Age‐associated myeloid malignancies – the role of STAT3 and STAT5 in myelodysplastic syndrome and acute myeloid leukemia

Abstract

In the last few decades, the increasing human life expectancy has led to the inflation of the elderly population and consequently the escalation of age‐related disorders. Biological aging has been associated with the accumulation of somatic mutations in the Hematopoietic Stem Cell (HSC) compartment, providing a fitness advantage to the HSCs leading to clonal hematopoiesis, that includes non‐malignant and malignant conditions (i.e. Clonal Hematopoiesis of Indeterminate Potential, Myelodysplastic Syndrome and Acute Myeloid Leukemia). The Janus Kinase‐Signal Transducer and Activator of Transcription (JAK–STAT) pathway is a key player in both normal and malignant hematopoiesis. STATs, particularly STAT3 and STAT5, are greatly implicated in normal hematopoiesis, immunity, inflammation, leukemia, and aging. Here, the pleiotropic functions of JAK–STAT pathway in age‐associated hematopoietic defects and of STAT3 and STAT5 in normal hematopoiesis, leukemia, and inflammaging are reviewed. Even though great progress has been made in deciphering the role of STATs, further research is required to provide a deeper understanding of the molecular mechanisms of leukemogenesis, as well as novel biomarkers and therapeutic targets for improved management of age‐related disorders.

The increase in the aging population has resulted in the rise of age‐related disorders, including MDS and AML. The JAK–STAT pathway is essential for several cellular functions in the HSPCs. Deregulation of the STAT3/5 signaling may occur at several pathway stages (indicated by exclamation points) and has been associated with MDS and AML development through various mechanisms. Thus, STAT3/5 factors constitute ideal targets for developing improved therapeutic management strategies.

Abbreviations

ALL, acute lymphocytic leukemia

AML, acute myeloid leukemia

CCD, coiled‐coil domain

CHIP, clonal hematopoiesis of indeterminate potential

CLL, chronic lymphocytic leukemia

CML, chronic myeloid leukemia

CMML, chronic myelomonocytic leukemia

DBD, DNA‐binding domain

DNA, deoxyribonucleic acid

ELN, European leukemia net

FDA, US Food and Drug Administration

HSC, hematopoietic stem cell

HSPCs, hematopoietic stem and progenitor cells

ICC, International Consensus Classification

IFN‐γ, interferon γ

IgE, immunoglobulin E

IgK, immunoglobulin kappa

IL2, interleukin 2

IL6, interleukin 6

Il6ra, interleukin 6 receptor alpha

IL7, interleukin 7

ITD, internal tandem duplication

JAK–STAT, Janus Kinase/signal transducer and activator of transcription

KSL, c‐Kit⁺ Sca‐1⁺ lineage⁻

LD, linker domain

LSC, leukemic stem cell

MDS, myelodysplastic syndrome

miRNA, microRNA

mRNA, messenger RNA

NGS, next‐generation sequencing

NIH, National Institutes of Health

NTD, N‐terminal domain

PIAS, protein inhibitors of activated

PROTAC, proteolysis‐targeting chimera

pSTAT, phosphorylated STAT

PTPs, protein tyrosine phosphatases

R/R, relapsed/refractory

RNA, ribonucleic acid

SH2, Src‐homology 2

SOCS, suppressor of cytokine signaling

TAD, carboxy‐terminal transactivation domain

TCA, tricarboxylic acid cycle

Teff, effector T cell

TFH, T follicular helper cell

TH, T helper cell

sTNF‐RII, soluble tumor necrosis factor receptor II

TPO, thrombopoietin

Treg, regulatory T cell

TYK, tyrosine kinase

UK, United Kingdom

US, United States

uSTAT, unphosphorylated STAT

VISTA, V‐domain Ig suppressor of T‐cell activation

WHO, World Health Organization

y.o., years old

Aging: current situation and why it matters

Significant scientific and technological developments have improved health and life conditions, increasing human lifespan. Median age continues to increase, yet there is a debate in the scientific community regarding a potential upper limit of 125 years in the life expectancy of humankind [1, 2, 3]. According to WHO, by 2050 the population over 60 years old (y.o.) will double worldwide, reaching 2.1 billion [4]. In Europe, the aging of the population has been an observed demographic trend in the last few decades, with the older population (considered 65 y.o. and over), accounting for 21.3% in 2023, expected to reach 32.5% by 2100 [5]. Although we are going through the WHO's Decade of Healthy Aging aiming to support age‐friendly healthcare and communities, and to promote aging‐associated research [6], several challenges linked to age‐related diseases and person‐oriented care remain to be addressed.

Aging has been connected to or considered a risk factor for several ailments, including neurodegenerative [7], cardiovascular [8], and several hematological disorders both malignant, such as myelodysplastic syndromes, leukemias, myelomas, lymphomas, and non‐malignant, such as anemia [9, 10, 11]. Hematologic cancers are considered age‐related diseases and are characterized by increased mortality rate in the elderly population. Additionally, they account for one of the most serious clinical and scientific challenges that despite the immense scientific progress, remain to be decoded.

Namely, leukemia is a malignant progressive disease defined by the overproduction of abnormal hematopoietic cells. Ιt is contemplated as an age‐related disorder as it is more prevalent in older patients, with a median age of diagnosis at 67 y.o. According to NIH data, 59 610 new cases of leukemia and 23 710 deaths have been estimated in 2023, while 490 875 patients were living with leukemia in the US in 2020 [12].

Leukemia has been classified regarding severity, as acute or chronic and based on the cellular origin, as myelogenous or lymphocytic. Hence, there are four broad leukemia categories: acute lymphocytic leukemia (ALL), acute myelogenous leukemia (AML), chronic lymphocytic leukemia (CLL), and chronic myelogenous leukemia (CML). AML remains the most common acute leukemia, accounting for approximately one‐third of the estimated leukemia cases in 2023 worldwide, with only a 31.7% 5‐year relative survival rate [13]. Additionally, a significant increase in myeloid leukemia incidents has been observed in the US since 2011, with AML deaths rising continuously in the past 20 years [14].

