CT-based deep learning model for improved disease-free survival prediction in clinical stage I lung cancer: a real-world multicenter study.
Authors
Affiliations (6)
Affiliations (6)
- Department of Radiation Oncology, Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
- Department of Oncological Surgery, Shanghai Lung Cancer Center, Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
- Department of Thoracic Radiotherapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, Hangzhou, Zhejiang, China.
- Department of Thoracic Radiotherapy, Zhejiang Cancer Hospital, Hangzhou Institute of Medicine (HIM), Chinese Academy of Sciences, Hangzhou, Zhejiang, China. [email protected].
- School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China. [email protected].
- Department of Radiation Oncology, Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China. [email protected].
Abstract
To develop a deep learning (DL) model for predicting disease-free survival (DFS) in clinical stage I lung cancer patients who underwent surgical resection using pre-treatment CT images, and further validate it in patients receiving stereotactic body radiation therapy (SBRT). A retrospective cohort of 2489 clinical stage I non-small cell lung cancer (NSCLC) patients treated with operation (2015-2017) was enrolled to develop a DL-based DFS prediction model. Tumor features were extracted from CT images using a three-dimensional convolutional neural network. External validation was performed on 248 clinical stage I patients receiving SBRT from two hospitals. A clinical model was constructed by multivariable Cox regression for comparison. Model performance was evaluated with Harrell's concordance index (C-index), which measures the model's ability to correctly rank survival times by comparing all possible pairs of subjects. In the surgical cohort, the DL model effectively predicted DFS with a C-index of 0.85 (95% CI: 0.80-0.89) in the internal testing set, significantly outperforming the clinical model (C-index: 0.76). Based on the DL model, 68 patients in the SBRT cohort identified as high-risk had significantly worse DFS compared to the low-risk group (p < 0.01, 5-year DFS rate: 34.7% vs 77.4%). The DL-score was demonstrated to be an independent predictor of DFS in both cohorts (p < 0.01). The CT-based DL model improved DFS prediction in clinical stage I lung cancer patients, identifying populations at high risk of recurrence and metastasis to guide clinical decision-making. Question The recurrence or metastasis rate of early-stage lung cancer remains high and varies among patients following radical treatments such as surgery or SBRT. Findings This CT-based DL model successfully predicted DFS and stratified varying disease risks in clinical stage I lung cancer patients undergoing surgery or SBRT. Clinical relevance The CT-based DL model is a reliable predictive tool for the prognosis of early-stage lung cancer. Its accurate risk stratification assists clinicians in identifying specific patients for personalized clinical decision making.