HBV Cccdna: the Stumbling Block for Treatment of HBV Infection Shousheng Liu and Yongning Xin

JCTH Associate Editors Dr. Masahiro Arai Prof. Junqi Niu OWNED BY THE SECOND AFFILIATED HOSPITAL Toshiba General Hospital Jilin University OF CHONGQING MEDICAL UNIVERSITY Tokyo, Japan Changchun, China Dr. Roopjeet Bath Dr. Kittichai Promrat University of Connecticut Alpert Medical School of Brown University Farmington, USA Providence, USA Dr. Timothy Billiar Dr. Farzin Roohvand Editors-in-Chief Presbyterian University Hospital Pasteur Institute of Iran Prof. Hong Ren Pittsburgh, USA Tehran, Iran General Editor-in-Chief Dr. John Birk Dr. Arielle Rosenberg The Second Affiliated Hospital of Chongqing University of Connecticut University Paris Descartes Medical University, China Farmington, USA Paris, France Dr. Harry Hua-Xiang Xia Dr. Wendy Cao Dr. Michael Schilsky Editor-in-Chief New York University School of Medicine Yale University Novartis Pharmaceuticals Corporation, USA New York, USA New Haven, USA Prof. George Y. Wu Prof. Limin Chen Dr. Ashwani Singal Comprehensive Editor-in-Chief Peking Union Medical College University of Alabama at Birmingham University of Connecticut Heath Center, USA Chengdu, China Birmingham, USA Prof. Aziz A. Chentoufi Dr. Tawesak Tanwandee Medicine King Fahad Medical College Siriraj Hospital, Mahidol University Managing Editors Riyadh, Saudi Arabia Bangkok, Thailand Huaidong Hu Prof. Mohamed A Daw Dr. Fusheng Wang Chongqing, China University of Tripoli 302 Miltitary Hospital of China Beijing, China Sandeep Kumar Karn Tripoli, Libya Chongqing, China Prof. Xiaoguang Dou Prof. Lai Wei Shengjing Hospital of China Medical Peking University People’s Hospital University Beijing, China Executive Editor Shenyang, China Prof. Catherine Wu Hua He Prof. Marko Duvnjak University of Connecticut Houston, USA Clinical Hospital Centre “Sestre milosrdnice” Farmington, USA Zagreb, Croatia Dr. Xuefeng Xia Prof. Jinlin Hou The Methodist Hospital Research Institute Technical Editor Nanfang Hospital of Southern Medical Houston, USA Wanchen Hu University Guangzhou, China Prof. Manfung Yuen Wuhan, China Prof. Keqin Hu The University of Hong Kong Queen Mary University of California Hospital Orange, USA Hong Kong, China Prof. Jidong Jia Prof. Dazhi Zhang Capital Medical University The Second Affiliated Hospital of Chongqing Beijing, China Medical University Chongqing, China Prof. Douglas LaBrecque University of Iowa Prof. Lanjing Zhang Contact information Lowa, USA University Medical Center of Princeton Plainsboro, USA Telephone +86 23 63727251 Dr. Joseph Lim Yale University Fax +86 23 63701383 New Haven, USA E-mail [email protected] Prof. Lungen Lu Shanghai Jiaotong University School of Address 74 Linjiang Road, Medicine Yuzhong District, Shanghai, China Chongqing, P. R. China, Prof. Yen-Hsuan Ni 400010 National Taiwan University Taipei, Taiwan

Publisher Xia & He Publishing Inc. 14090 Southwest Freeway, Suite 300, Sugar Land, Texas 77478, USA JOURNAL OF CLINICAL AND TRANSLATIONAL HEPATOLOGY

Beginning with 2018, JCTH has been indexed and abstracted in Emerging Sources Citation Index (ESCI).

CONTENTS 2019 7(3):195–283

Editorial

HBV cccDNA: The Stumbling Block for Treatment of HBV Infection Shousheng Liu and Yongning Xin ...... 195

Original Articles

Comparing the Efficacy and Safety of Treating Chronic Hepatitis B Infection during Pregnancy with Lamivudine, Telbivudine, and Tenofovir: A Meta-analysis Shahnaz Sali, Mohammad Darvishi, Mojtaba GhasemiAdl, Meisam Akhlaghdoust, Azin Mirzazadeh, Somayeh Elikaei Behjati, Hossein Sheikh-Zeinolabedini, Shervin Shokouhi and Soheil Tavakolpour...... 197

Efficacy and Safety of All-oral, 12-week Ravidasvir Plus Ritonavir-boosted Danoprevir and Ribavirin in Treatment-na€ıve Noncirrhotic HCV Genotype 1 Patients: Results from a Phase 2/3 Clinical Trial in China Xiaoyuan Xu, Bo Feng, Yujuan Guan, Sujun Zheng, Jifang Sheng, Xingxiang Yang, Yuanji Ma, Yan Huang, Yi Kang, Xiaofeng Wen, Jun Li, Youwen Tan, Qing He, Qing Xie, Maorong Wang, Ping An, Guozhong Gong, Huimin Liu, Qin Ning, Rui Hua, Bo Ning, Wen Xie, Jiming Zhang, Wenxiang Huang, Yongfeng Yang, Minghua Lin, Yingren Zhao, Yanhong Yu, Jidong Jia, Dongliang Yang, Liang Chen, Yinong Ye, Yuemin Nan, Zuojiong Gong, Quan Zhang, Peng Hu, Fusheng Wang, Yongguo Li, Dongliang Li, Zhansheng Jia, Jinlin Hou, Chengwei Chen, Jinzi J. Wu and Lai Wei ...... 213

Efficacy and Safety of 12-week Interferon-based Danoprevir Regimen in Patients with Genotype 1 Chronic Hepatitis C Lai Wei, Jia Shang, Yuanji Ma, Xiaoyuan Xu, Yan Huang, Yujuan Guan, Zhongping Duan, Wenhong Zhang, Zhiliang Gao, Mingxiang Zhang, Jun Li, Jidong Jia, Yongfeng Yang, Xiaofeng Wen, Maorong Wang, Zhansheng Jia, Bo Ning, Yongping Chen, Yue Qi, Jie Du, Jianning Jiang, Lixin Tong, Yao Xie and Jinzi J. Wu ...... 221

Hepatitis C Screening: Barriers to Linkage to Care Sammy Saab, Youssef P. Challita, Lisa M. Najarian, Rong Guo, Satvir S. Saggi and Gina Choi ...... 226

TIPS Is Not Associated with a Higher Risk of Developing HCC in Cirrhotic Patients: A Systematic Review and Meta-analysis Bin Chen, Long Pang, Hao-Bin Chen, Dong-Bo Wu, Yong-Hong Wang and En-Qiang Chen ...... 232 Effect of Autologous Bone Marrow Stem Cell Therapy in Patients with Liver Cirrhosis: A Meta-analysis Chuan-Xin Wu, Deng Wang, Ying Cai, Ao-Ran Luo and Hang Sun ...... 238

A Potential Functional Cure in Chinese HBeAg-negative Chronic Hepatitis B Patients Treated with Peg-interferon Alpha-2a Xinyue Chen, Qianguo Mao, Yao Xie, Xiaoguang Dou, Qing Xie, Jifang Sheng, Zhiliang Gao, Xiaoling Zhou, Yingxia Liu, Huanwei Zheng, Shuqin Zhang, Shibo Li, Fusheng Zhu, Yuqin Xu, Mingxiang Zhang, Yaoren Hu, Xiaoping Chen, Yan Huang, Hong Ren and Jidong Jia ...... 249

Review Articles

HBV cccDNA and Its Potential as a Therapeutic Target Anjing Zhu, Xinzhong Liao, Shuang Li, Hang Zhao, Limin Chen, Min Xu and Xiaoqiong Duan ...... 258

Complementary and Alternative Medicine-related Drug-induced Liver Injury in Asia Cyriac Abby Philips, Philip Augustine, Sasidharan Rajesh, Praveen Kumar Y and Deepak Madhu ...... 263

Obesity Paradox in Chronic Liver Diseases: Product of Bias or a Real Thing? Ines Bilic Curcic, Maja Cigrovski Berkovic, Lucija Kuna, Hrvoje Roguljic, Robert Smolic, Silvija Canecki Varzic, Lucija Virovic Jukic and Martina Smolic ...... 275

Case Report

Amoxicillin-clavulanate-induced Granulomatous Hepatitis: Case Report and Review of the Literature Avin Aggarwal, Neha Jaswal, Richa Jain and Hussien Elsiesy ...... 280 Editorial

HBV cccDNA: The Stumbling Block for Treatment of HBV Infection

Shousheng Liu1 and Yongning Xin*2

1Central Laboratories, Qingdao Municipal Hospital, Qingdao, China; 2Department of Infectious Disease, Qingdao Municipal Hospital, Qingdao, China

Citation of this article: Liu S, Xin Y. HBV cccDNA: The stum- After the HBV infection, the innate immune and adaptive bling block for treatment of HBV infection. J Clin Transl Hepatol immune systems are activated and initiate antiviral 2019;7(3):195–196. doi: 10.14218/JCTH.2019.00047. responses, like those involving interferon (IFN)-g and TNF-a that are produced by T cells, thereby decreasing the levels of Hepatitis B virus (HBV) infection leads to severe liver disease HBV cccDNA, but the elimination of HBV cccDNA is difficult to and is one of the main causes of hepatocellular carcinoma- achieve.7 As the intermediate of HBV replication, cccDNA con- related mortality. Although the prevalence of HBV infection tributes to the persistence of HBV replication due to the bio- has decreased markedly with the effective application of logical mechanism of cccDNA formation, which itself remains ’ vaccines, it s still an arduous task to clear the hepatitis B unclear. Eliminating the cccDNA is critical to cure the HBV surface antigen after routine antiviral therapy, as the viral infection.8 rebound often occurs after therapy withdrawal. Therefore, the Recently, Zhu et al.9 conducted a systemic review of HBV 1 situation of dealing with HBV infection is still challenging. As cccDNA and its potential role as the therapeutic target of HBV such, eliminating HBV infection has become an important task infection. Pegylated-IFNa and nucleotide analogues (NAs) are for governments all over the world. In 2016, the Global Health the standard therapy methods, currently. Tang et al.10 Sector Strategy on Viral Hepatitis was approved by the World reported that IFNa exerts a significant effect on the long- Health Assembly to eliminate the HBV infection by 2030.2 term and sustainable suppression of cccDNA transcription, Accumulated studies have illuminated the processes which may due to the alteration of epigenetic modifications underlying the HBV replication cycle, which has provided the of cccDNA minichromosomes. NAs can inhibit HBV replication important insight into the detailed action of the HBV lifestyle by targeting the viral RNA-dependent DNA polymerase, and that has promoted development of therapeutic strategies several NAs, such as lamivudine, entecavir, telbivudine and so against HBV infection. HBV infection begins with the low- on, have been approved for clinical use. Although an effective affinity interaction between HBV and the heparin sulfate antiviral response has been observed with the use of IFNa and proteoglycans located on the host hepatocytes, followed by NAs, the cccDNA still could not be cleared completely, at HBV binding to the sodium taurocholate cotransporting poly- 10 3 present. peptide and then its entry into the hepatocytes. The HBV 9 genome is a relaxed circular DNA with a length of 3.2 kb, Zhu et al. reviewed several gene therapy strategies that which can be transported into the nuclear compartment in have been proposed to disable HBV cccDNA (Fig. 1). The first its nucleocapsid form after entry into the hepatocytes. therapeutic strategy involves the gene editing techniques, The relaxed circular DNA can be converted into closed such as those based on zinc finger nucleases (ZFNs), tran- covalent circular DNA (cccDNA), which exists in the nuclei of scription activator-like effector nucleases (TALENs), and the infected cells and can generate offspring virus. In the nucleus, clustered regularly interspaced short palindromic repeats/ cccDNA can transcribe to HBV pregenomic RNA, S RNAs for S CRISPR associated system (CRISPR/Cas). Three ZFNs target- proteins, and X RNA for HBx protein. In the patients positive ing the HBV polymerase, gene X and core were designed and 11 for hepatitis B surface antigen (commonly referred to as delivered into the HepAD38 cells by Weber et al., and they HBsAb), the amounts of cccDNA could become reduced with found that sustained suppression of HBV could be achieved by treatment of antiviral therapy but the reduction will be the HBV-targeted ZFNs in the HepAD38 cells. TALENs are modest. What’s more, the cccDNA can be monitored in the similar to ZFNs and have been used in several studies to liver of patients who recovered from HBV infection decades investigate their antiviral effects. The results showed that – ago.4 6 TALENs can reduce the cccDNA molecules effectively in cultured cell lines. The CRISPR/Cas system, used as a genome-editing tool, was shown to cleave DNA with sequence-specificity. Abbreviations: cccDNA, closed covalent circular DNA; CRISPR/Cas, clustered Several studies have demonstrated that the CRISPR/Cas regularly interspaced short palindromic repeats/CRISPR associated system; NAs, system can efficiently destroy the HBV cccDNA. Importantly, nucleotide analogues; HBV, hepatitis B virus; TALENs, transcription activator-like effector nucleases; ZFNs, zinc finger nucleases. in this review, the authors pointed out some defects that Received: 15 September 2019; Revised: 20 September 2019; Accepted: 20 Sep- should be overcome before the gene editing techniques can tember 2019 be applied in the clinical treatment of HBV infection. First, the *Correspondence to: Yongning Xin, Department of Infectious Disease, Qingdao Municipal Hospital, 1 Jiaozhou Road, Qingdao, Shandong 266011, China. Tel: specificity of the ZFNs must be ensured. Second, the TALENs +86-532-82789463, Fax: +86-532-85968434, E-mail: [email protected] should be delivered into the cells effectively. Third, the

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 195–196 195

the future. Gene editing and epigenetic modification techniques have displayed their bright prospect, but the deficiency of these therapeutic methods should be resolved in further studies. In addition, new therapeutic strategies targeting the HBV cccDNA should be explored in order to realize the goal of eliminating HBV infection by 2030.

Conflict of interest

The authors have no conflict of interests related to this publication.

Author contributions

Article conception and design (YX), acquisition of data (SL and YX), drafting of the manuscript (SL), critical revision of the manuscript (YX).

References

[1] Nassal M. HBV cccDNA: viral persistence reservoir and key obstacle for a cure of chronic hepatitis B. Gut 2015;64:1972–1984. doi: 10.1136/gutjnl-2015- 309809. [2] Global prevalence, treatment, and prevention of hepatitis B virus infection in 2016: a modelling study. Lancet Gastroenterol Hepatol 2018;3:383–403. doi: 10.1016/S2468-1253(18)30056-6. [3] Liang TJ, Block TM, McMahon BJ, Ghany MG, Urban S, Guo JT, et al. Present Fig. 1. Pathway of HBV infection and possible therapeutic methods of and future therapies of hepatitis B: From discovery to cure. Hepatology HBV infection by targeting cccDNA. 2015;62:1893–1908. doi: 10.1002/hep.28025. Abbreviations: cccDNA, closed covalent circular DNA; HBV, hepatitis [4] Chan HL, Wong VW, Tse AM, Tse CH, Chim AM, Chan HY, et al. Serum hepatitis B virus. B surface antigen quantitation can reflect hepatitis B virus in the liver and predict treatment response. Clin Gastroenterol Hepatol 2007;5:1462–1468. doi: 10.1016/j.cgh.2007.09.005. [5] Gao Y, Li Y, Meng Q, Zhang Z, Zhao P, Shang Q, et al. Serum hepatitis B virus CRISPR/Cas9 system must be delivered into the cells effec- DNA, RNA, and HBsAg: Which correlated better with intrahepatic covalently closed circular DNA before and after nucleos(t)ide analogue treatment? J Clin tively and the off-target effect must be solved. Microbiol 2017;55:2972–2982. doi: 10.1128/JCM.00760-17. The second therapeutic strategy is to silence cccDNA [6] Testoni B, Durantel D, Zoulim F. Novel targets for hepatitis B virus therapy. transcription via epigenetic modifications. Results from a Liver Int 2017;37 Suppl 1:33–39. doi: 10.1111/liv.13307. previous study indicated that DNA methylation and histone [7] Xia Y, Stadler D, Lucifora J, Reisinger F, Webb D, Hösel M, et al. Interferon-g and tumor necrosis factor-a produced by T cells reduce the HBV persistence ’ acetylation are required for cccDNA, so it s promising to form, cccDNA, without cytolysis. Gastroenterology 2016;150:194–205. doi: regulate the DNA methylation or histone acetylation to 10.1053/j.gastro.2015.09.026. eliminate the cccDNA. Belloni et al.12 reported that the ace- [8] Ni Y, Lempp FA, Mehrle S, Nkongolo S, Kaufman C, Fälth M, et al. Hepatitis B tylation of cccDNA-bound histones could be decreased and and D viruses exploit sodium taurocholate co-transporting polypeptide for species-specific entry into hepatocytes. Gastroenterology 2014;146:1070– the transcription of HBV cccDNA could be repressed by 1083. doi: 10.1053/j.gastro.2013.12.024. IFNa. It has also been reported that CpG methylation can [9] Zhu A, Liao X, Li S, Zhao H, Chen L, Xu M, et al. HBV cccDNA and its potential regulate the transcription of HBV cccDNA during chronic as a therapeutic target. J Clin Transl Hepatol 2019;7:258–262. doi: 10. HBV infection, and that CpG island II methylation can 14218/JCTH.2018.00054. [10] Tang LSY, Covert E, Wilson E, Kottilil S. Chronic hepatitis B infection: A decrease cccDNA transcription and subsequent viral core review. JAMA 2018;319:1802–1813. doi: 10.1001/jama.2018.3795. 13 DNA replication. The acetylation status of HBV cccDNA- [11] Weber ND, Stone D, Sedlak RH, De Silva Feelixge HS, Roychoudhury P, bound histone 3 and histone can affect the HBV replication. Schiffer JT, et al. AAV-mediated delivery of zinc finger nucleases targeting Wei et al.14 reported that curcumin could repress HBV repli- hepatitis B virus inhibits active replication. PLoS One 2014;9:e97579. doi: 10.1371/journal.pone.0097579. cation by decreasing the cccDNA-bound histone acetylation. [12] Belloni L, Allweiss L, Guerrieri F, Pediconi N, Volz T, Pollicino T, et al. IFN-a These conclusions suggested that epigenetic modifications inhibits HBV transcription and replication in cell culture and in humanized could be regarded as the potential therapeutic target to elim- mice by targeting the epigenetic regulation of the nuclear cccDNA minichro- – inate HBV cccDNA. mosome. J Clin Invest 2012;122:529 537. doi: 10.1172/JCI58847. [13] Zhang Y, Mao R, Yan R, Cai D, Zhang Y, Zhu H, et al. Transcription of hepatitis In summary, HBV infection, representing the most severe B virus covalently closed circular DNA is regulated by CpG methylation during chronic liver disease in the world, cannot be ignored. chronic infection. PLoS One 2014;9:e110442. doi: 10.1371/journal.pone. Although some antiviral medicines, like pegylated-IFNa and 0110442. NAs, have been approved for the clinical treatment of HBV, [14] Wei ZQ, Zhang YH, Ke CZ, Chen HX, Ren P, He YL, et al. Curcumin inhibits hepatitis B virus infection by down-regulating cccDNA-bound histone acety- the persistent existence of cccDNA and the frequent rebound lation. World J Gastroenterol 2017;23:6252–6260. doi: 10.3748/wjg.v23. of HBV after treatment withdrawal still need to be overcome in i34.6252.

196 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 195–196 Original Article

Comparing the Efficacy and Safety of Treating Chronic Hepatitis B Infection during Pregnancy with Lamivudine, Telbivudine, and Tenofovir: A Meta-analysis

Shahnaz Sali1, Mohammad Darvishi*2, Mojtaba GhasemiAdl3, Meisam Akhlaghdoust4, Azin Mirzazadeh5, Somayeh Elikaei Behjati6, Hossein Sheikh-Zeinolabedini6, Shervin Shokouhi1 and Soheil Tavakolpour*1,3

1Infectious Diseases and Tropical Medicine Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran; 2Department of Aerospace and Subaquatic Medicine, Infectious Diseases and Tropical Medicine Research Center (IDTMRC), AJA University of Medical Sciences, Tehran, Iran; 3Pharmaceutical Sciences Research Center, Tehran University of Medical Sciences, Tehran, Iran; 4Pars Advanced and Minimally Invasive Medical Manners Research Center, Pars Hospital, Iran University of Medical Sciences, Tehran, Iran; 5Joint Bioinformatics Graduate Program, University of Arkansas Little Rock and University of Arkansas for Medical Sciences, Little Rock, AR, USA; 6The Genetics Department at Islamic Azad University, Science and Research Branch, Tehran, Iran

Abstract gested LdT’s high capability of preventing mother-to-child- transmission. However, TDF failed to show significant associ- Background and Aims: The perinatal transmission of hep- ations to a reduced risk of mother-to-child-transmission, atitis B virus (HBV) remains an important global health prob- probably due to the low number of patients included. Con- lem. Here, a systematic review and meta-analysis were clusions: Although using either lamivudine, LdT, or TDF could conducted to evaluate the evidence regarding the efficacy lead to more favorable maternal/fetal outcomes, LdT seemed and maternal/fetal safety of treating pregnant women to show more potential in resolving certain infant- and mater- with lamivudine, telbivudine (LdT), and tenofovir (TDF). nal-related outcomes. More studies on the safety profile of Methods: A PubMed and Scopus search resulted in 1,076 such treatments are required. records, which were reduced to 36, containing 7,717 preg- Citation of this article: Sali S, Darvishi M, GhasemiAdl M, nant women with chronic HBV infection and 7467 infants Akhlaghdoust M, Mirzazadeh A, Behjati SE, et al. Comparing meeting the inclusion criteria. The latest search was in August the efficacy and safety of treating chronic hepatitis B infection 2019. Results: Treatment with LdT, but not lamivudine and during pregnancy with lamivudine, telbivudine, and tenofovir: TDF, could significantly reduce the hepatitis B virus surface A meta-analysis. J Clin Transl Hepatol 2019;7(3):197–212. antigen-positive rate (odds ratio (OR) = 0.37) in infants; it doi: 10.14218/JCTH.2019.00021. also led to higher rates of hepatitis B e antigen loss (OR = 12.14), hepatitis B e antigen seroconversion (OR = 8.93), and alanine aminotransferase normalization in Introduction mothers (OR = 1.49). Each of these treatments was able to significantly reduce HBV DNA positivity at birth (total OR = Approximately 240 million people are chronically infected 0.19) and mother-to-child-transmission of HBV (total OR = with hepatitis B virus (HBV), which has a high rate of mortality 0.15), and to cause higher rates of HBV DNA suppression in annually.1 During recent decades, the epidemiology of HBV mothers (total OR = 25.53). However, nucleos(t)ide ana- infection had decreased, due to the impact of universal logues might also be involved in creatine kinase elevation infant vaccination programs. HBV vaccination is an effective (total OR = 7.48). In contrast, no significant association and safe approach, given on day 0 and at the end of 1 month was found between nucleos(t)ide analogue therapy and pre- and 6 months.2 However, this method is ineffective for term/premature births, congenital malformation, low birth patients already infected with HBV. weight, and abortion or fetal/infant death. The results sug- HBV can be acquired by contaminated blood product exposure, sexual activity, and perinatal transmission. Peri- natal transmission, or mother-to-child transmission (MTCT), Keywords: Hepatitis B virus; Antiviral therapy; Tenofovir; Telbivudine; Lamivudine; Nucleos(t)ide analogues; Pregnancy. remains a critical infection route in hepatitis B-endemic Abbreviations: CHB, chronic HBV infection; CI, confidence interval; CK, creatine countries. Regardless of the fact that pegylated-interferon kinase; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B virus surface antigen; alpha-2a can lead to high rates of hepatitis B virus surface HBV, hepatitis B virus; LAM, lamivudine; LdT, telbivudine; MTCT, mother-to-child- antigen (HBsAg) loss,3 nucleos(t)ide analogues (NAs), includ- transmission; NA, nucleos(t)ide analogue; TDF, tenofovir. Received: 7 June 2019; Revised: 13 August 2019; Accepted: 19 August 2019 ing lamivudine (LAM), telbivudine (LdT), entecavir, adefovir *Correspondence to: Soheil Tavakolpour,Infectious Diseases and Tropical Medicine and tenofovir (TDF), are unable to eradicate this chronic Research Center, Shahid Beheshti University of Medical Sciences, Tehran 198396- infection. However, they seem to be able to decrease the 3113, Iran. Tel/Fax: +98-2122267157, E-mail: [email protected]; risk of MTCT. Without prophylaxis, in mothers who are posi- Mohammad Darvishi, Department of Aerospace and Subaquatic Medicine, Infectious Diseases and Tropical Medicine Research Center (IDTMRC), AJA University of Medical tive for both HBsAg and hepatitis B e antigen (HBeAg), the 4 Sciences, Tehran, Iran. E-mail: [email protected] risk for transmission to the baby is high. In a considerable

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 197

A systematic literature search was conducted for all published Statistical analysis articles associated with NAs therapy for CHB during pregnancy, using the PubMed and Scopus databases, with no Statistical analyses were carried out using Review Manager limitation period. The last search update was on August 1, statistical software, version 5.3. Dichotomous data 2019. Selected keywords covered all studies associated with were expressed as odds ratio (OR) and 95% confidence LAM, LdT, and TDF therapies for CHB during pregnancy. The intervals (CIs). Mantel–Haenszel was used. Otherwise, the keywords employed were ((Tenofovir) OR (Telbivudine) OR meta-analysis was conducted using a fixed-effect model.14 (Adefovir) OR (Entecavir) OR (Lamivudine) OR (Nucleoside Specifically, the analysis was performed with the use of a analogues) OR (Nucleotide analogues) OR (Nucleos(t)ide random-effects model (Mantel–Haenszel) after exploring for analogues)) AND ((Hepatitis B virus) OR (HBV)) AND ((Preg- causes of heterogeneity or the fixed-effects models. Cochran nancy) OR (Pregnant) OR (Intrauterine transmission) OR Q test and the I2 statistic were used for examining heteroge- (Perinatal transmission) OR (Utero transmission) OR (Vertical neity among studies and were considered significant if transmission)). The references for the selected articles were p <0.10 or I2 >50%. When significant heterogeneity in the also checked for any articles missed. results was observed, the random effect model was employed. However, in homogeneous conditions, the fixed-effect model Selection criteria was used. During the entire study, a p-value of <0.05 was considered as statistically significant for all outcomes. Among the studies found, only controlled or comparative studies that enrolled pregnant women diagnosed with CHB Results infection (a persistence of HBsAg for more than 6 months), who received LAM, LdT, or TDF were considered for analysis. The initial search resulted in 1,076 records. Before starting As the current recommendation of NAs treatment for MTCT the primary screening, duplicate records, and non-English had been suggested to be initiated from week 24 of preg- articles were identified and excluded (n = 269). Checking nancy, studies that contained only patients treated before titles of the articles led to the omission of 529 records. The week 24 were excluded. The studies needed to include remaining studies (n = 278) were evaluated by reviewing

198 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 Sali S. et al: Treating chronic HBV during pregnancy their abstracts. As a result, 173 studies were identified as from documented patient data in the literature.36 Table 1 nonrelevant records. Finally, the full text of 105 studies was summarizes the characteristics of the studies. checked to select those matching the inclusion criteria. Seventy articles were excluded, due to different reasons Infant outcomes (lack of original data, not containing control group, insufficient data, acute HBV, combination therapy, treatment initia- Comparison of antiviral therapy with no treatment: tion in the first trimester, and case report), while one new Results adapted from 25 studies in the analysis revealed study was included which had not appeared among the orig- that NA (LAM, LdT, and TDF) therapies could significantly inal search results. Eventually, 36 studies were included for reduce the rate of HBsAg positivity at birth for infants born meta-analysis. The study selection process and reasons for from CHB mothers (OR [95% CIs] = 0.50 [0.38, 0.67]; I2 = exclusions are presented in Fig. 1. 61%; p-value <0.00001) (Fig. 2). As the results of treating CHB-positive pregnant women with these drugs, the risk of birth of an infant with positive HBV DNA was also reduced Studies’ selection and characteristics significantly (OR [95% CIs] = 0.19 [0.10, 0.36], I2 = 84%, p-value <0.00001) (Fig. 3). The rate of MTCT for any sepa- Thirty-six studies, containing 7,717 pregnant women (4,468 rated drug was extractable in almost all studies included. treated; 3,249 untreated) with CHB and 7,467 infants (4,317 Reports were analyzed from a total of 3,629 newborn from treated mothers; 3,150 from untreated mothers), were infants from CHB mothers and 3,245 controls, who had included. From these studies, there were 15 groups treated 4,6–9,15–26 received hepatitis B immune globulin and vaccine, and also with LAM, 17 groups treated with LdT, and 12 were followed for more than 6 months. Results from the 36 16,24,26–35 groups treated with TDF. Some of them covered studies revealed that starting antiviral therapy in the second 7,15,16,24,26,28,31 more than one NA (n = 7). Contrary to rela- or third trimester could significantly protect infants from CHB tively older studies, the majority of recent studies did not (OR [95% CIs] = 0.15 [0.11, 0.19], I2 = 12%, p-value cover LAM. In all studies for the group under treatment, anti- <0.00001) (Fig. 4). viral therapy was initiated in the second or third trimester, Following analysis of the risk of congenital malformation in while discontinuance occurred at different times. All the a total of 1,954 born babies from CHB mothers and 2,194 studies presented original data associated with the control controls, no statistical difference was obtained. However, group, except for one, where the control group was taken those who were exposed to NA therapy seemed to be more

Fig. 1. The study selection process.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 199 0 ora fCiia n rnltoa Hepatology Translational and Clinical of Journal 200 Table 1. Characteristics of the included studies

Hepatitis B immune Baseline Treatment Treatment Duration Participants, globulin Maternal HBV DNA Baseline alanine start, discontinuation. up to mothers: Interventions, and age in level, Log10 aminotransferase, gestational postpartum delivery, Author, year Region infants mothers vaccine# years IU/mL U/L week week week Study design

Li, 200340 China 43:43 Lamivudine N/A N/A 7.49 6 0.54 N/A 28 4 12 RCT 52:52 Control N/A 7.05 6 1.29 Zonneveld, Netherlands 8:8 Lamivudine Yes 21.25 6 9.30 32.4 6 34 200341 2.65 16.3 At delivery 6 Cohort 24:25 Control 23 (16–34) 9.39study N/A Su, 200436 China 38:12 Lamivudine Yes N/A N/A N/A N/A N/A N/A Cohort study 10:10 Control N/A N/A N/A Ni, 200542 Taiwan 29:29 Lamivudine N/A 14.7 6 10.95 214 6 195 N/A N/A N/A Prospective, 5.6 open-label, nonrandom 29:29 Control 14.0 6 9.32 165 6 123 5.8 Xu, 20095 United 89:56 Lamivudine Yes 26 (19– 2220.0 6 0.4 (0.1–5.3) 3 32 4 8 RCT Kingdom 32) 1610.9 ULN MEq/mL* 61:59 Control 25 (20– 2692.7 6 0.4 (0.1–6) 3 ULN 36) 1627.0 MEq/mL* Han, 20116 China 135:132 Telbivudine Yes 27 (20– 8.10 6 0.56 35.7 6 43 20–32 4 8–20 Cohort study 38) 94:94 Control 26 (20– 7.98 6 0.61 42.5 6 40.1 35) Yu, 201143 China 94:94 Lamivudine Yes 27.68 6 6.97 6 1.95 394.36 6 372.18 24–32 After childbirth, 8–16 Cohort study 3.65 till satisfactory

efficacy or drug S. Sali resistance 2019 mutation appeared tal et 6 6 6 o.7|197 | 7 vol. 91:91 Control 26.33 7.20 0.94 294.03 233.83 3.24 raigcrncHVdrn pregnancy during HBV chronic Treating : Yu, 201244 China 94:94 Lamivudine Yes 26.64 6 7.63 6 0.54 $40 24–32 Variable after 8–16 Cohort study 4.17 delivery 91:91 Control 25.78 6 7.71 6 0.71 45.0

– 3.89 212 Pan, 20124 China 53:54 Telbivudine Yes 27 (21– 8.08 (rage 60.40 (41.40– Second Variable after 12–27 Cohort study 34) 6.6–9.4) 422.00) trimester delivery 35:35 Control 27 (21– 8.08 (range 63.20 (42.40– 33) 6.76–9.08) 262.50) Celen, 201327 Turkey 21:21 Tenofovir Yes 28.2 6 >7 56 (22–71) 18–27 4 13–22 Retrospective 4.1 study 24:23 Control 26.9 6 >7 52 (19–77) 2.9

(continued) aiS. Sali Table 1. (continued )

Hepatitis B

immune Baseline Treatment Treatment Duration al et Participants, globulin Maternal HBV DNA Baseline alanine start, discontinuation. up to

mothers: Interventions, and age in level, Log10 aminotransferase, gestational postpartum delivery, pregnancy during HBV chronic Treating : Author, year Region infants mothers vaccine# years IU/mL U/L week week week Study design

Zhang, 20147 China 55:54 Lamivudine Yes 28.42 6 7.62 6 0.37 39.65 6 26.37 28–30 4 10–12 Prospective, 7.11 open-label, nonrandom 374:370 Control 28.97 6 7.58 6 0.45 29.53 6 20.72 64.59 ora fCiia n rnltoa Hepatology Translational and Clinical of Journal Zhang, 20147 China 263:262 Telbivudine Yes 29.78 6 7.69 6 0.44 30.06 6 28.86 28–30 4 10–12 Prospective, 6.31 open-label, nonrandom 374:370 Control 28.97 6 7.58 6 0.45 29.53 6 20.72 4.59 Ayres, 201445 Australia 21:18 Lamivudine Yes N/A >7 N/A 32 2 8 Cohort study 5:3 Control Greenup, Australia 58:43 Tenofovir Yes 30 6 8.5 7.9 6 08 28 (22–36) 32 12 8 Cohort study 201428 20:10 Control 28 6 586 04 25 (17–31) Greenup, Australia 52:44 Lamivudine Yes 28 6 5.3 7.7 6 06 22 (18–30) 32 12 8 Cohort study 201428 20:10 Control 28 6 586 04 25 (17–31) Nguyen, Australia 44:44 Telbivudine N/A 29.1 6 8.0 23.5 (20–31.3) 32 4 8 Cohort study 201415 4.9 14:14 Control 27.1 6 25.5 (18–35) 4.0 Nguyen, Australia 43:43 Lamivudine N/A 30.9 6 8.0 29.0 (17.5–40.5) 32 2–12 8 Cohort study 201415 4.5 14:14 Control 27.1 6 25.5 (18–35) 4.0 Han, 20158 China 362:365 Telbivudine Yes 27 (20– 8(6–9.1) 19.9 (5.2–513.5) N/A N/A N/A Prospective, 2019 38) open-labeled, nonrandom 92:92 Control 26 (20– 7.93 (6– 26.55 (8.1–262.5) 35) 9.5) o.7|197 | 7 vol. Chen, 201532 Taiwan 62:65 Tenofovir Yes 32.41 6 8.25 6 0.45 23.27 6 36.2 30 4 10 Prospective, 3.12 open-labeled, nonrandom 56:56 Control 32.45 6 8.24 6 0.35 16.59 6 14.43 3.2

– Tekin Koruk, Turkey 29:20 Tenofovir Yes 27.4 6 N/A N/A 22.2 6 8.5 N/A ;18 Retrospective 1 201 212 201516 4.7 (1–36) 54:54 Control 28.7 6 1.98 6 2.21 26.7 6 22.9 4.9 Tekin Koruk, Turkey 4:4 Lamivudine Yes 28.8 6 N/A N/A 22.2 6 8.5 N/A ;18 Retrospective 201516 5.1 (1–36) 54:54 Control 28.7 6 1.98 6 2.21 26.7 6 22.9 4.9 Tekin Koruk, Turkey 31:36 Telbivudine Yes N/A N/A 22.2 6 8.5 N/A ;18 Retrospective 201516 (1–36) 54:54 Control 28.7 6 1.98 6 2.21 26.7 6 22.9 4.9

(continued) 0 ora fCiia n rnltoa Hepatology Translational and Clinical of Journal 202 Table 1. (continued )

Wu, 20159 China 279:280 Telbivudine Yes 27 (17– 7.26 6 0.50 111 (45–282) 24–32 N/A 8–16 Cohort study 38 171:130 Control 28 (18– 7.40 6 0.65 134 (44–330) 40) Liu, 201618 China A: 50:50 Telbivudine Yes 27.88 6 7.67 6 0.79 46.64 6 58.74 before the 4 N/A Cohort study B: 32:32 3.73 7.46 6 0.73 28.91 6 38.48 third 8–12 28.31 6 trimester 3.81 28–32 78:78 Control 27.46 6 7.56 6 0.57 30.87 6 28.99 3.47 Pan, 201629 China 97:95 Tenofovir Yes 27.4 6 8.2 6 0.5 23.0 6 22.4 30–32 4 8–10 RCT 3.0 100:88 Control 26.8 6 8.0 6 0.7 20.5 6 15.4 3.0 Samadi Canada 23:24 Tenofovir Yes 30 (28– 7.7 (3.2– 30 (18–50) 28–32 12 8–12 Cohort study Kochaksaraei, 34) 8.1) 201630 138:146 Control 32 (29– 2.3 (1.6– 17 (12–24) 36) 3.1) Tan, 201617 China A: 34:34 Telbivudine Yes 29 2 (1.82– A: 18 (9–500) A: <14 28 N/A Cohort study B: 135:137 29 6.99) B: 37 (6–697) B: 14–28 12–26 7.69 (6.05– 8.98) 316:320 Control 29 7.67 (6– 22 (5–623) 8.91) Chen, 201722 China 43:43 Telbivudine Yes 28.1 6 7.2 6 0.7 89.3 6 104.2 13–32 At delivery 8–27 Cohort study 6.7 S. Sali

2019 A: 79:79 Control A: 27.2 A: 4.2 6 0.8 A: 47.8 6 57.9 B: 89:89 6 5.5 B: 7.2 6 0.6 B: 85.0 6 86.3

B: 26.2 al et o.7|197 | 7 vol. 6 4.5

Hu, 201719 China 149:128 Telbivudine Yes 25.9 6 7.43 6 1.26 N/A 28–32 3–48–12 Cohort study pregnancy during HBV chronic Treating : 3.7 179:156 Control 26.4 6 7.37 6 1.49 N/A 3.4 – 46 212 Pan, 2017 China A: 66:66 lamivudine Yes 27.65 6 7.22 6 0.61 68.6 6 103.6 13–26 At delivery 14–27 Retrospective B: 94:94 4.08 7.26 6 0.55 36.4 6 39.7 28–30 12 10–12 cohort 27.37 6 3.54 89:89 Control 27.08 6 7.33 6 0.47 28.0 6 35.4 4.22

(continued) aiS. Sali Table 1. (continued )

Hepatitis B

immune Baseline Treatment Treatment Duration al et Participants, globulin Maternal HBV DNA Baseline alanine start, discontinuation. up to

Sun, 201720 China A: 62:62 Telbivudine Yes 28.9 6 7.79 6 0.22 125.3 6 57.6 12 12 28 Cohort study B: 61:61 11.8 7.75 6 0.19 132.3 6 52.9 20–28 12–20 29.7 6 9.8 6 6 6 ora fCiia n rnltoa Hepatology Translational and Clinical of Journal 65:65 Control 27.5 7.74 0.22 128.5 48.7 12.9 Wakano, Japan 2:2 Tenofovir Yes 28–37 9.0 N/A 28–32 4–88–12 Cohort study 31 2017 3:3 Control 9.0 N/A Wakano, Japan 3:3 Lamivudine Yes 28–37 9.0 N/A 28–32 4–88–12 Cohort study 201731 3:3 Control 9.0 N/A Yi, 201721 China A: 41:41 Telbivudine Yes 31.54 6 1.50 6 0.62 15.19 6 8.53 Third 28 Up to 12 Cohort study B: 179:179 4.21 8.05 6 0.37 21.58 6 13.15 trimester 27.77 6 3.48 176:176 Control 28.27 6 7.94 6 0.62 18.85 6 9.83 3.65 Chang, 201933 Taiwan 110:115 Tenofovir Yes 32.84 6 8.25 6 0.48 20.88 6 28.94 30–32 Variable after 8–10 Cohort study 3.57 delivery 91:93 Control 32.69 6 8.29 6 0.49 19.10 6 23.85 3.36 Han, 201923 China 139:137 Telbivudine Yes 26 (20– 7.73 117 (56–1166) 12–34 Variable after 6–28 prospective 43) (6.04; delivery nonintervention 9.30) study 102:99 Control 26 (18– 7.72 (6.03;9.00) 164 (53–

2019 42) 1025)

35 o.7|197 | 7 vol. Lin, 2018 China 59:58 Tenofovir Yes 28.31 6 Not 54.62 6 105.7 24 28 ;16 Cohort study 3.56 mentioned 52:52 Control 28.06 6 7.44 6 0.80 57.5 6 103.3 3.42 Zeng, 201926 China A: 58:58 A: Telbivudine Yes A:27.2 6 A: 7.88 6 A: 127.3 6 72.2 20–28 12 N/A Retrospective

– B: 51:51 B: Tenofovir 10.8 0.65 B: 143.3 6 104.6 study 1 203 212 B: 26.5 B: 7.91 6 6 9.5 0.75 36:36 Control 25.7 6 7.69 6 0.53 132.3 6 78.3 10.9 Jourdain, China 168:147 Tenofovir Yes 25.5 7.6 6 1.5 N/A 28.3 8 N/A RCT 201834 (22.6– (27.9– 29.1) 28.6) 163:147 Control 26.7 7.3 6 1.7 28.1 (23.5– (27.9– 30.5) 28.6)

(continued) Sali S. et al: Treating chronic HBV during pregnancy

vulnerable to developing congenital malformation (OR [95% CIs] = 1.55 [0.80, 3.00], I2 =0%,p-value = 0.19). Regard- ing low birth weight, nine studies were available, which did not show a significant difference among the treated and untreated groups (OR [95% CIs] = 0.95 [0.57, 1.61], I2 = observation study multicenter study open label multicenter study 0%, p-value = 0.86). In order to evaluate the risk of NAs therapy threatening the life of a fetus/infant, abortion and fetal/infant death were analyzed separately. The results 18 Prospective, – 16 Prospective suggest a probable protective role of NAs therapy for each 12 Prospective – ; Duration up to delivery, week Study design of these factors, but they were not significantly different among the treated and untreated patients (abortion: OR [95% CIs] = 0.47 [0.11, 1.92], I2 = 31%, p-value = 0.29; fetal/infant death: OR [95% CIs] = 0.90 [0.54, 1.50], I2 = 10%, p-value = 0.44). There were 15 studies that compared premature/preterm 12 Treatment discontinuation. postpartum week birth rate among those who received NA, but significant associations were not found (OR [95% CIs] = 0.79 [0.58, 1.09], I2 = 32%, p-value = 0.16).

infants may be treated with hepatitis B immune globulin and/or Comparison of the results of LAM, LdT, and TDF: Fol- 28 12 12 32 12 8 – – lowing the use of LAM, LdT, and TDF, the risk of HBsAg pos- 22 trimester Treatment start, gestational week itivity of an infant at birth was reduced compared with the cases not using any NAs; however, only LdT showed a significant result (LAM: OR [95% CIs] = 0.63 [0.38, 1.06], I2 = 65%, p-value = 0.05; LdT: OR [95% CIs] = 0.37 [0.24, 71.87 66.34 2 6 6 0.57], I = 67%, p-value <0.00001; TDF: OR [95% CIs] = 62.46 8.32 24 6.51 2 6 6 6 0.53 [0.21, 1.33], I = 60%, p-value = 0.18) (Fig. 2). The results from 20 studies, containing 4,041 infants, demonstra- N/A N/A Third B: 53.34 A: 45.79 Baseline alanine aminotransferase, U/L ted a significant reduction of HBV DNA positivity at birth in babies of CHB-infected mothers, who were exposed to each of the NAs (LAM: OR [95% CIs] = 0.15 [0.06, 0.40], I2 = 6 6 0.79 41.16 0.82 26.53 1.04 23.62

1.8 23%, p-value = 0.0002; LdT: OR [95% CIs] = 0.23 [0.09, 6 6 6 – 0.57], I2 = 88%, p-value = 0.001; TDF: OR [95% CIs] = 2 (474 105) 7.71 A: 7.89 0.66 B: 7.68 0.70 Baseline HBV DNA level, Log10 IU/mL 8.15 8.09 0.19 [0.11, 0.33], I =0%,p-value <0.00001) (Fig. 3). The analysis implies a higher efficacy of LdT in reducing 2

6 the risk of MTCT (OR [95% CIs] = 0.10 [0.06, 0.15], I = 2.9 3.9 103 6 6 6

6 16%, p-value <0.00001). The next most effective is TDF 3.56 4.35 2 4.6 4.72 6 B: 28.35 6 Maternal age in years 4.17 3.66 (OR [95% CIs] = 0.17 [0.11, 0.27], I =0%,p-value <0.00001), then LAM (OR [95% CIs] = 0.24 [0.14, 0.39], I2 = 11%, p-value <0.00001) (Fig. 4). When associated # with the risk of congenital malformation, none of the NAs was higher than the others. Indeed, despite their nonsignifi- Hepatitis B immune globulin and vaccine Yes A: 27.78 cant differences, each of these drugs may be a risk factor for congenital malformation development (LAM: OR [95% CIs] = 1.33[0.38, 2.34], I2 =0%,p-value = 0.58; LdT: OR [95% CIs] = 1.70 [0.57, 5.03], I2 =0%,p-value = 0.34; TDF: OR [95% CIs] = 1.80 [0.43, 7.65], I2 =0%,p-value = 0.42). Interventions, mothers A: Telbivudine B: Tenofovir Maternal outcomes

Comparison of antiviral therapy with no treatment: Among the selected studies, 10 (LAM = 3, LdT = 6, TDF = 39:39 Control 27.4 136:136 Control 27.14 Participants, mothers: infants B: 325:325 21:21 Control 26.8 1) evaluated the capacity of NAs therapy in terms of sup- pressing HBV DNA in mothers. The overall results showed encouraging results (OR [95% CIs] = 25.53 [8.59, 75.92], I2 = 62%, p-value <0.00001) (Fig. 5A)). However, when HBeAg loss or seroconversion rates were analyzed, no signifi- China 91:79China Telbivudine A: 396:400 YesEgypt 34:34 27.8 Lamivudine Yes 27 ) cant differences were detected (HBeAg loss: OR [95% CIs] = 2.90 [1.58, 5.34], I2 = 58%, p-value = 0.0006; HBeAg sero- 2 47 48 conversion: OR [95% CIs] = 2.68 [1.59, 4.52], I = 53%, continued 24 ( p-value = 0.0002). Moreover, no significant difference was found in the total results for the normalization of alanine aminotransferase levels (OR [95% CIs] = 1.37 [0.88, 2.14], I2 = Sheng, 2018 Foaud, 2019 Author, year Region Liu, 2019 The control group in thisvaccine. Table represents the pregnant women whoAbbreviation: did RCT, not randomized receive controlled any trial. type of oral hepatitis B virus therapy during the pregnancy. Their Table 1. 95%, p-value = 0.17).

204 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 Sali S. et al: Treating chronic HBV during pregnancy

Fig. 2. Forest plots of infant HBsAg positivity at birth. Abbreviation: HBsAg, hepatitis B surface antigen.

For maternal side effects, two parameters were consid- NAs group and controls regarding postpartum hemorrhage ered: CK elevation and postpartum hemorrhage. Interest- (OR [95% CIs] = 0.94 [0.77, 1.14], I2 =0%,p-value = 0.52). ingly, among the 1,619 mothers monitored from the NAs Comparison of LAM, LdT, and TDF: The calculations group for CK elevation, 22 of them showed a high level of CK. showed that LdT probably had a greater capacity to suppress In contrast, none of the 994 mothers without NA therapy was HBV DNA in pregnant women, compared with LAM (LAM: OR reported. This could suggest NAs playing a role in CK [95% CIs] = 10.88 [0.61, 194.48], I2 = 79%, p-value = elevation during the pregnancy (OR [95% CIs] = 7.48 0.10; LdT: [95% CIs] = 61.15 [19.71, 189.74], I2 =0%,p- [2.41, 23.24], I2 =0%,p-value = 0.0005) (Fig. 5B). value <0.00001 (Fig. 5A)). Moreover, LdT was the only NA However, no significant differences were found among the which was capable to induce HBeAg loss and seroconversion

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 205 Sali S. et al: Treating chronic HBV during pregnancy

Fig. 3. Forest plots of infant HBV DNA positivity at birth. Abbreviation: HBV, hepatitis B virus. in a significant manner. HBeAg loss (LAM: OR [95% CIs] = compared to the LdT group, these results might be revised in 1.20 [0.62, 2.33], I2 = not applicable, p-value = 0.59; LdT: future analysis. OR [95% CIs] = 12.14 [2.17, 67.92], I2 =0%,p-value = 0.004; TDF: OR [95% CIs] = 3.26 [0.60, 17.73], I2 =61%, Publication bias p-value = 0.17) HBeAg seroconversion (LAM: OR [95% CIs] = 1.05 [0.54, 2.02], I2 =0%,p-value = 0.89; LdT: OR [95% In order to evaluate publication bias in the studies included, a CIs] = 8.93 [2.86, 27.90], I2 =7%,p-value = 0.0002; TDF: funnel plot was used. The shape of these plots for each analysis OR [95% CIs] = 1.20 [0.30, 4.85], I2 =61%,p-value = 0.80). suggests no evidence of publication bias among the studies. As Interestingly, the LdT groups also led to significant normal- an example, the funnel plot for MTCT is shown in Fig. 6. izations of alanine aminotransferase levels, as compared with off-therapy controls (OR [95% CIs] = 1.49[1.30, 1.72], I2 = Discussion 0%, p-value <0.00001), but not LAM (OR [95% CIs] = 2.47 [0.27, 22.52], I2 = 97%, p-value = 0.42). However, because The rate of new HBV infections has declined by approximately of the low number of mothers included in the TDF group, as 82% since 1991.37 However, women of childbearing age with

206 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 Sali S. et al: Treating chronic HBV during pregnancy

Fig. 4. Forest plots of mother-to-child transmission of HBV infection after 6–12 months. Abbreviation: HBV, hepatitis B virus.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 207 Sali S. et al: Treating chronic HBV during pregnancy

Fig. 5. Forest plots of the rate of making HBV DNA undetectable (A) and creatine kinase elevation, postpartum hemorrhage (B) in mothers following NA therapy during pregnancy. Abbreviations: HBV, hepatitis B virus; NA, nucleos(t)ide analogue.

