Recent advances in primary immunodeficiency: from molecular diagnosis to treatment

Introduction

Perhaps more than other medical disciplines, the field of primary immunodeficiency is expanding rapidly thanks to recent advances in sequencing, gene editing tools, and the introduction of new biological drugs and small molecules that target specific checkpoints relevant to immunity, inflammation, and cancer. We have witnessed, first of all, a rapid rise in the number of described monogenic inborn errors of immunity (IEIs), and more than 400 distinct defects were included in the International Union of Immunological Societies (IUIS) classification of 2019¹. This represents an increase of 200 new genes and diseases in 10 years and reveals a paradigm shift from “primary immunodeficiencies” as fundamental defects in the immune response to infection toward the broader concept of “inborn errors of immunity” as a comprehensive group of different phenotypes, including infection, autoinflammation, autoimmunity, allergy, and malignancy^1,2. Indeed, a relatively new concept in IEIs is immune dysregulation caused by the innate components of immunity, in juxtaposition with adaptive immunity-driven autoimmunity. Autoinflammatory disorders are caused by an over-activation of pro-inflammatory cytokines or pathways, mostly components of the inflammasomes. They typically manifest as fevers, skin rashes, systemic inflammation, and variable arthritis and lymphadenopathy, although some forms can also present with immunodeficiency and other clinical phenotypes³. A subgroup of these disorders is represented by interferonopathies, characterized by constitutively activated type I interferon (IFN) responses⁴.

New insight into the pathogenesis of IEIs has introduced targeted treatments next to substitution and symptomatic therapy (immunoglobulin replacement, antimicrobial and anti-inflammatory, or immunosuppressive treatments) on one side and replacement of the flawed immune system by hematopoietic stem cell transplantation (HSCT) or gene therapy on the other². Moreover, the latter have greatly benefited from the new advances in cell and gene manipulation, expanding the number of diseases to be successfully treated and increasing the survival chances of affected individuals⁵. In this review, we focus on some of the most recent advances over the last 3 years in the diagnosis and therapy of human IEIs.

Molecular diagnosis of inborn errors of immunity

Sequencing in general and next-generation sequencing (NGS) techniques in particular are becoming technically more accurate, fast, affordable and therefore widely available to researchers and physicians. Apart from the obvious result of increasing the sheer number of defined IEIs, this rapid expansion has highlighted the phenotypic heterogeneity and genetic pleiotropy of these disorders. On the other hand, many novel IEIs are studied in a single kindred or a small number of kindreds, thus providing little information on the penetrance and phenotype¹. Moreover, the pathophysiology often remains to be unraveled. For example, heterozygous mutations in cell division cycle 42 (CDC42), encoding a small GTP/GDP-binding protein involved in eukaryotic actin cytoskeleton dynamics, cause a wide range of different developmental phenotypes with or without autoinflammation, depending on the affected protein domain^6–12. Patients with biallelic mutations in lariat debranching enzyme 1 (DBR1) have a very rare brainstem viral encephalitis through a defect of cell-intrinsic immunity that is still entirely unexplained¹³. A relevant example of phenotypic heterogeneity is the deficiency of adenosine deaminase 2 (DADA2): originally reported as a small-vessel vasculitis manifesting with polyarteritis nodosa, livedo racemosa, stroke, and mild immunodeficiency, the phenotype has significantly expanded to include pure red cell aplasia, other cytopenias, lymphoproliferation, and lymphoma^14–30.

In at least 10 genes responsible for IEI, both gain-of-function (GOF) and loss-of-function (LOF) mutations have been described, resulting in different biological effects and clinical phenotypes (Table 1)^31–55.

