Investigating the transmission of baloxavir‐resistant influenza viruses from treated index patients to untreated household contacts in the BLOCKSTONE study

Abstract In a post‐hoc analysis of the phase 3 BLOCKSTONE study (JapicCTI‐184180), we investigated household transmission of baloxavir‐resistant (PA/I38X) influenza viruses. Using baloxavir resistance rates from prior clinical trials and the rate of influenza transmission observed in the study, the predicted number of PA/I38X transmission events was 4.8, assuming wild type and PA/I38X viruses were equally transmissible. However, no PA/I38X viruses were observed. These results suggest a low potential for baloxavir‐resistant influenza virus transmission from treated to untreated individuals, potentially due to reduced viral/transmission fitness for PA/I38X viruses and/or low viral titres at the time when resistant viruses arise.

The influenza virus is subject to constant evolutionary change due to its high replication rate and lack of proofreading activity. This change can lead to the emergence of mutations that make the virus less susceptible to influenza antiviral drugs, resulting in antiviral resistance. 1 In immunocompetent patients, influenza antiviral-resistant viruses are typically transient and do not cause lasting infection, and they do not impact future treatment with the same antiviral. 1 However, there is concern around the potential for widespread or sustained transmission of resistant viruses, which could result in an antiviral becoming no longer suitable for treatment of influenza cases in the community. 1 Baloxavir marboxil (baloxavir) is a first-in-class influenza antiviral that inhibits cap-dependent endonuclease activity of the influenza virus 2 and has been approved as both a single-dose treatment and postexposure prophylaxis for influenza in over 70 countries worldwide. 3 Resistance to baloxavir is most frequently caused by substitutions at isoleucine 38 in the viral polymerase acidic subunit of the influenza RNA polymerase (PA/I38X). 1 Treatment-emergent baloxavir resistant viruses have been observed in phase 3 pivotal clinical trials, with rates of 5% to 10% in adults and 19% in children. [4][5][6] To date, sustained transmission of baloxavir-resistant viruses has not been observed in any population, although there have been reports of sporadic transmission of baloxavir-resistant viruses in Japan. 7 There is a need for a greater understanding of the potential for transmission of baloxavirresistant viruses.
To examine the likelihood of baloxavir-resistant virus transmission relative to wild type, we conducted a post-hoc analysis of the BLOCKSTONE study (JapicCTI-184180). We compared the number of transmission events of baloxavir-resistant viruses in the study with those predicted to occur, using influenza household transmission rates from the study, baloxavir treatment-emergent resistance rates from across baloxavir clinical trials and the assumption of equivalent viral fitness for all influenza viruses. In the BLOCKSTONE study, household contacts (HHCs) who lived with an index patient (IP) with influenza were randomized to receive either baloxavir or placebo as prophylaxis.
As BLOCKSTONE was conducted in Japan where treating influenza with antivirals is standard of care, all IPs in the study were treated with antivirals including baloxavir. 8 (Table 1).
No PA/I38X viruses were identified in any IPs at day 1. As samples were not collected from IPs after day 1, it was not possible to know whether treatment-emergent PA/I38X viruses developed in these patients after this timepoint. Therefore, the incidence of treatment-emergent PA/I38X in IPs was predicted based on estimates from pooled baloxavir clinical trial data stratified by age group (<12 or ≥12 years of age) and virus type/subtype (A/H1N1, A/H3N2 and B) ( Table 2). This incidence was multiplied by the number of eligible IPs in each stratum to give a predicted number of IPs with PA/I38X viruses of 19.7 (Table 3).
To calculate the number of expected PA/I38X virus infections in HHCs, the transmission rate of wild-type viruses was first determined.
Among those included in the analysis, IPs were associated with a single infected HHC in most cases; therefore, the analysis focused on  Whilst ferret studies provide some insights into the fitness and transmissibility of influenza viruses, the translatability of the animal data to a human household/population is limited; therefore, learnings from human clinical studies such as this remain highly relevant.
A second possible explanation for the lack of transmission of baloxavir-resistant viruses is that these viruses arise later in the course of infection in treated IPs, when viral titres are lower than peak viral loads. 1 Therefore, the chances that these viruses are transmitted to others may be lower than if they arose when viral loads were high.
Surveillance studies continue to report very low or no incidence of baloxavir-resistant viruses being detected in untreated patients to date. [10][11][12][13][14] One limitation of this analysis is that the BLOCKSTONE study design did not allow for direct assessment of transmission of baloxavir-resistant viruses from treated IPs, as no virology samples were collected from IPs after day 1 of the study. This will be investigated in the ongoing CENTERSTONE trial (NCT03969212), which is evaluating the efficacy of baloxavir in reducing transmission of influenza from treated IPs to untreated HHCs. 15 Another potential limitation is that it was assumed that all transmission events occurred from the IP to the HHC (or from within the IP's household) and that no viruses were transmitted from outside the household.
In conclusion, the results of this post-hoc analysis suggest that baloxavir-resistant PA/I38X viruses may transmit less frequently than wild-type influenza viruses; however, further validation is needed.
The ongoing CENTERSTONE trial will provide additional insights into transmission of baloxavir-resistant viruses.

ACKNOWLEDGEMENTS
This work was supported by F. Hoffmann-La Roche Ltd. The

CONFLICTS OF INTEREST
The authors declare the following financial interests/personal relationships, which may be considered as potential competing Conceptualization; writing-original draft; writing-review and editing.

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/irv.13079.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.