Prognostic stratification for children with hepatoblastoma: The SIOPEL experience

PurposeTo identify factors relevant to long-term outcome in newly diagnosed hepatoblastoma, and define subgroups for clinical research on tailoring treatment to the individual patient.Patients and methodsBetween 1995 and 2006 the SIOPEL group conducted two clinical trials which established risk-adapted therapy for hepatoblastoma patients. Patients were stratified into high-risk (AFP < 100 ng/mL and/or PRETEXT IV and/or vascular invasion and/or extra-hepatic intra-abdominal disease (V+/P+/E+) and/or metastases) and standard-risk (all others). The hierarchy of these factors plus multifocality, PRETEXT III, AFP > 1,200,000 ng/mL, patient age, platelet count and histology were further explored. The outcome measure was event-free survival (EFS).ResultsIn 541 patients, reduced EFS correlated significantly with AFP < 100 ng/ml (hazard ratio [HR] 4.09, 95% confidence interval 2.16–7.75), AFP ⩾ 1.2 × 106 ng/mL (2.48, 1.47–4.17), metastatic disease (3.02, 2.05–4.44), PRETEXT IV (2.15, 1.19–3.87), multifocality (1.59, 1.01–2.50), age>5 years (2.76, 1.68–4.53); borderline with small cell undifferentiated (SCU) histology (2.29, 95% confidence interval 0.91–5.77); but not with PRETEXT III, age 30–60 months, platelet count or V+/P+/E+. By using the significant factors and SCU to stratify the population, we have identified three distinct prognostic groups: PRETEXT I/II/III, and no other factors, have 3 year EFS of 90%, PRETEXT IV and/or multifocal tumour and/or age > 5 years and/or AFP > 1.2 × 106 have 3 year EFS of 71% and SCU and/or AFP < 100 ng/mL and/or metastatic have a 3 year EFS of 49%.ConclusionPrognostic stratification for clinical research on newly diagnosed hepatoblastoma should take into consideration PRETEXT, metastatic disease, AFP, multifocality, age and SCU histology.