Background: Alpha-fetal protein (AFP) is an important conventional clinical diagnostic indicator of hepatocellular carcinoma (HCC). However, the utilization of AFP alone might yield deceptive results due to its limited sensitivity and accuracy.
Aims: Our study was designed to investigate latent diagnostic biomarkers that could function as auxiliary clinical indicators of HCC and enhance the accuracy of joint diagnosis with AFP.
Methods: We analyzed gene expression profiles and clinical data from HCC patients in The Cancer Genome Atlas database. Differentially expressed genes were identified, and a gene set enrichment analysis was conducted to uncover their biological functions and molecular processes. A weighted correlation network analysis and a protein-protein interaction analysis were performed to detect AFP-related biomarkers. The diagnostic performance of these biomarkers was verified using datasets from the GEO database. A diagnostic nomogram was established using the expression levels of potential biomarkers. Quantitative real-time PCR was performed using tissues from 16 HCC patients to validate the results.
Results: Five AFP-related diagnostic biomarkers, CDT1, MCM7, NUDT1, CENPM, and HDAC11, were discovered. The diagnostic performance of these biomarkers and the nomogram were demonstrated to be excellent according to receiver operating characteristic curves. CDT1, MCM7, and NUDT1 were shown to be up-regulated in HCC tissues through quantitative real-time PCR.
Conclusions: We discovered five diagnostic biomarkers and established a nomogram as a complement to AFP in the diagnosis of HCC. Our results provide a more accurate diagnostic plan for HCC patients based on next-generation sequencing compared with AFP alone.
Keywords: AFP; Biomarkers; Diagnosis; Hepatocellular carcinoma; Nomogram; TCGA.