208 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 Sali S. et al: Treating chronic HBV during pregnancy

recent years, or increases in the number of studies analyzed. Selected other outcomes reported in certain published meta- analyses are displayed in Table 2. According to the European Association for the Study of the Liver recommendations for pregnant women with CHB: (1) those with positive HBsAg should be screened in the first trimester of pregnancy (Evidence level 1, the grade of recommendation 1); (2) in a CHB-infected woman of childbearing age without advanced fibrosis who plans a pregnancy in the future, delaying therapy until the child is born was recommended (Evidence level II-2, the grade of recommendation 1); (3) for pregnant women with CHB and advanced fibrosis or cirrhosis, therapy with TDF is recommended (Evidence level II-2, the grade of recommendation 1); and (4) continuing TDF and switching to TDF in those under treatment with other NAs was also recommended (Evidence level II-2, the grade of recommendation 1; these are consistent with TDF and LdT being in a safer category than LAM Fig. 6. Funnel plot analysis to examine publication bias for mother-to- (Federal Drug Administration Pregnancy Category B vs. C), child transmission. and a higher barrier to resistance in TDF than LdT); (5) in pregnant women with either HBV DNA >200,000 IU/mL or CHB infections remain an important source of the continued HBsAg levels >4 log IU/mL, starting antiviral prophylaxis with spread of HBV. Hence, it is critical to prevent the maternal TDF at week 24–28 of gestation and continuing for up to 12 vertical transmission of HBV to reduce the overall number of weeks after delivery was recommended (Evidence level 1, the CHB patients. Pregnant women are vulnerable to several grade of recommendation 1). The recommendation to con- treatments and diseases. In the case of HBV and its associ- tinue for up to 12 weeks might be due to high risk of ated therapeutic options, several important points should be postpartum alanine aminotransferase level elevation in CHB considered. The capacity of medications to prevent MTCT as patients, especially mothers with elevated alanine amino- 38 well as the safety of both infant and mother are the upper- transferase or HBV DNA levels $5 log10 IU/mL at delivery. most considerations. In addition to the comparison of these Compared with other NAs, the number of TDF studies was factors between treatment groups and controls, it is impor- lower. This may affect the accuracy of analyses associated with this type of drug. During the analysis of factors that tant to identify the drugs with the highest efficacy and the contain a low number of studies, NA types were not distin- safest profiles for both mother and infant. As mentioned, guished. However, those with distinguished results support regardless of drug type, NAs have been shown to be beneficial the high efficacy of LdT. Indeed, in almost all the analyses, for pregnant women, while some of their side effects influence LdT was more effective in the reduction of undesirable out- both infant and mothers. comes associated with both infants and mothers but was not The results from the studies analyzed in this study showed an entirely safe drug.39 Drug-resistance is one of the most that the prevalence of positive HBsAg and/or HBV DNA is challenging issues related to the treatment of pregnant significantly lower in a newborn infant from CHB mothers who women with CHB. Interestingly, using LdT in such patients received antiviral therapy in the second or third trimester. rarely could lead to LdT-related resistance. In the reviewed Moreover, they have a greater chance to be non-HBV carriers studies, only Li et al.18 reported an HBV M204I drug resist- at 6 months. There are several studies that have reported ance mutation at the 40th week of treatment in one patient. lower immunoprophylaxis failure as the results of antiviral However, the others did not report any LdT-resistance devel- therapy during pregnancy. Some evidence is also available opment during the study periods.4,7–9,19,20 This could be that implies the roles of antiviral therapy during pregnancy in explained by the fact that short-term use of LdT is not preventing several other undesirable fetal outcomes, includ- enough to induce obvious resistance.9 Analyzing the risk of ing low birth weight, premature birth, abortion, and death. congenital malformation, no significant difference was found, Mothers may also have greater chances for the suppression of while neither LAM, LdT, nor TDF, could be presented as an HBV DNA, HBeAg loss/seroconversion, and alanine amino- utterly safe drug. Table 2 summarizes the results of previous transferase normalization. meta-analyses regarding the efficacy and safety of treat- However, there are some risks that could threaten both ments for CHB during pregnancy. infants and mothers. One of the most critical ones is the In spite of the multiple analyses conducted, this study has relatively increased but nonsignificant risk of congenital some limitations, which may affect the selection of drugs for malformations as a result of exposure to NAs. Additionally, an individual. First, it does not cover treatment and safety mothers exposed to NAs may experience more severe side predictive factors, such as positivity for HBsAg, baseline 12 effects, such as CK elevation. Recently, Brown et al. levels of HBV DNA, duration of disease, HBV genotypes, and assessed the risk of CK elevation as a result of NA therapy. so on. Second, drug resistance – a critical factor for drug In contrast to the current study, they could not find any sig- choice – was not considered. Third, only journal articles in nificant association between the CK elevation and NA therapy. English language that were indexed in PubMed and Scopus Hyun et al.13 also suggested a possible role of TDF therapy in were included in the study. The lack of analysis regarding NAs CK elevation in pregnant women but did not find any statisti- treatment duration is another limitation of the current study. cally significant association. The difference in the results In conclusion, it has been shown that NAs therapy is might be explained by the attention placed on CK during essential for pregnant women with CHB to prevent the

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 209 1 ora fCiia n rnltoa Hepatology Translational and Clinical of Journal 210

Table 2. Some of the most important reported outcomes in the selected published meta-analysis studies

HBV Fetal/ No of HBsAg DNA HBV DNA Postpartum Congenital Preterm infant Study studies positivity positivity MTCT suppression CK elevation hemorrhage malformation birth death

Hyun 10 0.87 0.16 0.23 [0.10, 8.49 [0.98, 73.28] 0.76 [0.27, 2.16] 1.60 [0.30, 2.39 1.28 et al.13 [0.31, [0.07, 0.52] 8.47] [0.84, [0.20, 2.40] 0.39] 6.81] 8.25] Chen 5 0.16 0.21 [0.07, 254.46 9.56 [1.17, 78.09] 0.73 [0.26, 2.07] 1.85 [0.42, 2.35 1.54 et al.49 [0.07, 0.61] [28.39, 8.18] [0.80, [0.25, 0.37] 2280.79] 6.94] 9.56] Shi 10 0.38 0.22 HBsAg: 0.31 et al.50 [0.15, [0.12, [0.15, 0.63] 0.94] 0.40] HBV DNA: 0.20 [0.10, 0.39] Lu 7 HBsAg:0.09 87.96 7.71 [1.99, 29.80] et al.51 [0.04, 0.22] [17.03, HBV DNA: 454.32] 0.07 [0.02, 0.22] aiS. Sali

2019 Brown 26 0.26 0.31 HBsAg: 0.3 LAM: 57.1 Not reported but Not reported but 0.88 [0.21, 0.73 et al.12 [0.16, [0.20, [0.2, 0.4] [3.5, the difference is the difference is 3.62] [0.35, 0.44] 0.49] HBV DNA: 921.4] not significant not significant 1.53] al et o.7|197 | 7 vol. 0.3 [0.2, LdT: RR 5 0.5] 52.8 [10.7, pregnancy during HBV chronic Treating : 261.8] TDF: 45.4

– [9.3, 212 222.5]

Data of the Hyun et al.,13 Chen et al.,49 Shi et al.,50 and Lu et al.51 studies were reported according to OR [95% confidence interval]. Data of the study conducted by Brown et al.12 were reported according to RR [95% confidence interval]. Abbreviations: CK, creatine kinase; HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; LAM, lamivudine; LdT, telbivudine; MTCT, mother-to-child transmission; OR, odds ratio; RR, relative risk; TDF, tenofovir. Sali S. et al: Treating chronic HBV during pregnancy

MTCT of HBV as well as to decrease various undesirable infant [14] Mantel N, Haenszel W. Statistical aspects of the analysis of data from retro- outcomes. However, mothers should be warned of the possi- spective studies of disease. J Natl Cancer Inst 1959;22:719–748. [15] Nguyen V, Tan PK, Greenup AJ, Glass A, Davison S, Samarasinghe D, et al. ble risk of elevated CK. Based on the findings, LdT therapy is Anti-viral therapy for prevention of perinatal HBV transmission: extending more effective than others, while more studies on TDF, which therapy beyond birth does not protect against post-partum flare. Aliment has a high barrier to resistance, are needed to clarify TDF Pharmacol Ther 2014;39:1225–1234. doi: 10.1111/apt.12726. efficacy and safety. [16] Tekin Koruk S, Batirel A, Kose S, Cetin Akhan S, Aygen B, Tulek N, et al. Evaluation of hepatitis B virus transmission and antiviral therapy among hepatitis B surface antigen-positive pregnant women. J Obstet Gynaecol Res 2015;41:1870–1876. doi: 10.1111/jog.12821. Conflict of interest [17] Tan Z, Yin Y, Zhou J, Wu L, Xu C, Hou H. Telbivudine treatment of hepatitis B virus-infected pregnant women at different gestational stages for the pre- The authors have no conflict of interests related to this vention of mother-to-child transmission: Outcomes of telbivudine treatment publication. during pregnancy. Medicine (Baltimore) 2016;95:e4847. doi: 10.1097/MD. 0000000000004847. [18] Liu Y, Wang M, Yao S, Yuan J, Lu J, Li H, et al. Efficacy and safety of telbivudine in different trimesters of pregnancy with high viremia for interrupting Author contributions perinatal transmission of hepatitis B virus. Hepatol Res 2016;46:E181– E188. doi: 10.1111/hepr.12525. Conceptualized the study (ST, SS), performed the literature [19] Hu Y, Xu C, Xu B, Hu L, Liu Q, Chen J, et al. Safety and efficacy of telbivudine search and data analysis (AM, MGA, SEB, MA, HSZ, SS), in late pregnancy to prevent mother-to-child transmission of hepatitis B virus: A multicenter prospective cohort study. J Viral Hepat 2018;25:429– drafted the manuscript (ST), critically revised the work (SS, 437. doi: 10.1111/jvh.12834. MD, ST). [20] Sun W, Zhao S, Ma L, Hao A, Zhao B, Zhou L, et al. Telbivudine treatment started in early and middle pregnancy completely blocks HBV vertical transmission. BMC Gastroenterol 2017;17:51. doi: 10.1186/s12876-017-0608-7. References [21] Yi W, Li MH, Xie Y, Wu J, Hu YH, Zhang D, et al. Prospective cohort study on the efficacy and safety of telbivudine used throughout pregnancy in blocking [1] World Health Organization. Guidance on prevention of viral hepatitis B and C mother-to-child transmission of hepatitis B virus. J Viral Hepat 2017;24 among people who inject drugs. 2012. Available from: https://apps.who. Suppl 1:49–56. doi: 10.1111/jvh.12788. int/iris/bitstream/handle/10665/75357/9789241504041_eng.pdf;jsessio- [22] Chen ZX, Gu GF, Bian ZL, Cai WH, Shen Y, Hao YL, et al. Clinical course and nid=62031292140D0CB41D0AFF31837AC872?sequence=1. Accessed July perinatal transmission of chronic hepatitis B during pregnancy: A real-world 2012. prospective cohort study. J Infect 2017;75:146–154. doi: 10.1016/j.jinf. [2] Das S, Ramakrishnan K, Behera SK, Ganesapandian M, Xavier AS, 2017.05.012. Selvarajan S. Hepatitis B vaccine and immunoglobulin: Key concepts. J Clin [23] Han LF, Zheng JM, Zheng LQ, Gao HB, Chen LX, Xu QL, et al. Telbivudine can Transl Hepatol 2019;7:165–171. doi: 10.14218/JCTH.2018.00037. safely reduce mother-to-child transmission in chronic hepatitis B women [3] Hu P, Shang J, Zhang W, Gong G, Li Y, Chen X, et al. HBsAg loss with peg- after 12 weeks of gestation. BMC Infect Dis 2019;19:614. doi: 10. interferon alfa-2a in hepatitis B patients with partial response to nucleos(t) 1186/s12879-019-4250-6. ide analog: New switch study. J Clin Transl Hepatol 2018;6:25–34. doi: 10. [24] Liu J, Wang J, Yan T, Du D, Qi C, Cao F, et al. Efficacy and safety of telbivudine 14218/JCTH.2017.00072. and tenofovir disoproxil fumarate in preventing hepatitis B vertical transmis- [4] Pan CQ, Han GR, Jiang HX, Zhao W, Cao MK, Wang CM, et al. Telbivudine sion: A real-life practice. J Viral Hepat 2019. doi: 10.1111/jvh.13156. prevents vertical transmission from HBeAg-positive women with chronic [25] Sheng Q, Ding Y, Li B, Han C, Li Y, Zhang C, et al. Efficacy and safety of hepatitis B. Clin Gastroenterol Hepatol 2012;10:520–526. doi: 10.1016/j. nucleos(t)ide analogues to prevent hepatitis B virus mother-to-child trans- cgh.2012.01.019. mission in pregnant women with high viremia: real life practice from China. [5] Xu WM, Cui YT, Wang L, Yang H, Liang ZQ, Li XM, et al. Lamivudine in late Int J Med Sci 2018;15:796–801. doi: 10.7150/ijms.25047. pregnancy to prevent perinatal transmission of hepatitis B virus infection: a [26] Zeng J, Zheng C, Li H. Effectiveness of tenofovir or telbivudine in preventing multicentre, randomized, double-blind, placebo-controlled study. J Viral HBV vertical transmission for pregnancy. Medicine (Baltimore) 2019;98: Hepat 2009;16:94–103. doi: 10.1111/j.1365-2893.2008.01056.x. e15092. doi: 10.1097/MD.0000000000015092. [6] Han GR, Cao MK, Zhao W, Jiang HX, Wang CM, Bai SF, et al. A prospective and [27] Celen MK, Mert D, Ay M, Dal T, Kaya S, Yildirim N, et al. Efficacy and safety of open-label study for the efficacy and safety of telbivudine in pregnancy for tenofovir disoproxil fumarate in pregnancy for the prevention of vertical the prevention of perinatal transmission of hepatitis B virus infection. J transmission of HBV infection. World J Gastroenterol 2013;19:9377–9382. Hepatol 2011;55:1215–1221. doi: 10.1016/j.jhep.2011.02.032. doi: 10.3748/wjg.v19.i48.9377. [7] Zhang H, Pan CQ, Pang Q, Tian R, Yan M, Liu X. Telbivudine or lamivudine use [28] Greenup AJ, Tan PK, Nguyen V, Glass A, Davison S, Chatterjee U, et al. in late pregnancy safely reduces perinatal transmission of hepatitis B virus in Efficacy and safety of tenofovir disoproxil fumarate in pregnancy to real-life practice. Hepatology 2014;60:468–476. doi: 10.1002/hep.27034. prevent perinatal transmission of hepatitis B virus. J Hepatol 2014;61: [8] Han GR, Jiang HX, Yue X, Ding Y, Wang CM, Wang GJ, et al. Efficacy and safety 502–507. doi: 10.1016/j.jhep.2014.04.038. of telbivudine treatment: an open-label, prospective study in pregnant [29] Pan CQ, Duan Z, Dai E, Zhang S, Han G, Wang Y, et al. Tenofovir to prevent women for the prevention of perinatal transmission of hepatitis B virus infec- hepatitis B transmission in mothers with high viral load. N Engl J Med 2016; tion. J Viral Hepat 2015;22:754–762. doi: 10.1111/jvh.12379. 374:2324–2334. doi: 10.1056/NEJMoa1508660. [9] Wu Q, Huang H, Sun X, Pan M, He Y, Tan S, et al. Telbivudine prevents vertical [30] Samadi Kochaksaraei G, Castillo E, Osman M, Simmonds K, Scott AN, transmission of hepatitis B virus from women with high viral loads: a pro- Oshiomogho JI, et al. Clinical course of 161 untreated and tenofovir- spective long-term study. Clin Gastroenterol Hepatol 2015;13:1170–1176. treated chronic hepatitis B pregnant patients in a low hepatitis B virus doi: 10.1016/j.cgh.2014.08.043. endemic region. J Viral Hepat 2016;23:15–22. doi: 10.1111/jvh.12436. [10] Brown RS Jr, Verna EC, Pereira MR, Tilson HH, Aguilar C, Leu CS, et al. [31] Wakano Y, Sugiura T, Endo T, Ito K, Suzuki M, Tajiri H, et al. Antiviral therapy Hepatitis B virus and human immunodeficiency virus drugs in pregnancy: for hepatitis B virus during second pregnancies. J Obstet Gynaecol Res 2018; findings from the Antiretroviral Pregnancy Registry. J Hepatol 2012;57: 44:566–569. doi: 10.1111/jog.13540. 953–959. doi: 10.1016/j.jhep.2012.06.031. [32] Chen HL, Lee CN, Chang CH, Ni YH, Shyu MK, Chen SM, et al. Efficacy of [11] Nachega JB, Uthman OA, Mofenson LM, Anderson JR, Kanters S, Renaud F, maternal tenofovir disoproxil fumarate in interrupting mother-to-infant et al. Safety of tenofovir disoproxil fumarate-based antiretroviral therapy transmission of hepatitis B virus. Hepatology 2015;62:375–386. doi: 10. regimens in pregnancy for HIV-infected women and their infants: A system- 1002/hep.27837. atic review and meta-analysis. J Acquir Immune Defic Syndr 2017;76:1–12. [33] Chang KC, Chang MH, Lee CN, Chang CH, Wu JF, Ni YH, et al. Decreased doi: 10.1097/QAI.0000000000001359. neonatal hepatitis B virus (HBV) viremia by maternal tenofovir treatment pre- [12] Brown RS Jr, McMahon BJ, Lok AS, Wong JB, Ahmed AT, Mouchli MA, et al. dicts reduced chronic HBV infection in children born to highly viremic mothers. Antiviral therapy in chronic hepatitis B viral infection during pregnancy: A Aliment Pharmacol Ther 2019;50:306–316. doi: 10.1111/apt.15321. systematic review and meta-analysis. Hepatology 2016;63:319–333. doi: [34] Jourdain G, Ngo-Giang-Huong N, Harrison L, Decker L, Khamduang W,Tierney C, 10.1002/hep.28302. et al. Tenofovir versus placebo to prevent perinatal transmission of hepatitis B. N [13] Hyun MH, Lee YS, Kim JH, Je JH, Yoo YJ, Yeon JE, et al. Systematic review Engl J Med 2018;378:911–923. doi: 10.1056/NEJMoa1708131. with meta-analysis: the efficacy and safety of tenofovir to prevent mother- [35] Lin Y, Liu Y, Ding G, Touqui L, Wang W, Xu N, et al. Efficacy of tenofovir in to-child transmission of hepatitis B virus. Aliment Pharmacol Ther 2017;45: preventing perinatal transmission of HBV infection in pregnant women with 1493–1505. doi: 10.1111/apt.14068. high viral loads. Sci Rep 2018;8:15514. doi: 10.1038/s41598-018-33833-w.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 211 Sali S. et al: Treating chronic HBV during pregnancy

[36] Su GG, Pan KH, Zhao NF, Fang SH, Yang DH, Zhou Y. Efficacy and safety of hepatitis B virus. Eur J Clin Microbiol Infect Dis 2012;31:2211–2218. doi: lamivudine treatment for chronic hepatitis B in pregnancy. World J Gastro- 10.1007/s10096-012-1557-2. enterol 2004;10:910–912. doi: 10.3748/wjg.v10.i6.910. [45] Ayres A, Yuen L, Jackson KM, Manoharan S, Glass A, Maley M, et al. Short [37] Daniels D, Grytdal S, Wasley A. Surveillance for acute viral hepatitis - United duration of lamivudine for the prevention of hepatitis B virus transmission in States, 2007. MMWR Surveill Summ 2009;58:1–27. pregnancy: lack of potency and selection of resistance mutations. J Viral [38] Yi W, Pan CQ, Li MH, Wan G, Lv YW, Liu M, et al. The characteristics and Hepat 2014;21:809–817. doi: 10.1111/jvh.12212. predictors of postpartum hepatitis flares in women with chronic hepatitis B. [46] Pan CQ, Yi W, Liu M, Wan G, Hu YH, Zhou MF. Lamivudine therapy during the Am J Gastroenterol 2018;113:686–693. doi: 10.1038/s41395-018-0010-2. second vs the third trimester for preventing transmission of chronic hepatitis B. [39] Tavakolpour S, Darvishi M, Mirsafaei HS, Ghasemiadl M. Nucleoside/nucleo- J Viral Hepat 2017;24:246–252. doi: 10.1111/jvh.12640. tide analogues in the treatment of chronic hepatitis B infection during preg- [47] Sheng QJ, Wang SJ, Wu YY, Dou XG, Ding Y. Hepatitis B virus serosurvey and nancy: a systematic review. Infect Dis (Lond) 2018;50:95–106. doi: 10. awareness of mother-to-child transmission among pregnant women in She- 1080/23744235.2017.1384957. nyang, China: An observational study. Medicine (Baltimore) 2018;97: [40] Li XM, Yang YB, Hou HY, Shi ZJ, Shen HM, Teng BQ, et al. Interruption of HBV e10931. doi: 10.1097/MD.0000000000010931. intrauterine transmission: a clinical study. World J Gastroenterol 2003;9: [48] Foaud HM, Maklad S, Gmal El Din A, Mahmoud F. Lamivudine use in pregnant 1501–1503. doi: 10.3748/wjg.v9.i7.1501. HBsAg-females effectively reduces maternal viremia. Arab J Gastroenterol [41] van Zonneveld M, van Nunen AB, Niesters HG, de Man RA, Schalm SW, 2019;20:8–13. doi: 10.1016/j.ajg.2019.02.003. Janssen HL. Lamivudine treatment during pregnancy to prevent perinatal [49] Chen JZ, Liao ZW, Huang FL, Su RK, Wang WB, Cheng XY, et al. Efficacy and transmission of hepatitis B virus infection. J Viral Hepat 2003;10:294–297. safety of tenofovir disoproxil fumarate in preventing vertical transmission of doi: 10.1046/j.1365-2893.2003.00440.x. hepatitis B in pregnancies with high viral load. Sci Rep 2017;7:4132. doi: 10. [42] Ni YH, Huang FC, Wu TC, Kong MS, Jeng YM, Chen PJ, et al. Lamivudine 1038/s41598-017-04479-x. treatment in maternally transmitted chronic hepatitis B virus infection [50] Shi Z, Yang Y, Ma L, Li X, Schreiber A. Lamivudine in late pregnancy to patients. Pediatr Int 2005;47:372–377. doi: 10.1111/j.1442-200x.2005. interrupt in utero transmission of hepatitis B virus: a systematic review 02101.x. and meta-analysis. Obstet Gynecol 2010;116:147–159. doi: 10. [43] Yu M, Ji Y, Jiang H, Ju L, Wu K, Kan N. Efficacy of peripartum antiviral treat- 1097/AOG.0b013e3181e45951. ment for hepatic failure due to hepatitis B virus. Int J Gynaecol Obstet 2011; [51] Lu YP, Liang XJ, Xiao XM, Huang SM, Liu ZW, Li J, et al. Telbivudine during the 114:33–36. doi: 10.1016/j.ijgo.2011.03.004. second and third trimester of pregnancy interrupts HBV intrauterine trans- [44] Yu M, Jiang Q, Ji Y, Jiang H, Wu K, Ju L, et al. The efficacy and safety mission: a systematic review and meta-analysis. Clin Lab 2014;60:571– of antiviral therapy with lamivudine to stop the vertical transmission of 586. doi: 10.7754/clin.lab.2013.130408.

212 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 197–212 Original Article

Efficacy and Safety of All-oral, 12-week Ravidasvir Plus Ritonavir-boosted Danoprevir and Ribavirin in Treatment-naïve Noncirrhotic HCV Genotype 1 Patients: Results from a Phase 2/3 Clinical Trial in China

Xiaoyuan Xu1, Bo Feng2, Yujuan Guan3, Sujun Zheng4, Jifang Sheng5, Xingxiang Yang6, Yuanji Ma7, Yan Huang8, Yi Kang9, Xiaofeng Wen10, Jun Li11, Youwen Tan12, Qing He13, Qing Xie14, Maorong Wang15, Ping An16, Guozhong Gong17, Huimin Liu18, Qin Ning19, Rui Hua20, Bo Ning21, Wen Xie22, Jiming Zhang23, Wenxiang Huang24, Yongfeng Yang25, Minghua Lin26, Yingren Zhao27, Yanhong Yu28, Jidong Jia29, Dongliang Yang30, Liang Chen31, Yinong Ye32, Yuemin Nan33, Zuojiong Gong34, Quan Zhang35, Peng Hu36, Fusheng Wang37, Yongguo Li38, Dongliang Li39, Zhansheng Jia40, Jinlin Hou41, Chengwei Chen42, Jinzi J. Wu43 and Lai Wei*44

1Peking University First Hospital, Beijing, China; 2Peking University People’s Hospital, Beijing, China; 3Guangzhou Eighth People’s Hospital, Guangzhou, China; 4Beijing YouAn Hospital, Capital Medical University, Beijing, China; 5The First Affiliated Hospital of Medical School of Zhejiang University, Hangzhou, China; 6Sichuan Provincial People’s Hospital, Chengdu, China; 7West China Hospital, Sichuan University, Chengdu, China; 8Xiangya Hospital, Central South University, Changsha, China; 9Henan Provincial People’s Hospital, Zhengzhou, China; 10Liuzhou People’s Hospital, Liuzhou, China; 11Jiangsu Province Hospital, Nanjing, China; 12Zhenjiang No.3 People’s Hospital, Zhenjiang, China; 13The Third People’s Hospital of Shenzhen, Shenzhen, China; 14Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China; 15People’s Liberation Army Bayi Hospital, Nanjing, China; 16Shenyang Sixth People’s Hospital, Shenyang, China; 17The 2nd Xiangya Hospital of Central South University, Changsha, China; 18Xixi Hospital of Hangzhou, Hangzhou, China; 19Tongji Medical College of Huazhong University of Science & Technology, Wuhan, China; 20The First Hospital of Jilin University, Changchun, China; 21Baoji Center Hospital, Baoji, China; 22Beijing Ditan Hospital, Beijing, China; 23Huashan Hospital Affiliated to Fudan University, Shanghai, China; 24Chongqing Medical University No.1 Affiliated Hospital, Chongqing, China; 25The Second Hospital of Nanjing, Nanjing, China; 26Fujian Fuzhou Municipal Infectious Disease Hospital, Fuzhou, China; 27No.1 Affiliated Hospital of Xi’an Jiaotong University, Xi’an, China; 28The First Affiliated Hospital of Guangxi Medical University, Nanning, China; 29Beijing Friendship Hospital, Capital Medical University, Beijing, China; 30Wuhan Union Hospital, Wuhan, China; 31Shanghai Public Health Clinical Center, Shanghai, China; 32Foshan No.1 People’s Hospital, Foshan, China; 33The Third Hospital of Hebei Medical University, Shijiazhuang, China; 34Wuhan University Renmin Hospital, Wuhan, China; 35The Affiliated Hospital of Guizhou Medical University, Guiyang, China; 36The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China; 37PLA 302 Hospital, Beijing, China; 38The First Affiliated Hospital of Harbin Medical University, Harbin, China; 39Fuzhou General Hospital of Nanjing Military Command, Fuzhou, China; 40Tang Du Hospital, Fourth military Medical University, Xi’an, China; 41Nanfang Hospital, Nanfang Medical University, Guangzhou, China; 42The 85 branch of the Chinese People’s Liberation Army Hospital, Shanghai, China; 43Ascletis BioScience Co., Ltd. Hangzhou, China; 44Tsinghua Changgeng Hospital, Beijing, China

Abstract confirming the efficacy and safety of RDV + ritonavir-boosted danoprevir + ribavirin for 12 weeks in treatment-naïve non- Background and Aims: Ravidasvir (RDV) is a new genera- cirrhotic patients with genotype 1 infection in a large population pangenotypic hepatitis C virus (HCV) NS5A inhibitor, with tion. Methods: In this multicenter, randomized, double- high barrier to baseline resistance-associated species. This blinded, placebo-controlled phase 2/3 trial (NCT03362814), is the first phase 2/3 study conducted in Mainland China we enrolled 424 treatment-naïve, noncirrhotic adult HCV genotype 1 patients. All patients were randomized at 3:1

Keywords: Ravidasvir; Danoprevir; HCV; SVR; Treatment-naïve; Noncirrhotic; ratio to receive a combination of RDV 200mg once daily plus GT1. ritonavir-boosted danoprevir 100mg/100mg twice daily and Abbreviations: AE, adverse event; ALT, alanine aminotransferase; AST, aspar- oral ribavirin 1000/1200mg/day (body weight <75/$75 kg) tate aminotransferase; CI, confidence interval; DAA, direct acting antiviral (n = 318) or placebo (n = 106) for 12 weeks. The primary therapy; DNVr, ritonavir-boosted danoprevir; GT1, genotype 1; HCV, hepatitis C virus; RBV, ribavirin; RDV, ravidasvir; SAE, serious adverse event; SVR12, sus- end-point was the rate of sustained virologic response tained virological response at 12 weeks after treatment. 12 weeks after the end of treatment, and the safety was Received: 30 July 2019; Revised: 12 September 2019; Accepted: 15 September evaluated and compared between treatment and placebo 2019 groups. Results: The overall rate of sustained virological *Correspondence to: Lai Wei, Tsinghua Changgeng Hospital, No. 168 Li Tang Road, Chang Ping District, Beijing 102218, China. Tel: +86-10-88326666, Fax: response at 12 weeks after treatment is 99% (306/309, +86-10-68318386, E-mail: [email protected] 95%, CI: 97%–100%) under per protocol set analysis.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 213–220 213

All patients harboring baseline NS5A resistance-associated biopsy indicating no cirrhosis [metavir score #3]); plasma species in the treatment group (76/76, per protocol set) HCV RNA $1 3 10,000 IU/mL; and never received antiviral achieved sustained virological response at 12 weeks after treatment for HCV infection. treatment. No treatment-related serious adverse events were reported. Laboratory abnormalities showed mild or Exclusion criteria moderate severity (grade 1 and grade 2) in liver function tests. Conclusions: In treatment-naïve, noncirrhotic HCV Mixed HCV genotypes; HCV genotypes other than GT-1; HBV Chinese patients infected with HCV genotype 1, all-oral reg- or HIV co-infection; alpha-fetoprotein >100 ng/mL; presence imen of RDV + ritonavir-boosted danoprevir + ribavirin for or suspected of other chronic liver illness or hepatocellular 12 weeks was highly efficacious, safe, and well tolerated. carcinoma; presence of cirrhosis as assessed by Fibroscan® Citation of this article: Xu X, Feng B, Guan Y, Zheng S, (>12.9 kPa) or liver biopsy (metavir score at 4); body mass Sheng J, Yang X, et al. Efficacy and safety of all-oral, 12-week index <18 or $30 kg/m2; anti-HAV (IgM), HBsAg, anti-HEV ravidasvir plus ritonavir-boosted danoprevir and ribavirin in (IgM) or anti-HIV positivity; ANC < 1.5 3 109/L; platelet treatment-naïve noncirrhotic HCV genotype 1 patients: Re- <100 3 109/L; hemobilirubin <110 g/L (female) or <120 g/L sults from a phase 2/3 clinical trial in China. J Clin Transl Hep- (male); international normalized ratio >1.5; alanine amino- atol 2019;7(3):213–220. doi: 10.14218/JCTH.2019.00033. transferase (ALT) or aspartate aminotransferase (AST) $5*ULN; total bilirubin $2*ULN (direct bilirubin $35% total bilirubin ); or creatinine $1.5*ULN. Introduction Study design In China, more than 10 million people are estimated to be chronically infected with hepatitis C virus (HCV),1 with geno- Eligible subjects were enrolled and randomly assigned at a type 1 (GT1) b as the most prevalent sub-genotype 3:1 ratio to active treatment (the treatment group) or placebo (56.8%).2 Currently, approximately 50–75% of the patients (the placebo group) arms (Fig. 1). During the double-blinded infected with GT1 HCV are able to achieve a sustained viro- phase, subjects in the treatment arm received a combination logical response (SVR) after treatment with a combination of of RDV at 200 mg once daily plus DNVr at 100 mg/100 mg pegylated-interferon plus ribavirin (RBV) for 48 weeks, the twice daily and oral RBV at 1000/1200 mg/day (1000 mg current standard of care for chronic HCV infection in China,3 daily if the body weight was <75 kg and 1200 mg daily if leaving strong needs for new direct acting anti-viral therapy the body weight was $75 kg) for 12 weeks. Subjects in the (DAA) with higher efficacy as warranted. placebo arm received placebo during the double-blinded Ravidasvir (RDV), a potent pangenotypic NS5A inhibitor, phase and entered into a deferred open-labelled active treat- has a favorable pharmacokinetic profile, with rapid absorp- ment arm for another 12 weeks. All subjected were followed tion, stable plasma distribution and high plasma trough up for 24 weeks after the end of active treatment (Fig. 1). concentration, allowing for continuous HCV inhibition with once-daily oral dosing.4 Danoprevir (DNV) is a potent, macro- Study oversight cyclic inhibitor of the HCV NS3/4A serine protease5,6 that can be co-administered with low-dose ritonavir to achieve thera- All the subjects provided written informed consent. The peutic trough concentrations throughout the dosing interval. protocol of this study conformed to the International Confer- Combination of RDV and DNV may serve as a potent regimen ence on Harmonisation guidelines, applicable regulations, against HCV infection. and ethical guidelines of the Declaration of Helsinki. The As reported in the phase 2 EVEREST study, all-oral RDV + research was approved by the ethical committee of each ritonavir-boosted danoprevir (DNVr) + RBV regimen has participating site and was registered at clinicaltrials.gov (trial demonstrated sustained virological response at 12 weeks registration: NCT03362814). All authors had full access to the after treatment (SVR12) at 100% (38/38) in treatment- database and vouch for the completeness and accuracy of the naïve noncirrhotic Taiwanese patients with HCV GT1 infection, data and analyses as presented. and was safe and well tolerated.7 This phase 2/3 study was designed to investigate the safety and efficacy of this combi- Efficacy assessments nation regimen in treatment-naïve HCV GT1 patients without cirrhosis in a larger population from Mainland China. Vital signs were measured and samples were collected and stored at screening and on study visits at weeks 1, 2, 4, 8, and Methods 12 and at post-treatment weeks 4, 12 and 24 for analyses biochemistry, hematology, viral load, and urinalysis. Patients The lower limit of quantitation of HCV RNA was defined as 15 IU/mL, which was determined by using CAP/CTM HCV 2.0 Patients with confirmed chronic HCV infection were screened assay. The primary end-point was SVR12 after the end of the and recruited from 42 sites in different provinces across treatment. Viral breakthrough was defined as a one log China, and eligible subjects were enrolled in the study increase in HCV RNA from postbaseline nadir, or a confirmed according to the inclusion and exclusion criteria as below: positivity in HCV RNA from undetectable level. Post-treatment relapse was defined as HCV RNA returning positivity from Inclusion criteria previously achieved HCV RNA negativity before or at end of treatment. Chinese adult of $18 years-old with chronic GT1 HCV- We performed resistance analysis by using population infection; absence of cirrhosis (defined as a liver stiffness sequencing on samples obtained at baseline, when patients measurement <9.6 kPa by Fibroscan® or a screening liver had experienced virologic breakthrough during treatment, or

214 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 213–220 Xu X. et al: Efficacy of RDV + DNV regimen in GT1 CHC patients

Fig. 1. Study design. Abbreviations: RDV, ravidasvir; DNVr, ritonavir-boosted danoprevir; RBV, ribavirin. relapse during follow up after the treatment, in order to control rate, which was based on response rates among evaluate the role and impact from substitutions potentially previously untreated noncirrhotic patients who received rendering resistance to NS5A or NS3/4A inhibitors. pegylated-interferon/RBV, was determined from a meta- analysis and experts’ clinical experience. Safety assessment All statistical tests and 95% CIs were two-sided, and p values <0.05 were considered statistically significant. Adverse events (AEs) were assessed at each study visit and Descriptive statistics for quantitative data used are reported documented for all patients who received at least one dose of as mean ± standard deviation and median (interquartile medications. Data for all treatment-emergent AEs, laboratory range), while qualitative data is summarized as counts and abnormalities, clinically significant changes of vital signs and percentages. Data analysis was performed using SAS Propri- electrocardiogram testing were obtained from the start of etary Software 9.4 (Licensed to PEKING UNIVERSITY HEALTH drug administration until 4 weeks after the end of study. AEs SCIENCE CENTER, Site 11202165). were categorized using the Medical Dictionary for Regulatory Activities. Results

Statistical comparison groups and subgroups Baseline characteristics and demographics

Analysis population included the FAS subgroup of all random- A total of 709 subjects were screened, and 425 qualified ized subjects who received at least one dose of the study drug patients were assigned to randomization. Among whom, 424 and provided at least one point of data; the per protocol (318 patients in the treatment arm and 106 in the placebo population of patients from FAS subjects who met the arm) received at least one dose of investigational medication inclusion criteria without reasons for exclusion and completed and were included in the FAS population for the primary the study treatment and follow-up; the safety population of efficacy and safety analyses. A total of 309 patients in the all randomized subjects who received at least one dose of the treatment arm and 101 patients in the placebo arm completed study drug and provided safety data for appraisal. the 12-week follow-up and fulfilled the inclusion criteria for the per protocol population (Fig. 2). The demographic and baseline Statistical methods characteristics of the study population are shown in Table 1.

Demographics and baseline characteristics were collected for Efficacy the FAS population. Efficacy analyses were performed on the FAS and per protocol set populations, and safety analyses were Overall, the SVR12 rate in the FAS analysis was 96% (306/ performed on the per protocol set population. For the primary 318, 95% CI, 94;98%) in the treatment arm and 95% (101/ efficacy end-point (SVR12), a two-sided 95% confidence 106, 95% CI, 93;100%) in the placebo arm. In the per interval (CI) was calculated using the normal approximation protocol population, the SVR12 reached 99% (306/309, 95% to the binomial distribution. Subgroup efficacy analyses of CI, 97;100%) in the treatment arm and 99% (100/101, 95% SVR12 (including stratification by genotype, baseline HCV RNA CI, 95;100%) in the placebo arm (Table 2). level, IL28B genotypes, and baseline polymorphisms in NS5A) All subjects receiving RDV + DNVr + RBV had a rapid were performed for the per protocol set population. decline of serum HCV RNA levels (Fig. 3). By week 4 during To evaluate the superiority of the RDV plus DNVr and RBV the double-blinded treatment phase, the HCV RNA level regimen to the historical control in HCV GT1 patients, the was below the lower limit of quantitation (15 IU/mL) in 98% lower boundary of the 95% CI for SVR12 in the treatment and of the patients who were treated by the active regimen (303 placebo group entering deferred active treatment had to of the 309 patients in the treatment group), and at the end of exceed the calculated historical control rate of 75%. The the treatment the proportion reached 99% (306/309).

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 213–220 215 Xu X. et al: Efficacy of RDV + DNV regimen in GT1 CHC patients

Fig. 2. Consolidated Standards of Reporting Trials (known as CONSORT) flow diagram showing participants flow through each stage of the randomized controlled trial (enrollment, intervention allocation, follow-up, and data analysis). Two hundred and eighty-four patients who failed to meet the standards were excluded from this clinical study and among 425 randomized patients, one voluntarily discontinued without being administered medication. One patient from the placebo arm discontinued due to relocation overseas. Eight patients from the intervention arm discontinued and only one was considered due to an adverse event related to the active treatment.

The SVR12 subanalysis in various subgroups showed that in the placebo-controlled arm (29%, 29/101, per protocol set) the SVR12 rates were similarly high in all subgroups, includ- at baseline (for details of the resistance-associated species see ing those defined by gender (99% with male and 99% with Table 3). The most common NS5A resistance-associated female, per protocol set), IL28B genotypes (99% with CC and species detected were R30Q (38/309, 12.3%), Y93H (21/ 100% with non-CC, per protocol set), age (99% among 309, 6.8%) and R30Q + Y93H (8/309, 2.6%) in the treatment people <65 years-old and 100% among people $65 years- arm and R30Q (21/101, 20.7%), Y93H (6/101, 5.9%) and old), and initial baseline HCV RNA levels (100% with a level R30Q + Y93H (1/101, 0.99%) in the placebo-controlled arm. <400,000 IU/mL, 97% with a level of 400,000;800,000 All patients with baseline NS5A resistance-associated species IU/mL, 99% with a level $800,000 IU/mL). from both arms achieved SVR12 (Fig. 4).

Resistance analyses Treatment failure

NS5A resistance-associated species were detected in 76 Among 309 patients in the treatment arm, the 76 patients patients (25%, 76/309) in the treatment arm and 29 subjects reported with baseline NS5A resistance-associated species all

Table 1. Baseline characteristics and demographic data of subjects

Parameter Treatment arm, n = 318 Placebo arm, n = 106 p value

Female, n (%) 165 (52%) 58 (55%) 0.613 Age, median (range), years 48 (21–73) 45 (23–72) 0.01 BMI, mean 6 SD, kg/m2 23.2 6 2.8 22.8 6 2.8 0.192 HCV RNA (log10 IU/mL), median 6.31 6.22 0.475 <400,000 IU/mL, n (%) 53 (16.67%) 21 (19.81%) 400,000 IU/mL-800,000 IU/mL, n (%) 32 (10.06%) 13 (12.26%) 0.569 $ 800,000 IU/mL, n (%) 233 (73.27%) 72 (67.92%) IL28B CC genotype, n (%) 259 (81%) 89 (84%) 0.559 HCV genotype 1a, n (%) 6 (2%) 2 (2%) 1 HCV genotype 1b, n (%) 312 (98%) 104 (98%) 1

Abbreviations: BMI, body mass index; HCV, hepatitis C virus.

216 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 213–220 Xu X. et al: Efficacy of RDV + DNV regimen in GT1 CHC patients

Table 2. Sustained virologic response at post-treatment week 12 among all patients after receiving the active regimen

Group Subgroup Response/total SVR 12 95% CI Historical control with PR p value

Treatment arm FAS 306/318 96% (94%, 98%) 75% <0.001 per protocol set 306/309 99% (97%, 100%) 75% <0.001 Placebo arm FAS 101/106 95% (93%, 100%) 75% <0.001 100/101 99% (95%, 100%) 75% <0.001

* SVR12, sustained response rate at post-treatment week 12 was defined as HCV RNA <15 IU/mL (lower limit of quantitation).

(76/76, per protocol set) achieved SVR12 showing potency of reported were anemia (in 40% of the subjects in the treat- this regimen. Three patients (0.97%) experienced a virologic ment arm and 5% in the placebo arm) and upper respiratory breakthrough during the treatment phase, and no patient tract infection (in 21% of the subjects in the treatment arm relapsed by post-treatment week 12. Serum samples from and 21% in the placebo arm). Hemobilirubin elevation (16% two of the three subjects who experienced virologic break- vs. 6%) was also seen among the patients with the most through showed positivity for resistance-associated species frequent treatment-related AEs that occurred but was in that confer resistance to both NS3/4A and NS5A. The <10% of subjects (Table 4). resistance-associated species were Q80k, V36M, R155K in Discontinuation due to any AE occurred in one patient in the the protein NS3/4A, and Q30K, Q30H, H58D in the region treatment arm, owing to drug allergy. One death was reported NS5A. The third patient with virologic breakthrough had no due to traffic accident but was not considered treatment- detectable RAS at both baseline and on rise, but the genotype related. Most of the liver function abnormalities were mild or turned out as genotype-2a instead of GT1b as initially moderate (grade 1 or 2), and no grade 4 abnormalities were diagnosed. This case could not be ruled out from potential observed. In the treatment arm, four of the three hundred- mechanisms such as a misdiagnosed genotype on enroll- eighteen patients had grade 3 treatment-related elevated total ment, viral superinfection with GT1b dominance, or re- bilirubin level and one had grade 3 elevated ALT. In the placebo infection with genotype-2a during treatment. In the placebo arm, three patients experienced grade 3 ALTelevation and two arm, one subject relapsed and showed resistance-associated with AST elevation, respectively. species R155K in the NS3/4A region and L31V in the NS5A region. Discussion

Safety and tolerability A cross-sectional observational study conducted in 2014 in China revealed HCV GT1b to be the most prevalent subtype at During the double-blinded phase, 94% of the subjects in the 56.8% and to be associated with low response to traditional treatment arm (298/318), experienced adverse events (AEs) treatment with pegylated-interferon plus RBV for 48 weeks. compared to 79% of those in the placebo arm (84/106). The Recently, some DAA regimens were approved in China but majority of AEs were mild. Only 15 of the 318 subjects in the with pre-requisition according to results from a baseline resist- treatment arm (15/318, 5%) and 9 subjects (9/103, 9%) in ance-associated species test. Baseline resistance-associated the placebo arm had grade 3 treatment-related AEs. The species in HCV are somehow prevalent and may pose a laboratory abnormality tests revealed evaluated liver func- necessity of intervention in either diagnosis or treatment. tions (ALT, AST, total bilirubin, direct bilirubin) in the treat- Hence, there is a strong desire for a new treatment option ment arm were mostly of grade 1 or 2 and no grade 4 AEs with higher cure rate and good resistance barrier profile. In this occurred. No treatment-related severe AEs or death occurred randomized, double-blinded, placebo-controlled phase 2/3 during the full treatment course. The most common AEs trial, we demonstrated the excellent efficacy of an interferon-

Fig. 3. Mean of HCV RNA log10 IU/mL to the end of follow-up (per protocol set). When HCV RNA <15 IU/mL or reported as target not detected, HCV RNA was counted as 1 IU/mL, or log101 IU/mL in the Figure.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 213–220 217 Xu X. et al: Efficacy of RDV + DNV regimen in GT1 CHC patients

Table 3. Baseline resistance-associated substitution analysis of HCV regimen, hence confirming the superiority hypothesis. Sensitiv- NS5A (per protocol set) ity analysis also reached a consistent conclusion. Baseline with NS5A Baseline resistance-associated substitutions (of resist- resistance-Associated Treatment Placebo arm, ance-associated species) in HCV NS5A are not uncommon species arm, n = 309 n = 101 in Chinese GT1b patients and may account for up to 19% prevalence with detrimental response in certain patients.10 R30Q, n (%) 38 (12.3%) 21 (20.8%) Heretofore, the first generation NS5A inhibitors (such as Y93H, n (%) 21 (6.8%) 6 (5.9%) daclatasvir, ledipasvir, and ombitasvir) have relatively lower NS5A resistance barriers and share cross-drug resist- R30Q + Y93H, n (%) 8 (2.6%) 1 (1.0%) ance, which undermines the efficacy of DAA therapies.11 L28M, n (%) 3 (1.0%) / According to drug labels and guidelines, the NS5A resist- R30H, n (%) 1 (0.3%) / ance-associated species test is recommended for elbasvir/ grazoprevir in GT1a, for daclatasvir plus asunaprevir in Y93C, n (%) 1 (0.3%) / GT1b, and even for sofosbuvir/velpatasvir treatment in R30Q + L31M, n (%) 1 (0.3%) / genotype 3 treatment-naive patients with cirrhosis and L28M + Y93C, n (%) 1 (0.3%) / treatment-experienced patients (with or without cirrhosis).12 Nonetheless, consensus from clinicians is still the L28M + Y93H, n (%) 1 (0.3%) / simplified testing as the preferable choice. Different from L28Q, n (%) 1 (0.3%) 1 (1.0%) the NS5A inhibitors aforementioned, RDV is a pangenotypic Total, n (%) 76 (24.6%) 29 (28.7%) HCV NS5A inhibitor with high resistance barrier and more potent antiviral activity in accordance with a higher administration dose at 200 mg per day boosting to garner greater exposure.4 Besides, the wild-type HCV virus, conversely, in free, all-oral antiviral therapy in treatment naïve, noncirrhosis, a previous study of RDV also demonstrated its potency GT1 subjects, without resistance-associated species test before to fully inhibit single NS5A substitution variants.13 In treatment initiation. The SVR12 rate in the treatment arm was this study, GT1 NS5A resistance-associated species 96% (306/318) in the FAS and 99% (306/309) the per protocol were detected in 76 of the participating subjects (24.6%, set models, respectively. The SVR12 for the placebo-controlled 76/309) in the treatment arm and 29 in the control arm arm was 95% (101/106) in the FAS and 99% (100/101) the per (29%, 29/101, per protocol set) at baseline (Fig. 2). All sub- protocol set sets. jects with baseline NS5A resistance-associated species from In the interferon and RBV era, SVR was associated with both arms achieved SVR12. These results support a clinical several host-, viral- and drug-related characteristics, such as use of this regimen without posing a baseline NS5A resist- age, sex, baseline viral load, treatment duration, etc.8,9 Our ance-associated species test in the studied population, results using subanalysis covering these factors showed which was treatment naïve, noncirrhotic, and GT1 HCV there was little impact from age, gender, IL28B genotypes, subjects. and baseline viral load, and the efficacy was similarly high As to safety, the RDV + DNVr + RBV combination was well among all subgroups. The traditional negative predictors herein tolerated in the studied cohort in this trial. Most AEs were mild no longer delineate different response in SVR12 rates, whereas or moderate in intensity and were considered less relevant to the potency of the investigational drug combination was demon- investigational medications, which is concordant to results strated concordantly. The lower boundary of the 95% CI for the reported from the previous phase 2 studies using the same SVR12 rates in both arms were marginally higher than the his- regimen. Severe AEs occurred in 2.8% (12/424) of the study torical control value (75%) from the pegylated-interferon + RBV population (7 in the treatment arm and 5 in the placebo arm), whereas none was assessed as related to study medication per discretion of the investigators. One patient died during the study due to a traffic accident and this event was considered not related to study drugs. One patient discontinued due to drug allergy. Most of the laboratory abnormalities in liver function were of mild to moderate severity (grade 1 and grade 2). Presence of the placebo arm in the study design hereby allowed for comparison of the safety profile with the active treatment arm. The resultant safety profile in the overall population was quite similar and comparable between the two arms, with exception of more anemia from the treatment arm which was not surprising due to the use of RBV. In pharmacology, both RDV and danoprevir are pangenotypic anti-HCV inhibitors,4,5 possessing potent antiviral activities against HCV GT1 through genotype 6 species from in vitro and in vivo results. In phase 1 trials, RDV monotherapy showed potent antiviral activity and excellent safety profile in treatment-naïve subjects infected with GT1, genotype 2, and Fig. 4. Virological response of patients with baseline resistance-associated genotype 3.12 species. The strength of this study is its being the first large scale, Abbreviation: SVR12, Sustained virologic response at post-treatment double-blinded, randomized, multicenter trial that confirms week 12. the efficacy and safety profiles of the new RDV + DNVr + RBV

218 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 213–220 Xu X. et al: Efficacy of RDV + DNV regimen in GT1 CHC patients

Table 4. Adverse Events and Laboratory Abnormalities During the Double-Blinded Treatment Phase (FAS)

Patients, n (%) Treatment arm, n = 318 Placebo arm, n = 106

Any AEs 298 (94%) 84 (79%) Treatment-related AEs leading to treatment discontinuation 1 (0.3%)* 0 y Any SAEs 7 (2%) 5 (5%)z Treatment-related SAEs 0 0

# Deaths 1 (0.3%) 0 Common adverse events ($10%), n (%) Anemia 128 (40%) 5 (5%) Upper respiratory tract infection 67 (21%) 22 (21%) Hemobilirubin rise 52 (16%) 6 (6%) Laboratory abnormalities (grade 3), n (%) Alanine aminotransferase rise 1 (0.31%) 3 (2.83%) Aspartate aminotransferase rise 0 2 (1.89%) Total bilirubin increase 4 (1.26%) 0

* Discontinuation due to any adverse event occurred in one patient in the treatment group owing to drug allergy. y Serious adverse events in the treatment arm were vasovagal syncope, multiple bilateral rib fractures, died from a car accident. Hemorrhoids, cerebral infarction, infectious diarrhea, and community-acquired pneumonia, occurred in one patient each. z Serious adverse events in the placebo arm were acute exacerbation of chronic bronchitis, acute appendicitis and traumatic ciliary body detachment of the right eye, occurring in one patient each; left clavicle fracture and left third rib fracture in one; and, acute pelvic inflammatory disease and urinary tract infection in one. # One patient died due to traffic accident, and the event was not considered treatment-related. Abbreviations: AEs, adverse events; SAEs, serious adverse events. combination. The SVR12 of 99% regardless of pre-existing Acknowledgments baseline NS5A resistance-associated species may provide a simple to use choice to clinicians in the defined study Ascletis BioScience Co., Ltd. provided financial support for population. Good tolerability and safety profile further this study. endorse application of this regimen in a large population. Nonetheless, this study may have some limitations. First, the Conflict of interest use of RBV resulted in more anemia. However, most of the AEs were of mild or moderate severity and returned to normal Lai Wei was consultant for AbbVie, Abbott, BMS, Gilead after proper handling or intervention, as indicated from the Sciences, Johnson & Johnson, MSD, Novartis, and Roche, trial results. The paradoxical role of RBV from the beneficial speaker for Abbott, BMS, Gilead Sciences, MSD, Novartis, side may be eliminating quiescent viruses under other mech- Roche, and Ascletis, and received grant support from BMS anisms apart from DAAs and reducing hepatocellular carci- and Roche. The other authors have no conflict of interests noma incidence as reported before. The prolonged duration of related to this publication. using pegylated-interferon + RBV for 24 to 48 weeks pref- erably once plagued clinicians with safety concerns. Under Author contributions this new combination, the duration was shortened to 12 weeks without interferon; therefore, the safety profile was Article conception and design (LW), acquisition of data (XX, very appreciable. Second, this study did not recruit cirrhosis BF, YG, SZ, JS, XY, YM, YH, YK, XW, JL, YT, QH, QX, MW, PA, subjects, due to insufficient proof of concept study results. We GG, HL, QN, RH, BN, WX, JZ, WH, YY, ML, YZ, YY, JJ, DY, LC, may look into the cirrhosis cohort further, including it in our YY, YN, ZG, QZ, PH, FW, YL, DL, ZJ, JH, CC, and JJW), drafting future study design. Third, the use of ritonavir may trigger and critical revision of the manuscript (LW). some potential concern in drug-drug interaction in clinical practice. Per the outcome of the safety results, there was limited discontinuation due to drug-drug interaction and little References drug dose adjustment. This may be partially due to the observation that noncirrhosis patients are less sick and not [1] Global prevalence and genotype distribution of hepatitis C virus infection in 2015: a modelling study. Lancet Gastroenterol Hepatol 2017;2:161–176. taking complex concomitant comedications, while a short- doi: 10.1016/S2468-1253(16)30181-9. ened treatment duration of only 12 weeks may also help. [2] Rao H, Wei L, Lopez-Talavera JC, Shang J, Chen H, Li J, et al. Distribution and Nevertheless, this may warrant a real-world observation to clinical correlates of viral and host genotypes in Chinese patients with chronic hepatitis C virus infection. J Gastroenterol Hepatol 2014;29:545–553. doi: further understand the outcome. 10.1111/jgh.12398. In conclusion, treatment with all-oral RDV and DNVr in [3] Ghany MG, Strader DB, Thomas DL, Seeff LB. Diagnosis, management, and combination with RBV for 12 weeks resulted in SVR12 of 96% treatment of hepatitis C: an update. Hepatology 2009;49:1335–1374. doi: (FAS subgroup) and 99% (per protocol set subgroup) and was 10.1002/hep.22759. [4] Lalezari JP, Farrell GC, Shah PS, Schwab C, Walsh D, Vig P, et al. PPI-668, a well tolerated in treatment-naïve noncirrhotic HCV GT1 potent new pan-genotypic HCV NS5A inhibitor: Phase 1 efficacy and safety. Chinese patients. Hepatology 2012;56:1065A.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 213–220 219 Xu X. et al: Efficacy of RDV + DNV regimen in GT1 CHC patients

[5] Rajagopalan R, Misialek S, Stevens SK, Myszka DG, Brandhuber BJ, [9] Petta S, Craxì A. How to optimize HCV therapy in genotype 1 patients: pre- Ballard JA, et al. Inhibition and binding kinetics of the hepatitis C dictors of response. Liver Int 2013;33 Suppl 1:23–29. doi: 10.1111/liv.12053. virus NS3 protease inhibitor ITMN-191 reveals tight binding and slow [10] Zhang Y, Cao Y, Zhang R, Zhang X, Lu H, Wu C, et al. Pre-existing HCV dissociative behavior. Biochemistry 2009;48:2559–2568. doi: 10. variants resistant to DAAs and their sensitivity to PegIFN/RBV in Chinese 1021/bi900038p. HCV genotype 1b patients. PLoS One 2016;11:e0165658. doi: 10. [6] Seiwert SD, Andrews SW, Jiang Y, Serebryany V, Tan H, Kossen K, et al. 1371/journal.pone.0165658. Preclinical characteristics of the hepatitis C virus NS3/4A protease inhibitor [11] Fujii H, Umemura A, Nishikawa T, Yamaguchi K, Moriguchi M, Nakamura H, ITMN-191 (R7227). Antimicrob Agents Chemother 2008;52:4432–4441. et al. Real-world efficacy of daclatasvir and asunaprevir with respect to doi: 10.1128/AAC.00699-08. resistance-associated substitutions. World J Hepatol 2017;9:1064–1072. [7] Kao JH, Yu ML, Chen CY, Peng CY, Chen MY, Tang H, et al. Twelve-week doi: 10.4254/wjh.v9.i25.1064. ravidasvir plus ritonavir-boosted danoprevir and ribavirin for non-cirrhotic [12] Hepatitis C. guidance 2018 update: AASLD-IDSA recommendations for HCV genotype 1 patients: A phase 2 study. J Gastroenterol Hepatol 2018; testing, managing, and treating hepatitis C virus infection. Clin Infect Dis 33:1507–1510. doi: 10.1111/jgh.14096. 2018;67:1477–1492. doi: 10.1093/cid/ciy585. [8] Ghany MG, Nelson DR, Strader DB, Thomas DL, Seeff LB. An update on treat- [13] Huang Q, Huang N, Huq A, Lau M, Peng E, Lalezari J, et al. Vast majority of ment of genotype 1 chronic hepatitis C virus infection: 2011 practice guide- detected NS5A resistant variants are not amplified in HCV patients during line by the American Association for the Study of Liver Diseases. Hepatology 3-day monotherapy with the optimized NS5A inhibitor PPI-668. J Hepatol 2011;54:1433–1444. doi: 10.1002/hep.24641. 2013;58:S485. doi: 10.1016/S0168-8278(13)61193-9.

