Erythropoietin beyond red blood cells: Systemic and oncological functions (Review)

Erythropoietin (Epo) as a cytokine, hormone and growth factor

Human Epo is a thermally stable hormone with numerous well-characterized biological functions, some of which remain unclear. During prenatal development, Epo is primarily synthesized by foetal hepatocytes, but postnatal production shifts to renal peritubular fibroblasts and is regulated by hypoxia-inducible transcription factors (HIFs). This hormone is essential for initiating the haematopoietic cascade and promoting erythroid differentiation in the bone marrow, a process that is coordinated by multiple cytokines, growth factors and transcriptional regulators (1).

The structure of this acidic glycoprotein, with a molecular weight of 30.4 kDa, is composed of a 165-amino-acid polypeptide stabilized by two disulfide bonds. Approximately 40% of its mass is contributed by glycans, including three N-linked and one O-linked structure, whose main function is to protect the Epo molecule from degradation (2).

The discovery of Epo represents more than a century-long scientific journey from early physiological observations to the elucidation of molecular oxygen-sensing mechanisms (Fig. 1). Following the initial proposal of a circulating erythropoietic factor by Carnot and Deflandre in 1906 (3,4), the kidney was identified as the primary source of Epo production (5). The synthesis is regulated by the partial pressure of oxygen in blood. The resulting hormone drives the proliferation and differentiation of erythroid progenitors in the bone marrow, thereby increasing erythrocyte mass and the oxygen-carrying capacity of blood (6,7). Erslev et al (8) explained the lack of a positive relationship between erythrocytosis, blood viscosity and hypoxia by demonstrating that renal oxygenation depends primarily on oxygen consumption associated with sodium reabsorption, which is proportional to glomerular filtration. Reduced renal blood flow therefore decreases oxygen demand, preserving tissue oxygenation and preventing inappropriate Epo upregulation, as observed in disorders such as polycythaemia vera (8).

Figure 1. Schematic figure summarizing the major historical discoveries and scientific recognition of Epo. The timeline shows the major developments leading from the initial physiological evidence of a circulating erythropoietic factor, through the identification of the kidney as the principal source of Epo, the purification and cloning of the Epo gene and elucidation of the HIF-mediated oxygen-sensing pathway regulating Epo expression, culminating in the 2019 Nobel Prize for discoveries of cellular oxygen-sensing mechanisms. Epo, erythropoietin; HIF-1, hypoxia-inducible factor 1.

Under hypoxia, renal peritubular interstitial cells increase Epo production by stabilizing HIF-1α (9), which dimerizes with HIF-1β and binds the hypoxia-responsive element in the Epo promoter (10). The increase in Epo levels then provide negative feedback by reducing HIF-1α activity (11). The discovery of HIF as the key transcriptional regulator of Epo expression under hypoxia was awarded the 2019 Nobel Prize in Physiology or Medicine (1).

In addition, various forms of organ injury can also induce Epo expression. Circulating Epo binds to cells expressing the Epo receptor (EPOR), which is present not only in haematopoietic tissues, skeletal muscle, endothelial and neural tissues, but also in numerous organs, including the kidney, heart, reproductive system and pancreas (12).

EPOR and its activation

The effects of Epo are mediated by binding to EPOR, leading to activation of multiple intracellular signalling pathways. Although Epo/EPOR signalling was originally characterized in erythroid precursor cells, EPOR is expressed in a variety of non-haematopoietic cells, including endothelial cells, cardiomyocytes, neurons, immune cells and several tumour cell types. EPOR activation therefore elicits context-specific biological responses that are cell type- and microenvironment-dependent (13,14).

Epo signalling in blood cells

Epo controls red blood cell production by interacting with EPOR, a type I cytokine receptor commonly observed as a preformed homodimer. Ligand binding induces receptor conformational change, which in turn triggers the activation of Janus kinase 2 (JAK2). JAK2 then phosphorylates particular tyrosine residues on the intracellular domain of EPOR (15).

Two principal signalling cascades mediate the cellular effects of Epo/EPOR engagement. In erythroid progenitor cells, the JAK2-signal transducer and activator of transcription 5 (STAT5) axis represents the dominant signalling pathway. Epo binding to EPOR induces autophosphorylation of JAK2, which subsequently phosphorylates the transcription factor STAT5. Phosphorylated STAT5 dimerizes and translocates to the nucleus, where it regulates the expression of genes responsible for erythroid cell survival, proliferation and differentiation, including BCL2L1 (Bcl-xL), c-MYC, members of the Pim kinase family and D-type cyclins (13). In parallel, the PI3K-AKT and MAPK-ERK pathways are activated, promoting proliferation, inhibiting apoptosis and ensuring the proper course of erythropoiesis (14,16).

While deletion of Epo, EPOR or JAK2 leads to embryonic mortality and termination of erythropoiesis, constitutive activation of STAT5 can partially rectify these deficiencies and underscore its critical effector function (17). Functional genomics studies (14,17) have demonstrated that STAT5 interacts with essential regulators of erythroid cells (GATA1, TAL1 and KLF1) at enhancer elements to regulate chromatin organization and transcriptional programs vital for the terminal differentiation of erythroid cells (18).

Epo signalling in non-haematopoietic cells

Although EPOR expression is relatively low in non-erythroid tissues, accumulating evidence (12,19,20) indicates that Epo exerts significant biological functions beyond haematopoiesis. The non-haematopoietic effects of Epo are mediated by a different receptor mechanism: Unlike the classic haematopoietic signalling pathway, tissue-protective effects are mediated by a heteromeric receptor consisting of monomeric EPOR units linked to a common target structure shared by other cytokines. This entire EPOR structure exhibits a lower affinity for Epo, and it is precisely this alternative receptor system that allows Epo to function as a cytoprotective molecule produced locally in response to current metabolic stress or injury (21).

Through this receptor architecture, EPOR activation in non-haematopoietic and malignant cells engages distinct combinations of the downstream cascades described previously, leading to angiogenesis, cytoprotection, neuroprotection, and in malignant settings, tumour progression and immune evasion (13,14). A summary of the principal Epo/EPOR signalling pathways across the major cell types in which EPOR is functionally characterized is provided in Table I and Fig. 2 (22–35).

Figure 2. Epo/EPOR signalling pathways in haematopoietic and non-haematopoietic cells. Created in https://BioRender.com. Epo, erythropoietin; EPOR, Epo receptor; βcR, β common receptor (CD131); ELK1, ETS transcription factor ELK1; CREB, cAMP response element-binding protein; JAK2, Janus kinase 2; eNOS, endothelial nitric oxide synthase; NO, nitric oxide; AP-1, activator protein 1.

Table I. Cell-type specific Epo/EPOR major downstream pathways activated by EPOR engagement, principal target genes and effectors, and resulting biological functions across major cell types in which EPOR has been functionally characterized.

Table I.

Cell-type specific Epo/EPOR major downstream pathways activated by EPOR engagement, principal target genes and effectors, and resulting biological functions across major cell types in which EPOR has been functionally characterized.

