A Machine Learning Algorithm that Predicts Patients at High Risk of Lung Adenocarcinoma Recurrence

By Robert Dillard - Last Updated: January 4, 2021

Researchers developed a machine learning model that may identify patients at high risk of lung adenocarcinoma (LUAD) recurrence. The findings were published in Biomed Research International.

Despite surgical resection, about 30-75% of LUAD patients suffer from recurrence, which carries grim survival outcomes. Therefore, the researchers wrote that the, “Identification of patients with high risk of recurrence to impose intense therapy is urgently needed.”

To conduct this study, researchers first obtained gene expression data on LUAD from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. They noted that prognostic genes associated with the recurrence-free survival (RFS) of LUAD patients were identified using univariate analysis. LASSO Cox regression and multivariate Cox analysis were then applied to identify key genes to build the machine learning algorithm.

According to the results, the model was able to detect 37 Differentially expressed genes (DEGs) between primary and recurrent LUAD tumors. Univariate analysis showed 31 DEGs which were significantly correlated with RFS. Moreover, the researchers noted, the efficiency of the prediction model was further confirmed in different clinical subgroups. The researchers noted that compared to clinicopathological features, the prediction model possessed higher accuracy to identify patients with high risk of recurrence (AUC=96.3%).

“Our recurrence-specific gene-based prognostic prediction model provides extra information about the risk of recurrence in LUAD, which is conducive for clinicians to conduct individualized therapy in clinic,” the researchers wrote in conclusion.