The emergence and rapid spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has triggered the coronavirus disease 2019 (COVID-19) pandemic. To date, over 118 million cases have been confirmed, and over 2.6 million have lost their lives. The sheer scale of the outbreak is perplexing, seconded only by the scale of heterogeneity in the clinical presentation of the infection.
A new preprint, released on the bioRxiv* server, discusses the pathogenesis of this condition with reference to autoimmune phenomena involving the key antiviral type I interferon pathway.
The unpredictability of COVID-19
Of every hundred individuals infected with SARS-CoV-2, only about 15% develop symptoms. Less than a tenth become severely ill, as characterized by shortness of breath, pulmonary infiltrates, and low blood oxygen saturation. About a quarter of hospitalized COVID-19 patients finally develop critical disease for which they are intubated and put on mechanical ventilation.
These patients contribute to most deaths due to COVID-19 and occupy most of the time of healthcare workers during the course of their illness. It is thus important to understand how these patients become so sick.
There is a need to disentangle the immunological consequences of SARS-CoV-2 infection and the underlying immunological causes of critical COVID-19, for stratifying patients early in their disease course and for targeting treatment using available or novel therapies.”
Interferon autoantibodies in COVID-19
The role of genes and of pre-existing immunologic dysfunction in the progression of disease following SARS-CoV-2 infection is a vital area of research. For instance, type I interferons (IFNs) are crucial in the earliest antiviral immune responses within the infected host cell.
Inborn errors of these cytokines, or the development of autoantibodies to these interferons, could predispose individuals to severe COVID-19. Research shows that anti-type I IFN autoantibodies are very rare in healthy individuals and absent in cases of asymptomatic COVID-19. However, they are found in a tenth or more of critical COVID-19 patients.
These may be of importance in the causation of critical COVID-19, having been already associated with these individuals before infection and in certain autoimmune, genetic disorders such as autoimmune polyglandular syndrome type 1 (APS-1).
The current study seeks to determine the occurrence of such autoantibodies in patients who do not have critical COVID-19, the levels of such antibodies over time, and their effects on circulating white blood cells.
The researchers also explore the dynamic white blood cell profile throughout this disease, comparing this with the changes seen in patients hospitalized with the same type of respiratory features due to other conditions.
And finally, they examine the potential predictive value of changes in innate and adaptive immunity in patients with and without such autoantibodies, with respect to an adverse outcome in COVID-19.
What were the results?
The study focuses on a cohort of 4,500 individuals, including those with SARS-CoV-2 infection showing various grades of severity, convalescent COVID-19 patients, and a community of mostly asymptomatic individuals.
The researchers found that, of 284 patients with documented infection, a fifth of those with the critical disease had type I IFN autoantibodies, compared to 6% and 0% in severe and moderate disease, respectively. The overall proportion of males in the group was ~70%, but in the critical group with autoantibodies, it was ~80%.
Of the community cohort, about a third each comprised individuals of Caucasian and Hispanic/Latinx origin, respectively, 2% and 9% being Black or Asian/Pacific Islanders, respectively. All were asymptomatic even if they were infected with the virus.
The prevalence of anti-IFN-α2 autoantibodies was only 0.3%, with no sex predilection or preferential ethnicity. Thus, such autoantibodies appear to be present at low frequencies before this infection is acquired.
Convalescent plasma analysis from 175 donors showed the absence of anti-IFN-α2 antibodies. This again turned up zero positive results. This could be good news in the context of the use of convalescent plasma to treat critically ill COVID-19 patients.
White cells in critical COVID-19
The researchers used data from the COVID-19 Multi-Phenotyping for Effective Therapies (COMET) cohort in San Francisco, California. This effort included the longitudinal follow-up of ~70 hospitalized patients by both peripheral blood mononuclear cells (PBMCs) and serum samples.
All these patients were hospitalized with COVID-19 symptoms, but only 54 were positive for the virus. Among the positives, 19 had critical disease, while 18 and 17 had moderate and severe disease, respectively.
Anti-IFN-α2 autoantibodies were recovered from a fifth of critical patients, all from the earliest sample onwards, but none of the others. All the critical patients showed increases in B cells, plasmablasts and classical monocytes, in circulating PBMCs.
CD8+ and gamma delta T cells (Tgd) were found to have decreased frequencies in critical cases. Interestingly, the presence of autoantibodies to IFN-α2 was related to still lower frequencies of these cells, while classical monocytes were higher than in critical COVID-19 patients without these autoantibodies.
These changes are probably not due to illness significant enough to warrant hospitalization but are specific to critical COVID-19, since the differences in the magnitude of change in these cells were significant between critical COVID-19 and other patients negative for the virus but hospitalized for COVID-19-like symptoms.
The cell frequencies tended to normalize over time except for an overshoot by classical monocytes.
The general comparable composition of circulating leukocytes in critical C19+ patients with and without anti-IFN autoantibodies suggests the presence of a broader, conserved mechanism underlying severe disease, such as additional IFN-related pathology particularly in the autoantibody negative patients.”
Impaired ISG expression in critical COVID-19
The researchers observed that type I and type II interferon-stimulated genes (ISGs) were expressed at lower levels in critical COVID-19 on the day of hospitalization compared to moderate or severe disease. The lowest levels were in the four critical patients with autoantibodies, and were comparable to that in healthy individuals.
In all cases of moderate or severe disease, type I ISG expression was high initially but rapidly declined.
The low ISG expression levels were associated inversely with the expression of surface proteins on myeloid cells. One example is the inhibitory leukocyte-associated immunoglobulin-like receptor 1 (LAIR-1) on classical monocytes. This protein showed the highest levels in the four critically ill COVID-19 patients who had autoantibodies to IFN-α2.
These findings may indicate that type 1 ISG suppression by a common underlying mechanism underlies critical COVID-19, irrespective of autoantibody presence to anti-IFN-α2. The absence of such autoantibodies may indicate a defect in the type I IFN response.
Moreover, LAIR-1 may be a specific biomarker of an impaired type I IFN response.
What are the implications?
The study shows the presence of type I IFN autoantibodies in about a tenth of patients with critical COVID-19 across multiple ethnicities, more often in males and with a mean age of 55 years.
These autoantibodies apparently predate infection with SARS-CoV-2 and may indicate that some individuals are at higher risk for critical disease because of their presence. The failure to detect these antibodies in asymptomatic infection or ambulatory patients with COVID-19, as well as convalescent plasma samples, indicates that their presence is universally correlated with critical disease.
The findings of high type I ISG expression early in mild disease with a rapid decline thereafter indicates, along with the impaired type I ISG response in critical COVID-19 patients, that “impaired type I IFN immunity during the first hours and days of infection may account for the protracted disease course including pulmonary and systemic inflammation.”
They suggest a two-step model of critical COVID-19. Here, impaired type I IFN response early in infection permits viral propagation as the first step. This, in turn, leads to the triggering of leukocyte-mediated hyperinflammatory pathways in the lungs and other organs during the second week after infection.
The study also brings out the potential importance of LAIR-1 as a biomarker of critical COVID-19 due to impaired type I IFN responses. In combination with assays for antibodies to IFN-α2, this could allow patients at high risk for critical disease to be identified at admission.
Finally, the existence of antibodies to IFN-α2 may favor treatment with beta-interferons, if these patients are identified early in the disease course. For all these reasons, this study points to the need to identify these patients early enough to improve their outcomes, or even to prevent infection in these individuals.
bioRxiv 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.