Utility of surgically documented minimal residual disease as a therapeutic target and early surrogate of frontline treatment outcomes in ovarian cancer

5549 Background: Cure rates for patients with advanced stage ovarian cancer have not changed over the last four decades likely due to inability to eradicate residual cancer cells after frontline therapy. Identifying clinically undetectable minimal residual disease (MRD) after frontline therapy by second look laparoscopy (SLL) has potential to a) uncover mechanisms of chemoresistance, b) identify MRD-specific therapeutic targets, and c) allow for earlier intervention with investigational therapies. The objective of this study is to describe the initial safety and feasibility of SLL and illustrate clinical pathologic correlates. Methods: This IRB-approved observational, single-center study includes patients with epithelial high-grade ovarian cancer with complete CA-125 and radiologic response after frontline treatment who underwent SLL from 4/2017-12/2022. Patients were subsequently offered standard of care or investigational maintenance therapy options based on homologous recombination deficiency (HRD) and MRD status. Summary statistics described the study population, chi-square tests compared categorical variables, and progression free survival (PFS) based on MRD status was estimated using Kaplan-Meier method and compared using log rank test. Results: 64 patients underwent SLL with a median of 8 biopsies (range 1-16). The majority of patients had stage III-IV disease (89%), and high-grade serous histology (89%). Approximately half (52%) of patients underwent neoadjuvant chemotherapy (NACT) with interval tumor reductive surgery (TRS). MRD was present in 28 patients (44%), and the presence of MRD was associated with a worse PFS (HR 3.1, 95% CI 1.5-6.3; 9.2 vs 24.6 months; P=0.002). MRD was more frequent among NACT recipients compared to those who underwent primary TRS (61% vs 26%, P=0.005), those with HRD negative tumors (63% vs 22%, P=0.003), and those with R1 versus R0 resections (71% vs 35%, P=0.012). Of patients with MRD, 20 patients (71%) received bevacizumab until progression as part of a clinical trial (NCT02884648), 3 (11%) received bevacizumab off trial, 4 (14%) received a PARP inhibitor (all HRD positive) and 1 received other therapy. There was one (1.6%) intraoperative complication requiring conversion to laparotomy. Conclusions: SLL is feasible, overall safe, and detected MRD in 44% of ovarian cancer patients who were in radiologic and CA-125 remission after frontline therapy. MRD rates were higher in patients who received NACT and HRD negative patients, and PFS was significantly longer in patients without residual disease. MRD rate by SLL may have utility as an early surrogate for efficacy in trials evaluating novel frontline investigational therapies. Knowledge of MRD status in this patient population enables interventional, observational, and translational investigations, as well as personalized prognostic counseling.