Nomogram Prediction Model for Overall Survival of Late-Stage Lung Cancer Patients Undergoing 125I Particle Implantation Brachytherapy

Abstract

Objective The purpose of this study was to explore the overall survival rate of late-stage lung cancer patients after receiving ¹²⁵I particle implantation brachytherapy and establish a corresponding Nomogram prediction model to provide reference for clinical decision-making. Methods We retrospectively analyzed 436 late-stage lung cancer patients who underwent ¹²⁵I radioactive particle brachytherapy in the Department of Nuclear Medicine of the General Hospital of Northern Theater Command of the Chinese People's Liberation Army from December 2013 to June 2019. The 436 patients were randomly divided into a training set and a validation set in a 7:3 ratio, with 305 patients in the training set and 131 patients in the validation set. Single-factor and multivariate Cox proportional hazards models were used to select independent factors affecting the prognosis of late-stage lung cancer patients. Based on these factors, a nomogram model was constructed to predict the overall survival at 1, 3, and 5 years after ¹²⁵I particle implantation brachytherapy for late-stage lung cancer, as well as the 1-year progression-free survival. The accuracy and predictive ability of the model were evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Results The 436 patients were included and randomly divided into the training and validation sets. The total survival time of the patients was 1113 ± 391.11 days, and the progression-free survival time was 200 ± 100.03 days. The results of the univariate and multivariate Cox proportional hazards model analyses of overall survival time (OS) showed that smoking, lung atelectasis, superior vena cava obstruction syndrome, and surgical time were significantly associated with the OS of the patients and were independent influencing factors. The results of the univariate and multivariate Cox proportional hazards model analyses of progression-free survival (PFS) showed that planning target volume, maximum dose, average dose, preoperative D90, V100 at 1 cm around the lesion, and surgical time were significantly associated with the PFS of the patients and were independent influencing factors. Based on these independent prognostic factors, nomogram models were constructed to predict the 1, 3, and 5-year overall survival and 1-year progression-free survival of late-stage lung cancer patients. The ROC curves showed that both the training and validation set nomogram prediction probabilities had good predictive ability. Decision curve results demonstrated good clinical applicability of the nomogram. The calibration curve showed a good correlation with the ideal curve, suggesting good accuracy of both models. Conclusion This study revealed the potential benefits of ¹²⁵I particle implantation brachytherapy for the overall survival rate of late-stage lung cancer patients and provided clinicians with a reliable tool to personalize the assessment of patient prognosis and formulate treatment plans.