Bladder Cancer

Introduction

• 83K/yr in the US, M:F ~3:1 incidence and ~2.6:1 mortality, age ~73 yrs. White ↑ incidence, Black ↑ mortality.
• Risk factors: tobacco use, occupational exposures, urinary tract diseases, drugs, familial.
  • Occupational: exposure to aromatic amines (particularly 2-naphthylamine, benzidine, and polycyclic aromatic hydrocarbons) is associated with ↑ incidence of bladder ca (eg, workers in dyestuff manufacturing and rubber and aluminum industries).
  • Infection: Schistosoma hematobium → both squamous and urothelial carcinomas. Other chronic UTI, including stones and cystitis → chronic inflammation → ↑ risk of bladder ca.
  • Drugs:
    • phenacetin-containing analgesics → tumors of the renal pelvis and ureter.
    • cyclophosphamide and ifosfamide → urothelial carcinoma.
  • HNPCC (Lynch syndrome) → bladder and other urothelial ca, most notably upper-tract tumors.
    • NCCN: consider germline testing or genetic referral if <45 y/o and FHx of colon/endometrial ca.
  • Balkan nephropathy → chronic tubulointerstitial disease found in Bosnia, Bulgaria, Croatia, Romania, and Serbia; occurs exclusively in farming villages along the Danube River. Aristolochia clematitis, a plant native to the endemic region, grows in cultivated fields where its seeds comingle with wheat grain during the annual harvest; bread prepared from these fields provides enough exposure to aristolochic acid.
• Polychronotropism. The entire urothelial tract is at risk from these carcinogenic exposures (field cancerization), so pts with urothelial carcinoma in one area are at risk of separate primary cancer in other areas and at other times.
• Genes:
  • TERT promoter mutations are most common overall; FGFR3 in low-grade NMIBC; TP53 common in MIBC (RB1 less common).
  • DNA damage-repair genes (ERCC2, ATM, RB1, FANCC) may predict response to chemo and ICI.
  • TMB associated with ↑ response to ICI; bladder ca has one of the highest TMB among cancers.
  • Subtypes of MIBC: luminal and basal subtypes based on genetic profiles.
  • FGFR alterations in ~20% of pts with locally advanced or metastatic bladder ca and 70% in NMIBC and upper-tract bladder ca; not associated with reduced ORR to immunotherapy (FGFR3-altered mUC responds to PD-1/PD-L1 blockade similarly to wild-type). Erdafitinib (FGFR inhibitor) is used in metastatic bladder ca (see below).

