Immune cells and signatures characterize tumor microenvironment and predict outcome in ovarian and endometrial cancers

We investigated immunogenomic signatures and correlated them with survival in ovarian cancer (OV) and endometrial cancer (EC). : We used whole transcriptome sequencing data from uterine serous cancer and The Cancer Genome Atlas data of OV and EC (n = 719). Gene expression score was calculated. Population abundance of immune cells were estimated. TGF-β, myeloid cells, IFN-γ, T cells, B cells and endothelial cells predicted overall survival. Whereas , neutrophils and endothelial cells predicted progression-free survival. In multivariate analyses, TGF-β, and monocytic cells predicted survival in high levels of microsatellite instability (MSI-H) EC whereas high IFN-γ trended toward improved survival in the MSI-S EC. High IFN-γ/low TGF-β and high IFN-γ/low signatures predicted longer overall survival. Low TGF-β/low signature predicted longer overall survival only in the MSI-H EC. Our data support the role of immune markers in predicting survival in OV/EC. We studied the association of immune markers and immune cells with survival outcome in ovarian and endometrial cancers. We performed gene expression analyses on tumor tissue of 719 patients. We generated signatures for the immune cells and markers and correlated them with survival outcome. We showed that specific immune cells and markers correlated strongly with survival even after we adjusted for other confounding factors. Further, we showed that these immune markers and immune cells have different correlation with survival when stratified in subtypes of endometrial cancer. Our data support the role of these immune markers in predicting survival outcome in ovarian and endometrial cancers.