The group of inborn errors of immunity (IEI) currently comprises nearly 500 heritable disorders of immune function and/or regulation (1). While individual IEI disorders are rare, IEIs as a group account for a significant disease and financial burden for healthcare systems (2–4). Current estimates are that around 1 in 2000 people are affected by IEI, with more individuals diagnosed due to increasing awareness amongst healthcare providers (5, 6). The International Union of Immunological Societies (IUIS) classifies IEIs into 10 groups according to the part of the immune system predominantly affected or associated features, e.g. into combined immunodeficiencies (affecting T- and B cell function), predominantly antibody deficiencies (affecting primarily B cell function), immune dysregulatory disorders or others (1).

An important milestone was the introduction of newborn screening for severe T-cell lymphopenia/severe combined immunodeficiencies (SCID). SCID is invariably lethal if left untreated, but can be cured with early hematopoietic stem cell transplantation (7). Some IEIs are included in umbrella terms such as common variable immunodeficiency (CVID), while others have been characterized as distinct entities including X-linked agammaglobulinemia (XLA) (8), chronic granulomatous disease (CGD) (9) or activated PI3Kδ syndrome (APDS) (10, 11). These diseases are not being routinely tested for in newborns, although a few countries include testing for XLA. For many patients, there is a significant delay between symptom onset and diagnosis, oftentimes leading to irreversible organ damage (12).

APDS is a very rare disease. It may be caused by point mutations or deletions in one of the two genes encoding the two phosphoinositide 3-kinase δ (PI3Kδ) subunits. Both gain-of-function (GOF) in PIK3CD and loss-of-function (LOF) in PIK3R1 contribute to the clinical phenotype (13), i.e. the two mechanisms are represented in the pathogenesis of APDS. The disease pattern of APDS has been largely deciphered and may serve as an example to explain how GOF and LOF mutations in a signaling pathway cause disease. The spectrum of symptoms and their severity greatly vary among APDS patients. APDS often manifests early in life with a combination of recurrent infections, lymphoproliferation, and autoimmune disease, and - later in life - with deteriorating lung function or lymphoma. However, many patients may be oligosymptomatic. Indeed, even asymptomatic individuals have been described in literature (14). Consequently, the first hurdle in patient care is diagnosis of affected individuals, which requires attentive general practitioners and efficient diagnostic algorithms. The heterogeneity of the APDS patient collective also poses a challenge for choosing the right therapeutic approach. While symptom-oriented therapy may be appropriate for some patients, others develop life-threatening infections or lymphoma. Accordingly, the consideration of whether to proceed with potentially curative hematopoietic stem cell transplantation (HSCT), associated with potential risks, is not straightforward. APDS patients thus have an unmet need for therapeutic options which address immunodeficiency and lymphoproliferation alike with limited side effects.

Whole exome sequencing (WES) and whole genome sequencing (WGS) for genetic testing will soon be within reach of (almost) every patient. This has helped identify distinct gene variants underlying APDS and many other IEIs. Genetic diagnosis may be of considerable importance for a patient’s therapeutic management: opening the possibility for targeted therapies, e.g. pharmacological inhibition of an overactivated enzyme such as PI3Kδ (15, 16).

The aim of this review is to describe the molecular mechanism underlying APDS and update the reader on the current state of diagnosis and treatment.

APDS is a very rare disease with an estimated prevalence of 1-2 affected people per million. The APDS registry of the European Society for Immunodeficiencies (ESID) has registered around 200 patients from Europe and the Middle East (17, 18). The exact number of APDS patients is unknown, as many patients are not registered or not diagnosed as having an IEI. APDS can be subdivided into APDS1 and APDS2, two very similar conditions which are caused by different point mutations or deletions in either of the two genes encoding PI3Kδ subunits: PIK3CD and PIK3R1 respectively (19).

APDS1 and APDS2 are characterized by similar symptoms, although some differences are observed, e.g. for susceptibility to certain infections or the age at onset of lymphoma (19–22). Frequent infections and autoimmunity limit patients’ quality of life while severe infections and lymphoma lower the average life expectancy of APDS patients (21, 23–25). An overview of major signs and symptoms is provided in Figure 1 .

