COLLABORATE: a precision-medicine platform trial to improve the care of sick and preterm newborn babies
- COLLABORATE is a paradigm-shifting approach to enable neonatal teams to contribute collectively to resolving important, longstanding practice uncertainties and generate high quality evidence to advance neonatal care more quickly, efficiently, and cost-effectively than has been hitherto possible.
- COLLABORATE is a planned neonatal precision-medicine platform trial to examine the efficacy of nutritional practices as well as new interventions to prevent or treat necrotising enterocolitis (NEC) in very preterm neonates and test mechanistic hypotheses relating to their mode of action.
- The infrastructure developed as part of COLLABORATE will be made available to other researchers and trials units to address additional areas of neonatal care.
- COLLABORATE co-chief Investigators are Victoria Cornelius, Professor of Statistics and Trial Methodology, Head of the Imperial Clinical Trials Unit; and Neena Modi, Professor of Neonatal Medicine, Imperial College London
- The COLLABORATE Advisory Board includes representation from the UK NEC Society, and Adult Preemie Advocacy Network; Bliss is a co-applicant.
- COLLABORATE welcomes the involvement of all neonatal units in the UK, Australia, and New Zealand
COLLABORATE is a precision-medicine platform trial. In a precision medicine trial, researchers randomise patients to treatments according to characteristics that are likely to affect the efficacy of the intervention in an individual. In a platform trial researchers test multiple therapies in a continuous manner, introducing or discarding interventions as data accumulate. We will use a trial design known as REMAP (Randomised, Embedded, Multifactorial, Adaptive Platform). In a REMAP trial:
- Patient are randomised to intervention or control, which may be active or placebo (this is the randomised element)
- The trial is conducted within a normal clinical care setting (this is the embedded element)
- Multiple questions can be evaluated simultaneously (this is the multifactorial element)
- Recruitment can be structured to increase statistical efficiency; the design of the trial can be altered; dose-outcome relationships can be assessed; the trial can be stopped when sufficient data have accrued rather than when a pre-specified sample size is reached; the trial can also be extended and/or have added arms incorporated (this is the adaptive element)
- Interactions between different interventions can be evaluated
- The precision medicine factors we will incorporate are sex, gestational age, and degree of intrauterine growth restriction.
- We will use routinely recorded clinical data available from the National Neonatal Research Database to reduce costs and burden of data collection and improve trial efficiency.
- We will employ a Bayesian approach to further improve trial efficiency; a Bayesian design does not incorporate a fixed sample size; instead, recruitment is stopped when sufficient data have accumulated to show benefit, harm, or futility; we have conducted extensive simulation studies to identify the likely number of recruits required.
- We are co-designing the trial with parents, former patients, clinicians, and if applicable, industry partners.
- Mechanistic studies will focus on intestinal health and brain development.
- We have received an “Accelerated Development Award” from the Efficacy and Mechanism programme of the UK National Institute for Health Research to enable us to design and prepare for delivering COLLABORATE. The one-year “Accelerated Development Award” commenced on 1st September 2022. We will submit a stage 1 application for full funding in May 2023, and if this is successful, a stage 2 application in September 2023, with a planned start date in 2024.
- We have held multiple webinars and focus group sessions with clinical teams, patients, and parents. We have invited proposals for interventions, defined research questions and drawn up a list of initial treatments to be evaluated. We are currently finalising co-applicants, advisors, and committee members.
- If you have an intervention you would like considered for inclusion in the COLLABORATE platform, please contact email@example.com or firstname.lastname@example.org.
Initial research questions in PICO (Population; Intervention; Comparator: Outcome) format, that will be addressed in COLLABORATE:
- In babies born below 29 weeks gestation (Population), when the availability of own mother’s milk volume is insufficient, does supplementation with pasteurised human donor milk (Intervention), compared with supplementation with preterm formula (Comparator), affect the likelihood of survival to 34 weeks postmenstrual age without NEC surgery (primary Outcome), neurodevelopment at age two-years, and other outcomes?
- In babies born below 29 weeks gestation (Population), does a daily probiotic* supplement (Intervention), compared with no supplementation (Comparator), affect the likelihood of survival to 34 weeks postmenstrual age without NEC surgery (primary Outcome), neurodevelopment at age two-years, and other outcomes? *We have received expressions of interest from multiple companies marketing neonatal probiotic products in the UK
- In babies born below 29 weeks gestation (Population), does routine macronutrient fortification of human milk feeds with a) cow-milk based fortifier (Intervention) or b) pooled human-milk based fortifier (Intervention), compared with fortification only when weight velocity falls below a predefined threshold (Comparator), affect the likelihood of survival to 34 weeks postmenstrual age without NEC surgery (primary Outcome), neurodevelopment at age two-years, and other outcomes?
- In babies born below 32 weeks gestation who require NEC surgery (Population), does fluorescence-imaging-guided surgery (Intervention), compared with conventional visually guided surgery (Comparator), affect the duration of total parenteral nutrition (primary Outcome) and other outcomes?
- In babies born below 32 weeks gestation commencing post-operative care following NEC surgery (Population), does a) enteral insulin (Intervention), or b) enteral arginine (Intervention), compared with placebo (Comparator), affect the duration of total parenteral nutrition (primary Outcome) and other outcomes?