A multicentre study was sponsored by the International Atomic Energy Agency (Vienna) to assess the safety and efficacy of trans-arterial rhenium-188 HDD conjugated lipiodol (radioconjugate) in the treatment of patients with inoperable hepatocellular carcinoma (HCC). The radioconjugate was prepared by using an HDD (4-hexadecyl 1-2,9,9-tetramethyl-4,7-diaza-1,10-decanethiol) kit developed in Korea, and lipiodol. Over a period of 18 months, 70 patients received at least one treatment of radioconjugate. Some patients were re-treated if there was no evidence of disease progression. The level of radioconjugate administered was based on radiation-absorbed dose to critical normal organs, calculated following a "scout" dose of radioconjugate. The organs at greatest risk for radiation toxicity are the normal liver, the lung and the bone marrow. An Excel spreadsheet was used to determine maximum tolerated activity (MTA), defined as the amount of radioactivity calculated to deliver no more than 12 Gy to lungs, or 30 Gy to liver, or 1.5 Gy to bone marrow. These doses have been found to be safe in multiple trials using external beam therapy, but this has not been confirmed for systemically administered radiopharmaceuticals. Patients were followed for at least 12 weeks after therapy, until recovery from all toxicity. The clinical parameters evaluated included toxicity, response as determined by contrast-enhanced computed tomography, palliation of symptoms, overall survival, performance status (Karnofsky) and hepatic function (Child's classification). Liver function tests, serum alpha-fetoprotein (AFP) levels and complete blood counts were done at each follow-up visit. In the majority of patients, the scout dose studies indicated the radiation absorbed dose to normal liver to be the limiting factor to the treatment dose, while in a few patients dose to lung was the limiting factor. Radiation dose to bone marrow was negligible and was thus not a factor for the MTA calculations. Side-effects were minimal and usually presented as loss of appetite, right hypochondrial discomfort and low-grade fever, even at high levels of administered radioactivity. The symptoms resolved with simple supportive therapy within 3 days of onset. Liver function tests at 24 and 72 h showed no significant changes and complete blood counts at 1 week, 4 weeks and 12 weeks showed no changes (no bone marrow suppression). Sixteen patients were treated in the dose escalation phase of the study, when the activities administered started at 1.8 GBq (50 mCi) and rose to 7.7 GBq (206 mCi). In the efficacy phase of the study a further 54 patients were treated. Both groups of patients are included in this paper. The treatment activity of 188Re-lipiodol administered transarterially ranged from 1.8 to 9.8 GBq (50-265 mCi), with a mean activity of 4.6 GBq (124 mCi). Survival at 3 months was 90%, and at 6 months, 60%; 19% survived for 1 year. Mean survival after treatment in the total treated group of 70 patients was 9.5 months, with a range of 1-18 months. The results of this multicentre study show that 188Re-lipiodol is a safe and cost-effective method to treat primary HCC via the transarterial route. In terms of efficacy, it is potentially a new therapeutic approach for further evaluation by treatment of larger numbers of patients.