Integrated multi-omics characterization across clinically relevant subgroups of long COVID

Abstract

When SARS-CoV-2 became regional epidemics, a substantial number of patients suffered from post-acute sequelae of COVID-19 (PASC, aka long COVID). Exploring the pathogenesis and especially the heterogenicity features of long COVID subgroups is of paramount importance for understanding its etiology. In this study, through integrative multi-omics analyses encompassing transcriptomics, proteomics, and metabolomics, long COVID patients exhibited overall elevated MAPK pathway activation, while patients who have recovered from long COVID showed down-regulation of this response. Long COVID heterogenicity is described by multi-omics distinct signatures for each subgroup. The Multisystemic (MULTI) symptom subgroup is characterized by enhanced glycerophospholipid and ether lipid metabolism, Neurological (NEU) by augmented glycoprotein synthesis metabolism, Cardio cerebral (CACRB) by increased pyruvate metabolism and suppressed macrophage polarization, Musculoskeletal + Systemic (MSK + SYST) by elevated glycerophospholipid metabolism, and Cardiopulmonary (CAPM) by inhibited NF-κB signaling pathways. ABHD17A, CSNK1D, PSME4 and SYVN1 were general long COVID combination biomarkers, while CRH (MULTI), FPGT (NEU), CBX6 (CACRB) and RBBP4 (CAPM) were selected as serum-specific subgroup proteins. Our study provides a commonly shared and distinct pathophysiological explanation underpinning PASC, paving the way for future diagnosis and therapeutic interventions.