Genotypic and phenotypic characterization of influenza A viral variants in study participants treated with pimodivir in the phase 2b TOPAZ study

Abstract

Background

Pimodivir is a first-in-class polymerase basic protein 2 (PB2) subunit inhibitor of the influenza A polymerase complex. The randomized double-blinded placebo-controlled phase 2b TOPAZ study demonstrated antiviral activity and safety of twice daily pimodivir alone (300 mg, 600 mg) or in combination with oseltamivir (pimodivir 600 mg, oseltamivir 75 mg) in adult study participants with acute uncomplicated influenza A. The detailed genotypic and phenotypic characterization of viral variants observed in this study are reported.

Methods

Population sequencing of PB2 and neuraminidase genes, and phenotypic susceptibility testing, were performed using baseline and last virus-positive post-baseline nasal swab samples.

Results

Sequencing of baseline samples in 206 of 223 (92.4%) randomized study participants with confirmed influenza A infection identified no polymorphisms at any predefined PB2 positions of interest for pimodivir and no phenotypic reduced susceptibility to pimodivir was observed. Post-baseline sequencing data for 105/223 (47.1%) participants identified emergence of PB2 mutations at amino acid positions of interest in 10 (9.5%) participants (pimodivir 300 mg: n = 3; 600 mg: n = 6; combination: n = 1; placebo: n = 0) and included positions S324, F325, S337, K376, T378, and N510. These emerging mutations were typically associated with decreased pimodivir susceptibility, but not viral breakthrough. No reduced phenotypic susceptibility was observed in the one (1.8%) participant with emerging PB2 mutations from the pimodivir plus oseltamivir group.

Conclusions

Participants with acute uncomplicated influenza A treated with pimodivir in the TOPAZ study infrequently developed reduced susceptibility to pimodivir and combining pimodivir with oseltamivir further decreased the risk of reduced susceptibility development.

Keywords

influenza A pimodivir viral variants antiviral therapy polymerase basic protein 2 subunit inhibitor clinical trial TOPAZ

Introduction

Emergence of reduced antiviral drug susceptibility remains a major concern for seasonal influenza with a significant annual incidence, hospitalization rate, and associated mortality, accompanied by limited treatment options in the event that drug class resistance develops.^1,2 The US Food and Drug Administration (FDA) approved anti-influenza drugs that are M2 ion channel inhibitors (M2Is), neuraminidase inhibitors (NAIs), and a cap-dependent endonuclease inhibitor for their potential to treat uncomplicated, self-limiting infections and for use only in the outpatient setting.^3-5

NAIs are the mainstay in the management of seasonal influenza, although they possess some limitations and concerns. Continuous antigenic drift, due to accumulation of influenza A virus mutations over time, contributes to the emergence of drug-resistant viral strains and can impact the efficacy of antivirals.⁵ The first generation M2Is (the adamantanes, amantadine, and rimantadine) previously showed antiviral activity against influenza A strains at micromolar concentrations, but the rapid and widespread emergence of drug-resistant viral strains has restricted their use in prevention and treatment of influenza.⁶ Neuraminidase inhibitors (zanamivir, peramivir, and oseltamivir), which stop the release of new virus particles from infected cells, have antiviral activity against influenza A and B viruses with a lower frequency of reduced susceptibility development compared to M2Is.⁷ However, oseltamivir-resistant influenza strains have emerged. Through 2007, fewer than 1% of circulating A (H1N1) strains were oseltamivir-resistant, but during the 2008–2009 influenza season this increased to 95% and was not associated with oseltamivir use, indicating a lack of drug pressure in emergence of oseltamivir-resistant A (H1N1) strains.^8-10 While this influenza strain was displaced by the emerging A (H1N1) variant in subsequent seasons, its identification demonstrated the potential for widespread NAI resistance to develop. Another limitation of NAIs is the narrow treatment window of 48 hours between symptom onset and treatment initiation.

Hence, the current therapeutic landscape indicates the need for new antiviral drugs effective against influenza, with activity against known NAI-resistant strains. The cap-dependent endonuclease inhibitor, baloxavir, recently approved for acute uncomplicated influenza, has demonstrated antiviral efficacy but there is evidence of emergence of viral mutants with approximately 10% of adult study participants treated with baloxavir showing I38 T/M in the polymerase acid (PA) protein in phase 3 studies.^11,12 There is also lack of benefit demonstrated on the efficacy of baloxavir alone or in combination with other antiviral therapies in hospital settings.¹³ Like NAIs, treatment initiation with baloxavir must begin within 48 hours of symptom onset.

