Increased induction of de novo serum ANCA and ANCA-associated vasculitis after mass vaccination against SARS-CoV-2.

Different immune-mediated diseases have been described after SARS-CoV-2 vaccination, with antineutrophil cytoplasmic antibodies (ANCA)-associated vasculitis (AAV) being one of the possible side effects. In this study, a total of 35 patients presented ANCA for the first time during 2021, with the number during 2019 being 15. Twenty-seven out of thirty-five patients developed ANCA after vaccination. Two of them developed these antibodies after receiving the first dose (7.4%), and 25 patients developed ANCA after the second dose of the vaccine (92.6%), with BNT162b2 being the main vaccine received by these patients. In 97.1% of the patients who developed ANCA during 2021, the positivity of ANCA was accompanied by systemic involvement, with renal and respiratory tracts being the main organs affected. Therefore, an increase in the development of AAV has been observed during 2021 in comparison with 2019, which could be due to the administration of SARS-CoV-2 vaccine.

Introduction

Mass vaccination against SARS-CoV-2 has been the most effective strategy to combat against SARS-CoV-2 infection, conferring a 95% protection against COVID-19 and reducing hospitalization and mortality (Mathieu et al., 2021). As of 10 January 2022, 90.4% of Spanish population over 12 years had received complete vaccination schedule against SARS-CoV-2, being BNT162b2 (Pfizer-BioNTech) (69.4%), mRNA-1273 (Moderna) (17.9%), the main vaccines administrated. Likewise, although to a lesser extent, AZD1222 (Oxford-AstraZeneca) (10.2%) and Ad26.CoV2.S (Janssen) (2.5%) vaccines have also been used (“https://www.mscbs.gob.es/profesionales/saludPublica/ccayes/alertasActual/nCov/vacunaCovid19.htm,” n.d.). The mentioned vaccines have been shown to be safe and effective, and the most prevalent short-term side effects have mostly involved injection site reaction. Severe related adverse events have been rare. However, different immune-mediated diseases, including cases of myocarditis or glomerulonephritis (GN) have been reported, antineutrophil cytoplasmic antibodies (ANCA)-associated vasculitis being one of the possible side effects caused by the mass-scale vaccination (Gargano et al., 2021; Sekar et al., 2021). Vasculitis is a disorder characterized by damage of mural structures of blood vessels caused by the infiltration of mononuclear cells in vessel walls. ANCA-associated vasculitis (AAV) is characterized by the development of autoantibodies against antigens in cytoplasmic granules of neutrophils affecting predominantly to small vessels, with myeloperoxidase (MPO) or proteinase 3 (PR3) being the main antigens toward which these autoantibodies are directed. Although the development of ANCA after influenza vaccination has been previously reported, most of the studies that analyze the possible relationship between SARS-CoV-2 vaccination and the development of ANCA and AAV correspond to case reports (Chen et al., 2021; Davidovic et al., 2021; Dube et al., 2021; Feghali et al., 2021; Felzer et al., 2021; Hakroush and Tampe, 2021; Okuda et al., 2021). Likewise, the few published works that include series of patients and review the temporal association between glomerular disease and SARS-CoV-2 vaccination only include very few cases of AAV, as these studies consider the effect of the vaccine on the development of different types of glomerulonephritis, such as immunoglobulin A (IgA) nephropathy, membranous nephropathy, minimal change disease, collapsing glomerulopathy, or lupus nephritis (Caza et al., 2021; Fenoglio et al., 2022; Klomjit et al., 2021). Therefore, to our knowledge, this is the largest study that include patients who debuted with ANCA and AAV during 2021 after receiving vaccination against SARS-CoV-2 without suffering from COVID-19, in comparison with 2019, before the COVID-19 pandemic.

Results

A total of 35 patients presented ANCA and/or anti-GBM antibodies for the first time during 2021, with the number during 2019 being 15. Throughout 2019 and 2021, a similar number of ANCA and/or anti-GBM antibodies determinations were carried out. Specifically, this study was performed in 1,287 patients in 2019 and 1,434 patients during 2021. Monitorization of the mentioned autoantibodies was not taken into account. Consequently, the percentage of patients who debuted with these autoantibodies in 2019 was 1.17%, whereas in 2021 this percentage increased to 2.44%, the differences observed being statistically significant (p = 0.020) (Figure 1). Mean age of the patients included was 65.8 and 63.1 years in 2019 and 2021, respectively. No statistically significant difference on gender was observed. In 2019, the most frequent ANCA was anti-MPO, being present in 80.0% of the patients. However, in 2021, an increase in the positivity for anti-PR3 antibody was observed, which was presented in 45.8% of the patients. Despite nonsignificant differences in the frequency of anti-PR3 antibodies between 2019 and 2021, an important increase of the detection of this autoantibody was observed in 2021 (Table 1). The median titer of anti-MPO antibodies in 2019 was 191.0 CU (interquartile range [IQR] 92.0–718.4 CU) and 115.2 CU (IQR 34.5–689.0 CU) in 2021, with no significant differences between both years (p = 0.481). In addition, the median titer of anti-PR3 antibodies in 2019 and 2021 were 100.0 CU (IQR 62.5–139.5 CU) and 53.2 CU (IQR 30.0–334.3 CU), respectively. No significant differences between both years (p = 0.824) were found.

