Highlight
– In SunRISe-4, neoadjuvant TAR-200 (gemcitabine intravesical system) plus the PD‑1 antibody cetrelimab produced a pathologic complete response (pCR) rate of 38% and pathological overall response (pOR; ≤ypT1N0) of 53% in patients with cT2–T4a N0M0 muscle‑invasive bladder cancer (MIBC) who were ineligible for or refused cisplatin.
– Cetrelimab monotherapy yielded lower pCR (28%) and pOR (44%) rates; 1‑year recurrence‑free survival (RFS) was 77% with the combination versus 64% with monotherapy.
– Urinary tumor DNA (utDNA) negativity and clearance before cystectomy were strongly associated with pCR; ctDNA negativity at baseline and week 12 associated with improved RFS, supporting complementary roles for utDNA and ctDNA as MRD biomarkers.
Background: Unmet need in neoadjuvant management of MIBC
Radical cystectomy (RC) with pelvic lymph node dissection remains the standard curative-intent treatment for localized muscle‑invasive bladder cancer (MIBC). Randomized and phase III data have established a survival benefit for neoadjuvant cisplatin‑based combination chemotherapy, which improves long‑term outcomes primarily through increased rates of pathological complete response (pCR) at RC (for example, SWOG 8710/Neoadjuvant MVAC; Grossman et al.). However, a substantial proportion of patients are ineligible for cisplatin because of comorbidities (renal dysfunction, hearing loss, poor performance status) or decline it because of toxicity concerns.
For cisplatin‑ineligible patients, optimal neoadjuvant strategies remain undefined. Immune checkpoint inhibitors (ICIs) have activity in advanced bladder cancer and limited neoadjuvant data suggest they can induce pCR in a subset of patients. Intravesical delivery of cytotoxic agents designed to provide high local concentrations with reduced systemic exposure is an attractive strategy to increase local tumor kill in the bladder while minimizing systemic toxicity.
Study design: SunRISe-4 (NCT04919512)
SunRISe-4 is an open‑label, multicenter, randomized (5:3), parallel‑cohort phase II trial enrolling adults with ECOG performance status 0–1 and clinical stage cT2–T4a N0M0 MIBC who were cisplatin‑ineligible or declined cisplatin and planned for RC. Randomization was stratified by completeness of transurethral resection of bladder tumor (TURBT; residual tumor permitted up to 3 cm) and T stage.
Two treatment arms were evaluated: Cohort 1 (C1) received TAR‑200 plus cetrelimab, and Cohort 2 (C2) received cetrelimab monotherapy. TAR‑200 is a gemcitabine intravesical system designed for sustained local delivery of gemcitabine to the bladder mucosa. Cetrelimab is an anti‑PD‑1 monoclonal antibody. Treatment proceeded with planned RC; the primary endpoint was pathologic complete response (pCR) at RC. Secondary endpoints included recurrence‑free survival (RFS) and safety. Exploratory biomarker analyses assessed molecular residual disease (MRD) using urinary tumor DNA (utDNA) and circulating tumor DNA (ctDNA) at baseline and prior to RC (week 12). The protocol prespecified a side‑by‑side descriptive efficacy summary between cohorts rather than a formal hypothesis test of superiority.
Key findings
At the data cutoff (May 9, 2025), 159 patients received study treatment; 88 in C1 and 46 in C2 proceeded to RC and were evaluable for the primary endpoint.
Pathologic outcomes and recurrence‑free survival
– pCR rates: 38% in the TAR‑200 plus cetrelimab cohort (C1) versus 28% in the cetrelimab monotherapy cohort (C2).
– Pathological overall response (pOR; ≤ypT1N0): 53% in C1 versus 44% in C2.
– One‑year RFS: 77% with the combination versus 64% with monotherapy.
pCR rates were reported as consistent across key subgroups (including T stage and TURBT completeness), although the report emphasizes a descriptive comparison rather than a formal comparative test for superiority.
Safety
No new safety signals were observed with either regimen in the neoadjuvant setting. The safety profile of cetrelimab was consistent with class‑related immune‑mediated effects expected for PD‑1 blockade. Intravesical TAR‑200 was not associated with unexpected local or systemic toxicity in the reported analysis. Detailed rates of specific adverse events and grade ≥3 toxicities should be consulted in the full manuscript for clinical decision making.
Biomarker results: utDNA and ctDNA as MRD indicators
Exploratory molecular analyses showed a strong association between urinary tumor DNA (utDNA) negativity prior to RC (week 12) and pathological complete response (pCR). Specifically, utDNA‑ status before RC and utDNA clearance correlated with pCR with high statistical significance (P < 10⁻⁵ and P < 10⁻³, respectively), indicating that local intravesical MRD measured in urine is tightly linked to residual tumor in the bladder at surgery.
