Systematic Review and Meta-Analysis of the Management of Relapsed Germ Cell Tumours Following Cisplatin-Based Chemotherapy.

INTRODUCTION: Salvage treatment options after cisplatin-based chemotherapy for germ cell tumours (GCT) are either conventional-dose chemotherapy (CDCT) or high-dose chemotherapy (HDCT). However, the optimal sequencing of these strategies remains debated owing to limited comparative data guiding the choice of first-line salvage. Here, we evaluated the efficacy and safety of CDCT and HDCT, focusing on survival rates, objective response rate (ORR) and treatment-related deaths (TRD). METHODS: The protocol was registered on PROSPERO (CRD42024526637). Three databases were searched and reference lists were manually screened. Two reviewers independently screened articles and extracted data, with quality and bias assessed using Risk of Bias in Non-randomised Studies of Interventions (ROBINS-I) tools. Studies of patients with metastatic GCT who relapsed after cisplatin-based chemotherapy, whether receiving first-line salvage or subsequent lines of treatment, were included in the study. In the meta-analysis, 1-, 2- and 3-year survival rates and ORR were pooled using logit transformation, back-transformed to proportions and compared between CDCT and HDCT groups. RESULTS: Out of 2435 studies screened, 65 (5745 patients) met the inclusion criteria. Only nine studies reported International Prognostic Factor Study Group (IPFSG) scores. Median overall survival (OS) was 27 months for CDCT and 21 months for HDCT (p = 0.4035). The 1-, 2-, and 3-year survival rates were higher with CDCT (81.7%, 61.6%, 47.0%) compared with HDCT (70.0%, 57.4%, 39.1%), though not statistically significant. The pooled ORR was 53% with CDCT and 61% with HDCT. TRD were significantly lower with CDCT (3%) compared with HDCT (7.5%) (p = 0.0002). In subgroup analysis of studies evaluating second-line treatment, median OS was 23.9 months for CDCT and 19.4 months for HDCT. CONCLUSIONS: CDCT was associated with higher survival rates compared with HDCT, though the differences were not statistically significant. Larger prospective studies are needed to directly compare these strategies and guide personalised treatment selection in relapsed GCT.