Background: There is ample evidence that macrophages play a crucial role in the biological processes of hepatocellular carcinoma (HCC). This study was designed to establish a novel macrophage-associated prognostic model for HCC.
Methods: Long non-coding RNA (lncRNA) microarrays and clinical data in The Cancer Genome Atlas (TCGA) database were analysed using a univariate Cox proportional regression model to select macrophage-associated prognostic lncRNAs. Multivariate Cox proportional regression models and survival analysis were used to establish a prognosis index (PI) model. Furthermore, to better understand the biological functions of differentially expressed macrophage-associated lncRNAs (MALs) in HCC, enrichment analysis was performed. Finally, the correlation between MALs and clinical features was further analysed in HCC.
Results: We identified eight MALs with significant prognostic values for HCC. Next, a PI model for HCC was developed, and patients were classified into the high-risk or low-risk group based on risk scores. The overall survival (OS) of high-risk patients was significantly shorter than that of low-risk patients (P < 0.001). Univariate and multivariate factors indicate that risk scores can be used as independent prognostic factors for patients with HCC. Multiple receiver operating characteristic (ROC) plots show that the area under the ROC curve (AUC) of the risk score is higher than that of other clinical features. The C-index of our nomogram was 0.768.
Conclusion: The PI model has a prognostic efficacy superior to that of other clinical features.
Keywords: macrophage-associated lncRNAs, The Cancer Genome Atlas, prognosis index model, liver hepatocellular carcinoma