Abstract
On March 28, 2025, the Food and Drug Administration approved durvalumab (Imfinzi, AstraZeneca) with gemcitabine and cisplatin as neoadjuvant treatment, followed by single agent durvalumab as adjuvant treatment following radical cystectomy (RC), for adults with muscle invasive bladder cancer (MIBC). Substantial evidence of effectiveness was obtained from NIAGARA (NCT03732677), a randomized, phase 3, open-label trial in cisplatin-eligible patients with MIBC who had not received prior systemic chemotherapy or immunotherapy. A total of 1063 patients were randomized (1:1) to receive neoadjuvant durvalumab + gemcitabine and cisplatin prior to RC, followed by adjuvant durvalumab (GC-D), or neoadjuvant gemcitabine and cisplatin (GC) prior to RC, with no subsequent adjuvant treatment. The dual primary endpoints were event free survival (EFS) and pathologic complete response (pCR), both per blinded independent central review (BICR). The key secondary endpoint (alpha-controlled) was overall survival (OS). GC-D demonstrated a statistically significant improvement in EFS compared to GC at the second interim analysis (IA2), with a hazard ratio (HR) of 0.68 (95% CI: 0.56 – 0.82; p < 0.0001). Median EFS was not reached (NR) for GC-D and was 46.1 months (95% CI: 32.3, NR) for GC. There was no statistically significant difference in pCR rate between the arms. A statistically significant improvement in OS was observed for GC-D compared with G+C, with a HR of 0.75 (95% CI: 0.59 - 0.93; 2-sided p = 0.0106). Median OS was NR in both arms. Safety appeared consistent with the safety profile demonstrated in prior trials of durvalumab in combination with platinum-based chemotherapy.
Keywords: durvalumab, muscle invasive bladder cancer, FDA
Introduction
The current standard of care for patients with muscle invasive bladder cancer (MIBC) who are fit for chemotherapy is cisplatin-based neoadjuvant chemotherapy (NAC) followed by radical cystectomy (RC). The combination of NAC and RC compared to RC alone has been shown to increase disease-free survival (DFS) and overall survival (OS) in this patient population (1-3).
Durvalumab is a human recombinant monoclonal IgG1 kappa-isotype antibody that binds to programmed cell death receptor ligand 1 (PD-L1) and blocks its interaction with programmed death cell protein 1 (PD-1), releasing PD-1 pathway-mediated inhibition of the immune response. Durvalumab is currently approved by the U.S. Food and Drug Administration (FDA) for multiple oncology indications.
On March 28, 2025, the FDA approved durvalumab (Imfinzi, AstraZeneca) with gemcitabine and cisplatin as neoadjuvant treatment, followed by single agent durvalumab as adjuvant treatment following RC, for adults with MIBC (4).
The FDA granted this approval based on the results from the randomized NIAGARA trial (NCT03732677) comparing neoadjuvant durvalumab plus gemcitabine and cisplatin prior to RC, followed by adjuvant durvalumab (GC-D) to neoadjuvant gemcitabine and cisplatin (GC) prior to radical cystectomy and no adjuvant treatment (5).
The FDA review team considered the statistically significant improvements in event-free survival (EFS) per blinded independent central review (BICR) and OS and an acceptable safety and tolerability profile to be clinically meaningful in an add-on trial. This article summarizes the FDA’s review and risk/benefit assessment in support of approval of durvalumab for this indication.
Study design
NIAGARA was a randomized, open-label trial that enrolled 1063 cisplatin-eligible patients with MIBC who were candidates for RC and who had not received prior systemic chemotherapy or immunotherapy for treatment of non-muscle invasive bladder cancer (NMIBC) or MIBC. Patients were randomized (1:1) to the GC-D or the GC arms. Randomization was stratified by clinical tumor stage (T2N0 vs > T2N0 [including T2N1, T3 and T4a] to reflect stage II vs IIIa, respectively); renal function (adequate renal function [≥60 ml/min/1.73m2] vs borderline renal function [40-59 ml/min/1.73m2]); and PD-L1 status (high [PD-L1 expression of tumor cell ≥ 25% and/or tumor infiltrating immune cell ≥ 25%] vs low/negative [results below the thresholds for PD-L1 high]).
