Orexin-A Trajectories in Newborn Brain Injury Reveal Cooling’s Hidden Signature

In the fragile hours after birth, when a newborn’s brain has been starved of oxygen, clinicians have remarkably few tools to tell them how deep the injury runs or how the brain is responding to treatment. Hypoxic-ischemic encephalopathy, or HIE, affects roughly one to three per thousand live births in high-income countries and remains a leading cause of lifelong neurological disability. The only established therapy, therapeutic hypothermia, involves deliberately cooling the infant’s body to around 33.5 degrees Celsius for seventy-two hours, a maneuver that has cut death and severe disability rates since its adoption two decades ago. Yet no blood-based biomarker has ever been validated for routine clinical use in this condition. A new prospective longitudinal study published in BMC Pediatrics now offers a tantalizing glimpse of one candidate: orexin-A, a small neuropeptide produced by a few thousand neurons in the hypothalamus, whose behavior in the bloodstream during the first days of life appears to track the brain’s functional state with unexpected sensitivity.

Orexin-A, also known as hypocretin-1, is best known to sleep scientists as the molecule whose loss causes narcolepsy. But its repertoire extends far beyond keeping animals awake. The peptide binds two G-protein-coupled receptors, OX1R and OX2R, distributed across brain regions that govern arousal, autonomic tone, and stress responses. In experimental models of cerebral hypoxia-ischemia, orexin signaling has repeatedly shown neuroprotective properties, apparently acting through pathways that include hypoxia-inducible factor 1-alpha, the master transcriptional switch that cells deploy when oxygen runs low. This dual identity, as both a sleep regulator and a stress-response molecule, made orexin-A an intriguing target for researchers hunting for a window into the injured newborn brain. If the peptide’s concentration in blood reflected the brain’s struggle to cope with oxygen deprivation, it might serve as a real-time reporter of neurological status in a way that static measures of tissue damage cannot.

The research team, led by pediatric neurologists Pınar Gençpınar and Gunce Basarir and spanning multiple institutions in İzmir and Istanbul, Türkiye, enrolled forty-one newborns born at thirty-six weeks of gestation or later, all diagnosed with HIE within the first six hours after birth. The design was deliberately longitudinal: blood samples were drawn twice, once between six and twelve hours of life and again between seventy-two and ninety-six hours, capturing the earliest window in which clinicians must make decisions and the period immediately after cooling ends. Twenty-two of the infants underwent therapeutic hypothermia according to standard criteria, while nineteen, whose severity did not meet cooling thresholds or for whom cooling was not indicated, were managed with supportive normothermic care. Serum orexin-A concentrations were quantified using enzyme-linked immunosorbent assay, and the researchers cross-referenced the peptide’s behavior with an unusually rich clinical dataset: amplitude-integrated electroencephalography, conventional video EEG, brain magnetic resonance imaging, and routine metabolic indices.

The headline finding is subtle but striking. At neither time point did absolute orexin-A concentrations differ significantly between the cooled and non-cooled groups, a result that on its own might have seemed like a dead end for the biomarker hypothesis. The story emerged only when the researchers examined each infant’s trajectory over time. Among babies who received therapeutic hypothermia, orexin-A levels were relatively preserved or even rose between the two measurements, with a median within-subject change of plus 31.2 picograms per milliliter. Among babies managed without cooling, the peptide fell sharply, with a median change of minus 54.5 picograms per milliliter. The difference between these trajectories was statistically robust, with a p-value of 0.002, and significantly more infants in the hypothermia group showed a higher value at the second sampling than at the first. In the normothermia group, the decline over time was itself significant, at p equal to 0.005.

What makes this pattern scientifically interesting is that it inverts the usual logic of injury biomarkers. Molecules such as neuron-specific enolase, glial fibrillary acidic protein, neurofilament light chain, and ubiquitin C-terminal hydrolase-L1 are released when cells die, so higher concentrations generally signal worse damage. Orexin-A appears to behave differently. The authors interpret it as a marker of functional state rather than tissue injury: a brain that is still actively regulating arousal, autonomic function, and stress responses keeps producing and releasing orexin, while a more profoundly depressed brain lets circulating levels drift downward. The longitudinal change also differed significantly by amplitude-integrated EEG background, with a p-value of 0.010, suggesting that the peptide’s trajectory carries electrophysiological meaning. However, the researchers note an important caveat here: abnormal EEG tracings were almost entirely confined to the hypothermia group, which complicates any attempt to disentangle the peptide’s relationship with brain function from the effects of illness severity itself.

