BOND-003 Cohort C: Study Design and Patient Population

The phase 3 BOND-003 trial (NCT04452591) was designed as a single-arm, international study to evaluate the activity of intravesical cretostimogene grenadenorepvec in patients with high-risk BCG-unresponsive NMIBC, with Cohort C focusing on patients with carcinoma in situ (CIS), with or without concomitant papillary Ta or T1 disease. In the third segment of this series, Mark D. Tyson II, MD, MPH, a professor of urology at Mayo Clinic in Phoenix, Arizona, reviews the study's design rationale, including the patient selection criteria, treatment schedule, and primary end point definition—each of which reflects both the regulatory guidance from the FDA and the biologic realities of immunotherapy in this disease setting.

All patients enrolled in Cohort C were required to meet the FDA's established definition of BCG unresponsiveness. Patients received a 6-week induction course of cretostimogene followed by maintenance dosing, and the protocol prospectively incorporated an option for re-induction in patients with persistent disease at 3 months. Tyson explains that this re-induction provision was not a concession to early failure but rather a deliberate design feature grounded in the biology of immune-mediated therapy: Unlike cytotoxic agents, where a lack of early response is generally predictive of futility, immunotherapy may require repeated antigen exposure to shift from innate to adaptive immune activation. Data from BOND-003 confirmed the clinical value of this approach, with 50% of the 28 patients who underwent re-induction converting to a complete response.

The primary end point was centrally confirmed complete response rate at any time point—a definition that accounts for the possibility of delayed responses in patients who received re-induction. Tyson notes that a patient with persistent disease at 3 months who subsequently achieves a complete response after a second induction course is appropriately captured by this end point at 6 months rather than being classified as a nonresponder. This design aligns with FDA guidance for single-arm, open-label trials in a disease setting where randomization to cystectomy or placebo is not feasible, and it was constructed to avoid penalizing an immunotherapy for the biologic delay that is sometimes inherent to its mechanism.