Abstract
Targeted metabolomics kit-based assays are widely used for quantitative metabolic profiling in clinical research. However, they are primarily validated for conventional matrices such as plasma and serum. Their direct application to dried blood spots (DBS) cannot be assumed and requires specific optimization due to intrinsic matrix-specific differences. This study aims to optimize the TMIC MTX MEGA assay for DBS analysis and matrix comparison. Validation was then conducted using paired longitudinal DBS and serum samples from 11 participants in a clinical study undergoing cardiac rehabilitation. A total of 323 compounds were quantitatively measured in DBS and 496 in serum. Principal component analysis revealed matrix-driven separation, highlighting clear distinctions of plasma/serum from DBS and confirming intrinsic matrix differences. Despite global differences, longitudinal trends across metabolite classes and at individual metabolite levels were largely concordant between DBS and serum. Class-specific differences were observed, particularly among lipid species and metabolites influenced by intracellular contributions, consistent with known biological factors. Although absolute concentrations differed for several metabolites, relative temporal changes were preserved. This proof-of-concept study demonstrates the feasibility of applying targeted metabolomics kit-based assays to DBS following workflow optimization. Overall, the optimized DBS workflow shows promise for longitudinal metabolic profiling in clinical contexts. These preliminary findings support further evaluation of DBS as a minimally invasive sampling alternative for targeted metabolomics applications in larger cohorts.