CT-based radiomics in ovarian and endometrial cancers: toward predictive biomarkers for targeted therapy

Abstract Ovarian and endometrial malignancies provide significant therapeutic hurdles owing to their molecular variability and inconsistent treatment responses. Identifying non-invasive predictive biomarkers is essential for tailoring treatment and improving patient care. Radiomics has emerged as a promising methodology, facilitating the systematic extraction of quantitative features from computed tomography (CT) images to assess tumor heterogeneity, genetic changes, and microenvironmental properties that may affect prognosis and therapy response. Nonetheless, the evidence substantiating radiomics as a validated prognostic biomarker for targeted therapy is still limited. Most contemporary models should be regarded as prognostic or hypothesis-generating tools, rather than as instruments intended to guide therapeutic decision-making. Recent studies integrating CT-derived radiomic characteristics with clinical and genetic data have demonstrated improved predictive accuracy, especially in stratifying patients likely to benefit from targeted therapies, including poly-ADP-ribose polymerase (PARP) inhibitors and immunotherapy. In gynecologic oncology, the application of diverse methodologies indicates that radiomics holds promise for enhancing existing biomarkers and propelling precision oncology forward. The primary obstacles to widespread implementation encompass variations in feature extraction and analysis techniques, the utilization of different imaging modalities, and the absence of comprehensive validation across diverse patient populations. To enable clinical adoption, these issues must be addressed through extensive, multi-institutional studies and the establishment of standardized methodologies. This review examines the current status of CT-based radiomics in ovarian and endometrial cancers, evaluating its role as a prognostic biomarker and its potential to guide customized therapy approaches. Further investigation to integrate radiomics into treatment protocols, with the aim of improving prognoses for women diagnosed with gynecologic cancers is needed.