Circulating tumor DNA (ctDNA) status also provided prognostic information: ctDNA‑negative status at baseline and at week 12 was associated with longer RFS compared with ctDNA‑positive status at the same timepoints (P = .04 and .01, respectively). This supports a model in which utDNA primarily reports local intravesical residual disease while ctDNA captures systemic micrometastatic risk.
Interpretation and clinical implications
The SunRISe‑4 primary analysis provides evidence that adding a local sustained‑release intravesical cytotoxic (TAR‑200) to systemic PD‑1 blockade (cetrelimab) can increase pCR and pOR rates, and yield numerically higher 1‑year RFS, in a population of cisplatin‑ineligible or cisplatin‑refusing patients with MIBC. For clinicians, these data are important because they address a real clinical gap: effective neoadjuvant options for patients who cannot receive cisplatin.
Key clinical inferences include:
- TAR‑200 plus an ICI may augment local tumor eradication in the bladder beyond ICI alone, potentially translating into improved short‑term oncologic endpoints.
- The absence of new safety signals suggests the combination is tolerable in the neoadjuvant window, although larger datasets and longer follow‑up are needed to fully characterize rare or late toxicities.
- MRD assays have practical translational potential: utDNA appears to be an effective biomarker of local bladder response (predicting pCR), whereas ctDNA identifies patients at higher risk of systemic recurrence and worse RFS. Together, they could help personalize decisions about the need for additional systemic therapy or intensified surveillance after RC.
Strengths and limitations
Strengths of the trial include a pragmatic design that enrolled a clinically relevant, cisplatin‑ineligible population and incorporation of contemporaneous tissue‑agnostic MRD assays (utDNA and ctDNA) with prespecified timing. The study also used a pathologic endpoint (pCR) known to correlate with long‑term survival in MIBC.
Important limitations must be emphasized:
- The study was open‑label and phase II with a descriptive side‑by‑side comparison rather than a randomized superiority design against a standard neoadjuvant comparator (e.g., cisplatin‑based chemotherapy or observation). Thus, although the combination yielded numerically higher pCR and RFS rates, causality and formal statistical superiority remain to be demonstrated in larger randomized trials.
- The randomized allocation was 5:3 favoring the combination arm; this unbalanced design may complicate comparative interpretation and limits statistical power for head‑to‑head comparisons.
- Follow‑up duration in the primary analysis is limited; longer follow‑up is required to determine whether pCR and short‑term RFS improvements translate into durable survival benefits.
- Although utDNA and ctDNA findings are compelling, standardized assays, thresholds for positivity, and prospective validation in independent cohorts are necessary before routine clinical adoption.
Expert commentary and next steps
For clinicians and trialists, SunRISe‑4 offers several actionable takeaways: (1) intravesical sustained‑release gemcitabine plus PD‑1 blockade is a promising neoadjuvant approach for cisplatin‑ineligible MIBC; (2) combining local and systemic modalities may achieve complementary control of local disease and micrometastatic risk; and (3) integrating utDNA and ctDNA into trial endpoints and clinical workflows could optimize patient selection for additional therapy and surveillance intensity.
Future work should prioritize randomized phase III testing against contemporary standards or best available alternatives, standardized MRD assay platforms, correlative studies linking MRD dynamics with immune and molecular tumor features, and evaluation of bladder‑preserving strategies when sustained complete responses are documented.
Conclusions
In this primary analysis of SunRISe‑4, TAR‑200 plus cetrelimab produced higher pCR, pOR, and 1‑year RFS rates than cetrelimab monotherapy in cisplatin‑ineligible/refusing MIBC, without new safety signals. The strong correlation of utDNA negativity/clearance with pCR and the prognostic value of ctDNA negativity for RFS support the complementary use of urinary and circulating MRD assays to monitor local and systemic disease. These findings justify further randomized evaluation of the combination and prospective validation of MRD assays to guide personalized neoadjuvant and adjuvant strategies in MIBC.
Funding and clinicaltrials.gov
ClinicalTrials.gov identifier: NCT04919512 (SunRISe‑4). Funding and detailed disclosures are reported in the primary publication (Necchi et al., J Clin Oncol. 2025).
References
1. Necchi A, Guerrero‑Ramos F, Crispen PL, et al. Gemcitabine intravesical system plus cetrelimab or cetrelimab alone as neoadjuvant therapy in patients with MIBC: primary analysis and biomarker results of SunRISe‑4. J Clin Oncol. 2025 Dec 3:101200JCO2502382. doi:10.1200/JCO-25-02382 . Epub ahead of print. PMID: 41337691 .
2. Grossman HB, Natale RB, Tangen CM, et al. Neoadjuvant chemotherapy plus cystectomy compared with cystectomy alone for locally advanced bladder cancer. N Engl J Med. 2003;349(9):859–866.
3. European Association of Urology Guidelines on Muscle‑Invasive and Metastatic Bladder Cancer. EAU Guidelines 2022 (latest available guidance on perioperative systemic therapy and management of MIBC).