Is Chemoembolisation of Value in Inoperable Primary Hepatocellular Carcinoma

primary treatment for unresectable hepatocellular carcinoma (HCC). In this unit, 185 patients with a new diagnosis of HCC not amenable to surgery were seen between 1988 and 1991. Intended therapy for these patients was chemoembolisation with doxorubicin (60 mg/m(2)) and lipiodol, repeated at six week intervals until it was technically no longer possible o.r until complete tumour response had been obtained. Chemoembolisation was possible in 67 of the 185 (37%). Reasons for exclusion were portal vein occlusion (n=36), decompensated cirrhosis (n 44), distant metastases (n=5), diffuse tumour or unsuitable anatomy (tumour or vasculature) (n=11), patient refusal (n=11), and other (n=11). Patients excluded from treatment survived for a median of 10 weeks (range 3 days-19 months). In patients treated, 18 had small HCC (4cm) and 49 had large or multifocal HCC. Chemoembolisation was carried out a median of two sessions for small and three sessions for large tumours. Ten of 18 patients with small HCC showed a 50% or greater reduction in tumour size. Five of 49 patients with large or multifocal tumours showed a response to treatment. Median overall survival for treated patients was 36 weeks (range 3 days–4 years). One patient has subsequently undergone liver transplantation with no recurrence and minimal residual disease at transplantation. Two other patients are alive three years after chemoembolisation, one with no evidence of recurrent disease. No patient was thought suitable for surgery after their response to chemoembolisation. Chemotherapy related complications were seen in 22%. Complications were significantly more common in patients with larger tumours and poor liver reserve. Five patients died as a result of chemotherapy related complications. In conclusion, only one third of UK patients with unresectable HCC are treatable by chemoembolisation. Results with small tumours are encouraging, with a high response rate and the possibility of surgical intervention in previously inoperable disease. Large tumours, however, show a poor response and significant incidence of side effects, suggesting that this treatment offers little benefit in advanced disease.