A, Erythroid progenitors (CFU-E, proerythroblasts)
Major downstream pathways	Principal target genes/effectors	Biological functions	(Refs.)
JAK2-STAT5 (dominant pathway)	BCL2L1 (Bcl-xL), c-MYC, Pim-1/ Pim-2, Cyclin D1/D2	Survival, proliferation, erythroid differentiation, inhibition of apoptosis	(13,14,22)
B, Erythroid progenitors
Major downstream pathways	Principal target genes/effectors	Biological functions	(Refs.)
PI3K-AKT	FOXO3, mTOR, GSK3β	Cell survival, proliferation, iron metabolism, erythroid maturation	(14,23)
MAPK-ERK	ELK1, AP-1 transcription factors	Cell-cycle progression and proliferation	(16)
C, Haematopoietic stem/progenitor cells
Major downstream pathways	Principal target genes/effectors	Biological functions	(Refs.)
JAK2-STAT5, PI3K-AKT	Bcl-xL, Pim kinases	Maintenance of erythropoiesis and progenitor expansion	(13)
D, Endothelial cells
Major downstream pathways	Principal target genes/effectors	Biological functions	(Refs.)
PI3K-AKT-eNOS	eNOS, NO signalling	Angiogenesis, endothelial survival, vascular protection	(24)
MAPK-ERK	VEGF-associated signalling molecules	Cell migration and angiogenesis	(25)
E, Cardiomyocytes
Major downstream pathways	Principal target genes/effectors	Biological functions	(Refs.)
PI3K-AKT	Bcl-2, Bcl-xL	Cytoprotection, anti-apoptotic effects, ischaemic protection	(26)
JAK2-STAT5	STAT5-dependent survival genes	Protection against ischaemia-reperfusion injury	(27)
F, Neurons
Major downstream pathways	Principal target genes/effectors	Biological functions	(Refs.)
JAK2-STAT5, PI3K-AKT	Bcl-xL, NF-κB-regulated genes	Neuroprotection, anti-apoptotic signalling, neuronal survival	(28)
MAPK-ERK	CREB-associated pathways	Neuronal differentiation and regeneration	(29)
G, Tumour cells
Major downstream pathways	Principal target genes/effectors	Biological functions	(Refs.)
JAK2-STAT5	Bcl-xL, c-Myc, Cyclin D	Cell survival, proliferation, therapy resistance	(30,36)
PI3K-AKT/mTOR	mTOR, BAD, survivin	Anti-apoptotic signalling, proliferation, resistance to therapy	(31)
MAPK-ERK	AP-1, cyclins	Tumour growth and progression	(32)
HIF-1α/VEGF axis	VEGF, angiogenic mediators	Angiogenesis, hypoxia adaptation, metastasis	(33,34)
H, Tumour-associated macrophages
Major downstream pathways	Principal target genes/effectors	Biological functions	(Refs.)
EPOR-NRF2 signalling	Immunosuppressive macrophage genes	Immune evasion and suppression of anti-tumour immunity	(35)

[i] CFU-E, colony-forming unit-erythroid; Epo, erythropoietin; EPOR, Epo receptor; NO, nitric oxide; eNOS, endothelial NO synthase.

Cardiac and vascular tissue

Epo protects endothelial cells by activating AKT1, maintaining mitochondrial integrity and modulating nitric oxide (NO) production via endothelial NO synthase (eNOS) (24,36,37). In experimental models, these effects have been associated with arterial strengthening and reduced ischaemic damage (37).

Renal epithelium

In vitro studies suggest that Epo may protect primary renal tubular epithelial cells from ischaemic damage. To protect such cells, the JAK2-STAT5 signalling pathway must be activated, which occurs via phosphorylation of EPOR at tyrosine 343. This increases Pim-3 gene expression, which is necessary for longer cell survival (38).

Nerve tissue

In hippocampal neurons devoid of STAT5, Epo retains its neuroprotective properties against glutamate-induced toxicity, whereas its neurotrophic effects are abolished. This indicates that PI3K-AKT signalling predominantly promotes neuroprotection, whereas STAT5 is crucial for trophic effects. Other neuroprotective methods use the NF-κB and ERK pathways to induce cellular responses that mitigate inflammation and cell death (39).

Regulation of metabolic processes and adipose tissue

In animal studies (19,22,40), Epo alters metabolism and inflammation in white adipose tissue, exhibiting sex-dependent effects. In male mice on a high-fat diet, Epo diminishes fat mass gain and inflammatory infiltration. Epo influences hypothalamic proopiomelanocortin (POMC) neurons through STAT3 signalling in the brain, thereby promoting leanness. In female mice, oestrogen receptor signalling seems to be more important than the metabolic effects of Epo (41).

Regeneration of tissue and remodelling of bone

Epo supports tissue repair and regeneration, including the restoration of skeletal muscle and changes in bone shape. The effects may pertain to EPOR expression on mesenchymal stem cells and its correlation with Wnt/β-catenin signalling; however, the molecular specifics are currently under investigation (41).

Together, these tissue-specific observations illustrate that EPOR engages overlapping but functionally distinct signalling programs depending on the cellular context.

Haematopoietic effects of Epo

Epo principally regulates erythrocyte production. The haematopoietic function of Epo is primarily localized to the bone marrow in adults and in the foetal liver during embryonic development (40). As aforementioned, this process occurs in adults under hypoxic conditions and is mediated by HIF transcription factors. Haematopoiesis begins with the differentiation of haematopoietic stem cells through several progenitor stages, including colony-forming units (CFUs) generating granulocytes, erythrocytes, monocytes and megakaryocytes, and burst-forming units-erythroid (BFU-E) cells. BFU-E cells differentiate into CFU-erythroid (CFU-E) cells, which express high levels of EPOR and require Epo for survival (2). The erythropoietic effect of Epo is initiated by its binding to the EPOR homodimer on these progenitors. This triggers signal transduction, primarily through activation and tyrosine phosphorylation of JAK2 kinase. JAK2 signalling then activates STAT5, which translocates to the nucleus and induces the expression of target genes required for proliferation and differentiation. The PI3K/AKT pathway is also activated, providing essential anti-apoptotic signals to erythroid cells (16,42). This signalling cascade supports CFU-E cell division, generating erythroblasts that ultimately extrude their nuclei to become reticulocytes. Mature erythrocytes lack EPOR, and the Epo/EPOR complex is internalised and degraded, thereby terminating Epo signalling (2).

Roles of Epo in the reproductive system

One of the most important discoveries in recent decades has been the identification of a fully autonomous Epo/EPOR signalling system in human reproductive tissues. Unlike classical hypoxia-driven renal Epo production, reproductive Epo acts primarily through local, oestrogen-dependent regulatory mechanisms and functions as a paracrine and autocrine growth factor (43). Although the role of Epo in female tissues has been extensively studied, the male reproductive system remains largely unexplored, representing a substantial gap in current understanding.

Female reproductive system

In the female reproductive system, Epo production is stimulated by oestradiol-17β, unlike the hypoxia-driven regulation observed in the kidney, liver and brain. The uterus is a unique site of physiological neovascularisation in adults, supported by a local paracrine Epo/EPOR system within endometrial tissues. Activation of this pathway promotes oestrogen-dependent angiogenesis through Epo expression in endometrial vascular endothelial cells. Thus, Epo functions as a local growth factor that contributes to cyclic remodelling of the endometrium across the menstrual cycle (33). The role of Epo as a growth factor is further supported by its autocrine mechanism of action in endometrial tissue. Epo and EPOR mRNA levels are higher during the secretory phase of the menstrual cycle than during the proliferative phase and are regulated by ovarian steroids (44,45).

