Abstract
Background/Aim: The EV-301 trial demonstrated the efficacy of enfortumab vedotin (EV) as a third-line treatment for metastatic urothelial carcinoma (mUC), showing significant improvement in overall survival (OS) and progression-free survival (PFS) compared to chemotherapy in patients previously treated with platinum-based therapy and immune checkpoint inhibitors. In real-world clinical practice, patients undergoing third-line treatment often have poor baseline health status, leading to the off-label use of EV in populations ineligible for clinical trials. This study aimed to evaluate the treatment outcomes of EV in both EV-301 trial-eligible and -ineligible patients.
Patients and Methods: Fifty-eight patients with mUC treated with EV across five Institutions were retrospectively evaluated and stratified based on the EV-301 trial eligibility criteria. Patients with an Eastern Cooperative Oncology Group performance status of ≥2, baseline hemoglobin level of <9 g/dl, creatinine clearance <30 ml/min, or other protocol-defined criteria were analyzed. Treatment outcomes were assessed for both groups.
Results: Of the 58 patients, 33 (56.9%) met the EV-301 trial eligibility criteria. No significant differences were observed in PFS (median: 9.2 vs. 7.1 months, for eligible vs. ineligible patients) and OS (15.4 vs. 8.9 months). Although the objective response rate was higher in the eligible group (54.6% vs. 28.0%), there was no significant difference in the disease control rate (78.8% vs. 80.0%). Adverse events (AEs) of any grade were more frequent in the eligible group (93.9% vs. 64.0%), but the incidence of grade ≥3 AEs did not differ significantly (12.1% vs. 8.0%).
Conclusion: The findings of this multi-institutional study highlight the feasibility of EV treatment in EV-301 trial-ineligible patients with mUC, supporting its potential applicability in both trial-eligible and -ineligible groups.
Introduction
The EV-301 trial demonstrated the efficacy of enfortumab vedotin (EV) in patients with locally advanced or metastatic urothelial carcinoma (mUC) who previously received platinum-based chemotherapy and subsequently experienced disease progression during or after treatment with immune checkpoint inhibitors (ICIs) targeting programmed cell death protein 1 or programmed death ligand 1 (1). In this randomized phase III trial, EV significantly improved overall survival (OS) (median: 12.9 vs. 9.0 months) and progression-free survival (PFS) (median: 5.6 vs. 3.7 months) compared with standard chemotherapy. Based on these findings, EV was granted regulatory approval in Japan in September 2021.
As a controlled clinical trial, the EV-301 trial excluded patients with poor baseline characteristics, including those with an Eastern Cooperative Oncology Group (ECOG) performance status (PS) score of ≥2, a baseline hemoglobin (Hb) level of <9 g/dl, or a creatinine clearance (CrCl) level of <30 ml/min. Additional exclusion criteria–such as severe comorbidities and specific organ dysfunction–were applied to ensure patient safety and consistency in trial outcomes. Notably, poor PS or renal impairment has historically been associated with limited prognosis and fewer therapeutic options.
Despite limited clinical evidence on the safety and efficacy of EV in these populations, real-world use of the drug in patients who do not meet EV-301 trial eligibility criteria has increased, highlighting a gap in understanding its outcomes in such groups. However, real-world evidence on its use remains limited, particularly in patients with reduced PS, low Hb levels, or advanced renal impairment. Therefore, this study aimed to assess the efficacy and safety of EV in patients who did not meet the eligibility criteria for the EV-301 trial.
Patients and Methods
Patient selection. We retrospectively evaluated 63 patients with mUC who were treated with EV at the Tokyo Women’s Medical University, Tokiwakai Jyoban Hospital, Saiseikai Kawaguchi General Hospital, Saiseikai Kazo General Hospital, and Tokyo Women’s Medical University Medical Center Adachi as third- or later-line therapy following platinum-based chemotherapy and ICI therapy between November 2021 and September 2024. Patients were excluded if they 1) had incomplete post-treatment clinical information, 2) had an observation period of less than 1 month, and 3) did not undergo imaging examination after the initiation of EV treatment. Hence, only 58 patients were included in this retrospective study. The study was approved by the institutional ethics review boards of the participating organizations (Ethics ID: 2024-0017). This study was conducted in accordance with the principles outlined in the 1964 Declaration of Helsinki and its later amendments. The need for informed consent was waived because of the retrospective observational nature of this study.
