Predictive Model for Recurrence in Postoperative Stage IA1-IIIA NSCLC Patients

Overview

This study developed a multi-dimensional nomogram integrating clinical, pathological, radiological, and genetic data to predict postoperative recurrence in stage ⅠA1-ⅢA NSCLC patients. Using a 56-gene LungCore panel alongside imaging and clinical features, the model aims to improve individualized risk stratification beyond traditional TNM staging.

Background

Lung cancer is the leading cause of cancer incidence and mortality globally, with NSCLC comprising over 85% of cases. Surgical resection is the standard treatment for stage I-III NSCLC, but recurrence rates vary widely, with 5-year recurrence-free survival ranging from 34% to 82%. While TNM staging guides prognosis, it lacks sufficient granularity for personalized risk prediction. Integrating clinical, radiological, and genetic factors may enhance prognostic accuracy and guide postoperative management.

Data Highlights

A total of 911 patients with stage ⅠA1-ⅢA NSCLC who underwent radical surgery and had 56-gene LungCore panel results were retrospectively analyzed. Patients were randomly split into training (70%) and validation (30%) cohorts. The study had a 14% loss to follow-up rate, with sensitivity analysis showing no significant baseline differences between lost and retained patients. Missing data accounted for less than 2% and were imputed. Follow-up extended to at least five years, with recurrence defined by new lesions on PET-CT or pathology.

Key Findings

• The developed nomogram incorporated demographic, clinical, pathological, radiological, and genetic mutation data from the 56-gene LungCore panel.
• LASSO regression with 10-fold cross-validation identified key predictive variables for postoperative recurrence.
• The model demonstrated improved risk stratification compared to TNM staging alone.
• Integration of radiological imaging features and targeted genetic profiling enhanced individualized prognostic prediction.
• The 56-gene panel provided a cost-effective genetic assessment correlated with tumor mutational burden relevant for recurrence risk.

Clinical Implications

This multi-dimensional predictive model can assist clinicians in identifying NSCLC patients at higher risk of postoperative recurrence, facilitating tailored surveillance and adjuvant treatment strategies. Incorporating genetic profiling with routine clinical and imaging data offers a practical approach to personalized prognosis without the cost and complexity of whole-genome sequencing. Such models may optimize resource allocation and improve patient outcomes through individualized care.

Conclusion

The study successfully developed and validated a comprehensive nomogram integrating clinical, radiological, and genetic factors to predict postoperative recurrence in stage ⅠA1-ⅢA NSCLC patients. This approach enhances personalized risk assessment beyond conventional staging systems.

References

1. Global and Chinese Lung Cancer Statistics and Burden
2. TNM Staging and Prognostic Limitations in NSCLC
3. Use of 56-gene LungCore Panel for Genetic Profiling in NSCLC
4. TRIPOD Reporting Standards for Prognostic Models