Antiviral Activity of Remdesivir and Obeldesivir Against SARS-CoV-2 Omicron Subvariants That Were Circulating from September 2023 Through June 2025

Abstract

With the ongoing emergence of SARS-CoV-2 variants, continued surveillance of antiviral susceptibility remains critical for detecting resistance that could compromise treatment efficacy. This study evaluated the activity of 2 SARS-CoV-2 RNA-dependent RNA polymerase (Nsp12) inhibitors against emerging Omicron variants: remdesivir (RDV), an approved antiviral for the treatment of COVID-19, and obeldesivir (ODV), an oral prodrug that shares the same parent nucleoside as RDV. Both RDV and ODV were shown to retain antiviral activity against the Omicron subvariants BA.2.86.1, JN.1.7, KP.2, KP.3.1.1, KP.3.3, LP.8.1, NB.1.8.1, XBB.2, XEC, and XFG compared with wild-type reference strains. Only 1 new lineage-defining Nsp12 substitution, D284Y (detected in NB.1.8.1), was observed. Phenotypic analysis demonstrated that a replicon containing this substitution remained susceptible to both RDV and ODV. These findings are consistent with previous studies showing that RDV and ODV retain potent activity against previously identified Omicron variants, support the continued clinical use of RDV against circulating SARS-CoV-2 variants, and reinforce the potential of ODV as an oral antiviral therapeutic.

1. Introduction

Since the start of the COVID-19 pandemic, new SARS-CoV-2 variants have continuously emerged [1], accompanied by the potential for higher transmissibility, enhanced immune evasion, and lower susceptibility to existing vaccines and antiviral therapeutics [2]. The Omicron variant (Pango lineage B.1.1.529) was first reported in November 2021 and was quickly classified by the World Health Organization as a circulating variant of concern (VOC). Since then, the Omicron variant and its subvariants have been the predominant VOCs, variants of interest (VOIs), and variants under monitoring (VUMs) classified by the World Health Organization [1,3]. The Omicron subvariant BA.2.86 and its descendant lineages, including BA.2.86.1, were classified as a VOI in November 2023. JN.1, a descendant lineage of BA.2.86.1, was designated as a separate VOI in December 2023, when it accounted for up to 66% of globally available sequences. Since then, JN.1 and its subvariants, including JN.1.7, KP.2, KP.3.1.1, LP.8.1, XEC, and XFG, have become the most prevalent VOIs and VUMs, along with NB.1.8.1. By April 2024, JN.1 accounted for up to 95% of globally available sequences. JN.1.7 and KP.2 were classified as VUMs in May 2024. The weekly global prevalence of JN.1.7 peaked at 9% in April 2024, while that of KP.2 peaked at 23% in May 2024. In September 2024, the weekly prevalence of KP.3.1.1 rose to 47%, and XEC, which accounted for 5% of sequences, was classified as a VUM. Weekly KP.3.1.1 prevalence peaked at 51% in October 2024, while the weekly prevalence of XEC continued to rise until January 2025, when it reached a prevalence of 46%. LP.8.1 was designated as a VUM in January 2025 and rose to a maximum weekly prevalence of 36% in April 2025. The NB.1.8.1 variant, derived from the recombinant variant XDV.1.5.1, was designated a VUM in May 2025 and reached a weekly prevalence of 28% in June 2025, when the newly designated VUM XFG became the most common variant globally. In the last week of December 2025, XFG accounted for 75% of submitted sequences [1].

Remdesivir (RDV) is an intravenous nucleotide analog prodrug that is approved to treat COVID-19 in adult and pediatric patients [4,5,6]. Obeldesivir (ODV) is an oral prodrug of the same parent nucleoside of RDV that has demonstrated activity against SARS-CoV-2 in clinical trials [7,8,9]. RDV and ODV, through different metabolic pathways, generate GS-443902, an active nucleoside triphosphate (NTP) that targets Nsp12, the highly conserved SARS-CoV-2 RNA-dependent RNA polymerase [4,7]. Both RDV and ODV have maintained potent antiviral activity against previous Omicron subvariants relative to an ancestral strain [7,10].

