박주리 ((재)씨젠의료재단)

백제현 ((재)씨젠의료재단)

박주리 ((재)씨젠의료재단)

김효진 ((재)씨젠의료재단)

이세영 ((재)씨젠의료재단)

양원석 ((재)씨젠의료재단)

백제현 ((재)씨젠의료재단)

Liquid chromatography–mass spectrometry (LC-MS) is increasingly useful in biomarker discovery and clinical diagnostics. However, efficient and consistent sample preparation remains a critical bottleneck, especially for large-scale clinical analysis, affecting analytical accuracy and reproducibility. In particular, implementing LC-MS assays in routine diagnostics requires robust automated sample preparation. In this study, upon a previously developed clinical assay, we implemented two automated sample preparation platforms using the KingFisher™ Flex Purification system (Thermo Scientific) and the BRAVO liquid handler (Agilent Technologies). Both platforms processed human serum samples for quantifying alpha-fetoprotein (AFP) and fucosylated AFP, established biomarkers for hepatocellular carcinoma (HCC). Samples were analyzed using an Evosep One LC coupled with a SCIEX Triple Quad 6500+ MS. We compared both automated systems with manual preparation across four criteria: quantitative performance, cost, processing time, and sample throughput. Both automated platforms outperformed manual preparation across all the criteria. While both showed similar quantitative performance, the KingFisher™ Flex-based platform was considered more suitable for routine clinical application in terms of its operational simplicity and cost advantages. We propose the automated LC-MS-based method developed in this study as a practical solution for large-scale clinical sample analysis.