Abstract 1061: Tumor-informed circulating tumor DNA identifies high-grade serous ovarian cancer patients at highest risk for recurrence despite optimal first-line treatment with primary macroscopic complete resection

Abstract Background: Complete macroscopic tumor resection estimation as visualized by the surgeon is the foremost prognostic factor in high-grade serous ovarian cancer (HGSOC). Objective measures of postoperative molecular residual disease (MRD) are lacking, including serologic and radiologic evaluation. Tumor-informed circulating tumor DNA (ctDNA) has shown promise as a new approach to identify patients at higher risk for relapse. Methods: This prospective bicentric study included advanced HGSOC patients (pts) undergoing primary debulking surgery (PDS), six cycles of chemotherapy and maintenance therapy (07/2021 - 09/2024). Whole-genome sequencing of tumor tissue identified structural variants, used to design personalized multiplex dPCR assays to detect MRD. Plasma samples were collected: perioperatively (preop, postop day[d]2 and d10), during chemotherapy (C1, C3, C6) and every 3 months (mo) until progression or 24 mo of follow up (FU). CA-125 was analyzed accordingly (cutoff 35 kU/L). Statistics included chi-square and log-rank tests, multivariate Cox model and Kaplan-Meier estimates for progression-free survival (PFS). Results: 33/84 pts (39%) experienced relapse with a median PFS of 12.0 mo during a median FU time of 15.4 mo. ctDNA was detected in 77/82 pts (94%) at baseline and in 63/74 pts (85%) postoperatively. At postop d10, pts with complete tumor resection had significantly lower ctDNA levels than pts with postop residual disease (p=0.0018). Median ctDNA level decreased by 94% between preop and postop d10 in pts with complete tumor resection (p=0.0247), but not in pts with postop tumor residuals (p=0.891). Within the pt subgroup with complete tumor resection, ctDNA clearance at C1 (20%) and C6 (70%) was associated with significantly lower recurrence risk compared to persistent ctDNA levels (C1: 80%, C6: 30%) with a stronger association at C6 (C1: HR=3.39, p=0.0470, C6: HR= 52.21, p<0.0001). Median time to recurrence in the C6 ctDNA positive group was 10.7 mo compared to 21.3 mo in the ctDNA negative group. This effect was not observed at postop d10 (p=0.273) or for pts with residual tumor. ctDNA outperformed CA-125 for prognosis, as CA-125 levels at C1 or C6 failed to predict recurrence (C1: p=0.189; C6: p=0.165). In a multivariate analysis, ctDNA persistence at C6 (HR=14.58, p<0.0001) and postoperative residual tumor were confirmed as independent prognostic markers. Conclusion: Tumor-informed ctDNA analysis can further accurately stratify HGSOC pts at high risk of recurrence despite primary surgery with complete macroscopic resection. ctDNA assessment at C6 may provide an opportunity window for risk-stratified treatment adjustments directly after chemotherapy, enabling new personalized maintenance strategies. Citation Format: Magdalena Postl, Christina Victoria Tauber, Valentina Glueck, Mira Maria Gliga, Nuria Segui, Karen Howarth, Miguel Alcaide, Lucia Oton, Gerda Hofstetter, Isabelle Maurer, Alexander Burges, Sven Mahner, Mirjana Kessler, Melina Danisch, Stephan Polterauer, Nicole Concin, Christoph Grimm, Fabian Trillsch. Tumor-informed circulating tumor DNA identifies high-grade serous ovarian cancer patients at highest risk for recurrence despite optimal first-line treatment with primary macroscopic complete resection [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 1061.