Association of tumor-informed ctDNA-based molecular residual disease (MRD) with clinical outcomes for upper tract urothelial cancer (UTUC).

4565 Background: Upper tract urothelial carcinoma (UTUC) accounts for 5–10% of all urothelial carcinomas. Given its aggressive phenotype compared to primary bladder tumors and the lack of efficient biomarkers to guide treatment decisions, disease management remains challenging. Herein, we evaluate the prognostic value of ctDNA-based molecular residual disease (MRD) detection in UTUC. Methods: We conducted a retrospective analysis of real-world data from commercial ctDNA testing in a multicenter cohort of patients with UTUC using a personalized, tumor-informed, mPCR-NGS ctDNA assay (SignateraTM, Natera, Inc.). ctDNA was evaluated during the 1) MRD window (2-16 weeks post-surgery, before adjuvant therapy [AT]) and 2) surveillance windows ( > 16 weeks post-surgery if no AT was given or after AT completion). The correlation between ctDNA status and patient outcomes (recurrence-free survival [RFS]) was assessed using Cox regression analysis. RFS was defined as the interval from surgery to the date of radiographic recurrence or any evidence of residual/persistent disease after the completion of surgery or AT. Results: A total of 349 plasma samples collected from 45 patients with stages I-IV UTUC between 4/20/2021-12/16/2024 were available for analysis. Neoadjuvant therapy (NAT) was administered for 9% (4/45) of the patients, while 51% (23/45) of patients received AT post-surgery/-diagnosis [chemotherapy 57% (13/23), immunotherapy 39% (9/23), chemoimmunotherapy: 4% (1/23)] and 22% (10/45) received treatment for metastatic disease [EV-immunotherapy or immunotherapy: 40% (4/10), chemotherapy: 40% (4/10), chemoimmunotherapy: 20% (2/10)]. With a median follow-up of 17 (range: 3-71) months, the ctDNA detection rate within the MRD (N = 24) and surveillance (N = 32) windows was 70.8% and 68.8%, respectively. Patients with ctDNA-positivity within the MRD and surveillance windows showed a significantly inferior RFS compared to ctDNA-negative patients (MRD: HR = 13.4, P = 0.012 and surveillance: HR = 14.46, P = 0.01). Notably, ctDNA-positive patients showed a 12-month RFS of 32.1% (95% CI: 11.83–54.6%) and 45.5% (95% CI: 24.4–64.3%), respectively for MRD (N = 17) and surveillance (N = 22) windows, compared to 100% 12-month RFS for ctDNA-negative patients (MRD, N = 7; surveillance, N = 10). Multivariate regression analysis during surveillance revealed ctDNA-positivity as the only factor significantly associated with poor RFS (HR = 17.6, P = 0.011) when adjusted for clinical stage, NAT, and AT. Conclusions: This is the first study utilizing longitudinal, tumor-informed ctDNA testing to assess patient outcomes and disease status in UTUC. Our hypothesis-generating results suggest that ctDNA-based MRD detection via tumor-informed ctDNA testing is prognostic of patient outcomes post-surgery in UTUC and warrants further investigation in larger prospective cohorts.