Urine-based testing for patient selection and genomic characterization of patients with FGFR alteration-positive non–muscle-invasive bladder cancer (NMIBC) treated with TAR-210.

676 Background: TAR-210 is an intravesical drug delivery system that is designed to provide local, continuous release of erdafitinib. Its safety and efficacy are being evaluated in a first-in-human clinical study (NCT05316155) of patients with bladder cancer whose tumors harbor select FGFR alterations (alt). Toovercome tissue-based challenges in identifying susceptible FGFRalt to select patients for treatment with TAR-210, Janssen partnered with Predicine to use a proprietary urine cell-free DNA diagnostic assay (PredicineCARE). Validation of the urine assay to detect FGFRalt was previously demonstrated using urine samples collected in a highly controlled manner via a collaboration with Stratifyer Molecular Pathology GmbH, Cologne, Germany (Kim et al. ASCO GU 2023). However, evaluation of the urine assay for diagnostic screening in a real-world setting is currently lacking. Here, we report preliminary results of the urine assay for patient selection. We also report on the characterization of the urine-defined genomic landscape in screened patients. Methods: Enrollment was based on detection of prespecified FGFRalt from either tumor tissue obtained from previous biopsies (Qiagen Therascreen FGFR assay) or urine samples obtained prior to enrollment (PredicineCARE next-generation sequencing assay). Results: As of Jun 20, 2023, urine assay performance was compared to the tissue test from all screened patients with NMIBC (N=178). The proportions of samples that yielded results were 60% and 58% from tissue and urine, respectively, while the FGFRalt detection rates in the subsets that yielded results were 62% and 42%, respectively. FGFR3 S249C was the most frequent alteration detected in both tissue (48%) and urine (61%). For 36% of urine samples in which FGFRalt were detected, there was no corresponding tissue result. Of the disease-evaluable patients with high-risk (HR) NMIBC (N=11) or intermediate-risk (IR) NMIBC (N=15),18.2% and 33%, respectively, were enrolled based on urine assay alone due to insufficient tissue samples. A recurrence-free (RF) rate of 82% and a complete response (CR) rate of 87% were achieved at the first disease evaluation amongst patients with HR NMIBC and IR NMIBC, respectively. All patients (HR NMIBC, N=2, and IR NMIBC, N=5) enrolled by “urine only” were RF or achieved a CR. Based on the PredicineCARE panel, comprehensive genomic assessment of urine samples from all screened patients with NMIBC was performed. The prevalence of alterations detected was similar to that described in prior studies using tissue-based testing. Conclusions: Implementing a urine-based assay expands the molecular testing methods to identify additional patients that may respond to erdafitinib. Results justify further study. Clinical trial information: NCT05316155 .