Pimodivir (JNJ-63623872, formerly VX-787), a novel non-nucleotide polymerase inhibitor of influenza A virus replication, blocks the polymerase basic protein (PB2) cap-snatching activity of the influenza A viral polymerase complex; thereby uniquely targeting a different viral protein than current FDA approved influenza A therapeutics.^14-16 Pimodivir has shown in vitro antiviral activity against influenza A virus, including pandemic A (H1N1), A (H5N1), and A (H7N9), as well as influenza A strains with reduced susceptibility to NAIs,^16,17 and has shown little activity against influenza B. In a phase 2a human challenge study, pimodivir 1200/600 mg dose demonstrated efficacy in reducing viral load and improved the clinical symptoms in volunteers inoculated with influenza A (H3N2) virus.¹⁴ In this proof-of-concept study, 9.7% of volunteers harboured viral strains with reduced pimodivir susceptibility. The only PB2 variant observed in multiple volunteers was M431I (n = 4). The PB2 amino acid substitutions S324C, K376R, and M431L/R/V, associated with decreased pimodivir susceptibility in vitro, were observed in viral strains obtained from individual volunteers at single time points.¹⁴

In the dose-response phase 2b TOPAZ study, the combination of pimodivir 600 mg and oseltamivir 75 mg given twice daily (bid) or pimodivir alone (300 mg or 600 mg bid) demonstrated a significant and dose-dependent decrease in viral load in study participants with acute uncomplicated influenza A infection.¹⁸ Summary data of pimodivir genotypic and phenotypic susceptibility was reported previously.¹⁸ Here we report a more detailed exploratory analysis of the genotypic and phenotypic changes observed in the influenza A viral strains obtained from study participants with acute, uncomplicated seasonal influenza from TOPAZ, treated with pimodivir.

Methods

Study design and study participants

The study design and participant population of the TOPAZ study were described previously.¹⁸ Briefly, this was a phase 2b, double-blind, placebo-controlled, parallel-group study in adult study participants infected with influenza A who were randomized 1:1:1:1 to receive a twice-daily oral dose of pimodivir 300 mg, pimodivir 600 mg, pimodivir 600 mg + oseltamivir 75 mg, or placebo for 5 days. The study had a planned interim analysis at the end of the second influenza season (2015–2016) in the northern hemisphere and was stopped early for success as the primary endpoint of viral load area under the curve was met. The study was conducted in compliance with the Declaration of Helsinki and approved by the Institutional Review Board (NCT02342249). Written informed consent was obtained from all study participants prior to enrolling into the study.

Viral load and viral titre assays

The viral load was determined using a quantitative reverse transcription–polymerase chain reaction (qRT-PCR, performed by DDL Diagnostic Laboratory, Rijswijk, The Netherlands). The limit of quantification (LOQ) for this assay was 4.0 log₁₀ copies/mL, and the limit of detection (LOD) was 3.48 log₁₀ copies/mL. Results below the LOQ and above the LOD (“target detected”) were imputed with 3.75 log₁₀ copies/mL; results below the LOD (“target not detected”) were imputed with zero. The viral titre was determined using a cell culture assay with MDCK cells and the read-out was haemagglutination inhibition (HAI) for A (H1N1) subtypes or nucleoprotein ELISA for A (H3N2) subtypes. The LOD and LOQ values for this assay were 0.75 log ₁₀TCID50/mL, and results below LOD were imputed with 0.375 log₁₀TCID50/mL.

Genotypic analysis of influenza A viral strains

The genotypic and phenotypic characterization of influenza A virus strains was performed using nasal swab samples from participants enrolled in the study. The genotypic analysis was carried out using Sanger sequencing of PB2 and NA genes with a nested polymerase chain reaction (PCR) technology (Viroclinics Biosciences, Rotterdam, NL). Sequencing of PB2 and NA genes was carried out for each study participant both at baseline and the last virus-positive post-baseline sample, and the assay generated near full-length sequences with >80% efficiency if the viral load was above the threshold >5.15 log₁₀ copies/mL (influenza A specific qRT-PCR determined by DDL Diagnostic Laboratory). Changes in PB2 and NA sequences were evaluated against the following reference strains: A (H1N1): A/California/07/2009 [GenBank CY121687] and A (H3N2): A/Texas/50/2012 [GenBank KJ942623] for PB2; A (H1N1): A/California/07/2009 [GenBank HM138502]; A (H3N2): A/Texas/50/2012 [GenBank KJ942618] (NH season 2014–2015) and A (H3N2): A/Switzerland/9715293/2013 [Gisaid Epiflu:EPI614443] (NH season 2015–2016) for NA. The reference sequence differences between A/Texas/50/2012 and A/Switzerland/9715293/2013 were minimal with no differences observed for PB2. For NA, three amino acids (at positions 150, 221, and 392) were different between the Texas and Switzerland strains but none of those were associated with reduced susceptibility to NAI.