Pathology and Location

• Urothelial carcinoma → 95% of all bladder ca
  • Urothelial carcinoma, aka transitional cell carcinoma (TCC), is ca of any structure lined by the urothelium (transitional cells) in the urinary tract.
  • Location: >90% in bladder/lower urinary tract, and 5% to 7% in the upper urinary tract (renal pelvis, majority, and ureter).
• Urothelial cancer in bladder diverticula
  • Difficult to dx and treat. Because of the limited presence of a muscularis propria layer, there is ↑ risk of bladder perforation during TURBT.
  • There is no T2 in diverticular ca; therefore staged as Ta, Tis, T1, or T3+ (indicating extradiverticular extension).
  • Treatment:
    • Without extradiverticular extension → diverticulectomy or partial cystectomy.
    • With extradiverticular extension → partial or radical cystectomy.
• Urothelial carcinoma in prostatic urethra and prostate gland
  • Difficult to dx and treat. Hard to treat with intravesical therapies because these are given via Foley catheter, which bypasses the prostatic urethra, leading to ↓ drug exposure.
  • Transurethral resection of the prostate → for staging and to ↑ exposure of the prostate to BCG.
  • If prostatic stromal invasion is found → cystoprostatectomy ± primary urethrectomy.
• Upper-tract urothelial carcinoma (associated with Lynch syndrome)
  • Ca of renal pelvis and ureter are also of urothelial origin and should be treated on the basis of primary histology; ↓ OS and different mutation profile from bladder ca.
  • Dx by imaging or retrograde pyelogram; can perform ureter cytology ± brushing during the pyelogram.
  • LN+ → considered stage IV, but resection is still performed because long-term remission is possible.
  • Ureter ca → endoscopic resection with mitomycin C gel (low grade) or nephroureterectomy with cuff of bladder resection and LN dissection ± neoadj chemo.
    • High risk (pT2 to pT4 and/or N+) → adj GC after nephroureterectomy. POUT study: 3 yr DFS 71% vs 46%; final 5-yr OS 66% vs 57% (HR 0.68); but after nephrectomy kidney function can ↓, so consider neoadj tx.
  • Neoadj chemo may be reasonable, extrapolating from bladder ca, but there are no RCTs yet.
  • ~20% to 50% will eventually develop bladder ca; surveillance cystoscopy is necessary.
• Nonurothelial bladder ca → 5% SqCC, 2% adenocarcinoma, 1% small-cell.
  • SqCC: chronic UTI, RT, cyclophosphamide, chronic indwelling catheters (spinal cord injury), and schistosomiasis (↑ in Africa) can cause SqCC and AC.
  • AC may originate from remnant urachal tissue, especially in the bladder dome.
    • Plasmacytoid: aggressive form of urothelial carcinoma associated with somatic CDH1 mutation. Histology shows discohesive single cells diffusely infiltrating tissue and spreading along fascial planes (E-cadherin loss from CDH1 mutation), appearing as vague infiltrative processes on imaging rather than discrete metastatic deposits. PD-L1 expression is variable; plasmacytoid histology does not predict PD-L1 negativity.
  • Treatment of nonurothelial bladder ca:
    • Localized nonurothelial bladder ca → cystectomy is primary tx. Role of chemo ± RT as neoadj or adj tx is not well defined. Neoadj RT may have a role for SqCC (↑ risk of local recurrence f/b metastatic spread).
    • Pure adenocarcinoma, squamous cell, or small-cell histology → generally not treated as urothelial ca.
    • Pure small-cell treated with lung ca regimens such as carbo or cis/etoposide for neoadj or metastatic disease. Consider MRI brain.
    • Pure adenocarcinoma including urachal AC: for localized urachal AC → partial or complete cystectomy with en bloc resection of the urachal ligament with umbilicus and LN dissection. Limited data on neoadj chemo and RT.
    • MIBC AC: no role for neoadj/adj chemo; if LN+ consider colorectal regimens (FOLFOX; or gemcitabine, 5-FU/leucovorin and cisplatin); if responding, consolidate with surgery.

Diagnosis and Staging

• Urine cytology → do in pts with irritative symptoms (especially heavy smokers).
• Cystoscopy and biopsy are the main tools to establish the dx.
• There is no screening for bladder ca per 2011 USPSTF guidelines.
• T staging
  • Papillary urothelial neoplasms of low malignant potential.
  • Ta → noninvasive papillary carcinoma; recurs commonly and can progress to lamina propria/muscle invasion, especially when high grade.
  • Tis or CIS → precursor for aggressive/lethal/invasive variant.
  • T1 → invade lamina propria or muscularis mucosa (subepithelial).
  • T2 → invade the muscle
    • pT2a: invade superficial muscle <50%.
    • pT2b: invade deep muscle >50%.
  • T3 → penetrates muscle and invades perivesical tissue (fat).
    • pT3a: evident microscopically.
    • pT3b: evident macroscopically (extravesical mass).
  • T4
    • T4a: extravesical tumor invading the prostate stroma, uterus, vagina, seminal vesicle.
    • T4b: extravesical tumor invading pelvic wall, abdominal wall → stage IV.
  • N1 → single LN metastasis in the true pelvis (perivesical, obturator, internal and external iliac, or sacral LN).
  • N2 → multiple LN metastases in the true pelvis.
  • N3 → LN metastasis to the common iliac LNs.
  • M1a → metastasis limited to LNs beyond the common iliacs.
• Three categories of disease:
  1. Non-muscle-invasive (NMIBC) → goal is to prevent recurrence and progression to muscle-invasive disease.
  2. Muscle-invasive (MIBC) → goal is to maximize the chance of cure with multimodality tx: chemo, RT, surgery.
  3. Metastatic disease.
• Staging workup
  • Results of cystoscopy and TURBT poorly correlate (only ~50%).
  • TURBT can only give information up to T2 level; it cannot stage more advanced disease because the urologist must avoid bladder perforation.
    • Muscularis propria should be present for accurate staging; if absent → repeat TURBT, except completely resected Ta low-grade tumors or primary CIS. T1 warrants repeat TURBT even if muscle is present.
    • CT or MRI is more reliable if done before TURBT and with a distended bladder; immediately after TURBT, adenopathy may be inflammatory.
    • PET/CT may have a role in staging muscle-invasive disease and can detect mets.
  • Histologic grading (low grade vs high grade) is more relevant for non-muscle-invasive tumors because all muscle-invasive neoplasms are high grade.

