Generating Intervals of Reference FFor Early Life Brain Biomarkers. (GIRAFFE)

With highly sensitive immunoassays becoming more accessible and widely available, there has not only been an urgent call for blood-derived biomarkers for neurological insult/disease, but there has also been a major increase in the number of potential biomarkers being discovered. Neuro-specific biomarkers are being widely used in adult medicine to track neurodegenerative disorders or acquired traumatic brain injury, with information available for both physiological and pathological levels, but little is known about normal physiological levels in the neonatal population. In fact, the most common type of acquired brain injury is perinatal. Perinatal insults leading to lifelong disability make up 44% of all state claims, but currently prediction of outcome remains difficult in the early neonatal period. Before the full potential of neuro-specific biomarkers can be utilised in neonatal populations for the prediction of outcome, a better understanding of physiological levels for neuro-specific biomarkers needs to be elucidated. Many current methods of analysis for neuro-specific biomarkers involve labour-intensive wet lab work conducted by highly skilled personnel. With recent advancements in main-line clinical chemistry analysers, neuro-specific biomarkers are set to become more available for routine measurement in clinical care.

From the limited literature available on homeostatic neuro-specific biomarker levels in paediatric populations, there have already been documented differences between those observed in adulthood. For this study we will focus on those neuro-specific biomarkers which are most promising for the translation into clinical care; GFAP and Tau and specific MicroRNA (miRNA) which have shown potential for the early diagnosis of Hypoxic Ischemic Encephalopathy (HIE) and poor neurodevelopmental outcomes.

GFAP is a protein associated with astrocyte cells in the brain. It plays a pivotal role in providing structural support to astrocytes and the blood-brain barrier. In physiological conditions, the expression of GFAP outside of the Central Nervous System (CNS) does not occur. This neuro-specific property of GFAP has enabled it to become a prime candidate biomarker for the detection of insult or disease within the CNS.

GFAP has become a popular biomarker for the screening of adults for mild Traumatic Brain Injury (mTBI) since the Food and Drug Administration (FDA) approval of the Abbott i-STAT TBI Plasma test. The i-STAT TBI Plasma test provides a point-of-care device for measuring GFAP and Ubiquitin Carboxy-terminal Hydrolase-L1 (UCH-L1) for screening adults presenting with mTBI for Computed Tomography (CT) scans. Elevations in GFAP levels in adults with mTBI have been shown to correlate well with abnormal CT findings. Using the Abbott i-STAT TBI Plasma test, the GFAP cut-off levels for determining if further investigation, such as head CT, is needed, for adult mTBI are 30 pg/mL. The reference interval determined by Abbott for an adult population ranging from 18 to 79 years of age is 2-51 pg/mL.

Limited literature is available on how physiological paediatric GFAP levels differ from those in an adult. However, from the available literature, clear differences are being documented between normal paediatric and adult GFAP levels. One study investigated the changes in GFAP levels in relation to age and included a population with ages ranging from 3 to 79 years. Cooper et al. demonstrated three distinct reference intervals for age. They determined that physiological levels of GFAP differ between the following age ranges: 3 - < 10 years, 10 - < 60 years, and 60 - < 80 years. Their results detail that an approximate 7% decrease is observed in GFAP levels per year between the ages of 3 and 20. This is followed by a 2% yearly increase in normal physiological GFAP levels between the ages of 20 and 60. With final physiological increases of 3% per year observed in adults >60 years of age. A study conducted by Mannix et al. observed that 100% of healthy children under the age of 3.5 years have a GFAP level greater than that of the cut-off level determined. The study only included n = 18 participants under the age of 3.5 years, with the youngest participant being 3 months old. A sub-study conducted by Puravet et al, in France determined a reference interval for GFAP serum levels using the Abbott i-STAT TBI platform. Their study included children younger than 6 months up to the age of 16 years. They demonstrated that in healthy control children <6 months old, the mean concentration of GFAP was 96.3 ng/L. This is approximately three times more than the cut-off level determined by Abbott for the investigations of mTBI. It is also much higher than the derived reference interval for the adult population. For children <6 months, the 95th percentile observed in the study conducted by Puravet et al. was 197.57 ng/L. It should be noted that this study also had a very limited number of children <6 months old, n = 18. In comparison, as mentioned, the 95th percentile for the adult population derived by Abbott was 51 pg/mL.

