Fitness of influenza A and B viruses with reduced susceptibility to baloxavir: A mini‐review


Influenza viruses are responsible for a contagious respiratory disease of significant public health and economic importance. Each year, sea- sonal influenza epidemics, that involve influenza A (H3N2) and A (H1N1) pdm09 strains as well as the Yamagata and Victoria lineages of influenza B, affect 10%–20% of the human population and contribute to substantial morbidity and mortality worldwide.1 Annual immunisa- tion programs are the most effective means to control seasonal influ- enza infections; however, current influenza vaccines are strain‐specific
and afford protection against viruses that optimally match the vaccine strains.2 In addition, the immunogenicity of influenza vaccines is generally lower in high‐risk groups such as the elderly and immuno- compromised subjects.3 Consequently, anti‐influenza agents consti- tute a valuable option for the treatment of severe cases of seasonal influenza infections. Due to the presence of the S31N matrix (M)‐2 substitution conferring resistance to adamantanes (i.e., amantadine and rimantadine) in current seasonal influenza A (H1N1) pdm09 and A (H3N2) strains, neuraminidase inhibitors (NAIs) have become the cornerstone of anti‐influenza therapy.4 Two NAIs: oseltamivir (Tami- flu; Hoffmann‐La Roche) and zanamivir (Relenza; GlaxoSmithKline) were licensed worldwide in late 1990s while two others, peramivir (Rapivab, BioCryst) and laninamivir (Inavir, Biota), have been subse- quently approved in some countries.5–8 NAIs target the sialidase ac- tivity of the NA protein, responsible for virion release from infected cells. The emergence of the influenza A (H1N1)‐H275Y NA mutation conferring high levels of resistance to oseltamivir and peramivir during the 2007–2009 period and its worldwide dissemination9 highlighted the need to develop alternative antiviral options.

Because of their pivotal role in the transcription/replication process, the polymerase acidic (PA) and polymerase basic (PB1 and PB2) proteins constitute interesting targets for the development of small molecule inhibitors. Baloxavirmarboxil (BXM), an inhibitor of the cap‐dependent endonuclease (CEN) activity of the PA protein, has been approved in the United States and Japan since 2018,10 whereas favipiravir, an inhibitor of PB1 acting as a purine nucleoside and pimodivir, a compound that targets the PB2 subunit of influenza A viruses (not active against B strains), are at various stages of clinical development.11 One of the major concerns of antiviral ther- apy is the emergence of resistant variants with a potential to outcompete the susceptible wild‐type (WT) virus and to disseminate, compromising the drug’s clinical use.

Previous reports showed that, despite a potent in vitro activity of baloxavir acid (BXA) against seasonal influenza viruses13 in addition to important benefits of BXM in clinical trials, there is a significant concern with this compound relative to the emergence of drug‐ resistant viruses.11,14 Amino acid substitutions resulting in reduced susceptibility to this drug was shown to occur mainly at the isoleu- cine 38 residue (I38T/M/F/L) within the PA protein of influenza A (H1N1) and A (H3N2) viruses.14,15 Other changes with a lower impact on drug susceptibility (E23G/K/R, A36V, A37T, E119D and E199G) have also been reported.The purpose of this manuscript is to review the impact of the main baloxavir resistance PA mutations on fitness properties of influenza A and B variants.


The polymerase of influenza viruses is a hetero trimer complex composed of three protein subunits that include the PA, PB1 and PB2 proteins. This complex binds the conserved 3′ and 5′ ends of each of the eight negative‐sense RNA genome segments and is responsible for transcription and replication of the genomic RNA in the nucleus of infected cells.17,18 X‐ray crystal structure analysis of the influenza A/B polymerase complexes bound to viral RNA (vRNA) were determined for influenza A/H7N10 and influenza B/Memphis/ 13/03 strains.19 These studies highlighted that the PA, PB1 and PB2 subunits interact with each other indifferent ways to mediate vRNA transcription and translation. While the PA protein appears to interact extensively with PB1 and only marginally with the PB2 subunit, there is extensive interaction between PB1 and PB2 subunits.19 The N‐terminus domain of the PA subunit encodes for the endonuclease activity of the viral polymerase whereas the C‐terminus domain of PA interacts with the PB1 subunit to form a tunnel where the vRNA promoter binds. The transcription of the vRNA is initiated by a ‘cap‐snatching’ process during which the PB2 subunit binds the cap structure (7 mG) of a host RNA, and the PA subunit subsequently cleaves it approximately 12–15 nu- cleotides from the cap.17,18 Then, the capped‐oligonucleotides serve as primers to copy the virus template. Finally, the elongation of vRNA is performed by the PB1 subunit which acts as a RNA‐ dependent RNA polymerase.20 Of interest, the overall structures of influenza A (H7N10) and B/Memphis/13/03 polymerases complexes were found to be very similar despite the limited sequence identities for PA (36%), PB1 (59.5%) and PB2 (37%) between the two strains.20 Thus, because of their essential role in the influenza virus replication cycle, the influenza polymerase complex subunits constitute a target of choice for the development of small molecule inhibitors.


