Statistics from Altmetric.com
Strategies aimed at expediting immunisation campaigns against COVID-19 include providing single vaccine doses to individuals with previous exposure to SARS-CoV-2 and delaying second doses. While such approaches are effective at the population level, immunogenicity yielded by one dose of vaccines in immunocompromised patients may be alarmingly low.1 2 Biological (b) and targeted synthetic (ts) disease-modifying antirheumatic drugs (DMARDs) interfere with the immune system at multiple levels and may variably reduce response to viral vaccines.3 Limited data on small numbers of rheumatic patients with variable diagnoses and treatments hamper definitive conclusions on the possible impact of immune-mediated inflammatory diseases, immunomodulatory drugs or both on the efficacy of the new generation of mRNA vaccines.4 5
Here we present interim data analysis on the immunogenicity of the BNT162b2 COVID-19 vaccine in 140 patients with chronic inflammatory arthritis all treated with b/tsDMARDs at the Division of Rheumatology of the IRCCS Policlinico San Matteo University Hospital of Pavia, receiving the first dose of vaccine between 24 and 31 March 2021. Patients were advised to discontinue both the b/tsDMARD and concomitant methotrexate around vaccination. In particular, the following suggestions were made: (1) for all the bDMARDs and methotrexate, withholding of therapy in the 7 days before and after vaccination; and (2) for tsDMARDs, withholding of therapy from the day before until day 7 after vaccination. For glucocorticoids and conventional synthetic DMARDs other than methotrexate, no modifications were advised. Blood samples were obtained immediately before vaccination and at day 21 after the first dose. Serum samples were tested using chemiluminescent immunoassay (LIAISON SARS-CoV-2 S1/S2 IgG; DiaSorin) for the quantitative characterisation of SARS-CoV-2 anti-S1 and anti-S2 IgG antibodies, with values >15 AU/mL indicating a positive result. Demographic and clinical variables were retrieved from the last available rheumatological assessment (median (IQR) 14 (5–19) days before vaccination) (table 1). The b/tsDMARD was predominantly an anticytokine therapy (67.1%), followed by CTLA4Ig (21.4%) and Janus kinase inhibitors (8.6%). Treatment also included low-dose glucocorticoids (mean (SD) prednisone dose 3.8 (1.9) mg/day; ≤5 mg/day in 98.6% of the cases) in 38.5% of the patients and conventional synthetic DMARDs (mostly methotrexate) in 56.6%. Arthritis was on average well controlled, with 74.8% of the patients being in low disease activity.
Fifty-five patients (39.3%) were non-responders based on anti-S IgG levels at day 21. As shown in table 1, non-responders were more frequently on methotrexate and/or glucocorticoids; among the different b/tsDMARDs, CTLA4Ig was more common and interleukin-17/23 inhibitors were less common in non-responders. Results were confirmed when patients with a known history of swab-confirmed COVID-19 (n=9) or prevaccine antibody levels indicative of previous SARS-CoV-2 infection (n=11) were excluded (online supplemental table S1). Collectively, seroconversion decreased from 85.4% among patients not receiving neither methotrexate nor glucocorticoids to 33.3% among those on both therapies (figure 1A); such significant trend was confirmed after exclusion of patients with previous COVID-19 (figure 1B). At multivariable analysis, methotrexate and glucocorticoids independently predicted failure to achieve immunogenicity with adjusted ORs (95% CI) of 7.46 (2.88 to 19.33) and 2.69 (1.04 to 5.89), even with the inclusion of patients with previous stimulation by SARS-CoV-2 (online supplemental figure 1S 1C, table S2). In contrast, the effect of CTLA4Ig was restricted to patients with no history of COVID-19 (figure 1C and D and online supplemental table S2). The lower rates of seroconversion observed in patients with RA compared with other arthritis were largely dependent on covariates such as age and type of immunomodulatory treatment. Of note, neither adherence to the recommendations on methotrexate withholding (followed by 50% of the patients) nor the interval of b/tsDMARD discontinuation significantly impacted on the results (online supplemental table S2). The negative impact of methotrexate and glucocorticoids was confirmed in the larger subgroup of patients on tumour necrosis factor inhibitors (online supplemental table S3). Importantly, both drugs also impaired the magnitude of the antibody response among patients who seroconverted (figure 1E and F). The negative association of methotrexate and glucocorticoids with antibody levels was dose dependent (online supplemental table S4). Although the vast majority of the patients was receiving prednisone doses ≤5 mg/day, differences were already seen between the group treated with >2.5 mg/day and the group treated with ≤2.5 mg/day (geometric mean (95% CI) anti-S IgG levels 69.81 (150.95) vs 39.85 (83.82) AU/mL, p=0.07).
Deeper characterisation of memory B cell and T cell responses after each of the two doses of mRNA vaccines is needed to assist the optimal vaccination strategy in rheumatic patients on immunosuppressive treatments. Furthermore, the impact of specific medications, such as those interfering with interferon-driven responses, needs to be more extensively evaluated in larger patient cohorts. Equally important, strategies of methotrexate withholding, alone or in combination with b/tsDMARDs, should be established through randomised clinical trials. Notwithstanding these limitations, the high rate of response (>80%) following a single dose of mRNA COVID-19 vaccine among patients not receiving neither methotrexate nor glucocorticoids found here approaches the immunogenicity reported in registration trials of BNT162b26 and confirms the low impact of most anticytokine therapies on vaccination.3 However, impaired humoral responses associated with methotrexate and glucocorticoids, even at low doses, impose caution before endorsing delayed second dose boosts in rheumatic patients.
Data availability statement
The authors confirm that the data supporting the findings of this study are available within the article. Raw data are available from the corresponding author (SBu) on reasonable request.
Patient consent for publication
The study was approved by the Local Ethical Committee of the IRCCS Policlinico San Matteo Foundation, Pavia, Italy.
We are grateful to our trainees Alessandra Milanesi, Fabio Brandolino, Sofia Chiricolo and Clarissa Rocca for their precious help in recruiting patients and collecting clinical data. We are equally grateful to our nurses Laura Vecchio, Michela Milanesi, Massimo Facchini, Marina Filippa Berlinese, Giovanna Strarosti, Maria Carmela Amoruso and Francesca Clarissa Lorato for their unconditioned assistance in the phases of vaccination and blood sampling. On top, we are grateful to our patients for their enthusiasm in the COVID-19 vaccination campaign.
Handling editor Josef S Smolen
Contributors SBu conceived the work, contributed to the analysis and interpretation of data and drafted the manuscript. LDS conceived the work, contributed to the analysis and interpretation of data, and drafted the manuscript. SBa contributed to the acquisition and interpretation of clinical and laboratory data and revised the manuscript critically. MIG, TL and IM contributed to the collection and handling of serum samples, to the interpretation of data and revised the manuscript critically. IC contributed to the analysis of serum samples, to the interpretation of the results and revised the manuscript critically. LB contributed to the acquisition and interpretation of data and revised the manuscript critically. BD, MdL and EM contributed to the acquisition of clinical data, to the collection of serum samples and revised the manuscript critically. DL contributed to the analysis of serum samples, to the interpretation of the results and revised the manuscript critically. FB contributed to the interpretation of data and revised the manuscript critically for important intellectual content. AM contributed to the analysis and interpretation of data and revised the manuscript critically for important intellectual content. CM conceived the work and revised the manuscript critically for important intellectual content. All the authors provided final approval of the version to be published.
Funding This study was supported in part by fundings from the IRCCS Policlinico San Matteo Foundation, Pavia, Italy.
Competing interests None declared.
Patient and public involvement statement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
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.
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.