Introduction

Several symptoms typical for autoimmune diseases have been described in patients with SARS-CoV-2 infection, including cutaneous rashes, vasculitis, Raynaud syndrome, central or peripheral neuropathy, myositis, and myocarditis.1,2 The immune system plays a central role in COVID-19 as it is responsible for its clinical manifestations and prognosis of affected patients.3,4 Autoantibodies against type I interferons in patients with life-threatening COVID-19 were reported.5 The principal damage is not caused by SARS-CoV-2 itself but by subsequent immune response triggered by the virus. The damage differs depending on an organ and is mainly detected in the lung and manifests as pulmonary vasculitis.6 Therefore, it would be of interest to investigate whether interstitial lesions in the lungs in the course of SARS-CoV-2 infection could be related to an autoimmune feature. Indeed, there is a consensus statement that provided uniform nomenclature, introducing the term “interstitial pneumonia with autoimmune features,” and a set of classification criteria.7 These criteria include: 1) presence of interstitial pneumonia detected by high-resolution computed tomography and / or surgical lung biopsy; 2) exclusion of alternative etiologies; and 3) lack of typical connective tissue disease according to the diagnostic criteria for a well-defined connective tissue disease. This paper discusses the background behind interstitial pneumonia with autoimmune feature phenotype in patients with COVID-19. Moreover, it offers insight into directions of future research in this area.

Patients and methods

In this study, we spontaneously evaluated serum antinuclear antibodies (ANA) in 15 patients hospitalized for COVID-19 pneumonia (Table 1). None of them suffered from any autoimmune diseases. At the onset of COVID-19, 6 patients had fever, 10 had cough, and 10 had dyspnea. On admission to the hospital, in 13 patients the disease severity was assessed as moderate, while in 2 as severe. Computed tomography showed ground-glass opacities and bilateral pulmonary infiltrates in all patients; crazy paving was observed in 6 patients. Five patients experienced pulmonary embolism during hospitalization. Before the COVID-19 pandemic, none of the patients had samples collected for autoimmunity testing. Moreover, none of the patients had any rheumatic symptoms or conditions independent of ANA screening test results. All hospitalized patients gave written informed consent for diagnostic procedures, including blood tests. The analysis of laboratory results was retrospective and the institutional ethics committee approval was not required.

Table 1. Demographic and clinical characteristics as well as laboratory findings of patients with COVID-19 (n = 15)

Variable

Value

Demographic and clinical characteristics

Age, y

64 (58–72)

Sex

Male

13 (86.7)

Female

2 (13.3)

BMI, kg/m2

26.2 (22.3–33)

Medical history

Autoimmune disease

0

Hypertension

11 (73.3)

Diabetes

2 (13.3)

Symptoms at disease onset

Fever

6 (40)

Cough

10 (66.7)

Dyspnea

10 (66.7)

Computed tomography imaging findings

Ground-glass opacity

15 (100)

Crazy paving

6 (40)

Bilateral pulmonary infiltrates

15 (100)

Findings on admission

Disease severity

Moderate

13 (86.7)

Severe

2 (13.3)

Critical

0

Blood oxygen saturation, %

94 (90–96)

Pulmonary embolism during hospitalization

5 (33.3)

Subsequent autoimmune testing

ANA titer

Not detected

3 (20)

0–160

4 (26.7)

>160

8 (53.5)

Median (IQR) for group >160

1:320 (1:320–1:1280)

c-ANCA (+)

0

p-ANCA (+)

1 (6.7)

Subsequent serologic testing

Anticardiolipin IgG antibodies (+)a

2 (13.3)

6.5 (4.6–10.8)

Anticardiolipin IgM antibodies (+)b

7 (46.7)

13.3 (9.4–24.6)

Anti–β2-glycoprotein I IgG antibodies >20c

15 (100)

1.7 (1.3–1.8)

Anti–β2-glycoprotein I IgM antibodies >20d

1 (6.7)

1.9 (1.5–2.6)

Lupus anticoagulant

Not detected

Laboratory findings on the day of autoimmune testing

White blood cell count, ×103/mm3

8.4 (6.4–10.9)

