In this observational case-control study, children with IBD and healthy age-and sex-matched children receiving two doses of messenger ribonucleic acid (mRNA) vaccine BNT162b2 (Cominaty-Pfizer/BioNTech) were enrolled at the University of Debrecen, Department of Pediatrics. Vaccination of children 12 years or older by the mRNA vaccine BNT162b2 (Cominaty-Pfizer/BioNTech) was approved by the European Medicine Agency in May 2021, and in Hungary it commenced in June 2021. All participants received vaccination according to current protocols. Patient enrollment was initiated in June 2021 and was completed in May 2022. Inclusion criteria of patients were the following: age 12-18 years with previously diagnosed IBD (CD or UC), parental consent to receiving the Pfizer-BioNTech BNT162b2 vaccine and to participation in the study. Exclusion criteria included surgical intervention or thrombotic event in the past 3 months, history of hemorrhagic/thrombotic diathesis, other autoimmune diseases, malignant disease, anticoagulant therapy, anaemia or thrombocytopenia, hepatic- or renal disease. IBD was diagnosed by means of clinical, endoscopic, and histological findings in accordance with the consensus guidelines of ECCO/ESPGHAN (14). Inclusion criteria of healthy controls were: age 12-18 years, parental consent to receiving the Pfizer-BioNTech BNT162b2 vaccine and to participation in the study. Exclusion criteria: chronic disorders (malignancy, known haemorrhagic or thrombotic disorders, autoimmune disorders, chronic kidney or liver disease, diabetes mellitus, etc.) acute illness within the past 4 weeks, thrombotic event, surgery or major trauma within the past 3 months, medications including anticoagulant use. Baseline demographic data of participants (age, sex, BMI) and clinical characteristics of patients (disease activity, medications) were registered before the first vaccine dose. The clinical activity of CD was characterized by the Pediatric Crohn’s Disease Activity Index (PCDAI) (15). The Pediatric Ulcerative Colitis Activity Index (PUCAI) was used to characterize the clinical activity of children with UC (16). Follow-up of patients and controls took place at two to four weeks after the second vaccine dose. A guided questionnaire was used to collect data on adverse reactions (AEs) after each vaccination dose, conducted by a trained interviewer. Vaccine AEs were classified as localized or systemic reactions as described earlier (11). Localized AEs included local swelling, pain/sensitivity, redness or other local reactions at the injection site. Systemic AEs included fever, chills, fatigue, dizziness, headache, nausea, loss of appetite, joint pain, muscle pain, enlargement of lymph nodes, sleep disturbances, allergic reaction, rash, or other. If an AE was reported, the rate of severity was asked (mild: did not interfere with daily activity; moderate: required medication, interfered with daily activity; severe: hospitalization or medical care by professional was required). At follow up, all patients were assessed again for the clinical activity of IBD using PCDAI or PUCAI. Flare of IBD was defined as reported earlier (11), based on the increase of disease activity index, including the worsening of at least one of the following symptoms: abdominal pain, bowel movement frequency, rectal bleeding, extraintestinal manifestations.

The study design was in accordance with the guiding principles of the Declaration of Helsinki and was approved by the Institutional Ethics Committee of the University of Debrecen and the Ethics Committee of the National Medical Research Council. All parents or caregivers provided written informed consent.

Venous blood was drawn on two occasions from the participants in this cohort: before the first dose of Pfizer-BioNTech BNT162b2 vaccination (up to 7 days before) and 2-6 weeks after the second vaccine dose (elapsed time 10.0±4.0 weeks). Routine laboratory tests (serum electrolytes, renal and liver function tests, lipid profile, serum iron, ferritin, high-sensitivity C-reactive protein measurement, complete blood count) were carried out immediately from the blood samples by standard laboratory methods (Roche Diagnostics, Mannheim, Germany and Sysmex Europe GmbH, Hamburg, Germany). For the examination of haemostasis tests, blood samples were collected to vacutainer tubes containing 0.109 M sodium citrate (Becton Dickinson, Franklin Lane, NJ) and were processed to obtain platelet-free plasma immediately (centrifugation twice at 1500 g, room temperature, 15 min). Screening tests of coagulation (prothrombin time, activated partial thromboplastin time, and thrombin time) were performed immediately on a BCS coagulometer using routine methods (Siemens Healthcare Diagnostic Products, Marburg, Germany). Fibrinogen levels were measured according to the method of Clauss on a BCS coagulometer (Siemens Healthcare Diagnostic Products, Marburg, Germany). Serum angiotensin convertase enzyme 2 (ACE2) activity was assessed using a specific quenched fluorescent substrate as described earlier (17). For the execution of thrombin generation assay, aliquots of citrated plasma were stored at −80°C until analysis.

