Early detection of hepatocellular carcinoma (HCC) remains a persistent worldwide challenge. Owing to its minimal invasiveness, liquid biopsy has emerged as a promising alternative for early screening. As key components of the tumor microenvironment (TME), platelets (PLTs) represent a rich source of biomolecular information that complements the data from conventional plasma and serum samples. Integrative multiomics analysis of such data offers a powerful strategy to deepen our understanding of hepatocarcinogenesis and accelerate the discovery of robust biomarker panels. Here, we described an integrative and ultrafast multiomics sample preparation (IAU-MOSP) strategy for the high-purity platelets. The optimized IAU-MOSP method shortened the multiomics workflow from over 24 to 6 h while yielding comparable biomolecule identifications to those of conventional methods. Then, the workflow was applied in an HCC cohort (n = 68) study. We quantified 6660 biomolecules with high reproducibility (median CVs: 0.31-0.39). The data exhibited strong cross-omics correlations, particularly between proteins and lipids (r = 0.75) as well as protein and metabolite (r = 0.68) groups. Differential analysis revealed 10 biomolecules significantly dysregulated in HCC platelets (TEK, citric acid, glycerol-3-phosphate (G3P), P2RX4, malic acid, ATP, PRG3, ITGAM, CXCR2, ITGB2) that participate in key pathways driving proliferation and metastasis. Accompanied by machine learning, the 10 biomolecules were ultimately identified as a potential biomarker panel for early diagnosis of HCC. It shows superior diagnostic efficacy (accuracy = 0.81, sensitivity = 0.74) over α-fetoprotein (AFP) (accuracy = 0.75, sensitivity = 0.45) for early HCC detection.
Shen et al. (2026) studied this question.