Articles

The effect of methotrexate and targeted

immunosuppression on humoral and cellular immune
responses to the COVID-19 vaccine BNT162b2: a cohort
study
Satveer K Mahil*, Katie Bechman*, Antony Raharja, Clara Domingo-Vila, David Baudry, Matthew A Brown, Andrew P Cope, Tejus Dasandi,
Carl Graham, Thomas Lechmere, Michael H Malim, Freya Meynell, Emily Pollock, Jeffery Seow, Kamila Sychowska, Jonathan N Barker,
Sam Norton, James B Galloway, Katie J Doores†, Timothy I M Tree†, Catherine H Smith†

Summary
Background Patients on therapeutic immunosuppressants for immune-mediated inflammatory diseases were Lancet Rheumatol 2021
excluded from COVID-19 vaccine trials. We therefore aimed to evaluate humoral and cellular immune responses to Published Online
COVID-19 vaccine BNT162b2 (Pfizer-BioNTech) in patients taking methotrexate and commonly used targeted July 8, 2021
https://doi.org/10.1016/
biological therapies, compared with healthy controls. Given the roll-out of extended interval vaccination programmes
S2665-9913(21)00212-5
to maximise population coverage, we present findings after the first dose.
See Online/Comment
https://doi.org/10.1016/
Methods In this cohort study, we recruited consecutive patients with a dermatologist-confirmed diagnosis of psoriasis S2665-9913(21)00217-4
who were receiving methotrexate or targeted biological monotherapy (tumour necrosis factor [TNF] inhibitors, *Joint first authors
interleukin [IL]-17 inhibitors, or IL-23 inhibitors) from a specialist psoriasis centre serving London and South East England. †Joint senior authors
Consecutive volunteers without psoriasis and not receiving systemic immunosuppression who presented for vaccination St John’s Institute of
at Guy’s and St Thomas’ NHS Foundation Trust (London, UK) were included as the healthy control cohort. All Dermatology, Guy’s and
participants had to be eligible to receive the BNT162b2 vaccine. Immunogenicity was evaluated immediately before and St Thomas’ NHS Foundation
Trust (S K Mahil PhD,
on day 28 (±2 days) after vaccination. The primary outcomes were humoral immunity to the SARS-CoV-2 spike
A Raharja MBBS, D Baudry MSc,
glycoprotein, defined as neutralising antibody responses to wild-type SARS-CoV-2, and spike-specific T-cell responses T Dasandi MSc, F Meynell MSc,
(including interferon-γ, IL-2, and IL-21) 28 days after vaccination. Prof J N Barker MD,
Prof C H Smith MD), Centre for
Rheumatic Diseases
Findings Between Jan 14 and April 4, 2021, 84 patients with psoriasis (17 on methotrexate, 27 on TNF inhibitors, 15 on
(K Bechman PhD,
IL-17 inhibitors, and 25 on IL-23 inhibitors) and 17 healthy controls were included. The study population had a median Prof M A Brown PhD,
age of 43 years (IQR 31–52), with 56 (55%) males, 45 (45%) females, and 85 (84%) participants of White ethnicity. Prof A P Cope PhD,
Seroconversion rates were lower in patients receiving immunosuppressants (60 [78%; 95% CI 67–87] of 77) than in J B Galloway PhD), Department
of Immunobiology, Faculty of
controls (17 [100%; 80–100] of 17), with the lowest rate in those receiving methotrexate (seven [47%; 21–73] of 15). Life Sciences & Medicine
Neutralising activity against wild-type SARS-CoV-2 was significantly lower in patients receiving methotrexate (median (C Domingo-Vila BSc,
50% inhibitory dilution 129 [IQR 40–236]) than in controls (317 [213–487], p=0·0032), but was preserved in those E Pollock BSc, K Sychowska MSc,
receiving targeted biologics (269 [141–418]). Neutralising titres against the B.1.1.7 variant were similarly low in all Prof T I M Tree PhD),
Department of Infectious
participants. Cellular immune responses were induced in all groups, and were not attenuated in patients receiving Diseases, School of
methotrexate or targeted biologics compared with controls. Immunology and Microbial
Sciences (C Graham MRes,
Interpretation Functional humoral immunity to a single dose of BNT162b2 is impaired by methotrexate but not by T Lechmere MSc,
Prof M H Malim PhD, J Seow PhD,
targeted biologics, whereas cellular responses are preserved. Seroconversion alone might not adequately reflect K J Doores PhD), and Psychology
vaccine immunogenicity in individuals with immune-mediated inflammatory diseases receiving therapeutic Department, Institute for
immunosuppression. Real-world pharmacovigilance studies will determine how these findings reflect clinical Psychiatry Psychology and
effectiveness. Neuroscience (S Norton PhD),
King’s College London, London,
UK
Funding UK National Institute for Health Research. Correspondence to:
Prof Catherine H Smith, St John’s
Copyright © 2021 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 Institute of Dermatology, Guy’s
license. and St Thomas’ NHS Foundation
Trust, King’s College London,
London SE1 9RT, UK
Introduction (TNF), interleukin (IL)-17, and IL-23, are highly effective at catherine.smith@kcl.ac.uk
Immune-mediated inflammatory diseases including attenuating the shared pathogenic immune pathways
psoriasis, rheumatoid arthritis, and inflammatory bowel across immune-mediated inflammatory diseases, but they
disease collectively affect 3–7% of European and North also increase the risk of serious infections, particularly
American populations.1–3 Drugs such as methotrexate, and with respiratory pathogens.4,5 Notably, in many countries,
biologics targeting the cytokines tumour necrosis factor including the UK, individuals with immune-mediated

