RATIONALE AND OBJECTIVES
To develop and validate a machine learning model combining multiparametric Magnetic Resonance Imaging (MRI) radiomics and clinical indicators for predicting lymphovascular space invasion (LVSI) in endometrial cancer (EC).
MATERIALS AND METHODS
This retrospective study enrolled EC patients who underwent preoperative MRI at two centers. Of 567 initially screened patients, 408 were included per inclusion/exclusion criteria, divided into training and validation sets by hospital. Clinical risk factors and intratumoral/peritumoral radiomic features were identified. Six machine learning algorithms were used to build models; the one with the highest validation Area Under the Curve (AUC) was optimal. Five additional models were developed, and performance was evaluated via AUC, calibration curves, and decision curve analysis (DCA).
RESULTS
Logistic regression identified CA125 and tumor diameter as independent LVSI risk factors. Six machine learning models were built with CA125, tumor diameter, Rad_Score1 and Rad_Score2; the NeuralNetwork performed best (validation AUC=0.803). The combined clinical-intratumoral-peritumoral radiomics model achieved the highest AUC(training AUC = 0.863, validation AUC = 0.803), with good calibration (Hosmer-Lemeshow test, P > 0.05) and favorable net clinical benefit (threshold 0.1-0.7). SHapley Additive exPlanations (SHAP) analysis enhanced model interpretability.
CONCLUSION
This study systematically compared the predictive performance of six machine learning models for LVSI in EC, identifying the NeuralNetwork model as the most optimal. The combined clinical-intratumoral-peritumoral radiomics model, alongside its SHAP visualization tool, enhanced the accuracy (ACC) of non-invasive preoperative LVSI prediction, demonstrating certain potential for clinical application.
In the cancer microenvironment, stromal and immune cells hold clinical importance. In lung squamous cell carcinoma (LSCC), the aim of this study was to find immune‐linked gene expression that has prognostic pertinence.
From the Cancer Genome Atlas, this study obtained the LSCC gene expression profile. The Estimation of Stromal and Immune cells in Malignant Tumor tissues using Expression data algorithm was applied to derive stromal and immune scores for the included cases.
Stromal and immune scores were not statistically associated with sex or tumor stage, whereas a borderline significant association was observed between the immune score and smoking status (
p
= 0.0614). Interestingly, the immune score showed an independent association with overall survival in LSCC as shown by multivariate analyses. There were 517 differentially expressed genes (DEGs) linked to immune scores in total, 42 of which were upregulated. The DEGs were commonly linked to inflammatory and immune responses, extracellular exosomes, and chemokine activities. Cox analyses revealed that 21 DEGs were significantly related to overall survival (OS) in LSCC. Through validation, four genes (AP1S2, CLEC10A, FBXO2, and IQGAP2) were found to be significant OS predictors in independent Gene Expression Omnibus (GEO) datasets. However, immunohistochemistry (IHC) validation showed inconsistent prognostic trends for AP1S2 compared with bioinformatic predictions, whereas CLEC10A exhibited no differential expression between tumor and adjacent tissues. FBXO2 and IQGAP2 maintained consistent associations with OS. Multivariate Cox regression confirmed several of these genes as independent prognostic factors for LSCC.
The study refines immune score‐associated DEGs and identifies independent prognostic factors in LSCC, and it provides tissue‐level evidence that may support future prognostic evaluation.
Weijia Jiang, Jing Xu, Wenwen Ma et al.· iNew Medicine· 0 citations