Impact of durable response of first-line systemic treatment for patients with locally advanced or metastatic urothelial carcinoma: data update of EV-302/KEYNOTE-A39 trial Authors: Makito Miyake Published: 2025 Urothelial carcinoma (UC) originates from the urothelial lining of the renal pelvis, ureter, bladder, and urethra. Upper urinary tract UC is more prevalent in Asian countries and accounts for only approximately 5% of all urinary tract cancers (1). According to the Global Cancer Statistics 2020 report, bladder cancer (BCa) is the 7th and 13th most frequent malignancy among men worldwide, with approximately 573,000 new cases and 213,000 deaths in 2020 (2). By 2040, the annual number of new BCa cases and deaths is projected to rise to 991,000 (a 72.8% increase) and 397,000 (an 86.6% increase), respectively (3). Therefore, to reduce the global burden, there is an urgent need to accelerate control initiatives for high-risk populations with UC, including those with BCa. Locally advanced or metastatic UC (la/mUC) is a progressive disease with poor prognosis. The median overall survival (OS) is approximately 19–26 months following first-line platinum-based chemotherapy and subsequent administration of immune checkpoint inhibitors (ICIs), including anti-programmed cell death 1 (PD-1) and anti-programmed death-ligand 1 (PD-L1) inhibitors (4-7). A substantial proportion of the population responds poorly to chemotherapy and ICIs, which results in poor survival rates. Novel drug combinations, including chemotherapy, ICIs, antibody-drug conjugates (ADCs), and tyrosine kinase inhibitors (TKIs), have been explored for a long time to overcome inherent drug resistance and improve antineoplastic activity (8). This unmet need was addressed by two pivotal trials, CheckMate-901 and EV-302/KEYNOTE-A39 (hereinafter referred to as EV-302), presented at the European Society for Medical Oncology (ESMO)-2023 plenary session. The approval of nivolumab (anti-PD-1 inhibitor) in combination with gemcitabine and cisplatin (GC) and pembrolizumab (anti-PD-1 inhibitor) with enfortumab vedotin (EV) based on the positive results has revolutionized the first-line treatment of la/mUC in many countries (9,10). In the CheckMate-901 trial, which compared conventional GC with and without nivolumab in previously untreated, cisplatin-eligible la/mUC, the primary endpoint was met with a median OS benefit favoring the GC plus nivolumab (GCN) arm [21.7 vs. 18.9 months, hazard ratio (HR) 0.75; 95% confidence interval (CI): 0.63–0.96; P=0.017] (9). The GCN achieved an objective response rate (ORR) of 58%, with a complete response (CR) rate of 22%, compared to 43% ORR and 12% CR of GC alone (9). The EV-302 trial compared EV with pembrolizumab (EVP) with platinum-based combination chemotherapy [comparator arm: either GC or gemcitabine plus carboplatin (GCarbo)] and demonstrated a marked improvement in median OS to 31.5 months, nearly doubling that of the comparator arm (16.1 months; HR 0.47; P<0.00001), with 68% ORR and 29% CR compared to 44% ORR and 12% CR in the comparator arm (10). A similar benefit was observed in progression-free survival (PFS), with a 55% reduction in the EVP arm than in the comparator arm (HR 0.45; 95% CI: 0.38−0.54; P<0.001). Notably, the superiority of EVP was consistent irrespective of PD-L1 status and cisplatin eligibility. Recently updated guidelines released by the ESMO (11), European Association of Urology (EAU) (12), and National Comprehensive Cancer Network (NCCN) (13) recommend the EVP regimen as the preferred frontline treatment for previously untreated la/mUC. Patients who are ineligible for or cannot receive EVP may be treated with alternative regimens such as GCN or induction platinum-based chemotherapy, followed by maintenance avelumab, according