ABSTRACT
Background: Remdesivir (RDV) is a broad-spectrum nucleotide analog antiviral approved for the treatment of COVID-19 in patients who are hospitalized or non-hospitalized and at risk of progressing to severe disease. Here we present SARS-CoV-2 resistance analyses from the Phase 3 PINETREE trial. Method(s): PINETREE was a double-blind, placebo-controlled trial of nonhospitalized participants (N=562) with COVID-19 and >=1 risk factor for disease progression, randomized to receive RDV or placebo once-daily for 3 days. The whole genome of SARS-CoV-2 was sequenced from nasopharyngeal swabs collected at days 1 (baseline), 2, 3, 7, and 14 using next-generation sequencing. Emergent amino acid substitutions in SARS-CoV-2 from participants treated with RDV were tested in a replicon system to determine if they alter sensitivity to RDV. Result(s): Resistance analysis criteria included all participants in the RDV group and 50% in the placebo group with viral load above the lower limit of detection for the viral load assay. Of 281 participants who met these criteria, baseline and postbaseline sequencing data were available for 115/130 (88.5%) participants in the RDV group and 129/151 (85.4%) participants in the placebo group (Table 1). Among these, emergent substitutions in Nsp12 were observed in 8/115 (7.0%) in the RDV group and 7/129 (5.4%) in the placebo group. A total of 7 emergent amino acid substitutions in Nsp12 were observed in the RDV group, but not in the placebo group. Among these, only one substitution from one participant (A376V;first detected at day 14), showed reduced in vitro susceptibility to RDV, with a half-maximal effective concentration (EC50) fold-change of 12.6 compared with a wildtype reference. The participant achieved clinical recovery by day 14. None of the other substitutions impacted RDV susceptibility (EC50 fold-change <=1.4). Emergent substitutions in Nsp8, Nsp10, Nsp13, or Nsp14 were detected in 10/115 (8.7%) of participants in the RDV group and 10/129 (7.8%) in the placebo group, with substitutions in the RDV group showing similar susceptibility to RDV as the wildtype reference (EC50 fold-change <=2.3). Conclusion(s): Overall, emergent substitutions in the SARS-CoV-2 replication complex including Nsp12 were observed with similar frequency in the RDV and placebo groups, with only one participant developing a substitution associated with reduced in vitro RDV susceptibility, indicating a high barrier to the development of RDV resistance in COVID-19 patients.
ABSTRACT
Background: Recent SARS-CoV-2 variants of concern (VOCs) have shown a progressive loss of sensitivity to monoclonal antibody therapeutics. Remdesivir (RDV) is a nucleotide analog prodrug that targets the viral RNA-dependent RNA polymerase (RdRp) Nsp12 and is approved to treat COVID-19 in hospitalized and non-hospitalized patients. Nsp12 is highly conserved across VOCs to date and RDV antiviral activity against previous VOCs (Alpha to Omicron BA.1) has been maintained. Here, we conduct a structural analysis of Nsp12 substitutions observed in recent Omicron subvariants (BA.2, BA.2.12.1, BA.4, BA.5 and BA.2.75) and assess RDV antiviral activity against clinical isolates and sitedirected mutants (SDMs) in a replicon system. Method(s): The prevalence of Nsp12 substitutions in Omicron subvariants was evaluated by analysis of sequences from the Global Initiative on Sharing Avian Influenza Data (GISAID) EpiCoV database. Structural analysis of identified substitutions was conducted on a prior cryo-electron microscopy-based model of the replication-transcription complex. Antiviral activity against subvariant clinical isolates was assessed by nucleoprotein ELISA in A549-hACE2-TMPRSS2 cells and by SDMs in the replicon system. Result(s): Genomic analysis of >1.4 million Omicron subvariant sequences revealed unique substitutions in Nsp12 compared to the ancestral WA1 strain. Besides P323L, present in all subvariants, G671S was observed in 95.9% of BA.2.75 sequences, F694Y was observed in <=1.9% of BA.4, BA.5 and BA.2.75 sequences, and Y521C was observed in 1.7% of BA.5 sequences. As anticipated, structural analysis of these substitutions showed no direct interaction with the incoming RDV nucleotide triphosphate or the viral RNA. Phenotyping of clinical isolates of Omicron subvariants BA.2, BA.2.12.1, BA.4, BA.5, and BA.2.75 consistently resulted in mean RDV EC50 values of 24.5 nM (BA.2) to 106.0 nM (BA.5). This represented 0.15-to 0.66-fold changes compared to WA1, indicating no loss of in vitro RDV antiviral activity against these VOCs. P323L, G671S, and F694Y were shown previously to have no impact on RDV antiviral activity. Similarly, the individual substitution Y521C showed no change in RDV susceptibility in the SARS-CoV-2 replicon system. Conclusion(s): RDV retained potent in vitro antiviral activity against all tested Omicron VOCs with potencies comparable to the WA1 isolate. These data support the continued use of RDV in patients infected with Omicron subvariants.
ABSTRACT
Background: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the cause of the current global pandemic of the COVID-19, which has persisted partly through the emergence of new variants. A non-infectious, convenient, and reproducible in vitro system is needed to assess drug susceptibility of new variants of concern and potential drug resistance mutations. Method(s): The SARS-CoV-2 replicon protocol was adapted and optimized based on {Zhang 2021}. The replicon RNA was produced by in vitro transcription of full-length replicon DNA assembled by ligation of plasmid fragments encoding for the SARS-CoV-2 non-structural proteins (Nsps), nucleoprotein and gaussia luciferase reporter protein. Wild-type and mutant replicon RNAs were transfected into Huh7-1CN cells by electroporation and treated with remdesivir (RDV). To determine EC50 values, luciferase activity was determined at 48 hours post transfection. A recombinant SARS-CoV-2 virus rescue system {Xie 2020} was used to generate matching Nsp mutants for comparison with the replicon system. Result(s): The selected substitutions reflective of Omicron BA.5 sub-lineage BF.7 variant: the triple mutants (Nsp12 (P323L) +Nsp13 (R392C) + Nsp14 (I42V), and a single Nsp12 L247F mutant as well as several specific Nsp12 mutations identified by in vitro resistance selection with RDV or RDV parent nucleoside analog GS-441524 were cloned into the replicon and tested for susceptibility to RDV. RDV inhibited the SARS-CoV-2 wild-type replicon with a mean EC50 value of 14.7 +/- 3.5 nM (N=9). The Nsp12 P323L substitution, a common polymorphism in all major variants of concern including Omicron, was fully susceptible to RDV with a 0.6-fold change in EC50 from the wild-type. The Omicron BF.7 triple mutants and L247F were also fully susceptible to RDV with 0.5- and 0.4-fold changes, respectively. Nsp12 substitutions F480L, V557L, V792I, S759A+V792I, and C799F resulting from in vitro resistance selections with RDV showed minimal to moderate levels of reduced susceptibility to RDV (1.8 to 18.3-fold change) (Table 1). The RDV EC50 fold changes correlated between the noninfectious replicon and recombinant infection virus system (Table 1). Conclusion(s): The replicon system is a convenient and reproducible model to test the susceptibility of SARS-CoV-2 mutant variants to RDV and potentially other antivirals. The common Nsp12 polymorphisms in all variants including the highly transmissible Omicron variant were fully susceptible to RDV.