An External-Validated Prediction Model to Predict Lung Metastasis among Osteosarcoma: A Multicenter Analysis Based on...

Area of Science:

• Oncology
• Medical Informatics
• Machine Learning

Background:

• Lung metastasis significantly complicates osteosarcoma treatment.
• Accurate prediction of lung metastasis is crucial for effective therapeutic strategies.

Purpose of the Study:

• To develop and validate a clinical prediction model for lung metastasis risk in osteosarcoma patients.
• To utilize machine learning (ML) algorithms for enhanced predictive accuracy.

Main Methods:

• Retrospective data collection from SEER database and Chinese hospitals.
• Application of six ML algorithms (LR, GBM, XGBoost, RF, DT, MLP) for model development.
• Internal validation using 10-fold cross-validation and external validation with Chinese data.

Main Results:

• 17.73% of osteosarcoma patients developed lung metastasis.
• Sex, N stage, T stage, surgery, and bone metastasis were identified as independent risk factors.
• The extreme gradient boosting (XGBoost) model achieved the highest AUC (0.738 internal, 0.729 external).

Conclusions:

• The XGBoost model demonstrates superior prediction for lung metastasis in osteosarcoma.
• An online calculator based on the XGBoost model can aid clinicians in risk assessment.
• This tool facilitates individualized medical strategies for osteosarcoma patients.