During the neoadjuvant phase, patients in the GC-D arm with adequate renal function received durvalumab 1500 mg IV plus cisplatin 70 mg/m2 and gemcitabine 1000 mg/m2 on day 1 followed by gemcitabine 1000 mg/m2 on day 8 every 21 days for a total of 4 cycles. Patients with borderline renal function received durvalumab 1500 mg IV plus cisplatin 35 mg/m2 and gemcitabine 1000 mg/m2 on day 1 followed by cisplatin 35 mg/m2 and gemcitabine 1000 mg/m2 on day 8 every 21 days for 4 cycles. The dosages of cisplatin and gemcitabine were the same in the neoadjuvant phase for patients in the GC arm.
For the adjuvant phase, patients in the GC-D arm received durvalumab 1500 mg IV every 28 days for total of 8 cycles, while patients in the GC arm did not receive any further treatment.
Patients on either treatment arm who refused to undergo RC and who had a complete clinical response could enter a non-cystectomy extension. Patients enrolled on the GC-D arm who entered the non-cystectomy extension phase could be administered durvalumab 1500 mg as monotherapy every 28 days for a maximum of 8 doses (corresponding to a maximum exposure of 12 months).
The dual primary endpoints were EFS per BICR and pathologic complete response (pCR) by central review. EFS was defined as time from randomization to the first of these events: recurrence of disease after RC, first documented progression in patients who were medically precluded from RC, time of expected surgery in patients who refused RC, failure to undergo RC in participants with suspected or documented residual disease after neoadjuvant therapy, or death due to any cause. pCR rate was defined as the proportion of patients whose pathological staging was T0N0M0 at RC. OS was a key secondary endpoint. The efficacy analyses of EFS, pCR and OS were performed in the full analysis set, which included all randomized patients.
A multiple testing procedure with gatekeeping strategy was used across the dual primary endpoints and OS to control the overall type I error rate (α) at 5% (2-sided) level for the study. Initially, the 5% overall significance level was assigned to pCR per central pathology review (0.1%) and EFS per BICR (4.9%).The primary endpoint of pCR per central pathology review was tested at one time point (IA1, the planned final analysis of pCR), and the primary endpoint of EFS and the key secondary endpoint of OS were to be tested at two interim analyses (IAs) and the final analysis. The interim and final EFS analyses were planned to be performed at the time of the final pCR analysis (IA1 of the study), when 410 EFS events and 451 EFS events have occurred, respectively. The interim and final OS analyses were planned to be performed at the time of the second interim EFS analysis (IA1), at 318 OS events (IA2), and at 428 OS events (final analysis), respectively.
Efficacy Results
NIAGARA enrolled 1063 patients: 533 on the GC-D arm and 530 on the GC arm. Overall, the baseline characteristics were well-balanced between arms. Key demographic and baseline disease characteristics of the enrolled patients are presented in the supplementary Table S1.
NIAGARA demonstrated a statistically significant improvement in EFS per BICR favoring the GC-D arm over the GC arm at IA2, with HR of 0.68 (95% CI: 0.56–0.82) and p-value < 0.0001. Median EFS was not reached in the GC-D arm and was estimated to be 46.1 months (95% CI: 32.2–not reached) in the GC arm. In the primary analysis of EFS, there were 40 patients (7.5%) in the D+ G+C arm vs. 60 patients (11.3%) in the GC arm who refused or who failed to undergo RC despite the presence of residual disease. A sensitivity analysis was performed in which these patients were censored at the time of expected surgery instead of counting these as EFS events in the primary analysis, as per the protocol definition. The HR based on this sensitivity analysis was 0.69 (95% CI:0.56–0.86), consistent with the primary analysis.
At IA2, 305 deaths occurred (71% information fraction) and a statistically significant improvement in OS was observed for the GC-D arm compared to the GC arm with HR of 0.75 (95% CI: 0.59–0.93) and a 2-sided p-value = 0.0106. Median OS was not reached for either treatment arm. Efficacy results for EFS and OS are shown in Table 1. The Kaplan-Meier (KM) curves of EFS per BICR and OS are provided in Figure 1 and 2, respectively.
Table 1:
Efficacy Results in NIAGARA
| Durvalumab plus gemcitabine and cisplatin (N=533) |
Gemcitabine and cisplatin (N = 530) |
|
|---|---|---|
| EFS * | ||
| Number of events (%) | 187 (35.1) | 246 (46.4) |
| Median EFS (months) (95% CI) | NR (NR, NR) | 46.1 (32.2, NR) |
| HR (95% CI)† | 0.68 (0.56, 0.82) | |
| 2-sided p-value§,# | <0.0001 | |
| Overall Survival (OS) | ||
| Number of events (%) | 136 (25.5) | 169 (31.9) |
| Median OS (months) (95% CI) | NR (NR, NR) | NR (NR, NR) |
| HR (95% CI)† | 0.75 (0.59, 0.93) | |
| 2-sided p-value§,# | 0.0106 | |
Source: table adapted from the label (8); there are no restrictions on its use
Even Free Survival (EFS) is defined as the time from randomization to the first recurrence of disease after radical cystectomy, the time of first documented progression in patients who are medically precluded from a radical cystectomy, or time of expected surgery in patients who refuse to undergo a radical cystectomy or failure to undergo a radical cystectomy in participants with residual disease, or the time of death due to any cause, whichever occurs first.
