Daily mosnodenvir dosing as dengue prophylaxis in a human infection model

Abstract

Background:

Antiviral prophylaxis or treatment is unavailable for dengue. We evaluated the antiviral activity, safety and pharmacokinetics of repeated oral doses of mosnodenvir (JNJ-64281802), a pan-serotype dengue antiviral, as pre-exposure prophylaxis against dengue virus serotype 3 (DENV-3) infection in a controlled human infection model (CHIM).

Methods:

In this phase 2a, double-blind study, healthy adults were randomized to receive once daily oral mosnodenvir at different dose levels or placebo for 26 days (5 days loading dose [LD], 21 days maintenance dose [MD]). An under-attenuated DENV-3 strain (rDEN3Δ30) was subcutaneously injected on the day (D) of first MD (D1). Safety, pharmacokinetics, virology and serology parameters were evaluated up to D85.

Results:

The proportion of participants without signs of DENV-3 infection was 0% (0/6), 17% (1/6), 60% (6/10) in low-, medium-, high-dose regimens, respectively, versus 0% (0/7) in the placebo arm. A Tobit analysis of variance showed a statistically significant reduction on log₁₀ AUC_D1-D29 VL in mosnodenvir high-dose versus placebo participants (2-sided p<0.001). In this small study, mosnodenvir did not have any apparent safety concerns. Mosnodenvir plasma concentrations increased from D-5 to D1 and were maintained up to D21. Emergent amino acid variations in the NS4B region were detected in all 14 participants with available NS4B sequencing data in the mosnodenvir arms, while none were observed in the placebo arm

Conclusion:

Mosnodenvir significantly reduced DENV-3 viral load versus placebo in a dose-dependent manner without relevant clinical safety findings in a CHIM.

Dengue is an acute disease caused by four antigenically different serotypes of dengue virus (DENV-1–4), transmitted by Aedes mosquitoes.¹ Dengue is a growing public health threat, with about half of the world population being considered at risk by the World Health Organization.² Dengue continues to increase due to population growth and climate changes allowing vector expansion in endemic areas, such as the Americas and Southeast Asia and is spreading to temperate regions in the USA and Europe.^3–7

Dengue disease can manifest as undifferentiated fever or non-severe dengue fever.⁸ In a small proportion of cases, the disease can progress to severe manifestations such as dengue hemorrhagic fever or dengue shock syndrome, with a potentially fatal outcome.⁹ Licensed dengue treatments are currently unavailable and management consists of supportive measures, such as antipyretics and volume repletion.¹⁰ Two live attenuated tetravalent vaccines, CYD-TDV (which recently stopped production) and TAK-003 (in certain countries), are currently approved for use.¹¹ A third, Butantan-DV, has filed for licensure in Brazil.^12–14 Vaccines can take weeks to provide protective immunity and have varying efficacy by baseline DENV immunity.^13,15 In contrast, an oral antiviral could be deployed in outbreaks, could limit cases in endemic areas through a prophylactic approach, and may be used for travelers and those who cannot receive a vaccine.^16–18

Mosnodenvir (also known as JNJ-64281802) is an oral, pan-serotype dengue small-molecule antiviral, that blocks viral replication by inhibiting de novo DENV nonstructural protein 3 (NS3)-NS4B interaction.^19,20 Mosnodenvir exhibits picomolar to nanomolar in vitro antiviral potency against a representative DENV genotype panel and has antiviral efficacy in mice and non-human primates.¹⁹ In a first-in-human study, no safety concerns were identified with mosnodenvir.²¹ Here, we report the initial results of a phase 2a study assessing the prophylactic antiviral activity, safety, and pharmacokinetics of different daily doses of mosnodenvir against DENV-3 infection (under-attenuated strain) in healthy participants.