In addition to its presence in the endometrium, Epo mRNA has been confirmed in other female reproductive organs, including the ovary and normal cervix, using various analytical methods. Epo was subsequently identified in glandular and surface epithelial cells, vascular endothelial cells, decidual cells of the endometrium and ovarian follicles during oogenesis. These findings suggest a potential association between Epo and pathological processes related to endometriosis (43,45).

Epo expression in normal, unaltered endometrial tissue is regulated by oestrogen and progesterone. In neoplastic tissue, by contrast, this expression appears to be mediated independently of steroid hormone action and is instead associated with tumour-related hypoxia. This mechanism may contribute to increased tumour aggressiveness and disease progression (46).

During pregnancy, Epo produced in the maternal body does not cross the placental barrier, and Epo is therefore synthesised not only in hepatic cells but also in trophoblast cells of the human placenta. EPOR expression in trophoblasts may be related to mechanisms that support foetal growth, regulate proliferation and protect against hypoxic stress. In pregnancy complications such as gestational hypertension or restricted foetal growth, elevated Epo levels have been detected in the plasma and amniotic fluid of patients. When normal and complicated pregnancies were compared, EPOR expression was higher in villous trophoblasts from abortive tissue, confirming increased Epo synthesis in response to hypoxic conditions (47).

Male reproductive system

The existence of a functional Epo/EPOR system in male reproductive organs has been demonstrated in recent years by several studies. EPOR has been identified in various tissues and cell types within the testes, including Sertoli cells and germ cells, as well as in other sites such as the epididymis and the membranes of mature spermatozoa (48,49). The hormone itself has also been detected in human seminal plasma, where its protective role and active growth factor function have been demonstrated (50).

The wide distribution of Epo may directly influence key reproductive processes, particularly spermatogenesis and sperm physiology. The principal paradigm of its action involves its anti-apoptotic, antioxidant and trophic properties, which are reflected in its effects on sperm vitality, motility and fertilisation capacity. Moreover, Epo stimulates testosterone production by Leydig cells in the testes. The overall beneficial effect of Epo on male reproductive cells may be exploited to improve outcomes in assisted reproductive technologies (49).

The antioxidant and anti-apoptotic effects of Epo on rat sperm following exposure to the toxic effects of doxorubicin were studied by Akman et al (51). Following administration of darbepoetin, an Epo analogue, a marked reduction in oxidative stress markers in testicular tissue was observed, accompanied by partial improvement in sperm parameters and a decrease in the number of abnormal male germ cells. The importance of Epo as a protective and regulatory factor essential for testicular cell survival and sperm integrity was confirmed. The combination of Epo and ramipril, an angiotensin-converting enzyme inhibitor, provided greater protection against doxorubicin-induced gonadotoxicity than darbepoetin alone, as ramipril also exerted antioxidant activity (51).

Regarding the therapeutic potential of this growth factor, studies have focused on its positive effects on sperm vitality and motility. These beneficial effects of Epo have been confirmed using in vitro animal models (49,50,52,53). Lower levels of Epo, along with other growth factors, in seminal plasma have been associated with male infertility. Therefore, Epo appears to be a promising agent for treating hypogonadism, promoting spermatogenesis and improving sperm preservation techniques, making it an attractive target for the treatment of male reproductive disorders (52). As EPORs are expressed on the plasma membrane of spermatozoa, the protective function of Epo is of fundamental importance after ejaculation. This may be crucial for maintaining the viability of these germ cells in the female reproductive system and supporting fertilisation (53).

Overall, the reproductive Epo system exemplifies a paradigm shift in the understanding of Epo biology, from a hormone whose function was traditionally considered limited to oxygen homeostasis to a pleiotropic regulator closely integrated into reproductive physiology, angiogenesis and disease.

Non-haematopoietic effects of Epo

The discovery of EPOR in non-haematopoietic tissues, including the central nervous system (brain and spinal cord), the cardiovascular system (myocardium) and the skin, reveals a much broader biological role for this cytokine than the regulation of erythrocyte production alone (Fig. 3) (54). Among its several pleiotropic effects across various organ systems, Epo plays a crucial role in the nervous system by supporting the development, protection, repair and maintenance of neural functions (55). Epo also exerts cardioprotective and renoprotective effects in ischaemia and various forms of tissue damage (56). Moreover, Epo has been shown to contribute to neurogenesis, angiogenesis, immune response regulation and metabolic homeostasis, including glucose regulation and adipose tissue inflammation (57).

Figure 3. Epo/EPOR expression in various tissues throughout the body. In the bone marrow, Epo drives the formation of red blood cells, which is its primary role. Beyond this, Epo contributes to neuroprotection in the brain, supports cardioprotection and vascular regulation in the heart and blood vessels, aids repair and maintenance in the ocular system, participates in wound healing, and shapes metabolic activity in white adipose tissue, including sex-specific control of fat mass and inflammation. Epo/EPOR signalling also has distinctive functions in the female and male reproductive system. Together, these sites illustrate the broad protective and metabolic influence of Epo beyond haematopoiesis. Created in https://BioRender.com. Epo, erythropoietin; EPOR, Epo receptor.

A substantial body of research has demonstrated the anti-apoptotic, regenerative and tissue-protective actions of Epo. The discovery of these properties has generated considerable interest in developing Epo and its derivatives as therapeutic molecules for neurological and cardiovascular diseases (19,54). However, clinical application remains problematic, as extensive clinical trials in conditions such as coronary reperfusion, heart failure, stroke and acute kidney injury have often failed to demonstrate a clear therapeutic benefit (20). These functions of Epo in non-haematopoietic tissues may be induced by circulating Epo produced by the kidneys or endocrine tissues, as well as by locally produced Epo in specific non-haematopoietic organs (58).

Nervous system

Epo serves as a multifunctional neuroprotective factor in the nervous system, promoting neuronal development, survival and repair through receptor-mediated anti-apoptotic and anti-inflammatory mechanisms (59).

In the nervous system, Epo is produced primarily by astrocytes, whereas EPOR is expressed across the neuronal surface. Stimulated EPOR initiates several signalling pathways, including JAK/STAT, PI3K/Akt and MAPK, which elicit various physiological processes. These include neurogenesis, astrocyte formation and neurite outgrowth, as well as the prevention of oxidative stress, excitotoxicity, apoptosis and inflammation (60).

In addition to endogenous, locally synthesised Epo, some peripherally administered Epo variants can cross the blood-brain barrier and protect nervous tissue. The principal mechanisms include stimulation of neurogenesis, activation of neurotrophic, anti-apoptotic and anti-inflammatory effects in the brain, and promotion of neuronal differentiation. Preclinical evidence suggests that these properties may be exploitable in the treatment of certain neuropsychiatric disorders, although clinical validation remains limited (61). Epo provides neuroprotection in various neurological impairments, including neurodegenerative diseases such as Parkinson's and Alzheimer's disease, ischaemic injury, ethanol-induced neurotoxicity and N-methyl-D-aspartate-mediated excitotoxicity (62).