Clinical management of leukemia represents a major challenge for the European Healthcare System, considering that despite the extensive advancements, leukemia cases continue to increase alongside the aging population. Moreover, the economic burden of leukemia treatment remains substantial, with AML treatment costs amounting to a mean of $25 243 per patient per month over the follow‐up period and incremental cost‐effectiveness ratio (ratio of the cost difference to the difference in effectiveness) ranging from $50 000 to $150 000 per patient in the US and $20 000 to $30 000 in the UK [15, 16]. Therefore, the above‐indicated facts stipulate the prominent need for improved and personalized therapeutic approaches for age‐related hematopoietic defects. Although remarkable strides have been achieved through extensive research in this field, further research is required to unveil the underlying molecular mechanism of leukemogenesis.

Age‐associated defects in hematopoiesis: CHIP, MDS, and AML

Biological aging has been associated with several cellular and systemic defects, at least to some extent, due to the accumulation of somatic mutations in the Hematopoietic Stem Cell (HSC) compartment. As a result, mutations provide a fitness advantage to the mutated HSCs, leading to the expansion of identical mutated cells, called clones, giving rise to clonal hematopoiesis [17].

A subset of elderly with clonal hematopoiesis (10–40%), progresses to Clonal Hematopoiesis of Indeterminate Potential (CHIP), which is identified by the presence of cancer‐associated mutated clones in >4% of enucleated blood cells, but with no existing neoplastic features. CHIP is characterized by an increased risk of development of hematologic malignancies, including MDS and AML after the incorporation of additional mutations [17, 18].

MDS is a very heterogeneous group of clonal hematopoietic disorders, which are mainly characterized by ineffective hematopoiesis, dysplasia and the accumulation of immature blood cells (Fig. 1). Next‐generation sequencing (NGS) has allowed the extensive characterization of the molecular pathology of MDS through the determination of driver mutations [19]. The characterized driver mutations have been involved in fundamental cellular processes, namely RNA splicing, DNA modification, chromatin regulation, and cell signaling, whereas the number of the oncogenic aberrations in combination with the subclonal mutations are indicative of disease progression [19]. Although the aforementioned NGS studies have provided a great tool for prognosis and clinical management, further investigation is needed to elucidate the mechanisms underlying disease progression. Various additional factors, including the differential contribution of the identified mutations, chromosomal abnormalities, and the MDS subtype, might be of great importance.

Fig. 1.

The leukemic transformation from MDS to AML. The term MDS refers to a heterogeneous group of clonal hematopoietic disorders, characterized by impaired differentiation and hematopoietic dysfunctions, resulting in reduced erythrocytes, white blood cells and platelets. MDS commonly progresses to AML due to the accumulation of genomic aberrations that promote unrestrained proliferation. AML constitutes a very diverse group of aggressive hematological malignancies characterized by the accumulation of immature myeloblasts in the bone marrow and the peripheral blood.

MDS is commonly referred to as a pre‐leukemic stage due to the increased risk of transformation to AML, a highly aggressive type of leukemia. A two‐hit model had been initially proposed as the underlying mechanism in AML transformation that denotes the accumulation of Class I mutations (i.e. in FLT3, c‐KIT, etc.) providing the advantage of excessive proliferation and Class II mutations (i.e. in CEBPA) blocking the normal hematopoietic differentiation [20, 21]. Additional mutations associated with the promotion of epigenetic alterations (i.e mutations in TET2, DNMT3A, and ASXL1) and mutations at the tumor suppressors level (i.e. TP53 mutations) have been also identified and can cooperatively lead to leukemia development [22, 23, 24]. However, more recent studies have provided contradicting evidence, proposing an alternative model of non‐linear parallel clonal evolution with distinctive subclones within stem cell compartments leading to MDS and AML progression [25]. This indicates the importance for the investigation of the higher subclonal complexity of the distinct stem cell compartments in MDS to AML transformation, which challenges the belief that the increased clonal complexity arises in the blast compartment [25].

AML is the most common acute leukemia in adults, and it has been defined by uncontrolled clonal expansion of undifferentiated myeloid precursors in the bone marrow and peripheral blood, leading to ineffective hematopoiesis and bone marrow failure. The molecular heterogeneity of the disease, the complex pathways involved in the establishment and maintenance of AML and the vigorous clonal landscape render AML a complex dynamic disease. Enduring progress has been made in profiling the genomic scenery in AML. Gene mutations associated with leukemia have been cataloged and categorized by the Cancer Genome Atlas initiative and at least one potential driver mutation has been identified per case [26]. Potent mutational association in AML pathogenesis has been revealed resulting in the determination of three additional driver mutation sub‐categories, namely chromatin‐spliceosome, TP53‐aneuploidy, and IDH2 ^R172 mutation, as well as in the mapping of clonal trajectories, providing novel insights into the role of clonal complexity through disease progression [27, 28].

In 2022, the 5th edition of the WHO Classification of Hematolymphoid Tumors, the European Leukemia Net (ELN) 2022 risk classification, and a new classification system, the International Consensus Classification (ICC) of Myeloid Neoplasms and Acute Leukemias, have been published [29, 30, 31]. All the aforementioned classification systems have highlighted the significance of the incorporation of molecular analysis in everyday clinical practice and have emphasized the need for personalized therapeutic approaches, arising from the deeper understanding of the pathophysiology underlying AML complexity and heterogeneity, as well as the multiple novel therapeutic agents developed in the last decade. Although exceptional progress has been made, the rudimental mechanisms of leukemic transformation have not been elucidated yet and further research is needed to provide extended knowledge of the molecular pathways governing disease progression for more efficient patient stratification and treatment.

JAK–STAT signaling: a pleiotropic pathway

The Janus Kinase‐Signal Transducer and Activator of Transcription (JAK–STAT) signaling pathway has been of major significance for cellular function. Since its discovery more than 30 years ago [32, 33, 34], JAK–STAT represents one of the most conserved pathways that mediates signal transduction from the extracellular environment to the nucleus and leads to the modification of transcription. Activation of the pathway is induced by a plethora of molecules including cytokines, growth factors, and hormones, and regulates several important cellular functions, such as proliferation, differentiation, migration, and cell survival depending on the stimuli from the environment, thus providing a direct communication diode between the extracellular niche and the nucleus [35].