NMIBC (80% of newly dx bladder cancer)

• Non-muscle-invasive disease includes: papillary urothelial neoplasm of low malignant potential, CIS, and low- and high-grade urothelial ca.
• 70% confined to the mucosa (Ta or Tis) and 30% involving the submucosa (T1).
• Dx: positive urinary cytology → cystoscopy → TURBT with random biopsies of normal bladder.
• After TURBT, pts receive constant bladder irrigation (CBI) and are discharged with a catheter for ~1 wk.
• Risk stratification. Rule of thumb: low-grade and small tumor → low risk; high-grade and large tumor → high risk.
  • Low risk: low-grade, <3 cm, solitary, Ta without CIS or Tis → surveillance.
  • High risk: high-grade, >3 cm, multiple lesions, Tis/CIS, T1 → induction BCG.
  • Intermediate risk: ↓ grade but >3 cm and multifocal, or ↑ grade but <3 cm and solitary → induction BCG.
  • Very high risk: large or multiple T1 high-grade lesions, cooccurring T1 high-grade with CIS or involving prostatic urethra, or micropapillary/plasmacytoid/sarcomatoid variant histology, LVI → cystectomy.

Repeat TURBT, surveillance, and low-risk NMIBC treatment

• Once T1 is diagnosed with TURBT, pts need repeat TURBT in 4 to 6 weeks, as upstaging on repeat TURBT is ~15% to 20% if muscle was present and ~40% to 50% if muscle was absent (~30% overall). T1 is high-risk NMIBC: cystoscopy + cytology q3 to 4 mo x2 yr, then q6 mo yrs 3 to 4, then annually. (Low-risk: after a negative initial cystoscopy, repeat at 6 to 9 mo then annually; routine cytology unnecessary.)
• Immediate postoperative intravesical single-dose chemo: gemcitabine and mitomycin C are most commonly used, within 24 h of TURBT; ↓ recurrence by 35% but no effect on progression or mortality.
  • Recommended for pts with suspected or known low- or intermediate-risk bladder ca.
  • Not preferred if there was extensive resection during TURBT or suspected perforation.

