Constructing a Risk Prediction Model for Lung Cancer Recurrence by Using Gene Function Clustering and Machine Learning.

Abstract

A significant proportion of patients with early non-small cell lung cancer (NSCLC) can be cured by surgery. The distant metastasis of tumors is the most common cause of treatment failure. Precisely predicting the likelihood that a patient develops distant metastatic risk will help identify patients who can further intervene, such as conventional adjuvant chemotherapy or experimental drugs. Current molecular biology techniques enable the whole genome screening of differentially expressed genes, and rapid development of a large number of bioinformatics methods to improve prognosis. The genes associated with metastasis do not necessarily play a role in the pathogenesis of the disease, but rather reflect the activation of specific signal transduction pathways associated with enhanced migration and invasiveness. In this study, we discovered several genes related to lung cancer resistance and established a risk model to predict high-risk patients.