The JAK kinase family is composed of four members, JAK1, JAK2, JAK3, and TYK2 [36, 37, 38, 39]. STAT protein family in mammals includes seven members, STAT1, STAT2, STAT3, STAT4, STAT5A, STAT5B, and STAT6, all of which contain a conserved common structure, that includes the N‐terminal domain, the coiled‐coil domain, the Src‐homology 2 domain that is important for STAT dimerization, the linker domain, the DNA‐binding domain and the transactivation domain [40, 41, 42, 43] (Fig. 2). The pathway has been first discovered within the context of research for interferons, which are major players in anti‐tumor immune response [44, 45] and interleukin‐6 (IL6) signaling [46, 47]. Since its discovery, thorough research has uncovered its activation by over 50 cytokines, resulting in its involvement in many cellular functions, including cell cycle progression, development, proliferation, differentiation, survival, and apoptosis through ligand‐mediated signal transduction [35, 48, 49, 50, 51].

Fig. 2.

Structure of STAT protein family. The STAT family comprises seven highly similar members, STAT1, STAT2, STAT3, STAT4, STAT5A, STAT5B and STAT6 that share a structural topological similarity and are organized in six highly conserved structural domains. The schematic illustration shows the structural domains of the STAT proteins, including the N‐terminal domain (NTD), followed by the coiled‐coil (CCD), the DNA‐binding (DBD), the linker (LD), the Src‐homology 2 (SH2) and the carboxy‐terminal transactivation (TAD) domains. The NTD and SH2 domains are involved in protein–protein interactions and dimerization, the CCD is associated with nuclear localization, and the DBD mediates the DNA binding on specific sequences. Finally, the TAD C‐terminal domain includes essential tyrosine and serine residues for phosphorylation. The total number of amino acids constituting each STAT member is shown.

JAK–STAT canonical activation is achieved by the binding of an extracellular ligand to its transmembrane receptor that is noncovalently linked to JAKs, which undergo autophosphorylation and mediate tyrosine phosphorylation of the receptors acting as docking sites for the recruitment of STATs [32, 52] (Fig. 3). Subsequently, STATs are phosphorylated by JAKs at a single tyrosine residue in the C‐terminus, resulting in STAT homo‐ or/and hetero‐dimerization and translocation to the nucleus, where they bind to their target genes and other genomic positions, and regulate transcription. STATs are then dephosphorylated and translocated to the cytoplasm. Negative regulation of the JAK–STAT signaling pathway is mainly achieved by the Protein Inhibitors of Activated STATs (PIASs), the Protein Tyrosine Phosphatases (PTPs), and the Suppressors of Cytokine Signaling (SOCS) proteins. PIASs inhibit STAT phosphorylation and translocation, PTPs control STAT dephosphorylation, while SOCS are regulated by STATs and inhibit the JAK–STAT signaling via various mechanisms, creating a negative feedback loop for STAT activation [53, 54, 55]. Hence, the JAK–STAT pathway provides a simple yet sophisticated, tightly regulated and highly precise system, as different ligands result in the activation of specific STATs, which consequently regulate divergent pathways.

Fig. 3.

Canonical JAK–STAT pathway. The JAK–STAT pathway constitutes a highly conserved system that arbitrates the communication between the nucleus and the extracellular stimuli prompting necessary transcriptional changes and controlling important cellular functions including differentiation, proliferation, and survival. Several ligands, including cytokines, hormones, and growth factors, bind to their transmembrane receptor leading to JAK engagement that mediates the phosphorylation of JAKs and the creation of the docking site for STATs. Specific STAT proteins are recruited and activated by Tyrosine residue phosphorylation mediated by JAKs creating active dimers. Subsequently, activated STAT dimers translocate into the nucleus, where they either induce or inhibit transcription of their target genes through various direct or indirect mechanisms. The main negative regulators of the JAK–STAT pathways are PIASs, PTPs and SOCS proteins. The first two categories are implicated in STAT dephosphorylation, while SOCS are important JAK–STAT inhibitors and known STAT targets, creating a negative feedback loop for the regulation of the STAT function.

Recent studies have added further complexity to the role of STAT proteins by shedding light on novel aspects of non‐canonical STAT function. In contrast to the canonical JAK–STAT pathway, in which unphosphorylated STATs (uSTATs) are considered inactive, recent data have revealed that nuclear colocalization of uSTAT5 with CTCF in the absence of thrombopoietin (TPO), has repressed a megakaryocytic transcriptional program, identifying a novel cytokine‐dependent mechanism for megakaryocytic differentiation [56]. Moreover, STAT localization has been shown to be of great importance. Cytoplasmic localization of STATs has been typically considered to be equivalent to the inactive state, however active STAT5 has exhibited distinct effects in the cytoplasm related to the PI3K‐AKT pathway activation and leukemogenesis. Specifically, protein–protein interaction between STAT5 and p85 (the regulatory subunit of PI3K), mediated by the scaffolding protein GAB2, has been essential for the activation of PI3K‐AKT pathway [57]. In accordance to these findings, constitutively activated STAT5 localized in the cytoplasm in leukemia patient samples have led to PI3K‐AKT pathway activation, assigning an additional role to STAT5 as a cytoplasmic signaling effector molecules [58, 59].

In connection with aging, the JAK–STAT pathway might exhibit different and sometimes opposite functions depending on several factors including the organism, the tissue, the developmental stage, etc. Reduced JAK–STAT signaling has led to early aging‐associated deficiencies in cytokinesis of Drosophila stem cells, through the control of F‐actin, which is implicated in the germline stem cell cytokinesis, leading to abscission failure [60]. Also, reduced JAK–STAT signaling and cellular levels of pSTAT in leukocytes have been associated with increased physical weakness of the elderly [61]. However, age‐related increased JAK–STAT signaling has led to the promotion of senescence in tendon stem and progenitor cells during tendon aging. This effect had been reversed after JAK2 or STAT3 inhibition [62].

As mentioned, irregular cell signaling is positioned on the basis of the pathophysiology of the age‐related hematopoietic defects, such as MDS and AML. Amongst the most common mutations in AML and several other hematologic malignancies, numerous activating mutations in the JAK–STAT signaling pathway have been identified. STATs and particularly, aberrant activation of STAT3, STAT5A, and STAT5B have been implicated in aging and the development of numerous hematopoietic malignancies and solid cancers including MDS and AML, but their role has not been fully delineated yet [63].