In addition to studying all changes versus the reference strain, a directed analysis was performed using a list of PB2 amino acid positions of interest which are associated with reduced pimodivir susceptibility based on in vitro data,¹⁶ clinical data,¹⁴ or known contact residues for the compound with the cap-binding region of PB2.¹⁵ At the time of the statistical analysis of the clinical study, this list included the following amino acid positions: Q306, F323, S324, F325, S337, H357, F363, K376, F404, Q406, M431, T378, and N510. Known oseltamivir NA amino acid substitutions associated with reduced susceptibility included the following (non-exhaustive list): D199 N, I223 R, H275Y, and N295S for N1 strains, and E119V, H274Y, R292K, and N294S for N2 strains.¹⁹ Emergence was defined as the presence of an amino acid substitution at the post-baseline time point but not at baseline.

Phenotypic analysis of influenza A viral variants

The baseline and last virus-positive post-baseline sample was used for the phenotypic analysis. Pimodivir susceptibility, expressed as fold change in 50% effective concentration (EC₅₀) values, was assessed using the cell-culture–based assay (Influenza ViroSpot assay, Viroclinics Biosciences, Rotterdam, The Netherlands) with Madin-Darby canine kidney (MDCK) cells.²⁰ In brief, MDCK cell monolayers were inoculated with virus (aiming at 100 infectious units per well) and incubated in absence and presence of increasing concentrations of pimodivir. After incubation, cells were fixed with formalin (10%) and treated with Triton X-100 (0.5%) to permeabilize. Cells were immunostained with a murine monoclonal antibody directed against influenza A virus nucleoprotein (HB65, EVL, Woerden, The Netherlands), HRP-conjugated goat-anti-mouse as secondary antibody (Invitrogen, Foster City, CA, USA), and TrueBlue substrate (KPL, Gaitherburg, MD, USA). Plates were scanned using the ImmunoSpot Image Analyzer (Cellular Technology Limited ImmunoSpot, Cleveland, OH, USA). Well areas covered by spots per well at each antiviral compound concentration were quantified using BioSpot ®software (Cellular Technology Limited, Cleveland, OH, USA). EC₅₀ values are calculated using nonlinear regression analysis with GraphPad (GraphPad Software, Boston, MA, USA). Oseltamivir susceptibility, expressed as fold change in 50% inhibitory concentration (IC₅₀) values, was determined using the influenza NA Inhibitor Reagent kit NA Star^® (Applied Biosystems, performed by Viroclinics Biosciences) using a chemiluminescent substrate to read-out NA enzyme activity according to the manufacturer’s instructions. The fold change (FC) in EC₅₀ values for pimodivir versus the reference strains A (H1N1) A/California/07/2009, or A (H3N2) A/Texas/50/2012 (season 1) or A (H3N2) A/Switzerland/9715293/2013 (season 2) for the Virospot assay, determined in 4 replicates, was calculated in each experiment. The mean (SD) pimodivir EC50 values for the A (H1N1) and A (H3N2) reference strains were 0.17 nM (0.07 nM) and 0.24 nM (0.13 nM), respectively. The FC in IC50 value in the NA Star assay versus the reference strain (A (H1N1) A/PR8/8/34) was calculated similarly, with the mean (SD) oseltamivir IC50 values for the A (H1N1) and A (H3N2) reference strains being 0.43 nM (0.06 nM) and 0.41 nM (0.07 nM), respectively. Decreased phenotypic susceptibility to pimodivir and oseltamivir was defined using a preliminary biological cutoff of FC >4.0 (based on assay variation using repeat testing of wild-type strains).

Site-directed mutants

Site-directed mutant strains with selected mutations introduced at selected positions in the PB2 gene were constructed using a plasmid-based reverse genetics system with an A (H1N1) A/WSN/1933 background.²¹ 293T cells were transfected with each plasmid vector to generate the mutant virus strains which were then used to infect MDCK cells. After virus titration and viral sequencing to confirm the presence of the mutation, each viral strain was tested for pimodivir susceptibility in a 3-day cytopathic effect (CPE-based) assay using MDCK cells, in at least 3 independent experiments. Each of the selected PB2 amino acid substitutions was tested individually. The replication capacity (RC) of each site-directed mutant virus was evaluated by infecting the MDCK cells with the respective mutant or wild-type strain at a multiplicity of infection of 0.01 and the supernatants were titrated over 62 hours, in the absence of compound. The RC was then calculated as the percentage of the mutant strain mean titre divided by wild-type mean titre at 48-hours following infection.

