An independent evaluation of modern prognostic scores in a central European cohort of 120 patients with hepatocellular carcinoma.

Background And Objectives: Hepatocellular carcinoma (HCC) is a common cancer worldwide. As prognosis of HCC patients depends not only on tumour extension but also on liver function, TNM staging of HCC is of limited value. The Okuda score incorporating the variables of liver function and tumour extension is used widely. However, among patients with an intermediate Okuda score, survival varies considerably. Several newer scores promise to perform better than the Okuda score in stratifying HCC patients. We therefore tested the ability of several newer scores to predict survival in comparison to the Okuda score in a European cohort of HCC patients.

Patients And Methods: A total of 120 patients with sufficient follow-up data were identified retrospectively among the 130 patients with HCC first seen between 1997 and 2000 in our department. Child-Pugh score, Okuda score, Vienna survival model for HCC (VISUM-HCC) score, Chevret score, Barcelona clinic liver cancer (BCLC) classification and cancer of the liver Italian programme (CLIP) score were calculated. Survival analysis was performed for all eligible patients stratified according to each scoring system. Receiver operating characteristics analysis was performed using six months survival as the outcome measure. Univariate and stepwise logistic regression analyses were performed to identify prognostic factors.

Results: Survival times of HCC patients grouped according to all scores were significantly different. All scores performed similarly to the Okuda score in the receiver operating characteristic analysis. Prognostic factors for survival were albumin concentration and the presence of portal obstruction.

Conclusion: In our central European cohort, there was no advantage of using the newer scores instead of the Okuda score.