Introduction and Unmet Need
Urothelial carcinoma (UC), encompassing bladder cancer and upper tract urothelial carcinoma, represents a heterogeneous disease requiring stage-specific and biomarker-driven therapeutic strategies. The past 12 months (February 2025–February 2026) have witnessed transformative shifts in the treatment paradigm, particularly in locally advanced/metastatic UC (la/mUC), with the emergence of antibody-drug conjugate (ADC) combinations superseding platinum-based chemotherapy as first-line standard of care457. Despite these advances, substantial unmet needs persist across all disease stages: approximately two-thirds of patients with advanced UC do not receive second-line therapy, indicating significant attrition and highlighting the critical importance of optimizing first-line selection31. In non-muscle-invasive bladder cancer (NMIBC), BCG-unresponsive disease affects 30–40% of high-risk patients, necessitating bladder-sparing alternatives to radical cystectomy4649. This review synthesizes the current therapeutic landscape, mapping molecular targets to approved/late-stage agents and comparing efficacy and safety outcomes across disease stages and lines of therapy.
Target Landscape and Biomarker Definitions
Immune Checkpoints: PD-1/PD-L1 and Beyond
PD-L1 expression remains the most extensively validated predictive biomarker for immune checkpoint inhibitors (ICIs), though responses occur in PD-L1-negative tumors, limiting its binary predictive utility1. European Association of Urology (EAU), National Comprehensive Cancer Network (NCCN), and European Society for Medical Oncology (ESMO) guidelines recommend PD-L1 testing for locally advanced bladder cancer or upper tract urothelial carcinoma to guide ICI therapy selection2. Tumor mutational burden (TMB) and molecular UC subtypes (luminal vs basal/squamous) represent emerging biomarkers, though clinical validation remains incomplete1. Emerging immune checkpoints including TIGIT and LAG-3 have demonstrated efficacy in other solid tumors—relatlimab plus nivolumab achieved 10.1-month median progression-free survival (PFS) versus 4.6 months with nivolumab monotherapy in melanoma (hazard ratio [HR] 0.75)10—but UC-specific phase 2/3 results are not available in the retrieved materials for the specified timeframe.
FGFR2/3 Alterations
FGFR2/3 alterations occur in approximately 15–20% of advanced UC, encompassing point mutations (most commonly S249C in FGFR3), fusions/rearrangements, and amplifications1113. FGFR testing via next-generation sequencing (NGS), fluorescence in situ hybridization (FISH), and immunohistochemistry (IHC) is recommended by EAU, NCCN, and ESMO guidelines, with positive results indicating erdafitinib therapy213. Resistance mechanisms include on-target FGFR kinase domain mutations (37% of cases: N540K, V555L/M, E587Q) and off-target PI3K–mTOR pathway alterations (58% of cases: PIK3CA mutations, TSC1/2 inactivation)11.
HER2 (ERBB2)
HER2 overexpression/amplification occurs in 10–22% of advanced UC, with significant prognostic implications3041. A Mexican cohort study reported 54% of muscle-invasive UC exhibited HER2 scores ≥1+ (22% IHC 3+, 10% IHC 2+, 22% IHC 1+), with HER2-positive (3+) patients experiencing median overall survival (OS) of 9.0 months versus 31.0 months for HER2-negative patients (p=0.0029)41. HER2-low (IHC 1+ or 2+ FISH-negative) has emerged as a distinct category: disitamab vedotin achieved 31.6% ORR (95% confidence interval [CI]: 12.6–56.6%) with median PFS 5.5 months in HER2-negative/low metastatic UC61. Non-invasive HER2 assessment via [68Ga]Ga-HER2 affibody PET/CT demonstrated strong correlation with IHC (Spearman R=0.727, p=0.001) and identified HER2-positive metastatic lesions in 92.86% of cases56.