Methods

Study oversight

This is a phase 2a, randomized, double-blind, placebo-controlled human infection model (CHIM) study conducted at the Johns Hopkins School of Public Health and the University of Vermont, United States (US), starting in February 2022 (ClinicalTrials.gov number, NCT05048875). The study consists of two cohorts conducted in a staggered manner: Cohort 1 is dose finding with 2 dose escalation groups and Cohort 2 assesses three different regimens based on Cohort 1’s findings, including weekly dose regimens. Cohort 1, Group 1 is high dose 600 mg daily (QD) loading dose (LD)/200 mg QD maintenance dose [MD] and placebo; and Cohort 1, Group 2 is medium dose 200 mg QD LD/50 mg QD MD and low dose 40 mg QD LD/10 mg QD MD and placebo (Figure 1). The data from Cohort 1 are presented here.

Figure 1.

*Group 1 was split into a sentinel group (Group 1a) with N=4 (2:2 randomization between placebo and 600 mg LD/200 mg MD mosnodenvir) who were enrolled first and the remaining participants (Group 1b) (4:8 randomization between placebo and 600 mg QD LD/200 mg QD MD mosnodenvir).

†One participant in the mosnodenvir arm withdrew consent on D-2 (before inoculation) because of moderate photosensitivity (considered related to study drug by the investigator) and was replaced according to protocol.

DENV, dengue virus; LD, loading dose; MD, maintenance dose; PK, pharmacokinetics; QD, daily; rDEN3Δ30, under-attenuated DENV-3 virus.

The study is conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki. Written informed consent was obtained from each participant prior to any study-related activities. Mosnodenvir was supplied as 10 mg, 50 mg, and 100 mg oral capsules and administered under fasted conditions. The study protocol (available with the full text of this article at NEJM.org) was reviewed and approved by Independent Ethics Committees at each site and written informed consent was obtained from each participant. Data were gathered by the study site investigators and analyzed at Johnson & Johnson in collaboration with the National Institutes of Health (NIH). All the authors vouch for the accuracy and completeness of the data presented and for the fidelity of the study to the protocol. Medical writing assistance was funded by Johnson and Johnson.

Participants

We enrolled healthy individuals 18–55 years of age, who were confirmed to be seronegative to DENV and Zika virus (ZIKV) prior to enrollment, had not traveled to any dengue-endemic region within 4 weeks from enrollment nor planned to do so and had not received any live attenuated vaccines within 28 days before and after study drug intake. For more details on the inclusion and exclusion criteria, see the protocol.

Study procedures

All participants attended screening visits between day (D)-65 and D-6, received either mosnodenvir or placebo (oral QD and under fasted conditions) as a LD from D-5 to D-1, followed by a MD from D1 to D21, with a challenge of 3 log10 plaque-forming units (PFU) of the under-attenuated virus rDEN3Δ30,²² administered subcutaneously on D1. Participants were randomized to receive either high-dose mosnodenvir; (N=10) or matching placebo (N=6) (Group 1) or medium- (N=6) and low-dose (N=6) mosnodenvir or matching placebo (N=2) (Group 2) (Figure 1). All participants were admitted to the inpatient unit for the first two dosing days (D-6 through D-4) and observed for at least 30 minutes after initial dosing and/or inoculation to ensure their safety. Safety parameters were monitored throughout the study (clinical laboratory tests, electrocardiogram, vital signs, and physical examinations) and solicited and unsolicited adverse events (AEs) were evaluated. Blood samples were taken at regular time points for virology and pharmacokinetic assessments. Participants were followed up through D85 (64 days after the last dose). More information about the study procedures is available in the Supplementary Appendix and protocol.

Study objectives

The primary objective was to assess the antiviral activity of mosnodenvir versus placebo in terms of reduction of DENV-3 RNA by evaluating the area under the DENV-3 RNA viral load (VL) concentration-time curves from immediately before inoculation (D1) until D29 (AUCD1-D29). Secondary objectives included safety and tolerability, occurrence and severity of DENV infection-associated AEs, other virology parameters, antibody responses, pharmacokinetics of mosnodenvir and the characterization of the relationship between pharmacokinetics and antiviral activity of mosnodenvir under different QD dose regimens.