Based on stratified Cox proportional hazard model.
Based on stratified log-rank test.
Alpha allocated at the interim analysis was 0.0412 for EFS and 0.0154 for OS.
CI = Confidence Interval, HR = Hazard Ratio, NR = Not Reached
Figure 1. Kaplan-Meier Plot of Event-Free Survival in NIAGARA.
Source: FDA generated
D = Durvalumab; G = Gemcitabine; C = Cisplatin
Note: Figure adapted from the label; there are no restrictions on its use
Figure 2. Kaplan-Meier Plot of Overall Survival in NIAGARA.
Source: FDA generated
D = Durvalumab; G = Gemcitabine; C = Cisplatin
At the final analysis for pCR per central pathology review, there was no statistically significant difference in the pCR rate for the GC-D arm (pCR rate: 33.8%, 95% CI: 29.8% - 38.0%) compared with the GC arm (pCR rate: 25.8%, 95% CI: 22.2% - 29.8%). However, per Applicant, there were 59 evaluable pCR assessments that had the wrong data cut-off date. Consequently, these 59 additional patients were all considered non-responders for the pCR calculation in the protocol-specified analysis. The pCR analyses were repeated including the actual results of the 59 pCR assessments, where 28 patients were assessed as having pCR. The updated analysis showed a difference in pCR rate of approximately 10% between the GC-D arm compared to the GC arm (37.3% vs 27.5%). These results were considered exploratory only.
Safety results
The safety population in NIAGARA consisted of 530 patients in the GC-D arm and 526 patients in the GC arm. The FDA analyzed safety data for the overall study, including analyses specific to the neoadjuvant period (randomization until 90 days after the date of surgery or until the first subsequent anticancer therapy), which was the only time on study where therapy was administered to patients in both treatment arms. The comparison of safety findings during the neoadjuvant period therefore allowed for a direct comparison of the safety profile of neoadjuvant GC-D vs. neoadjuvant GC.
The most common adverse reactions, including laboratory abnormalities, in the overall study (occurring in ≥ 20% of patients) were decreased hemoglobin, decreased neutrophils, increased blood creatinine, decreased sodium, nausea, increased ALT, decreased calcium, decreased platelets, fatigue, increased potassium, decreased lymphocytes, increased AST, constipation, decreased magnesium, decreased appetite, increased alkaline phosphate, rash, pyrexia, diarrhea, vomiting, and abdominal pain.
The incidence of all-cause serious adverse events (SAEs) was similar between the two arms. The higher incidence of SAEs of diarrhea, renal dysfunction, and pneumonitis in the durvalumab arm may reflect immune-mediated toxicity.
Overall, safety data from NIAGARA did not suggest that the addition of durvalumab markedly increased the overall toxicity of the treatment regimen in either the neoadjuvant period or in the overall study period (Table 2).