3.1 | Antiviral mechanism

Crépin and colleagues have demonstrated that the amino acid region 1–209 of the influenza PA protein contains highly conserved metal ion binding (H41, E80, D108 and E119) and catalytic (K134) residues, typical of type II endonucleases.21 Based on the metal ion binding feature of dolutegravir (the human immunodeficiency virus integrase inhibitor), researchers from the Shionogi Inc group have computa- tionally designed compounds that specifically inhibit the influenza PA endonuclease. This resulted in the discovery of BXM, a prodrug that is metabolised by the arylacetamide deacetylase enzyme into BXA (Figure 1) which inhibits the CEN activity of the PA protein, thereby preventing the cleavage of mRNA during the ‘cap‐snatching’ process.

3.2 | In vitro activity of BXA against influenza viruses

In enzymatic assays performed with ribonucleoproteins purified from influenza A and B viral particles, BXA selectively inhibited the endonuclease activity of influenza A and B PAs with mean IC50 (the drug concentration decreasing the PA activity by 50%) values of 1.4–3.1 nM and 4.5–8.9 nM, respectively.13 BXA also demonstrated a significant impact on viral replication against influenza A (H1N1), A (H3N2) and B strains, as assessed by virus yield reduction and plaque reduction assays.13 In yield reduction assays, BXA showed high po- tency against influenza A and B viruses with mean EC90 (the drug concentration decreasing virus titres to one‐tenth of the untreated control)of 0.46–0.98 nM and 2.2–3.4 nM, respectively.13 Noteworthy, Such BXA EC90 values were 2–3 log lower compared to those determined for oseltamivir, zanamivir and laninamivir.13 When tested against current seasonal influenza A (H1N1) pdm09, A (H3N2) and B viruses, BXA demonstrated activity against all tested influenza sub- types, including NAI‐ and adamantane‐resistant viruses, with EC50 (the concentration decreasing the virus titres to half of the untreated control) values that were within the nanomolar range.16,22 Influenza B EC50 values are approximately five‐fold higher compared to influ- enza A counterparts.13,22 BXA is also active against zoonotic influ- enza A strains of various subtypes (H1N2, H5N1, H5N2, H6N6, H7N9 and H9N2)13 as well as against influenza C and D strains.23 The broad spectrum of activity of BXA could be attributed to the high conservation of amino acids forming the CEN active centre.24 In cell culture experiments, the combination of BXA with oseltamivir, zanamivir, peramivir or laninamivir against influenza A/Puerto Rico/8/1934 (H1N1) infections resulted in synergistic effects.

FIGURE 1 Chemical structure of baloxavir acid (BXA: C24H19F2N3O4S), the metabolic active compound, baloxavir marboxil (BXM: C27H23F2N3O7S), the prodrug and the related polymerase acidic inhibitor compound (RO‐7: C24H20F3N3O3S).

3.3 | Activity of BXA against influenza viruses in animal studies

Oral administration of BXM has also demonstrated beneficial effects in experimental influenza A and B infections of mice. The BXM oral doses used in mouse studies ranged from 0.05 to 50 mg/kg BID. Fukao and collaborators indicated that 15 mg/kg of BXM BID for 5 days given to mice could mimic the plasma concentration of BXA in humans.26 Single‐day administration of BXM (0.5 or 5 mg/kg BID for influenza A and 5 or 50 mg/kg bid for influenza B), starting at 72 h post‐infection (p.i.) provided a significant decrease in lung viral titres and completely prevented mortality in BALB/c mouse models of le- thal influenza A/Puerto Rico/1934 (H1N1) and B/Hong Kong/5/1972 infections.26 The therapeutic efficacy of BXM was also demonstrated in mice that received lethal doses of avian influenza A/Anhui/1/2013 (H7N9) virus where oral administration of BXM at 5 or 50 mg/kg twice a day, for 5 days, protected animals and reduced lung viral titres by more than 2–3 log compared to vehicle.27 In a BALB/c mouse model of lethal influenza A/Puerto Rico/1934 (H1N1) infec- tion, BXM therapy (15 mg/kg twice daily, for 5 days, starting at 96 h, p.i.) prevented mortality and reduced lung viral titres.25 In addition, combining a suboptimal dose of BXM with oseltamivir phosphate afforded synergistic benefits over monotherapies in terms of mor- tality, reduction of pro‐inflammatory cytokine/chemokine levels and lung pathological changes.25 In immunocompromised mice, BXM treatment, initiated at 120 h p.i., significantly reduced lung viral titres within 24 h after initial administration, and then significantly pre- vented body weight loss in influenza A/Puerto Rico/1934 (H1N1)‐ infected animals.

