Radiomics-based prognostic model for progesterone resistance in endometrial cancer: insights into extracellular matrix and collagen type III.

Background: Progestin resistance in fertility-preserving endometrial cancer (EC) patients remains a significant challenge, and radiomics has not yet been used to predict progestin therapy in these cases.

Results: In this study, we constructed a radiomics model to predict progestin resistant for fertility preservation patients. Distribution of clinical features have significant differences in high and low risk of progestin resistant subgroups, which we call predicting-sensitive (PS) vs predicting-resistant (PR) subgroups. The radiomics model achieved high predictive accuracy with an area under the ROC curve (AUC) of 0.841 in the training cohort. We further validate this model in validation and whole cohorts. As a result, the AUCs are 0.873 and 0.852, respectively. Key biological pathways identified include cellular response to external stimulus, collagen metabolic processes, and extracellular matrix (ECM) remodeling. PS was strongly linked to higher collagen type III content and changes in ECM stiffness, which were reflected in altered tumor microenvironment dynamics. Furthermore, fibroblast-epithelial interactions, cytoskeletal organization, and collagen binding were validated by atomic force microscope (AFM) and microfluid equipment with progestin resistance. These findings highlight the influence of ECM remodeling on treatment outcomes.

Conclusion: Our radiomics model provides a promising, non-invasive tool for predicting progestin resistance in EC. This approach paves the way for personalized therapeutic strategies for fertility-preserving patients.