Is the HLA System Really Involved in Implantation, Pregnancy, or Abortion of the Fetus?: Strengths, Weaknesses, and Controversial Points

ABSTRACT

Problems

Pregnancy is a sensitizing event for human leukocyte antigen (HLA) alleles, as the fetus, acting as a semi‐allograft, expresses maternal and paternal haplotypes in a codominant manner. Consequently, the mother may develop anti‐HLA antibodies directed against the paternal haplotype. However, unlike in organ transplantation, this sensitization is not usually associated with fetal rejection in multiparous women, raising a central question in reproductive immunology: How is maternal‐fetal tolerance achieved and maintained?

Method of Study

This review examines the influence of the HLA system on the immunological mechanisms underlying maternal–fetal tolerance.

Results

Accumulating evidence suggests that this phenomenon results from a complex network of cellular and molecular interactions. The roles of non‐classical HLA molecules, particularly HLA‐G, the modulation of NK activity via KIR receptors, and the expansion of regulatory T cells, which contribute to the maintenance of systemic immunological tolerance, are highlighted. In parallel, the relevance of anti‐HLA antibodies and polymorphisms in HLA regulatory regions in susceptibility to gestational complications has been investigated.

Conclusion

It considers how insights from reproductive immunology may inform related fields such as transplantation and oncology.

Abbreviations

ACII

Activator of class II (transcription factor)

APCs

Antigen‐presenting cells

C4d

Complement component 4d

CD4+/8+

Cluster of Differentiation 4 and 8 positive T cells

CD73

Cluster of Differentiation 73

cfDNA

Cell‐free DNA

CpG

Cytosine phosphate guanine motifs

DSA

Donor‐specific antibodies

EVT

Extravillous trophoblast

FcRn

Neonatal Fc receptor

Foxp3

Forkhead box P3

FR4

Folate receptor 4

HLA‐I

Human leukocyte antigen class I

IDO

Indoleamine 2,3‐dioxygenase‐1

IFN‐γ

Interferon gamma

IgG

Immunoglobulin G

IL‐10

Interleukin 10

IL‐8

Interleukin 8

KIR

Killer immunoglobulin‐like receptors

LILRB1

Leukocyte immunoglobulin‐like receptor subfamily B member 1

M2

Alternatively activated (anti‐inflammatory) macrophages

MMP‐9

Matrix metalloproteinase 9

NK cells

Natural killer cells

PD‐1

Programmed cell death protein 1

PD‐L1

Programmed cell death ligand 1

RPL

Recurrent pregnancy loss

sHLA‐G

Soluble HLA‐G

TGF‐β

Transforming growth factor beta

TLR9

Toll‐like receptor 9

TNF‐α

Tumor necrosis factor alpha

TRAIL

TNF‐related apoptosis‐inducing ligand

uNK

Uterine natural killer cells

VEGF

Vascular endothelial growth factor

1. Introduction

Successful gestation requires a unique immunological phenomenon: the maternal immune system adapts and coexists functionally with the fetus, a semi‐allograft that expresses human leukocyte antigens (HLA), both self (maternal) and non‐self (paternal) [1, 2, 3]. During this process, the cytotoxic adaptive immune response is inhibited, while the regulatory functions of the adaptive immune system are preferentially enhanced [4, 5]. In contrast, innate immunity remains functionally competent, fulfilling a dual role: On the one hand, it maintains protection against pathogens and, on the other, it actively participates in the immunological signaling and remodeling necessary for placentation and proper gestational development [6, 7, 8, 9]. Maternal IgG antibodies cross the placenta and provide passive immunity to the fetus during the first months of life [10]. IgG transfer occurs through the neonatal Fc receptor (FcRn) present on syncytiotrophoblast cells. Factors influencing transplacental transfer include maternal IgG levels, placental integrity, gestational age (higher transfer at week 13 and during the third trimester), IgG subclass (especially IgG1 and IgG3), and antigenic nature [11]. However, under normal conditions, this transfer does not occur, or at least not significantly, with anti‐HLA antibodies generated against the paternal haplotype in multiparous mothers.