ADC Targets: Nectin-4 and Trop-2
Nectin-4 exhibits limited normal tissue expression but is overexpressed in 97% of non-invasive papillary UC (pTa), with NECTIN4 amplification predicting enhanced clinical responses1. Enfortumab vedotin (EV), targeting Nectin-4 with monomethyl auristatin E (MMAE) payload, demonstrated 44% ORR (12% complete response [CR]) in platinum-refractory, PD-1/PD-L1-inhibitor-exposed UC in the EV-201 trial1.
Trop-2 is expressed in 98% of UC (median H-score 215 on 0–300 scale in TROPHY-U-01 analysis), with efficacy observed across expression levels1. Sacituzumab govitecan (SG), conjugating anti-Trop-2 antibody to SN-38 (topoisomerase I inhibitor), achieved 27% ORR and 10.9-month median OS in platinum-refractory, PD-1/PD-L1-exposed UC (TROPHY-U-01 cohort 1)129. However, TROPiCS-04 phase III trial failed to demonstrate statistically significant OS improvement versus chemotherapy (median OS 10.3 vs 9.0 months; HR 0.86, p=0.087)1.
Modality-Specific Evidence by Disease Stage and Line of Therapy
Non-Muscle-Invasive Bladder Cancer (NMIBC): BCG-Unresponsive Disease
Nadofaragene firadenovec-vncg (Adstiladrin), an adenoviral vector-based gene therapy, received FDA approval on December 16, 2022, for high-risk BCG-unresponsive NMIBC with carcinoma in situ (CIS). The CS-003 trial demonstrated 51% CR at any time (95% CI: 40.7–61.3%), with median duration of response (DoR) 9.7 months (range: 3–52+ months) and 46% of responders maintaining CR for ≥12 months49. Urinary tumor DNA (utDNA) profiling at 3 months post-induction predicted 12-month recurrence-free survival (RFS) of 100% for MRD-negative versus 38% for MRD-positive patients (p=0.038)42.
Pembrolizumab achieved 41% CR in BCG-unresponsive CIS (KEYNOTE-057) with median DoR 16.2 months1. Cost-effectiveness analyses at $100,000 USD willingness-to-pay threshold favored nadofaragene firadenovec (incremental cost-effectiveness ratio [ICER] $10,014 per QALY) and pembrolizumab (dominated strategy, both less costly and more effective) over radical cystectomy46.
Durvalumab plus BCG (NCT03528694) met its primary disease-free survival (DFS) endpoint in high-risk, BCG-naïve NMIBC with statistically significant and clinically meaningful improvement versus BCG alone (specific HR not reported in interim analysis); durvalumab plus BCG induction-only failed to meet DFS endpoint10.
Muscle-Invasive Bladder Cancer (MIBC): Perioperative and Adjuvant Strategies
NIAGARA (durvalumab plus gemcitabine-cisplatin [GemCis] perioperative): In 1,063 patients with cT2-T4aN0/1M0 MIBC, durvalumab plus GemCis followed by durvalumab adjuvant achieved median event-free survival (EFS) not reached versus 46.1 months (HR 0.68, p<0.0001) and median OS HR 0.75 (p=0.0106), with pathologic complete response (pCR) rate 37% versus 28%. Grade ≥3 treatment-related adverse events (TRAEs) occurred in 42.3% versus 43.4%10.
CheckMate 274 (nivolumab adjuvant): Updated analysis with 36.1-month follow-up demonstrated DFS benefit in intent-to-treat population (HR 0.71) and PD-L1 ≥1% subset (HR 0.52), with emerging OS trends (ITT HR 0.76; PD-L1 ≥1% HR 0.56)1. In Europe, nivolumab is indicated for adjuvant treatment of adults with MIUC with tumor PD-L1 expression ≥1% at high risk of recurrence after radical resection 6.