Intermediate- or high-risk NMIBC treatment

• Durvalumab + BCG (POTOMAC): BCG-naive high-risk NMIBC; durvalumab x13 cycles added to BCG induction + maintenance improved DFS vs BCG alone (HR 0.68). FDA approved 2026.
• Induction BCG: within 4 wks after TURBT, do induction BCG weekly x6 (± intravesical INF) → CR 70%.
  • Cystoscopy and urine cytology at the first response assessment (~3 mo after TURBT); biopsy or repeat TURBT only for suspicious lesions or suspected persistent disease.
• If CR → maintenance BCG: ~1 yr for intermediate-risk, 3 yr (SWOG schedule) for high-risk.
  • CR defined as negative cytology and no evidence of residual or previous disease.
  • BCG maintenance (SWOG protocol): weekly for 3 weeks at months 3, 6, 12, 18, 24, 30, and 36. In contrast to induction, maintenance is given weekly for 3 wk.
  • Repeat cystoscopy ~1 mo after each maintenance tx.
  • 75% unable to tolerate/complete maintenance tx.
  • Relapse can occur at the ureteral orifices, ureter, or urethral areas → less accessible for intravesical tx.
  • AEs of BCG:
    • Local: cystitis, frequency, dysuria, hematuria, fever.
    • Systemic: arthralgia, osteomyelitis, sepsis, multiorgan dysfunction with granulomatous involvement.
• Failure to achieve CR after one induction is not automatically BCG-unresponsive; that requires adequate BCG and specific criteria. Persistent HG Ta or CIS may get a second induction, but HG T1 at the first post-induction evaluation qualifies as BCG-unresponsive.
  • Adequate BCG therapy: at least 5 of 6 instillations of induction BCG plus at least 2 of 3 doses of maintenance therapy, or 2 to 6 doses of a second induction.
  • BCG refractory: stage progression at 3 mo after BCG induction (Ta → T1) or persistent NMIBC at 6 mo.
  • BCG-unresponsive (after adequate BCG): recurrent high-grade Ta/T1 within 6 mo, or CIS (± papillary) within 12 mo, or persistent HG T1 at first eval. BCG-relapsing: high-grade recurrence after an initial disease-free response (may be >12 mo). Low-grade recurrence is not BCG-unresponsive.
  • Treatment:
    • If muscle-invasive → definitive bladder tx (cystectomy or RT).
    • If non-muscle-invasive:
      • Cystectomy (preferred).
      • Second induction BCG is for selected persistent/recurrent Ta or CIS after only one induction (before adequate-BCG/unresponsive criteria are met). For confirmed BCG-unresponsive disease, favor cystectomy or a bladder-preserving alternative (pembrolizumab, nadofaragene, N-803, TAR-200); do not repeat BCG monotherapy. Use second-line intravesical chemo such as gemcitabine, docetaxel, valrubicin (CR 21%), or clinical trial. Can use gem/docetaxel combo.
      • Pembrolizumab (select pts ineligible for or refusing cystectomy), KEYNOTE-057 → 41% CR.
      • Nadofaragene firadenovec-vncg (Adstiladrin): non-replicating adenoviral IFN-alpha2b gene therapy.
      • Nogapendekin alfa inbakicept (Anktiva, N-803) + BCG: IL-15 superagonist fusion protein. FDA Apr 22, 2024 for BCG-unresponsive NMIBC with CIS (± papillary). QUILT-3.032 (Chamie NEJM Evid 2023): CR 71%, median DOR 26.6 mo. Alternative to cystectomy in BCG-unresponsive CIS.
      • TAR-200 (Inlexzo, gemcitabine intravesical releasing system): pretzel-shaped Foley-inserted gemcitabine device with sustained intravesical release. FDA Sep 9, 2025 for BCG-unresponsive high-risk NMIBC with CIS ± papillary. SunRISe-1 (Necchi ASCO GU 2024): CR 82.8%, 12-mo DOR maintained in most responders. Well tolerated (mostly urinary symptoms).

MIBC (20% to 25% of newly dx bladder cancer)

Radical cystectomy + PLND (after perioperative systemic therapy when eligible)

• Cisplatin-eligible cT2N0: neoadjuvant/perioperative cisplatin-based chemo (or perioperative durvalumab, NIAGARA) f/b radical cystectomy + PLND. Cisplatin-ineligible: perioperative EV + pembrolizumab (EV-303) is an approved option; cystectomy alone if systemic therapy unsuitable or declined.
  • Radical cystectomy includes removal of:
    • Bladder, prostate, seminal vesicles, and proximal urethra for men.
    • Bladder, urethra, and uterus (including BSO), ± portion of the anterior vaginal wall for women.
  • Pelvic LN dissection is a necessary part of the radical cystectomy → ↑ outcomes.
  • Three main types of urinary diversion:
    • Ileal conduit that drains to a pouch on the anterior abdominal wall.
    • Continent cutaneous reservoir constructed from detubularized bowel segments.
    • Orthotopic neobladder.
    • AEs: similar QOL between conduits; ↓K, ↑Cl metabolic acidosis, and ↓B12 if ileum is used.
  • Poor prognosis with higher T stage, LVI, and LN+.
  • ~2/3 of pts remain recurrence free after cystectomy on long-term follow-up.
  • Partial cystectomy can provide adequate local control in select pts.