More recently, the potential of BXM to impact influenza virus transmission was assessed in ferrets experimentally infected with influenza A (H1N1) pdm09 virus.28 The authors reported that administration of a single dose of BXM (4 mg/kg) reduced viral shedding in the upper respiratory tract of ferrets compared to pla- cebo and also reduced the frequency of airborne and contact trans- mission of virus to healthy ferrets, even when treatment was delayed until 2 days p.i.

3.4 | Activity of BXM against influenza viruses in clinical studies

In a phase I clinical study which evaluated safety, tolerability and pharmacokinetics, BXM was found to be well tolerated with a favourable safety profile and pharmacokinetic characteristics. In particular, BXM demonstrated a plasma half‐life of 49–91 h allowing this antiviral to be used as a single dose.29 A phase II trial enrolling Japanese adults with uncomplicated influenza A or B infections showed that a single dose (10, 20 or 40 mg) of oral BXM reduced the median time of alleviating influenza symptoms by 23.5, 26.7 and 28.2 h, respectively, compared to placebo.14 The phase III clinical trial, conducted in patients (aged ≥12 years) with uncomplicated influenza infection (CAPSTONE‐1), showed that administration of a single dose (40 or 80 mg) of oral BXM reduced the duration of influenza symptoms to an extent similar to standard (75 mg twice daily for 5 days) oseltamivir treatment (53.5 h vs. 53.8 h).14 BXM afforded greater reductions in infectious virus and viral RNA titres than oseltamivir, in particular during early time points. A phase III clinical trial enrolling outpatients with co‐morbidities (CAPSTONE‐2) showed that BXM significantly reduced time to improvement of influenza symptoms by 29.1 h compared to the placebo group.30 Moreover, BXM treatment was associated with a significant reduc- tion in the use of antibiotics and the incidence of influenza‐related complications.30 Most cases of influenza infections in CAPSTONE‐1 and 2 trials involved influenza A (H1N1) pdm09 and A (H3N2) viruses, respectively. When evaluated against influenza B infections, in the CAPSTONE‐2 study, BXM also reduced the median time of alleviation of symptoms by 25.8 and 27.0 h compared to placebo and oseltamivir, respectively.30 In a recent post‐exposure prophylactic report, BXM demonstrated efficacy in preventing influenza infections in household contacts of patients with influenza.

The US Food and Drug Administration approved BXM on 24 October 2018. The treatment is currently recommended for acute, uncomplicated influenza in people ≥12 years of age who have been symptomatic for no more than 48 h and who are at high risk of developing flu‐related complications. The oral regimen consists on a single weight‐based dose of 40 or 80 mg. In Japan, BXM was licensed on 23 February 2018 and it is also indicated for use in children younger than 12 years of age who can receive a single oral dose of up to 20 mg depending on age and weight. Currently, BXM is being studied in phase III development programs, including children under the age of one as well as severely ill, hospitalised patients.

3.5 | Mechanisms of resistance to BXA

As for other antivirals, influenza A/B variants with reduced suscepti- bility to BXA may emerge after in vitro passages, under BXA pressure and/or among BXM‐treated patients. In vitro passages of influenza A/ WSN/33 (H1N1) virus in MDBK (Madin–Darby bovine kidney) cells treated with increasing concentrations of BXA selected for the I38T PA mutation that appeared after 6–9 passages in different variants.13 This mutation conferred a 30–40 fold increase in BXA EC50 values compared to the WT. A similar phenotype was observed in a reverse genetics (rg) rescued A/WSN/33 virus containing the PA I38T sub- stitution confirming the role of this change on altered susceptibility.13 Of interest, the I38T PA substitution was also selected in influenza A/ California/04/2009 and A/PR8/1934 (H1N1) strains during in vitro passages in presence of RO‐7, another investigational PA inhibitor.24 In another study, in vitro passages of contemporary influenza A iso- lates under BXA pressure resulted in the emergence of the I38T substitution in the PA of influenza A/Bangladesh/3007/2017 (H3N2) and A/Illinois/08/2018 (H1N1) viruses.

