Interpreting the 39% clinical complete response rate in SURE-02: why the assessment pathway matters

Necchi and colleagues report SURE-02, a single-arm phase 2 trial of neoadjuvant sacituzumab govitecan plus pembrolizumab, followed by radical cystectomy or redo transurethral resection of the bladder tumor (re-TURBT) and then postsurgical pembrolizumab, in patients with muscle-invasive bladder cancer who were ineligible for or declined cisplatin-based neoadjuvant chemotherapy (1). Among 49 treated patients, 19 achieved a clinical complete response (cCR), corresponding to an intention-to-treat rate of 39% [95% confidence interval (CI): 25–54%]. Grade 3 treatment-related adverse events occurred in 16%, with no grade 4 or 5 treatment-related events. The trial met its protocol-defined primary endpoint.

SURE-02 is notable not only for the regimen studied but also for incorporating patient preference into the trial pathway. Patients who declined cystectomy after systemic therapy could undergo re-TURBT after multidisciplinary discussion and continue pembrolizumab. This pragmatic design addresses a real clinical dilemma. At the same time, it links the primary endpoint to the route used to assess response. The 39% is correct. The key question is what that percentage represents. It should be read as a result of the treatment-and-assessment strategy, not as a response rate attributable to the drugs alone.

Two denominators, two questions

At enrollment, radical cystectomy was planned for all patients. After the first 11 patients had been enrolled, a protocol amendment allowed re-TURBT for patients who declined cystectomy (1). The protocol defined cCR as negative imaging and no viable residual tumor at re-TURBT among patients not undergoing radical cystectomy. Under this definition, only patients in the re-TURBT pathway could be classified as having cCR; patients who proceeded to cystectomy were not assessed by this cCR endpoint. Of the 49 treated patients, 29 underwent re-TURBT, 16 underwent cystectomy, and four did not undergo a local procedure because of disease progression. All 19 cCRs were therefore documented in the re-TURBT group.

The two denominators answer different questions. Nineteen of 49 patients (39%) is the intention-to-treat estimate and the appropriate primary efficacy result. Nineteen of 29 patients (66%) is the proportion among patients who underwent re-TURBT. Because this group was selected after treatment, partly according to imaging findings and patient preference, 66% is descriptive and should not be treated as a second efficacy estimate. Its value is explanatory: it shows that the ability to document cCR depended strongly on the post-treatment pathway. In practical terms, the reported 39% summarizes the combined performance of treatment, decision-making, and response assessment.

This distinction is consistent with the Milan consensus, which emphasized standardized cCR assessment and intention-to-treat time-to-event outcomes in bladder-sparing trials (2). SURE-02 highlights a related consideration: a cCR rate cannot be interpreted fully without knowing how many patients underwent all components needed to define cCR.

Why the order of assessment matters

The comparison with HCRN GU16-257 is informative because the sequence was different. That study reported cCR in 33 of 76 patients (43%) after gemcitabine, cisplatin, and nivolumab (3). In HCRN GU16-257, the treatment pathway placed protocol-directed clinical restaging before the choice between bladder preservation and cystectomy. In SURE-02, imaging findings and patient preference influenced the local-treatment pathway, and the pathological component of cCR was then established only within the re-TURBT pathway. The numerical similarity between 43% and 39% should therefore not be taken to mean that the two rates measure the same thing.

Other features also shape the SURE-02 estimate. Thirty-six of 49 patients (73%) had an incomplete initial transurethral resection, and both clinical tumor (T) stage and resection completeness were associated with cCR in univariable analyses (1). These associations do not establish causality, but they suggest that baseline disease and the initial resection may influence subsequent cCR classification. Imaging was assessed qualitatively by experienced radiologists at the study center, and an independent central review was not reported. The investigators also stated that refusal of cystectomy was driven primarily by imaging. Imaging therefore influenced both the assessment pathway and the endpoint itself. These features do not invalidate the result, but they limit direct comparison with trials that use a uniform restaging sequence.

The nodal component should also remain explicit. In SURE-02, cCR was defined as ypT0Nx with negative imaging rather than cystectomy-confirmed ypT0N0 (1). This uncertainty is inherent to a bladder-preserving strategy, not unique to SURE-02. Nevertheless, complete response in the bladder and pathological confirmation of nodal status are distinct clinical questions and should be reported separately when cCR is used to support omission of cystectomy.

Reading the early survival results

No distant metastases occurred in the re-TURBT cohort during a median follow-up of 14 months (1). This is encouraging, but it cannot distinguish whether the favorable outcome reflects maintenance pembrolizumab, selection of patients entering the re-TURBT pathway, or both.