Treatment regimens and method of evaluation. All patients were histologically diagnosed with urothelial carcinoma (UC) and received platinum-based chemotherapy and ICIs prior to the initiation of EV therapy. An EV dose of 1.25 mg/kg body weight was administered via intravenous infusion on days 1, 8, and 15 of a 28-day cycle, following the protocol outlined in the EV301 trial (1). The initial dose was reduced at the discretion of the attending physician, depending on the patient’s condition and clinical response. Additionally, dose reductions were implemented in response to the occurrence of adverse events (AEs). The interval between EV administrations was adjusted based on the patient’s condition or the severity of AEs. After initiation EV, computed tomography (CT) of the thorax, abdomen, and pelvis with simple contrast was performed. Post-treatment CT was performed at regular intervals of 4 to 12 weeks, depending on the patient’s condition. EV treatment was continued until radiographic or clinical disease progression or the occurrence of intolerable AEs.
Outcome assessment. The following outcomes were assessed and compared between eligible and ineligible patients based on the EV-301 trial protocol (1). The primary endpoints of this study were PFS and OS, while the secondary endpoints were objective response rate (ORR) and disease control rate (DCR). The ORRs and DCRs were calculated based on the Response Evaluation Criteria in Solid Tumors version 1.1 (2). The AEs associated with EV were assessed in both groups using the Common Terminology Criteria for Adverse Events version 5.0. A subgroup analysis was conducted by stratifying the patients based on the number of ineligible criteria met. The best tumor response, as well as the ORRs and DCRs, was evaluated and compared between the groups.
Eligibility assessment. The eligibility criteria were based on the EV-301 trial protocol, which specified that patients with an ECOG PS score of 0 or 1 were eligible. Additionally, the following baseline characteristics were required, and patients who did not meet these criteria were excluded: 1) an absolute neutrophil count of ≥1,500/mm3; 2) a platelet count of ≥100 × 109/l; 3) an Hb level of ≥9 g/dl; 4) a serum bilirubin level of ≤1.5 times the upper limit of normal (ULN) or ≤3 times ULN for patients with Gilbert’s disease; 5) a CrCl level of ≥30 ml/min, estimated using institutional standards or measured via a 24-hour urine collection (with glomerular filtration rate used as an alternative parameter); and 6) and alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels of ≤2.5 times the ULN or ≤3 times the ULN for patients with liver metastases. Patients were stratified into eligible and ineligible groups based on these criteria (1). To ensure a comprehensive evaluation, all inclusion and exclusion criteria outlined in the original EV-301 protocol were reviewed and systematically compared against the clinical data of our cohort. This approach enabled precise stratification of patients into eligible and ineligible groups based on real-world parameters.
Statistical analysis. Continuous variables were analyzed using the Mann-Whitney U test, whereas categorical variables, including ORRs, DCRs, and AE incidence, were compared using Fisher’s exact test. PFS was calculated from EV initiation to disease progression or death, whichever occurred first. OS was calculated from EV initiation until death from any cause. Patients who were lost to follow-up were censored at the time of their last contact. PFS and OS were calculated using the Kaplan-Meier method and compared using the log-rank test. Univariate analyses of PFS and OS were performed based on the Cox proportional hazard model. The risks were expressed as hazard ratios (HRs) with their corresponding 95% confidence intervals (CIs). All statistical analyses were conducted using JMP version 17 (SAS Institute Inc., Cary, NC, USA), and a p-value of <0.05 was considered significant.
Results
Patient characteristics. Based on the EV-301 protocol (1), 33 (56.9%) and 25 (43.1%) patients were eligible and ineligible for the trial, respectively (Table I). Of the 58 patients who were treated with EV, 35 (60.3%) experienced disease progression, whereas 30 (51.7%) died from any cause during follow-up (median: 8.4 months). No significant differences were observed between the groups in terms of the number of prior therapies, number of metastatic organs, history of radical surgery for primary lesion, histology, and liver metastasis status (all p>0.05). However, the eligible group exhibited a significantly higher body mass index (eligible: median 22.6 kg/m2 vs. ineligible: 21.1 kg/m2, p=0.006) and a higher prevalence of metastatic involvement (90.9% vs. 52.0%, p=0.001).
