Carryover effects of baloxavir acid in human nasopharyngeal/pharyngeal swabs on infectious titer testing of influenza virus.

Abstract Baloxavir marboxil (BXM) demonstrated a rapid and profound decline in infectious viral titer 1 day after BXM administration. Rapid reduction in virus titer is a characteristic of BXM. There may be a possibility that drug carryover effects have impacts on the observed antiviral effects due to the poor correlation that was observed between viral titer reduction and alleviation of influenza symptoms. Here, we report possible carryover effects of baloxavir acid (BXA), an active form of BXM, on infectious titer testing. Our findings indicate that there is little impact of BXA carryover on the infectious titer testing.


Abstract
Baloxavir marboxil (BXM) demonstrated a rapid and profound decline in infectious viral titer 1 day after BXM administration. Rapid reduction in virus titer is a characteristic of BXM. There may be a possibility that drug carryover effects have impacts on the observed antiviral effects due to the poor correlation that was observed between viral titer reduction and alleviation of influenza symptoms. Here, we report possible carryover effects of baloxavir acid (BXA), an active form of BXM, on infectious titer testing. Our findings indicate that there is little impact of BXA carryover on the infectious titer testing.

K E Y W O R D S
baloxavir acid, baloxavir marboxil, cap-dependent endonuclease, drug carryover, influenza virus 2 | MATERIAL S AND ME THODS

| Study design
The CAPSTONE-1 study was conducted in the United States and Japan as a double-blind, placebo-and OTV-controlled, randomized trial. The details of the trial have been reported previously. 3

| Quantification of BXA concentration in the nasal swab samples
Nasopharyngeal/pharyngeal swabs were placed into the universal transport medium (Puritan UniTranz-RT TM ) and immediately stored at 2 to 8°C. 4 The swab samples were shipped to the central laboratory under cooled conditions, then subdivided, and stored at − 80°C.
A total of 48 swab samples were randomly selected according to the following criteria: (a) collected from subjects at Day 2, (b) samples from the subjects whose plasma BXA concentration at Day 2 was more than 40.0 ng/mL, which was higher than 25th percentile for the total 589 of BXM-treated subjects (34.3 ng/mL), and (c) stored at − 80°C within 96 hours after collection. To quantify the BXA concentration, 50 μL of the inoculated transport medium was mixed with acetonitrile and formic acid at the ratio of 1000:1 (v/v) for protein precipitation. The supernatants were subjected to liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis.
The analytical method was validated for a quantification range of 0.0500 to 50.0 ng/mL.

| Determination of BXA concentration that interferes with infectious titer testing
BXA was synthesized at Shionogi & Co., Ltd. (Osaka, Japan). A vaccine strain A/Victoria/361/2011 (H3N2) 5 was diluted in the Puritan transport medium to reach the final viral titers at 400 or 40 000 50% tissue culture infectious dose (TCID 50 )/mL and mixed with BXA at the final concentration from 0 to 300 ng/mL; five-independent portions were prepared for each sample ( Figure 1A). The samples were stored at − 80°C according to the procedures used in the clinical trial. Viral titers were determined in MDCK-SIAT1 cells 6,7 (European Collection of Cell Cultures) in the same way as in the CAPSTONE-1 study. 3 Briefly, the cells were inoculated with 10 0 -to 10 7 -fold of diluted samples. The cells were absorbed by spin inoculation, followed by washing of the cells to remove unabsorbed viruses and BXA in the inoculum, and then, the cells were incubated at 33°C in a CO 2 incubator for 3 days. The presence of virus-induced cytopathic effects was evaluated, and the viral titers were determined by the Behrens-Karber method ( Figure 1B). 8 The virus titer of less than the lower limit of quantification (LLOQ) was set as 0.5 log 10 TCID 50 /mL. The comparisons of virus titers between the baseline without BXA and each BXA concentration were conducted by Welch's t test using SAS version 9.2 at P values of less than .05 significance level.

| Comparison of declines between viral titers and RNA loads
Changes from the baseline in viral titers and RNA in the CAPSTONE-1 study were quantified as described previously. 3 Comparison of declines was conducted using data with viral titers and RNA higher than LLOQ (0.7 log 10 TCID 50 /mL, or 2.18 log 10 viral particle/mL).
Correlations were analyzed by Pearson's product using SAS version 9.2 at P values of less than .05 significance level.

| Quantification of BXA concentration in nasopharyngeal/pharyngeal swab samples from BXMtreated subjects
The median plasma BXA concentration for the 48 selected subjects (72.1 ng/mL) was higher than that for the total 589 BXM-treated subjects (51.5 ng/mL), indicating that this sample selection would not lead to underestimation of BXA concentration in the swab samples. LC-MS/MS analysis determined that the median BXA concentration in the selected swab samples was 0.9655 ng/mL (5-95 percentile, 0.08745-2.4375 ng/mL) ( Table 1).

| Effects of BXA in the transport medium on influenza virus titer testing
When tested with the inoculum of 400 TCID 50 /mL, 30 ng/mL or higher BXA concentrations resulted in statistically significant reductions in influenza A(H3N2) viral titer compared with the sample without BXA, and 100 ng/mL or higher resulted in negative in viral titer (<LLOQ) ( Figure 1B). With an inoculum of 40,000 TCID 50 /mL, 10 ng/mL or higher BXA concentration significantly affected the viral titer, and 300 ng/mL resulted in negative ( Figure 1B).

| Comparison of declines between viral titers and RNA loads
We next compared declines in infectious viral titers with RNA at

| D ISCUSS I ON
In this study, we found that the median BXA concentration in the swab samples was 0.9655 ng/mL at Day 2. Considering that approximately 0.1 mL was collected on a swab and diluted 40-fold in 4.0 mL of the transport medium, it was estimated that roughly 40 ng/mL of BXA was delivered in the swabs, which is comparable to a plasma level of 71.5 ng/mL, in the selected samples. In addition, there is little possibility that BXA was degraded during the storage. In the previous report, a significant decline was observed in both viral titers and viral RNA in the BXM-treated group compared with the placebo and OTV groups. 3 Comparative analysis confirmed a significant correlation between viral titers and RNA in total BXM group at Day 2 (r = .69). In addition, we confirmed the absence of a significant correlation between BXA concentration in the swab samples and the viral titer reduction in the selected subjects (r = .032; Figure   S2). These data indicate that BXA in the swab samples does not interfere with the infectious titer testing. In addition, these findings provide the rationale to continue to evaluate BXM's antiviral effects in samples collected from other clinical trials, such as from high-risk influenza patients, and in a trial investigating the suppression of viral  Takao Shishido https://orcid.org/0000-0002-2549-0945 Toru Ishibashi https://orcid.org/0000-0003-3383-9980