the SafeBoosC-III trial investigates the benefit and harms of treatment based on near-infrared spectroscopy monitoring compared with treatment as usual. The hypothesis is that treatment based on near-infrared spectroscopy monitoring for extremely preterm infants during the first 72 hours of life will result in a reduction in severe brain injury or death at 36 weeks postmenstrual age.

Background Each year, 50,000 extremely preterm infants are born in high-income countries with access to neonatal intensive care. Of these, 10,000 will die and a further 10,000 will suffer cerebral palsy or moderate-to-severe neuro-cognitive disability. Time spent outside normal cerebral oxygenation ranges (time with hypoxia or hyperoxia) is associated with poor neurological outcome in children and adults. Monitoring of cerebral oxygenation may reduce the risk of cerebral complications, but no such effects have yet been demonstrated in preterm infants in large randomised clinical trials. The recently completed SafeBoosC phase II trial was conducted at eight sites in eight European countries. 166 extremely preterm infants were randomised to visible monitoring of cerebral oxygenation by near-infrared spectroscopy (NIRS) combined with an evidence-based treatment guideline (experimental group) versus blinded NIRS and treatment as usual (control group). The trial found that NIRS monitoring in combination with an evidence-based treatment guideline successfully reduced the burden of hypoxia and hyperoxia from 81% to 36%hours during the first three days of life (p<0.001). Furthermore, the proportion of severe brain injury assessed by central reading of serial cranial ultrasound was 12.5% in the experimental group versus 23.4% in the control group, RR 0.53 (95% CI: 0.26 to 1.08). Mortality was 14% in the experimental versus 25% in the control group, RR 0.50 (95% CI: 0.29 to 1.00). No other evidence has been identified. Based on these preliminary findings, we are planning a phase III randomised clinical trial; the SafeBoosC-III trial.

Objectives The overall objective of the SafeBoosC-III trial is to investigate the benefit and harms of treatment based on near-infrared spectroscopy monitoring compared with treatment as usual. The hypothesis is that treatment based on near-infrared spectroscopy monitoring for extremely preterm infants during the first 72 hours of life will result in a reduction in severe brain injury or death at 36 weeks postmenstrual age.

Trial design The SafeBoosC-III trial will be an investigator-initiated, randomized, multinational, pragmatic phase-III clinically open trial with a two-parallel group design. Sixteen-hundred extremely preterm infants will be included within 24 months at 50 neonatal intensive care units (NICUs) across 20 countries (less than two children per month per unit). Data management and statistical analysis will be blinded.

Eligibility Eligible infants will be born before 28 weeks of postmenstrual age; decision is made to provide full life support; signed informed consent (unless the NICU has chosen to use 'opt-out' or deferred consent as consent method); and cerebral NIRS oximeter placed within 6 hours after birth.

Interventions Participants in the experimental group will be monitored during the first 72 hours of life with a cerebral NIRS oximeter, placed within six hours after birth, and treated according to an evidence-based treatment guideline. Participants in the control group will not undergo cerebral oxygenation monitoring and will be treated as usual. Each participant will be followed up at 36 weeks postmenstrual age.

Outcomes The primary outcome will be a composite of severe brain injury (cerebral haemorrhage grade III or IV, cystic periventricular leukomalacia, cerebellar haemorrhage, post-haemorrhagic ventricular dilatation or cerebral atrophy) detected on any of the serial cranial ultrasound scans that are routinely performed in these infants up to that age or death until 36 weeks of postmenstrual age.

Exploratory outcomes will be a count of the presence of bronchopulmonary dysplasia, retinopathy of prematurity stage 3+, and severe brain injury as defined in the primary outcome, as well as bronchopulmonary dysplasia, retinopathy of prematurity stage +3, necrotizing enterocolitis stage 2 or higher using the modified Bell's staging and/or focal intestinal perforation, and late-onset sepsis (>72 hours after birth) defined as treatment with antibiotics for at least five days, as isolated outcomes.

Sample size We have calculated our sample size based on the primary outcome with an alpha of 5%, a power of 90%, and a ratio of 1:1 between intervention groups. In the SafeBoosC-II trial, the proportion of trial participants in the control group with death or severe brain injury was approximately 34%. Assuming the same proportion in the SafeBoosC phase III trial control group and using 22% relative risk reduction as anticipated intervention effect, we will need to randomise a total of 1,600 participants.

Safety Oximeters approved for clinical use in newborn infants will be used. Predefined serious adverse events and serious adverse reactions will be reported separately. An independent Data Monitoring and Safety Committee will be established to monitor the safety of the trial participants during the trial.

Ethical considerations Approval from the relevant ethics committees and/or Institutional Review Boards is mandatory for every participating centre. Written parental informed consent will be obtained prior to randomisation unless the ethics committee has granted permission to use deferred informed consent or prior informed assent. The trial will be conducted in compliance with the guidelines of the Declaration of Helsinki in its latest form and of the International Conference on Harmonisation Good Clinical Practice. Procedures will be established to prevent and minimise risk of complications for participants, such as complications related to the device handling. The evidence-based treatment guideline includes only interventions that are commonly used during intensive care for this population.

Trial duration Recruitment will begin in April of 2019 and will be completed within 24 months (April of 2021).

Perspectives If the experimental intervention proves successful, we may save at least 2,000 extremely preterm infants every year from death or a life with a handicap due to brain injury in high-income countries. The ensuing health economics impact running into billions of Euros saved annually.

Inclusion Criteria:

• Infants born with postmenstrual age less than 28 weeks
• Signed informed consent, unless the NICU has chosen to use 'opt-out' or deferred consent as consent method.

Exclusion Criteria:

• Missing written parental informed consent (if the 'opt-out' method is used for consent, lack of a record that the clinical staff have explained the trial and the 'opt-out' consent process to parents and/or a record in the infant's clinical file of parents' decision to opt-out, are exclusion criteria)
• Decision not to conduct full life support
• No possibility to place cerebral NIRS oximeter within six hours after birth

Due to the nature of the experimental intervention, no blinding can be done for the clinical staff and the parents. Outcome assessment of mortality will not be blinded, but the diagnosis and classification of brain injury and the entry of this into the eCRF, will be done by a person that is blinded to group allocation. Principal investigators must develop a local procedure description that describe how this is done. The local procedure description must be approved by the sponsor. Furthermore, mortality will be checked by Good Clinical Practice (GCP) through source data verification in all patients. The data managers, statisticians and those drawing conclusions will be blinded to treatment allocation. Data management will be blinded.