www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5 1

 Articles
Research in context
Evidence before this study
Individuals with immune-mediated inflammatory diseases
receiving therapeutic immunosuppression were excluded from
COVID-19 vaccine trials; therefore, characterisation of vaccine
efficacy in this vulnerable population is important. Given the
roll-out of extended interval vaccination programmes, the
effect of immunosuppression on the immunogenicity of a
single dose of vaccine is of major public, and personal, health
importance. Methotrexate, but not targeted biological
therapies, can impair serological responses to influenza and
pneumococcal vaccines, but whether these drug-specific effects
can be generalised to COVID-19 vaccines is not known.
We searched PubMed for peer-reviewed studies published up to
April 16, 2021, using the terms “immunosuppression”,
“COVID-19”, “vaccination immune response”, and “vaccine
immunogenicity”, with no language restrictions. Studies on
both humoral and cellular immunogenicity of the COVID-19
vaccine in patients with immune-mediated inflammatory
diseases receiving immunosuppression had not been done.
Early evidence relating to a range of cancer therapies suggests
that both humoral and cellular responses to the first dose of the
COVID-19 vaccine BNT162b2 (Pfizer-BioNTech) are severely
attenuated.
Added value of this study
We evaluated the effect of methotrexate and biologics
targeting tumour necrosis factor, interleukin (IL)-17, and IL-23
on humoral and cellular immune responses to the first dose of
the COVID-19 vaccine BNT162b2 in patients with psoriasis.
Seroconversion rates were lower in patients receiving
immunosuppressants than in healthy controls, with the
inflammatory diseases receiving therapeutic immuno­
suppressants were advised to undertake stringent public
health risk-mitigating measures (shielding) early in the
pandemic due to concerns over drug-related risks of severe
illness from COVID-19.6 Shielding has led to reduced
natural acquisition of protective immunity to SARS-CoV-2,
and substantial psychological, social, and economic costs.
Although mass vaccination promises a return to
normality, there is a paucity of data on vaccine
immunogenicity in patients with immune-mediated
inflammatory diseases receiving immunosuppressants
because they were excluded from trials of COVID-19
vaccines, including the now widely used BNT162b2
(Pfizer-BioNTech) vaccine. This vaccine comprises a lipid-
nanoparticle-formulated nucleoside-modified mRNA
encoding a full-length, stabilised SARS-CoV-2 spike
glycoprotein.7 Studies evaluating immunogenicity to
influenza and pneumococcal vaccines indicate that
methotrexate, but not targeted therapies such as
TNF inhibitors, impairs humoral responses, based on
lower antibody titres in patients receiving this drug than in
healthy controls.8 Successful host protection induced by
2 www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5
lowest rate identified in those receiving methotrexate.
Neutralising titres against wild-type SARS-CoV-2 were also
attenuated in patients receiving methotrexate compared with
healthy controls but were preserved in those receiving
targeted biologics. Neutralising activity against the
B.1.1.7 variant was uniformly low among all participants,
including healthy controls. By contrast, spike-specific T-cell
responses (including interferon-γ, IL-2, and IL-21 production)
were induced in patients receiving methotrexate and all
classes of biologics, and were detectable at similar levels to
those in healthy controls.
Implications of all the available evidence
These findings indicate that seroconversion alone might not
adequately reflect vaccine immunogenicity in individuals with
immune-mediated inflammatory diseases receiving therapeutic
immunosuppression, and caution against routine use of
seroconversion data in isolation in clinical practice. When taking
into account functional humoral immunity and T-cell
responses, our data suggest that targeted biologics do not
impair vaccine responses and provide some reassurance to this
vulnerable population. Notably, although methotrexate
attenuated humoral immunity, cellular responses were
preserved. Real-world pharmacovigilance studies will determine
how these data reflect clinical effectiveness. Findings in relation
to the B.1.1.7 variant indicate that those on
immunosuppression, in common with the general population,
might remain vulnerable after the first dose of BNT162b2,
and therefore risk mitigation (in addition to vaccination)
remains a key strategy in the pandemic.
vaccinations is mediated by a complex interplay between
innate, humoral, and cellular immunity, all of which are
likely to be required for effective and enduring defence
against SARS-CoV-2.9 Thus, there is an urgent need to
determine both functional antibody and antigen-specific
cell-mediated immunogenicity to COVID-19 vaccines in
individuals receiving different types of therapeutic
immunosuppression to inform and, if necessary, revisit
public health policy decisions. Assessment of immune
responses to a single dose of vaccine is particularly
pertinent given the roll-out of extended-interval vaccination
programmes in countries such as the UK to maximise
population coverage.
We studied patients with the immune-mediated
inflammatory disease psoriasis to address this gap in
knowledge, since therapeutic immunosuppression is
generally prescribed as monotherapy, without concurrent
oral corticosteroids.10 We aimed to investigate the
effect of monotherapy with methotrexate, and biologics
targeting TNF, IL-17, and IL-23, on immunogenicity after
the first dose of BNT162b2. We sought to characterise
both humoral and cellular immune responses.

Methods
Study design and participants
In this cohort study, we screened consecutive patients who
presented to a specialist psoriasis centre (Severe Psoriasis
Service, St John’s Institute of Dermatology, Guy’s and St
Thomas’ NHS Foundation Trust, London, UK) serving
London and South East England. To be included in the
study, patients had to have a dermatologist-confirmed
diagnosis of psoriasis and had to be established on
monotherapy with one of the following immuno­
suppressants: metho­trexate, a TNF inhibitor (adalimumab,
infliximab, etanercept, or certolizumab), an IL-17 inhibitor
(ixekizumab or secukinumab), or an IL-23 inhibitor
(guselkumab, risankizumab, or ustekinumab). Volunteers
who did not have a diagnosis of psoriasis and were not
receiving immunosuppressant drugs were also recruited
as a healthy control cohort. All participants had to be
eligible to receive the BNT162b2 vaccine. Given the role of
previous infection on priming vaccine responses11 and the
reported variation in intensity of cellular responses
observed with different COVID-19 vaccines,12 individuals
with past SARS-CoV-2 infection or those who received the
ChAdOx1 nCoV-19 vaccine were excluded from the main
analysis.
This study was approved by the London Bridge
Research Ethics Committee (REC reference 06/
Q0704/18). All participants provided written informed
consent before enrolment.
Procedures
Clinical and safety data and blood samples were collected
at three study visits: baseline (same day of and immediately
before vaccination), 28 days (±2 days) after the first dose of
BNT162b2, and 14 days (±2 days) after the second dose
(which was planned to be administered 12 weeks after the
first dose as per UK Government guidelines). Data on
immune responses after the second vaccine dose will be
reported elsewhere.
Clinical and safety data included demographics,
comorbidities (including psoriatic arthritis), details of
psoriasis (including disease severity measured by the
Psoriasis Area and Severity Index13), and local or
systemic adverse events after vaccination. C-reactive
protein was not assessed in any participants at the
study visits. Plasma and peripheral blood mononuclear
cells were isolated from blood samples as previously
described.14 Details of immuno­ genicity assays are
presented in the appendix (pp 2–3). Humoral immuno­
genicity was based on seroconversion, assessed using
ELISAs for IgG specific for the SARS-CoV-2 spike
glycoprotein, and the functional capacity of participants’
plasma to neutralise both the prototypic (wild-type)
strain of SARS-CoV-2 and the B.1.1.7 variant
(VOC 202012/01 lineage), which has become highly
prevalent in the UK following its emergence in late 2020.
We first identified and excluded individuals with
evidence of previous COVID-19 based on baseline IgG
www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5 3
Articles
reactivity to both the SARS-CoV-2 nucleoprotein and
spike protein.
To assess cellular immune responses to BNT162b2, we
quantified interferon-γ, IL-2, and IL-21 responses to
stimulation with two peptide pools spanning the entire
length of the SARS-CoV-2 spike glycoprotein to ensure we
captured a wider range of vaccine-induced responses than
would be possible by detecting a single cytokine. Receiver
operator characteristic curve analysis was used to establish
threshold values to determine a positive response for the
total T-cell response (combined interferon-γ, IL-2, and
IL-21) and for each individual cytokine. Cytokine responses
to peptides derived from commonly encountered viruses
and fungal antigens were also assessed as controls. Given
the association between T helper (Th) 17 cell activation and
severe COVID-19,15,16 we also analysed IL-17A and IL-22
responses to total spike peptide pools. Direct FluoroSpot
assays were used to detect interferon-γ and IL-2, and IL-17A
and IL-22 (Th17 cytokine profile), and direct ELISpot assays
were used to detect IL-21.
Outcomes
The primary outcomes were humoral and cellular
immunity to the SARS-CoV-2 spike glycoprotein 28 days
(±2 days) after the first dose of BNT162b2. Specifically,
humoral immunity was defined as the presence of
antibodies with neutralising capacity against the spike
glycoprotein of wild-type SARS-CoV-2, and cellular
immunity was defined as the presence of T cells secreting
interferon-γ, IL-2, or IL-21 in response to stimulation
with two peptide pools spanning the entire length of the
SARS-CoV-2 spike glycoprotein.
Secondary outcomes (all of which were measured 28 days
(±2 days) after the first dose of BNT162b2) were antibody
seroconversion rates (presence of IgG antibodies specific
for the spike glycoprotein, without assessment of
neutralising capacity); the presence of antibodies with
neutralising capacity against the spike glycoprotein of the
SARS-CoV-2 pseudotype expressing the spike B.1.1.7 variant;
the presence of T cells secreting IL-17A and IL-22 in
response to stimulation with two peptide pools spanning
the entire length of the spike glycoprotein; safety (adverse
events); and patient-reported worsening of psoriasis.
Adverse events were classified as local or systemic, and
graded as mild (no or minimal interference with activity
and no or minimal medical intervention required),
moderate (limitation in activity and medical intervention See Online for appendix
required), severe (marked limitation in activity and
medical intervention required).
Results following the second dose will be presented
elsewhere, and the primary and secondary outcomes will
remain the same as the current study (but related to the
second dose).
Statistical analysis
Demographic and clinical characteristics, adverse events,
and humoral and cellular immunogenicity were
 Articles

