In 2013, nearly four million babies were born in the U.S., among whom 447,875 were born preterm. Approximately 30,000 of these infants were born before 28 weeks of gestation. These infants, termed Extremely Low Gestational Age Neonates (ELGANs), experience high morbidity and mortality despite modern therapies: approximately 20 % of ELGANs admitted to an NICU die before discharge, 20 % of survivors have severe, and 20 % moderate neurodevelopmental impairment (NDI). New approaches are needed to improve neonatal outcomes. Recombinant erythropoietin (Epo) is a promising neuroprotective agent that is widely available, affordable, and has been used safely in neonates to stimulate erythropoiesis. There are extensive preclinical data to support its use as a neuroprotective intervention: Epo promotes normal brain maturation by increasing neurogenesis, angiogenesis, and by protecting oligodendrocytes. Epo also decreases acute brain injury following hypoxia ischemia by decreasing inflammation, oxidative and excitotoxic injury, resulting in decreased apoptosis. Despite the availability of both preclinical and safety data there has not been a definitive clinical evaluation of the benefit of Epo, and a large phase III trial is necessary to provide evidence to support potential changes in practice guidelines. We first review the preclinical data motivating further clinical trials, and then describe in detail the design of the PENUT study (Preterm Epo Neuroprotection). PENUT is a phase III study evaluating the effect of neonatal Epo treatment on the combined outcome of death or severe NDI among ELGANS. 940 subjects will be randomized to determine: 1) whether Epo decreases the combined outcome of death or NDI at 22–26 months corrected age; 2) the safety of high dose Epo administration to ELGANs; 3) whether Epo treatment decreases serial measures of circulating inflammatory mediators, and improves biomarkers of brain injury; and 4) whether Epo treatment improves brain structure at 36 weeks postmenstrual age as measured by MRI. Epo neuroprotection is an exciting new approach to preterm neuroprotection, and if efficacious, will provide a much-needed therapy for this group of vulnerable infants.

Infants born extremely preterm (<28 weeks' gestation) are at high risk of neurodevelopmental impairment (NDI) with 50% of survivors showing moderate or severe NDI when at 2 years of age. We sought to develop novel models by which to predict neurodevelopmental outcomes, hypothesizing that combining baseline characteristics at birth with medical care and environmental exposures would produce the most accurate model. Using a prospective database of 692 infants from the Preterm Epo Neuroprotection (PENUT) Trial, which was carried out between December 2013 and September 2016, we developed three predictive algorithms of increasing complexity using a Bayesian Additive Regression Trees (BART) machine learning approach to predict both NDI and continuous Bayley Scales of Infant and Toddler Development 3rd ed subscales at 2 year follow-up using: 1) the 5 variables used in the National Institute of Child Health and Human Development (NICHD) Extremely Preterm Birth Outcomes Tool, 2) 21 variables associated with outcomes in extremely preterm (EP) infants, and 3) a hypothesis-free approach using 133 potential variables available for infants in the PENUT database. The NICHD 5-variable model predicted 3-4% of the variance in the Bayley subscale scores, and predicted NDI with an area under the receiver operator curve (AUROC, 95% CI) of 0.62 (0.56-0.69). Accuracy increased to 12-20% of variance explained and an AUROC of 0.77 (0.72-0.83) when using the 21 pre-selected clinical variables. Hypothesis-free variable selection using BART resulted in models that explained 20-31% of Bayley subscale scores and AUROC of 0.87 (0.83-0.91) for severe NDI, with good calibration across the range of outcome predictions. However, even with the most accurate models, the average prediction error for the Bayley subscale predictions was around 14-15 points, leading to wide prediction intervals. Higher total transfusion volume was the most important predictor of severe NDI and lower Bayley scores across all subscales. While the machine learning BART approach meaningfully improved predictive accuracy above a widely used prediction tool (NICHD) as well as a model utilizing NDI-associated clinical characteristics, the average error remained approximately 1 standard deviation on either side of the true value. Although dichotomous NDI prediction using BART was more accurate than has been previously reported, and certain clinical variables such as transfusion exposure were meaningfully predictive of outcomes, our results emphasize the fact that the field is still not able to accurately predict the results of complex long-term assessments such as Bayley subscales in infants born EP even when using rich datasets and advanced analytic methods. This highlights the ongoing need for long-term follow-up of all EP infants. Supported by the National Institute of Neurological Disorders and StrokeU01NS077953 and U01NS077955.