Clinical Gastroenterology and Hepatology
Volume 11, Issue 11 , Pages 1500-1502, November 2013
Suna Yapali, MD, Anna S. Lok, MD, FRCP
published online 25 July 2013.
The ultimate goal of hepatitis B treatment is to prevent progression to cirrhosis, liver failure, and hepatocellular carcinoma (HCC). This goal can be achieved with sustained suppression of hepatitis B virus (HBV) replication and hepatic inflammation. Current treatment guidelines for hepatitis B recommend antiviral therapy for patients with active or advanced liver disease and high serum HBV DNA levels.1, 2, 3, 4 For patients who do not have cirrhosis, these guidelines recommend that the decision to start treatment or to monitor should be based not only on the stage of liver fibrosis but also on the activity of liver disease and the predicted risk of cirrhosis and HCC. Unlike hepatitis C (HCV) where the vast majority of HCC cases occur in patients with cirrhosis, 3%–47% of HBV-related HCC occurs in the absence of cirrhosis.5, 6, 7 One major difference between HBV and HCV is that HBV is a DNA virus, and HBV DNA can be integrated into the host genome and may have direct oncogenic effect through activation of oncogenic pathways or down-regulation of tumor suppressor pathways.8, 9 Furthermore, ample evidence mainly generated from the population-based Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer-Hepatitis B Virus study shows that high levels of serum HBV DNA are associated with increased risk of cirrhosis, HCC, and liver-related mortality.10
Professional society guidelines agree that treatment should be initiated in non-cirrhotic patients with serum HBV DNA >20,000 IU/mL and alanine aminotransferase (ALT) levels higher than 2 times upper limit of normal (ULN) or histologic evidence of moderate-to-severe inflammation or fibrosis. The cutoff values of ALT, HBV DNA, and the need for liver biopsy in determining treatment indications vary slightly among the guidelines.11 Because HBV DNA levels are generally lower in hepatitis B e antigen (HBeAg)–negative patients, cutoff values for initiating treatment in HBeAg-negative patients are lower than in HBeAg-positive patients. Recognizing that liver biopsy is not performed on all patients with chronic hepatitis B, the guidelines of the American Association for the Study of Liver Diseases (AASLD) and the Asian Pacific Association for the Study of the Liver (APASL) primarily rely on ALT levels to guide treatment decisions.1, 2
For HBeAg-negative patients with HBV DNA 2000–20,000 IU/mL, the AASLD and APASL guidelines recommend treatment for patients with ALT levels higher than 2 times ULN and liver biopsy to guide treatment decisions for patients with ALT level 1–2 times ULN, particularly if they are older than the age of 40. United States Panel Algorithm (USPA) recommends liver biopsy in patients older than age 35–40 who have serum HBV DNA ≥2000 IU/mL and normal ALT levels.4 USPA and AASLD suggest using the updated ULN for ALT, 30 U/L for men and 19 U/L for women, whereas the European Association for the Study of the Liver (EASL) and APASL suggest the traditional definition of ULN for ALT (40 U/L). The EASL guidelines place more emphasis on liver histology; they recommend treatment of patients with ALT above ULN if liver biopsy shows at least a metavir activity grade of A2 (range, 0–3) or a metavir fibrosis score of F2 (range, 0–4).3 The 2012 EASL guidelines indicate that validated noninvasive assessment may replace liver biopsy to stage liver fibrosis. APASL, USPA, and EASL guidelines recommend an HBV DNA threshold of 2000 IU/mL for HBeAg-negative patients, whereas AASLD recommends a cutoff of 20,000 IU/mL. The recommendations of AASLD, APASL, EASL, and USPA are summarized in Figure 1. All guidelines specifically addressed that lower ALT and HBV DNA thresholds should be used in older patients and patients with a family history of HCC. Moreover, because of the fluctuating nature of chronic HBV infection, serial monitoring of HBV DNA is more important than a single measurement in determining the prognosis and the need for therapy.
Figure 1 Recommendations for HBeAg-negative noncirrhotic patients. ULN for ALT is 30 U/L for men, 19 U/L for women. †ULN for ALT is 40 U/L.
In this issue of the journal, Sanai et al12 assessed the accuracy of the guideline-defined thresholds for initiating treatment in HBeAg-negative patients in identifying those with metavir fibrosis stage ≥F2 (range, 0–4). The authors retrospectively reviewed the records of 366 consecutive HBeAg-negative patients who underwent liver biopsy at 4 centers in Saudi Arabia between January 2006 and January 2012. Patients coinfected with HCV, human immunodeficiency virus, or hepatitis D virus, those with other causes of liver disease, decompensated cirrhosis, and patients on immunosuppressive or antiviral therapy were excluded. A criterion for inclusion was the availability of multiple ALT and HBV DNA results before the biopsy. The last 3 serum ALT and HBV DNA levels before the liver biopsy were used to determine whether each patient did or did not meet guideline threshold for initiating antiviral therapy. Patients needed to have 3 normal ALT values or 3 HBV DNA <20,000 IU/mL to be considered normal or below threshold and 2 elevated ALT values or 2 HBV DNA ≥20,000 IU/mL to be considered above threshold. Different assays and different ULNs were used to measure ALT at the 4 study sites; therefore, ALT levels as a ratio of the ULN for each laboratory were used in the analysis. To determine whether the ALT value of any given patient is above or below the updated ULN proposed by Prati et al,13 the authors assumed that the proposed ULNs of 30 U/L for men and 19 U/L for women are equivalent to 0.75-fold and 0.50-fold, respectively, the ULN assigned by any laboratory because the laboratory-assigned ULN in their study was 40 U/L. All the biopsies were staged for fibrosis according to the metavir system by 2 pathologists who were blinded to all clinical information.
