1 HYPOTHESES

An allergic T-helper lymphocyte type-2 (Th-2) pattern of immune response plays a role in protecting against severe COVID-19 infection, but the same immune mechanisms also increase the risks of long-COVID.

This composite hypothesis arises from publications suggesting that increasing Th-2 activity and eosinophil counts are associated with recovery from acute infection and that pre-existing allergic asthma is associated with reduced severity of COVID infection. More recent studies have highlighted disproportionately more patients suffering from long-COVID who also have asthma and allergy. Investigation of the hypotheses could aid understanding of mechanisms and identify targets for novel therapeutic interventions both during acute infection and to prevent later sequalae.

2 BACKGROUND

Numerous studies have highlighted biological risk factors for severe COVID-19 infections beyond those of age and pre-existing medical disorders. These have included neutrophilia, lymphopenia and increases in many pro-inflammatory and coagulation promotors. The one feature that has received rather less attention has been persistent eosinopaenia associated with poorer outcomes from the acute infection.1 Eosinopaenia is a more sensitive marker of severity than lymphopenia. Reduced eosinophil counts have been negatively associated with the duration of ICU admission, mechanical ventilation, oxygen supplementation and CRP levels. In a hospitalized severely affected cohort, lower eosinophil counts were associated with death and recovery with progressively increasing counts. There are credible biological mechanisms to explain these clinical observations.

Eosinophils have properties indicating that they can act as anti-viral agents. These include expression of Toll-Like Receptor-7 (TLR-7), which recognizes single stranded RNA viruses such as coronaviruses; expression of Major Histocompatibility Complex (MHC) molecules so that they can act as antigen presenting cells; the preformed granule pools include Th-1 cytokines such as Interferon gamma (IFN-γ) and Il-12; and they generate super-oxide and nitric oxide. Resident eosinophils, distinct from pro-inflammatory eosinophils, in airways and gut may reduce risks of a hyper-inflammatory state by having regulatory activity. However, it remains to be established whether eosinophils are directly involved with virus control or are merely a marker of increased Th-2 responses, which could counter-balance excessive Th-1 activity thereby reducing COVID induced hyper-inflammatory states.

3 MECHANISMS

Innate immune responses to viral respiratory infections are associated with activation of innate lymphocytic cells with a type-2 phenotype (ILC-2) through infected epithelial cell release of the alarmins, Interleukins (IL-)25,33 and Thymic stromal lymphopoetin (TSLP). ILC-2 release a range of cytokines associated with allergic disease (IL4, 5, 9 and 13), as well as amphiregulin, which stimulates epithelial remodelling. IL-5 has an impact at many levels on eosinophils by inducing eosinophil proliferation, maturation, egress into the circulation and prolonged survival by impairing apoptosis. This immune sequence has been demonstrated in COVID-19 infected patients and associated with more favourable outcomes in those requiring respiratory support.2 The Th-2 cytokines from ILC-2 create a Th-2 biased adaptive immune response to infection and might also be expected to increase pre-existing allergic disease severity. We have anecdotal experience that this may sometimes be the case but further research is required. One of our patients with pre-existing birch pollen allergy and mouth itch when eating fresh apple had an anaphylactic reaction to apple 3 months after a relatively mild COVID-19 infection.

4 ASTHMA, ALLERGY AND RISKS OF COVID INFECTION

There is an evidence from several epidemiological studies that pre-existing asthma is associated with reduced COVID infection severity. A meta-analysis of COVID-19 and asthma from over half a million individuals showed that the risk reduction ratio for acquiring COVID-19 in those with asthma was 14% (RR = 0.86, 95% CI = 0.80–0.94), and the risk reduction ratio for hospitalization with COVID-19 was 13% (RR = 0.87, 95% CI = 0.77–0.99).3 Atopic status as judged by positive skin prick tests in patients with airway allergic disease was associated with reduced risk of severe COVID-19 related pneumonia and higher blood eosinophil count in asthmatics with improved survival.4 Inhaled corticosteroids used in the treatment of persistent asthma could have a protective effect and a recent randomized open label phase 2 trial of twice daily budesonide within 7 days of onset of mild symptoms of COVID-19 infection showed a significant reduction in requirement for subsequent emergency care (p = .004). However, review of publications on this topic has produced conflicting evidence indicating the need for more research.5

