Association between anti-interferon-alpha autoantibodies and COVID-19 in systemic lupus erythematosus: correspondence on ‘Clinical course of coronavirus disease 2019 (COVID-19) in a series of 17 patients with systemic lupus erythematosus under long-term treatment with hydroxychloroquine’ by Mathian et al

Mathian et al described the clinical course and heterogeneity of COVID-19 in 17 systemic lupus erythematosus (SLE) patients.¹ SLE subjects could be at higher risk of developing COVID-19, with more severe symptomatology and need for hospitalization due to multiple underlying risk factors.^1,2 Type-I interferons (IFN), including IFNα, play fundamental roles in immunity and are crucial in anti-viral responses.³ Defects in IFN signaling pathways, secondary to monogenic inborn errors or to blocking autoantibodies, promote immunodeficiency and recurrent infections.^4,5 Dysregulation in type-I IFN pathway also plays key pathogenic roles in SLE.⁶ A recent report showed association between anti-type-I IFN autoantibodies in 10% of subjects with life-threatening COVID-19 in the general population.⁷ A comprehensive evaluation of multiple anti-cytokine autoantibodies showed presence of anti-type-I IFN autoantibodies in 11% of SLE subjects in the pre-COVID-19 era.⁸ We hypothesized that SLE patients having anti-IFNα autoantibodies at baseline (prior to 2020) may be at higher risk of developing COVID-19, and that the presence of these autoantibodies may help in guiding management and preventive strategies.

Ten SLE females who developed COVID-19 between April 1^st to October 1^st, 2020 were identified among lupus subjects followed at the National Institutes of Health, Bethesda, MD, USA under IRB-approved SLE natural history protocol 94-AR-0066 (Supplemental methods, Table 1). Seven patients had mild to moderate COVID-19 symptoms that were managed at home with supportive care. Three patients had severe symptoms requiring hospitalization, supplemental oxygen, and/or steroids and convalescent plasma infusion. All patients had full recovery. Eight patients were on daily prednisone (range 5–20mg/day) when COVID-19 symptoms developed. Seven patients were taking hydroxychloroquine prior to COVID-19 and continued it during the infection. One patient (Patient 2) had received rituximab in February 2020 and developed COVID-19 in May 2020. Another patient (Patient 9) developed COVID-19 while on belimumab.

Table 1:

Clinical characteristics of SLE subjects with confirmed COVID-19

	Age	Gender	Race	COVID-19 Dx Method	COVID-19 Symptoms	Admission	Tx for COVID-19	Clinical manifestations of SLE/Other Autoimmune disease	Serologies	SLE Medications	Other Comorbidities
1	49	F	C	RT-PCR	cough	No	HCQ Zinc	LN, pleuritis, anemia, lymphopenia, SS	ANA, anti-dsDNA, anti-smith, anti-RNP, ACA, hypocomplementemia	Azathioprine, prednisone (5mg/day)	Obese (BMI 40)
2	48	F	H	Rapid Antigen	SOB, diarrhea	Yes	Oxygen, Convalescent Plasma, Azithromycin	LN, neuropsychiatric lupus, APLS	ANA, anti-dsDNA, anti-SSA, ACA, hypocomplementemia	Prednisone (10mg/day), coumadin, Rituximab infusion on 2/2020	Overweight (BMI 26)
3	40	F	H	RT-PCR	cough, fever, SOB, chest pain	Yes	Oxygen	arthritis, malar rash, pleuritis, alopecia, APLS	ANA, anti-dsDNA, anti-RNP, anti-Smith, anti-SSA, LA, ACA, anti-B2GP1, hypocomplementemia	Prednisone (6mg/day), azathioprine, HCQ, coumadin	Obese (BMI 32)
4	49	F	AA	RT-PCR	fever, chills, cough	No	Supportive care	arthritis, alopecia, LN, anemia, leukopenia, thrombocytopenia	ANA, anti-dsDNA, anti-RNP, anti-smith, anti-SSA, anti-SSB, ACA, hypocomplementemia	Prednisone (7.5mg/day), rivaroxaban (not APLS), MMF, HCQ	Obese (BMI 42)
5	56	F	C	Rapid Antigen	headache	No	Supportive care	malar rash, photosensitivity, alopecia, LN, thrombocytopenia	ANA, anti-SSA, hypocomplementemia	Prednisone (5mg/day), Azathioprine, HCQ	None
6	49	F	H	RT-PCR	fever, chills, headaches, vomiting, diarrhea, loss of smell, sore throat, cough	No	Supportive care	arthritis, alopecia, photosensitivity, ITP	ANA, anti-dsDNA, LA	HCQ	Obese (BMI 30)
7	48	F	H	Antibody	fever, cough, fatigue, myalgias, nasal congestion	No	Supportive care	malar rash, photosensitivity, alopecia, Raynaud’s, arthritis, neuropsychiatric lupus	ANA, anti-dsDNA, anti-smith, anti-RNP, anti-SSA, anti-chromatin, hypocomplementemia	MMF, HCQ	Dyslipidemia, HTN, DVT, PE, obese (BMI 35), ILD
8	48	F	H	RT-PCR	cough, SOB, URI symptoms	Yes	Oxygen, Steroids, Convalescent Plasma	arthritis, Raynaud’s, photosensitivity, alopecia, oral ulcers, malar rash, SS	ANA, anti-dsDNA, anti-SSA, anti-B2GP1, hypocomplementemia	Prednisone (5mg/day), HCQ	Overweight (BMI 28)
9	48	F	H	N/A	fever, cough, fatigue,	No	Supportive care	arthritis, lymphopenia, LN	ANA, anti-dsDNA, anti-RNP, anti-smith, anti-SSA, anti-SSB, LA, ACA, anti-B2GP1, hypocomplementemia	Prednisone (20mg/day), HCQ, MMF, belimumab	None
10	26	F	H	RT-PCR	loss of taste and smell	No	Supportive care	malar rash, alopecia, arthritis, photosensitivity	ANA, anti-dsDNA, anti-RNP, anti-smith, anti-SSA, LA, ACA, hypocomplementemia	Prednisone (5mg/day), HCQ	None