Table 2:
Summary of Safety for NIAGARA
| AE category | Number (%) of patients | |||||
|---|---|---|---|---|---|---|
| Neoadjuvant period | Adjuvant period | Overall period | ||||
| GC-D (N=530) |
GC (N=526) |
GC-D (N=383) |
GC (N=383) |
GC-D (N=530) |
GC (N=526) |
|
| Any AE | 520 (98.1) | 515 (97.9) | 331 (86.4) | 273 (71.3) | 527 (99.4) | 525 (99.8) |
| Any AE of CTCAE Grade 3 or 4 | 249 (47.0) | 271 (51.5) | 119 (31.1) | 91 (23.8) | 368 (69.4) | 355 (67.5) |
| grade 5 AEs | 6 (1.1) | 10 (1.9) | 7 (1.8) | 6 (1.6) | 27 (5.1) | 29 (5.5) |
| Any SAE (including events with outcome of death) | 125 (23.6) | 118 (22.4) | 101 (26.4) | 85 (22.2) | 326 (61.5) | 287 (54.6) |
| Any AE leading to discontinuation of study treatment | 79 (14.9) | 80 (15.2) | 30 (7.8) | 0 | 112 (21.1) | 80 (15.2) |
| Any AE leading to discontinuation of durvalumab | 50 (9.4) | NA | 30 (7.8) | NA | 86 (16.2) | NA |
| Any AE leading to discontinuation of G or C | 72 (13.6) | 80 (15.2) | NA | NA | 72 (13.6) | 80 (15.2) |
| Any AE leading to dose interruption or reduction of study treatment | 269 (50.8) | 247 (47.0) | 75 (19.6) | 0 | 305 (57.5) | 247 (47.0) |
| Any AE leading to dose interruption of durvalumab | 132 (24.9) | NA | 75 (19.6) | NA | 192 (36.2) | NA |
| Any AE leading to dose interruption or reduction of G or C (at least one component) | 260 (49.1) | 247 (47.0) | NA | NA | 260 (49.1) | 247 (47.0) |
Source: table 44, NIAGARA clinical study report (U.S. FDA. Clinical Review–BLA 761069, supplements 050; not public).
AE: adverse Events, SAE: Serious Adverse Event
Regulatory Insights
NIAGARA evaluated the addition of peri-operative (neoadjuvant plus adjuvant) durvalumab to standard-of-care neoadjuvant chemotherapy in patients with MIBC eligible for RC. NIAGARA demonstrated an improvement in both EFS and OS, although no statistically significant improvement was observed in pCR. Median OS had not been reached for either treatment arm and the FDA has requested a postmarketing commitment (PMC) for submission of updated OS results at the pre-specified final analysis.
A limitation of the NIAGARA design is that the trial was not designed to isolate the effect of the neoadjuvant compared to the adjuvant phase of treatment. It remains unclear whether durvalumab needs to be given in both phases to achieve the overall treatment effect and, if so, what the optimal duration of therapy should be for each phase.
The failure to evaluate contribution of phase was previously discussed in a 2024 meeting of the Oncologic Drugs Advisory Committee (ODAC) regarding the AEGEAN trial of the addition of peri-operative durvalumab to neoadjuvant chemotherapy in non-small cell lung cancer, which demonstrated an improvement in EFS. Similarly to NIAGARA, AEGEAN was not designed to isolate the contribution of each phase of treatment, but unlike NIAGARA, did not demonstrate an OS improvement. The ODAC opined that while future trials should be designed to better address contribution of each phase, the AEGEAN results were meaningful. Given the totality of data from NIAGARA, including a demonstrated OS benefit, and concordant with the discussion regarding the AEGEAN approval, the FDA considered the overall benefit: risk assessment to be favorable for adding durvalumab to gemcitabine and cisplatin regimen in the proposed indication. A statement was added to Section 14 of product labeling stating that NIAGARA was not designed to isolate the effect of durvalumab in each phase (neoadjuvant or adjuvant) of treatment.
Although pCR was included as a formally tested efficacy endpoint in NIAGARA, the utility of pCR to support efficacy for regulatory purposes remains in question. pCR as a potential ‘early’ endpoint in neoadjuvant trials in bladder cancer was discussed in a workshop convened by FDA on August 8, 2019 (6). During the workshop and in the publication that followed (7), participants expressed concern that individual and trial-level association with long-term outcome for pCR had not been established, that there was potential heterogeneity in the evaluation of pCR, and that the magnitude of benefit in pCR that would translate into clinical benefit was not clear. Although there were concerns expressed during the prior workshop about potential variability in pathologic evaluation, NIAGARA demonstrated 89.3% overall concordance between local and central pathology review for pCR assessment. In general, validating an early endpoint would require multiple trials demonstrating both patient- and trial-level association with clinically meaningful long-term outcomes. As the first randomized phase 3 clinical trial in bladder cancer that included centrally-assessed pCR as a dual primary endpoint with EFS, NIAGARA could potentially contribute to future meta-analyses when additional trial results become available.
Conclusion
The FDA review team considered the observed statistically significant improvements in EFS and OS with perioperative GC-D treatment compared to neoadjuvant GC treatment to be clinically meaningful and the safety and tolerability to be acceptable for patients with MIBC eligible for neoadjuvant chemotherapy and RC (Table 3). In accordance with the 2024 ODAC discussion around trial design in the perioperative setting, future trials should be designed to better address contribution of phase. However, given the totality of data from NIAGARA, including a demonstrated OS benefit, the FDA considered the overall benefit: risk assessment to be favorable for the regimen as studied for the proposed indication.