The maternal‐fetal interface, represented primarily by the decidua and extravillous trophoblast (EVT), is the setting in which the regulation of the decidual immune response begins. In this environment, multiple mechanisms collaborate to promote active tolerance to paternal antigens, including modulation of maternal immune cells, such as decidual NK cells, tolerogenic dendritic cells, and regulatory T lymphocytes, as well as the expression of non‐classical HLA molecules by the trophoblast. These mechanisms not only prevent the activation of effector responses but also induce a local and systemic anti‐inflammatory environment, in which soluble factors such as interleukin‐10 (IL‐10), tumor growth factor β (TGF‐β), or indoleamine 2,3‐dioxygenase‐1 (IDO) may play a fundamental role [12]. However, maternal‐fetal tolerance is not absolute. In certain circumstances, such as preeclampsia [13, 14, 15], premature birth [16], or recurrent fetal loss [17] Alterations in these immune tolerance mechanisms have been identified, along with the presence of anti‐HLA antibodies, altered HLA expression, and uncontrolled immune activation. Furthermore, it has been shown that pregnancy can induce immunological sensitization in the mother to fetal HLA antigens, with important clinical implications in later contexts, such as solid organ transplantation [18, 19, 20, 21]

2. Tolerance Mechanisms at the Materno‐Fetal Interface

The maternal‐fetal interface is carefully designed for the exchange of gases and nutrients, thanks to the coating of the chorionic villi by villous cytotrophoblast and syncytiotrophoblast, respectively, and the remodeling of the uterine arteries by EVT, to ensure fetal development [22].

It is noteworthy that the EVT is the first barrier between maternal and fetal tissues; as a result, it is characterized by its ability to evade maternal immunosurveillance. Among its immune evasion strategies, the antigenic silencing of classical HLA molecules stands out [23]. The EVT protects the fetus by presenting classical HLA‐I molecules, specifically HLA‐C, to uterine NK cells (uNK). These interactions are independent of the peptide and are mediated by immunoglobulin‐like receptors on NK cells (KIR), which determine NK cell function. The KIR system is highly polymorphic and exhibits notable interindividual variability. In addition, the specificity of the interaction between the KIR and the HLA‐C molecule depends on variation at amino acid residue 80 of the HLA‐C molecule. Thus, HLA‐C1 alleles, characterized by the presence of asparagine, are recognized mainly by the KIR2DS2 (activator) and KIR2DL2/3 (inhibitors) receptors, while HLA‐C2 alleles, with lysine in that position, interact with KIR2DS1 (activator) and KIR2DL1 (inhibitor) [24]. This recognition system is an essential mechanism for balancing immunological tolerance and defense against potential threats during pregnancy.

In addition, EVTs characteristically express non‐classical HLA‐I (HLA‐G, E, and F). Unlike classical HLA genes, genes encoding non‐classical HLA molecules exhibit limited genetic diversity, resulting in a restricted repertoire of peptides and non‐protein antigens presented to T lymphocytes and NK cells [25, 26]. Genetic variations in untranslated regulatory regions have been shown to significantly influence the expression levels of these molecules. In this context, even small genetic differences can modulate the selective activation of immune cells against allogeneic cells, either by recognizing non‐classical HLA molecules that transmit stimulatory signals or by preventing interactions necessary to generate inhibitory signals [27].

Also notable is the production of programmed cell death ligand 1 (PD‐L1) by EVT (its binding to PD‐1 expressed on T lymphocytes inhibits their activation) [28, 29]. With a similar objective, syncytiotrophoblasts lack HLA expression, express PD1‐L, and release exosomes rich in Fas‐L and TRAIL (programmed cell death inducers) onto activated T lymphocytes [30, 31].

