Development and Consequences of RBC autoantibodies: Warm Autoimmune Hemolytic Anemia

Abstract

Autoimmune Hemolytic Anemia (AIHA) is a rare disorder caused by loss of tolerance to red blood cell (RBC) antigens, leading to their destruction by autoantibodies. AIHA can occur as a primary condition or secondary to infections, malignancies, or immune-modulating therapies, such as immune checkpoint inhibitors. This review focuses on the roles of B and T cells in disease initiation and progression of warm AIHA (wAIHA). We discuss recent studies highlighting the importance of dysregulated CD4+ T cells in driving autoreactive B cell responses and autoantibody production and highlight a new role for purinergic signaling in contributing to T cell dysfunction. This dysfunction results in an imbalance between T regulatory cells and proinflammatory Th17 cells, further exacerbating the autoimmune response. Treatment strategies have variable success, with relapse rates up to 50% and mortality in ~11%. As such, we also discuss emerging therapeutic strategies, which may potentially lead to more effective and targeted treatments for this serious condition.

Autoimmune hemolytic anemia

Autoimmune hemolytic anemia (AIHA) is a complex disease driven by pathogenic autoantibodies targeting red blood cells (RBCs), leading to their destruction. AIHA is a collective term for several diseases, which are classified according to the monospecific direct antiglobulin test, the autoantibody class, the optimal temperature of the antigen-antibody reaction, and the absence or presence of underlying disease. In this review, we focus on warm AIHA (wAIHA), which represent ~70% of AIHA cases. In wAIHA, RBC-autoantibodies are mostly IgG and bind optimally at 37°C. wAIHA affects 1–3 per 100,000 individuals annually, with half of cases classified as primary, while others arise secondary to conditions including other autoimmune diseases, viral infections, or immune checkpoint inhibitor (ICPi) therapy. Current treatments show variable success, with relapse rates reaching 50%, highlighting the need for better therapies [1]. Recent advances have improved our understanding of how immune tolerance to RBCs is lost. This review examines the roles of B and T cells in wAIHA, emphasizing mechanisms driving disease initiation and perpetuation.

Central tolerance mechanisms against RBC autoantigens

Central and peripheral tolerance are essential immune mechanisms that prevent autoimmunity by eliminating or inactivating autoreactive lymphocytes in primary lymphoid organs (i.e., bone marrow, thymus) and peripheral tissues. B cell central tolerance in the bone marrow is achieved through deletion, anergy, and B cell receptor editing. Approximately 50-75% of immature B cells in both humans and mice are autoreactive, and up to 90% are either eliminated or rendered nonfunctional by central tolerance mechanisms [2-7]. In line with these findings, studies using mouse models of RBC autoreactivity show that tolerance mechanisms induce a 75-80% reduction of RBC autoreactive B cells [8]. Despite the development of both B cells and RBCs in close proximity, and the existence of central tolerance mechanisms, some RBC autoreactive B cell clones still escape into the periphery. This escape aligns with the observed incidence of autoimmune diseases in approximately 5% of the population [9], and the presence of detectable RBC autoantibodies in 0.1% of all healthy blood donors and up to 8% of hospitalized patients [10 11].

In many cases, T cell help is required for B cell activation and antibody secretion, making T cell tolerance essential for preventing autoimmunity. Within the thymus, T cells are exposed to autoantigens, including tissue-specific autoantigens, through the action of the transcription factor AIRE. Autoreactive T cells are eliminated, rendered nonfunctional, or redirected into becoming regulatory T cells (Tregs), which can prevent autoimmunity by suppressing autoreactive T cells. In humans and mice,~95% of autoreactive T cells are eliminated from the repertoire by these mechanisms. In stark contrast to most autoantigens, studies with mouse models of RBC autoreactivity show that autoreactive T cells do not undergo deletion, T cell receptor rearrangement, or redirection into Tregs, suggesting that RBC-derived autoantigens are unique and are likely not presented in the thymus [12]. Consistent with these data, very few patients with AIRE mutations, who manifest with Autoimmune-Polyendocrinopathy-Candidiasis-Ectodermal Dystrophy (APECED), develop secondary AIHA [4 13 14].