Here we are reviewing the role of the STATs, and more specifically STAT3 and STAT5 factors in leukemic transformation to provide further insights into the mechanistic role of STATs in the process of age‐related leukemogenesis.

STAT3 and STAT5 in normal hematopoiesis and leukemia

Amongst the STAT family, STAT3 and STAT5 possess crucial roles in normal hematopoiesis and their abnormal function leads to leukemogenesis. Both STAT3 and STAT5 genes are located on human chromosome 17. STAT3 has four splicing isoforms, STAT3α, STAT3β, STAT3γ, and STAT3δ [64], while STAT5 comprises two highly related proteins, encoded by two separate genes (STAT5A and STAT5B) that are major regulators with pleiotropic roles in normal and pathologic hematopoiesis, including leukemia [65, 66] (Fig. 3). Despite the genomic proximity and the high peptide sequence similarity (more than 90%), STAT5A and STAT5B appear to have distinct roles [67, 68, 69]. However, the respective contribution of STAT3, STAT5A and STAT5B proteins, as regulators in the HSCs and leukemia, has not been fully elucidated.

STAT3 and STAT5 in normal hematopoiesis

SΤΑΤ3 has been an important factor at several levels for the development, as well as the maintenance and the regulation of a plethora of biological processes, including normal hematopoiesis. In greater detail, STAT3 has been essential in early embryonic development, and its ablation has led to early embryonic lethality in mice [70]. Specifically, IL6‐mediated STAT3 activation has prompted an anti‐apoptotic benefit through the high expression of BCL‐2 in human CD34⁺ cells [71]. Additionally, STAT3 has been important for Hematopoietic Stem and Progenitor Cells (HSPCs) sustainability and lineage balanced hematopoiesis. Conditional Stat3 inhibition in mice has resulted in a reduction of the Lin⁻Sca1⁺c‐Kit⁺CD150⁺CD48⁻ HSPC subset, myeloid skewing, and increased DNA damage in HSPCs. This phenotype has been rescued by concomitant Ube2n deletion [72], which encodes the E2 ubiquitin‐conjugating enzyme UBC13 implicated in proinflammatory signaling and NF‐kB activation. Ube2n transcription has been suppressed by STAT3, influencing NF‐kB activity. These findings imply that STAT3 holds a protective role in bone homeostasis and control of inflammation [72, 73]. Regarding STAT5, its role in hematopoiesis has been even more extensively researched than STAT3, but its function has not been fully elucidated yet, due to its high complexity and the different levels of control. However, several studies have provided significant insights into STAT5 function. Specifically, STAT5 deficiency in HSCs has led to impaired long‐term repopulating activity of the HSCs in the bone marrow of lethally irradiated mice [74], while its activation has been determined as a major regulator of the cytokine‐induced HSC repopulating ability, but not for the self‐renewal capacity [75]. This effect could not be fully attributed to the lack of TPO responsiveness upstream STAT5, implying that STAT5 has been activated through different pathways to stimulate differentiation and it is essential at several steps of hematopoiesis [76]. Further research findings have indicated that STAT5 has been necessary in the perpetuation of transcriptional heterogeneity of HSPCs, acting as a moderator for clonal balance. Thus reduction of STAT5 activation during aging might be linked to increased clonality observed in the elderly and development of leukemias [77].

Moreover, STAT5 has been involved in differentiation and quiescence of HSCs through multiple mechanisms. STAT5 has been identified as a moderator in both primitive and definitive erythropoiesis and has exhibited inhibitory features in early embryonic erythropoiesis in xenopus models [78]. Besides, HSC quiescence in steady‐state hematopoiesis has been preserved through STAT5 signaling and has maintained the hematopoietic reservoir [79]. Accordingly, constitutive STAT5 activation and increased levels of BCLXL expression has led to induction of endogenous erythroid colony formation in human‐derived primary cells [80]. Further studies have provided evidence about the critical role of the STAT5 cellular level that has influenced HSC lineage commitment. Specifically, reduced STAT5 signaling in CD34⁺ cells has stimulated megakaryocytic growth, whilst STAT5 activation has promoted erythropoiesis [81]. Also, other evidence has suggested that the absence of Stat5 expression in adult bone marrow and fetal liver, resulted in decreased lymphomyeloid repopulating activity in mice [82]. Further investigation has indicated that STAT5 has been essential for the lymphomyeloid equilibrium and the repopulating capacity in hematopoiesis and has also underlined the significance specifically of the STAT5 N‐terminal domain in HSCs [83]. Moreover, STAT5 has been essential for myeloid differentiation through the regulation of Bclx resulting in promotion of survival and proliferation of the myeloid differentiating progenitors [84]. Another mechanism that implicates STAT5 in myeloid differentiation has been uncovered through the STAT5‐mediated CEBPA downregulation that has prompted self‐renewal and impaired myelopoiesis in human HSPCs [85].

Furthermore, STAT5 has been associated with stem cell proliferation through its regulation by microRNAs (miRNAs). Specifically, miR‐193b upregulation has restricted uncontrolled HSCs expansion through a STAT5‐mediating mechanism upon self‐renewal via the promotion of TPO‐MPL‐STAT5 signaling. This mechanism has provided negative feedback for unrestricted HSCs proliferation [86].

In addition, STATs interact with other proteins creating complexes to directly or indirectly activate or repress the transcription of target genes depending on the interacting partners, the binding affinity, the underlying motifs, and the genomic localization of the binding [87]. This differential regulation through STAT3 and STAT5 factors is at least partially achieved through the interaction with other proteins and/or transcription factors allowing broad functionality for STATs, with distinctive roles depending on the cell type, the differentiation stage, and the cellular status in healthy and malignant conditions.

The role of STAT3 and STAT5 in leukemia

Several factors, including upstream regulation, activation, mutations, protein stability, and post‐translational modifications, can lead to either hyperactivation or inactivation of STAT3 and STAT5A/B that result in various cancers, suggesting that the quote of Ancient Greeks ‘Metron Ariston’, which means moderation is best, is the necessary approach when considering STAT‐mediated signaling (Fig. 4).

