 |
|
|
|
|
eViralHepatitis Review
VOLUME
3, ISSUE 3
SHOULD THIS PATIENT BE TREATED? BEST PRACTICES IN IDENTIFYING CIRRHOSIS AND HCC IN PATIENTS INFECTED WITH HBV – ROLE OF PREDICTION MODELS
In this Issue...
The goal of antiviral treatment is to prevent progression to cirrhosis, liver failure, and hepatocellular carcinoma and can be achieved with sustained suppression of HBV replication and hepatic inflammation. Ability to predict the risk of disease progression leading to HCC can help to identify patients who need antiviral treatment or enhanced HCC surveillance.
In this issue we focus on recent investigations into predictive models for cirrhosis and HCC in patients with HBV, including reports on the utility of HBV surface antigen levels, host and virus profiles, noninvasive tests for fibrosis, and determining HCC risk scores for patients receiving antiviral therapy.
|
|
|
|
|
LEARNING
OBJECTIVES |
|
| |
|
After participating in this activity, the participant will demonstrate the ability to:
 |
|
Identify which patients with chronic hepatitis B virus (HBV) infection are at high risk for cirrhosis and hepatocellular carcinoma (HCC)
|
|
|
|
Outline the utility and limitations of prediction models for cirrhosis and HCC in patients with chronic HBV infection |
|
|
|
Describe how to incorporate predicted risk of HCC into decisions regarding hepatitis B treatment |
The Johns Hopkins University School of Medicine takes responsibility for the content, quality, and scientific integrity of this CME activity. |
|
| |
|
|
| |
|
| IMPORTANT CME/CE INFORMATION |
 |
|
|
|
|
|
Accreditation Statements
Physicians
This activity has been planned and implemented in accordance with the Essential Areas and Policies of the Accreditation Council for Continuing Medical Education through the joint sponsorship of The Johns Hopkins University School of Medicine and The Institute for Johns Hopkins Nursing. The Johns Hopkins University School of Medicine is accredited by the ACCME to provide continuing medical education for physicians.
Nurses
The Institute for Johns Hopkins Nursing is accredited as a provider of continuing nursing education by the American Nurses Credentialing Center's Commission on Accreditation.
The Institute for Johns Hopkins Nursing and the American Nurses Credentialing Center do not endorse the use of any commercial products discussed or displayed in conjunction with this educational activity.
credit designations
Physicians
eNewsletter: The Johns Hopkins University School of Medicine designates this enduring material for a maximum of 1.0 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
Podcast: The Johns Hopkins University School of Medicine designates this enduring material for a maximum of 0.5 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
Nurses
eNewsletter: This 1 contact hour Educational Activity is provided by the Institute for Johns Hopkins Nursing. Each Newsletter carries a maximum of 1 contact hour, or a total of 6 contact hours for the six newsletters in this program. To obtain contact hours, you must complete this Education Activity and post-test before August 27, 2015.
Podcast: These paired 0.5 contact hour Educational Activities are provided by the Institute for Johns Hopkins Nursing. Each podcast carries a maximum of 0.5 contact hour, or a total of 3 contact hours for the 6 podcasts in this program. To obtain contact hours, you must complete this Education Activity and post-test before July 29, 2015.
successful completion
To successfully complete this activity, participants must read the content, then link to The Johns Hopkins University School of Medicine's or The Institute for Johns Hopkins Nursing's website to complete the post-test and evaluation. Once you receive a passing grade, you can access and print your certificate of credit.
NOTE: If you have already registered for other Hopkins CME programs on their prospective websites simply enter the requested information when prompted.
There are no fees or prerequisites for this activity.
This activity is supported by educational grants from AbbVie, Inc., Boehringer Ingelheim Pharmaceuticals, Inc. and Genentech, Inc.
Launch Date
August 27, 2013; activities expire 2 years from the date of publication.
Internet CME Policy
The Office of Continuing Medical Education (CME) at the Johns Hopkins University School of Medicine is committed to protecting the privacy of its members and customers. The Johns Hopkins University SOM CME maintains its Internet site as an information resource and service for physicians, other health professionals, and the public.
Continuing Medical Education at the Johns Hopkins University School of Medicine will keep your personal and credit information confidential when you participate in an Internet-based CME program. Your information will never be given to anyone outside of the Johns Hopkins University School of Medicine's CME program. CME collects only the information necessary to provide you with the services you request.
Disclaimer Statement
The opinions and recommendations expressed by faculty and other experts whose input is included in this program are their own. This enduring material is produced for educational purposes only. Use of The Johns Hopkins University School of Medicine name implies review of educational format design and approach. Please review the complete prescribing information of specific drugs or combination of drugs, including indications, contraindications, warnings, and adverse effects before administering pharmacologic therapy to patients.
|
|
|
|
|
|
|
Statement of Responsibility
The Johns Hopkins University School of Medicine takes responsibility for the content, quality, and scientific integrity of this CME activity.
