GC-MS urinary steroid metabotyping of aldosterone deficient states

Huijie Li; Joern Pons-Kuehnemann; Michaela Hartmann; Stefan Wudy

doi:10.1530/endoabs.110.P83

P83

Prev Next

Section Contents Cite

Endocrine Abstracts (2025) 110 P83 | DOI: 10.1530/endoabs.110.P83

ECEESPE2025 Poster Presentations Adrenal and Cardiovascular Endocrinology (169 abstracts)

GC–MS urinary steroid metabotyping of aldosterone deficient states

Huijie Li ¹ , Joern Pons-Kuehnemann ² , Michaela Hartmann ¹ & Stefan Wudy ¹

Author affiliations

¹Justus Liebig University, Center of Child an Adolescent Medicine, Division of Pediatric Endocrinology & Diabetology, Laboratory for Translational Hormone Analysis, Steroid Research & Mass spectrometry Unit, Giessen, Germany; ²Institute of Medical Informatics, Justus Liebig University, Medical Statistics, Giessen, Germany

JOINT1642

Introduction: Disorders of isolated deficient aldosterone action involve insufficient production of aldosterone (aldosterone synthase defects type 1 and type 2 (corticosterone methyl oxidase (CMO) I and II)), as well as pseudohypoaldosteronism (PHA) featuring end-organ hormone resistance. Aldosterone is a key regulator of sodium–potassium homeostasis and blood pressure. Deficient action is characterized by hypotension, hyponatremia, hyperkalemia, and dehydration. We investigated whether gas chromatographic–mass spectrometric (GC–MS) urinary steroid metabolome analysis allows for delineation of these entities.

Method: We quantified 44 urinary steroid metabolites from spot urine (μg/l) by targeted GC–MS from 124 infants with aldosterone deficient states (24 CMO I, 26 CMO II, 74 PHA; aged 3–348 days) and 138 matched controls. Relative enzymatic activities were calculated from precursor/product metabolite ratios. Data preprocessing included log2 transformation, Z-score standardization, and quantile normalization, followed by logistic regression and decision tree analysis.

Results: Male patients dominated in all diseases (68%). All entities peaked around the end of the neonatal period (wk4) with CMO manifesting up to six months and PHA extending until the end of the first year of life. Elevated corticosterone metabolite levels distinguished patients best from controls. PHA showed grossly elevated metabolites of aldosterone and its precursors. The ratio between corticosterone and aldosterone metabolites discriminated best between CMO subtypes and PHA. The ratio between 18-hydroxylated corticosterone and aldosterone metabolites differentiated CMO II from CMO I. Decision tree (rpart) analysis identified various sequential classifiers distinguishing controls, PHA, CMO I and CMO II, with high specificity (94%) and sensitivities (96%, 92% and 77%), respectively.

Conclusions: GC–MS urinary steroid metabotyping from spot urine provides a non-invasive and highly reliable new diagnostic tool for delineating aldosterone deficient states in young infants. Various metabolites and metabolite ratios effectively discerned controls, patients with CMO I, CMO II and PHA. The quantitative biomarkers we found allow for a steroid metabolomics based precision medicine approach.