The authors found that 53 (14.5%), 173 (47.3%), 171 (46.7%), and 87 (23.8%) patients met the definite treatment criteria of the AASLD, USPA, EASL, and APASL guidelines, respectively. Among 113 patients (30.9%) who had significant fibrosis defined as metavir fibrosis score ≥F2, 79 (69.9%) met at least one guideline definite treatment criterion. The sensitivity of AASLD, USPA, EASL, and APASL definite treatment criteria in predicting metavir ≥F2 was 58.5%, 45.7%, 45.6%, and 56.3%, respectively, and the specificity was 73.8%, 82.4%, 82.1%, and 77.1%, respectively. The authors concluded that in general, the threshold HBV DNA and ALT levels for definite treatment were relatively good at correctly identifying individuals without significant fibrosis, but these threshold values had dismally low positive predictive values and would lead to treatment initiation in a high proportion of individuals without significant disease.
The study by Sanai et al12 has several strengths. The study included a reasonably large number of patients who underwent liver biopsy, and all the patients had more than one ALT and HBV DNA levels before liver biopsy. In addition, all the biopsies were scored by 2 central pathologists who were blinded to clinical data.
However, there are major problems with this study. First and foremost, the study was misguided. The authors mistakenly assumed that ALT and HBV DNA were used as markers of fibrosis to guide treatment decision. This is far from the truth. Although there is no doubt that hepatitis B patients with advanced fibrosis should receive antiviral treatment, treatment is also recommended for patients with high levels of HBV DNA and active liver disease (on the basis of serum ALT or hepatic inflammation). The importance of persistently high viral load and hepatic inflammation on the outcomes of patients with chronic HBV infection is the basis for using HBV DNA and ALT levels as criteria for initiating treatment in the AASLD, USPA, EASL, and APASL guidelines. As discussed earlier, 18%–47% of HBV-related HCC occurs in the absence of cirrhosis, and high levels of HBV DNA have been shown to be a strong predictor of cirrhosis, HCC, and liver-related mortality; therefore, indications for treatment of hepatitis B should not be based solely on liver fibrosis.5, 6, 10
There are other limitations in this study. The authors used 2 or 3 most recent ALT and HBV DNA levels before the biopsy to determine whether definite treatment criteria are met. The interval between the most recent ALT and HBV DNA levels and the liver biopsy was not specified; however, the authors stated that the median (interquartile range) periods of ALT and HBV DNA observations before the biopsy were 21 (8–46) months and 9 (5–18) months, respectively, suggesting that the interval may be quite long in some patients. A long interval between HBV DNA and ALT levels and the biopsy and fluctuations in HBV DNA and ALT levels during that period would lower the accuracy of these markers in predicting liver fibrosis. Indeed, the authors indicated that 22.7% and 15% of patients had HBV DNA levels fluctuating above or below 2000 and 20,000 IU/mL. It is possible that a similar proportion of patients had ALT levels fluctuating above or below the thresholds for initiating treatment. The authors did not indicate how patients with fluctuating HBV DNA or ALT levels were categorized.
An interesting aspect of this study was the adjustment of ALT values reported to the updated ULN proposed by Prati et al.13 The adjustment method assumes that the absolute ALT values obtained by different assays are equivalent, and a ULN of 40 in one laboratory is the same as a ULN of 60 in another laboratory. Neither assumption has been validated.
Determination of fibrosis stage is an important step in the management of hepatitis B; however, fibrosis is not the only indication for hepatitis B treatment. Age of the patient, HBV replication status, activity of liver disease, and family history of HCC should also be considered in deciding whom to treat and when to initiate treatment. Because currently available treatment for hepatitis B does not eradicate the virus, necessitating long durations and in many instances lifelong treatment, we agree with Sanai et al12 that criteria for initiating hepatitis B treatment in guidelines must be carefully weighed to avoid unnecessary treatment.
References
Conflicts of interest This author discloses the following: Anna S. Lok received research grants from Bristol-Myers Squibb, Gilead, and Merck and served as advisor for Gilead, GlaxoSmithKline, and Merck. The remaining author discloses no conflicts.
Funding Suna Yapali received support from Turkish Association for The Study of the Liver. Suna Yapali and Anna S. Lok received support from the Tuktawa Foundation through the Alice Lohrman Andrews Professorship.
PII: S1542-3565(13)01052-5
doi:10.1016/j.cgh.2013.07.012
© 2013 AGA Institute. Published by Elsevier Inc. All rights reserved.
Refers to article:
Accuracy of International Guidelines for Identifying Significant Fibrosis in Hepatitis B e Antigen–Negative Patients With Chronic Hepatitis , 01 July 2013
Faisal M. Sanai, Mohammed A. Babatin, Khalid I. Bzeizi, Fahad AlSohaibani, Waleed Al–Hamoudi, Khaled O. Alsaad, Hadeel Al Mana, Fayaz A. Handoo, Hamad Al–Ashgar, Hamdan AlGhamdi, Abeer Ibrahim, Abdulrahman Aljumah, Abduljaleel Alalwan, Ibrahim H. AlTraif, Hussa Al–Hussaini, Robert P. Myers, Ayman A. Abdo
Clinical Gastroenterology and Hepatology November 2013 (Vol. 11, Issue 11, Pages 1493-1499.e2)Abstract Full Text PDF (754 KB)
Source