summarised using descriptive statistics. Statistical
imbalance across the treatment groups was calculated with
the Kruskal-Wallis or χ² test. The key exposure measure
was immunosuppressive treatment type, comprising
five mutually exclusive groups: patients receiving
methotrexate, TNF inhibitors, IL-17 inhibitors, or
IL-23 inhibitors, and healthy controls (individuals not
receiving immunosuppressive drugs).
We tested two pre-determined hypotheses: that patients
taking methotrexate would have lower humoral, cellular,
or both humoral and cellular immunogenicity to the first
dose of BNT162b2 compared with healthy controls; and
that patients taking methotrexate would have lower
humoral, cellular, or both humoral and cellular immuno­
genicity to the first dose of BNT162b2 compared with
patients taking targeted biological therapy (TNF inhibitors,
IL-17 inhibitors, or IL-23 inhibitors).
Comparisons were made using linear regression of
log-transformed immunogenicity measures, with robust
estimates of variance to derive 95% CIs and p values. A
5% α level was used for significance testing. No correction
for multiple hypothesis testing was made. Correlations
were derived using Pearson’s method. Data were
analysed in Stata (version 16).
Role of the funding source
The funder of the study had no role in study design, data
collection, data analysis, data interpretation, or writing of
the manuscript.
Results
Between Jan 14 and April 4, 2021, 98 patients with
psoriasis receiving immunosuppressive drugs and
23 healthy volunteers (the control group) were recruited.
We excluded ten participants (six controls, one patient on
methotrexate, and three patients on an IL-23 inhibitor)
from the final analysis on the basis of previous COVID-19
(appendix p 8) and ten individuals because they had been
immunised with the ChAdOx1 nCoV-19 vaccine. The
final analysis included 17 controls and 84 patients with
psoriasis receiving monotherapy with methotrexate
(17 [20%] patients), TNF inhibitors (27 [32%]),
IL-17 inhibitors (15 [18%]), or IL-23 inhibitors (25 [30%]),
all without concurrent glucocorticoids (figure 1). All
participants included in the analysis followed the study
schedule protocol.
The median age of participants was 43 years (IQR 31–52),
56 (55%) participants were male, 45 (45%) were female,
and 85 (84%) were of White ethnicity (table 1). Among the
84 participants with psoriasis, the median duration of
psoriasis was 21 years (IQR 15–33), 20 (24%) had
concurrent psoriatic arthritis, and the median body-mass
index was 28·7 kg/m² (IQR 26·4–33·1). Adherence to
immunosuppression therapy was confirmed with all
participants, and no participants reported pausing or
stopping their immunosuppressants before or during
vaccination. The median dose of methotrexate was 15 mg
(IQR 15–20) per week. The median duration of the current
immuno­ suppressive treatment was 3·3 years
(IQR 1·3–5·3), and participants in all study groups had
well controlled psoriasis, as indicated by their baseline
disease severity. Baseline characteristics disaggregated by
sex are shown in the appendix (p 4).
63 (75%) of 84 patients with psoriasis receiving
immunosuppression therapy reported mild adverse
events within 28 days after the first dose of the BNT162b2
vaccine, compared with 16 (94%) of 17 controls
(appendix p 6). Adverse event data were missing for four
patients. 17 (20%) patients and one (6%) healthy control
reported no adverse effects, and no participants reported

121 participants enrolled

23 healthy controls 98 patients with psoriasis

receiving
immunosuppressive
therapies

6 excluded due to previous COVID-19 14 excluded

4 previous COVID-19
10 vaccinated with ChAdOx1 nCoV-19

17 healthy controls assessed
for single-dose
immunogenicity 28 days
after vaccination
with BNT162b2
17 patients on methotrexate
assessed for single-dose
immunogenicity 28 days
after vaccination with
BNT162b2
27 patients on a TNF
inhibitor assessed for
single-dose
immunogenicity 28 days
after vaccination with
BNT162b2
15 patients on an IL-17 25 patients on an IL-23
inhibitor assessed for inhibitor assessed for
immunogenicity single-dose
28 days after vaccination immunogenicity 28 days
with BNT162b2 after vaccination with
BNT162b2

Figure 1: Study overview

IL=interleukin. TNF=tumour necrosis factor.

4 www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5

 Articles

Healthy controls Patients on Patients on TNF Patients on IL-17 Patients on IL-23 Total (n=101) p value
(n=17) methotrexate inhibitors (n=27) inhibitors (n=15) inhibitors (n=25)
(n=17)
Age, years 34·0 (27·0–46·0) 48·0 (41·0–56·0) 36·0 (28·0–52·0) 45·0 (38·0–49·0) 50·0 (33·0–56·0) 43·0 (31·0–52·0) 0·053*
Sex ·· ·· ·· ·· ·· ·· 0·84†
Male 9 (53%) 11 (65%) 13 (48%) 8 (53%) 15 (60%) 56 (55%) ··
Female 8 (47%) 6 (35%) 14 (52%) 7 (47%) 10 (40%) 45 (45%) ··
Body-mass index, kg/m² 23·0 (21·5–30·5) 27·1 (26·4–29·7) 31·2 (26·2–39·4) 27·6 (25·1–30·9) 28·7 (27·0–33·1) 28·2 (25·1–32·7) 0·032*
Ethnicity ·· ·· ·· ·· ·· ·· 0·72*
White 14 (82%) 13 (76%) 24 (89%) 13 (87%) 21 (84%) 85 (84%) ··
Black 0 1 (6%) 0 0 0 1 (1%) ··
South Asian 3 (18%) 3 (18%) 3 (11%) 2 (13%) 3 (12%) 14 (14%) ··
Mixed 0 0 0 0 1 (4%) 1 (1%) ··
Disease severity measure (Psoriasis Area Severity Index13) ·· 2·2 (1·3–3·5) 1·1 (0·6–2) 0·6 (0–1·8) 1·2 (0–2·8) 1·2 (0·6–2·8)‡ 0·11†
Concomitant psoriatic arthritis ·· 2 (12%) 5 (19%) 8 (53%) 5 (20%) 20/84 (24%) 0·027*
Data are median (IQR), n (%), or n/N (%) unless otherwise specified. Statistical imbalance of the baseline characteristics across the treatment groups is presented by either the Kruskal-Wallis or χ² test.
IL=interleukin. TNF=tumour necrosis factor. *χ² test. †Kruskal-Wallis test. ‡Median for the 84 patients with psoriasis.