APDS1 is caused by heterozygous GOF variants in the PIK3CD gene (p110δ protein) (10, 30, 47), and APDS2 by heterozygous LOF variants in the PIK3R1 gene (p85α protein) (11, 20). These variants likely occur sporadically but may be passed on in families for several generations or else occur de novo (19). A founder effect is unlikely since APDS1-affected families do not share long-range haplotypes or other markers indicating such an effect (10). APDS inheritance follows an autosomal dominant pattern (20, 30), and is characterized by high penetrance (23). Furthermore, LOF mutations of the phosphatase and tensin homolog gene (PTEN) cause a syndrome with similar characteristics as APDS (APDS-like) (48).

PI3Ks are a family of lipid kinases with important roles in all eukaryotic cells. They synthesize phosphatidylinositol 3,4,5-trisphosphate (PIP3) from phosphatidylinositol-4,5-bisphosphate (PIP2). PIP3 in turn activates downstream proteins like Akt, PDK1 and mTOR. While PI3Kα and -β are active in all cells of the body, PI3Kγ and -δ are mostly - though not exclusively - active in leukocytes. PI3Kδ and its two subunits are highly relevant for the function of the adaptive immune system (49–52). p110δ is catalytically active, while p85α serves as a regulatory subunit (53). Together, these core components of the PI3K/Akt signaling pathway regulate the growth and differentiation of lymphocytes in concert with other molecules ( Figure 2 ). In B cells, p110δ can be activated via the B-cell receptor or via cytokine signals. In T cells, the main activator for p110δ is T-cell receptor signaling. p85α interacts with p110δ in its resting state, inhibiting catalytic activity. Most PI3Kδ-activating variants in PIK3CD affect the C-terminal part of its gene product, primarily amino acid E1021 ( Figures 3A, B ) (10, 52). These variants increase baseline activity by either impairing inhibitory interactions with p85α, or increasing the affinity of p110δ for the plasma membrane, or both (55). Interaction between the nSH2 and iSH2 (SH2 = Src-homology 2) domains of p85α and between the core domain, helical domain and catalytic domain of p110δ is necessary for the inhibition of p110δ in its resting state ( Figure 3C ). Moreover, the p85α iSH2 domain is needed for establishing a solid interaction with the N-terminal p110δ adaptor binding domain (ABD) (49, 55, 56). Virtually all APDS2 patients carry a variant in the splice donor site for intron 10 of PIK3R1, leading to deletion of exon 10 (aa434–475) (11, 52) and to LOF in p85α ( Figures 3A, C ). As a dedicated regulator of p110δ, such LOF effectively results in GOF of the PI3Kδ complex which is the unifying characteristic of all APDS-associated gene variants. Thus, APDS elegantly illustrates how GOF in effector enzymes and LOF in their regulators have similar effects. Accordingly, LOF in regulators with primary inhibitory effects are considered GOF mutations by some authors (13).

Early diagnosis of IEI, including APDS, is essential to define suitable treatments that may delay or prevent irreversible organ damage. Unfortunately, diagnosis in IEI patients is often delayed. For example, analyses of the ESID and United Kingdom Primary Immunodeficiency (UKPID) registries suggested minimum prevalences of 8.8/100000 for the Island of Ireland in 2022, and 4.4/100000 in France in 2017 (57, 58), which are well below the estimated ratio of 1/2000 in the general population. A total of 3100 IEI patients were recorded in the German immunodeficiency register PID-NET in 2019, and 4700 in 2021 (59, 60). In Germany, APDS and other IEIs are most likely underdiagnosed, however, the exact prevalence is unknown. Additionally, some patients may have been diagnosed but have not been reported to the registry (61). As a first step in IEI diagnosis, the treating pediatrician or general practitioner must appraise an unusual pattern of symptoms in their patient ( Figure 4 ). Although so-called “warning signs” of IEI (62) may help identify immunologic disorders, awareness of IEI among doctors may be a limiting factor (63, 64). Such typical warning signs were proposed by the North American Jeffrey Modell Foundation (https://info4pi.org), some of which have been integrated into diagnostic guidelines. Unfortunately, these do not include signs like autoimmunity or lymphoproliferation. Expertise in diagnosing IEIs is mostly confined to physicians specialized in managing primary immunodeficiencies, and diagnosing immune-related disorders may take up to ten years, as is the case with APDS (12, 14, 65, 66), the discovery of the causative gene variants underlying APDS has helped shorten this delay. Given the sparsity and geographical distribution of clinical immunologists, there is a high probability of patients not being directly referred to a specialist. In Germany, only 220 clinical immunologists practice compared to 16.100 pediatricians, 1900 hemato-oncologists, and 900 rheumatologists (67). The General Medical Council of the United Kingdom currently (early 2023) lists 182 immunologists, as opposed to 1954 hemato-oncologists and 1426 rheumatologists (68). Patients with signs of lymphoproliferation may see an oncologist, a gastroenterologist when either enteropathy or failure to thrive are the leading symptoms, or else a rheumatologist when autoimmunity prevails. Oftentimes, patients see multiple subspecialists, and the underlying primary immunodeficiency remains unrecognized for long periods of time.

