Abstract
BACKGROUND: The prognosis of patients with stage I non-small cell lung cancer (NSCLC) is often uncertain. This study aims to investigate a new prognostic tool to classify stage I NSCLC patients more accurately. METHODS: CD68 and CD163 macrophages were quantified by immunohistochemical analyses of the center of the tumor and the invasive margin of the 339 tumors, which were used to construct the macrophage immunoscore (MI). Cox proportional hazards models determined the effects of multiple factors on disease-free survival (DFS) and overall survival (OS). One nomogram was developed to predict DFS and OS of stage I patients. RESULTS: The multivariate Cox analysis identified MI (P<0.001), lymphocyte-to-monocyte ratio (LMR, P=0.006), and TNM stage (P=0.046) as independent prognostic factors for DFS. Compared with MI, TNM stage, and LMR alone, the nomogram improved the prediction accuracy of both DFS and OS in terms of the Harrell concordance index in the training cohort (0.812, P<0.001 for DFS; 0.810, P<0.001 for OS) and the external validation cohort (0.796, P<0.001 for DFS; 0.791, P<0.001 for OS). In addition, net reclassification (Nomogram vs. TNM-stage, P<0.001 for DFS and OS) and the integrated discrimination (Nomogram vs. TNM stage, P<0.001 for DFS and OS) also validated this improvement. CONCLUSIONS: The immunoscore-based prognostic nomogram could effectively predict DFS and OS of stage I NSCLC patients and enhance the predictive value of the TNM stage system.