Cell-free DNA (cfDNA) and plasma proteomics reveal inflammatory heterogeneity in vaso-occlusive crises in sickle cell disease (SCD). Free

Background: SCD is characterized by chronic hemolytic anemia and recurrent vaso-occlusive crises (VOCs), which are a leading cause of hospitalization and morbidity. The mechanisms driving VOC are complex, with a subset induced by acute inflammation, while others reflect exacerbations of chronic pain. This clinical heterogeneity complicates treatment and underscores the need for biomarkers that can distinguish between VOC subtypes to facilitate more effective interventions. cfDNA, a marker of innate immune activation, has been observed to be elevated in VOC. We hypothesize that plasma levels of cfDNA may help identify inflammation-associated VOC episodes.

Aims: To perform high-throughput proteomic analysis of matched plasma samples collected from individuals with SCD at baseline and during VOC, and integrate these data with cfDNA, to both identify molecular pathways activated during crisis and distinguish inflammation-driven episodes from non-inflammatory VOC.

Methods: We enrolled 46 individuals with SCD who were hospitalized or receiving outpatient infusion therapy for VOC. EDTA plasma samples were collected at enrollment and again at a follow-up hematology clinic visit during clinical baseline for 39 of these individuals. Samples were stored at –80°C prior to analysis. Proteomic profiling of the 39 matched samples was performed using the Olink® Explore platform to quantify normalized protein expression levels for over 1,000 plasma proteins. Paired student t-tests were performed to identify differentially expressed proteins with the threshold of P < 0.01. cfDNA concentrations were measured using a SYTOX™ Green fluorescent plate assay. Demographics, current medications, and clinical laboratory parameters obtained at the time of sample collection were abstracted from the electronic medical record.

Results: Proteomic analysis identified expected sex differences, as well as increased levels of Arginase-1 (a known marker of hemolysis) during VOC, supporting the validity of the assay.

A total of 38 proteins were significantly differentially expressed (P < 0.01) between VOC and baseline. Components of the ubiquitin–proteasome system (UPS), including multiple 19S and 20S proteasome subunits and regulatory proteins were elevated during VOC, as were post-translational modification proteins, suggesting increased ubiquitination and SUMOylation during VOC. Immune-related cytokines and modulators, including Oncostatin M and Interleukin 32, were significantly upregulated, consistent with an inflammatory state.

Pathway enrichment analysis of differentially expressed proteins confirmed significant upregulation (FDR < 0.01) of several key biological pathways during VOC. These included proteasome-mediated protein degradation, cytokine-mediated signaling, regulation of apoptosis, and oxidative stress response. While some plasma proteins may have been released due to increased hemolysis or cell turnover, the overlap of related pathways supports a coordinated inflammatory response during VOC.

Although cfDNA levels trended higher during VOC, these differences were not statistically significant and showed considerable overlap between time points, suggesting inflammatory heterogeneity. However, principal component analysis (PCA) was performed across the full VOC cohort (n=46) and revealed distinct clustering of individuals with high cfDNA (>0.75 µg/mL) versus low cfDNA (<0.75 µg/mL) at VOC. Patients with high cfDNA had significant changes in proteins related to inflammation, vascular stress, and vesicle trafficking, suggesting a coordinated systemic response during cfDNA elevation. Complementing these findings, Spearman correlation analysis across all samples showed that cfDNA levels were significantly correlated (P < 0.01) with over 60 plasma proteins involved in pathways related to cell adhesion, inflammation, and epithelial integrity.

Conclusion: VOC is associated with increased plasma levels of proteins related to proteasome activity, cytokine signaling, and apoptosis. cfDNA correlated with a broad inflammatory proteomic signature, supporting its potential as a marker of inflammatory VOC phenotype. To our knowledge, this is the first study to integrate high-throughput plasma proteomics with cfDNA measurement to stratify VOC inflammatory subtypes. These findings enhance our understanding of VOC pathobiology and highlight potential targets for biomarker development and therapeutic intervention.