Centers specialized in IEI will ask individuals with suspected IEI for a detailed history including a family tree. Patients will then undergo a comprehensive physical examination before laboratory assays are initiated. The results of this basic assessment may help decide how likely an IEI is, and whether in-depth analyses are necessary to determine the exact nature of a suspected IEI ( Figure 4 ) (64, 69–71). Phenotypical overlaps are common in IEI as is the case for APDS, which may fit under the umbrella entity of common variable immunodeficiency (CVID), although APDS is probably better classified as a combined immunodeficiency (CID). Only in exceptional cases would the clinical presentation (symptoms and lymphocyte phenotyping) suggest APDS to the experienced immunologist prior to genetic testing.

Current guidelines recommend performing advanced immunological diagnostics of suspected IEI patients prior to molecular testing (69, 70). Genetic sequencing is the accepted confirmatory step for most suspected IEIs. The treating physician may choose between gene panels, where predefined sets of genes are sequenced, or whole exome sequencing (WES). Additionally, whole genome sequencing (WGS) will complement - or even supersede - both panel and exome approaches. Given the heterogeneous clinical presentation of APDS patients, there is a potential risk of selecting the “wrong” gene panel: at worst missing the mutated gene. WES and WGS have become more affordable and are now being covered by some health insurances in developed countries, meaning many specialists may directly choose WES - or, a few, WGS - in search of a monogenic cause of the suspected IEI. It should be noted however, that only a limited number of genes from WES/WGS may be analyzed by geneticists (referred to as panel-in-exome or virtual panel). It is therefore of great value that the treating physician and the geneticist directly communicate before the genetic analysis is finalized to allow for re-evaluation of the sequencing data, and targeted filtering according to the human phenotype ontology terms (e.g. when no pathogenic variant has been identified in a given panel to explain the phenotype) (72). If a specific gene variant is identified, targeted Sanger sequencing may help segregate the variant within the family, possibly identifying other carriers or mildly affected family members.

Thanks to the availability and affordability of gene and genome sequencing, an increasing number of IEIs are being mapped to the underlying genetic causes. The first variant known to cause APDS1, the E1021K variant in p110δ, was first published in 2006 (73), but APDS1 was not described as a distinct disease entity until 2013 (10). Soon thereafter, the aa434–475 deletion in p85α was recognized to cause a similar condition (11). Extensive databases on disease-causing variants, including those associated with IEI are maintained on platforms such as OMIM (www.omim.org), OrphaNET (www.orpha.net), ClinVar (www.ncbi.nlm.nih.gov/clinvar), and DisGeNET (disgenet.org) (74). However, identifying a disease-causing gene variant in the depths of the human genome remains challenging for many patients. For the identification of disease-causing variants, experts cross-check variant databases for neutral variants and use pathogenic variant prediction algorithms (75). Besides the few variants in the PIK3CD and PIK3R1 genes functionally proven to be pathogenic (52), there are hundreds of variants of unknown significance (VUS), whose causality for an immunodeficiency has not yet been determined. Patients with an APDS-like presentation and abnormal PI3K/Akt signalling but without a pathogenic variant in the two APDS genes (e.g. PTEN deficiency) require further investigations. Databases like Franklin (https://franklin.genoox.com) and varsome (https://varsome.com) collect information on the pathogenicity of variants. The online tool LitVar (www.ncbi.nlm.nih.gov/CBBresearch/Lu/Demo/LitVar) offers assistance with literature research for previously reported variants. However, geneticists should not solely rely on published information as the field is constantly evolving and new information is being added to the various databases. In APDS - as is the case generally for rare hereditary diseases - the exchange between experts and the use of common platforms will greatly facilitate linking gene variants to patient phenotypes, and furthermore enrich the currently known spectrum of IEI (76). Genetic tests have the potential of making diagnostics for monogenetic diseases accessible to patients who do not yet benefit fully from modern medicine, for example due to the economic situation in their home country.