Statement of Need
HCV |
 |
Clinicians do not adequately identify which of their patients are at highest risk for HCV infection or effectively interpret testing results. |
|
Clinicians need to understand best practices in how to identify and manage HCV treatment-related side effects. |
|
Clinicians need improved awareness of how newly emerging therapies impact therapeutic decision-making in HCV infected and HIV/HCV co-infected patients. |
|
Clinicians are unaware of new non-invasive techniques to stage liver disease. |
|
Clinicians need to understand best practices in identifying cirrhosis and HCC in patients infected with HCV. |
HBV |
|
Clinicians do not effectively identify their patients at risk for HBV. |
|
Clinicians are unaware of new non-invasive techniques to stage liver disease. |
|
Clinicians need to understand best practices in identifying cirrhosis and HCC in patients infected with HBV. |
Intended Audience
The target audience (clinicians) for this initiative includes: OB/GYNs, NPs, PAs, hepatologists, gastroenterologists, infectious disease physicians, community gastroenterologists and others who care for patients of Asian and West African descent in areas of high HBV prevalence.
Planner Disclosure
As a provider approved by the Accreditation Council for Continuing Medical Education (ACCME), it is the policy of the Johns Hopkins University School of Medicine Office of Continuing Medical Education (OCME) to require signed disclosure of the existence of financial relationships with industry from any individual in a position to control the content of a CME activity sponsored by OCME. Members of the Planning Committee are required to disclose all relationships regardless of their relevance to the content of the activity. Faculty are required to disclose only those relationships that are relevant to their specific presentation. The following relationships have been reported for this activity:
 |
Mark S. Sulkowski, MD, discloses that he has served as a consultant for AbbVie, Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Bristol Myers-Squibb, Gilead, Janssen, Merck and Vertex Pharmaceuticals Incorporated. He has received grant/research funding from AbbVie, Inc., Boehringer Ingelheim Pharmaceuticals, Inc., Bristol Myers-Squibb, Gilead, Janssen, Merck and Vertex Pharmaceuticals Incorporated, and has served on a steering committee for Pfizer, Inc |
|
Raymond T. Chung, MD, discloses that he has served as a consultant for AbbVie, Inc. and Idenix and has received grant/research funding from Gilead and Mass Biologics.
|
| No other planners have indicated that they have any financial interest or relationships with a commercial entity. |
|
| Guest
Authors Disclosures |
Confidentiality Disclaimer for Conference Attendees
I certify that I am attending a Johns Hopkins University School of Medicine CME activity for accredited training and/or educational purposes.
I understand that while I am attending in this capacity, I may be exposed to "protected health information," as that term is defined and used in Hopkins policies and in the federal HIPAA privacy regulations (the Privacy Regulations). Protected health information is information about a person's health or treatment that identifies the person.
I pledge and agree to use and disclose any of this protected health information only for the training and/or educational purposes of my visit and to keep the information confidential.
I understand that I may direct to the Johns Hopkins Privacy Officer any questions I have about my obligations under this Confidentiality Pledge or under any of the Hopkins policies and procedures and applicable laws and regulations related to confidentiality.
The contact information is Johns Hopkins Privacy Officer, telephone: 410-735-6509, e-mail: HIPAA@jhmi.edu.
"The Office of Continuing Medical Education at The Johns Hopkins University School of Medicine, as provider of this activity, has relayed information with the CME attendees/participants and certifies that the visitor is attending for training, education and/or observation purposes only."
For CME Questions, please contact the CME Office at (410) 955-2959 or e-mail cmenet@jhmi.edu.
For CME Certificates, please call (410) 502-9634.
Johns Hopkins University School of Medicine
Office of Continuing Medical Education
Turner 20/720 Rutland Avenue
Baltimore, Maryland 21205-2195
Reviewed and Approved by
General Counsel, Johns Hopkins Medicine (4/1/03)
Updated 4/09
Hardware & Software Requirements
Pentium 800 processor or greater, Windows 98/NT/2000/XP/7 or Mac OS 9/X, Microsoft Internet Explorer 5.5 or later, 56K or better modem, Windows Media Player 9.0 or later, 128 MB of RAM, sound card and speakers, Adobe Acrobat Reader, storage, Internet connectivity, and minimum connection speed. Monitor settings: High color at 800 x 600 pixels. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
GUEST
AUTHORs OF THE MONTH |
|
|
|
|
|
|
|
Guest
Faculty Disclosures
Anna Suk-Fong Lok, MD has disclosed that she has received research grants from Bristol-Myers Squibb, Gilead Sciences, Inc. and Merck. She has served on advisory panels for Gilead Sciences, Inc. and Merck.
Suna Yapali, MD has indicated that she has no financial interests or relationships with a commercial entity whose products or services are relevant to the content of her presentation.
Unlabeled/Unapproved Uses
The authors have indicated that there will be no references to unlabeled/unapproved uses of drugs or products.
Program
Directors' Disclosures |
|
|
|
|
|
|
|
|
|
|
TO ACCESS THE
POST-TEST
Step
1.
Review the CE Information and study the educational content.
Step
2.
Click the post-test link at the end of the newsletter.
Step
3.
Follow the instructions to access a post-test.
|
|
|
|
|
|
|
|
|
|
|
|
|
The goal of hepatitis B treatment is to prevent progression to cirrhosis, liver failure, and hepatocellular carcinoma (HCC). As these endpoints take decades to evolve, suppression of HBV DNA to undetectable levels, hepatitis B e antigen (HBeAg) seroconversion to hepatitis B e antibody (anti-HBe), normalization of alanine aminotransferase (ALT) levels, hepatitis B surface antigen (HBsAg) loss, and improvement in histology have been used to determine the efficacy of treatment.
