Home / GU oncology / Bladder / urothelial / Adjuvant urothelial

Bladder / urothelial → Adjuvant urothelial

Last reviewed: 2026-09-22 · Next review: 2026-12-21

clinic-ready — perioperative EV+pembro and NIAGARA now change who still needs classic adjuvant IO; ctDNA MRD selects adjuvant atezo

Evidence page

Who this applies to

  • Post-cystectomy (or appropriately selected post-nephroureterectomy) with high-risk pathologic features (CheckMate 274: ypT2–T4a and/or ypN+ after neoadjuvant cisplatin, or pT3–T4a and/or pN+ without neoadjuvant chemo — exact eligibility in the protocol/label; restage from source if needed)1.
  • Note neoadjuvant / perioperative exposure: prior EV+pembro or durvalumab changes adjuvant counseling (overlap, cumulative neuropathy, irAE)34.
  • Cisplatin eligibility, residual disease, margins, histology/variant, surgical recovery, and ctDNA MRD status if considering IMvigor011-path atezolizumab56.

Standard options

Clinical contextOptionKey inclusionTypical duration
High-risk disease after radical surgery without perioperative ICINivolumab or observation1CheckMate 274-type risk and recovery1.One year in trial1.
High-risk disease after radical surgeryPembrolizumab or observation2AMBASSADOR DFS evidence; OS was not reported in the retrieved abstract2.One year in trial2.
Post-cystectomy ctDNA MRD-positive diseaseAtezolizumab pathway5IMvigor011/FDA companion-diagnostic setting5.Up to one year in trial5.
Prior perioperative ICI/ADC or outside evidenceIndividualize surveillance or trial34Avoid untested second adjuvant ICI sequencing35.Individualized5.
  • If perioperative EV+pembro or NIAGARA already given or planned as the MIBC pathway: adjuvant counseling is part of that perioperative package, not a separate CheckMate 274 default34.
  • If high-risk after RC and no prior perioperative ICI: adjuvant nivolumab 240 mg q2wk ×1 year (CheckMate 274) vs observation vs trial1. AMBASSADOR pembrolizumab 200 mg q3wk ×1 year is DFS-positive; OS not in the retrieved abstract2.
  • If post-cystectomy ctDNA MRD+ by Signatera CDx (serial testing window per label): adjuvant atezolizumab (IMvigor011 / FDA 15 May 2026) — distinct from the negative unselected IMvigor010 trial5.
  • Do not use adjuvant atezolizumab based on IMvigor010 (unselected high-risk)5.
  • Observation remains reasonable when risk is lower, recovery is poor, ctDNA remains negative, or irAE risk is prohibitive56.
  • Upper-tract: CheckMate 274 enrolled MIUC including UTUC; do not invent UTUC-specific HRs — not separately reported in the retrieved abstracts1.

Therapy sequencing

Prior treatment or postoperative stateNext treatment orderBranch point / evidence boundary
Resectable MIBC selected for EV-303/EV-304Neoadjuvant EV plus pembrolizumab → cystectomy → protocol-defined adjuvant EV plus pembrolizumab4.This is one perioperative package; do not add separate adjuvant nivolumab by default after prior perioperative ICI/EV41.
Cisplatin-eligible MIBC selected for NIAGARANeoadjuvant durvalumab plus GC → cystectomy → adjuvant durvalumab3.NIAGARA likewise tests one perioperative package, not subsequent adjuvant-ICI switching3.
High-risk disease after radical surgery without perioperative ICIAdjuvant nivolumab or observation; pembrolizumab is a DFS-positive alternative in AMBASSADOR12.Choose after recovery and immune-risk assessment; the trials do not establish serial PD-1/PD-L1 therapy12.
ctDNA MRD-positive after cystectomyAtezolizumab up to one year in the IMvigor011/FDA companion-diagnostic pathway5.This is distinct from unselected IMvigor010, which was negative5.
High-risk residual disease after perioperative EV plus pembrolizumab or NIAGARASurveillance or clinical trial discussion34.A preferred salvage-adjuvant sequence is not established in the retrieved sources35.