Furthermore, rodent models suggest that Epo may support cognitive function and memory retrieval, including in models of schizophrenia and mood disorders, although these findings have not been consistently replicated in clinical settings (58). These neurocognitive improvements may be partly attributed to Epo-induced increases in hippocampal pyramidal neurons and oligodendrocytes. This effect is sustained only under ongoing functional demand and may also be related to Epo-associated enhancement of general motivation (63).

In the pathological processes underlying peripheral neuropathies, Epo acts as a neuronal survival factor, as evidenced by its protective effects against axonal degradation. Preclinical studies suggest that the neuroprotective properties of Epo may have therapeutic potential for peripheral nerve injuries, chemotherapy-induced polyneuropathies caused by cisplatin or paclitaxel and diabetic neuropathy, although robust clinical evidence is currently lacking (19).

Ocular system

Within the human eye, Epo and its receptor are strongly expressed in the synaptic terminals and inner segments of photoreceptor cells. They are also present throughout the inner nuclear and ganglion cell layers, including neurons and glial cells, in the retinal pigment epithelium, and in the choroidal vasculature and stroma (64). The anti-apoptotic effects of Epo are primarily mediated by activation of EPOR or interleukin β-common receptors, which are present in retinal ganglion cells, ocular stem cells, retinal photoreceptors and other ocular structures. The anti-inflammatory, antioxidant and angiogenic effects of Epo are also important in ocular physiology and pathophysiology, and are mediated through activation of various signalling pathways (65).

Preclinical evidence shows that, in oxygen-induced retinopathy, Epo promotes dose-dependent pathological neovascularisation through upregulation of VEGF, eNOS and neuronal NOS (66). Another preclinical study suggested that, in blast-induced traumatic ocular injury, Epo has bimodal temporal effects. Early treatment within 3 days exacerbates cell death and oxidative stress due to iron accumulation, whereas delayed treatment protects against axonal degradation and remains effective when initiated up to 3 weeks after trauma (67). These findings challenge the assumption that Epo is uniformly protective and suggest that, in angiogenesis-driven pathologies, Epo may exacerbate disease. Conversely, preclinical data indicate that, in acute neural trauma, appropriately timed treatment or non-erythropoietic variants may offer therapeutic benefits (62,68), although this requires clinical confirmation.

Cardiovascular system

Epo acts as a vascular-protective cytokine; it modulates angiogenesis in female reproductive organs, during embryonic development and during wound healing. The capacity of Epo to promote endothelial cell migration and proliferation arises under hypoxic conditions, thereby protecting relevant tissues from anoxia. Increased neovascularisation may be crucial in conditions such as myocardial infarction or hypoxia-induced pulmonary hypertension, although whether this translates into prevention of chronic vascular insufficiency and meaningful recovery of blood flow in clinical settings remains to be established (58).

EPOR expression in cardiac tissue and isolated primary cardiomyocytes is associated with Epo-dependent activation of specific signal-transduction pathways, attenuation of ischaemia-induced cardiomyocyte apoptosis and improved restoration of left ventricular function in ex vivo-perfused myocardium (68). Certain experimental models have shown a marked reduction in infarct size and attenuation of left ventricular dysfunction, suggesting a cardioprotective role for Epo (69). Endogenous Epo signalling in the myocardium may trigger a protective response during acute ischaemic injury. In line with this interpretation, patients experiencing their first myocardial infarction show higher endogenous Epo levels, which correlates with reduced infarct extent, as determined by creatine kinase release (70).

Anaemic patients with chronic heart failure exhibit elevated plasma levels of Epo, which serves as an independent predictor of mortality and morbidity (71). In experimental settings, Epo treatment has been associated with reduced myocardial levels of pro-inflammatory cytokines, including IL-1, IL-6 and TNF-α, as well as TGF-β1, suggesting a potential role in preventing cardiac fibrosis (19).

Renal system

Renal cortical fibroblasts are the primary source of endogenous Epo, whereas EPOR is widely expressed in other renal structures, including proximal tubule and tubular epithelial cells, glomerular mesangial cells and fibroblasts. The principal pleiotropic function of Epo is the activation of anti-apoptotic pathways and the suppression of tubular cell death (72).

In ischaemia-reperfusion models (72,73), high-dose Epo markedly reduces tubular apoptosis, accelerates epithelial regeneration and decreases cast formation, resulting in marked acute cytoprotection. In preclinical models, Epo has also been shown to mobilise endothelial progenitor cells and promote angiogenesis, thereby contributing to vascular repair. In addition to its anti-apoptotic effects, Epo exerts angiogenic, regenerative, and anti-inflammatory effects by modulating fibroblast activity and myofibroblast differentiation. These effects may enhance tissue repair but could also contribute to chronic interstitial fibrosis if the stimulus is sustained (73).

Tissue regeneration and bone remodelling

EPOR is expressed in several cell types, including mesenchymal stem cells (MSCs), including dental and bone marrow MSCs, as well as skeletal muscle cells, enabling regulation of bone homeostasis. Preclinical data suggest that Epo-driven osteoblast differentiation from MSCs is dose- and age-dependent, with potential implications for fracture repair and acceleration of bone healing (1).

Preclinical investigations have shown that Epo modulates the physiological processes involved in wound repair (19). In rodent models, local Epo administration into subcutaneous fibrin chambers stimulated granulation tissue formation and angiogenesis, and increased inducible nitric oxide synthase (iNOS) expression, thereby markedly supporting the overall wound-healing process (74). In burn models, Epo treatment increased granulation tissue thickness, collagen deposition, and VEGF, iNOS, eNOS and nitric oxide levels. In addition, Epo was shown to accelerate bone repair in a murine femur-fracture model (75).

Adipose tissue and energy homeostasis

Epo acts as a systemic metabolic regulator that promotes lipolysis, improves glucose homeostasis and protects against diet-induced obesity (40). In murine models, mice lacking EPOR in adipose tissue exhibit excessive fat accumulation and glucose intolerance, whereas transgenic or pharmacological elevation of Epo enhances lipolysis and suppresses lipogenic genes in white adipose tissue, where EPOR is expressed, as well as in brown adipose tissue, skeletal muscle and liver tissue, resulting in a lean phenotype and improved insulin sensitivity. These metabolic actions are coordinated by the Epo-EPOR-runt-related transcription factor 1 (RUNX1) axis. Epo stabilises the transcription factor RUNX1, which activates lipolytic gene promoters and represses lipogenic promoters, while concurrently engaging peroxisome proliferator-activated receptor α, situin 1, AKT and STAT pathways to enhance mitochondrial biogenesis and fatty acid oxidation (76).

In preclinical studies, non-erythroid EPOR agonists have been shown to reduce body weight and fat mass while improving glucose tolerance without increasing haematocrit, suggesting a potential therapeutic strategy for separating metabolic from erythropoietic effects (77). Epo attenuates inflammation in white adipose tissue, but its regulation of body weight and fat mass is sex specific; its anti-obesity effect is masked in females through oestrogen-dependent pathways, with higher plasma Epo associated with weight loss in males but weight gain in females (57).

