What question did this study set out to answer?

To develop a model integrating 3D intratumoral heterogeneity scores with clinical features for predicting VPI in lung adenocarcinoma.

May 16, 2026Frontiers in Oncology0 citationsOpen Access

3D intratumoral heterogeneity-based quantitative score from chest CT for preoperative prediction of visceral pleural invasion in lung adenocarcinoma: a multicenter study

QOQunzhi OuyangCounty Hospital YWYanping WuWuhan University LZLiuhan ZhouXiangyang Central Hospital

Puntos clave

To develop a model integrating 3D intratumoral heterogeneity scores with clinical features for predicting VPI in lung adenocarcinoma.
Multicenter retrospective study with 1,301 lung adenocarcinoma patients from three medical centers.
Development cohort from Centers 1 and 2, validation cohort from Center 3; model created using machine learning based on CT imaging and clinical features.
Model performance evaluated using AUC from receiver operating characteristic analysis and SHAP for feature importance.
Stacking ensemble classifier achieved an AUC of 0.878 for predicting VPI in lung adenocarcinoma.
XGBoost showed competitive performance with several metrics; however, stacked model outperformed conventional radiomics (AUC = 0.841) and clinicoradiologic model (AUC = 0.776).
3D ITH score was identified as the most influential predictor, highlighting its significance in the model.

Resumen

Introduction Visceral pleural invasion (VPI) is a critical adverse prognostic factor in lung adenocarcinoma (LUAD). This study aimed to develop a stacking ensemble model that integrates three-dimensional intratumoral heterogeneity (3D ITH) scores with clinicoradiologic features to achieve accurate preoperative prediction of VPI in LUAD. Methods This multicenter retrospective study included 1,301 patients with LUAD from three medical centers. Patients from Centers 1 and 2 were assigned to the development cohort, whereas those from Center 3 constituted the fixed external validation cohort. To calculate the 3D ITH score, we integrated local radiomic descriptors with global pixel distribution characteristics derived from whole tumor CT volumes. Clinicoradiologic features and 3D ITH scores were then used to construct six base machine learning models and a final stacking ensemble classifier. Model performance was primarily assessed using receiver operating characteristic analysis and the area under the curve (AUC). SHapley Additive exPlanations (SHAP) were used to quantify feature contributions and to interpret the final model. Results The stacking ensemble classifier achieved the highest AUC for preoperative prediction of VPI in LUAD (AUC = 0.878), whereas XGBoost showed competitive performance on several threshold dependent metrics. SHAP analysis identified the 3D ITH score as the most influential predictor, followed by nodule size and CT density. Comparative experiments further showed that the stacking ensemble model outperformed the conventional radiomics signature (AUC = 0.841) and the clinicoradiologic comparative model (AUC = 0.776). Conclusion The model integrating 3D ITH scores with clinicoradiologic features showed strong discrimination for preoperative prediction of VPI in LUAD. This approach may serve as a useful adjunct for preoperative risk stratification and individualized treatment planning.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo