Title: A Highly Sensitive and Robust Method for Hepatitis B Virus Covalently Closed Circular DNA Detection in Single Cells and Serum
Abstract: Despite implications of persistence of hepatitis B virus (HBV) covalently closed circular DNA (cccDNA) in the development of hepatocellular carcinoma (HCC), little is known about serum cccDNA in HBV-infected diseases. We developed a cccDNA-selective droplet digital PCR (ddPCR) to assess cccDNA content and dynamics across different stages of HCC development. One hundred forty-seven serum samples and 35 formalin-fixed, paraffin-embedded tumor tissues were derived from patients with HCC or HBV hepatitis/cirrhosis. After specific amplification and selective digestion, probe-based ddPCR was used to quantify cccDNA copy numbers in single cells and clinical samples. The cccDNA in single HepG2.2.15 cells ranged from 0 to 10.8 copies/cell. Compared with non-HCC patients, HCC patients showed a higher cccDNA-positive rate (89.9% versus 53.2%; P = 4.22 × 10−6) and increased serum cccDNA contents (P = 0.002 and P = 0.041 for hepatitis and cirrhosis patients, respectively). Serum cccDNA ranged from 84 to 1.07 × 105 copies/mL. Quantification of serum cccDNA and HBV-DNA was an effective way to discriminate HCC patients from non-HCC patients, with areas under the curve of receiver operating characteristic of 0.847 (95% CI, 0.759–0.935; sensitivity, 74.5%; specificity, 93.7%). cccDNA-selective ddPCR is sensitive to detect cccDNA in single cells and different clinical samples. Combined analysis of serum cccDNA and HBV-DNA may be a promising strategy for HBV-induced HCC surveillance and antiviral therapy evaluation. Despite implications of persistence of hepatitis B virus (HBV) covalently closed circular DNA (cccDNA) in the development of hepatocellular carcinoma (HCC), little is known about serum cccDNA in HBV-infected diseases. We developed a cccDNA-selective droplet digital PCR (ddPCR) to assess cccDNA content and dynamics across different stages of HCC development. One hundred forty-seven serum samples and 35 formalin-fixed, paraffin-embedded tumor tissues were derived from patients with HCC or HBV hepatitis/cirrhosis. After specific amplification and selective digestion, probe-based ddPCR was used to quantify cccDNA copy numbers in single cells and clinical samples. The cccDNA in single HepG2.2.15 cells ranged from 0 to 10.8 copies/cell. Compared with non-HCC patients, HCC patients showed a higher cccDNA-positive rate (89.9% versus 53.2%; P = 4.22 × 10−6) and increased serum cccDNA contents (P = 0.002 and P = 0.041 for hepatitis and cirrhosis patients, respectively). Serum cccDNA ranged from 84 to 1.07 × 105 copies/mL. Quantification of serum cccDNA and HBV-DNA was an effective way to discriminate HCC patients from non-HCC patients, with areas under the curve of receiver operating characteristic of 0.847 (95% CI, 0.759–0.935; sensitivity, 74.5%; specificity, 93.7%). cccDNA-selective ddPCR is sensitive to detect cccDNA in single cells and different clinical samples. Combined analysis of serum cccDNA and HBV-DNA may be a promising strategy for HBV-induced HCC surveillance and antiviral therapy evaluation. CME Accreditation Statement: This activity ("JMD 2018 CME Program in Molecular Diagnostics") has been planned and implemented in accordance with the accreditation requirements and policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint providership of the American Society for Clinical Pathology (ASCP) and the American Society for Investigative Pathology (ASIP). ASCP is accredited by the ACCME to provide continuing medical education for physicians.The ASCP designates this journal-based