220 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 213–220 Original Article

Efficacy and Safety of 12-week Interferon-based Danoprevir Regimen in Patients with Genotype 1 Chronic Hepatitis C

Lai Wei*1, Jia Shang2, Yuanji Ma3, Xiaoyuan Xu4, Yan Huang5, Yujuan Guan6, Zhongping Duan7, Wenhong Zhang8, Zhiliang Gao9, Mingxiang Zhang10, Jun Li11, Jidong Jia12, Yongfeng Yang13, Xiaofeng Wen14, Maorong Wang15, Zhansheng Jia16, Bo Ning17, Yongping Chen18, Yue Qi19, Jie Du20, Jianning Jiang21, Lixin Tong22, Yao Xie23 and Jinzi J. Wu24

1Beijing Key Laboratory of Hepatitis C and Immunotherapy for Liver Diseases, Beijing, China; 2People’s Hospital of Henan Province, Henan, China; 3West China Hospital, Sichuan University, Chengdu, Sichuan, China; 4Peking University People’s Hospital, Beijing, China; 5Xiangya Hospital of Central South University, Changsha, Hunan, China; 6Guangzhou Eighth People’s Hospital, Guangzhou, Guangdong, China; 7Artificial Liver Center, Beijing YouAn Hospital, Capital Medical University, Beijing, China; 8Huashan Hospital, Fudan University, Shanghai, China; 9The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, Guangdong, China; 10The Sixth People’s Hospital of Shenyang, Shenyang, Liaoning, China; 11People’s Hospital of Jiangsu Province, Nanjing, Jiangsu, China; 12Beijing Friendship Hospital, Capital Medical University, Beijing, China; 13Nanjing Medical University Affiliated Second Hospital, Nanjing, Jiangsu, China; 14Liuzhou People’s Hospital, Liuzhou, China; 15Liver Disease Center of PLA, The 81st Hospital of PLA, Nanjing, Jiangsu, China; 16Tangdu Hospital, Air Force Medical University, Xi’an, Shaanxi, China; 17Baoji Central Hospital, Baoji, Shaanxi, China; 18The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, China; 19The First Hospital Affiliated to Jilin University, Changchun, Jilin, China; 20The First Hospital of Changsha, Changsha, Hunan, China; 21The First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi, China; 22The First Hospital of Hebei Medical University, Shijiazhuang, Hebei, China; 23Beijing Ditan Hospital, Capital Medical University, Beijing, China; 24Ascletis BioScience Co., Ltd., Hangzhou, Zhejiang, China

Abstract the majority of patients were of the IFNL4 rs12979860 CC genotype (87.2%, 123/141). Overall, 140 patients completed Background and Aims: Genotype (GT) 1 remains the pre- the 12-week treatment, and 97.1% (136/140) patients dominant hepatitis c virus (HCV) GT in Chinese patients. Over achieved sustained virologic response at 12 weeks (per proto- 80% of those Chinese patients harbor the interferon-sensitive col population group, 95% confidence interval: 92.9–99.2%). CC allele of IFNL4rs12979860, which is favorable for interfer- Only drug-related serious adverse event occurred. Most of the on-based treatment regimens. This phase III clinical trial adverse events were grade 1 and grade 2 alanine aminotrans- aimed to evaluate the efficacy and safety of the ritonavir- ferase elevation or liver dysfunction. One patient discontinued boosted danoprevir plus pegylated-interferon a-2a and riba- treatment because of severe head injury in a car accident. virin regimen for 12 weeks in treatment-naïve mainland Conclusions: The triple regimen of ritonavir-boosted dano- Chinese patients infected with HCV GT1 without cirrhosis. previr plus pegylated-interferon a-2a and ribavirin produced Methods: One hundred and forty-one treatment-naïve, a sustained virologic response rate of 97.1% after 12 weeks non-cirrhotic HCV GT1 Chinese patients (age $18 years) treatment in noncirrhotic HCV GT1-infected Chinese patients, were enrolled for this single-arm, multicenter, phase III MAN- and was safe and well tolerated. Trial Registration Clinical- ASA study (NCT03020082). Patients received a combination Trials.gov Identifier: NCT03020082 of ritonavir-boosted danoprevir (100 mg/100 mg) twice a day Citation of this article: Wei L, Shang J, Ma Y, Xu X, Huang Y, plus subcutaneous injection of weekly pegylated-interferon Guan Y, et al. Efficacy and safety of 12-week interferon-based a-2a (180 mg) and oral ribavirin (1000/1200 mg/day body danoprevir regimen in patients with genotype 1 chronic hepati- weight <75/$75 kg) for 12 weeks. The primary end-point tis C. J Clin Transl Hepatol 2019;7(3):221–225. doi: 10.14218/ was sustained virologic response rate at 12 weeks after the JCTH.2019.00018. end of treatment. The secondary end-points were safety outcomes, tolerability, virologic response over time and relapse rate. Results: All enrolled patients were HCV GT1-infected, and most among them (97.9%, 123/141) had the HCV GT1b Introduction subtype. Single-nucleotide polymorphism test showed that Chronic hepatitis C virus (HCV) infection is associated with chronic liver diseases, including cirrhosis and hepatocellular Keywords: Danoprevir; Hepatitis C virus (HCV); Sustained virological response carcinoma.1 It is estimated that over 10 million Chinese indi- (SVR); Treatment-naïve. 2 Abbreviations: AE, adverse event; DAA, direct-acting antiviral agent; DNVr, viduals are chronically infected with HCV. The annual report ritonavir-boosted danoprevir; GT, genotype; HCV, hepatitis C virus; SAE, serious on HCV issued by the China Ministry of Health showed a adverse event; SVR, sustained virological response. gradual increase of HCV cases from year 2003 (20,000 Received: 13 May 2019; Revised: 12 June 2019; Accepted: 30 June 2019 cases) to 2012 (200,000 cases).3 *Correspondence to: Lai Wei, Beijing Key Laboratory of Hepatitis C and Immu- notherapy for Liver Diseases, Beijing 100191, China. Tel: +86-1-88326666, Fax: Studies have shown that the genotype (GT)1b (56.8%) 4 +86-1-68318386, E-mail: [email protected] subtype of HCV is the most prevalent subtype in China,

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 221–225 221

Methods All data were analyzed by SAS 9.2 software. A sample size of 71 was required to provide a power of 0.95 with a two-sided Study design significance level of 0.05 to demonstrate the lower boundary of SVR12 of DNVr plus pegylated-interferon a-2a and ribavirin This phase III, single-arm, multicenter study (termed superiority to the pegylated-interferon a-2a and ribavirin ‘MANASA’) aimed to evaluate the efficacy and safety of regimen. A p-value <0.05 was considered statistically signifi- DNVr 100 mg/100 mg twice a day in combination with the cant. Our primary meta-analysis (data not shown) indicated regimen in treatment-naïve noncirrhotic GT1 patients, with a that the average SVR24 rate in mainland Chinese HCV superior design to compare the binomial distribution of SVR12 patients infected with GT1 for the pegylated-interferon a-2a against pegylated-interferon a-2a and ribavirin historical data and ribavirin regimen was 73.9%, and the expected SVR12 (SVR24 up to 73.9%). The study was conducted in full rate of the DNVr plus pegylated-interferon a-2a and ribavirin conformance to the ethical principles of the Declaration of regimen was 90%. Helsinki (2013), and the institutional review board of all participating institutions approved the study (available upon Results written request to the Corresponding Author). Written informed consent was obtained from all participants accord- Study population ing to local laws and regulations of the particular institutions. A total of 141 patients received $1 dose of the DNVr plus Patients and treatment pegylated-interferon a-2a and ribavirin regimen, among them 138/141 (87.8%) had IFNL4 rs12979860 (a single This open-label study enrolled 141 adult patients from 23 nucleotide polymorphism on chromosome 19, rs12979860) medical centers of China from June 2016 to July 2017. CC GT, which is favorable for interferon-based treatment, and Eligibility criteria are recorded in detail in the study registra- three patients were infected with HCV GT1a (3/141, 2.1%) tion information (NCT03020082). In short, eligible patients (Table 1). In total, 140/141 (97.1%) patients completed the were aged $18 years and having chronic HCV GT1 infection, treatment. One (1.4%) patient discontinued treatment due to with plasma HCV RNA level >10,000 IU/mL (prior to screening treatment-unrelated serious adverse events (SAEs).

222 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 221–225 Wei L. et al: Danoprevir regimen efficacy in genotype 1 CHC

Table 1. Patient demographics and disease characteristics

Variables n = 141

Age in years, mean (range) 42.1 (19.0;72.2) Gender as m/f, n 71/70 BMI in kg/m2, mean (range) 23.0 (18.1;29.7) Nationality of Han/other, n 126/49

Baseline HCV RNA in log10 6.04 (2.91;7.08) IU/mL, median (range) Baseline HCV RNA in IU/mL, n (%) <400,000 33 (23.4) 400,000-800,000 14 (9.9) $800 000 94 (66.7) IFNL4 rs12979860, n (%) CC 123 (87.2) Fig. 1. Mean plasma HCV RNA change during the treatment of all the non-CC 18 (12.8) patients who completed the treatment. The HCV RNA levels of all individual HCV genotype, n (%) patients declined rapidly after initiation of treatment with DNVr plus pegylated- interferon a-2a and ribavirin on week 1. Plasma HCV RNA was quantified at 1a 3 (2.1) screening and at each visit thereafter to assess the virologic response to treatment. Plasma HCV RNA was recorded as 1 IU/mL (equal to 1 log10 IU/mL) if HCV 1b 138 (97.9) RNA <15 IU/mL (lower limit of quantitation) or target not detected. FibroScan (kPa), mean (range) 6.1 (2.7;12.2) Abbreviations: HCV, hepatitis C virus; DNVr, ritonavir-boosted danoprevir; EOT, end of treatment; SVR12, sustained virologic response at 12 Abbreviations: BMI, body mass index; HCV, hepatitis C virus; kPa, kilopascals. weeks after treatment; SVR24, sustained virologic response at 24 weeks after treatment.

Virologic response over time Safety At week 1, the HCV RNA levels of all individual patients declined rapidly after treatment initiation with DNVr plus pegylated- The safety profiles of DNVr plus pegylated-interferon a-2a interferon a-2a and ribavirin (Fig. 1). Among 141 patients and ribavirin was comparable to that of pegylated-interferon enrolled, 140 patients completed the 12-week therapy and a-2a and ribavirin therapy. At least one adverse event (AE) one patient withdrew due to treatment-unrelated SAEs. A was reported in all patients. Most AEs were of mild to moderate total of 136 of 141 patients (intention-to-treat group, 96.5%, in severity. AEs that were observed in more than 10% of [95% confidence interval (CI): 91.9–98.8%]) achieved SVR12 the patients during treatment and follow-up are shown in Table 2. Anemia, fever, fatigue, leukopenia, neutropenia and (Fig. 2), with the lower boundary of the 95% CI for the DNVr plus pegylated-interferon a-2a and ribavirin regimen higher than the SVR12 of 73.9% in another pegylated-interferon a-2a and ribavirin regimen study (based on historical results of the pegylated-interferon a-2a and ribavirin regimen in Chinese chronic hepatitis C patients, p<0.001). Overall, among the 140 patients who completed the treatment, 118/122 (96.7%) with the IFNL4 rs12979860 CC genotype and 18/18 (100.0%) with the CT genotype achieved SVR12. A total of 139 [98.6%, (95% CI: 96.1–100%)] patients achieved virologic responses by the end of the treatment (HCV RNA <15 IU/mL). One patient discontinued the treatment on day 10 due to car accident.

Resistance

No virologic breakthrough occurred during the study. Four cases (2.9%, 4/140) experienced relapse, among which three were detected at week 4 after end of treatment and one at Fig. 2. SVR 12 weeks after the end of treatment (SVR12) in PP group week 12 after end of treatment. Baseline gene sequencing of (n = 140) and ITT group (n = 141). One patient discontinued treatment due to amino acid polymorphisms at NS3 positions revealed that car accident. three relapsed patients exhibited NS3-D168A polymorphisms, Abbreviations: SVR, sustained virologic response; PP, per protocol pop- while one patient had NS3-R155K polymorphisms. ulation; ITT, intention-to-treat.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 221–225 223 Wei L. et al: Danoprevir regimen efficacy in genotype 1 CHC

Table 2. Adverse events and laboratory abnormalities Table 3. Abnormal liver function test results

Event or abnormality n (%) Grade 1, Grade 2, Grades 3 and Feature n (%) n (%) 4, n (%) Any adverse events 141 (100.0) Severe adverse events* 5 (3.5) High ALT 10 (7.1) 3 (2.1) 0

# Adverse events leading to discontinuation 1 (0.0) High AST 10 (7.1) 1 (0.7) 0 Adverse events in ≥10% of patients High TBIL 13 (9.2) 0 0 Anemia 76 (53.9) High DBIL 3 (2.1) 0 0

Fever 53 (37.6) Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; Fatigue 52 (36.9) TBIL, total bilirubin; DBIL, direct bilirubin. Headache 31 (22.0) Dizziness 28 (19.9) regimens. This phase III treatment regimen produced 97% SVR12, and thus was safe and effective. Overall, this finding Anorexia 26 (18.4) matches that of phase II (96% SVR12) and the DAPSANG Diarrhea 25 (17.7) study (94.1% SVR12). In the sofosbuvir registration study in Flu-like symptoms 25 (17.7) China, sofosbuvir plus pegylated-interferon a-2a and ribavirin Hypertriglyceridemia 21 (14.9) 12-week therapy showed 94% SVR12, and in the simeprevir registration study in China, simeprevir plus pegylated-inter- Erythra 17 (12.1) feron a-2a and ribavirin 24/48-week therapy showed 91% Urinary tract infection 16 (11.34) SVR12. Compared with these two other DAA plus pegylated- Nausea 13 (19.5) interferon a-2a and ribavirin regimens, danoprevir shows better efficacy with a short treatment duration. In addition, it Laboratory abnormalities in ≥10% of patients also showed a better efficacy and shorter treatment duration Leukopenia 118 (83.7) than the pegylated-interferon a-2a and ribavirin regimen [first Neutropenia 103 (73.1) generation protease inhibitor plus pegylated-interferon a-2a and ribavirin regimens (67–75%)11] and other DAA plus pegy- Thrombocytopenia 51 (36.2) lated-interferon a-2a and ribavirin regimens (sofosbuvir plus Lymphocytosis 26 (14.2) pegylated-interferon a-2a and ribavirin, simeprevir plus pegy- High ALT 24 (17.0) lated-interferon a-2a and ribavirin, daclatasvir plus pegylated- 9,12–14 Hypochromia 22 (16.3) interferon a-2a and ribavirin). According to the literature, virologic breakthrough and High AST 19 (13.5) relapse are concerns during the treatment. Interestingly, in Erythropenia 19 (13.5) our phase II and III trials, no virologic breakthrough occurred

* but four patients with previously documented baseline NS3- One severe adverse event of the pegylated-interferon and ritonavir regimen, but not the ritonavir-boosted danoprevir regimen, and three severe adverse events D168 or NS3-R155 polymorphisms experienced relapse at not related to treatment regimen. weeks 4 and 12 after the end of the treatment. Notably, # One car crash death was not related with treatment and the patient was excluded patients enrolled in our phase II and III studies of the DNVr from the study. plus pegylated-interferon a-2a and ribavirin regimen without Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase. baseline resistant-associated substitutions screening achieved >95% SVR12, suggesting that there was no need thrombocytopenia occurred in more than 30% patients, which was most likely to be consistent with AEs of ribavirin and pegy- Table 4. Summary of severe adverse events during and after the lated-interferon treatment. Abnormal liver function related to treatment pegylated-interferon and ribavirin were mainly classified into grades 1 and 2 (Table 3). Relation to DNVr, Six SAEs occurred in five patients (Table 4), which included Events n (%) PEG-IFN, RTV and RBV recurrence of tuberculosis, neutropenia, ventricular tachyar- Neutropenia 1 (0.7) PEG-IFN, RBV, RTV, DNVr rhythmia, transient ischemic attack, and digital fibrokeratoma. All five patients continued treatment despite the Tuberculosis 1 (0.7) N/A SAEs. A 50-year-old patient withdrew from the study Ventricular 1 (0.7) PEG-IFN, RBV because of severe head injury in a car accident. Only neutro- tachyarrhythmia penia was considered to be related with pegylated-interferon, Transient 1 (0.7) N/A ribavirin, danoprevir, or ritonavir. ischemic attack Acquired digital 1 (0.7) N/A Discussion fibrokeratoma Craniocerebral 1 (0.7) Possibly not In this phase III study, 141 treatment-naïve mainland injury Chinese patients infected with HCV GT1 without cirrhosis was evaluated for 12-weeks to determine the efficacy and Abbreviations: PEG-IFN, pegylated-interferon; DNVr, ritonavir-boosted danopre- safety of DNVr plus pegylated-interferon a-2a and ribavirin vir; RBV, ribavirin; RTV, ritonavir; N/A, not applicable.

224 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 221–225 Wei L. et al: Danoprevir regimen efficacy in genotype 1 CHC for baseline resistant-associated substitutions screening speaker for Abbott, BMS, Gilead Sciences, MSD, Novartis, before using this regimen. Roche and Ascletis, and received grant support from BMS and The DNVr plus pegylated-interferon a-2a and ribavirin Roche. Lan Zhang, Huoling Tang, Jinzi J. Wu were employees regimen was safe and well tolerated. Most AEs were related of Ascletis Pharmaceuticals Company. The others have no to pegylated-interferon a-2a and ribavirin and only grade 1–2 conflict of interests related to this publication. liver dysfunctions were observed (Table 3). The most frequently reported AEs (in >30%) were anemia, fever, Author contributions fatigue, leukopenia, neutropenia and thrombocytopenia, of which only neutropenia was considered to be caused by dano- Study concept and design (LW and JJW), performed the study, previr, while others were consistent with AEs of ribavirin and recruited patients and collected specimens (LW, JS, YM, XX, pegylated-interferon treatment. One patient withdrew from YH, YG, ZD, WZ, ZG, MZ, JL, JJ, YY, XW, MW, ZJ, BN, YC, YQ, the study because of brain injury (and ultimate death) follow- JD, JJ, LT and YX), and drafted the manuscript (LW). ing a serious car accident during the treatment. No patient discontinued the study because of AEs. Evidence suggests that the all-oral DAA regimen from References foreign pharmaceutical companies that are currently about to be launched may cost USD $8376 to $17,092 for each [1] Guidelines for the screening care and treatment of persons with chronic hep- course,15 but the per capita disposable income of Chinese atitis C infection: Updated version. Geneva: World Health Organization, 2016. 16 residents in 2017 was only USD $3763.6. In addition, a [2] Global prevalence and genotype distribution of hepatitis C virus infection in medical insurance scheme for DAAs therapy is only valid in 2015: a modelling study. Lancet Gastroenterol Hepatol 2017;2:161–176. three provinces of China. Thus, most patients are still unable doi: 10.1016/S2468-1253(16)30181-9. to afford the imported all-oral DAA regimens. Danoprevir, [3] Duan Z, Jia JD, Hou J, Lou L, Tobias H, Xu XY, et al. Current challenges and the management of chronic hepatitis C in mainland China. J Clin Gastroenterol being the first DAA developed by a Chinese company (Ascle- 2014;48:679–686. doi: 10.1097/MCG.0000000000000109. tis) and having the more affordable price tag of USD $965.9/2 [4] Rao H, Wei L, Lopez-Talavera JC, Shang J, Chen H, Li J, et al. Distribution and weeks, seems affordable, with the cost of each 1% increase in clinical correlates of viral and host genotypes in Chinese patients with chronic – SVR of USD $77.6 (danoprevir plus pegylated-interferon a-2a hepatitis C virus infection. J Gastroenterol Hepatol 2014;29:545 553. doi: 10.1111/jgh.12398. and ribavirin regimen), which is lower than that of foreign [5] Rao HY, Li H, Chen H, Shang J, Xie Q, Gao ZL, et al. Real-world treatment made all-oral DAA regimens (USD $85.8 to $179.9 for each patterns and clinical outcomes of HCV treatment-naive patients in China: an 1% increase in SVR) and pegylated-interferon a-2a and rib- interim analysis from the CCgenos study. J Gastroenterol Hepatol 2017;32: – avirin regimens (USD $89.3 to $133 for each 1% increase in 244 252. doi: 10.1111/jgh.13467. [6] Manns MP, McHutchison JG, Gordon SC, Rustgi VK, Shiffman M, Reindollar R, SVR). The overall treatment cost may become lower, if the et al. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b cost of pegylated-interferon a-2a and ribavirin (USD plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. $288.4/2 weeks) gets reimbursed by a medical insurance Lancet 2001;358:958–965. doi: 10.1016/s0140-6736(01)06102-5. scheme in all provinces.15 [7] Jiang Y, Andrews SW, Condroski KR, Buckman B, Serebryany V, Wenglowsky S, et al. Discovery of danoprevir (ITMN-191/R7227), a highly selective and The limitation of our study includes small sample size potent inhibitor of hepatitis C virus (HCV) NS3/4A protease. J Med Chem restricted to the inclusion and exclusion criteria, as defined by 2014;57:1753–1769. doi: 10.1021/jm400164c. clinical trial protocols. Another concern was the weekly [8] Gottwein JM, Scheel TK, Jensen TB, Ghanem L, Bukh J. Differential efficacy of injection, which leads to poorer adherence in real-world protease inhibitors against HCV genotypes 2a, 3a, 5a, and 6a NS3/4A protease recombinant viruses. Gastroenterology 2011;141:1067–1079. doi: practices. Finally, real-world studies on large-scale should 10.1053/j.gastro.2011.06.004. be initiated for additional information. [9] Everson G, Cooper C, Hézode C, Shiffman ML, Yoshida E, Beltran-Jaramillo T, et al. DAUPHINE: a randomized phase II study of danoprevir/ritonavir plus Conclusions peginterferon alpha-2a/ribavirin in HCV genotypes 1 or 4. Liver Int 2015;35: 108–119. doi: 10.1111/liv.12471. [10] Kao JH, Tung SY, Lee Y, Thongsawat S, Tanwandee T, Sheen IS, et al. In all, the treatment goal for hepatitis C is not only to clear Ritonavir-boosted danoprevir plus peginterferon alfa-2a and ribavirin in HCV but also to improve disease outcome and long-term Asian chronic hepatitis C patients with or without cirrhosis. J Gastroenterol Hepatol 2016;31:1757–1765. doi: 10.1111/jgh.13374. prognosis. The DAA plus pegylated-interferon a-2a and [11] Wei L, Lok AS. Impact of new hepatitis C treatments in different regions of ribavirin regimens, which produce dual effect on immunomo- the world. Gastroenterology 2014;146:1145–1150.e4. doi: 10.1053/j. dulation and antivirus activity, could improve the efficacy and gastro.2014.03.008. shorten the chronic hepatitis C virus treatment course, and [12] Lawitz E, Mangia A, Wyles D, Rodriguez-Torres M, Hassanein T, Gordon SC, et al. Sofosbuvir for previously untreated chronic hepatitis C infection. N Engl thereby reduce the occurrence of cirrhosis, hepatocellular J Med 2013;368:1878–1887. doi: 10.1056/NEJMoa1214853. carcinoma, and HCV-related death. [13] Wei L, Han T, Yang D, Heo J, Shang J, Cheng J, et al. Simeprevir plus peginterferon/ribavirin for HCV genotype 1-infected treatment-naïve patients in China and South Korea. J Gastroenterol Hepatol 2016;31:912–920. doi: 10. Acknowledgments 1111/jgh.13288. [14] Hézode C, Hirschfield GM, Ghesquiere W, Sievert W, Rodriguez-Torres M, Ascletis Pharmaceuticals Co., Ltd. provided financial support Shafran SD, et al. Daclatasvir plus peginterferon alfa and ribavirin for treatment-naive chronic hepatitis C genotype 1 or 4 infection: a randomised for this study (MANASA). study. Gut 2015;64:948–956. doi: 10.1136/gutjnl-2014-307498. [15] Hu SL, Lu J. A comparative cost-effectiveness analysis on direct-acting anti- Conflict of interest HCV agents. China Health Insurance 2018;8:53–58. [16] National Bureau of Statistics of China. Statistical Communiqué of the People’s Republic of China on the 2017 National Economic and Social Devel- Lai Wei is a former consultant for AbbVie, Abbott, BMS, Gilead opment. Available from: http://www.stats.gov.cn/english/PressRelea- Sciences, Johnson & Johnson, MSD, Novartis and Roche, se/201802/t20180228_1585666.html. Accessed February 28, 2018.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 221–225 225 Original Article

Hepatitis C Screening: Barriers to Linkage to Care

Sammy Saab*1,2, Youssef P. Challita1, Lisa M. Najarian1, Rong Guo3, Satvir S. Saggi1 and Gina Choi1,2

1Departments of Surgery, University of California at Los Angeles, Los Angeles, CA, USA; 2Departments of Medicine, University of California at Los Angeles, Los Angeles, CA, USA; 3Departments of Medicine Statistics Core, University of California at Los Angeles, Los Angeles, CA, USA

Abstract liver transplantation.2 Moreover, its extra-hepatic manifestations include varied expressions of glomerulonephritis, cryo- Background and Aims: HepatitisC(HCV)isamedicaland globulinemia, and lymphoma.3–5 public health concern. Once infected individuals are identi- The largest cohort of individuals infected with hepatitis C fied, management includes not only education but also the virus (HCV) are the baby-boomers, people born between the use of antiviral therapy. Although screening for HCV is read- years 1945 and 1965.6 Although there are multiple risk ily available, barriers exist which prevent assessment and factors for this blood-borne infection, many individuals may treatment in individuals potentially infected with HCV. have been infected through blood transfusions before HCV Methods: This is a retrospective study of patients screened was identified and blood products were screened for the for HCV within the University of California, Los Angeles virus.7 Thus, a significant number of infected individuals are Health Care System between February 22 and July 9, unaware of their infection status.8 The treatment of HCV has 2018. We defined linkage to care as: 1) confirmatory HCV evolved from therapies that were associated with significant RNA test after screening HCV antibody test found a positive adverse effects to the development of all-oral direct acting result; and 2) follow-up appointment for treatment was estab- therapies that are highly effective, safe, and tolerable.9 Con- lished with a specialist. Demographic and baseline laboratory sequently, major health care societies recommend screening values were collected. Factors potentially associated with pro- patients for HCV based on their age, in addition to risk factors hibiting linkage of care were evaluated. Results: During the and recommended therapy to all infected individuals regard- study period, 17,512 individuals were screened for HCV. A total less of severity of fibrosis as long as their prognosis is at least of 238 (1.35%) were found to have detectable HCV antibodies. 1 year.10 Identifying those infected is the first step toward the Of the individuals with detectable HCV antibodies, 48 (20%) did elimination of HCV.11,12 not undergo confirmatory testing with viral levels. Of the 190 There is a cascade of events that begin with identifying individuals who underwent further testing, 70 patients were HCV infected individuals and continue toward treatment. The noted to be viremic. Among them, 17 of the 70 (24%) were biggest limitation appears to be identifying those infected.13 not linked to a specialist for further care. Younger patients (p = Screening for HCV relies on testing for antibodies of the virus 0.02) and people who inject drugs (p = 0.02) were less likely to and then confirming the presence of infection by measuring be referred for specialty care. Conclusions: The results of our the viral load. After HCV infection is confirmed, patients study highlight that younger patients and people who inject should be linked to care with a provider experienced in treat- drugs are less likely to be referred to specialty care for HCV ing HCV.10 However, there are a number of limitations that treatment. Efforts are needed to engage these populations. can exist in every stage of care. The hypothesis of our study Citation of this article: Saab S, Challita YP, Najarian LM, Guo is that most barriers to HCV treatment are modifiable. We R, Saggi SS, Choi G. Hepatitis C screening: Barriers to linkage reviewed the cascade of events following screening for HCV to care. J Clin Transl Hepatol 2019;7(3):226–231. doi: in order to identify barriers for linking patients infected with 10.14218/JCTH.2018.00063. HCV to care.

Methods Introduction

Hepatitis C is a viral infection with significant hepatic and Screening extra-hepatic manifestations.1 Not only is hepatitis C a common cause of cirrhosis, it is a major risk factor for hep- All patients found to have detectable HCV antibodies for atocellular carcinoma and one of the leading indications for hepatitis C at the University of California, Los Angeles Health- care System during the study period between February and July 2018 were identified using an administrative database.14 Keywords: Hepatitis C; Screening; Linkage. Abbreviations: CI, confidence interval; HCV, hepatitis C virus; IQR, interquartile Ordering providers were alerted of the results by the labora- range; PWID, people who inject drugs. tory. We also contacted providers of patients with positive Received: 13 December 2018; Revised: 4 February 2019; Accepted: 1 March enzyme-linked immunosorbent assay results. Our coordina- 2019 tor recommended additional laboratory testing in the form of *Correspondence to: Sammy Saab, Pfleger Liver Institute, UCLA Medical Center, 200 Medical Plaza, Suite 214, Los Angeles, CA 90095, USA. Tel: +1-310-206- an HCV RNA quantitative PCR with reference to genotype and 6705, Fax: +1-310-206-4197, E-mail: [email protected] offered to facilitate a clinic appointment with a Hepatologist.

226 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 226–231

Fig. 1. Cascade of care.

We collected data on patients’ age, gender, health insurance primary providers until a response was obtained. The health type, health care setting of antibody screening, home zip code, care setting was stratified according to whether testing homeless status, injection drug usage and laboratory test occurred during hospitalization, emergency department visit, results. The zip code was used to estimate the patients’ socio- or clinic visit. Clinics were further stratified if there were single economic status.15 We also recorded the time between anti- specialty (by Primary Care Providers) or multispecialty (not body and RNA testing, and the number of emails sent to the including Gastroenterology or Hepatology physicians).

Table 1. Overall demographics and hepatitis C viral load assessment

Variable not tested Total, n = 238 HCV RNA tested, n = 190 HCV RNA, n =48

Median age (IQR), years 59 (47–66) 59 (46–66) 61 (51–67) Gender – M/F, % 150/88 (63/37%) 118/72 (62/38%) 32/16 (67/33%) Insurance type Private 97 (40%) 81(43%) 16 (33.3%) Medicare 59 (25%) 45 (24%) 14 (29.2%) HMO 59 (25%) 45 (24%) 14 (29.2%) Medi-Cal 19 (8%) 16 (8%) 3 (6.3%) No insurance 4 (2%) 3 (1%) 1 (2%) Health care setting Hospital 84 (35%) 69 (36%) 15 (31%) ER 17 (7%) 14 (7%) 3 (6%) Clinic Single specialty 65 (27%) 51 (27%) 14 (29%) Multispecialty 72 (30%) 56 (30%) 16 (33%) Median distance (IQR), milesy 9.0 (3.6–24.3) 9.3 (3.6–22.8) 7.4 (3.6–30.8) Median household income (IQR), thousand $72 (53k-86) $72 (51k-86) $71 (55k-90) Homeless 34 (14%) 26 (14%) 8 (17%) PWID 33 (14%) 27 (14%) 6 (13%)

Abbreviations: IQR, interquartile range; HCV, hepatitis C virus; yrs, years; M/F, male/female; HMO, Health Maintenance Organization; ER, emergency room; PWID, people who inject drugs. y Distance from home to clinic.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 226–231 227 Saab S. et al: Barriers to HCV care

Table 2. Median laboratory values of hepatitis C antibody-positive patients

Variable Total, n = 238 HCV RNA tested, n = 190 HCV RNA not tested, n =48

Platelet count (IQR),1/L 221 (182–279) 218 (178–279) 228 (194–278) AST (IQR), U/L 28 (19–44) 28 (19–45) 28 (20–37) ALT (IQR), U/L 25 (17–49) 25 (16–50) 32 (18–43) Total bilirubin (IQR), mg/dL 0.4 (0.3–0.7) 0.4 (0.3–0.7) 0.6 (0.3–0.7) APRI (IQR) 0.3 (0.2–0.5) 0.3 (0.2–0.6) 0.3 (0.2–0.4)

IQR, interquartile range; AST, aspartate aminotransferase; ALT, alanine aminotransferase; U, units; HCV, hepatitis C Virus; APRI, AST to platelet ratio index.

Successful linkage to care was assessed in two separate using the Statistical Analysis Software version 9.4 (SAS Insti- ways: 1) if the patient underwent testing for RNA (testing tute Inc., Cary, NC, USA). linkage); and 2) if the patient was referred to a hepatitis C specialist if they were viremic (specialty linkage). Patients Results were given 3 weeks from their positive antibody result to undergo HCV RNA testing. Patients who tested for HCV A total of 17,512 patients were screened during the study viremia within the 3-week period met the criteria for suc- period, among whom 238 were found to have detectable HCV cessful testing linkage to care. Barriers to linkage of care were antibodies (Fig. 1). The median (IQR) age of patients found to recorded for both assessments. be HCV antibody positive was 59 (47–66) years (Table 1). Most of the patients were male and had commercial insurance. Most of the patients were tested in primary or multi- Statistics specialty clinics, and lived within 10 miles of the site of testing. Few patients were homeless or people who inject Outcomes at both the testing linkage and specialty linkage drugs (PWID). Median (IQR) liver-associated factors and pla- levels were assessed. Standard descriptive analyses, median telet test results are shown in Table 2. The median (IQR) (interquartile range [IQR]), n (%), and univariate analysis aspartate aminotransferase to platelet ratio index was 0.3 were conducted using Microsoft Excel 15.2 (Microsoft, (0.2–0.5). Seattle, WA, USA) by testing linkage or specialty linkage. A total of 190 (79%) out of the 238 patients with Further analysis was conducted in the form of chi-squared, detectable HCV antibodies were further assessed for HCV Fischer’s exact tests, and multivariate logistic regression viremia, of which 70 (37%) had detectable viral levels.

Table 3. Demographics of patients referred and not referred to specialty care

Variable Referred, n = 53 Not referred, n =17

Median age (IQR) 60 yrs (49–67) 48 yrs (30–58) Gender – M/F (%) 33/20 (62%/38%) 11/6 (65%/35%) Insurance type Private 24 (45%) 6 (35%) Medicare 13 (25%) 2 (12%) HMO 9 (17%) 7 (41%) Medi-Cal 7 (13%) 2 (12%) No insurance 0 0 Medical setting Hospital 23 (43%) 8 (46%) ER 3 (6%) 4 (24%) Clinic Single specialty 13 (25%) 1 (6%) Multispecialty 14 (26%) 4 (24%) Distance from home to clinic (IQR), miles 3.6 (3.4–15.9) 8.9 (3.5–18.0) Median household income (IQR), thousand $72 (51–82) $57 (56–73) Homeless 8 (15%) 6 (35%) PWID 9 (17%) 8 (47%)

Abbreviations: IQR, interquartile range; yrs, years; M/F, male/female; HMO, Health Maintenance Organization; ER, emergency room; PWID, people who inject drugs.

228 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 226–231 Saab S. et al: Barriers to HCV care

Table 4. Median laboratory values according to specialty referral in Discussion hepatitis C viremic patients

Referred, Not referred, The results of this study highlight the disparity between HCV Variable n =53 n =17 screening and linkage of care. In our study, approximately 27% of patients who were found to have detectable HCV Platelet count 204 (175–259) 256 (183–270) antibodies were not linked to either further testing or spe- (IQR), 1/L cialty care. We identified barriers to care and offering sug- AST (IQR), U/L 38 (29–53) 35 (26–63) gestions to mitigate these limitations. The definition of linkage to care varies across studies. Studies have defined ALT (IQR), U/L 40 (22–61) 39 (24–75) linkage as confirmatory HCV RNA testing in patients found to Total bilirubin (IQR), 0.5 (0.3–0.8) 0.5 (0.3–0.8) have HCV antibodies,16–19 referral to specialty clinics among mg/dL viremic patients,8,20–29 or both.30–35 Two studies defined APRI (IQR) 0.4 (0.3–0.6) 0.3 (0.2–0.5) linkage to care as receiving HCV RNA testing, referral to specialty care and attending the first appointment with the spe- 36,37 Abbreviations: IQR, interquartile range; AST, aspartate aminotransferase; ALT, cialist. Several studies defined linkage to care as alanine aminotransferase; U, Units; HCV, hepatitis C virus; APRI, AST to platelet attending the first appointment with a provider and starting ratio index. viral treatment,38–40 whereas two studies defined linkage to care as being seen by a specialist.41–43 In our study we defined linkage separately so we can better understand the Forty-eight patients with detectable HCV antibodies did not limitations of care. undergo viral load testing. Of the 48 patients who did not HCV infection possesses a number of characteristics that undergo HCV RNA testing, the test was not ordered by the make elimination feasible, including the lack of a non-human provider for 27 patients. For the remaining 21 of the 48 reservoir, easy diagnosis, and effective curative treatment.11 patients, the test was ordered but not completed by the Indeed, the World Health Organization proposes that viral patient. Demographic comparison of the patients who were infection can be eliminated within the next two decades pro- and were not checked for HCV viremia is shown in Table 1, vided that the diagnosis rate is 90%, the treatment rate is and laboratory results in Table 2. There were no statistically 80%, and the curative rate is 90%.44,45 However, the significant differences in the demographics and lab results results of our study highlight barriers to HCV elimination. between these two groups. Not only are there significant deficiencies in screening in the 46 A total of 53 out of the 70 (75%) patients with HCV USA but a major gap exists between the diagnosis and viremia were referred to specialty care for further treatment. treatment. The results of our study, as well as of others, Demographics and laboratory values of patients who were suggest that the biggest obstacle to curing infected patients is linkage of care. Concentrated efforts are essential to bridge and were not referred for specialty care are shown in Tables the gap. 3 and 4. Patients who were not referred for specialty care This study demonstrated the important predictors of lack were younger (median age of 48 vs 60, p =0.02),and of linkage to specialty care for hepatitis C were being a young more likely to be PWID (8/17, 47% vs 9/53 17%, p = adult and a PWID. The odds of receiving specialty care 0.02). In a multivariate logistic regression analysis including decrease by 70% for PWID (odds ratio = 0.31, 95% CI: age and PWID as independent variables, the adjusted odds 0.087–1.099, p = 0.07). Similarly, for every unit decrease in ratio for PWID was 0.31 (p = 0.07, 95% confidence interval age (years) the odds of being linked to specialty care [CI]: 0.09–1.09) and the adjusted odds ratio for age was decreases by 3% (odds ratio = 1.032, 95% CI: 0.996– 1.03 (p = 0.09, 95% CI: 1.00–1.07). In addition, we exam- 1.072, p = 0.09). This finding is of particular concern as ined the interaction between age and PWID; the interaction PWIDs likely represent one of the fastest growing cohorts of effect was not significant and thus excluded from the model HCV infection in the USA. In fact, in over a dozen states in this (p = 0.50). The reasons provided by laboratory ordering pro- country, the HCV prevalence among young individuals is as viders for not referring are listed in Table 5. The most high or higher than among baby-boomers.47 If HCV elimina- common reason for lack of specialty care was that the tion is truly a goal, we must better engage this population. patient did not attend the appointment. Several proposals to include HCV linkage to care among PWID include utilization of a multidisciplinary approach, fostering better communication, and increasing awareness of the impli- 22,23,48 Table 5. Failures in Linkage to care by specialty linkage definition, n =17 cations of HCV infection. There are several limitations to our study. The most Unsuccessful linkage reasons n important is its retrospective nature. For many of our patients who were not linked to care, the reasons could only be Physician deferred treatment 6 projected. Nevertheless, thorough review of the medical Upcoming kidney transplant 1 record and discussions with the primary providers important Primary care physician will proceed with treatment 1 insights were provided. Another important limitation is that we could not exclude selective screening in our cohort. The Physician states patient will not follow through with 3 current practice for screening involves an opt–in approach, treatment whereby a provider is alerted that their patient is a baby- Patient fled facility 1 boomer and should be considered for HCV screening, accord- Physician did not respond to linkage request via email 4 ing to the Center for Disease Control guidelines.7 Moreover, providers also screened patients based on risk factors for Patient noncompliant to referral 7 viral hepatitis C. Nonetheless, our entire cohort of patients

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 226–231 229 Saab S. et al: Barriers to HCV care screened consisted of almost 18,000 patients. Another limi- and Nutrition Examination Survey 2003 to 2010. Ann Intern Med 2014;160: tation is that our study cohort may not reflect the general 293–300. doi: 10.7326/M13-1133. [8] Castrejón M, Chew KW, Javanbakht M, Humphries R, Saab S, Klausner JD. population of patients with HCV. For instance, the viremic Implementation of a large system-wide hepatitis C virus screening and rate in patients with detectable HCV antibodies was 37%, linkage to care program for baby boomers. Open Forum Infect Dis 2017;4: which is substantially lower than described by others in the ofx109. doi: 10.1093/ofid/ofx109. literature. One possible reason is that patients with HCV anti- [9] Kohli A, Shaffer A, Sherman A, Kottilil S. Treatment of hepatitis C: a system- – bodies, but undetectable HCV RNA viral levels, may have atic review. JAMA 2014;312:631 640. doi: 10.1001/jama.2014.7085. [10] Hepatitis C. guidance: AASLD-IDSA recommendations for testing, manag- been previously treated. We do not think this accounts for a ing, and treating adults infected with hepatitis C virus. Hepatology 2015;62: substantial number of our aviremic patients since in that sit- 932–954. doi: 10.1002/hep.27950. uation there would be no medical indication for screening. [11] Saab S, Le L, Saggi S, Sundaram V, Tong MJ. Toward the elimination of – Moreover, there has been a trend noted in the literature of hepatitis C in the United States. Hepatology 2018;67:2449 2459. doi: 10. 34 1002/hep.29685. lower viremic rates in screening programs. Another limita- [12] World Health Organization. Combating hepatitis B and C to reach elimination tion to our study is that patients were not directly contacted by 2030. Advocacy brief. Assessed May 2016. Available from: https://apps. for reasons for not following up with viral load testing or who.int/iris/bitstream/handle/10665/206453/WHO_HIV_2016.04_eng.pdf; appointment for treatments. Prospective studies may help jsessionid=95C4FA3F13009E69C8212FC46AF1F8BA?sequence=1. [13] Moorman AC, Xing J, Ko S, Rupp LB, Xu F, Gordon SC, et al. Late diagnosis of clarify additional patient level barriers than what we described hepatitis C virus infection in the Chronic Hepatitis Cohort Study (CHeCS): in our retrospective study. Missed opportunities for intervention. Hepatology 2015;61:1479–1484. doi: The results of our study highlight that deficiencies exist in 10.1002/hep.27365. extending HCV-related care in the fastest growing infected [14] Chou R, Cottrell EB, Wasson N, Rahman B, Guise JM. Screening for hepatitis cohort in the USA, particularly for the young and PWID. For C virus infection in adults: a systematic review for the U.S. Preventive Serv- ices Task Force. Ann Intern Med 2013;158:101–108. doi: 10.7326/0003- HCV elimination policy to be effective, these cohorts must be 4819-158-2-201301150-00574. engaged. [15] United States Census Bureau. 2012–2016 American community survey 5-year estimates. Available from: https://factfinder.census.gov/faces/table- services/jsf/pages/productview.xhtml?src=bkmk. Acknowledgments [16] Young KL, Huang W, Horsburgh CR, Linas BP, Assoumou SA. Eighteen- to 30- year-olds more likely to link to hepatitis C virus care: an opportunity to The research described was supported by NIH/National decrease transmission. J Viral Hepat 2016;23:274–281. doi: 10.1111/jvh. 12489. Center for Advancing Translational Science (NCATS) UCLA [17] Assoumou SA, Huang W, Horsburgh CR Jr, Drainoni ML, Linas BP. Relation- CTSI Grant Number UL1TR001881. ship between hepatitis C clinical testing site and linkage to care. Open Forum Infect Dis 2014;1:ofu009. doi: 10.1093/ofid/ofu009. Conflict of interest [18] Blackburn NA, Patel RC, Zibbell JE. Improving screening methods for hepatitis C among people who inject drugs: Findings from the HepTLC initiative, 2012-2014. Public Health Rep 2016;131 Suppl 2:91–97. doi: 10. The authors have no conflict of interests related to this 1177/00333549161310S214. publication. [19] Geboy AG, Mahajan S, Daly AP, Sewell CF, Fleming IC, Cha HA, et al. High hepatitis C infection rate among baby boomers in an urban primary care clinic: Results from the HepTLC initiative. Public Health Rep 2016;131 Author contributions Suppl 2:49–56. doi: 10.1177/00333549161310S209. [20] MacLean CD, Berger C, Cangiano ML, Ziegelman D, Lidofsky SD. Impact of Contributed to study concept and design (SS), acquisition of electronic reminder systems on hepatitis C screening in primary care. J Viral Hepat 2018;25:939–944. doi: 10.1111/jvh.12885. data (YPC, LMH, SSS), analysis and interpretation of data [21] Franco RA, Overton ET, Tamhane AR, Forsythe JM, Rodgers JB, Schexnayder (SS, YPC, LMN, GC), drafting of the manuscript (SS, YPC), JK, et al. Characterizing failure to establish hepatitis C care of baby boomers critical revision of the manuscript for important intellectual diagnosed in the emergency department. Open Forum Infect Dis 2016;3: content (SS, YPC, GC), statistical analysis (RG, YPC), admin- ofw211. doi: 10.1093/ofid/ofw211. [22] Bajis S, Dore GJ, Hajarizadeh B, Cunningham EB, Maher L, Grebely J. Inter- istrative, technical, or material support (SS), study super- ventions to enhance testing, linkage to care and treatment uptake for hep- vision (SS, GC), and guarantor of the article (SS). All authors atitis C virus infection among people who inject drugs: A systematic review. approved the final version of the manuscript. Int J Drug Policy 2017;47:34–46. doi: 10.1016/j.drugpo.2017.07.002. [23] Morano JP, Zelenev A, Lombard A, Marcus R, Gibson BA, Altice FL. Strategies for hepatitis C testing and linkage to care for vulnerable populations: point- References of-care and standard HCV testing in a mobile medical clinic. J Community Health 2014;39:922–934. doi: 10.1007/s10900-014-9932-9. [1] Webster DP, Klenerman P, Dusheiko GM. Hepatitis C. Lancet 2015;385: [24] Seña AC, Willis SJ, Hilton A, Anderson A, Wohl DA, Hurt CB, et al. Efforts at 1124–1135. doi: 10.1016/S0140-6736(14)62401-6. the frontlines: Implementing a hepatitis C testing and linkage-to-care [2] Kim WR, Lake JR, Smith JM, Skeans MA, Schladt DP, Edwards EB, et al. program at the local public health level. Public Health Rep 2016;131 Suppl – OPTN/SRTR 2013 Annual Data Report: liver. Am J Transplant 2015;15 2:57 64. doi: 10.1177/00333549161310S210. Suppl 2:1–28. doi: 10.1111/ajt.13197. [25] Meyer JP, Moghimi Y, Marcus R, Lim JK, Litwin AH, Altice FL. Evidence-based [3] Younossi Z, Park H, Henry L, Adeyemi A, Stepanova M. Extrahepatic mani- interventions to enhance assessment, treatment, and adherence in the – festations of hepatitis C: A meta-analysis of prevalence, quality of life, and chronic Hepatitis C care continuum. Int J Drug Policy 2015;26:922 935. economic burden. Gastroenterology 2016;150:1599–1608. doi: 10.1053/j. doi: 10.1016/j.drugpo.2015.05.002. gastro.2016.02.039. [26] Younossi ZM, LaLuna LL, Santoro JJ, Mendes F, Araya V, Ravendhran N, et al. [4] Negro F, Forton D, Craxì A, Sulkowski MS, Feld JJ, Manns MP. Extrahepatic Implementation of baby boomer hepatitis C screening and linking to care in morbidity and mortality of chronic hepatitis C. Gastroenterology 2015;149: gastroenterology practices: a multi-center pilot study. BMC Gastroenterol 1345–1360. doi: 10.1053/j.gastro.2015.08.035. 2016;16:45. doi: 10.1186/s12876-016-0438-z. [5] Wong RJ, Saab S, Ahmed A. Extrahepatic manifestations of hepatitis C virus [27] Tait JM, McIntyre PG, McLeod S, Nathwani D, Dillon JF. The impact of a after liver transplantation. Clin Liver Dis 2017;21:595–606. doi: 10.1016/j. managed care network on attendance, follow-up and treatment at a hepatitis cld.2017.03.013. C specialist centre. J Viral Hepat 2010;17:698–704. doi: 10.1111/j.1365- [6] Smith BD, Morgan RL, Beckett GA, Falck-Ytter Y, Holtzman D, Teo CG, et al. 2893.2009.01227.x. Recommendations for the identification of chronic hepatitis C virus infection [28] Evon DM, Simpson KM, Esserman D, Verma A, Smith S, Fried MW. Barriers to among persons born during 1945-1965. MMWR Recomm Rep 2012;61:1–32. accessing care in patients with chronic hepatitis C: the impact of depression. [7] Denniston MM, Jiles RB, Drobeniuc J, Klevens RM, Ward JW, McQuillan GM, Aliment Pharmacol Ther 2010;32:1163–1173. doi: 10.1111/j.1365-2036. et al. Chronic hepatitis C virus infection in the United States, National Health 2010.04460.x.