Cisplatin eligible → perioperative neoadjuvant chemotherapy f/b cystectomy

• Cisplatin ineligibility (any of):
  1. WHO or ECOG ≥ 2.
  2. CrCl <60 mL/min per Galsky criteria (40 to 50 is borderline/minimum; <40 is contraindicated).
  3. Hearing loss CTCAE (v4) grade ≥ 2.
  4. Peripheral neuropathy CTCAE (v4) grade ≥ 2.
  5. NYHA class III heart failure.
• As 30% to 50% of pts have CrCl <60, roughly half of patients are ineligible.
• A 24 h CrCl may provide a more accurate estimate in borderline renal function. Cisplatin-based tx may be feasible in borderline renal function; some clinicians use split-dose regimens, although without strong evidence that this is safer or as efficacious as standard dosing.

Neoadjuvant chemo: two options, ddMVAC or Gem/cis (pt must be cisplatin eligible)

• ddMVAC with G-CSF x3 to 6 cycles is SOC → ↑ pCR in pts with cT2 to cT4a, N0 to N1, M0 MIBC.
  • ↓ Myelotoxicity and ↓ time to deliver neoadj chemo vs classic MVAC.
  • For pts with borderline kidney function, some split the cisplatin dose (eg, 35 mg/m2 on days 1 and 2, or days 1 and 8).
  • Classic MVAC (SWOG 87-10) in T2 to T4a, N0, M0 bladder → ↑ mOS 77 mo vs 46 mo; pCR 38% vs 15%.
  • Meta-analysis of neoadj platinum-based chemo → 5% absolute OS benefit at 5 yrs.
• Gem/cis (GC): frequently used instead of ddMVAC, although phase III VESPER favored ddMVAC (neoadjuvant subgroup 5-yr OS 66% vs 57%, HR 0.71).
  • Phase II SWOG S1314: pT0 30% vs 28% and downstaging (<pT2) 40% vs 47% for GC vs ddMVAC.
  • ddMVAC is shorter than GC but carries anthracycline cardiotoxicity.
• ERCC2 and ATM/RB1/FANCC (DNA repair genes) associated with ↑ pCR to cisplatin-based chemo.
  • RETAIN trial: DDR-altered pts (ATM, RB1, FANCC, ERCC2) achieving a clinical CR (cT0, not pCR) after neoadj accelerated MVAC go to active surveillance (bladder sparing) vs radical cystectomy for pts with >pT1.

• Carboplatin is inferior to cisplatin in the neoadjuvant setting; do not substitute.
• Neoadjuvant ICI (older/pending data):
  • Neoadj pembrolizumab (PURE trial): pCR 42% in both cisplatin-eligible and -ineligible pts; ↑ pCR with ↑ PD-L1 positivity or high TMB.
  • Neoadj atezolizumab: pCR 29%.
  • Neoadj durva/tremi (Gao, Nat Med 2020) and ipi/nivo (NABUCCO, van Dijk, Nat Med 2020) have phase II data; still investigational for routine perioperative use.
• NIAGARA, perioperative durvalumab (new SOC for cisplatin-eligible MIBC): Powles et al, NEJM 2024; PMID 39282910. Cisplatin-eligible MIBC randomized to neoadj durvalumab + GC x4 → cystectomy → adj durvalumab x8 cycles, vs neoadj GC alone → cystectomy. EFS NR vs 46.1 mo (HR 0.68); OS HR 0.75 (25% reduction in death). FDA Mar 28, 2025. NCCN Cat 1 perioperative regimen for cisplatin-eligible MIBC; first FDA-approved perioperative IO regimen for MIBC.
• Perioperative enfortumab vedotin + pembrolizumab (EV+P): EV-303/KEYNOTE-905 (cisplatin-ineligible or refusing, vs cystectomy alone): EFS HR 0.40, OS HR 0.50, pCR 57.1% vs 8.6%; FDA Nov 2025. EV-304/KEYNOTE-B15 (cisplatin-eligible, vs neoadj GC): EFS HR 0.53, OS HR 0.65, pCR 55.8% vs 32.5%; FDA Jul 2026. Current Padcev label: MIBC regardless of cisplatin eligibility.

