Premium
Confirmation of neurometabolic diagnoses using age‐dependent cerebrospinal fluid metabolomic profiles
Author(s) -
Peters Tessa M. A.,
Engelke Udo F. H.,
Boer Siebolt,
Heeft Ed,
Pritsch Cynthia,
Kulkarni Purva,
Wevers Ron A.,
Willemsen Michèl A. A. P.,
Verbeek Marcel M.,
Coene Karlien L. M.
Publication year - 2020
Publication title -
journal of inherited metabolic disease
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.462
H-Index - 102
eISSN - 1573-2665
pISSN - 0141-8955
DOI - 10.1002/jimd.12253
Subject(s) - metabolomics , metabolome , biomarker , cerebrospinal fluid , medicine , neurology , biomarker discovery , diagnostic biomarker , medical diagnosis , bioinformatics , diagnostic accuracy , pathology , chemistry , proteomics , biology , biochemistry , psychiatry , gene
Timely diagnosis is essential for patients with neurometabolic disorders to enable targeted treatment. Next‐Generation Metabolic Screening (NGMS) allows for simultaneous screening of multiple diseases and yields a holistic view of disturbed metabolic pathways. We applied this technique to define a cerebrospinal fluid (CSF) reference metabolome and validated our approach with patients with known neurometabolic disorders. Samples were measured using ultra‐high‐performance liquid chromatography‐quadrupole time‐of‐flight mass spectrometry followed by (un)targeted analysis. For the reference metabolome, CSF samples from patients with normal general chemistry results and no neurometabolic diagnosis were selected and grouped based on sex and age (0‐2/2‐5/5‐10/10‐15 years). We checked the levels of known biomarkers in CSF from seven patients with five different neurometabolic disorders to confirm the suitability of our method for diagnosis. Untargeted analysis of 87 control CSF samples yielded 8036 features for semiquantitative analysis. No sex differences were found, but 1782 features (22%) were different between age groups ( q < 0.05). We identified 206 diagnostic metabolites in targeted analysis. In a subset of 20 high‐intensity metabolites and 10 biomarkers, 17 (57%) were age‐dependent. For each neurometabolic patient, ≥1 specific biomarker(s) could be identified in CSF, thus confirming the diagnosis. In two cases, age‐matching was essential for correct interpretation of the metabolomic profile. In conclusion, NGMS in CSF is a powerful tool in defining a diagnosis for neurometabolic disorders. Using our database with many (age‐dependent) features in CSF, our untargeted approach will facilitate biomarker discovery and further understanding of mechanisms of neurometabolic disorders.