Currently, seven drugs have been approved for treating hepatitis B including two formulations of interferon alfa (IFN-α) [standard IFNα and pegylated (PEG)-IFNα], and 5 nucleos(t)ide analogues (NUCs): lamivudine, telbivudine, entecavir, adefovir dipivoxil, and tenofovir disoproxil fumarate. These therapies suppress but do not eradicate hepatitis B virus (HBV). Thus, most patients require long-term and often lifelong treatment to maintain viral suppression.
The decision to initiate treatment is simple when a patient presents with life-threatening HBV-related liver disease such as acute liver failure, decompensated cirrhosis, or severe exacerbation of chronic hepatitis B, as well as in patients with compensated cirrhosis and detectable HBV DNA. The decision when to start treatment in patients who have no evidence of cirrhosis is difficult because of the fluctuating course of chronic HBV infection.
Professional guidelines from the American, European, and Asian Pacific Liver Associations (AASLD, EASL, APASL) recommend initiation of treatment in noncirrhotic patients with serum HBV DNA > 20,000 IU/mL and ALT > 2X upper limit of normal (ULN) or histologic evidence of moderate to severe inflammation or fibrosis, with lower thresholds in older patients and patients with family history of HCC.1-3
It has been argued that these treatment criteria do not identify all patients who are at risk of cirrhosis or HCC.4, 5 Availability of prediction models for cirrhosis and HCC could help in deciding which patients should receive antiviral treatment. The best known models were derived from the REVEAL-HBV Study.6 The initial model for predicting HCC included age, gender, family history of HCC, alcohol consumption, ALT level, HBeAg status, HBV DNA level, and HBV genotype.7 The correlation coefficients between the observed HCC risk and the predicted risk were greater than 0.90 in all model derivation and validation sets. In the study by Yang et al reviewed in this issue, this model was later simplified and validated in three external cohorts of hospital-based patients in Hong Kong and South Korea. The new model, REACH-B score, included sex, age, HBeAg status, ALT level, and HBV DNA level, and the predicted risk correlated well with the observed HCC risk.
Recently, HBsAg level has been shown to reflect immune control of chronic HBV infection. Two studies by Tseng et al found that HBsAg level was an independent predictor of HCC in patients negative for HBeAg with low or intermediate viral load (HBV DNA < 20,000 IU/mL) but not in HBeAg-positive patients or patients negative for HBeAg with high viral load.8, 9 In the subset of patients who were HBeAg-negative with low viral load (HBV DNA < 2,000 IU/mL) and normal ALT, HBsAg level was a better predictor of HCC than HBV DNA level, a finding likely related to the narrow range of HBV DNA levels in these patients. Furthermore, HBsAg level was a predictor of HBeAg-negative hepatitis, ALT flares and cirrhosis (Tseng and colleagues, reviewed in this issue). These results led to the incorporation of HBsAg level in prediction models of disease progression and HCC for patients negative for HBeAg with low viremia, as pointed out in the reviewed study by Lee et al.
A key question is whether these models can help determine which patients with chronic HBV infection and no evidence of cirrhosis at presentation should be started on antiviral therapy based on predicted risk of cirrhosis and HCC. Although the models described above were derived from large cohorts of patients followed for more than 10 years, they focused on patients at low risk — HBeAg-negative with normal ALT and low viral load. One can argue that prediction models for the patients at highest risk — HBeAg positive or negative with elevated ALT and high viral load — are unnecessary because they should receive antiviral treatment anyway. It would, however, be extremely valuable to have accurate prediction models for patients positive for HBeAg in the immune tolerance phase (high viral load but normal ALT) and for patients negative for HBeAg in the gray zone (moderate viral load and normal or minimally elevated ALT).
These models have other limitations. They were derived from data in Asian patients with predominantly genotype B or C infection, so their applicability to patients in western countries (most of whom acquired HBV infection in adult life and are infected with other HBV genotypes) is unknown. Even among Asian patients, these models do not apply to those younger than 30 or older than 65 and have limited applicability to patients positive for HBeAg. The majority of the prediction models have focused on baseline values; however, given the fluctuating nature of chronic HBV infection, models that incorporate changes during follow-up would be helpful.
The study by Wong et al discussed herein attempted to apply prediction models of patients with HCC currently receiving antiviral therapy. Surprisingly, the risk score at the start of entecavir therapy was more accurate in predicting the risk of HCC than the risk score after two years of entecavir; however, this finding may be limited by the heterogeneity of the patients (some were switched from other treatment to entecavir), short duration of treatment, and small number of events.
In summary, models that can accurately predict the risk of cirrhosis and HCC will be of immense help in deciding which patients who have no immediate indications for treatment should be started on antiviral therapy. The proliferation of models in recent years is an important first step, but most of these models have limitations, and further validation is needed before they can be applied in clinical practice or incorporated into treatment guidelines.