Peripheral T cell tolerance prevents RBC autoimmunity

Using murine models, much has been learned about peripheral tolerance to RBC autoantigens. When RBC autoreactive T cells emigrate from the thymus (i.e., recent thymic emigrants [RTEs]), they retain functionality, proliferating in response to autoantigen exposure and providing help to B cells to elicit RBC autoantibody production [12]. Over a three-week maturation timeframe, RTEs undergo progressive tolerization by upregulating inhibitory receptors (e.g., PD-1, Lag-3, CTLA-4), anergy-associated markers (e.g., CD73+FR4+ co-expression), purinergic signaling molecules (e.g., CD39, Adora2a), and exhaustion-related transcription factors (e.g., EGR2) [12 15]. This also leads to expansion and/or conversion of autoreactive T cells into Tregs. As a result, RBC autoreactive CD4+ T cells persist, but become functionally inert [16]. However, B cell tolerance remains incomplete, as some autoreactive B cells can still be stimulated to secrete RBC-specific autoantibodies [16]. Stringent T cell tolerization serves as a critical checkpoint in preventing RBC autoimmunity, and its insufficiency or reversal may drive autoantibody production.

Human leukocyte antigen (HLA) presentation of autoantigens by antigen presenting cells (APCs) plays a key role in inducing peripheral T cell tolerance. When autoantigens are presented in the absence of inflammatory signals or co-stimulation, autoreactive T cells recognizing these peptides become anergic, undergo apoptosis, or differentiate into Tregs [17 18]. High levels of HLA molecules are present on the surface of most cells; however, RBCs express little or no HLA, although HLA expression has been reported to be upregulated in patients with certain diseases. Because RBCs generally do not have detectable HLA molecules, they likely do not play an active role in presenting RBC-derived autoantigens directly to T cells for tolerance induction [19]. Instead, APCs responsible for RBC clearance (e.g., macrophages) are most likely responsible for inducing T cell tolerance. This indirect pathway means that tolerance to RBC antigens may be less robust than autoantigens expressed on other cell types.

RBC antigens are expressed by glycoproteins, glycolipids, and/or proteins on the RBC membrane, many of which serve critical biological functions beyond RBC structure, function, or blood group compatibility. Although some RBC antigens, such as those in the ABH, Kell, Lutheran, and Duffy blood group systems, are also expressed on other tissues (e.g., endothelial cells), others are highly restricted to the erythroid lineage. Notably, Rh proteins and Rh-associated glycoproteins, Band 3, and Glycophorin A (GPA) are almost exclusively expressed on RBCs. In AIHA, the primary autoantibody targets are in the Rh system, Band 3, and GPA [20-22]. Because RBC-restricted antigens rely heavily on peripheral tolerance, this represents a potential vulnerability in immune regulation, whereas RBC antigens that are also expressed on other tissues are less likely to provoke autoreactive responses.

Th17:Treg CD4+ T cell imbalance promotes AIHA

Tregs play an essential role in preventing autoimmunity by regulating humoral immune responses through immunosuppressive cytokine secretion, metabolic disruption (e.g., ATP hydrolysis), APC modulation, and cytolysis. In contrast, Th17 cells contribute to inflammation and are implicated in the promoting immunemediated diseases, largely through secretion of the proinflammatory cytokine, IL-17. In autoimmune diseases, the balance between Tregs and Th17 cells is often disrupted, with a higher Th17:Treg ratio, resulting in an overactive immune response and autoimmunity (Figure 1).

Figure 1: Tolerance maintenance and failure in the context of AIHA.

Peripheral tolerance prevents autoimmunity by suppressing T cell responses to RBC antigens through (1) upregulation of checkpoint molecules (e.g., PD-1, CTLA-4, and Lag-3) and (2) differentiation into regulatory T cells (FoxP3+ Tregs and Tr1 cells), which express high levels of CD39 and CD73 to convert proinflammatory extracellular ATP (eATP) into immunosuppressive adenosine. Breakdown of these regulatory mechanisms leads to tolerance failure, marked by a Treg:Th17 imbalance. Pathogenic Th17 cells upregulate CD39 and promote B cell differentiation into plasma cells (PCs) and/or long-lived plasma cells (LLPCs) that secrete pathogenic polyclonal IgG antibodies. These antibodies bind to RBCs and facilitate FcγR-mediated RBC-clearance by splenic APCs and/or extravascular hemolysis. Hemolysis releases eATP, amplifying inflammation and perpetuating this pathogenic loop. Trogocytosis and RBC-derived extracellular vesicles (RBC-EVs) may contribute to ongoing hemolysis and disease persistence. Infections and/or immune checkpoint inhibitors (ICPis) may induce secondary AIHA by increasing eATP or disrupting peripheral tolerance mechanisms. Apyrase treatment disrupts the pathogenic loop by hydrolyzing eATP, mitigating AIHA symptoms and disease progression. Other therapeutic strategies include B cell depletion (anti-BAFF/APRIL, CAR T cells), neonatal Fc receptor (FcRn) blockade (pathogenic IgG degradation), and Fostamatinib (inhibition of erythrophagocytosis). Image created with BioRender.

