Predicting Time to First Rejection Episode in Lung Transplant Patients Using a Comprehensive Multi-Indicator Model.

Abstract

Background: Rejection hinders long-term survival in lung transplantation, and no widely accepted biomarkers exist to predict rejection risk. This study aimed to develop and validate a prognostic model using laboratory data to predict the time to first rejection episode in lung transplant recipients.

Methods: Data from 160 lung transplant recipients were retrospectively collected. Univariate Cox analysis assessed the impact of patient characteristics on time to first rejection episode. Kaplan-Meier survival analysis, LASSO regression, and multivariate Cox analysis were used to select prognostic indicators and develop a riskScore model. Model performance was evaluated using Kaplan-Meier analysis, time-dependent ROC curves, and multivariate Cox regression.

Results: Patient characteristics were not significantly associated with the time to the first rejection episode. Six laboratory indicators-Activated Partial Thromboplastin Time, IL-10, estimated intrapulmonary shunt, 50% Hemolytic Complement, IgA, and Complement Component 3-were identified as significant predictors and integrated into the riskScore. The riskScore demonstrated good predictive performance. It outperformed individual indicators, was an independent risk factor for rejection, and was validated in the validation dataset.

Conclusion: The riskScore model effectively predicts time to first rejection episode in lung transplant recipients.