The objectives of this study were to evaluate the in vitro antiviral activity of RDV and ODV against recent (September 2023–June 2025) SARS-CoV-2 Omicron subvariants using clinical isolates and Nsp12 site-directed mutants in a replicon system and to conduct a structural analysis of Nsp12 substitutions observed in recent Omicron subvariants to assess their impact on RDV and ODV susceptibility.

2. Materials and Methods

2.1. In Vitro Antiviral Activity Analysis

This study evaluated the in vitro antiviral activity of RDV and ODV against the following Omicron VOIs and VUMs: JN.1.7, KP.2, KP.3.1.1, LP.8.1, NB.1.8.1, XEC, and XFG. Other evaluated Omicron subvariants included KP.3.3 (another variant of KP.3 closely related to KP.3.1.1) and BA.2.86.1 and XBB.2, which were locally circulating variants at the time of the study (San Francisco Bay Area; September 2023–March 2024).

To evaluate the antiviral activity (half-maximal effective concentration [EC₅₀]) of RDV and ODV against clinical isolates of Omicron subvariants BA.2.86.1, JN.1.7, KP.2, KP.3.1.1, KP.3.3, XBB.2, and XEC (Table A1), a nucleoprotein enzyme-linked immunosorbent assay (ELISA) in A549-hACE2-TMPRSS2 cells (a549-hace2tpsa; InvivoGen, San Diego, CA, USA) was used, as described previously [7,10,11]. Detailed information on the materials and methods used in this study can be found in the references [7,10].

For the Omicron subvariants that did not have available clinical isolates (LP.8.1, NB.1.8.1, and XFG), a noninfectious SARS-CoV-2 replicon system (4 DNA fragments isolated from SARS-CoV-2/human/CHN/SH01/2020 [SARS-CoV-2 SH01; GenBank MT121215]) was used to assess the antiviral activity of RDV and ODV. This replicon method has been described previously [12] and applied in phenotypic analyses of SARS-CoV-2 site-directed mutants of components of the replication complex [10,13,14]. The variant-defining substitutions identified in the replication complex genes (Nsp8 to Nsp14) were introduced into the luciferase-expressing replicon. The replicon was then transfected into Huh7-1CN cells, which stably express the SARS-CoV-2 nucleoprotein, by electroporation [15]. These cells were treated with eight 4-fold serial dilutions of RDV (final concentration range: 10 μM–0.6 nM) or ODV (final concentration range: 200 μM–12 nM). Following 48 h of incubation, supernatants were harvested, and luciferase signal was measured to calculate EC₅₀ values for each compound.

EC₅₀ values were defined as the compound concentration at which there was a 50% reduction in nucleoprotein expression (ELISA) or luciferase signal (replicons) relative to infected cells with dimethyl sulfoxide alone (0% inhibition) and uninfected control cells (100% inhibition). EC₅₀ values for RDV and ODV were determined by nonlinear regression curve fitting (GraphPad Prism v8.1.2). For each experiment, the EC₅₀ fold change for variants relative to the relevant WA1 reference was calculated. Fold change across all experiments was then averaged to obtain the final reported values. Experiments were performed twice with technical triplicates for each clinical isolate and replicon. ELISA variability was previously estimated through repeated testing of the wild-type reference (RDV, n = 40; ODV, n = 11) [10]. For RDV, EC₅₀ values for the wild-type clinical isolate ranged from 44.2 nM to 196.8 nM, with a mean of 112.1 nM, a median of 103.4 nM, and a standard deviation of 35.2 nM, corresponding to an estimated variability of 2.8-fold. For ODV, EC₅₀ values for the wild-type clinical isolate ranged from 1.43 μM to 4.17 μM, with a mean of 2.29 μM, a median of 2.11 μM, and a standard deviation of 0.75 μM, corresponding to an estimated variability of 2.9-fold. Variability of the subgenomic replicon assay was also previously estimated through repeated testing of the wild-type reference (RDV, n = 44; ODV, n = 36). RDV EC₅₀ values for the wild-type replicon ranged from 3.7 nM to 30.8 nM, with a mean of 11.3 nM, a median of 10.1 nM, and a standard deviation of 4.6 nM; the estimated assay variability was 2.5-fold. ODV EC₅₀ values for the wild-type replicon ranged from 0.38 μM to 0.97 μM, with a mean of 0.62 μM, a median of 0.61 μM, and a standard deviation of 0.14 μM; the estimated assay variability was 2.3-fold. Variants exhibiting fold-change values below these assay-specific thresholds for RDV or ODV were considered within the inherent variability of the assay.