For the quantitative determination of antibodies against SARS-CoV-2 nucleocapsid (N) and spike (S) proteins, two Elecsys^® anti-SARS-CoV-2 tests were used on an electro-chemiluminescent immunoassay instrument (Cobas^® 602, Roche Diagnostics, Mannheim, Germany). The anti-SARS-CoV-2 N assay uses a recombinant protein representing the nucleocapsid antigen in a double-antigen sandwich assay format. The test is intended as an aid in the determination of the specific humoral immune reaction to SARS-CoV-2. The anti-SARS-CoV-2 S assay uses a recombinant RBD protein in a double-antigen sandwich assay format, which favors the quantitative determination of high affinity antibodies against SARS-CoV-2. The test is intended as an aid to assess the adaptive humoral immune response, including neutralizing antibodies, to the SARS-CoV-2 S protein after natural infection with SARS-CoV-2 or in vaccine recipients. During the measurement, the total amount of anti- SARS-CoV-2 S immunoglobulins (IgG or IgM) was calibrated against the WHO international standard and expressed as binding antibody units (BAU)/mL. In case of anti-SARS-CoV-2 N assay, the result of a sample was given in the form of a cut-off index (COI; signal sample/cut-off). According to the manufacturer, seropositivity can be pronounced above the limit value of 0.8 BAU/mL of anti-SARS-CoV-2 S total Igs.

Thrombin generation test was carried out from stored platelet-free plasma as described previously using the Thrombinoscope CAT (Calibrated Automated Thrombogram, Maastricht, The Netherlands) assay, based on the manufacturer’s instructions (Diagnostica Stago, Asnieres, France) (18). Briefly, 80 μL of plasma was incubated with 20 μL PPP-Reagent™ (containing 5 pM recombinant tissue factor and 4 μM phospholipids) in black round-bottomed 96-well microplates (Greiner Bio, One North America Inc., Monroe, MI, USA) for 10 minutes. For each sample, a calibrator (α2-macroglobulin-thrombin complex with known thrombin activity: Thrombin Calibrator™) was run in parallel, to correct the fluorescence signal for substrate consumption and to compensate plasma color variability. Thrombin generation was initiated by the addition of 20 μL of FluCa-Kit™ (fluorescent substrate and a buffer solution with CaCl₂: Fluo Buffer). All measurements were performed in duplicates. Fluorescence was detected by a Fluoroskan Ascent^® fluorimeter (Thermo Fischer Scientific, Waltham, MA) and the TG curves were analysed by the Thrombinoscope software (Thrombinoscope BV, Maastricht, The Netherlands). Thrombin generation curves were characterised by the following parameters (calculated and presented by the Thrombinoscope software): 1/Lagtime: the moment from the initiation of the test until thrombin generation starts (the time needed for the first amounts of thrombin to be generated), 2/Endogenous Thrombin Potential (ETP): the area under the curve, 3/Peak Thrombin: the highest thrombin concentration formed, 4/Time to Peak: the time until the Peak Thrombin.

Statistical analysis was performed by Graphpad Prism 8.0 software (GraphPad Prism Inc., La Jolla, CA). Normality of data was studied using the Shapiro-Wilk and Kolmogorov-Smirnov tests. Normally distributed data is presented as mean ± standard deviation (SD), non-parametric data as median and interquartile range (IQR). For two-group analyses, Student’s t-test or Mann-Whitney U-test was performed, depending on the result of the normality tests. Wilcoxon signed-rank-test was used to analyse paired data sets. ANOVA using the Bonferroni correction or Kruskal-Wallis test using the Dunn’s post-hoc test were applied for multiple comparisons. Spearman’s or Pearson’s correlation coefficient was used to determine the strength of correlation between continuous variables. Differences between categorical variables were assessed by χ² test or by Fisher’s exact where appropriate.

