Background
Phase 3, double-blind, randomized (1:1), placebo-controlled. N=907 cisplatin-eligible patients with cT2–T4aN0M0 or T1–T4aN1M0 MIBC (≥50% urothelial histology).
Results
Interventions and follow up: Arm A: pembrolizumab + cisplatin/gemcitabine q3w x4, cystectomy + PLND, then adjuvant pembrolizumab q3w x13 (n=453)
Arm B: placebo + cisplatin/gemcitabine, cystectomy, adjuvant placebo (n=454)
Primary endpoint: EFS (investigator)
mFollow up: 60.9mo (range 40.7–78.7)
Results: EFS (60mo): 63.5% vs 61.9%, HR 0.88 (95% CI 0.71–1.10), one-sided P=.1366 (not met)
pCR: 44.4% vs 35.8% (difference 8.9%, 95% CI 2.5–15.1)
OS: median NR in both arms, HR 1.00 (95% CI 0.79–1.27)
Arm B: placebo + cisplatin/gemcitabine, cystectomy, adjuvant placebo (n=454)
Primary endpoint: EFS (investigator)
mFollow up: 60.9mo (range 40.7–78.7)
Results: EFS (60mo): 63.5% vs 61.9%, HR 0.88 (95% CI 0.71–1.10), one-sided P=.1366 (not met)
pCR: 44.4% vs 35.8% (difference 8.9%, 95% CI 2.5–15.1)
OS: median NR in both arms, HR 1.00 (95% CI 0.79–1.27)
Adverse events
Treatment-emergent AEs (any): 99.8% vs 98.7%
Grade ≥3: 80.2% vs 75.1%
Other specifics: NR in abstract
Grade ≥3: 80.2% vs 75.1%
Other specifics: NR in abstract
Conclusions
Adding perioperative pembrolizumab to neoadjuvant cisplatin/gemcitabine and cystectomy did not significantly improve EFS or OS despite a higher pCR rate. Safety was manageable and consistent with prior reports.
Key Limitations
Negative on the primary endpoint; pCR gain did not translate into EFS/OS (limits pCR as surrogate). Cannot exclude small benefit; no biomarker-selected analyses in abstract. Standard of care has since shifted toward EV-pembrolizumab, so comparator is dated.
Clinical Context
Contrasts with NIAGARA (perioperative durvalumab + GC, positive EFS/OS) and with EV-303/KEYNOTE-905 and EV-304 (perioperative EV + pembrolizumab). Supports NIAGARA or EV-based regimens over adding pembrolizumab to GC; no regulatory approval from this trial.
