We read with interest the article by Yeo et al. published in a recent issue of HEPATOLOGY.1 The authors investigated the prognostic predictors in 125 patients with hepatocellular carcinoma (HCC) undergoing systemic chemotherapy. Interestingly, among all the clinical variables, the bilirubin level, the presence of a hepatitis C virus infection, and the hepatitis B virus (HBV) DNA load were significant independent predictors determining survival. The amount of serum HBV DNA has been recently identified as an important risk factor linked to the long-term outcome of patients with a chronic HBV infection. However, some of the results in this article are debatable and could raise concern if the HBV viral load can truly have an impact on the survival of HCC patients.
First, tumoral factors, such as the number and size of the tumors and the presence of vascular invasion, have been shown to predict the survival of HCC patients in many reports. Surprisingly, the prognostic significance of tumoral factors is not established in the current study by Yeo et al.1 One of the possible explanations is that the size and number of the tumors and other important clinical staging systems, such as the Cancer of the Liver Italian Program system and the Barcelona Clinic Liver Cancer system, which has been suggested as the primary staging system for HCC,2–4 were not included in the survival analysis. This strategy might introduce substantial statistical bias, so less influential prognostic predictors may prevail in the statistical model. Second, to confirm a direct link between the HBV viral load and mortality, the cause of death, which was not reported in the current study, should be analyzed. In most circumstances, patients die of tumor progression when no active treatment modality is given. This deserves special attention because patients in this study predominantly belonged to an intermediate or advanced cancer stage. In fact, as shown in figure 3 in the article by Yeo et al., there was no significant survival difference between patients with and without “severe hepatitis”, and this suggested that there was no direct relationship between a hepatitis flare and mortality. Lastly, about half of the study patients underwent combination chemotherapy, which combined interferon and a traditional chemotherapeutic regimen. Given that these patients had received an interferon treatment, the effect seemed minimal, without any apparent benefit of suppressing HBV viral activity. It would be interesting to know if the patients receiving the combination therapy had a lower HBV viral load and less severe hepatitis in comparison with those receiving single-agent chemotherapy during or after the treatment. In summary, this study does not provide convincing evidence that an increased HBV viral load can induce a worse outcome in HCC patients. Tumoral factors, which are stronger prognostic predictors, should be incorporated into the survival analysis for competition. If most patients eventually die of cancer death but not a hepatitis flare, the association of the viral load and patient survival cannot be ascertained.