Development and validation of a nomogram based on preoperative variables for predicting recurrence-free survival in stage IA lung adenocarcinoma.

Abstract

This study aimed to establish a nomogram for predicting risk of recurrence and provide a model for decision-making between lobectomy and sublobar resection in patients with stage IA lung adenocarcinoma. Patients diagnosed with stage IA lung adenocarcinoma (LUAD) between December 2010 and October 2018 from Cancer Hospital Chinese Academy of Medical Sciences were included. Patients were randomly assigned to training and validation cohorts, accounting for 70% and 30% of the total cases, respectively. We collected laboratory variables before surgery. Univariate and multivariate analyses were performed in the training cohort to identify variables significantly associated with recurrence-free survival (RFS) which were subsequently used to construct a nomogram. Validation was conducted in both cohorts. A receiver operating characteristic curve was used to determine the optional cutoff values of the scores calculated from the nomogram. Patients were then divided into low- and high-risk groups. Survival was performed to determine if the nomogram could guide the operation method. A total of 543 patients were included in this study. Gender, albumin level, carcinoembryonic antigen level and cytokeratin-19-fragment level were included in the nomogram. In both cohorts, the nomogram stratified the patients into high- and low-risk groups in terms of RFS. In particular, there was a significant difference in RFS between lobectomy and sublobar resection in the high-risk group. Gender, albumin level, carcinoembryonic antigen level and cytokeratin-19-fragment level are valuable markers in predicting recurrence and can guide surgical practice in patients with stage IA LUAD.