Beyond the well‐characterized role of PD‐L1 and apoptotic exosomes at the maternal‐fetal interface, recent studies have expanded the concept of immune checkpoint regulation in pregnancy, suggesting that tolerance is maintained by a coordinated network rather than by isolated pathways. In normal decidua, the canonical PD‐1/PD‐L1 axis remains a central regulator of T‐cell quiescence and trophoblast immune cell crosstalk; spatial and single‐cell transcriptomic analyses have demonstrated its abundant and compartmentalized expression in EVT, decidual T cells, and macrophages, linking balanced PD‐1/PD‐L1 signaling to controlled trophoblast invasion and physiological spiral artery remodeling [29, 32]. Consistently, dysregulation of this pathway in pathological pregnancies, particularly preeclampsia, has been associated with altered decidual immune landscapes, skewed macrophage polarization, and heightened inflammatory signatures, suggesting that loss of checkpoint equilibrium may contribute to placental dysfunction [33].

In parallel, inhibitory receptors, such as TIGIT, are increasingly recognized as modulators of uterine NK cells and other decidual immune cells. Experimental and transcriptomic studies indicate that TIGIT contributes to fetomaternal tolerance and restrains excessive cytotoxicity, and that its expression on NK cells changes across gestation [34, 35]. Recent single‐cell studies of decidual NK EVT crosstalk also support a role for inhibitory receptor networks that include TIGIT in regulating trophoblast invasion and immune homeostasis at the interface [36]. LAG‐3 has emerged as an additional regulatory layer in the decidua; its expression on decidual T cells has been linked to tolerogenic phenotypes in normal pregnancy, whereas altered LAG‐3/FGL‐1 signaling has been reported in early‐onset preeclampsia and recurrent pregnancy loss [37, 38]. In addition, studies up to 2025 have begun to implicate TIM‐3/Galectin‐9 signaling in the expansion of decidual regulatory T cells and the promotion of a Th2‐biased environment, further supporting the notion that multiple checkpoints act in concert rather than in isolation [39, 40] Overall, multi‐omic and spatial approaches indicate that these checkpoints form an integrated and dynamic regulatory network that adapts across gestation and varies between decidual niches, rather than representing static inhibitory signals [32]. Collectively, accumulating evidence suggests that a coordinated ensemble of immune checkpoints, including PD‐1/PD‐L1, TIGIT, LAG‐3, and TIM‐3, contribute to the maintenance of maternal‐fetal immune equilibrium, whereas perturbations in this network may predispose to complications such as recurrent pregnancy loss, preeclampsia, or preterm birth [41, 42].

2.1. HLA Molecules and the Immune‐Tolerance Materno‐Fetal Interface

HLA‐G has five isoforms: HLA‐G1 (complete reading of the primary transcript), HLA‐G2 (without exon 3), which encodes a membrane‐bound isoform associated with beta‐2 microglobulin, HLA‐G3 (without exons 3 and 4), HLA‐G4 (without exon 4), and HLA‐G5 (with intron 4), which encodes a soluble form of the HLA‐G antigen [43].

Notably, HLA‐G molecules play a central role in maternal‐fetal tolerance [42, 44, 45, 46, 47], contributing to both the maintenance of immunosuppression and the remodeling of spiral arteries. Its expression is concentrated in EVT cells, cells with a high adhesive and invasive capacity, justifying its use as a marker in trophoblastic neoplasia [48]. The initial physiological hypoxia of placentation constitutes a potent inducer of HLA‐G. In contrast, prolonged hypoxia can activate inhibitory factors that limit its expression, suggesting a control mechanism in trophoblast proliferation and invasion [49].

