The effect of immunomodulators on seroconversion after BNT162b2 and AZD1222 vaccines in patients with immune-mediated inflammatory diseases: a prospective cohort study.

Background: Biologic and nonbiologic immunomodulators, used to treat immune-mediated inflammatory diseases (IMIDs), could impair the immune response to COVID-19 vaccines and thus vaccine effectiveness.

Objectives: Our objective was to investigate the association between biologic and nonbiologic immunomodulators and seroconversion following the first and second dose of COVID-19 vaccines in patients with IMIDs.

Methods: Serum samples were collected following the first or second dose of the BNT162b2 or AZD1222 vaccines from patients receiving biologic and/or nonbiologic immunomodulators for one or more of psoriasis, psoriatic arthritis, rheumatoid arthritis, inflammatory bowel disease or systemic lupus erythematosus. Seroconversion was defined as a positive Roche Elecsys® Anti-SARS-CoV-2 S (spike protein subunit S1/receptor binding domain) immunoassay (≥ 0.8 U mL-1). Association between immunomodulator exposure and seroconversion was assessed using logistic regression, adjusting for age and sex.

Results: After excluding those with prior COVID-19, post-first vaccine dose samples from 193 participants and post-second dose samples from 312 participants were included in the analysis. Following the first vaccine dose, 17.6% (n = 34) of participants did not seroconvert. Seroconversion was reduced for those on nonbiologic [adjusted odds ratio (OR) 0.29, 95% confidence interval (CI) 0.12-0.69] or combined nonbiologic and biologic treatment (adjusted OR 0.14, 95% CI 0.045-0.45) compared with those on biologic monotherapy. Subgroup analysis demonstrated reduced odds of seroconversion in those on methotrexate (adjusted OR 0.097, 95% CI 0.19-0.49) or prednisolone treatment (adjusted OR 0.044, 95% CI 0.002-1.00) relative to tumour necrosis factor-α inhibitor monotherapy. No participants receiving rituximab (n < 5) seroconverted after the first vaccine dose. Following the second vaccine dose, 1.6% of all participants did not seroconvert. Non-seroconversion was associated with receiving rituximab (n = 3 of 4) compared with those receiving other therapies (n = 2 of 308, P < 0.001). Post hoc analyses demonstrated that non-seroconversion was associated with age [adjusted OR 0.18, 95% CI 0.037-0.93 for those aged 60 years and over (reference category age 18-39 years)], but not sex, ethnicity or vaccine type.

Conclusions: Treatment with nonbiologics, particularly methotrexate, is associated with impaired seroconversion following two BNT162b2 or AZD1222 vaccine doses, in patients with IMIDs. These findings are consistent with those of other published studies. While this could indicate reduced protection against COVID-19, the immunological parameters that correlate most closely with vaccine effectiveness need to be defined to reach this conclusion.