Table 1: Baseline characteristics of study participants

moderate or severe adverse events. The commonest local

Anti-SARS-CoV-2 IgG (serological) T-cell vaccine response
adverse effect was pain at the injection site, which was vaccine response
reported by 55 (65%) patients and 14 (82%) controls.
Number of Proportion of Number of Proportion of
Systemic adverse effects were reported by responders responders (95% CI) responders responders (95% CI)
36 (43%) patients and seven (41%) controls. The
Healthy controls 17/17 100% (80–100) 11/16 69% (41–89)
commonest systemic adverse events were fatigue and
Patients on 60/77 78% (67–87) 65/77 84% (74–92)
headache. No participants reported COVID-19 during the immunosuppressants
study. Nine (11%) patients reported that their psoriasis Patients on methotrexate 7/15 47% (21–73) 14/15 93% (68–100)
worsened after vaccination: four were on IL-23 inhibitors,
Patients on TNF inhibitors 19/24 79% (58–93) 19/24 79% (58–93)
two on methotrexate, and three on TNF inhibitors.
Patients on IL-17 inhibitors 15/15 100% (78–100) 14/15 93% (68–100)
Humoral and cellular data were missing for two patients
Patients on IL-23 inhibitors 19/23 83% (61–95) 18/23 78% (56–93)
on methotrexate, three patients on TNF inhibitors, and
A threshold EC50 value of 25 was used for anti-SARS-CoV-2 IgG titres, at which serological responses were classified as
two patients on IL-23 inhibitors. No cellular data were
positive. A threshold value of 30 cytokine-secreting cells per million peripheral blood mononuclear cells was
available for one healthy control. In the control group, established for total T-cell responses (interferon-γ, IL-2, or IL-21), at which the T-cell response was classified as positive.
17 [100%; 95% CI 80 to 100) of 17 participants had evidence EC50=half maximal effective concentration. Humoral and cellular data were missing for two patients on methotrexate,
of seroconversion after vaccination with BNT162b2, three patients on TNF inhibitors, and two patients on IL-23 inhibitors. No cellular data were available for one healthy
control. IL=interleukin. TNF=tumour necrosis factor.
compared with 60 (78%; 67 to 87) of 77 patients receiving
immunosup­pressants (table 2). The lowest seroconversion Table 2: Immunogenicity of the first dose of the COVID-19 vaccine BNT162b2
rate was identified in patients receiving methotrexate (seven
[47%; 95% CI 21 to 73] of 15 patients seroconverted).
Seroconversion in patients receiving TNF inhibitors controls. Median spike-specific IgG titres were
(19 [79%; 95% CI 58 to 93] of 24) and IL-23 inhibitors numerically, but not significantly, lower in patients
(19 [83%; 61 to 95] of 23) was also lower than in controls, receiving methotrexate than in those receiving targeted
but to a lesser extent than in patients on methotrexate. In biological therapy (48 [IQR 25–174], p=0·26).
summary, methotrexate was associated with lower To further interrogate the observed differences in
seroconversion rates than either healthy controls (β=–0·76 serological responses to BNT162b2 between individuals
[95% CI –1·05 to –0·48]; p=0·0001) or targeted biological receiving methotrexate versus targeted immunosup­
therapy (β=–0·52 [–0·77 to –0·26]; p=0·0001). pression, we next assessed the functional effect of
Of those who mounted a serological response, patients seroconversion. Of those who had evidence of neutrali­
with psoriasis receiving immunosuppres­sants had lower sation activity, individuals receiving TNF inhibitors,
median spike-specific IgG titres (half maximal effective IL-17 inhibitors, and IL-23 inhibitors had neutralisation
concentration [EC50] 43 [IQR 25–162]) 28 days after the titres (50% inhibitory dilution [ID50]) against wild-type
first dose of BNT162b2 than controls (101 [55–200]; figure 2; SARS-CoV-2 that were similar to controls (figure 3A). By
appendix p 9). At 28 days after vaccination, patients contrast, patients receiving methotrexate had lower
receiving methotrexate had significantly lower median median neutralisation titres (ID50 129 [IQR 40–236]) than
anti-spike IgG titres (31 [IQR 25–48], p=0·015) than healthy controls (317 [213–487], p=0·0032) or patients

www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5 5

 Articles
104
SARS-CoV-2 spike-specific titres (EC50 [log])
103
102
101
10
Control Methotrexate TNF
group group inhibitor
(n=17) (n=15) group (n=24)
Figure 2: Serological immune responses to COVID-19 vaccine BNT162b2
Spike-specific IgG titres (EC50) in plasma samples on day 28 after vaccination in
healthy controls and patients with psoriasis receiving methotrexate or targeted
biological monotherapy. The circles represent individual values. The red
diamonds and range lines indicate the median and IQR. In the IL-23 inhibitor
group, filled circles represent participants receiving IL-23p19 inhibitors and
hollow circles represent participants receiving an IL-12/23p40 inhibitor.
The horizontal dashed line indicates the seroconversion threshold. EC50=half
maximal effective concentration. IL=interleukin. TNF=tumour necrosis factor.
receiving targeted biological therapy (269 [141–418],
p=0·011).
Neutralisation activity against the B.1.1.7 variant was low
across all study participants after the first dose of BNT162b2
(figure 3B), and there were no differences between controls
and patients receiving immunosup­ pressants (median
ID50 61 [IQR 36–195] vs 54 [38–108], p=0·92).
Although seroconversion was determined by the
presence of IgG binding to spike (without assessment of
neutralising capacity), plasma neutralisation can be
mediated by IgM, IgA, and IgG. Nonetheless, anti-spike
IgG titres positively correlated with neutralisation titres
for both the wild-type and B.1.1.7 strains in each study
group (figure 3C). Neutralisation titres for wild-type and
B.1.1.7 strains were also correlated (appendix p 10).
The frequency of T cells with a Th1 cytokine profile
(characterised by IL-2 and interferon-γ) is widely used
to assess vaccine-induced cellular immunity, including
memory T-cell responses,17 and T follicular helper
(Tfh)-derived IL-21 is crucial for memory B-cell responses.18
Receiver operator characteristic curve analysis established
threshold values to determine a positive response for
the total T-cell response (Th1 or Tfh; a combination of
interferon-γ, IL-2, and IL-21) and for each individual
cytokine (appendix p 11). All study groups showed a wide
range of cytokine responses to peptides derived from
commonly encountered viruses and fungal antigens, with
no major differences in the magnitude or polarisation of
response in any group (appendix p 12).
6 www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5
All study groups showed positive spike-specific T-cell
responses 28 days after the first dose of BNT162b2, as
evidenced by production of interferon-γ, IL-2, or IL-21
(figure 4A; appendix p 13). T-cell response rates were
similar in patients receiving methotrexate and targeted
biologics and in controls (table 2). In regression analyses,
methotrexate was not associated with lower cellular
immunogenicity compared with healthy controls (β=0·65
[95% CI 0·07 to 1·24], p=0·029) or targeted biological
therapy (β=0·12 [–0·28 to 0·53], p=0·53). The magnitude
of T-cell responses was also similar between all study
groups, both with respect to the combined interferon-γ,
IL-2, and IL-21 response (figure 4A) and for individual
cytokines (appendix p 14).
T-cell reactivities were also evident in serological non-
responders (individuals without evidence of serocon­
IL-17 IL-23 version) who were receiving methotrexate (seven of eight),
inhibitor inhibitor
group group
TNF inhibitors (three of five), and IL-23 inhibitors
(n=15) (n=23) (one of four). Thus, T-cell responses were successfully
induced following vaccination in patients receiving either
methotrexate or targeted biologics, with similar levels of
response across treatment groups, independently of
detectable humoral immunogenicity.
Given the central role of Th17 immunity in the
pathogenesis of psoriasis, the therapeutic efficacy of
targeted IL-17 blockade,10 and the association between Th17
cell activation and severe COVID-19,15,16 we next analysed
IL-17A and IL-22 cytokine responses to total spike peptide
pools across the study groups. IL-17A and IL-22 responses
were low across all four immunosuppressant groups, and
similar to responses seen in controls (figure 4B).
Collectively, interferon-γ, IL-2, and IL-21 T cell responses
(in particular, interferon-γ) broadly correlated with
serological and functional humoral immune responses
(figure 4C). By contrast, no correlation was observed
between Th17 type responses and humoral immunity.
However, there was variation in the strength of correlation
across cytokines and drug categories (appendix p 15).
Finally, we assessed humoral and cellular immune
responses of four individuals receiving immuno­
suppression therapy (one on methotrexate and three on
IL-23 inhibitors) who received the first dose of BNT162b2
but were excluded from the main study due to previous
confirmed natural SARS-CoV-2 infection. On analysis of
their humoral immunity, there was evidence of robust
serological and neutralising responses against both wild-
type SARS-CoV-2 and the B.1.1.7 variant 28 days after
vaccina­ tion (appendix p 16). Similarly, robust cellular
immune responses were evident at day 28 (appendix p 16).
These humoral and cellular immune responses exceeded
that of our infection-naive study participants receiving the
same immunosuppression therapy (and controls).
Discussion
This study characterises the humoral and cellular immune
response to the first dose of BNT162b2 vaccine in
individuals receiving monotherapy with methotrexate,
 Articles