230 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 226–231 Saab S. et al: Barriers to HCV care

[29] Taylor BS, Hanson JT, Veerapaneni P, Villarreal R, Fiebelkorn K, Turner BJ. [39] Kronfli N, Linthwaite B, Kouyoumdjian F, Klein MB, Lebouché B, Sebastiani G, Hospital-based hepatitis C screening of baby boomers in a majority hispanic et al. Interventions to increase testing, linkage to care and treatment of hep- south texas cohort: Successes and barriers to implementation. Public Health atitis C virus (HCV) infection among people in prisons: A systematic review. Rep 2016;131 Suppl 2:74–83. doi: 10.1177/00333549161310S212. Int J Drug Policy 2018;57:95–103. doi: 10.1016/j.drugpo.2018.04.003. [30] Falade-Nwulia O, Mehta SH, Lasola J, Latkin C, Niculescu A, O’Connor C, et al. [40] Zuckerman A, Douglas A, Nwosu S, Choi L, Chastain C. Increasing success and Public health clinic-based hepatitis C testing and linkage to care in Baltimore. evolving barriers in the hepatitis C cascade of care during the direct acting anti- J Viral Hepat 2016;23:366–374. doi: 10.1111/jvh.12507. viral era. PLoS One 2018;13:e0199174. doi: 10.1371/journal.pone.0199174. [31] Tohme RA, Xing J, Liao Y, Holmberg SD. Hepatitis C testing, infection, and [41] Trooskin SB, Poceta J, Towey CM, Yolken A, Rose JS, Luqman NL, et al. linkage to care among racial and ethnic minorities in the United States, Results from a geographically focused, community-based HCV screening, 2009-2010. Am J Public Health 2013;103:112–119. doi: 10.2105/AJPH. linkage-to-care and patient navigation program. J Gen Intern Med 2015; – 2012.300858. 30:950 957. doi: 10.1007/s11606-015-3209-6. [32] Schoenbachler BT, Smith BD, Seña AC, Hilton A, Bachman S, Lunda M, et al. [42] Islam MM, Topp L, Conigrave KM, White A, Reid SE, Grummett S, et al. Linkage Hepatitis C virus testing and linkage to care in North Carolina and South into specialist hepatitis C treatment services of injecting drug users attending a needle syringe program-based primary healthcare centre. J Subst Abuse Carolina jails, 2012-2014. Public Health Rep 2016;131 Suppl 2:98–104. Treat 2012;43:440–445. doi: 10.1016/j.jsat.2012.07.007. doi: 10.1177/00333549161310S215. [43] Miller LS, Rollin F, Fluker SA, Lundberg KL, Park B, Quairoli K, et al. High-yield [33] Bourgi K, Brar I, Baker-Genaw K. Health disparities in hepatitis C screening birth-cohort hepatitis C virus screening and linkage to care among under- and linkage to care at an integrated health system in southeast Michigan. served African Americans, Atlanta, Georgia, 2012-2013. Public Health Rep PLoS One 2016;11:e0161241. doi: 10.1371/journal.pone.0161241. 2016;131 Suppl 2:84–90. doi: 10.1177/00333549161310S213. [34] Konerman MA, Thomson M, Gray K, Moore M, Choxi H, Seif E, et al. Impact of [44] World Health Organization. Global health sector strategy on viral hepatitis 2016– an electronic health record alert in primary care on increasing hepatitis c 2021. Available from: https://apps.who.int/iris/bitstream/handle/10665/ screening and curative treatment for baby boomers. Hepatology 2017;66: 246177/WHOHIV-2016.06-eng.pdf;jsessionid=BDB9D6116C68A2E10213BD8- 1805–1813. doi: 10.1002/hep.29362. CAB2F2E39?sequence=1. Accessed June 2016. [35] de la Torre AN, Castaneda I, Ahmad M, Ekholy N, Tham N, Herrera IB, et al. [45] World Health Organization. Global Hepatitis Report 2017. Available from: Audio-computer-assisted survey interview and patient navigation to https://apps.who.int/iris/bitstream/handle/10665/255016/9789241565455- increase chronic viral hepatitis diagnosis and linkage to care in urban eng.pdf?sequence=1. – health clinics. J Viral Hepat 2017;24:1184 1191. doi: 10.1111/jvh.12744. [46] Mehta D, Mccombs J, Sanchez Y, Marx S, Sammy S. Effectiveness of hepatitis [36] Patel RC, Vellozzi C, Smith BD. Results of hepatitis C birth-cohort testing and C virus screening laws in United States: Evidence from paid claims data from linkage to care in selected U.S. sites, 2012-2014. Public Health Rep 2016; 2010 to 2016. J Hepatol 2018;68:S177–S178. doi: 10.1016/S0168-8278 131 Suppl 2:12–19. doi: 10.1177/00333549161310S203. (18)30566-X. [37] Coyle C, Viner K, Hughes E, Kwakwa H, Zibbell JE, Vellozzi C, et al. Identi- [47] Morse A, Barritt AS 4th, Jhaveri R. Individual state hepatitis C data supports fication and linkage to care of HCV-infected persons in five health centers - expanding screening beyond baby boomers to all adults. Gastroenterology Philadelphia, Pennsylvania, 2012-2014. MMWR Morb Mortal Wkly Rep 2015; 2018;154:1850–1851.e2. doi: 10.1053/j.gastro.2018.02.035. 64:459–463. [48] Day E, Hellard M, Treloar C, Bruneau J, Martin NK, Øvrehus A, et al. Hepatitis [38] Akyar E, Seneca KH, Akyar S, Schofield N, Schwartz MP, Nahass RG. Linkage C elimination among people who inject drugs: Challenges and recommen- to care for suburban heroin users with hepatitis C virus infection, New Jersey, dations for action within a health systems framework. Liver Int 2019;39:20– USA. Emerg Infect Dis 2016;22:907–909. doi: 10.3201/eid2205.151980. 30. doi: 10.1111/liv.13949.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 226–231 231 Original Article

TIPS Is Not Associated with a Higher Risk of Developing HCC in Cirrhotic Patients: A Systematic Review and Meta-analysis

Bin Chen1, Long Pang2, Hao-Bin Chen2, Dong-Bo Wu1, Yong-Hong Wang1 and En-Qiang Chen*1

1Center of Infectious Diseases, West China Hospital, Sichuan University, Chengdu, China; 2West China School of Medicine, Sichuan University, Chengdu, China

Abstract hypertension-associated complications because the procedure effectively reduces portal pressure.2–5 With the development Background and Aims: The association between portal- of medical technology, surgical portal-systemic shunt has been systemic shunt and hepatocellular carcinoma (HCC) develop- replaced by TIPS, and bare-stent TIPS has been replaced by ment in patients who have cirrhosis is still controversial. This covered-stent TIPS with extended polytetrafluoroethylene.6 systematic review with meta-analysis was performed to sys- Early application of TIPS in cirrhotic patients with variceal tematically clarify the potential role of portal-systemic shunt bleeding has been recommended to reduce the overall mortal- in the development of HCC. Methods: The PubMed, Embase, ity.7 Moreover, TIPS has played an increasingly essential role in and Cochrane Library databases were searched for potentially the management of decompensated liver cirrhosis. eligible literature. Meta-analysis with random-effects model Hepatocellular carcinoma (HCC) remains one of the leading was performed to combine the incidence rates of HCC after causes of mortality among cirrhotic patients. Early accurate portal-systemic shunt. Finally, seven studies were included. diagnosis is beneficial for improving prognosis of this tumor, In the present review, we mainly focused on 859 patients and a regular detection program for HCC has been suggested (365 in the transjugular intrahepatic portal-systemic shunt to high-risk patients.8,9 Portal-systemic shunt could generate (TIPS) group and 494 in the non-TIPS group) from five studies circulatory, hemodynamic, and functional changes, which to analyze incidence rates after TIPS. Results: At the end of might increase the incidence of HCC as suggested in some follow-up, there were 66 (18%, 66/365) patients who devel- published studies.10 In 1985, Bjorneboe et al.11 reported a oped HCC after TIPS intervention and 63 (13%, 63/494) pa- higher incidence of developing HCC in shunted patients who tients who developed HCC after non-TIPS treatments. Pooled survived more than 6 months after operation (relative risk: estimates with random-effects model did not demonstrate a 3.28; 95% confidence interval [CI]: 1.52–7.45) compared significant increase of incidence of HCC after TIPS (risk ratio: with non-shunted patients. In 2005, Banares et al.12 reported 1.37 [confidence interval (CI): 0.96 to 1.97]; p =0.08)com- an observational study which suggested cirrhotic patients with pared with non-TIPS treatments. Subgroup analyses for those application of TIPS may suffer a higher incidence of HCC. As patients with transplanted liver also did not detect a significant such, a formal HCC monitoring program is highly recommen- difference between the TIPS group and non-TIPS group (risk ded for these patients, especially for those are not able to ratio: 1.10 [CI: 0.59 to 2.07]; p =0.75).Conclusions: Current receive a short-period or medium-period liver transplantation. evidence suggests that portal-systemic shunt is not associated However, incompatible results have been reported regard- with a higher risk of HCC development in cirrhotic patients. ing the incidence of developing HCC in patients with portal- Citation of this article: Chen B, Pang L, Chen HB, Wu DB, systemic shunt. In 2005, Libbrecht et al.13 published another Wang YH, Chen EQ. TIPS is not associated with a higher risk study which showed that neither the progress of HCC nor the of developing HCC in cirrhotic patients: A systematic review and existence and amount of dysplastic nodules was in associa- meta-analysis. J Clin Transl Hepatol 2019;7(3):232–237. doi: tion with the application of TIPS. In 2013, De Santis and his 10.14218/JCTH.2019.00007. colleagues14 reported a retrospective case–control study with 101 cirrhotic patients in different groups. The median length of follow-up was 56.7 months (range: 8.2–174.5) and 67.8 – Introduction months (range: 8.3 183.1) for TIPS patients and controls respectively (p = 0.08). The cumulative incidence of HCC at Portal-systemic shunt, especially the transjugular intrahe- 1, 3, 5 and 10 years was 2%, 7%, 18% and 46% in the TIPS patic portal-systemic shunt (TIPS) introduced in 1989,1 has group and 3%, 10%, 19% and 39% in the control group (p = become one of the most acceptable treatments of portal 0.19). These results suggested that TIPS did not account for the increased risk of HCC, and unmodified ultrasound super- vising was recommended for these patients. Recently, in Keywords: Transjugular intrahepatic portal-systemic shunt (TIPS); Hepatocellular 17 carcinoma (HCC); Hepatic cirrhosis; Systematic review. 2015, Borentain et al. reported another retrospective Abbreviations: CI, confidence interval; HCC, hepatocellular carcinoma; TIPS, study which suggested that TIPS may increase the incidence transjugular intrahepatic portosystemic shunt. of liver dysplasia; however, the risk of developing HCC in a Received: 27 January 2019; Revised: 2 April 2019; Accepted: 5 May 2019 transplanted liver was not related to TIPS. *Correspondence to: En-Qiang Chen, Center of Infectious Diseases, West China Hospital, Sichuan University, No. 37 Guo Xue Xiang, Wuhou District, Chengdu The present systematic review with meta-analysis aims to 610041, China. E-mail: [email protected] provide a comprehensive understanding of whether the

232 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 232–237

Copyright: © 2019 Authors. This article has been published under the terms of Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0), which permits noncommercial unrestricted use, distribution, and reproduction in any medium, provided that the following statement is provided. “This article has been published in Journal of Clinical and Translational Hepatology at DOI: 10.14218/JCTH.2019.00007 and can also be viewed on the Journal’s website at http://www.jcthnet.com”. Chen B. et al: TIPS is not related to HCC incidence rates of HCC development are statistically different Statistical analysis in cirrhotic patients who received TIPS than in those who received non-TIPS treatments, based upon a thorough search The primary outcome was the incidence rate of HCC at the of the electronic databases, including PubMed, Cochrane end of follow-up of each included study. Discontinuous out- Library, and Embase. comes were evaluated by odds ratio and a corresponding 95% confidence interval (CI). The curative effect in continuous variables was expressed as weighted mean difference Methods with corresponding variances. The results were identified as statistically significant when the p value was less than 0.05. 2 Literature search The heterogeneity among the studies was assessed by the I statistic (I2 > 50% suggested substantial heterogeneity) and A systematic literature search of three databases was inde- the chi-square test (p < 0.10 indicated significant statistical 16 pendently conducted, including PubMed, Cochrane Library, heterogeneity). Publication bias was investigated by the and Embase. This meta-analysis followed the standard Egger’s test. RevMan v5.3 (The Cochrane Collaboration, Soft- reporting guidelines.15 The search was done from the incep- ware Update, Oxford, UK) was the tool used to extract, pool tion of the databases to 20 March 2018. The following key- and analyze the data. words were used: TIPS; portal-systemic shunt; surgical shunt; and transjugular intrahepatic portal-systemic shunt. Results All potential eligible studies were manually searched to find possible relevant publications. No language or publication Literature search and study characteristics type or date restriction was set. Of 253 records, 9 met selection criteria11–14,17–20 (Fig. 1). Inclusion and exclusion criteria Two records were excluded due to no relevant data being available.20 Finally, seven studies were included in this review.Onewasarandomizedtrial,19 and six were observatio- Inclusion criteria were as follows: 1) the publications 11–14,17,18 described randomized controlled trials or observational nal studies. Two studies evaluated the incidence rate of HCC after surgical portal-systemic shunts,11,18 and five studies studies; 2) the participants in the experimental groups were 12–14,17,19 treated with portal-systemic shunt, while the participants in evaluated the incidence rate of HCC after TIPS. We can the control groups were treated with other treatments instead only make a descriptive review regarding the incidence rate after surgical portal-systemic shunts due to lack of available full-texts. of portal-systemic shunts; 3) reporting on the outcome of 11 incidence rate of HCC after treatment. If the data of any paper In 1985, Bjorneboe et al. reported a high risk of HCC in overlapped or were duplicated among two or more studies by shunted patients alive more than 6 months postoperatively (rel- – the same study team, only the study with the more complete ative risk: 3.28; 95% CI: 1.52 7.45) compared with non- data or one earlier study were included. shunted patients. However, another study reported that after a mean follow-up of 50 months, no differences were observed between the two groups in relation to the prevalence of HCC Study selection and data extraction after portal-systemic shunt.18 In this review, we mainly focused on the analysis of The titles and abstracts of identified articles were perused by incidence rate after TIPS due to the available data. The full- two of the authors (B.C. and L.P.) working independently after texts of the selected five studies were obtained,12–14,17,19 and the removal of duplications. Then, full articles were retrieved if the main characteristics are displayed in Table 1. Two studies further assessment was needed. Data collection was inde- were performed in Spain and another three were conducted in pendently carried out by two reviewers. Two authors made Belgium, Italy, and France respectively. The five total studies their own assessment of the risk of bias of enrolled studies involved 859 patients (365 in the TIPS group and 494 in the according to the Risk of Bias Tool conferred by the Cochrane non-TIPS group). The mean age in each study was over Handbook. Divergence was resolved by third opinion. 50 years old, and more than 60% of participants were male. The extracted and summarized data include first author, The main cause of cirrhosis was alcohol abuse. Of note, two publication year, country, study design, publication type, studies assessed the incidence rate of HCC after TIPS in trans- sample size, type of portal-systemic shunt (surgical shunt planted livers.13,17 According to the methodological assess- or TIPS), duration of follow-up, demographic data (age and ment, only one randomized controlled trial was considered sex), and the incidence rate of HCC after treatment. high quality.19 Most observational studies were rated as good quality.12,14,17 However, one study with abstract avail- Quality assessment able only was considered to be of unclear quality.13 The funnel plots demonstrated no publication bias, as shown in Fig. 2. The quality of observational studies was evaluated with the Newcastle–Ottawa Scale. The Newcastle–Ottawa Scale Meta-analyses of incidence rates of HCC measures quality in the three domains of selection, comparability, and exposure. High-quality studies were considered The total five studies involving 859 patients were included for to have a score of 7 or greater, consistent with other meta- the meta-analysis of incidence rates of HCC after TIPS or analyses. The Cochrane Risk of Bias Tool for assessing risk of other treatments (365 in the TIPS group and 494 in the non- bias was used to assess the quality of randomized controlled TIPS group). All of these five studies reported the number of trials. Study quality was assessed independently by two patients with de novo HCC in the TIPS group or non-TIPS investigators (C.B. and L.P.) and any discrepancies were group. The mean or median follow-up time was reported in addressed by a joint evaluation of the original article. three of the studies, ranging from 14.4 to 67.8 months. At the

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 232–237 233 Chen B. et al: TIPS is not related to HCC

Fig. 1. Flow chart of study selection. last follow-up, there were 66 (18%, 66/365) patients with de group and the non-TIPS group (relative risk: 1.10 [95% CI: novo HCC in the TIPS group and 63 (13%, 63/494) patients 0.59–2.07]; p = 0.75, random-effects model) (Fig. 4). with de novo HCC in the non-TIPS group. Although the incidence rate of HCC was higher in the TIPS group than in the non-TIPS group, the pooled results did not demonstrate a Discussion significant difference between the two groups (relative risk: 1.37 [95% CI: 0.96–1.97]; p = 0.08, random-effects model) There is still controversy regarding the influence of portal- (Fig. 3). No significant heterogeneity between studies was systemic shunt on the incidence rate of HCC after shunt detected (p = 0.36; I2 =8%). procedures. Our present study systematically collected those published studies which evaluated the influence of portal- systemic shunt on development of HCC. This is the first Subgroup analyses of incidence rates of HCC systematic review and meta-analysis focusing on this topic. Due to no relevant data available, meta-analysis was not Subgroup analysis was performed according to the percent- performed for those studies which assessed whether surgical age of cirrhosis patients due to alcohol. Pooled estimates from portal-systemic shunt increased the risk of HCC in cirrhosis. studies that enrolled more than 50% of cirrhosis due to alcohol So, this study mainly reported the pooled results with regard suggested that TIPS was associated with a significant increase to the influence of TIPS on the development of HCC. of development of HCC (relative risk: 2.02 [95% CI: 1.05– In total, five studies were included for meta-analysis. We 3.9]; p = 0.04, random-effects model) (Fig. 3). However, combined the results with random-effects model, and the pooled estimates from studies which enrolled less than 50% pooled estimates did not detect a significant difference of cirrhosis due to alcohol patients did not demonstrate a sig- regarding the incidence rate of HCC between the TIPS group nificant difference between the two groups (relative risk: 1.11 and the non-TIPS group. Of note, subgroup analysis sug- [95% CI: 0.73–1.69]; p = 0.63, random-effects model). We gested that TIPS was positively associated with incident HCC further performed a subgroup analysis assessing the inci- in those studies which enrolled more than 50% of cirrhosis dence rate of HCC in those patients with transplanted liver, patients due to alcohol. However, there is no other evidence and there was also no significant difference between the TIPS to implicate TIPS as a risk factor for HCC development in

234 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 232–237 Chen B. et al: TIPS is not related to HCC

Table 1. Main characteristics of included studies (TIPS vs. non-TIPS group)

Etiology of Publication Sample alcohol/ Follow-up Study Year Country type Study design size, n Age (y) Male other, n time

Escorsell 2002 Spain Full-text Randomized 47/44 5769/ 70%/ 24:23/ 15.4/14.4 et al.19 controlled trial 56611 80% 25:19 months Banares 2005 Spain Full-text Observational 138/ 56611/ 73%/ 70:68/ 30.3/31.4 et al.12 study 138 56611 73% 70:68 months Libbrecht 2005 Belgium Letter Observational 11/65 5468/ 64%/ 8:3/22:43 Not et al.13 study 55610 63% reported De Santis 2014 Italy Full-text Observational 101/ 5969/ 64%/ 27:74/ 67.8/56.7 et al.14 study 101 59610 64% 27:74 months Borentain 2015 France Full-text Observational 68/146 5167/ 77%/ 43:25/ Not et al.17 study 5068 78% 59:87 reported

patients with alcoholic cirrhosis. Additionally, subgroup anal- 5-year period, the actuarial risk of developing HCC was 34% in ysis also did not demonstrate that TIPS increased the risk of the TIPS group and 25% in the non-TIPS group. Of note, the HCC in transplanted liver. In order to improve the accuracy rate of liver transplantation was much higher in the non-TIPS and reliability of pooled estimates, we did not show the pooled group than in the TIPS group in this study, which meant that estimates with fixed-effects model in the section of results. the number of patients at risk of HCC over time was lower in However, the pooled results suggested that TIPS was asso- this group. Furthermore, as mentioned in the section of dis- ciated with a higher incidence rate of HCC, compared with cussion by authors, selection bias cannot be completely other treatments (relative risk: 1.41 [95% CI: 1.01–1.96]; excluded due to the retrospective nature of this study, p = 0.04, fixed-effects model). although researchers of this study performed a careful selec- Of those five included studies, only one retrospective cohort tion of the non-TIPS cohort to match the TIPS cohort as well as study suggested that the creation of TIPS may be associated possible. with a higher risk of developing HCC in patients with cirrho- Some researchers have suggested that the marked sis.12 Their results suggested that patients with TIPS had a decrease of portal blood flow due to TIPS placement might higher risk of developing HCC than matched controls. Over a lead to malignant transformation of the cirrhotic liver,

Fig. 2. Funnel plot showed no publication bias.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 232–237 235 Chen B. et al: TIPS is not related to HCC

Fig. 3. Pooled estimates with random-effects model did not demonstrate a significant increase of incidence of HCC after TIPS, compared with other treatments. Abbreviations: HCC, hepatocellular carcinoma; TIPS, transjugular intrahepatic portosystemic shunt. because changes in hepatic blood flow have been suggested to our meta-analysis. First, there have been many trials to be associated with the development of nodular regenerative evaluating the efficacy of portal-systemic shunt in the man- hyperplasia in patients with chronic Budd-Chiari syndrome.21 agement of variceal bleeding;7,23–27 however, they mainly Borentain and his colleagues17 further demonstrated that the focused on the outcome of survival and did not pay attention percentage of patients presenting with dysplastic nodules was to the association between portal-systemic shunt and devel- higher in their TIPS group than the non-TIPS group (33.9% vs. opment of HCC. So, only five studies were included in this 20.1%, p = 0.047). However, another study reported by Lib- meta-analysis. Second, the study designs varied from brecht et al.13 showed that the presence and number of dys- randomized controlled trials to observational cohort studies. plastic nodules was about 2 to 2.5 times lower in patients with Although most were of high quality, the inclusion criteria for a TIPS. To date, no special study has been conducted to patients varied, with some studies including only patients explore the mechanisms through which TIPS placement with transplanted liver. Third, the follow-up duration varied could initiate the development of HCC. One published study significantly, which might have influenced the incidence rate suggested that TIPS was associated with the augmentation of HCC after treatment. We pooled the incidence rate of HCC of hepatic iron deposits and the vascular and parenquimal at the last follow-up time from each included study, which changes.12 Another study found pseudointimal hyperplasia in would introduce some bias. So the random-effects model the lumen of bare-stent TIPS, which may also play a role in the was used in our meta-analysis, in order to minimize the influ- development of HCC.22 However, this problem could be ence of heterogeneity between studies on the pooled esti- avoided in the era of covered–stent utilization in the TIPS mates. Lastly, we only enrolled one randomized controlled procedure. trial, which makes future studies needed to fully elucidate Although there is no significant heterogeneity between the association between portal-systemic shunt and develop- studies, and no publication bias, there are several limitations ment of HCC.

Fig. 4. Subgroup analysis for those patients with transplanted liver also did not detect a significant difference between TIPS group and non-TIPS group. Abbreviations: HCC, hepatocellular carcinoma; TIPS, transjugular intrahepatic portosystemic shunt.

236 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 232–237 Chen B. et al: TIPS is not related to HCC

Conclusions [10] Hanafy AS, Soliman S, Abd-Elsalam S. Rescue therapy for chronic hepatitis C virus infection after repeated treatment failures: Impact on disease progression and risk of hepatocellular carcinoma. Hepatol Res 2019;49:377–384. Creation of TIPS is not associated with increased risk of HCC doi: 10.1111/hepr.13303. development in patients with cirrhosis. This meta-analysis [11] Bjørneboe M, Andersen JR, Christensen U, Skinhøj P, Jensen OM. Does a may further reassure physicians of this lack of association. portal-systemic shunt increase the risk of primary hepatic carcinoma in cir- – Nevertheless, considering the natural history of cirrhosis and rhosis of the liver? Scand J Gastroenterol 1985;20:59 64. doi: 10. 3109/00365528509089633. inherent residual risk of HCC development, which is not [12] Bañares R, Núñez O, Escudero M, Fernández C, Vaquero J, Beceiro I, et al. addressed by the use of TIPS, clinicians should remain vigilant Patients with cirrhosis and bare-stent TIPS may have increased risk of hep- to the HCC risk in these patients. atocellular carcinoma. Hepatology 2005;41:566–571. doi: 10.1002/hep. 20576. [13] Libbrecht L, Maleux G, Verslype C, Nevens F, Roskams T. Influence of TIPS on Conflict of interest development of hepatocellular carcinoma in cirrhosis. Hepatology 2005;42: 236. doi: 10.1002/hep.20745. The authors have no conflict of interests related to this [14] De Santis A, Iegri C, Kondili L, Riggio O, Salvatori FM, Catalano C, et al. publication. Hepatocellular carcinoma in cirrhotic patients with transjugular intrahepatic portosystemic shunt: a retrospective case-control study. Dig Liver Dis 2014; 46:726–730. doi: 10.1016/j.dld.2014.04.009. Author contributions [15] Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med 2009;151:264–269. doi: 10.7326/0003-4819-151-4-200908180- Searched literature and assessed methodological quality 00135. (BC), selected studies for inclusion and extracted data (LP), [16] Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in analyzed and interpreted the data (BC, DBW, YHW, and HBC), meta-analyses. BMJ 2003;327:557–560. doi: 10.1136/bmj.327.7414.557. and wrote the manuscript (BC and LP), designed the study, [17] Borentain P, Garcia S, Gregoire E, Vidal V, Ananian P, Ressiot E, et al. Trans- jugular intrahepatic porto-systemic shunt is a risk factor for liver dysplasia provided administrative support, and final approval of manu- but not hepatocellular carcinoma: a retrospective study of explanted livers. script (EQC), all authors approved the final manuscript. Dig Liver Dis 2015;47:57–61. doi: 10.1016/j.dld.2014.09.009. [18] Elizalde JI, Castells A, Planas R, Rodríguez-Iglesias MP, Bruix J, Brú C, et al. Prevalence of hepatocellular carcinoma in cirrhotic patients with portosyste- References mic shunt. Cohort analysis. Gastroenterol Hepatol 1996;19:189–193. [19] Escorsell A, Bañares R, García-Pagán JC, Gilabert R, Moitinho E, Piqueras B, [1] Richter GM, Noeldge G, Palmaz JC, Roessle M, Slegerstetter V, Franke M, et al. TIPS versus drug therapy in preventing variceal rebleeding in advanced – et al. Transjugular intrahepatic portacaval stent shunt: preliminary clinical cirrhosis: a randomized controlled trial. Hepatology 2002;35:385 392. doi: results. Radiology 1990;174:1027–1030. doi: 10.1148/radiology.174.3. 10.1053/jhep.2002.30418. 174-3-1027. [20] Wei J, Li H, Li C. Investigating the risk factors of hepatocellular carcinoma [2] Pomier-Layrargues G, Villeneuve JP, Deschênes M, Bui B, Perreault P, and survival analysis for cirrhosis after transjugular intrahepatic portosyste- – Fenyves D, et al. Transjugular intrahepatic portosystemic shunt (TIPS) mic shunt in treating portal hypertension. J Cancer Res Ther 2018;14:826 versus endoscopic variceal ligation in the prevention of variceal rebleeding 832. doi: 10.4103/jcrt.JCRT_930_17. in patients with cirrhosis: a randomised trial. Gut 2001;48:390–396. doi: 10. [21] Cazals-Hatem D, Vilgrain V, Genin P, Denninger MH, Durand F, Belghiti J, 1136/gut.48.3.390. et al. Arterial and portal circulation and parenchymal changes in Budd- – [3] Jalan R, Forrest EH, Stanley AJ, Redhead DN, Forbes J, Dillon JF, et al. Chiari syndrome: a study in 17 explanted livers. Hepatology 2003;37:510 A randomized trial comparing transjugular intrahepatic portosystemic 519. doi: 10.1053/jhep.2003.50076. stent-shunt with variceal band ligation in the prevention of rebleeding from [22] Sanyal AJ, Mirshahi F. Endothelial cells lining transjugular intrahepatic por- esophageal varices. Hepatology 1997;26:1115–1122. doi: 10.1002/hep. tasystemic shunts originate in hepatic sinusoids: implications for pseudoin- 510260505. timal hyperplasia. Hepatology 1999;29:710–718. doi: 10.1002/hep. [4] Gülberg V, Schepke M, Geigenberger G, Holl J, Brensing KA, Waggershauser T, 510290323. et al. Transjugular intrahepatic portosystemic shunting is not superior to [23] Orozco H, Mercado MA, Chan C, Guillén-Navarro E, López-Martínez LM. endoscopic variceal band ligation for prevention of variceal rebleeding in cir- A comparative study of the elective treatment of variceal hemorrhage with rhotic patients: a randomized, controlled trial. Scand J Gastroenterol 2002;37: beta-blockers, transendoscopic sclerotherapy, and surgery: a prospective, 338–343. doi: 10.1080/003655202317284255. controlled, and randomized trial during 10 years. Ann Surg 2000;232:216– [5] Axley P, Ahmed Z, Ravi S, Singal AK. Hepatitis C virus and hepatocellular 219. doi: 10.1097/00000658-200008000-00011. carcinoma: A narrative review. J Clin Transl Hepatol 2018;6:79–84. doi: [24] Henderson JM, Kutner MH, Millikan WJ Jr, Galambos JT, Riepe SP, Brooks WS, 10.14218/JCTH.2017.00067. et al. Endoscopic variceal sclerosis compared with distal splenorenal shunt to [6] Ma KW, Chok KSH, Fung JYY, Lo CM. Liver transplantation for hepatitis B prevent recurrent variceal bleeding in cirrhosis. A prospective, randomized virus-related hepatocellular carcinoma in Hong Kong. J Clin Transl Hepatol trial. Ann Intern Med 1990;112:262–269. doi: 10.7326/0003-4819-112-4- 2018;6:283–288. doi: 10.14218/JCTH.2017.00058. 262. [7] García-Pagán JC, Caca K, Bureau C, Laleman W, Appenrodt B, Luca A, et al. [25] Cello JP, Grendell JH, Crass RA, Weber TE, Trunkey DD. Endoscopic sclero- Early use of TIPS in patients with cirrhosis and variceal bleeding. N Engl J Med therapy versus portacaval shunt in patients with severe cirrhosis and acute 2010;362:2370–2379. doi: 10.1056/NEJMoa0910102. variceal hemorrhage. Long-term follow-up. N Engl J Med 1987;316:11–15. [8] Elwan N, Salem ML, Kobtan A, El-Kalla F, Mansour L, Yousef M, et al. High doi: 10.1056/NEJM198701013160103. numbers of myeloid derived suppressor cells in peripheral blood and ascitic [26] Cello JP, Grendell JH, Crass RA, Trunkey DD, Cobb EE, Heilbron DC. Endo- fluid of cirrhotic and HCC patients. Immunol Invest 2018;47:169–180. doi: scopic sclerotherapy versus portacaval shunt in patients with severe cirrho- 10.1080/08820139.2017.1407787. sis and variceal hemorrhage. N Engl J Med 1984;311:1589–1594. doi: 10. [9] Abdelfattah AAM, Rizk F, Hawash N, Hanafy A, El-Kalla F, Abd-Elsalam S. 1056/NEJM198412203112501. Randomized trial of preoperative administration of oral pregabalin for post- [27] Lo GH, Liang HL, Chen WC, Chen MH, Lai KH, Hsu PI, et al. A prospective, operative analgesia in patients scheduled for radiofrequency ablation of focal randomized controlled trial of transjugular intrahepatic portosystemic shunt lesions in the liver. Int J Hyperthermia 2018;34:1367–1371. doi: 10. versus cyanoacrylate injection in the prevention of gastric variceal rebleed- 1080/02656736.2018.1424946. ing. Endoscopy 2007;39:679–685. doi: 10.1055/s-2007-966591.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 232–237 237 Original Article

Effect of Autologous Bone Marrow Stem Cell Therapy in Patients with Liver Cirrhosis: A Meta-analysis

Chuan-Xin Wu#1, Deng Wang#1, Ying Cai2, Ao-Ran Luo2 and Hang Sun*2

1Department of Hepatobiliary Surgery, The Second Affiliated Hospital, Chongqing Medical University, Chongqing, China; 2Key Laboratory of Molecular Biology for Infectious Diseases (Ministry of Education), Institute for Viral Hepatitis, Department of Infectious Diseases, The Second Affiliated Hospital, Chongqing Medical University, Chongqing, China

Abstract tients with liver cirrhosis: A meta-analysis. J Clin Transl Hep- atol 2019;7(3):238–248. doi: 10.14218/JCTH.2019.00008. Background and Aims: Although autologous bone marrow stem cell (BMSC) transplantation is an effective treatment for Introduction liver cirrhosis, there are few reports describing the optimal delivery route and number of injected BMSCs. Methods: A Liver cirrhosis is a severe disease of the digestive system, literature search was conducted using PubMed, ISI Web of associated with many complications, poor prognosis, and a Science, Cochrane Central Register of Controlled Trials, and high rate of morbidity and mortality worldwide. The main EBSCO. A meta-analysis was performed to assess the effect causes of chronic liver disease include infection by hepatitis B of BMSCs on liver and coagulation function indices. Subgroup virus and hepatitis C virus, excessive alcohol consumption, analysis was performed based on number of injected BMSCs, primary biliary cirrhosis, and autoimmune liver disease.1 delivery route, and length of follow-up. Results: A total of Chronic liver disease frequently progresses to liver cirrhosis, 15 studies were selected from among 1903 potential studies following different processes that involve liver cell degenera- for analysis. Autologous BMSC transplantation significantly tion and extensive necrosis.2 improved aspartate aminotransferase, total bilirubin, albu- The key to achieving a curative effect against cirrhosis is to min, prothrombin time, prothrombin activity, prothrombin effectively improve the rate of regeneration of damaged liver concentration, Child-Pugh score, and model for end-stage liv- cells and reduce the accumulation of fibrous tissues. While er disease. In the subgroup analysis of cell numbers, all four liver transplantation is a treatment option for those with liver of the indices were significantly improved when the number of 8 cirrhosis, a multitude of contraindications, such as smoking BMSCs was >4 3 10 . The subgroup analysis referring to the history, alcohol abuse, and other risky lifestyle habits, com- delivery route showed that arterial infusion increased the bined with general donor liver shortages, exorbitant trans- therapeutic effect over venous infusion. Finally, in the sub- plant costs, and other complicating factors, limit its group analysis of follow-up length, the results showed that utilization.3–5 Therefore, it is crucial at this time to develop BMSC therapy significantly improved liver function at 2 weeks new strategies for treatment of liver cirrhosis. after transplantation. In addition, this therapy improved co- Recent clinical studies have shown that cellular therapy has agulation 4 weeks after the transplant, with a maintenance of the potential to enhance liver regeneration and modulate the efficacy for up to 24 weeks. Conclusions: Autologous BMSC disease course, thus representing an alternative treatment therapy is beneficial for liver improvement and coagulation in strategy to organ transplantation.6 Many recent studies have patients with liver cirrhosis. The therapeutic effect was gen- reported that autologous bone marrow stem cells (BMSCs) par- erated at 2–4 weeks after transplantation. The effect lasted ticipate in the repair, reconstruction, and restoration of liver for 24 weeks but no more than 48 weeks. The greatest benefit 7–11 8 function. The antifibrosis effects of autologous BMSC to patients was observed with a 4 3 10 autologous BMSC therapy in the injured liver have been clearly demonstrated in transplant via the hepatic artery. animal models.12,13 Moreover, clinical trials have shown that Citation of this article: Wu CX, Wang D, Cai Y, Luo AR, Sun autologous BMSC transfusion can quickly improve liver func- H. Effect of autologous bone marrow stem cell therapy in pa- tion, without significant side effects.10,14–16 Although many of these studies are still in the pilot stage, preliminary results have Keywords: Autologous bone marrow stem cells; Liver cirrhosis; Cell therapy; suggested that autologous BMSC transplantation is a safe and Meta-analysis. effective treatment for liver cirrhosis.5,16–18 Autologous BMSC Abbreviations: ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMSC, bone marrow stem cell; CI, confidence interval; INR, therapy avoids the limiting factor of donor liver availability, and international normalized ratio; MELD, model for end-stage liver disease; PC, pro- there is no risk of transplant rejection or surgical complications. thrombin concentration; PT, prothrombin time; PTA, prothrombin activity; SMD, These factors present a significant advancement over previous standardized mean difference; TBIL, total bilirubin; WMD, weighted mean differ- procedures. ence. Received: 31 January 2019; Revised: 14 July 2019; Accepted: 4 August 2019 However, studies of autologous BMSC therapy have #These two authors contributed equally to this study. revealed several limitations. First, outcomes have been exten- *Correspondence to: Hang Sun, Key Laboratory of Molecular Biology for Infec- sively described in animal models, while existing data regard- tious Diseases (Ministry of Education), Institute for Viral Hepatitis, Department of ing adult human patients remain deficient. Secondly, there Infectious Diseases, The Second Affiliated Hospital, Chongqing Medical University, No. 76 Linjiang Road, Yuzhong District, Chongqing 400010, China. Tel: +86- have been several published articles reporting on safety, 13527599558, Fax: +86-23-63829191, E-mail: [email protected] feasibility, and therapeutic efficacy but few reports on the

238 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 238–248

Data extraction Data source Three reviewers (Chuan-Xin Wu, Deng Wang, and Ying Cai) We performed a literature search to identify eligible studies independently extracted relevant data according to previous published on or before August 15, 2018, that explored the data extraction methods. The extraction results were eval- therapeutic effects of autologous BMSC therapy for liver uated by two other reviewers (Hang Sun and Ao-Ran Luo). cirrhosis, and using the databases of PubMed (http://www. Disagreements were resolved by discussion. The extracted ncbi.nlm.nih.gov/pubmed), ISI Web of Science (http:// data included the first author, year of publication, country of wokinfo.com/), Cochrane Central Register of Controlled Trials origin, type of disease, study design, sample size, dose and (http://community.cochrane.org/), and EBSCO (http://www. route of cell administration, and length of follow-up. ebscohost.com/). The search terms were “liver cirrhosis” or “cirrhosis” combined with “bone marrow stem cell” or “autologous bone marrow mononuclear cell” or “BMSCs”. Moreover, Quality appraisal we supplemented our search by screening the reference lists of all relevant studies, including original articles, reviews, and Methodological quality was evaluated using the Delphi criteria for quality assessment of randomized clinical trials for con- meta-analyses. References to all identified publications were 20 entered into reference-managing software (EndNote, version ducting systematic reviews with additional items. Items X6; http://endnote.com/). specifically important for evaluation of the curative effect of patients with cirrhosis were also added, the detailed items of Delphi are described in Table 1. Each item in this quality list Inclusion criteria had the same weight. For each publication, a quality score “ ” The initial screening of titles and abstracts was independently was calculated, where yes received a score of 1 point for a certain quality item and “no” and “do not know” received a performed by two reviewers (Chuan-Xin Wu and Deng Wang). A score of 0 points. For scoring quality items on masking, allo- second screening of the full-text was conducted by the same cation concealment, and intention-to-treat analysis, we used reviewers. Then, the included studies were evaluated to deter- 21–24 mine whether they were in accordance with the cross-checking the recommendations of Berger et al., with a total achievable score of 14 points. Low-quality research received method. When necessary, disagreements were settled by a score between 0 and 7 points, while high-quality research reaching a consensus with a third party (Hang Sun). All studies included in this meta-analysis met the following criteria: received a score between 8 and 14 points. The quality of each study was independently assessed by four reviewers (Chuan- (1) Randomized controlled trial study or case-control Xin Wu, Deng Wang, Ying Cai, and Hang Sun). Disagreements study in design. on rating were resolved through discussions among members (2) Topic of autologous BMSC therapy for liver cirrhosis. of the research group until a consensus was reached. (3) Patients age >15 years. (4) Clinical symptoms of both cases and controls avail- Statistical analysis able to calculate the weighted mean difference (WMD), standardized mean difference (SMD), and The pooled WMD or SMD with corresponding 95% CIs were corresponding 95% confidence interval (CI). calculated to evaluate the curative effect of autologous BMSC (5) All patients received a definitive diagnosis of cirrhosis transplantation in patients with cirrhosis, and the x2 and I2 by abdominal ultrasound and portal hypertension indices were calculated to assess possible heterogeneity with abnormal serum albumin (ALB) and/or total between individual studies.25 A fixed-effects model was bilirubin (TBIL) levels and/or prothrombin time (PT), applied to calculate the pooled WMD/SMD with its 95% CI platelet count $30,000/mm3, and were able to give when there was no obvious heterogeneity between studies; informed consent. otherwise, the random-effects model was used.26,27 Sensitiv- (6) Availability of liver function parameters (levels of ity analyses were conducted by omitting certain studies each alanine aminotransferase (ALT), aspartate amino- time, such as studies that combined BMSC transplantation transferase (AST), TBIL and ALB; PT; prothrombin with other drug therapies. Moreover, we performed subgroup activity (PTA); prothrombin concentration (PC); inter- analysis stratified by the number of infused BMSCs, delivery national normalized ratio (INR); Child-Pugh score and route, and length of follow-up. Publication bias was assessed model for end-stage liver disease (MELD) score). using funnel plots and the Egger’s linear regression test.28

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 238–248 239 Wu C.X. et al: BMSC therapy for liver cirrhosis

Table 1. Quality items included for quality assessment, source from which the quality item was obtained, and number of publications that had a positive quality score, per quality item

Item code Source Quality item

A Delphi list Was a method of randomization used? B Added by Was the period of outcome y authors measurements equal among all groups? C Considered Is it unlikely that compliance for Delphi list may explain differences between groups? D Added by Are side effects reported? authors z E Delphi list Were inclusion criteria specified? z F Delphi list Were exclusion criteria specified? G Considered Are the interventions described for Delphi list explicitly? H Delphi list Were the groups similar at baseline regarding the most important prognostic indicators? I Delphi list Was the patient masked to the treatment? Fig. 1. Flow diagram (modified from The PRISMA Flow Diagram) and G Considered Was calculation of statistical results of literature review. for Delphi list power reported after allocation to the treatment? K Delphi list Was an intention-to-treat Study characteristics analysis performed? All of the studies that were analyzed were published between L Delphi list Was the treatment allocation 2010 and 2017. All of these studies used an injection of concealed? 107–1012 cells in the treatment group. The studies were con- M Delphi list Was the outcome assessor ducted in six countries (China, Iran, Egypt, Switzerland, blinded? Austria, and Japan). Nine of the studies were randomized 14,17,29,32,33,35,37–39 N Delphi list Was the care provider blinded? controlled trials, while the other six were 15,16,30,31,34,36 y case-control studies. The authors added items to improve interpretations of liver cirrhosis measure- A study by Liu et al.33 used a combination of lamivudine and ments. z adefovir dipivoxil with autologous BMSC transplantation in the Item split into inclusion and exclusion criteria. treatment group, while the study by Peng et al.15 administered reduced glutathione, glycyrrhizin, ademetionine, and human serum ALB to both the BMSC group and the control group. All Statistical analyses were performed using Review Manager of the patients in the study by Liao et al.28 received a required 5.3 and Stata/SE software. A two-sided probability (p) value devascularization and splenectomy. None of the other trials of <0.05 was considered statistically significant. employed any additional treatments. All cases were diagnosed based on clinical testing, evalua- Results tion of consolidation and infiltrates by abdominal ultrasound, and laboratory examinations. Controls were matched with cases according to sex, age, ethnicity, and residential area. Literature search After assessment of risk bias using the Delphi list27 with additional items, most studies were considered high-quality A total of 1903 potential and relevant publications up to research (Delphi list score $9). The detailed characteristics August 15, 2018 were systematically retrieved from the and risks of the included studies are described in Table 2 and electronic databases. After screening of titles and abstracts, Table 3. 32 studies were identified after excluding duplicate studies, reviews, and reports not pertinent to the effect of autologous BMSC therapy on liver cirrhosis. Among these, 17 studies Therapeutic effect of autologous BMSC therapy for were excluded after full-text screening because they did not treatment of liver cirrhosis match the inclusion criteria. Finally, 15 studies14–17,29–39 were included in this meta-analysis. A flow chart of the A total of 15 studies, which included 750 participants (383 article selection process is shown in Fig. 1. cases and 367 controls), were included in this meta-analysis

240 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 238–248 Table 2. Characteristics of studies included in the meta-analysis C.X. Wu

Age (years, mean 6 SD) Child- Type of Patients and al et Case/ Sex Treated Control Pugh infused cells disease Follow-up StudyRef Year Country Design controls (male %) group group class and volume Cell dose Route etiology, n length cirrhosis liver for therapy BMSC :

Salama et al.29 2010 Egypt RCT 90/50 117 50.2 6 50.9 6 B-C Autologous 5*107 PV HBV + HCV 6 months (78%) 6.05 7.23 CD34+ and CD133+ stem cells ora fCiia n rnltoa Hepatology Translational and Clinical of Journal (250 mL) Lyra et al.14 2010 Austria RCT 15/15 NA 56.7 6 50.0 6 A-C Autologous (3.78 6 HA 30 12 9.23 10.37 BMCs (20 2.69)* alcoholic months mL) 108 Amer et al.16 2011 Egypt Case- 20/20 33 NA NA C Autologous 2 3 108 Intrasplenic 40 HCV 6 months control (82.5%) BM-MSCs or (95 6 25 intrahepatic mL) Peng et al.15 2011 China Case- 39/77 NA 42.19 6 42.22 6 NA Autologous (3.4– HA 116 HBV 3 months control 10.80 1.37 BM-MSCs 3.8)*108 (100–120 mL) El-Ansary 2012 Egypt Case- 10/15 19 48.0 6 51.6 6 NA Autologous 106/kg PV 25 HCV 6 months et al.30 control (76%) 7.4 7.2 BM-MSCs (90 mL) Mohamadnejad 2013 Iran RCT 11/14 13 43.1 6 34.6 6 B-C Autologous (1.2– PV 2 HBV + 1 12 et al.32 (52%) 17.6 13.8 BM-MSCs 2.95)* HCV + 2 months (100 mL) 108 PBC + 9 AIH + 11 2019 cryptogenic Liao et al.31 2013 China Case- 6/6 9 (75%) NA NA NA Autologous (9.08 6 HA 12 HBV 3 months

o.7|238 | 7 vol. control BMDLSCs 1.77)* (30–40 108 mL) Xu et al.17 2014 China RCT 20/19 24 44 6 12 45 6 10 NA Autologous (8.45 6 HA 39 HBV 6 months (64.9%) BM-MSCs 3.28) 8 – (130–150 310 4 241 248 mL) Bai et al.34 2014 China Case- 32/15 29 46.4 6 47.4 6 B-C Autologous NA HA 43 HBV + 4 24 control (61.7%) 11.6 11.1 BM-MNCs alcoholic months (80–100 mL) Liu et al.33 2014 China RCT 40/37 70 51.6 6 50.4 6 B-C Autologous (3.2 6 HA 77 HBV 3 months (91%) 9.2 8.5 BMSCs (10 1.6)* mL) 1011

(continued) 4 ora fCiia n rnltoa Hepatology Translational and Clinical of Journal 242

Table 2. (continued ) Age (years, mean ± SD) Child- Type of Patients and Case/ Sex Treated Control Pugh infused cells disease Follow-up StudyRef Year Country Design controls (male %) group group class and volume Cell dose Route etiology, n length

Salama et al.35 2014 Egypt RCT 20/20 117 50.27 6 50.9 6 B-C Autologous 106/kg PV 40 HCV 6 months (78%) 6.05 7.23 CD34+ and CD133+ stem cells (100 mL) Al Tayeb et al.36 2015 Egypt Case- 10/10 20 49.20 6 49.20 6 B-C Autologous (0.25– HA 20 HBV 6 months control (100%) 3.27 3.27 BMSCs (50 1.91)* mL) 108 Suk et al.38 2016 Korea RCT 18/18 32 53.1 6 53.7 6 B-C Autologous 5*107 HA 48 12months (89%) 8.7 8.2 BMSCs (10 alcoholic mL) Mohamadnejad 2016 Tehran RCT 10/9 12 43.90 6 46.22 6 B-C Autologous (7.62 6 PV 8 AIH + 1 6 months 37 2019 et al. (63%) 14.84 15.31 BMSCs (20 5.53)* PBC+ 10 mL) 108 cryptogenic uC.X. Wu Wu et al.39 2017 China RCT 42/42 49 49 6 11 50 6 10 B-C Autologous 106/kg HA 84 HBV 6 months o.7|238 | 7 vol. (58%) BM-MSCs (80 mL) tal et

Abbreviations: AIH, autoimmune hepatitis; BMCs, bone marrow cells; BMDLSCs, bone marrow-derived liver stem cells; BMSCs, bone marrow stem cells; BM-MNCs, bone marrow mononuclear cells; BM-MSCs, bone

marrow mesenchymal stem cells; HA, hepatic artery; HBV, hepatitis B virus; HCV, hepatitis C virus; PBC, primary biliary cirrhosis; PV, portal vein or peripheral vein; RCT, randomized controlled trial. cirrhosis liver for therapy BMSC : – 248 Wu C.X. et al: BMSC therapy for liver cirrhosis

Table 3. Risk of bias of included trials

Publication Withdrawals and Quality Quality criteria not StudyRef Design type dropouts score fulfilled

Salama et al.29 RCT Article No 10 i, l, m, n Lyra et al.14 RCT Article No 10 i, l, m, n Amer et al.16 Case- Article No 9 a, i, l, m, n control Peng et al.15 Case- Article No 9 a, i, l, m, n control El-Ansary et al.30 Case- Article No 9 a, i, l, m, n control Mohamadnejad RCT Article No 12 m, n et al.32 Liao et al.31 Case- Article No 9 a, i, l, m, n control Xu et al.17 RCT Article No 12 m, n Bai et al.34 Case- Article No 9 a, i, l, m, n control Liu et al.33 RCT Article No 10 i, l, m, n Salama et al.35 RCT Article No 10 i, l, m, n Al Tayeb et al.36 Case- Article No 9 a, i, l, m, n control Suk et al.38 RCT Article No 10 i, l, m, n Mohamadnejad RCT Article No 14 No* et al.37 Wu et al.39 RCT Article No 10 i, l, m, n

No*, the study got 14 “Yes” in the risk test. Abbreviation: RCT, randomized controlled trial.

