Background: Data on long-term prognosis of metastatic GCT (mGCT) is scant. The frequency of spontaneous regressions (SRs) is unknown. We aimed to estimate the prognosis of mGCT.
Methods: We searched electronic scientific literature databases and generic Internet from January 1980 to August 2017. After identifying eligible studies we performed descriptive analyses and meta-analyses to estimate overall survival (OS), disease specific survival (DSS) and frequency of SRs in the years before the widespread use of denosumab. We performed pre-specified subgroup analyses of studies published before and after 2000 and of those with more and less than 10 years of follow-up.
Results: After retrieving and combining data from 26 relevant retrospective case-series totaling 242 patients with a median follow-up of 6.9 years, the estimated pooled OS was 86.9% (95% CI 78.0-94.2). Pooled DSS was 88.0% (95% CI 79.7-94.7). SRs were observed in 4.5% of patients. In the subgroup of studies published after 2000 mGCT was the only cause of death of affected subjects. In case-series with a follow-up longer than 10 years pooled DSS was 69.7% (95% CI 25.5-99.8).
Conclusions: To our knowledge this is the first study to derive estimated pooled OS and DSS of mGCT based on a large dataset. SRs were not exceptional phenomena. In a long run the disease could impact in a significant way on the life expectancy of affected subjects.