CME activity ("JMD 2018 CME Program in Molecular Diagnostics") for a maximum of 18.0 AMA PRA Category 1 Credit(s)™. Physicians should claim only credit commensurate with the extent of their participation in the activity.CME Disclosures: The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. CME Accreditation Statement: This activity ("JMD 2018 CME Program in Molecular Diagnostics") has been planned and implemented in accordance with the accreditation requirements and policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint providership of the American Society for Clinical Pathology (ASCP) and the American Society for Investigative Pathology (ASIP). ASCP is accredited by the ACCME to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity ("JMD 2018 CME Program in Molecular Diagnostics") for a maximum of 18.0 AMA PRA Category 1 Credit(s)™. Physicians should claim only credit commensurate with the extent of their participation in the activity. CME Disclosures: The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Hepatitis B virus (HBV) infection has led to >686,000 deaths worldwide per year.1MacLachlan J.H. Locarnini S. Cowie B.C. Estimating the global prevalence of hepatitis B.Lancet. 2015; 386: 1515-1517Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar Approximately 2 billion individuals worldwide are infected with HBV and are at risk of developing cirrhosis and hepatocellular carcinoma (HCC),2Trepo C. Chan H.L. Lok A. Hepatitis B virus infection.Lancet. 2014; 384: 2053-2063Abstract Full Text Full Text PDF PubMed Scopus (1043) Google Scholar which ranks fifth in terms of malignant cancer mortality. The global prevalence of chronic hepatitis B (CHB) is distributed unevenly and is most concentrated in Africa and southeast Asia.2Trepo C. Chan H.L. Lok A. Hepatitis B virus infection.Lancet. 2014; 384: 2053-2063Abstract Full Text Full Text PDF PubMed Scopus (1043) Google Scholar Currently, clinical therapies of CHB are still limited to pegylated interferon-α or the five approved nucleos(t)ide analog (NUC) treatments. However, only 10% of patients achieve hepatitis B s antigen (HBsAg) loss by pegylated interferon-α and/or NUC after a follow-up of 5 years.3Zoulim F. Durantel D. Antiviral therapies and prospects for a cure of chronic hepatitis B.Cold Spring Harb Perspect Med. 2015; 5: a021501Crossref PubMed Scopus (123) Google Scholar Increasing evidence shows that long-term NUC administration for CHB patients is mainly attributable to the persistence of viral covalently closed circular DNA (cccDNA) and defective immune responses.4Nassal M. HBV cccDNA: viral persistence reservoir and key obstacle for a cure of chronic hepatitis B.Gut. 2015; 64: 1972-1984Crossref PubMed Scopus (541) Google Scholar, 5Durantel D. Zoulim F. New antiviral targets for innovative treatment concepts for hepatitis B virus and hepatitis delta virus.J Hepatol. 2016; 64: S117-S131Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar The cccDNA is formed during the HBV life cycle, which serves as a template for the transcription of HBV RNAs and causes progeny viruses.6Lucifora J. Protzer U. Attacking hepatitis B virus cccDNA: the holy grail to hepatitis B cure.J Hepatol. 2016; 64: S41-S48Abstract Full Text Full Text PDF PubMed Scopus (129) Google Scholar Therefore, physical elimination of cccDNA in hepatocytes is regarded as a crucial step in developing a cure for HBV. In 2005, a study suggested that long-term lamivudine treatment could reduce serum cccDNA in CHB patients.7Yuen M.F. Wong D.K. Sum S.S. Yuan H.J. Yuen J.C. Chan A.O. Wong B.C. Lai C.L. Effect of lamivudine therapy on the serum covalently closed-circular (ccc) DNA of chronic hepatitis B infection.Am J Gastroenterol. 