The naturally occurring frequency of viruses with amino acid substitutions associated with reduced susceptibility to pimodivir was evaluated using a panel of approximately 8900 influenza A PB2 amino acid sequences available in Genbank.²²

Results

Patient disposition and baseline characteristics

Of 967 study participants screened, 223 (23.1%) with confirmation of influenza A infection by PCR received either placebo (n = 51, 22.9%), pimodivir 300 mg (n = 58, 26.0%), pimodivir 600 mg (n = 57, 25.6%), or pimodivir 600 mg + oseltamivir 75 mg (n = 57, 25.6%) and were included in the full analysis set.¹⁸ A negative rapid influenza diagnostic test for influenza A virus was the most common criteria for exclusion. Of the full analysis set of 223 participants, 113 (50.7%) were female, median age was 41 years, and 118 (84.3%) were white. The demographics and baseline characteristics were balanced across the treatment groups.¹⁸ At baseline, 23 (47.9%), 29 (52.7%), 28 (50.0%), and 30 (54.5%) participants were infected with A (H3N2) in the placebo, pimodivir 300 mg, pimodivir 600 mg, and pimodivir 600 mg plus oseltamivir 75 mg groups, respectively. A (H1N1) infection at baseline was present in 25 (52.1%), 26 (47.3%), 28 (50.0%), and 25 (45.5%) participants in the placebo, pimodivir 300 mg, pimodivir 600 mg, and pimodivir 600 mg plus oseltamivir 75 mg groups, respectively.¹⁸

Viral resistance at baseline: Genotype and phenotype

At baseline, 206 (92.4%) of the 223 study participants had available sequencing results for PB2 and NA genes. Baseline polymorphisms in PB2 and NA regions that were present in the viruses from ≥10% participants are shown in Supplementary Table 1. None of these PB2 baseline polymorphisms were at any of the predefined positions of interest. At baseline, two participants harboured a virus strain with a known oseltamivir resistance mutation in NA: one participant in the 300 mg pimodivir group showed H275Y and one participant in the 600 mg pimodivir group showed D199 N (Supplementary Table 1).

At baseline, pimodivir and oseltamivir FC in EC₅₀/IC₅₀ values were available for 80.7% (180 of 223) and 73.5% (164 of 223) of study participants, respectively (Table 1). All baseline pimodivir FC values were <4.0 across the treatment groups, while 3 participants showed a baseline oseltamivir FC value >4.0. One of these 3 participants harboured a virus strain with H275Y at baseline and with an oseltamivir FC value of 218, the second participant (with oseltamivir FC value = 261) had no baseline genotype available, but showed H275Y in the Day 3 sample, and the last participant (with oseltamivir FC = 8.8) did not show any known oseltamivir mutation.

Table 1.

Mean (SD) baseline and post-baseline fold-change in EC₅₀/IC₅₀ (full analysis set).

	Placebo BID	Pimodivir 300 mg BID	Pimodivir 600 mg BID	Pimodivir 600 mg + oseltamivir 75 mg BID
Full analysis set	51	58	57	57
Pimodivir baseline EC₅₀ fold change
N	42	48	46	44
Mean (SD)	0.82 (0.51)	0.79 (0.43)	0.78 (0.70)	0.70 (0.36)
Median	0.76	0.74	0.63	0.60
Range	0.1, 2.7	0.1, 2.1	0.1, 3.3	0.2, 2.2
N with FC > 4	0	0	0	0
Pimodivir post-baseline EC₅₀ fold change
N	22	9	8	5
Mean (SD)	0.66 (0.31)	55.86 (82.52)	73.39 (125.61)	0.49 (0.23)
Median	0.67	2.90	27.36	0.53
Range	0.1, 1.5	0.1, 240.2	0.1, 372.1	0.2, 0.7
N with FC > 4	0	4	5	0
OST baseline IC₅₀ fold change
N	36	44	44	40
Mean (SD)	1.01 (0.39)	5.91 (32.68)	1.02 (0.42)	7.73 (41.10)
Median	0.89	0.89	0.90	0.95
Range	0.5, 2.4	0.4, 217.7	0.3, 2.4	0.4, 261.1
N with FC > 4	0	1	0	2
OST post-baseline IC₅₀ fold change
N	18	4	4	4
Mean (SD)	0.98 (0.31)	1.09 (0.67)	1.00 (0.30)	1.08 (0.33)
Median	0.89	0.90	0.88	1.06
Range	0.6, 1.8	0.6, 2.0	0.8, 1.4	0.7, 1.5
N with FC > 4	0	0	0	0

BID, twice-daily; EC₅₀, 50% effective concentration; FC, fold change; IC₅₀, 50% inhibitory concentration; OST, oseltamivir; SD, standard deviation.

