Ad Hoc Analysis Supports Contributions of Both Enfortumab Vedotin and Pembrolizumab in Cisplatin-Ineligible MIBC

Perioperative enfortumab vedotin-ejfv (Padcev) plus pembrolizumab (Keytruda) improved event-free survival (EFS) and overall survival (OS) vs radical cystectomy and pelvic lymph node dissection (RC + PLND) alone in patients with muscle-invasive bladder cancer (MIBC) who were ineligible for or declined cisplatin, according to a descriptive ad hoc analysis of the phase 3 KEYNOTE-905/EV-303 trial (NCT03924895) presented at the 2026 ESMO Congress.¹ Perioperative pembrolizumab monotherapy was associated with numerically better EFS and OS vs surgery alone, and the pattern of outcomes across the 3 arms was consistent with contributions from both agents to the activity of the combination.

In the ad hoc intention-to-treat (ITT) population of patients concurrently randomly assigned to all 3 arms, the median EFS was not reached (NR; 95% CI, 41.2-NR) with enfortumab vedotin plus pembrolizumab (n = 85) vs 28.7 months (95% CI, 13.9-NR) with pembrolizumab alone (n = 83) and 15.7 months with RC + PLND alone (n = 87). The HRs vs control were 0.40 (95% CI, 0.25-0.65) for the combination and 0.71 (95% CI, 0.47-1.07) for pembrolizumab monotherapy. The 36-month EFS rates were 69.5%, 49.6%, and 37.4%, respectively. All comparisons were descriptive, and no formal statistical testing was performed.

The median OS was NR (95% CI, NR-NR) with the combination, 51.0 months (95% CI, 32.0-NR) with pembrolizumab, and 41.7 months with surgery alone, with HRs vs control of 0.48 (95% CI, 0.29-0.80) and 0.76 (95% CI, 0.48-1.20), respectively. The respective 36-month OS rates were 74.5%, 61.4%, and 55.7%.

"Overall, the pattern of efficacy outcomes across all 3 KEYNOTE-905 study arms was consistent with the contribution of both [enfortumab vedotin] and [pembrolizumab] to the observed activity of the perioperative [combination] regimen," presenting author Anders Ullén, MD, PhD, of Karolinska University Hospital and Karolinska Institutet in Stockholm, Sweden, said. "However, and importantly, these observations should be interpreted with caution due to the descriptive ad hoc nature of this analysis."

In the previously reported primary analysis of 344 patients concurrently randomly assigned to enfortumab vedotin plus pembrolizumab or surgery alone, the combination significantly improved EFS (HR, 0.40; 95% CI, 0.28-0.57; P < .001), OS (HR, 0.50; 95% CI, 0.33-0.74; P < .001) and pathologic complete response (pCR) rate (57.1% vs 8.6%; estimated difference, 48.3%; 95% CI, 39.5-56.5; P < .001).¹

Prior data from KEYNOTE-905 supported the November 2025 FDA approval of enfortumab vedotin plus pembrolizumab or pembrolizumab and berahyaluronidase alfa-pmph (Keytruda Qlex) as neoadjuvant treatment followed by adjuvant treatment after cystectomy for patients with MIBC who are ineligible for cisplatin.² The indication was expanded to include all patients with MIBC, irrespective of cisplatin eligibility, in July 2026, backed by data from the phase 3 KEYNOTE-B15/EV-304 study (NCT04700124).³

How was the 3-arm ad hoc analysis designed?

KEYNOTE-905 opened in 2019 with 2 arms, evaluating perioperative pembrolizumab plus RC + PLND vs RC + PLND alone, in patients who were cisplatin ineligible.¹ An enfortumab vedotin plus pembrolizumab arm was added in August 2020, and randomization was updated to 1:1:1. In 2022, eligibility expanded to patients who were cisplatin eligible but declined cisplatin-based chemotherapy; randomization to the pembrolizumab arm stopped in November 2022, and adjuvant nivolumab (Opdivo) was permitted in the control arm when clinically indicated and regionally available.

Eligible patients had clinical stage T2-T4aN0M0 or T1-T4aN1M0 MIBC by central assessment, at least 50% urothelial histology, and an ECOG performance status of 0 to 2, and were cisplatin ineligible per Galsky criteria or declined cisplatin.

