Needle biopsy of hepatocellular carcinoma: Assessment of prognostic contribution of histologic parameters including proliferating cell nuclear antigen labeling and correlations with clinical outcomes

Abstract

BACKGROUND. Although many studies concerning the prognostic factors of resected hepatocellular carcinoma have been performed, the prognostic information provided by needle biopsy of these tumors has not yet been extensively studied. METHODS. One hundred twenty-four white patients with histologically-confirmed hepatocellular carcinoma were entered into a prospective study from 1990 to 1992 with followed-up through October 1994. Most of the patients had cirrhosis (79%). Patients were classified according to Okuda's staging: Stage I (50%), Stage II (47%), and Stage III (3%). The one-year survival rate was estimated at 58% in Okuda's Stage I, and 15% in Stages II and III. Edmondson's grading, mitotic index, and proliferating cell nuclear antigen labeling index were studied on each biopsy. Univariate and multivariate analyses were performed to determine the prognostic significance of the different parameters. RESULTS. According to Edmondson's criteria, 13 of the hepatocellular carcinomas were Grade I, 63 Grade II, 42 Grade III, and 6 Grade IV. There was a significant correlation between Edmondson's grade, mitotic index, and proliferating cell nuclear antigen labeling index (P < 0.01). Univariate analysis showed both the histologic grade (P < 0.001) and mitotic index (P < 0.05) correlated with survival. Cox's model indicated that Edmondson's grading yields additional prognostic information for Okuda's staging (P < 0.001). The one-year survival rate for Okuda's Stage I was 76% for Edmondson's Grades I + II versus 23% for Grades III + IV. CONCLUSIONS.Edmondson's grading appears to be an important prognostic parameter for patients with hepatocellular carcinoma, and could help stratify patients for clincial trials. Cancer 1996;77:864-71.