Surveillance and treatment de-escalation

Post-cystectomy imaging schedule (EAU practice framework)

Time after cystectomyChest/abdomen/pelvis CT (incl. upper tract when indicated)6Notes6
Every 6 months through 36 months6CT6Panel framework; intensify upper-tract monitoring if multifocal disease, CIS, or positive ureteral margins6
Annually thereafter through ~60 months6CT6Consider stopping routine CT after 5 years in majority of patients without ongoing high-risk features6
PhaseMonitoring cadenceStop, hold, or de-escalateRe-escalation / next line
Adjuvant nivolumab or pembrolizumab ×1 yearLabs and immune-symptom review each cycle; ctDNA serial testing if using IMvigor011 pathway125.Complete 1-year adjuvant course unless grade ≥3/4 immune toxicity; hold for immune events per label12. ctDNA-negative patients should not receive unselected adjuvant atezolizumab (IMvigor010 negative)5.ctDNA conversion or radiographic recurrence → systemic therapy; avoid untested second adjuvant ICI after perioperative EV/pembro or NIAGARA345.
ctDNA MRD-positive adjuvant atezolizumabctDNA per companion-diagnostic schedule; immune monitoring5.Stop at 1 year or progression; hold for immune toxicity5.Progression → mUC systemic options56.
  • Prior perioperative EV plus pembrolizumab or NIAGARA completes the intended adjuvant component; separate CheckMate 274-style adjuvant ICI is not the default34.
  • Observation remains reasonable when risk is lower, recovery is poor, or prior perioperative ICI makes adjuvant PD-1 inappropriate; imaging follows the EAU schedule above156.
CheckMate 274 published DFS landmarks

Published DFS landmarks only at 36, 48, and 60 months (nivolumab 45.1%, 41.2%, 36.4% vs placebo 35.3%, 33.0%, 30.9%); not a full KM reconstruction8.

Landmark evidence

These rows mix perioperative (changes adjuvant need) and true adjuvant trials. Indirect comparison only5.

TrialPopulationIntervention armControl armKey outcomes (both arms)Key HREvidence status
EV-303 / KEYNOTE-905 NCT03924895Cisplatin-ineligible or declined MIBC; RC+PLND candidates; N=3444Neoadjuvant EV 1.25 mg/kg d1,8 + pembro 200 mg q3wk ×3 → RC+PLND → adjuvant EV ×6 + pembro (17 total pembro cycles)4RC+PLND alone42-y EFS 74.7% vs 39.4%; 2-y OS 79.7% vs 63.1%; pCR 57.1% vs 8.6%; grade ≥3 AE 71.3% vs 45.9%4EFS HR 0.40 (0.28–0.57); OS HR 0.50 (0.33–0.74)4Vulsteke 2026 NEJM; FDA Padcev label cisplatin-ineligible4
EV-304 / KEYNOTE-B15 NCT04700124Cisplatin-eligible MIBC; N=8084Neoadjuvant EV 1.25 mg/kg d1,8 + pembro 200 mg ×4 → RC+PLND → adjuvant EV ×5 + pembro ×134Neoadjuvant cisplatin 70 mg/m² d1 + gemcitabine 1000 mg/m² d1,8 ×4 → RC4Median EFS NR vs 48.5 mo; 2-y EFS 79.4% vs 66.2%; 2-y OS 86.9% vs 81.3%; pCR 55.8% vs 32.5%; grade ≥3 AE 75.7% vs 67.2%4EFS HR 0.53 (0.41–0.70); OS HR 0.65 (0.48–0.89)4FDA Padcev label / EV-304 labeled perioperative use4
NIAGARA NCT03732677Cisplatin-eligible MIBC; N=533 vs 5303Neoadjuvant durvalumab + GC q3wk ×4 → RC → adjuvant durvalumab q4wk ×83Neoadjuvant GC ×4 → RC (no adjuvant)324-mo EFS 67.8% vs 59.8%; 24-mo OS 82.2% vs 75.2%; TRAE grade 3–4 40.6% vs 40.9%; RC 88.0% vs 83.2%. pCR % not in NEJM abstract; FDA summary: pCR not statistically different3EFS HR 0.68 (0.56–0.82); OS HR 0.75 (0.59–0.93)3Powles 2024 NEJM PMID 392829103
CheckMate 274High-risk MIUC after radical surgery; N=7091Adjuvant nivo 240 mg q2wk ×1 y1Placebo ×1 y1Primary median DFS 20.8 vs 10.8 mo1. At ≥5-y minimum follow-up: median DFS 21.9 vs 11.0 mo (not a 5-y DFS %); DFS landmarks 45.1%/41.2%/36.4% vs 35.3%/33.0%/30.9% at 36/48/60 mo; median OS 75.0 vs 50.1 mo8.DFS HR 0.70 (98.22% CI 0.55–0.90) primary1; 5-y DFS HR 0.74 (0.61–0.90); OS HR 0.83 (0.67–1.02)8Bajorin 2021 NEJM PMID 340776431; Galsky 2026 Ann Oncol PMID 411106948
AMBASSADOR A031501 NCT03244384High-risk MIUC after radical surgery; N=7022Adjuvant pembro 200 mg q3wk ×1 y2Observation2Median DFS 29.6 vs 14.2 mo; grade ≥3 AE 50.6% vs 31.6%. OS not in abstract2DFS HR 0.73 (0.59–0.90)22025 NEJM PMID 392829022
IMvigor010 NCT02450331High-risk MIUC after radical surgery; N=8095Adjuvant atezolizumab 1200 mg q3wk ×16 / 1 y5Observation5Primary DFS not met. Both-arm DFS medians not in retrieved abstract5DFS HR not in retrieved abstract5Unselected; do not use — guideline/practice boundary5
IMvigor011 NCT04660344Post-cystectomy MIBC with ctDNA MRD+ (serial); N=250 (2:1)5Atezolizumab 1680 mg q4wk ≤12 cycles / 1 y5Placebo5Median DFS 9.9 vs 4.8 mo; median OS 32.8 vs 21.1 mo5DFS HR 0.64 (0.47–0.87); OS HR 0.59 (0.39–0.90)5FDA 15 May 2026; companion Signatera CDx — practice framing per guideline5