Fig. 4.

JAK–STAT effects in Leukemia. The JAK–STAT pathway has been involved in several essential cellular functions including survival, apoptosis, proliferation, and differentiation. Additionally, biological processes such as inflammation and aging strongly affect hematopoiesis, through the dysregulation of bone marrow niche and clonal hematopoiesis, which have been mediated by JAK–STAT signaling. Impairment in the JAK–STAT pathway, mostly involving constitutive activation of STAT3 and STAT5 has been implicated in leukemogenesis and increased relapse rates, through numerous molecular mechanisms resulting in blockade of differentiation, and induced survival and proliferation. This indicates the necessity to delineate the role of STAT3 and STAT5 in leukemia development to allow design of novel therapeutic strategies.

STAT3 has been extensively researched in terms of its role in MDS and AML and multiple mechanisms have been uncovered, commonly being appointed an oncogenic character [88]. However, controversial evidence has also been presented, supporting STAT3 tumor suppressing activity in various cancers [89, 90].

In addition, STAT5, as already stated, is an essential regulator for several important cellular functions such as cell proliferation, differentiation, and survival. Ostensively divergent STAT5 signaling has been associated with numerous malignancies, including MDS and AML. Aberrant STAT3 and STAT5A/B signaling has been associated with leukemogenesis. Although several mechanisms and signaling pathways in MDS and AML implicating STAT3 and STAT5 activation and deregulation have been unveiled, the molecular pathogenesis has not been fully elucidated yet.

Concerning STAT3, it has been associated with negative disease outcome in AML via various mechanisms. Specifically, its activation has been connected with poor disease outcome in AML, through increased levels of LAPTM4B that interacts with RPS9 causing increased protein stability of RPS9, thus enhancing AML cell progression [91]. Moreover, expression of the immune checkpoint molecule, V‐domain Ig suppressor of T‐cell activation (VISTA) has been transcriptionally controlled by STAT3 via direct binding. Hence, constitutive STAT3 has caused increased VISTA expression in AML cells, which has been linked to unfavorable disease prediction [92].

Additionally, STAT3 pathway has been implicated in important cellular functions in AML cells including proliferation, apoptosis, and survival. For example, STAT3 has regulated autophagy downstream of the oncogenic KIT D816V mutation in AML cells resulting in increased proliferation and survival [93]. Besides, suppressed STAT3 expression by RNA interference in HL‐60 AML cells, has restrained proliferation and promoted apoptosis, assigning an oncogenic feature to STAT3 [94]. An additional mechanism has been identified, implicating STAT3 in MDS and AML immune resistance. In detail, maturation of myeloid cells has led to STAT3‐mediated atypical IFN‐γ signaling and PD‐1 ligands (PD‐L1 and PD‐L2) upregulation, thus increasing the responsiveness of MDS and AML cells to IFN‐γ. However, MDS and AML cells have utilized STAT3 pathway and PD‐1 ligands to surpass the immune response caused by IFN‐γ and resulted in secondary immune resistance [95].

Concerning STAT5, it plays a significant role in the survival and proliferation of leukemia. In greater detail, STAT5 has been involved in the long‐term preservation of both normal and leukemic HSPCs [96], while its activation in CD34⁻ KSL cells has led to an increase in the multipotent progenitors population and has promoted HSC self‐renewal ex vivo [97]. On the other hand, STAT3 has been non‐essential for the HSC maintenance in vivo, and has stimulated differentiation in vitro providing a better understanding of the distinct functions of STAT3 and STAT5 in normal hematopoiesis and leukemogenesis [97].

An additional level of control for STATs lies in the activation via phosphorylation of specific tyrosine residues, which when perturbed can lead to aberrant activation of STATs that has been strongly associated with leukemias. Interestingly, STAT5A activation through serine phosphorylation can be critical for leukemogenesis. For instance, STAT5A phosphorylation on Ser725 and Ser779 residues in human leukemic cell lines and patient‐derived primary samples has been essential for hematopoietic cell transformation [98].

The complexity of the pleiotropic function of STATs is, at least to some extent, owed to the large number of signaling molecules that lead to their activation. In accordance with the described role of STATs thus far in leukemia, knockdown of SPATS2L, a poor prognosis marker in AML, has led to decreased JAK2, STAT3, and STAT5 protein levels and increased apoptosis and cell growth suppression, highlighting the significance of JAK–STAT signaling in AML [99]. However, abnormally low STAT3 and STAT5 activation has been linked to dismal diagnosis through expression of inflammation‐associated genes in pediatric AML [100].

Another mechanism responsible for irregular STAT5 signaling has been identified involving a constitutively activated STAT5A mutant that formed enhanced levels of stable tetramers. STAT5 tetramers have accumulated in excess to dimers in human leukemias and have been linked to leukemogenesis [101]. The tetrameric STAT5 has been demonstrated to possess the ability of recruiting the chromatin modifier EZH2, leading to inhibition of gene expression of the IgK locus. However, this repressive role has not been identified in STAT5 dimers [102]. This further supports recent findings that provide an epigenetic regulator role to STAT3 and STAT5 through the control and recruitment of epigenetic factors to specific gene loci [103]. Specifically, STAT5A and STAT5B have been shown to induce the expression of the Dpf3 gene that encodes for an epigenetic factor, through direct binding to its promoter causing increased expression that has been linked to leukemia [104]. Additionally, DNMT3A expression, an epigenetic modifier that is commonly mutated in both MDS and AML, has been positively controlled by STAT5A, leading to epigenetic inhibition of tumor‐suppressor genes in CD34⁺CD38⁻ AML cells [105].

