Immune checkpoint inhibitor–based combination therapy for platinum-resistant or platinum-refractory recurrent ovarian cancer: a systematic review and meta-analysis of prospective trials

Background: The clinical value of immune checkpoint inhibitor (ICI)-based combinations in platinum-resistant or platinum-refractory recurrent ovarian cancer remains regimen-dependent and uncertain. We updated the evidence base and separated nonrandomized response evidence from randomized comparative evidence. Methods: PubMed/MEDLINE, Embase, Web of Science, and the Cochrane Central Register of Controlled Trials (CENTRAL) were searched from inception to 23 July 2026, supplemented by reference and citation searching. Eligible reports were peer-reviewed prospective phase II or III trials of an ICI-containing combination in adults with platinum-resistant or platinum-refractory recurrent epithelial ovarian, fallopian tube, or primary peritoneal cancer. Front-line, maintenance-only, ICI-monotherapy, retrospective, phase I-only, protocol-only, and mixed platinum-status cohorts without extractable resistant/refractory data were excluded. Objective response rate (ORR) and disease control rate (DCR) from compatible nonrandomized cohorts were synthesized as proportions using a logit random-effects model with Paule-Mandel variance estimation and Hartung-Knapp confidence intervals. Randomized hazard ratios were synthesized descriptively and were not pooled. Results: Twenty-one eligible prospective reports were included, comprising 14 nonrandomized reports and 7 randomized trials. Of the 14 nonrandomized reports, 13 provided compatible confirmed response data for the pooled ORR analysis, yielding 112 objective responses among 490 participants and a pooled ORR of 21.9% (95% CI, 13.1%-34.3%; I² = 70.7%). The remaining nonrandomized report was retained for narrative synthesis only because it reported an unconfirmed response. Ten nonrandomized reports contributed 211 disease-control events among 354 participants, yielding a pooled DCR of 59.9% (95% CI, 46.4%-72.1%; I² = 71.7%). The seven randomized trials were summarized separately and were not pooled with the nonrandomized reports. KEYNOTE-B96 demonstrated improved progression-free survival (HR, 0.70; 95% CI, 0.58-0.84) and overall survival (HR, 0.82; 95% CI, 0.69-0.97), whereas JAVELIN Ovarian 200, EORTC 1508, NRG-GY023, and AGO-OVAR 2.29 did not meet their primary efficacy objectives. Conclusions: ICI-based combinations show measurable but highly heterogeneous activity in platinum-resistant/refractory ovarian cancer. The evidence does not support a uniform class effect: a survival benefit is established for pembrolizumab plus weekly paclitaxel in KEYNOTE-B96, whereas several other randomized strategies were negative. Future trials should prioritize regimen-specific validation, biomarker enrichment, and harmonized toxicity reporting.