Abstract
BACKGROUND: Some patients with non-small cell lung cancer (NSCLC) experience early relapse within 2 years post-surgery. Screening patients who are prone to recurrence is crucial. This study aimed to determine factors influencing early recurrence within 2 years of surgery for stage IA-IIA NSCLC and to establish a prediction model. METHODS: We retrospectively analyzed the hematological indices and imaging indicators of patients with stage IA-IIA NSCLC who underwent surgery at our hospital, and relevant clinical data were obtained through long-term follow-up from September 2019 to September 2020. Least absolute shrinkage and selection operator (LASSO) regression and univariate and multivariate Cox regression analyses were used to identify high-risk factors influencing postoperative recurrence, establish a predictive model, and construct a nomogram associated with recurrence-free survival. RESULTS: Among 186 patients (90 male and 96 female), 29 (15.6%) experienced recurrence or metastasis during the follow-up period. Univariate analysis identified several significant factors, including tumor size, direct bilirubin, indirect bilirubin, albumin, globulin, serum creatinine, platelet-lymphocyte ratio, lymphocyte-monocyte ratio, prognostic nutrition index, albumin-alkaline phosphatase ratio, marginal lobulation, air bronchogram sign, pathological type of squamous cell carcinoma, tumor stage IB, and solid nodules. LASSO regression was used to further select variables and construct a multivariate Cox model showing globulin levels, air bronchogram signs, and solid nodules as independent prognostic factors for early recurrence within 2 years in patients with stage IA-IIA NSCLC. The Cox model stratified patients into high- and low-risk groups and was validated by Kaplan-Meier survival analysis, which demonstrated that high-risk patients had a significantly lower survival rate than low-risk patients, demonstrating the robust discriminative power of the predictive model. CONCLUSION: Globulin content, air bronchogram signs, and solid nodules were independent prognostic factors for early recurrence within 2 years in patients with stage IA-IIA NSCLC. The proposed model, developed based on the above factors and the albumin-alkaline phosphatase ratio, can effectively predict recurrence risk, potentially aiding clinicians in quantifying prognostic risk, making personalized survival assessments, and devising the most effective treatment plans.