Studies in humans with AIHA show significantly reduced numbers of FoxP3+Tregs and increases in Th17 cells and IL-17 levels, as compared to healthy matched controls [23-25]. Increased Th17 cell frequency and IL-17 levels correlate with higher RBC autoantibodies and severe hemolysis, as indicated by elevated serum lactate dehydrogenase and decreased hemoglobin levels, suggesting that Th17 cell expansion is linked to AIHA disease severity [25]. Patients with AIHA also exhibit elevated levels of IL-6 and IL-21 cytokines, which promote Th17 differentiation and expansion [25-27]. Although some studies report increased levels of TNFα, IL-10, IL-12 cytokines, and IL-8/CXCL8 and IP-10 chemokines, others show no change [23 24 26 27]. These discrepancies may stem from differences in disease stage, severity, or therapeutic interventions at the time of sampling. Nonetheless, recent studies support the role of TNFα signaling in disrupting Treg function [24], showing that FoxP3 protein expression is downregulated in Tregs from patients with AIHA. Indeed, transcriptomic profiling linked this downregulation to upregulated TNFα signaling genes. Thus, these results implicate TNFα as a potential driver of Treg dysregulation, leading to Th17 polarization, increased IL-17 secretion, and, ultimately, AIHA onset.

Similar patterns have been observed in AIHA mouse models [28]. The Playfair and Marshall-Clarke model, where AIHA is induced by immunizing mice with rat RBCs, provides direct evidence of Th17 cell involvement. Other studies also show that mice with AIHA have increased numbers of Th17 cells and increased IL-17 levels [25]. Adoptive transfer of Th17 cells before immunization can enhance RBC autoantibody levels, whereas neutralization of IL-17, or using IL-17-deficient mice, prevent AIHA development. These studies underscore the critical role of Th17 cells, and of IL-17, in driving disease progression.

Other regulatory T cell populations are also implicated in AIHA pathogenesis; follicular helper T cells (TFH) promote B cell activation and antibody production within germinal centers, whereas T follicular regulatory cells (TFR), a subset of FoxP3+Tregs, suppress TFH activity regulating humoral responses [29 30]. Dysregulation of TFH:TFR ratio is associated with autoimmune diseases [31]. In the Playfair and Marshall-Clarke model, mice with AIHA exhibit elevated TFH:TFR ratios along with increased serum IL-6 and IL-21 levels [31].

The role of Tregs in AIHA is complex. Although an imbalance between Tregs and Th17 cells is consistently observed in both human patients and mouse models of AIHA, conflicting data remain. For instance, only 27% of patients with IPEX (Immune dysregulation-polyendocrinopathy-enteropathy-X-linked), a severe autoimmune disease caused by FoxP3 gene mutations that impair Treg development, develop AIHA [32]. In the Playfair and Marshall-Clarke model, Treg depletion with anti-CD25 antibodies before rat RBC immunization increases AIHA incidence from 30% to 90%, suggesting that Tregs are critical for preventing disease onset [33]. However, in HOD-transgenic mice, which express an RBC-specific antigen, anti-CD25 treatment fails to induce AIHA, even when combined with immunization using RBC autoantigens in complete Freund’s adjuvant [34]. This discrepancy may be due to cross-reactivity between rat and murine RBCs, whereby AIHA is induced through epitope spreading and linked recognition [35]. These findings also suggest that redundant regulatory mechanisms, or other immune cell types, may compensate for Treg loss under certain conditions. Notably, most studies in humans and mice focus on FoxP3+Tregs as the primary regulatory population; however, Type 1 regulatory T (Tr1) cells, which significantly contribute to peripheral tolerance by producing IL-10 and TGFβ, are detectable in patients with AIHA [36]. Notably, Tr1 cells, isolated from IPEX patients, exhibit suppressive activity [37]. Thus, disruptions in the number and function of multiple Treg subsets likely contribute to the Th17:Treg imbalance observed in AIHA. Further research is needed to delineate the relative contributions of different Treg subsets in AIHA prevention.

The role of purinergic signaling in Th17:Treg imbalance

The etiology of primary AIHA remains largely unknown due to a significant research gap: most studies focus on disease progression after onset rather than on preclinical events leading to tolerance failure. This limitation arises because patients typically require medical care only after the disease has developed, making it challenging to study early, predisposing, immunologic events. However, studies using mouse models of secondary AIHA have provided significant insight into the breakdown of immune tolerance.