Figure 1

STROBE flow chart for patient identification and selection

Table 1

Demographic characteristics and laboratory findings of ANCA-positive patients in 2019 and 2021

	2019 (n = 15)	2021 (n = 35)	p
Age, years, mean (SD)	65.8 (14.6)	63.1 (16.4)	0.969
Sex, females, n (%)	6 (40.0%)	19 (54.3%)	0.537
ANCA or anti-GBM Ab, n (%)
Anti-MPO Ab	11 (73.3%)	16 (45.7%)	0.137
Anti-PR3 Ab	2 (13.3%)	15 (42.9%)	0.090
Anti-GBM Ab	1 (6.7%)	2 (5.7%)	0.603
Anti-MPO + anti-PR3 Ab	1 (6.7%)	1 (2.9%)	0.875
Anti-MPO + anti-GBM Ab	—	1 (2.9%)	—

SD, standard deviation; ANCA, antineutrophil cytoplasmic antibodies; MPO, myeloperoxidase; PR3, proteinase 3; GBM, glomerular basal membrane; Ab, antibody.

Only 8 out of 35 patients detected in 2021 developed ANCA before receiving SARS-CoV-2 vaccine. Of the remaining 27 patients, 2 patients developed ANCA after receiving the first dose (7.4%), being BNT162b2 the vaccine administrated in all cases, and 25 patients developed ANCA after the second dose of the vaccine (92.6%), being the vaccination schedule received at that moment: two doses of BNT162b2, two doses of mRNA-1273, two doses of AZD1222, and one dose of Ad26.CoV2.S plus one dose of mRNA-1273, in 87.5%, 4.2%, 4.2%, and 4.2%, respectively. The median time for ANCA detection was 24 days after the first dose (IQR 6–42 days) and 102 days (IQR 46.5–130.5 days) after the second dose of the vaccine.

Four patients included in the present study presented SARS-CoV-2 infection during the follow-up. However, in all four cases the developed of ANCA occurred after vaccination and before the viral infection.

In 97.1% of the patients who developed ANCA during 2021, the positivity of ANCA was accompanied by systemic involvement (Table 2). The activity index following BVAS score of the patients who debuted with AAV during 2021 was 18.8 (SD 6.5) in comparison with those patients who debuted in 2019 in whom a mean of 11.4 (SD 5.2) was observed, with no statistically significant differences (p = 0.741). Renal and respiratory tracts were the main organs affected. Kidney biopsy was carried out in 8 of the patients who presented renal manifestations during 2019 (80%) and 2021 (47%). In the rest of the cases, the kidney biopsy could not be carried out due to the clinical status of the patients. In all of the biopsied patients who debuted with ANCA during 2019 and 2021, the biopsy showed data of pauci-immune extracapillary glomerulonephritis. The data regarding renal function and the need for renal replacement therapy in patients with renal involvement can be seen in Table 3.

Table 2

Clinical characteristics of ANCA-positive patients in 2019 and 2021

	2019 (n = 15)	2021 (n = 35)	p
SARS-CoV-2 infection, n (%)	—	4 (11.4%)	—
Vaccination schedule, n (%)
2 doses BNT162b2 (Pfizer)	—	12 (34.3%)	—
3 doses BNT162b2 (Pfizer)	—	19 (54.3%)	—
2 doses mRNA-1273 (Moderna) + 1 dose BNT162b2 (Pfizer)	—	1 (2.9%)	—
3 doses mRNA-1273 (Moderna)	—	1 (2.9%)	—
2 doses AZD1222 (AstraZeneca) + 1 dose mRNA-1273 (Moderna)	—	1 (2.9%)	—
1 dose Ad26.CoV2.S (Janssen) + 1 dose mRNA-1273 (Moderna)	—	1 (2.9%)	—
ANCA positivity occurs, n (%)
Before vaccination	—	8 (22.9%)	—
After 1st dose	—	2 (5.7%)	—
After 2nd dose	—	25 (71.4%)	—
After 3rd dose	—	—	—
Organ involvement, n (%)
Renal	10 (66.7%)	17 (48.6%)	0.386
Respiratory	3 (20.0%)	21 (60.0%)	0.022
Sinuses	2 (13.3%)	7 (20.0%)	0.872
Ophthalmic	—	4 (11.4%)	—
Neural	1 (6.7%)	—	—
Gastrointestinal	—	2 (5.7%)	—
Cutaneous	—	—	—
Musculoskeletal	1 (6.7%)	4 (11.4%)	1.000
Constitutional syndrome	6 (60.0%)	12 (34.3%)	0.949
None	1 (6.7%)	2 (5.7%)	0.603

