e17066 Background: Identifying high-risk early-stage EOC patients (p) is a major issue. Also, controversy about the role of neoadjuvant chemotherapy prior to surgery in advanced EOC is still ongoing. A high MLR is related to poorer outcomes in several cancers, and also in ovarian cancer according to limited retrospective series. We explored the clinical value of MLR in a cohort of EOC p. Methods: We included all EOC p diagnosed in our center, from January 2008 to April 2017, who had undergone radical treatment. Overall survival (OS) was assessed by Kaplan Meyer. Maximally selected rank statistics was used to determine the MLR cut-off value for the greatest OS discrimination. The hazard ratio (HR) of death of MLR was explored by multivariate Cox regression models, adjusted by age at diagnosis, stage I/II versus (vs) III/IV, high grade serous carcinomas (HGSC) vs other, primary surgery vs interval surgery after neoadjuvant chemotherapy (PS, IS), residual disease < 1 vs > 1cm, year of diagnosis, and hemoglobin, neutrophils, platelets and MLR at diagnosis. Results: 124 p were included. Median age at diagnosis was 58.7 years; 55.6% were HGSC;67.7% stage III/IV; 69.3% received PS; 14.3% had residual disease > 1cm. Median OS was 75 months. MLR was the laboratory parameter that more significantly impacted on OS in our model (p = 0.02). Optimal MLR cut-off was 0.38. The MLR > 0.38 group (MLR-hi) had a higher-risk of death (HR = 2.9; 95% IC: 1.3–6.5) with respect to those with MLR < = 0.38 (MLR-lo). Median OS were 100 and 35 months, respectively. In the MLR-lo group, median age at diagnosis was 56.9 years (vs 61.9 in the MLR-hi, p 0.047), 44.8% were HGSC (vs 61.2%, p 0.134), and 49.97% were stages I/II (vs 14.2%, p 0.001). MLR-hi p who had undergone PS or IS had similar OS (p = 0.63). In contrast, among MLR-lo p, PS showed significantly higher OS than IS (p < 0.0001). Conclusions: In our cohort, MLR was an independent prognostic factor in EOC. Its potential role to identify high-risk early-stage p and to predict worse outcomes after PS among advanced-stage p need to be explored in larger cohorts.