2.2. Identification and Characterization of Amino Acid Substitutions

The prevalence of Nsp12 substitutions (≥75% of sequences) was evaluated using SARS-CoV-2 sequences from the Global Initiative on Sharing All Influenza Data EpiCoV database. The impact of Nsp12 substitutions on the antiviral activity of RDV and ODV was assessed using the same replicon system described above with site-directed mutants bearing the identified Nsp12 substitutions.

2.3. Structural Analysis of Nsp12 Amino Acid Polymorphisms

Structural analysis of the identified Nsp12 substitutions was conducted on a composite model of cryo-electron microscopy structures of the replication-transcription complex retrieved from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (ID#: 6XEZ and 7UO4) [16,17].

3. Results

3.1. Susceptibility of Clinical Isolates of Omicron Subvariants to RDV and ODV

Mean RDV EC₅₀ values for all clinical isolates of Omicron subvariants (BA.2.86.1, JN.1.7, KP.2, KP.3.1.1, KP.3.3, XBB.2, and XEC) ranged from 21.8 nM (BA.2.86.1) to 87.3 nM (KP.2), with a fold-change range of 0.14 (BA.2.86.1) to 0.63 (KP.2) compared with the WA1 reference (Table 1, Figure 1a). Mean ODV EC₅₀ values ranged from 357 nM (BA.2.86.1) to 1924 nM (KP.2), with a fold-change range of 0.14 (BA.2.86.1) to 0.86 (KP.2) compared with the WA1 reference strain (Table 1, Figure 1b). All EC₅₀ fold-change values were within the variability of the assay.

Table 1.

Susceptibility of Omicron subvariants to RDV and ODV.

	Lineage-Defining Substitutions in the Replication Complex a	RDV		ODV
		Mean ± SD EC50 (nM) b	Mean ± SD Fold Change c from Wild-Type Reference d	Mean ± SD EC50 (nM) b	Mean ± SD Fold Change c from Wild-Type Reference d
Variant lineage (clinical isolates)
SARS-CoV-2 reference strain (WA1; lineage A)		Range, 72–191	1.00	Range, 1512–3061	1.00
BA.2.86.1	Nsp9 T35I Nsp12 P323L Nsp13 R392C Nsp14 I42V	21.8 ± 13.7	0.14 ± 0.04	357 ± 124	0.14 ± 0.07
JN.1.7	Nsp9 T35I Nsp12 P323L Nsp13 R392C Nsp14 I42V	57.2 ± 7.4	0.48 ± 0.09	1107 ± 52	0.50 ± 0.03
KP.2	Nsp9 T35I Nsp12 P323L Nsp13 R392C Nsp14 I42V	87.3 ± 2.0	0.63 ± 0.17	1924 ± 84	0.86 ± 0.19
KP.3.1.1	Nsp9 T35I Nsp12 P323L Nsp13 R392C Nsp14 I42V	40.2 ± 9.1	0.45 ± 0.02	904 ± 56	0.49 ± 0.19
KP.3.3	Nsp9 T35I Nsp12 P323L Nsp13 R392C Nsp14 I42V	63.8 ± 2.9	0.46 ± 0.11	1099 ± 410	0.51 ± 0.27
XBB.2	Nsp12 P323L Nsp12 G671S Nsp13 S36P Nsp13 R392C Nsp14 I42V	43.4 ± 26.4	0.27 ± 0.08	605 ± 267	0.24 ± 0.14
XEC	Nsp9 T35I Nsp12 P323L Nsp13 R392C Nsp14 I42V	25.5 ± 1.6	0.30 ± 0.10	504 ± 78	0.26 ± 0.05
Variant lineage (replicon system) e
SARS-CoV-2 reference (SH01; lineage B)		Range, 10.9–14.1	1.00	Range, 376–658	1.00
LP.8.1 XFG	Nsp9 T35I Nsp12 P323L Nsp13 R392C Nsp14 I42V	12.6 ± 1.0	1.14 ± 0.09	465 ± 45	1.05 ± 0.15
NB.1.8.1	Nsp9 P57S Nsp9 P80L Nsp12 D284Y Nsp12 P323L Nsp12 G671S Nsp13 S36P Nsp13 R392C Nsp14 I42V	16.6 ± 0.3	1.19 ± 0	762 ± 244	1.17 ± 0.38