Reference strains (H1N1 A/California/07/2009, or H3N2 A/Texas/50/2012 (season 1) or H3N2 A/Switzerland/9715293/2013 (season 2).

Mean (SD) pimodivir EC50 values for the A(H1N1)pdm09 and A(H3N2) reference strains were 0.17 nM (0.07 nM) and 0.24 nM (0.13 nM), respectively. The mean (SD) oseltamivir IC50 value for the H1N1 A/PR8/8/34 and H3N2 reference strains were 0.43 nM (0.06 nM) and 0.41 nM (0.07 nM), respectively.

Genotypic changes in influenza A viral variants

Paired baseline/post-baseline data for PB2 and NA genes were available for 105/223 (47.1%) and 109/223 (48.9%) study participants, respectively (PB2 gene: n = 30, 30, 28, and 17, NA gene n = 31, 30, 30, and 18, for the placebo, pimodivir 300 mg, pimodivir 600 mg, and combination groups, respectively). The only reason for not obtaining a sequence was a too low viral load, which occurred more often in the combination group. Emergence of PB2 amino acid substitutions known to be associated with reduced susceptibility to pimodivir was observed in virus strains from 10/105 (9.5%) participants (pimodivir 300 mg [n = 3], pimodivir 600 mg [n = 6], and combination group [n = 1]) and at the following PB2 positions: S324, F325, S337, K376, T378, and N510 (Table 2). The study participant in the combination group who showed the emergence of a F325L substitution also showed the presence of V504V/A. 9 of 10 participants with emerging PB2 substitutions were infected with A (H1N1), and one with A (H3N2). One participant (pimodivir 600 mg) had a viral strain with N295S in NA, a known oseltamivir resistance associated substitution, in the absence of oseltamivir treatment (data not shown). No other emerging amino acid substitutions in NA were observed.

Table 2.

Emerging genotypic and phenotypic reduced susceptibility to pimodivir.

Study participant	Type of influenza A	Treatment group	Sample day (genotypic/phenotypic)	Emerging PB2 mutations^a	FC valuePimodivir^b
01	H1N1pdm09	Pimodivir 600 mg + oseltamivir 75 mg	4/4	F325F/L^a, V504V/A	0.5
02	H1N1pdm09	Pimodivir 600 mg	4/4	S337P^a	>372.0
03	H1N1pdm09	Pimodivir 600 mg	6/6	K376R^a	90.3
04	H1N1pdm09	Pimodivir 600 mg	3/3	S324S/K/N/R^a, N510N/K^a	69.0
05	H1N1pdm09	Pimodivir 600 mg	3/4	S324N^a	45.3
06	H1N1pdm09	Pimodivir 600 mg	6/4	K376K/R^a	9.4
07	H3N2	Pimodivir 600 mg	6/-	K376N^a	Not available
08	H1N1pdm09	Pimodivir 300 mg	4/4	T378S^a	240.2
09	H1N1pdm09	Pimodivir 300 mg	8/4	S324S/N^a	127.2
10	H1N1pdm09	Pimodivir 300 mg	-/4	No sequence available	70.8
11	H1N1pdm09	Pimodivir 300 mg	6/6	S324N^a	60.4

FC, fold change; n/a, not available; PB2, polymerase basic protein 2.

^aPB2 positions of interest.

^bEmergence of phenotypic reduced susceptibility to pimodivir is defined as FC >4.0.

PB2 positions of interest: Q306, F323, S324, F325, S337, H357, F363, K376, T378S, F404, Q406, M431, and N510.

Oseltamivir amino acid substitutions associated with reduced susceptibility: D199N, I223R, H275Y, and N295S for N1; E119V, H274Y, R292K, and N294S for N2.

Phenotypic changes in influenza A viral variants

Reduced phenotypic susceptibility occurred in 9 study participants (pimodivir 600 mg [n = 5], pimodivir 300 mg [n = 4]) and was associated with genotypic changes in PB2. Emergence of the PB2 substitutions S324K/N/R, S337P, K376N/R, T378S, or N510K was associated with increased pimodivir fold change in EC50 values (FC value range 9.4 to >372.0). The one participant in the combination group who showed F325F/L, along with V504V/A, did not show reduced susceptibility to pimodivir (FC = 0.5) (Table 2). Individual viral load profiles are shown in Figure 1 and demonstrate that the emergence of genotypic or phenotypic reduced susceptibility to pimodivir is not clearly associated with a virologic rebound.

