Background/Purpose: Systemic lupus erythematosus (SLE) poses a significant challenge at the diagnostic and treatment levels due to its severe multisystem nature and notable heterogeneity in clinical course and therapeutic response. We investigated the serum proteomic and metabolomic profiles of SLE, to uncover new mechanisms underlying relevant clinical patterns.

Methods: Proteomic and metabolomic approaches were used to assess the serum levels of 184 inflammation and organ damage-related proteins [proximity extension immunoassay (PEA, Olink)] and 168 serum metabolites, [nuclear magnetic resonance (NMR, Nightingale)] in consecutive SLE patients (n = 135) and age-matched healthy donors (HD) (n = 27). In parallel, an extensive clinical and analytical profile was conducted. To evaluate the contribution of molecular profiles to the disease severity, unsupervised hierarchical clustering analyses were developed.

Results: Fifty-three proteins related to inflammation and organ damage, and 32 metabolites, were significantly altered (p = < 0.05, FC= 1.5, FDR) in the serum of SLE patients compared to HD. Unsupervised hierarchical clustering of proteomic data differentiated 2 patient clusters, where patients belonging to one cluster were characterized by elevated disease activity (SLEDAI >5), and increased incidence of atheroma plaques, dyslipidaemia, and hypertension. The top features driving classification were CD40, PGF, PDL1 and CSF1, with area under the curve >0.9, known to be associated with immune activation. At the molecular level, these patients showed 49 overexpressed proteins (p = < 0.05, FC= 1.5, FDR), enriched in biological pathways linked to chemotaxis and lipid metabolism.

Unsupervised hierarchical clustering of metabolomic data also identified 2 patient clusters with patients in one cluster showing a greater preponderance of lupus nephropathy (LN). The top features driving classification included short chain fatty acids, total fatty acids, total lipids and total free cholesterol, with area under the curve >0.9. Molecularly, these patients showed increased levels of 60 proatherogenic VLDL and LDL subsets (p = < 0.05, FC= 1.5, FDR) and metabolite enrichment analysis identified ketone body and butyrate pathways.

Integrating proteomic and metabolomic patient clusters identified two patient subgroups with the most (N=41) and least (N=36) severe molecular and clinical profiles. The group with more severe profiles ( SLEDAI >5, prevalence of LN, and higher incidence of atheroma plaques, hypertension, and dyslipidaemia) exhibited higher levels of 56 proteins associated with chemotaxis, inflammatory response and cytokine signalling, as well as 53 proatherogenic lipoproteins, (p = < 0.05, FC= 1.5, FDR) confirming the original molecular stratification.

Conclusion: The simultaneous application of proteomic and metabolomic approaches for molecular profiling in SLE patients offers valuable complementary information enabling a more accurate definition of their clinical profile, providing novel therapeutic targets and biomarkers of disease.

Acknowledgements: Supported by ISCIII (PI21/0591, CD21/00187 and RICOR-21/0002/0033) co-financed by FEDER

« Back to ACR Convergence 2023

ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-multi-omic-landscape-of-systemic-lupus-erythematosus-patients-highlights-metabolic-and-immune-pathways-associated-with-distinctive-clinical-profiles/