Background: Post-recurrence survival (PRS) of children and adolescents with malignant germ cell tumors (MGCTs) has not previously been described, due to the limited sample size of relapsed patients in clinical trials. Methods: Data from 7 pediatric extracranial GCT trials conducted by the Children’s Oncology Group (COG, United States) or the Children’s Cancer and Leukemia Group (CCLG, United Kingdom) between 1985-2009 were merged. The COG and CCLG trials differed mainly in the type of platinum agent used for upfront treatment (cisplatin or carboplatin, respectively), and event-free survival after either treatment were not significantly different, as previously reported. Kaplan-Meier survival curves and Cox regression were used to analyze the effects of potential predictor variables on PRS. Results: Among 1,107 children in the pooled dataset, 700 MGCT patients received treatment with platinum-based chemotherapy. 72 of these patients experienced a relapse or refractory disease. For these 72 patients, the 5-year PRS was 38% (95% confidence interval (CI) 26% - 49%). The 5-year PRS was significantly higher for 29 children who had been treated upfront with carboplatin regimens (51% , CI 31% - 67%) compared to 43 children treated upfront with cisplatin regimens (28%, CI 15% - 43%; log-rank p=0.046). PRS was not significantly associated with age at diagnosis, tumor site, stage, initial AFP, or time to relapse. In a multivariable model where the known prognostic factors for upfront pediatric GCT (age, stage, and site) were added, the effect of chemotherapy regimen remained significant (hazard ratio 2.06, CI 1.03 - 4.10; p=0.041). Data on post-recurrence treatment regimens were not available. Conclusions: Though limited by the lack of data on type of salvage therapy, we observed a significantly higher PRS after upfront carboplatin regimens compared to upfront cisplatin regimens. This observation should be explicitly tested in a prospective trial. This finding also suggests that overall survival, rather than relapse-free survival, should be utilized as the primary end-point in reduction-of-therapy trials for treatments with differences in PRS.