Antibody, not Cellular, Immune Responses to SARS-CoV-2 Vaccination Outperform Infection in Inflammatory Bowel Disease Patients

This is the authors' abstract. We don't add key points for this paper.

Following the successful development of the first vaccines to protect against SARS-CoV-2 infection, the International Organization for the Study of Inflammatory Bowel Disease (IOIBD) recommended full vaccination of all patients with inflammatory bowel disease (IBD) already early in the pandemic.1 Subsequent studies have reported the effects of immunosuppressive medications on immune responses to SARS-CoV-2 infection and vaccination in patients with IBD. The most consistent findings are of reduced serological responses in patients on infliximab and an uncoupling of cellular and humoral responses.2–5 Although previous studies assessed aspects of the immune response in IBD, a direct longitudinal comparison of immunity induced by primary SARS-CoV-2 infection compared with vaccination in patients with IBD has not been reported. As part of the ICARUS-IBD study at the Icahn School of Medicine at Mount Sinai, patients with IBD were followed from early in the pandemic (May 2020) prior to vaccine development through March 2022 when full vaccination schedules, including booster doses, were available. We prospectively obtained clinical data, serum, and peripheral blood mononuclear cells (PBMCs; Supplementary Figure 1). This biobank allows the first detailed examination of the relative immune response to naturally acquired infection compared with vaccination-induced immunity. Of 616 patients with IBD in the ICARUS-New York City cohort, 62 patients were positive for anti-SARS-CoV-2 Spike (S) or anti-SARS-CoV-2 Nucleocapsid (N) antibodies prior to vaccination. Control patients (n = 48) were matched by sex, biological therapy, vaccine, and comparable lengths of longitudinal follow-up. Patients were treated with infliximab, vedolizumab, or ustekinumab and received either an mRNA vaccine (from Pfizer-BioNTech or Moderna) or the viral vector vaccine from Johnson & Johnson (Supplementary Table 1). In this study, the analysis of male and female patient responses did not reveal significant differences at any time point, whether postnatural infection or following vaccination. Furthermore, patients with ulcerative colitis (UC; 3365 U/mL ± 6453, geometric mean titer [GMT] = 642.7 ± 11.81) demonstrated a significantly elevated humoral immune response compared with those with Crohn’s Disease (CD; 1408 U/mL ± 6677, GMT = 38.554 ± 36.84, P = .0004) in the context of natural infection. However, this distinction was not observed postvaccination, a finding that could potentially be attributed to a reduced sample size during follow-up. Importantly, no differences were detected in cellular immune responses between UC and CD patients (Supplementary Figure 2). Uninfected individuals who received 2 vaccine doses (43.22 U/mL ± 32.96, GMT = 22.41 ± 4.91) had a higher SARS-CoV-2 spike (S) specific antibody concentration than naturally infected individuals (22.19 U/mL ± 42.68, GMT = 5.66 ± 7.05, P = .0022; Figure 1A). Double-vaccinated individuals had greater neutralization against WIV04/2019 (IC50 of 1914 ± 3024, GMT = 281.3 ± 24.66) than the B.1.1.529 (Omicron; IC50 of 268 ± 380, GMT = 10.57 ± 51.13, P = .0018) variant (Supplementary Figure 3A). Neutralization responses were maintained for 9 months following natural infection and double vaccination and 6 months after triple vaccination. Upon third vaccination dose, the neutralization responses were greater in naïve (IC50 of 6616 ± 8396, GMT = 2785 ± 4.13) compared with naturally infected patients (IC50 of 2242 ± 6601, GMT 127.9 ± 31.4, P = .0002; Figure 1D). When comparing responses of patients on infliximab vs vedolizumab vs ustekinumab therapy after natural infection and vaccination, there were no significant differences between the groups (Supplementary Figure 3B). Additionally, individuals who were exposed to 3 antigens—2 doses of vaccine and natural infection—did not demonstrate a significant difference in immune response compared with those who solely received 2 vaccine doses (P = .4176). This observation held true across various treatment regimens (Supplementary Figure 3). Consequently, although vaccination alone yielded a superior immune response compared with natural infection, we did not observe an additive response in patients who experienced both vaccination and natural infection. Kinetics of antibody responses, neutralization responses, and T-cell immunity assessed by ELISPOT against SARS-CoV-2 spike in individuals with IBD at different time points. A, COVT2 antibody response between naïve, naturally infected and vaccinated individuals. B-C, Neutralizing responses to SARS-CoV-2 WIV04/2019 (B) and B.1.617 (C) at different time points. D, Neutralizing responses to SARS-CoV-2 WIV04/2019 over a time course following natural infection and second dose vaccination. E-F, IFNγ secreting cellular response to S1 + S2 (E) and M/NP (F) in IBD patients at different time points following natural infection and vaccination. COV2T, SARS-CoV-2 total antibody assay. Neutralization responses following SARS-CoV-2 infection and vaccination in patients with IBD measured by pseudotyped microneutralization assay. Magnitude of the T-cell response, as determined by IFNγ+ Spot Forming Units (SFU)/106 PBMCs, to SARS-CoV-2. Kruskal-Wallis test was performed for A, B, C, E and F. Mann Whitney test was performed for D. The SARS-CoV-2 anti-S T-cell responses were similar between naturally infected (254.3 spot forming units [SFU]/106 