Reappraisal of Cisplatin vs Carboplatin Treatment for Solid Tumors: An Umbrella Meta-Analysis.

Importance: Cisplatin and carboplatin are central pillars of chemotherapy in solid tumor oncology, but their comparative efficacy and tolerability remain variably defined in the literature across tumor types and settings. Objective: To quantify and grade the certainty of evidence comparing the efficacy and safety outcomes of cisplatin- and carboplatin-based systemic therapies through an umbrella review of published meta-analyses. Data Sources: Web of Science Core Collection databases from inception through January 11, 2026. Study Selection: Eligible studies were original meta-analyses of patients with solid tumors directly comparing both platinum-based regimens and reporting outcomes consisting of objective response rate, complete response, overall survival (OS), progression-free survival, disease-free survival, mortality or treatment failure ratio, and selected adverse events. Data Extraction and Synthesis: Effect sizes from individual studies were pooled and harmonized as equivalent odds ratios (eORs) through random-effects models. Certainty of evidence was classified with an algorithmic GRADE (Grading of Recommendations, Assessment, Development and Evaluation) approach. The umbrella meta-analysis followed the Preferred Reporting Items for Overviews of Reviews (PRIOR) reporting guidelines. Main Outcomes and Measures: Outcomes were summarized as pooled eORs. Results: Eleven meta-analyses (including 52 unique studies and 12 683 unique patients) were included across 7 tumor types. High-certainty evidence supported a higher objective response rate with cisplatin vs carboplatin in advanced urothelial carcinoma (8 studies; eOR, 1.38 [95% CI, 1.24-1.53]; P < .001) and no progression-free survival difference in advanced ovarian carcinoma (3 studies; eOR, 0.91 [95% CI, 0.81-1.03]; P = .09). Moderate-certainty evidence supported an increased complete response with cisplatin in early-stage cervical carcinoma (3 studies; eOR, 2.03 [95% CI, 1.27-3.24]; P = .02), while 3-year disease-free survival and OS did not show significant differences. In advanced non-small cell lung cancer, 1-year OS (10 studies; eOR, 1.07 [95% CI, 0.89-1.27]) and mortality ratio (14 studies; eOR, 0.99 [95% CI, 0.98-1.01]) did not show differences, with moderate GRADE class. Peripheral neurotoxicity did not differ significantly, with moderate-certainty evidence. Ototoxicity, nephrotoxicity, and emesis were significantly higher with cisplatin, while hematologic toxicity, particularly thrombocytopenia, was significantly higher with carboplatin; heterogeneity was high and certainty was very low. Conclusions and Relevance: In this umbrella meta-analysis comparing cisplatin- and carboplatin-based chemotherapy, the findings clarified the certainty of evidence regarding their comparative efficacy and safety. High-certainty findings were confined to urothelial and ovarian cancers. For most other tumor types, particularly for survival outcomes, estimates suggested no meaningful differences, with moderate to low certainty, highlighting important gaps in high-level evidence.