Alternative pathway dysregulation in tissues drives sustained complement activation and predicts outcome across the disease course in COVID-19

Abstract

Complement, a critical defence against pathogens, has been implicated as a driver of pathology in COVID-19. Complement activation products are detected in plasma and tissues and complement blockade considered for therapy. To delineate roles of complement in immunopathogenesis, we undertook the largest comprehensive study of complement in an COVID-19 to date, a comprehensive profiling of 16 complement biomarkers, including key components, regulators and activation products, in 966 plasma samples from 682 hospitalised COVID-19 patients collected across the hospitalisation period as part of the UK ISARIC4C study. Unsupervised clustering of complement biomarkers mapped to disease severity and supervised machine learning identified marker sets in early samples that predicted peak severity. Compared to heathy controls, complement proteins and activation products (Ba, iC3b, terminal complement complex) were significantly altered in COVID-19 admission samples in all severity groups. Elevated alternative pathway activation markers (Ba and iC3b) and decreased alternative pathway regulator (properdin) in admission samples associated with more severe disease and risk of death. Levels of most complement biomarkers were reduced in severe disease, consistent with consumption and tissue deposition. Latent class mixed modelling and cumulative incidence analysis identified the trajectory of increase of Ba to be a strong predictor of peak COVID-19 disease severity and death. The data demonstrate that early-onset, uncontrolled activation of complement, driven by sustained and progressive amplification through the alternative pathway amplification loop is a ubiquitous feature of COVID-19, further exacerbated in severe disease. These findings provide novel insights into COVID-19 immunopathogenesis and inform strategies for therapeutic intervention.

Rights

This content is not covered by the Open Government Licence. Please see source record or item for information on rights and permissions.

Cite as

Downloadable citations

Identifiers

DOI

https://doi.org/10.1111/imm.13585

Repository URI

https://www.research.ed.ac.uk/en/publications/d36d494e-07aa-4d5b-a9d1-c2396e90995f