At the maternal‐fetal interface, uNK cells represent the predominant leukocyte population in early stages. These cells, attracted by endometrial chemokines, surround the spiral arteries and establish a close dialogue with the ETV. These interactions regulate trophoblast invasion dually: They promote migration and survival via IL‐8 and matrix metalloproteinase 9 (MMP‐9), but also restrict excessive proliferation via TNF‐α, TGF‐β, and IFN‐γ. HLA‐G molecules, in turn, interact with inhibitory NK receptors (LILRB1, KIR2DL4), reducing cytotoxicity and inducing IL‐10 secretion. Furthermore, phenomena such as HLA‐G trogocytosis confer an immunosuppressive phenotype on NK cells without requiring NK cell gene expression, thereby reinforcing maternal‐fetal tolerance [50]. In addition, HLA‐G molecules exert immunomodulatory effects by inhibiting the activity of T cells, dendritic cells, and decidual macrophages. These immunomodulatory characteristics have led to the investigation of this molecule in other fields, such as the prevention of acute graft rejection in solid‐organ transplantation [51, 52], hematopoietic progenitor cell transplantation [53] and the regulation of autoinflammatory diseases [54]. In addition, its expression has been studied as a mechanism of immune evasion in the tumor microenvironment [55]. Of note are the studies by Morandi et al., Tronik‐Le Roux et al., and Benitez Fuentes et al. that propose HLA‐G blockade as a therapeutic target for future immunomodulatory drugs in solid organ neoplasms [56, 57, 58]. HLA‐F and HLA‐E, along with HLA‐G, interact with uNKs, macrophages, and CD8+ T cells, activating inhibitory signaling pathways that protect the fetus from the maternal immune response and promote placentation and EVT migration. It should be noted that HLA‐E expression is limited to the first trimester of pregnancy, while HLA‐C, HLA‐F, and HLA‐G expression progressively decrease throughout pregnancy [59, 60].

2.2. Immune Cells and Tolerance at the Interface Materno‐Fetal

On the other hand, innate immune response cells are reprogrammed at the decidual level. NK cells adopt a reduced cytotoxic capacity phenotype, secreting cell growth factors and cytokines that contribute to tissue homeostasis [9, 61, 62]. In addition, recent studies have identified a unique subset of decidual NK cells “trained by pregnancy” in multiparous women. These cells have an increased capacity to produce IFN‐γ and VEGF, suggesting that they promote successful embryo implantation in subsequent pregnancies, especially when the father is the same [63]. In addition, macrophages are polarized toward the M2 phenotype. They are characterized by the production of anti‐inflammatory cytokines, uterine tissue remodeling, and the release of the enzyme IDO, which reduces cytotoxic T cell activation and promotes the recruitment of regulatory T cells by modulating the tryptophan metabolic pathway [8, 64]. For their part, dendritic cells exhibit an attenuated capacity to activate T lymphocytes because the antigen‐presenting model is restricted to indirect allorecognition [65]. Decidual stromal cells actively participate in generating a placental microenvironment suitable for gestation by cooperating with the maternal hormonal axis through chemokine production [66]. Some authors even report that, in a murine model, stromal cells can locally inhibit the recruitment of effector Th1 lymphocytes by epigenetically silencing CXCL9 and CXCL10 expression [67]. Finally, regulatory T cells are the main players in maintaining maternal‐fetal tolerance, thanks to their ability to block the activation of effector T cells and their anti‐inflammatory action [4, 5]. From the first contact with paternal antigens during conception, regulatory T cells are activated. Subsequently, fetal microchimerism, the transfer of fetal cells into the maternal circulation, perpetuates the recruitment of regulatory T cells. It should be noted that the percentage of fetal‐specific regulatory T cells increases with each successive pregnancy of the same father, suggesting an immunological advantage conferred by the first child to its siblings [12, 68, 69].

3. Mechanisms of Maternal‐Fetal Tolerance at the Systemic Level

Inevitably, throughout pregnancy, fetal cells and placental debris are released into the maternal circulation. Therefore, the maternal systemic immune response undergoes a series of tolerogenic adaptations to preserve fetal viability. Porret et al. describe that fetal cells circulate in maternal blood until they reach secondary lymphoid organs, where paternal antigens are presented by maternal antigen‐presenting cells (APCs) and recognized by maternal naive CD4+ and CD8+ T lymphocytes (indirect allorecognition). A visual graph of the fetal antigen recognition model is shown in Figure 1. In this figure we can see how the classic response to unknown fetal proteins involves the activation of follicular T lymphocytes (blue) and their collaboration with B lymphocytes, through the T‐dependent pathway, which culminates in the production of incompatible anti‐paternal HLA antibodies, carried by the fetus, with maternal antigens, among others, and the generation of memory effector lymphocytes (purple).

FIGURE 1.