The PA I38 residue is highly conserved among seasonal influenza A (H1N1), A (H3N2) and B viruses.24 Co‐crystal structures of influ- enza A and B endonucleases with BXA showed that the I38T substitution alters van der Waals contacts between the endonuclease and BXA resulting in reduced stability of the drug‐target linkage.15 According to in vitro predictions, the role of amino acid substitutions at residue 38 of the PA protein was confirmed in phase II and III trials where A (H1N1) and/or A (H3N2) I38T/M/F variants with reduced susceptibility emerged in 2.2% and 9.7% of BXM‐treated patients, respectively.14 BXM‐resistant variants, mainly due to I38X changes, were also detected in 18 out of 77 (23.3%) children who received BXM therapy.15 Other changes (such as E23K/G, A37T, E119D and E199G)were also detected in BXM‐treated individuals.

3.6 | Viral fitness of BXA‐resistant influenza A and B viruses

The fitness cost of reduced susceptibility is an important feature from the virological and clinical points of view. Numerous in vitro and animal studies, which assessed the fitness cost of NA mutations mediating resistance to NAIs, have shown that some mutations could severely compromise the viral fitness whereas unaltered fitness could be observed for other drug‐resistant mutants.33,34 The following sections summarize the different studies related to the fitness of influenza A (Table 1) and B (Table 2) viruses harbouring the main PA substitutions responsible for reduced susceptibility to BXA.

3.6.1 | In vitro fitness of BXA‐resistant influenza A viruses

Noshi and collaborators found that the influenza A/WSN/1933 (H1N1) I38T PA variant, selected by in vitro passages in presence of BXA, grew at lower titres when compared to the WT, particularly in the early phase of viral replication kinetics using MDBK, MDCK, RPMI2650 and A549 cell lines.13 The effect of the PA I38T substi- tution was confirmed by using the rg‐rescued I38T PA mutant.13 Similar impacts on viral growth were observed for rg‐rescued A/ WSN/33 (H1N1) and A/Victoria/3/1975 (H3N2) variants containing I38T, I38F and I38M PA substitutions.15 When evaluated in recent viral backgrounds, our group found that the I38T substitution did not significantly alter the replication kinetics of rg‐rescued A/Quebec/ 144147/2009 (H1N1) I38T and A/Switzerland/9715293/2013 (H3N2) I38TPA variants in ST6‐GalI‐MDCK cells.35 However, in competition experiments, we observed that an initial 50%:50% WT/ mutant mixture evolved to 70%:30% for A (H1N1) and 88%:12% for A (H3N2) viruses after a single‐cell passage.35 In addition, the sub- stitution was maintained in the two viral backgrounds after four in vitro passages. Such genetic stability was also described for an influenza A/California/04/2009 (H1N1) I38T PA variant selected in vitro with RO‐7.24 A mild impact on in vitro replication was also observed for contemporary influenza A/Bangladesh/3007/2017 (H3N2)‐I38T, A/Louisiana/49/2017 (H3N2)‐I38M and A/Illinois/08/ 2018 (H1N1)‐I38T/S PA isolates.32 Indeed, despite the slower growth of mutant isolates compared to their respective control at early time points of replication kinetics, peak viral titres of compa- rable levels could be reached for all mutant viruses and their respective WT controls at 48 h p.i.32 In another report, the viral fitness of two A (H1N1) and two A (H3N2) I38T PA variants recov- ered in 2018–2019 from BXM‐treated patients was assessed, in comparison with their respective pre‐therapy isolates.36 In a repli- cation kinetics experiment using hCK cells (an MDCK cell line expressing high levels of α2,6‐sialoglycans),37 the four I38T mutant isolates grew to titres that were comparable to those obtained for their respective WT (pre‐therapy) isolates.36 In competition experi- ments, the WT appeared to outcompete the I38T variants for the two A (H1N1) pairs; by contrast, the two A (H3N2) I38T mutants grew at higher proportions versus the respective WT virus.36 In a recent report, the I38S PA A (H1N1) pdm09 variant, selected by in vitro passages with baloxavir also appeared to have a delayed growth, compared to the control in both MDCK and MDCK‐SIAT1 cells.

The replication of the PA E23K A (H1N1) pdm09 mutant virus, recovered from a Japanese untreated child, was also significantly reduced compared to the wild‐type virus in both native MDCK and humanized (hCK) cells suggesting that the PA E23K substitution altered viral growth at least in vitro.38 Such in vitro replication default attributable to the E23K PA change was previously described by Jones and colleagues who performed in vitro serial passages of A (H1N1) pdm09 virus under RO‐7 pressure and detected the E23K substitution in a nonviable variant (viral RNA could be detected, but virus titre was below detection limits).24

