Is the Search for a Single Comprehensive Staging System for Hepatocellular Carcinoma an Elusive One?

Abstract

Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide and there are more than 500,000 attributable deaths each year.1,2 Accumulating evidence suggests that the incidence is rising.3,4 Most patients with HCC have cirrhosis from chronic viral hepatitis, either hepatitis B or C. The prognosis of HCC depends on both tumor characteristics and residual liver function. The role of tumor biology may also prove to be an important predictor of outcome. Clinical staging systems should ideally group patients together to define prognosis and thereby allow for selection of appropriate treatment options. HCC was initially classified by the Okuda staging system and tumor, node, metastasis staging. However, the Okuda system does not distinguish between early and late stage HCC and the tumor, node, metastasis staging does not employ variables of liver function. Consequently, several staging systems have been subsequently developed including the Cancer of the Liver Italian Program system, the Barcelona Clinic Liver Cancer system, the Tokyo scoring system, and the Japan Integrated Scoring system. However, the optimal staging system for HCC remains controversial and lacks worldwide consensus. In this issue of the journal, Huo et al5 attempt to evaluate a short-term prognostic model by examining 5 HCC staging systems as well as the applicability of the model for end-stage liver disease (MELD) and the addition of sodium to the model (MELD-Na). The authors suggest that MELD and MELD-Na are better prognostic models in predicting short-term mortality for surgical patients. This is conceivable but the tumor-related factors were already removed from the analysis as the patients were deemed eligible for surgical resection. MELD has been validated as an accurate predictor of survival among different populations of patients with end-stage liver disease.6,7 Its application postresection would be based upon the severity of cirrhosis and would have accurate prognostic implications. The potential for recurrence of tumor after resection is high and in one series, the median disease-free survival postsurgery was only 3 months.8 Therefore, it would be of interest to see if the MELD or MELD-Na could be useful as prognostic indicators, if the study was extended for a longer period of time. In this study, only 27% of the patients underwent surgical resection and a very small percentage, 0.7%, underwent liver transplantation. Consequently, it is difficult to make conclusions from this limited data. Majority of patients in the study (73%) received medical therapy, which combined local regional therapies with best supportive care. Patients who have smaller tumors and receive ablative, potentially curative therapies will certainly have better survival than patients with extensive tumor burden who receive palliative care. Therefore, it may have been more informative to separate local regional therapies from supportive care given the potential disparity in prognosis. The authors conclude that the Cancer of the Liver Italian Program and Tokyo systems are better in predicting short-term mortality in nonsurgical and high-risk patients. It would stand to reason that MELD and MELD-Na would perform poorly in this scenario, as these systems do not take tumor characteristics into account. As the authors point out, the results from this cohort of Taiwanese patients with hepatitis B virus-related HCC may not be applicable to patients in other countries or to patients with other etiologies of cirrhosis, such as hepatitis C or alcohol. In addition, the applicability of staging systems for patients with HCC is most likely dependent on the actual parameters employed in the staging system itself and the availability of different treatment modalities. This underscores the need for better understanding of tumor biology and its incorporation potentially into a simple, well-designed staging system that would have widespread use. The applicability of the staging system will then depend on the surgical and local regional therapies that would be available to the patient.