Mutations of signal transducer and activator of transcription 1 (STAT1) are exemplary. Biallelic LOF mutations cause either complete or partial STAT1 deficiency: the first impairs type I and II IFN responses and produces a severe phenotype of mycobacterial and viral susceptibility, which invariably is fatal if not corrected by HSCT; the second has a similar but milder presentation^{46,47,56–60}. Heterozygous mutations with a dominant-negative LOF effect impair mainly type II IFN signaling and cause mendelian susceptibility to mycobacterial disease (MSMD) and salmonellosis^48,57,61. Finally, heterozygous STAT1 GOF mutations affect interleukin-17 (IL-17) immunity and cause autosomal dominant (AD) chronic mucocutaneous candidiasis (CMC), bacterial and viral infections, autoimmunity, and cerebral aneurysms^49,50,57,62. A similar situation can be seen in defects of caspase recruitment domain family 11 (CARD11). Homozygous null mutations in fact cause a profound combined immunodeficiency, heterozygous dominant-negative LOF mutations cause combined immunodeficiency with severe atopic disease, and heterozygous GOF mutations instead cause B-cell expansion with nuclear factor kappa B (NF-κB) and T-cell anergy (BENTA) disease^{51–55,63,64}. These examples of additional phenotypes progressively connected to different mutations in the same genes highlight the risk of labeling heterozygous variants as irrelevant to the observed clinical phenotype without functional testing.

With NGS methods routinely available for diagnosis and with a shift from the fixed gene panel and whole exome sequencing (WES) toward whole genome sequencing (WGS), we also learned that deep intronic variants can be disease-causing^65,66. For example, various deep intronic mutations in UNC13D, underlying familial hemophagocytic lymphohistiocytosis (HLH), are very commonly found in patients of European, North American, or Asian origin^67–71. Similarly, intronic mutations were found in a number of IEI-causing genes: in IL7R and Janus kinase 3 (JAK3) in patients with severe combined immunodeficiency (SCID); in ZAP70, encoding the zeta chain-associated protein kinase, in a child with severe T-cell immunodeficiency; in STAT3 in a patient with hyper IgE syndrome; in the NF-κB essential modulator (NEMO) in two patients with ectodermal dysplasia and immunodeficiency (EDA-ID); in POLA1, encoding DNA polymerase α, in patients with X-linked recessive reticulate pigmentary disorder (XLPDR); in CYBB, encoding p91-PHOX, in patients with chronic granulomatous disease (CGD); in ATM in patients with ataxia-telangiectasia; and in CD40LG in patients with hyper IgM syndrome^72–85. RNA sequencing can be an invaluable tool in the validation of these deep intronic variants but also of synonymous and splice-site variants^75,76,86,87. In particular, it can highlight the partial or complete loss of gene expression in the proband compared with controls, which would be missed by traditional sequencing. The increase in diagnostic rate of rare diseases obtained by WGS versus WES can be as high as 6 to 11%, and the use of tissue-specific RNA sequencing can aid the diagnostic process and help clarify the pathophysiology of the disease^86,87. Interestingly, in the case of monogenic rare diseases, RNA sequencing as the first molecular approach led to a diagnosis in 7 to 17% of cases⁸⁷. No hematological/immunological disorders were among those successfully diagnosed in this cohort, which consisted mostly of patients affected by neuromuscular disorders. A lower success rate in IEIs could also be a reflection of the difficulty of comparing normal controls with patients with significant blood cell abnormalities.

The benefit of diagnostic NGS comes at a cost: the necessity of unequivocal functional validation of new variants, albeit in known IEI-related genes. Still too often the pathogenicity of a given variant is based only on the available in silico prediction tools, which can be misleading. Indeed, the validation of variants of unknown significance represents one of the biggest hurdles in making WES and NGS technologies the norm in clinical practice. This is because of not only the financial cost but also the time and competence required first to analyze and study a variant and afterwards to interpret the results and translate them into clinical treatment and care. More than anything, this calls for collaboration and sharing of data for the benefit of the patients. With limited research resources, the further development of more robust prediction tools/validation models becomes a necessity. Finally, as physician-scientists, we wish to stress the importance of the clinical phenotype as a beacon in the era of so-called unbiased sequencing approaches.