3.6.2 | In vitro fitness of BXA‐resistant influenza B viruses

In contrast to influenza A laboratory strains, the I38T and I38M PA substitutions did not alter the in vitro replication of rg‐rescued influenza B/Maryland/1/1959 variants whereas the I38F substitution resulted in lower titres versus the WT, at 24 h p.i.13 We recently characterized rg‐rescued contemporary Yamagata‐like influenza B/Phuket/3073/2013 viruses with I38T/M and E23K PA sub- stitutions.39 In replication kinetics experiments using ST6‐GalI‐MDCK cells, the rg‐rescued WT virus had peak viral titres comparable to those of I38T and I38M recombinants whereas replication was significantly lower for the E23K mutant. Another group recently described rg‐rescued Victoria‐like influenza B/Brisbane/60/2008 viruses harbouring I38T/M/F PA substitutions. While the I38T and I38F substitutions resulted in altered replication in vitro, the I38M change had a minimal impact.40

3.6.3 | In vivo fitness of BXA‐resistant influenza A viruses

Our group assessed the effect of I38T PA substitution on contem- porary influenza A (H1N1) pdm09 and A (H3N2) strains in mice. The WT and its I38T mutant induced similar weight loss with comparable lung titres in both viral subtypes and no T38I reversions were detected in lung samples.35 Imai and colleagues assessed the fitness of two A (H1N1) and two A (H3N2) I38T PA variants recovered in 2018–2019 from BXM‐treated patients in different animals models.36 For any of the four pairs of viruses, no significant differ- ence could be seen between the WT and the mutant viruses in Syrian hamsters.36 Moreover, there was no substantial difference in viral titres from nasal turbinates and lungs of experimentally infected BALB/c and DBA/2 mice, between the A (H1N1) WT and mutant viruses.36 The A (H1N1) and A (H3N2) isolates were also evaluated in ferrets and, again, similar behaviours were observed for the mutants and their respective WT parental viruses. For both of the two pairs of viruses, no significant differences in nasal wash viral titres were observed between the WT and the I38T mutant and the mutant vi- ruses could be transmitted through respiratory droplets with a similar efficiency than their respective WT parental viruses.

In ferrets experimentally infected with influenza A/Bangladesh/ 3007/2017 (H3N2)‐I38T and A/Louisiana/49/2017 (H3N2)‐I38M isolates, infectious viruses were shed for 7 consecutive days in both mutant and WT viruses.32 There were no substantial differences in nasal wash viral titres and disease symptoms between the paired virus groups and no reversion to I38 was observed in nasal wash samples collected from ferrets inoculated with either I38M or I38T, suggesting the genetic stability of such mutants.32 However, when different proportions of the WT and I38T or I38M mutants were evaluated in competition experiments using ferrets, the WT out- competed the mutants although the latter was detectable until the last time point (i.e., day 7 p.i.).32 Another group recently reported that I38T/M PA substitutions did not alter direct contact (DC) and airborne contact (AC) transmission in ferrets infected with the re- combinant influenza A/California/04/2009 (H1N1) and A/Texas/71/ 2017 (H3N2) viruses.40 The WT virus and I38T/M variants were shed by all donors and DC/AC animals.

In patients enrolled in BXM trials, the median time to alleviation of symptoms was longer in treated patients infected with I38T/M PA variants than in those without such substitutions.14 In particular, PA‐ I38T/M substitutions were also associated with prolonged shedding, and rebound in viral replication, with titres increasing from days 3 to 6 post‐treatment, making such changes clinically relevant drug resis- tance markers.41,42 Recently, influenza A (H1N1) pdm09 and A (H3N2) viruses harbouring the PA I38S or I38T amino acid substitutions, recovered frombaloxavir‐treated children, were also found to be associated with viral titre rebounds.43 Noteworthy, in two patients infected with the PA I38S A (H1N1) pdm09 variant after baloxavir treatment, both the viral RNA loads and virus titres increased be- tween days 3 and 4, and the population of I38S after virus isolation in hCK cells was higher than in the original nasopharyngeal samples recovered on day 4, suggesting a good fitness of this variant.43 Finally, it should be indicated that, an influenza A (H3N2) virus containing the I38T PA substitution was detected in a hospitalized child who had not received BXM therapy.44 In another report, Takashita and colleagues also detected a PA E23K mutant A (H1N1) pdm09 virus from a child without baloxavir treatment.45 Morever, a recent clinical report suggested the possible association of E23K/G, A37T and E199G substitutions with A (H1N1) pdm09 viral rebound despite lower levels of reduced susceptibilities to BXA for such variants.46

3.6.4 | In vivo fitness of BXA‐resistant influenza B viruses

Few reports are available on the fitness of BXM‐resistant B variants. In our study using recombinant influenza B/Phuket/3073/2013 vi- ruses, infections of mice with the WT, I38T and I38M recombinants induced mortality rates of 60%, 40% and 100%, respectively, and similar lung viral titres were obtained for the three groups at days 3 and 6 p.i.39 In another report, the B/Brisbane/60/2008 I38T and I38M variants disseminated to naive ferrets by contact and airborne transmission, whereas the I38F variant failed to transmit via the airborne route.40