2005; 100: 1099-1103Crossref PubMed Scopus (41) Google Scholar Recently, several new antiviral strategies targeting cccDNA have been established to improve HBV clearance.5Durantel D. Zoulim F. New antiviral targets for innovative treatment concepts for hepatitis B virus and hepatitis delta virus.J Hepatol. 2016; 64: S117-S131Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar For instance, RNA interference has been widely used to inhibit the replication of HBV in vitro; three siRNA molecules are under clinical trial, including ARC-520, TKM-HBV, and ALN-HBV.8Gish R.G. Yuen M.F. Chan H.L. Given B.D. Lai C.L. Locarnini S.A. Lau J.Y. Wooddell C.I. Schluep T. Lewis D.L. Synthetic RNAi triggers and their use in chronic hepatitis B therapies with curative intent.Antiviral Res. 2015; 121: 97-108Crossref PubMed Scopus (114) Google Scholar In hepatitis B e antigen (HBeAg)–positive chimpanzees, NUC administration decreased cccDNA by 0.7 ± 0.6 log10; additional treatment with ARC-520 resulted in a reduction of cccDNA levels to 1.4 ± 0.7 log10 compared with the baseline. Conversely, cccDNA levels were not reduced remarkably in HBeAg-negative animals.5Durantel D. Zoulim F. New antiviral targets for innovative treatment concepts for hepatitis B virus and hepatitis delta virus.J Hepatol. 2016; 64: S117-S131Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar High-dose interferon-α or lymphotoxin-β receptor agonist activation could degrade cccDNA in a nonhepatotoxic way via up-regulated apolipoprotein B mRNA editing enzyme, catalytic polypeptide (APOBEC) 3A and APOBEC3B cytidine deaminases, respectively.9Lucifora J. Xia Y. Reisinger F. Zhang K. Stadler D. Cheng X. Sprinzl M.F. Koppensteiner H. Makowska Z. Volz T. Remouchamps C. Chou W.M. Thasler W.E. Huser N. Durantel D. Liang T.J. Munk C. Heim M.H. Browning J.L. Dejardin E. Dandri M. Schindler M. Heikenwalder M. Protzer U. Specific and nonhepatotoxic degradation of nuclear hepatitis B virus cccDNA.Science. 2014; 343: 1221-1228Crossref PubMed Scopus (663) Google Scholar In addition, interferon-γ and tumor necrosis factor α could decay cccDNA in a noncytolytic way via activation of nuclear APOBEC3 deaminases.10Xia Y. Stadler D. Lucifora J. Reisinger F. Webb D. Hosel M. Michler T. Wisskirchen K. Cheng X. Zhang K. Chou W.M. Wettengel J.M. Malo A. Bohne F. Hoffmann D. Eyer F. Thimme R. Falk C.S. Thasler W.E. Heikenwalder M. Protzer U. Interferon-gamma and tumor necrosis factor-alpha produced by T cells reduce the HBV persistence form, cccDNA, without cytolysis.Gastroenterology. 2016; 150: 194-205Abstract Full Text Full Text PDF PubMed Scopus (225) Google Scholar Recently, zinc-finger nucleases, the RNA-guided clustered regularly interspaced short palindromic repeats/Cas system, and transcription activator-like endonucleases specifically targeted to the HBV cccDNA are being engineered.11Dong C. Qu L. Wang H. Wei L. Dong Y. Xiong S. Targeting hepatitis B virus cccDNA by CRISPR/Cas9 nuclease efficiently inhibits viral replication.Antiviral Res. 2015; 118: 110-117Crossref PubMed Scopus (186) Google Scholar With the increasing development of innovative strategies, a new method for evaluating the efficiency of cccDNA clearance in a sensitive and specific manner is urgently required. Indeed, the infectious HBV virions contain only partially double-stranded DNA [alias relaxed circular DNA (rcDNA)], deriving from reverse transcription of a pregenomic RNA.12Seeger C. Mason W.S. Molecular biology of hepatitis B virus infection.Virology. 