The BOND-003 trial, a phase 3 study of intravesical cretostimogene monotherapy, has demonstrated that urinary genomic disease burden, as measured by the UroAmp platform, can predict patient response in BCG-unresponsive non-muscle invasive bladder cancer (NMIBC). The analysis, presented at the 2024 International Bladder Cancer Network (IBCN) Annual Meeting, highlights the potential of using genomic profiling to improve patient selection and treatment strategies. Cretostimogene grenadenorepvec is an oncolytic adenovirus designed to selectively replicate in bladder cancer cells with alterations in the Retinoblastoma-E2F pathway and expresses the GM-CSF transgene, resulting in a potent oncolytic immunotherapy mechanism of action.
Genomic Disease Burden and Treatment Response
The study evaluated the association between cretostimogene treatment and urinary genomic disease burden in patients with BCG-unresponsive NMIBC. The BOND-003 trial included patients with CIS +/- Ta/T1, with all Ta/T1 disease having been resected prior to treatment. A mandatory biopsy was conducted at the 12-month assessment. Earlier results from BOND-003 presented at AUA 2024 showed complete response rates of 68% at 3 months and 76% at any time point, with 54% of re-induced patients converting to complete response. The 12-month cystectomy-free rate was 92%, and the 12-month progression-free survival rate was 97%. The treatment was well-tolerated, with no grade 3-5 treatment-related adverse events.
UroAmp Platform and Mutation Analysis
The UroAmp platform, developed by Convergent Genomics, uses next-generation sequencing (NGS) to quantify mutations and DNA alterations from urine cell-free DNA. It employs deep NGS of a 60-gene bladder cancer-specific panel and low-pass whole-genome sequencing to identify aneuploidy. The platform quantifies minimal residual disease by genomic disease burden, ranking the variant allele frequency of a sample relative to its percentile ranking within an established training set.
Baseline mutation profiles were available for 64 patients, with post-treatment profiles analyzed for 51 patients at 3 months and 35 patients at 6 months. The most frequent variants in BCG-unresponsive NMIBC were found in TERT, TP53, genes coding for chromatin-modifying enzymes, and proliferation pathways. A subset of aneuploidy-positive tumors, sharing a genetic signature with muscle-invasive bladder cancer, was also identified. RB1 mutations were relatively rare.
Clinical Outcomes and Predictive Value
Patients achieving a complete response at 3 months showed lower genomic disease burden, with a significant reduction in genomic disease burden observed in re-induced patients who achieved a complete response at 6 months. Changes in variant allele frequency paralleled changes in genomic disease burden. The molecular response to cretostimogene was characterized by a reduction in the prevalence of tumors with aneuploidy and alterations in ERBB2, TP53, and RB1, suggesting a reduced risk of disease progression.
Notably, the 3-month UroAmp signature predicted 12-month recurrence-free survival, with 80% recurrence-free survival for low-risk patients and 33% for high-risk patients (p = 0.012). Longitudinal changes in genomic disease burden were informative to clinical outcomes, with 1/13 low-risk patients (8%) and 3/6 high-risk patients (50%) at 6 months experiencing documented recurrence at 12 months (p = 0.041).
Implications for Future Trials
Dr. Colin Dinney from MD Anderson Cancer Center, concluded that cell-free urine DNA profiling with the UroAmp platform enables quantitative assessment of the clinical response to cretostimogene. The UroAmp platform stratified patients treated with cretostimogene as low risk or high risk and predicted durable 12-month recurrence-free survival. The reduction in variant allele frequency and genomic disease burden in patients salvaged by re-induction reinforces the value of re-induction. Longitudinal molecular disease burden assessment may support the stratification of control and intervention arms in future clinical trials.
