Response Evaluation Criteria in Tumors of the Bladder (RECIT-Bladder) for future window of opportunity trials as a platform to accelerate drug development.

571 Background: There are abundant potential permutations of novel or combination therapeutics that could be tested in non-muscle invasive bladder cancer (NMIBC), but the ability to study these therapies is limited. Window of opportunity trials (WOOT) may accelerate otherwise costly and slow drug evaluation by testing response to treatment during the time between the surveillance office cystoscopy that diagnoses a recurrent tumor and removal by transurethral resection (TURBT), but the accuracy of endoscopic tumor measurement is unknown. Through an iterative process modifying the RECIST criteria used for metastatic cancer response assessment; we developed RECIT-Bladder as a short-term endpoint for NMIBC WOOTs. Here, we report the feasibility and reproducibility of paired tumor measurement at office cystoscopy and TURBT in “untreated” patients as the development cohort for RECIT-Bladder. Methods: Bladder tumors were measured at office cystoscopy and again at formal TURBT in a standardized fashion compared to a calibrated object (flexible biopsy forceps). If multiple tumors were present, the two largest, measurable, representative tumors were measured. Patients were excluded if tumor location did not allow for accurate measurement (e.g., bladder neck) or if the lesion was not amenable to measurement (e.g., diffuse CIS or papillary carpeting). Results: We characterized 69 papillary tumors from 34 patients (4 patients with recurrent tumors included ≥2 times). 38% (13/34) had LGTa and the remainder had high grade (HG) bladder cancer (1 CIS, 15 HGTa, 2 HGT1, and 1 HGT2). 98.6% (68/69) were measurable at TURBT. There was a median 23 (16-37 IQR) days between cystoscopy and TURBT during which lesion size increased by a median 0.25 mm (0-0.1mm IQR, p=0.4). Measurements on cystoscopy and at TURBT were highly correlated (correlation coefficient 1.04, p<0.01, r2 = 0.88). There was no correlation between tumor growth and time between cystoscopy and TURBT (correlation coefficient 0.00, p=0.5, r2 = 0.00). MSK-IMPACT targeted exome sequencing was available in 62% of patients (21/34) of whom 71% (15/21) had an FGFR3 mutation (median growth 1 mm over median 23 days). Conclusions: Using a standardized protocol (RECIT-Bladder), tumor measurement at office cystoscopy and follow-up TURBT appears precise and reproducible. RECIT-Bladder may offer a practical short-term clinical trial endpoint for evaluating progression, stable disease, or response to therapy. If validated, RECIT-Bladder offers a pathway to rapidly discover novel therapeutic agents. We have begun a prospective multi-surgeon protocol to validate these results and evaluate potential pharmacodynamic endpoints. RECIT-Bladder is also currently being assessed in the context of an ongoing investigator-initiated WOOT testing oral erdafitinib in FGFR3 altered recurrent NMIBC (NCT04917809).