Safety and efficacy of AK0529 in respiratory syncytial virus‐infected infant patients: A phase 2 proof‐of‐concept trial

Abstract Background Respiratory syncytial virus (RSV) infection is a cause of substantial morbidity and mortality in young children. There is currently no effective therapy available. Methods This was a Phase 2 study of the oral RSV fusion protein inhibitor AK0529 in infants aged 1–24 months, hospitalized with RSV infection. In Part 1, patients (n = 24) were randomized 2:1 to receive a single dose of AK0529 up to 4 mg/kg or placebo. In Part 2, patients (n = 48) were randomized 2:1 to receive AK0529 at 0.5, 1, or 2 mg/kg bid or placebo for 5 days. Sparse pharmacokinetic samples were assessed using population pharmacokinetics modelling. Safety, tolerability, viral load, and respiratory signs and symptoms were assessed daily during treatment. Results No safety or tolerability signals were detected for AK0529: grade ≥3 treatment‐emergent adverse events occurring in 4.1% of patients in AK0529 and 4.2% in placebo groups, respectively, and none led to death or withdrawal from the study. In Part 2, targeted drug exposure was reached with 2 mg/kg bid. A numerically greater reduction in median viral load with 2 mg/kg bid AK0529 than with placebo at 96 h was observed. A −4.0 (95% CI: −4.51, −2.03) median reduction in Wang Respiratory Score from baseline to 96 h was observed in the 2 mg/kg group compared with −2.0 (95% CI: −3.42, −1.82) in the placebo group. Conclusions AK0529 was well tolerated in hospitalized RSV‐infected infant patients. Treatment with AK0529 2 mg/kg bid was observed to reduce viral load and Wang Respiratory Score. Clinical Trials Registration NCT02654171.

his team received additional funding from the Ministry of Health and Welfare, Taiwan (MOHW110-TDU-B-211-124002). The authors had full control over the analysis and interpretation of the data and of the decision to publish. All authors reviewed and approved the final manuscript. placebo. In Part 2, patients (n = 48) were randomized 2:1 to receive AK0529 at 0.5, 1, or 2 mg/kg bid or placebo for 5 days. Sparse pharmacokinetic samples were assessed using population pharmacokinetics modelling. Safety, tolerability, viral load, and respiratory signs and symptoms were assessed daily during treatment.

Results:
No safety or tolerability signals were detected for AK0529: grade ≥3 treatment-emergent adverse events occurring in 4.1% of patients in AK0529 and 4.2% in placebo groups, respectively, and none led to death or withdrawal from the study. In Part 2, targeted drug exposure was reached with 2 mg/kg bid. A numerically greater reduction in median viral load with 2 mg/kg bid AK0529 than with placebo at 96 h was observed. A À4.0 (95% CI: À4.51, À2.03) median reduction in Wang Respiratory Score from baseline to 96 h was observed in the 2 mg/kg group compared with À2.0 (95% CI: À3.42, À1.82) in the placebo group. in-hospital deaths in children younger than 5 years. 1 Safe and effective easily dosed therapies are largely absent. Monoclonal antibodies in development may provide certain protection from infection to infants, 2 but for those infected, the standard of care is currently limited to supportive therapy.
A promising target for RSV drug development is the class I viral trimeric fusion (F) glycoprotein, which mediates RSV entry in response to binding of the attachment (G) glycoprotein to a host receptor. Several molecules targeting F protein have been evaluated in clinical challenge models in adults as well as in children with an acute RSV infection but with limited success. [3][4][5][6] Other compounds, such as RSV replication inhibitors, have also been tested in early stage clinical trials. 7

| Study design
This was a double-blind, placebo-controlled, randomized, multicenter study with the primary objective to evaluate the safety and tolerability of single and multiple oral doses of AK0529. Secondary objectives were to evaluate the effect of AK0529 on Wang Respiratory Score, to determine the effects of treatment on viral load, and to characterize the pharmacokinetics (PK) of single and multiple AK0529 doses.
The ethics committee at each trial center approved the trial. All subjects' parents or legal guardians provided written informed consent.
The study consisted of two parts. In both parts, subjects were randomized in a 2:1 ratio to receive active treatment or placebo. Subjects were stratified by age (6 to 24 months in Cohorts 1 and 3, and 1 to <6 months in Cohorts 2 and 4).
Part 1 was a single-dose study with a 7-day follow-up. The doses administered to patients in Cohort 1 included 4, 2, and 1 mg/kg. The Safety Review Committee (SRC) reviewed PK and safety data after each of the sets of three patients, before further enrollment of subjects, and recommended 0.5 and 1 mg/kg as the doses for patients in Cohort 2.
In Part 2, subjects received AK0529 or placebo over 5 days and were followed until day 14 post-initial dose. The doses in Cohort 3 included 1 and 0.5 mg/kg bid. In Cohort 4, the doses included 2, 1 and 0.5 mg/kg bid.
Details of the rationale for the dose selection, subject dosing, and safety review schemes in Parts 1 and 2 are detailed in the Supporting Information.

