Article Text

Third COVID-19 vaccine dose with BNT162b2 in patients with ANCA-associated vasculitis
  1. Claudius Speer1,2,
  2. Maximilian Töllner1,
  3. Louise Benning1,
  4. Katrin Klein1,
  5. Marie Bartenschlager3,
  6. Christian Nusshag1,
  7. Florian Kälble1,
  8. Paula Reichel1,
  9. Paul Schnitzler3,
  10. Martin Zeier1,
  11. Christian Morath1,
  12. Wilhelm H Schmitt4,
  13. Raoul Bergner5,
  14. Ralf Bartenschlager3,6,
  15. Matthias Schaier1
  1. 1Nephrology, University Hospital Heidelberg, Heidelberg, Germany
  2. 2Molecular Medicine Partnership Unit Heidelberg, European Molecular Biology Laboratory, Heidelberg, Baden-Württemberg, Germany
  3. 3Department of Infectious Diseases, Virology Heidelberg, Heidelberg, Germany
  4. 4Department of Nephrology and Rheumatology, Kidney Center Weinheim, Weinheim, Germany
  5. 5Department of Internal Medicine A, Clinical Center Ludwigshafen, Ludwigshafen, Germany
  6. 6Division of Virus-Associated Carcinogenesis, German Cancer Research Centre, Heidelberg, Baden-Württemberg, Germany
  1. Correspondence to Dr Claudius Speer, Nephrology, University Hospital Heidelberg, Heidelberg, Germany; claudius.speer{at}

Statistics from

Humoral and cellular immune responses after standard two-dose COVID-19 vaccination are reduced in immunosuppressed patients with antineutrophil cytoplasmic antibodies associated vasculitis (AAV).1–3 Emerging variants such as B.1.617.2 (delta) are of particular concern because of their higher transmissibility and partial immune escape.4 AAV patients with lower neutralising antibody levels may become particularly susceptible to these variants of concern and additional booster vaccination may be required.

We performed a prospective observational study at three different German vasculitis centres to investigate humoral responses against the variant of concern B.1.617.2 after a third vaccine dose with BNT162b2 in 21 patients with AAV on immunosuppressive maintenance therapy. All individuals met the 2017 provisional American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) criteria for AAV. We investigated antispike S1 IgG and surrogate neutralising antibodies a median (IQR) of 23 (21–58) days after standard two-dose COVID-19 vaccination, immediately before a third vaccine dose, as well as a median (IQR) of 21 (21–21) days after third vaccination (online supplemental material). The third vaccine dose was administered a median (IQR) of 103 (72–126) days after second vaccination. In addition, neutralisation activity against B.1.617.2 was analysed in vitro in SARS-CoV-2-infected VeroE6 cells after second vaccination and before and after the third vaccine dose (online supplemental methods).5 Patients were also stratified according to whether or not they had received rituximab treatment as maintenance therapy in the last year. Baseline characteristics and individual immunosuppressive regimens are given in (online supplemental tables S1 and S2).

After second COVID-19 vaccine dose, the median (IQR) anti-S1 IgG index was 1.6 (0.1–3.0) and the median (IQR) per cent inhibition of surrogate neutralising antibodies 34 (31–70; figure 1A). A median (IQR) of 103 (72–126) days after the second vaccine dose, both anti-S1 IgG and neutralising surrogate antibodies decreased to 0.1 (0.1–1.8) and 9 (0–35), respectively, and a third vaccine dose with BNT162b2 was subsequently administered (figure 1A). Anti-S1 IgG and surrogate neutralising antibodies significantly increased to a median (IQR) index of 5.6 (0.5–150) and a median (IQR) per cent inhibition of 56 (4–94) 3 weeks after the third vaccine dose (for both p<0.01; figure 1A). Most importantly, after second vaccination, only 6/16 (38%) patients showed neutralising activity against B.1.617.2 and this number decreased to 3/16 (13%) directly before third vaccination (figure 1B). Even patients with detectable antibodies in commercially available anti-S1 IgG or surrogate neutralising assays had no neutralisation against B.1.617.2. The number of patients with neutralising antibody activity against B.1.617.2 significantly increased to 12/21 (57%) 3 weeks after the third vaccine dose with a median (IQR) ID50 of 40 (0–160) compared with 0 (0–20) after second vaccination and to 0 (0–0) before third vaccination (p<0.05 and p<0.001; figure 1B). Individual courses of anti-S1 IgG, surrogate neutralising and B.1.617.2 neutralising antibodies before and after third vaccination are shown in detail in online supplemental table S3.

