# Untargeted Lipidomics in Fabry Disease of Urine Samples by Low-Resolution Flow Injection Mass Spectrometry (ESI(±)-LTQ MS)

## TL;DR

This study explores using urine samples and lipidomics with mass spectrometry to diagnose Fabry disease, a genetic disorder, with high accuracy.

## Contribution

The novel use of urine samples and multivariate analysis in lipidomics for early diagnosis of Fabry disease is proposed.

## Key findings

- Combining positive and negative ionization modes in PLS-DA achieved 92% accuracy in differentiating Fabry disease cases from controls.
- Urine-based lipidomics combined with multivariate analysis is a promising tool for early diagnosis of Fabry disease.
- The method enables improved healthcare outcomes through early detection of the disease.

## Abstract

Background: Fabry
disease (FD) is a lysosomal storage disease caused
by genetic mutations related to the coding of the enzyme α-galactosidase
A, which is responsible for the metabolism of glycosphingolipids such
as globotriaosylceramide and globotriaosylsphingosine. The accumulation
of these and other metabolites can occur in various types of cells
and impair the functioning of multiple organs and systems, such as
the heart, brain, and kidneys. However, with early diagnosis and appropriate
therapeutic intervention, the clinical outcome can be significantly
improved. This study aimed to analyze the performance of new diagnostic
methods for FD using the broad field of lipidomics combined with multivariate
analyses, proposing the use of urine as a specimen. Materials and
Methods: urine samples were collected from patients with both confirmed
(Case) and negative (Control) diagnoses of FD, which were later processed
for specific lipid extraction. After extraction, 81 samples (44 cases
and 37 controls) were subjected to mass spectrometry analysis, with
direct infusion and electrospray ionization in both positive and negative
modes (ESI(±)). After spectral acquisition, the data were processed
and analyzed using multivariate analysis methods such as Principal
Component Analysis (PCA) and Partial Least Squares Discriminant Analysis
(PLS-DA). Results: the combination of both ionization modes for PLS-DA
was able to differentiate between the Case and Control groups with
92% accuracy. Conclusion: this paper suggests that the proposed method
of application of lipidomics combined with multivariate analyses as
a tool for early diagnosis of FD is promising, enabling and contributing
to the improvement of healthcare for these patients.