A B
104 104

B.1.1.7 SARS-CoV-2 variant neutralisation titre (ID50 [log])

Wild-type SARS-CoV-2 neutralisation titre (ID50 [log])

103 103

102 102

101 101

10 10
Control group Methotrexate TNF inhibitor IL-17 inhibitor IL-23 inhibitor Control group Methotrexate TNF inhibitor IL-17 inhibitor IL-23 inhibitor
(n=17) group group group group (n=17) group group group group
(n=15) (n=24) (n=15) (n=23) (n=15) (n=24) (n=15) (n=23)

C
Control group (n=17) Methotrexate group (n=15) TNF inhibitor group (n=24) IL-17 inhibitor group (n=15) IL-23 inhibitor group (n=23)
r=0·57, 95% CI 0·11 to 0·83 r=0·79, 95% CI 0·46 to 0·93 r=0·87, 95% CI 0·72 to 0·95 r=0·51, 95% CI –0·00 to 0·81 r=0·91, 95% CI 0·81 to 0·96
105
Wild-type SARS-CoV-2 neutralisation

104
titre (ID50 [log])

103

102

101

10

r=0·31, 95% CI –0·22 to 0·70 r=0·89, 95% CI 0·67 to 0·97 r=0·90, 95% CI 0·75 to 0·96 r=0·51, 95% CI –0·06 to 0·83 r=0·92, 95% CI 0·83 to 0·97
105
B.1.1.7 SARS-CoV-2 variant neutralisation

104
titre (ID50 [log])

103

102

101

10
101 102 103 104 105 101 102 103 104 105 101 102 103 104 105 101 102 103 104 105 101 102 103 104 105
Spike-specific IgG (EC50 [dilution]) Spike-specific IgG (EC50 [dilution]) Spike-specific IgG (EC50 [dilution]) Spike-specific IgG (EC50 [dilution]) Spike-specific IgG (EC50 [dilution])

Figure 3: Functional humoral immunogenicity of the COVID-19 vaccine BNT162b2

(A) Neutralisation titres against wild-type SARS-CoV-2 on day 28 after the first dose of BNT162b2. (B) Neutralisation titres against B.1.1.7 SARS-CoV-2 variant on day 28 after the first dose of BNT162b2.
The circles represent individual values. The red diamonds and range lines indicate the median and IQR. In the IL-23 inhibitor group, filled circles represent participants receiving IL-23p19 inhibitors and
hollow circles represent participants receiving an IL-12/23p40 inhibitor. The horizontal dashed line indicates neutralisation activity threshold. (C) Correlation between spike-specific IgG and
neutralisation titres against wild-type or the B.1.1.7 variant. Blue diamonds show individual patients, shading indicates the 95% CI. EC50=half maximal effective concentration. ID50=50% inhibitory
dilution. IL=interleukin. TNF=tumour necrosis factor.

TNF inhibitors, IL-17 inhibitors, or IL-23 inhibitors. vaccine immunogenicity. We show that seroconversion
Our homogeneous cohort of individuals receiving and neutralising capacity against the wild-type SARS-
immunosuppressant monotherapy for psoriasis, without CoV-2 strain is lower in patients receiving methotrexate
concurrent corticosteroids, has enabled the interrogation than in those receiving targeted biological therapy or
of drug-specific (rather than disease-specific) effects on healthy controls. Neutralising titres against the

www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5 7

 Articles

B.1.1.7 variant were low for all groups. Notably, however,

A Combined interferon-γ, IL-2, and IL-21 responses cellular immunity following BNT162b2, including Th1 and
104 Tfh responses, was shown across all classes of
immunosuppression and was similar to that of controls
and also similar to levels observed in other control
cohorts.11,14
Cytokine-secreting cells per 106 PBMCs

103 Although our findings on vaccine immunogenicity in

the context of targeted biologics are encouraging, the
results from patients treated with methotrexate highlight
a disparity between humoral and cellular immuno­
102
genicity. The clinical significance of this finding warrants
further research. Our study builds on recent research,
which has primarily assessed only serological responses
to COVID-19 vaccines in immunosuppressed patients,
101
without considering neutralising antibody responses
Figure 4: Cellular
(the gold standard measure of humoral immunogenicity)
immunogenicity of the
COVID-19 vaccine BNT162b2 or cellular immunity.19–22 Consistent with our data
(A) Total T-cell response, as 10 indicating lower levels of spike-specific IgG in patients
determined by combined receiving TNF inhibitors versus controls, a study of
interferon-γ, IL-2, and IL-21 B
responses to stimulation with
IL-17A and IL-22 responses 1293 patients with inflammatory bowel disease showed
104 lower antibody titres and seroconversion rates following
peptides from total spike
peptide pools, reported as a single dose of vaccine in those receiving TNF blockade
number of cytokine-secreting (infliximab) than in a reference cohort of patients
cells per 10⁶ cells in PBMC
receiving vedolizumab (gut-selective integrin inhibitor).19
Cytokine-secreting cells per 106 PBMCs

