Predicting persistent SARS-CoV-2 shedding in immunocompromised patients: a probability-based approach

Background: Although recommended isolation periods for Coronavirus disease 2019 (COVID-19) have been shortened as the pandemic has subsided, prolonged Severe Acute Respiratory Syndrome-Coronavirus-2 (SARS-CoV-2) shedding remains common in immunocompromised patients. This study estimated the probability of viral clearance in these patients based on elapsed days and specific risk factors.

Methods: We prospectively enrolled immunocompromised patients with a confirmed COVID-19 diagnosis from January 2022 to May 2023 during the Omicron variant era. We collected weekly respiratory specimens for viral load measurement and culture. We identified significant predictors of viral culture negative conversion through univariate and multivariate analyses and estimated viral clearance probabilities using a Cox time-varying proportional hazard model.

Results: Among 70 patients with serial 319 respiratory specimens with positive SARS-CoV-2 genomic polymerase chain reaction results that underwent cell culture, ∼69% (48) had haematologic malignancies and 31% (22) underwent solid organ transplants. B-cell depleting agents and viral copy number significantly influenced viral culture negative conversion. The probability of culture-negative conversion for immunocompromised patients not treated with B-cell-depleting agents increased over time, with over 90% achieving negative conversion by Day 84. Patients treated with B-cell depleting agents showed lower conversion rates. By Day 84, <90% of patients with cycle threshold values 23-28 [4.85-6.35 log copies/mL] achieved culture-negative conversion. The results indicate more prolonged shedding than in patients without B-cell depletion.

Conclusion: Estimating SARS-CoV-2 clearance probabilities based on specific risk factors can guide individualised isolation decisions for immunocompromised patients, tailoring policies to each patient's delayed viral clearance risk.