Adjuvant chemo after cystectomy

• Upfront cystectomy (no neoadj chemo) with pT3, pT4a, or pN+ → ddMVAC x3 to 6, Gem/cis x4, or nivolumab x1 yr.
• Neoadj chemo → cystectomy → residual disease (ypT2 to ypT4a or ypN+) → adjuvant nivolumab x1 yr.
• Adj chemo is preferred in pts with LN+ or extravesical extension, but no definitive trials.
• CheckMate 274 5-yr update (Ann Oncol 2026; PMID 41110694): DFS benefit preserved (HR 0.74); mOS 75.0 vs 50.1 mo (HR 0.83, 95% CI 0.67 to 1.02) in all randomized pts, not statistically significant; detectable post-op ctDNA identified pts who benefited most. Solidifies adj nivolumab as SOC for high-risk urothelial post-radical surgery, in both chemo-untreated and post-neoadj residual disease pts.
• Adj nivolumab x1 yr (CheckMate 274):
  • T3+ or N+ bladder ca without cisplatin neoadj chemo, or
  • Residual ypT2+ or N+ bladder ca after neoadj cisplatin-based chemo.
    • ↑ mDFS 20.8 mo vs 10.8 mo. Approved for high-risk MIBC post-cystectomy.
• Adj atezolizumab (IMvigor010): same DFS and OS vs observation; ctDNA+ pts seem to benefit.
• Adj pembrolizumab (AMBASSADOR / A031501, Apolo NEJM 2025; PMID 39282902): high-risk MIBC post-cystectomy → adj pembrolizumab x1 yr vs observation; met co-primary DFS endpoint (mDFS 29.6 vs 14.2 mo, HR 0.73); OS not yet significant at interim. Provides an alternative to adj nivolumab.
• Current guidelines limit perioperative (neoadj or adj) chemo to cisplatin-eligible pts. From NCCN: for pts who are not cisplatin candidates, there are no data to support perioperative cytotoxic (carboplatin-based) chemo; perioperative EV + pembrolizumab (EV-303) is the evidence-based option.
• Adj chemo vs chemo at relapse (EORTC 30994): pT3 to pT4 or N+, M0 radical cystectomy f/b immediate adj chemo (4 cycles GC, ddMVAC, or classic MVAC) vs 6 cycles chemo at relapse. Closed after only 284 of 660 planned pts.
  • Nonsignificant OS; ↑ PFS 3.11 yrs vs 0.99 yrs for adj vs delayed chemo.
  • Key point: adj chemo may be offered to high-risk pts who did not receive neoadj chemo.

Bladder preservation

• pCR rate with neoadj chemo alone is only ~20% to 40%; chemo alone cannot replace definitive tx (surgery or RT).
• Trimodality tx (TURBT f/b definitive RT plus radiosensitizing chemo) → ↑ DFS and OS approaching radical cystectomy. Appropriate for trimodality tx:
  • Solitary tumor.
  • <5 cm.
  • Visibly complete TURBT.
  • No extensive CIS (no mass on exam under anesthesia).
  • No T4 disease or tumor-associated hydronephrosis.
  • Good bladder function and capacity.
• Ideally, tumors should be maximally debulked via TURBT prior to chemoradiation.
• No phase III data on cystectomy vs trimodality; MGH trimodality series: 5 yr disease-specific survival 64%, OS 52%. Propensity-matched multi-institutional comparison (Zlotta, Lancet Oncol 2023): similar 5 yr metastasis-free survival, 75% TMT vs 74% radical cystectomy.
• Best radiosensitizing regimen is unknown; options include:
  • Weekly cisplatin monotherapy.
  • Paclitaxel.
  • Gemcitabine.
  • 5-FU and mitomycin C + RT vs RT alone (BC2001): 2 yr locoregional DFS 67% vs 54% and 5 yr OS 48% vs 35% (P=0.16) for chemoRT vs RT alone.
• Role of induction chemo (eg, GC) before chemoRT is unknown.
• Follow-up cystoscopy after RT; if less than CR → salvage cystectomy, required in ~1/3 of pts.
• RT can ↓ QOL due to ↓ function of bladder, anal sphincter, and large bowel.
• Pts with bladder preservation require lifelong surveillance.