samples on day 28 after the 103

first dose of BNT162b2. Multivariable models highlighted an association of age
The horizontal line indicates of 60 years or older and co-therapy (azathioprine,
the T-cell response threshold. mercaptopurine, or metho­ trexate) with lower
(B) IL-17A and IL-22 responses
to stimulation with peptides
102 SARS-CoV-2 antibody titres; however, neutralising titres
from total spike peptide pools, or cellular immunity (which were similar in our TNF
reported as number of inhibitor cohort versus controls) were not explored.
cytokine-secreting cells per Serological responses to a single dose of vaccine were
10⁶ cells in PBMC samples on 101
day 28 after the first dose of
also lower in recipients of solid organ transplants
COVID-19 vaccine BNT162b2. receiving anti-metabolite immunosuppression (myco­
The horizontal line indicates phenolate mofetil, mycophenolic acid, or azathioprine)
the T helper 17 response 10 than in those not receiving such immuno­suppression.22
threshold. The circles represent
individual values. The red Finally, both cellular and humoral immunogenicity of a
Control Methotrexate TNF IL-17 IL-23
diamonds and range lines group group inhibitor inhibitor inhibitor single dose of vaccine was low in immunocompromised
indicate the median and IQR. (n=16) (n=15) group (n=24) group (n=15) group (n=23) individuals with solid and haematological cancers.14 This
In the IL-23 inhibitor group, finding contrasts with our data showing intact cellular
filled circles represent
participants receiving C responses in immuno­­ suppressed individuals with
IL-23p19 inhibitors and hollow immune-mediated inflammatory diseases and might be
Combined interferon-γ,
Spike-specific IgG EC50

circles represent participants attributable to the effect of the underlying malignancy

receiving an
IL-17A and IL-22

and type or burden of therapeutic immunosuppression.14

IL-2, and IL-21
Wild-type ID50

IL-12/23p40 inhibitor.
Interferon-γ

In our dataset, the first dose of BNT162b2 induced

B1.1.7 ID50

(C) Spearman correlation

between T-cell responses favourable T-cell responses (interferon-γ, IL-2, and IL-21)
IL-21
IL-2

(cytokine-secreting cells per to total spike peptide pools in the majority of controls
1·0
10⁶ PBMCs) and humoral IL-17A and IL-22
0·8 and patients receiving immuno­suppressants, with no
immune responses as
determined by ELISA and Spike-specific IgG EC50 0·6 attenuation in cellular immunity found in those
neutralisation assays. 0·4 receiving methotrexate or targeted biological mono­
Wild-type ID50
The colour scale indicates the therapy compared with controls. Further­more, T cells
0·2
Spearman R values; all p values B1.1.7 ID50
0 from patients receiving methotrexate or targeted
are less than 0·01. EC50=half
maximal effective IL-2
–0·2
biologics who were serological non-responders
concentration. IL-21 –0·4
demonstrated T-cell reactivities following the vaccine,
ID50=50% inhibitory dilution. thus highlighting a disparity between humoral and
Interferon-γ –0·6
IL=interleukin.
PBMCs=peripheral blood Combined interferon-γ, –0·8
cellular responses.
mononuclear cells. IL-2, and IL-21 –1·0
The true in-vitro correlates of immune protection after
TNF=tumour necrosis factor. vaccination, and the relative importance of humoral and