Figure 1.

Individual load and viral titre profiles of eleven study participants with genotypic changes or phenotypic reduced susceptibility to pimodivir. Influenza viral load was determined by qRT-PCR in nasopharyngeal swab specimens from days 1, 3, 4, 6, and 8 of TOPAZ. The limit of quantification (LOQ) was 4.0 log₁₀ copies/mL and the limit of detection (LOD) was 3.48 log₁₀ copies/mL. Results below the LOQ and above the LOD were imputed with 3.75 log₁₀ copies/mL; results below the LOD were imputed with zero. Viral titre was determined using an MDCK cell culture assay with LOD 0.75 log₁₀ copies/mL, results below LOD were imputed with 0.375 log₁₀ copies/mL.

Site-directed mutants in PB2 were available for S324N, S324R, S337P, K376R, T378S, M431R, N510K, and N510T and demonstrated pimodivir FC values of 84, 57, 4.9, 100, 1.2, >3333, 514, and 77.3, respectively, when tested in a phenotypic assay (Table 3). The phenotypic resistance observed in the viral strain with T378S could not be replicated using the site-directed mutant as no viral strain could be recovered after transfection. The replication capacity of these mutant viral strains was low. The percentage of viral titres as compared to wild-type virus ranged between 1% and 18% (S324N [18%], S324R [18%], S337P [1%], K376R [10%], N510K [1%]). The generation of other mutant strains using reverse genetics was not possible due to the inability of the resulting viral strains to replicate in cell culture.

Table 3.

Phenotypic susceptibility to pimodivir and replication capacity of PB2 site-directed mutants.

PB2 variant	Mean FC for pimodivir for site-directed mutants	Titre (% of WT)	Frequency in Genbank (%)
S324N	84	18%	0.003
S324R	57	18%	0.003
S337P	4.9	1%	0.004
K376R	100	10%	0.050
T378S	1.2	ND	0.008
M431R	>3333	ND	0.003
N510K	514	1%	0.010
N510T	77.3	ND	0.010

FC, fold change; ND, not determined; PB2, polymerase basic protein 2; WT, wild-type.

The Genbank frequency of the PB2 genes containing mutations encoding for the selected PB2 amino acid substitutions was 0.003% for S324N, S324R, and M431R; 0.004% for S337P; 0.008% for T378S; 0.01% for N510K and N510T; and 0.05% for K376R (Table 3).

Discussion

This analysis evaluated the genotypic and phenotypic changes in the influenza A viral strains isolated from infected study participants, treated with pimodivir with or without oseltamivir, using population sequencing analysis of PB2 and NA genes as well as phenotypic susceptibility testing using pimodivir and oseltamivir. At baseline, none of the participants harboured viral variants with PB2 amino acid substitutions associated with reduced susceptibility to pimodivir. This finding is expected as the PB2 gene is well conserved and its function is crucial for viral replication, and there is no selective pressure on this gene by treatment with any of the currently available antivirals. Reduced susceptibility to oseltamivir at baseline was infrequent: three participants had reduced phenotypic susceptibility to oseltamivir (one of those showed H275Y). This is in line with the observation that approximately 1% of the influenza A strains circulating during the 2015–2016 seasons showed reduced susceptibility to oseltamivir.²³

Evidence of emerging resistance to pimodivir using genotype and/or phenotype was observed in 11 of 172 study participants (6.4%) treated with pimodivir. Importantly, most of the genotypic and phenotypic changes were observed in the pimodivir monotherapy groups (pimodivir 300 mg, n = 4/58; pimodivir 600 mg, n = 6/57), whereas only 1/57 participants treated with the combination of pimodivir and oseltamivir developed resistance. In a phase 2a human challenge study in healthy volunteers (NCT01561807), similar frequencies of resistant variants developing on pimodivir monotherapy treatment were observed: PB2 resistance mutations were observed in 7 of 72 participants (9.7%) treated with pimodivir.¹⁴ In another phase 2b study, OPAL (NCT03376321), where study participants were treated with a combination of pimodivir (600 mg BID) and oseltamivir, there were no PB2 substitutions identified at any of the positions of interest in post-baseline samples.²⁴ The observations from both the present study and the OPAL study suggest that combination therapy may reduce the frequency of viruses with reduced pimodivir susceptibility as compared to pimodivir monotherapy.

The most frequently observed PB2 mutations in the human challenge study occurred at position M431¹⁴ but these were not observed in the present study. PB2 mutations at positions S324 and K376 were detected in both studies. Differences in genetic background between the challenge strain and the influenza strains circulating at the time of conduct of natural infection in studies TOPAZ and OPAL could potentially explain this observation.