to explore the therapeutic effect of autologous BMSC therapy 95% CI, −1.46 to −0.15, p = 0.02; TBIL: SMD, −0.62, 95% for treatment of liver cirrhosis. After injection of autologous CI, −0.96 to −0.28, p = 0.0003; ALB: SMD, 0.64, 95% CI, BMSCs, most liver function and coagulation indices were 0.19 to 1.09, p =0.02;MELD:WMD,−1.39, 95% CI, −2.70 significantly improved, except for ALT and INR. (ALT: SMD, to −0.07, p < 0.0001). The data are shown in Fig. 2. −0.30, 95% CI, −1.00 to 0.40, p = 0.4; AST: SMD, −6.26, 95% CI, −11.97 to −0.54, p = 0.03; TBIL: SMD, −0.58, 95% Subgroup meta-analyses of delivery route CI, −0.76 to −0.41, p < 0.0001; ALB: SMD, 0.62, 95% CI, 0.22 to 1.02, p = 0.002; PT: WMD, −2.53, 95% CI, −4.27 to Of the 15 studies examined, 14 clearly indicated that the −0.79, p = 0.004; PTA: WMD, 4.12, 95% CI, 0.28 to 7.96, transfusion route was either arterial or venous, with the p = 0.04; PC: WMD, 16.24, 95% CI, 11.30 to 21.19, p < 0.0001; exception of Amer et al.16 where BMSCs were directly injected INR: WMD, −0.05, 95% CI, −0.24, 0.13, p = 0.57; Child-Pugh into the liver or spleen. Therefore, a subgroup analysis was score: WMD, −0.80, 95% CI, −1.52 to −0.09, p = 0.03; MELD: performed based on differences in transfusion route. AST, WMD, −1.80, 95% CI, −2.97 to −0.64, p = 0.02). TBIL, ALB, and MELD were selected as the evaluation To identify factors related to the efficacy of autologous indices. Compared with the control group, all of the four BMSC therapy, subgroup meta-analyses were conducted to indices in the hepatic arterial group were significantly analyze the effect of the number of cells injected, the infusion increased (AST: SMD, −0.86, 95% CI, −1.29 to −0.43, route, and the follow-up time on efficacy. p = 0.0001; TBIL: SMD, −0.47, 95% CI, −0.67 to −0.27, p < 0.0001; ALB: SMD, 0.59, 95% CI, 0.4 to 0.79, p < 0.0001; Subgroup meta-analyses of number of injected BMSCs MELD: WMD, −2.4, 95% CI, −3.31 to −1.48, p < 0.0001). However, only the TBIL level was significantly increased in the Based on the studies analyzed, the number of BMSCs injected portal or peripheral vein group (SMD, −0.97, 95% CI, −1.34 was separated into three groups: BMSCs #5 3 107;53 107 < to −0.6, p < 0.0001). The data are shown in Fig. 3. BMSCs #4 3 108; BMSCs >4 3 108. ALT, TBIL, ALB, and MELD were chosen as the evaluation indices because the others Subgroup meta-analyses of follow-up length lacked literature support. The results indicated that when BMSCs >4 3 108, all of the four indices were significantly To explore the efficacy of BMSC therapy, 2, 4, 8, 12, 24 and 48 improved compared to the control group (ALT: SMD, −0.8, weeks were chosen as the time points to perform subgroup

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 238–248 243 Wu C.X. et al: BMSC therapy for liver cirrhosis

Fig. 2. Forest plot of the number of injected BMSCs. Red indicates significant improvement compared with the control group; blue indicates no significant improvement. In the group of BMSCs #5 3 107, TBIL, ALB and MELD were significantly improved (p < 0.05). In the group of 5 3 107 < BMSCs #4 3 108, only MELD was significantly improved (p < 0.05). In the group of BMSCs >4 3 108, ALT, TBIL, ALB and MELD were significantly improved (p < 0.05). Abbreviations: ALB, albumin; ALT, alanine aminotransferase; BMSCs, bone marrow stem cells; F, fixed effects model; MELD, model for end-stage liver disease; Phet, p value of heterogeneity; R, random effects models; TBIL, total bilirubin. analyses. The results showed that after injection of autolo- Sensitivity analysis gous BMSCs, liver function significantly improved at 2 weeks. In addition, coagulation function significantly improved at Sensitivity analyses were conducted by sequentially omitting 4 weeks, and the efficacy could be maintained for 24 weeks. individual studies and performing comparisons between the Three studies carried out observation for up to 48 weeks. The results of pooled WMD/SMD for the random and fixed effects results indicated that ALB did not show an improvement models. We found no material changes in any liver function compared to controls (SMD, −0.00, 95% CI, −1.29 to 1.29, indices under some conditions, while the indicators for p = 1.00). The other indices could not be analyzed because of heterogeneity were reduced. an insufficient number of studies. The data are shown in Fig. 4. Publication bias

Side effects Publication bias among the included studies was assessed using the Begg’s and Egger’s tests, as these tests are often There were no significant adverse effects after autologous used to provide evidence of publication bias. There was no BMSC transfusion. There were no serious side effects or obvious asymmetry among the risk factors. TBIL is an complications observed in short-term or long-term follow-up important laboratory parameter to evaluate improvement in in 12 reports, which included 264 cases. However, in the liver function, thus a representative funnel plot for TBIL is study by Mohamadnejad et al.,32 which included 14 patients, shown in Fig. 5. 3 in the autologous BMSCs group died of liver failure after cellular infusion. The clinical study by Lyra et al.,14 which Discussion included 15 patients, reported that 2 experienced mild pain at the sites of bone marrow puncture. There were no other Autologous BMSC therapy is less expensive and relatively complications or specific side effects related to the infusion. easier to administer than liver transplantation, and results in In four reports, by Amer et al.,16 Xu et al.,17 Salama a lower rate of transplant rejection and surgical complica- et al.,29 and Liu et al.,33 which included 30 patients in the tions. The results of several uncontrolled studies suggested autologous BMSCs groups, fever was observed within 24 h. that infusion of autologous BMSCs might transiently improve In summary, there were no significant side effects following liver function in some patients with cirrhosis.8,40–42 Existing autologous BMSC therapy in the treatment of chronic liver meta-analyses of BMSC therapy for liver cirrhosis are insuffi- disease. cient in regard to their included studies, and only a single

244 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 238–248 Wu C.X. et al: BMSC therapy for liver cirrhosis

Fig. 3. Forest plot of the delivery route. Red indicates significant improvement compared with the control group; blue indicates no significant improvement. AST, TBIL, ALB and MELD were significantly improved (P < 0.05) in group HA. Only TBIL was significantly improved (P < 0.05) in group PV. Abbreviations: ALB, albumin; AST, aspartate aminotransferase; F, fixed effects model; HA, hepatic artery; MELD, model for end-stage liver disease; Phet, p value of heterogeneity; PV, portal vein or peripheral vein; R, random effects models; TBiL, total bilirubin. subgroup analysis has been performed. This is limited to a through subgroup analysis whether the number of cells study of the changes of liver and coagulation function at dif- injected affected the therapeutic effect. The results showed ferent follow-up times.43,44 However, our study included more that when BMSCs >4 3 108, all indices were significantly publications than previous studies, with a larger sample size improved compared to the control group, but the other two and more comprehensive and reliable analysis results. groups (BMSCs #5 3 107 and 5 3 107 < BMSCs #4 3 108) had In terms of subgroup analysis, we also performed sub- no such effect. This finding suggested that the number of cells group analysis of the number of injected BMSCs and delivery injected was an important factor influencing the efficacy of route, and discussed their impact on efficacy. Therefore, this autologous BMSC therapies. In this study, we found that study fills the gaps in previous studies and provides more BMSCs >4 3 108 were more beneficial to patients. At robust and accurate statistical evidence to evaluate the present, there are few clinical trials that have investigated efficacy of autologous BMSC therapy for treatment of cir- the relationship between efficacy and BMSC number, thus rhosis. Because meta-analyses have larger sample sizes, more clinical trial results are needed to verify this conclusion. differences caused by random errors are reduced and test Regarding the route of infusion, most clinical trials chose efficiency is improved. Furthermore, such analyses are intravenous infusion or arterial infusion. Although intrave- helpful to clarify current controversies and provide the best nous infusion is simple to perform and manage, arterial evidence for clinical practice. infusion contributes to homing of marrow stem cells to Our meta-analysis of 15 studies included 383 patients who avoid phagocytosis by reticuloendothelial cells.45 The results underwent autologous BMSC therapy and 367 controls. The showed that all four indices were significantly improved in the results showed that the indicators of liver and coagulation arterial injection group; however, only the TBIL was improved function except for ALT and INR might be significantly in the intravenous injection group compared to controls. Arte- improved after autologous BMSC therapy, which was consis- rial infusion increased the beneficial therapeutic effect. This is tent with the reports of Ma et al.44 In this study, BMSC therapy consistent with the findings from a study by Zhao et al.44–46 did not improve the function of ALT and INR. The reason may Kwak et al.47 reported that the improvement of liver func- be that only some of the 15 studies we included observed tion decreased with time. It was further observed that the changes of ALT and INR, and the results were different in duration of liver function improved after autologous BMSC different studies. Therefore, the current analysis results can therapy. The results showed that after injection of autologous only be used as a reference, and more studies with larger BMSCs, liver function significantly improved at 2 weeks, samples are needed. Perhaps ALT and INR will be significantly coagulation function significantly improved at 4 weeks, and improved when more studies are included. the efficacy could be maintained for 24 weeks. Therefore, it Of the 15 selected articles, 14 indicated the number of was speculated that the efficacy of BMSC therapy was no cells injected, but the reasons for choosing the numbers were more than 48 weeks. However, this conclusion requires not given in these articles. The search intended to explore more long-term follow-up clinical trial results to substantiate.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 238–248 245 Wu C.X. et al: BMSC therapy for liver cirrhosis

Fig. 4. Forest plot of the length of follow-up. Red indicates significant improvement compared with the control group; blue indicates no significant improvement. Liver function indices, such as ALB and MELD, were significantly improved at 2 weeks and maintained for 24 weeks (p < 0.05). Coagulation function indices, such as PT and PC, were significantly improved at 4 weeks and maintained for 24 weeks (p < 0.05). ALB was non-significantly improved at 48 weeks (p > 0.05). Abbreviations: ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; F, fixed effects model; INR, international normalized ratio; MELD, model for end-stage liver disease; PC, prothrombin concentration; PT, prothrombin time; PTA, prothrombin activity; Phet, p value of heterogeneity; R, random effects models; TBIL, total bilirubin.

These data raised the question of whether 48 weeks would be decreased in sensitivity analyses after omitting studies with an appropriate time point for a second infusion of autologous small sample sizes. We carried out three different subgroup BMSCs. In a study by Suk et al.,38 the authors compared the analyses, and found that the heterogeneity of some indicators efficacy of BMSC treatment between one-time and two-time was significantly reduced, so the number of cells injected, the injections. The second injection took place at 30 days after infusion route, and the follow-up time could influence hetero- the first one, with a total of 5 3 107 BMSCs injected per treat- geneity. Therefore, it appears as though the above-mentioned ment. The results showed that there was no significant differ- factors contributed to overall heterogeneity. ence between the two groups. According to our analysis, the There were several limitations to this study that may have time chosen by the author for the second treatment was still affected the results. First, only four databases were explored. within the effective range of the first treatment, therefore no Relevant articles published in other databases, as well as better outcome was achieved. unpublished studies, may have been overlooked. Lack of a It is also important to mention that heterogeneity existed in complete repertoire of manuscripts may have altered our our study. For liver function tests, significant heterogeneity of results. Second, there may have been clinical heterogeneity some indicators was detected in the combination studies. One between studies. For example, we had strict enrollment study30 assessed a combination of lamivudine, adefovir dipi- criteria of references which included studies that were con- voxil, and autologous BMSC transplantation in the treatment ducted in different countries. As such, the diagnostic group, while a second25 evaluated reduced glutathione, glycyr- methods, such as abdominal ultrasound and liver biopsy, rhizin, ademetionine, and human serum ALB administered to and the basic condition of eligible patients may have varied both the BMSC and control groups, which may have added to greatly. Also, these factors may have contributed to greater the heterogeneity. However, this heterogeneity was effectively heterogeneity between the included studies. Third, some

246 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 238–248 Wu C.X. et al: BMSC therapy for liver cirrhosis

Acknowledgments

This work was supported by a grant from the National Natural Science Foundation of China (No. 81871608).

Conflict of interest

The authors have no conflict of interests related to this publication.

Author contributions

Conception of study design (HS), data collection (CXW, DW), data analysis (CXW, DW), statistical analysis (YC, ARL) and drafting the article (CXW). All authors have read and approved the final version of the manuscript. Fig. 5. A Begg’s plot of TBIL. The horizontal line in the funnel plot indicates the fixed effects summary estimate, while the sloping lines indicate the expected 95% confidence intervals for a given standard error, assuming no heterogeneity be- References tween studies. No publication bias was observed among studies using Egger’s (p = 0.91) test, which suggested that there was no evidence of publication bias. [1] Bosch FX, Ribes J, Borràs J. Epidemiology of primary liver cancer. Semin Liver Abbreviation: TBIL, total bilirubin. Dis 1999;19:271–285. doi: 10.1055/s-2007-1007117. [2] Lee WM. Hepatitis B virus infection. N Engl J Med 1997;337:1733–1745. doi: 10.1056/NEJM199712113372406. [3] Dai LJ, Li HY, Guan LX, Ritchie G, Zhou JX. The therapeutic potential of bone studies had small patient cohorts, which may have affected marrow-derived mesenchymal stem cells on hepatic cirrhosis. Stem Cell Res – the statistical significance of the publication bias. Although 2009;2:16 25. doi: 10.1016/j.scr.2008.07.005. ’ [4] Goldberg D, French B, Trotter J, Shetty K, Schiano T, Reddy KR, et al. Under- publication bias in the meta-analysis according to the Egger s reporting of liver transplant waitlist removals due to death or clinical deteri- test was not significant, there was relatively little bias in the oration: results at four major centers. Transplantation 2013;96:211–216. summary effect size estimate; thus, these test results must doi: 10.1097/TP.0b013e3182970619. [5] Terai S, Takami T, Yamamoto N, Fujisawa K, Ishikawa T, Urata Y, et al. Status be interpreted with caution. Due to the limitation of the and prospects of liver cirrhosis treatment by using bone marrow-derived cells literature, no subgroup analysis of different etiologies of and mesenchymal cells. Tissue Eng Part B Rev 2014;20:206–210. doi: 10. liver cirrhosis and BMSC cell types was carried out in this 1089/ten.TEB.2013.0527. [6] Lee Z, Dennis JE, Gerson SL. Imaging stem cell implant for cellular-based thera- study. All these need further research in the future. pies. Exp Biol Med (Maywood) 2008;233:930–940. doi: 10.3181/0709-MR-234. At present, clinical trials of BMSC therapy for liver cirrhosis [7] Pai M, Zacharoulis D, Milicevic MN, Helmy S, Jiao LR, Levicar N, et al. Autol- have been carried out at home and abroad. Extraction, ogous infusion of expanded mobilized adult bone marrow-derived CD34+ cells into patients with alcoholic liver cirrhosis. Am J Gastroenterol 2008; expansion and transplantation of stem cells can be carried 103:1952–1958. doi: 10.1111/j.1572-0241.2008.01993.x. out in large general hospitals without major technical diffi- [8] Terai S, Ishikawa T, Omori K, Aoyama K, Marumoto Y, Urata Y, et al. Improved culties. The main problem is that there is no unified standard liver function in patients with liver cirrhosis after autologous bone marrow cell infusion therapy. Stem Cells 2006;24:2292–2298. doi: 10.1634/stem- therapy process at present, such as the number of injected cells.2005-0542. BMSCs, delivery route, the time and frequency of infusion, [9] Levicar N, Pai M, Habib NA, Tait P, Jiao LR, Marley SB, et al. Long-term clinical and so on. In addition, the current research mainly relies on results of autologous infusion of mobilized adult bone marrow derived CD34+ cells in patients with chronic liver disease. Cell Prolif 2008;41 Suppl 1:115– the changes of liver and coagulation function to evaluate the 125. doi: 10.1111/j.1365-2184.2008.00491.x. curative effect, which is relatively limited. It is hoped that [10] Mohamadnejad M, Alimoghaddam K, Mohyeddin-Bonab M, Bagheri M, there will be more judgement indices to evaluate the curative Bashtar M, Ghanaati H, et al. Phase 1 trial of autologous bone marrow mesenchymal stem cell transplantation in patients with decompensated liver effect comprehensively. cirrhosis. Arch Iran Med 2007;10:459–466. [11] Khan AA, Parveen N, Mahaboob VS, Rajendraprasad A, Ravindraprakash HR, Venkateswarlu J, et al. Safety and efficacy of autologous bone marrow stem Conclusions cell transplantation through hepatic artery for the treatment of chronic liver failure: a preliminary study. Transplant Proc 2008;40:1140–1144. doi: 10. 1016/j.transproceed.2008.03.111. The findings of this meta-analysis indicated that autologous [12] Zhao DC, Lei JX, Chen R, Yu WH, Zhang XM, Li SN, et al.Bonemarrow-derived BMSC therapy may be beneficial to improve liver and coag- mesenchymal stem cells protect against experimental liver fibrosis in rats. World J Gastroenterol 2005;11:3431–3440. doi: 10.3748/wjg.v11.i22.3431. ulation function in patients with liver cirrhosis. The therapeu- [13] Aquino JB, Bolontrade MF,García MG, Podhajcer OL, Mazzolini G. Mesenchymal tic effect was generated at 2–4 weeks after transplantation, stem cells as therapeutic tools and gene carriers in liver fibrosis and hepato- and lasted for 24 weeks. However, the effects lasted no more cellular carcinoma. Gene Ther 2010;17:692–708. doi: 10.1038/gt.2010.10. [14] Lyra AC, Soares MB, da Silva LF, Braga EL, Oliveira SA, Fortes MF, et al. than 48 weeks. There were few symptoms and no serious side Infusion of autologous bone marrow mononuclear cells through hepatic effects or complications among patients after the 48-week artery results in a short-term improvement of liver function in patients follow-up period. It was shown that 4 3 108 autologous BMSC with chronic liver disease: a pilot randomized controlled study. Eur J Gastro- enterol Hepatol 2010;22:33–42. doi: 10.1097/MEG.0b013e32832eb69a. transplant via the hepatic artery was more beneficial to [15] Peng L, Xie DY, Lin BL, Liu J, Zhu HP, Xie C, et al. Autologous bone marrow patients. In the future, larger multicentered, randomized, mesenchymal stem cell transplantation in liver failure patients caused by controlled and double-blinded studies are needed to substan- hepatitis B: short-term and long-term outcomes. Hepatology 2011;54: 820–828. doi: 10.1002/hep.24434. tiate the findings of this meta-analysis and to design new [16] Amer ME, El-Sayed SZ, El-Kheir WA, Gabr H, Gomaa AA, El-Noomani N, et al. protocols to prolong the therapeutic effects. Clinical and laboratory evaluation of patients with end-stage liver cell failure

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 238–248 247 Wu C.X. et al: BMSC therapy for liver cirrhosis

injected with bone marrow-derived hepatocyte-like cells. Eur J Gastroenterol decompensated hepatitis B cirrhosis. Genet Mol Res 2014;13:9336–9342. Hepatol 2011;23:936–941. doi: 10.1097/MEG.0b013e3283488b00. doi: 10.4238/2014.February.21.13. [17] Xu L, Gong Y, Wang B, Shi K, Hou Y, Wang L, et al. Randomized trial of [34] Bai YQ, Yang YX, Yang YG, Ding SZ, Jin FL, Cao MB, et al. Outcomes of autologous bone marrow mesenchymal stem cells transplantation for hep- autologous bone marrow mononuclear cell transplantation in decompen- atitis B virus cirrhosis: regulation of Treg/Th17 cells. J Gastroenterol Hepatol sated liver cirrhosis. World J Gastroenterol 2014;20:8660–8666. doi: 10. 2014;29:1620–1628. doi: 10.1111/jgh.12653. 3748/wjg.v20.i26.8660. [18] Kharaziha P, Hellström PM, Noorinayer B, Farzaneh F, Aghajani K, Jafari F, et al. [35] Salama H, Zekri AR, Medhat E, Al Alim SA, Ahmed OS, Bahnassy AA, et al. Improvement of liver function in liver cirrhosis patients after autologous mes- Peripheral vein infusion of autologous mesenchymal stem cells in Egyptian enchymal stem cell injection: a phase I-II clinical trial. Eur J Gastroenterol HCV-positive patients with end-stage liver disease. Stem Cell Res Ther 2014; Hepatol 2009;21:1199–1205. doi: 10.1097/MEG.0b013e32832a1f6c. 5:70. doi: 10.1186/scrt459. [19] Spahr L, Chalandon Y, Terraz S, Kindler V, Rubbia-Brandt L, Frossard JL, et al. [36] Al Tayeb H, El Dorry A, Amer N, Mowafy N, Zimaity M, Bayoumy E, et al. Autologous bone marrow mononuclear cell transplantation in patients with Autologous stem cells transplantation in egyptian patients with liver cirrhosis decompensated alcoholic liver disease: a randomized controlled trial. PLoS on top of hepatitis C virus. Int J Stem Cells 2015;8:209–218. doi: 10. One 2013;8:e53719. doi: 10.1371/journal.pone.0053719. 15283/ijsc.2015.8.2.209. [20] Verhagen AP, de Vet HC, de Bie RA, Kessels AG, Boers M, Bouter LM, et al. The [37] Mohamadnejad M, Vosough M, Moossavi S, Nikfam S, Mardpour S, Akhlagh- Delphi list: a criteria list for quality assessment of randomized clinical trials for poor S, et al. Intraportal infusion of bone marrow mononuclear or CD133+ conducting systematic reviews developed by Delphi consensus. J Clin Epide- cells in patients with decompensated cirrhosis: A double-blind randomized – miol 1998;51:1235 1241. doi: 10.1016/S0895-4356(98)00131-0. controlled trial. Stem Cells Transl Med 2016;5:87–94. doi: 10.5966/sctm. [21] Berger VW. A review of methods for ensuring the comparability of compar- 2015-0004. – ison groups in randomized clinical trials. Rev Recent Clin Trials 2006;1:81 [38] Suk KT, Yoon JH, Kim MY, Kim CW, Kim JK, Park H, et al. Transplantation with 86. doi: 10.2174/157488706775246139. autologous bone marrow-derived mesenchymal stem cells for alcoholic cir- [22] Berger VW, Weinstein S. Ensuring the comparability of comparison groups: is rhosis: Phase 2 trial. Hepatology 2016;64:2185–2197. doi: 10.1002/hep. randomization enough? Control Clin Trials 2004;25:515–524. doi: 10. 28693. 1016/j.cct.2004.04.001. [39] Wu YZ, Yang L, Zhai YF, Zhang HH, Huo LY. Therapeutic effect of autologous [23] Berger VW, Ivanova A, Knoll MD. Minimizing predictability while retaining bone marrow mesenchymal stem cells on hepatic fibrosis, liver function, balance through the use of less restrictive randomization procedures. Stat MELD score and 1-year survival rate in patients with decompensated hepa- Med 2003;22:3017–3028. doi: 10.1002/sim.1538. titis B. Chinese Journal of Tissue Engineering Research 2017;21:2049– [24] Berger VW, Bears JD. When can a clinical trial be called ‘randomized’? Vaccine 2055. doi: 10.3969/j.issn.2095-4344.2017.13.014. 2003;21:468–472. doi: 10.1016/s0264-410x(02)00475-9. [40] Han Y, Yan L, Han G, Zhou X, Hong L, Yin Z, et al. Controlled trials in hepatitis B [25] Bagos PG. A unification of multivariate methods for meta-analysis of genetic virus-related decompensate liver cirrhosis: peripheral blood monocyte trans- association studies. Stat Appl Genet Mol Biol 2008;7:Article31. doi: 10. plant versus granulocyte-colony-stimulating factor mobilization therapy. 2202/1544-6115.1408. Cytotherapy 2008;10:390–396. doi: 10.1080/14653240802129901. [26] Bagos PG. Genetic model selection in genome-wide association studies: [41] Kim JK, Park YN, Kim JS, Park MS, Paik YH, Seok JY, et al. Autologous bone robust methods and the use of meta-analysis. Stat Appl Genet Mol Biol marrow infusion activates the progenitor cell compartment in patients with 2013;12:285–308. doi: 10.1515/sagmb-2012-0016. advanced liver cirrhosis. Cell Transplant 2010;19:1237–1246. doi: 10. [27] Kavvoura FK, Ioannidis JP. Methods for meta-analysis in genetic association studies: a review of their potential and pitfalls. Hum Genet 2008;123:1–14. 3727/096368910X506863. [42] Takami T, Terai S, Sakaida I. Novel findings for the development of drug doi: 10.1007/s00439-007-0445-9. [28] Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis therapy for various liver diseases: Current state and future prospects for detected by a simple, graphical test. BMJ 1997;315:629–634. doi: 10. our liver regeneration therapy using autologous bone marrow cells for – 1136/bmj.315.7109.629. decompensated liver cirrhosis patients. J Pharmacol Sci 2011;115:274 [29] Salama H, Zekri AR, Bahnassy AA, Medhat E, Halim HA, Ahmed OS, et al. 278. doi: 10.1254/jphs.10R13FM. Autologous CD34+ and CD133+ stem cells transplantation in patients with [43] Pankaj P, Zhang Q, Bai XL, Liang TB. Autologous bone marrow transplanta- end stage liver disease. World J Gastroenterol 2010;16:5297–5305. doi: 10. tion in decompensated liver: Systematic review and meta-analysis. World J – 3748/wjg.v16.i42.5297. Gastroenterol 2015;21:8697 8710. doi: 10.3748/wjg.v21.i28.8697. [30] El-Ansary M, Abdel-Aziz I, Mogawer S, Abdel-Hamid S, Hammam O, Teaema S, [44] Ma XR, Tang YL, Xuan M, Chang Z, Wang XY, Liang XH. Transplantation of et al. Phase II trial: undifferentiated versus differentiated autologous mesen- autologous mesenchymal stem cells for end-stage liver cirrhosis: a meta- chymal stem cells transplantation in Egyptian patients with HCV induced liver analysis based on seven controlled trials. Gastroenterol Res Pract 2015; cirrhosis. Stem Cell Rev 2012;8:972–981. doi: 10.1007/s12015-011-9322-y. 2015:908275. doi: 10.1155/2015/908275. [31] Liao X, AnCheng JY, Zhou QJ, Liao C. Therapeutic effect of autologous bone [45] Walczak P, Zhang J, Gilad AA, Kedziorek DA, Ruiz-Cabello J, Young RG, et al. marrow-derived liver stem cells transplantation in hepatitis B virus-induced Dual-modality monitoring of targeted intraarterial delivery of mesenchymal liver cirrhosis. Hepatogastroenterology 2013;60:406–409. doi: 10. stem cells after transient ischemia. Stroke 2008;39:1569–1574. doi: 10. 5754/hge12821. 1161/STROKEAHA.107.502047. [32] Mohamadnejad M, Alimoghaddam K, Bagheri M, Ashrafi M, Abdollahzadeh L, [46] Zhao L, Chen S, Shi X, Cao H, Li L. A pooled analysis of mesenchymal stem Akhlaghpoor S, et al. Randomized placebo-controlled trial of mesenchymal cell-based therapy for liver disease. Stem Cell Res Ther 2018;9:72. doi: 10. stem cell transplantation in decompensated cirrhosis. Liver Int 2013;33: 1186/s13287-018-0816-2. 1490–1496. doi: 10.1111/liv.12228. [47] Kwak KA, Cho HJ, Yang JY, Park YS. Current perspectives regarding stem cell- [33] Liu L, Yan Y, Zhou J, Huang LW, He CP, Ling K, et al. Curative effect of based therapy for liver cirrhosis. Can J Gastroenterol Hepatol 2018;2018: combined lamivudine, adefovir dipivoxil, and stem cell transplantation on 4197857. doi: 10.1155/2018/4197857.

248 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 238–248 Original Article

A Potential Functional Cure in Chinese HBeAg-negative Chronic Hepatitis B Patients Treated with Peg-interferon Alpha-2a

Xinyue Chen#1, Qianguo Mao#2, Yao Xie#3, Xiaoguang Dou4, Qing Xie5, Jifang Sheng6, Zhiliang Gao7, Xiaoling Zhou8, Yingxia Liu9, Huanwei Zheng10, Shuqin Zhang11, Shibo Li12, Fusheng Zhu13, Yuqin Xu14, Mingxiang Zhang15, Yaoren Hu16, Xiaoping Chen17, Yan Huang18, Hong Ren*19 and Jidong Jia*20

1International Medicine Ward, Beijing YouAn Hospital, Capital Medical University, Beijing, China; 2Department of Infectious Diseases, Xiamen Hospital of Traditional Chinese Medicine, Xiamen, Fujian, China; 3Department of Liver Disease, Beijing Ditan Hospital, Capital Medical University, Beijing, China; 4Department of Infectious Diseases, Shengjing Hospital of China Medical University, Shenyang, Liaoning, China; 5Department of Infectious Diseases, Rui Jin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China; 6Department of Infectious Diseases, The First Affiliated Hospital of Zhejiang University, Hangzhou, Zhejiang, China; 7Department of Infectious Diseases, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China; 8Department of Gastroenterology, Liuzhou Traditional Chinese Medical Hospital, Liuzhou, Guangxi, China; 9Department of Infectious Diseases, Shenzhen Third People’s Hospital, Shenzhen, Guangdong, China; 10Department of Infectious Diseases, The Fifth Hospital of Shijiazhuang, Shijiazhuang, Hebei, China; 11Department of Infectious Diseases, Hepatology Hospital of Jilin Province, Changchun, Jilin, China; 12Department of Infectious Diseases, Zhoushan Hospital of Zhejiang Province, Zhoushan, Zhejiang, China; 13Department of Infectious Diseases, General Hospital of Dagang Oilfield, Tianjin, China; 14Department of Infectious Diseases, The 211 Hospital of People’s Liberation Army, Harbin, Heilongjiang, China; 15Department of Integrated Traditional and Western Medicine on Liver Diseases, The Sixth People’s Hospital of Shenyang, Shenyang, Liaoning, China; 16Department of Liver Disease, Ningbo No. 2 Hospital of Zhejiang Province, Ningbo, Zhejiang, China; 17Department of Infectious Diseases, Guangdong General Hospital, Guangzhou, Guangdong, China; 18Department of Medical Science, Shanghai Roche Pharmaceuticals Ltd., Shanghai, China; 19Department of Infectious Diseases, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China; 20Department of Liver Disease, Beijing Friendship Hospital, Capital Medical University, Beijing, China

Abstract and treatment duration according to Chinese clinical practice. Effectiveness of peg-IFNa treatment was measured Background and Aims: Data are limited on the use of pe- by the percentage of: patients with HBV DNA <2000 IU/mL gylated-interferon alpha-2a (peg-IFNa) in Chinese patients and loss of hepatitis B surface antigen (commonly known with chronic hepatitis B virus (HBV) infection (CHB). We eval- as HBsAg); HBV DNA level at end of treatment (EOT), and uated the effectiveness and safety of peg-IFNa in Chinese 6 months and 1 year posttreatment; and time course change patients with hepatitis B envelope antigen-negative CHB in in quantitative HBV DNA and HBsAg. Results: At EOT, 6 routine clinical practice. Methods: In this prospective, multi- months posttreatment, and 1 year posttreatment, the per- center, observational, non-interventional cohort study, centage of patients with HBV DNA <2000 IU/mL was patients were assessed for up to 1 year after peg-IFNa treat- 90.0%, 81.8%, and 82.2%, and that of patients with HBsAg ment cessation. Treating physicians established the dosing loss was 6.5%, 9.4%, and 9.5%, respectively. The HBV DNA level decreased from 5.61 log IU/mL at baseline to 2.48 log Keywords: Chronic hepatitis B; Prospective studies; Observational study; IU/mL at EOT and 2.67 log IU/mL at 1 year posttreatment. Interferon alpha. The HBsAg level decreased from 3.08 log IU/mL at baseline to Abbreviations: AEs, adverse events; ALT, alanine aminotransferase; CHB, 2.24 log IU/mL at EOT and 2.10 log IU/mL at 1 year chronic hepatitis B; CI, confidence interval; EOT, end of treatment; FAS-MSC, posttreatment. The incidence of adverse events was 52.0%. full analysis set of those who meet selection criteria; HBeAg, hepatitis B envelope antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; NUC, nucleo- Conclusions: Peg-IFNa has the potential to provide function- side analogue; peg-IFNa, pegylated-interferon alpha-2a; SD, standard deviation; al cure (HBsAg loss) for CHB and is well tolerated in hepatitis ULN, upper limit of normal. B envelope antigen-negative CHB patients in routine clinical Received: 4 April 2019; Revised: 27 June 2019; Accepted: 19 July 2019 practice in China. Clinical Trial Registration: ClinicalTrials. #These authors contributed equally to this study. *Correspondence to: Hong Ren, Department of Infectious Diseases, The Second gov (NCT01730508). Affiliated Hospital of Chongqing Medical University, No. 76, Linjiang Road, Yuzhong Citation of this article: Chen X, Mao Q, Xie Y, Dou X, Xie Q, District, Chongqing 400010, China. Tel: +86-13983888786, Fax: +86-23- Sheng J, et al. A potential functional cure in Chinese HBeAg- 63711527, E-mail: [email protected]; Jidong Jia, Department of Liver ’ negative chronic hepatitis B patients treated with peg-inter- Disease, Beijing Friendship Hospital, Capital Medical University, No. 95 Yong an – Road, Xicheng District, Beijing 100069, China. Tel: +86-13501378269, Fax: +86- feron alpha-2a. J Clin Transl Hepatol 2019;7(3):249 257. 10-62704289, E-mail: [email protected] doi: 10.14218/JCTH.2019.00016.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 249–257 249

Introduction and safety of peg-IFNa in Chinese patients with HBeAg-negative CHB in routine clinical practice. According to the World Health Organization Global Hepatitis Report in 2017, 257 million people worldwide were living with Methods chronic hepatitis B virus (HBV) infection (CHB),1 among whom one third (86 million) were located in China.2 In Study design and patients China, surveys conducted in 2006 and 2014 showed a decreasing trend in hepatitis B surface antigen (HBsAg) prev- This was a prospective, observational, noninterventional alence after the start of a vaccination program in 1992.3 cohort study. Dosing and treatment duration were deter- However, in 2016, the prevalence of HBsAg was reported to mined at the discretion of the investigator and reflect actual be relatively high, namely 6.0% in men aged 21–49 years in Chinese clinical practice. Patients were followed up for 1 year rural China4 and 6.1% in Northeastern China.5 Among after treatment cessation. Data on treatment outcomes (i.e. HBsAg-positive individuals, the lifetime relative risk for hep- HBV DNA, HBsAg, quantitative HBsAg, hepatitis B surface atocellular carcinoma was reported to be 15- to 20-fold higher antibody, and alanine aminotransferase [ALT]) were collected compared with that of HBsAg-negative individuals,6 and the from medical records and documented in electronic case risk of hepatocellular carcinoma significantly decreased in report forms. CHB patients with HBsAg clearance.7 HBeAg-negative CHB patients from 79 study sites in China Hepatitis B envelope antigen (HBeAg)-negative CHB rep- (Supplemental Table 1) who received peg-IFNa therapy from resents a late phase in the natural history of CHB that November 2012 to April 2015 were consecutively enrolled. develops immediately after HBeAg seroconversion or after a The key eligibility criteria included serum ALT >upper limit 3 long inactive chronic hepatitis B virus carrier phase.8,9 of normal (ULN) but #10 ULN, and HBV DNA $2000 IU/ HBeAg-negative CHB patients often require treatment mL according to Chinese peg-IFNa-2a labeling and HBV clin- 20 because spontaneous remission rarely occurs, and these ical practice guidelines. Those who did not meet the eligi- patients have more advanced liver disease compared with bility criteria were excluded from the effectiveness analysis. HBeAg-positive patients.8–10 All procedures followed were in accordance with the The therapeutic goal of CHB is to achieve a “functional” ethical standards of the responsible committee on human cure, which is characterized by sustained HBsAg loss, but is experimentation (institutional and national) and with the almost impossible to achieve with nucleoside analogues Helsinki Declaration of 1975, as revised in 2008. Informed consent in writing was obtained from all patients included in (NUCs) (including entecavir, tenofovir disoproxil fumarate, the study. This study was registered at ClinicalTrials.gov and tenofovir alafenamide).11 The efficacy of pegylated-inter- (NCT01730508). feron alpha (peg-IFNa) in terms of HBsAg clearance and/or improvement in sustained off-treatment virologic response End-points was shown in several interventional multi-national studies in Caucasian patients with HBeAg-negative CHB.9,12–15 In a The following effectiveness end-points were evaluated as phase 3 study investigating 177 HBeAg-negative CHB percentage of patients with: HBV DNA <2000 IU/mL, <400 patients who received peg-IFNa in combination with lamivu- IU/mL, and <200 IU/ml; HBsAg <10 IU/mL, <100 IU/mL, dine for 48 weeks, HBsAg clearance was achieved by 5% of and <1000 IU/mL; HBsAg loss; and HBsAg seroconversion at patients at 1 year posttreatment.9,16 This rate increased to end of treatment (EOT), 6 months posttreatment, and 1 year 12% at 5 years posttreatment.16 Among patients with HBV posttreatment. The time course change in quantitative HBV DNA <2000 IU/mL at 1 year posttreatment, 28% achieved 16 DNA and HBsAg during the observation period and baseline HBsAg loss at 5 years posttreatment. The PegBeLiver and on-treatment predictors of response were also evaluated. study demonstrated that extended treatment (96 weeks) An analysis of predefined subgroups was performed accord- with peg-IFNa was well tolerated, and significantly improved ing to treatment pattern (peg-IFNa monotherapy, NUC add- sustained response rates measured by HBsAg loss in HBeAg- on during peg-IFNa treatment, and NUC add-on during negative patients predominantly infected with HBV genotype follow-up), baseline ALT level (#2, >2 and #5, and >5 12 D (6%, 1 year posttreatment). In a study on a small sample ULN), age (<35 and $35 years), treatment duration (48, of Chinese HBeAg-negative patients, a significantly greater 72, and 96 weeks), and early treatment response (HBV DNA HBsAg clearance rate at 48 weeks posttreatment was decline >2 log plus ALT increase at week 12 and HBV DNA reported among those who received extended treatment decline #2 log or no ALT increase at week 12) in patients who with peg-IFNa (72 weeks) compared with standard treatment received Peg-IFNa monotherapy. For safety, reported adverse 17 (48 weeks) (35.7% vs. 10.5%, respectively; p < 0.05). events (AEs) and laboratory data were evaluated. The dominant HBV genotypes are B and C in Asia and A and D in Europe. As responses to interferon treatment have Statistical analysis been found to vary depending on the HBV genotype,18 and these genotypes follow a geographical distribution, it is The full analysis set (FAS) was defined as the subjects who important to evaluate Asian patients separately. Although underwent at least one dose of peg-IFNa treatment and was HBeAg-negative CHB is less common in China than in used for the safety analysis. The FAS of those who meet Europe, the incidence of HBeAg-negative disease has selection criteria (FAS-MSC) was defined as FAS subjects who increased in China.19 Considering the limited data on peg- met all the selection criteria of this study, and was used for IFNa in Asian/Chinese patient populations, the difference in the effectiveness analysis. Descriptive statistics were used for response to treatment based on the dominant HBV genotype baseline demographic and clinical characteristics, with n (%) and the increasing incidence of HBeAg-negative disease in for categorical variables and mean ± standard deviation (SD) China, the present study aimed to evaluate the effectiveness for continuous variables. Statistical tests and 95% confidence

250 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 249–257 Chen X. et al: Peg-IFNa 2a in HBeAg-negative CHB intervals (CIs) were two-sided. The significance level was set mean age was 37.9 years and most patients were male at p # 0.05. The response rate of HBV DNA suppression, ALT (81.0%). The mean HBV DNA and HBsAg levels were 5.6 normalization, HBsAg loss, and seroconversion was calcu- log IU/mL and 3.1 log IU/mL, respectively. lated, and the exact 95% (two-sided) CIs from the binomial Among patients with known HBV genotype, genotypes B distribution were provided. The effectiveness analyses were and C were the dominant genotypes. performed in patients with measurements at the corresponding time point. The statistical software used for the statistical Effectiveness analysis was SAS® (software package version 9.2; SAS Inc., Cary, NC, USA). Fig. 2A shows the percentage of patients with HBV DNA <2000 IU/mL, <400 IU/mL, and <200 IU/mL, that of patients Results with a combined response (HBV DNA <2000 IU/mL and ALT normalization), and that of patients with HBsAg loss and Patients HBsAg seroconversion at EOT, 6 months posttreatment, and 1 year posttreatment. At 1 year posttreatment, the percent- The patient population is shown in Fig. 1. In total, 930 age of patients with HBV DNA <2000 IU/mL was 82.2% (95% patients from 79 sites were enrolled and treated with at CI 77.4, 86.3) in FAS-MSC subjects with available results at 1 least one dose of peg-IFNa, and were included in the FAS year posttreatment. The percentage of patients with suppres- (safety analysis). However, 268 patients did not meet the sion of HBV DNA to <2000 IU/mL was 90.0% at EOT and eligibility criteria, so only 662 patients were included in the 81.8% at 6 months posttreatment. The percentage of FAS-MSC (effectiveness analysis). Of the 662 patients in the patients with suppression of HBV DNA to <400 IU/mL and FAS-MSC, 33.7% did not complete the 1 year follow-up <200 IU/mL at EOT, 6 months posttreatment, and 1 year (Fig. 1). A total of 476 (71.9%) patients completed 48 posttreatment ranged between 35.6% and 45.5%. The per- weeks of treatment, 167 (25.2%) completed 72 weeks of centage of patients with a combined response (HBV DNA treatment, and 71 (10.7%) completed 96 weeks of treat- <2000 IU/mL and ALT normalization) increased from 51.0% ment. Among patients in the FAS-MSC, the most common at EOT to 71.6% and 73.4%, respectively, at 6 months and 1 treatment pattern was peg-IFNa monotherapy (80.4%), fol- year posttreatment. The percentage of patients with HBsAg lowed by NUC add-on during peg-IFNa treatment (14.7%), loss was 6.5% at EOT, 9.4% at 6 months posttreatment, and and NUC add-on during follow-up (5.0%) (Fig. 1). The 9.5% at 1-year posttreatment; the percentage of patients number of patients with each measurement at each time with HBsAg seroconversion was 5.2% at EOT, 7.6% at 6 point is shown in Supplemental Table 2. months posttreatment, and 7.1% at 1 year posttreatment. The baseline demographic and clinical characteristics in all The change of HBsAg category throughout the observation patients who met eligibility criteria are shown in Table 1. The period is shown in Fig. 2B. The percentage of patients with

Fig. 1. Patient disposition. The predefined date to stop data collection was at 1 year after the last enrolled patient completed the treatment. At the final cut-off (date to stop data collection), 13 patients who received prolonged treatment were still within the 1 year follow-up window. Abbreviations: AE, adverse event; FAS, full analysis set; FAS-MSC, full analysis set of those who meet selection criteria; NUC, nucleoside analogue; peg-IFNa, pegylated-interferon alpha-2a; tx, treatment.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 249–257 251 5 ora fCiia n rnltoa Hepatology Translational and Clinical of Journal 252 Table 1. Baseline demographic and clinical characteristics in all patients and by subgroups

Treatment pattern ALT levelÃ AgeÃ Treatment durationÃ Treatment responseÃ

NUC add- on >2 HBV DNA HBV DNA NUC add- during and decline >2 decline #2 on during follow- #2 #5 >5 <35 $35 48 72 96 log plus ALT log or no ALT Total Peg-IFNa peg-IFNa up ULN ULN ULN years years weeks weeks weeks increase at increase at (n = monotherapy treatment (n = (n = (n = (n = (n = (n = (n = (n = (n = week 12 week 12 662) (n = 532) (n = 97) 33) 184) 256) 92) 209) 323) 249) 111) 37) (n = 48) (n = 338)

Age in 37.9 37.6 6 9.3 39.4 6 37.6 6 38.3 37.0 37.7 28.3 43.6 37.6 38.6 35.6 38.5 6 7.0 37.4 6 9.4 years 6 9.4 9.6 9.9 6 9.2 6 9.4 6 9.2 6 4.1 6 6.2 6 8.9 6 9.7 6 7.6 Male 536 426 (80.1) 83 (85.6) 27 154 205 67 157 269 187 94 32 37 (77.1) 273 (80.8) (81.0) (81.8) (83.7) (80.1) (72.8) (75.1) (83.3) (75.1) (84.7) (86.5) HBV DNA 5.6 6 5.6 6 1.2 5.7 6 1.2 6.0 6 5.2 6 5.7 6 5.9 6 5.7 6 5.5 6 5.6 6 5.5 6 5.6 6 5.5 6 1.0 5.7 6 1.2 as log IU/ 1.2 1.3 1.1 1.2 1.2 1.3 1.1 1.2 1.2 1.1 m ALT, ULN 3.2 6 3.2 6 2.0 2.7 6 1.7 3.7 6 1.5 6 3.1 6 6.9 6 3.2 6 3.2 6 3.3 6 3.2 6 3.5 6 1.7 6 0.7 3.5 6 2.1 2.0 2.3 0.3 0.8 1.3 1.8 2.1 2.0 2.2 2.3 HBsAg as 3.1 6 3.0 6 0.8 3.3 6 0.6 3.2 6 2.9 6 3.1 6 3.0 6 3.1 6 3.0 6 3.1 6 3.0 6 3.0 6 2.9 6 0.9 3.1 6 0.8 log IU/mL 0.8 0.8 0.9 0.8 0.8 0.9 0.7 0.9 0.8 0.7 Known 117.5 116.9 6 125.4 6 104.7 121.2 107.9 133.1 92.8 132.5 120.3 117.8 119.4 138.2 6 121.6 6 hepatitis B 6 100.9 90.8 6 93.8 6 6 6 6 6 6 6 6 86.3 106.4 disease 99.1 99.1 98.2 110.3 82.0 108.8 106.6 90.4 108.2 course in months HBV genotype A/E/F/G 0 0 0 0 0 0 0 0 0 0 0 0 0 0 hnX. Chen

2019 B3225 (4.7) 6 (6.2) 1 (3.0) 4 17 4 9 16 4 12 2 1 (2.1) 18 (5.3) (4.8) (2.2) (6.6) (4.3) (4.3) (5.0) (1.6) (10.8) (5.4) C4634 (6.4) 11 (11.3) 1 (3.0) 9 20 5 15 19 10 10 5 3 (6.3) 19 (5.6) o.7|249 | 7 vol. (6.9) (4.9) (7.8) (5.4) (7.2) (5.9) (4.0) (9.0) (13.5) al et

D11 (0.2) 0 0 0 0 1 1 00000 0 Peg-IFN : (0.2) (1.1) (0.5) H11 (0.2) 0 0 0 0 1 010100 0 – (0.2) (1.1) (0.3) (0.9) 257

Unknown 577 467 (87.8) 79 (81.4) 31 167 219 81 183 284 234 87 30 43 (89.6) 300 (88.8) a ai BA-eaieCHB HBeAg-negative in 2a (87.2) (93.9) (90.8) (85.5) (88.0) (87.6) (87.9) (94.0) (78.4) (81.1) Others 5 4 (0.8) 1 (1.0) 0 4 0013 1 1 0 1 (2.1) 1 (0.3) (0.8) (2.2) (0.5) (0.9) (0.4) (0.9) With anti- 147 101 (19.0) 43 (44.3) 3 (9.1) 34 49 18 38 63 49 23 6 7 (14.6) 67 (19.8) HBV (22.2) (18.5) (19.1) (19.6) (18.2) (19.5) (19.7) (20.7) (16.2) history

Data are shown as n (%) or mean ± SD. *In patients who received peg-IFNa monotherapy. Abbreviations: ALT, alanine aminotransferase; FAS-MSC, analysis set of those who meet selection criteria; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; NUC, nucleoside analog; peg-IFNa, pegylated- interferon alpha-2a; ULN, upper limit of normal. Chen X. et al: Peg-IFNa 2a in HBeAg-negative CHB

Fig. 2. Percentage of patients with HBV DNA <2000 IU/mL, <400 IU/mL, and <200 IU/mL, that of patients with a combined response (HBV DNA <2000 IU/mL and ALT normalization), and that of patients with HBsAg loss and HBsAg seroconversion at EOT, 6 months posttreatment, and 1 year posttreatment (A), as well as changes in HBsAg category (B), time course change of mean HBV DNA (C), and time course change of mean HBsAg (D). The error bars in 2A indicate 95% confidence intervals and those in C and D indicate standard errors. The mean values in C and D were calculated using the data of the FAS-MSC. Abbreviations: ALT, alanine aminotransferase; EOT, end of treatment; FAS-MSC, full analysis set of those who met the selection criteria; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; ULN, upper limit of normal.

