Between 1982 and 1987, selective regional cancer chemotherapy using Lipiodol plus an anticancer drug (lipiodolization) was prescribed for 200 patients with hepatocellular carcinoma. One hundred forty-nine patients were given lipiodolization alone, and the remaining 51 underwent hepatic resection following lipiodolization. The grades of deposits of the oily contrast medium in the neoplastic tissue seen on the plain X-ray correlated well with the antitumor effect. In the resected specimens of 17 patients treated with lipiodolization prior to surgery, concentrations of adriamycin in the malignant liver tissues were 13.2 ± 18.2 μg per gm, whereas the adjacent liver parenchyma contained 1.4 ± 2.0 μg per gm of adriamycin; the difference was statistically significant (p < 0.05). In the 149 patients treated with lipiodolization, 1-, 2-, 3- and 4-year survival rates were 56.1, 28.9, 17.3 and 7.4%, respectively. Thus, lipiodolization was considerably more effective, compared to the results achieved by hepatic artery ligation and cannulation into the hepatic artery for patients in Stages I and II. In this sequential nonrandomized study, the survival rates for patients undergoing hepatic resection were superior to those for patients in Stage I and treated with lipiodolization. The significant difference appeared to depend on incomplete killing of tumor cells, which are most often present in the fibrous capsule, by lipiodolization. We conclude from these data that lipiodolization is an effective treatment for hepatocellular carcinoma when the tumor is not curatively resectable. When the clinical status is good, then surgery is warranted.