Do eosinophils themselves confer protection or is an enhanced Th-2 immune response involved? Reduced Angiotensin Converting Enzyme -2 (ACE-2) gene expression, the key binding molecule facilitating COVID-19 infectivity, has been associated with increased Th-2 gene expression in bronchial brushings from adults with and without asthma. This study also showed a direct correlation between ACE-2 gene expression and Th-17 gene expression.6 In another study of nasal brushings from children with asthma, decreasing levels of epithelial cell expression of ACE-2 expression was associated with increasing degrees of allergic sensitization and higher IgE sensitization.7 In addition, the latter study also showed that Il-13 (a Th-2 cytokine) reduced ACE-2 expression in cultured nasal and bronchial epithelial cells from adults. Nasal and segmental bronchial allergen challenge was followed 48 h later by significant reductions in ACE-2 expression in nasal epithelial brushings in those with allergic rhinitis and bronchial brushings from those with asthma.7 The direct association between allergy/Th-2 activity and reduced ACE-2 provides a compelling explanation for the association between asthma and reduced severity of COVID-19 infection. However, it is not consistent with the benefits of inhaled and systemic corticosteroids.

5 LONG-COVID AND EOSINOPHILS

There is a potential down-side to Th-2 and eosinophilic responses. While eosinophil cytotoxic granule proteins are important for elimination of gut parasites, they are also toxic to host tissues and contribute to the chronic airway remodelling that occurs in more severe asthma. The relatively common occurrence of persistent symptoms after recovery from acute infection known as ‘long-COVID’ is beginning to be characterized, and there is an urgent need to investigate likely mechanisms. One feature of acute and long-COVID is chilblain like lesions known as pernio on fingers and toes. A biopsy study of pernio lesions has highlighted the presence of eosinophilic microthrombi in dermal venules and eosinophils are prominent in several COVID-19 induced skin eruptions. Hyper-eosinophilic diseases are characterized by cardiomyopathy; thromboembolic disease; pulmonary disease; neuropathy and skin lesions (including pernio). Thrombotic events in humans are associated with increased levels of serum ECP as evidence of eosinophil activation. In murine models, eosinophils are seen within experimentally induced thrombi and they have strong thrombin generating capacity through enzymatic lipid oxidation. Long-COVID has very similar features including myocardial injury; thromboembolic consequences; chronic pulmonary disease; neural dysfunction and skin rashes.

6 LONG-COVID AND ASTHMA

We have conducted two follow-up studies of persisting symptoms fulfilling the current definition of long-COVID in previously hospitalized patients in Moscow. The first on 518 hospitalized children has shown an association between long-COVID respiratory and other symptoms (affecting 24.3%) with pre-existing allergic disease defined as asthma, allergic, eczema and/or food allergy OR 1.67 (1.04–2.67).8 This is mirrored by a similar association between asthma and long-COVID (47.1%) in our study of 2,649 hospitalized Russian adults. The odds ratios (OR) and 95% confidence intervals for symptoms associated with asthma were: Neurological 1.95 (1.25–2.98); Mood/behaviour 2.02 (1.24–3.18); Respiratory 1.68 (1.21–2.32).9 A UK/USA/Sweden cohort of 4182 COVID swab positive symptomatic patients found a significantly higher prevalence of pre-existing asthma in the 189 (13.3%) with persistent symptoms 28 days after acute COVID-19 infection compared to those who recovered within 10 days (OR 2.14 [1.55–2.96]). As persistent respiratory symptoms were common the association could have been due to incitement of asthma exacerbations. A Norwegian cohort (n = 312) of mixed hospitalized and home isolated COVID-19 infected patients identified 61% with persistent symptoms 6 months after the acute infection, of whom the highest significant associated co-morbidity was chronic respiratory disease defined as either asthma and/or COPD. The pre-existing chronic respiratory problems were associated with an increased OR fatigue score 1.22 (1.11–1.34) and 2.00 (1.33–3.07) for a higher number of different long-COVID symptoms. This study also showed a significant association between higher convalescent COVID IgG antibodies and both long-COVID fatigue and the number of on-going symptoms.

Could the latter observation point to an auto-immune response in long-COVID? One of the problems being reported amongst children with long-COVID is new onset behavioural problems and tics. This is reminiscent of Paediatric Auto-immune Neuro -psychiatric Disorders Associated with Streptococcal infection (PANDAS).

The diverse definitions of long-COVID and the wide variation in prevalence from epidemiological studies indicates the need for better standardization of diagnosis and core outcomes. However, the consistent association with asthma is striking and the suggestion that auto-immunity may also be involved mandates the need for more mechanistic research. Future studies should record the eosinophil count, total IgE, COVID IgG antibodies and an auto-antibody screen as a minimum component of the blood tests within the data set.