| Study assessments
The Wang Respiratory Score assesses the severity of respiratory rate, wheezing, retraction of respiratory muscles, and general condition on a scale from 0 to 3 except for the general condition, which is scored 0 for normal and 3 for irritability or lethargy. The sum score, ranging between 0 and 12 (most severe), reflects the severity of the respiratory system disorders in children. 10

| Safety
Safety and tolerability was evaluated by clinical assessment of safety data, including laboratory data, electrocardiographic data, vital signs, and adverse events (AEs).

| PK
Individual plasma concentrations of the active drug were measured, and PK were determined using nonlinear mixed-effect models. In all cases, when new data were available, the updated model was used for evaluation under the assumption of an allometric coefficient of 0.75 and incorporation of CYP maturation. Using a Population PK (Pop-PK) analysis approach, AUC 0-∞ , clearance (CL/F), volume of distribution (Vc/F), and other PK parameters were estimated for AK0529.
Maximum plasma concentration (C max ) was computed from the predicted profiles.

| Statistical analysis
Clinical safety and tolerability were assessed on the safety population, comprising all subjects who received ≥1 dose of study drug. Clinical variables were evaluated in the full analysis population, which included all subjects who received a dose of study drug and for whom ≥1 post-treatment assessment was available. Pharmacokinetic analysis was performed on those subjects in the safety population for whom both pre-and post-dose samples were available.
All data are summarized by descriptive statistics with 95% confidence intervals for comparisons. As respiratory viral infections may coexist with potentially pathogenic respiratory bacteria that may affect the severity of RSV infection, 13,14 a post-hoc analysis of clinical efficacy was performed, excluding patients with bacterial pneumonia diagnosed by the treating physicians. There was no general screening for bacterial co-infection.
In study Part 1, no formal sample size calculation was performed.
For Part 2, it was calculated that a sample size of 16 patients in each active treatment group would provide >90% power to detect a 50% reduction (mean ratio of 0.5) in RSV viral load AUC with AK0529 as compared with placebo, using a one-sided significance level of 0.05, two-sample equal variance t-test, and assuming a 0.3 variation coefficient of the placebo for viral load. These assumptions were based on viral challenge models of early treatment in adults. 3 For the transaminase shift analysis, the Chi-square test and Fisher's exact test were used, as appropriate, to test for differences between active treatment and placebo.

| Study population
The study flow is shown in Figure 1. Seventy-three subjects were enrolled in the study, including 20 subjects from Australia, 39 from Taiwan, eight from Malaysia, three from Israel, and three from Turkey.
Detailed distribution of subjects in each country or region by age group and gender is provided in the Supportive Information. One randomized subject in Part 2 did not receive study medication because of early discharge by the study site and was excluded from all analyses.
One subject in Part 2 received unblinded treatment with AK0529 F I G U R E 1 Study flow chart. a One randomized patient in Part 2 received unblinded treatment with AK0529 (2 mg/kg bid) on a compassionate use basis for the best interest of the patient and was excluded from the efficacy analysis unless specified.
T A B L E 1 Demographic characteristics. Percentages do not always add up to 100% because of rounding. One subject received unblinded treatment with AK0529 (2 mg/kg bid) on a compassionate use basis and was excluded from the full analysis population.

| Reduction in viral load and S&S scores
In Part 1, reductions in viral load at 24 h after a single dose AK0529 administration were <1 log 10 PFUe/mL relative to placebo. In Part 2, the multiple 2 mg/kg bid dose was associated with 1.25 log 10 PFUe/mL greater reductions in median viral load than placebo at 72 h and 1.73 log 10 PFUe/mL greater reductions at 96 h post-dose ( Figure 2A). These differences are not statistically significant.
In detected a mixed T400I mutation in the RSV-A F gene. No associated clinical virological failure was detected. T400 is located in the intervening region between HR1 and HR2 of F protein, and its T400I mutation was known to cause a resistance to an RSV fusion inhibitor. 12

| Safety
The safety profile of AK0529 is shown in Table 3 Table 4.
Two serious TEAEs (pneumonia) were reported: one of Grade 4 15 (also counted as a serious AEs [SAE], 2 mg/kg bid, Cohort 4) and one Grade 2 (0.5 mg/kg bid, Cohort 4). The relationships to study treatment were respectively considered "not related" and "unlikely", and both AEs were resolved without sequelae.

| PK
Pharmacokinetic parameters evaluated in Part 1 are shown in Table 5.
There was a dose-dependent increase in AUC with no apparent pla-    16  A problem with some candidate molecules has been the resistance mutations in RSV strains. 4 Only one indication of resistance to AK0529 was detected in a single subject. It is debatable whether resistance is a serious concern in immune-competent subjects with a pathogen such as RSV, which has high replication rates but is rapidly cleared by the body. 23 Vigilance is nevertheless warranted.