Patient characteristics.
Eligibility profiles. Although some criteria overlapped, 14 (56.0%) patients had an ECOG PS score of 2 or higher, 11 (44.0%) had an Hb level below 9 g/dL, and 8 (32.0%) had a CrCl level <30 ml/min. Additionally, one (4.0%) patient required prednisolone treatment at a dose of 20 milligrams or more per day at the initiation of EV therapy due to the occurrence of immune-related AEs (irAEs) triggered by prior ICI treatment, which was another exclusion criterion. As a result, no patients met the inclusion criteria (Table II).
Tumor response according to the number of ineligibility criteria factors.
Survival according to the eligibility criteria. After the initiation of EV treatment, PFS did not differ significantly between the eligible and ineligible groups [eligible: median 9.2 months (95% CI=5.1-13.3) vs. ineligible: 7.1 months (95% CI=2.7-13.0), p=0.448; HR=0.77 (95% CI=0.40-1.50), p=0.447] (Figure 1a). Similarly, OS was not significantly different between these groups [eligible: 15.4 months (95% CI=12.0-N.R.) vs. ineligible: 8.9 months (5.1-19.4), p=0.221; HR=0.64 (95% CI=0.31-1.31), p=0.225] (Figure 1b).
Kaplan–Meier curves for PFS and OS in eligible and ineligible patients. (a) PFSl and (b) OS after enfortumab vedotin initiation. PFS, Progression-free survival; OS, overall survival; CI, confidence interval; median, months.
Tumor response according to the eligibility criteria. Next, the magnitude of tumor response was compared between the groups based on the eligibility criteria. The ORRs were significantly higher in the eligible group compared to the ineligible group (eligible: 54.5% vs. ineligible: 28.0%, p=0.043), whereas the DCRs were comparable between the two groups (eligible: 78.8% vs. ineligible: 80.0%, p=0.910). In the subgroup analysis, 21 (36.2%) patients met only one ineligibility criterion, whereas 4 met two or more criteria. Due to the small sample size in these subgroups, formal statistical analysis was not performed (Table II).
Tumor response according to the eligibility criteria. The AEs associated with EV treatment are summarized in Table III. The overall incidence of any EV treatment-associated AEs was significantly different between the groups (eligible: 93.9% vs. ineligible: 64.0%, p=0.004). However, the incidence of grade 3 or higher AEs was not significantly different between these groups (12.1% vs. 8.0%, p=0.610). For major AEs, no significant difference was found in the incidence rate of maculopapular rash (30.3% vs. 28.0%, p=0.849). However, a significant difference was observed in the incidence rate of neuropathy (30.3% vs. 8.0%, p=0.030) between the groups. Among the grade 3 or higher AEs, neuropathy occurred in two patients, rash in one patient, and Stevens-Johnson syndrome in another patient from the eligible group. In the ineligible group, interstitial pneumonia developed in one patient, whereas Stevens-Johnson syndrome occurred in another patient (Table III).
Summary of adverse events observed after EV initiation.
Discussion
This multi-institutional retrospective study found no significant differences in PFS, OS, or grade ≥3 AEs rates between trial-eligible and ineligible mUC patients treated with EV. However, the eligible group demonstrated superior ORRs. Patients were stratified using the full eligibility criteria of the EV-301 trial to reflect real-world practice, where EV is often administered to patients typically excluded from trials due to poor performance status, anemia, or renal impairment. This approach helps contextualize trial data within the broader real-world population.
Despite advancements in second-line therapies, such as pembrolizumab (3) and avelumab maintenance therapy (4), survival outcomes for patients with mUC remain poor. In clinical practice, many patients with mUC experience impaired renal function due to nephroureterectomy (5), tumor obstruction (6), or chemotherapy-induced nephrotoxicity (7), in addition to poor PS resulting from disease progression or prior treatments. As the EV-301 trial excluded such patients, real-world data on the efficacy and safety of EV in this population remain limited.