The post hoc event-free survival curves were stratified by cCR status but measured from the start of systemic therapy, whereas cCR was determined several months later. To be classified as a responder, a patient first had to remain event-free until response assessment. The responder group therefore includes an early event-free period by construction, introducing guarantee-time bias (4). A landmark analysis beginning at restaging, as used in HCRN GU16-257, or a time-dependent model would provide a less biased prognostic comparison (3). As presented, the curves are most appropriately viewed as descriptive. Bladder-intact event-free survival also counts cystectomy itself as an event; therefore, the difference between the cCR and non-cCR groups reflects the local-treatment pathway as well as subsequent disease control.

A signal for further study, not a basis for ranking regimens

SURE-02 used a Simon MinMax two-stage design in which a cCR rate of 30% or lower would not support further study, 45% was the target rate, and at least 19 responses among 48 planned patients were required (1). The planned decision rule was met. At the same time, the observed 95% CI of 25–54% includes both 30% and 45%. These statements are not contradictory: the trial crossed its formal phase 2 threshold, while the activity estimate remains imprecise.

The result supports further evaluation of sacituzumab govitecan plus pembrolizumab within a bladder-sparing strategy that incorporates patient preference. It does not establish that cystectomy can be omitted safely, that maintenance pembrolizumab caused the favorable outcomes in the re-TURBT cohort, or that this regimen is better or worse than other perioperative regimens. Direct comparisons of cCR and pathological complete response rates can be misleading when the assessment sequence and the role of cystectomy differ across trials. Differences in eligibility, treatment, and endpoint definitions further weaken any ranking based on a single response percentage.

Trimodality therapy remains an established bladder-preserving option with mature long-term data in selected patients (5). New response-adapted strategies should ultimately be judged by durable local and distant disease control, the frequency and outcomes of salvage treatment, preservation of a functional bladder, treatment burden, and patient-reported outcomes, rather than by cCR alone.

What future trials should report

Future studies should define the assessment pathway as precisely as the response definition. Where feasible, protocol-directed restaging should be completed before the final local-treatment decision, using a consistent combination of cystoscopy, resection or biopsy, urine cytology, and imaging. The intended extent and actual completeness of the initial transurethral resection should be documented. Incomplete assessments and their reasons should be reported, and central review should be considered for key imaging and pathological components.

Patient preference should also be recorded prospectively because it is clinically important and may influence which local procedure and which response assessments are performed. Reports may present both the intention-to-treat cCR rate and the proportion among patients who completed the required assessment, provided that the latter is clearly labeled as descriptive. Bladder response and nodal status should be reported separately, and survival analyses by cCR should use landmark or time-dependent methods. Long-term intention-to-treat outcomes should remain the main reference for judging clinical value.

SURE-02 provides a clinically relevant and encouraging phase 2 signal. Its broader lesson is methodological: in response-adapted bladder-preservation trials, the assessment pathway determines who can be labeled a responder and therefore becomes part of the endpoint. The 39% is valid, but its clinical meaning depends on the pathway, denominator, and timing that produced it. Transparent reporting of these elements will be essential as perioperative bladder-preservation trials move forward.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Gland Surgery. The article did not undergo external peer review.

Funding: None.

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-2026-0415/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement:This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See:https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Necchi A, Maiorano BA, de Jong JJ, et al. Neoadjuvant sacituzumab govitecan plus pembrolizumab, followed by adjuvant pembrolizumab, in patients with muscle-invasive bladder cancer (SURE-02): a single-arm, phase 2 study. Lancet Oncol 2026;27:442-51. [Crossref] [PubMed]
2. Necchi A, Galsky MD, Dizman N, et al. End Points for the Next-Generation Bladder-Sparing Perioperative Trials for Patients With Muscle-Invasive Bladder Cancer. J Clin Oncol 2025;43:3536-44. [Crossref] [PubMed]
3. Galsky MD, Daneshmand S, Izadmehr S, et al. Gemcitabine and cisplatin plus nivolumab as organ-sparing treatment for muscle-invasive bladder cancer: a phase 2 trial. Nat Med 2023;29:2825-34. [Crossref] [PubMed]
4. Giobbie-Hurder A, Gelber RD, Regan MM. Challenges of guarantee-time bias. J Clin Oncol 2013;31:2963-9. [Crossref] [PubMed]
5. Giacalone NJ, Shipley WU, Clayman RH, et al. Long-term Outcomes After Bladder-preserving Tri-modality Therapy for Patients with Muscle-invasive Bladder Cancer: An Updated Analysis of the Massachusetts General Hospital Experience. Eur Urol 2017;71:952-60. [Crossref] [PubMed]