Integration of RASSF1A and TSPYL5 methylation and AFP protein as plasma biomarker for hepatocellular carcinoma diagnosis.

[BACKGROUND] Early detection of hepatocellular carcinoma (HCC) represents a significant clinical challenge due to the lack of effective biomarkers. Previous studies have revealed abnormal hypermethylation of RASSF1A and TSPYL5 in HCC tissues, prompting further investigation into their utility as potential biomarkers for HCC diagnosis. In this study, we explored the clinical value of RASSF1A/TSPYL5 methylation alone and in combination with AFP protein as diagnostic biomarkers for early HCC detection.

[METHODS] Bioinformatics analysis using MethSurv, LinkedOmics, and Wanderer was conducted to assess the methylation levels of RASSF1A and TSPYL5 in HCC tissue samples from the public database. The methylation status of these two genes was validated in cell lines and plasma samples from HCC patients. Furthermore, the ARTHCC model, comprising the measurement of RASSF1A and TSPYL5 methylation and AFP protein, was developed and validated using plasma samples from HCC patients.

[RESULTS] Our comprehensive bioinformatic analysis revealed that TSPYL5 CpG sites were hypermethylated in HCC tissue samples, particularly among CpG island and shore regions. Approximately 40% of CpG sites of RASSF1A also exhibited hypermethylation. The hypermethylation status of TSPYL5 may be associated with portal vein tumor thrombus (PVTT) in HCC patients. Combining analysis of RASSF1A and TSPYL5 methylation levels with AFP protein, the ARTHCC model demonstrates superior sensitivity and specificity for detecting HCC in both the discovery and validation cohorts. Notably, the ARTHCC model consistently exhibited robust performance across different subgroups.

[CONCLUSIONS] This study provides compelling evidence that the combination of RASSF1A/TSPYL5 methylation and AFP protein is a promising biomarker panel for the early detection of HCC. The ARTHCC model demonstrates remarkable sensitivity and specificity, outperforming AFP.