A PRELIMINARY STUDY OF THE RELATIONSHIP BETWEEN PLASMA MICROBIAL CELL-FREE DNA AND DISEASE ACTIVITY IN PATIENTS WITH LUPUS

PT001 / #268 Topic: AS04 - Biomarkers POSTER TOUR 02: RECENT INSIGHTS ON THE PATHOGENESIS OF LUPUS NEPHRITIS 23-05-2025 10:00 AM - 10:40 AM Background/Purpose Systemic lupus erythematosus (SLE) is a chronic autoimmune disease characterized by widespread tissue inflammation and damage in association with antinuclear antibody production. Emerging research suggests that disturbances in the microbiome (dysbiosis) can interact with the immune system to drive pathogenesis. Microbial cell-free DNA (mcfDNA) in plasma, analogous to human cell-free DNA, is thought to originate from microbial organisms undergoing cellular turnover. These microbial derived DNA fragments can transverse into the bloodstream and may be processed by circulating DNases. However, these degraded fragments may also be readily detected, identified, and quantified in plasma using advanced molecular and bioinformatics methodologies. The purpose of this pilot study was to explore a possible relationship between plasma mcfDNA and disease activity in patients with lupus. Methods Plasma samples from patients with lupus were collected at 2 clinical centers. Patients were clustered into 3 groups: complete remission, remission with a positive anti-dsDNA titre, and active disease (SLEDAI greater > 6) with a positive anti-dsDNA titre. Specimens from a healthy cohort were derived from an independent collection center. Plasma was collected, processed, and stored in K2-EDTA tubes. Cell-free DNA was extracted from plasma via the Karius Discovery assay and sequenced at a depth of 400M paired-end reads per sample. A set of analytical filters was applied to control for contamination, separating biological signals from background. Differential abundance analysis, correlation analysis, and principal coordinate analysis were conducted to identify microbial signatures that discriminated between the healthy and lupus patient populations as well as the disease activity groupings. Identified features were incorporated into a gradient-boosted machine learning classifier to assess their predictive power. Results Our study included 54 patients with SLE (median age 37.5 years, 46% had a history of lupus nephritis, 85% female) and 36 healthy controls (median age 45 years, 61% female). Our analysis indicated specific elevated microbial species, estimated in molecules per microliter (MPM), with concordant findings observed across both clinical centers. The mcfDNA that were identified were associated with the oral (Streptococcus, Prevotella, Porphyromonas, and Veillonella species); gastro-intestinal (Bacteroides, Alcaligenes, Streptomyces, and Campylobacter species); and skin (Staphylococcus, Corynebacterium, and Acintobacter species) microbiomes (Figure 1). Principal coordinate analysis and preliminary machine learning classifiers suggested a possible partition between the healthy individuals and those with lupus (Figure 2). The analysis also indicated that a subset of the microbial signatures may differentiate between disease activity groupings. Figure 1. Figure 2. Conclusions Our pilot study provides preliminary data suggesting an increase in mcfDNA concentration from signature microbial species that can distinguish patients with SLE from controls and differentiate between disease subgroups. Further studies, including longitudinal analyses of larger and more diverse patient cohorts, will be needed to determine the utility of plasma mcfDNA as a biomarker for disease activity and delineate mechanisms by which increased mcfDNA may arise and contribute to pathogenesis.