High dose EPO – friend or foe in the newborn?
H. Liley
DOI: https://doi.org/10.1113/jphysiol.2014.274522
2014-05-01
Journal of Physiology
Abstract:Most infants who receive neonatal intensive care survive and do well. Nevertheless, rates of cognitive impairment are inversely proportional to gestation at birth. Mitigating brain injury in newborns, both in vulnerable extremely preterm infants, and in those who have suffered complications during term birth, is a pre-eminent ambition, because it has the potential for lifelong impact for babies and their families. The causal pathways and patterns of brain injury after preterm birth and after hypoxia–ischaemia during term birth differ, though there are some shared pathophysiological pathways and downstream consequences, leading to the possibility that some neuroprotective and neuroregenerative treatments may work for both.
The cytokine erythropoietin (EPO) is a glycoprotein comprising 165 amino acids, heavily glycosylated at four sites. It is well known as an essential growth factor for the recruitment of erythropoietic progenitor cells towards a haematopoietic fate and their subsequent survival, and it exerts these effects sequentially in the yolk sac, the fetal liver and then the bone marrow. EPO is produced in the kidney and liver in response to hypoxia, and serves to increase red cell mass and therefore oxygen-carrying capacity. EPO binds dimeric receptors, activating several intracellular pathways, including Janus kinase-signal transduction, transcription activation, phosphatidylinositol 3-kinase and mitogen-activated kinase pathways (Elliott et al. 2008).
Synthetic erythropoiesis-stimulating agents (ESAs), either recombinant human EPO (rhEPO, known by the international non-proprietary name epoietin) or its analogues are widely used for the treatment of the erythropoietin-deficiency anaemia that can complicate chronic renal failure. In addition, ESAs have also been tested for other types of anaemia, including that due to prematurity. Very premature infants typically have insufficient erythropoiesis to cope with growth and phlebotomy losses, and some have significant comorbidities which limit nutrition or further suppress erythropoiesis; ESAs can reduce transfusion requirements. However, as with many other hormones, the most visible or first-described role is not the whole of the EPO story.
EPO receptors have been reported in most tissues, albeit at low levels compared to the bone marrow, indicating the potential for EPO effects throughout the body, although some of these results have been questioned because of recent contradictory results using monoclonal antibodies (Sinclair et al. 2010). EPO mRNA is also found in a variety of organs other than the kidney, including the brain and the lung, and it may be translated and have paracrine effects in these organs.
EPO and EPO receptors are also expressed in the central nervous system throughout life, but especially in the developing brain, and brain EPO expression is induced by hypoxia (Bernaudin et al. 2002). Because of a wealth of supportive preclinical research, ESAs are the subject of several current and planned trials as neuroprotective agents after premature birth or hypoxia–ischaemia during term birth. Data from two human trials and a primate study show benefit in hypoxic ischaemic encephalopathy at term, and several non-randomised studies suggest benefit in preterm babies. High systemic doses are considered necessary to achieve meaningful effects within the brain. Neuroprotective roles for EPO in mitigation of adult stroke and in depressive illness have also been proposed.
Human trials of epioetin for anaemia of prematurity have shown moderate reductions, although not prevention, of the need for transfusion. Early epoietin may increase the risk of retinopathy of prematurity, but there has been a notable lack of other recognised adverse events. Neither the increased mortality risk nor the anti-EPO antibody-associated pure red cell aplasia that have been seen in some adult studies have been found in infants. Previous human and animal studies suggest that epoietin may have beneficial effects on premature infant lung.
To date, studies of epoietin in human infants have not suggested lung toxicity. However, in an article in this issue of The Journal of Physiology, Polglase and colleagues raise concern that high dose epoietin may have harmful effects on the developing lung, exacerbating ventilator-induced lung injury (Polglase et al. 2014). In their study of premature lambs ventilated with high tidal volumes and no positive end-expiratory pressure, epoietin alfa at 5000 IU kg−1 exacerbated physiological, biochemical and histological evidence of injury in the lungs as well as serum amyloid A-3 levels in the liver.
However, adverse effects of epoietin could be highly dose, agent and context specific (Horl, 2013). The doses used in the Polglase study were high when compared with endogenous EPO levels, even in hypoxic states, and when compared with the doses used to prevent or treat anaemia of prematurity (typically 90–1400 IU week−1, often used subcutaneously rather than intravenously, so resulting in lower peak levels) and the doses in current trials for neuroprotection.
Effects on other organs may depend on whether the copious numbers of receptors in the bone marrow have been saturated. Furthermore, in both adults and infants, epoietin clearance is dose dependent and saturable. The main sites of clearance are unknown, but may include uptake into EPO receptor-bearing cells, followed by degradation. These factors could strongly influence the dose dependence of adverse effects.
There are several forms of epoietin; alfa, beta, omega and delta and other biosimilar epoietins have been developed. Most have identical amino acid sequences, but variable glycans, and there is no research to show whether they have different pharmacokinetic or pharmacodynamic effects in premature human newborns.
Lastly, it is hard to think of a hormone or cytokine that does not have precise effects at different times during embryonic, fetal and subsequent development, so the adverse or beneficial effect profile of most hormones used for treatment is likely to be highly specific to gestational and post-natal age.
Nevertheless, these results in premature lambs raise concerns that high dose epoietin might have previously unrecognised adverse effects in human infants, and underscore the need for ongoing research on ESAs in infants to include pharmacokinetics and consideration of a range of potential adverse outcomes.