References
1. Lok AS, McMahon BJ. Chronic hepatitis B: update 2009. Hepatology. 2009;50:61-62.
2. EASL clinical practice guidelines: Management of chronic hepatitis B virus infection. J Hepatol. 2012;57:167-185.
3. Liaw YF, Kao JH, Piratvisuth T, et al. Asian-Pacific consensus statement on the management of chronic hepatitis B: a 2012 update. Hepatol Int. 2012;6:531-561.
4. Tong MJ, Hsu L, Chang PW, et al. Evaluation of current treatment recommendations for chronic hepatitis B: a 2011 update. J Gastroenterol. Hepatol 2011;26:829-835.
5. Tong MJ, Hsien C, Hsu L, et al. Treatment recommendations for chronic hepatitis B: an evaluation of current guidelines based on a natural history study in the United States.Hepatology. 2008;48:1070-1078.
6. Chen CJ, Yang HI, Su J, et al. Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B virus DNA level. JAMA. 2006;295:65-73.
7. Yang HI, Sherman M, Su J, et al. Nomograms for risk of hepatocellular carcinoma in patients with chronic hepatitis B virus infection. J Clin Oncol. 2010;28:2437-2444.
8. Tseng TC, Liu CJ, Yang HC, et al. High levels of hepatitis B surface antigen increase risk of hepatocellular carcinoma in patients with low HBV load. Gastroenterology. 2012;142:1140-1149.
9. Tseng TC, Liu CJ, Chen CL, et al. Risk Stratification of Hepatocellular Carcinoma in Hepatitis B Virus e Antigen-Negative Carriers by Combining Viral Biomarkers. J Infect Dis. 2013;Aug;208(4):584-593. Epub 2013 May 8.
|
|
| back
to top |
|
|
|
|
|
|
|
|
|
|
UTILIZATION OF MODELS FOR PREDICTING THE RISK OF HBV-RELATED HCC |
|
|
|
|
Yang HI, Yuen MF, Chan HLY, et al. Risk estimation for hepatocellular carcinoma in chronic hepatitis B (REACH-B): development and validation of a predictive score. Lancet Oncol 2011;12(6):568-574.
(For non-subscribers to this journal, an additional fee may apply to obtain full-text articles.)
|
|
The goal of treatment of hepatitis B is to prevent cirrhosis and hepatocellular carcinoma (HCC). For patients who do not have cirrhosis, practice guidelines recommend antiviral treatment based on hepatitis B virus (HBV) DNA and alanine aminotransferase (ALT) levels; however, other factors contribute to the risk of HCC. Yang et al developed several prediction models for HCC with data obtained from the Risk Evaluation of Viral Load Elevation and Associated Liver disease/Cancer-HBV (REVEAL-HBV) study.1,2 In this community-based study in Taiwan, 3653 hepatitis B surface antigen (HBsAg)-positive persons aged 30-65 years were followed for a median of 12 years. At enrollment, 94% had normal ALT, 85% were negative for hepatitis B e antigen (HBeAg), and 2% had cirrhosis. Factors associated with HCC included: sex, age, alcohol consumption, family history of HCC, HBeAg status, HBV DNA level, ALT level, and HBV genotype.
The authors developed three regression models to predict HCC. The correlation coefficients between the observed HCC risk and the nomogram-predicted risk were greater than 0.90 in all model derivation and validation sets. In this study, the authors developed a simplified model using objective, readily available data from a derivation cohort of 3584 persons without cirrhosis in the REVEAL study. This REACH-B (Risk Estimation for Hepatocellular Carcinoma in Chronic Hepatitis B) risk score has a range of 0-17 and comprised gender, age, ALT, HBeAg status, and HBV DNA level. The predicted HCC risk at three, five, and 10 years was 0-23.6%, 0-47.4%, and 0-81.6% for persons with the lowest and highest risks, respectively. This model was validated in an external cohort of 1505 patients from three hospitals in Hong Kong and South Korea. The area under receiving operating characteristic (AUROC) curves for predicting HCC risk in the validation cohort at three, five, and 10 years was 0.811 (95% confidence interval [CI], 0.790-0.831), 0.796 (95% CI, 0.775-0.816), and 0.769 (95% CI, 0.747-0.790), respectively. Because the validation cohort included a higher percentage (19.4%) with cirrhosis, analysis was repeated after excluding patients with cirrhosis, and AUROC for predicting HCC was slightly higher.
The REACH-B score is a simple-to-use model, but its utility in determining which patients should undergo enhanced HCC surveillance or receive antiviral treatment is unclear. The REACH-B model has several limitations. It was derived from and validated in Asian patients, and its accuracy in patients with different genetic background, HBV genotypes other than B and C, or adult-acquired infection remains to be confirmed. Even among Asian patients, this model cannot be applied to people younger than 30 years or older than 65 years.
References
1. Chen CJ, Yang HI, Su J, et al. Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B virus DNA level. JAMA. 2006;295:65-73.
2. Yang HI, Sherman M, Su J, et al. Nomograms for risk of hepatocellular carcinoma in patients with chronic hepatitis B virus infection. J Clin Oncol. 2010;28:2437-2444. |
|
| back
to top |
|
|
|
|
|
|
|
|
|
|
|
CAN HEPATITIS B SURFACE ANTIGEN LEVELS PREDICT RISK OF DISEASE PROGRESSION AND HCC? |
|
|
|
Tseng CT, Liu CJ, Yang HC, et al. Serum hepatitis B surface antigen levels help predict disease progression in patients with low hepatitis B virus loads. Hepatology 2013;57:441-450.
