Background: PROC describes a heterogeneous group of patients (pts) with a variable but generally poor prognosis and low response to chemotherapy. Predicting OS in PROC would help stratify pts in trials and guide treatment decisions. We developed a nomogram to predict OS in PROC. Methods: Data from two randomized phase III trials evaluating chemotherapy for PROC – CARTAXHY [Lortholary Ann Oncol 2012] and AURELIA [Pujade-Lauraine JCO 2014] (chemotherapy-alone arm) – were combined to form the training cohort (N = 331) for nomogram development. Baseline variables significantly associated with OS were identified using Cox regression analysis. Pts were assigned scores based on the weighted sum of the relative importance of each variable in the multivariate model. The nomogram was then validated in the chemotherapy plus bevacizumab arm of the AURELIA trial (N = 166). Nomogram performance was assessed by calculating the c statistic. A classification based on the nomogram’s score was generated to group pts according to prognosis. Results: Poor performance status, CA125 ≥ 100 U/mL, ascites, platinum-free interval < 3 months, primary platinum resistance, and largest tumor > 5cm were associated with shorter OS. The nomogram predicted OS with a c statistic of 0.69 (training) and 0.67 (validation). In the training cohort, the median OS in good (N = 93), intermediate (N = 162), and poor (N = 76) prognostic groups was 25.3, 15.6, and 6.9 months, respectively (P < 0.0001). In the validation cohort, median OS in good (N = 50), intermediate (N = 79), and poor (N = 37) prognostic groups was 26.7, 13.8, and 9.0 months, respectively (P < 0.0001). Conclusions: This nomogram combining six baseline factors accurately predicts OS in PROC pts treated with chemotherapy either alone or in combination with bevacizumab. It could help to select pts for treatment, counsel pts regarding prognosis, and stratify according to risk in trials. Clinical trial information: NCT00976911