Perioperative Nivolumab Plus Neoadjuvant Chemotherapy Misses Co-Primary End Points in Muscle-Invasive Bladder Cancer

The addition of perioperative nivolumab (Opdivo) to neoadjuvant gemcitabine and cisplatin (GC) did not significantly improve pathologic complete response (pCR) rates or event-free survival (EFS) vs neoadjuvant GC alone in patients with previously untreated muscle-invasive bladder cancer (MIBC), according to final results from the phase 3 CA017-078 (ENERGIZE) trial (NCT03661320) presented at the 2026 ESMO Congress.¹

Although pCR rates were numerically higher with the nivolumab-containing regimen, neither co-primary end point met its prespecified threshold for statistical significance, and overall survival (OS), which was analyzed descriptively, showed no clear difference between the arms. Exploratory analyses identified circulating tumor DNA (ctDNA) status before and during neoadjuvant therapy as prognostic in both arms.

Findings showed that at a median follow-up of 54.5 months (range, 39.4-77.0), the median EFS was not reached (95% CI, not estimable [NE]-NE) with nivolumab plus GC (n = 325) vs 67.6 months (95% CI, 63.0-NE) with GC alone (n = 328; HR, 0.94; 95% CI, 0.72-1.24; P = .6697); the threshold for significance was P < .0426. The respective pCR rates were 39.1% and 29.6%, translating to an absolute difference of 9.5% (95% CI, 2.3%-16.7%; 99.906% CI, –2.6% to 21.6%), which did not establish a significant difference after multiplicity adjustment.

"The analysis in the combined biomarker population corroborated the robustness of circulating tumor DNA [ctDNA] as a prognostic biomarker, either at baseline or after neoadjuvant therapy,” lead study author Andrea Necchi, MD, of IRCCS San Raffaele Hospital in Milan Italy, said during the presentation. “Although the core primary end points were not met, ENERGIZE, like [the phase 3 KEYNOTE-866 trial (NCT03924856)], is practice-informing for me, in particular with regard to the use of chemotherapy or chemotherapy as a backbone therapy for newer combination therapy.”

How was ENERGIZE designed?

ENERGIZE enrolled patients with MIBC of predominant urothelial histology, clinical stage cT2-T4a N0 M0 disease, a glomerular filtration rate (GFR) of at least 50 mL/min, and an ECOG performance status of 0 or 1.

Patients were randomly assigned 1:1 to receive 4 cycles of GC every 3 weeks plus nivolumab at 360 mg every 3 weeks, followed by radical cystectomy and up to 9 cycles of adjuvant nivolumab at 480 mg every 4 weeks; or 4 cycles of GC followed by radical cystectomy and no further treatment. Randomization was stratified by clinical stage, GFR, and PD-L1 status. A third arm evaluating nivolumab and linrodostat with GC was terminated in December 2021 per an independent data monitoring committee recommendation and was excluded from formal testing.

The co-primary end points were pCR rate and EFS per blinded independent central review. OS and safety were secondary end points, and ctDNA analyses were exploratory.

The median age was 66.0 years in the nivolumab arm and 67.0 years in the GC arm, 76.6% and 77.4% of patients, respectively, had cT2 disease, and 47.1% and 47.3%, respectively, had a PD-L1 expression of at least 1%. Radical cystectomy was performed in 85.8% and 86.9% of patients, respectively. Among treated patients in the nivolumab arm (n = 318), 75.2% initiated adjuvant nivolumab, and 55.0% completed it.

What did the additional efficacy data show?

The respective EFS rates with nivolumab plus GC vs GC alone were 82.6% vs 78.6% at 12 months, 72.3% vs 71.6% at 24 months, and 68.1% vs 68.5% at 36 months. Deaths occurred in 21.8% and 24.7% of patients, respectively; the median OS was not reached in either arm (HR, 0.88; 95% CI, 0.64-1.21). The 36-month OS rates were 79.8% and 80.5%, respectively.

