There are no reliable clinical bio-markers of survival prognosis, patients risk stratification and treatment prediction for epithelial ovarian cancers(EOC). The most common type of the human EOC is a high grade serous EOC. This cancer is characterized with one of the lowest survival rates compared to other cancers. The present invention relates to an method for a prognosis of survival of a subject diagnosed with EOC, the method comprising determining in a sample of the subject gene expression level of at least one gene in the list of Evi1 pathway genes and/or copy number of at least one gene in the MECOM locus wherein the level against at least one expression threshold value will define the risk group of the subject and/or a risk of the disease progression after surgery treatment, and/or an effectiveness of post-surgery chemotherapy. The quantification method of Evi1/MECOM locus regulatory pathway provides a set of multigene prognostic signatures representing EVI1 pathway modules, which collectively provided a framwork of high-confidence, sensitive and specific prognosis assay(s) of EOC and stratification method for the EOC patient stratification according to disease relapse.