Indeed, the study is unusually candid about its central limitation. Treatment allocation in HIE is inseparable from disease severity, because cooling is prescribed precisely for the sickest infants. The preserved or rising orexin-A trajectories seen under hypothermia therefore cannot be attributed to cooling itself; they may simply reflect that cooled infants, despite being more severely affected, occupied a different physiological regime, or that the interplay between severity and treatment produced the observed divergence. The authors explicitly state that these trajectories cannot be ascribed to the cooling intervention, a level of interpretive restraint that is rare and welcome in biomarker research. Similarly, higher concentrations at the second time point tracked clinical severity only in outlier-sensitive statistical analyses, hinting that a small number of extreme values may have driven part of the apparent signal, and that the peptide’s relationship with severity is weaker and less linear than a simple biomarker model would predict.

Equally informative are the relationships that failed to materialize. Neither the absolute concentrations nor the longitudinal changes of orexin-A correlated with video EEG findings, with abnormalities on brain MRI, or with routine metabolic indices. This null pattern reinforces the idea that orexin-A is not reporting on structural damage. MRI lesions and epileptiform discharges reflect injury to neural tissue, whereas the peptide’s trajectory seems to reflect something closer to the brain’s ongoing regulatory activity. If this distinction holds up in future work, it would position orexin-A as a complementary rather than redundant marker, one that could in principle answer a question that damage markers cannot: not how much brain has been lost, but how much functional capacity remains and how it is evolving hour by hour.

The technical execution of the study deserves attention as well. Measuring orexin-A reliably in neonatal serum is demanding, since the peptide is present at low concentrations and ELISA-based assays are sensitive to pre-analytical handling. The two-sample design, anchored to clinically meaningful windows, allowed the researchers to compute within-subject changes, which are far more informative than cross-sectional comparisons in a condition as dynamic as HIE. The study was funded by the Scientific and Technological Research Council of Türkiye under its 1002 support module, and the results were previously presented at the fifteenth Congress of the European Paediatric Neurology Society in Prague in June 2023, indicating that the findings have circulated through the specialist community before formal publication.

For clinicians and families, the practical message for now is one of tempered promise. Orexin-A is not ready for the neonatal intensive care unit. The authors are explicit that serial sampling in larger, severity-matched cohorts, ideally with long-term neurodevelopmental follow-up, is required before any clinical role can be defined. Such studies would need to disentangle the effects of cooling from the effects of severity, perhaps by comparing infants with matched severity who did and did not receive cooling, or by tracking orexin-A in animal models where treatment can be randomized. They would also need to establish whether the trajectory of the peptide predicts outcomes such as cerebral palsy, cognitive impairment, or epilepsy, which is the endpoint that ultimately matters to families.

Still, the conceptual contribution of this work may prove durable. In a field dominated by markers of destruction, a molecule that reports on the brain’s living regulatory activity offers a genuinely different lens on neonatal brain injury. The finding that the direction of change, rather than any single concentration, carries the signal is a methodological lesson that extends beyond orexin-A: biomarkers of dynamic physiological states may only reveal their value when sampled repeatedly and interpreted as trajectories. As the search for clinically usable biomarkers in HIE continues, this small Turkish study of forty-one newborns adds a compelling hypothesis to the pipeline, and a reminder that in the injured newborn brain, the most telling numbers may be the ones that move.

Subject of Research: Serum orexin-A dynamics in neonatal hypoxic-ischemic encephalopathy and therapeutic hypothermia

Article Title: Temporal profile of serum orexin-A in neonatal hypoxic-ischemic encephalopathy: a prospective longitudinal study

Article References: Gençpınar, P., Basarir, G., Ersen, A., Engür, D., Olgaç Dündar, N., Üstün, H., Kefeli Demirel, M., Sarıoğlu, F. C., & İşbilen Başok, B. (2026). Temporal profile of serum orexin-A in neonatal hypoxic-ischemic encephalopathy: a prospective longitudinal study. BMC Pediatrics, 26(1), Article 968. https://doi.org/10.1186/s12887-026-07727-0

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07727-0

Keywords: orexin-A, hypoxic-ischemic encephalopathy, therapeutic hypothermia, biomarker, newborn, hypothalamus, neuropeptide, electroencephalography, brain MRI, neonatology, neuroprotection, hypocretin