Predictive performance of a centrosome-associated prognostic model in prognosis and immunotherapy of lung adenocarcinoma.

Abstract

In recent years, mounting investigations have highlighted the pivotal role of centrosomes in cancer progression. In this study, we employed bioinformatics and statistics to establish a 13-centrosome-associated gene prognostic model for lung adenocarcinoma (LUAD) utilizing transcriptomic data from TCGA. Based on the Riskscore, patients were stratified into high- and low-risk groups. Through survival analysis and receiver operating characteristic curve analysis, our model demonstrated a consistent and robust prognostic capacity, which was further validated using the GEO database. Univariate/multivariate Cox regression analyses identified our model as an independent prognostic factor for LUAD patients. Subsequently, immunoinfiltration analysis showed that immune cell infiltration levels of aDCs, iDCs, Mast cells, and Neutrophils, as well as immune functionalities such as HLA, Type I IFN Response and Type II IFN Response, were markedly reduced in the high-risk group compared to the low-risk group. Finally, we conducted a drug screening to identify potential treatments for patients with different prognoses. We utilized the GDSC database and molecular docking techniques to identify small molecule compounds targeting the prognostic genes. In conclusion, our prognostic model exhibits robust and reliable predictive capability, and it may have important clinical implications in guiding treatment decisions for LUAD patients.