Model of fetal antigen recognition by the maternal immune system. First, fetal antigens escape from the decidua and extravasate into the lymph nodes, where they are presented to the mother's immune cells. Several cell types participate in the presentation of paternal antigens and produce anti‐paternal antigen antibodies that are incompatible with maternal antigens. Figure developed by the author through BioRender.com.

However, this model describes that the presence of paternal alloantigens induces the activation of a pool of T lymphocytes that favors an attenuated immune response. PD‐1 + CTLA4 + CD8 + T lymphocytes with attenuated effector functions (red), CD4 + Foxp3 + regulatory T lymphocytes (green), and CD4 + FR4 + CD73 + anergic T lymphocytes (yellow) (Figure 1). This mechanism of natural systemic tolerance could be of interest in other fields such as solid organ transplantation, since pregnant women often have reduced access to transplantation due to the presence of anti‐paternal HLA antibodies in serum. If this adaptation of the T response is maintained over time, these women would be less likely to develop acute graft rejection, since their antigen‐specific lymphocytes would be hypofunctional and accompanied by reduced immunological memory [18]. In contrast, Higgins et al. They show that preformed donor‐specific antibodies (DSA) secondary to pregnancy respond more intensely than DSA from previous transplants in renal transplant patients [70]. On the other hand, Tafuri et al. describe similar adaptations in the hypofunctional maternal T response [71], and it has even been shown that the presentation of paternal antigens via an indirect route causes clonal deletion of CD8+ T lymphocytes, thereby preventing their cytotoxic capacity [65]. In addition to the above, LeMaoult et al. postulate that the HLA‐G1 molecule can regulate the immune response by interacting with CD4+ T lymphocytes. This hypothesis was studied by transfecting APCs with the HLA‐G1 gene; upon interaction with CD4+ T lymphocytes, these APCs acquire an anergic profile, induce proapoptotic signals, and favor their conversion to regulatory T lymphocytes, thereby contributing to the maintenance of maternal‐fetal tolerance [72].

It is worth mentioning that the soluble isoform of HLA‐G (sHLA‐G) has shown immunomodulatory effects at the systemic level. Kotze et al. reported an association between increased sHLA‐G levels and successful fetal implantation [73]. Furthermore, Morandi et al. demonstrated in an in vitro study that sHLA‐G reduces the expression of chemokine receptors on T lymphocytes [74].

Likewise, the B‐cell‐mediated response also undergoes several important modifications. During pregnancy, an increase in regulatory B cells has been detected, especially throughout the first trimester. These cells are characterized by the release of the cytokine IL‐10, which, as previously mentioned, has a potent anti‐inflammatory effect due to the inhibition of TNF‐α and the ability to block antigen presentation to CD4+ T cells by preventing the maturation of dendritic cells, which favors the generation of a suitable microenvironment for fetal development. Furthermore, Ashwin Ajith et al. describe the activation of multiple cell‐signaling pathways mediated by HLA‐G molecules and their effect on the population of regulatory B cells. This interaction favors the proliferation of memory Breg cells and the production of IL‐10 [75]. It is noteworthy that the study of this cell population is limited, as it is not a B lymphocyte subpopulation per se and lacks its own transcription factor [76, 77, 78]. Regarding antibodies, some authors argue that the presence of asymmetric antibodies in plasma and the placenta increases during pregnancy. These antibodies exhibit a structural anomaly, an oligosaccharide residue, that prevents them from activating effector immune mechanisms, such as complement fixation and phagocytosis. However, they maintain their affinity for the antigen, acting as blocking antibodies that could contribute to fetal acceptance [79]. A study reports the presence of anti‐idiotype antibodies from the eighth week of pregnancy, indicating that the maternal immune system can self‐regulate the production of anti‐HLA alloantibodies to prevent fetal damage [79]. To facilitate understanding of the transition from tolerance to pathology, an integrated overview is shown in Figure 2.

FIGURE 2.

