A translational LC-MS/MS framework for lipid biomarker identification and quantification in human plasma

Abstract Altered lipid metabolism is a breast cancer hallmark, yet translating lipidomic discoveries into clinical biomarkers remains constrained by analytical variability and limited validated frameworks. This challenge is compounded by a chicken-and-egg problem: analytical standards required for lipid identification and quantitation are expensive, yet the diagnostic relevance of target lipids remains uncertain until quantified reliably. Previous work indicated 52 lipid biomarker species with potential for breast cancer prediction. This study verified 48 of these lipid species and developed a quantitative LC-MS/MS method, providing the basis for clinical assay development. We established a curated library using authentic lipid standards to verify plasma lipids through retention-time matching and high-resolution spectral comparison. We confidently identified 41 lipids in plasma based on co-elution with standards and diagnostic fragment ions. We performed method qualification across 48 lipids in parallel with identification, assessing accuracy, precision, recovery, and linearity. Ultimately, 46 lipids met all predefined qualification criteria. Practical constraints - including time, cost, and availability of authentic standards - necessitated parallel identification and targeted method development, highlighting inherent challenges in translating lipidomics into clinical assays. This workflow provides a reproducible framework for integrating lipidomics into biomarker discovery and clinical applications.