(For non-subscribers to this journal, an additional fee may apply to obtain full-text articles.) |
|
|
|
A high level of HBV DNA (> 2,000 IU/mL) had been shown to be an independent predictor of cirrhosis, hepatocellular carcinoma (HCC), and liver-related mortality in the Risk Evaluation of Viral Load Elevation and Associated Liver disease/Cancer-HBV (REVEAL-HBV) study, which included predominantly hepatitis B e antigen (HBeAg) negative persons. Tseng et al investigated the association between levels of hepatitis B surface antigen (HBsAg), HBV DNA, and alanine aminotransferase (ALT) with the risk of HCC in a retrospective cohort of 2688 patients followed for a mean of 14.7 years in the Elucidation of Risk Factors for Disease Control or Advancement in Taiwanese Hepatitis B Carriers (ERADICATE-B) Study.1 In this hospital-based cohort, 60.8% of the patients were men, 29.1% had ALT >= 40 U/L, 19.5% were positive for hepatitis B e antigen (HBeAg), 59.7% had HBV DNA level ≥ 2,000 IU/mL, and 59.1% had HBsAg level ≥ 1000 IU/mL. All the patients had a minimum follow-up of three years.
The authors found that HBV DNA and ALT levels were better predictors of HCC than HBsAg level in the overall cohort. In the subset of patients negative for HBeAg who had HBV DNA < 2,000 IU/mL, HBsAg level but not HBV DNA level predicted HCC development. The adjusted hazard ratio (HR) of HCC was 13.7 (95% CI 4.8-39.3) for those with HBsAg level ≥ 1,000 IU/mL vs those with level < 1,000 IU/mL. Subsequently, they examined whether HBsAg level was related to hepatitis activity, hepatitis flares, and cirrhosis in 1068 patients negative for HBeAg who had baseline HBV DNA levels < 2,000 IU/mL. The authors found that the incidence of HBeAg-negative hepatitis was 2.0%/year, and there was an association between HBeAg-negative hepatitis and cirrhosis. Patients who developed HBeAg-negative hepatitis were older, more likely to be male, and more likely to have higher baseline ALT levels. After adjustment for age, sex, ALT, and HBV DNA levels, HBsAg ≥ 1,000 IU/mL was an independent predictor for HBeAg-negative hepatitis, hepatitis flare, and cirrhosis, with HR of 1.7 (95% CI, 1.3-2.3), 2.3 (95% CI, 1.5-3.5), and 4.1 (95% CI, 2.0-8.3), respectively.
The authors concluded that HBeAg-negative patients with HBV DNA < 2,000 IU/mL, ALT < 40 U/L and HBsAg < 1,000 IU/mL had minimal risk of disease progression, which may explain their lower risk of HCC. They further suggested that the monitoring interval for these patients might be increased to 12 months. However, the data were solely obtained from Chinese patients with genotype B and C infection, and the results should be validated in patients infected with other HBV genotypes. In this study, the authors did not investigate whether HBsAg level can further stratify the risk of progressive disease among patients negative for HBeAg who have HBV DNA 2,000 to 20,000 IU/mL. However, in another study,2 they found that HBsAg levels help stratify risk of HCC in patients negative for HBeAg who have intermediate levels of HBV DNA.
References
1. Tseng TC, Liu CJ, Yang HC, et al. High levels of hepatitis B surface antigen increase risk of hepatocellular carcinoma in patients with low HBV load. Gastroenterology. 2012;142:1140-1149.
2. Tseng TC, Liu CJ, Chen CL, et al. Risk stratification of hepatocellular carcinoma in hepatitis B virus e antigen-negative carriers by combining viral biomarkers. J Infect Dis. 2013, Jun 4 [Epub ahead of print]
|
|
|
| back
to top |
|
|
|
|
|
|
|
|
|
|
THE INTEGRATION OF HOST AND VIRUS PROFILES IN PREDICTION MODELS OF CIRRHOSIS AND HCC |
|
|
|
| Lee MH, Yang HI, Liu J, et al. Prediction models of long-term cirrhosis and HCC risk in chronic hepatitis B patients: Risk scores integrating host and virus profiles. Hepatology. 2013; Aug;58(2):546-554 |
|
The Risk Evaluation of Viral Load Elevation and Associated Liver disease/Cancer-HBV (REVEAL-HBV) study had been used to generate prediction models for hepatocellular carcinoma development.1,2 The REACH-B model included sex, age, alanine aminotransferase, hepatitis B e antigen status, and HBV DNA level.3 Recent studies showed that hepatitis B surface antigen (HBsAg) level is an indicator of immune control of HBV infection and an independent predictor of HCC as well as disease progression in patients negative for HBeAg who had low viremia (HBV DNA < 2,000 IU/mL).4-6 In this study, Lee et al used the REVEAL-HBV data to develop predictive models for cirrhosis. In addition, they developed a revised model incorporating HBsAg level to predict HCC.