The EV-301 trial only included patients with an ECOG PS score of 0-1, reflecting a population with a relatively good functional status. However, in real-world clinical practice, patients with a poor PS are commonly encountered. In first-line treatment, a PS score of >1 is considered unsuitable for cisplatin-based chemotherapy, and carboplatin-based regimens are typically used as alternatives (8). The Bellmunt risk score, a key prognostic model for second-line ICI therapy, identified a PS score of >0, an Hb level of <10 g/dl, and liver metastasis as adverse prognostic factors (9, 10). Across various treatment settings, a PS score of ≥2 consistently correlates with poor outcomes due to reduced immune and organ function, limited treatment tolerance, increased risk of complications, and disease progression. Real-world data have also shown very poor PFS outcomes (median: 1.0 month) in patients with a PS score ≥2 (11). A multicenter study conducted in Australia demonstrated a significant difference in both PFS (PS0-1: median 5.3 months vs. PS ≥2: 2.4 months, p<0.001) and OS (11.3 months vs. 2.8 months, p=0.048) between patients with a PS score of 0-1 and those with a PS score of ≥2. Additionally, this study reported a significant difference in the ORR between the two groups (40% vs. 12%, p=0.007) (12). In Japan, Minato et al. reported that patients with a PS score of 3 had significantly poorer PFS compared with those with a PS score of 0. This finding indicates that poor PS serves as a negative prognostic factor for treatment outcomes (HR=4.54, p=0.019) (13). However, the Urothelial Cancer Network to Investigate Therapeutic Experiences (UNITE) study reported no significant difference in ORRs between patients with PS scores of 0–1 and those with PS scores of 2–3 (56% vs. 34%, p=0.18) (14). Studies comparing PS have shown some variability in statistical significance, but a consistently lower ORR has been observed in patients with a PS score of ≥2. In this study, a significant difference in ORRs was identified (eligible: 54.6% vs. ineligible: 28.0%; p=0.043). Notably, 14 patients (56.0%) in this study had a PS score of ≥2, a higher proportion than that included in other studies. Interestingly, a PS score of ≥2 did not show the expected negative impact on PFS, possibly due to the fact that most patients had a PS score of 2 (11 patients, 44.0%). This finding suggests a more nuanced relationship between PS and treatment efficacy.
Anemia in mUC is either chronic (due to cancer progression) or acute (due to hematuria). A low Hb level is recognized as a prognostic factor in the Bellmunt model (9). However, its direct impact on EV outcomes remains unclear (15). A study comparing treatment outcomes of EV following various ICI therapies reported a significant difference in OS in the univariate analysis of patients with an Hb level of <11 g/dl (HR=2.1, p=0.049). However, as multivariate analysis was not conducted, the causal relationship remains unclear (16). In our study, 11 patients with an Hb level of <9 g/dl, considered ineligible based on the EV-301 trial criteria, showed no significant differences in efficacy or safety outcomes when compared with eligible patients.
In patients with mUC, a CrCl level of <30 ml/min often indicates underlying chronic kidney disease. Although impaired renal function is a common challenge in managing mUC, the UNITE study reported no significant difference in ORRs using an estimated GFR cutoff level of 30 ml/min/1.73 m2 (14). Although evidence remains limited to case-level observations, several reports have documented the use of EV in patients undergoing dialysis, suggesting that both the safety and efficacy profiles are comparable to those in patients not requiring dialysis, although the evidence remains limited to case-level observations (17-19).
In our cohort, the incidence of treatment-related AEs significantly differed between the eligible and ineligible groups (93.9% vs. 64.0%, p=0.004), whereas no significant difference was found in the incidence of grade 3 or higher AEs between the two groups (12.1% vs. 8.0%, p=0.610). The overall incidence of AEs of any grade in the eligible group was consistent with the findings of the EV-301 trial (93.9%), although the incidence of grade 3 or higher AEs was lower than that reported in the EV-301 trial (51.4%) (1). This discrepancy may reflect the differences in AE reporting focus (clinically important events) or shorter follow-up in real-world settings. Notably, no specific risk factors have been identified in any cohort analysis (11, 20-23).
