Breakthrough SARS-CoV-2 infections and prediction of moderate-to-severe outcomes during rituximab therapy in patients with rheumatic and musculoskeletal diseases in the UK: a single-centre cohort study

Abstract

Background

Concerns have been raised regarding the reduced immunogenicity of vaccines against COVID-19 in patients with autoimmune diseases treated with rituximab. However, the incidence and severity of breakthrough infections in unbiased samples of patients with specific rheumatic and musculoskeletal diseases are largely unknown. We aimed to assess the incidence of breakthrough SARS-CoV-2 infection, compare rates of moderate-to-severe COVID-19 with any severe infection event, and evaluate predictors of moderate-to-severe COVID-19 outcomes in patients treated with rituximab.

Methods

We did a retrospective cohort study in all rituximab-treated patients with rheumatic and musculoskeletal diseases in a single centre in Leeds, UK between March 1, 2020 (the index date), and April 1, 2022. Adults aged 18 years and older, who fulfilled classification criteria for established rheumatic and musculoskeletal diseases, and received therapy with at least one rituximab infusion between Sept 1, 2019 (6 months before the pandemic in the UK), and April 1, 2022, were eligible for inclusion in the study. SARS-CoV-2 infection was defined by antigen test or PCR. COVID-19 outcomes were categorised as mild (from ambulatory to hospitalised but not requiring oxygen support) or moderate-to-severe (hospitalised and requiring oxygen support or death). The primary outcome was breakthrough COVID-19 infection, which was defined as an infection occurring 14 days or more after the second vaccine dose. Predictors of moderate-to-severe COVID-19 outcomes were analysed using Cox regression proportional hazards.

Findings

Of the 1280 patients who were treated with at least one cycle of rituximab since Jan 1, 2002, 485 (38%) remained on rituximab therapy on April 1, 2022. Of these patients, 400 fulfilled all inclusion criteria and were included in our final analysis. The mean age at the index date was 58·9 years (SD 14·6), 288 (72%) of 400 patients were female and 112 (28%) were male, 333 (83%) were White, and 110 (28%) had two or more comorbidities. 272 (68%) of 400 patients had rheumatoid arthritis, 48 (12%) had systemic lupus erythematosus, 48 (12%) had anti-neutrophil cytoplasmic antibody-associated vasculitis, and 46 (12%) had other rheumatic and musculoskeletal diseases. During the study, 798 rituximab cycles were administered. Of the 398 (>99%) of 400 patients with vaccine data, 372 (93%) were fully vaccinated. Over the 774·6 patient-years of follow-up, there was an incremental increase in all SARS-CoV-2 severity types over the three pandemic phases (wild-type or alpha, delta, and omicron), but most infections were mild. The rates of moderate-to-severe COVID-19 were broadly similar across these three variant phases. Of 370 patients who were fully vaccinated and with complete data, 110 (30%) had all severity type breakthrough COVID-19, 16 (4%) had moderate-to-severe breakthrough COVID-19, and one (<1%) died. In the post-vaccination phase (after Dec 18, 2020), the incidence rates of all severity type and moderate-to-severe COVID-19 were substantially lower in those who were fully vaccinated compared with unvaccinated or partially vaccinated individuals (22·83 per 100 person-years [95% CI 18·94–27·52] in those who were fully vaccinated vs 89·46 per 100 person-years [52·98–151·05] in those who were partially vaccinated or unvaccinated for infections of all severities, and 3·32 per 100 person-years [2·03–5·42] in those who were fully vaccinated vs 25·56 per 100 person-years [9·59–68·10] in those who were partially vaccinated or unvaccinated for moderate-to-severe infections). The rate of moderate-to-severe COVID-19 was broadly similar to other severe infection events in this cohort (5·68 per 100 person-years [95% CI 4·22–7·63]). In multivariable Cox regression analysis, factors associated with an increased risk of moderate-to-severe COVID-19 were the number of comorbidities (hazard ratio 1·46 [95% CI 1·13–1·89]; p=0·0037) and hypogammaglobulinaemia (defined by a pre-rituximab IgG concentration of <6 g/L; 3·22 [1·27–8·19]; p=0·014). This risk was reduced with each vaccine dose received (0·49 [0·37–0·65]; p<0·0001). Other factors, including concomitant prednisolone use, rituximab-associated factors (eg, rituximab dose and time to vaccination since last rituximab dose), and vaccine-associated factors (eg, vaccine type and peripheral B-cell depletion) were not predictive of moderate-to-severe COVID-19 outcomes.

Interpretation

This study presented detailed analyses of rituximab-treated patients during various phases of the COVID-19 pandemic. In later stages of the pandemic, the SARS-CoV-2 breakthrough infection rate was high but severe COVID-19 rates were similar to any severe infection event rate in patients who were vaccinated. The risk–benefit ratio might still favour rituximab in vaccinated patients with severe rheumatic and musculoskeletal diseases who have few other treatment options. Increased vigilance is needed in the presence of comorbidities and hypogammaglobulinaemia for all infection types.

Funding

Wellcome Trust and Eli Lilly.