Table 3:
Benefit-Risk Assessment
| Dimension | Evidence and Uncertainties | Conclusions and Reasons |
|---|---|---|
| Analysis of Condition | Bladder cancer is the 6th leading cause of mortality from cancer in the US.
|
Patients with MIBC are at high risk of recurrence after RC and have a serious and life-threatening condition. |
| Current Treatment Options | The current standard-of-care treatment options for patients with stage II-IIIA MICB are:
In the adjuvant setting, patients can receive cisplatin-based chemotherapy if they have residual disease at the time of cystectomy (pT3-4 or N+) and did not receive neoadjuvant cisplatin-based chemotherapy. FDA approved nivolumab in 2021 as an adjuvant treatment for patients with high-risk bladder cancer (pT2-pT4a or ypN+ in patients with neoadjuvant or pT3-pT4a in patients without neoadjuvant) based on the results of CHECKMATE-274 trial and improved DFS. |
There are no current FDA-approved peri-operative treatment options for patients with MIBC and, as of the current status of CHECKMATE-274, no adjuvant therapies that have demonstrated improved OS. Due to the high recurrence rate and mortality of patients with MIBC even after neoadjuvant and adjuvant treatments, effective and tolerable FDA-approved treatment options are needed. |
| Benefit |
|
The observed EFS and OS improvements in the NIAGARA trial are statistically significant and clinically meaningful to patients with MIBC who are cisplatin-eligible and candidates for RC. |
| Risk and Risk Management |
|
The risk-benefit profile of nivolumab is acceptable in the approved patient population |
Source: FDA’s multi-disciplinary review (U.S. FDA. Clinical Review–BLA 761069, supplements 050; not public).
MIBC: Muscle invasive bladder cancer, NMIBC: Non-muscle invesive bladder cancer, RC: radical cystectomy, TURBT: Transurethral resection of bladder tumor, EFS: event free survival, OS: overal survival, IA: interim analysis, REMS: risk evaluation and mitigation strategy
Supplementary Material
Footnotes
Note: William F. Pierce completed work on this publication while an employee at the FDA. At the time of publishing, Dr. Pierce is an employee at Resilient Pharmaceuticals, Inc. Dr. Pazdur completed work on this publication while he was an employee at the FDA. At the time of publishing, he is no longer at the FDA.
Note: This is a U.S. Government work. There are no restrictions on its use.
Disclosure of Potential Conflicts of Interest:
No potential conflicts of interest were disclosed.
References:
- 1.Grossman HB, Natale RB, Tangen CM, et al. Neoadjuvant chemotherapy plus cystectomy compared with cystectomy alone for locally advanced bladder cancer. N Engl J Med 2003;349:859–66. [DOI] [PubMed] [Google Scholar]
- 2.Sonpavde G, Goldman BH, Speights VO, et al. Quality of pathologic response and surgery correlate with survival for completely resected bladder cancer following neoadjuvant chemotherapy. Cancer 2009;115(18):4104–09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Vale CL. Advanced Bladder Cancer (ABC) Meta-analysis Collaboration. Neoadjuvant chemotherapy in invasive bladder cancer: update of a systematic review and meta-analysis of individual patient data advanced bladder cancer (ABC) meta-analysis collaboration. Eur Urol. 2005;48(2):202–5; discussion 205-6 [DOI] [PubMed] [Google Scholar]
- 4.FDA approves durvalumab for muscle invasive bladder cancer. Available from: https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-durvalumab-muscle-invasive-bladder-cancer [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Powles T, Catto JWF, Galsky MD, et al. Perioperative durvalumab with neoadjuvant chemotherapy in operable bladder cancer. NEJM. 2024; 391(19): 1773–1786 [DOI] [PubMed] [Google Scholar]
- 6.FDA-BCAN Workshop: Endpoints for the Development of Neoadjuvant Systemic Therapy Regimens for Muscle-Invasive Bladder Cancer. Aug 8, 2019, [Google Scholar]
- 7.Chang E, Apolo AB, Bangs R, et al. Refining neoadjuvant therapy clinical trial design for muscle-invasive bladder cancer before cystectomy: a joint US Food and Drug Administration and Bladder Cancer Advocacy Network workshop. Nat Rev Urol. 2022. Jan;19(1):37–46. [DOI] [PubMed] [Google Scholar]
- 8.US Food and Drug Administration: Imfinzi US Prescribing Information. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/label/2025/761069s050lbl.pdf [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.