A number of variants have been functionally ascertained to underlie APDS, based on the correlation between genetics and symptomatology of patients (10, 11). However, to confirm a direct link between genotype and clinical phenotype, a functional proof of pathogenicity in the affected molecular pathomechanism is crucial. The primary characteristic of APDS-causing variants lies in increased activity of PI3Kδ and downstream proteins, including Akt and mTOR. Confirming a suspected diagnosis of APDS requires matching of the clinical phenotype and a (likely) pathogenic variant. This would leave patients who do not have one of the confirmed APDS variants without a diagnosis, regardless of their phenotype. Therefore in some academic settings, the overactivation of the PI3K/Akt pathway may be investigated in order to characterize the biology of APDS patient cells. This includes analysis of phosphorylation levels of downstream proteins like Akt, FOXO1, and of the ribosomal protein S6 (11, 20, 30). However, these functional tests have not found their way into clinical practice yet, which would require fresh patient samples to be shipped to a designated diagnostic center (30). The phospho-S6 ribosomal protein, however, has been proposed as a predictive marker in sarcoma (77), so it is conceivable that a phospho-S6 assay could be standardized as one possible confirmatory test for APDS. Developing functional testing for VUS in APDS could serve as a proof-of principle for IEIs characterized by GOF or LOF variants in other genes. In the future, functional testing could help assess the efficacy of PI3Kδ inhibitors in APDS and other diseases with abnormal PI3K/Akt signaling, e.g. PTEN deficiency, which could help provide existing therapy options for more patients and to monitor therapy.

The present review has addressed the clinical, immunological, and molecular characteristics of APDS, and discussed the current challenges in the diagnosis and treatment of this complex disease. Based on our review of literature, we have identified the following key challenges in the diagnosis and treatment of APDS:

There is a need for continuing education across specialties caring for IEI patients. Multidisciplinary cooperation is essential in order to a) overcome current underdiagnosis of IEI and b) spread awareness that early and precise diagnosis usually has a direct impact on treatment decisions, including targeted therapies. In light of an increasing number of molecules for targeted treatment of specific IEI, defining the genotype of IEI should be part of diagnostic routine. Still, patients may present with typical signs of APDS but have inconclusive genetics. To characterize so-called VUS, and to identify more variants that result in PI3K pathway hyperactivity, functional testing procedures within reach of specialty clinics need to be validated. Traditional management of APDS is based on antibiotic or antiviral treatment and prophylaxis for recurrent infections, IRT as well as immunosuppressive agents to control symptoms of immune dysregulation (21, 23, 52). Since the current therapeutic options are sometimes insufficient to adequately control symptoms, HSCT as a potentially curative treatment could be an alternative for selected patients. Especially in younger patients at risk for severe course of disease, HSCT can improve immune function, and reduce frequency of infections and severity of lymphoproliferation (85, 86). However, markers predicting severity of disease and individual risk to develop lymphoma are currently missing, so risk vs. benefit assessment remains a true challenge. Large observational studies such as the ESID APDS registry are best suited to identify such markers. The promising results for targeted treatment with leniolisib suggest that targeted therapy could become a viable option in APDS and other monogenetic IEI. Leniolisib is a convincing example of a targeted disease-modifying drug developed for the treatment of a distinct IEI. The drug was approved by the FDA (95) and is awaiting approval by the EMA.

Since both HSCT and targeted therapies have advantages and shortcomings, both approaches need to be discussed depending on a patient’s condition and disease severity. Better understanding of the pathomechanisms underlying other IEIs holds the prospect of making novel or repurposed targeted therapeutics available for patients, either newly diagnosed or with progressive disease. The relative simplicity of their application could also make them a reasonable option for patients living in countries where facilities for stem cell transplantation are not yet available. Targeted treatments may prevent or delay disease progression and contribute to a better quality of life in individuals with IEI.

SV is the first author and was the main person responsible for the literature search and writing of the manuscript text. Furthermore, he was responsible for drafting the outline. VW supported the first author in the conceptualization of the article, contributed content ideas, checked the content for scientific correctness, and assisted in the structuring of the article. CS contributed her expert clinical and scientific knowledge to the content of the article, provided additional literature sources, and helped harmonize the structure of the content. Furthermore, she provided critical feedback on the messaging of the article regarding methods of diagnosis and treatment options for APDS.