In the absence of a universal influenza vaccine and due to limited therapeutic options, anti‐influenza agents with a broad spectrum activity may play a significant role against severe seasonal and pandemic influenza infections. Baloxavir demonstrated potent in vitro and/or in vivo activities against strains from the four (A, B, C and D) influenza types. In addition, a single oral dose of BXM pro- vided clinical benefits that were equal or superior to the standard (5 days, twice daily) regimen of oseltamivir.14 These advantages could lead to increased prescriptions of BXM in future clinical settings. Consequently, there is a need to remain vigilant with regard to the emergence of BXM resistance mutations and to document the fitness characteristics of mutants with a high potential to emerge.
This review highlights the significant role of amino acid sub- stitutions at the PA I38 residue (I38T/M/F) in conferring resistance to baloxavir both in vitro and in clinic. In vitro, 5–10 passages in presence of BXA seems to be required to select for drug‐resistant variants, whereas clinical resistance to BXM occurs a few days after a single‐ dose treatment. In clinic, BXM resistance occurred at higher rates in the paediatric population.42 Also, it is likely that the resistance‐ associated PA substitutions may emerge at higher rates in H3N2 vi- ruses than in the H1N1 background.47 The I38T substitution induced the highest increase in BXA IC50s in influenza A and B backgrounds, compared to the other reported substitutions. As for NAIs, BXA IC50 values are approximately fivefold higher in influenza B viruses than influenza A strains. Notably, there is a need to select and standardise a reliable technical method for assessing BXA resistance.16

The in vitro fitness of influenza viruses containing resistance‐associated PA mutations appears to be influenced by the genetic background of the viral strain and the cell line utilized. While the I38T substitution clearly impacted the replication of old (A/WSN/33 and A/Puerto Rico/8/1934) influenza strains, a more nuanced effect was described for recent A (H1N1) pdm09 I38T variants. The impact of PA substitution may also vary between influenza A subtypes and B lineages.

With regard to the different PA substitutions reported in vivo and in clinic, our review suggests a pattern of viral fitness that differs significantly depending on whether the isoleucine 38 residue is replaced by threonine, methionine or phenylalanine. I38F influenza A and B mutants were the most impacted variants whereas a milder ef- fect was seen for I38T variants.40 On the other hand, the I38M sub- stitution could even increase the viral fitness (virulence) of influenza A and B variants.39,40 Whether the delay in replication observed for some PA mutants would be clinically relevant remains to be investigated.

In summary, this review confirms that seasonal influenza A and B viruses harbouring PA mutations mediating reduced susceptibility to BXA may retain a significant level of viral fitness. This raises the concern that widespread use of BXM may favour the circulation of influenza variants with drug‐resistant mutations in future influenza seasons. To ensure the utility of this new polymerase inhibitor in the near future, there is a need fora rational use of BXM and continuous monitoring for the emergence of seasonal influenza A and B variants. Improving BXM dosages and/or combining this drug to NAI or other polymerase inhibitors could also help to maintain maximal benefits from BXM.


This work was supported by a Canadian Institutes of Health Research (CIHR) foundation grant to GB (grant No. 229733) for a research program on the pathogenesis, treatment and prevention of respira- tory and herpes viruses.


Guy Boivin and Yacine Abed designed the mini review. Yacine Abed and Amel Saim‐Mamoun reviewed the literature, selected the articles to be included in the mini‐review and prepared the manuscript. Guy Boivin revised and corrected the manuscript.


Data sharing not applicable to this mini‐review as no datasets were generated during the preparation of this manuscript.


