RESUMEN
PURPOSE: We describe the results of a risk-adapted, response-based therapeutic approach from the Brazilian GCT-99 study on germ cell tumors. PATIENTS AND METHODS: From May 1999 to October 2009, 579 participants were enrolled in the Brazilian GCT-99 study. Treatment, defined as specific chemotherapy regimen and number of cycles, was allocated by means of risk-group assignment at diagnosis with consideration for stage and primary tumor site. Patients at low risk received no chemotherapy. Patients at intermediate risk (IR) with a good response (GR) received four cycles of platinum and etoposide (PE), for total doses of platinum 420 mg/m(2) and etoposide 2,040 mg/m(2). Patients at IR with a partial response (PR) received three cycles of PE plus three cycles of ifosfamide, vinblastine, and bleomycin. Patients at high risk (HR) with a GR received four cycles of PE and ifosfamide (PEI) at total doses of platinum 420 mg/m(2), etoposide 1,200 mg/m(2), and ifosfamide 30 g/m(2). Patients at HR with a PR received six cycles of PEI. RESULTS: The risk-group distribution was 213 LR, 138 IR, and 129 HR for 480 evaluable patients. Overall survival (OS) and event-free survival (EFS) rates at 10 years were, respectively, 90% and 88.6% in the IR-GR group (n = 126) and 74.1% and 74.1% in the IR-PR group (n = 12). Ten-year rates for the HR-GR group (n = 86) were an OS of 66.8% and an EFS of 62.5%. The HR-PR group (n = 43) had an OS of 74.8% and an EFS of 73.4%. In univariable and multivariable analysis, increased serum lactate dehydrogenase level and histology for a metastatic immature teratoma were prognostic of a worsened outcome. CONCLUSION: Reduction of therapy to two drugs did not compromise survival outcomes for patients in the IR-GR group, and escalation of therapy with PEI did not significantly improve OS and EFS in patients at HR.