Biomarkers

MarkerCurrent clinic roleEmerging / investigationalHow it changes decision nowLimits
PD-L1Not required for CheckMate 274 ITT DFS benefit1Larger DFS effect in PD-L1 ≥1% subgroup (primary DFS HR 0.55; 5-y DFS HR 0.58)18Do not withhold nivolumab solely for PD-L1-negative disease based on ITT1.Subgroup enrichment is not a registration selector for adjuvant nivolumab18.
ctDNA MRD (Signatera)Companion diagnostic selecting adjuvant atezolizumab after cystectomy (IMvigor011 / FDA)5CheckMate 274 5-year exploratory baseline ctDNA correlates with recurrence risk8Serial testing until positive or end of the recommended 12-month window selects the atezolizumab pathway5.Exploratory CheckMate 274 ctDNA is not the registration companion for nivolumab18.
Nectin-4Not an adjuvant IO selector in these trials15Perioperative EV plus pembrolizumab context on neoadjuvant/mUC pages4No change to classic adjuvant IO choice on this page15.Not used to select adjuvant nivolumab, pembrolizumab, or atezolizumab15.
HER2Not an adjuvant IO selector15Metastatic / tumor-agnostic HER2 context elsewhere5No adjuvant decision change on this page15.No validated adjuvant assay role here15.
FGFRNot an adjuvant IO selector15Metastatic FGFR3 / erdafitinib pathway5No adjuvant decision change on this page15.Adjuvant setting does not use FGFR to select IO15.

Upcoming research

Field direction: perioperative EV plus pembrolizumab and NIAGARA are reshaping who still needs classic adjuvant IO, while ctDNA MRD (IMvigor011) adds a biomarker branch; residual disease after perioperative therapy and dual EV/IO then MRD-positive pathways remain open questions345.

Program or questionWhy it may change careEvidence status
AMBASSADOR overall survivalClarifies whether adjuvant pembrolizumab DFS benefit translates to mature OS versus observation2.Peer-reviewed DFS report; OS follow-up pending in this pack2.
High-risk yp residual disease after perioperative EV plus pembrolizumab or NIAGARADefines adjuvant or next systemic options when pathologic residual disease persists34.Not answered in retrieved sources; treat as a care-changing evidence gap34.
Perioperative EV plus pembrolizumab then later ctDNA MRD positivitySequencing when both pathways apply45.Not defined in retrieved sources45.

Toxicity

Regimen and comparatorTrial-level safety outcomesCounseling focus
Nivolumab versus placebo, CheckMate 274Grade ≥3 treatment-related AEs 17.9% vs 7.2%1.Promptly report cough/dyspnea, diarrhea, abdominal pain, rash, jaundice, fatigue, or endocrine symptoms; immune toxicity may require a hold, corticosteroids, or hormone replacement1.
Pembrolizumab versus observation, AMBASSADORGrade ≥3 AEs 50.6% vs 31.6%2.Use the same immune-toxicity counseling; coordinate with recovery after radical surgery2.
Perioperative EV plus pembrolizumabGrade ≥3 AEs 71.3% vs 45.9% in EV-303 and 75.7% vs 67.2% in EV-3044.Report rash, neuropathy, visual symptoms, thirst/polyuria, or immune symptoms promptly4.
Perioperative durvalumab plus GCGrade 3–4 treatment-related AEs 40.6% vs 40.9% in NIAGARA3.Review fever/infection, cytopenias, renal and neuropathy symptoms, and immune toxicity3.

