Analysis of pivotal Phase III immunotherapy trials in mCRPC: efficacy outcomes and future implications.
| NCT number | Trial name | Phase | Description | Updated results | Key insight & implication for future development | Reference | |
|---|---|---|---|---|---|---|---|
| 1 | NCT00861614 | CA184-043 | III | Ipilimumab + RT vs. placebo + RT in mCRPC | Median OS: 11.5 months vs. 10.3 months. HR: 0.88, p = 0.049 | The narrow survival benefit, which barely missed statistical significance, suggests CTLA-4 inhibition may only be effective in a subset with more favorable prognoses (e.g., no visceral disease, normal ALP). Future efforts require robust biomarkers for patient selection. | [19] |
| 2 | NCT01057810 | CA184-095 | III | Ipilimumab vs. placebo in mCRPC | Median OS: 29.1 months vs. 28.0 months. HR: 1.09, p = 0.34 | Lack of efficacy in an earlier-line, less heavily pretreated population indicates that a mere “cold” TME is not the only barrier. Intrinsic resistance mechanisms or the immunosuppressive role of continuous androgen signaling may need concurrent targeting. | [19] |
| 3 | NCT02312557 | Single-Arm Phase II Study | II | Pembrolizumab + enzalutamide in mCRPC progressing on enzalutamide alone | Primary endpoint (ORR): Limited activity observed in this biomarker-unselected, post-enzalutamide population | Demonstrated that PD-1 inhibition after ARSI progression has minimal efficacy without patient selection, highlighting the need for biomarkers and questioning combination timing. | [15] |
| 4 | NCT03834519 | KEYLYNK-010 | III | Pembrolizumab + olaparib vs. NHA in mCRPC | Median OS: 16.0 months vs. 14.7 months. HR: 0.95, p = 0.24 | The failure in a biomarker-unselected population confirms that the immunogenic potential of PARP inhibition is likely restricted to tumors with HRR deficiencies. Future trials must be strictly biomarker-driven. | [30] |
| 5 | NCT03834506 | Keynote-921 | III | Pembrolizumab + docetaxel vs. docetaxel in mCRPC | Median OS: 20.1 months vs. 18.9 months. HR: 0.89, p = 0.15 | Unlike in NSCLC, chemo-immunotherapy synergy is not universal. In mCRPC, docetaxel may not induce sufficient immunogenic cell death or may concurrently deplete immune cells, negating the benefit of PD-1 blockade. | [31] |
| 6 | NCT03016312 | IMbassador250 | III | Atezolizumab + enzalutamide vs. placebo + enzalutamide in mCRPC | Median OS: 15.8 months vs. 16.7 months. HR: 1.14, p = 0.21 | The trend toward worse survival in the combination arm raises a critical safety flag. It suggests the potential for antagonism between PD-L1/PD-1 blockade and enzalutamide, necessitating a deeper investigation into their interplay on the immune landscape. | [32] |
| 7 | NCT03338790 | CheckMate 9KD (Cohort B) | II | Nivolumab + docetaxel in chemotherapy-naive mCRPC | Primary endpoint (ORR): Encouraging anti-tumor activity and acceptable safety, warranting further investigation. | This Phase II signal contrasted with subsequent Phase III failures, suggesting chemotherapy combination efficacy may be context-dependent and not sufficient as a broad strategy in mCRPC. | [33] |
NCT: National Clinical Trial; NHA: next-generation hormonal agent monotherapy; OS: overall survival; ORR: objective response rate; mCRPC: metastatic castration-resistant prostate cancer; RT: radiotherapy; ARSIs: androgen receptor signaling inhibitors; HR: hazard ratio; TME: tumor microenvironment; ALP: alkaline phosphatase; HRR: homologous recombination repair. Adapted from Meng et al. [6] with added critical analysis (CC-BY).
DKN: Conceptualization, Writing—original draft, Writing—review & editing, Software, Methodology. VB: Investigation, Software. IJK: Visualization, Investigation. XZ: Supervision, Validation, Writing—review & editing. All authors read and approved the submitted version.
The authors declare that they have no conflicts of interest.
Not applicable.
Not applicable.
Not applicable.
Not applicable.
Not applicable.
© The Author(s) 2026.
Open Exploration maintains a neutral stance on jurisdictional claims in published institutional affiliations and maps. All opinions expressed in this article are the personal views of the author(s) and do not represent the stance of the editorial team or the publisher.