HBsAg level <1000 IU/mL increased from 39.0% at baseline Subgroup analysis to 71.4% at 1 year posttreatment. That of patients with HBsAg level <100 IU/mL increased from 6.1% at baseline The main baseline demographic and clinical characteristics to 29.1% at 1 year posttreatment, and that of patients with by subgroup according to treatment pattern (peg-IFNa HBsAg level <10 IU/mL increased from 1.0% at baseline to monotherapy, NUC add-on during peg-IFNa treatment, and 15.2% at 1 year posttreatment. NUC add-on during follow-up), ALT level (#2, >2 and #5, The time course change of HBV DNA and HBsAg level and >5 ULN), age (<35 and $35 years), treatment duration throughout the study period is shown in Fig. 2C and 2D, (48, 72, and 96 weeks), and treatment response (HBV DNA respectively. The HBV DNA level decreased from 5.61 log decline >2 log plus ALT increase at week 12 and HBV DNA IU/mL at baseline to 2.48 log IU/mL at EOT and 2.67 log IU/ decline #2 log or no ALT increase at week 12) are shown in mL at 1 year posttreatment. The HBsAg level decreased from Table 1. HBV DNA and HBsAg levels were generally similar 3.08 log IU/mL at baseline to 2.24 log IU/mL at EOT and 2.10 between subgroups. log IU/mL at 1 year posttreatment. The results of the effectiveness end-points by treatment pattern are shown in Table 2. The percentage of patients with Predictors of response HBV DNA <2000 IU/mL at 1 year posttreatment was higher in patients who received NUC add-on (>90%) compared with We assessed baseline predictors of HBV DNA <2000 IU/mL at that in patients who received peg-IFNa monotherapy the end of a 1-year follow-up period, including sex, age, body (79.4%). The percentage of patients with HBsAg loss at mass index, method of HBV transmission, hepatitis B disease EOT, 6 months posttreatment, and 1 year posttreatment course, anti-HBV history, HBV genotype, HBV DNA level, who received peg-IFNa monotherapy (6.9%, 10.6%, and HBsAg level, and ALT level. However, univariate and multi- 10.6%, respectively) and NUC add-on during peg-IFNa treat- variate logistic regression analyses did not reveal any statisti- ment (6.0%, 6.3%, and 7.1%, respectively) was greater as cally significant relationships (data not shown). compared with patients who received NUC add-on during Similarly, early HBV DNA and HBsAg response were not follow-up (0, 0, and 0, respectively). In Supplemental Table 3, found to be significant predictors of HBV DNA suppression at 1 we summarize the data at EOT, 6 months posttreatment, and year posttreatment. The receiver operating characteristic 1 year posttreatment by subgroup in patients receiving peg- curves for HBV DNA and HBsAg change from baseline (log) IFNa monotherapy. at week 12 showed an area under the curve of 0.521 (p = The changes of HBsAg were evaluated by subgroups 0.626) and 0.505 (p = 0.929), respectively (data not shown). according to ALT level, age, treatment duration, and treatment

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 249–257 253 Chen X. et al: Peg-IFNa 2a in HBeAg-negative CHB

Table 2. Effectiveness end-points by treatment pattern

6 months 1 year EOT posttreatment posttreatment

HBV DNA <2000 IU/ Peg-IFNa monotherapy 91.8 (88.5, 94.5) 80.6 (75.6, 85.0) 79.4 (73.6, 84.4) mL NUC add-on during peg-IFNa 88.9 (78.4, 95.4) 90.0 (78.2, 96.7) 90.2 (78.6, 96.7) treatment NUC add-on during follow-up 63.6 (40.7, 82.8) 78.3 (56.3, 92.5) 94.7 (74.0, 99.9) HBV DNA <2000 IU/ Peg-IFNa monotherapy 53.1 (47.6, 58.5) 70.2 (64.4, 75.6) 70.6 (64.1, 76.5) mL NUC add-on during peg-IFNa 41.7 (29.1, 55.1) 77.1 (62.7, 88.0) 83.7 (70.3, 92.7) 3 and ALT <1 ULN treatment NUC add-on during follow-up 42.9 (21.8, 66.0) 76.2 (52.8, 91.8) 81.3 (54.4, 96.0) HBsAg loss Peg-IFNa monotherapy 6.9 (4.5, 10.1) 10.6 (7.2, 14.8) 10.6 (6.9, 15.5) NUC add-on during peg-IFNa 6.0 (1.7, 14.6) 6.3 (1.3, 17.2) 7.1 (1.5, 19.5) treatment NUC add-on during follow-up 0 0 0 HBsAg Peg-IFNa monotherapy 5.1 (3.0, 8.2) 8.2 (5.2, 12.3) 7.0 (3.9, 11.5) seroconversion NUC add-on during peg-IFNa 7.3 (2.0, 17.6) 5.9 (0.7, 19.7) 9.4 (2.0, 25.0) treatment NUC add-on during follow-up 0 0 0

Data are shown as percentage (95% confidence interval). Abbreviations: ALT, alanine aminotransferase; EOT, end of treatment; FAS-MSC, analysis set of those who meet selection criteria; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; NUC, nucleoside analogue; peg-IFNa, pegylated-interferon alpha-2a; ULN, upper limit of normal. response (Fig. 3). In patients with ALT >5 ULN, HBsAg levels AEs with an incidence $10% were decreased white blood decreased at a slower rate during treatment compared with the (24.7%), platelet (23.4%), and neutrophil counts (21.9%). subgroups with ALT #2 and those with ALT >2 and #5ULN,but a greater decrease in HBsAg level was shown at 1 year post- Discussion treatment compared with the other two subgroups. Patients aged <35 years showed a better response in HBsAg decrease This is the largest observational study of peg-IFNa therapy in over time compared with patients aged >35 years. A marked Asian HBeAg-negative CHB patients whose predominant HBV decrease in HBsAg level was observed in the subgroup receiv- genotypes are B and C. A cut-off of 1 year posttreatment was ing treatment for 96 weeks at the EOT compared with those chosen in the present study because missing HBV laboratory receiving treatment for 48 and 72 weeks, and this tendency testing data (e.g., HBV DNA and HBsAg, especially quantita- remained at 1 year posttreatment. A greater decrease in tive testing data) are unavoidable in observational studies, HBsAg was observed in the subgroup with early response and the amount of missing data tends to increase with a (HBV DNA decline >2 log at week 12 and ALT increase) at longer follow-up period. Moreover, the association between 1- EOT and 1 year posttreatment compared with the subgroup year posttreatment response and sustained off-treatment 16 with HBV DNA decline #2 log at week 12 or no ALT increase response were reported in a phase 3 study of peg-IFNa. at week 12. In the present study, the percentage of patients with HBsAg The percentages of patients with HBV DNA <2000 IU/mL loss at EOT, 6 months posttreatment, and 1 year posttreatment and HBsAg loss at 1 year posttreatment were not significantly increased from 6.5% to 9.4% and 9.5%. In an observational different between subgroups according to baseline ALT level, study of Korean HBeAg-negative CHB patients who received 21 age, and treatment response at week 12 (data not shown). peg-IFNa therapy for 24 and 48 weeks (the TRACES study), The percentage of patients with HBV DNA <2000 IU/mL at 1 only one patient (1.4%) in the 48-week group presented HBsAg loss at EOT and 6 months posttreatment. In the PegBe- year posttreatment was 75.6% (95% CI 67.2, 82.8), 84.6% 22 (95% CI 71.9, 93.1), and 89.5% (95% CI 66.9, 98.7), and Liver study, Caucasian HBeAg-negative patients who that of patients with HBsAg loss was 8.5% (4.2, 15.2), 10.0% received 96 weeks of treatment achieved an HBsAg loss rate (3.3, 21.8), and 27.8% (9.7, 53.5) among the subgroups with of 5.8% compared with 0% in those who received 48 weeks of treatment. Besides the demographic, clinical, and genotype dif- treatment duration of 48 weeks, 72 weeks, and 96 weeks, ferences between the study populations, the higher HBsAg loss respectively. rate at 1 year posttreatment in the present study compared with the TRACES and PegBeLiver studies could have been Safety because our study included patients who received treatment beyond 48 weeks. Our findings were similar to those of The incidence of AEs was 52.0% and that of drug-related AEs another subgroup analysis of Chinese HBeAg-negative patients was 47.3% (Table 3). Twelve patients (1.3%) had serious (the S-COLLATE study),23 in which the HBsAg loss rate AEs. One patient (0.1%) died, but this event was not increased from 9% to 12% and 13% at EOT, 6 months post- related to the study drug. treatment, and 3 years posttreatment, respectively. In another

254 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 249–257 Chen X. et al: Peg-IFNa 2a in HBeAg-negative CHB

Fig. 3. Changes in HBsAg by subgroup according to ALT (FAS-MSC) (A), age (B), treatment duration (C), and treatment response (D). The error bars indicate standard errors. Abbreviations: ALT, alanine aminotransferase; FAS-MSC, full analysis set of those who met the selection criteria; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; ULN, upper limit of normal. study of Asian HBeAg-negative patients,17 HBsAg loss at 48 response (i.e. HBsAg decline >1 log at week 24, but not weeks posttreatment was 10.5% in patients who received 48 achieving HBsAg loss at week 48), extended treatment duration weeks of treatment (which was similar to the HBsAg loss rate at to 72 or 96 weeks helps patients achieve functional cure.24 The 1 year posttreatment in the present study, 10%) and 33.3% in present study showed a marked HBsAg level decrease from 24 those who received 72 weeks of treatment. weeks to 96 weeks in patients who received 96 weeks of treat- The Chinese consensus on peg-IFNa in treatment of CHB ment compared with those who received 48 weeks of treat- published recently emphasizes that for patients with suboptimal ment, and provides further evidence for the extended treatment duration preferred by medical practitioners in China. In the present study, the percentage of patients with HBV Table 3. Adverse events DNA <2000 IU/mL was 90.0% at EOT, 81.8% at 6 months Adverse event Total, n = 930 posttreatment, and 82.2% at 1 year posttreatment. In the TRACES study,21 the percentage of patients with HBV DNA Any 484 (52.0%) <2000 IU/mL was 87.8% at EOT and 47.3% at 6 months Any drug-related 440 (47.3%) posttreatment. In a study of Asian HBeAg-negative patients,17 the HBV DNA suppression rate at 48 weeks post- Any SAE 12 (1.3%) treatment was 60.5% in patients who received 48 weeks of Any drug-related SAE 4 (0.4%) treatment and 83.3% in those who received 72 weeks of Any leading to discontinuation from 81 (8.7%) treatment. The longer treatment duration in the present treatment study may have contributed to the higher HBV DNA suppression rate at 6 months and 1 year posttreatment in our study Any leading to dose adjustment 114 (12.3%) versus previous studies. However, undocumented NUC add- Any leading to death 1 (0.1%) on treatment is suspected (e.g., patients may have self-pre- Any drug-related leading to death 0 scribed NUC add-on or it may have been prescribed by other health care providers when patients sought medical attention White blood cell count decreased 230 (24.7%) in other hospitals). Unlike HBsAg, NUC add-on treatment has Platelet count decreased 218 (23.4%) a significant effect on HBV DNA suppression, and thus, NUC Neutrophil count decreased 204 (21.9%) add-on treatment may have confounded the results. Regarding predictors of response, in HBeAg-negative CHB Data are shown as n (%). patients with genotype D, a combination of no decrease in Abbreviation: SAE, serious adverse event. HBsAg levels and <2 log IU/mL reduction in serum HBV DNA

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 249–257 255 Chen X. et al: Peg-IFNa 2a in HBeAg-negative CHB levels at 12 weeks of Peg-IFNa therapy is associated with no Ltd. The study was funded by Shanghai Roche Pharmaceuti- response to treatment, and these characteristics should be cals Ltd. considered as criteria for peg-IFNa treatment discontinua- 11 tion. However, no robust treatment discontinuation criteria Conflict of interest have been developed for HBeAg-negative CHB patients with 11 genotype B or C. Based on clinical experience, Chinese clin- Qianguo Mao has received grants and Jidong Jia has received ical experts recommend peg-IFNa treatment discontinuation grants and personal fees from Shanghai Roche Pharmaceut- at week 24 if HBsAg decreases to <1 log IU/mL and HBV DNA icals Ltd. during the conduct of this study. Yan Huang is an 24 decreases to <2 log IU/mL. Unfortunately, no statistically employee of Shanghai Roche Pharmaceuticals Ltd. The other significant relationship was found between baseline factors authors have no conflict of interests related to this and HBV DNA suppression at 1 year posttreatment in the publication. present study, and no evidence was generated to support treatment discontinuation at week 12 or week 24 for HBeAg-negative patients dominated by genotype B or C. Author contributions Regarding the changes in HBsAg by subgroups, in patients with ALT >5 ULN, HBsAg levels decreased at a slower rate Contributed to the design of the study and/or collection and during treatment compared with the subgroups with ALT #2 analysis of the data, drafting/critical revision of the manu- and those with ALT >2 and #5 ULN, but a greater decrease in script for intellectual content, played a role in final approval HBsAg level was shown at 1 year posttreatment in patients for publication of the manuscript, and agrees to be account- with ALT >5 ULN compared with the other two subgroups. able for the accuracy and integrity of the published work (YH), Reportedly, in patients with CHB, an ALT level $200 IU/L is and contributed to the design of the study and/or collection associated with HBsAg seroclearance.14,25,26 Such increased and analysis of the data, played a role in final approval for levels of ALT indicate that HBV-infected hepatocytes have publication of the manuscript, and agree to be accountable for triggered a strong host immune response which is likely a the accuracy and integrity of the published work (XC, QM, YX, result of the immunomodulating effects of peg-IFN27 that XD, QX, JS, ZG, XZ, YL, HZ, SZ, SL, FZ, YX, MZ, YH, XC, HR, will eventually lead to anti-HBe seroconversion and HBV and JJ). DNA reduction. Thus, it is easier for patients with high ALT to achieve HBsAg clearance. Additionally, the effect of ALT References seemed to be more obvious after discontinuation, which may be related to the mobilization of the immunity. [1] World Health Organization. Global hepatitis report, 2017. Availbale from: Among the 930 patients in the FAS who were evaluated for https://apps.who.int/iris/bitstream/handle/10665/255017/WHO-HIV-2017. 06-eng.pdf;jsessionid=FC414C553CBB7236A3FE29BA3A68C275? safety in the present study, more than half presented AEs and sequence=1. the most common AEs were decreased white blood, platelet, [2] Global prevalence, treatment, and prevention of hepatitis B virus infection in and neutrophil counts. These findings were similar to the peg- 2016: a modelling study. Lancet Gastroenterol Hepatol 2018;3:383–403. IFNa safety profile identified in peg-IFNa labeling.28. doi: 10.1016/S2468-1253(18)30056-6. [3] Cui F, Shen L, Li L, Wang H, Wang F, Bi S, et al. Prevention of chronic hepatitis The present study has some limitations, including those B after 3 decades of escalating vaccination policy, China. Emerg Infect Dis inherent to observational studies and inadequately controlled 2017;23:765–772. doi: 10.3201/eid2305.161477. confounders. Although the study was designed to document [4] Liu J, Zhang S, Wang Q, Shen H, Zhang M, Zhang Y, et al. Seroepidemiology potential confounders and adjustments for these potential of hepatitis B virus infection in 2 million men aged 21-49 years in rural China: a population-based, cross-sectional study. Lancet Infect Dis 2016;16:80–86. confounders were made in the statistical analysis, residual doi: 10.1016/S1473-3099(15)00218-2. confounding may still exist. In clinical practice, patients [5] Zhang Q, Qi W, Wang X, Zhang Y, Xu Y, Qin S, et al. Epidemiology of hepatitis cannot attend visits at predefined times as in interventional B and hepatitis C infections and benefits of programs for hepatitis prevention studies. If we had considered patients with missing data as in northeastern China: A cross-sectional study. Clin Infect Dis 2016;62:305– 312. doi: 10.1093/cid/civ859. nonresponders, this would have resulted in a considerable [6] El-Serag HB. Epidemiology of viral hepatitis and hepatocellular carcinoma. Gas- underestimation of treatment outcomes. Thus, analyzing troenterology 2012;142:1264–1273.e1. doi: 10.1053/j.gastro.2011.12.061. patients with available data is more reasonable in this [7] Simonetti J, Bulkow L, McMahon BJ, Homan C, Snowball M, Negus S, et al. situation. Clearance of hepatitis B surface antigen and risk of hepatocellular carcinoma in a cohort chronically infected with hepatitis B virus. Hepatology 2010;51: In conclusion, peg-IFNa showed good effectiveness and 1531–1537. doi: 10.1002/hep.23464. was well tolerated by HBeAg-negative CHB Chinese patients [8] Hadziyannis SJ, Papatheodoridis GV. Hepatitis B e antigen-negative chronic in routine clinical practice in China. Additionally, our results hepatitis B: natural history and treatment. Semin Liver Dis 2006;26:130– suggest that a certain proportion of HBeAg-negative patients 141. doi: 10.1055/s-2006-939751. [9] Marcellin P, Lau GK, Bonino F, Farci P, Hadziyannis S, Jin R, et al. Peginterferon have the potential to achieve functional cure (HBsAg loss) alfa-2a alone, lamivudine alone, and the two in combination in patients with with the use of peg-IFNa; however, 48 weeks of treatment HBeAg-negative chronic hepatitis B. N Engl J Med 2004;351:1206–1217. may not be sufficient. doi: 10.1056/NEJMoa040431. [10] Papatheodoridis GV, Manolakopoulos S, Dusheiko G, Archimandritis AJ. Ther- apeutic strategies in the management of patients with chronic hepatitis B virus infection. Lancet Infect Dis 2008;8:167–178. doi: 10.1016/S1473- Acknowledgments 3099(07)70264-5. [11] EASL 2017 Clinical Practice Guidelines on the management of hepatitis B virus infection. J Hepatol 2017;67:370–398. doi: 10.1016/j.jhep.2017.03.021. The authors wish to thank all of the investigators and [12] Lampertico P, Vigano M, Di Costanzo G, Sagnelli E, Fasano M, Di Marco V, et al. participating study sites, which are listed in Supplemental Extended (2 years) treatment with peginterferon alpha-2a [40kD] improves Table 1, as well as all the patients who participated in this sustained response rates in genotype D patients with HBeAg negative chronic study. The authors also wish to thank Michelle Belanger, MD, hepatitis B. J Hepatol 2010;52:S45. doi: 10.1016/S0168-8278(10)60100-6. [13] Rijckborst V, Ferenci P, Akdogan M, Pinarbasi B, ter Borg MJ, Simon K, et al. of Edanz Medical Writing for providing medical writing assis- Long-term follow-up of hepatitis B e antigen-negative patients treated with tance, which was funded by Shanghai Roche Pharmaceuticals peginterferon a-2a: progressive decrease in hepatitis B surface antigen in

256 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 249–257 Chen X. et al: Peg-IFNa 2a in HBeAg-negative CHB

responders. Eur J Gastroenterol Hepatol 2012;24:1012–1019. doi: 10. 2a (pegasys): TRACES study. Medicine (Baltimore) 2016;95:e3026. doi: 10. 1097/MEG.0b013e3283557e23. 1097/MD.0000000000003026. [14] Marcellin P, Bonino F, Lau GK, Farci P, Yurdaydin C, Piratvisuth T, et al. Sus- [22] Lampertico P, Viganò M, Di Costanzo GG, Sagnelli E, Fasano M, Di Marco V, tained response of hepatitis B e antigen-negative patients 3 years after et al. Randomised study comparing 48 and 96 weeks peginterferon a-2a treatment with peginterferon alpha-2a. Gastroenterology 2009;136:2169– therapy in genotype D HBeAg-negative chronic hepatitis B. Gut 2013;62: 2179.e1-4. doi: 10.1053/j.gastro.2009.03.006. 290–298. doi: 10.1136/gutjnl-2011-301430. [15] Brunetto MR, Moriconi F, Bonino F, Lau GK, Farci P, Yurdaydin C, et al. Hep- [23] Wei L, Xie Y, Chen X, Li X, Chen Y, Zhang J, et al. Effectiveness of pegylated- atitis B virus surface antigen levels: a guide to sustained response to pegin- interferon alpha-2a (40KD) therapy in HBeAg negative Chinese patients with terferon alfa-2a in HBeAg-negative chronic hepatitis B. Hepatology 2009;49: chronic hepatitis B at 3 years posttreatment: sub-analysis of the prospec- 1141–1150. doi: 10.1002/hep.22760. tive, global, observational S-collate study. Hepatol Int 2017;11(Suppl 1):44. [16] Marcellin P, Bonino F, Yurdaydin C, Hadziyannis S, Moucari R, Kapprell HP, doi: 10.1007/s12072-016-9783-9. [24] Zhang W, Zhang D, Dou X, Xie Q, Jiang J, Chen X, et al. Consensus on pegy- et al. Hepatitis B surface antigen levels: association with 5-year response to lated interferon alpha in treatment of chronic hepatitis B. J Clin Transl peginterferon alfa-2a in hepatitis B e-antigen-negative patients. Hepatol Int Hepatol 2018;6:1–10. doi: 10.14218/JCTH.2017.00073. 2013;7:88–97. doi: 10.1007/s12072-012-9343-x. [25] Kim GA, Lim YS, An J, Lee D, Shim JH, Kim KM, et al. HBsAg seroclearance [17] Chen X, Chen X, Chen W, Ma X, Huang J, Chen R. Extended peginterferon after nucleoside analogue therapy in patients with chronic hepatitis B: clin- alfa-2a (Pegasys) therapy in Chinese patients with HBeAg-negative chronic ical outcomes and durability. Gut 2014;63:1325–1332. doi: 10.1136/gutjnl- – hepatitis B. J Med Virol 2014;86:1705 1713. doi: 10.1002/jmv.24013. 2013-305517. [18] Lin CL, Kao JH. Hepatitis B viral factors and treatment responses in chronic [26] Nagaoka S, Abiru S, Komori A, Sasaki R, Bekki S, Hashimoto S, et al. Hepatic – hepatitis B. J Formos Med Assoc 2013;112:302 311. doi: 10.1016/j.jfma. flares promote rapid decline of serum hepatitis B surface antigen (HBsAg) in 2013.02.001. patients with HBsAg seroclearance: A long-term follow-up study. Hepatol [19] Hou J, Liu Z, Gu F. Epidemiology and prevention of hepatitis B virus infection. Res 2016;46:E89–E99. doi: 10.1111/hepr.12533. Int J Med Sci 2005;2:50–57. doi: 10.7150/ijms.2.50. [27] Masaki K, Suzuki F, Hara T, Kawamura Y, Sezaki H, Hosaka T, et al. Long-term [20] Hou J, Wang G, Wang F, Cheng J, Ren H, Zhuang H, et al. Guideline of pre- effects of peginterferon alfa-2a therapy in Japanese patients with chronic vention and treatment for chronic hepatitis B (2015 update). J Clin Transl hepatitis B virus infection. Virol J 2015;12:225. doi: 10.1186/s12985-015- Hepatol 2017;5:297–318. doi: 10.14218/JCTH.2016.00019. 0453-7. [21] Chon YE, Kim DJ, Kim SG, Kim IH, Bae SH, Hwang SG, et al. An observatio- [28] PEGASYS® (pegylated-interferon alpha-2a) injection, for subcutaneous use. nal, multicenter, cohort study evaluating the antiviral efficacy and safety in Available from: https://www.gene.com/download/pdf/pegasys_prescribing. Korean patients with chronic hepatitis B receiving pegylated interferon-alpha pdf.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 249–257 257 Review Article

HBV cccDNA and Its Potential as a Therapeutic Target

Anjing Zhu, Xinzhong Liao, Shuang Li, Hang Zhao, Limin Chen, Min Xu* and Xiaoqiong Duan*

Institute of Blood Transfusion, Chinese Academy of Medical Sciences and Peking Union Medical College, Chengdu, China

Abstract Integration of the HBV DNA into the host genome begins immediately after infection of hepatocytes.11 After transcrip- Chronic hepatitis B virus infection continues to be a major tion and translation, pgRNA is packaged and assembled in the health burden worldwide. It can cause various degrees of liver nucleocapsid, where it is transcribed to relaxed circular DNA. damage and is strongly associated with the development of Finally, the nucleocapsid is enveloped and the virion is liver cirrhosis and hepatocellular carcinoma. Covalently secreted from the hepatocyte.12,13 closed circular DNA in the nucleus of infected cells cannot The current standard therapy for HBV infection includes be disabled by present therapies which may lead to HBV PEGylated interferon a and nucleotide analogues (NAs). For persistence and relapse. In this review, we summarized the example, a recent study showed that interferon a induces a current knowledge on hepatitis B virus covalently closed long-term and sustainable suppression of cccDNA transcrip- circular DNA and its potential role as a therapeutic target. tion, possibly by altering epigenetic modification of cccDNA Citation of this article: Zhu A, Liao X, Li S, Zhao H, Chen L, minichromosomes.13 The inhibition of HBV replication by NAs Xu M, et al. HBV cccDNA and its potential as a therapeutic is mediated by targeting the viral RNA-dependent DNA poly- target. J Clin Transl Hepatol 2019;7(3):258–262. doi: merase which catalyzes the reverse transcription of pgRNA to 10.14218/JCTH.2018.00054. mature viral DNA. Five NAs have been approved for clinical use since 1998, including lamivudine, entecavir, telbivudine, adefovir dipivoxil and tenofovir.14 Although NAs have been considered as first-line therapy for the treatment of chronic Hepatitis B virus (HBV) life cycle and therapies HBV infection due to their high efficiency,13 they have failed to cure HBV in most cases because they cannot clear cccDNA.15 HBV was discovered by Blumberg and colleagues1 in 1965. As such, virus rebound occurs following the termination of Today, there are about 257 million people chronically infected NAs treatment. Obviously, disabling the cccDNA is key in with HBV, which causes 650000 deaths worldwide every year. terms of curing HBV infection.16 According to World Health Organization reports, HBV caused 887000 deaths, mostly from complex diseases (including cir- HBV cccDNA: Formation and modification rhosis and hepatocellular carcinoma) in 2015. HBV infection brings a huge burden to people and even society. cccDNA is stable and acts as a virus transcription template. The HBV genome is a circular, partially double-stranded There are 3–50 copies of cccDNA per infected cell, and the DNA, which is enveloped by an outer lipoprotein with an inner number of copies decreases when the host cell divides. Thus, nucleocapsid core. HBV DNA is only 3.2kb in length and recycling of the new relaxed circular form to the nucleus contains four overlapping open reading frames, known as S, occurs in order to maintain the relatively stable cccDNA C, P and X. The four open reading frames code for 7 viral copy number.17 There are three main steps in conversion of proteins: pre-S1, pre-S2, S, C, pre-C, X protein (commonly – the relaxed circular form to cccDNA: (1) unlocking of the known as HBX), and HBV polymerase.2 4 Once infected, HBV protein, which is covalently linked to the 5′ end of the enters host hepatocytes by binding to the sodium taurocho- (–)-DNA; (2) removal of the 5′ end of the (+)-strand consist- late cotransporting polypeptide NTCP and being endocy- ing of an RNA oligonucleotide; and (3) covalently ligation tosed.5 After entry, the virus releases its DNA-containing of both strands.18,19 Recent studies have shown that the nucleocapsid into the cytoplasm, which is then transported host DNA damage response is involved in the formation of to the nucleus. In the nucleus, the viral DNA is converted cccDNA, but the exact mechanism remains to be clarified.20 from its relaxed circular form to closed covalent circular – Epigenetic modifications, like histone acetylation and DNA DNA (cccDNA).6 9 The cccDNA is transcribed into HBV prege- methylation, play an important role in the transcriptional nomic RNA (pgRNA) and several subgenomic RNAs.6,10 activity of cccDNA.21–24 Six CpG islands have been reported in the HBV genome. The CpG islands distribute differently in Keywords: Hepatitis B virus (HBV); Covalently closed circular DNA (cccDNA); Therapeutic target. HBV genotypes, overlapping some functional genes. Three Abbreviations: cccDNA, covalently closed circular DNA; CRISPR/Cas, clustered conventional CpG islands (I, II, III) are potential targets for regularly interspaced short palindromic repeats/CRISPR associated; HBV, chronic HBV DNA methylation (Fig. 1).25 hepatitis B virus; NA, nucleotide analogue; pgRNA, pregenomic RNA; TALEN, transcription activator-like effector nuclease; ZFN, zinc finger nuclease. Received: 26 September 2018; Revised: 2 April 2019; Accepted: 10 July 2019 Novel strategies for eradicating HBV cccDNA *Correspondence to: Min Xu and Xiaoqiong Duan, Institute of Blood Transfusion, Chinese Academy of Medical Sciences and Peking Union Medical College, Chengdu 610052, China. Tel: +86-135-4080-7307, E-mail: [email protected] (MX) HBV cure depends on disabling of the HBV cccDNA, which is or [email protected] (XD) very difficult because cccDNA resides in the nucleus as an

258 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 258–262

gene editing can be used in combination with gene-silencing technologies for antiHBV therapy. Using TALEN-mediated homology directed recombination to introduce artificial primary miRNAs into HBV genome could boost the antiviral efficacy of TALENs.35 The CRISPR/Cas system is the adaptive immunity of bacteria and archaea, and acts against invading foreign DNA via RNA-guided DNA cleavage.36–40 The CRISPR/Cas9 system is a newly developed programmable genome-editing tool and allows for sequence-specific cleavage of DNA. Compared to other genome-editing tools, the advantage of the CRIPSR/ Cas9 system lies in its simplicity and flexibility in design. Several studies have demonstrated that the CRISPR/Cas9 system can efficiently destroy HBV cccDNA.41–45 Using the CRISPR/Cas9 system, scientists have completely excised a full-length 3175bp integrated HBV DNA from the host genome and disrupted the HBV cccDNA in a stable HBV cell line.44,45 Therefore, gene editing seems highly promising for disabling HBV cccDNA.29,34 These three genome-editing technologies work similarly by targeting and modifying DNA sequences using engineered nucleases, thereby inducing a targeted DNA double-strand break that stimulates the cellular DNA repair mechanisms. However, the engineered nucleases, recognition area, and the molecular mechanism are different.46 Engineered ZFNs and TALENs are composed of a DNA binding domain and a Fok I nuclease motif. The CRISPR-Cas9 genome editing system Fig. 1. Distribution diagram of CpG islands. Island I overlaps the start site of consists of two components: a “guide” RNA and a nonspecific the S gene; island II contains enhance I and the X gene promoter, close to the core gene promoter and enhance II; island III overlaps the Sp1 promoter and the start CRISPR-associated endonuclease (the Cas9). The recognition codon of the P gene. areas of the ZFNs, TALENs, CRISPR/Cas system are zinc- finger proteins, NLS, and sgRNA, respectively. ZFNs and TALENs use the Fok I enzyme to cleave target DNA, while episomal plasmid-like molecule to produce progeny CRISPR/Cas uses the Cas protein. The target sizes among 15,26 virus. Several gene therapy strategies have been pro- the three genome-editing tools are also different, that is posed to disable HBV cccDNA. (9-12bp)*2 for ZFNs and TALENs, 20bp+ NGG for CRISPR/ Cas.29,46,47 Silencing cccDNA expression by gene editing There are some limitations for these technologies. First, techniques specificity can be altered by context-dependent effects caused by interactions among neighboring zinc fingers of , Several genomic editing technologies, including zinc finger the DNA binding domain Second, the large size of TALENs nucleases (ZFNs), transcription activator-like effector makes it difficult to deliver. Finally, the CRISPR/Cas9 system nucleases (TALENs) and the clustered regularly interspaced faces several challenges like in vivo delivery efficiency and 48,49 short palindromic repeats/CRISPR associated (CRISPR/Cas) off-target cleavage. Cutting integrated HBV genomes by system, have been used to disrupt HBV cccDNA. CRISPR/Cas9 also raises serious concerns, because this ZFNs, a custom DNA endonuclease, are used to create a manipulation can cause genome instability. Compared to DNA double-strand break in a specific target site and repair by ZFNs and TALENs, the two advantages of the CRISPR/Cas9 creating sequence alterations at the cleavage sites.27–29 system have contributed to the advancement of the new tech- Through a proof-of-concept experiment, scientists found nology and generated widespread interest, according to its that anti-HBV ZFNs (cognate 6L and 6R ZFN pair) disrupted simplicity and flexibility. Thus, we believe that the CRISPR/ 36% of plasmid-derived viral sequences in a cell culture Cas9 system is very promising for curing chronic HBV once 50 model. Weber et al.30 designed three ZFNs targeting HBV several challenges are solved. polymerase, gene X and core, and delivered them into the HepAD38 cells by self-complementary adeno-associated Silencing cccDNA transcription by epigenetic viral vectors, respectively. They found these HBV-targeted modifications ZFNs produced a sustained suppression of HBV levels over the course of the experiment.31 HBV cccDNA exists in the nucleus as multiple copies of TALENs are similar to ZFNs. However, the DNA-binding nucleosome-decorated minichromosomes, which indicates domain of TALENs is highly repeated and derived from tran- that epigenetic modifications may influence HBV replication scription activator-like III effectors, which are proteins and persistent infection.51 It has been shown that DNA meth- secreted by Xanthomonas bacteria.32,33 The efficiency of ylation and histone acetylation are required for cccDNA TALENs in reducing HBV productions in cell culture was first formation.22 Therefore, regulation of DNA methylation or described by Bloom et al.34 in 2013. The investigators found histone acetylation is a potential method to reduce that targeting the S and C regions of cccDNA led to 35% cccDNA.21,52–55 Interferon a treatment has been adopted to decrease in cccDNA molecules in HepG2 cell lines. Moreover, silence cccDNA through epigenetic modification. It has been

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 258–262 259 Zhu A. et al: HBV cccDNA as a therapeutic target found that interferon a inhibits HBV replication by regulating problem and finding ways to deliver in vivo. What is more, for the epigenetic modification of HBV cccDNA. Interferon a a long time, lack of robust, reliable and quantifiable HBV represses the transcription of HBV cccDNA through recruit- cccDNA models has delayed the development of cccDNA ment of the histone deacetylases HDAC1 and hSirt1 and therapies. Recently, Yuan et al.68 established a cell line decreasing the acetylation of cccDNA-bound histones.56 through integrating 2–60 copies of the monomeric HBV Hyun et al.24 found that short hairpin RNA induced HBV genome into HepG2-derived cell lines, where the cccDNA cccDNA methylation to inhibit its transcription in human hep- could be produced and detected with specific primers. The atoma cells. HBV cccDNA transcription was regulated by CpG establishment of a cell line will provide a proper model for methylation during chronic HBV infection.21 CpG island II evaluation of drugs or therapies targeting on cccDNA. methylation has been shown to significantly decrease Though there are challenges to overcome for both gene cccDNA transcription and subsequent viral core DNA replica- editing- and epigenetic modification-based strategies and it tion, while CpG island III methylation has been shown to be there remains a long road from basic research to clinical appli- associated with low serum HBsAg titers (Fig. 1).21,53,54,57–59 cation, it is very likely that we will conquer HBV similar to the Interestingly, HBx can reduce chromatin-mediated transcrip- hepatitis C virus cure in the future. tional repression of HBV cccDNA caused by SETDB1 histone methyltransferase and therefore allow the establishment of Conflict of interest active chromatin.60 Recently, it has also been reported that HB core carboxyl-terminal domain arginine residues reduced The authors have no conflict of interests related to this acetylation of cccDNA-bound histones and thus reduced the publication. interaction of HBc with cccDNA.61 Two enzymes have been found to regulate the methylation.62,63 Protein arginine methyltransferase 5 can regulate symmetric dimethylation of argi- Author contributions nine 3 on histone 4 of cccDNA.62 The silent mating type information regulation 2 homolog 3 SIRT3 restricts HBV tran- Collected the data and wrote the manuscript (AZ, XD), scription and replication via epigenetic regulation of cccDNA, revised the manuscript and answered the reviewersquestions involving SUV39H1 and SETD1A histone methyltransferases.63 (AZ, MX), directed and wrote the manuscript with comments HBV replication is also regulated by the acetylation status (LC), helped to edit the manuscript (XL, SL, HZ). All authors of HBV cccDNA-bound histone 3 and histone 4.22 The Np95/ have seen and approved the content of this manuscript. ICBP90-like RING finger protein NIRF, a novel E3 ubiquitin ligase, has been found to inhibit HBV DNA replication and References hepatitis B e antigen secretion in HepG2 cells through reduc- 64 ing HBV cccDNA-bound histone 3. Curcumin can inhibit HBV [1] Blumberg BS, Gerstley BJ, Hungerford DA, London WT, Sutnick AI. A serum replication via down-regulation of cccDNA-bound histone ace- antigen (Australia antigen) in Down’s syndrome, leukemia, and hepatitis. tylation.65 Retinoid X receptor a was also reported to be able Ann Intern Med 1967;66:924–931. doi: 10.7326/0003-4819-66-5-924. to regulate the replication of HBV and modulate the HBV [2] Tiollais P, Pourcel C, Dejean A. The hepatitis B virus. Nature 1985;317:489– 66 495. doi: 10.1038/317489a0. cccDNA epigenetically. In addition, basal core promoter [3] Arzumanyan A, Reis HM, Feitelson MA. Pathogenic mechanisms in HBV- and mutations were found to inhibit viral replication through mod- HCV-associated hepatocellular carcinoma. Nat Rev Cancer 2013;13:123– ulating the acetylation and deacetylation of cccDNA-bound 135. doi: 10.1038/nrc3449. histones, while preCore mutations have been shown to have [4] Doo EC, Ghany MG. Hepatitis B virology for clinicians. Clin Liver Dis 2010;14: 67 397–408. doi: 10.1016/j.cld.2010.05.001. no effect on viral replication. These collective findings indi- [5] Li W. The hepatitis B virus receptor. Annu Rev Cell Dev Biol 2015;31:125– cate that regulation of the DNA methylation and histone ace- 147. doi: 10.1146/annurev-cellbio-100814-125241. tylation may inactivate cccDNA transcription and thus inhibit [6] Seeger C, Mason WS. Hepatitis B virus biology. Microbiol Mol Biol Rev 2000; HBV replication. 64:51–68. doi: 10.1128/mmbr.64.1.51-68.2000. [7] Bock CT, Schranz P, Schröder CH, Zentgraf H. Hepatitis B virus genome is Disabling cccDNA is the key to curing hepatitis B. Studies organized into nucleosomes in the nucleus of the infected cell. Virus Genes have shown the great potential of gene editing technologies 1994;8:215–229. doi: 10.1007/bf01703079. for disrupting the cccDNA in cultured cells and in hydro- [8] Newbold JE, Xin H, Tencza M, Sherman G, Dean J, Bowden S, et al. The dynamically-injected mice. However, in the case of HBV covalently closed duplex form of the hepadnavirus genome exists in situ as a heterogeneous population of viral minichromosomes. J Virol 1995;69: infection, there are some disadvantages of using gene 3350–3357. editing technologies. First, HBV DNA often integrates into [9] Zoulim F. New insight on hepatitis B virus persistence from the study of intra- the host genome and undesirable gene cleavage may occur. hepatic viral cccDNA. J Hepatol 2005;42:302–308. doi: 10.1016/j.jhep. Second, off-target effects has been observed in some studies. 2004.12.015. [10] Block TM, Guo H, Guo JT. Molecular virology of hepatitis B virus for clinicians. Finally, the means by how to deliver in vivo remains a chal- Clin Liver Dis 2007;11:685–706. doi: 10.1016/j.cld.2007.08.002. lenge for therapeutic application of these technologies. The [11] Tu T, Budzinska MA, Vondran FWR, Shackel NA, Urban S. Hepatitis B Virus epigenetic modifications mentioned above are related to the DNA Integration Occurs Early in the Viral Life Cycle in an In Vitro Infection replication and transcription of HBV cccDNA. While they can Model via Sodium Taurocholate Cotransporting Polypeptide-Dependent Uptake of Enveloped Virus Particles. J Virol 2018;92:e02007–17. doi: 10. regulate the HBV cccDNA, they cannot eliminate the HBV 1128/JVI.02007-17. cccDNA. [12] Huovila AP, Eder AM, Fuller SD. Hepatitis B surface antigen assembles in a post-ER, pre-Golgi compartment. J Cell Biol 1992;118:1305–1320. doi: 10. 1083/jcb.118.6.1305. Conclusions [13] Tang LSY, Covert E, Wilson E, Kottilil S. Chronic hepatitis B infection: A review. JAMA 2018;319:1802–1813. doi: 10.1001/jama.2018.3795. In conclusion, disabling or complete inactivation of cccDNA is [14] Liaw YF, Sung JJ, Chow WC, Farrell G, Lee CZ, Yuen H, et al. Lamivudine for the desired end-point of HBV treatment. However, the most patients with chronic hepatitis B and advanced liver disease. N Engl J Med 2004;351:1521–1531. doi: 10.1056/NEJMoa033364. urgent issues should be addressed before using gene editing [15] Zhou T, Guo H, Guo JT, Cuconati A, Mehta A, Block TM. Hepatitis B virus e technologies for HBV treatment include solving the off-target antigen production is dependent upon covalently closed circular (ccc) DNA in

260 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 258–262 Zhu A. et al: HBV cccDNA as a therapeutic target

HepAD38 cell cultures and may serve as a cccDNA surrogate in antiviral [39] Garneau JE, Dupuis MÈ, Villion M, Romero DA, Barrangou R, Boyaval P, et al. screening assays. Antiviral Res 2006;72:116–124. doi: 10.1016/j.antiviral. The CRISPR/Cas bacterial immune system cleaves bacteriophage and 2006.05.006. plasmid DNA. Nature 2010;468:67–71. doi: 10.1038/nature09523. [16] Seeger C, Mason WS. Molecular biology of hepatitis B virus infection. Virol- [40] Manica A, Zebec Z, Teichmann D, Schleper C. In vivo activity of CRISPR- ogy 2015;479–480:672–686. doi: 10.1016/j.virol.2015.02.031. mediated virus defence in a hyperthermophilic archaeon. Mol Microbiol [17] Ji M, Hu K. Recent advances in the study of hepatitis B virus covalently closed 2011;80:481–491. doi: 10.1111/j.1365-2958.2011.07586.x. circular DNA. Virol Sin 2017;32:454–464. doi: 10.1007/s12250-017-4009-4. [41] Seeger C, Sohn JA. Complete Spectrum of CRISPR/Cas9-induced Mutations [18] Qi Y, Gao Z, Xu G, Peng B, Liu C, Yan H, et al. DNA polymerase k is a key on HBV cccDNA. Mol Ther 2016;24:1258–1266. doi: 10.1038/mt.2016.94. cellular factor for the formation of covalently closed circular DNA of hepatitis [42] Ramanan V, Shlomai A, Cox DB, Schwartz RE, Michailidis E, Bhatta A, et al. B virus. PLoS Pathog 2016;12:e1005893. doi: 10.1371/journal.ppat. CRISPR/Cas9 cleavage of viral DNA efficiently suppresses hepatitis B virus. 1005893. Sci Rep 2015;5:10833. doi: 10.1038/srep10833. [19] Königer C, Wingert I, Marsmann M, Rösler C, Beck J, Nassal M. Involvement [43] Karimova M, Beschorner N, Dammermann W, Chemnitz J, Indenbirken D, of the host DNA-repair enzyme TDP2 in formation of the covalently closed Bockmann JH, et al. CRISPR/Cas9 nickase-mediated disruption of hepatitis circular DNA persistence reservoir of hepatitis B viruses. Proc Natl Acad Sci B virus open reading frame S and X. Sci Rep 2015;5:13734. doi: 10. U S A 2014;111:E4244–E4253. doi: 10.1073/pnas.1409986111. 1038/srep13734. [20] Schreiner S, Nassal M. A role for the host DNA damage response in hepatitis [44] Liu X, Hao R, Chen S, Guo D, Chen Y. Inhibition of hepatitis B virus by the B virus cccDNA formation-and beyond? Viruses 2017;9:E125. doi: 10. CRISPR/Cas9 system via targeting the conserved regions of the viral 3390/v9050125. genome. J Gen Virol 2015;96:2252–2261. doi: 10.1099/vir.0.000159. [21] Kim JW, Lee SH, Park YS, Hwang JH, Jeong SH, Kim N, et al. Replicative [45] Li H, Sheng C, Wang S, Yang L, Liang Y, Huang Y, et al. Removal of integrated activity of hepatitis B virus is negatively associated with methylation of cova- hepatitis B virus DNA using CRISPR-Cas9. Front Cell Infect Microbiol 2017;7: lently closed circular DNA in advanced hepatitis B virus infection. Intervirol- 91. doi: 10.3389/fcimb.2017.00091. ogy 2011;54:316–325. doi: 10.1159/000321450. [46] Gaj T, Gersbach CA, Barbas CF 3rd. ZFN, TALEN, and CRISPR/Cas-based [22] Pollicino T, Belloni L, Raffa G, Pediconi N, Squadrito G, Raimondo G, et al. methods for genome engineering. Trends Biotechnol 2013;31:397–405. Hepatitis B virus replication is regulated by the acetylation status of hepatitis doi: 10.1016/j.tibtech.2013.04.004. B virus cccDNA-bound H3 and H4 histones. Gastroenterology 2006;130: [47] Holkers M, Maggio I, Liu J, Janssen JM, Miselli F, Mussolino C, et al. Differ- 823–837. doi: 10.1053/j.gastro.2006.01.001. ential integrity of TALE nuclease genes following adenoviral and lentiviral [23] Liu F, Campagna M, Qi Y, Zhao X, Guo F, Xu C, et al. Alpha-interferon sup- vector gene transfer into human cells. Nucleic Acids Res 2013;41:e63. presses hepadnavirus transcription by altering epigenetic modification of doi: 10.1093/nar/gks1446. cccDNA minichromosomes. PLoS Pathog 2013;9:e1003613. doi: 10. [48] Mussolino C, Morbitzer R, Lütge F, Dannemann N, Lahaye T, Cathomen T. A 1371/journal.ppat.1003613. novel TALE nuclease scaffold enables high genome editing activity in combi- [24] Park HK, Min BY, Kim NY, Jang ES, Shin CM, Park YS, et al. Short hairpin RNA nation with low toxicity. Nucleic Acids Res 2011;39:9283–9293. doi: 10. induces methylation of hepatitis B virus covalently closed circular DNA in 1093/nar/gkr597. human hepatoma cells. Biochem Biophys Res Commun 2013;436:152– [49] Tsai SQ, Zheng Z, Nguyen NT, Liebers M, Topkar VV, Thapar V, et al. GUIDE- 155. doi: 10.1016/j.bbrc.2013.04.108. seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas [25] Zhang Y, Li C, Zhang Y, Zhu H, Kang Y, Liu H, et al. Comparative analysis of nucleases. Nat Biotechnol 2015;33:187–197. doi: 10.1038/nbt.3117. CpG islands among HBV genotypes. PLoS One 2013;8:e56711. doi: 10. [50] Yang HC, Chen PJ. The potential and challenges of CRISPR-Cas in eradication 1371/journal.pone.0056711. of hepatitis B virus covalently closed circular DNA. Virus Res 2018;244:304– [26] Werle-Lapostolle B, Bowden S, Locarnini S, Wursthorn K, Petersen J, Lau G, 310. doi: 10.1016/j.virusres.2017.06.010. et al. Persistence of cccDNA during the natural history of chronic hepatitis B [51] Levrero M, Pollicino T, Petersen J, Belloni L, Raimondo G, Dandri M. Control of and decline during adefovir dipivoxil therapy. Gastroenterology 2004;126: cccDNA function in hepatitis B virus infection. J Hepatol 2009;51:581–592. 1750–1758. doi: 10.1053/j.gastro.2004.03.018. doi: 10.1016/j.jhep.2009.05.022. [27] Cathomen T, Keith Joung J. Zinc-finger nucleases: The next generation [52] Miller RH, Robinson WS. Integrated hepatitis B virus DNA sequences specify- emerges. Mol Ther 2008;16:1200–1207. doi: 10.1038/mt.2008.114. ing the major viral core polypeptide are methylated in PLC/PRF/5 cells. Proc [28] Kim YG, Cha J, Chandrasegaran S. Hybrid restriction enzymes: zinc finger Natl Acad Sci U S A 1983;80:2534–2538. doi: 10.1073/pnas.80.9.2534. fusions to Fok I cleavage domain. Proc Natl Acad Sci U S A 1996;93:1156– [53] Guo Y, Li Y, Mu S, Zhang J, Yan Z. Evidence that methylation of hepatitis B 1160. doi: 10.1073/pnas.93.3.1156. virus covalently closed circular DNA in liver tissues of patients with chronic [29] Cradick TJ, Keck K, Bradshaw S, Jamieson AC, McCaffrey AP. Zinc-finger hepatitis B modulates HBV replication. J Med Virol 2009;81:1177–1183. nucleases as a novel therapeutic strategy for targeting hepatitis B virus doi: 10.1002/jmv.21525. DNAs. Mol Ther 2010;18:947–954. doi: 10.1038/mt.2010.20. [54] Vivekanandan P, Thomas D, Torbenson M. Methylation regulates hepatitis B [30] Weber ND, Stone D, Sedlak RH, De Silva Feelixge HS, Roychoudhury P, viral protein expression. J Infect Dis 2009;199:1286–1291. doi: 10. Schiffer JT, et al. AAV-mediated delivery of zinc finger nucleases targeting 1086/597614. hepatitis B virus inhibits active replication. PLoS One 2014;9:e97579. doi: [55] Curradi M, Izzo A, Badaracco G, Landsberger N. Molecular mechanisms of 10.1371/journal.pone.0097579. gene silencing mediated by DNA methylation. Mol Cell Biol 2002;22:3157– [31] Bloom K, Maepa MB, Ely A, Arbuthnot P. Gene therapy for chronic 3173. doi: 10.1128/mcb.22.9.3157-3173.2002. HBV-Can we eliminate cccDNA? Genes (Basel) 2018;9:E207. doi: 10. [56] Belloni L, Allweiss L, Guerrieri F, Pediconi N, Volz T, Pollicino T, et al. IFN-a 3390/genes9040207. inhibits HBV transcription and replication in cell culture and in humanized [32] Maeder ML, Thibodeau-Beganny S, Osiak A, Wright DA, Anthony RM, Eich- mice by targeting the epigenetic regulation of the nuclear cccDNA minichro- tinger M, et al. Rapid “open-source” engineering of customized zinc-finger mosome. J Clin Invest 2012;122:529–537. doi: 10.1172/JCI58847. nucleases for highly efficient gene modification. Mol Cell 2008;31:294–301. [57] Zhang Y, Mao R, Yan R, Cai D, Zhang Y, Zhu H, et al. Transcription of hepatitis doi: 10.1016/j.molcel.2008.06.016. B virus covalently closed circular DNA is regulated by CpG methylation during [33] Boch J, Bonas U. Xanthomonas AvrBs3 family-type III effectors: discovery chronic infection. PLoS One 2014;9:e110442. doi: 10.1371/journal.pone. and function. Annu Rev Phytopathol 2010;48:419–436. doi: 10. 0110442. 1146/annurev-phyto-080508-081936. [58] Vivekanandan P, Thomas D, Torbenson M. Hepatitis B viral DNA is methylated [34] Bloom K, Ely A, Mussolino C, Cathomen T, Arbuthnot P. Inactivation of hep- in liver tissues. J Viral Hepat 2008;15:103–107. doi: 10.1111/j.1365-2893. atitis B virus replication in cultured cells and in vivo with engineered tran- 2007.00905.x. scription activator-like effector nucleases. Mol Ther 2013;21:1889–1897. [59] Vivekanandan P, Kannangai R, Ray SC, Thomas DL, Torbenson M. Compre- doi: 10.1038/mt.2013.170. hensive genetic and epigenetic analysis of occult hepatitis B from liver tissue [35] Dreyer T, Nicholson S, Ely A, Arbuthnot P, Bloom K. Improved antiviral effi- samples. Clin Infect Dis 2008;46:1227–1236. doi: 10.1086/529437. cacy using TALEN-mediated homology directed recombination to introduce [60] Rivière L, Gerossier L, Ducroux A, Dion S, Deng Q, Michel ML, et al. HBx artificial primary miRNAs into DNA of hepatitis B virus. Biochem Biophys Res relieves chromatin-mediated transcriptional repression of hepatitis B viral Commun 2016;478:1563–1568. doi: 10.1016/j.bbrc.2016.08.152. cccDNA involving SETDB1 histone methyltransferase. J Hepatol 2015;63: [36] Wiedenheft B, Sternberg SH, Doudna JA. RNA-guided genetic silencing 1093–1102. doi: 10.1016/j.jhep.2015.06.023. systems in bacteria and archaea. Nature 2012;482:331–338. doi: 10. [61] Chong CK, Cheng CYS, Tsoi SYJ, Huang FY, Liu F, Seto WK, et al.Roleof 1038/nature10886. hepatitis B core protein in HBV transcription and recruitment of histone ace- [37] Shah SA, Garrett RA. CRISPR/Cas and Cmr modules, mobility and evolution tyltransferases to cccDNA minichromosome. Antiviral Res 2017;144:1–7. of adaptive immune systems. Res Microbiol 2011;162:27–38. doi: 10. doi: 10.1016/j.antiviral.2017.05.003. 1016/j.resmic.2010.09.001. [62] Zhang W, Chen J, Wu M, Zhang X, Zhang M, Yue L, et al. PRMT5 restricts [38] Jore MM, Brouns SJ, van der Oost J. RNA in defense: CRISPRs protect pro- hepatitis B virus replication through epigenetic repression of covalently karyotes against mobile genetic elements. Cold Spring Harb Perspect Biol closed circular DNA transcription and interference with pregenomic RNA 2012;4:a003657. doi: 10.1101/cshperspect.a003657. encapsidation. Hepatology 2017;66:398–415. doi: 10.1002/hep.29133.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 258–262 261 Zhu A. et al: HBV cccDNA as a therapeutic target

[63] Ren JH, Hu JL, Cheng ST, Yu HB, Wong VKW, Law BYK, et al. SIRT3 restricts [66] Song M, Sun Y, Tian J, He W, Xu G, Jing Z, et al. Silencing retinoid X receptor hepatitis B virus transcription and replication through epigenetic regulation alpha expression enhances early-stage hepatitis B virus infection in cell cul- of covalently closed circular DNA involving suppressor of variegation 3-9 tures. J Virol 2018;92:e01771–17. doi: 10.1128/JVI.01771-17. homolog 1 and SET domain containing 1A histone methyltransferases. Hep- [67] Koumbi L, Pollicino T, Raimondo G, Stampoulis D, Khakoo S, Karayiannis P. atology 2018;68:1260–1276. doi: 10.1002/hep.29912. Hepatitis B virus basal core promoter mutations show lower replication [64] Qian G, Hu B, Zhou D, Xuan Y, Bai L, Duan C. NIRF, a novel ubiquitin ligase, fitness associated with cccDNA acetylation status. Virus Res 2016;220: inhibits hepatitis B virus replication through effect on HBV core protein and 150–160. doi: 10.1016/j.virusres.2016.04.022. H3 histones. DNA Cell Biol 2015;34:327–332. doi: 10.1089/dna.2014.2714. [68] Wu M, Li J, Yue L, Bai L, Li Y, Chen J, et al. Establishment of Cre-mediated [65]WeiZQ,ZhangYH,KeCZ,ChenHX,RenP,HeYL,et al. Curcumin inhibits HBV recombinant cccDNA (rcccDNA) cell line for cccDNA biology and antiviral hepatitis B virus infection by down-regulating cccDNA-bound histone acetylation. screening assays. Antiviral Res 2018;152:45–52. doi: 10.1016/j.antiviral. World J Gastroenterol 2017;23:6252–6260. doi: 10.3748/wjg.v23.i34.6252. 2018.02.007.

