Diagnosis of systemic Lupus erythematosus Nephritis using Nuclear Magnetic Resonance Metabolomics
Low-grade invasive method for early stage detection and continuous monitoring of SLEN
Kidney involvement in systemic lupus erythematosus (SLE) patients is a severe negative prognostic factor associated with substantial direct, indirect and intangible costs. The current WHO diagnostic guidelines stipulate that SLE-nephritis (SLEN) is to be diagnosed through kidney biopsy. Monitoring of SLEN is usually carried out by routine measurement of several urine parameters like proteinuria or the presence of abnormal urinary sediments. However, these indicators are only positive at advanced stages of SLEN, whereas at early stages they can be within the norm. Thus, early stage SLEN can only be accurately detected and monitored by very invasive and risk-loaded biopsy interventions. Therefore, the aim of our study was to test the potential of using 1H nuclear magnetic resonance (NMR) based metabolomics and lipidomics of serum and urine as a low-grade invasive method to detect, grade and monitor SLEN. For our project we collected paired serum and urine samples from SLE patients (n=109), patients with granulomatosis with polyangiitis (GPA) or micropolyangiitis (MPA) as a positive control group (n=31), and healthy individuals (n=10). We have also standardized the preparation of the urine samples to lower the interference of the different pH, which is major confounder when doing 1H NMR metabolomics. We defined a set of 26 small metabolites and 11 lipid-cores in the serum (Fig 1A) and 46 small metabolites in the urine (Fig 1B). Over 400 clinical and demographic parameters have also been collected for each patient.
Analysis of the serum metabolome and lipidome between patients with no SLEN (G0), low-grade (G1+2) and high-grade SLEN (G3+4) shows marked differences in metabolite concentrations (Fig 2). The quantification of the urine spectra is currently underway. After all metabolites have been quantified we will integrate clinical, demographic and metabolomic data to calculate a multivariate diagnostic model for SLEN. This will establish a non-invasive method for continuous monitoring of SLEN, allowing a swift adaptation of therapeutic regimens and improve patient’s kidney health and overall well-being.
Example of the metabolites identified in the serum (A) and in the urine (B) of the same SLE patient.
Important features identified by PLS-DA. The colored boxes on the right indicate the relative concentrations of the corresponding metabolite in each group under study.