Overall, there was a good correlation between the emergence of PB2 substitutions and the emergence of reduced phenotypic susceptibility to pimodivir. Based on this study and previous observations using site-directed mutants¹⁶ (data on file), many of these PB2 substitutions showed a markedly decreased replication capacity as compared to wild-type virus. This indicates a low viral fitness and a potential impact on transmissibility of resistant virus, but further studies are required to confirm this.

In this study most, of the emerging reduced susceptibility was observed in study participants with A (H1N1) (10/11 participants). Pimodivir has shown potent and comparable antiviral activity against various influenza A (H1N1) and A (H3N2) strains in vitro and in vitro selection experiments using A (H1N1) or A (H3N2) as starting strain did not reveal any significant difference in the frequency and nature of the amino acid substitutions observed¹⁶ (data on file). It is not clear whether the higher frequency of reduced susceptibility observed in A (H1N1) strains is related to an intrinsic difference with A (H3N2) strains, or due to a difference in seasonal prevalence during the enrolment periods of the clinical study.

This study was the first clinical study with pimodivir in a natural infection setting and had several limitations. First, the number of study participants with reduced susceptibility is relatively small to draw firm conclusions on the importance of specific PB2 amino acid substitutions for reduced susceptibility to pimodivir. Further exploration using reverse genetics as well as clonal analyses are warranted to determine the contribution and relative importance of each mutation in the susceptibility profile of pimodivir. A second limitation is that genotypic analysis was done using population-based Sanger sequencing. Even though genotypic and phenotypic data correlated well, it cannot be excluded that minority variants were present and had an impact on reduced susceptibility to pimodivir.

In summary, these data demonstrate that in study participants with acute uncomplicated influenza A, treated with pimodivir monotherapy in the phase 2b study TOPAZ, the development of reduced susceptibility to pimodivir was infrequent. Combination treatment of pimodivir with oseltamivir further decreased the risk of developing reduced susceptibility.

Supplemental Material

Supplemental Material - Genotypic and phenotypic characterization of influenza A viral variants in study participants treated with pimodivir in the phase 2b TOPAZ study

Supplemental Material for Genotypic and phenotypic characterization of influenza A viral variants in study participants treated with pimodivir in the phase 2b TOPAZ study by Johan Vingerhoets, Ilse Van Dromme, Wilbert van Duijnhoven, David Anderson, Sandra De Meyer and Lorant Leopold by Antiviral Therapy

Footnotes

Acknowledgments

All authors met ICMJE criteria and all those who fulfilled those criteria are listed as authors. All authors had access to the study data, provided direction and comments on the manuscript, had final approval of the document, and made the final decision about where to publish these data.

We acknowledge Ineke Seghers for her contributions to the analyses. Sandra J. Saouaf, PhD (InSeption Group, LLC) and Lakshmi Kasthurirangan, PhD (SIRO Clinpharm Pvt. Ltd) provided medical writing assistance and Robert Achenbach and Bradford Challis, PhD (Janssen Janssen Global Services, LLC, LLC) provided additional editorial support for this manuscript. The authors thank the study participants without whom this study would not have been accomplished, and also thank the investigators for their participation in this study.

Declaration of conflicting interests

All authors are or were employees of Janssen Research and Development, LLC and may own stock/stock options in Johnson & Johnson.

All authors participated in the original design of the studies, supervising recruitment and monitoring of data quality, and contributed to the data interpretation, development and review of this manuscript and confirm that they have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. All authors met ICMJE criteria and all those who fulfilled those criteria are listed as authors. All authors had access to the study data, provided direction and comments on the manuscript, made the final decision about where to publish these data and approved submission to this journal.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Janssen Research & Development, LLC.

ORCID iD

Lorant Leopold

Supplemental Material

Supplemental material for this article is available online.

References

Centers for Disease Control and Prevention (CDC) . Weekly US influenza surveillance report. (Accessed 27 January 2017.) Available from https://www.cdc.gov/flu/weekly/index.htm

World Health Organization . WHO estimate of respiratory deaths due to seasonal influenza. (Accessed 27 January 2019.) Available from https://www.who.int/influenza/surveillance_monitoring/bod/FAQsInfluenzaMortalityEstimate.pdf

Centers for Disease Control and Prevention (CDC) . Influenza Antiviral drug resistance. (Accessed 27 January 2017.) Available from https://www.cdc.gov/flu/treatment/antiviralresistance.htm

Fiore

Fry

Shay

, et al. Antiviral agents for the treatment and chemoprophylaxis of influenza-recommendations of the advisory committee on immunization practices (ACIP). MMWR. Recommendations and Reports : Morbidity and Mortality Weekly Report. Recommendations and Reports 2011; 60(1): 1–24.