Integrated overview of HLA‐dependent mechanisms in pregnancy. (A) At the maternal–fetal interface, EVT expresses classical (HLA‐C) and non‐classical (HLA‐G/E/F) molecules that interact with uterine NK cells and other immune cells to balance tolerance and defense. At the same time, PD‐L1 and apoptotic exosomes restrain T‐cell activity. (B) Systemically, indirect presentation of fetal antigens induces attenuated T‐cell responses, expansion of regulatory T and B cells, and immunomodulatory effects of soluble HLA‐G. (C) Breakdown of these mechanisms through altered HLA expression, dysregulated uNK activity, anti‐HLA antibodies, reduced sHLA‐G, or inflammatory triggers such as cfDNA has been associated with recurrent pregnancy loss, preeclampsia, and preterm birth.

4. HLA System and its Involvement in Fetal Loss

If the mechanisms described above fail, the risk of fetal harm or loss increases considerably. First, studies argue that anti‐HLA antibodies can harm the fetus. Thus, a case‐control study shows that the average fluorescence intensity of anti‐HLA antibodies during pregnancy is associated with an increased risk of premature birth [16]. Premature birth is probably due to maternal antibodies crossing the placenta and harming the fetus through complement activation. Thus, other studies find antibodies in the umbilical cord and C4d deposits in the umbilical vein [80]. However, the meta‐analysis by Lashley et al. concludes that no significant differences are found between the presence of anti‐HLA class I and/or anti‐HLA class II antibodies and the development of gestational complications [81]. On the other hand, it has been described that alterations in the restricted expression of HLA molecules at the maternal‐fetal interface could contribute to gestational loss [17]. Likewise, the role of uNK cells is essential for successful fetal implantation [82], which is why different biomarkers in blood and urine, based on cytokine profiles, have been investigated to enable early prediction of alterations in this leukocyte population [83, 84] In previous sections, the importance of the interaction between HLA‐C [85, 86] and non‐classical HLA‐I [87] molecules with KIR receptors is described, specifically highlighting the interaction between sHLA‐G and KIR2DL4 in NK cells, which promotes remodeling of maternal spiral arteries in the early stages of gestation [88]. If this immunological adaptation does not occur, fetal viability could be severely compromised. In fact, several studies have investigated the relationship between plasma sHLA‐G levels and recurrent pregnancy loss (RPL) [89, 90]. sHLA‐G levels have been observed to increase during pregnancy compared with non‐pregnant women, although they exhibit substantial intra‐individual variability.

Longitudinal studies indicate that women who experience spontaneous abortion exhibit significantly lower sHLA‐G concentrations before the event, even lower than those of non‐pregnant women. However, in non‐pregnant women with a history of RPL, results are inconsistent; some studies show no differences compared with healthy controls, although a trend towards lower sHLA‐G levels is observed in cases of recurrent RPL [91, 92]. According to the literature, autoimmune diseases (e.g., systemic lupus erythematosus or antiphospholipid syndrome) [93] and pharmacological treatments (such as heparin) [94] can alter sHLA‐G levels, so their exclusion from studies is key to avoiding bias. In placental tissue, HLA‐G expression in trophoblasts does not differ significantly between women with and without RPL. However, higher levels have been observed in the decidua in patients with RPL. Furthermore, certain polymorphisms in the 3′ UTR of HLA‐G, such as nucleotide deletions/insertions, have been associated with decreased sHLA‐G production and increased risk of RPL [95, 96]. Other studies, such as those by Pabalan et al. and Nilsson et al., did not observe such an association [97, 98]. Tang et al., on the other hand, argued that hypermethylation of the HLA‐G promoter reduces HLA‐G expression in women with preeclampsia [99]. Taken together, these findings suggest that sHLA‐G could constitute a potential marker of miscarriage risk, although its clinical utility requires further validation.