One hundred children were included in the study, 38 IBD patients (18 CD and 20 UC) and 62 age- and sex-matched healthy controls. Baseline characteristics of the study cohort are reported in Table 1. Based on the anti-SARS-CoV-2 N and anti-SARS-CoV-2 S assay, seropositivity was found at similar rates in healthy controls (60%, n=37) as compared to children with IBD (48%, n=16; p=0.2813). Among the baseline routine laboratory parameters, CRP levels, platelet count, and fibrinogen levels were significantly higher in patients with IBD as compared to controls. ACE2 activity was also significantly elevated in children with IBD as compared to controls. Iron levels and haemoglobin concentrations were significantly lower in patients. Baseline TG parameters did not differ between patients and controls. Except for BMI, which was significantly lower in children with CD, baseline characteristics of CD and UC patients did not differ significantly (Table 2). Six children had active (mild or moderate) disease, while most children had clinically quiescent IBD as assessed by PCDAI (median: 5, IQR: 0-15) and PUCAI (median: 0, IQR: 0-8). Thrombin generation parameters did not differ between CD and UC patients at baseline. Seropositivity for anti-SARS-CoV-2 total Ig (IgG/IgM) was not associated with any baseline TG parameters in either group (data not shown). Most patients received 5-ASA as monotherapy or in combination with azathioprine and/or TNFα inhibitor therapy (Table 2).

Although the extent of TG was not significantly different between patients and healthy controls, in patients with IBD, ETP showed a significant positive correlation with markers of inflammation including hsCRP, WBC count, platelet count and fibrinogen and a significant negative correlation with hemoglobin levels at baseline (Table 3). Such associations were not found in healthy control children (Supplementary Table 1). In both groups, the extent of TG (ETP and peak thrombin) negatively correlated with PT and APTT (Table 3, Supplementary Table 1).

Routine laboratory parameters did not differ significantly between the samples obtained up to 7 days before the first dose of BNT162b2 mRNA vaccination and 2-6 weeks after the second dose in either group, except for a marginal decrease in fibrinogen and ACE2 levels in controls (Supplementary Tables 2–4). TG parameters of the cohort before and after vaccination are shown on Figure 1. Although in some cases individual changes in TG parameters could be observed over time, paired data were not significantly different in any group.

According to the anti-SARS-CoV-2 S total Ig (IgG/IgM) measurement, vaccination significantly increased antibody levels in all three investigated groups (Figure 2, Supplementary Tables 2–4). In healthy control children, the obtained response was significantly higher as compared to children with CD and UC (anti-SARS-CoV-2 S total Ig median:15057; IQR: 11344-30449 vs. median:7475; IQR: 358-27495 and median:7606; IQR: 2480-12781 BAU/mL, respectively, p=0.0005). No significant difference was observed in post-vaccination anti-SARS-CoV-2 S IgG/IgM levels between CD and UC patients (p=0.9999). Postvaccination anti-SARS-CoV-2 S IgG/IgM levels were below the 5^th percentile value of healthy children in more than one third of patients (CD: 37.5%, n=6/16 and UC: 41.1%, n=7/17). Patients receiving TNFα inhibitor therapy presented significantly lower SARS-CoV-2 S IgG/IgM levels as compared to those receiving other immunosuppressive regimens (Figure 3).

IBD disease activity scores did not increase significantly following vaccination (Supplementary Tables 3 and 4). In children with CD, ETP showed significant association with PCDAI before vaccination (Figure 4A). After vaccination, only the lag time parameter of TG showed significant association with PCDAI (Figure 4B), although this correlation was not present before vaccination. None of the CD patients experienced disease flare within 6 weeks of vaccination, on the contrary, in two patients, a decrease in disease activity was observed post-vaccination, that was accompanied by a decrease in ETP (2575 to 2134 nM*min and 1693 to 1497 nM*min) (Figure 4C, Supplementary Table 3). In these two cases, glucocorticoid therapy was initiated in addition to the previous immunosuppressive treatment before the first blood draw, and the decrease in disease activity is most probably attributed to the intensified treatment. Two UC patients had IBD flare (one already with active IBD before vaccination) while one UC patient had a decrease in disease activity post-vaccination, but these changes were not accompanied by parallel changes in TG parameters (data not shown).

Systemic AEs, after the first dose and the second dose of BNT162b2 mRNA vaccination as self-reported by a questionnaire, did not differ between patients and controls (Figures 5A, B). Interestingly, local symptoms (pain, swelling) were found to be significantly less frequent in children with IBD as compared to healthy control children after the first and second dose of the vaccine (Figures 5A, B). No thrombotic events occurred in the cohort during the study period.