to cisplatin or platinum eligibility, and other patient-specific considerations. The extended follow-up results of the EV-302 trial were reported by Powles et al. in a recent issue of Annals of Oncology (14). Overall, the EVP arm continued to show superior efficacy and safety compared with conventional chemotherapy, with a median follow-up of 29.1 months (approximately 2.5 years), providing an additional year of follow-up since the primary analysis. In the EV-302 trial, the dual primary endpoints comprised PFS per Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 by blinded independent central review and OS, analyzed in all patients randomly assigned to a treatment group. At the data cutoff (August 8, 2024), 12.2% of the patients in the EVP arm and none of the patients in the chemotherapy arm remained on treatment. The median PFS was 12.5 months (95% CI: 10.4–16.6) for the EVP arm and 6.3 months (95% CI: 6.2–6.5) for the chemotherapy arm (HR 0.48; 95% CI: 0.41–0.57). The median OS was 33.8 months (95% CI: 26.1–39.3) for the EVP arm and 15.9 months (95% CI: 13.6–18.3) for the chemotherapy arm (HR 0.51; 95% CI: 0.43–0.61) (Figure 1A,1B). Moreover, the safety profile with an additional year of follow-up remained consistent with that of the primary analysis. Kaplan-Meier estimates of (A) PFS in the overall population, (B) OS in the overall population, (C) duration of response for patients with the best overall response of CR or PR, and (D) duration of response for patients with the best overall response of CR in the EV-302 trial. Image reproduced with permission from Elsevier (14). a, P value is nominal and descriptive. CI, confidence interval; CR, complete response; EV + P, enfortumab vedotin plus pembrolizumab combination therapy; HR, hazard ratio; OS, overall survival; PD, progessive disease; PFS, progression-free survival; PR, partial response; NE, not estimable; NR, not reached. Powles et al. reported an exploratory analysis of the duration of response (DOR) in patients who responded to EVP or chemotherapy (14,15). The confirmed ORR [CR rate (CRR) + PR rate (PRR)] was 67.5% (295 of 437 patients; 95% CI: 62.9–71.9) for the EVP arm and 44.2% (195 of 441 patients; 95% CI: 39.5–49.0) for chemotherapy arm. The baseline characteristics of responders were consistent with those of the overall population. The median DOR was 23.3 months [95% CI: 17.8–not estimable (NE)] in the EVP arm, substantially longer than in the chemotherapy arm [7.0 months (95% CI: 6.2–9.0)] (Figure 1C). In the EVP and chemotherapy arms, CRR was achieved in 30.4% and 14.5% of the patients, respectively. However, the median DOR was not reached in the EVP arm, whereas it reached 15.2 months (95% CI: 10.3–NE) in the chemotherapy arm (Figure 1D). The probabilities of responders without progessive disease (PD) or death at 12 months were 84.3% and 60.0% for the EVP and chemotherapy arms, respectively, decreasing to 74.3% and 43.2% at 24 months. Additionally, the median DOR was only 10.6 months (95% CI: 8.4–14.6) and 5.4 months (95% CI: 4.5–6.1) in the EVP and chemotherapy arms, respectively. A substantial difference in favorable survival outcomes was observed between patients achieving CR versus those with PR in the EVP arm. The authors concluded that EVP reduced the risk of progression or death in patients achieving CR, provided appropriate dose modifications were applied. A durable response to systemic therapy is one of the key outcomes for achieving long-term survival in cancer treatment. Table 1 lists the oncological outcomes of first-line EVP and GCN for la/mUC, based on pivotal clinical trials (9,14-20). EV-103 dose escalation/cohort A and EV-103 cohort K, included cisplatin-ineligible patients with la/mUC (16-18). The ORR for the EVP treatment in cohort