Clinical Report: Developing and Validating a Model to Forecast 5-Year Survival in NSCLC

Overview

This study developed and validated a deep learning model to predict 5-year survival in non-small-cell lung cancer (NSCLC) using data from the Korean Central Cancer Registry. The model integrates various clinical features and aims to enhance prognostic accuracy for individual patients.

Background

Non-small-cell lung cancer (NSCLC) is a leading cause of cancer mortality globally, with prognosis heavily influenced by factors such as disease stage and molecular biomarkers. Accurate survival prediction is essential for optimizing treatment strategies and improving patient outcomes. The use of machine learning, particularly deep learning, offers a promising avenue for enhancing prognostic models in oncology.

Data Highlights

The study utilized data from the Korean Central Cancer Registry, which includes comprehensive clinical information from over 50 medical centers across South Korea. The dataset comprised patients diagnosed with NSCLC between 2014 and 2017.

Key Findings

• A deep learning model was developed to predict 5-year mortality in NSCLC patients.
• The model incorporated a grouped-input architecture to enhance interpretability and clinical relevance.
• Permutation-based importance was used to assess feature contributions transparently.
• The study addressed challenges related to data preprocessing and hyperparameter tuning to ensure reproducibility.
• Results indicated that the model could effectively utilize routine clinical variables for survival prediction.

Clinical Implications

The developed model provides a clinically applicable tool for predicting 5-year survival in NSCLC patients, which can aid in personalized treatment planning. By utilizing routinely collected clinical data, the model enhances the feasibility of integrating advanced predictive analytics into everyday clinical practice.

Conclusion

This study demonstrates the potential of deep learning models in improving survival predictions for NSCLC, emphasizing the importance of utilizing large, well-curated datasets for robust clinical applications.

Related Resources & Content

1. Frontiers in Oncology, 2026 -- Construction and verification of 5-year survival prediction model for post-op ESCC patients
2. European Radiology, 2023 -- A predictive model for assessing progression-free survival in advanced non-small cell lung cancer patients following image-guided microwave ablation combined with chemotherapy
3. The ASCO Post, 2025 -- External Validation Confirms Ability of AI Model to Stratify Recurrence Risk in Early-Stage Lung Cancer
4. The ASCO Post — External Validation Confirms Ability of AI Model to Stratify Recurrence Risk in Early-Stage Lung Cancer
5. IASLC Staging Project: Lung Cancer, Thymic Tumors, and Mesothelioma | IASLC
6. Ninth Edition of the Tumor, Node, and Metastasis Classification of Lung Cancer - PubMed
7. CAP Publishes Guideline for PD-L1 Testing of Patients with Lung Cancer
8. Early and locally advanced non-small-cell lung cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up - PubMed
9. Overall Survival with Neoadjuvant Nivolumab plus Chemotherapy in Lung Cancer | New England Journal of Medicine
10. KEYTRUDA® (pembrolizumab) Demonstrates Long-Term Survival Benefit in Certain Patients With Earlier or Advanced Stages of Non-Small Cell Lung Cancer (NSCLC) - Merck.com
11. Overall Survival with Osimertinib in Resected EGFR-Mutated NSCLC | New England Journal of Medicine
12. Survival with Osimertinib plus Chemotherapy in EGFR-Mutated Advanced NSCLC | New England Journal of Medicine
13. Amivantamab Plus Lazertinib vs Osimertinib in EGFR-Mutated Advanced NSCLC - The ASCO Post
14. Therapy for Stage IV Non–Small Cell Lung Cancer With Driver Alterations: ASCO Living Guideline, Version 2025.1 | Journal of Clinical Oncology
15. Lung Cancer Survival Rates | 5-Year Survival Rates for Lung Cancer | American Cancer Society