To further investigate the role of STAT3, the different STAT3 isoforms in leukemogenesis have been investigated, but their function has not been delineated yet. Nonetheless, there is evidence for distinct roles of STAT3α and STAT3β isoforms in hematologic malignancies. Specifically, it has been recently revealed that the α and β splicing isoform ratio has been significant in AML, with the higher STAT3β/α mRNA ratio having been identified as a prediction marker for favorable disease outcome [106]. Additionally, STAT3β has been characterized as a tumor suppressor in an AML transgenic mouse model, leading to decreased leukemia progression [106]. Accordingly, for STAT5 there is increasing evidence about the distinct role of STAT5A and STAT5B in leukemia, with STAT5B possibly holding a more dominant role. Importantly, STAT5B has been identified to control self‐renewal in both normal HSCs and leukemic stem cells (LSCs), through the STAT5B/CD9 axis [107]. Further evidence provides significant insights about the distinct function of STAT5A and STAT5B in MDS and AML, highlighting the significance of distinguishing between the two STAT5 factors [108].

Furthermore, STAT3 and STAT5 mediate the effects of key mutations in genes such as FLT3 and SF3B1 that have been associated with AML. FLT3 activating mutations are amongst the most common genetic aberrations in AML [26] and they result in the induction of STAT target genes [109, 110]. Several direct and indirect mechanisms involving the interplay between STAT5 and FLT3 mutations have been determined in AML. The FLT3 D835/I836 mutations have led to constitutive activation of STAT5 in pediatric ALL with hyperdiploidy and infant ALL with MLL rearrangements [111]. Additionally, internal tandem duplications (ITD) in FLT3 gene occur in approximately 25% of AML patients [112, 113]. The FLT3‐ITD mutation constitutively activated STAT5, leading to upregulated expression of the crucial factor for survival MCL‐1, providing a survival advantage in LSCs, irrespective of the wild‐type FLT3 signaling [114]. Also, in FLT3‐ITD AML, CDC25A, a phosphatase important for proliferation and differentiation, is directly regulated by STAT5 through the STAT5/miR‐16 axis [115]. Furthermore, STAT5 signaling in FLT3‐ITD mutated cells has caused increased PIM kinase expression and mTORC1/MCL1 signaling, which has served as a resistance mechanism for the PI3K/AKT pathway inhibitors [116]. Further evidence supports the involvement of STAT5 signaling in resistance mechanisms, provided that FLT3‐ITD mutations have caused CXCR4 upregulation in AML patients, involving also the signaling components STAT5 and PΙΜ1 that have been associated with therapy resistance and relapse [117].

Nevertheless, the inactivation of STAT5 by a type I tyrosine kinase inhibitor (gilteritinib), has resulted in the diminishment of FLT3‐ITD/D835 mutant cells through reduction of the expression of the downstream target of STAT5, BCL2A1, preventing treatment resistance [118]. Moreover, STAT5 aberrant activation downstream of FLT3‐ITD, has resulted in induced PIM kinases expression and protection of the mTORC1/4EBP1/S6K/MCL1 signaling axis providing drug resistance to the PI3K/AKT inhibitors. Hence, suppression of the STAT5/PIM signaling axis has resulted in the reinforcement of the cytotoxic effects of proteasome inhibitors in FLT3‐ITD AML cells, providing more efficient treatment [119]. STAT5 has been also shown to play a crucial role in the resistance mechanism to IDH inhibitors through differentiation blockade in leukemic cells, signifying the necessity for targeting STAT5 signaling selectively, in combination with IDH inhibition, for more efficient treatment [120].

In addition, SF3B1 mutations are commonly present in AML and they may be linked to a MDS pre‐phase. SF3B1 is frequently mutated in MDS or MDS/myeloproliferative neoplasms (MPN), being linked to a favorable and slow disease progression [121]. Recent studies indicate that STAT3 has been essential for HSPCs in homozygous Sf3b1 mutant zebrafish. Heterozygosity of SF3B1 has been examined, due to its prevalence in hematologic malignancies, resulting in augmented sensitivity to STAT3 inhibition in zebrafish, mouse, and human HSPCs. STAT3 inhibition has increased aberrant splicing in SF3B1 mutant cells, which has provided a selective target in hematologic malignancies [122].

Also, recent findings have increasingly associated STAT3 with the function of miRNAs in AML. In the last decade, it has been shown that miRNAs are involved in leukemogenesis, either as tumor suppressors or as oncogenes, and can constitute precious targets for biomarker development and/or novel therapies [123, 124]. Namely, miR‐548ac delivered by AML cell‐derived extracellular vesicles to CD34⁺ HSCs have induced TRIM28 expression and downstream activation of STAT3 that resulted in impeded hematopoiesis [125]. In addition, miR‐4532, delivered into CD34⁺ HSCs via AML cell‐released exosomes, has suppressed normal HSC hematopoiesis through the stimulation of the LDOC1‐mediated STAT3 signaling pathway [126]. Moreover, the STAT3 pathway has been activated by AML cell‐derived extracellular vesicles containing miR‐1246 through the direct regulation of the tumor‐suppressor gene LRIG1, hence leading to increased LSC survival [127]. Collectively, these data suggest that miRNAs derived by AML cells activate STAT3 to promote leukemia and inhibit normal hematopoiesis.

To add another level of complexity, an important role has been attributed to STAT3 regarding cellular metabolism in AML. Recent studies have revealed that STAT3 controls oxidative phosphorylation through numerous mechanisms, a procedure that has been the metabolic pathway of choice for LSCs, and is essential for their proliferation and survival [128]. Upon further research, it has been discovered that STAT3 has mediated oxidative phosphorylation in LSCs through the expression of MYC, which has successively regulated SLC1A5, an important factor for glutamine transport in LSCs. Consequently, SLC1A5 inhibition has resulted in decreased glutamine, glutathione, and TCA cycle metabolites, thus truncating the TCA cycle and repressing oxidative phosphorylation [129]. This provides a direct connection between STAT3 and metabolism in leukemia [129].

Finally, these findings confirm the functional involvement of mutated/activated STAT3 and STAT5 in hematologic malignancies, as well as their distinct roles in each condition. Several studies have identified target genes regulated by STATs and protein complexes of STATs in normal hematopoiesis and leukemia [87, 104]. Although such technological and scientific developments have contributed to vast progress in comprehending the pathways and the processes involved in leukemogenesis, further research is needed. The complexity and diversity of STAT mechanisms in leukemia render the development of efficient therapeutic agents very challenging, requiring deeper investigation to fully depict the molecular mechanisms through which STAT3, STAT5A/B, and their target genes are implicated in leukemogenesis, as well as their crosstalk with other interacting partners and their effects in transcription.