ANCA, antineutrophil cytoplasmic antibodies.

Table 3

Renal function in ANCA-positive patients in 2019 and 2021 with renal involvement

	2019 (n = 10)	2021 (n = 17)	p
Serum creatinine (mg/dL), mean (SD)	2.9 (1.6)	4.1 (3.5)	0.331
GFR (mL/min), median (IQR)	19.5 (10.8–42.5)	17.5 (12.8–30.8)	0.763
Proteinuria (mg/24 h), median (IQR)	1,265 (644–2050)	1,365 (320–3588)	0.977
Acute hemodialysis, n (%)	3 (30%)	4 (23.5%)	0.933
Requirement chronic hemodialysis, n (%)	2 (20%)	3 (17.6%)	0.718

GFR, glomerular filtration rate; SD, standard deviation; IQR, interquartile range.

Besides, a significant increase of respiratory manifestations was observed in patients who developed ANCA in 2021 in comparison with those patients who presented ANCA for the first time during 2019 (p = 0.022). Taking this into account, the most frequent respiratory manifestations are depicted in Table 4. Likewise, although in 2019 none of the patients presented ophthalmic or gastrointestinal manifestations, in 2021 11.4% and 5.7% of new ANCA positive patients suffered from ophthalmic and gastrointestinal involvement, respectively.

Table 4

Respiratory manifestations of ANCA-positive patients in 2019 and 2021 with lung involvement

	2019 (n = 3)	2021 (n = 21)	p
ILD, n (%)	2 (66.6%)	12 (57.1%)	0.754
Haemoptysis/alveolar hemorrhage, n (%)	1 (33.3%)	5 (23.8%)	0.722
Nodules or cavities, n (%)	—	2 (9.5%)	—
Others, n (%)	—	2 (9.5%)	—

ILD, interstitial lung disease.

In relation to the treatments used, no differences were observed between both years (Table 5).

Table 5

Treatments administrated to ANCA positive patients in 2019 and 2021

Treatment, n (%)	2019 (n = 15)	2021 (n = 35)	p
Steroid	14 (93.3%)	24 (68.6%)	0.129
Plasmapheresis	3 (20.0%)	3 (8.6%)	0.506
IVIG	3 (20.0%)	4 (11.4%)	0.722
Cyclophosphamide	6 (40.0%)	6 (17.1%)	0.170
Rituximab	2 (13.3%)	5 (14.3%)	0.722
MMF	—	2 (5.7%)	—

IVIG, intravenous immunoglobulin; MMF, mycophenolate mofetil.

Discussion

To date, only a case series including patients who developed new or relapsing glomerulonephritis (Caza et al., 2021; Fenoglio et al., 2022; Klomjit et al., 2021), or case reports of unique patients who presented AAV following different types of SARS-CoV-2 vaccines (Anderegg et al., 2021; Baier et al., 2022; Garcia et al., 2022; Gupta and Ellis, 2022; Ibrahim et al., 2022; Obata et al., 2021; Prabhahar et al., 2022; Prema et al., 2021; Shakoor et al., 2021; Villa et al., 2021), have been published.

Our case series is the largest series reported on new cases of patients with ANCA after SARS-CoV-2 vaccination. Importantly, the finding of ANCA positivity remained specific as in 2019 and was associated with AAV and systemic involvement with a significant predominance of respiratory manifestations. The two most common vaccines currently in use in our country include mRNA-based (BNT162b2 and mRNA-1273) and adenovirus vector (AZD1222 and Ad26.CoV2.S) vaccines.

Nevertheless, the BNT162b2 (Pfizer-BioNTech) vaccine was the one received by most of the patients included in the study. This vaccine was administrated in 88.6% of the patients included in the study and in 88.9% of the patients who developed ANCA after receiving SARS-CoV-2 vaccine. The results obtained could be attributable to the administration of such vaccine. However, because majority of subjects are vaccinated with BNT162b2, it cannot be excluded a secondary effect from the polyclonal activation induced by the vaccine that can include autoreactive clones specific for MPO or PR3. In any case, the ANCA production was observed at the time of clinical debut and clearly associated with AAV development. Thus, it cannot be considered as an epiphenomenon.

