SICIENTIFIC BACKGROUND OF THE STUDY

Primary malignant epithelial liver tumours, hepatoblastoma(HB) or hepatocellular carcinoma(HCC) are one of the rarest of solid tumours in childhood with a world wide incidence of 0.5-1.5 per million children. The prognosis for HB has dramatically improved since the introduction in the 1980's of effective chemotherapy - primarily Cisplatin and Doxorubicin - capable of reducing tumour volume and making previously unresectable tumours resectable. For these patients the 3 year overall survival has improved from about 25% to around 70%.

In the first SIOPEL study patients with HB or HCC were treated with 4-6 courses of chemotherapy, Cisplatin and Doxorubicin (PLADO) before surgical resection of the primary tumour was attempted. Using the data collected in this study it has been possible to identify various prognostic factors for patients with HB. Lung metastases at diagnosis and pre-treatment tumour extension (PRETEXT) were statistically significant factors associated with 3 year overall survival. Patients with PRETEXT IV tumours (i.e. tumour in all 4 liver sections ) had the worst survival -68%- of the 4 PRETEXT groups, and patients with lung metastases had an overall survival of 66% compared to 82% for patients without lung metastases.

Based on these findings patients in SIOPEL 2 were treated according to extent of disease at diagnosis. 'Standard risk' patients, those whose tumour was confined to part of the liver, were treated with Cisplatin alone. The rationale for this was to avoid the risk of cardiomyopathy which may occur many years after treatment with Doxorubicin. Those patients whose tumour involved the whole of the liver or had spread beyond the liver, ('high risk') were treated with more intensive chemotherapy, using 3 drugs, in an effort to improve the prognosis. Patients with HCC, who have a poor prognosis were treated with the 'high risk' protocol.

The preliminary results of SIOPEL 2 suggest that Cisplatin alone is effective chemotherapy for 'standard risk' children. The purpose of the randomised study is to prove that Cisplatin alone is less toxic but as effective as PLADO. It is not possible to run a randomised study for the 'high risk' HB and HCC patients because of the small number of patients. The response to chemotherapy and the resection rate in this group of patients will be compared to that seen in similar patients treated with Cisplatin and Doxorubicin in SIOPEL-1.