Patients received pembrolizumab at 200 mg every 3 weeks for 3 neoadjuvant and 14 adjuvant cycles, with or without enfortumab vedotin at 1.25 mg/kg on days 1 and 8 every 3 weeks for 3 neoadjuvant and 6 adjuvant cycles.

The primary end point was EFS by blinded independent central review; OS and pCR were key secondary end points, and pathologic downstaging (pDS) and safety were additional secondary end points.

The ad hoc ITT population comprised patients randomly assigned between August 2020 and November 2022, with a median time from randomization to data cutoff of 42.0 months (range, 12.9-53.7). The median ages were 73.0, 71.0, and 72.0 years in the combination, pembrolizumab, and control arms, respectively; 94.1%, 96.4%, and 92.0% of patients were cisplatin ineligible, and 18.8%, 25.3%, and 13.8% had an ECOG performance status of 2.

What did the response analyses show?

The pCR rates were 58.8% (95% CI, 47.6%-69.4%;) with enfortumab vedotin plus pembrolizumab, 25.3% (95% CI, 16.4%-36.0%) with pembrolizumab alone, and 10.3% with surgery alone, for estimated differences vs control of 48.6% (95% CI, 35.6%-60.1%) and 15.7% (95% CI, 4.4%-27.4%), respectively. Patients who did not undergo surgery, including those with a clinical complete response after neoadjuvant therapy, were counted as nonresponders.

The pDS rates were 65.9% (95% CI, 54.8%-75.8%), 33.7% (95% CI, 23.7%-44.9%), and 14.9%, respectively, with estimated differences vs control of 51.0% (95% CI, 37.5%-62.5%) and 19.6% (95% CI, 6.9%-32.1%).

What did the safety analysis show?

The median duration of therapy across all phases was 5.3 months (range, 0.3-19.7) with the combination and 6.5 months (range, 0.7-17.3) with pembrolizumab. In the neoadjuvant phase, adverse effects (AEs) of any grade occurred in 95.2% of patients in the combination arm (n = 84) vs 81.7% of those in the pembrolizumab arm (n = 82); grade 3 or higher AEs in 44.0% vs 20.7%; serious AEs in 26.2% vs 19.5%; and AEs leading to discontinuation of any drug in 23.8% vs 9.8%. Two patients (2.4%) in each arm died of AEs during this phase. Drug-related deaths included myasthenia gravis and toxic epidermal necrolysis (n = 1 each) in the combination arm and diabetic ketoacidosis (n = 1) in the pembrolizumab arm.

In the adjuvant phase, grade 3 or higher AEs occurred in 60.0% of patients in the combination arm (n = 45) vs 44.9% of those in the pembrolizumab arm (n = 49), and AEs led to discontinuation of any drug in 37.8% vs 18.4%. Five patients (11.1%) in the combination arm died of AEs during adjuvant treatment; none of these deaths were considered drug related.

The most common all-cause AEs with the combination were pruritus (47.6%), diarrhea (33.3%), anemia (29.8%), decreased appetite (29.8%), and fatigue (29.8%). The most common AEs with pembrolizumab were anemia (34.1%), urinary tract infection (34.1%), pruritus (18.3%), and hypothyroidism (17.1%). Ullén said no new safety signals were identified.

References

1. Ullén A, Danchaivijitr P, Ku JH, et al. Perioperative enfortumab vedotin plus pembrolizumab or pembrolizumab in participants with muscle-invasive bladder cancer who are cisplatin-ineligible: ad hoc analysis from KEYNOTE-905. Presented at: 2026 ESMO Congress; October 23-27, 2026; Madrid, Spain. Abstract 4733O.
2. FDA approves pembrolizumab with enfortumab vedotin-ejfv for muscle invasive bladder cancer. FDA. November 21, 2025. Accessed October 23, 2026. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-pembrolizumab-enfortumab-vedotin-ejfv-muscle-invasive-bladder-cancer
3. FDA approves pembrolizumab or pembrolizumab and berahyaluronidase alfa-pmph each with enfortumab vedotin-ejfv for muscle invasive bladder cancer. FDA. July 10, 2026. Accessed October 23, 2026. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-pembrolizumab-or-pembrolizumab-and-berahyaluronidase-alfa-pmph-each-enfortumab-vedotin