Virology and pharmacokinetic assessments

DENV-3 RNA serum levels were assessed using a validated quantitative DENV reverse transcriptase polymerase chain reaction (RT-qPCR) assay. DENV-3 viremia was determined on DENV-3 RNA positive samples using a plaque assay and anti-DENV immunoglobulin (Ig)G and IgM antibodies were measured by enzyme-linked immunosorbent assay (ELISA; Euroimmun). DENV-3 viral sequencing was performed using Illumina sequencing technology to characterize emerging DENV-3 genetic variations. Emerging amino acid variations were defined as having a sequence read frequency ≥15% at a post-baseline visit while absent and a read frequency <3% in the inoculated rDEN3Δ30 strain sequence. Blood samples were obtained over 24 hours on D5 and D21 to measure mosnodenvir plasma concentrations using a validated, liquid chromatography-tandem mass spectrometry method.²³ Pharmacokinetic parameters, including maximum plasma concentration (C_max), time to C_max (t_max), average concentration (C_avg), terminal elimination half-life (t_½) and area under the plasma concentration-time curve (AUC), were determined using the validated software Phoenix (Certara, Princeton, NJ, USA). Additional details about these and other assessments (e.g., Saint-Louis encephalitis virus [SLEV], ZIKV, reporter virus particle neutralization tests (RVPNT)) are available in the Supplementary Appendix and protocol.

Sample size calculation

Data were simulated using a Bernoulli distribution for the infection rate (assuming 90% under placebo) and using a normal distribution for the log₁₀ AUC_D1-D29 (VL) in the infected participants (mean 5.5 log₁₀ copies/mL/28 days, standard deviation 0.70). Based on these simulations, the power to detect a relevant reduction of ≥ 30% on log₁₀ AUC_D1-D29 (VL) at the 2-sided 10% significance level was calculated to be more than 85% with 6 participants in the placebo arm and 10 participants in the mosnodenvir high-dose arm. The number of participants in Cohort 1 is considered sufficient for an initial characterization of the relationship between pharmacokinetics and antiviral activity of mosnodenvir based on DENV-3 RNA. Details can be found in the protocol.

Statistical analysis

The primary efficacy analysis included all participants from Cohort 1 Group 1 who were inoculated with rDEN3Δ30. A Tobit analysis of variance with log₁₀ AUC_D1-D29 (VL) as dependent variable and the study drug as a fixed covariate was performed to test whether a significant difference between mosnodenvir and placebo was observed, at the 2-sided 10% significance level, provided that at least 65% of the inoculated participants in the placebo arm had detectable DENV-3 RNA at any of the assessments up to D29. Values were left censored for participants with undetectable DENV-3 RNA up to D29. The exact Wilcoxon rank sum test was also performed. Descriptive statistics are provided on the primary and secondary endpoints. Analyses were performed with SAS 9.04 (SAS Institute Inc., Cary, NC, USA). Graphs were created in R version 4.2.0 (Comprehensive R Network, http://cran.r-project.org/).

Results

Participants

In Cohort 1, 31 participants were recruited between February 2022 and February 2023 and were included in the safety analysis set. Thirty (30) participants were randomized to receive either mosnodenvir (N=22) or placebo (N=8) (Table 1, Figure 1, Figure S1), 1 additional participant was enrolled as a replacement for a participant in the mosnodenvir group and 29 inoculated participants were included in the efficacy analysis. The baseline characteristics were similar between the placebo and mosnodenvir as well as across different dosing arms (Table 1) except for more females included in the medium-dose arm (83%) when compared to the other arms (Table 1). In the placebo arm, one participant missed a single dose on Day 18 and another missed a single dose on Day 21 and took this dose on Day 22 instead. All other participants from Cohort 1 were 100% medication compliant. The representativeness of the study population is shown in Table S1.

