A Recurrence-Specific Gene-Based Prognosis Prediction Model for Lung Adenocarcinoma through Machine Learning...

Area of Science:

• Oncology
• Genomics
• Bioinformatics

Background:

• Lung adenocarcinoma (LUAD) has a high recurrence rate (30-75%) after surgery, leading to poor survival.
• Identifying patients at high risk of recurrence is crucial for timely and intensive therapeutic interventions.

Purpose of the Study:

• To develop and validate a gene expression-based prediction model for recurrence-free survival (RFS) in LUAD patients.
• To identify key genes and pathways associated with LUAD recurrence.

Main Methods:

• Analysis of gene expression data from TCGA and GEO databases for LUAD.
• Identification of differentially expressed genes (DEGs) between primary and recurrent tumors.
• Application of LASSO Cox and multivariate Cox regression to build a prognostic model based on key genes.

Main Results:

• A 13-gene prediction model for RFS was established, accurately classifying patients into high- and low-risk groups.
• The high-risk group demonstrated significantly worse RFS in training and validation cohorts.
• The model showed high accuracy (AUC=96.3%) in predicting recurrence, outperforming clinicopathological features. The G2M checkpoint pathway was implicated in recurrence.

Conclusions:

• A novel gene-based prognostic model effectively predicts LUAD recurrence risk.
• This model aids clinicians in stratifying patients for individualized therapy, potentially improving survival outcomes.