Metastatic Bladder Cancer

First line

1L preferred for ALL pts (cisplatin-eligible and -ineligible): enfortumab vedotin + pembrolizumab (EV+P)

• EV-302 / KEYNOTE-A39 (Powles NEJM 2024; PMID 38446675): EV+P vs platinum-doublet chemo in 1L mUC (cisplatin-eligible and -ineligible).
  • mOS 31.5 vs 16.1 mo (HR 0.47); mPFS 12.5 vs 6.3 mo (HR 0.45); ORR 67.7% vs 44.4%, CR 29.1% vs 12.5%.
  • FDA Dec 15, 2023, first regimen to outperform platinum chemo in 1L mUC; new SOC for ALL 1L mUC pts (no PD-L1 selection required).
  • Toxicity: peripheral neuropathy (50%), skin rash, hyperglycemia (avoid in uncontrolled DM), pneumonitis/ILD (can be severe or fatal; per PADCEV label withhold for grade 2, permanently discontinue for grade 3 to 4), neutropenia.

If EV+P not feasible: cisplatin-eligible → nivolumab + GC (CheckMate 901) OR GC/ddMVAC + avelumab maintenance

• Nivolumab + GC (CheckMate 901, van der Heijden NEJM 2023; PMID 37870949): cisplatin-eligible mUC; mOS 21.7 vs 18.9 mo (HR 0.78), mPFS 7.9 vs 7.6 mo (HR 0.72), CR 21.7% vs 11.8%. FDA Mar 6, 2024, first immunochemo combo in 1L mUC. Note: the CheckMate 901 ipi+nivo arm vs chemo was NEGATIVE; only the nivo+GC arm is positive/approved.
• GC or ddMVAC (older approach): GC ORR ~50%, similar to MVAC, mOS ~14 to 15 mo, less toxic than MVAC (mOS 14 vs 15.2, P=0.44).
• Avelumab maintenance (JAVELIN Bladder 100, Powles NEJM 2020; PMID 32945643): in pts with CR/PR/SD after 4 to 6 cycles of platinum chemo, mOS 21.4 vs 14.3 mo. Standard maintenance after chemo if EV+P not used; start within 4 to 10 weeks after chemo.

Cisplatin-ineligible (and EV+P not feasible) → gem/carbo f/b avelumab maintenance. (Gem/carbo alone historical mOS ~9 mo, EORTC 30986; avelumab maintenance improves OS, JAVELIN Bladder 100 mOS 21.4 mo overall.)

• Gem/carbo: in contrast to the neoadjuvant setting, in metastatic disease carboplatin can substitute for cisplatin.
• Gemcitabine + taxanes (good ECOG but platinum ineligible, or fluctuating renal function).
• Pembrolizumab monotherapy: use in pts ineligible for ANY platinum, regardless of PD-L1; KEYNOTE-052 phase 2: ORR 24%, mOS 11.3 mo. Largely supplanted by EV+P.
• Historical EV-103 (phase 1b/2): EV+pembro in cisplatin-ineligible mUC, mPFS 12.3 mo, ORR 73%, 2 yr OS 56%; precursor to EV-302.
• Therapies that did NOT work in 1L (negative trials):
  • KEYNOTE-361 and IMvigor130: the chemo+IO combination arms missed their prespecified OS bars, so single-agent IO arms could not be formally tested; IMvigor130 did meet its co-primary PFS for atezo+chemo.
  • DANUBE: durvalumab ± tremelimumab vs chemo, negative.
  • CheckMate 901 ipi+nivo arm: vs chemo, negative for OS (the separate nivo+GC arm was positive, see above).