EC₅₀, half-maximal effective concentration; ODV, obeldesivir; RDV, remdesivir. ^a The characteristic mutations for a lineage are defined as nonsynonymous substitutions or deletions that occur in ≥75% of sequences within that lineage. ^b Mean EC₅₀ values are based on the results of ≥2 independent experiments. ^c Mean ± SD fold-change values are based on the results of ≥2 independent experiments. Fold change was calculated for each experiment, and a mean fold change was calculated with these values. ^d For clinical isolates, the wild-type SARS-CoV-2 reference strain was the WA1 strain (lineage A). For the replicon system, the wild-type SARS-CoV-2 reference strain was the SH01 replicon (lineage B). ^e EC₅₀ values from subvariants tested using the replicon system cannot be directly compared with variants tested using clinical isolates due to differences in assay reagents, conditions, and detection technology. Fold-change values, which are independent of the assay used, offer a more accurate basis for comparison.

Figure 1.

Percent inhibition against clinical isolates of the Omicron subvariants BA.2.86.1, JN.1.7, KP.2, KP.3.1.1, KP.3.3, XBB.2, and XEC compared with the WA1 reference strain by (a) RDV and (b) ODV and against replicons containing lineage-defining substitutions of the Omicron subvariants LP.8.1, NB.1.8.1, and XFG compared with the SH01 reference strain by (c) RDV and (d) ODV. Each subvariant was tested 2 times with technical triplicates. Data are presented as mean ± SD. The range of RDV or ODV activity against the WA1 reference strain (8 replicates) or the SH01 reference strain (4 replicates) is shaded in gray. Percent inhibition is relative to DMSO. DMSO, dimethyl sulfoxide; ODV, obeldesivir; RDV, remdesivir; WT, wild-type.

3.2. Susceptibility of Omicron Subvariants in the Replicon System to RDV and ODV

Clinical isolates for the Omicron subvariants LP.8.1, NB.1.8.1, and XFG were not available. Therefore, the lineage-defining substitutions observed in the replication complex were introduced into a replicon by site-directed mutagenesis to assess RDV and ODV antiviral activity (Table 1). The replicon containing the lineage-defining substitutions for LP.8.1 and XFG had mean RDV and ODV EC₅₀ values of 12.6 nM and 465 nM, respectively, corresponding to an EC₅₀ fold change of 1.14 for RDV and 1.05 for ODV (Table 1). The replicon containing the lineage-defining substitutions for NB.1.8.1 had mean RDV and ODV EC₅₀ values of 16.6 nM and 762 nM, respectively, corresponding to an EC₅₀ fold change of 1.19 for RDV and 1.17 for ODV. All EC₅₀ fold-change values were within the variability of the assay.

3.3. Characterization of Nsp12 Amino Acid Substitutions Observed in Omicron Subvariants

Unique substitutions in Nsp12 were observed following genomic analysis of >17 million SARS-CoV-2 sequences compared with the WA1 reference strain. One Nsp12 substitution, D284Y, was observed in ≥75% of Nsp12 sequences of the Omicron subvariant NB.1.8.1. D284Y was the only new lineage-defining substitution in Nsp12 identified in BA.2.86.1, JN.1.7, KP.2, KP.3.1.1, KP.3.3, LP.8.1, NB.1.8.1, XBB.2, XEC, or XFG compared with earlier Omicron subvariants [10]. This Nsp12 substitution was introduced in the replicon system and resulted in mean RDV and ODV EC₅₀ values of 11.8 nM and 570 nM, respectively, corresponding to EC₅₀ fold changes of 0.85 and 0.87, respectively (Table 2). Lineage-defining substitutions in other regions of the replication complex (Nsp9, Nsp10, Nsp13, and Nsp14) were observed in the clinical isolates tested; these substitutions did not impact the susceptibility to RDV and ODV (Table 1).