On-treatment monitoring

SOC/option regimenHallmark monitoring toxicity and available ratePatient action / clinic response
Nivolumab, pembrolizumab, or atezolizumab adjuvant pathwayCheckMate 274 nivolumab: grade ≥3 treatment-related AEs 17.9% vs 7.2% with placebo1. AMBASSADOR pembrolizumab: grade ≥3 AEs 50.6% vs 31.6% with observation2. A comparable IMvigor011 named immune-event rate was not retrieved5.Before each dose, review bowel, lung, liver, thyroid/adrenal, and skin symptoms. Persistent diarrhea/abdominal pain, cough/dyspnea, jaundice, marked fatigue, or palpitations needs prompt contact; immune events may require holding treatment, corticosteroids, or lasting hormone replacement12.
Perioperative EV plus pembrolizumabEV-303 FDA-label treatment-arm rates (single-arm label rates; no paired comparator % in the retained label table): rash 54%, grade 3–4 7%; neuropathy 39%, grade 3–4 3%; increased glucose 72%, grade 3–4 12%4. EV-304: rash 63%, grade 3–4 12%; neuropathy 43%, grade 3–4 3%; hyperglycemia 20%, grade 3–4 7%4. Trial-level grade ≥3 AEs 71.3% vs 45.9% with RC+PLND alone (EV-303) and 75.7% vs 67.2% with neoadjuvant GC (EV-304)4.Monitor cumulative neuropathy, glucose symptoms, skin, vision, renal function, and immune toxicity before postoperative doses; new numbness/weakness, thirst/polyuria, rash, vision change, dyspnea, or persistent diarrhea requires prompt assessment4.
Perioperative durvalumab plus GCNIAGARA grade 3–4 treatment-related AEs 40.6% vs 40.9% with neoadjuvant GC alone; a comparable named monitoring rate was not extracted3.Monitor CBC, renal function, electrolytes, neuropathy/hearing, and immune symptoms; fever, infection symptoms, reduced urine output, dyspnea, or severe diarrhea needs urgent assessment3.

Epic phrase

#Adjuvant urothelial carcinoma

  • Counseling: CheckMate 274 adjuvant nivolumab improved disease-free survival versus placebo in high-risk MIUC (DFS HR 0.70 vs placebo). AMBASSADOR pembrolizumab improved DFS versus observation (HR 0.73 vs observation); overall survival was not reported in the retrieved abstract. IMvigor011 ctDNA MRD-positive adjuvant atezolizumab improved DFS (HR 0.64 vs placebo) and OS (HR 0.59 vs placebo)—distinct from negative unselected IMvigor010. Prior perioperative EV/pembro or NIAGARA changes who still needs separate adjuvant ICI.
  • Counseling: Reviewed surveillance and de-escalation: post-cystectomy chest/abdomen/pelvis CT every 6 months through 36 months then annually through ~60 months (EAU framework; intensify upper-tract monitoring if multifocal disease, CIS, or positive ureteral margins); adjuvant PD-1 for up to 1 year then stop unless progression/toxicity; ctDNA-negative patients should not receive unselected adjuvant atezolizumab; recurrence → metastatic urothelial pathway.
  • Monitoring: With adjuvant nivolumab, pembrolizumab, or atezolizumab, review new diarrhea/abdominal pain, cough/dyspnea, jaundice, rash, marked fatigue, palpitations, or temperature/weight intolerance before every treatment. With perioperative EV plus pembrolizumab, additionally review numbness/weakness, vision change, thirst/polyuria, and rash; with durvalumab plus GC, monitor CBC, renal function, electrolytes, neuropathy/hearing, and immune symptoms. Promptly contact oncology for these symptoms. *** after-hours/urgent-assessment instructions reviewed.
  • Discussed pathology and staging: *** (ypT / pT, N, margins, histology, variant features).
  • Discussed perioperative course: neoadjuvant / perioperative * (EV+pembro vs durvalumab+GC vs GC vs none; cycles / response); surgery date *. Discussed that perioperative EV+pembrolizumab is FDA-approved for MIBC cystectomy candidates regardless of cisplatin eligibility (EV-303 cisplatin-ineligible; EV-304 cisplatin-eligible, FDA 10 Jul 2026) and NIAGARA (durvalumab + GC) remains an alternative for cisplatin-eligible patients — these change adjuvant counseling. Adjuvant nivolumab (CheckMate 274) is for high-risk patients who did not already receive perioperative ICI. For post-cystectomy ctDNA MRD+ (Signatera CDx), adjuvant atezolizumab (IMvigor011) is FDA-approved — distinct from negative unselected IMvigor010.
  • Discussed sequence: perioperative EV+pembrolizumab or durvalumab+GC already includes its postoperative component; do not layer separate adjuvant nivolumab by default after prior perioperative ICI/ADC. If high-risk after surgery without perioperative ICI, consider adjuvant nivolumab or observation; if ctDNA MRD-positive, use the separate atezolizumab pathway. ***
  • Discussed adjuvant options: observation vs adjuvant immunotherapy (nivolumab per CheckMate 274 criteria; pembrolizumab per AMBASSADOR; atezolizumab if ctDNA MRD+) vs other per *** guidelines / trial.
  • Eligibility / rationale: * (risk features that drive adjuvant Rx; prior perioperative ICI / ADC: *; ctDNA MRD: ***).
  • Discussed prognosis and goals.
  • Discussed toxicities of planned adjuvant therapy: ***.
  • He / she opts to proceed with ***.
  • Labs: CBC, CMP; ***
  • Imaging surveillance plan: ***
  • Referrals: ***
  • Follow-up: ***