8 www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5


cellular immunity, are unknown. So far research on
responses to the COVID-19 vaccine has overwhelmingly
focused on serological responses, and there is a wide
variation in laboratory methodology used (limiting
comparability of assays) and no indication of the level of
antibody required for protection (ie, clinical effective­
ness). Furthermore, the importance of cellular immune
responses in preventing SARS-CoV-2 transmission in the
context of diminished humoral immunity is still unclear.
Vaccine-induced cell-mediated immunity might be a better
surrogate than humoral immunity for protection against
respiratory viruses such as influenza in patients with
compromised immune responses.23,24 Cellular immunity
has been shown to be crucial for clearance of SARS-CoV-2
and recovery from COVID-19.25,26 SARS-CoV-2-specific
CD4 and CD8 T-cell responses were observed in patients
with COVID-19,27 and memory B cells and Tfh cells were
identified following recovery. T cells recognising peptides
derived from the spike glycoprotein, nucleoprotein, and
matrix of SARS-CoV-2 were found in the convalescent
phase, and low numbers of T cells, T-cell exhaustion, and
increased levels of pro-inflammatory cytokines are
hallmarks of severe COVID-19. Thus, cellular responses
after vaccination might affect overall vaccine efficacy
against SARS-CoV-2. Given the known effect of
methotrexate and biological therapies on T-cell
proliferation, activation, and cytokine production, our
data on cell-mediated responses after vaccination are
encouraging.
Our findings support existing data showing that
methotrexate attenuates humoral immunity; metho­
trexate is used therapeutically to prevent anti-drug
antibody development and it decreases vaccine serological
responses to seasonal influenza (particularly novel strains)
and pneumococcus.8,28 Trial data in rheumatoid arthritis
indicate that temporary methotrexate discontinuation for
2 weeks after influenza vaccination might increase sero­
protection rates compared with continuing methotrexate.29
However, only humoral immunogenicity was assessed,
and the clinical significance of higher antibody responses
on the incidence of influenza infections is unknown.
Registry data suggest that methotrexate or biological
monotherapy for immune-mediated inflammatory
diseases is not associated with adverse COVID-19
outcomes,30,31 so it is unclear whether the reduced humoral
responses to BNT162b2 in the context of methotrexate
translate to an increased SARS-CoV-2 infection risk.
These data also highlight variation in vaccine immuno­
genicity against different variants of SARS-CoV-2. Novel
variants harbouring mutations in the spike glycoprotein
might evade vaccine-induced protective immunity,
and the highly transmissible B.1.1.7 variant harbouring
nine amino acid changes in the spike protein has been
shown to have increased resistance to neutralisation by
antibodies generated following vaccination or
infection.32–34 While assessing responses following the
second vaccine dose is vital, we identified low
www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5 9
Articles
neutralisation titres against B.1.1.7 after vaccination in
both controls and patients with psoriasis receiving
immunosuppressants, which underlines the need for
ongoing risk mitigating measures following the first dose
of vaccine. Emerging evidence indicates that neutralising
antibody titres against B.1.1.7 generated by vaccination
might not correlate with clinical vaccine effectiveness
against symptomatic COVID-19,35 so the contribution of
cellular immunity (T-cell and natural killer cell responses
and antibody-dependent cellular cytotoxicity or
phagocytosis) in vaccine-induced protection against new
lineages of SARS-CoV-2 warrants investigation.
Our study addresses several limitations of existing
datasets. We have interrogated the effect of different
immunosuppressant treatment types on both humoral
and cellular vaccine immune responses in a relatively
young and homogeneous study cohort. Our study has
limited confounding since participants had the same
underlying immune-mediated inflammatory disease
(psoriasis) with low to absent disease activity, and they
were not receiving combination therapy with other
immunosuppressants or glucocorticoids. The major
classes of biologics used across immune-mediated
inflammatory diseases were included, in addition to
the widely prescribed first-line immunosuppressant
methotrexate. Limitations of our study include a limited
numbers of participants and absence of immunogenicity
data following the second (booster) dose of vaccine.
Although the number of participants was small, study
groups were well matched with respect to baseline
demographics and multiple components of the immune
response are profiled, which will inform future larger-scale
studies. Longer-term data are not presented since this
interim analysis focused on characterising immune
efficacy of the priming dose of the vaccine. Encouragingly,
the robust humoral and cellular responses observed
following the first dose of BNT162b2 in the four patients
receiving immunosuppression with past SARS-CoV-2
infection indicates infection priming of immune
responses. This is in keeping with observations in
immunocompetent individuals.11 Important follow-up
immunogenicity data following the (extended 12 week
interval) booster dose of vaccine in our infection-naive
participants receiving immunsuppression are anticipated.
While global mass COVID-19 vaccination programmes
are underway, there remains concern over vaccine efficacy
in immunosuppressed patients, including against novel
SARS-CoV-2 variants that threaten immune escape.
Measures of the immune response that correspond to risk
of COVID-19 after vaccination are currently unknown and
emerging research in immunosuppressed patients has
focused on serological responses alone. We show
that serological responses are not representative of the
complex immune response to vaccines, and thus might
have limited clinical utility when considered in isolation.
Although our data indicating preserved cellular
immunogenicity across biological classes and
 Articles
methotrexate, independent of neutralising activity, are
reassuring, ongoing pharmacovigilance studies will
determine whether this finding translates into clinical
effectiveness of vaccines. The magnitude and quality of
T-cell responses will also be affected by further vaccine
doses, so analysing cellular dynamics following booster
doses is important. Nonetheless, the observed attenuated
functional humoral immunogenicity of the first dose of
BNT162b2 in individuals receiving methotrexate and low
neutralising activity against B.1.1.7 in all study participants
indicate a reduced time interval between initial and booster
vaccine doses might be favourable. These data underscore
the need for ongoing stringent risk mitigating measures
after a first dose of vaccine to reduce exposure to
SARS-CoV-2.36
Contributors
SKM, CHS, TIMT, KJD, MAB, JBG, and JNB contributed to study
design. SKM, CHS, TIMT, and KJD developed the study protocol.
DB, AR, CD-V, CG, TL, EP, JS, KS, KJD, and TIMT contributed to
sample preparation and performed the assays. SKM, CHS, AC, TIMT,
MHM, KJD, MAB, JBG, and JNB contributed to clinical interpretation of
study findings. KB, SN, and JG proposed statistical analysis methods
and undertook the analysis. FM, SKM, AR, and TD contributed to all
aspects of governance and study delivery. TD and AR contributed to data
capture methods. SKM, CHS, TIMT, KJD, JBG, KB, JNB, MAB, MM,
CD-V, AC, and SN contributed to drafting and editing the final report.
TIMT, KJD, JBG, and KB were responsible for verifying the underlying
data. CHS and JBG secured the funding for the study. All authors had
full access to all the data in the study and had final responsibility for the
decision to submit for publication.
Declaration of interests
CHS reports grants from AbbVie, Sanofi, Novartis, and Pfizer, outside
the submitted work. JBG reports personal fees from AbbVie, Sanofi,
Novartis, Pfizer, Janssen, and UCB, and grants from Eli Lilly, outside the
submitted work. SKM reports departmental income from AbbVie,
Celgene, Eli Lilly, Janssen-Cilag, Novartis, Sanofi, and UCB, outside the
submitted work. MAB reports departmental income from Clementia,
Eli Lilly, Ipsen, Novartis, Pathios Therapeutics, Regeneron, and UCB,
outside the submitted work. JNB reports grants and personal fees from
AbbVie, Novartis, Lilly, and J&J, outside the submitted work. SN reports
personal fees from Pfizer and Janssen, outside of the submitted work.
APC reports grants from Bristol Myers Squibb and Janssen, and speaker
bureau and consultancy fees from AbbVie, Bristol Myers Squibb, and
UCB. TIMT reports grants from GlaxoSmithKline, Sanofi, and Imcyse,
and consultancy fees from GlaxoSmithKline, Novartis, UCB, and
Quell Therapeutics, outside the submitted work. All other authors
declare no competing interests.
Data sharing
The authors are happy to share data on request to the corresponding
author.
Acknowledgments
This research was funded by the UK National Institute for Health
Research (NIHR) Biomedical Research Centre based at Guy’s and
St Thomas’ NHS Foundation Trust and King’s College London and the
NIHR Clinical Research Facility. The views expressed are those of the
authors and not necessarily those of the UK National Health Service,
the NIHR, or the UK Department of Health. This work was supported
by a grant from the Psoriasis Association to CHS; a King’s Together
Rapid COVID-19 Call award to MHM and KJD; Fondation Dormeur,
Vaduz for funding equipment to KJD; a Huo Family Foundation Award
to MHM and KJD; a grant from the Chronic Disease Research
Foundation (CDRF-22/2020) to MAB, MHM, and KJD; part of the
EDCTP2 programme supported by the European Union
(RIA2020EF-3008 COVAB); UK Medical Research Council (MRC)
through the Genotype-to-Phenotype UK National Virology Consortium
(MR/W005611/1) to MHM and KJD; and a grant from Isaac Schapera
10 www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5
Research Trust to AR. CG was supported by the MRC-KCL Doctoral
Training Partnership in Biomedical Sciences (MR/N013700/1). SKM is
funded by an MRC Clinical Academic Research Partnership award
(MR/T02383X/1).
References
1 Safiri S, Kolahi AA, Hoy D, et al. Global, regional and national
burden of rheumatoid arthritis 1990–2017: a systematic analysis of
the Global Burden of Disease study 2017. Ann Rheum Dis 2019;
78: 1463–71.
2 Mehrmal S, Uppal P, Nedley N, Giesey RL, Delost GR. The global,
regional, and national burden of psoriasis in 195 countries and
territories, 1990 to 2017: a systematic analysis from the Global
Burden of Disease Study 2017. J Am Acad Dermatol 2021;
84: 46–52.
3 GBD 2017 Inflammatory Bowel Disease Collaborators. The global,
regional, and national burden of inflammatory bowel disease in
195 countries and territories, 1990–2017: a systematic analysis for the
Global Burden of Disease Study 2017. Lancet Gastroenterol Hepatol
2020; 5: 17–30.
4 Kalb RE, Fiorentino DF, Lebwohl MG, et al. Risk of serious
infection with biologic and systemic treatment of psoriasis: results
from the Psoriasis Longitudinal Assessment and Registry
(PSOLAR). JAMA Dermatol 2015; 151: 961–69.
5 Yiu ZZN, Ashcroft DM, Evans I, et al. Infliximab is associated with
an increased risk of serious infection in patients with psoriasis in
the U.K. and Republic of Ireland: results from the British
Association of Dermatologists Biologic Interventions Register
(BADBIR). Br J Dermatol 2019; 180: 329–37.
6 Williamson EJ, Walker AJ, Bhaskaran K, et al. Factors associated
with COVID-19-related death using OpenSAFELY. Nature 2020;
584: 430–36.
7 Polack FP, Thomas SJ, Kitchin N, et al. Safety and efficacy of the
BNT162b2 mRNA Covid-19 vaccine. N Engl J Med 2020; 383: 2603–15.
8 Hua C, Barnetche T, Combe B, Morel J. Effect of methotrexate, anti-
tumor necrosis factor α, and rituximab on the immune response to
influenza and pneumococcal vaccines in patients with rheumatoid
arthritis: a systematic review and meta-analysis. Arthritis Care Res
2014; 66: 1016–26.
9 Jeyanathan M, Afkhami S, Smaill F, Miller MS, Lichty BD, Xing Z.
Immunological considerations for COVID-19 vaccine strategies.
Nat Rev Immunol 2020; 20: 615–32.
10 Griffiths CEM, Armstrong AW, Gudjonsson JE, Barker JNWN.
Psoriasis. Lancet 2021; 397: 1301–15.
11 Angyal A, Longet S, Moore S, et al. T-cell and antibody responses to
first BNT162b2 vaccine dose in previously SARS-CoV-2-infected and
infection-naive UK healthcare workers: a multicentre, prospective,
observational cohort study. SSRN 2021; published online March 25.
https://dx.doi.org/10.2139/ssrn.3812375 (preprint).
12 Parry HM, Bruton R, Tut G, et al. Single vaccination with BNT162b2
or ChAdOx1 in older people induces equivalent antibody generation
but enhanced cellular responses after ChAdOx1. SSRN 2021;
published online April 13. https://dx.doi.org/10.2139/ssrn.3825573
(preprint).
13 Mahil SK, Wilson N, Dand N, et al. Psoriasis treat to target: defining
outcomes in psoriasis using data from a real-world, population-
based cohort study (the British Association of Dermatologists
Biologics and Immunomodulators Register, BADBIR).
Br J Dermatol 2020; 182: 1158–66.
14 Monin L, Laing AG, Muñoz-Ruiz M, et al. Safety and
immunogenicity of one versus two doses of the COVID-19 vaccine
BNT162b2 for patients with cancer: interim analysis of a prospective
observational study. Lancet Oncol 2021; 22: 765–78.
15 Orlov M, Wander PL, Morrell ED, Mikacenic C, Wurfel MM. A case
for targeting Th17 cells and IL-17A in SARS-CoV-2 infections.
J Immunol 2020; 205: 892–98.
16 Hotez PJ, Bottazzi ME, Corry DB. The potential role of Th17
immune responses in coronavirus immunopathology and vaccine-
induced immune enhancement. Microbes Infect 2020; 22: 165–67.
17 Panagioti E, Klenerman P, Lee LN, van der Burg SH, Arens R.
Features of effective T cell-inducing vaccines against chronic viral
Infections. Front Immunol 2018; 9: 276.
18 Rankin AL, MacLeod H, Keegan S, et al. IL-21 receptor is critical for
the development of memory B cell responses. J Immunol 2011;
186: 667–74.

