Immune-coagulation dynamics in severe COVID-19 revealed by autoantibody profiling and multi-omics integration

Abstract

Severe COVID-19 is characterized by immune-coagulation dysregulation, yet the contribution of related autoantibodies remains poorly understood. We investigated relationships between plasma autoantibody reactivities, whole-blood transcriptomics, plasma proteomics, and clinical laboratory parameters in a cohort of hospitalized COVID-19 patients. Transcriptomic analysis revealed that 42 curated coagulation and complement cascade genes were upregulated in severe cases compared to healthy controls, with 15 genes, including CR1L, ELANE, ITGA2B, ITGB3, VWF, TFPI, PROS1, MMRN1, and SELP (> 1.2 log2 fold-change), also significantly different from mild cases. Autoantibody profiling against eight coagulation-related proteins (ADAMTS13, Factor V, Protein S, SERPINC1, Apo-H, PROC1, Prothrombin, and PF4) showed reactivities below positivity thresholds across all groups. Using an exploratory approach, in severe cases, subthreshold autoantibody candidates (FDR < 0.25) showed negative correlation trends with select gene expressions and inflammatory markers (Factor V with IL-6 and CXCL10), suggesting potential disease-specific immunomodulatory associations. In contrast, while mild cases exhibited stronger gene-protein correlations, they showed limited associations with antigen reactivities or clinical laboratory parameters. Additionally, no correlations were observed between autoantibodies and platelet-counts or Fibrin-D-dimer levels. Age-associated increases in antigen reactivities were noted in severe disease, implying a role for immunosenescence. These findings support further investigation into the role of subthreshold autoantibody candidates in thromboinflammatory COVID-19 pathogenesis.