The early and accurate diagnosis of hepatocellular carcinoma (HCC) remains a major clinical challenge, particularly in patients with chronic hepatitis C virus (HCV) infection, as standard biomarkers such as alpha-fetoprotein often lack sensitivity and specificity. To address this issue, we have implemented a two-tiered metabolomics workflow that combines high-resolution untargeted discovery with rapid targeted validation. The aim is to identify robust circulating biomarkers that can distinguish HCC from HCV-related conditions, including mixed cryoglobulinemia (MC).
Plasma samples from 102 patients (69 with HCC, 23 with HCV and 10 with MC) were analyzed using ultra-high-performance liquid chromatography coupled to high-resolution mass spectrometry, integrating both hydrophilic interaction chromatography (HILIC) and reverse-phase separations. Untargeted profiling, enhanced by trapped ion mobility spectrometry (TIMS), revealed distinct molecular patterns among the groups. HCC samples were characterized by elevated levels of short- and long-chain acylcarnitines (CARs), whereas medium-chain CARs remained specific to HCV. Lipidomic analysis showed a dramatic reduction in lysophosphatidylcholines (LPCs) and a concomitant increase in phosphatidylcholines in HCC samples, indicating membrane remodeling during tumor progression.
To translate these signatures into clinical utility, we developed a fast, sensitive, six-minute HILIC-MS assay that employs Multiplexed Single Ion Monitoring (SIM) on a quadrupole-Orbitrap platform. The method used a 100 mm narrow-bore column with sub-2 µm particles and optimized chromatographic conditions to achieve high peak quality and reproducibility. A panel of 24 CARs and LPCs was quantified with excellent analytical performance, demonstrating sensitivity (LODavg: 0.0558 ng/mL; LOQavg: 0.667 ng/mL), accuracy (98.75%), linearity (R² = 0.999) and stability (intra-day CV: 0.39%; inter-day CV: 1.46%). The robustness of the approach was further confirmed by minimal matrix effects (10%) and high recovery rates (95.82%) using isotope-labelled standards.
To further strengthen these findings, an independent external cohort including both HCC and healthy control plasma samples was analyzed using the same targeted and untargeted assay. The results confirmed the previously observed metabolic trends, with consistent alterations in CARs and LPCs profiles across disease groups. This validation underscores the reproducibility and translational relevance of the identified biomarker panel.
The targeted panel validated the molecular signature identified through a discovery approach. This integrated approach bridges the gap between discovery and validation, showcasing the potential of TIMS-enhanced untargeted analysis and Orbitrap-based targeted quantification in the development of precision diagnostics in oncology.