Aging, immunity and inflammaging: the role of STAT3 and STAT5, and their association with leukemogenesis

Leukemia incidents including AML are augmented in the elderly population. Biological aging is accompanied by several cellular changes including accumulation of somatic mutations, epigenetic alterations, stem cell malfunction, cellular senescence, etc. that have been associated with leukemogenesis [130]. However, the exact mechanisms that connect the aging process with leukemia have not been elucidated.

Aging has been strongly correlated with changes in the innate and adaptive immune system [131, 132], possibly creating a connection between aging and leukemogenesis. The JAK–STAT pathway might be a significant link between immunity and aging and thus leukemia development. STATs have been implicated in immunity through various mechanisms [133]. STAT5 has been widely involved in lymphoid development. In particular, STAT5 has been shown to control B‐cell development, albeit it has not been essential for B‐cell maturation and Ig production [134]. STAT5 has been shown to interact with LSD1 and HDAC3 in pro‐B cells regulating STAT5‐dependent transcription, signifying the involvement of STAT5 in multi‐protein complexes that modify chromatin in B‐cell development [87]. Additionally, JAK–STAT5 has been involved in the early development of B cells, being regulated by COX1‐derived thromboxane A2 that is important for the immune effects [135]. Regarding T cells, both STAT5A and STAT5B have controlled cell cycle and proliferation through IL2 signaling, which revealed a functional overlap between the two factors, provided that defects in T‐cell development have not been observed in deficient mice of STAT5A or STAT5B alone [136]. Furthermore, the interplay between mTOR and STAT5 signaling influences the mutual differentiation of effective and regulatory T cells (Teff and Treg cells), leading to different proliferation activity. This finding offers novel insights into the regulation of the Treg‐Teff equilibrium [137]. In Tregs maturation, STAT5 has been bound directly to Foxp3, being necessary for its induction, whereas STAT3 has been necessary for the reduction of FOXP3 through IL6 signaling, presenting opposite roles via differential control of FOXP3 [138]. Additionally, Stat5 expression has been required for mast cell survival, development, and IgE‐induced activation, the latter partly through post‐translational cytokine regulation [139, 140]. Recent studies provide evidence for the existence of interaction between STATs and the Ikaros family, which includes five highly homologous members IKZF1‐5 and has been strongly linked to immunity and lymphocyte development [141, 142]. Research on the Ikaros transcription factor family in T helper (TH) cells has indicated that there is possible cooperation between STATs and IKZFs [143]. Specifically, in T Follicular Helper (TFH) cells STAT3 has been shown to interact with Aiolos (IKZF3), creating a complex for the upregulation of the expression of Bcl6 and TFH cells‐associated cytokine receptor Il6ra [144]. However, in the same cell type the IL2/STAT5 signaling has hindered the Aiolos/STAT3 pathway by dual inhibition of Aiolos (IKZF3) expression and STAT3 activity [143]. In pre‐B cells, Ikaros (IKZF1) has been shown to directly compete for STAT5 DNA binding for common targets with GGAA motifs, as induced Ikaros activity has resulted in rapid extensive loss of STAT5 genome‐wide binding. This has provided an understanding of the mechanism behind the opposite effects between Ikaros and IL7/STAT5 pathways in B‐cell development [145].

All these studies provide continually extended evidence for STAT and Ikaros (IKZF) family engagement in mutual mechanisms, suggesting that interaction between STATs and IKZF factors may transpire to key players for the control of lymphomyeloid development.

Furthermore, the immune system's response to potentially dangerous stimuli, known as inflammation, is amongst the seven pillars of aging, which also include macromolecular damage, epigenetics, adaptation to stress, proteostasis, stem cell regeneration, and metabolism [146]. The term inflammaging has been used to describe a sterile, chronic, low grade inflammatory state that is linked to cellular senescence, dysregulation of mitochondrial function, autophagy, mitophagy, ubiquitin‐proteasome system, and activation of the DNA damage response mechanisms and dysbiosis (in host microbiota). Inflammaging contributes to a plethora of age‐associated diseases, including hematologic malignancies and cancer [147]. Aging has been associated with increased inflammatory signaling pathways that lead to senescence and conversely, senescent cells promote inflammation that accelerates biological aging, creating a vicious cycle. Indeed, inflammation stimulates HSCs aging that affects several cellular functions, such as differentiation, self‐renewal and energy metabolism [148]. However, the causality between inflammation and aging has not been elucidated yet. The JAK–STAT signaling plays an instrumental role in hematopoiesis, inflammation, and immunity, with nearly all of the immune responses triggered by cytokines being STAT‐dependent [149]. Expectedly, dysregulation of the JAK–STAT pathway has been associated with inflammation, autoimmune diseases, and cancers, including hematologic malignancies such as leukemia [150]. Increased inflammatory signaling in the human AML niche has caused aberrant JAK–STAT signaling in AML blasts and bone marrow stromal cells and consequently, escalation of leukemic proliferation [151].

Augmented STAT1, STAT3 and STAT5 activation has been positively correlated with age, alongside with increased inflammatory signaling‐associated molecules, including IL6 and sTNF‐RII [152]. In addition, STAT3 mitochondrial localization has been enhanced with age in CD4⁺ T cells leading to increased proinflammatory cytokine production and mitochondrial function creating a possible link between metabolism and inflammaging mediated by STAT signaling [153]. Also, cytokine signaling by both proinflammatory and anti‐inflammatory cytokines has been STAT3‐ and STAT5‐mediated [154, 155]. Hence, the deregulated STAT signaling might be the connecting link between inflammation and aging leading to leukemia development. Furthermore, the activation of STAT3 and STAT5 signaling pathways plays a crucial role in the regulation of innate receptor‐induced inflammatory cytokines, with the level of signaling establishing distinct patterns of regulation. In greater detail, following human monocyte‐derived macrophages stimulation by pattern‐recognition receptors, reduced expression of STAT3, STAT5A, and STAT5B has resulted in a decline in anti‐inflammatory cytokines. Initially, proinflammatory cytokines decline, but then increase, when the expression of STAT3 or STAT5 falls below a certain level [155].