PBMCs ± 351.6, GMT = 63.46 ± 11.54) and double vaccinated individuals without prior infection (489.7 SFU/106 PBMCs ± 1020, GMT = 19.97 ± 48.83, P => .999; Figure 1E). Importantly, anti-M/NP responses were not induced in vaccinated patients without prior infection, as all vaccines available in the United States target the S protein (Figure 1F). There were no significant differences between treatment groups (Supplementary Figure 3C). This study provides the longest longitudinal surveillance of patients with IBD in a single center with analysis of both neutralizing antibody and T-cell responses. We were uniquely able to compare SARS-CoV-2 antibody and T-cell responses in naturally infected and vaccinated IBD patients over time, including within the same individual. Comparing the natural course of antibody responses to SARS-CoV-2 infection to vaccination has important implications for recommendations regarding management and potential vaccination strategies after SARS-CoV-2 infection in patients with IBD. Our findings largely correlate with previous findings from larger studies, such as the CLARITY and VIP groups, showing increasing antibody titres with subsequent doses and greater neutralization responses to the SARS-CoV-2 reference sequence than variants. While these and other studies clarified differences in serological responses based upon medications, notably that patients with IBD treated receiving infliximab or tofacitinib exhibited lower anti-SARS-CoV-2 S protein neutralizing antibodies than those receiving vedolizumab and healthy controls, our study reveals that antibody responses after 2- and 3-dose vaccinations are more robust than infection.3,4,6 This has previously been shown in healthy controls but not in patients with IBD or any other immune-mediated diseases receiving immune-modulating medications. Regarding cellular responses, previous studies in immunocompromised cohorts have shown a clinically protective role of cell-mediated immune responses independent of humoral responses after SARS-CoV-2 infection.5 In individuals with IBD on antitumor necrosis factor (TNF) therapy, an increased T-cell clonal depth compared with untreated groups was shown.7,8 Others have reported uncoupling of the immune response in IBD, as evidenced by differences between antibodies directed against the receptor binding domain (RBD) of the spike and T-cell responses against the full spike8. On the other hand, similar T-cell responses in double vaccinated patients under infliximab and vedolizumab therapies have been shown.9 Recently, the CLARITY-IBD study investigators found no difference in anti-S T-cell fractions or interferon gamma-producing T cells between infliximab and vedolizumab after one or 2 SARS-CoV-2 vaccination doses.6 They did, however, find a reduced T-cell response in patients taking tofacitinib.3,10 In this study, we show that anti-S T cell responses were comparable across medications in both naturally infected and vaccinated patients. Importantly, we show that in uninfected individuals with IBD on immunosuppressive medications, 2 or 3 doses of SARS-CoV-2 vaccine outperform natural infection in terms of the serological response. Although this difference was not reflected in the cellular responses, we argue that full vaccination schedules in IBD patients on immunosuppressive treatments should be recommended, even if they have had prior SARS-CoV-2 infections. Limitations of the study include that it is a single center study with a small cohort of patients on ustekinumab and lack of patients receiving janus kinase inhibitors and other biological therapies. On the other hand, our study provides unique insights on combined serological and cellular responses in SARS-CoV-2-infected and vaccinated IBD individuals. These data support full vaccination schedules in IBD patients, regardless of therapy or infection history. ICARUS-IBD working group: Paul Klenerman, Vicky Martinez Pazos, Marcia Mukanga Lange4, Emily Spiera, Lodoe Sangmo, Joshua Lacoste, Drew Helmus, and Jean-Frederic Colombel. The authors wish to thank The Mount Sinai Therapeutic Infusion Center nursing team and Siemens Healthineers. E.G.C.: conceptualization: equal; data curation: equal; formal analysis: equal; investigation: equal; methodology: equal; visualization: lead; writing, reviewing and editing: equal. M.E.: conceptualization: equal; data curation: equal; formal analysis: equal; investigation: equal; visualization: equal; methodology: equal; writing, reviewing and editing: equal. J.W.: conceptualization: equal; data curation: equal; formal analysis: equal; investigation: equal; visualization: equal; methodology: equal; writing, reviewing and editing: equal. C.T.: conceptualization: equal; data curation: equal; formal analysis: equal; methodology: equal; resources: equal; writing, reviewing and editing: equal. K.C.: conceptualization: equal; methodology: equal; resources: equal; writing, reviewing and editing: equal. J.S.: conceptualization: equal; funding acquisition: lead; methodology: equal; resources: equal; writing, reviewing and editing: equal. S.Y.W.: conceptualization: lead; data curation: equal; formal analysis: supporting; funding acquisition: lead; methodology: equal; project administration: lead; resources: equal; visualization: supporting; writing, reviewing and editing: equal. This work was supported by The Leona M. and Harry B. Helmsley Charitable Trust as part of the ICARUS study (grant number 2107- 04731). J.W. was supported by the Research Foundation Flanders (FWO), Belgium by a PhD Fellowship strategic basic research (SB) grant (1S06023N).