Oncology · Medical Literature
[Articles] Intravesical cretostimogene grenadenorepvec oncolytic immunotherapy in high-risk, BCG-unresponsive, non-muscle-invasive bladder cancer with carcinoma in situ (BOND-003 Cohort C): a single-arm, phase 3 trial
Abstract
Cretostimogene showed clinically meaningful anti-tumour response, with an adverse event profile characterised predominantly by low-grade, transient events. Cretostimogene shows promise as an innovative bladder-sparing treatment for high-risk, BCG-unresponsive, non-muscle-invasive bladder cancer with carcinoma in situ.
Clinical relevance
In adults with high-risk, BCG-unresponsive, non-muscle-invasive bladder cancer (NMIBC) with carcinoma in situ (CIS), does intravesical cretostimogene grenadenorepvec monotherapy achieve clinically meaningful complete response rates while preserving the bladder and avoiding progression to muscle-invasive disease? __SNAPSHOT__ {"studyDesign":"Single-arm, Phase 3 trial (BOND-003 Cohort C; NCT04452591)","setting":"41 academic and community urology sites across North America, Asia, and Australia (international, multi-center)","sampleSize":"n=112 enrolled; 110 evaluable patients reported across efficacy analyses","population":"Adults with pathologically confirmed high-risk, BCG-unresponsive NMIBC with CIS, with or without concomitant resected high-grade Ta or T1 papillary disease; heavily pretreated (median 12 prior BCG instillations [IQR 9–15]); median age 74 years; prior therapies including intravesical gemcitabine–docetaxel and systemic pembrolizumab were permitted and represented","intervention":"Intravesical cretostimogene grenadenorepvec (oncolytic adenovirus with GM-CSF transgene that selectively replicates in tumor cells with Rb-E2F pathway alterations), administered once weekly for 6 weeks (induction), followed by maintenance therapy; reinduction permitted for persistent disease at 3 months","comparator":"Not applicable (single-arm design; no concurrent control group)","primaryOutcome":"Complete response (CR) at any time, with centrally confirmed pathology (serial cystoscopy, urine cytology, axial imaging, and mandatory biopsy at month 12)","followUp":"Median follow-up of approximately 22–27+ months (ongoing; median DOR not yet reached at data cutoff); one patient disease-free beyond 51 months","fundingSource":"Industry (CG Oncology, Inc.; NASDAQ: CGON)"}
Key findings
The trial met its primary endpoint with statistical significance. A complete response at any time was achieved in 75.5% of patients (95% CI 66.3–83.2; n=83/110). Duration of response (DOR) at 12 months was 64.2% (95% CI 52.2–73.8) and at 24 months was 60.1% (95% CI 48.2–70.0). The median DOR was reported as at least 27.9 months and was ongoing at the time of data cutoff, with approximately 90% of 12-month responders maintaining response at 24 months. Cystectomy-free survival (bladder preservation) was approximately 89% at 12 months and 81% at 24 months. Freedom from progression to muscle-invasive bladder cancer (MIBC) was 96.6% at both 48 and 96 weeks; only 4 patients progressed to MIBC (3.4%). Of 18 patients who ultimately underwent radical cystectomy, 15 (83%) had non-muscle-invasive or pathologically confirmed T0 disease on final pathology. NNT is not calculable from this single-arm design. Regarding safety, treatment-related adverse events (TRAEs) occurred in 63% of patients and were predominantly Grade 1–2 lower urinary tract symptoms; no Grade 3 or higher TRAEs, no treatment-related discontinuations, and no treatment-related deaths were reported. The median time to TRAE resolution was 1 day (IQR 0–7). The most common TRAEs (≥10%) were bladder spasm, pollakiuria, micturition urgency, dysuria, and hematuria. Two Grade 2 serious TRAEs were reported: non-infective cystitis and urinary bladder hemorrhage. No systemic exposure to the viral vector was detected.
Practice implications
BOND-003 Cohort C establishes that cretostimogene grenadenorepvec monotherapy can produce high rates of durable complete response with a notably favorable safety profile in patients with high-risk, BCG-unresponsive NMIBC with CIS — a population facing a binary choice between additional intravesical salvage therapy and radical cystectomy. The CR rate of 75.5% and a median DOR of at least 27.9 months compare favorably with historical benchmarks for this indication (e.g., pembrolizumab CR ~41%, nadofaragene firadenovec CR ~53% at any time). The bladder preservation rate (81% at 24 months) and the extremely low MIBC progression rate (3.4%) are clinically important given the morbidity of cystectomy. Responses were consistent across subgroups, including patients who had previously received gemcitabine–docetaxel or pembrolizumab. However, several important caveats apply: (1) this is a single-arm trial without a randomized comparator, so relative efficacy against other approved agents cannot be formally quantified; (2) the full manuscript is published but this summary is based on publicly available press releases and secondary reporting — the complete dataset, subgroup analyses, and statistical methodology require direct review of the primary publication; (3) the trial enrolled a predominantly elderly, highly pretreated population at specialty centers, which may limit generalizability to community or primary care urology settings; (4) the trial is ongoing and the median DOR has not yet been fully reached, meaning long-term durability data are still maturing; (5) the agent remains investigational — FDA approval has not yet been granted (as of August 2026), and its safety and efficacy have not been formally established by any regulatory authority; (6) industry funding (CG Oncology) introduces potential for sponsorship bias in reporting and interpretation. A direct randomized comparison against other approved bladder-sparing agents is needed to definitively establish comparative superiority. __LIMITATIONS__ 1. Single-arm design without a randomized comparator precludes causal inference about superiority over other approved salvage therapies. 2. Industry-sponsored trial (CG Oncology) with attendant risk of reporting and framing bias; all summary data currently available via press release, and full methodological details require verification from the primary publication. 3. Response was assessed using composite criteria (cystoscopy, cytology, imaging, biopsy), and it is not yet clear how results compare under differing response definition frameworks used in other trials. 4. Median DOR is reported as 'at least 27.9 months and ongoing' — the true median has not been reached, meaning durability estimates are subject to change with further follow-up. 5. Population was predominantly elderly (median age 74 years) with very high BCG exposure burden (median 12 prior instillations), potentially limiting generalizability to less-pretreated or younger patients. 6. No active comparator arm; cross-trial comparisons with pembrolizumab and nadofaragene firadenovec are confounded by differences in patient populations, CR definitions, and follow-up duration. 7. Subgroup analyses are exploratory; the trial was not powered for individual subgroup comparisons. 8. Long-term overall survival and cancer-specific survival data are not reported. 9. The agent is not yet FDA-approved; any clinical use outside a trial or expanded access program would be off-label/investigational. 10. Sites were predominantly academic or specialty urology centers, which may not reflect the outcome achievable in broader community practice. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In this single-arm Phase 3 trial, intravesical cretostimogene grenadenorepvec achieved a complete response rate of 75.5% (95% CI 66.3–83.2) with a median duration of response of at least 27.9 months and no Grade 3+ treatment-related adverse events, representing a compelling efficacy and tolerability profile for high-risk, BCG-unresponsive NMIBC with CIS. Bladder preservation at 24 months was 81% and progression to muscle-invasive disease occurred in only 3.4% of patients. The agent remains investigational and pending FDA approval; while these results are among the most favorable seen in this indication, confirmatory randomized data against active comparators are needed before practice-changing conclusions can be drawn.