7 CONCLUSIONS

The evidence leads to the generation of hypotheses, which require further investigation. Eosinopenia is a common feature of sepsis, which is suggested to be a ‘simple, convenient, fast and inexpensive biomarker’. What is missing is any consideration of the potential for poor eosinophil responses being causally associated with severe infection. We hypothesize that there is a U-shaped curve of increased morbidity and mortality from COVID-19 infection, which in the acute phase is at least in part a consequence of eosinopaenia; avoidance of hyper-inflammatory states and recovery is due to the regulatory effects of tissue resident eosinophils, enhanced Th-2 activity and regulatory T-cells; and long-COVID is in part a consequence of eosinophilia and/or Th-2 up-regulation (Figure 1). Do eosinophils and/or enhanced Th-2 immune responsiveness confer some degree of protection against acute COVID-19 infection? If so those with other atopic diseases or parasitosis should have less severe infections. Could the lower death rates from COVID-19 in many developing countries be in part due to parasitosis? Does allergy with or without asthma increase the risks of long-COVID? Are eosinophils, mast cells or Th-2 responses involved in the immunopathology of long-COVID? If so trials of interventions to reduce responses would be indicated. Dismissing the associations as epi-phenomena is not acceptable without further investigation. Research is needed establish whether eosinophils and/or Th-2 responses play any part in COVID-19 syndromes, which could lead to identification of targets for treatment. Additional references are available in Supporting information.

Schematic representation of hypotheses implicating asthma and allergy in relation to eosinophilic responses as contributors to the immunopathology of COVID-19 infection from acute presentation to recovery and for post-infection persistent symptoms often known as long-COVID

CONFLICT OF INTEREST

The authors have no direct conflicts of interest related to the subject of the hypothesis paper. JOW and DM are members of the ISARIC paediatric long-COVID working group and have published papers on various aspects of COVID-19 infection. In the last 3 years, JOW has received research grants and bursaries for lectures from Airsonett, Danone/Nutricia; Friesland-Campina.

AUTHOR CONTRIBUTIONS

The original hypothesis was formulated by the JOW. It was modified and the basic immunology clarified by JAW. DM ran the studies in Moscow which generated the evidence for a relationship between asthma, allergy and long-COVID. All the authors worked on the manuscript and approved the submitted version.

Filename Description cea14021-sup-0001-Supinfo.docxWord 2007 document , 16.3 KB Supplementary Material

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

REFERENCES

1Gao Y-D, Ding M, Dong X, et al. Risk factors for severe and critically ill COVID-19 patients: a review. Allergy. 2021; 76: 428- 455. https://doi.org/10.1111/all.14657 2Gomez-Cadena A, Spehner L, Kroemer M, et al. Severe COVID-19 patients exhibit an ILC2 NKG2D+ population in their impaired ILC compartment. Cell Molec Immunol. 2021; 18: 484- 486. 3Sunjaya AP, Alllida SM, Di Tanna GL, Jenkins C. Asthma and risk of infection, hospitalization, ICU admission and mortality from COVID-19: systematic review and meta-analysis. J Asthma. 2021; 9: 1- 14. https://doi.org/10.1080/02770903.2021.1888116 4Schultze A, Douglas I. COVID-19 and inhaled corticosteroids—another piece in an expanding puzzle. Lancet Respir Med. 2021; 9(7): 674- 675. https://doi.org/10.1016/S2213-2600(21)00076-X 5Ferastraoaru D, Hudes G, Jerschow E, et al. Eosinophilia in asthma patients is protective against severe COVID-19 illness. J Allergy Clin Immunol. 2021; 9: 1152- 1162.e3. 6Bradding P, Richardson M, Hinks TSC, et al. ACE2, TMPRSS2, and furin gene expression in the airways of people with asthma—implications for COVID-19. J Allergy Clin Immunol. 2020; 146: 208- 211. 7Jackson DJ, Busse W, Bacharier LB, et al. Association of respiratory allergy, asthma, and expression of the SARS-CoV-2 receptor ACE2. J Allergy Clin Immunol. 2020; 146: 203- 206.e3. https://doi.org/10.1016/j.jaci.2020.04.009 8Osmanov IM, Spiridonova E, Bobkova P, et al. Risk factors for long covid in previously hospitalised children using the ISARIC Global follow-up protocol: a prospective cohort study. Europ Respir J. 2021. https://doi.org/10.1183/13993003.01341-2021 9Munblit D, Bobkova P, Spiridonova E, et al. Incidence and risk factors for persistent symptoms in adults previously hospitalized for COVID-19. Clin Exp Allergy. 2021; 51(9): 1107- 1120. https://doi.org/10.1111/cea.13997