In humans, AIHA secondary to ICPi is the most frequently reported hematologic immune-related adverse event [38 39]. The most-commonly used ICPis target PD-1,CTLA-4, and Lag-3, which are checkpoint molecules that murine models have identified as critical for maintaining tolerance to RBC antigens. Inhibiting these pathways in mice leads to a dramatic loss of tolerance, characterized by Th17:Treg imbalance, increased IL-17 production, and AIHA onset (Figure 1). These studies also revealed a novel CD4+ T cell subset, termed CD39 single-positive (CD39SP) cells, which emerges before T cell dysfunction and RBC autoantibody production. CD39SP cells (i.e.,CD39+CD73-FoxP3-CD25-CD4+) are pathogenic, secrete proinflammatory IFNγ and IL-17, and provide B cell help to promote autoantibody production [15].

FoxP3+ Tregs typically co-express high levels of CD39 and CD73. CD39 is an ectonucleotidase that converts extracellular ATP (eATP) and/or ADP into AMP. CD73 then converts AMP into adenosine. ATP and ADP are proinflammatory molecules and chronically high eATP levels promote inflammation, inhibit FoxP3 expression, and induce conversion of Tregs into Th17 cells, further exacerbating the Th17:Treg imbalance [40]. In contrast, adenosine can be immunosuppressive, promote Treg generation, and prevent T cell activation and proliferation. Thus, the co-expression of CD39 and CD73 facilitates conversion of proinflammatory eATP into immunosuppressive adenosine [41]. CD39SP cells retain the ability to hydrolyze eATP to AMP, but the absence of CD73 prevents further conversion to adenosine, thereby limiting immunosuppressive signaling. While the enzymatic activity of CD39 in these cells remains incompletely characterized, their dominant proinflammatory function is evident and contributes to AIHA pathogenesis. Supporting this ATP-driven mechanism, recent data demonstrate that apyrase, an enzyme that degrades ATP, mitigates ICPi-induced AIHA in mice [15].

Notably, increased numbers of CD39SP cells are detectable in several AIHA mouse models and in patients with AIHA, demonstrating relevance in both idiopathic and secondary disease. Further, when NZB mice, a lupus-prone strain, are treated with ICPis, they exhibit accelerated autoimmune disease, including AIHA, highlighting broader implications for autoimmunity. ICPi-therapy also leads to reduced numbers of FoxP3+Tregs and Tr1 cells, potentially explaining the disease severity and high mortality seen with ICPi-induced AIHA in humans [15]. Taken together, this suggests that FoxP3+ Tregs and Tr1 cells may provide redundant tolerance mechanisms, with either subset alone being sufficient to prevent AIHA.

Given that CD39SP cells are found in both murine and human AIHA, as well as in lupus, it is crucial to identify the source of eATP driving their expansion. For example, RBCs are rich in ATP, and hemolysis could elevate eATP levels. Infection-driven increases in ATP could also lead to the generation of CD39SP cells. When AIHA occurs in patients infected with SARS-CoV-2, it typically develops shortly after infection and is associated with cytokine storm-induced inflammation [42]. Because inflammation, driven by proinflammatory cytokines, can oxidatively damage RBCs, leading to hemolysis and subsequent eATP release [43], this process may establish a positive feedback loop that sustains inflammation and promotes AIHA onset. Given that infections are frequently implicated in cases of secondary AIHA, chronically high eATP levels may be a common thread underlying AIHA initiation [44 45].

AIHA perpetuation: The role of extracellular vesicles and trogocytosis

RBC extracellular vesicles (RBC-EVs) are membrane-bound microparticles released by RBCs, that are involved in inflammation, coagulation, and intercellular communication. Most research has focused on RBC-EVs produced in vitro in stored blood products, linking them to transfusion-related complications. However, emerging evidence highlights the importance of endogenously released RBC-EVs in pathophysiological processes like thrombosis, hemostasis, infection, cancer, and inflammation. In patients with hemolytic disorders, EVs arise in vivo from oxidatively damaged RBCs and platelets, and correlate with increased inflammation, oxidative stress, and thromboembolic complications [46]; elevated RBC-EV levels correlate with the severity of anemia and hemolysis. Similarly, elevated RBC-EV levels in patients with AIHA correlate with anemia severity and hemolysis [47]. RBC-EVs may exacerbate disease progression by promoting inflammation and transferring bioactive molecules to immune cells, potentially contributing to the perpetuation of anemia through intercellular signaling.