Finally, studies by Small et al. [100] and Tersigni et al. [101] also link aberrant HLA‐DR expression in syncytiotrophoblasts to preeclampsia in women. This dysregulation of the immune response could be explained by the inflammatory context inherent to the disease, since interferon‐γ induces the expression of the ACII (activator of class II) transcription factor [102]. Although the repercussions of this immune alteration are unclear, it is suspected to affect responses mediated by CD4+ T cells and regulatory T cells. In this sense and to add more points of immune regulation, by orchestrating trophoblast invasion, vascularization, immune tolerance, and tissue remodeling, TGF‐β ligands also contribute to the proper functioning of a healthy maternal–fontal interface. However, dysregulation of TGF‐β signaling has been implicated in the pathogenesis of preeclampsia, in which shallow trophoblast invasion, defective vascular remodeling, decreased uteroplacental perfusion, and endothelial cell and immune dysfunction observed in pre‐eclampsia are all affected by altered TGF‐β signaling, as previously published [103]. Another study demonstrates the role of different lymphoid cell populations in the pathophysiology of preeclampsia [104].

Finally, fetal cell‐free DNA (cfDNA) may play an immune‐regulatory role in pregnancy and in materno‐fetal tolerance. cfDNA is a source of genetic material from the placenta that enters the maternal circulation during pregnancy and can trigger an immune response, potentially contributing to conditions like preterm labor. While cfDNA plays a role in immune regulation in healthy individuals, its presence in a foreign host (the mother) and its structural features, such as CpG motifs and epigenetic modifications, can activate immune cells via pathways like Toll‐like receptor 9 (TLR9). Studies have observed higher levels of cfDNA associated with adverse outcomes and an inflammatory response [105], though a direct causal link to specific pregnancy complications is still being researched [106]. Thus, cfDNA also appears to be essential for regulating the immune system by influencing immune homeostasis and inflammation, and further studies are needed to clarify its role.

To clarify the distinction between mechanistic evidence and associative clinical observations, particularly regarding anti‐HLA antibodies and sHLA‐G levels, and to clarify whether proposed biomarkers (SHLA‐G, cfDNA) could be likely pathogenic mediators or surrogate indicators of placental dysfunction, we will now expand on some interesting clinical studies regarding both parameters. Regarding anti‐HLA antibodies, it has been reported that the thrombocytopenic neonates had a significantly lower adjusted birth weight and significantly higher odds of being small for gestational age compared with controls, and increasing anti‐HLA class I antibody levels in the mother were significantly associated with lower birth weight and placental weight among thrombocytopenic neonates, but not among controls. These results indicate that maternal anti‐HLA class I antibodies in thrombocytopenic neonates are associated with reduced fetal growth [107]. However, other studies do not support the concept that maternal anti‐HLA class I antibodies represent a risk factor of foetal and neonatal alloimmune thrombocytopenia [108]. These points should be clarified with more studies in the future.

Regarding sHLA‐G, there is also a matter for debate. Firstly, in a study, no significant difference in sHLA‐G trajectories during pregnancy was observed between mothers delivering Small for gestational age (SGA) and those delivering appropriate for gestational age (AGA) [109]. A trend towards higher sHLA‐G levels at the first trimester of pregnancy (< 14 weeks of gestation) was observed in mothers delivering SGA neonates as compared to those delivering AGA neonates, indicating that during pregnancy, sHLA‐G trajectories do not differ significantly between mothers delivering SGA and those delivering AGA neonates. However, sHLA‐G levels tend to increase during early pregnancy in mothers delivering SGA neonates.

In other intesting article, sHLA‐G may be important for placental function [110]. In pregnancy, plasma sHLA‐G was significantly lower both in the early‐ and late‐onset preeclampsia groups compared to controls. In women with preeclampsia or gestational hypertension, sHLA‐G was inversely correlated with serum sFlt‐1. Postpartum, plasma sHLA‐G levels were significantly higher in women who had had early‐onset preeclampsia compared to controls. Unexpectedly, sHLA‐G was elevated up to 3 years after early‐onset preeclampsia, suggesting an excessively activated immune system following this severe preeclampsia form, potentially contributing to future cardiovascular disease risk.