19 Kennedy NA, Lin S, Goodhand JR, et al. Infliximab is associated with
attenuated immunogenicity to BNT162b2 and ChAdOx1 nCoV-19
SARS-CoV-2 vaccines in patients with IBD. Gut 2021; published online
April 26. https://doi.org/10.1136/gutjnl-2021-324789gutjnl-2021-324789.
20 Wong S-Y, Dixon R, Martinez Pazos V, et al. Serological response
to COVID-19 vaccination in IBD patients receiving biologics.
medRxiv 2021; published online March 20.
https://doi.org/10.1101/2021.03.17.21253848 (preprint).
21 Deepak P, Kim W, Paley MA. Glucocorticoids and B cell depleting
agents substantially impair immunogenicity of mRNA vaccines to
SARS-CoV-2. medRxiv 2021; published online April 7.
https://doi.org/10.1101/2021.04.05.21254656 (preprint).
22 Boyarsky BJ, Werbel WA, Avery RK, et al. Immunogenicity of a
single dose of SARS-CoV-2 messenger RNA vaccine in solid organ
transplant recipients. JAMA 2021; 325: 1784–86.
23 McElhaney JE, Ewen C, Zhou X, et al. Granzyme B: correlates with
protection and enhanced CTL response to influenza vaccination in
older adults. Vaccine 2009; 27: 2418–25.
24 McElhaney JE, Xie D, Hager WD, et al. T cell responses are better
correlates of vaccine protection in the elderly. J Immunol 2006;
176: 6333–39.
25 Cox RJ, Brokstad KA. Not just antibodies: B cells and T cells mediate
immunity to COVID-19. Nat Rev Immunol 2020; 20: 581–82.
26 Peng Y, Mentzer AJ, Liu G, et al. Broad and strong memory CD4+
and CD8+ T cells induced by SARS-CoV-2 in UK convalescent
individuals following COVID-19. Nat Immunol 2020; 21: 1336–45.
27 Grifoni A, Weiskopf D, Ramirez SI, et al. Targets of T cell responses
to SARS-CoV-2 coronavirus in humans with COVID-19 disease and
unexposed individuals. Cell 2020; 181: 1489–501.e15.
28 Ribeiro ACM, Guedes LK, Moraes JC, et al. Reduced seroprotection
after pandemic H1N1 influenza adjuvant-free vaccination in
patients with rheumatoid arthritis: implications for clinical practice.
Ann Rheum Dis 2011; 70: 2144–47.
www.thelancet.com/rheumatology Published online July 8, 2021 https://doi.org/10.1016/S2665-9913(21)00212-5 11
Articles
29 Park JK, Lee YJ, Shin K, et al. Impact of temporary methotrexate
discontinuation for 2 weeks on immunogenicity of seasonal
influenza vaccination in patients with rheumatoid arthritis:
a randomised clinical trial. Ann Rheum Dis 2018; 77: 898–904.
30 Strangfeld A, Schäfer M, Gianfrancesco MA, et al. Factors
associated with COVID-19-related death in people with rheumatic
diseases: results from the COVID-19 Global Rheumatology Alliance
physician-reported registry. Ann Rheum Dis 2021; published online
Jan 27. https:// doi.org/10.1136/annrheumdis-2020-219498.
31 Ungaro RC, Brenner EJ, Gearry RB, et al. Effect of IBD medications
on COVID-19 outcomes: results from an international registry. Gut
2021; 70: 725–32.
32 Supasa P, Zhou D, Dejnirattisai W, et al. Reduced neutralization of
SARS-CoV-2 B.1.1.7 variant by convalescent and vaccine sera. Cell
2021; 184: 2201–11.e7.
33 Collier DA, De Marco A, Ferreira IATM, et al. Sensitivity of
SARS-CoV-2 B.1.1.7 to mRNA vaccine-elicited antibodies. Nature
2021; 593: 136–41.
34 Garcia-Beltran WF, Lam EC, St Denis K, et al. Multiple SARS-CoV-2
variants escape neutralization by vaccine-induced humoral
immunity. Cell 2021; 184: 2372–83.e9.
35 Emary KRW, Golubchik T, Aley PK, et al. Efficacy of
ChAdOx1 nCoV-19 (AZD1222) vaccine against SARS-CoV-2 variant
of concern 202012/01 (B.1.1.7): an exploratory analysis of a
randomised controlled trial. Lancet 2021; 397: 1351–62.
36 Public Health England. COVID-19: the green book, chapter 14a.
Nov 27, 2020. https://www.gov.uk/government/publications/covid-
19-the-green-book-chapter-14a (accessed June 24, 2021).