A Randomised, Embedded, Multi-factorial, Adaptive Platform Trial for Community-Acquired Pneumonia (REMAP-CAP) - Influenza

-Research questions

Overall platform: What are the most clinically and cost-effective pharmaceutical interventions for patients of all ages hospitalised with confirmed influenza? Domain specific (at outset): In adults and children hospitalised with confirmed influenza • What is the optimal anti-viral treatment strategy?

• Do corticosteroids improve outcomes? • What is the optimal immune modulation strategy in the severely ill?

-Design: An established UK multifactorial Bayesian adaptive platform trial for community acquired pneumonia, also recruiting internationally, using response adaptive randomisation.

-Setting: 150 hospitals throughout the UK admitting patients with influenza including under-served regions and communities. -Target population: Adults and children admitted to hospital acutely unwell with confirmed influenza. Exclusion criteria: Patient is expected to be discharged from hospital today or tomorrow

Previous participation in this trial within the last 90 days -Interventions being assessed at outset Influenza antiviral domain • Oseltamivir daily for 5 days • Oseltamivir daily for 10 days • Baloxavir on days 1 and 4 • Combination of Oseltamivir daily for 5 days and Baloxavir on days 1 and 4

• Combination of Oseltamivir daily for 10 days and Baloxavir on days 1 and 4 Initial comparator is no antiviral treatment. Corticosteroid domain • Dexamethasone for 10 days Comparator is no corticosteroid treatment Immune modulation domain • Tocilizumab • Baricitinib Initial comparator is no immune modulation treatment.

The design allows comparative effectiveness between interventions in a domain to be estimated. The multi-factorial design allows interactions between treatments in different domains to be evaluated. Measurement of outcomes & costs: Primary: 28-day organ support-free days, including mortality

Secondary: Symptom severity, WHO ordinal scale, need for intensive care (ICU) admission, need for ventilation, duration of ICU and hospital stay, 90- and 180-day mortality, serious adverse events / reactions, health-related quality of life (EQ-5D-5L), disability (WHODAS), cost effectiveness at 6 months. Development of anti-viral resistance.

Statistical methods: A Bayesian hierarchical model, with pre-defined triggers for efficacy and futility (to control), superiority and equivalence (to other treatments). Stratification will allow separate treatment effects to be estimated for critically ill and non-critically ill patients, as well as adults and children.

Timelines for delivery: International recruitment has begun (antiviral n=150, corticosteroids n=144). Approvals for UK ICU recruitment are in place. The adaptive Bayesian design allows conclusions as soon as triggers are reached rather than fixed sample sizes. International participation (Southern hemisphere) will ensure year-round recruitment and rapid results for a seasonal illness ahead of rises in UK cases.

Anticipated impact and dissemination: REMAP-CAP changed practice globally, saving lives in severe COVID-19, and can impact influenza treatment in the same way.