Effect of antecedent B-cell depletion therapy for patients hospitalized with COVID-19 within a single health system: a propensity score analysis

B-cell depletion therapy is employed in a variety of clinical contexts from auto-immune diseases to malignancy. Prior research on patients with prior B-cell depletion treatment has suggested a mortality risk in patients hospitalized with COVID-19 however previous case-control studies have differed in their methods of patient comparison. Patients previously treated with B-cell-depletion hospitalized with COVID-19 were compared to matched controls in the Johns Hopkins Health System between March 1, 2020 and November 30, 2021. The primary outcome was 30-day all-cause mortality. Secondary outcomes included time to severe illness or death and time to clinical improvement. To eliminate bias due to imbalanced covariates, each patient who had previously received B-cell depletion therapy was matched with patients who had not received therapy based on age, sex, race, WHO severity score, admission date, COVID-19 specific treatment, and vaccination status. Propensity scores were calculated from a multivariable logistic regression model and performed on the matched sets, using B-cell depletion as the outcome, where the propensity score was the probability of receiving B-cell depletion therapy. The propensity score included matched covariates as well as smoking status, medical comorbidities, and vaccination status. Cox proportional-hazards regression models were applied on the matched sets to perform time to death, time to severe illness or death, and time to clinical improvement analyses. 50 patients were identified who had received B-cell depletion therapy and were compared to 186 matched controls. Patients treated with B-cell depletion experienced 30-day mortality of 6.0% compared to 3.8% in controls, adjusted hazard ratio (aHR) 1.45 (95% CI 0.30 to 6.95). B-cell-depleted patients experienced a longer time to clinical improvement, aHR 0.65 (95% CI 0.45-0.94). In this cohort, patients treated with B-cell depletion experienced a higher mortality rate compared to matched controls however this was not statistically significant. This group also experienced a prolonged time to clinical improvement based on WHO severity score.