Table 1.

Analysis sets and baseline characteristics of study participants

	Mosnodenvir
	Placebo arm N=8	Low dose* N=6	Medium dose* N=6	High dose* N=11	Combined N=23	Total N=31
Randomized, n	8	6	6	10 (+ 1 replacement ‡)	22 (+ 1 replacement ‡)	30 (+1 replacement)
Received study drug, n	8	6	6	11	23	31
Inoculated with rDEN3Δ30, n (%)	7 (87.5)°	6 (100.0)	6 (100.0)	10 (90.9)	22 (95.7)	29 (93.5)
Completed study drug, n (%)†	7 (87.5)	6 (100.0)	6 (100.0)	10 (90.9)	22 (95.7)	29 (93.5)
Completed study, n (%)	7 (87.5)	6 (100.0)	6 (100.0)	10 (90.9)	22 (95.7)	29 (93.5)‡
Included in the safety analysis set, n (%)§	8 (100.0)	6 (100.0)	6 (100.0)	11 (100)	23 (100)	31 (100.0)
Included in the efficacy analysis set, n (%)§	7 (87.5)°	6 (100.0)	6 (100.0)	10 (90.9)‖	22 (95.7)	29 (93.5)
Included in the PK analysis set, n (%)§	0⁋	6 (100.0)	6 (100.0)	11 (100)	23 (100)	23 (74.2)
Mean age (SD), years	30.0 (7.29)	29.5 (4.59)	36.0 (10.68)	34.5 (9.08)	33.6 (8.64)	32.6 (8.35)
Female sex, n (%)	4 (50.0)	4 (66.7)	5 (83.3)	7 (63.6)	16 (69.6)	20 (64.5)
Race, n (%)**
N	8	5	6	11	22	30
American Indian or Alaska Native	1 (12.5)	0	0	0	0	1 (3.3)
Asian	0	0	1 (16.7)	1 (9.1)	2 (9.1)	2 (6.7)
Black or African American	3 (37.5)	3 (60.0)	3 (50.0)	5 (45.5)	11 (50.0)	14 (46.7)
White	3 (37.5)	2 (40.0)	2 (33.3)	3 (27.3)	7 (31.8)	10 (33.3)
Multiple	1 (12.5)	0	0	2 (18.2)	2 (8.79.1)	3 (10.0)
Ethnicity, n (%)‖
N	8	5	6	11	22	30
Hispanic or Latino	1 (12.5)	0	0	0	0	1 (3.3)
Not Hispanic or Latino	7 (87.5)	5 (100)	6 (100.0)	11 (100.0)	22 (100)	29 (96.7)
Site, n (%)
Johns Hopkins University, Baltimore, MD	5 (62.5)	3 (50.0)	3 (50.0)	7 (63.6)	13 (56.5)	18 (58.1)
University of Vermont, Burlington, VT	3 (37.5)	3 (50.0)	3 (50.0)	4 (36.4)	10 (43.5)	13 (41.9)

^*Actual treatments received: Mosnodenvir low dose 40 mg QD loading dose (LD)/10 mg QD maintenance dose (MD); medium dose 200 mg QD LD/50 mg QD MD; high dose 600 mg QD LD/200 mg QD MD.

^†Study drug intake was considered completed if study drug was taken at D21, irrespective of missed doses before.

^‡One participant in the mosnodenvir arm withdrew consent on D-2 (before inoculation) because of moderate photosensitivity (considered related to study drug by the investigator) and was replaced according to protocol.

^§All participants from Cohort 1 who received at least 1 dose of mosnodenvir were included in the safety analysis. The efficacy analysis included all participants from Cohort 1 who were inoculated with rDEN3Δ30. There were no inoculated participants with protocol deviations that could affect the antiviral activity assessment. The PK analysis included all participants with at least one available PK result after dosing.