Second line

• If progressed on pembro/avelumab:
  • Gem/carbo.
  • Enfortumab vedotin.
  • Single-agent chemo with docetaxel or paclitaxel.
• If progressed on chemotherapy:
  • Pembrolizumab (KEYNOTE-045): ORR 21%, mOS 10.3 mo.
  • Avelumab: ORR 17%, mOS 6.5 mo.
  • Nivolumab (CheckMate 275): ORR 19.6%, mOS 8.7 mo.
  • Atezolizumab (IMvigor211) and durvalumab are withdrawn.

Third line and biomarker-directed options

• Erdafitinib (Balversa): pan-FGFR (FGFR1 to 4) TKI; FGFR2/3 alterations in ~8% to 20% of mUC (higher in upper tract).
  • BLC2001 phase 2: ORR 40%, CR 3%; FDA accelerated approval Apr 2019.
  • THOR cohort 1 (Loriot NEJM 2023; PMID 37937786): erdafitinib vs investigator-choice chemo (docetaxel/vinflunine) in FGFR3-altered mUC after PD-1/PD-L1, mOS 12.1 vs 7.8 mo (HR 0.64). FDA full approval Jan 19, 2024 for FGFR3-altered locally advanced/metastatic UC after at least one prior systemic therapy. THOR cohort 2 (vs pembro) was negative; do not use erdafitinib instead of IO in the IO-naive setting.
  • TRAEs: ↑ phosphate (very common, ~75%; monitor and manage per label), ↓Na, stomatitis, skin/nail changes, asthenia, and central serous retinopathy / retinal detachment (baseline eye exam, then monthly x4 mo, then every 3 mo thereafter, and urgently for any visual symptoms; include OCT).
• Enfortumab vedotin: ADC targeting Nectin-4.
  • EV-301 phase III: ↑ PFS 5.55 mo vs 3.71 mo and mOS 12.88 mo vs 8.97 mo; ORR 40.6% vs 17.9%.
  • TRAEs: neuropathy (PADCEV has no formal contraindications; pivotal trials excluded baseline grade >=2 neuropathy; hold for grade 2, discontinue for grade >=3), rash, alopecia, hyperglycemia (uncontrolled DM excluded from the trial), neutropenia.
• Sacituzumab govitecan (Trodelvy): TROP-2 ADC. TROPHY-U-01 phase 2: ORR 27%, CR 5%. Initially FDA accelerated Apr 2021 for pretreated mUC, but the urothelial indication was WITHDRAWN (announced Oct 2024; FDA withdrawal Nov 22, 2024) after confirmatory TROPiCS-04 phase 3 (vs chemo) failed its primary OS endpoint with a concerning safety signal. No longer used for mUC.
• Disitamab vedotin (HER2 ADC, RC48): investigational in the US for HER2+ mUC; approved in China. RC48-C014 supports activity.
  • Disitamab vedotin + toripalimab (RC48-C014): 1L HER2-expressing mUC, ORR 76%; supports HER2-directed ADC/IO combo strategy. Global phase 3 (DV-001) ongoing. HER2 IHC is becoming clinically actionable in urothelial ca.

Survivorship (Last Remarks)

• NMIBC → risk-adapted surveillance: high-risk needs lifelong cystoscopy and cytology ± upper-tract imaging; low-risk needs less intensive follow-up (no routine cytology or upper-tract imaging).
• MIBC → pts need radical cystectomy with urinary diversion → ↓ QOL. Pts receiving neoadj chemo: monitor for neuropathy, hearing loss, and renal dysfunction.
• Bladder-sparing/trimodality tx is an alternative to cystectomy, especially in the elderly.
• Cisplatin should be offered to any eligible patient regardless of age, but this requires careful assessment of baseline neuropathy, hearing, and kidney function, with serial reassessment.
• Consider immunotherapy and ADC for cisplatin-ineligible pts.