A total of 3342 persons were included. Patients were randomly divided in a ratio of 2:1 to derivation and validation sets. Diagnosis of cirrhosis was determined by a quantitative scoring system based on ultrasound findings.7 After 39,016 person-years of follow-up, 327 persons developed cirrhosis. Persons who developed cirrhosis were more likely to be older than age 50, male, alcohol consumers, positive for HBeAg, infected with HBV genotype C, and to have higher ALT, HBV DNA > 10,000 copies/mL, and HBsAg > 100 IU/mL. One hundred-sixty-four incident HCC cases were observed. In addition to the risk factors for cirrhosis, family history of HCC was predictive of HCC development. The multivariate-adjusted HR of cirrhosis was 1.68 (95% CI, 1.12-2.54) and 2.20 (95% CI, 1.48-3.27) for serum HBsAg levels 100-999 and ≥ 1000 IU/mL, compared to those with HBsAg < 100 IU/mL. Multivariate adjusted HR of HCC was 2.83 (95% CI, 1.55-5.18) and 4.06 (95% CI, 2.24-7.36) for HBsAg levels 100-999 and ≥ 1000 IU/mL. HBsAg levels were significantly associated with risk of cirrhosis and HCC in the subset of patients negative for HBeAg, particularly those who had HBV DNA levels < 6 log10 copies/mL but not in the subset of those positive for HBeAg.
The authors developed a risk score for cirrhosis comprising age, sex, ALT, and different combinations of HBeAg status/HBV DNA level/HBsAg level/HBV genotype, and a risk score for HCC comprising the same variables plus family history of HCC. The cirrhosis risk model predicted three, five, and 10-year risk of cirrhosis with area under receiving operating characteristic curve (AUROC) of 0.86, 0.86, and 0.83 in the derivation set, and 0.79, 0.80, and 0.82 in the validation set, respectively. The AUROC for prediction of HCC risk at five, 10, and 15 years was 0.89, 0.85 and 0.86 in the derivation set and 0.84, 0.86, 0.87 in the validation set.
This study confirmed that HBsAg level predicts HCC and disease progression; however, the authors did not show whether the current model for predicting HCC is superior to their previous models, which did not include HBsAg level, HBV genotype, and family history of HCC. The novelty of this study is the predictive model for cirrhosis, but the diagnosis of cirrhosis was based on ultrasonography (which is subjective and insensitive for making that diagnosis). The two models discussed in this paper have not been externally validated, and the inclusion of variables that are not widely available may limit their utility in clinical practice.
References
1. Chen CJ, Yang HI, Su J, et al. Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B virus DNA level. JAMA. 2006;295:65-73.
2. Yang HI, Sherman M, Su J, et al. Nomograms for risk of hepatocellular carcinoma in patients with chronic hepatitis B virus infection. J Clin Oncol. 2010;28:2437-2444.
3. Yang HI, Yuen MF, Chan HLY, et al.Risk estimation for hepatocellular carcinoma in chronic hepatitis B (REACH-B): development and validation of a predictive score. Lancet Oncol. 2011; 12(6):568-574.
4. Tseng CT, Liu CJ, Yang HC, et al. Serum hepatitis B surface antigen levels help predict disease progression in patients with low hepatitis B virus loads. Hepatology. 2013; 57:441-450.
5. Tseng TC, Liu CJ, Yang HC, et al. High levels of hepatitis B surface antigen increase risk of hepatocellular carcinoma in patients with low HBV load. Gastroenterology .2012;142:1140-1149.
6. Tseng TC, Liu CJ, Chen CL, et al. Risk stratification of hepatocellular carcinoma in hepatitis B virus e antigen-negative carriers by combining viral biomarkers. J Infect Dis. 2013, Jun4 [Epub ahead of print]
7. Iloeje UH, Yang HI, Su J, et al. Predicting cirrhosis risk based on the level of circulating hepatitis B viral load.Gastroenterology. 2006;130:678-866.
|
|
|
| back
to top |
|
|
|
|
|
|
|
|
|
|
THE PROGNOSTIC VALUE OF NON-INVASIVE TESTS FOR FIBROSIS AND LIVER STIFFNESS TO PREDICT OUTCOME IN PATIENTS WITH CHRONIC HBV INFECTION |
|
|
|
| de Lédinghen V, Vergniol J, Barthe C, et al. Non-invasive tests for fibrosis and liver stiffness predict 5-year survival of patients chronically infected with hepatitis B virus. Aliment Pharmacol Ther. 2013; 37: 979-88. |
|
|
Assessment of liver fibrosis is important in predicting the risk of hepatocellular carcinoma in patients with chronic hepatitis B virus infection. Liver fibrosis may be assessed by histology or noninvasive tests based on quantification of serum biomarkers of fibrosis or by measurement of liver stiffness. While the accuracy of noninvasive tests in predicting fibrosis has been validated in patients with chronic hepatitis C, performing these tests in patients with chronic HBV infection is less well studied and maybe confounded by liver inflammation and HBV replication status. This investigation by de Lédinghen et al compared the prognostic value of liver stiffness measurement, FibroTest, aspartate aminotransferase (AST) to platelet ratio index (APRI), FIB-4 score, and liver biopsy in predicting five-year survival in patients with chronic hepatitis B infection. The authors prospectively enrolled 600 patients from a hospital-based cohort: 64% were men, mean age was 42.5 years, 16.3% were positive for hepatitis B e antigen (HBeAg), and 56.5% had normal alanine aminotransferase levels. Liver biopsy was available in 214 (35.7%) patients. During the course of the study, 24.7% of the patients received antiviral treatment.