Locally Advanced/Metastatic UC: First-Line Therapy
EV-302/KEYNOTE-A39 (enfortumab vedotin plus pembrolizumab): The pivotal trial establishing first-line standard of care randomized 886 cisplatin-eligible and -ineligible patients. With 2.5-year median follow-up, EV plus pembrolizumab achieved median OS 33.8 months (95% CI: 26.1–39.3) versus 15.9 months (95% CI: 13.6–18.3) for chemotherapy (HR 0.51, 95% CI: 0.43–0.61), median PFS 12.5 versus 6.3 months (HR 0.48, 95% CI: 0.41–0.57), and ORR 67.7% versus 44.4% (p<0.001), with CR rate 30.4% versus 14.5%157. Grade ≥3 TRAEs: 55.9% (EV+pembrolizumab) versus 69.5% (chemotherapy); discontinuation rate 11%10.
CheckMate 901 (nivolumab plus GemCis): In 608 cisplatin-eligible patients, nivolumab plus GemCis followed by nivolumab maintenance achieved median OS 21.7 versus 18.9 months (HR 0.78, p=0.02), median PFS 7.9 versus 7.6 months (HR 0.72, p=0.001), ORR 57.6% versus 43.1%, and CR rate 21.7% versus 11.8% (median CR duration 37.1 vs 13.2 months). Grade ≥3 TRAEs: 61.8% versus 51.7%110. A sensitivity analysis accounting for differential censoring suggested perceived-inferiority censoring confounded randomization, with adjusted OS HR 0.95 (95% CI 0.77–1.17; p=0.64)8.
Platinum-ineligible populations: Pembrolizumab monotherapy demonstrated consistent activity across variable ineligibility definitions (ECOG PS 2 ± comorbidities), achieving ORR 23.5–33.3%, median PFS 2.1–4.4 months, and median OS 9.0–10.6 months in pooled KEYNOTE-052 and LEAP-011 analysis (n=612)44.
Post-Platinum Maintenance and Later-Line Therapy
JAVELIN Bladder 100 (avelumab maintenance): In 700 patients without progression after 4–6 cycles of first-line platinum-based chemotherapy, avelumab plus best supportive care achieved median OS 21.4 versus 14.3 months (HR 0.69, p<0.001) overall, and not reached versus 17.1 months (HR 0.56, p<0.001) in PD-L1+ patients. Median PFS: 3.7 versus 2.0 months (HR 0.62). Grade ≥3 TRAEs: 16.6%; 33.7% received ≥1 year of treatment110. Exploratory analyses showed particular benefit in nonvisceral metastases (median OS 31.4 vs 17.1 months, HR 0.60) though lymph node-only disease did not reach statistical significance (HR 0.86)32.
JAVELIN Bladder Medley (avelumab plus sacituzumab govitecan maintenance): Interim analysis demonstrated median PFS 11.17 versus 3.75 months (HR 0.49, 95% CI 0.31–0.76) favoring the combination, though any-grade TRAEs occurred in 97.3% versus 63.9% and grade ≥3 TRAEs in 69.9% versus 0%3.
FGFR-altered UC: THOR trial cohort 1 (post-ICI progression) showed erdafitinib achieved median OS 12.1 versus 7.8 months (HR 0.64, p=0.005) and median PFS 5.6 versus 2.7 months (HR 0.58, p<0.001) versus chemotherapy. Grade 3–4 adverse events (AEs): 45.9% versus 46.4%17. THOR cohort 2 (anti-PD-[L]1-naive) demonstrated erdafitinib and pembrolizumab had similar median OS (10.9 vs 11.1 months, HR 1.18, p=0.18), though erdafitinib achieved higher ORR (40.0% vs 21.6%) with shorter median DoR (4.3 vs 14.4 months). Grade 3–4 AEs: 64.7% versus 50.9%19.
Pemigatinib (FIGHT-201) in FGFR3-altered UC achieved ORR 17.8–23.3%, median DoR 6.2 months, median PFS 4.0–4.3 months, and median OS 6.8–8.9 months, with diarrhea, alopecia, stomatitis, and hyperphosphatemia (42.7% each) as most common TRAEs20.