Shin

Seong

. Novel antiviral drug discovery strategies to tackle drug-resistant mutants of influenza virus strains. Expert Opinion on Drug Discovery 2019; 14(2): 153–168.

Krol

Rychlowska

Szewczyk

. Antivirals-current trends in fighting influenza. Acta Biochim Pol 2014; 61(3): 495–504.

Samson

Pizzorno

Abed

, et al. Influenza virus resistance to neuraminidase inhibitors. Antiviral Research 2013; 98(2): 174–185.

Dharan

Gubareva

Meyer

, et al. Infections with oseltamivir-resistant influenza A(H1N1) virus in the United States. Jama 2009; 301(10): 1034–1041.

Kramarz

Monnet

Nicoll

, et al. Use of oseltamivir in 12 European countries between 2002 and 2007-lack of association with the appearance of oseltamivir-resistant influenza A(H1N1) viruses. Euro surveillance: bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin 2009; 14(5): 19112.

10.

World Health Organization . Influenza A(H1N1) virus resistance to oseltamivir, 2008. Available from https://www.who.int/influenza/patient_care/antivirals/oseltamivir_summary/en/ (Accessed 8 August 2020).

11.

Hayden

Sugaya

Hirotsu

, et al. Baloxavir marboxil for uncomplicated influenza in adults and adolescents. New England Journal of Medicine 2018; 379(10): 913–923.

12.

Omoto

Speranzini

Hashimoto

, et al. Characterization of influenza virus variants induced by treatment with the endonuclease inhibitor baloxavir marboxil. Scientific Reports 2018; 8(1): 9633.

13.

Roche

H-L

. NCT03684044 study results (Accessed 1 April 2021.) ClinicalTrials.gov:NIH US National Library of Medicine; 2020 updated January 2021 Available from https://clinicaltrials.gov/ct2/show/results/NCT03684044?view=results

14.

Trevejo

Asmal

Vingerhoets

, et al. Pimodivir treatment in adult volunteers experimentally inoculated with live influenza virus: a Phase IIa, randomized, double-blind, placebo-controlled study. Antiviral Therapy 2018; 23(4): 335–344.

15.

Gaelings

Söderholm

, et al. JNJ872 inhibits influenza A virus replication without altering cellular antiviral responses. Antiviral Research 2016; 133: 23–31.

16.

Byrn

Jones

Bennett

, et al. Preclinical activity of VX-787, a first-in-class, orally bioavailable inhibitor of the influenza virus polymerase PB2 subunit. Antimicrobial Agents and Chemotherapy 2015; 59(3): 1569–1582.

17.

Clark

Ledeboer

Davies

, et al. Discovery of a novel, first-in-class, orally bioavailable azaindole inhibitor (VX-787) of influenza PB2. Journal of Medicinal Chemistry 2014; 57(15): 6668–6678.

18.

Finberg

Lanno

Anderson

, et al. Phase 2b study of pimodivir (JNJ-63623872) as monotherapy or in combination with oseltamivir for treatment of acute uncomplicated seasonal influenza A: TOPAZ trial. The Journal of Infectious Diseases 2019; 219(7): 1026–1034.

19.

World Health Organization . Summary of neuraminidase amino acid substitutions associated with reduced inhibition by neuraminidase inhibitors, 2018. Available from https://www.who.int/influenza/gisrs_laboratory/antiviral_susceptibility/NAI_Reduced_Susceptibility_Marker_Table_WHO.pdf?ua=1 (Accessed 6 August 2020).

20.

van Baalen

Jeeninga

PendersPenders

GHWM

, et al. ViroSpot microneutralization assay for antigenic characterization of human influenza viruses. Vaccine 2017; 35: 46–52.

21.

Neumann

Fujii

Kino

, et al. An improved reverse genetics system for influenza A virus generation and its implications for vaccine production. Proceedings of the National Academy of Sciences 2005; 102(46): 16825–16829.

22.

National Center for Biotechnology Information (NCBI) . Genbank. (Accessed 20 March 2013.) Available from https://www.ncbi.nlm.nih.gov

23.

Gubareva

Besselaar

Daniels

, et al. Global update on the susceptibility of human influenza viruses to neuraminidase inhibitors, 2015-2016. Antiviral Research 2017; 146: 12–20.

24.

O’Neil

Ison

Hallouin-Bernard

, et al. A phase 2 study of pimodivir (JNJ-63623872) in combination with oseltamivir in elderly and nonelderly adults hospitalized with influenza A infection: OPAL study. J Infect Dis 2020; 226: 109–118.