2015; 479-480: 672-686Crossref PubMed Scopus (535) Google Scholar The structure of rcDNA is rather similar to cccDNA, which presents a great challenge to distinguish cccDNA from rcDNA. Besides, the low copy number of cccDNA generates another obstacle for cccDNA detection. HBV-infected ducks had, on average, 10 cccDNA copies per hepatocyte, and the cccDNA in CHB patients seems to be much lower, averaging 0.1 to 1 copy per hepatocyte.13Volz T. Allweiss L. Ben M.M. Warlich M. Lohse A.W. Pollok J.M. Alexandrov A. Urban S. Petersen J. Lutgehetmann M. Dandri M. The entry inhibitor Myrcludex-B efficiently blocks intrahepatic virus spreading in humanized mice previously infected with hepatitis B virus.J Hepatol. 2013; 58: 861-867Abstract Full Text Full Text PDF PubMed Scopus (255) Google Scholar, 14Volz T. Lutgehetmann M. Wachtler P. Jacob A. Quaas A. Murray J.M. Dandri M. Petersen J. Impaired intrahepatic hepatitis B virus productivity contributes to low viremia in most HBeAg-negative patients.Gastroenterology. 2007; 133: 843-852Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar, 15Werle-Lapostolle B. Bowden S. Locarnini S. Wursthorn K. Petersen J. Lau G. Trepo C. Marcellin P. Goodman Z. Delaney W.E. Xiong S. Brosgart C.L. Chen S.S. Gibbs C.S. Zoulim F. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy.Gastroenterology. 2004; 126: 1750-1758Abstract Full Text Full Text PDF PubMed Scopus (762) Google Scholar Southern blotting can detect cccDNA by a few picograms of virus DNA per lane with 32P-labeled probes. The limit of detection (LOD) is approximately 10 cccDNA copies per infected/transfected cell,4Nassal M. HBV cccDNA: viral persistence reservoir and key obstacle for a cure of chronic hepatitis B.Gut. 2015; 64: 1972-1984Crossref PubMed Scopus (541) Google Scholar suggesting this gold standard method is not sensitive enough to detect low copy numbers of cccDNA.15Werle-Lapostolle B. Bowden S. Locarnini S. Wursthorn K. Petersen J. Lau G. Trepo C. Marcellin P. Goodman Z. Delaney W.E. Xiong S. Brosgart C.L. Chen S.S. Gibbs C.S. Zoulim F. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy.Gastroenterology. 2004; 126: 1750-1758Abstract Full Text Full Text PDF PubMed Scopus (762) Google Scholar Rolling circle amplification is a potential alternative to increase the sensitivity of cccDNA detection,16Margeridon S. Carrouee-Durantel S. Chemin I. Barraud L. Zoulim F. Trepo C. Kay A. Rolling circle amplification, a powerful tool for genetic and functional studies of complete hepatitis B virus genomes from low-level infections and for directly probing covalently closed circular DNA.Antimicrob Agents Chemother. 2008; 52: 3068-3073Crossref PubMed Scopus (51) Google Scholar, 17Huang J.T. Liu Y.J. Wang J. Xu Z.G. Yang Y. Shen F. Liu X.H. Zhou X. Liu S.M. Next generation digital PCR measurement of hepatitis B virus copy number in formalin-fixed paraffin-embedded hepatocellular carcinoma tissue.Clin Chem. 2015; 61: 290-296Crossref PubMed Scopus (51) Google Scholar but this has not been widely accepted.6Lucifora J. Protzer U. Attacking hepatitis B virus cccDNA: the holy grail to hepatitis B cure.J Hepatol. 2016; 64: S41-S48Abstract Full Text Full Text PDF PubMed Scopus (129) Google Scholar, 18Rybicka M. Woziwodzka A. Romanowski T. Stalke P. Dreczewski M. Bielawski K.P. Differences in sequences between HBV-relaxed circular DNA and covalently closed circular DNA.Emerg Microbes Infect. 2017; 6: e55Crossref PubMed Scopus (6) Google Scholar To increase the specificity, a cccDNA PCR, using specific primers and probes to selectively amplify cccDNA and exclude rcDNA, was developed. This is achieved as a result of using primers spanning the nick of rcDNA and hybridizing the probe to its gap region.19Kock J. Schlicht H.J. Analysis of the earliest steps of hepadnavirus replication: genome repair after infectious entry into hepatocytes does not depend on viral polymerase activity.J Virol. 