Table 2.

Genotypic and phenotypic characterization of Nsp12 substitutions observed in Omicron subvariants at ≥75% frequency.

Nsp12 Lineage-Defining Substitution	Omicron Lineage	Frequency (%) a	RDV		ODV
			Mean ± SD EC50 (nM) b	Mean ± SD Fold Change from SH01 Reference Strain c	Mean ± SD EC50 (nM) b	Mean ± SD Fold Change from SH01 Reference Strain c
SH01 reference strain	B	–	13.9 ± 0.3	1.00	654 ± 5	1.00
D284Y	NB.1.8.1	99.8	11.8 ± 0.5	0.85 ± 0.05	570 ± 28	0.87 ± 0.05

EC₅₀, half-maximal effective concentration; ODV, obeldesivir; RDV, remdesivir. ^a Frequency in the GISAID (Global Initiative on Sharing All Influenza Data) database as of 4 September 2025. ^b Mean EC₅₀ values are the mean of ≥2 independent experiments. ^c Fold change was calculated for each experiment, and a mean fold change was calculated with these values.

3.4. Structural Analysis of Identified Nsp12 Polymorphisms

The Nsp12 lineage-defining polymorphisms D63N, Y273H, P323L, G671S, and G823insD were identified in previous studies [10,11]. Prior analyses identified P323L in Alpha, Delta, and Omicron variants; G671S in Delta and earlier Omicron variants; and D63N, Y273H, and G823insD in earlier Omicron variants. The Nsp12 lineage-defining substitution D284Y, along with these previously identified Nsp12 lineage-defining polymorphisms, showed no direct interaction with the incoming active NTP metabolite of RDV and ODV or the viral RNA based on the structural analysis (Figure 2).

Figure 2.

Map of Nsp12 amino acid polymorphisms observed in ≥75% of Omicron subvariants. The structure is a model of the active NTP in the SARS-CoV-2 active site based on the cryo-electron microscopy structures 6XEZ [16] and 7UO4 [17]. The pre-incorporated active NTP metabolite of RDV and ODV is shown in cyan. Nsp12 is shown in green, Nsp7 is shown in white, Nsp8 is shown in yellow, and Nsp13 is shown in orange. Amino acid substitutions located in this view of the structure are shown in magenta. The template RNA strand is shown in blue and the nascent RNA strand is shown in red. NTP, nucleoside triphosphate; ODV, obeldesivir; RDV, remdesivir.

4. Discussion

The emergence of new SARS-CoV-2 variants continues to coincide with periodic surges in COVID-19 cases, hospitalizations, and deaths [1,18]. COVID-19–related deaths and hospitalizations have decreased markedly worldwide since 2021 [1], primarily due to population-level immunity against the prevalent SARS-CoV-2 variants and improved clinical management [2]. However, as SARS-CoV-2 continues to evolve, there is an ever-present risk of a more virulent and highly transmissible variant emerging; in a worst-case scenario, existing vaccines and pre-existing immunity would not be as effective at reducing the severity of infection with such a variant [2]. For this reason, it is critical to regularly test the activity of antiviral medications such as RDV to confirm that newer SARS-CoV-2 variants remain susceptible to treatment.

In this study, phenotypic analyses of recent Omicron subvariants demonstrated that RDV and ODV retained potent in vitro antiviral activity against the SARS-CoV-2 Omicron subvariants BA.2.86.1, JN.1.7, KP.2, KP.3.1.1, KP.3.3, LP.8.1, NB.1.8.1, XBB.2, XEC, and XFG relative to wild-type reference strains. The lineage-defining Nsp12 substitution D284Y, identified in the Omicron subvariant NB.1.8.1, remained susceptible to RDV and ODV.