^°One participant in the placebo arm of Group 2 arm discontinued the dosing before being challenged because of severe elevated liver function tests and life-threatening elevated creatine kinase tests. This participant was unblinded and not replaced.

^‖One participant in the high-dose mosnodenvir arm had a missing sample on D29 and was excluded from the primary analysis since the missing value could not be imputed according to pre-specified statistical analysis plan rules.

^⁋PK samples from participants assigned to placebo (except for samples taken pre-dose on D-5) were not analyzed.

^**Race and ethnicity were reported by the participants.

N, number of participants included in each arm, reflecting non-missing values; n (%), number (percent) of participants, PK, pharmacokinetic; SD, standard deviation.

Primary endpoint analysis

A Tobit analysis of variance showed a statistically significant reduction on the log₁₀ AUC_D1-D29 VL in the high-dose mosnodenvir arm versus the placebo arm (2-sided p<0.001) (Figure 2A and Table S2). The statistically significant result was also seen by the exact Wilcoxon rank sum test (2-sided p=0.001).

Figure 2.

DENV-3 RNA log10 AUC_D1-D29 (VL) (A) Kaplan-Meier plot of time to first onset of detectable DENV-3 RNA from Day 1 to Day 29 (B) Efficacy analysis set

A, Solid black dots represent the individual log10 AUCD1-D29 (viral load) values. Box plots represent the 25th, 50th and 75th percentile. Dots at the censoring (target not detected, TND) line represent participants with undetectable DENV-3 RNA throughout D29. One participant in the mosnodenvir high-dose arm had a missing sample on D29 and was excluded from the primary analysis AUC, area under the DENV-3 RNA viral load (VL) concentration-time curves from D1 to D29; VL, viral load.

B, All participants from Group 1 with one participant (mosnodenvir arm) censored on D25. The time to first detectable DENV-3 RNA was defined as the number of days between the date of inoculation and the date of the first occurrence of the event + 1. If a detectable event was not measured, the participant was censored at the last available sample with a valid result. Shaded area indicates 95% confidence intervals.

Daily dose regimens

A dose-dependent antiviral activity, as measured by log₁₀ AUC values, across the different mosnodenvir regimens was observed (Figure 2). The proportion of participants with all available DENV-3 RNA measurements being undetectable was 0% (0/6), 17% (1/6), 60% (6/10) in the low-, medium-, and high-dose arms, respectively, versus 0% (0/7) in the placebo arm (Figure S2). In all mosnodenvir-dosed participants without detectable DENV-3 RNA, no anti-DENV IgM/IgG nor neutralizing antibodies (nAbs) were observed until D85 (Figure S2, Table S3).

For participants with detectable DENV-3 RNA in the mosnodenvir dose regimens, the peak DENV-3 RNA levels were comparable to placebo, with the exception of 3 participants with peak DENV-3 RNA levels detectable below or around the lower limit of quantification (LLOQ) (Figure S2); and the median time to first onset of detectable DENV-3 RNA was delayed in a dose-dependent manner (Figure 2). In all participants with detectable DENV-3 RNA, infectious virus was detected, except for 1 participant in the medium-dose arm and 1 participant in the high-dose arm (Figure S2), and positive anti-DENV IgM and/or positive anti-DENV IgG and/or positive nAbs on ≥1 assessment after baseline were observed, except for 1 participant in the medium-dose arm with detectable DENV RNA (<LLOQ) at a single-timepoint (Figure S2 and Table S3).

A DENV-associated rash, defined as rash in combination with detectable DENV-3 RNA, was reported in 100% (7/7) of the participants in the placebo arm versus 83% (5/6), 50% (3/6), and 30% (3/10) of participants in low-, medium-, and high-dose arm, respectively (Figure S2). Most rashes were reported within 2 days after peak VL.

