Comparison of tumor pathology with duration of survival of North American patients with hepatocellular carcinoma.

BACKGROUND: Several of the large studies addressing prognosis and survival from hepatocellular carcinoma (HCC) have been from Europe and Asia, but only a few have emanated from North America. Survival statistics from other parts of the world may not be applicable to the North American population because of etiologic, demographic, cultural, lifestyle, and intangible differences. The current study investigated the survival experience and histologic correlates of North American patients with HCC and compared findings with similar studies from North America and other parts of the world.
METHODS: One thousand sixty-three patients examined during a 14-year period, met inclusion criteria for this study. Each patient was placed in one of three categories based on the duration of survival from date of diagnosis of HCC. Each tumor was examined histologically and classified according to World Health Organization criteria. Patient's sex and age at diagnosis were obtained from case records. Survival analyses and comparisons were performed using appropriate methods. Variables were tested for association using chi-square tests and randomization tests as appropriate.
RESULTS: Age, sex, tumor growth pattern, Edmondson and Steiner's nuclear grades, mitotic index, and presence or absence of tumor giant cells or portal venous invasion, were found to have statistically significant (P < 0.05) relationships with the duration of patient survival. Significantly better survival was found to be associated with female sex, low nuclear grade, low mitotic index, age at diagnosis younger than 50 years, absence of giant cells, and absence of portal venous invasion.
CONCLUSIONS: Certain histopathologic features of HCC may be useful for predicting patient survival and, thus, for empiric prognostication of these patients.