STAT3 function in inflammation has been extensively researched and an anti‐inflammatory effect in HSPCs has been uncovered through transcriptional downregulation of the Ube2n gene, involved in proinflammatory signaling. This has indicated a STAT3‐dependent protection of the hematopoietic system preventing uncontrolled proinflammatory signaling by Ubc13 [72]. However, contrary evidence indicated that chronic inflammation reduced the HSC fitness through the stimulation of the JAK–STAT3 signaling pathway, which was inverted after Stat3 inhibition, in mice with chronic multifocal osteomyelitis [156].

As described in this chapter, STAT3 and STAT5 have been implicated in the regulation of innate immunity and inflammation, two functions that have been strongly associated with aging and age‐related conditions, such as hematopoietic malignancies. Albeit, further research is required to better outline the link between STATs in immunity and inflammation with aging and by extension to leukemogenesis. Elucidation of the role of STAT3 and STAT5 in age‐related leukemogenesis will provide novel targets for therapeutic approaches to improve management and treatment of various leukemias.

JAK–STAT targeted therapies in MDS and AML

The continuously increasing evidence about the significance of deregulated JAK–STAT signaling in myeloid disorders, such as MDS and AML, has augmented the interest in the scientific community about direct or indirect targeting for therapeutic purposes. Several therapeutic biological agents targeting JAK–STAT signaling components have been developed, including antibodies targeting cytokines and/or receptors, JAK and STAT inhibitors. Ruxolitinib is a JAK1 and JAK2 inhibitor approved by the US Food and Drug Administration (FDA) for treating myelofibrosis, polycythemia vera, and steroid‐refractory graft‐versus‐host disease [157, 158, 159, 160]. Fedratinib, a JAK–STAT inhibitor has been presented as a potential therapeutic molecule, with 78% tumor efficacy response in an ex vivo tumor explant AML model for patients who are not responding to cytarabine or azacitidine [161]. Still, several adverse effects have been identified by the use of JAK inhibitors highlighting the urge for enhanced understanding of the JAK–STAT pathway [162].

Although direct STAT inhibitors of clinical grade are not currently available, the use of direct or indirect inhibitors at preclinical level has provided further information about the mechanisms underlying STAT3 and STAT5 roles. For STAT3, chidamide, a potent HDAC inhibitor, has been exhibited to decrease MDS and AML cell survival through SOCS3‐mediated JAK2‐STAT3 inhibition [163]. AZD9150, an antisense STAT3 inhibitor, has enhanced apoptosis, hindered survival in leukemic cell lines and promoted differentiation in primary MDS/AML stem and progenitor cells, further confirming the significance of STAT3 as a leukemia‐promoting factor [164].

The possibilities for STAT3 and STAT5 direct targeting have been extensively reviewed for the development of cancer therapies [165, 166]. Despite the challenges in directly targeting STAT3 and STAT5 factors, the imminent technology of the proteolysis‐targeting chimera (PROTAC) degraders provides a promising strategy to subdue the constraints of conventional small molecule inhibitors. In 2023, the development of AK‐2292, a small‐molecule STAT5 PROTAC degrader, was reported [167]. AK‐2292 has selectively led to the degradation of STAT5A and STAT5B proteins achieving STAT5 inhibition in normal and leukemic cells, and resulted in tumor regression in human chronic myeloid leukemia (CML) xenografts in mice and AML cell lines providing a promising new component for STAT5 inhibition in leukemia and a possible therapeutic agent [167, 168].

Additionally, SD‐36 is a highly selective PROTAC degrader that potently induces STAT3 protein degradation, leading to cell cycle arrest and/or apoptosis in AML cell lines and xenograft models [169, 170].

A small number of direct and indirect STAT3/5 inhibitors have been used in clinical trials for MDS and AML therapy (Table 1). Even though the potential of STAT3 and STAT5 for the development of targeted therapies is continuously emerging and many molecules are currently in preclinical trials, further clinical research is required for their utilization in clinical practice and their establishment as therapeutic agents.

Table 1.

Clinical trials using direct or indirect STAT3 and STAT5 inhibitors. Clinical trials from https://www.clinicaltrials.gov/ for the treatment of MDS and AML, using direct and indirect STAT3 and STAT5 inhibitors are shown (R/R, Relapsed/Refractory).

Inhibitor	Clinical trial ID	Target	Phase	Disease/info in clinicaltrials.gov	Current status
Danvatirsen (AZD9150)	NCT05986240	STAT3	I	R/R MDS and AML/Selective, high‐affinity, antisense oligonucleotide inhibitor of STAT3	Recruiting
Pyrimethamine	NCT03057990	STAT3	I	Intermediate/High‐risk MDS R/R after treatment with azanucleosides/STAT3 Inhibitor	Withdrawn
Pacritinib (SB1518) in combination with decitabine or cytarabine	NCT02532010	JAK2 FLT3	II	Elderly with AML/Reduces pJAK2, pSTAT3 or pSTAT5 in cell lines	Terminated

Conclusions

Here the role of aging in hematopoietic defects, the pleiotropic JAK–STAT pathway and the role of STAT3 and STAT5 in normal hematopoiesis, leukemia, immunity and inflammaging have been reviewed. It has been established that the JAK–STAT pathway is a major contributor to several processes in normal hematopoiesis and malignancies. Specifically, STAT3 and STAT5 deregulation has been heavily involved in leukemia development and maintenance through several mechanisms, hence they represent very attractive therapeutic targets.

Despite extensive research in the hematology field that has provided new insights into the JAK–STAT function in hematologic malignancies, the role of STAT3 and STAT5 in leukemia remains largely unclear. This is partly due to the increased complexity of the molecular mechanisms of STAT3 and STAT5 in the molecular pathogenesis of leukemia, as they can act as inducers and repressors militating transcription changes and epigenetic alterations. To develop novel treatments to maximize efficacy and minimize adverse effects, it is necessary to gain a deeper knowledge of the STAT3 and STAT5 function in leukemogenesis. Furthermore, the identification of novel biomarkers is of major importance for leukemia stratification and therapeutic management, to contribute to clinical decisions.

Author contributions

ESF and EK wrote and edited the manuscript. EK supervised manuscript preparation. Both authors contributed to the article and approved the submitted version.