1993; 67: 4867-4874PubMed Google Scholar Reducing rcDNA copy number is another way to increase the accuracy of cccDNA detection. Pretreatment of HBV-DNA with Plasmid-Safe ATP-Dependent DNase (PSAD) (Epicentre, Madison, WI)15Werle-Lapostolle B. Bowden S. Locarnini S. Wursthorn K. Petersen J. Lau G. Trepo C. Marcellin P. Goodman Z. Delaney W.E. Xiong S. Brosgart C.L. Chen S.S. Gibbs C.S. Zoulim F. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy.Gastroenterology. 2004; 126: 1750-1758Abstract Full Text Full Text PDF PubMed Scopus (762) Google Scholar or T5 exonuclease (NEB, Beverly, MA)18Rybicka M. Woziwodzka A. Romanowski T. Stalke P. Dreczewski M. Bielawski K.P. Differences in sequences between HBV-relaxed circular DNA and covalently closed circular DNA.Emerg Microbes Infect. 2017; 6: e55Crossref PubMed Scopus (6) Google Scholar to degrade DNA with free ends can improve discriminatory power up to 1000-fold.15Werle-Lapostolle B. Bowden S. Locarnini S. Wursthorn K. Petersen J. Lau G. Trepo C. Marcellin P. Goodman Z. Delaney W.E. Xiong S. Brosgart C.L. Chen S.S. Gibbs C.S. Zoulim F. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy.Gastroenterology. 2004; 126: 1750-1758Abstract Full Text Full Text PDF PubMed Scopus (762) Google Scholar Although cccDNA-selective PCR provides a relative discrimination between cccDNA and rcDNA, additional means are still required to reduce the false-positive amplification of rcDNA. In recent years, droplet digital PCR (ddPCR) has shown great potential in laboratory research and clinical utility owing to its high sensitivity and accuracy in detecting trace molecules,20Hindson C.M. Chevillet J.R. Briggs H.A. Gallichotte E.N. Ruf I.K. Hindson B.J. Vessella R.L. Tewari M. Absolute quantification by droplet digital PCR versus analog real-time PCR.Nat Methods. 2013; 10: 1003-1005Crossref PubMed Scopus (941) Google Scholar such as viruses, circulating tumor DNA, and single-nucleotide polymorphisms.21Ruelle J. Yfantis V. Duquenne A. Goubau P. Validation of an ultrasensitive digital droplet PCR assay for HIV-2 plasma RNA quantification.J Int AIDS Soc. 2014; 17: 19675Crossref PubMed Google Scholar, 22Ayturk U.M. Couto J.A. Hann S. Mulliken J.B. Williams K.L. Huang A.Y. Fishman S.J. Boyd T.K. Kozakewich H.P. Bischoff J. Greene A.K. Warman M.L. Somatic activating mutations in GNAQ and GNA11 are associated with congenital hemangioma.Am J Hum Genet. 2016; 98: 789-795Abstract Full Text Full Text PDF PubMed Scopus (113) Google Scholar, 23Dahmcke C.M. Steven K.E. Larsen L.K. Poulsen A.L. Abdul-Al A. Dahl C. Guldberg P. A prospective blinded evaluation of urine-DNA testing for detection of urothelial bladder carcinoma in patients with gross hematuria.Eur Urol. 2016; 70: 916-919Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar, 24Wan J.C. Massie C. Garcia-Corbacho J. Mouliere F. Brenton J.D. Caldas C. Pacey S. Baird R. Rosenfeld N. Liquid biopsies come of age: towards implementation of circulating tumour DNA.Nat Rev Cancer. 