Viral genome sequencing

Emergent amino acid variations in the NS4B region were detected in each of the 14 participants with available NS4B sequencing data in the mosnodenvir dose arms, while none were observed in the placebo arm. The most frequent emergent NS4B variations were V91A (N=9), V91G (N=2), P104L (N=2), T215S (N=3), and A233P (N=4). All those emerging NS4B variant frequencies were consistently >99% in the high-dose arm, with lower frequencies (15% to >99%) observed in the low/mid-dose arms. Emergent variations outside the NS4B region, which occurred in ≥2 participants, were A20T in the 2K region (Table S4).

Safety

All participants reported at least one AE starting from first study drug intake until study termination (Table 2). Most AEs were mild (grade 1) to moderate (grade 2), occurred with similar frequency across all dosing arms and resolved without complications. There were 2 mosnodenvir-dosed participants with severe AEs, a participant in the medium-dose mosnodenvir arm with severe increase of lipase and glycemia and a participant in the high-dose mosnodenvir arm with severe COVID-19 infection. These severe AEs were reported during follow-up. One participant withdrew consent after receiving 4 days of mosnodenvir and before inoculation, due to moderate photosensitivity, considered related to mosnodenvir. None of the mosnodenvir dosed participants had serious AEs, and none died. All out-of-range laboratory findings were isolated and fully reversible.

Table 2:

Overview clinical safety data

A: Adverse events from D-5 to D85 (Safety analysis set)
		Mosnodenvir
	Placebo N=8	Low Dose* N=6	Medium Dose*N=6	High Dose* N=11	Combined N=23
Participants with at least 1 adverse event, n (%)#
Any	8 (100.0)	6 (100.0)	6 (100.0)	11 (100.0)	23 (100.0)
Related to study drug†	3 (37.5)	3 (50.0)	4 (66.7)	5 (45.5)	12 (52.2)
Related to rDEN3Δ30 inoculation†	5 (62.5)	5 (83.3)	5 (83.3)	3 (27.3)	13 (56.5)
Serious AEs	1 (12.5)	0	0	0	0
AEs leading to discontinuation of any study treatment	1 (12.5)	0	0	0	0
Grade 1 as worst severity (mild)	3 (37.5)	2 (33.3%)	2 (33.3)	5 (45.5)	9 (39.1)
Grade 2 as worst severity (moderate)	4 (50.0)	4 (66.7%)	3 (50.0)	5 (45.5)	12 (52.2)
Grade 3 as worst severity (severe)	1 (12.5)	0	1 (16.7)	1 (9.1)¤	2 (8.7)
Grade 4 as worst severity (life-threatening)	0	0	0	0	0
Participants with the most common adverse events reported, n (%)‡
Investigations	3 (37.5)	5 (83.3)	4 (66.7)	7 (63.6)	16 (69.6)
Hemoglobin decreased	0	1 (16.7)	0	3 (27.3)	4 (17.4)
General disorders and administration site conditions	1 (12.5)	4 (66.7)	3 (50.0)	5 (45.5)	12 (52.2)
Vessel puncture site bruise	0	4 (66.7)	3 (50.0)	4 (36.4)	11 (47.8)
Skin And Subcutaneous Tissue
Disorders	2 (25.0)	2 (33.3)	4 (66.7)	6 (54.5)	12 (52.2)
Ecchymosis	0	1 (16.7)	1 (16.7)	2 (18.2)	4 (17.4)
Nervous System Disorders	3 (37.5)	3 (50.0)	2 (33.3)	4 (36.4)	9 (39.1)
Headache	3 (37.5)	2 (33.3)	1 (16.7)	4 (36.4)	7 (30.4)
Blood And Lymphatic System
Disorders	2 (25.0)	2 (33.3)	3 (50.0)	0	5 (21.7)
Lymphadenopathy	1 (12.5)	2 (33.3)	3 (50.0)	0	5 (21.7)
B: Dengue infection-associated adverse events reported during challenge phase from D1 to D43 (Safety analysis set)
		Mosnodenvir
	Placebo N=8	Low Dose* N=6	Medium Dose* N=6	High Dose* N=11	Combined N=23
Participants with at least 1 DENV infection-associated adverse event§ by worst severity, n (%)
Grade 1 as worst severity (mild)	3 (37.5)	2 (33.3)	3 (50.0)	6 (54.5)	11 (47.8)
Grade 2 as worst severity (moderate)	4 (50.0)	4 (66.7)	1 (16.7)	1 (9.1)	6 (26.1)
Grade 3 as worst severity (severe)	0	0	0	0	0
Grade 4 as worst severity (life-threatening)	0	0	0	0	0
Participants with at least 1 DENV infection-associated adverse event§ by preferred term, n (%)
Headache	6 (75.0)	6 (100.0)	3 (50.0)	4 (36.4)	13 (56.5)
Rash Maculo-Papular	7 (87.5)	5 (83.3)	4 (66.7)	3 (27.3)	12 (52.2)
Fatigue	5 (62.5)	4 (66.7)	1 (16.7)	3 (27.3)	8 (34.8)
Myalgia	3 (37.5)	5 (83.3)	3 (50.0)	0	8 (34.8)
Eye Pain	3 (37.5)	3 (50.0)	2 (33.3)	0	5 (21.7)
Nausea	2 (25.0)	3 (50.0)	1 (16.7)	1 (9.1)	5 (21.7)
Abdominal Pain	1 (12.5)	3 (50.0)	1 (16.7)	0	4 (17.4)
Arthralgia	2 (25.0)	2 (33.3)	1 (16.7)	1 (9.1%)	4 (17.4)
Decreased Appetite	3 (37.5)	1 (16.7)	2 (33.3)	0	3 (13.0)
Diarrhea	1 (12.5)	2 (33.3)	0	0	2 (8.7)
Pyrexia	1 (12.5)	0	1 (16.7)	0	1 (4.3)
Vomiting	2 (25.0)	1 (16.7)	0	0	1 (4.3)