Yacine Abed


1. Tokars JI, Olsen SJ, Reed C. Seasonal incidence of symptomatic influenza in the United States. Clin Infect Dis. 2018;66:1511‐1518.
2. Palese P. Making better influenza virus vaccines? Emerg Infect Dis. 2006;12:61‐65.
3. Soema PC, Kompier R, Amorij JP, Kersten GF. Current and next generation influenza vaccines: formulation and production strate- gies. Eur J Pharm Biopharm. 2015;94:251‐263.
4. Samson M, Pizzorno A, Abed Y, Boivin G. Influenza virus resistance to neuraminidase inhibitors. Antivir Res. 2013;98:174‐185.
5. Birnkrant D, Cox E. The emergency use authorization of peramivir for treatment of 2009 H1N1 influenza. N Engl J Med. 2009; 361:2204‐2207.
6. Kubo S, Tomozawa T, Kakuta M, Tokumitsu A, Yamashita M. Laninamivir prodrug CS‐8958, a long‐acting neuraminidase inhibitor, shows superior anti‐influenza virus activity after a single administration. Antimicrob Agents Chemother. 2010;54:1256‐ 1264.
7. Watanabe A, Chang SC, Kim MJ, Chu DW, Ohashi Y. Long‐acting neuraminidase inhibitor laninamivir octanoate versus oseltamivir for treatment of influenza: a double‐blind, randomized, noninferiority clinical trial. Clin Infect Dis. 2010;51:1167‐1175.
8. FDA News Release. FDA Approves Rapivab to Treat Flu Infection. ucm427755.htm. Accessed August 19, 2016.
9. Meijer A, Lackenby A, Hungnes O, et al. Oseltamivir‐resistant influenza virus A (H1N1), Europe 2007–08 season. Emerg Infect Dis. 2009;15:552‐560.
10. Heo YA. Baloxavir: first global approval. Drugs. 2018;78:693‐697.
11. Hayden FG, Shindo N. Influenza virus polymerase inhibitors in clinical development. Curr Opin Infect Dis. 2019;32:176‐186.
12. Hussain M, Galvin HD, Haw TY, Nutsford AN, Husain M. Drug resistance in influenza A virus: the epidemiology and management. Infect Drug Resist. 2017;10:121‐134.
13. Noshi T, Kitano M, Taniguchi K, et al. In vitro characterization of baloxavir acid, a first‐in‐class cap‐dependent endonuclease inhibitor of the influenza virus polymerase PA subunit. Antivir Res. 2018;160:109‐117.
14. Hayden FG, Sugaya N, Hirotsu N, et al. Baloxavir marboxil for uncomplicated influenza in adults and adolescents. N Engl J Med. 2018;379:913‐923.
15. Omoto S, Speranzini V, Hashimoto T, et al. Characterization of influenza virus variants induced by treatment with the endonuclease inhibitor baloxavir marboxil. Sci Rep. 2018;8:9633.
16. Gubareva LV, Mishin VP, Patel MC, et al. Assessing baloxavir susceptibility of influenza viruses circulating in the United States during the 2016/17 and 2017/18 seasons. Euro Surveill. 2019;24(3):1800666.
17. Fodor E, Pritlove DC, Brownlee GG. The influenza virus panhandle is involved in the initiation of transcription. J Virol. 1994;68:4092‐4096.
18. Krug RM, Broni BA, Bouloy M. Are the 5′ ends of influenza viral mRNAs synthesized in vivo donated by host mRNAs? Cell. 1979;18:329‐334.
19. Zhang J, Hu Y, Musharrafieh R, Yin H, Wang J. Focusing on the influenza virus polymerase complex: recent progress in drug discovery and assay development. Curr Med Chem. 2019;26: 2243‐2263.
20. Stevaert A, Naesens L. The influenza virus polymerase complex: an update on its structure, functions, and significance for antiviral drug design. Med Res Rev. 2016;36:1127‐1173.
21. Crepin T, Dias A, Palencia A, Swale C, Cusack S, Ruigrok RW. Mutational and metal binding analysis of the endonuclease domain of the influenza virus polymerase PA subunit. J Virol. 2010;84: 9096‐9104.
22. Koszalka P, Tilmanis D, Roe M, Vijaykrishna D, Hurt AC. Baloxavir marboxil susceptibility of influenza viruses from the Asia‐Pacific, 2012‐2018. Antivir Res. 2019;164:91‐96.
23. Mishin VP, Patel MC, Chesnokov A, et al. Susceptibility of influenza A, B, C, and D viruses to baloxavir(1). Emerg Infect Dis. 2019;25:1969‐1972.
24. Jones JC, Kumar G, Barman S, et al. Identification of the I38T PA substitution as a resistance marker for next‐generation influenza virus endonuclease inhibitors. mBio. 2018;9(2).
25. Fukao K, Noshi T, Yamamoto A, et al. Combination treatment with the cap‐dependent endonuclease inhibitor baloxavir marboxil and a neuraminidase inhibitor in a mouse model of influenza A virus infection. J Antimicrob Chemother. 2019;74:654‐662.
26. Fukao K, Ando Y, Noshi T, et al. Baloxavir marboxil, a novel cap‐dependent endonuclease inhibitor potently suppresses influ- enza virus replication and represents therapeutic effects in both immunocompetent and immunocompromised mouse models. PLoS One. 2019;14:e0217307.