262 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 258–262 Review Article

Complementary and Alternative Medicine-related Drug-induced Liver Injury in Asia

Cyriac Abby Philips*1, Philip Augustine2, Sasidharan Rajesh3, Praveen Kumar Y4 and Deepak Madhu5

1The Liver Unit and Monarch Liver Lab, Cochin Gastroenterology Group, Ernakulam Medical Centre, Kochi, Kerala, India; 2Gastroenterology, Cochin Gastroenterology Group, Ernakulam Medical Centre, Kochi, Kerala, India; 3Interventional Radiology, Cochin Gastroenterology Group, Ernakulam Medical Centre, Kochi, Kerala, India; 4Gastroenterology, Government Medical College, Thrissur, Kerala, India; 5Gastroenterology, Aster MIMS, Calicut, Kerala, India

Abstract (EBM). Complementary medicine is that which is utilized along with EBM, while alternative medicine wholly replaces The use of complementary and alternative medicines (CAMs) EBM. Traditional medicine (TM), which differs from CAM, is for treatment of acute and chronic diseases is on the rise defined by the World Health Organization as the “sum total of world over, especially in Asian countries, and mostly in China the knowledge, skill, and practices based on the theories, and India. Drug-induced liver injury (DILI) due to CAM is beliefs, and experiences indigenous to different cultures, increasingly reported in the literature from multiple centers used in the maintenance of health as well as in the preven- all around the world and with large-number patient series tion, diagnosis, improvement or treatment of physical and published from the West, mostly based on nation-wide DILI mental illness.”1,2 networks and multicenter collaboration. Comprehensive DILI Most CAM systems incorporate TM practices. In Asia, the networks are lacking among major Asian countries with high major CAM practices include traditional Chinese medicine incidence of CAM practices. Chinese medical societies dealing (TCM) in Mainland China, Ayurveda in India, traditional with drug toxicity, CAM practice and hepatobiliary disease Korean medicine in Korea, and TCM-based TM in Japan have adopted an integrated approach to establishing identi- (known as Kampo). A mixture of influences from these fication, diagnosis and treatment of CAM-related DILI, repre- major systems along with indigenous practices form part of senting a systematic approach that could be iterated by other CAM in other countries in the South-East Asian region. Siddha countries for improving patient outcomes. In this exhaustive medicine, a system of traditional medicine originating in review, we provide published data on CAM-related DILI in ancient Tamil Nadu (a Southern state in India) is followed in Asia, with detail on incidences along with analysis of patient South India and Sri Lanka. The Persian-Arabic traditional population and their clinical outcomes. Concise and clear medicine system known as Unani, which was practiced in discussion on commonly implicated CAM agents in major Mughal India, is followed in modern-day Central Asia. Asian countries and associated chemical and toxicology In India, six CAM systems have official recognition (Ayur- analyses as well as descriptions of liver biopsy findings are veda, Yoga, Naturopathy, Unani Medicine, Siddha, and discussed with future directions. Homeopathy, known as AYUSH) with institutionalized educa- Citation of this article: Philips CA, Augustine P, Rajesh S, tion systems, governed by the ministry of AYUSH and under Kumar Y P, Madhu D. Complementary and alternative medi- regulation by the Central Council for Indian Medicine, which is cine-related drug-induced liver injury in Asia. J Clin Transl Hep- independent from the Indian Medical Council, the latter which atol 2019;7(3):263–274. doi: 10.14218/JCTH.2019.00024. regulates modern medicine. In China, TCM is fully integrated into the main healthcare system, with TCM institutions governed by the same national legislation on conventional 3,4 Introduction or mainstream medicine. There is strong general public, political, cultural and social acceptance to CAM in almost all Complementary and alternative medicine (CAM) comprise countries in Asia, but the lack of scientific evidence-based various health care practices and products that do not form knowledge, paucity in adequate basic science as well as clin- part of conventional (modern) evidence-based medical care ical trial-based research on CAM and ignorance regarding potential drug toxicity with CAM practices and associated

Keywords: CAM; Herbals; Ayurveda; Chinese; DILI; Hepatitis; Cirrhosis; ACLF; products have led to a drift in general acceptance in utilizing Liver biopsy. these systems for healthcare among conventional medical Abbreviations: ACLF, acute on chronic liver failure; AHM, Ayurvedic herbal med- practitioners and specialists. ication; ALF, acute liver failure; AYUSH, Ayurveda, Yoga, Naturopathy, Unani Med- While conventional modern medicine drugs require pre- icine, Siddha, and Homeopathy; CAM, complementary and alternative medicine; CAM-DILI, CAM-related drug-induced liver injury; EBM, evidence-based medical clinical and phased clinical trials before approval of marketing care; TCM, traditional Chinese medicine; PM, Polygonum multiflorum; PMP, and use, CAM drugs do not have to undergo this process. P. multiflorum praeparata; TM, traditional medicine. Wide word-of-mouth and social media popularity on assumed Received: 28 June 2019; Revised: 12 August 2019; Accepted: 13 August 2019 but an ‘assured’, ‘safer’ profile associated with CAM has led to *Correspondence to: Cyriac Abby Philips, Philip Augustine Associates, 35/194 B, Symphony, Automobile Rd, Palarivattom, Kochi, Kerala 682025, India. E-mail: a massive surge in use of these practices among the general [email protected] population and patients with acute as well as chronic

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 263

Fig. 1. Retrieved Ayurvedic drugs from two patients. The absence of labeling, regarding ingredient and component detail, is evident in both. The first patient took the decoction and dried herbs for management of diabetes mellitus (A) while the second patient ingested the drugs for cure from hepatitis B virus infection. Both patients developed severe drug-induced liver injury, leading to death in the latter due to acute on chronic liver failure.

264 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 Philips C.A. et al: CAM-related DILI in Asia

Eosinophilia was less frequent and the death rate was lower in of actual and recommended dosage had excessive consump- cases in which Chinese herbal medicines were implicated.12 tion of TCM, assuming safety associated with natural prod- In a study by Liem et al.13 on use of TCM among patients ucts. Liver enzyme elevation was higher with CAM compared with chronic hepatitis B virus infection, it was found that 74% to prescription drugs, and the hepatocellular type of liver injury of the patients were Asian, 60% were males and only 46% was common; ninety-six point three percent of patients used antivirals. Seventy-one percent of these patients were recovered. using CAM that contained potential hepatotoxic agents, such Lai and coworkers21 reported on the clinical characteristics as green tea extracts and St. John’s wort. Family history of and liver histological changes in 138 patients with DILI from a chronic hepatitis B virus infection, lower levels of viral load, single center in China. They found that Chinese herbal and and higher socioeconomic status were independently associ- proprietary medicines and health products accounted for ated with CAM use in this cohort of patients. The authors shed almost 60% of DILI in their series. Wang and colleagues22 light on important factors associated with CAM use among summarized the current literature with regards to clinical pre- patients with chronic disease. sentation, disease course, and prognosis of TCM-related DILI. The proportion of TCM-related DILI out of all DILI ranged from Korea 22%–34% in most of the regions in China. Women and elderly were the most common users, and gynecologic and Suk et al.14 performed a prospective nationwide study of DILI orthopedic diseases were the most common indications in Korea, in which the most common causative agent impli- for CAM use. Male gender, higher levels of direct bilirubin, cated was TCM (27.5%). Dietary supplements, medicinal aspartate aminotransferase, and hypoalbuminemia were plants, folk remedies, and poly-herbal preparations were the predictors of mortality. Higher risk of death or liver trans- found to cause DILI in 13.7%, 9.4%, 8.6%, and 3.2%, plantation was notable in patients with cholestatic-type TCM- respectively. The frequency of DILI due to herbs was signifi- related DILI, and the mortality rate associated with TCM-DILI cantly higher among female patients, and liver enzyme ele- was 4.67%. 23 vation was higher in patients who consumed CAM. The acute Ou et al. reported that the crude annual incidence rate of hepatocellular type of DILI was the most common pattern of DILI was 92.95 cases per 100,000 patients, with Chinese liver injury. Three out of five patients who died in this series herbal medicines identified as the primary cause of DILI in had DILI due to CAMs and one who underwent liver trans- 36% of the patients and an overall mortality of approximately plantation also had herbal medicine-related DILI. Recently, 9% (out of which, 70% died of progressive liver failure). The Cho and colleagues15 evaluated the risk of herbal medicine- model for end-stage liver disease score and hypoalbumine- related DILI in a nationwide prospective study in Korea. Out of mia at baseline were found to be independent predictors of 1001 in-patients, 6 were found to have CAM-DILI with death. Roussel Uclaf Causality Assessment Scoring Method scores ranging from 4 to 7. All were women with the hepatocellular India type of DILI. One patient fulfilled Hy’s law, and all recovered after drug withdrawal. Udayakumar et al.24 showed that the use of traditional indigenous herbal medicine (South-Indian Tamil native healers) in China patients with acute liver failure (ALF) increased risk of death. In the first detailed single center study on DILI (n = 313) from Chinese herbal medicine, medicines for tuberculosis, and India, Devarbhavi and colleagues25 found that DILI associ- antibiotics were described as leading causes for DILI in ated with CAM (Indian Ayurvedic medicines) was seen only China, a different pattern compared to the West. A meta- in 1.3% patients, of whom 50% died due to progressive liver analysis that followed demonstrated that from 2006 to 2012, failure. Philips and colleagues26 were the first to study clinical the most common cause was due to TCM (in 30.8%). In two outcomes and analyze component toxicology of patients systematic analyses consisting of 9,335 and 24,112 patients developing acute liver injury due to use of Ayurvedic herbal with DILI respectively, TCM was found responsible for DILI in medications (AHMs). Out of 1440 liver disease patients, 94 18.6% and 21.2% cases.16 Woo et al.17 showed that, among were found to have a severe liver injury and associated AHM a total of 1169 in-patients, only 0.43% cases were attributed intake, of whom, 33 patients had AHM-DILI by the Roussel to TCM alone. In a study on 300 patients with DILI, Hou Uclaf Causality Assessment Scoring Method. One-third of the et al.18 found that Chinese herbal medicines were the patients consumed AHM from unregistered traditional Ayur- leading cause of liver injury (in 40.3%), and almost 50% of vedic healers. Males predominated (70.4%) and median these patients used herbal medications for treatment of der- duration of drug intake to symptom development was matological and orthopedic diseases. 28 days. The most common reason for use of CAMs was for In a most recent study, the largest ever, on DILI from gastrointestinal symptoms. Overall mortality was approxi- Mainland China, Shen et al.19 demonstrated that TCM and mately 19%. Independent predictors of mortality included herbal and dietary supplements were the leading causes for hepatic encephalopathy and hypoalbuminemia at presenta- liver injury, in almost 27% (n = 25,927), and most commonly tion. Another unique feature of the patient population was among women, showing a rising and alarming trend in hep- the identification of autoantibodies in patients, revealing the atotoxicity associated with CAM in China. In 16% of the impli- possibility of a drug-induced autoimmune-like hepatitis asso- cated TCMs, formulation components were unknown. Zhang ciated with CAM-DILI. et al.20, in 2016, systematically reviewed data on TCM- The same study group also described characteristics and induced liver injury, with particular focus on clinical character- clinical outcomes in cirrhosis patients ingesting CAMs. The istics and causative herbs. Among 58 cases, the majority authors found that out of 1,666 patients with cirrhosis, 68% being male, with DILI latency the median period was had used CAM and 35.7% patients presented with CAM- 30 days. Nine of eleven cases with well-defined information related DILI leading to acute on chronic liver failure (ACLF).

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 265 Philips C.A. et al: CAM-related DILI in Asia

In patients with CAM-DILI-related ACLF, 63% self-medicated TCM consumption were chronic skin diseases, followed by based on social media sharing, 83% were male patients and diseases of the respiratory tract. Women predominated, overall 53% of patients died, with a median survival 194 days. with the average age being 39.7 years. The mortality of Baseline overt hepatic encephalopathy and CLIF-C-organ patients with TCM-related ALF was 60%, in whom, one-third failure score, total bilirubin, hyponatremia and leukocytosis, died of severe coagulation failure-related symptoms and 78% and grade of ACLF predicted 1-, 3-, 6- and 12-month mortal- with overt hepatic encephalopathy grade >2 having died. The 27 ity, respectively. majority of patients in this series consumed Tripterygium wilfordii (known as Lei Gong Teng in TCM, or Thunder-God Vine). CAM-related chronic liver disease and ACLF Histopathological patterns were not studied in this series of patients. Recently, a large multicenter study demonstrated that ACLF in In the study by Teo and coworkers,9 the most common Asia-Pacific countries was predominantly due to CAMs (in CAM components/agents associated with DILI included Fu approximately 72% of patients; n = 3132) and that the most Ling (Sclerotium poriae cocos), Huang Qin (Radix scutellaria common cause for chronic liver disease was alcohol. The baicalensis), Gan Cao (Radix glycyrrhizae), Ze Xie (Alisma authors found that encephalopathy, bilirubin, blood urea, orientalis), and Chuan Xiong (Rhizoma ligustici). Most of the lactate, and international normalized ratio predicted mortality patients used at least 10 different ingredients, and the dura- 28 29 in drug-induced ACLF. Zhu and coworkers studied predic- tion of use ranged from a few months to 3 years. tors of poor outcomes in 488 patients with herb-induced liver Lei et al.32 performed a systematic review of case reports injury, at a single tertiary center in China. The hepatocellular and series (450 cases) on liver damage associated with Poly- type of liver injury was most common, with median age gonum multiflorum (PM; known as He Shou Wu). In TCM, raw 45 years. Eighty-two percent of patients had complete nor- PM is used for detoxification, to treat skin infections, to malization of clinical and biochemical abnormalities, while prevent malaria, and to improve gut motility, while the pro- fourteen percent developed chronic liver disease, and cirrho- cessed form (known as P. multiflorum praeparata or PMP) is sis developed in 7.6%. Approximately 4% of patients in this utilized for improving liver and kidney health, as a hematinic, series died without provision for liver transplantation. Argu- for blackening hair, and as a hypolipidemic agent. Even ably, it is possible that CAM-DILI is higher in the Chinese though some pharmacological studies have shown the antiin- patient population (probably due to better surveillance and flammatory and hypolipidemic nature of PM, no standard, integrated reporting) but outcomes are worse in those devel- high-quality, controlled and rigorously-designed human oping CAM-DILI in the Indian subcontinent. Furthermore, the trials exist. Symptomatic liver injury occurred approximately presence of chronic liver disease or cirrhosis significantly 30-days after intake of PM with prominence of jaundice. Two increases the risk of death in patients who develop ACLF. patients underwent liver transplantation and another seven patients died due to liver failure. CAMs associated with liver injury Chinese authors, in general, have considered PMP relatively safe, with raw form PM extracts showing higher Knowledge on DILI associated with CAMs has mostly come potential for liver injury. It was suggested that the ethanol through three sources: case reports, case series and descrip- extract produced hepatic lesions more easily than that of tions as part of large retrospective or prospective studies on water formulation, and the decreasing order of toxic potential DILI. In patients with CAM-DILI, the type of liver injury can be was described as PM ethanol extract > PM water extract > classified into hepatocellular, mixed or cholestatic types, PMP ethanol extract > PMP water extract. However, other based on the R value (https://www.mdcalc.com/r-factor- authors33 have mentioned the toxicity potential differently liver-injury; in which a value >5 is considered hepatocellular, (PM water extract > PM acetone extract > PMP acetone <2 as cholestatic, and 2–5 as mixed type of liver injury). In extract), which has added to the confusion in identification the majority of published data on CAM-DILI, the hepatocel- of complex toxicity profiles associated with whole herb lular pattern of injury is most common. Repositories of DILI preparations. are useful sources for identifying causative agents. These also Recently Jing et al.34 performed a case-control study on provide important information on DILI associated with medical treatments, showing regional differences, such as the association between the concurrence of pre-existing for CAMs. Most of the published data on CAM-DILI comprise chronic liver disease and prognosis in patients with PM- case reports and small case series, with high-quality, large induced liver injury. Among these, 22.8% had pre-existing cohort studies available for certain CAMs only, the latter chronic liver disease, with the majority having (52%) alco- discussed succinctly below. holic liver cirrhosis. On comparing PM-DILI with and without Zhao and colleagues30 performed a retrospective analysis chronic liver disease, the latter group showed higher mortal- of patients diagnosed with ALF in seven hospitals in different ity (0.9% vs. 9.1%) and more prolonged course of liver injury areas of China, from January 2007 to December 2012. Out of (12.5% vs. 30.3%). Four patients (2.8%) in the entire cohort 177 patients, 63.28% died. The common causes for liver developed ACLF. The concurrent presence of underlying failure were drug toxicity (43.50%), indeterminate etiology chronic liver disease was found to be an independent risk (29.38%), and acute viral hepatitis (11.30%). Among those factor for both of chronicity and mortality in those with PM- 17 who developed DILI-ALF, TCM was the cause in 39%. The DILI. In the Woo et al. study from Korea, herbal medicines- authors found that age, baseline evidence of overt hepatic related DILI were seen in only <1% of cases reviewed; encephalopathy, coagulation failure, and blood ammonia however, the common causative herbal agents associated were independently associated with death due to DILI-ALF. with hepatotoxicity were Ephedrae Herba and Scutellariae The same group31 later published separately on the 30 Radix (skullcap). In the study by Ou and colleagues, among cases of ALF due to TCM, describing the patient characteris- 36% of patients with CAM-DILI, polyherbal mixtures, Radix tics and clinical outcomes in detail. The common reasons for polygoni multiflora, Panax pseudo-ginseng, Tripterygium

266 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 Philips C.A. et al: CAM-related DILI in Asia wilfordii, saffron and shenbao mixture were found to be 52 in the USA; however, it remains heavily utilized among common causes for liver injury, in descending order of patients in Asian countries currently. incidence.23 In the study by Philips et al.,26 almost 40% of CAMs Thyagarajan and coworkers35 described more than 300 retrieved from patients with CAM-DILI were proprietary preparations for the treatment of jaundice and chronic liver Indian Ayurvedic herbal medicines in the form of bottled diseases in Indian systems of medicine, using more than 87 branded decoctions, pills, tablets, and powders. One-third of Indian medicinal plants. However, among them, only four ter- the CAMs were unlabeled polyherbal mixtures, prescribed by restrial plants had undergone good quality studies adhering to traditional Ayurvedic healers. On chemical and toxicology the internationally acceptable scientific protocols, which analysis, lead was detected in 73% (maximum detected: remains so, even now, almost 16 years later. Reports on 7318.1 mg/kg), mercury in 64% (751.5 mg/kg), and Indian Ayurvedic medicine-related hepatotoxicity are sparse arsenic in 58% (111 mg/kg). Apart from heavy metals, multi- in the literature. Centella asiatica (Asiatic pennywort) or Gotu ple volatile organic compounds and industry grade chemicals Kola in Ayurvedic medicine has been reported to produce ALF, were also detected in the more than 70% of the samples granulomatous hepatitis and severe necro-inflammatory liver analyzed. However, in the study on CAM-DILI-related ACLF, injury (Fig. 2).36,37 Teschke and Bahre38 presented the case the most common CAM utilized by the cirrhosis patient pop- of a 64-year-old female patient on Ayurvedic medications for ulation was that of Ayurveda (77%), of which, the majority treatment of vitiligo. The implicated drugs included Bakuchi were traditional indigenous herbal medicines prescribed by tablets containing extracts from Psoralea corylifolia leaves, local healers for diabetes mellitus management. These Khadin tablets containing extracts from Acacia catechu included aloe vera, passionflower and guava leaf extracts, leaves, Brahmi tablets containing Eclipta alba, and Usheer Phyllanthus niruri, Polygonum aviculare (wire weed), tea prepared from Vetivexia zizaniodis. Malabar nut tree leaf extracts, Datura stramonium (devil’s Fleig et al.39 performed a randomized controlled trial on weed) and Chlorophytum borivilianum (safed musli).26,27 the Ayurvedic medicine Liv 52 in a large cohort of patients Zhu et al.40 analyzed the causes, clinical, laboratory and with alcoholic liver disease. They found that Liv 52 did not pathological features, and outcomes of DILI in 69 children (up improve liver-related events or survival in treated patients, to 14 years of age) and compared the differences between and that its use in patients with Child C cirrhosis led to TCM and Western medicine-related liver injury. Autoantibod- increased mortality. This study led to nationwide ban of Liv ies were detected in 32 children with DILI, in their series.

Fig. 2. Ayurvedic medication prescribed by a registered practitioner for a patient with vague digestive complaints. Multiple medicines (A) were taken over a period of 1 month, leading to severe bilateral leg edema and ascites; transjugular liver biopsy revealed features of extensive lymphocytic and eosinophilic infiltration of sinusoids associated with sinusoidal dilatation (B, hematoxylin and eosin stain, 403); giant cell transformation of hepatocytes (C, H&E stain, 4003); multiple scattered granulomas in the lobular and portal areas (D and E, H&E stain, 4003) suggestive of granulomatous hepatitis with sinusoidal obstruction. Taken together, the Ayurvedic medications had close to 50 different types of herbal and nonherbal components. Abbreviation: H&E, hematoxylin and eosin.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 267 Philips C.A. et al: CAM-related DILI in Asia

Western medicine-related DILI was noted in 56%, TCM use in that TCM–CAM-related DILI resulted in high mortality in 22%, and combinations in 22%. The major implicated herbs chronic hepatitis B patients and that prospective randomized- were Ephedra sinica and PM, followed by Andrographis pan- controlled trials with the same stringent criteria as Western iculate, Xanthium sibiricum, Mentha haplocalyx, Smilax medicine clinical trials are required for CAMs for appropriate glabra, and Tripterygium wilfordii. Chronicity was noted documentation of efficacies and safety before marketing. In more in the Western medicine group, while development of similar lines, Philips and coworkers demonstrated that use of ALF and death rates were higher in children with DILI due to AHMs could lead to reactivation of chronic hepatitis B akin to CAM. that seen with chemotherapy, corticosteroid or biological use in In the recent study by Zhu and colleagues29 on outcomes predisposed patients, due to presence of unknown and unre- in 488 patients with herb-induced liver injury, the major searched immunomodulating components in whole herbs implicated agent was polyherbal TCMs which contained in (Fig. 3).44 common, Fructus psoraleae, Gynura segetum and PM, apart A compilation of important CAMs of Asian origin implicated from a multitude of other components. However, these three in DILI and utilized in Asian countries is shown in herbs as single agents were also responsible for ALF leading Table 1.5,45–48 to death or liver transplantation in three patients. Similarly, those herbs, as single agents, associated with chronic herb- induced liver injury in the study patients, included Radix poly- Histopathological characteristics of CAMs associated goni multiflora, Tripterygium wilfordii, Semen cassiae, and with liver injury Smilax scobinicaulis. Lee et al.41 found that Chinese herbal medicines such as Histopathology of CAM-related DILI is nonspecific and Xiao-Chai-Hu-Tang, Long-Dan-Xie-Gan-Tang, or others, con- includes acute or chronic hepatitis, cholestatic liver disease taining more than 19 gram of radix bupleuri (made from dried (including vanishing bile duct syndrome), vascular liver root of Bupleurum falcatum L. or B. falcatum L. var. scorzo- disease (mostly sinusoidal obstruction syndrome), autoim- nerifolium) prescribed for management of chronic hepatitis mune features, granulomatous hepatitis, giant cell hepatitis (mainly hepatitis B virus-related) could lead to severe liver and the syndrome of ALF with massive hepatic necrosis. Most injury. The relative risk of developing liver injury by subse- of the histopathological descriptions on CAM-DILI are based quent addition of 19 g of radix bupleuri was 2.19. Jakkula on isolated case reports, and specific findings related to CAM et al.42 found that a regimen of Chinese herbal medicines on liver biopsy remain difficult to describe in view of multiple did not improve quality of life, liver chemistry results, or CAM use at a time, each with multiple ingredients in a single viral load in a cohort of patients with hepatitis C virus, in a patient. Liver histopathological examination has shown that randomized trial. CAM-DILI, especially related to TCM and Ayurvedic herbals, Yuen et al.43 found that in approximately 16% of patients mimics severe nonalcoholic steatohepatitis, alcoholic hepati- with chronic hepatitis B virus-related liver disease, acute liver tis, acute hepatitis with varying severity of necrosis, hepato- injury was attributable to traditional Chinese medicine, with cellular and canalicular cholestatic features, veno-occlusive all patients showing clinical presentation of acute exacerba- disease and cirrhosis, associated pigment changes, and tion of chronic hepatitis B. There were two deaths and one neutrophilic, lymphoplasmacytic, eosinophilic or mixed patient underwent liver transplantation. The identified hepa- inflammatory patterns affecting lobular, portal and interface totoxic components included Polygonum multiflorum Thunb, areas.49–53 Cassia obtusifolia L, Melia toosendan Sieb., Rheum palmatum Large studies on descriptive histopathology associated L., Scolopendra subspinipes mutilans L, Alisma orientale with CAM-related DILI are infrequent in the published liter- Juzepe, Glycyrrhiza uralensis Fisch., and Mentha haplocalyx ature. Lai et al.21 in their study on DILI in 138 patients found Briq, while one TCM medicine was adulterated with a highly that the most common cause for liver injury was CAM, mainly hepatotoxic agent, N-nitrosofenfluramine. The authors concluded those of Chinese herbal medicine. The liver biopsy features

Fig. 3. Ayurvedic medicines prescribed by a registered practitioner for a patient with chronic hepatitis B-related cirrhosis with low viral load. Multiple medications (A), most of which unlabeled and with directions for use, written over the bottles themselves, were handed over to the patient. After consuming the medicines for almost 1 month, the patient developed cholestatic jaundice, ascites and hepatic encephalopathy within 2 weeks, suggestive of acute on chronic liver failure. Transjugular liver biopsy revealed severe mixed portal inflammation with ductular reaction (B, H&E, 403) associated with extensive periportal necrosis and cholestasis with cholangitis (C, H&E, 4003). He subsequently died on the liver transplant wait-list. Abbreviation: H&E, hematoxylin and eosin.

268 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 Philips C.A. et al: CAM-related DILI in Asia

Table 1. Complementary and alternative medicines used in Asian countries implicated in drug-induced liver injury

No Region Name of complementary and alternative drug Salient features

1 Pan-Asian Aloe perfoliata var. vera (aloe vera) · Idiosyncratic hepatotoxicity · Rechallenge proven liver injury · Spotty necrosis, hepatocyte ballooning, portal and lobular inflammation on liver histopathology · Causes toxic syndrome with gastrointestinal, bone marrow, 2 China Podophyllum pleianthum (Ba Jiao Lian or neurological and hepatic injury Chinese may apple) · Toxicity due to glycoside podophyllotoxin 3 China/Taiwan Breynia vitis-idaea (Chi R Yun or Large calyx · Known as ‘7-days of dizziness’ · Neurological, liver and respiratory toxicity breynia) · Usually part of herbal and dietary supplements 4 China Jin Bu Huan · Used as sedative and analgesic · Associated with acute and chronic hepatitis Polyherbal formulation containing Lycopodium · Implicated agent is the alkaloid 1-tetrahydropalmatine serratum (club moss), Panax pseudo ginseng (ginseng) and Polygala chinensis (milkwort) 5 China Ephedra sinica (Ma Huang or yellow hemp) · Idiosyncratic hepatotoxicity · Associated cardiovascular toxicity · Acute and chronic hepatitis, vanishing bile duct syndrome · Reported to cause acute liver failure · Used to treat liver fibrosis and prevent hepatocellular 6 China/Japan Sho-saiko-to or Xiao Chai Hu Tang (polyherbal) carcinoma Bupleurum chinense (thorowax), Scutellaria · Idiosyncratic, acute hepatitis, self-limited with drug withdrawal baicalensis (Chinese skullcap), Pinellia ternate · Scutellaria (skullcap) most likely hepatotoxic component (crow-dipper), Zingiber officinale (ginger), Glycyrrhiza glabra (licorice), Codonopsis pilosula (Poor man’s ginseng), Ziziphus jujube (Chinese date) 7 China Polygonum multiflorum Thunb. (Shou Wu Pian · Herbal drug on which largest series on liver toxicity published · Mixed or hepatocellular type of injury or Chinese climbing knotweed) · Reported cases of fatality · Toxicity due to anthraquinones, such as emodin and the stilbene glycoside, tetrahydroxy stilbene-glucopyranoside 8 India Garcinia cambogia (Malabar tamarind) · Common component of herbal weight loss products · Linked to clinically apparent acute liver injury that can be severe and even fatal · Associated with serotonin syndrome and rhabdomyolysis · Hydroxy citric acid, the active component suppresses appetite and is hepatotoxic 9 China/Korea Panax ginseng (ginseng) · No direct hepatotoxic potential · Affects cytochrome P450 and 3A4 enzyme system leading to drug-drug interactions · Ginseng potentiating prescription drug related liver injury with concomitant use which reversed after stopping ginseng reported 10 Pan-Asian Camellia sinensis (green tea) · Common constituent of weight loss dietary supplements · Dose-dependent liver injury · Liver injury typically within 3 months, with latency to onset of symptoms ranging from 10 days to 7 months · Autoimmune hepatitis phenomenon described · Fatal liver failure reported · Epigallocatechin-3-gallate (known as EGCG) implicated component in liver injury · A safe intake level of 338 mg EGCG/day for adults 11 China/India Piper methysticum (kava, intoxicating pepper) · Used as anxiolytics · Contains kavapyrones (kavalactones) with effects similar to alcohol · Can lead to severe hepatitis leading to fulminant hepatic failure · Idiosyncratic or immune-allergic type of liver injury noted in the published literature 12 South-East Mitragyna speciose (kratom) · Sedative, anxiolytic · Chronic use associated with acute liver injury Asia · Onset usually from 2 to 8 weeks of starting regular use · Liver injury is cholestatic or mixed · May be accompanied by acute renal failure and bone marrow toxicity · For hypertension/gastrointestinal complaints as herbal extract/ 13 China/Japan/ Viscum album and Loranthus ferrugineus infusion or tea preparation India (mistletoe) · Found to cause chronic hepatitis on long-term consumption 14 India/China Morinda citrifolia (Indian mulberry/noni/Ba Ji · Idiosyncratic acute liver injury · Hepatocellular pattern Tian) · Can lead to acute liver failure · Associated with high levels of autoantibodies

(continued)

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 269 Philips C.A. et al: CAM-related DILI in Asia

Table 1. (continued ) No Region Name of complementary and alternative drug Salient features

15 India Cassia angustifolio (senna) · Commonly used laxative · Chronic use leads to transient acute hepatocellular injury · Re-exposure leading to recurrence of liver injury in published literature · Acute hepatomyoencephalopathic syndrome, akin to Reyes 16 India Cassia occidentalis (coffee senna) syndrome · Outbreaks reported from India during flowering season in children consuming beans · Used in Ayurveda for cough/colds and digestion ‘imbalance’ in children and adults · Common components of dietary supplements 17 China/India Scutellaria lateriflora (skullcap) Utilized for treatment of arthralgia/arthritis ‘ ’ · Known as huang qin in China · Liver injury hepatocellular type Component of anti-inflammatory proprietary · Can cause acute liver failure drug marketed as Univestin (Unigen®) Macvestin (Macleods®)/Cardivestin or Ostivestin (Zydus Cadilla®) 18 India/China/ Arthrospira platensis (spirulina, blue-green · Not proven to be efficacious in any medical condition · Contamination with other blue-green algae that produce Korea algae; cyanobacteria) hepatotoxins (microcystins) is common · Contamination responsible for acute hepatocellular type of liver injury · Isolated case reports of liver injury due to turmeric-containing 19 India Curcuma longa (curcumin/turmeric) dietary supplements · Idiosyncratic liver injury mostly due to herb-herb interaction or herb-drug interaction 20 India/South- Azadirachta indica (margosa oil or Indian neem · Reports of acute severe metabolic acidosis and liver failure · Injury and presentation similar to Reye syndrome (in children) East Asia oil) · Causes microvesicular steatosis (Malaysia) · Causes cellular mitochondrial injury · Active components include pentacyclic triterpenic 21 India Centella asiatica (Asiatic pennywort or Gotu saponosides (asiaticoside, madecassoside) Kola) · Liver histology features include granulomatous hepatitis, marked necroinflammatory activity, chronic hepatitis 22 India Liv 52 and Liv 52 DS (proprietary Ayurvedic · Marketed specifically for the treatment of liver diseases · In a randomized controlled trial, high mortality noted in drug manufactured by Himalaya Drug patients with Child C disease Company®) · Withdrawal of the drug from the USA market in the early Capparis spinosa, Cichorium intybus, Mandur 2000s. Bhasma, Solanum nigrum, Terminalia arjuna, Cassia occidentalis, Achillea millefolium, Tamarix gallica Processed in extracts of Eclipta alba, Phyllanthus amarus, Boerhaavia diffusa, Tinospora codifolia, Raphanus sativus, Emblica officinalis, Plumbago zeylanica, Embelia ribes, Terminalia chebula, Fumaria officinalis

included macro- and microvesicular steatosis, cholestasis, Another recent study from India on Ayurvedic medicine- hepatocyte apoptosis, epithelioid granulomas, eosinophilic, related liver injury among the general population, was the neutrophilic and lymphoplasmacytic infiltration, and iron dep- first to detail liver histopathological characteristics linked to osition. The inflammation, necrosis and Ishak fibrosis score in patient outcomes. Chronic hepatitis was the most common patients with cholestatic and mixed type of liver injury were type of inflammation, seen in 81.5%. Portal-based inflamma- higher than those with predominantly hepatocellular type of tion predominated (89%), and necrosis was noted in almost injury. 56% of patients with AHM-related injury. Fibrosis was seen Zhu et al.41 described liver biopsy of 55 children in whom in 77.8% and cholestasis in 67%. The presence of necrosis the major cause for DILI was TCM. They found that liver cell and steatosis correlated positively and negatively with degeneration, necrosis (bridging, confluent and submassive mortality respectively. The same study group also identified type), and lobular inflammation were common findings. and described liver histopathology findings among patients Approximately 29% had varying degrees of interface hepati- with cirrhosis consuming Ayurvedic medications and devel- tis, while hepatocellular and canalicular cholestasis were oping ACLF. Out of 30 patients with Roussel Uclaf Causality observed in 27%. Almost half of the liver biopsies showed Assessment Scoring Method scores suggestive of probable eosinophil infiltration, while in a quarter of the children, andpossibleCAM-DILI-relatedACLF,12consentedtotrans- after complete normalization of liver biochemistries, histolog- jugular liver biopsy. Pertinent findings included portal- ical features of moderate portal inflammation, interface hep- based moderate to severe mixed inflammation, neutrophilic atitis and fibrosis prevailed, suggesting chronicity. predominance, eosinophilic infiltration of portal tracts,

270 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 Philips C.A. et al: CAM-related DILI in Asia

Table 2. Published large patient series on liver histopathology related to CAM-related liver injury

Patient AuthorRef Year population Most common CAM Number Salient findings

Lai 2012 Adults, Traditional Chinese herbal 74/138 Macro and microvesicular steatosis, cholestatic et al.21 general medicines hepatitis, eosinophilic and neutrophilic population predominant portal and lobular inflammation and severe grades of hepatic siderosis Higher grades of inflammation, necrosis and Ishak fibrosis score in cholestatic and mixed type of liver injury No separate analysis on CAM-DILI-related histology and outcomes Zhu 2015 Children, Traditional and proprietary 15/69 Severe grades of apoptosis and necrosis, et al.40 general Chinese medicine lobular inflammation, varying degrees of population interface hepatitis Lesser incidence of hepatocellular and canalicular cholestasis Evidence for chronic DILI noted in children on liver histology, after normalization of biochemical parameters Separate analysis on CAM-DILI subgroup not provided Philips 2018 Adults, Proprietary and traditional 33 Presence of necrosis linked to mortality et al.26 general Ayurveda and herbal Presence of massive and submassive necrosis population medicines worst prognosis Predominant steatotic type of liver histopathology favorable outcome Arsenic and mercury in CAM associated with poor survival First to provide linkage on histopathology related outcomes in CAM-DILI and chemical and toxicology analysis in CAM related to DILI Philips 2019 Adult Traditional Ayurveda and 30 First to discuss histopathology in CAM-DILI et al.27 patients herbal medicines leading to ACLF with Portal-based moderate to severe mixed cirrhosis inflammation, neutrophilic/eosinophilic predominance, varying grades of hepatocellular and canalicular cholestasis with cholangitis and extensive siderosis notable

Abbreviations: ACLF, acute on chronic liver failure; CAM, complementary and alternative medicine; DILI, drug-induced liver injury. cholangitis, intracanalicular and hepatocellular cholestasis, classified as foods and dietary supplements and 2) considered varying grades of ductular reaction, and extensive safe as per ancient practices forming part of a separate siderosis.26,27 pharmacopeia deep-rooted in religious, cultural and social In summary (Table 2), liver biopsy is not mandatory for acceptance, and are thus exempt from rigorous clinical confirmation of diagnosis of CAM-DILI since histological fea- testing and quality control. For example, in India, CAM tures are nonspecific and can mimic any other acute or prepared and marketed as per classical Ayurvedic literature chronic liver disease etiology. However, the patterns and does not require preclinical and phased-clinical trials based on severity of injury on liver biopsy helps prognosticate patient the fact that these ‘drug recipes’ are knowledge passed down outcomes in specific situations and should be performed if through thousands of years of practice and as such, are taken adequate expertise in both interventional radiology and liver as granted to be safe and efficacious. histopathology is available. This would also help in identifica- Utilization of modern research tools in such instances can tion of patients who require liver transplantation at the outset help identify toxic components and beneficial compounds for for early listing, so as to prevent certain futility on medical multiple disease conditions. Much of current CAM-based management. research studies are of poor quality, inclusive of paltry hypotheses, suffering from inadequate methodology, Current perspectives and future directions meddled with impecunious statistical analysis, and ultimately lost to predatory or low-grade nonindexed journals, being Problems associated with CAM use in many countries stem further showcased as clinical data to an unknowing patient from the fact that these practices and products are 1) mostly population for better for-profit sale of the product.54,55

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 271 Philips C.A. et al: CAM-related DILI in Asia

Fig. 4. Herbal medicine and decoctions without any identifiable components (A) prescribed by a TAH, for management of complications of diabetes mellitus in a middle-aged man. After 3 weeks of medicine use, the patient presented with severe cholestatic liver injury. He went back to the TAH for management of cholestatic jaundice and was advised to continue the same drugs at half dose with dietary restrictions (only rice porridge with boiled vegetables, as required). After 1 week, severe clinical worsening with malnutrition ensued, and the patient was brought to the emergency unit. Transjugular liver biopsy revealed hepatic lobular distortion due to severe mixed inflammation at the portal and lobular regions (B, H&E, 403), severe ballooning of hepatocytes with apoptosis (C, H&E, 4003), and periportal and perivenular necrosis with severe hepatocellular and canalicular cholestasis (D, H&E stain, 1003). The patient recovered completely with drug withdrawal and aggressive supportive care, including eight sessions of plasma-exchange. Abbreviations: H&E, hematoxylin and eosin; TAH, traditional Ayurvedic healer.

A prime example is of the Ayurvedic antimalarial drug AYUSH- inserted into the product package in conventional medicine, 64, on which a non-controlled clinical trial was published (in which is almost nonexistent or immensely vague in propriet- the year 1981; without ethical statements and disclosures; ary and nonproprietary CAM. alarmingly showcasing complete confidential patient details First and foremost, treating primary physicians and spe- in Table 1 of the published study) in a journal (Journal of cialists who care for liver disease patients must realize and Research in Ayurveda and Siddha, Vol 2, No 4, pages 309– accept the fact that use of CAM is here to stay and will almost 26), which is currently not traceable to any valid scientific certainly rise among the general and patient population. data indexing repositories. Properly conducted clinical trials Secondly, they have to realize that CAM is not without side on the same drug decades later showed absence of blood effects, and liver injury due to CAM is a real concern. The schizonticide activity on preclinical testing, with efficacy physicians must educate the patient and the family, who will <50% in curing malaria in humans when compared to stand- in turn educate the locality and thereby the community ard drug chloroquine.56,57 regarding CAM and associated adverse events. Industry, The need for evidence of quality, quality tests, and government agencies, and practitioners of CAM need to production standards that are currently without homogeneity understand the importance of bench-to-bedside work and and control in CAM has to be addressed by the industry and should be motivated to perform studies that are rich in appropriate government agencies. An important example in scientific methodology. This ensures 1) identifying efficacious this aspect is CAM prescribed by indigenous and traditional and toxic components associated with CAM and 2) integration health practitioners, the majority being uncertified and not of a refined logic in CAM that can ultimately be used along licensed but enjoying public and social acceptance for provid- with conventional medicine. ing healthcare, and has been associated with severe DILI as This aspect is beautifully undertaken in China where an well as death due to progressive liver failure (Fig. 4). Practi- integrated guideline approach to diagnosis and management tioners and producers of CAM must be trained and regulated of CAM-DILI was recently published.58 The guidelines specify to provide exhaustive information, such as component evaluation, identification and treatment of CAM-DILI in China descriptions, indications, precautions, usage directions, side according to the strength of evidence through a structured effects, storage details, and regulatory information regarding diagnostic workflow. It also recommends either identifying the drugs. This information is usually provided on a leaflet the species of Chinese or herbal medicine or excluding

272 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 Philips C.A. et al: CAM-related DILI in Asia adulterations and toxin contaminants associated with DILI, to [4] Agarwal V. Complementary and alternative medicine provider knowledge dis- improve the level of evidence for a clinical diagnosis. This course on holistic health. Front Commun 2018;3:15. doi: 10.3389/fcomm. 2018.00015. acceptance, as seen in the Chinese Guidelines, for improving [5] Amadi CN, Orisakwe OE. Herb-induced liver injuries in developing nations: patient outcomes through an integrative medical approach is An update. Toxics 2018;6:E24. doi: 10.3390/toxics6020024. an unmet need in countries entrenched in promotion and [6] Harris PE, Cooper KL, Relton C, Thomas KJ. Prevalence of complementary practice of CAM in Asia. and alternative medicine (CAM) use by the general population: a systematic review and update. Int J Clin Pract 2012;66:924–939. doi: 10.1111/j.1742- 1241.2012.02945.x. Conclusions [7] Wai CT. Presentation of drug-induced liver injury in Singapore. Singapore Med J 2006;47:116–120. CAM use is widely prevalent in Asia and is associated with, [8] Wai CT, Tan BH, Chan CL, Sutedja DS, Lee YM, Khor C, et al. Drug-induced – among other adverse effects, hepatotoxicity. Both proprietary liver injury at an Asian center: a prospective study. Liver Int 2007;27:465 474. doi: 10.1111/j.1478-3231.2007.01461.x. as well as nonproprietary or traditional CAMs have been [9] Teo DC, Ng PS, Tan SH, Lim AT, Toh DS, Chan SY, et al. Drug-induced liver implicated in hepatotoxicity. Acute hepatocellular pattern of injury associated with Complementary and Alternative Medicine: a review of liver injury is the most common type of liver injury seen, and adverse event reports in an Asian community from 2009 to 2014. BMC Com- the spectrum of liver-related adverse events range from plement Altern Med 2016;16:192. doi: 10.1186/s12906-016-1168-z. [10] Bhalerao S, Munshi R, Tilve P, Kumbhar D. A survey of the labeling informa- simple elevation of liver enzymes to the very serious ALF tion provided for ayurvedic drugs marketed in India. Int J Ayurveda Res and ACLF, which may, at times, require liver transplant. 2010;1:220–222. doi: 10.4103/0974-7788.76785. CAM-related liver injury is one among the major causes for [11] Devarbhavi H. Ayurvedic and herbal medicine-induced liver injury: It is time hepatotoxicity, including ALF and ACLF worldwide, with high to wake up and take notice. Indian J Gastroenterol 2018;37:5–7. doi: 10. 1007/s12664-018-0820-6. incidence among Asian countries. Patient outcomes associ- [12] Takikawa H, Murata Y, Horiike N, Fukui H, Onji M. Drug-induced liver injury in ated with CAM-DILI are generally poor, with very high mortal- Japan: An analysis of 1676 cases between 1997 and 2006. Hepatol Res ity rates in those with chronic liver disease. Stringent 2009;39:427–431. doi: 10.1111/j.1872-034X.2008.00486.x. regulations, at par with that of conventional modern medi- [13] Liem KS, Yim C, Ying TD, Zanjir WR, Fung S, Wong DK, et al. Prevalence and cine, are required, and may help improve safety of patients predictors of complementary and alternative medicine modalities in patients with chronic hepatitis B. Liver Int 2019;39:1418–1427. doi: 10.1111/liv. seeking CAM for their health needs. Regional surveillance 14105. including postmarketing analysis from government agencies [14] Suk KT, Kim DJ, Kim CH, Park SH, Yoon JH, Kim YS, et al. A prospective associated with drug regulation and control in tandem with nationwide study of drug-induced liver injury in Korea. Am J Gastroenterol – national as well as regional level hepatology societies are 2012;107:1380 1387. doi: 10.1038/ajg.2012.138. [15] Cho JH, Oh DS, Hong SH, Ko H, Lee NH, Park SE, et al. A nationwide study of important for understanding the true prevalence of DILI the incidence rate of herb-induced liver injury in Korea. Arch Toxicol 2017; associated with CAM. An integrated approach used by practi- 91:4009–4015. doi: 10.1007/s00204-017-2007-9. tioners combining conventional and traditional medicine to [16] Wang GQ, Deng YQ, Hou FQ. Overview of drug-induced liver injury in China. identify safety and efficacy of CAMs is an unmet need in most Clin Liver Dis (Hoboken) 2014;4:26–29. doi: 10.1002/cld.386. [17] Woo HJ, Kim HY, Choi ES, Cho YH, Kim Y, Lee JH, et al. Drug-induced liver of the Asian countries. Endorsement of scientific methodology injury: A 2-year retrospective study of 1169 hospitalized patients in a single with good quality preclinical and clinical trials and abolish- medical center. Phytomedicine 2015;22:1201–1205. doi: 10.1016/j. ment of unhealthy publication practices is an area that needs phymed.2015.10.002. immediate attention in CAM practice. Such holistic standard [18] Hou FQ, Zeng Z, Wang GQ. Hospital admissions for drug-induced liver injury: – science-based approaches could help ameliorate liver disease clinical features, therapy, and outcomes. Cell Biochem Biophys 2012;64:77 83. doi: 10.1007/s12013-012-9373-y. burden in the general and patient population. [19] Shen T, Liu Y, Shang J, Xie Q, Li J, Yan M, et al. Incidence and Etiology of Drug-Induced Liver Injury in Mainland China. Gastroenterology 2019;156: 2230–2241.e11. doi: 10.1053/j.gastro.2019.02.002. Conflict of interest [20] Zhang P, Ye Y, Yang X, Jiao Y. Systematic review on Chinese herbal medicine induced liver injury. Evid Based Complement Alternat Med 2016;2016: 3560812. doi: 10.1155/2016/3560812. The authors have no conflict of interests related to this [21] Lai RT, Wang H, Gui HL, Ye MZ, Dai WJ, Xiang XG, et al. Clinical and patho- publication. logical features in 138 cases of drug-induced liver injury. Zhonghua Gan Zang Bing Za Zhi 2012;20:185–189. doi: 10.3760/cma.j.issn.1007-3418. 2012.03.009. [22] Wang R, Qi X, Yoshida EM, Méndez-Sánchez N, Teschke R, Sun M, et al. Author contributions Clinical characteristics and outcomes of traditional Chinese medicine- induced liver injury: a systematic review. Expert Rev Gastroenterol Hepatol Designed the research study (CAP, PA), collected the data 2018;12:425–434. doi: 10.1080/17474124.2018.1427581. (CAP, SR, DM), wrote the manuscript (CAP, PA, DM, PKY) and [23] Ou P, Chen Y, Li B, Zhang M, Liu X, Li F, et al. Causes, clinical features and outcomes of drug-induced liver injury in hospitalized patients in a Chinese critically reviewed and revised the manuscript (CAP, PA, SR, tertiary care hospital. Springerplus 2015;4:802. doi: 10.1186/s40064-015- DM, PKY). 1600-8. [24] Udayakumar N, Subramaniam K, Umashankar L, Verghese J, Jayanthi V. Predictors of mortality in hepatic encephalopathy in acute and chronic liver References disease: a preliminary observation. J Clin Gastroenterol 2007;41:922–926. doi: 10.1097/01.mcg.0000225639.45157.ee. [1] PDQ Integrative, Alternative, and Complementary Therapies Editorial Board. [25] Devarbhavi H, Dierkhising R, Kremers WK, Sandeep MS, Karanth D, Adarsh Topics in Integrative, Alternative, and Complementary Therapies (PDQ®): CK. Single-center experience with drug-induced liver injury from India: Health Professional Version. In: PDQ Cancer Information Summaries [Inter- causes, outcome, prognosis, and predictors of mortality. Am J Gastroenterol net]. Bethesda (MD): National Cancer Institute (US); 2002-. Available from: 2010;105:2396–2404. doi: 10.1038/ajg.2010.287. https://www.ncbi.nlm.nih.gov/books/NBK126745. [26] Philips CA, Paramaguru R, Joy AK, Antony KL, Augustine P. Clinical outcomes, [2] World Health Organization. WHO traditional Medicine strategy 2014–2023. histopathological patterns, and chemical analysis of Ayurveda and herbal Available from: http://www.searo.who.int/entity/health_situation_trends/- medicine associated with severe liver injury-A single-center experience who_trm_strategy_2014-2023.pdf?ua=1. from southern India. Indian J Gastroenterol 2018;37:9–17. doi: 10. [3] World Health Organization. WHO global report on traditional and complemen- 1007/s12664-017-0815-8. tary medicine 2019. Available from: https://apps.who.int/iris/bitstream/- [27] Philips CA, Paramaguru R, Augustine P, Rajesh S, Ahamed R, George T, et al. handle/10665/312342/9789241515436-eng.pdf?ua=1. A single-center experience on outcomes of complementary and alternative

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 273 Philips C.A. et al: CAM-related DILI in Asia

medicine use among patients with cirrhosis. Hepatol Commun 2019;3: of symptomatic hepatitis C. Arch Intern Med 2004;164:1341–1346. doi: 1001–1012. doi: 10.1002/hep4.1355. 10.1001/archinte.164.12.1341. [28] Devarbhavi H, Choudhury AK, Sharma MK, Maiwall R, Al Mahtab M, Rahman [43] Yuen MF, Tam S, Fung J, Wong DK, Wong BC, Lai CL. Traditional Chinese S, et al. Drug-induced acute-on-chronic liver failure in Asian patients. Am J medicine causing hepatotoxicity in patients with chronic hepatitis B infec- Gastroenterol 2019;114:929–937. doi: 10.14309/ajg.0000000000000201. tion: a 1-year prospective study. Aliment Pharmacol Ther 2006;24:1179– [29] Zhu Y, Niu M, Wang JB, Wang RL, Li JY, Ma YQ, et al. Predictors of poor out- 1186. doi: 10.1111/j.1365-2036.2006.03111.x. comes in 488 patients with herb-induced liver injury. Turk J Gastroenterol [44] Philips CA, Augustine P, Padsalgi G. Herbal medicines and reactivation of – 2019;30:47 58. doi: 10.5152/tjg.2018.17847. chronic hepatitis B virus infection, Hepat Mon 2018;18:e81000. doi: 10. [30] Zhao P, Wang C, Liu W, Chen G, Liu X, Wang X, et al. Causes and outcomes of 5812/hepatmon.81000. acute liver failure in China. PLoS One 2013;8:e80991. doi: 10.1371/journal. [45] Jing J, Teschke R. Traditional Chinese medicine and herb-induced liver injury: pone.0080991. Comparison with drug-induced liver injury. J Clin Transl Hepatol 2018;6:57– [31] Zhao P, Wang C, Liu W, Wang F. Acute liver failure associated with traditional 68. doi: 10.14218/JCTH.2017.00033. Chinese medicine: report of 30 cases from seven tertiary hospitals in China. [46] Frenzel C, Teschke R. Herbal hepatotoxicity: Clinical characteristics and listing Crit Care Med 2014;42:e296–e299. doi: 10.1097/CCM.0000000000000136. compilation. Int J Mol Sci 2016;17:E588. doi: 10.3390/ijms17050588. [32] Lei X, Chen J, Ren J, Li Y, Zhai J, Mu W, et al. Liver damage associated with [47] Liu C, Fan H, Li Y, Xiao X. Research advances on hepatotoxicity of herbal polygonum multiflorum Thunb.: A systematic review of case reports and case series. Evid Based Complement Alternat Med 2015;2015:459749. medicines in China. Biomed Res Int 2016;2016:7150391. doi: 10. doi: 10.1155/2015/459749. 1155/2016/7150391. [33] Liu Y, Wang Q, Yang J, Guo X, Liu W, Ma S, et al. Polygonum multiflorum [48] Teschke R, Eickhoff A. Herbal hepatotoxicity in traditional and modern med- Thunb.: A review on chemical analysis, processing mechanism, quality eval- icine: actual key issues and new encouraging steps. Front Pharmacol 2015; uation, and hepatotoxicity. Front Pharmacol 2018;9:364. doi: 10. 6:72. doi: 10.3389/fphar.2015.00072. 3389/fphar.2018.00364. [49] Ma X, Peng JH, Hu YY. Chinese herbal medicine-induced liver injury. J Clin [34] Jing J, Wang RL, Zhao XY, Zhu Y, Niu M, Wang LF, et al. Association between Transl Hepatol 2014;2:170–175. doi: 10.14218/JCTH.2014.00009. the concurrence of pre-existing chronic liver disease and worse prognosis in [50] Philips CA, Paramaguru R, Augustine P. Severe alcoholic hepatitis in a tee- patients with an herb- Polygonum multiflorum thunb. induced liver injury: a totaler. Am J Gastroenterol 2018;113:1260–1261. doi: 10.1038/s41395- case-control study from a specialised liver disease center in China. BMJ Open 018-0154-0. 2019;9:e023567. doi: 10.1136/bmjopen-2018-023567. [51] Philips CA, Augustine P, Paramaguru R, Ahamed R. Homeopathy-medicine [35] Thyagarajan SP, Jayaram S, Gopalakrishnan V, Hari R, Jeyakumar P, Sripathi induced severe alcoholic hepatitis. BMJ Case Rep 2019;12:e229627. doi: MS. Herbal medicines for liver diseases in India. J Gastroenterol Hepatol 10.1136/bcr-2019-229627. 2002;17 Suppl 3:S370–S376. doi: 10.1046/j.1440-1746.17.s3.30.x. [52] Philips CA, Paramaguru R, Augustine P. Ayurveda metallic-mineral ‘Bhasma’- [36] Jorge OA, Jorge AD. Hepatotoxicity associated with the ingestion of Centella associated severe liver injury. BMJ Case Rep 2018;2018:bcr-2018-225590. – asiatica. Rev Esp Enferm Dig 2005;97:115 124. doi: 10.4321/S1130- doi: 10.1136/bcr-2018-225590. 01082005000200006. [53] Philips CA, Augustine P, Paramaguru R, Ahamed R, Padsalgi G. Ayurveda [37] Dantuluri S, North-Lewis P, Karthik SV. Gotu Kola induced hepatotoxicity in a herbal medicine-induced liver cirrhosis. Cureus 2019;11:e4122. doi: 10. child - need for caution with alternative remedies. Dig Liver Dis 2011;43: 7759/cureus.4122. 500. doi: 10.1016/j.dld.2010.12.012. [54] Kotecha VR. Ayurveda research publications: A serious concern. Ayu 2015; [38] Teschke R, Bahre R. Severe hepatotoxicity by Indian Ayurvedic herbal prod- 36:1–2. doi: 10.4103/0974-8520.169004. ucts: a structured causality assessment. Ann Hepatol 2009;8:258–266. doi: [55] Samal J, Dehury RK. The need and importance of incorporating academic 10.1016/S1665-2681(19)31777-6. research results in to the curricula of Ayurveda in India. J Clin Diagn Res [39] Fleig W, Morgan MY, Hölzer M. The Ayurvedic drug LIV.52 in patients with 2017;11:KA01–KA03. doi: 10.7860/JCDR/2017/26742.10112. alcoholic cirrhosis. Results of a prospective, randomized, double-blind, [56] Kazim M, Puri SK, Dutta GP, Narasimham MV. Evaluation of Ayush-64 for placebo-controlled clinical trial. J Hepatol 1997;126:127. [40] Zhu Y, Li YG, Wang JB, Liu SH, Wang LF, Zhao YL, et al. Causes, features, and blood schizontocidal activity against rodent and simian malaria parasites. – outcomes of drug-induced liver injury in 69 children from China. Gut Liver Indian J Malariol 1991;28:255 258. 2015;9:525–533. doi: 10.5009/gnl14184. [57] Valecha N, Devi C, Joshi H, Shahi V, Sharma V, Lal S. Comparative efficacy of [41] Lee CH, Wang JD, Chen PC. Risk of liver injury associated with Chinese herbal Ayush-64 vs chloroquine in vivax malaria. Current Science 2000;78:1120– products containing radix bupleuri in 639,779 patients with hepatitis B virus 1122. infection. PLoS One 2011;6:e16064. doi: 10.1371/journal.pone.0016064. [58] Wang JB, Zhu Y, Bai ZF, Wang FS, Li XH, Xiao XH. Guidelines for the diagnosis [42] Jakkula M, Boucher TA, Beyendorff U, Conn SM, Johnson JE, Nolan CJ, and management of herb-induced liver injury. Chin J Integr Med 2018;24: et al. A randomized trial of Chinese herbal medicines for the treatment 696–706. doi: 10.1007/s11655-018-3000-8.

