Despite clinical advances and decades of research, the ability to reliably predict the course of respiratory viral diseases such as influenza and coronavirus infections remains poor. The aim of this project is to develop a platform for identifying and developing predictive tests by combining physiological data and correlates of severity in influenza-like infections so that progression to severe pulmonary involvement can be anticipated during respiratory viral infection. This would then permit safe discharge of patients with self-limiting disease or more rapid intensification of treatment as appropriate.
Respiratory infections are among the most important causes of severe disease worldwide, with the major respiratory viruses responsible for overwhelming pressure on health services each winter due to annual surges in incidence. The two most common viral causes of severe lung disease, influenza and respiratory syncytial virus (RSV), are responsible for \~50% of hospital admissions in children and 22% in adults, with mortality greatest in older people. As the population ages, this burden of disease is steadily increasing. Furthermore, the continual risk of newly emergent pandemic influenza strains that arise unpredictably is universally considered one of the most critical threats to global health and socioeconomic stability. This has been demonstrated by the recent COVID-19 pandemic.
Risk factors for severe influenza have been investigated extensively in clinical cohorts, with older age, co-morbidities, obesity and pregnancy all increasing the likelihood of severe disease. However, accurate prognostic markers remain elusive and the dynamics of the response to respiratory viral infection has not been explored in naturally-infected patients. Furthermore, biomarker discovery has been limited by heterogeneity in virus strain and dose; delays in timing of presentation; and patient-level confounders. To address these issues, the investigators have conducted controlled human infections with influenza and RSV since 2010, to investigate mechanisms of immunopathogenesis with a particular focus on disease in the human respiratory tract. Recent preliminary data from a cohort of volunteers infected with the influenza A(H1N1)2009 strain showed that rapid changes in the transcriptome of whole blood occurred within 2 days of virus exposure. During the 2009 influenza pandemic, similar studies were also performed with hospitalised patients. There, transcriptomic analysis of blood showed similar antiviral signatures in less severely unwell individuals but divergent signatures associated with poor clinical outcomes.
The aim of this project is to identify and test predictors of disease progression and clinical deterioration in patients with influenza-like illness, in order to develop novel methods to more accurately determine the need for hospital admission and treatment intensification during respiratory viral infection. To further develop and test these biomarkers in an independent cohort of naturally-infected patients, hospitalised adults with influenza-like illness will be recruited within 24 hours of admission and samples obtained from blood and nose at 3 subsequent time-points.
Using these data, predictive transcriptomic signatures will be identified. Longitudinal samples and clinical data will then be used to test, validate and refine them in affected local populations. These findings will then be translated into novel diagnostic tools and a biobank established for further investigation of the virology and immunopathogenesis of severe respiratory viral infections.