In this sense, another study notes that normal physiological pregnancy is also associated with elevated serum concentrations of the sHLA‐G molecule [111]. The increased concentration of sHLA‐G in mid‐gestation could suggest a role for the protein in the second phase of a physiological invasion by extravillous cytotrophoblasts into spiral arteries. Furthermore, the results could suggest a role for decreased serum sHLA‐G concentrations in the pathogenesis of pre‐eclampsia, in a manner like that reported in the previous article.

To shed some light on this or increase the shadows, a recent meta‐analysis (comparing nine studies with altogether 567 preeclampsia cases and 1132 normal pregnancy controls) by Bhattarai and colleagues [112] reaches the conclusion that although the first‐ and third‐trimester sHLA‐G levels in pre‐eclampsia are significantly lower, given the inconsistent results from the sensitivity analysis, the findings support the need for more studies with larger sample sizes to provide solid ground for pooled evidence on the predictive role of sHLA‐G in pre‐eclampsia.

Finally, a study in women with scleroderma shows that these maintain significantly higher quantities of persisting fetal microchimerism from complete or incomplete pregnancies in their peripheral blood and that peripheral sHLA‐G is at a lower concentration in plasma from scleroderma women compared to healthy women [113]. Women with scleroderma display less sHLA‐G expression independently of the eight HLA‐G polymorphisms tested. This decreased production correlates with higher quantities of persisting fetal microchimerism commonly observed in blood from women with scleroderma. These results shed some light on the contribution of the maternal HLA‐G protein to long‐term persistence of the fetus.

In summary, inflammation can lead to insufficient placental development or dysfunction, or low birth weight. Various innate and adaptive immune cells and mediators have been implicated in the development of placental dysfunction, in which oxidative stress is associated with activation of the maternal inflammatory response. Immune cells such as Treg cells, macrophages, NK cells, and neutrophils are known to play major roles in the pathology of placental dysfunction, but the contributions of additional immune cells, such as B cells and allogenic antibodies, as well as inflammatory cytokines, soluble mediators, and autoantibodies, are now recognized [114]. These points provide an overview of the immune responses involved in the pathogenesis of placental dysfunction, including the roles of innate and adaptive immune cells and mediators.

5. Conclusions

The role of the HLA system in embryo implantation and fetal viability is a complex field of research, with numerous questions still to be elucidated. Current scientific evidence suggests that non‐classical HLA molecules are essential at multiple levels for the maintenance of maternal‐fetal tolerance, with the influence of HLA‐G particularly noteworthy. In parallel, the regulation of uNK cells via their KIR receptors is essential for the success of gestation, while their role in maternal protection against infections is key. This is further compounded by the clonal expansion of hypofunctional T lymphocyte subsets in the presence of paternal antigens, thereby regulating the systemic immune response. And adaptations in the B lymphocyte response, through regulatory B cells and the production of anti‐idiotype and asymmetric antibodies, are also evident during pregnancy. However, the correlation between paternal anti‐HLA antibodies and the onset of pregnancy complications has not yielded conclusive results. All of this highlights the need to continue researching the intricacies of reproductive immunology, as its potential expansion into other fields, such as cancer and organ transplantation, could be extremely useful for developing new immunomodulatory therapies. These points also have important and interesting translational implications and should be priorities for future validation studies, which will help elucidate the mechanisms of placental dysfunction and inform decisions to prevent risks to the mother and the newborn.

Funding

This research was funded by Virgen de la Arrixaca University Clinic Hospital and Biomedical Research Institute of Murcia (IMIB) (Murcia, Spain) within the framework of the PI19/01194 National Research Project funded by the Carlos III Health Institute (Madrid, Spain) and Mutua Madrileña Foundation project AP183152023.

Ethics Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (Ethics Committee) of the Biomedical Research Institute of Murcia (IMIB) (Murcia, Spain) for studies involving humans.

Informed Consent Statement

Not applicable.

Conflicts of Interest

The authors declare no conflicts of interest.

Hita A., Botella C., Galian J. A., et al. “Is the HLA System Really Involved in Implantation, Pregnancy, or Abortion of the Fetus?: Strengths, Weaknesses, and Controversial Points.” American Journal of Reproductive Immunology 96, no. 2 (2026): e70309. 10.1111/aji.70309

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.