An Analytical Strategy for Reliable Metabolome Analysis of Clinical Leftover Sera Using Timed Aliquoting

Accurate metabolome analysis depends not only on advanced analytical techniques but also on strict control of preanalytical variables. This study presents an analytical strategy for reliable untargeted metabolomics using clinical leftover sera, focusing on the impact of timed aliquoting during short-term storage. Leftover serum samples from routine clinical testing, typically stored at 4 °C for up to 7 days, offer a valuable and accessible resource for biomarker discovery. However, variable delays in sample aliquoting and storage can compromise metabolite stability. We used high-coverage ¹²C-/¹³C-dansylation LC-MS to profile the amine/phenol submetabolome in serum samples collected from healthy individuals at multiple time points postdraw. The study included 630 LC-MS runs (105 subjects × 6 time points) in the discovery set and 280 runs (70 subjects × 4 time points) in the validation set, quantifying 1382 and 1352 metabolites, respectively. Although time-dependent changes in metabolite abundances were observed, these shifts were relatively small between adjacent time points. Notably, clear sex-based metabolic separation was observed when using samples aliquoted at the same time or within 24-h intervals, whereas discriminatory power diminished when samples with longer storage time differences were combined. These findings demonstrate that, with carefully timed aliquoting, clinical leftover sera can be reliably used for metabolomics. Our study establishes a practical and scalable workflow to control preanalytical variation─specifically by minimizing storage time differences─thereby enabling broader use of clinical samples in biomarker discovery and population-scale metabolomics.