative bilirubin, albumin, and INR). Patients with portal hypertension (platelet count 200 ng/ml) is more likely to be detected in HCC patients with positive serum HBsAg [Peng et al., 2004]. HBV transcription, including the transcription of HBsAg, requires host cell transcription factors; therefore, it is possible that HBsAg and AFP share common transcription factors. Both the preS1 and preS2/S promoter and the AFP promoter have binding sites for transcription factors including hepatocyte nuclear factor 1 (HNF-1), nuclear factor 1 (NF-1), and CAAT/enhancer binding protein (C/EBP) [Bois-Joyeux and Danan, 1994; Crowe et al., 1999; Quasdorff and Protzer, 2010]. Moreover, enhanced HBV DNA integration events into host genome correlates with higher AFP expression [Sung et al., 2012], and may result in increased HBsAg production. Another viral protein, HBx, has also been shown to alleviate p53-mediated repression of AFP expression [Ogden et al., 2000], adding another layer of complexity to the possible interaction of HBV viral proteins and AFP expression. Because a role for AFP in HCC prognosis has been demonstrated in the literature [Personeni et al., 2012; Blank et al., 2013], future studies to elucidate the interaction of HBV and AFP will be important to advance further understanding of the prognostic capacity of AFP in HBVassociated HCC. Our previous studies have demonstrated that intrahepatic HBV DNA and viral replicative activity was associated with necro-inflammation and ALT, while
TABLE II. Correlations Between Circulating HBsAg (log10 IU/ml) and Intrahepatic HBV Viral Load and Activity in Tumor or Non-Neoplastic Liver Tissue (n ¼ 89) Correlation with HBsAg (log10 IU/ml)
Tumor Total HBV DNA (log10 copies/cell) cccDNA (log10 copies/cell) preC-pgRNA (log10 copies/cell) cccDNA/HBV DNA ratio pgRNA/cccDNA ratio Non-neoplastic liver tissue Total HBV DNA (log10 copies/cell) cccDNA (log10 copies/cell) preC-pgRNA (log10 copies/cell) cccDNA/HBV DNA ratio pgRNA/cccDNA ratio
N
Range
76 86 55 76 54
4.48–3.65 6.35–0.43 3.70–2.10 12.49–17.03 2.61–5.61
85 82 58 80 53
4.00–1.24 5.59 to 0.80 4.00–1.80 13.07–31.18 1.80–7.63
Pearson R 0.11 0.28 0.15 0.08 0.10 0.15 0.46 0.34 0.10 0.17
P-value 0.35 0.01 0.28 0.58 0.40 0.17