Most patients in our series developed ANCA after the second dose of the vaccine (68.6%), but the detection of these autoantibodies varied from 6 days after the first dose to more than 5 months after the second dose.

Unlike expected, in Southern European countries including Spain, the incidence of anti-MPO antibodies usually is higher than anti-PR3 antibodies. However, in the present study a relevant increase in the frequency of anti-PR3 antibodies in 2021 (45.8%) compared with 2019 (20.0%) was observed. There is no clear explanation for such a finding. A possible limitation of the present study could be that ANCA were only tested using chemiluminescent methods for anti-MPO and anti-PR3 antibodies. ANCA were not tested using IIF, so other ANCA specificities different from those mentioned could not be detected.

New-onset autoimmune manifestations following SARS-CoV-2 vaccination are being reported extensively, molecular mimicry, the production of particular autoantibodies, and the role of certain vaccine adjuvants being the main mechanisms through which these vaccines trigger autoimmune phenomena. Previous studies indicate that the use of human papillomavirus, hepatitis B, and influenza vaccines may trigger the onset or exacerbations of autoimmune diseases by molecular mimicry inducing autoimmunity (Segal and Shoenfeld, 2018). However, whether the association between SARS-CoV-2 vaccine and autoimmune manifestations is coincidental or causal remains to be elucidated. Further studies and longer-term follow-up studies are necessary to elucidate the underlying mechanisms and identify the exact cause of these autoimmune manifestations.

Limitations of the study

A possible limitation of the present study could be that ANCA were only tested using chemiluminescent methods for anti-MPO and anti-PR3 antibodies that are the same routine tests in 2019 and 2021. ANCA were not tested using indirect immunofluorescence on human neutrophils, so other ANCA specificities different from those mentioned could not be detected.

STAR★Methods

Key resources table

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed and will be fulfilled by the Lead Contact, Marcos López-Hoyos (marcos.lopez@scsalud.es).

Materials availability

This paper does not generate any new material.

Experimental model and subject details

Human study

All patients from University Hospital Marqués de Valdecilla, a tertiary referral hospital for the Autonomous Community of Cantabria, who debuted with the presence of ANCA (anti-MPO, anti-PR3), and/or anti-glomerular basement membrane (GBM) antibodies during 2021 with suspected vasculitis, discarding the presence of inflammatory bowel disease, and who received vaccination against SARS-CoV-2 were included. This hospital is the unique center that performances the determinations of autoantibodies in the Autonomous Community of Cantabria,with a reference population of 580,000 inhabitants. Likewise, patients who debuted with these autoantibodies in 2019 were also included in order to stablish the corresponding comparisons.

Study approval

The study was conducted according to the guidelines of the Declaration of Helsinki and it was approved by the Regional Ethics Committee (CEIm, internal code 2020.167).

Method details

ANCA and anti-GBM determination

Serum measurement of ANCA and anti-GBM antibodies was performed by chemiluminescence assay using BIO-FLASH (Inova, San Diego, CA) according to the instructions of the manufacturer A value of 20 chemiluminescent units (CU) was stablished as cut-off point to consider positive results for ANCA or anti-GBM antibodies. The measuring range for anti-MPO antibodies was 3.2–739.8 CU, for anti-PR3 antibodies was 2.3–3285.3 CU, and for anti-GBM antibodies was 2.9–1437.8 CU. ANCA were not tested by indirect immunofluorescence (IIF).

BVAS score

Disease activity was determined using the scoring system Birmingham Vasculitis Activity Score (BVAS) (Mukhtyar et al., 2009).

Quantification and statistical analysis

Statistical analysis

Retrospective observational single-center study. Parametric and non-parametric tests were chosen as appropriate for descriptive comparisons of continuous variables, and chi-squared test for categorical variables. Fisher exact test was used for the comparisons in smaller groups. Comparisons were based on the unpaired T-Student test or U-Mann–Whitney U test for parametric and non-parametric continuous data, respectively. A p-value < 0.05 was considered to be significant. All p-values are reported two-sided. Analyses were performed using IBM SPSS Statistics 19.0 (IBM Corp. Armonk, NY, USA).

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Critical commercial assays
QUANTA Flash® MPO	Werfen	701133
QUANTA Flash® PR3	Werfen	701138
QUANTA Flash® GBM	Werfen	701143