A and cohort K were 73% and 64.5%, respectively, with median DORs of 25.6 months (95% CI: 8.3–NE) and not reached (95% CI: 10.3–NE), respectively; these results were comparable to those of EV-302 trial (14). The oncologic efficacy in pre-specified subgroups, including those defined by primary disease site of origin and metastatic site, was presented at the American Society of Clinical Oncology (ASCO) 2025 meeting, showing that PFS, OS, DOR, and ORR continued to demonstrate the sustained benefit of EVP versus conventional chemotherapy across pre-specified subgroups after long-term follow-up (19). Therefore, we focused on exploratory subgroup analysis of patients with lymph node (LN)-only metastasis disease. Among the 866 patients in the EV-302 trial, 207 patients (24%) had LN-only disease and were randomly grouped to 103 patients in the EVP arm and 104 in the conventional chemotherapy arm. The confirmed ORR were 77.5% and 53.4% in the EVP and chemotherapy arms, respectively, which were higher than those in the overall population (67.5% and 44.2%, respectively). Recently, a similar exploratory post hoc analysis of CheckMate-901, reported by Galsky et al. (20), revealed notable data. Among the randomized 608 patients, 102 (17%) achieved CR, including 66 (22%) of 304 patients treated with GCN and 36 (12%) of the 304 patients treated with GC alone. When enriched by patients with LN-only disease among those with a CR, 54 patients in each arm received systemic treatment. In patients with LN-only disease, the median OS for patients treated with GCN versus those treated with GC alone were 46.3 and 24.9 months, respectively (HR 0.58; 95% CI: 0.34–1.00). Similarly, the median PFS for patients treated with GCN versus those treated with GC alone were 30.5 and 8.84 months, respectively (HR 0.38; 95% CI: 0.22–1.00). This post hoc analysis of the CheckMate-901 trial demonstrated that GCN provided considerable benefits for OS and PFS compared to GC alone in patients with mUC and LN-only disease. Table 1Oncologic outcomes of first-line enfortumab vedotin plus pembrolizumab and GC plus nivolumab for la/mUC: exploratory analyses of responders and lymph node only diseaseTrial nameTarget populationRegimenNo. of patientsMedian PFS (months) (95% CI)HR for progressionMedian OS (months) (95% CI)HR for deathOS rate (%)Median DOR in responder (months) (95% CI)Responders without PD or death (%)Citation12 months24 months12 months24 monthsEV-103 DE/A (cis-ineligible)Overall populationEVP4512.3 (8.0–NE)–26.1 (15.5–NE)––––––(16)Responder (CR + PR)EVP33 (ORR 73%)––––63.9 (95% CI: 44.2–78.2)47.0 (95% CI: 27.6–64.3)25.6 (8.3–NE)63.9 (95% CI: 44.2–78.2)47.0 (95% CI: 27.6–64.3)(17)EV-103 cohort K (cis-ineligible) Overall populationEVP76NR (8.31–NE)–22.3 (19.1–NE)––––––(18)Responder (CR + PR)EVP49 (ORR 64.5%)––NA–––NR (10.3–NE)65.4–Overall populationPembro mono738.0 (6.1–10.4)–21.7 (15.2–NE)––––––Responder (CR + PR)Pembro mono33 (ORR 45.3%)––NA–––13.2 (6.1–16.0)56.3–EV-302 (platinum-eligible) Overall populationEVP44212.5 (10.4–16.6)0.48 (0.41–0.57)33.8 (26.1–39.3)0.51 (0.43–0.61)77.760.1–––(14)GC or GCarbo4446.3 (6.2–6.6)Ref15.9 (13.6–18.3)Ref61.135.4–––LN only subgroupEVP10322.1 (15.3–NE)0.47 (0.32–0.70)NR (39.3–NE)0.51 (0.33–0.79)–––––(19)GC or GCarbo1048.3 (6.2–13.1)Ref24.4 (18.3–NE)Ref–––––LN only responderEVP79 (ORR 77.5%)––––––NR (19.9–NE)––GC or GCarbo55 (ORR 53.4%)––––––12.4 (8.6–24.9)––Responder (CR + PR)EVP295 (ORR 67.5%)––––91.176.323.3 (17.8–NE)67.549.4(14)GC or GCarbo195 (ORR 44.2%)––––85.159.87.0 (6.2–9.0)35.124.0CR subgroupEVP133 (CRR 30.4%)––––97.795.4NR (NE–NE)84.374.3GC or GCarbo64 (CRR 14.5%)––––98.485.815.2 (10.3–NE)60.043.2PR subgroupEVP162 (PRR 37.1%)––––NANA10.6 (8.4–14.6)49.023.4(15)GC or GCarbo131 (PRR 29.7%)––––NANA5.4 (4.5–6.1)11.38.2CheckMate-901 (cis-eligible)Overall populationGC + Nivo3047.9 (7.6–9.5)0.72 (0.55–0.94)21.7 (18.6–26.4)0.78 (0.63–0.96)70.246.9–––(9)GC3047.6 (6.1–7.8)Ref18.9 (14.7–22.4)Ref62.740.7–––LN only subgroupGC + Nivo5430.5 (9.6–NE)0.38 (0.22–0.66)46.3 (18.6–26.4)0.58 (0.34–1.00)NANA–––(20)GC568.8 (7.5–10.9)Ref24.9 (14.7–22.4)RefNANA–––LN only CR subgroupGC + Nivo34 (CRR 63%)––––NANANR (22.0–NE)70 (95% CI: 51–82)65 (95% CI: 45–79)GC19 (CRR 34%)––––NANA8.7 (6.7–15.6)32 (95% CI: 10–57)0CI, confidence interval; CRR, complete response rate; DE/A, dose escalation/cohort A; DOR, duration of response; EVP, enfortumab vedotin plus pembrolizumab combination therapy; GCarbo, gemcitabin plus carboplatin combination chemotherapy; GC, gemcitabin plus cisplatin combination chemotherapy; GC + Nivo, GC plus nivolumab combination therapy; HR, hazard ratio; la/mUC, locally advanced or metastatic urothelial carcinoma; LN, lymph node; NA, not available; NE, not estimable; NR, not reached; ORR, objective response rate; OS, overall survival; PD, progessive disease; PFS, progression-free survival; PR, partial response; PRR, partial response rate; Ref, reference. Oncologic outcomes of first-line enfortumab vedotin plus pembrolizumab and GC plus nivolumab for la/mUC: exploratory analyses of responders and lymph node only disease CI, confidence interval; CRR, complete response rate; DE/A, dose escalation/cohort A; DOR, duration of response; EVP, enfortumab vedotin plus pembrolizumab combination therapy; GCarbo, gemcitabin plus carboplatin combination chemotherapy; GC, gemcitabin plus cisplatin combination chemotherapy; GC + Nivo, GC plus nivolumab combination therapy; HR, hazard ratio; la/mUC, locally advanced or metastatic urothelial carcinoma; LN, lymph node; NA, not available; NE, not estimable; NR, not reached; ORR, objective response rate; OS, overall survival; PD, progessive disease; PFS, progression-free survival; PR, partial response; PRR, partial response rate; Ref, reference. The biological and clinical mechanisms underlying the particularly favorable response to GCN in patients with LN-only mUC remain unclear. Several studies have shown that LN-only mUC demonstrated an enhanced durable response to both cisplatin-based chemotherapy and ICI therapy (21,22). Moreover, LN-only mUC may be characterized by tumor microenvironments that are particularly susceptible to GCN (23,24). Additional biomarkers and basic biological studies are required to address these considerations. In summary, more than 2 years have passed since Powles et al. first presented the striking EV-302 data at the EMSO Congress (October 20–24, 2023, Madrid, Spain). The updated EV-302 data have highlighted multiple aspects of the clinical benefit of first-line EVP treatment, such as the sustained efficacy and safety, durable response, and consistent benefit across prespecified subgroups, including those defined by the primary disease site of origin and metastatic site. There is a significant lack of evidence regarding potential biomarkers, such as Nectin-4 expression, PD-1/PD-L1 expression, and circulating tumor DNA, which are useful for selecting the optimal regimen (25,26). Future clinical trials and post hoc analysis are needed to deliver a precision medicine to patients with la/mUC. Advances in antineoplastic drugs, along with an improved understanding of the pathogenesis, biological characteristics, and genetic alterations of malignant diseases, have driven the development of new combination therapies for la/mUC management. However, a radical cure remains unachievable for most patients with la/mUC; thus, further translational research and clinical trials are required. Supplementary The article’s supplementary files as Available at https://tau.amegroups.com/article/view/10.21037/tau-2025-587/prf