2017; 17: 223-238Crossref PubMed Scopus (1319) Google Scholar Nevertheless, little is known of serum cccDNA and its clinical significance in patients with CHB, cirrhosis, and HCC. ddPCR and selective PCR were combined to detect cccDNA in serum, single cells, and formalin-fixed, paraffin-embedded (FFPE) tumor tissues, and the association of cccDNA copy numbers with patients' clinical features was also investigated. A total of 168 patients with HBV infection (HBsAg/HBV-DNA positive) were enrolled in this study between December 2013 and November 2015 at the Zhongnan Hospital of Wuhan University (Wuhan, China), including 79 cases of non-HCC patients [56 without and 23 with cirrhosis; 50 males and 29 females; mean (SD) age, 41.7 (12.8) years] and 89 cases of HCC patients [75 males and 14 females; mean (SD) age, 53.4 (15.3) years]. HCC patients were confirmed by pathology, without other cancers. Informed consent was obtained from each participant. The ethics committee of Zhongnan Hospital, Wuhan University, has approved this study. In total, 79 serum samples were collected from non-HCC patients, 68 preoperative serum samples on the first day after hospital admission and 14 paired FFPE tumor tissues, as well as 21 nonpaired FFPE tumor tissues from HCC patients. For non-HCC patients, 92.4% (73/79) had received NUC treatment, 5.1% (4/79) were treated with pegylated interferon-α, and 2.5% (2/79) were first diagnosed, without any antiviral treatment. Most HCC patients [78/89 (87.6%)] had received NUC treatment, and 12.3% (11/89) HCC patients did not receive antiviral therapy. Serum HBV-DNA copy number was assayed with COBAS TaqMan 48 kit (Roche Diagnostics, Basel, Switzerland). The serological HBsAg, hepatitis B s antibody (HBsAb), HBeAg, hepatitis B e antibody (HBeAb), hepatitis B c antibody (HBcAb), and anti-HCV were detected by Architect chemiluminescent enzyme immunoassays (Abbott Architect i system; Abbott Diagnostics, Lake Bluff, IL). Clinical biochemical parameters and tumor biomarkers were measured by an automatic chemistry analyzer (Abbott-Aeroset; Abbott Diagnostics) with commercial kits, including aspartate aminotransferase (AST), alanine aminotransferase (ALT), γ-glutamyl-transpeptidase, alkaline phosphatase, cholinesterase, total bilirubin, direct bilirubin, indirect bilirubin, total biliary acid, total protein, albumin, globulin, glucose, blood urea nitrogen, creatinine, uric acid, cystatin C, β2-microglobulin (β2-MG), carcinoembryonic antigen, and α-fetoprotein (AFP), together with lipid profile (triglyceride, total cholesterol, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol). DNA was extracted with commercial kits, TIANamp Virus DNA/RNA Kit (TIANGEN, Beijing, China) for fresh serum, DEXPAT Easy kit (Takara, Dalian, China) and AxyPrep DNA Gel Extraction Kit (Axygen, Wujiang, China) for FFPE tumor tissues, and TIANamp Genomic DNA Kit (TIANGEN) for fresh tumor tissues and cultured cells. Human cell lines HepG2 and HepG2.2.15 were cultured in Dulbecco's modified Eagle's medium (Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum and 1× penicillin-streptomycin (Gibco); they were maintained in a humidified incubator with 5% CO2 at 37°C. To get single HepG2.2.15 cells, cell pellets were resuspended in phosphate-buffered saline at 105 cells/mL after washing twice with 1× phosphate-buffered saline. Single cells were individually micromanipulated using a 10-μm transfer tip on a Transferman NK2 micromanipulator (Eppendorf, Hamburg, Germany). After the captured cells were injected into PCR vials, and incubated at 90°C for 10 minutes for lysis, 1 μL of proteinase K (20 mg/mL) was added and incubated at 58°C for another 1 hour for protein digestion and at 95°C for 15 minutes for DNA denaturing as well as proteinase K inactivation. The supernatant served as DNA template for ddPCR (Figure 1A). HBV cccDNA was amplified from HepG2.2.15 cell line DNA with the cccDNA-selective primers [forward, 5′-GGGGCGCACCTCTCTTTA-3′ (position 1523 to 1540); reverse, 5′-AGGCACAGCTTGGAGGC-3′ (position 1886 to 1870)]; then, it was inserted into pMD18-T vector to obtain a cccDNA-positive control. To increase the sensitivity and