Trogocytosis involves the transfer of membrane fragments between cells. This process occurs through two main pathways: adhesion molecule-mediated and IgG-FcγR (Fc gamma receptor) mediated. In adhesion molecule-mediated trogocytosis, cell-cell interactions lead to the exchange of membrane patches, potentially transferring RBC autoantigens to APCs. IgG-FcγR-mediated trogocytosis involves the transfer of IgG-opsonized RBC membrane fragments to FcγR-expressing cells, which can promote RBC removal by intravascular hemolysis or erythrophagocytosis [48 49]. This process may also perpetuate AIHA by continuously exposing the immune system to RBC autoantigens.

New and emerging treatments

As conventional therapies often remain insufficient, novel treatments targeting immune pathways are under active investigation.

Targeting purinergic signaling represents a promising therapeutic strategy for AIHA by reducing proinflammatory eATP levels, which may be elevated due to hemolysis. In an ICPi-induced AIHA mouse model, apyrase treatment significantly alleviated disease severity [15]. Apyrase has also shown protective effects in a colitis model of inflammatory bowel disease [50], supporting its broader anti-inflammatory potential. Several purinergic signaling modulators are currently being evaluated in human clinical trials, laying the groundwork for potential translation to patients with AIHA.

IgG-mediated extravascular hemolysis contributes to AIHA progression and remains a major challenge in AIHA treatment. Novel therapies target the neonatal Fc receptor (FcRn), which regulates IgG homeostasis by preventing lysosomal degradation and extending IgG half-life. Inhibiting FcRn-mediated IgG recycling reduces pathogenic IgG levels, which may help mitigate AIHA severity. FcRn inhibitors, such as nipocalimab (NCT04119050) and orilanolimab (NCT03075878), are currently being evaluated in clinical trials for AIHA.

Another promising approach focuses on fostamatinib, a SYK inhibitor that prevents RBC phagocytosis by monocytes and reduces proinflammatory TNFα production [24]. Ongoing clinical trials are evaluating fostamatinib in relapsed/refractory AIHA (NCT03764618, NCT04138927). Although a Phase 2 study demonstrated efficacy in 45% of patients, Phase 3 results showed a response rate of only 35.6%, failing to meet the specified expected endpoint [51]. These findings highlight the need for alternative therapies or combinatorial treatment approaches to improve response rates in refractory AIHA.

First-line AIHA treatment relies on corticosteroids followed by rituximab, an anti-CD20 monoclonal antibody that depletes B cells but causes general immunosuppression [52]. Despite immunosuppressive agents and B cell targeting immunotherapies, fewer than 40% of patients achieve long-term remission. Relapses and refractory AIHA are associated with long-lived autoreactive splenic plasma cells (LLPCs) and elevated B cellactivating factor (BAFF) levels [53 54]. Elevated BAFF and APRIL (a proliferation-inducing ligand) have been correlated with increased anemia severity and are reduced after glucocorticoid treatment, representing a marker of disease progression [55]. Since B cell depletion is often incomplete, novel therapies targeting LLPCs and BAFF/APRIL signaling pathways are being developed. Ianalumab, an anti-BAFF monoclonal antibody (NCT05648968), and povetacicept, an engineered dual inhibitor of BAFF and APRIL, are currently in Phase Ib trials (NCT05757570). Additionally, given the incomplete response to B cell depletion therapies, chimeric antigen receptor (CAR) T cell therapy is emerging as a potential option for severe or refractory AIHA. A phase I clinical trial is currently recruiting patients to evaluate the safety and efficacy of this innovative treatment in AIHA and Evans syndrome (NCT06231368).

Conclusion

Recent insights into AIHA pathogenesis highlight the pivotal role of T cell dysfunction, particularly an imbalance between Tregs and Th17 cells, in driving disease progression. This dysregulation results from the breakdown of peripheral T cell tolerance mechanisms, which can be facilitated by immune checkpoint inhibitor therapy and infections, and is characterized by a reduction in Treg numbers and/or function, and an expansion of pathogenic Th17 cells. New data from preclinical AIHA models reveal an unexpected role of purinergic signaling in disease perpetuation and show that targeting this pathway with apyrase has therapeutic potential. These studies further identify a novel biomarker (i.e., CD39SP) that predicts AIHA onset and may serve as an indirect marker of treatment efficacy. As AIHA is a complex hematologic disease with limited durable treatment options, apyrase and other emerging therapeutics offer promising avenues for improving patient outcomes.