^*Actual treatments received: Mosnodenvir low dose 40 mg QD loading dose (LD)/10 mg QD maintenance dose (MD); medium dose 200 mg QD LD/50 mg QD MD; high dose 600 mg QD LD/200 mg QD MD.

^#Solicited systemic AEs events reported from challenge up to and including the D43 visit are excluded from Table 2A as they are considered as DENV infection-associated AEs and are described in Table 2B. The headaches included in Table 2A are reported after D43.

^†Whether an AE was possibly related to mosnodenvir or to rDEN3Δ30 inoculation was determined by the principal investigator.

^‡The most common AEs were those that occurred in 4 or more participants who received mosnodenvir.

^¤A severe COVID-19 infection during the follow-up period

^§DENV infection-associated AEs are defined as solicited systemic AEs events reported from challenge up to and including the D43 visit. The challenge phase used is extended until visit D43, considering the half-life (~10 days) of the study intervention.

Participants were counted only once for any given event, regardless of the number of times they experienced the event. The event experienced by the participant with the worst severity was used.

AEs are coded using the Medical Dictionary for Regulatory Activities Version 25.0.

AE, adverse event; N, number of participants included in each arm; n, number of participants.

Pharmacokinetics

Mosnodenvir plasma concentrations rapidly increased with the LD from D-5 to D1 and were maintained up to D21, after which concentrations declined slowly consistent with a long t_1/2 (7.5–11 days) observed in this study (Table S5) and 6.3–9.2 days in the first-in-human study²¹ (Figure 3). Individual maintenance concentrations were well separated between the different dose regimens. On D-5 and on D21, median t_max was approximately 8 hours post-dose with individual t_max ranging from 2.02 to 12.10 hours. All pharmacokinetic parameters (e.g., C_max and AUC_24h) on D-5 and D21 are summarized in Figure 3 and Table S5.