Hypothalamic regulation

Epo activity decreases adrenocorticotropic hormone (ACTH) secretion, as EPORs are localised in the hypothalamus, specifically in POMC neurons. POMC is an essential intermediate product in ACTH synthesis and is involved in the pathogenesis of metabolic syndrome and obesity (40,78). In animal studies, high-dose Epo treatment has been shown to increase hypothalamic POMC expression, reduce food intake and attenuate fat mass accumulation (78).

Pulmonary, hepatic and immune effects

Normal lung tissues, including bronchial epithelium and type II alveolar cells, express Epo and its tissue-protective receptor complex, EPOR/β-common receptor (βcR), thereby allowing direct pulmonary actions. In experimental models of acute lung injury, recombinant human Epo (rHu-Epo) markedly reduces interstitial and alveolar oedema and limits leukocyte infiltration, while modulating inflammation by suppressing TNF-α and monocyte chemoattractant protein-1, and upregulating the anti-inflammatory cytokine IL-10. Epo has also been shown to enhance antioxidant defences and vascular repair and to decrease epithelial apoptosis, suggesting a supportive role in lung tissue repair after diverse insults (42).

Epo markedly improves hepatic function in thioacetamide-induced injury by lowering serum alanine aminotransferase and aspartate aminotransferase activities while restoring albumin concentrations. In preclinical studies (57,79), Epo treatment dose-dependently reduced hepatic pro-fibrotic and inflammatory cytokines, as well as oxidative stress markers. Histopathological examination showed dose-related regression of fibrous tissue, restoration of normal hepatic architecture and a notable decrease in collagen deposition (79).

Regarding the immune system, Epo directly influences various immune cell populations. This cytokine modulates macrophage phenotype and infiltration within human tissues. In white adipose tissue, Epo reduces the infiltration of pro-inflammatory macrophages and promotes the development of anti-inflammatory subtypes (80). Conversely, in cancer models, tumour-derived Epo promotes efferocytosis, the phagocytic clearance of apoptotic cells, in tumour-associated macrophages (TAMs) and drives their differentiation towards an immunoregulatory Kupffer cell-like phenotype, thereby suppressing antitumour T-cell immunity (35). In vitro and animal studies suggest that Epo administration generally inhibits the proliferation and expansion of both CD4+ and CD8+ T cells while simultaneously facilitating the expansion of regulatory T cells. In animal models, Epo treatment reduces the severity of autoimmune conditions such as lupus nephritis and experimental autoimmune encephalomyelitis (81).

Clinical applications of Epo

The applications of recombinant human Epo are extensive, encompassing both its primary haematological role and its diverse pleiotropic effects in non-haematopoietic tissues (Fig. 4). Epo exerts protective effects across various organs, a property that has been effectively utilised in clinical practice. However, it has also been used off-label by elite athletes as a performance-enhancing agent (57).

Figure 4. Different applications of rHu-Epo. The administration of Epo or its derivatives as active substances in clinical indications has broadened their use, although many other potential uses are currently being investigated using laboratory animal models or in clinical studies. Created in https://BioRender.com. Epo, erythropoietin; rHu-Epo, recombinant human Epo; HIV, human immunodeficiency virus.

Anaemia

Clinical evidence indicates that Epo is most widely used to treat anaemia associated with chronic kidney disease (CKD); its use has also expanded to patients with cancer, with approximately one-third of those with chemotherapy-induced anaemia receiving treatment. Epo can be administered by both the subcutaneous and intravenous routes. rHuEpo is an antianaemic agent that stimulates erythropoiesis and increases haematocrit, thereby reducing the need for blood transfusions and the associated risks in patients with renal failure. rHuEpo is indicated for patients receiving regular haemodialysis as well as those with CKD who have not yet commenced dialysis (82). In oncology, Epo is primarily used to manage chemotherapy-induced anaemia, although it is associated with important safety considerations. Clinical trials have consistently demonstrated that Epo increases haemoglobin concentrations, reduces transfusion requirements and improves quality of life in patients with cancer (83).

Another indication for Epo in the management of anaemia is in patients with human immunodeficiency virus (HIV) infection. HIV-associated anaemia may arise from multiple causes, including the infection itself, antiretroviral therapy and opportunistic infections. In this setting, Epo provides an alternative to blood transfusion for treating anaemia (84).

Emerging and investigational clinical applications

Recent clinical investigations have expanded the use of Epo and its non-erythropoietic derivatives beyond the correction of anaemia. According to a 2025 review of Epo-derived therapeutics for acute ischaemic stroke, high-dose recombinant Epo has raised marked safety concerns. However, several modified Epo variants have demonstrated strong neuroprotective efficacy in preclinical studies and are currently being evaluated in early-phase clinical trials to minimise erythropoiesis-related adverse effects (85,86).

A non-haematopoietic Epo analogue, ARA 290, has been investigated as a protective agent in pancreatic islet transplantation. The study confirmed that ARA 290 protects pancreatic islets against apoptosis, inflammatory responses and cytokine-induced damage, thereby improving transplantation outcomes (87).

To explore potential benefits in the management of acute respiratory distress syndrome, erythropoiesis-stimulating agents (ESAs) have been administered as adjunctive therapy in critically ill patients with COVID-19. Anaemia has been observed in mechanically ventilated patients, partly due to reduced endogenous Epo concentrations. Based largely on preclinical evidence, the pleiotropic effects of ESAs have been hypothesised to protect lung integrity, restore homeostasis, improve respiratory function, and potentially exert anti-inflammatory and antioxidant effects (88). Several clinical studies and expert reviews have reported clinical improvement and favourable outcomes following ESA administration in critically ill patients with COVID-19, particularly with respect to respiratory function, anaemia and systemic inflammation (88–91). By increasing haemoglobin concentrations, ESAs enhance oxygen-carrying capacity, potentially improving tissue oxygenation and reducing transfusion requirements. However, a multicentre retrospective cohort study reported that ESA administration was associated with prolonged Intensive Care Unit and overall hospital stays, possibly owing to a higher incidence of liver injury in patients with hypoalbuminaemia and increased rates of nosocomial infection during hospitalisation. ESA use was also associated with fewer ventilator-free days, which was one of the study outcome measures (88).

Epo may also have prognostic value as a biomarker in sepsis. Elevated Epo concentrations have been associated with a higher incidence of septic shock and independently linked to increased in-hospital and 28-day mortality rates. These elevated levels may reflect a compensatory response to hypoxia and anaemia, both of which are common in severe infections. During sepsis, systemic inflammation and impaired tissue perfusion result in widespread hypoxia, which stimulates Epo production as a physiological response. These findings suggest that elevated Epo levels in sepsis may indicate disease severity rather than a protective physiological response (92).

Safety concerns and limitations of Epo therapy

Despite its well-established clinical benefits in the management of anaemia, ESA use carries substantial adverse effects and risks related to mortality and patient survival. Strong evidence suggests that ESAs increase mortality during the active study period and worsen overall survival in patients with cancer, particularly those with head and neck, lymphoid, and non-small cell lung malignancies (93,94). Other limitations include an increased risk of thromboembolic complications, induced hypertension, cardiac failure and arrhythmias, particularly with aggressive dosing (95). An additional critical concern is that the tissue-protective effects of Epo may potentially stimulate cancer cell proliferation, which could offset the benefits of treating anaemia (96). Several clinical studies have shown that Epo administration is associated with an increased risk of cancer progression and mortality, depending on cancer type and stage, as EPOR expression has been documented in various malignancies, including breast, ovarian, cervical, endometrial, head and neck, non-small cell lung, gastric, hepatocellular, renal cell, and glioblastoma tumours (12).