The Nsp12 substitution P323L was observed at a frequency of ≥75% in all Omicron subvariants evaluated in the current study. This substitution was first observed as early as January 2020 [19] and has been observed at a frequency of >99% in a previous analysis of the B.1.1.7 (Alpha), B.1.617.2 (Delta), and B.1.1.529 (Omicron) variants [11]. Additionally, the current analysis observed the Nsp12 substitution G671S at a frequency of ≥75% in the XBB.2 and NB.1.8.1 variants. G671S has been detected in earlier variants, including in 98% of B.1.617.2 variants [11]. Phenotypic analyses in the current study suggest minimal impact of P323L and G671S on the efficacy of RDV or ODV, in agreement with prior studies [10,11].

These findings are consistent with previous studies that showed RDV and ODV retained potent activity against earlier Omicron subvariants [7,10,11]. These results provide evidence to support the continued use of RDV for the treatment of COVID-19 arising from infection with Omicron subvariants.

Abbreviations

The following abbreviations are used in this manuscript:

BEI Resources	Biodefense and Emerging Infections Research Resources Repository
DMSO	dimethyl sulfoxide
EC50	half-maximal effective concentration
ELISA	enzyme-linked immunosorbent assay
GISAID	Global Initiative on Sharing All Influenza Data
JHU	Johns Hopkins University
NTP	nucleoside triphosphate
ODV	obeldesivir
RDV	remdesivir
VOI	variant of interest
VUM	variant under monitoring
WT	wild-type

Appendix A

Table A1.

SARS-CoV-2 clinical isolates.

Variant	Isolate	Source
BA.2.86.1	hCoV-19/USA/CA-GS140831/2024	Gilead Sciences, Inc.
JN.1.7	hCoV-19/USA/MD-HP51675- PIDFZBZXSK/2024	JHU, Pekosz Lab
KP.2	hCoV-19/USA/MD-HP51511- PIDEACTNPM/2024	JHU, Pekosz Lab
KP.3.1.1	hCoV-19/USA/MD-HP51826- PIDNPNPNON/2024	JHU, Pekosz Lab
KP.3.3	SARS-CoV-2/USA/MD-HP51771/2024	JHU, Pekosz Lab
WA1 (lineage A)	hCoV-19/USA-WA1/2020	BEI Resources (Catalog # NR-52281)
XBB.2	hCoV-19/USA/CA-GS136871/2024	Gilead Sciences, Inc.
XEC	HP50294	JHU, Pekosz Lab

BEI Resources, Biodefense and Emerging Infections Research Resources Repository, Bethesda, MD, USA; JHU, Johns Hopkins University, Baltimore, MD, USA.

Author Contributions

Conceptualization, L.R. and C.H.; methodology, L.R., J.L., D.H., J.K.P. and C.H.; validation, L.R., J.L. and C.H.; formal analysis, L.R., J.L., D.H., N.P., C.M., P.Y.H., J.L.R.Z. and J.K.P.; investigation, L.R. and J.L.; resources, L.R., J.L. and C.H.; data curation, L.R., J.L., D.H., N.P., C.M., P.Y.H. and J.L.R.Z.; writing—original draft preparation, L.R. and C.H.; writing—review and editing, L.R., J.L., D.H., N.P., C.M., P.Y.H., J.L.R.Z., R.M., J.P.B., J.K.P. and C.H.; visualization, L.R. and J.K.P.; supervision, C.H.; project administration, C.H.; funding acquisition, C.H. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Gilead Sciences shares anonymized individual patient data upon request or as required by law or regulation with qualified external researchers based on submitted curriculum vitae and reflecting non-conflict of interest. The request proposal must also include a statistician. Approval of such requests is at Gilead Sciences’ discretion and is dependent on the nature of the request, the merit of the research proposed, the availability of the data, and the intended use of the data. Data requests should be sent to datasharing@gilead.com.

Conflicts of Interest

L.R., J.L., D.H., N.P., C.M., P.Y.H., R.M., J.K.P. and C.H. are stockholders and employees of Gilead Sciences, Inc. J.L.R.Z. and J.P.B. are stockholders and former employees of Gilead Sciences, Inc. The funders contributed to the design of the study; the collection, analyses, or interpretation of data; the writing of the manuscript; and the decision to publish the results.

Funding Statement

This research was funded by Gilead Sciences, Inc.