Laboratory tests required for FibroTest, APRI, and FIB-4 scores were performed at the time of liver stiffness measurement. Study endpoint was defined as any cause of death, including liver transplantation. After a median of 49.7 months, 29 (4.8%) patients reached study endpoint, including 17 liver-related events and 12 deaths not liver-related. The five-year overall survival was 97.1% in patients with liver stiffness < 9 kPa and 61.5% in patients with liver stiffness > 20 kPa, and 96.8% for FibroTest ≤0.73, and 49.2% for FibroTest > 0.85. FibroTest and liver stiffness had higher prognostic values compared to liver histology, APRI, and FIB-4, even after adjustment for age, necro-inflammation assessed by ActiTest, HBV DNA level, and treatment status.
While this study showed that noninvasive tests of fibrosis may be used to predict survival of patients with chronic HBV infection, it has several limitations. First, the study endpoint was a combination of liver and nonliver- related deaths, and liver-related deaths included both deaths and liver transplantation. Second, the short duration of follow-up limits the number of events and the accuracy of the prediction of the study endpoint. Third, the study focused on noninvasive tests of fibrosis and did not include other factors that had been shown to be predictive of HCC, such as gender, HBeAg status, and ALT level. Fourth, the study population was heterogeneous, with patients who had hepatic decompensation or HCC at baseline lumped together with inactive carriers. Fifth, 24.7% patients received antiviral therapy. Finally, although the authors claimed that Fibrotest and liver stiffness measurement had better prognostic value than liver biopsy, only 35.7% of patients had liver biopsies.
Determining the utility of noninvasive tests of fibrosis in predicting outcomes of patients with chronic HBV infection requires longer follow-up of a large cohort of patients with compensated liver disease and no HCC at enrollment. Other factors that influence disease progression such as ALT, HBeAg status, HBV DNA level, and HBsAg level should also be tested in the prediction models, and the potential impact of antiviral therapy must be accounted for in these models.
|
|
|
| back
to top |
|
|
|
|
|
|
|
|
|
|
SHOULD HCC RISK MODELS BE INCORPORATED INTO CLINICAL PRACTICE GUIDELINES? |
|
|
|
| Chen TM, Chang CC, Huang PT, et al. Performance of risk estimation for hepatocellular carcinoma in chronic hepatitis B (REACH-B) score in classifying treatment eligibility under 2012 Asian Pacific Association for The Study of the Liver (APASL) guideline for chronic hepatitis B patients. Aliment Pharmacol Ther. 2013; 37:243-251. |
|
|
Current treatment guidelines for hepatitis B recommend antiviral therapy for patients with active or advanced liver disease and high serum HBV DNA levels. The decision about when to start treatment should also consider the predicted risk of cirrhosis and HCC. The REACH-B score that predicts the risk of HCC in patients with chronic HBV infection was derived from the community-based REVEAL-HBV Study.1 Chen et al investigated the performance of REACH-B score in determining antiviral treatment eligibility according to the Asian Pacific Association for the Study of the Liver (APASL) 2012 treatment guideline.2
The investigators retrospectively reviewed the data from 4762 consecutive HBsAg-positive patients in the Bureau of National Health Insurance in Taiwan database between January 2006 and May 2012. A total of 904 patients between age 30-65 who did not have cirrhosis, HCC, or HCV coinfection at presentation and who had at least one HBV DNA test result were included. Of these, 438 patients, 71% of patients positive for HBeAg and 41.3% of patients negative for HBeAg met the APASL 2012 guideline criteria for treatment. REACH-B score was higher in the patients who met treatment criteria than those who did not [10.1 ± 1.9 vs 7.8 ± 2.9, 10.5 ± 1.8 vs 6.9 ± 2.6, respectively (P < 0.001)]. The minimum REACH-B score to meet the treatment criteria was 7 for patients positive for HBeAg and 6 for patients negative for HBeAg. Increase in REACH-B score predicted treatment eligibility in patients negative for HBeAg but not in patients positive for HBeAg. Using a cutoff score of 8 in patients positive for HBeAg age < 40 and patients negative for HBeAg age < 45, a cutoff score of 11 in patients positive for HBeAg age ≥ 40 and in patients negative for HBeAg age ≥ 45, the authors found that the sensitivity and specificity of REACH-B score in identifying patients meeting treatment criteria were more than 80% for patients negative for HBeAg. The sensitivity and specificity for patients positive for HBeAg < 40 years were 80% and 74.3%, respectively. For patients positive for HBeAg ≥ 40, the sensitivity was 78.4%, but the specificity of the REACH-B score was only 53.6%. The projected HCC risk estimated by REACH-B score for patients who met the treatment criteria of the APASL guideline at three, five, and 10 years was 0.9%, 2%, and 5.2%, respectively, and 0.2%, 0.5% and 1.2%, respectively for those who did not meet treatment criteria.
