Abstract

Background: The extent of the SARS-CoV-2 short-term evolution under Remdesivir (RDV) exposure and whether it varies across different upper respiratory compartments are not fully understood. Methods: Patients hospitalized for COVID-19, with or without RDV therapy, were enrolled and completed up to three visits, in which they provided specimens from four respiratory compartments. Near full-length genome SARS-CoV-2 sequences were obtained from viral RNA, standard lineage and variant assignments were performed, and viral mutations in the RNA-dependent RNA polymerase (RdRp) region—the RDV target gene—were detected and compared between participants with and without RDV, across the four compartments, within participants across visits, and versus a larger sequence dataset. The statistical analysis used a generalized linear mixed-effects model. Results: A total of 139 sequences were obtained from 37 out of the 44 (84%) enrolled participants. The genotyping success varied across respiratory compartments, which ranged from 42% with oropharyngeal specimens to 67% with nasopharyngeal specimens and showed improvement with higher viral loads. No RdRp mutations known to be associated with RDV resistance were identified, and for 34 detected mutations at 32 amino acid positions that are not known as RDV-associated, there was no evidence of any associations with the RDV exposure, respiratory compartment, or time. At least 1 of these 34 mutations were detected in all participants, and some differed from the larger sequence dataset. Conclusions: This study highlighted the SARS-CoV-2 short-term genomic stability within hosts and across upper respiratory compartments, which suggests a lack of evolution of RDV resistance over time. This contributes to our understanding of SARS-CoV-2 genomic dynamics.