Numerically higher pCR rates with nivolumab were observed across the PD-L1–positive (46.4% vs 35.5%) and PD-L1–negative (33.3% vs 24.3%) subgroups, as well as in patients with cT2 (40.6% vs 31.5%) and cT3/cT4a (35.1% vs 23.9%) disease. The combined response rate (≤ypT1N0) was 51.7% for nivolumab plus GC vs 41.8% for GC alone (difference, 9.9%; 95% CI, 2.3%-17.5%). Descriptive EFS and OS subgroup analyses by PD-L1 status, clinical stage, and creatinine clearance showed no clear differences between arms.

A greater proportion of patients in the GC arm received subsequent therapy (23.8% vs 15.7%), including subsequent checkpoint inhibitors (14.9% vs 3.4%). No patients in the GC arm received adjuvant anti–PD-1 or –PD-L1 therapy. "

What did the ctDNA analyses reveal?

Furthermore, ctDNA was assessed with a tumor-informed assay at cycle 1 day 1 and cycle 4 day 1; 53.0% of randomly assigned patients (n = 653) had results at both time points. Among patients who were ctDNA positive at baseline, clearance was achieved in 81.3% with nivolumab plus GC (n = 107) vs 71.7% with GC alone (n = 92; Fisher exact P = .130). Among patients who cleared ctDNA, pCR rates were 43.7% vs 33.3%, respectively (P = .242), compared with 5.0% vs 7.7% among those who did not clear ctDNA.

In pooled analyses across arms, pretreatment ctDNA detection was associated with shorter EFS and OS, and ctDNA clearance was associated with longer EFS and OS. No clear treatment difference between arms was observed regardless of pretreatment ctDNA or clearance status.

"There was no formal hypothesis testing for the biomarker, so the biomarker results are just presented descriptively," Necchi noted.

What did the safety analysis show?

In the safety population (nivolumab plus GC, n = 318; GC, n = 325), grade 3 or higher adverse effects (AEs) of any cause occurred in 73.0% vs 56.3% of patients, respectively, and serious AEs occurred at respective rates of 45.9% vs 18.5%. Treatment-related AEs (TRAEs) occurred in 93.4% vs 93.8% of patients, respectively, including respective rates of grade 3 or higher TRAEs at 50.0% vs 49.5%, despite the longer AE observation period with adjuvant nivolumab. The most common TRAEs in the nivolumab arm were nausea (48.7%), anemia (30.8%), neutropenia (27.7%), and fatigue (25.2%). One treatment-related death (0.3%), due to pneumonitis, occurred in the nivolumab arm.

How do these results fit into the MIBC treatment landscape?

Perioperative immunotherapy has gained ground in MIBC; in both the European Union and United States, the combination of pembrolizumab (Keytruda) plus enfortumab vedotin (Padcev) as neoadjuvant and adjuvant treatment for patients with resectable MIBC, regardless of cisplatin eligibility.² ENERGIZE was one of 2 long-awaited phase 3 trials of perioperative checkpoint inhibition with cisplatin-based chemotherapy presented in the same ESMO session, alongside KEYNOTE-866.³

References

1. Necchi A, Gupta S, van der Heijden MS, et al. Results from a phase 3 trial of neoadjuvant chemotherapy alone vs in combination with nivolumab followed by postsurgical adjuvant nivolumab in patients with muscle-invasive bladder cancer. Presented at: 2026 ESMO Congress; October 23-27, 2026; Madrid, Spain. Abstract 4736O.
2. Pembrolizumab Plus Enfortumab Vedotin Earns EC Approval in Resectable MIBC. OncLive. Published October 21, 2026. Accessed October 23, 2026. https://www.onclive.com/view/pembrolizumab-plus-enfortumab-vedotin-earns-ec-approval-in-resectable-mibc
3. VOLGA, DASL-HiCaP Data Highlight Key Genitourinary Cancer Presentations Arriving at ESMO 2026. OncLive. Published October 16, 2026. Accessed October 23, 2026. https://www.onclive.com/view/volga-dasl-hicap-data-highlight-key-genitourinary-cancer-presentations-arriving-at-esmo-2026