Lung cancer is the leading cause of cancer-related death worldwide. Even early-stage patients might encounter disease recurrence with relative high risk. Effective postoperative therapy is based on an accurate assessment of treatment failure after surgery. The aim of this study is to construct a disease-free survival (DFS) prediction model and stratify patients into different risk score groups.A total of 356 pathological stage I patients (7th American Joint Committee on Cancer) who underwent lung resection from January 2005 through June 2011 were retrospectively reviewed. Of these patients, 63 patients were eliminated for this study. A total of 293 p-stage I patients were included for further univariate and multivariate analysis. Clinical, surgical, and pathological factors associated with high risk of recurrence were analyzed, including age, gender, smoking status, additional primary malignancy (APM), operation method, histology, visceral pleural invasion, angiolymphatic invasion, tumor necrosis, and tumor size.Of the 293 p-stage I non-small cell lung cancer (NSCLC) patients examined, 143 were female and 150 were male, with a mean age of 62.8-years old (range: 25–83-years old). The 5-year DFS and overall survival rates after surgery were 58.9% and 75.3%, respectively. On multivariate analysis, current smoker (hazards ratio [HR]: 1.63), APM (HR: 1.86), tumor size (HR: 1.54, 2.03), nonanatomic resections (HR: 1.81), adenocarcinoma histology (HR: 2.07), visceral pleural invasion (HR: 1.54), and angiolymphatic invasion (HR: 1.53) were found to be associated with a higher risk of tumor recurrence. The final model showed a fair discrimination ability (C-statistic = 0.68). According to the difference risk group, we found patients with intermediate or higher risk group had a higher distal relapse tendency as compared with low risk group (P = 0.016, odds ratio: 3.31, 95% confidence interval: 1.21–9.03).Greater than 30% of disease recurrences occurred after surgery for stage I NSCLC patients. That is why we try to establish an effective DFS predicting model based on clinical, pathological, and surgical covariates. However, our initial results still need to be validated and refined into greater population for better application in clinical use.
展开▼
