Combined transcatheter arterial chemoembolization and local radiotherapy of unresectable hepatocellular carcinoma

PURPOSE: The best prognosis in hepatocellular carcinoma (HCC) can be achieved with surgical resection; however, the number of resected cases are limited due to advanced lesions or associated liver disease. The purpose of this study was to investigate the efficacy and toxicity of a prospective trial of combined transcatheter arterial chemoembolization (TACE) and local radiotherapy (RT) in unresectable HCC. METHODS AND MATERIALS: Patients with histologically proven unresectable HCC due to either advanced lesions or associated cirrhosis were eligible. From March 1992 to August 1994, 30 patients were entered into this study. TACE was performed with Lipiodol (5 ml) and doxorubicin (Adriamycin ; 50 mg), followed by gelatin sponge particle (Gelfoam) embolization. Local RT was started within 7-10 days following TACE. Mean tumor dose was 44.0+/-9.3 Gy in daily 1.8 Gy fractions. Response was assessed by computerized tomography (CT) scan 4-6 weeks following completion of the treatment and then at 1-3-month intervals. Survival was calculated from the start of TACE using the Kaplan-Meier method. RESULTS: An objective response was observed in 19 patients, giving a response rate of 63.3%. Distant metastasis occurred in 10 patients, with 8 in the lung only and 2 in both lung and bone. Survival rates at 1, 2, and 3 years were 67%, 33.3%, and 22.2%, respectively. Median survival was 17 months. There were 6 patients surviving more than 3 years. Toxicity included transient elevation of liver function tests in all patients, fever in 20, thrombocytopenia in 4, and nausea and vomiting in 1. There was no treatment-related death. CONCLUSION: Combined TACE and local RT is feasible and tolerable. It gives a 63.3% response rate with median survival of 17 months. We feel that this regimen would be a new promising modality in unresectable HCC. Further study is required to compare the therapeutic efficacy of this regimen to TACE alone.