Edit markdown

Guideline references

Sources

  1. Bajorin et al. (2021). CheckMate 274. *New England Journal of Medicine*. PMID: 34077643
  2. Apolo et al. (2025). AMBASSADOR. *New England Journal of Medicine*. PMID: 39282902
  3. Powles et al. (2024). NIAGARA. *New England Journal of Medicine*. PMID: 39282910
  4. FDA Padcev label, revised 2026
  5. NCCN Bladder Cancer
  6. EAU Muscle-invasive and metastatic bladder cancer guideline
  7. ESMO bladder cancer guideline
  8. Galsky et al. (2026). CheckMate 274 5-year efficacy and ctDNA. *Annals of Oncology*. PMID: 41110694

Changelog

  • 2026-09-22: Replaced surveillance with EAU post-cystectomy CT every 6 months ×3 years then annual through ~60 months; added CheckMate 274 DFS landmark figure and Source #8; clarified median DFS vs landmark % in the landmark table; expanded Biomarkers to five-column table; mirrored EAU cadence in paired dotphrase.
  • 2026-09-22: Renamed Upcoming trial results to Upcoming research; reframed for ongoing trials and field direction.
  • 2026-09-22: Named placebo/observation/GC/surgery comparators in monitoring rates and labeled EV FDA rates as single-arm; mirrored comparator-named counseling in the paired dotphrase.
  • 2026-09-22: Added line-level source superscripts ([^n]) linked to numbered Sources.
  • 2026-09-22: Added Surveillance and treatment de-escalation section; strengthened quantified benefit counseling in dotphrase.
  • 2026-09-21: Added EV-303 and EV-304 FDA-label neuropathy, glucose, and rash monitoring rates to the perioperative row.
  • 2026-09-21: Added treatment sequencing and regimen-level on-treatment monitoring, including the IMvigor011 safety-reporting boundary.
  • 2026-09-21: Renamed the prospective-evidence section to Upcoming trial results.
  • 2026-09-21: Removed Bottom line; added a compact Guidelines table and disease → setting page framing. No treatment recommendations or trial statistics changed.
  • 2026-09-21: filled EV-304 FDA approval (10 Jul 2026 — all MIBC cystectomy candidates); added IMvigor011 / Signatera ctDNA MRD adjuvant atezolizumab (FDA 15 May 2026) and clarified IMvigor010 remains negative for unselected adjuvant atezo
  • 2026-09-14: expanded from stub; landscape table includes EV-303, EV-304 (now published), NIAGARA, CheckMate 274 5-year, AMBASSADOR, IMvigor010; perioperative SOC explicitly changes adjuvant counseling
  • 2026-09-14: initial stub

Edit evidence markdown