HER2-targeted ADCs: Disitamab vedotin in HER2-positive (IHC 2+ or 3+) UC achieved confirmed ORR 50.5% (95% CI: 40.6–60.3), median DoR 7.3 months, median PFS 5.9 months, and median OS 14.2 months. Grade ≥3 TRAEs: 54.2%, with peripheral sensory neuropathy (68.2% any-grade, 18.7% grade ≥3) being dose-limiting30. Combination with toripalimab (anti-PD-1) in HER2-unselected la/mUC achieved confirmed ORR 73.2%, median PFS 9.3 months, and median OS 33.1 months, with grade ≥3 TRAEs in 51.2%47.
Sacituzumab govitecan sequencing: Real-world data post-enfortumab vedotin showed limited efficacy (ORR 11%, median PFS 2.1 months, median OS 6.0 months) in heavily pretreated patients26, contrasting with TROPHY-U-01 cohort 3 (ICI-naive, platinum-refractory) where SG plus pembrolizumab achieved 41% ORR (20% CR rate), median DoR 11.1 months, median PFS 5.3 months, and median OS 12.7 months. Grade ≥3 TRAEs: 61%, with neutropenia (37%), leukopenia (20%), and diarrhea (20%) most common27.
Comparative Efficacy and Safety Summary
| Trial | Regimen | Setting/Line | N | ORR (%) | CR (%) | mPFS (mo) | mOS (mo) | Grade ≥3 TRAEs (%) |
|---|---|---|---|---|---|---|---|---|
| EV-302 | EV + pembro vs chemo | 1L la/mUC | 886 | 67.7 vs 44.4 | 30.4 vs 14.5 | 12.5 vs 6.3 | 33.8 vs 15.9 | 55.9 vs 69.5 |
| CheckMate 901 | NIVO + GemCis vs GemCis | 1L la/mUC (cis-eligible) | 608 | 57.6 vs 43.1 | 21.7 vs 11.8 | 7.9 vs 7.6 | 21.7 vs 18.9 | 61.8 vs 51.7 |
| JAVELIN Bladder 100 | Avelumab vs BSC | Maintenance post-1L platinum | 700 | — | — | 3.7 vs 2.0 | 21.4 vs 14.3 | 16.6 |
| NIAGARA | Durva + GemCis vs GemCis | Perioperative MIBC | 1,063 | — | 37 vs 28 (pCR) | EFS HR 0.68 | OS HR 0.75 | 42.3 vs 43.4 |
| THOR C1 | Erdafitinib vs chemo | Post-ICI, FGFR-altered | 266 | 40 vs — | — | 5.6 vs 2.7 | 12.1 vs 7.8 | 45.9 vs 46.4 |
| RC48-C005/C009 | Disitamab vedotin | Post-chemo, HER2+ | 107 | 50.5 | 1.9 | 5.9 | 14.2 | 54.2 |
| TROPHY-U-01 C3 | SG + pembro | Post-platinum, ICI-naive | 41 | 41 | 20 | 5.3 | 12.7 | 61 |
| FIGHT-201 | Pemigatinib | Post-treat, FGFR3-altered | 204 | 17.8–23.3 | — | 4.0–4.3 | 6.8–8.9 | Not specified |
Class-Specific Safety Considerations
ICIs: Immune-related AEs (irAEs) include pneumonitis, colitis, hepatitis, and endocrinopathies, managed via grade-based corticosteroid algorithms. Grade 1: continue ICI with monitoring; Grade 2: suspend ICI, initiate corticosteroids 0.5–1 mg/kg/day; Grade 3–4: interrupt/discontinue ICI, high-dose corticosteroids 1–2 mg/kg/day1.