As a result of these safety concerns, current recommendations restrict ESA use to specific indications during palliative chemotherapy and require careful assessment of the risk-benefit ratio (97). In certain contexts, such as head and neck tumours treated with radiation therapy, the addition of Epo has been associated with significantly poorer overall survival (98). Non-erythropoietic Epo derivatives are being investigated for their potential to provide protective benefits without stimulating erythropoiesis (57).

A rare but serious complication of long-term ESA therapy is pure red cell aplasia (PRCA), an immune-mediated condition in which neutralising antibodies against recombinant Epo cross-react with endogenous Epo, block the interaction between Epo and its receptor, and result in severe transfusion-dependent anaemia. ESA-associated PRCA was first reported in 1998, with most cases occurring in patients receiving epoetin alfa subcutaneously. Although its incidence has declined markedly following improvements in ESA storage, handling and administration methods, PRCA remains a recognised complication, including with newer third-generation ESA formulations, and requires immediate discontinuation of ESA therapy after diagnosis, followed by immunosuppressive treatment (99).

Haemoglobin overshoot, defined as haemoglobin levels exceeding the recommended target of 12–13 g/dl, substantially increases cardiovascular risk and represents a critical dosing consideration in both renal and oncological settings. Current guidelines therefore recommend strict haemoglobin monitoring throughout ESA therapy to avoid dose-dependent adverse effects (100).

Beyond these established safety signals, accumulating evidence has introduced a further oncological concern. Tumour-derived Epo may promote an immunosuppressive tumour microenvironment (TME) by acting on TAMs, potentially undermining antitumour immunity and reducing responsiveness to checkpoint inhibitor immunotherapy (35,100). This is particularly concerning given the increasing use of immune checkpoint inhibitors in oncology, as co-administration of ESAs with immunotherapy may paradoxically compromise the efficacy of the latter. Retrospective analyses investigating whether ESA intake adversely affects immunotherapy outcomes are therefore urgently needed (100).

Next-generation therapeutics

Accumulating safety concerns surrounding conventional ESAs, particularly the risks of thromboembolism, tumour progression and the more recently described immunosuppressive remodelling of the TME, have driven the development of next-generation therapeutics intended to preserve the clinical benefits of Epo signalling while avoiding its liabilities (100). Current translational strategies fall into three main modalities: i) Orally bioavailable HIF prolyl hydroxylase inhibitors (HIF-PHIs); ii) non-erythropoietic Epo derivatives that retain tissue-protective activity; and iii) variant-selective biologicals targeting tumour-specific Epo isoforms without affecting canonical Epo function (100,101).

HIF-PHIs represent the most clinically advanced alternative to recombinant Epo for the management of CKD-associated anaemia. By inhibiting prolyl hydroxylase domain enzymes, these agents stabilise HIF-α subunits and thereby induce endogenous Epo transcription, along with upregulation of iron transporters and suppression of hepcidin, thereby restoring erythropoiesis even under inflammatory conditions in which conventional ESAs frequently lose efficacy (102). The class currently comprises six agents, namely, roxadustat, daprodustat, vadadustat, molidustat, enarodustat and desidustat, and a recent network meta-analysis of 45 randomised trials including >32,000 patients clarified that these agents are not interchangeable. Roxadustat and daprodustat emerged as the most effective haemoglobin-raising agents overall, with roxadustat performing best in non-dialysis-dependent populations and daprodustat showing additional advantages in iron mobilisation, particularly in dialysis-dependent patients (102). The regulatory landscape reflects ongoing uncertainty regarding long-term safety. Agent-level safety patterns, including vascular occlusion with roxadustat, gastrointestinal events with daprodustat, residual cardiovascular concerns with vadadustat and a uniquely reduced risk of hyperkalaemia with molidustat, have shaped these divergent decisions and strongly argue against treating HIF-PHIs as a homogeneous class (102,103).

Of relevance to the oncological focus, HIF-PHIs raise a safety question that conventional ESAs do not. As HIF-a stabilisation activates the entire hypoxia transcriptional programme, including angiogenesis, metabolic reprogramming, apoptosis resistance and immune evasion, sustained pathway activation could theoretically accelerate tumour progression beyond effects attributable to elevated Epo alone (104). Clinical studies (104,105) have shown that roxadustat achieves efficacy comparable to conventional ESAs in chemotherapy-induced anaemia, but treatment-emergent serious adverse events, including tumour progression, have been reported, and observation periods to date are insufficient to establish or exclude causality (104).

A second strategy seeks to dissociate the tissue-protective effects of Epo from its erythropoietic activity by structurally modifying the molecule itself. Carbamylated Epo, asialo-Epo and the small peptide ARA290 (cibinetide) preferentially bind to the heterodimeric tissue-protective receptor formed by EPOR and the β-common subunit (CD131), rather than to the homodimeric EPOR responsible for erythropoiesis (105). These derivatives retain anti-apoptotic, anti-inflammatory and cytoprotective effects across multiple tissues without raising haemoglobin or increasing thromboembolic risk (101).

A third, conceptually distinct approach targets tumour-specific Epo splice variants rather than canonical Epo itself. Navone et al (106) recently demonstrated that glioblastoma cells, particularly glioblastoma stem-like cells, markedly overexpress an Epo splice variant known as EV-3, which lacks exon 3 and therefore differs structurally from canonical Epo. EV-3 exerts potent tumorigenic, angiogenic and immunomodulatory effects in patient-derived glioblastoma cells, promoting proliferation, invasion, stemness and immune cell exclusion through autocrine and paracrine mechanisms. EV-3 levels are specifically enriched in hypoxic tumour regions co-localising with HIF1-α expression, indicating hypoxia-driven amplification as a key regulatory mechanism, consistent with the broader role of HIF1-α in Epo gene regulation. A novel monoclonal antibody, AND-C4, developed with high affinity specifically for EV-3, antagonises all these pro-tumorigenic effects in vitro and markedly reduces tumour growth while improving survival in orthotopic glioblastoma mouse models, without anti-erythropoietic effects or toxicity in normal brain cells (106). This variant-selective paradigm differs fundamentally from receptor-selective Epo derivatives. Rather than modifying Epo to engage a tissue-protective receptor, it neutralises a tumour-specific isoform while leaving the entire physiological Epo/EPOR axis intact (106).

Performance enhancement and anti-doping considerations

RHu-Epo improves endurance performance beyond what is explained by increased oxygen-carrying capacity alone. Non-haematopoietic mechanisms, including enhanced mitochondrial enzyme activity, increased fat oxidation and reduced lactate production, are likely to contribute. Administration of this recombinant form to athletes induces both haematological and tissue-level adaptations, with the latter playing a crucial role in markedly increasing time to exhaustion (107).

