Background/purpose: The aim of this study was to retrospectively evaluate and compare the clinical features, treatment strategy, pathology, and outcome of all patients with hepatoblastoma treated at an African hospital over a 31-year period (1970 to 2001).
Methods: Forty patients with hepatoblastoma were divided into 3 groups according to the treatment given. Group I (1970 to 1983, 14 patients) had no protocol therapy; group II (1984 to 1988, 6 patients) received protocol treatment according to Children's Study Group (CCSG) guidelines; group III (1989 to 2001, 20 patients) received SIOPEL protocol therapy. All available clinical, surgical, radiologic, and pathologic data were reviewed and analyzed.
Results: Overall patient survival was as follows: group I, 14%; group II, 50%, and group III, 80%. Deaths in group II were caused by chemotherapy-induced immunosuppression only. Prognostic data for group III showed that all tumor-related deaths could be predicted by identifying multifocal disseminated growth patterns (P =.001) or vascular invasion (P =.001) in resected tumors. Of the 40 diagnostic tumor biopsies performed, 2 significant complications (1 death, 1 intraperitoneal tumor seeding) occurred. Histologic criteria evaluating these biopsies were not predictive of overall survival.
Conclusions: The introduction of protocol therapy has resulted in a marked improvement in survival. Immunosuppression-related sepsis in our setting resulted in unacceptable mortality in patients treated according to CCSG guidelines. A diagnostic biopsy in hepatoblastoma is of value but not without complications. Preoperative chemotherapy followed by complete surgical excision according to International Society of Paediatric Oncology guidelines yields excellent results with a current survival rate of 80%.