RESUMEN
PURPOSE: The aim of this study was: 1. to evaluate treatment results of combined therapy (surgery, postoperative craniospinal radiotherapy with or without chemotherapy) and 2. to assess factors affecting prognosis (extend of tumor removal, involvement of the brain stem, extent of disease, postoperative meningitis, shunt placement, age, sex and time interval from surgery to start of postoperative radiotherapy). PATIENTS AND METHODS: During the period 1986-1996, 78 patients with medulloblastoma, aged 1-22 years (median 8.6 years), were treated with combined modality therapy and 72 of them were evaluable for the study endpoints. Entry criteria were histologically proven diagnosis, age under 22 years, and no history of previous malignant disease. The main characteristics of the group are shown in Table 1. Twenty-nine patients (37.2%) have total, 8 (10.3%) near total and 41 (52.5%) partial removal. Seventy-two of 78 patients were treated with curative intent and received postoperative craniospinal irradiation. Radiotherapy started 13-285 days after surgery (median 36 days). Only 13 patients started radiotherapy after 60 days following surgery. Adjuvant chemotherapy was applied in 63 (80.7%) patients. The majority of them (46; 73%) received chemotherapy with CCNU and Vincristine. The survival rates were calculated with the Kaplan-Meier method and the differences in survival were analyzed using the Wilcoxon test and log-rank test. RESULTS: The follow-up period ranged from 1-12 years (median 3 years). Five-year overall survival (OS) was 51% and disease-free survival (DFS) 47% (Graph 1). During follow-up 32 relapses occurred. Patients having no brain stem infiltration had significantly better survival (p = 0.0023) (Graph 2). Patients with positive myelographic findings had significantly poorer survival compared to dose with negative myelographic findings (p = 0.0116). Significantly poorer survival was found in patients with meningitis developing in the postoperative period, with no patient living longer than two years (p = 0.0134) (Graph 3). By analysis of OS and DFS in relation to presence of the malignant cells in liquor, statistically significant difference, i.e. positive CSF cytology was not obtained, which was of statistical importance for survival (p = 0.8207). Neither shunt placement nor shunt type showed any impact on survival (p = 0.5307 and 0.7119, respectively). Children younger than three years had significantly poorer survival compared to those older than 16 years (p = 0.0473). Although there was a better survival rate in females than in males this was not statistically significant (p = 0.2386). The analysis results of treatment showed that significantly better survival occurred in patients in whom total or subtotal tumor removal was possible (p = 0.0022) (Graph 4). Patients who started radiotherapy within two months after surgery have better survival, but again this was not statistically significant, probably due to the small number of patients receiving delayed radiotherapy (p = 0.2231) (Graph 5). CONCLUSION: Based on this factors standard and high risk group could be defined. Combined chemotherapy should to be investigated particularly for high risk subgroup. Future research should be done to define new therapeutic modalities (gene therapy, compounds active in tumor antiangiogenesis etc).