27. Taniguchi K, Ando Y, Nobori H, et al. Inhibition of avian‐origin influenza A(H7N9) virus by the novel cap‐dependent endonuclease inhibitor baloxavir marboxil. Sci Rep. 2019;9:3466.
28. Lee LYY, Zhou J, Frise R, et al. Baloxavir treatment of ferrets infected with influenza A(H1N1) pdm09 virus reduces onward transmission. PLoS Pathog. 2020;16:e1008395.
29. Koshimichi H, Ishibashi T, Kawaguchi N, Sato C, Kawasaki A, Wajima T. Safety, tolerability, and pharmacokinetics of the novel anti‐influenza agent baloxavir marboxil in healthy adults: phase I study findings. Clin Drug Invest. 2018;38:1189‐1196.
30. Ison MG. Improving delivery of early treatment to influenza‐infected patients. Clin Infect Dis. 2018;66:1042‐1044.
31. Ikematsu H, Hayden FG, Kawaguchi K, et al. Baloxavir marboxil for prophylaxis against influenza in household contacts. N Engl J Med. 2020;383:309‐320.
32. Chesnokov A, Patel MC, Mishin VP, et al. Replicative fitness of seasonal influenza A viruses with decreased susceptibility to baloxavir. J Infect Dis. 2020;221:367‐371.
33. Govorkova EA. Consequences of resistance: in vitro fitness, in vivo infectivity, and transmissibility of oseltamivir‐resistant influenza A viruses. Influenza Other Respir Viruses. 2013;7(suppl 1):50‐57.
34. McKimm‐Breschkin JL. Influenza neuraminidase inhibitors: antiviral action and mechanisms of resistance. Influenza Other Respir Viruses. 2013;7(suppl 1):25‐36.
35. Checkmahomed L, M’Hamdi Z, Carbonneau J, et al. Impact of the baloxavir‐resistant polymerase acid I38T substitution on the fitness of contemporary influenza A(H1N1)pdm09 and A(H3N2) strains. J Infect Dis. 2020;221:63‐70.
36. Imai M, Yamashita M, Sakai‐Tagawa Y, et al. Influenza A variants with reduced susceptibility to baloxavir isolated from Japanese patients are fit and transmit through respiratory droplets. Nat Microbiol. 2020;5:27‐33.
37. Takada K, Kawakami C, Fan S, et al. A humanized MDCK cell line for the efficient isolation and propagation of human influenza viruses. Nat Microbiol. 2019;4:1268‐1273.
38. Takashita E, Kawakami C, Ogawa R, et al. Influenza A(H3N2) virus exhibiting reduced susceptibility to baloxavir due to a polymerase acidic subunit I38T substitution detected from a hospitalised child without prior baloxavir treatment, Japan, January 2019. Euro Surveill. 2019;24(12).
39. Abed Y, Fage C, Checkmahomed L, Venable MC, Boivin G. Characterization of contemporary influenza B recombinant viruses harboring mutations of reduced susceptibility to baloxavir marboxil, in vitro and in mice. Antivir Res. 2020;179:104807.
40. Jones JC, Pascua PNQ, Fabrizio TP, et al. Influenza A and B viruses with reduced baloxavir susceptibility display attenuated in vitro fitness but retain ferret transmissibility. Proc Natl Acad Sci U S A. 2020;117:8593‐8601.
41. Uehara T, Hayden FG, Kawaguchi K, et al. Treatment‐Emergent influenza variant viruses with reduced baloxavir susceptibility: impact on clinical and virologic outcomes in uncomplicated influenza. J Infect Dis. 2020;221:346‐355.
42. Hirotsu N, Sakaguchi H, Sato C, et al. Baloxavir marboxil in Japanese pediatric patients with influenza: safety and clinical and virologic outcomes. Clin Infect Dis. 2020;71:971‐981.
43. Sato M, Takashita E, Katayose M, et al. Detection of variants with reduced baloxavir marboxil susceptibility after treatment of children with influenza A during the 2018‐2019 influenza season. J Infect Dis. 2020;222:121‐125.
44. Takashita E, Ichikawa M, Morita H, et al. Human‐to‐human transmission of influenza A(H3N2) virus with reduced susce- ptibility to baloxavir, Japan. Emerg Infect Dis. 2019;25:2108‐ 2111.
45. Takashita E, Abe T, Morita H, et al. Influenza A(H1N1)pdm09 virus exhibiting reduced susceptibility to baloxavir due to a PA E23K substitution detected from a child without baloxavir treatment. Antivir Res. 2020;180:104828.
46. Ince WL, Smith FB, O’Rear JJ, Thomson M. Treatment‐emergent influenza virus PA substitutions independent of those at I38 asso- ciated with reduced baloxavir susceptibility and virus rebound in trials of baloxavir marboxil. J Infect Dis. 2020;22.
47. Gubareva LV, Fry AM. Baloxavir and treatment‐emergent resis- tance: public health insights and next steps. J Infect Dis. 2020;221: 337‐339.