Growth factors, peptide hormones, mimetics and related substances, including Epo and its derivatives, are prohibited both in and out of competition by the World Anti-Doping Agency. Products that affect blood parameters, such as haematocrit levels, erythrocyte mass and oxygen uptake, are mainly used to enhance endurance in sports such as cycling, long-distance running, triathlon and skiing. The products improve aerobic activity by increasing haematocrit levels, maximal oxygen uptake and blood viscosity, while decreasing ferritin levels (108). Epo and its recombinant forms are often administered at supraphysiological doses, which can lead to adverse cardiovascular effects, including arrhythmias, hypertension, thromboembolic events, acute myocardial infarction and oedema (109). Athletes currently often use microdoses of ESAs, as these dosing strategies are more difficult to detect. Detection generally relies on ELISA-based assays and liquid chromatography-mass spectrometry methods (108), as well as isoelectric focusing and immunoblotting (2).

Roles of Epo in cancer

The roles of Epo and EPOR have been described in various solid malignancies, with increased expression often observed in hypoxic regions of the tumour (30,110,111). However, the interpretation of these observations remains controversial. Although several studies have demonstrated EPOR expression on tumour or endothelial cells, the biological relevance and functionality of this expression have not been clearly confirmed (112,113). Major limitations of several earlier studies include methodological problems in EPOR detection, particularly the limited specificity of some antibodies used in immunohistochemical analyses, the low abundance of the receptor outside erythroid cells, and the unclear relationship among mRNA expression, protein presence and functional signalling (114,115). A pan-cancer bioinformatics analysis of EPOR expression across 33 tumour types confirmed this heterogeneous prognostic impact, identifying EPOR as a risk factor in mesothelioma, lung squamous cell carcinoma and prostate cancer, while associating it with a better prognosis in lung adenocarcinoma and pancreatic cancer (116,117). The clinical and translational evidence linking Epo/EPOR signalling with cancer progression and patient outcomes across major tumour types is summarised in Table II (118–134).

Table II. Clinical and translational evidence linking Epo/EPOR signalling with cancer progression and patient outcomes.

Table II.

Clinical and translational evidence linking Epo/EPOR signalling with cancer progression and patient outcomes.

First author, year	Cancer type	Study design	Epo/EPOR-related finding	Molecular mechanisms	Clinical outcome	(Refs.)
Henke et al, 2003	HNSCC	Phase III randomized trial (ENHANCE trial)	Epoetin β during radiotherapy in EPOR-positive tumours	EPOR activation, JAK2/STAT5, hypoxia-related signalling	Worse locoregional progression-free survival and overall survival	(118)
Overgaard et al, 2018	HNSCC	Randomized clinical trial analysis (DAHANCA-10)	EPOR expression in tumours treated with darbepoetin alfa	HIF-1α signalling, angiogenesis	Reduced locoregional control and poorer prognosis	(119)
Winter et al, 2005	HNSCC	Tumour expression/ prognostic study	High intratumoral Epo and EPOR expression	HIF-1α/VEGF, PI3K/AKT	Reduced overall survival and relapse-free survival	(120)
Seibold et al, 2013	HNSCC	Retrospective prognostic cohort study (n=144)	Tumour Epo and EPOR expression evaluated in locally advanced HNSCC	Epo/EPOR signalling, hypoxia-associated pathways	Poorer locoregional control, metastasis- free survival and overall survival	(121)
Leyland-Jones et al, 2005	Metastatic breast cancer	Phase III randomized trial (BEST trial)	Epoetin alfa targeting normal haemoglobin levels	Anti-apoptotic signalling, EPOR activation	Increased mortality, disease progression and thromboembolic events	(122)
Larsson et al, 2009	Breast cancer	Clinical prognostic study	High EPOR expression in tumour tissue	JAK2/STAT5, PI3K/AKT, MAPK/ERK	Reduced benefit from tamoxifen therapy and worse recurrence outcomes	(123)
Pirker et al, 2008	NSCLC	Randomized clinical trial	ESA treatment for cancer-related anaemia	PI3K/AKT/mTOR, MAPK/ERK	Improved anaemia control without consistent survival benefit	(124)
Tsuboi et al, 2009	NSCLC	Phase III randomized study	Darbepoetin alfa during chemotherapy	Hypoxia-related signalling pathways	Improved haematologic parameters; no significant survival improvement	(125)
Tong et al, 2024	Lung cancer	Meta-analysis of RCTs	ESA use in lung cancer patients	Hypoxia-associated signalling pathways	Improved anaemia management; thromboembolic risk remained clinically relevant	(126)
Ma and Luo, 2025	Lung cancer	Systematic review and meta-analysis	ESA efficacy and safety in lung cancer	ESA-related survival pathways	Benefit for anaemia correction; safety concerns persist	(127)
Wilkinson et al, 2006	Ovarian cancer	Randomized clinical study	Epoetin alfa during platinum-based chemotherapy	VEGF-associated angiogenic signalling	Improved haemoglobin levels and quality of life; no survival benefit	(128)
Acs et al, 2003	Cervical cancer	Observational prognostic studies	EPOR expression in hypoxic tumours	HIF-1α/VEGF, JAK2 signalling	Radioresistance and poorer clinical outcomes	(110)
Wang et al, 2011	Gastric adenocarcinoma	Prognostic cohort study (n=172)	Increased Epo and EPOR expression in tumours	HIF-1α/VEGF, angiogenesis, microvessel density	Associated with lymph-node metastasis, advanced stage and reduced overall survival; Epo identified as an independent prognostic factor	(34)
Lin et al, 2012	Oral squamous cell carcinoma	Prognostic cohort study (n=256)	High EPOR expression in tumour tissue	Hypoxia-related signalling, EPOR-mediated survival pathways	Associated with advanced TNM stage, nodal metastasis, and poorer 5-year overall and disease-specific survival	(129)
Bennett et al, 2008	Multiple solid tumours	Meta-analysis (JAMA)	ESA exposure during anticancer treatment	Anti-apoptotic and proliferative signalling pathways	Increased mortality signal across pooled trials	(130)
Bohlius et al, 2009; Bohlius et al, 2019	Multiple solid tumours	Individual patient data meta-analysis (53 RCTs; n≈13,900)	ESA therapy	Multiple Epo/EPOR survival pathways	Increased on-study mortality and slightly reduced overall survival	(94,131)
Tonelli et al, 2004	Multiple solid tumours	Systematic review/meta-analysis	ESA treatment	Tumour progression pathways evaluated	No significant increase in objective tumour progression	(132)
Glaspy et al, 2004	Multiple solid tumours	Study-level meta-analysis	ESA use during anticancer therapy	Disease-progression signalling assessed	No significant effect on disease progression	(133)
Adams et al, 2022	Multiple malignancies	Systematic review and network meta-analysis	ESA ± iron supplementation	Erythropoietic signalling and treatment-response pathways	Improved haematological response and reduced transfusion requirements; no survival benefit demonstrated	(134)
Zhang et al, 2022	Pan-cancer (33 tumour types)	TCGA-based prognostic analysis	EPOR expression across cancers	JAK/STAT signalling, immune regulation, tumour microenvironment pathways	EPOR associated with survival, immune infiltration and prognosis across several malignancies	(116)

[i] Epo, erythropoietin; EPOR, Epo receptor; HNSCC, head and neck squamous cell carcinoma; ESA, erythropoiesis-stimulating agent; TCGA, The Cancer Genome Atlas.

