Association of ctDNA status with upstaging, pathologic outcomes, and genomic alterations in high-risk NMIBC.

Abstract

831 Background: The role of circulating tumor DNA (ctDNA) in non–muscle-invasive bladder cancer (NMIBC) remains incompletely defined. In high-risk and Bacillus Calmette–Guérin (BCG)–exposed disease, ctDNA detection and genomic profiling may provide insight into disease biology and guide management. We evaluated ctDNA positivity rates and their association with genomic alterations, clinical upstaging, and pathologic outcomes in patients with high-risk NMIBC. Methods: We retrospectively reviewed 44 patients with histologically confirmed high-risk NMIBC (per AUA criteria) treated at a single institution between 2022-2025. Among BCG-exposed patients, BCG status was categorized as unresponsive per FDA definitions. ctDNA testing was performed using a personalized, tumor-informed assay (Signatera, Natera) at baseline and every 3 months; positivity was defined as detection of ≥2 tumor-specific variants. Comprehensive genomic profiling (CGP) with the Altera assay (Natera) was incorporated when available. Clinical upstaging was defined as progression to ≥cT2 or cN+ disease, and pathologic outcomes after radical cystectomy were evaluated by final stage and nodal status. Results: A total of 44 patients with HR-NMIBC were included (median age 69 years, IQR 59–75); 85% were male and 70% had pure urothelial carcinoma. Median follow-up was 8.2 months (IQR 4.7–11.7). BCG-naïve and BCG-exposed cohorts comprised 55% and 45% of patients, respectively, with 65% of the latter classified as BCG-unresponsive. At presentation, 93% had T1, 5% Tis, and 2% Ta disease; concomitant CIS was observed in 19%. ctDNA positivity was detected in 30% (6/24) of BCG-naïve and 40% (8/20) of BCG-exposed patients. Following clinical and radiologic assessment, upstaging occurred in 36% (5/14) of ctDNA-positive versus 10% (3/30) of ctDNA-negative patients. Overall, 36% (16/44) underwent radical cystectomy; 44% (7/16) were ctDNA-positive, of whom 86% (6/7) had muscle-invasive or locally advanced disease (≥pT2b or pN+), while one had pT1pN0. In contrast, all ctDNA-negative patients (9/9) had negative nodal status (4 pT0N0, 4 < pT1N0, and 1 pT2aN0). Among the 23 patients with available CGP, 8 were concurrently ctDNA-positive. TP53 mutations were more common in ctDNA-positive versus ctDNA-negative patients (88% vs 33%), whereas FGFR3 alterations were less frequent (0% vs 40%). Conclusions: ctDNA positivity in HR-NMIBC correlated with higher rates of clinical upstaging and adverse pathologic findings, including muscle-invasive and nodal disease at cystectomy, whereas ctDNA negativity was largely confined to non-invasive pathology. TP53 enrichment among ctDNA-positive patients suggests a biologically distinct and potentially more aggressive subset. These findings support the potential role of ctDNA as a biomarker for disease biology and risk stratification in HR-NMIBC.