accuracy of cccDNA detection, PSAD (Epicentre) was used to eliminate rcDNA, single-stranded DNA, and double-stranded DNA. cccDNA copy number was quantified using the QX200 Droplet Digital PCR system (Bio-Rad, Hercules, CA). Briefly, 20 μL ddPCR mixture consisted of 10 μL of 2× ddPCR supermix for probes (Bio-Rad), 950 nmol/L of cccDNA-selective primers, 250 nmol/L of cccDNA probe (5′-FAM-TCACCTCTGCCTAATCATCTC-TAMRA-3′), and 5.9 μL of DNA template. The mixture was loaded into the DG8 cartridge, 70 μL of droplet generation oil was added, and droplets (approximately 1 nL per droplet) were then formed in the droplet generator (Bio-Rad). Generally, each sample could generate up to 20,000 stable water-in-oil droplets. Next, the droplets were transferred to a 96-well PCR plate (Eppendorf) and amplified on a C1000 thermal cycler (Bio-Rad) with a thermal profile beginning at 95°C for 5 minutes, followed by 45 cycles of 94°C for 30 seconds and 60°C for 60 seconds, 1 cycle of 98°C for 10 minutes, and ending at 4°C. After amplification, the plate was loaded on the droplet reader (Bio-Rad), and data were analyzed by QuantaSoft analysis software version 1.7.4 (Bio-Rad) on the basis of Poisson distribution (Figure 1A). For cell counting and normalization, β-actin copy number in each DNA sample without PSAD treatment was also simultaneously determined with a pair of primers (5′-ACTGTGCCCATCTACGAGG-3′ and 5′-CAGGCAGCTCGTAGCTCTT-3′) and a probe (5′-FAM-CGGGAAATCGTGCGTGAC-TAMRA-3′) in neighboring wells. All data were analyzed with SPSS version 19.0 (IBM, Armonk, NY). t-Test and U test were used to analyze the mean difference for normally distributed data and skewed data, respectively, and the results were presented as means (SEM) or median (interquartile range). Spearman rank-order correlation was used to investigate the association between the cccDNA copy number and clinical parameters. All statistical tests were two sided, and the level of statistical significance was set at P < 0.05. The serologic tests showed that 46 of 168 patients (27.3%) were HBeAg positive, 65 of 168 patients (38.7%) were HBeAb positive, no patients were infected with hepatitis C virus, and 7 of 168 patients (6.4%) had hepatitis A virus infection. The HBeAg-positive HCC patients [9/89 (10.1%)] were fewer than non-HCC patients [37/79 (46.8%)]. Among 168 patients, 122 had Child-Pugh A liver function, 27 had Child-Pugh B liver function, and 19 had Child-Pugh C liver function. In non-HCC patients, HBV-DNA copy number was 1.95 × 106 (5.48 × 104 to 2.31 × 107) copies/mL and cccDNA copy number was 169.5 (0 to 491.5) copies/mL. In HCC patients, HBV-DNA copy number was 1.17 × 104 (1.01 × 103 to 1.29 × 105) copies/mL and cccDNA copy number was 423.7 (224.6 to 792.4) copies/mL. Patients' demographic and clinical pathologic information are presented in Tables 1 and 2.Table 1Clinical Characteristics of Study SubjectsVariableHepatitis and cirrhosis patients (n = 79)Hepatocellular carcinoma patients (n = 89)Age, means±SD, years41.7 ± 12.853.4 ± 15.3∗P < 0.05.Males/females, n (%)50 (63.3)/29 (36.7)75 (84.3)/14 (15.7)∗P < 0.05.Etiology of liver disease, n (%)HBV79 (100)89 (100) HBsAg positive72 (91.1)81 (91.0) HBV-DNA positive65 (82.3)57 (64.0) HBeAg positive37 (46.8)9 (10.1)∗P < 0.05. HBeAb positive31 (39.2)34 (38.2) HBV + HAV7 (8.9)0 (0.0) HCV0 (0.0)0 (0.0)Stage of cirrhosis (Child-Pugh class), n (%) A43 (54.4)79 (88.8)∗P < 0.05. B17 (21.5)10 (11.2) C19 (24.1)0 (0.0)∗P < 0.05.Difference between two groups was tested using the χ2 test for proportions and unpaired t-test for continuous variables.HAV, hepatitis A virus; HBV, hepatitis B virus; HBeAb, hepatitis B e antibody; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B s antigen; HCV, hepatitis C virus.