This study showed that the majority of patients regardless of HBeAg status, below the age of 40 who are predicted to have a high risk of HCC based on REACH-B score would meet the APASL guideline treatment criteria, but nearly half of the patients positive for HBeAg ≥ 40 years who are predicted to have high risk of HCC based on REACH-B score would be excluded from treatment (according to the APASL guideline). In this study, a large number of patients (66.5% of the entire cohort) were excluded because of the unavailability of a HBV DNA result. Further, the patients included in this study differed from the REACH-B cohort in that a higher percentage of patients were HBeAg positive and a higher percentage had elevated ALT. Another limitation of this study is that it did not consider histological evidence of significant fibrosis as a criterion for treatment (as recommended by the APASL guidelines) in patients who had ALT levels 1-2X ULN, particularly those above age 40.
References
1. Yang HI, Yuen MF, Chan HLY, et al. Risk estimation for hepatocellular carcinoma in chronic hepatitis B (REACH-B): development and validation of a predictive score. Lancet Oncol. 2011;12(6):568-574.
2. Liaw YF, Kao YH, Piratvisuth T, et al. Asian-Pacific consensus statement on the management of chronic hepatitis B: a 2012 update. Hepatol Int. 2012;6:531-561
|
|
|
| back
to top |
|
|
|
|
|
|
|
|
|
|
THE ACCURACY OF RISK SCORES FOR PATIENTS RECEIVING ENTECAVIR |
|
|
| |
| Wong GL, Chan HL, Chan HY, et al. Accuracy of risk scores for patients with chronic hepatitis B receiving entecavir treatment. Gastroenterolology. 2013;144:933-944. |
|
|
|
Most models developed to predict the risk of HCC in patients with chronic HBV infection have been based on patients who never received treatment or who were censored at the start of antiviral treatment. With the availability of safe and efficacious antiviral treatment, patients with moderate or high risk for HCC are expected to be receiving antiviral therapy. Antiviral therapy has been shown to decrease but not eliminate the risk of HCC; therefore, it is important to determine whether any of the previously published models are accurate in predicting HCC in patients receiving antiviral therapy.
In this study, Wong et al investigated the accuracy of HCC risk scores using values at baseline and during treatment, including a retrospective-prospective cohort of chronic hepatitis B patients who received entecavir 0.5 mg daily for at least 12 months. Patients with preexisting HCC and those whose HCC was diagnosed in the first year of entecavir treatment were excluded. The authors examined the performance of three HCC risk scores: CU-HCC score comprising age, albumin, bilirubin level, HBV DNA level, and cirrhosis; GAG-HCC score comprising sex, age, HBV DNA level, and cirrhosis; and REACH-B score consisting of sex, age, ALT level, HBeAg status, and HBV DNA level.1-3 They applied the risk scores at the time of initiation of entecavir treatment and again at year 2 and assessed the three- and five-year HCC risk based on the reported cut-off values in treatment-naive patients.
Of 1531 patients, 47 (2.9%) patients developed HCC during a mean follow-up of 42 months. Overall cumulative incidence rates of HCC at three and five years were 2.9% (95% CI, 2.4% - 3.4%) and 4.3% (95% CI, 3.6% - 5.0%), respectively. Hypoalbuminemia, presence of cirrhosis, and duration of virologic remission less than 24 months were independent predictors of HCC. The sensitivity of CU-HCC, GAG-HCC, and REACH-B scores in predicting HCC was 93.6%, 55.3% and 95.2%, respectively; while the specificity was 47.8%, 78.9%, and 16.5%.
The authors concluded that CU-HCC was superior to the other two models. This is not surprising because there is partial overlap between the original cohort from which CU-HCC was derived and the current cohort. Patients who had risk scores above the cutoff values both at baseline and at year 2 had a higher risk of HCC than those with baseline scores below the cutoff values at both time points. Patients with high baseline CU-HCC score that declined at year 2 had lower risk of HCC than those with persistently high score, but the risk remained higher than for those with low baseline score.
This is the first study showing that HCC risk scores derived in untreated patients are also accurate in predicting risk of HCC in patients receiving antiviral treatment. It is surprising that risk scores derived from values when HBV DNA was suppressed were not more accurate in predicting HCC risk than scores derived from values at baseline. This may be related to the short duration of treatment and the heterogeneity of patients, with 31% having been on other antiviral therapy prior to the start of entecavir treatment. In these latter patients, baseline lab values were obtained after varying durations of treatment with the other antiviral therapy.
References
1. Wong VW, Chan SL, Mo F, et al. Clinical scoring system to predict hepatocellular carcinoma in chronic hepatitis B carriers. J Clin Oncol. 2010;28:1660-1665.
2. Yuen MF, Tanaka Y, Fong DY, et al. Independent risk factors and predictive score for the development of hepatocellular carcinoma in chronic hepatitis B. J Hepatol. 2009;50:80-88.
3. Yang HI, Yuen MF, Chan HL, et al. Risk estimation for hepatocellular carcinoma in chronic hepatitis B (REACH-B): development and validation of a predictive score. Lancet Oncol. 2011;12:568-574.
|
|
|
| back
to top |
|
©
2013 JHUSOM, IJHN and eViralHepatitis Review
Presented by JHUSOM and IJHN in collaboration with DKBmed. |
|
|
|
|
|
|
COMPLETE
THE
POST-TEST
Step
1.
Click on link to download instructions for the post-test and evaluation
 |
|
|
|
|
|
|
