Influenza virus infection and aerosol shedding kinetics in a controlled human infection model

Abstract

Establishing effective mitigation strategies to reduce the spread of influenza virus requires an improved understanding of the mechanisms of transmission. We evaluated the use of a controlled human infection model using an H3N2 seasonal influenza virus to study critical aspects of transmission, including symptom progression and the dynamics of virus shedding. Eight volunteers were challenged with influenza A/Perth/16/2009 (H3N2) virus between July and September 2022 at Emory University Hospital. Viral shedding in the nasopharynx, saliva, stool, urine, and respiratory aerosols was monitored over the quarantine period, and symptoms were tracked until day 15. In addition, environmental swabs were collected from participant rooms to examine fomite contamination, and participant sera were collected to assess seroconversion by hemagglutination inhibition or microneutralization assays. Among the eight participants, influenza virus infection was confirmed in six (75%). Infectious virus or viral RNA was found in multiple physiological compartments, fecal samples, aerosol particles, and on surfaces in the immediate environment. Illness was moderate, with upper respiratory symptoms dominating. In participants with the highest viral loads, antibody titers rose by day 15 post-inoculation, while in participants with low or undetectable viral loads, there was little or no increase in functional antibody titers. These data demonstrate the safety and utility of the human infection model to study features critical to influenza virus transmission dynamics in a controlled manner and will inform the design of future challenge studies focused on modeling and limiting transmission.CLINICAL TRIALSThis study is registered with ClinicalTrials.gov as NCT05332899.

Importance: We use a controlled human infection model to assess respiratory and aerosol shedding kinetics to expand our knowledge of influenza infection dynamics and help inform future studies aimed at understanding human-to-human transmission.

Keywords: aerosols; antibody response; controlled human infection model; influenza; influenza-like illness; virus shedding.

Fig 1

Participant enrollment breakdown and study schedule of events. (A) General overview of participant enrollment from recruitment to study conclusion. A total of eight participants were enrolled from July to September 2022. (B) Study timeline includes quarantine at the hospital for 7 days post-inoculation. Collection days for different sample types are denoted below the study timeline. Symptom scores from FLU-PRO were collected daily from days 1 to 15. Stool was provided by the participants based on availability. Outpatient visits were conducted on day 15 (±3), 30 (±3), and 60 (±7) post inoculation.

Fig 2

Viral shedding in nasal and oral cavities. (A) qPCR Ct values for NP swabs. Each sample was run with two distinct influenza A virus primers (see Materials and Methods for details). Amount of infectious virus present in (B) NP swabs, (C) saliva, and (D) NLF was determined from participants who tested PCR positive. The dashed horizontal line is the LOD for plaque assay, and the dashed vertical line represents the day the participants were inoculated.

Fig 3

Daily participant symptom scores and NP viral RNA shedding. The columns show symptom scores for each participant with each color indicating a specific symptom. The solid lines represent Ct values for qPCR from the NP swabs with the symbols representing two separate gene targets.

Fig 4

Viral RNA and aerosol particles in respiratory emissions. The number of viral gene copies in aerosol expulsions was determined using droplet digital PCR during (A) breathing and (B) speaking. Aerosol particle concentrations measured during (C) breathing and (D) speaking. Measured particle concentrations (for particles of 0.3–25 µm size range) were normalized to background particle concentrations in the corresponding participant’s room. The dashed horizontal line indicates a ratio of 1 with respect to the background.

Fig 5

Environmental conditions in participant rooms. (A) Average air change rates in individual participant rooms during the course of their hospital stay. (B) Daily average temperature and RH in participants’ rooms, measured next to the bed, during the course of their hospital stay. Error bars represent the SD across study days 2–7.

Fig 6

Serological responses to A/Perth/16/2009 (H3N2) and A/California/07/2009 (H1N1) virus. Sera from the study participants were tested for antibodies capable of (A and C) inhibiting hemagglutination of red blood cells or (B and D) neutralizing infection of cultured cells against A/Perth/16/2009 (H3N2) or A/California/07/2009 (H1N1), respectively. The HAI assays were run in triplicates, and MN assays were run as one replicate. The dashed line shows the limit of detection for each assay. The geometric mean for the three HAI replicates was plotted, and the SD is shown.