274 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 263–274 Review Article

Obesity Paradox in Chronic Liver Diseases: Product of Bias or a Real Thing?

Ines Bilic Curcic1,2, Maja Cigrovski Berkovic3, Lucija Kuna4, Hrvoje Roguljic1,5, Robert Smolic6,7, Silvija Canecki Varzic2,8, Lucija Virovic Jukic9 and Martina Smolic*1,4

1Department of Pharmacology, Faculty of Medicine, J. J. Strossmayer University of Osijek, Osijek, Croatia; 2Department of Diabetes, Endocrinology and Metabolism Disorders, Osijek University Hospital Centre, Faculty of Medicine, Josip Juraj Strossmayer University of Osijek, Osijek, Croatia; 3Department for Endocrinology, Diabetes and Metabolism University Hospital Center, “Sestre Milosrdnice”, Zagreb, Croatia; 4Department of Pharmacology and Biochemistry, Faculty of Dental Medicine and Health, J. J. Strossmayer University of Osijek, Osijek, Croatia; 5Department for Cardiovascular Disease, Osijek University Hospital, Osijek, Croatia; 6Department of Gastroenterology and Hepatology, Osijek University Hospital Centre, Faculty of Medicine, Josip Juraj Strossmayer University of Osijek, Osijek, Croatia; 7Department of Pathophysiology, Faculty of Medicine, Josip Juraj Strossmayer University of Osijek, Osijek, Croatia; 8Department of Internal Medicine, Faculty of Medicine, Josip Juraj Strossmayer University of Osijek, Osijek, Croatia; 9Department of Gastroenterology and Hepatology, Sisters of Mercy University Hospital, University of Zagreb School of Medicine, Zagreb, Croatia

Abstract liver diseases: Product of bias or a real thing? J Clin Transl Hep- atol 2019;7(3):275–279. doi: 10.14218/JCTH.2019.00029. In recent years, evidence supporting the theory of obesity paradox has increased, showing that obese/overweight people with prevalent chronic diseases experience lower mortality Introduction compared with patients of normal weight. So far, evidence is most comprehensive in cardiovascular and chronic renal dis- Obesity, among other diseases and conditions, increases eases; however, published studies are prone to many biases, morbidity from cardiovascular disease, diabetes, some enabling us to reach a definite conclusion. Available data in cancer types, biliary disease, and arthrosis, and impairs chronic liver disease is scarce and ambiguous. Obesity is quality of life, representing one of the most important traditionally associated with nonalcoholic fatty liver disease causes of mortality.1 Global prevalence of obesity (defined and steatosis in viral hepatitis and as such one would not by body mass index (BMI) of $30 kg/m2) and overweight expect the obesity paradox to be a real possibility in liver (defined by BMI of 25.0 to 29.9 kg/m2) has reached epidemic disease. Yet, there seem to be new data indicating the opposite proportions. According to 2016 World Health Organization – the obesity paradox exists in severe and end-stage liver data, 39% of adults were overweight and 13% (over 650 cirrhosis, which could be attributed to a better lean mass in million) were obese; the numbers, although already disturb- patients with higher body mass index, meaning that sarcope- ing, might be an underestimation of the true prevalence.2 nia, as one of the most important prognostic factors of survival, On the other hand, evidence is increasingly showing that is less likely to be present. Nonetheless, the problem of various obese/overweight people with prevalent chronic diseases methodological problems addressing the association between experience lower mortality compared with patients of normal body weight and mortality, which is present both in liver weight during acute illnesses. For instance, in one meta- disease and other chronic diseases, are preventing us from analysis, in patients with chronic heart failure, obesity was attaining an unanimous conclusion. Still, we should be aware associated with a 33% lower risk of mortality relative to those that the obesity paradox might be true, especially in severe and with normal weight, while in patients with ischemic heart end-stage illness. This suggests focusing our efforts toward disease, similar risk reductions have also been observed.3–6 preserving or building up fat-free mass and decreasing inflam- This paradoxical observation, known as the “obesity matory activity responsible for catabolism and sarcopenia, and paradox,” is rather confusing since lifestyle intervention with implying that the underlaying cause should be treated. a goal of achieving weight reduction is a fundamental issue in Citation of this article: Curcic IB, Berkovic MC, Kuna L, the management of these conditions.7–9 Possible consequen- Roguljic H, Smolic R, Varzic SC, et al. Obesity paradox in chronic ces of the obesity paradox in chronic diseases could have quite an impact on public health and clinical practice. If the obesity paradox proves to be true, the recommendation of change of Keywords: Obesity paradox; Chronic liver disease; NAFLD; NASH; Cirrhosis. lifestyle with the goal of weight loss in the current guidelines Abbreviations: BMI, body mass index; HCC, hepatocellular carcinoma; NAFLD, regarding those conditions, could actually do more harm than nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; NHANES, 10 National Health and Nutrition Examination Survey. good. Also, in the national guidelines regarding weight loss in Received: 11 July 2019; Revised: 8 September 2019; Accepted: 9 September the overall population, no distinction has been made between 2019 various conditions in which obesity could be protective. *Correspondence to: Martina Smolic, Department of Pharmacology, J. J. Stross- So, it is important to elucidate the true obesity-related mayer University of Osijek Faculty of Medicine Osijek, J. Huttlera 4, Osijek 31000, Croatia. Tel: + 385-31-512-800, Fax: +385-31-512-833, E-mail: martina.smolic@ survival benefit, especially in the era of the current wisdom of 11 mefos.hr weight loss. One of the crucial problems we are facing today

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 275–279 275

Copyright: © 2019 Authors. This article has been published under the terms of Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0), which permits noncommercial unrestricted use, distribution, and reproduction in any medium, provided that the following statement is provided. “This article has been published in Journal of Clinical and Translational Hepatology at DOI: 10.14218/JCTH.2019.00029 and can also be viewed on the Journal’s website at http://www.jcthnet.com”. Curcic I.B. et al: Obesity paradox in chronic liver diseases is understanding the definition of obesity, or rather address- Additionally, morbid obesity presents an increased risk for ing lack of the same. It is of utter importance to revisit the type 2 diabetes and cardiovascular disease, and is a signifi- definition of obesity as made by the sole use of BMI as a cant risk factor for numerous malignant diseases, particularly measure, since it lacks information on body composition, hepatocellular carcinoma (HCC).21 It has been demonstrated such as sarcopenia, which might be clinically relevant in a that the increase in BMI >30 kg/m2 has influence on raising condition within the obesity paradox.12–14 Furthermore, the risk of HCC development, particularly in the male popula- when discussing the obesity paradox, one must take into tion where HCC mortality rates are 5 times higher than in account different important confounders that lead to bias those with a normal BMI.22 Numerous essential studies when interpreting data on the obesity-related health protec- related to control of inflammatory cytokine regeneration, tive role in different disease settings. adipose factors, oxidative stress, and lipid storage have The majority of available data related to the obesity paradox explained NASH as a metabolic syndrome. Moreover, recent have been extracted from populations with overt diseases, studies have also demonstrated the important role of genetic wherein weight loss can be a sign of more severe disease or factors in contributing to NASH pathogenesis and to inducing advanced disease stage. In a study examining myocardial malignancy in the liver.22 Furthermore, a connection has been infarction, congestive heart failure, stroke, and ischemic heart made between obesity and NAFLD, such that obesity enhan- disease in a population-based study, the obesity paradox was ces the risk of progression of fibrosis and hepatic inflamma- observed to be prevalent in cardiovascular disease; however, tion, leading to NASH-related cirrhosis.23 incidence of cardiovascular disease provided no evidence of a The pathophysiology of NASH is complex and still of major survival advantage for obesity.6 In the case of the normal research interest. It is accepted that numerous mechanisms weight category including people who lost weight due to and pathways are included in the pathogenesis of NASH. disease severity, then the adverse effects of obesity on mortal- Insulin resistance, dietary factors, genetic polymorphisms, ity could be systematically and severely underestimated.15,16 and lipotoxicity form the group of mechanisms found to date Additionally, there is an issue of survivor bias, which relates to and characterized as playing key roles in NASH pathogene- those who are obese and with more severe disease who die sis.24 It remains uncertain, however, whether NASH always early, leaving behind more robust obese persons at the time precedes NAFLD. Pathogenic driving factors are unlikely to be of data analysis for obesity paradox investigation. So, in order the same for all patients. Thus, the mechanisms that cause to minimize and control the mentioned biases, additional the disease and their clinical manifestations are greatly studies examining persons at time of the incident disease, or heterogeneous.25 using the prediagnosis weight or BMI data, would be more rel- A link between obesity, NAFLD and HCC is well established. evant prior to suggesting broad population guidelines on weight Moreover, chronic infection with hepatitis B virus or hepatitis loss, without distinguishing between the persons with condi- C virus is strongly connected with hepatic steatosis. Com- tions/diseases where obesity might be protective.17 pounded forms of metabolic syndrome involving type 2 diabetes mellitus, obesity and/or dyslipidemia are a frequent Obesity paradox in nonalcoholic steatohepatitis – cause of viral hepatitis-associated hepatic steatosis. Due to Possibility or improbability? that, damaged glucose metabolism enhances the possibility of hepatic failure and death in patients with recompensated 26 One particular clinical entity requiring special attention when hepatitis C virus-related cirrhosis. In turn, hepatitis C virus it comes to obesity and liver health is certainly nonalcoholic infection is essential as a risk factor, as it has significant influ- 27,28 fatty liver disease (NAFLD), which has become an increas- ence on development of insulin resistance. Therefore, ingly common cause of chronic liver disease in the developing based on these collective lines of evidence, the logical ques- world. In general, NAFLD has been found in 17–30% of the tion to ask is whether it is possible that the obesity paradox populations of Western nations and is estimated to be 2–4% even exists in the condition of liver disease. worldwide. Metabolic disorders such as hypertension, type 2 diabetes mellitus, dyslipidemia and obesity show strongly Obesity paradox in liver cirrhosis – Is it time for a associations with NAFLD. Moreover, NAFLD manifests as paradigm shift? numerous series of liver disease, ranging from common steatosis to nonalcoholic steatohepatitis (NASH) and cirrhosis Answer to the question asked, could be provided by the in a subset of patients. It is important to emphasize that not data reported in a recently published study conducted by all patients with NASH will suffer from cirrhosis. Indeed, the Kargozian et al.29 in which clinical outcomes, such as terms of collective NAFLD patients in who the disease will lead to NASH mortality, length of stay, and total hospital charges, in hospi- and require therapy is a hot area of investigation. Currently, talized obese cirrhotic patients were compared with those of various different therapeutic targets and numerous therapies nonobese cirrhotic patients. The results were in alignment are being investigated for NASH.18 with the obesity paradox as indicated to be present in other Over the past few decades, obesity has emerged as a intensive care patients.30,31 Obesity itself was associated with global epidemic, with significant influence on the increasing a decrease in inpatient mortality but with higher costs and prevalence of obesity-related systemic disorders and involv- lengthy hospitalization in patients with cirrhosis. However, ing NAFLD.19 This particularly refers to modifications in life- the Kargozian study had several limitations, as it was of ret- style, especially in physical activity and nutrition. In fact, it is rospective design and therefore susceptible to biases and well known that half of the adult population around the world, with a lack of anthropometric data and BMI values for each particularly in developed Western countries, have a problem patient. However, similar results were found in patients with overweight or obesity. Although, obesity is one of the with hepatocellular carcinoma; specifically, lower BMI was a causes of a disease that can be prevented and which, if is predictive factor for higher rates of mortality and shorter sur- uncontrolled, can lead to metabolic complications and conse- vival. Although, 20% of patients with lower BMI were actually quently to premature mortality.20 undernourished, which could have produced alternate study

276 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 275–279 Curcic I.B. et al: Obesity paradox in chronic liver diseases

Fig. 1. Schematic presentation of obesity as a risk factor for chronic liver diseases, such as NASH and NAFLD. The latter are denoted by blue arrows (increasing impact). This risk factor could also have beneficial effects (denoted by red arrows; decreasing impact) on liver disease-associated mortality, which might be hampered when associated with sarcopenia, due to impact of sarcopenia on malnutrition and weight loss. Abbreviations: NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis. results considering that malnutrition in terminally ill patients thus affecting the final outcome. In addition, there is the is a poor prognostic risk factor.32 factor of body composition; in older patients, BMI is a better Previously published studies have suggested the complete indicator of lean body mass than the adipose tissue, whereas opposite of the above data. In a study comprising data from in patients with chronic illness obesity correlates with the National Health and Nutrition Examination Survey (oft increased muscle mass and improved nutritional status.40,41 referred to as NHANES) database for patients without pre- A well-known fact is that patients with advanced liver vious liver disease, cirrhosis-related deaths and hospital- disease lose weight more prominently due to loss of appetite izations were found to be positively associated with obesity and inadequate nourishment caused by ascites and impaired (follow-up being 13 years).33 Recently, it was also found gut motility; absorption of nutrients is also impaired due to that increased BMI was positively associated with clinical bacterial overgrowth, while the inflammatory state induces decompensation in cirrhosis (with a mean follow-up of 59 anorexia, and glycogen stores decrease, thereby increasing months).34 One of the possible explanations of those seem- gluconeogenesis. Those disturbances all lead to one common 42 ingly opposite results is regarding obesity as a consequence denominator: sarcopenia. A recent study on cirrhotic rather than a cause. This is supported by obese patients being patients demonstrated that sarcopenic patients had increased 43 more aggressively treated according to guidelines than mortality rates compared to nonsarcopenic patients. normal body weight people, possibly contributing to a better However, the prognosis was significantly worse in patients outcome in obese patients. For example, obese patients are with sarcopenic obesity, followed by sarcopenia and visceral more likely to receive statins and oral antihyperglycemic obesity; meanwhile, elderly (aged $64 years) subjects had agents, such as metformin, pioglitazone or liraglutide, contri- poorer prognosis that for sarcopenic obesity than did the sub- buting to a decrease in the inflammation processes and being jects with normal body composition, suggesting once again associated with better prognosis in liver diseases.35–37 In one that visceral obesity has a beneficial effect on prognosis as 13 study including patients with chronic hepatitis C virus-associ- long there is no sarcopenia present. ated liver disease and severe liver fibrosis or compensated This leave us with the question of how the potential biases cirrhosis, 43% were obese and 32% were overweight. A lurking around the corner could significantly affect the inter- 14% increase in the risk of worsening of liver fibrosis or pretation of the results from these collective studies? One of decompensation was observed to be related to an increase many possible answers is that among patients with already in BMI (with a follow-up of 3.5 years).38 developed chronic liver disease, those with more serious Support for the hypothesis of obesity paradox in liver conditions lose weight more prominently, thus creating a false image of the protective effect of obesity. Moreover, the answer disease was recently obtained by a study that demonstrated could be that obese people with severe disease die earlier and a liver-related complications to be higher among patients with selectively overweight population of healthier people remain.6 hepatitis C cirrhosis than among patients with NAFLD cirrhosis, but overall mortality was the same.40 This finding was probably due to the higher BMI among NAFLD patients influ- Conclusions encing a more prolonged course of illness compared to hepatitis C cirrhosis but with no difference in mortality. Also, Available data suggest that obesity prior to onset of liver NAFLD-related cirrhosis is factor of more slowly progressing disease could have detrimental effects, while higher BMI in disease than the other causes of cirrhosis (alcohol, chronic the advanced liver disease may represent a positive prog- hepatitis C). In the initial stages, the nonobese patients nostic factor; the latter is supported by evidence obtained may have other more aggressive cofactors for liver disease, from studies of other chronic illnesses, such as renal and

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 275–279 277 Curcic I.B. et al: Obesity paradox in chronic liver diseases cardiovascular disease, heart failure and rheumatoid arthri- [7] Martín-Ponce E, Santolaria F, Alemán-Valls MR, González-Reimers E, Martínez- tis.6,44–47 Thus, we can conclude that the fat mass itself is Riera A, Rodríguez-Gaspar M, et al. Factors involved in the paradox of reverse epidemiology. Clin Nutr 2010;29:501–506. doi: 10.1016/j.clnu.2009.12.009. probably a factor associated with a better nutritional state [8] Han SJ, Boyko EJ. The evidence for an obesity paradox in type 2 diabetes positively affecting survival but not a relevant factor stipulat- mellitus. Diabetes Metab J 2018;42:179–187. doi: 10.4093/dmj.2018.0055. ing a better outcome. Although a protective role of obesity [9] Curtis JP, Selter JG, Wang Y, Rathore SS, Jovin IS, Jadbabaie F, et al. The regarding survival or mortality rate cannot be established obesity paradox: body mass index and outcomes in patients with heart failure. Arch Intern Med 2005;165:55–61. doi: 10.1001/archinte.165.1.55. with certainty, the observation of patients with more severe [10] Smith SC Jr, Benjamin EJ, Bonow RO, Braun LT, Creager MA, Franklin BA, and advanced liver disease without sarcopenia presenting et al. AHA/ACCF secondary prevention and risk reduction therapy for higher BMI could mean a better prognosis (Fig. 1). patients with coronary and other atherosclerotic vascular disease: 2011 To summarize, there is an extensive body of evidence update: a guideline from the American Heart Association and American College of Cardiology Foundation endorsed by the World Heart Federation supporting the obesity paradox regarding other chronic ill- and the Preventive Cardiovascular Nurses Association. J Am Coll Cardiol nesses, although some confounders may be present; 2011;58:2432–2446. doi: 10.1016/j.jacc.2011.10.824. however, data on liver diseases are still vague and scarce. [11] Schmidt DS, Salahudeen AK. Obesity-survival paradox-still a controversy? Ambiguous results from the data published so far are pre- Semin Dial 2007;20:486–492. doi: 10.1111/j.1525-139X.2007.00349.x. [12] Pichard C, Kyle UG, Morabia A, Perrier A, Vermeulen B, Unger P. Nutritional venting us from obtaining a unanimous conclusion. This assessment: lean body mass depletion at hospital admission is associated difference is probably caused by several factors, including with an increased length of stay. Am J Clin Nutr 2004;79:613–618. doi: 10. the widely used BMI as an imperfect tool for defining obesity, 1093/ajcn/79.4.613. the use of prevalent models in published studies instead of [13] Roubenoff R. Sarcopenic obesity: the confluence of two epidemics. Obes Res 2004;12:887–888. doi: 10.1038/oby.2004.107. incident ones since prevalent illness is inevitably influenced [14] Lavie CJ, De Schutter A, Patel DA, Romero-Corral A, Artham SM, Milani RV. by weight loss caused by disease, and selective survival Body composition and survival in stable coronary heart disease: impact of causing impartiality. However, we should be open to the lean mass index and body fat in the “obesity paradox”. J Am Coll Cardiol possibility that the obesity paradox may be true and focus 2012;60:1374–1380. doi: 10.1016/j.jacc.2012.05.037. our efforts toward preserving or generating lean mass, such [15] Mehta NK, Chang V. The Oxford Handbook of the Social Science of Obesity. Cawley JH, editor. New York: Oxford University Press; 2011. as that of muscle and bone, and decreasing inflammatory [16] Banack HR, Kaufman JS. Does selection bias explain the obesity paradox activity responsible for catabolism and sarcopenia; thus, the among individuals with cardiovascular disease? Ann Epidemiol 2015;25: cause of the primary disease should be treated. 342–349. doi: 10.1016/j.annepidem.2015.02.008. [17] World Health Organization. Obesity: preventing and managing the global epidemic. Report of a WHO. consultation (WHO Technical Report Series 894) Conflict of interest 2000. Available from: https://www.who.int/nutrition/publications/obesity/ WHO_TRS_894/en/. [18] Sarwar R, Pierce N, Koppe S. Obesity and nonalcoholic fatty liver disease: The authors have no conflict of interests related to this current perspectives. Diabetes Metab Syndr Obes 2018;11:533–542. doi: publication. 10.2147/DMSO.S146339. [19] Renehan AG, Tyson M, Egger M, Heller RF, Zwahlen M. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective Author contributions observational studies. Lancet 2008;371:569–578. doi: 10.1016/S0140- 6736(08)60269-X. Conceived of and designed the article, and critically revised [20] Larsson SC, Wolk A. Excess body fatness: an important cause of most cancers. Lancet 2008;371:536–537. doi: 10.1016/S0140-6736(08)60247-0. the manuscript (IBC, MS), performed critical revision of the [21] Wolk A, Gridley G, Svensson M, Nyrén O, McLaughlin JK, Fraumeni JF, et al.A manuscript for important intellectual content, obtained prospective study of obesity and cancer risk (Sweden). Cancer Causes funding, and provided administrative, technical and material Control 2001;12:13–21. doi: 10.1023/A:1008995217664. support (MS), performed literature searches and critical [22] Yu J, Shen J, Sun TT, Zhang X, Wong N. Obesity, insulin resistance, NASH and hepatocellular carcinoma. Semin Cancer Biol 2013;23:483–491. doi: 10. revision of the manuscript for important intellectual content 1016/j.semcancer.2013.07.003. (RS, SCV, LVJ), performed literature searches, wrote the [23] Jaycox SH. A fatty world: Exploring racial disparity in NAFLD/NASH. J Hep manuscript, and updated the text of the paper (IBC, MCB, 2016;2:2. doi: 10.21767/2471-9706.100014. LK), performed literature searches and wrote the manuscript [24] Chalasani N, Younossi Z, Lavine JE, Charlton M, Cusi K, Rinella M, et al. The diagnosis and management of nonalcoholic fatty liver disease: Practice guid- (HR). ance from the American Association for the Study of Liver Diseases. Hepa- tology 2018;67:328–357. doi: 10.1002/hep.29367. [25] Alonso C, Fernández-Ramos D, Varela-Rey M, Martínez-Arranz I, Navasa N, References Van Liempd SM, et al. Metabolomic identification of subtypes of nonalcoholic steatohepatitis. Gastroenterology 2017;152:1449–1461.e7. doi: 10.1053/j. [1] Lonardo A, Bellentani S, Argo CK, Ballestri S, Byrne CD, Caldwell SH, et al. gastro.2017.01.015. Epidemiological modifiers of non-alcoholic fatty liver disease: Focus on high- [26] Calzadilla-Bertot L, Vilar-Gomez E, Torres-Gonzalez A, Socias-Lopez M, risk groups. Dig Liver Dis 2015;47:997–1006. doi: 10.1016/j.dld.2015.08. Diago M, Adams LA, et al. Impaired glucose metabolism increases risk of 004. hepatic decompensation and death in patients with compensated hepatitis [2] de Herder WW, Eng C. Obesity, diabetes mellitus, and cancer. Endocr Relat C virus-related cirrhosis. Dig Liver Dis 2016;48:283–290. doi: 10.1016/j. Cancer 2012;19:E5–E7. doi: 10.1530/ERC-12-0264. dld.2015.12.002. [3] Niedziela J, Hudzik B, Niedziela N, Gąsior M, Gierlotka M, Wasilewski J, et al. [27] Mehta SH, Brancati FL, Sulkowski MS, Strathdee SA, Szklo M, Thomas DL. The obesity paradox in acute coronary syndrome: a meta-analysis. Eur J Prevalence of type 2 diabetes mellitus among persons with hepatitis C virus Epidemiol 2014;29:801–812. doi: 10.1007/s10654-014-9961-9. infection in the United States. Ann Intern Med 2000;133:592–599. doi: 10. [4] Oreopoulos A, Padwal R, Norris CM, Mullen JC, Pretorius V, Kalantar-Zadeh K. 7326/0003-4819-133-8-200010170-00009. Effect of obesity on short- and long-term mortality postcoronary revascula- [28] Sanyal AJ, Chalasani N, Kowdley KV, McCullough A, Diehl AM, Bass NM, et al. rization: a meta-analysis. Obesity (Silver Spring) 2008;16:442–450. doi: Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis. N Engl J 10.1038/oby.2007.36. Med 2010;362:1675–1685. doi: 10.1056/NEJMoa0907929. [5] Romero-Corral A, Montori VM, Somers VK, Korinek J, Thomas RJ, Allison TG, [29] Karagozian R, Bhardwaj G, Wakefield DB, Baffy G. Obesity paradox in et al. Association of bodyweight with total mortality and with cardiovascular advanced liver disease: obesity is associated with lower mortality in hospi- events in coronary artery disease: a systematic review of cohort studies. talized patients with cirrhosis. Liver Int 2016;36:1450–1456. doi: 10. Lancet 2006;368:666–678. doi: 10.1016/S0140-6736(06)69251-9. 1111/liv.13137. [6] Chang VW, Langa KM, Weir D, Iwashyna TJ. The obesity paradox and incident [30] Hogue CW Jr, Stearns JD, Colantuoni E, Robinson KA, Stierer T, Mitter N, et al. cardiovascular disease: A population-based study. PLoS One 2017;12: The impact of obesity on outcomes after critical illness: a meta-analysis. e0188636. doi: 10.1371/journal.pone.0188636. Intensive Care Med 2009;35:1152–1170. doi: 10.1007/s00134-009-1424-5.

278 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 275–279 Curcic I.B. et al: Obesity paradox in chronic liver diseases

[31] Akinnusi ME, Pineda LA, El Solh AA. Effect of obesity on intensive care mor- [40] Oreopoulos A, Ezekowitz JA, McAlister FA, Kalantar-Zadeh K, Fonarow GC, bidity and mortality: a meta-analysis. Crit Care Med 2008;36:151–158. doi: Norris CM, et al. Association between direct measures of body composition 10.1097/01.CCM.0000297885.60037.6E. and prognostic factors in chronic heart failure. Mayo Clin Proc 2010;85:609– [32] Li Q, Xing H, Liu D, Li H. Negative impact of low body mass index on liver 617. doi: 10.4065/mcp.2010.0103. cirrhosis patients with hepatocellular carcinoma. World J Surg Oncol 2015; [41] Casas-Vara A, Santolaria F, Fernández-Bereciartúa A, González-Reimers E, 13:294. doi: 10.1186/s12957-015-0713-4. García-Ochoa A, Martínez-Riera A. The obesity paradox in elderly patients [33] Ioannou GN, Weiss NS, Kowdley KV, Dominitz JA. Is obesity a risk factor for with heart failure: analysis of nutritional status. Nutrition 2012;28:616–622. cirrhosis-related death or hospitalization? A population-based cohort study. Gas- doi: 10.1016/j.nut.2011.10.006. troenterology 2003;125:1053–1059. doi: 10.1016/s0016-5085(03)01200-9. [42] Kim HY, Jang JW. Sarcopenia in the prognosis of cirrhosis: Going beyond the [34] Berzigotti A, Garcia-Tsao G, Bosch J, Grace ND, Burroughs AK, Morillas R, MELD score. World J Gastroenterol 2015;21:7637–7647. doi: 10.3748/wjg. et al. Obesity is an independent risk factor for clinical decompensation in v21.i25.7637. patients with cirrhosis. Hepatology 2011;54:555–561. doi: 10.1002/hep. [43] Hara N, Iwasa M, Sugimoto R, Mifuji-Moroka R, Yoshikawa K, Terasaka E, 24418. et al. Sarcopenia and sarcopenic obesity are prognostic factors for overall [35] Souk K, Al-Badri M, Azar ST. The safety and benefit of statins in liver cirrho- survival in patients with cirrhosis. Intern Med 2016;55:863–870. doi: 10. sis: a review. Exp Clin Endocrinol Diabetes 2015;123:577–580. doi: 10. 2169/internalmedicine.55.5676. 1055/s-0035-1564093. [44] Lavie CJ, Milani RV, Ventura HO. Obesity and cardiovascular disease: risk [36] Donadon V, Balbi M, Mas MD, Casarin P, Zanette G. Metformin and reduced risk factor, paradox, and impact of weight loss. J Am Coll Cardiol 2009;53: of hepatocellular carcinoma in diabetic patients with chronic liver disease. 1925–1932. doi: 10.1016/j.jacc.2008.12.068. Liver Int 2010;30:750–758. doi: 10.1111/j.1478-3231.2010.02223.x. [45] Thomas F, Bean K, Pannier B, Oppert JM, Guize L, Benetos A. Cardiovascular [37] Blaslov K, Bulum T, Zibar K, Duvnjak L. Incretin based therapies: a novel mortality in overweight subjects: the key role of associated risk factors. Hyper- treatment approach for non-alcoholic fatty liver disease. World J Gastroen- tension 2005;46:654–659. doi: 10.1161/01.HYP.0000184282.51550.00. terol 2014;20:7356–7365. doi: 10.3748/wjg.v20.i23.7356. [46] De Schutter A, Lavie CJ, Patel DA, Milani RV. Obesity paradox and the heart: [38] Everhart JE, Lok AS, Kim HY, Morgan TR, Lindsay KL, Chung RT, et al. Weight- which indicator of obesity best describes this complex relationship? Curr related effects on disease progression in the hepatitis C antiviral long-term Opin Clin Nutr Metab Care 2013;16:517–524. doi: 10.1097/MCO. treatment against cirrhosis trial. Gastroenterology 2009;137:549–557. doi: 0b013e328363bcca. 10.1053/j.gastro.2009.05.007. [47] Elkan AC, Engvall IL, Cederholm T, Hafström I. Rheumatoid cachexia, central [39] Bhala N, Angulo P, van der Poorten D, Lee E, Hui JM, Saracco G, et al. The obesity and malnutrition in patients with low-active rheumatoid arthritis: natural history of nonalcoholic fatty liver disease with advanced fibrosis or feasibility of anthropometry, Mini Nutritional Assessment and body compo- cirrhosis: an international collaborative study. Hepatology 2011;54:1208– sition techniques. Eur J Nutr 2009;48:315–322. doi: 10.1007/s00394-009- 1216. doi: 10.1002/hep.24491. 0017-y.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 275–279 279 Case Report

Amoxicillin-clavulanate-induced Granulomatous Hepatitis: Case Report and Review of the Literature

Avin Aggarwal*1, Neha Jaswal1, Richa Jain2 and Hussien Elsiesy3

1Department of Medicine, University of Arizona, Tucson, AZ, USA; 2Department of Pathology, University of Arizona, Tucson, AZ, USA; 3Liver consultants of Texas, Baylor All Saints Medical Center, Fort Worth, TX, USA

Abstract rebound tenderness. Initial laboratory examination revealed elevated total bilirubin at 3.2 mg /dL, alkaline phosphatase at Amoxicillin-clavulanate (AC) is a common cause of drug- 510 IU/L, alanine aminotransferase at 388 IU/L, and aspar- induced liver injury, either cholestatic or mixed with hepatitis tate aminotransferase at 171 IU/L. Lipase level, international pattern. Rarely, AC causes granulomatous hepatitis. We normalized ratio, and platelet count were within normal report a new case of AC-induced granulomatous hepatitis limits. Total leukocyte count was also normal at 6.6 documented by liver biopsy, with complete resolution of any (1000/mL) but differential count showed eosinophilia histological sequelae on a follow-up liver biopsy after AC was (8.3%), which trended up to 12% with absolute eosinophil withdrawn. count of 0.76 (1000/mL). The liver enzymes and bilirubin Citation of this article: Aggarwal A, Jaswal N, Jain R, Elsiesy trend throughout the hospitalization is shown in Fig. 1. Inter- H. Amoxicillin-clavulanate-induced granulomatous hepatitis: national normalized ratio continued to be within normal range Case report and review of the literature. J Clin Transl Hepatol throughout the hospitalization. 2019;7(3):280–283. doi: 10.14218/JCTH.2019.00027. Computed tomography of the abdomen showed normal liver with a low-density lesion in the right lobe of the liver, probably representing a cyst. Magnetic resonance imaging of the abdomen showed heterogenous arterial enhancement of Introduction the liver without biliary obstruction. Serological work-up, including testing for human immunodeficiency virus, rapid Amoxicillin-clavulanate (AC) is identified as a common cause plasma reagin, hepatitis panel (hepatitis A antibody subtype of nonacetaminophen drug-induced liver injury (DILI)1,2 as IgM, hepatitis surface antigen, hepatitis B core IgM, hepatitis well as the consequent hospitalizations for adverse hepatic reactions.2 The Drug-Induced Liver Injury Network (DILIN) C antibody), antineutrophilic antibody, antimitochondrial anti- cohort was established in 2004 in the USA and currently con- body, and anti-smooth muscle antibody, was negative. Cer- sists of over 1450 cases of DILI, of which 11% are due to AC. uloplasmin, ferritin, and immunoglobulin panel were normal. Classically, AC-DILI presents as a spectrum of cholestatic Urine drug screen was positive for cannabinoids. Acetamino- liver injury, with pure intra-hepatic cholestasis and/or mixed phen level was negative at presentation. hepatic/cholestatic injury, but there are only rare reports of a Liver biopsy showed portal granulomas centered around granulomatous hepatitis (GH) pattern.3–5 Our case demon- the bile ducts, with a significant biliary epithelial injury. strates GH caused by AC-DILI, with a follow-up complete res- Fungal, acid-fast bacilli and Giemsa stainings were negative, olution of the histological lesion once the drug was withdrawn. and no fibrosis or periportal copper deposition was seen to suggest chronic cholestatic disease. Abundant eosinophils with mononuclear cells were observed. Cytokeratin 19 Case report showed increased biliary profiles in some portal tracts. However, there was an absence of marginal ductular reaction A 58-year-old woman with a history of polysubstance abuse or ductopenia. PAX-5 staining did not show any B cell (alcohol, methamphetamine, cocaine, and marijuana) pre- aggregates (Fig. 2). Other differential diagnosis considered sented with 1-week history of epigastric discomfort, malaise, (and ruled out) included primary biliary cholangitis, sarcoido- and nausea with dark urine and pale colored stools, following sis, Hodgkin’s disease, and Langerhans cell histiocytosis. It a 7-day course of AC for Bartholin abscess. She denied any was not clear if the granuloma in the liver biopsy was pre- recent travel or intravenous drug use. She was afebrile and existing undiagnosed or related to the use of AC. hemodynamically stable on presentation. Physical exam was Follow-up at 1 month showed continuous improvement in significant for scleral icterus and epigastric tenderness but no the liver enzymes, but they did not normalize. Four months later, the liver panel showed a total bilirubin of 1.2 mg/dL, Keywords: Granulomatous hepatitis; Amoxicillin; Clavulanate; Drug-induced aspartate aminotransferase of 43 IU/L, alanine aminotrans- liver injury; Liver biopsy. Abbreviations: AC, amoxicillin-clavulanate; DILI, drug-induced liver injury; GH, ferase of 104 IU/L, and alkaline phosphatase of 249 IU/L. A granulomatous hepatitis; HLA, human leukocyte antigen. repeat liver biopsy was performed at 6 months and showed Received: 8 July 2019; Revised: 8 September 2019; Accepted: 10 September nearly normal liver parenchyma, with no granulomas, fibro- 2019 sis, steatohepatitis, steatosis, fibrosis or cirrhosis, (evaluated *Correspondence to: Avin Aggarwal, Department of Medicine, University of Arizona, 1501 N Campbell Ave, Tucson, AZ 85724, USA. Tel: +1-520-626-4555, with trichrome and reticulin stainings) along with intact bile Fax: +1-520-874-7105, E-mail: [email protected] ducts (Fig. 3). Eventually, at 7 months, her liver enzymes

280 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 280–283

Fig. 1. Liver panel trend throughout the course of liver injury. Abbreviations: T. Bilirubin, total bilirubin; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase. normalized completely. This strongly suggested that her combinations of 250 to 875 mg amoxicillin with 125 mg of granulomas had been induced by the AC, and they were clavulanate, taken 2–3 times a day. AC is currently the most found to be completely resolved in the follow-up liver biopsy. common cause of nonacetaminophen DILI in the USA and Spain.1,2 Injury onset is reported within a few days to up to Discussion 8 weeks after therapy initiation and can also occur after a delay of days to up to 6 weeks after completion of the anti- AC is an antibiotic combination consisting of amoxicillin, a biotic course. third-generation penicillin and a beta-lactam clavulanate that Hepatotoxicity is typically characterized by a cholestatic acts against beta-lactamase producing penicillin-resistant pattern with marked elevations in alkaline phosphatase and bacteria. AC is one of the most common antibiotic regimens gamma-glutamyltranspeptidase but a mixed or hepatocellu- prescribed in the USA and is available in various dose lar pattern has been reported as well, particularly in younger

Fig. 2. All portal tracts were expanded by mononuclear inflammatory infiltrate. A, H&E staining (x40). B-C, Bile ducts in multiple portal tracts showed epithelioid granulomata centered on the bile duct (B, H&E, B, x100; C, x200). D, Marginal ductular reaction was only mild and focal; foci of lobular inflammation, including sinusoidal infiltrate (D, H&E, x200) and endothelitis (not shown) were observed. E, There was no ductopenia and at least a portion of the bile duct in each portal tract was consistently preserved, as shown by immunostain for cytokeratin 7 (x40).

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 280–283 281 Aggarwal A. et al: AC induced granulomatous hepatitis

Fig. 3. Complete resolution of inflammation. A-B, Nearly normal liver parenchyma (A) and no fibrosis (B) were observed. C, The bile ducts were intact, without any evidence of persistent injury. D, A single focus of periportal copper deposition and focal hepatocyte cholestasis were identified, which will likely be resorbed in due course of time (rhodanine stain, x400). patients.6 The hepatic injury is estimated to occur in 1 out of In a more recent single-center report, hepatic granulomas 2500 prescriptions and to be more common in men, the were found in 2–15% of unselected liver biopsies and out of elderly, and after multiple courses.3,7 A genetic predisposition these, autoimmune liver diseases (primary biliary cirrhosis, with human leukocyte antigen (HLA) type has also been overlap syndrome, and autoimmune hepatitis) accounted identified, with a significant association of the HLA class II for the majority of cases (68%), followed by sarcoidosis haplotype DRB1*1501-DRB5*0101-DQBI*0602 with the (7.5%), chronic hepatitis B virus and hepatitis C virus infection susceptibility to AC-associated hepatotoxicity.8 (7.5%), idiopathic (6%), drugs (3%), and other miscellaneous The exact mechanism of hepatotoxicity is unknown but causes (7.5%).11 Numerous medications, like phenylbuta- an immune-allergic reaction to the clavulanate component zone, allopurinol, sulfonamides, phenytoin, carbamazepine, (rather than amoxicillin) is proposed. Initial presenting chlorpropamide, quinidine, methyldopa, nitrofurantoin, isonia- symptoms include fatigue, nausea, and abdominal pain, zid, amiodarone, and diazepam, have been linked to hepatic followed by pruritus and jaundice, except in children, where granulomas. In a large study with 117 patients with confirmed the injury is typically anicteric. Immune-allergic manifes- AC-DILI, histopathology was available for 32, most of which tations, like fever, rash and eosinophilia, are not always showed cholestatic patterns of injury. There were eosinophils present. The antineutrophilic and antimitochondrial anti- and granulomas noted in 21 and 28 cases, respectively; only 4 body have been reported in 18% and 15% of cases, cases showed large epithelioid granulomas and none of the respectively.3 Granulomatous injury, especially when asso- cases were classified as GH.3 GH due to AC-DILI was reported ciated with eosinophilia suggests a hypersensitivity the first time by Silvain et al.4 but no follow-up liver biopsy was immune-allergic phenomenon. Hepatic injury can be available after normalization of liver enzymes over 75 days. In severe and prolonged (4–24 weeks) but rarely results in another case series of 15 patients with AC-DILI, confirmed lasting injury or death. In the largest study of AC-DILI, histology was available for 7 patients only and out of those with 117 cases, only 12 (11%) continued to have ongoing only 1 had features of GH on initial biopsy.5 GH pattern of evidence of liver injury at 6 months after the onset and to AC-DILI is unusual and only reported in a few reports in the have developed chronic DILI, and 5 of those normalized ami- literature.3–5 notransferases in 2–3years.3 Our case demonstrated classical findings of GH on liver The most common causes of granulomatous liver disease histology and a clear temporal relation of these findings to include sarcoidosis, primary biliary cirrhosis, and tuberculo- abnormal aminotransferases and AC intake. After 6 months, sis.9 Numerous drugs have been implicated in the develop- the aminotransferases markedly improved but did not nor- ment of GH. In a retrospective review of 88 cases of GH at malize completely; therefore, a repeat biopsy was performed. Mayo Clinic, comprehensive work-up for etiology revealed It showed a complete histological resolution of GH, establish- idiopathic GH confined to the liver in 50%, sarcoidosis with ing this case as being an unusual pattern of AC-induced liver the confirmed extra-hepatic disease in 22%, drug-related injury. To our knowledge, this is the first case of AC GH that GH in 6%, tuberculosis in 3%, and other causes in 19%.10 has a follow-up liver biopsy confirming the resolution of GH.

282 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 280–283 Aggarwal A. et al: AC induced granulomatous hepatitis

Conflict of interest [3] deLemos AS, Ghabril M, Rockey DC, Gu J, Barnhart HX, Fontana RJ, et al. Amoxicillin-Clavulanate-Induced Liver Injury. Dig Dis Sci 2016;61:2406– 2416. doi: 10.1007/s10620-016-4121-6. The authors have no conflict of interests related to this [4] Silvain C, Fort E, Levillain P, Labat-Labourdette J, Beauchant M. Granuloma- publication. tous hepatitis due to combination of amoxicillin and clavulanic acid. Dig Dis Sci 1992;37:150–152. doi: 10.1007/bf01308359. [5] Larrey D, Vial T, Micaleff A, Babany G, Morichau-Beauchant M, Michel H, et al. Author contributions Hepatitis associated with amoxycillin-clavulanic acid combination report of 15 cases. Gut 1992;33:368–371. doi: 10.1136/gut.33.3.368. Conception, design, and drafting of the manuscript (AA), [6] Mohi-ud-din R, Lewis JH. Drug- and chemical-induced cholestasis. Clin Liver manuscript preparation with substantial contributions to Dis 2004;8:95–132. doi: 10.1016/S1089-3261(03)00124-7. [7] National Institutes of Health. Drug record amoxicillin-clavulanate. Available intellectual content (NJ, RJ), critical editing and revisions of from: https://livertox.nih.gov/AmoxicillinClavulanate.htm. the manuscript (HE). [8] Lucena MI, Molokhia M, Shen Y, Urban TJ, Aithal GP, Andrade RJ, et al. Sus- ceptibility to amoxicillin-clavulanate-induced liver injury is influenced by multiple HLA class I and II alleles. Gastroenterology 2011;141:338–347. References doi: 10.1053/j.gastro.2011.04.001. [9] Coash M, Forouhar F, Wu CH, Wu GY. Granulomatous liver diseases: a review. [1] Chalasani N, Fontana RJ, Bonkovsky HL, Watkins PB, Davern T, Serrano J, J Formos Med Assoc 2012;111:3–13. doi: 10.1016/j.jfma.2011.11.023. et al. Causes, clinical features, and outcomes from a prospective study of [10] Sartin JS, Walker RC. Granulomatous hepatitis: a retrospective review of 88 drug-induced liver injury in the United States. Gastroenterology 2008;135: cases at the Mayo Clinic. Mayo Clin Proc 1991;66:914–918. doi: 10. 1924–1934, 1934.e1-4. doi: 10.1053/j.gastro.2008.09.011. 1016/s0025-6196(12)61578-x. [2] Andrade RJ, Lucena MI, Fernández MC, Pelaez G, Pachkoria K, García-Ruiz E, [11] Dourakis SP, Saramadou R, Alexopoulou A, Kafiri G, Deutsch M, Koskinas J, et al. Drug-induced liver injury: an analysis of 461 incidences submitted to et al. Hepatic granulomas: a 6-year experience in a single center in Greece. the Spanish registry over a 10-year period. Gastroenterology 2005;129: Eur J Gastroenterol Hepatol 2007;19:101–104. doi: 10.1097/01.meg. 512–521. doi: 10.1016/j.gastro.2005.05.006. 0000243882.09820.d2.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 280–283 283 Editorial Board

Avin Aggarwal Kazuhiko Hayashi Qing Mao Verona, Italy Tucson, USA Nagoya, Japan Chongqing, China Piero Luigi Almasio Song-Bing He Matthew McMillin Rolf Teschke Palermo, Italy Suzhou, China Temple, USA Hanau, Germany Gianfranco D. Alpini Peng Hu Fanyin Meng Claudio Tiribelli Bryan, USA Chongqing, China Temple, USA Rome, Italy Moustafa El Awady Jing Hua Ahmed Mesalam Vladimir Maximovich Tsyrkunov Cairo, Egypt Shanghai, China Cairo, Egypt Grodno, Republic Belarus Sina Aziz Yan Huang Albert D. Min Zhengkun Tu Karachi, Pakistan Changsha, China San Diego, USA Changchun, China Gyorgy Baffy Yuehua Huang Paul Naylor David Victor Boston, USA Guangzhou, China Detroit, USA New York, USA Mahmoud Mohamed Bahgat Trana Hussaini Calvin Q. Pan Genshu Wang Cairo, Egypt Vancouver, Canada New York, USA Guangzhou, China Fernando Bessone Hartmut Jaeschke James S. Park Leyi Wang Rosario, Argentina Kansas City, USA New York, USA Columbu, USA Kalyan Ram Bhamidimarri Wasim Jafri María Teresa Pérez-Gracia Yongning Xin Miami, USA Karachi, Parkistan València, Spain Qingdao, China Peter Buch Tatsuo Kanda Cyriac Abby Philips Ming Yan Heidelberg, USA Chiba, Japan Palarivattom, India Jinan, China Chalermrat Bunchorntavakul Lindsey Kennedy Nikolaos T. Pyrsopoulos Dongliang Yang Bangkok, Thailand Temple, USA Newark, USA Wuhan, China Phunchai Charatcharoenwitthaya John Koskinas Alok Ranjan Li Yang Bangkok, Thailand Athens, Greece Washington, USA Cincinnati, USA Enqiang Chen Anastasios Koulaouzidis Sammy Saab Tian Yang Chengdu, China Edinburgh, USA Los Angeles, USA Shanghai, China Po-Hung Chen Anand V. Kulkarni Behnam Saberi Eric M. Yoshida Baltimore, USA Hyderabad, India Baltimore, USA Vancouver, Canada Ashok Kumar Choudhury Ashish Kumar Regina Santella Samar Samir Youssef New Delhi, India New Delhi, India New York, USA Cairo, Egypt Jianqiang Ding Manoj Kumar Moinak Sen Sarma Yufeng Yuan Foshan, China New Delhi, India Lucknow, India Wuhan, China Yuchen Fan Jun Li Keqing Shi Maysaa El Sayed Zaki Jinan, China Hangzhou, China Wenzhou, China Cairo, Egypt Jiangao Fan Bing Liu Gamal Shiha Yuanyuan Zhang Shanghai, China Guangzhou, China Mansoura, Egypt Chengdu, China Eduardo Fernández-Martínez Xiangming Lao Surajit Sinha Xinxin Zhang Hidalgo, Mexico Guangzhou, China Rockville, USA Shanghai, China Heather L Francis Chaohong Liu Coleman Smith Minghua Zheng Bryan, USA Wuhan, China Washington, USA Wenzhou, China Catherine Frenette Chenghai Liu Robert Smolic Shicheng Zheng La Jolla, USA Shanghai, China Osijek, Croatia Chengdu, China George Boon-Bee Goh Manqing Liu Martina Smolic Huiping Zhou Singapore Wuhan, China Osijek, Croatia Richmond, USA Yanhang Gao Fengmin Lu Jonathan G. Stine Hong Zhou Jilin, China Beijing, China Charlottesville, USA Nanjing, China Ahmet Gurakar Mingqin Lu Qingfeng Sun Baltimore, USA Wenzhou, China Wenzhou, China Steven-Huy Bui Han Mazyar Malakouti Giovanni Targher Los Angeles, USA San Antonio, USA Ying Han Alessandro Mantovani Xi’an, China Verona, Italy JCTH