Real-world data-enabled, 2-randomisation, adaptive trial to determine the clinical efficacy and safety of widely used enteral feeds in preterm babies: the COLLABORATE trial – protocol

Introduction Enteral feed type is widely believed to affect brain growth and development directly as well as influencing the risk of necrotising enterocolitis (NEC), a leading cause of death and neuro-impairment in extremely preterm babies. Own mother’s milk (OMM) enhances neurodevelopment and reduces NEC. However, most babies born below 29 weeks gestation require some supplementary feeds during their neonatal unit stay because their mothers are unable to express sufficient milk. The available options, pasteurised human donor milk (pHDM) and preterm formula (PTF), differ markedly in nutrient and non-nutrient composition and cost. Non-definitive data suggest superiority of pHDM over PTF in relation to NEC risk, but no benefits have been shown in important corroboratory outcomes: surgical NEC, death, blood stream infection and neurodevelopmental impairment. Human milk macronutrient content is variable and often low, hence some clinicians practise routine protein-carbohydrate fortification. Others fear that cow-milk derived fortifiers may increase NEC risk or may provide excessively high protein intakes dangerous to neurodevelopment and metabolic health. These uncertainties compromise patient care and safety and are strong justification for randomised evaluation. Methods and analysis COLLABORATE is an efficient UK-wide, real-world data-enabled, 2-randomisation, group-sequential adaptive trial embedded in routine clinical practice. The primary outcome is survival to 34 weeks postmenstrual age without NEC surgery. The aim is to resolve two longstanding nutritional uncertainties for babies born below 29 weeks gestation: whether pHDM or PTF is the optimal supplement should there be an insufficiency of OMM; and whether OMM and pHDM require routine protein-carbohydrate fortification. Based on the assumption that 50% of participants will take part in both randomisations, total recruitment is targeted at 3252 (5% between-arm primary outcome difference; α=0.05; minimum 87% power). Ethics and dissemination COLLABORATE is funded by the UK National Institute for Health and Care Research, sponsored by Imperial College London and approved by the UK Health Research Authority (reference 25/LO/0697). Recruitment commenced in April 2026. Results will be shared through academic and lay publications, conferences, workshops, social media and strong personal networks. Trial registration number ISRCTN10443084 .