Abstract Background: Although 80% of high grade serous ovarian cancer (HGSOC) patients are initially responsive to platinum (Pt)-based chemotherapy, the majority of them experience relapse with a progressive Pt-resistant disease. One of the most challenging issues that hampers the possibility to effectively treat relapsed disease is the lack of biological information about tumor recurrence. To date, the Pt-free interval (PFI), an empirical measure of the time lagging between the end of front-line chemotherapy and relapse, is the only and widely accepted parameter to predict patient outcome and sensitivity to Pt second-line chemotherapy. We have previously demonstrated that analysis of circulating tumor DNA (ctDNA) by low pass whole genome sequencing (sWGS) is a suitable tool to predict patient outcome and to intercept early signs of relapsed disease (Paracchini et al. PMID:33323403). In the present study, we quantified the percentage of tumor fraction (TF) in plasma of patients enrolled within the frame of a phase III clinical trial (MITO-16A/MaNGO-OV2A-EUDRACT number: NCT01706120) to demonstrate, at baseline, the prognostic value of untargeted ctDNA analysis. Methods: 172 eligible patients enrolled with the frame of the MITO-16A/MaNGO-OV2A clinical trial, for whom plasma sample was collected at baseline before chemotherapy, were selected for the study. For each plasma sample, circulating-free DNA (cfDNA) was purified and whole genome libraries (HyperPlus, Roche) sequenced (mean coverage 0.5X). ichorCN algorithm was used to infer the percentage of TF, as previously published (Paracchini et al. PMID:33323403). Results: In 168 out of 172 plasma samples analyzed, the TF was detectable, and it ranged from 2.4% to 47%. Cox-model univariate analysis with patients’ overall survival (OS) and Progression-Free Survival (PFS), indicate association of TF levels with both OS and PFS (p<0.001). The prognostic relevance of TF was also confirmed in multivariate analysis, where, considering CA-125, residual tumor, stage of disease, histology and age as co-variables, TF maintained its statistical significance in PFS (HR = 2.1, CI 95%, p-value < 0.003). Conclusion: Retrospective analysis of baseline plasma samples collected within the frame of the MITO-16A/MaNGO-OVtrialnical trial, demonstrated that the percentage of TF is an independent prognostic marker of relapse. This finding confirms the importance of liquid biopsy analysis based on untargeted sWGS tool as an innovative approach to predict response to front line therapy and patient’s outcome. Citation Format: Sergio Marchini, Lara Paracchini, Laura Arenare, Giovanni Scambia, Carmela Pisano, Riccardo Zadro, Laura Mannarino, Annamaria Ferrero, Paolo Chiodini, Daniela Califano, Piergiacomo Di Gennaro, Sandro Pignata, Maurizio D'Incalci. Untargeted analysis of plasma tumor fraction is an independent prognostic marker of response to front line chemotherapy in high-grade serous epithelial ovarian cancer: the MITO-16A/MANGO-OV2A clinical trial experience [abstract]. In: Proceedings of the AACR Special Conference on Ovarian Cancer; 2023 Oct 5-7; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(5 Suppl_2):Abstract nr A062.