MSACL 2026 Abstract

Optimized Extraction of Dried Blood Spots for Untargeted HILIC Metabolomics: A Factorial Comparison with Independent Confirmation

Étienne Ljóni Poisson (1), Freyr Jóhannsson (2), Jón Jóhannes Jónsson (2, 3), Óttar Rolfsson (1)
(1) Faculty of Biochemistry and Molecular Biology, University of Iceland, Reykjavík, Iceland, (2) Department of Genetics and Molecular Medicine, Landspítali University Hospital, Reykjavík, Iceland, (3) Faculty of Medicine, University of Iceland, Reykjavík, Iceland

Etienne Ljoni Poisson, MsSci (Presenter) University of Iceland

Relevant Financial Disclosures (within past 24 months, reported on Apr 09, 2026) No relevant financial relationship(s) to disclose.

INTRODUCTION:
Untargeted metabolomics of dried blood spots holds transformative potential for clinical diagnostics, newborn screening, and rare disease biomarker discovery. Clinical translation is currently impeded by a lack of systematically validated, matrix-optimized analytical pipelines. This study details the development and characterization of a UPLC-QTOF platform integrating hydrophilic interaction liquid chromatography (HILIC) in positive and negative electrospray ionization modes, anchored by a 13-component isotopically labelled internal standard cocktail and a 387-compound spectral library of clinically relevant small molecules.

METHODS:
A robust factorial design was executed to evaluate extraction parameters for 3.2 mm dried blood spot punches. The matrix compared four extraction solvents (varying ratios of methanol, water, and acetonitrile), two post-extraction processing strategies (direct supernatant versus vacuum-dried reconstitution), and two extraction durations (45 versus 90 minutes). A subsequent independent experiment evaluated multiple reconstitution solvent compositions to isolate the variables driving analyte loss. Data processing relied on targeted peak integration against the established spectral library, normalized to the internal standard cocktail to ensure high data fidelity, consistency, and structural reproducibility.

RESULTS:
Buffer composition functioned as the overwhelmingly dominant variable influencing both metabolite coverage and signal intensity. Aqueous methanol systems systematically outperformed pure organic solvents, specifically the reference standard of 100% methanol, for the recovery of highly polar, multiply charged species such as nucleotides and phosphorylated sugars. The 45-minute direct supernatant protocol utilizing 4:1 methanol to water achieved optimal analytical performance, recovering 24.8% of negative and 24.4% of positive tracked library entries with exceptional within-group reproducibility (median SD(log10) equal to 0.042). Vacuum-drying and pellet reconstitution systematically reduced both coverage and precision. The targeted reconstitution experiment confirmed this evaporative loss could not be rescued by downstream solvent optimization. Testing of sequential solvent addition indicated modest precision gains but did not alter the core set of detectable metabolites.

CONCLUSION:
Aqueous methanol direct supernatant extraction fundamentally outperforms standard reconstituted pellet protocols for untargeted HILIC analysis of dried blood spots. Eliminating the drying step removes a critical point of variance and minimizes the irreversible loss of labile or adsorption-prone species. By establishing an optimized, highly reproducible analytical workflow, this framework satisfies the rigorous quality control and throughput parameters necessary for large-scale clinical implementation, immediately underpinning a prospective 1,000-newborn baseline cohort and ongoing rare disease biomarker investigations.