Revisiting the treatment recommendations for chronic hepatitis B #

I read with great interest the work by Tong et al.1 on the evaluation of current treatment recommendations based on a longitudinal cohort of 369 chronic hepatitis B patients followed for a mean of 84 months. Under the published guidelines, in 2003-2007, 40% to 80% of patients who developed hepatocellular carcinoma (HCC) or died of liver-related problems did not fall into the criteria for antiviral treatment. The authors have proposed adding low platelet count and low serum albumin to the guidelines so that up to 100% of patients who develop HCC or die can be identified. I have a different angle of interpretation on the results of this study. Low platelet count and low serum albumin are surrogate markers of portal hypertension and hepatic dysfunction, respectively. Patients who have these laboratory findings are likely suffering from early liver cirrhosis. In the report by Tong et al.,1 five of nine patients who developed HCC were diagnosed as having chronic hepatitis by histology and therefore did not fulfill the recommended treatment criteria. These patients probably had normal alanine aminotransferase (ALT) and/or intermediate hepatitis B virus (HBV) DNA levels (between 10,000 and 100,000 copies/mL). This observation revealed the insufficiency of liver biopsy in determining the true liver histologic staging, particularly when the biopsy core is not long enough.2 It would be interesting to know the length and number of portal tracts of the liver biopsies among patients with histologic chronic hepatitis in this report. Furthermore, 5% to 18% of hepatitis B e antigen–negative chronic hepatitis B patients with normal ALT (<30 IU/L in men and <19 IU/L in women) might already have developed liver cirrhosis.3 In the current report,1 7 of 15 patients with cirrhosis who developed HCC could not be identified by the treatment recommendations. The reason for missing these patients was most likely their low HBV DNA levels (lower than 10,000 copies/mL). Previous large-scale longitudinal studies have confirmed that HBV DNA and liver cirrhosis are independent risk factors of HCC.4, 5 In other words, patients with cirrhosis have a significant risk of developing HCC even when their HBV DNA levels are not high. In the cohort studied by Tong et al.,1 two-thirds of the patients who had HBV DNA levels lower than 10,000 copies/mL in fact had HBV DNA levels lower than 1000 copies/mL. The number of patients with cirrhosis who had undetectable HBV DNA but developed HCC on subsequent follow-up was not reported. It remains dubious whether antiviral treatment can reduce the risk of liver-related complications in patients with cirrhosis with very low viremia. In summary, to improve current treatment recommendations, high-quality liver biopsy of adequate length should be encouraged when there is clinical suspicion of liver cirrhosis (e.g., middle-age patients with low platelet count and/or low serum albumin), regardless of the ALT and HBV DNA levels. Antiviral treatment should be recommended for all patients with cirrhosis with detectable HBV DNA in order to maximize the possible benefit of treatment. Henry L. Y. Chan*, * The Chinese University of Hong Kong, Hong Kong SAR, China.