From the limited available literature, it is clear that age specific normative ranges for GFAP are required. GFAP appears to exhibit a U-shaped curve with respect to age. Higher physiological levels are seen in infancy and childhood, with plateaus observed in adolescence and adulthood. Followed by physiological increases in later adulthood. As mentioned, in adult mTBI, elevations of GFAP levels indicate injury to the brain and CNS. In neonates, elevations in GFAP levels have also been demonstrated to be associated with brain abnormalities such as white matter lesions and injury to the basal ganglia and cortex. GFAP has been proposed as an early biomarker for the prediction of abnormal neurodevelopmental outcomes such as cerebral palsy, with elevations of GFAP concentrations correlating with poor neuropsychomotor outcomes.

With GFAP becoming a biomarker of interest for the early prediction of abnormal neurodevelopmental outcomes, the need for a physiological reference interval for the neonatal population is of growing importance. A reference interval for the neonatal population may further our understanding of physiological GFAP levels in early life. This may provide a foundation for the accurate interpretation of GFAP elevations in neonatal brain injury, aiding in the development of biomarker-based screening tools for early diagnosis and intervention.

Tau is a microtubule-associated protein critical for regulating neuronal functions. Its roles encompass stabilising microtubules, modulating synaptic plasticity, and facilitating axonal transport processes. Tau can also undergo many post-translational modifications, including phosphorylation. Tau protein is released when neuronal damage occurs and a strong relationship between brain injury and the measurement of Tau outside of the CNS exists in a number of conditions. Tau is well known and documented for its use in Alzheimer's research, which has a primary focus on adult and ageing populations. However, evidence is emerging that it may be a useful biomarker for mTBI and for the prediction of neurodevelopmental outcome in neonates who have suffered a brain injury at birth.

There is limited literature available on the physiological levels of Tau in early life. One study by Stukas et al examined the relationship between serum total Tau and age. They reported that serum total Tau decreases with age. Their study highlighted that there are three significant age partitions for physiological Tau: 1 - < 4 years, 4 - < 16 years, and 16 - < 19 years. This was the largest study encountered in the literature, with a total of n = 416 participants composing the control group for the establishment of the reference intervals. Another study examined the physiological levels of phosphorylated Tau-181 and determined that age-specific reference ranges were also needed for this phosphorylated derivative of Tau. They determined that there are significant differences in normative pTau-181 levels for the age groups: 3 - < 12, 12- 60, and 60 - < 80.

In addition to protein-based biomarkers, MicroRNAs (miRNAs) have emerged as promising candidates for the early detection of brain injury and prediction of neurodevelopmental outcomes. miRNAs are small endogenous RNA molecules that are released into the extracellular space. They regulate translation in eukaryotic cells at the post-transcriptional level. Through binding their target messenger RNA (mRNA) sites, they can induce translational repression or degradation. They are detectable at very small concentrations, and several miRNAs are involved in normal brain development. Our group has previously demonstrated that altered miRNA expression is present in the umbilical cord blood of neonates with Perinatal Asphyxia or HIE, two conditions known to increase the risk of poor neurodevelopmental outcomes. However, little is currently known about the physiological expression profiles of miRNA in healthy term neonates. Establishing normative comparison data for miRNA expression in this population is therefore essential to enable their clinical utility in distinguishing pathological changes from normal biological variation.

The vast majority of the current published work on paediatric neuro-specific biomarker concentrations contains much smaller sample numbers than what is recommended for the development of a reference interval. Some studies discussed only have a population size of n = 18 when the Clinical and Laboratory Standards Institute (CLSI) guidelines call for a minimum of n = 120 for the development of a reference interval. The youngest participant in the studies discussed was 3 months, meaning there is a present gap in the knowledge surrounding normal neuro-specific biomarker levels in neonates, limiting their translation into clinical care.