Biomarker discovery boosts long-COVID prediction accuracy to 78.5%

In a recent paper uploaded to the medRxiv preprint* server, researchers tested the hypothesis that altered regulation of the complement cascade might result in long-COVID, and the biomarkers of this dysregulation may predict disease presence and outcome. They analyzed plasma samples of patients with long-COVID and controls who did not suffer from the condition despite prior severe SARS-CoV-2 infection. Their analysis revealed significant differences in the complement pathways of cases and controls. These findings suggest that testing for just four clinically traceable biomarkers is sufficient to predict long-COVID with 78.5% accuracy.

Study: Complement dysregulation is a predictive and therapeutically amenable feature of long COVID. Image Credit: tilialucida / Shutterstock

*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

Long-COVID and the need for disease diagnoses and prediction

The coronavirus disease 19 (COVID-19) pandemic remains one of the worst in human history, infecting more than 771 million people and claiming almost 7 million lives since its emergence in late 2019. Despite extensive global vaccination campaigns significantly reducing the disease's burden, a large proportion of the 763+ million survivors suffered from chronic symptoms long after 'recovering' from the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) mediated condition.

Clinically referred to as "post-acute sequelae of SARS-CoV-2", the colloquial umbrella term 'long-COVID' refers to persistent COVID-19-like symptoms that persist 12 or more weeks following recovery from acute COVID-19 infections. Symptoms usually mirror those observed during infection and include cognitive blunting ("brain fog"), chest pain, severe fatigue, sensory dysregulation (mainly auditory and olfactory), and dyspnoea.

Studies have shown that long-COVID has a significant detrimental effect on daily routines and the overall quality of life of affected individuals, resulting in national-scale work leave and socioeconomic loss. Research has estimated that between 41-45% of all COVID-19 patients experience some form of long-COVID, with global estimates at 313+ million patients. More than 40% of patients report symptoms persisting for two years or more.

Unfortunately, long-COVID remains poorly understood – condition diagnosis is based on patient-reported symptoms, and no clinical diagnostic test has hitherto been identified. Numerous hypotheses to explain the mechanisms of long-COVID have been proposed, including viral persistence, coagulation defects, and endothelial- and immune dysregulation. Studies aimed at verifying these hypotheses, however, remain inconclusive.

Recent work on patients suffering from long-COVID has identified persistent inflammation as a shared characteristic between afflicted individuals. An inflammatory response is a highlight of the complement system dysregulation, found in many diseases (including severe COVID-19) and clinically identified by elevated C-reactive protein (CRP) and proinflammatory cytokines. These findings suggest that complement system dysregulation may play a role in long-COVID pathogenesis and, more importantly, might help predict the disease in patients currently suffering from acute SARS-CoV-2 infection.

About the study

In the present study, researchers aimed to verify the hypothesis of long-COVID-mediated persistent inflammation being associated with elevated plasma levels of the complement system biomarkers. This, in turn, would allow for the development of tests to predict COVID-19 risk in patients presenting high concentrations of key CRP and cytokines.

The study cohort comprised healthy convalescent controls (n = 79) and long-COVID patients (cases; n = 166), matched for age, ethnicity, infection severity, gender, and vaccination type. All participants had experienced at least one bout of severe COVID-19 infection, with infection status confirmed by clinical molecular evidence. Participants comprised adult (>18 years) men and non-pregnant women with no current alternative disease diagnosis/medication.

Data collection involved ethylenediaminetetraacetic acid (EDTA) preserved blood samples for plasma analysis. Disease and symptom severity were patient-self-reported on a scale of 0-10, with 0 indicating no symptoms (for controls) and 10 indicating the worst possible symptoms. General health was clinically scored on an inverse scale, with 0 indicating poor health and 10 indicating good/normal health. Additionally, demographic and anthropometric measurements were acquired from participants who either self-reported (for demographic) or collected in tandem with blood sample collection.

Immunoassays comprised of enzyme-linked immunosorbent assays (ELISAs) were used to identify and quantify complement proteins, regulators, and activation products. The ELISA method was also used to detect antibodies against RBD, a SARS-CoV-2 spike protein instrumental in the virus' infection capabilities. Finally, hemolytic assays employing sheep erythrocytes pretreated with rabbit anti-sheep erythrocyte antiserum were used to measure classical pathway hemolytic activity.

Study findings

Results from this study verified complement dysregulation in long-COVID cases versus controls. Importantly, markers of complement activation across classical (C1s-C11NH), terminal (MASP1-C11NH), and alternative (iC3b, Ba) pathways were identified as being significantly upregulated in long-COVID patients when compared to normal controls. In contrast, no differences were observed in the lectic pathway. Convalescent samples presented elevated concentrations of iC3b and TCC for up to 21 days following infection termination, suggesting COVID-19's role in complement activation, but these concentrations rapidly declined thereafter.

Analyses of plasma complement components revealed elevated C3, C4, C5, and C9 concentrations, suggesting that long-COVID does cause inflammation via positive phase reactant upregulation. Similarly, C11NH, FD, properdin, clusterin, and FH were upregulated in cases versus controls.

Most notably, nine of the 21 complement products analyzed in this study were found to predict long-COVID. C11NH was the most predictive component, with an area under the curve (AUC) of 0.746. The most accurate predictions were derived from a combination of Ba, C1q, C11NH, C4, C5, properdin, TCC, and FD biomarkers. However, just four activation markers (Ba, iC3b, C5a, and TCC) were sufficient to attain an AUC of 0.785. Given that these markers are easily tested in a clinical setting, these findings highlight a novel tool for identifying and predicting long-COVID patients currently undergoing COVID-19 treatment.

Conclusions

In the present preprint, researchers verified hypotheses of long-COVID-associated inflammatory response arising due to complement system dysregulation. They identified nine complement system biomarkers that could be used to predict long-COVID in the plasma samples of patients currently undergoing COVID-19 treatment. While C11NH was found to have the greatest individual prediction accuracy, it was also detected in plasma samples of convalescent controls for up to 21 days following discharge, and hence cannot be used in isolation.

This study's highlight is identifying four critical complement biomarkers – Ba, iC3b, C5a, and TCC – which can predict future long-COVID with a 78.5% accuracy. Given that these activation markers can easily be measured in most clinical settings, this study forms the basis for future diagnostic tests capable of identifying long-COVID. Furthermore, this study provides insights into the mechanisms underlying long-COVID and may form the basis for future therapeutic interventions to treat patients already suffering from the condition.

*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

Written by

Hugo Francisco de Souza

Hugo Francisco de Souza is a scientific writer based in Bangalore, Karnataka, India. His academic passions lie in biogeography, evolutionary biology, and herpetology. He is currently pursuing his Ph.D. from the Centre for Ecological Sciences, Indian Institute of Science, where he studies the origins, dispersal, and speciation of wetland-associated snakes. Hugo has received, amongst others, the DST-INSPIRE fellowship for his doctoral research and the Gold Medal from Pondicherry University for academic excellence during his Masters. His research has been published in high-impact peer-reviewed journals, including PLOS Neglected Tropical Diseases and Systematic Biology. When not working or writing, Hugo can be found consuming copious amounts of anime and manga, composing and making music with his bass guitar, shredding trails on his MTB, playing video games (he prefers the term ‘gaming’), or tinkering with all things tech.