These limitations are particularly relevant given the marked heterogeneity of solid tumours and the complexity of the TME, both of which can markedly affect the interpretation of biomarker studies (135). The potential role of Epo/EPOR signalling in carcinogenesis therefore remains debated. Experimental studies suggest that in some EPOR-expressing tumour cell lines, including neuroblastoma, Ewing's sarcoma and colorectal carcinoma, Epo may promote cell survival by upregulating anti-apoptotic pathways (136,137). However, it is not always clear whether this reflects a direct consequence of functional EPOR signalling in tumour cells or an indirect effect mediated by the TME (32).

It is important to distinguish between the direct effects of Epo on tumour cells and indirect mechanisms involving angiogenesis, hypoxic responses or immunomodulation (32,138). Current understanding of solid tumour biology emphasises that the TME, including hypoxia, inflammatory pathways and vascular remodelling, can markedly modulate tumour cell responses to growth factors and cytokines (139,140). Epo is an established proangiogenic factor that stimulates the proliferation of endothelial and progenitor cells via the JAK2/STAT5 and ERK1/2 pathways, which may, in turn, promote tumour perfusion and growth (141,142). Therefore, some antitumour or protumour effects attributed to Epo may not necessarily reflect autonomous signalling within tumour cells themselves (138).

Notably, Epo has been shown to accelerate the growth of tumour xenografts lacking EPOR expression by enhancing angiogenesis in vivo, demonstrating that protumorigenic effects can occur entirely through indirect vascular mechanisms without direct receptor-mediated action on tumour cells (100). This distinction between direct and indirect mechanisms is clinically important, as it implies that even tumours with low or absent EPOR expression may be adversely affected by Epo administration through its angiogenic properties alone.

Further ambiguity arises from conflicting findings across tumour types and clinical studies. For example, in melanoma and glioma, a functional Epo/EPOR loop associated with angiogenesis and tumour progression has been described (143,144). Lai et al (145) observed an association between increased Epo expression and higher morbidity, mortality and cervical lymph node metastases in head and neck squamous cell carcinoma, suggesting a possible contribution of Epo/EPOR signalling to invasiveness and metastatic spread. Conversely, other studies in laryngeal squamous cell carcinoma confirmed the presence of Epo and EPOR in tumour tissue but did not demonstrate an association with disease progression or response to oncological treatment (146). These discrepancies indicate that the mere presence of EPOR may not be sufficient evidence of biologically relevant signalling (112).

A further dimension of the oncological significance of Epo has emerged from the recent discovery that alternatively spliced Epo variants may exert independent oncogenic effects in specific malignancies. In glioblastoma, the splice variant EV-3 has been identified as a tumour-derived driver of proliferation, angiogenesis and immune evasion, acting independently of the classical Epo/EPOR axis (106). This finding broadens the oncological relevance of Epo biology beyond canonical Epo signalling and identifies tumour-specific isoforms as potentially actionable therapeutic targets, as discussed previously.

Taken together, the oncological implications of Epo/EPOR signalling extend well beyond the correction of cancer-related anaemia. Accumulating evidence from pan-cancer expression analyses, genetic epidemiological studies, mechanistic preclinical models, newly characterised splice variant biology and the recently described immunosuppressive TME mechanism collectively reframes Epo as a multifaceted contributor to tumour biology (100). The functional consequences of Epo/EPOR signalling in cancer are highly context-dependent and vary substantially according to tumour type, signalling architecture and TME conditions. Malignancies of the female reproductive system, including ovarian cancer, are of particular concern due to their frequent intrinsic or acquired resistance to treatment. Emerging evidence suggests that Epo signalling may contribute to the mechanisms underlying chemoresistance, highlighting the need for careful evaluation of therapeutic strategies involving Epo derivatives in this patient population. The ongoing development of non-erythropoietic Epo derivatives and variant-targeting approaches such as AND-C4 offers a promising strategy for separating the beneficial tissue-protective and anaemia-correcting effects of Epo from its potentially harmful oncological consequences (106). Nevertheless, further clinical and mechanistic studies are needed to fully elucidate the interplay between Epo signalling, tumour biology, and the efficacy of both chemotherapy and immunotherapy across diverse cancer types.

Conclusions

Epo is widely recognised as one of the primary regulators of erythrocyte production within the haematopoietic system (19). As research in this field has advanced, multiple studies have shown that the potential of this molecule extends beyond erythrocyte production. The expression of Epo or EPOR in numerous tissues has been shown to mediate the non-haematopoietic functions of Epo, affecting several biological processes in the body.

Despite compelling evidence across multiple organ systems, translation of these tissue-protective effects into clinical practice has so far been limited, and several randomized clinical trials have failed to demonstrate consistent therapeutic benefit. The tissue-protective properties of Epo have shown considerable promise in experimental models; however, translation into routine clinical practice has been limited. Current indications for Epo, including recombinant forms and derivatives, mainly include anaemia associated with CKD or chemotherapy, in which its haematopoietic properties are exploited.

In the context of potential future applications as a tissue-protective agent, derivatives lacking haematopoietic properties would be preferred. The primary rationale is to avoid the adverse effects of mildly induced erythropoiesis, which increases blood viscosity and may lead to thromboembolic complications or adversely affect the course and treatment of malignant disease. Based on current preclinical and early clinical data, the most promising future applications are expected to involve the treatment of ischaemically damaged tissues, including tissues affected by stroke (85,86), myocardial infarction (26,70) or ischaemic kidney disease (72,73). Other prospective applications include use in wound healing, tissue remodelling and healing processes within the musculoskeletal system (74,75). Owing to their strong neuroprotective effects, Epo derivatives could represent a further step in the treatment of various neurodegenerative and cognitive disorders, as well as ocular degenerative conditions and mental illnesses.

Translation of these findings into clinical practice will require appropriate patient selection, accurate dosing and mitigation of adverse effects, particularly those that could negatively affect cancer progression or compromise therapeutic efficacy. Malignancies of the female reproductive tract frequently exhibit intrinsic or acquired treatment resistance. Emerging evidence suggests that Epo signalling may contribute to mechanisms underlying chemoresistance, highlighting the need for careful evaluation of therapeutic strategies.

Beyond pharmacological applications, targeted modulation of the Epo/EPOR signalling pathway through structural modification of the Epo molecule or via targeted delivery systems could enable tissue-specific activity while minimising haematopoietic side effects. Nevertheless, further studies are required to validate these approaches and to elucidate the interplay between Epo signalling and tumour biology, particularly its impact on chemotherapy efficacy in tumours of the female reproductive system and beyond.

Acknowledgements

Not applicable.

Funding

The present review was supported by the Scientific Grant Agency of the Ministry of Education of the Slovak Republic (contract nos. VEGA 1/0500/23 and KEGA 024UPJŠ-4/2024).

Availability of data and materials

Not applicable.

Authors' contributions

KD, ZS and PS were responsible for preparing the study design and manuscript preparation. KD, PH and PS were responsible for data collection, and KD, PH, MŠ and PS were responsible for data interpretation. KD and PS were responsible for literature searching, reviewing and editing. MŠ and PS were responsible for funding acquisition and supervision. Data authentication is not applicable. All authors read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References