Inborn errors of immunity in other diseases

As previously mentioned, the concept of what constitutes an immunodeficiency is steadily shifting from a focus on defects of hematopoietic immune system components, such as leukocytes and immunoglobulins, to conditions affecting immunity at the organism level⁸⁸. Classic examples are cystic fibrosis and sickle cell disease: the first is a chloride channel defect with primarily pulmonary and digestive manifestations, the second is a red blood cell disease caused by mutations in the β-globin gene, and both have recurrent and life-threatening infections as major symptoms (secondary to functional asplenia in the case of sickle cell disease)^88–93. All cells and tissues exert essential host defense functions that range from physical and chemical barriers to pathogen sensing, cytokine production, and protein secretion and activation of the immune response upon infection, representing cell-intrinsic immunity. Infection and inflammation in fact can arise in several disorders of organs other than the classic hematopoietic immune system. The importance of these cell-specific immune responses is apparent when we consider those IEI characterized by the involvement of a single non-hematopoietic tissue, such as the central nervous system (CNS)-specific lack of resistance against herpesviruses in case of mutations of the Toll-like receptor 3/IFN pathway, or the keratinocyte-restricted susceptibility to beta-papillomavirus infections in epidermodysplasia verruciformis (due to null mutations in transmembrane channel-like protein 6 [TMC6], TMC8 [encoding EVER1 and EVER2, respectively], or CIB1, encoding calcium- and integrin-binding protein 1)^94–104. More generally, inherited defects of the skin or mucosal barriers cause at the very least a detrimental secondary effect on immune protection, as well exemplified by genodermatoses and inflammatory bowel diseases.

Several defects of the skin barrier frequently cause secondary infection of the affected epithelia and can also present with additional features resembling known IEIs¹⁰⁵, such as Netherton syndrome (caused by autosomal recessive [AR] defects in SPINK5), hyper IgE syndrome (caused by defects in STAT3, DOCK8, IL6ST, ZNF341, or phosphoglucomutase 3 [PGM3]), or immunodysregulation polyendocrinopathy enteropathy X-linked (IPEX) syndrome (caused by X-linked FOXP3 defects). Defects in desmoglein 1 (DSG1) or desmoplakin (DSP), structural desmosomal proteins, cause severe dermatitis, multiple allergies, and metabolic wasting (SAM) syndrome, a severe multisystemic disease resembling IPEX syndrome at least clinically^106,107. Recent reports have highlighted the complex clinical and immunological phenotype of this disease, including recurrent sepsis and mucocutaneous HSV-1 infection in a patient and T helper 1 (Th1)/Th17/IL-23 skewing in the skin and Th17/IL-22 skewing in the blood of another patient, which merit further research^108–110.

The immunological effects of glycosylation defects and metabolic diseases have also recently gained interest in the IEI field. Glycosylation is crucial in many mechanisms of the immune response, such as pathogen recognition and cell–cell interaction, and glycoimmunology is a rapidly expanding field of research^111–119. Among the currently known congenital disorders of glycosylation (CDGs) (around 133 heterogenous diseases), 23 show a relevant degree of immune impairment; in 10 of these, immunodeficiency is a prominent trait of the disease¹¹⁸. Two of these CDGs manifest as severe congenital neutropenia caused by defects in Jagunal homolog 1 (JAGN1) and glucose-6-phosphatase catalytic 3 (G6PC3), one is a leukocyte adhesion deficiency type II due to defects in solute carrier family 35 member C1 (SLC35C1), one is a glycogen storage disease type I with neutropenia or neutrophil dysfunction (or both) caused by defects in SLC37A4, and the others show various degrees of lymphocyte and immunoglobulin impairment^{118,120–125}. Additionally, AR PGM3 mutations cause glycosylation defects that lead to atopy, immune deficiency with hyper IgE, autoimmunity, and neurocognitive impairment¹²⁶. Finally, a recent study highlighted the role of impaired glycosylation in the pathogenesis of X-linked immunodeficiency with magnesium defect, Epstein–Barr virus (EBV) infection, and neoplasia (XMEN) disease, caused by hemizygous LOF mutations in the magnesium transporter 1 (MAGT1) gene^127,128. On the other side, the study of differentially glycosylated antibodies and their diverse antigen reactivity will teach us more about the fine tuning of the immune responses by glycosylation¹²⁹.