The EV-301 trial demonstrated a significant improvement in both PFS (median: 5.6 months) and OS (median: 12.9 months) in patients with mUC. While our study addressed eligibility-based differences, other real-world studies have examined factors such as prior treatment sequence. A recent Japanese reported significantly longer OS with EV following avelumab maintenance compared to pembrolizumab (24), while another multicenter analysis did not find a significant difference between the two groups (16). In this study, the eligible group showed relatively favorable outcomes, with a median PFS of 9.2 months and a median OS of 15.4 months, compared to the results reported in the EV-301 trial. Similarly, the ineligible group demonstrated PFS (median: 7.1 months) results comparable to those observed in the EV-301 trial. Although the eligible group in this study had a higher median age and a greater proportion of patients with liver metastases compared with the EV-301 cohort, the proportion of patients with a PS score of 0 was higher in our study. An Australian study reporting real-world experience with EV found a median PFS of 4.8 months and a median OS of 10.8 months (12). In the UNITE study, the median PFS was 6.8 months, whereas the median OS was 14.4 months; both outcomes were comparable with those in the EV-301 trial (14). In Japanese studies, Miyake et al. reported a median PFS of 9 months and a median OS of 16 months (22), whereas Endo et al. reported a median PFS of 10.5 months and a median OS of 12.9 months (23). The favorable outcomes observed in the present study align with those in these previous Japanese studies, supporting the efficacy of EV in Japanese patients with mUC.
Study limitations. First, the small sample size may have introduced inherent bias, and the study’s retrospective design imposed additional limitations. In particular, selection bias may have occurred because patients with incomplete clinical data or without post-treatment imaging were excluded, potentially favoring those with better performance status or longer treatment duration. Moreover, information bias is possible, as AEs and tumor response assessments were based on retrospective chart review without centralized imaging review, and assessment timing varied across institutions. Confounding bias cannot be excluded either, since no multivariable analyses were performed, and the eligible and ineligible groups differed in several baseline characteristics such as ECOG PS, Hb levels, and CrCl level. Although multivariable analysis was not performed due to limited sample size, we provided hazard ratios with 95% confidence intervals to support cautious interpretation. However, the limited statistical power to detect small differences should also be acknowledged. Therefore, larger, prospective studies are needed to confirm our findings. Additionally, the retrospective nature of the study and reliance on medical records may have resulted in incomplete or inaccurate data. Furthermore, this study did not capture full details of post-EV treatment, and their impact on OS remains uncertain, which has also been highlighted by Furubayashi et al. (25). Finally, our study did not include any biomarker analyses, such as measurement of programmed death ligand 1 and immunohistochemical staining for nectin-4. These limitations should be considered when interpreting the results of this study.
Notably, the findings from this study underscore the clinical relevance of assessing eligibility status, as they suggest that even patients deemed ineligible under stringent trial conditions may still derive meaningful benefit from EV therapy. This highlights the importance of real-world evidence in informing treatment decisions and expanding therapeutic options beyond rigid trial populations. Our findings suggest that EV is feasible and efficacy in both trial-eligible and ineligible patients; however, the precise impact of eligibility criteria on treatment outcomes remains unclear. Further prospective studies are warranted to validate these results and optimize treatment strategies for diverse patient populations.
Conclusion
This multicenter retrospective study demonstrated no significant differences in PFS, OS, or the incidence of grade ≥3 AEs between the EV-301 trial-eligible and - ineligible groups in patients with mUC. These findings suggest that EV treatment may be effective and safe for a broader range of patients, including those considered ineligible.
Acknowledgements
The Authors would like to thank the clinical staff of the participating institutions who played an indispensable role in patient care and data collection.
Footnotes
Authors’ Contributions
All Authors contributed to the study conception and design. Kazutaka Nakamura, Hanae Kondo, Toru Inakawa, Yuki Nemoto, and Yuki Kobari were responsible for performing material preparation, data collection, and data analysis. Kazutaka Nakamura wrote the first draft of the manuscript. All Authors provided feedback on previous versions of the manuscript. All Authors have read and approved the final manuscript.
Conflicts of Interest
The Authors declare no conflicts of interest.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received May 2, 2025.
- Revision received June 14, 2025.
- Accepted June 23, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
Jump to section
Related Articles
Cited By...
- No citing articles found.