∗ P < 0.05. Open table in a new tab Table 2Serum Characteristics of the Study PatientsParameterHepatitis and cirrhosis patients (n = 79)Hepatocellular carcinoma patients (n = 68)P valueHBV-DNA copy number, copies/mL∗Nonnormally distributed data are expressed as median (interquartile range).1.9 × 106 (5.4 × 104–2.3 × 107)1.2 × 104 (1.0 × 103–1.3 × 105)1.1 × 10−8†P < 0.05.cccDNA copy number, copies/mL∗Nonnormally distributed data are expressed as median (interquartile range).169.5 (0–491.5)423.7 (224.6–792.4)0.001†P < 0.05.ALT, U/L∗Nonnormally distributed data are expressed as median (interquartile range).136.0 (42.0–421.0)38.1 (22.9–49.5)1.13 × 10−8†P < 0.05.AST, U/L∗Nonnormally distributed data are expressed as median (interquartile range).108.0 (43.5–281.0)40.9 (30.1–72.7)1.37 × 10−5†P < 0.05.γGT, U/L∗Nonnormally distributed data are expressed as median (interquartile range).66.0 (35.4–137.0)75.5 (33.1–183.3)0.250ALP, U/L∗Nonnormally distributed data are expressed as median (interquartile range).109.0 (74.0–143.0)110.0 (80.3–182.0)0.091Cholinesterase, U/L‡Normally distributed data are described as means (SEM).5544.9 (346.7)5973.9 (279.7)0.332Total bilirubin, μmol/L∗Nonnormally distributed data are expressed as median (interquartile range).30.2 (18.8–168.5)17.7 (13.3–25.6)3.42 × 10−7†P < 0.05.Direct bilirubin, μmol/L∗Nonnormally distributed data are expressed as median (interquartile range).8.9 (4.5–81.7)7.3 (5.9–11.7)0.132Indirect bilirubin, μmol/L∗Nonnormally distributed data are expressed as median (interquartile range).19.2 (14.4–66.0)10.4 (7.0–13.8)1.23 × 10−6†P < 0.05.Total bile acid, μmol/L∗Nonnormally distributed data are expressed as median (interquartile range).37.6 (8.1–182.6)11.1 (5.3–27.9)0.002†P < 0.05.Total proteins, g/L‡Normally distributed data are described as means (SEM).65.4 (1.0)66.6 (0.8)0.330Albumin, g/L‡Normally distributed data are described as means (SEM).36.3 (0.8)38.7 (0.7)0.028†P < 0.05.Globin, g/L‡Normally distributed data are described as means (SEM).29.0 (0.7)27.9 (0.7)0.296Glucose, mmol/L∗Nonnormally distributed data are expressed as median (interquartile range).4.7 (4.1–5.6)5.0 (4.4–5.7)0.252BUN, mmol/L∗Nonnormally distributed data are expressed as median (interquartile range).4.1 (3.3–5.2)5.1 (4.3–6.0)2.33 × 10−4†P < 0.05.Creatinine, μmol/L∗Nonnormally distributed data are expressed as median (interquartile range).60.8 (50.6–77.6)63.2 (56.0–71.8)0.45Uric acid, μmol/L‡Normally distributed data are described as means (SEM).245.9 (11.5)316.3 (9.9)8.83 × 10−6†P < 0.05.Cystatin C, mg/L∗Nonnormally distributed data are expressed as median (interquartile range).0.9 (0.8–1.2)0.6 (0.5–0.7)8.47 × 10−10†P < 0.05.β2-Microglobulin, μg/L‡Normally distributed data are described as means (SEM).2512.6 (172.2)2054.6 (536.3)0.493Total cholesterol, mmol/L‡Normally distributed data are described as means (SEM).3.6 (0.18)3.9 (0.15)0.125Triglycerides, mmol/L‡Normally distributed data are described as means (SEM).1.1 (0.12)1.3 (0.16)0.328HDL cholesterol, mmol/L‡Normally distributed data are described as means (SEM).0.9 (0.11)1.1 (0.06)0.072LDL cholesterol, mmol/L‡Normally distributed data are described as means (SEM).1.9 (0.15)2.2 (0.12)0.101AFP, ng/mL∗Nonnormally distributed data are expressed as median (interquartile range).30.7 (4.6–143.5)81.4 (4.8–227.7)0.40CEA, ng/mL‡Normally distributed data are described as means (SEM).2.9 (0.3)1.9 (0.2)0.002†P < 0.05.P values of normally distributed data and nonnormally distributed data were derived from unpaired t-test and unpaired U test, respectively.γGT, γ-glutamyl-tran