Jiashan Ding

Background: High-grade serous ovarian cancer (HGSOC) remains one of the most challenging gynecological malignancies, with over 70% of ovarian cancer patients ultimately experiencing disease progression. The current prognostic tools for progression-free survival (PFS) in HGSOC patients have limitations. This study aims to develop an explainable machine learning (ML) model for predicting PFS in HGSOC patients. Methods: Nine ML algorithms for PFS prediction were developed using a prospective cohort of 310 HGSOC patients consecutively enrolled from a large Chinese tertiary hospital between January 2017 and December 2020. The optimal model was internally validated using the 1000 bootstrap method. The SHapley Additive exPlanations (SHAP) method was employed to interpret the model in terms of feature importance and feature effects. The final model, constructed with the optimal feature subset, was deployed as an interactive web-based Shiny app. Results: The random survival forest (RSF) model demonstrated superior predictive performance compared to other ML models, the RFS model constructed with an optimal feature subset in the optimal imputed dataset achieved a superior 1000 bootstrap C-index of 0.755 (95% CI: 0.750–0.780) and a Brier score of 0.183 (95% CI: 0.175–0.190). SHAP analysis identified tumor residual, HE4, FIGO stage, T stage, CA125, age, ascites volume, platelet counts, and BMI as the top nine contributing factors. It also revealed potential nonlinear relationships and important thresholds between HE4, CA125, age, ascites volume, platelet counts, the body mass index, and PFS risk. Additionally, interaction effects were found between tumor residual and age, HE4, and CA125. Finally, an interactive web-based Shiny app for the model was developed and accessible at https://rsfmodels.shinyapps.io/ocRSF/. Conclusion: An explainable ML model for PFS prediction in HGSOC patients was developed with superior results. The publicly accessible web tool based on the optimized model facilitates its utility in clinical settings, potentially improving individualized patient management and treatment decision-making in HGSOC.

Background: Endometrial cancer (EC) as one of the most common gynecologic malignancies is increasing in incidence during the past 10 years. Genome-Wide Association Studies (GWAS) extended to metabolic and protein phenotypes inspired us to employ multiomics methods to analyze the causal relationships of plasma metabolites and proteins with EC to advance our understanding of EC biology and pave the way for more targeted approaches to its diagnosis and treatment by comparing the molecular profiles of different EC subtypes. Methods: Two-sample mendelian randomization (MR) was performed to investigate the effects of plasma metabolites and proteins on risks of different subtypes of EC (endometrioid and nonendometrioid). Pathway analysis, transcriptomic analysis, and network analysis were further employed to illustrate gene-protein-metabolites interactions underlying the pathogenesis of distinct EC histological types. Results: The authors identified 66 causal relationships between plasma metabolites and endometrioid EC, and 132 causal relationships between plasma proteins and endometrioid EC. Additionally, 40 causal relationships between plasma metabolites and nonendometrioid EC, and 125 causal relationships between plasma proteins and nonendometrioid EC were observed. Substantial differences were observed between endometrioid and nonendometrioid histological types of EC at both the metabolite and protein levels. The authors identified seven overlapping proteins (RGMA, NRXN2, EVA1C, SLC14A1, SLC6A14, SCUBE1, FGF8) in endometrioid subtype and six overlapping proteins (IL32, GRB7, L1CAM, CCL25, GGT2, PSG5) in nonendometrioid subtype and conducted network analysis of above proteins and metabolites to identify coregulated nodes. Conclusions: Our findings observed substantial differences between endometrioid and nonendometrioid EC at the metabolite and protein levels, providing novel insights into gene-protein-metabolites interactions that could influence future EC treatments.