FGFR inhibitors: Hyperphosphatemia (58–89%), ocular toxicity (retinopathy), nail/skin changes, and stomatitis require phosphate binders, ophthalmologic monitoring, and dose modifications11172022.
Nectin-4 ADCs (enfortumab vedotin): Skin reactions (37–67%, grade ≥3 15.5%), peripheral neuropathy (37–60%, grade ≥3 6.8%), and hyperglycemia (grade ≥3 6.1%) necessitate dermatology consultation, neuropathy dose reductions for grade ≥2, and glucose monitoring110. Routine dexamethasone premedication (6.6 mg IV) demonstrated feasibility with low severe nausea (0% grade ≥3) and cutaneous toxicity rates (31.7%, all grade 1–2)50.
Trop-2 ADCs (sacituzumab govitecan): Neutropenia (34–64%, febrile neutropenia ~7–10%), diarrhea (62%, grade ≥3 8–20%), nausea, and fatigue require G-CSF support, loperamide/atropine, and UGT1A1 genotype consideration (though discontinuation rates remain consistent across genotypes)12729.
HER2 ADCs: Peripheral sensory neuropathy (68.2% any-grade, 18.7% grade ≥3 for disitamab vedotin), hepatotoxicity, and myelosuppression mandate dose modifications and neurologic assessments3047.
Treatment Sequencing and Future Directions
Proposed sequencing algorithm:
- First-line la/mUC: EV + pembrolizumab (standard for cisplatin-eligible and -ineligible); nivolumab + GemCis (alternative for cisplatin-eligible).
- Biomarker-selected first-line: FGFR-altered patients may consider erdafitinib (particularly if anti-PD-[L]1-naive per THOR cohort 2 equipoise); HER2+ patients represent candidates for disitamab vedotin ± toripalimab.
- Maintenance post-platinum: Avelumab (if not on EV+pembrolizumab first-line).
- Post-ICI progression: FGFR-altered patients receive erdafitinib; HER2+ patients receive HER2-targeted ADC; unselected patients receive EV monotherapy.
- Post-EV progression: Limited efficacy of SG in real-world post-EV setting suggests platinum rechallenge or clinical trial consideration2631.
Testing recommendations: Baseline tissue NGS for FGFR2/3 alterations, ERBB2 amplification/mutation, PD-L1 (CPS or IC scoring per assay), and TMB. Circulating tumor DNA (ctDNA) at progression for acquired resistance mutations (FGFR kinase domain mutations, PIK3CA)11. HER2 PET/CT under investigation as non-invasive monitoring56.
Near-term catalysts: PROOF302 (infigratinib adjuvant in FGFR3-altered resected UC)52, novel intravesical delivery systems (TAR-200, UGN-102)54, and expansion of perioperative ICI strategies to FGFR/HER2-targeted perioperative regimens represent active areas of investigation.
Conclusion
The urothelial carcinoma therapeutic landscape has undergone paradigm transformation, with EV + pembrolizumab establishing median OS exceeding 30 months in first-line la/mUC (HR 0.51 vs chemotherapy)57, perioperative durvalumab plus GemCis achieving EFS HR 0.68 in MIBC10, and nadofaragene firadenovec providing bladder-sparing option with 51% CR in BCG-unresponsive CIS49. Biomarker-driven strategies targeting FGFR2/3 (erdafitinib: median OS 12.1 months post-ICI)17 and HER2 (disitamab vedotin: 50.5% ORR in HER2+; 31.6% ORR in HER2-low)3061 expand therapeutic options in molecularly defined subsets. However, sequential therapy attrition31, emerging resistance mechanisms11, and modest activity of later-line agents post-ADC exposure26 underscore the imperative for upfront optimization, rational sequencing based on molecular testing, and development of resistance-directed combinations. The integration of ctDNA-guided MRD monitoring42, non-invasive imaging biomarkers56, and perioperative ctDNA clearance strategies may further refine treatment selection and intensification approaches across the UC continuum.