Shaded subjects had anti-interferon α autoantibodies. F=female, C=Caucasian, AA= African-America, H=Hispanic, SLE: systemic lupus erythematosus, APLS: anti-phospholipid syndrome, LN: lupus nephritis, SS: Sjögren’s syndrome, HCQ: hydroxychloroquine, MMF: mycophenolate mofetil, SOB: shortness of breath, ITP: Idiopathic thrombocytopenic purpura, BMI: Body mass index, DVT: deep vein thrombosis, PE: pulmonary embolism, HTN: hypertension, ILD: interstitial lung disease, URI: upper respiratory infection, ANA: anti-nuclear antibody, anti-dsDNA: anti-double stranded DNA, anti-SSA: anti–Sjögren’s-syndrome-related antigen A, anti-SSB: anti–Sjögren’s-syndrome-related antigen B, anti-RNP: anti-ribonucleoprotein, LA: lupus anticoagulant, ACA: anti-cardiolipin antibody, anti-B2GP1: anti-beta2 glycoprotein 1 antibody, N/A: Not available

Biobanked plasma from healthy controls (HC; n=119) and the ten SLE subjects was tested for anti-IFNα IgG autoantibodies by ELISA (Supplemental methods). Values two standard deviations above mean in HC samples were considered positive. Anti-IFNα autoantibodies was detected in 4 out of the 10 SLE patients (patients 2, 3, 9, 10) who developed COVID-19 (40%; Figure 1A). Longitudinal assessments of lupus plasma samples confirmed the presence of anti-IFNα autoantibodies preceding the infection as far back as 2017 (Figure 1A). Patients with anti-IFNα autoantibodies had higher rates of hospitalization requiring oxygen (2 out of 4) compared to those without (1 out of 6). Of the two patients (Patient 2 and 9) who had received anti-B-cell therapy in the prior years, both had persistent anti-IFNα autoantibodies. These results suggest that the prevalence of anti-IFNα autoantibodies is higher in those patients with confirmed COVID-19 than what has been previously reported in SLE.⁸

Figure 1: Presence of blocking autoantibodies to IFNα in SLE subjects.

(A) Bar graph depicts arbitrary units (AU) of anti-IFNα measured by ELISA in 10 SLE subjects who developed RT-PCR- confirmed COVID-19 between April 1^st and October 1^st, 2020. Horizontal dotted line shows 2 standard deviations above mean of 119 healthy controls (55 AU); individual subjects are separated by vertical dotted line, missing plasma samples are represented by X. Plasma samples from Patient 3 and 9 (boxed) had blocking antibodies. (B) Representative example of detection of blocking anti-IFNα. Healthy control PBMCs were incubated with 10% plasma from healthy controls or from autoantibody- positive or negative SLE subjects with COVID-19, and then left unstimulated or stimulated with recombinant human IFNα. IFN-induced phosphorylation of STAT1 was measured by flow cytometry.

We evaluated if the plasma positive for anti-IFNα autoantibodies could block IFNα signaling in vitro (Supplemental methods). Out of the 4 SLE subjects with anti-IFNα autoantibodies, half of the samples (2 subjects; patients 3 and 9) blocked recombinant human IFNα induced signal transducer and activator of transcription 1 (STAT1) phosphorylation in healthy control PBMCs at 10% concentration (Figure 1B). These patients had the highest anti-IFNα autoantibodies titers. None of the COVID-19 SLE plasma samples negative for anti-IFNα autoantibodies (n=6) inhibited STAT1 phosphorylation by rhIFNα.

In this initial assessment, 40% of SLE patients that developed confirmed COVID-19 were positive for anti-IFNα IgG autoantibodies in samples obtained prior to SARS-CoV-2 infection. In general, positive autoantibodies were present several years before and in some patients persisted despite B-cell targeted therapy. Previous reports in the same cohort showed that SLE subjects had anti-IFNα autoantibodies prevalence of 11%.⁸ Therefore, those SLE patients that developed confirmed COVID-19 during this initial wave of the pandemic had enrichment in anti-IFNα autoantibodies. Plasma samples with the highest titers of anti-IFNα autoantibodies inhibited signaling of IFNα in vitro, suggesting that levels of these autoantibodies may affect their blocking ability. A worse outcome in COVID-19 patients positive for anti-IFNα autoantibodies in the general population was recently reported, and suggested that these antibodies may precede infection based on two prestored plasma samples.⁷ Our findings support this hypothesis, as SLE patients who developed confirmed COVID-19 had anti-IFNα autoantibodies detected prior to the infection, suggesting a potential pathogenic role for these autoantibodies in increasing susceptibility to SARS-CoV-2 infection.

Our study is limited by the small sample size. Whether the presence of autoantibodies will contribute to modulating the severity and outcome of the SARS-CoV-2 infection in SLE requires systematic assessment in larger numbers of patients. The natural history of these autoantibodies should also be further evaluated in longitudinal studies.

This report highlights the key role that IFNα and autoantibodies against this cytokine may play in both SARS-CoV-2 infection and in SLE pathogenesis. The presence of anti-IFNα autoantibodies may prove a helpful prognostic marker to predict which SLE patients may develop COVID-19 and could inform preventive measures and management of this subset of patients.