Figure 1

Humoral responses after a third COVID-19 vaccine dose with BNT162b2 in patients with ANCA-associated vasculitis (AAV) on maintenance therapy. (A) SARS-CoV-2 anti-S1 IgG and surrogate neutralising antibody levels were measured after the second COVID-19 vaccine dose (N=21 and N=16), immediately before a third vaccine dose with BNT162b2 (N=16) and after third vaccination (N=21) in patients with AAV on maintenance therapy. Anti-S1 IgG antibody levels are shown logarithmically as an anti-S1-IgG index. The dashed red line represents the cut-off for detection. A semiquantitative index of ≥1 was classified as positive. Surrogate neutralising antibodies are given as per cent binding inhibition. A cut-off of <30% binding inhibition (dashed red line) indicates the cut-off for detection of this assay. (B) Titers of neutralising antibodies against the B.1.617.2 (delta) variant were determined in a live virus SARS-CoV-2 infection assay using VeroE6 target cells and serial twofold serum dilutions after the second (N=16), directly before a third (N=16) and after (N=21) a third vaccine dose with BNT162b2. Neutralisation titers refer to the serum dilution that inhibits 50% of the infectivity (ID50). The results of the three different time points in (A) and (B) were compared using Friedman’s test for paired samples with Dunn’s post-test. (C) Humoral responses of patients with AAV who had received a rituximab (monoclonal anti-CD20 antibody) dose <1 year before third COVID-19 vaccination (N=8) were analysed separately with the Mann-Whitney U test. (D) The correlation between the anti-S1-IgG index or the surrogate neutralisation assay and the neutralisation of B.1.617.2 (delta) was examined in patients with AAV using Spearman’s correlation analysis, respectively. ANCA, antineutrophil cytoplasmic antibodies; RTX, rituximab; sVNT, surrogate neutralisation antibodies; *p<0.05; **p<0.01; ***p<0.001.

Patients receiving rituximab maintenance therapy had significantly lower anti-S1 IgG, surrogate neutralising and B.1.617.2 neutralising antibody levels after third vaccination compared with patients not receiving rituximab treatment (online supplemental table S3; figure 1C). Of note, 12/13 (92%) patients without rituximab treatment showed neutralising activity against B.1.617.2, whereas none of those treated with rituximab showed neutralising activity after a third vaccine dose (figure 1C).

Both anti-S1 IgG index and neutralising surrogate antibody activity correlated well with the ID50 value of neutralising B.1.617.2 activity of patients with AAV (figure 1D). However, exceeding the cut-off value for detection in both commercially available assays did not necessarily imply neutralising activity against B.1.617.2 at the same time.

Local adverse events occurred significantly more often after third vaccine dose compared with the first or second vaccination (for both p<0.001; online supplemental figure S1). However, systemic adverse events occurred infrequently after all vaccine doses and no patient experienced a disease flare during follow-up (online supplemental figures S1 and S2).

Consistent with other studies on the immunogenicity of COVID-19 mRNA vaccines in immunosuppressed patients with autoimmune diseases, our data indicate that most individuals have detectable antibody levels in commercially available assays after standard two-dose vaccination, but at significantly lower levels as compared with healthy individuals.1 6 7 Notably, patients treated with rituximab had particularly low seroconversion rates6 7 without detectable neutralising antibody activity against B.1.617.2 in our study. In a first case series on a third vaccine dose in three patients with AAV treated with rituximab, the booster dose was only associated with detectable humoral response in one patient.8 In our study, no patient treated with rituximab in the last year showed neutralising activity against B.1.617.2 after a third vaccine dose. However, in patients with AAV not treated with rituximab, a third mRNA vaccine dose resulted in significantly higher B.1.617.2 neutralisation compared with standard two-dose mRNA vaccination.

Summarised, this study suggests that immunosuppressed patients with AAV may not be adequately protected against B.1.617.2 after standard two-dose COVID-19 vaccination. A third vaccine dose with BNT162b2 induced a strong neutralising antibody activity against B.1.617.2 in most individuals; however, patients receiving rituximab maintenance therapy showed no humoral vaccine response even after a third vaccine dose.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by Ethics Committee of Uniklinik Heidelberg: S-416/2021 Participants gave informed consent to participate in the study before taking part.


The authors thank Iris Arnold and Sabine Bönisch from the Department of Nephrology, Heidelberg University Hospital, Verena Backendorf from the Department of Immunology, Heidelberg University Hospital and Heeyoung Kim from the Department of Infectious Diseases, Molecular Virology, Heidelberg University Hospital for their technical support.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Handling editor Josef S Smolen

  • Contributors Contributed to the manuscript by planning the study: CS, LB, KK and MS, performed the experiments and collected the data: CS, MT, LB, MB, CN, FK, PR and PS, analysis and interpretation of data: CS, MT, LB, PS, CM and MS and preparation and revision of the manuscript: CS, MT, LB, PR, MZ, CM, WHS, RBergner, RBartenschlager and MS. All authors contributed to the article and approved the submitted version.

  • Funding Funding for this study has been received from Dietmar Hopp Stiftung. CS was funded by the Physician Scientist Program of the Heidelberg Faculty of Medicine. LB was funded by the Rahel Goitein-Strauss Program of the Heidelberg Faculty of Medicine. RBartenschlager was supported by the program for surveillance and control of SARS-CoV-2 mutations of the State of Baden-Württemberg, the German Federal Research Network Applied Surveillance and Testing (BFAST) within the Network University Medicine, the DKFZ@fightCOVID initiative and the Helmholtz Association’s Initiative and Networking Fund Project ‘Virological and immunological determinants of COVID-19 pathogenesis—lessons to get prepared for future pandemics (KA1-Co-02 ‘COVIPA’)’.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.