Figure 3.

Individual mosnodenvir plasma concentration-time profiles (PK analysis set) through Day 85 (A), and for Day −5 (B).

LD, loading dose; MD, maintenance dose. Mosnodenvir plasma concentrations below the lower limit of quantification (LLOQ, 5 ng/mL) are not displayed.

Discussion

Mosnodenvir demonstrated dose-dependent antiviral activity against rDEN3Δ30 in healthy participants establishing inhibition of NS3-4B interaction as a viable target. While CHIMs have been successfully employed to study dengue vaccine candidates,^24–28 we now extend this model to prevention of DENV infection in humans with a dengue antiviral. Consistent with earlier studies in this CHIM^24–28, the infection rate in our study was 100% in placebo participants with uniform viral profiles, immune responses, and mild DENV-associated symptoms. The primary objective of this study was met, Tobit ANOVA analysis showed a statistically significant decrease in the log₁₀ AUC_D1-D29 [VL] in the high-dose mosnodenvir arm when compared to placebo. Furthermore, mosnodenvir administered in low, medium and high dose prevented DENV-3 infection (undetectable DENV RNA, no anti-DENV IgM/IgG/nAb seroconversion and no DENV-associated rash) in a dose-dependent manner, demonstrating that mosnodenvir can reduce the incidence of DENV-3 infection and associated symptomatology, prophylactically in a dengue-naïve population. Further evaluation of the concentration-effect relationship is ongoing, including viral kinetic modeling.²⁹ The pharmacokinetic results of mosnodenvir were consistent with those observed in the first-in-human study.²¹ Five days of daily loading doses allowed mosnodenvir plasma concentrations to reach steady-state levels at the time of DENV-3 inoculation.

Emerging variations in NS4B were detected in all (14/14) participants with NS4B sequencing data available in the mosnodenvir arms, as compared to none (0/7) of the placebo participants. The associated individual risks based on these limited data are considered minor as the peak and duration of the DENV-3 RNA levels were comparable or lower than those in the placebo arm. In addition, the frequency and severity of associated symptoms were not increased. The variations that emerged under mosnodenvir dosing were consistent with preclinical findings, pointing to NS4B as the target for mosnodenvir. Although certain variations in NS4B detected in this study have been noted in local outbreaks over time,³⁰ widespread occurrence among contemporary virus strains has not been described.^31,32 An increasing diversity of circulating DENV strains has been observed, exemplified by the co-circulation of the 4 DENV serotypes during the recent dengue outbreaks in the Americas.^33–36 Future field studies might provide insights in the emergence of variations in contemporary virus strains, considering this high diversity and the natural patterns of dengue seasonality.^37,38 The conditions of a CHIM are carefully controlled, using a well-characterized under-attenuated DENV-3 strain with consistent DENV RNA/viremia profiles and an acceptable safety profile. Those studies are therefore well positioned to assess novel DENV interventions in early clinical studies, also given the relatively small sample size needed. Findings from those studies must be confirmed in larger safety and efficacy studies in the target population since CHIM studies do not reflect field conditions. Taken together, the favorable pre-clinical profile (pan-serotype activity in vitro and prophylactic antiviral activity against RNAemia in mice/NHPs¹⁹) together with the data from this CHIM study support further development of mosnodenvir. In a phase 2 clinical field study (NCT05201794), it will be further explored whether prophylactic administration of mosnodenvir has an impact on the number of (symptomatic) DENV infections in endemic regions.

In this initial analysis, we show that mosnodenvir can prevent DENV-3 infection and associated symptoms in a dose-dependent manner in a controlled human infection setting. The advancement to Cohort 2 of the CHIM study will allow us to further characterize the relationship between mosnodenvir pharmacokinetics and antiviral activity for daily and weekly maintenance dosing. These results will complement the findings of the prophylaxis phase 2 clinical field study (NCT05201794).