Abstract
The ovarian cancer (OC) is the deadliest gynecologic cancer and it is highly heterogeneous with common cases being diagnosed at an advanced stage. The mouse models still cannot be ignored as they have been used to study the pathogenesis of ovarian cancer, yet, their translational potential has been hampered especially in their morphological characterization. This review describes the spectrum of available models, which are genetically engineered models, syngeneic systems and patient-derived xenografts with focus on their applicability in the recapitulation of human disease. We emphasize the synergy between morphological phenotyping and the use of sophisticated imaging modalities and artificial intelligence (AI) as one of the prospects of becoming a multi-dimensional model to increase its usefulness. Histopathological and genetic measurements give information on tumor development and heterogeneity, and high-resolution imaging allows non-invasive surveillance. Moreover, artificial intelligence technologies can help to detect the tumor more accurately, characterize the microenvironment of the tumor, and develop individual treatment plans. Taken together, this review highlights the opportunities of integrating experimental models with new computational approaches to enhance early disease-detection, tumor-classification, and clinical translation of preclinical discoveries into clinical applications of ovarian cancer.