Platelet count, ×103/mm3

209 (165–338)

Hemoglobin, g/l

12.9 (11–13.9)

Creatinine, μmol/l

70.9 (52.9–84.4)

eGFR <⁠60 ml/min/1.73 m2

1 (6.7)

High-sensitivity cardiac troponin I, ng/l

19.9 (5.3–63.8)

Fibrinogen, g/l

4.3 (3.3–5.5)

D-dimer, mg/l

1.5 (0.6–2.5)

Serum ferritin, μg/l

797 (501–1482)

Procalcitonin, ng/ml

0.02 (0.02–0.18)

High-sensitivity C-reactive protein, mg/l

19.9 (5.3–63.8)

Data are presented as median (interquartile range) or number (percentage).

a Phospholipid unit G

b Phospholipid unit M

c Standard β2 glycoprotein unit G

d Standard β2 glycoprotein unit M

Abbreviations: ANA, antinuclear antibodies; c-ANCA, “classic” antineutrophil cytoplasmic antibodies; BMI, body mass index; eGFR, estimated glomerular filtration rate; Ig, immunoglobulin; p-ANCA, perinuclear antineutrophil cytoplasmic antibodies

Indirect immunofluorescence (IIF) was the first-line tool for ANA screening (ANA1) using the HEp-2 cells (Euroimmun, Lübeck, Germany). Autoantibodies detected on IIF were confirmed by additional specific tests (enzyme-linked immunosorbent assay, line blot ANA3; Euroimmun).8

Statistical analysis

Quantitative data were described as medians and interquartile ranges (IQRs). Patient characteristics were compared using the Fisher exact test for categorical data, and the Mann–Whitney test for continuous data. A P value of less than 0.05 was considered significant. The statistical analysis was performed using GraphPad Prism 9 (GraphPad Software, San Diego, California, United States, www.graphpad.com).

Results and discussion

Approximately 53% of our patients with interstitial pneumonia were positive for ANA1, with an ANA titer greater than 1:160, as detected by IIF. Additionally, 27% of participants were positive for ANA1, with an ANA titer of 1:160. Among patients with ANA titer greater than 1:160, the median (IQR) was 1:320 (1:320 to 1:1280). A line blot assay did not detect autoantibodies typical for various autoimmune rheumatic diseases. We did not find any differences between COVID-19 patients with interstitial pneumonia with autoimmune features and those without.

In more than half of the patients with COVID-19, interstitial pneumonia was associated with a positive autoimmune feature, with none of the cases fulfilling the diagnostic or classification criteria for specific autoimmune diseases.9 It is still unknown whether these autoimmune conditions represent transitory postinfectious phenomena, which we termed “interstitial pneumonia with autoimmune feature phenotype”, in COVID-19 patients or whether autoantibodies are detectable in serum years before the onset of autoimmune disease. Autoantibody seropositivity may also be found in healthy individuals.10 The reason for the higher frequency of ANA positivity in our COVID-19 patients (53%) with interstitial pneumonia as compared with the general population9 is poorly understood. Indeed, up to 20% of healthy women and up to 10% of healthy men are ANA1-positive and most of them will never develop clinical symptoms. It was proposed that a novel immune signature identified in healthy ANA1-positive individuals may protect from T-cell expansion, heightened activation of interferon pathways, and disease transition.11 It was reported that interleukin 6 levels were elevated in healthy ANA1-positive individuals and that increased T-cell count correlated with elevated gene expression in types I and II interferon signaling pathways in patients with systemic lupus erythematosus, which suggests that dysregulation of these cytokines contributes to autoimmune pathogenesis.11 A number of reports describing dysregulation of these cytokines in patients with COVID-19 was published.

The current study is not without limitations and the main one is that it was a single time-point study on consecutive patients. The observation concerning the association of COVID-19 interstitial pneumonia with positive autoimmune features should be confirmed in a larger cohort of COVID-19 patients, and a longitudinal assessment of patients with interstitial pneumonia with autoimmune features is necessary to note changes in this immune profile over time. Small cohort studies should be viewed as observations only rather than recommendations for patient evaluation or treatment.