Targeted therapies for inborn errors of immunity

Recent developments in molecular studies have allowed the identification of several possible targets for specific therapeutic interventions. Targeted therapies comprise monoclonal antibodies (mAbs) and small molecules, such as cytokines or cytokine inhibitors, employed to up- or down-regulate a particular pathway, depending on the need. They can be used instead of or in combination with traditional immunosuppressant/immunomodulating agents, also as a bridge to definitive treatment, such as HSCT or gene therapy. Well-known targeted therapies are rituximab (anti-CD20) to treat autoimmune and lymphoproliferative manifestations; enzyme replacement therapy with pegylated bovine adenosine deaminase (ADA) to treat ADA-SCID; IFN-γ to increase superoxide production in the monocytes/macrophages of patients with CGD and to enhance anti-mycobacterial immunity in patients with defects of the IFN-γ/IL-12R pathway; anti-tumor necrosis factor alpha (anti-TNFα) to treat defects of the immunoproteasome causing an autoinflammatory syndrome; IL-1 signal antagonists to treat inflammasome disorders; and finally mechanistic target of rapamycin (mTOR) inhibitors to control aberrant proliferation of effector T cells in various immune dysregulation disorders, such as in IPEX, or to downregulate increased mTOR signaling, such as in activated phosphoinositide 3-kinase δ (PI3Kδ) syndrome (APDS, or PASLI)^130–136.

The latest molecular defects to be targeted in the context of IEI therapy are PI3Kδ activating mutations, cytotoxic T lymphocyte-associated antigen 4 (CTLA4) haploinsufficiency and the closely related lipopolysaccharide (LPS)-responsive beige-like anchor (LRBA) deficiency, STAT1 GOF mutations, STAT3 GOF mutations, IFN-γ activation in the context of HLH, and activating defects in NLR family CARD domain-containing 4 (NCLR4).

Recent advances in hematopoietic stem cell transplantation and gene therapy

Since they are largely due to intrinsic defects of hematopoietic cells or their descendant, the mature blood cells, HSCT represents the treatment of choice for many IEIs. Thanks to better and faster molecular diagnosis, improvement in conditioning practices, better donor selection and graft manipulation, and development of more successful supportive therapies to guarantee engraftment while fending off infections and graft-versus-host disease (GvHD), the overall survival of immunodeficient patients after HSCT has steadily improved up to over 80%^176–179. What then are the challenges and future perspectives in this field?

Conclusions

In this review, we have tried to provide a timely overview of recent advances in the diagnosis and treatment of primary immunodeficiencies/IEIs. With rapidly evolving molecular techniques as the leading force, the end of progress is not yet in sight. Indeed, it can be expected that we are on the verge of a further breakthrough of knowledge in primary immunodeficiency.

Challenges will lie in making these treatments and diagnostics tools available to as many patients as possible and to tailor them to specific needs. Undoubtedly, primary immunodeficiencies/IEIs as experiments of nature will continue to teach us about the magnificent and still-underestimated complexity of the human immune system.

Abbreviations

AD, autosomal dominant; ADA, adenosine deaminase; APDS, activated phosphoinositide 3-kinase δ syndrome; AR, autosomal recessive; CARD11, caspase recruitment domain family 11; CDG, congenital disorder of glycosylation; CGD, chronic granulomatous disease; CMC, chronic mucocutaneous candidiasis; CNS, central nervous system; CTLA4, cytotoxic T lymphocyte-associated antigen 4; DOCK8, dedicator of cytokinesis 8; EBV, Epstein–Barr virus; FOXP3, forkhead box P3; GOF, gain of function; GvHD, graft-versus-host disease; HLH, hemophagocytic lymphohistiocytosis; HSCT, hematopoietic stem cell transplantation; IEI, inborn error of immunity; IFN, interferon; IL, interleukin; IL6ST, interleukin 6 signal transducer; IPEX, immunodysregulation polyendocrinopathy enteropathy X-linked; JAK, Janus kinase; LOF, loss of function; LRBA, lipopolysaccharide-responsive beige-like anchor; mAb, monoclonal antibody; mTOR, mechanistic target of rapamycin; NF-κB, nuclear factor kappa B; NGS, next-generation sequencing; PGM3, phosphoglucomutase 3; PI3Kδ, phosphoinositide 3-kinase δ; SCID, severe combined immunodeficiency; SLC37A4, solute carrier family 37 member A4; SPINK5, serine protease inhibitor Kazal type 5; STAT, signal transducer and activator of transcription; Th, T helper; TMC, transmembrane channel-like protein; WES, whole exome sequencing; WGS, whole genome sequencing; ZNF341, zinc finger 341