NK cell deficiency syndromes: Clinical manifestations and diagnosis

INTRODUCTION — Natural killer (NK) cells are innate lymphocytes that are critical in defense against virally infected cells and in tumor surveillance [1-4]. These cells are considered part of the innate immune system because they do not require previous exposure to foreign, pathogenic, or dangerous antigens to act. NK cells are the original members of a larger family of "innate lymphoid cells" (ILCs), each having particular roles and functions [5,6]. NK cells are the longest studied and best established of all the ILCs.

The importance of NK cells in human health and disease is illustrated by a small number of human diseases in which NK cells are absent or defective. These conditions are primarily characterized by severe, recurrent, or atypical infections with herpes viruses.

The biology of NK cells, the clinical manifestations of isolated NK cell deficiency syndromes, the evaluation of patients suspected to have these disorders, and the differential diagnosis will be discussed here [7]. The management of patients with NK cell deficiency syndromes is presented separately. (See "NK cell deficiency syndromes: Treatment".)

BIOLOGY OF NK CELLS — NK cells are lymphocytes that originate in the bone marrow and comprise 2 to 18 percent of the peripheral blood lymphocyte pool [8-11]. Although low in number, they are the main lymphocytes of the innate immune system in the peripheral blood. NK cells account for a minority of lymphocytes within organs in most cases, although there are important exceptions. In the duodenal epithelia and in the uterine decidua during pregnancy, NK cells account for up to 45 and 70 percent of lymphocytes, respectively [12,13].

Phenotypic characteristics — NK cells do not display the same cell surface markers as other populations of lymphocytes. Specifically, they do not have CD3, CD4, or CD19. Instead, the majority of mature NK cells (although not all) display the following surface molecules [14,15]:

●CD16 – The Fc-gamma-RIIIa receptor for immunoglobulin (Ig)G

●CD56 – A neural cell adhesion molecule

●CD158 – A member of the killer cell immunoglobulin-like receptor (KIR) family, many of which serve as receptors for class I major histocompatibility complex (MHC) molecules.

In standard lymphocyte flow cytometry panels, NK cells are typically characterized as CD3-, CD16/56+. Some laboratories will report CD56+ and CD16+ NK cells separately, although it is still very common for the "either/or" approach to CD16/CD56 to be used. Approximately one-half of NK cells express CD8, although these are CD8 alpha/alpha homodimers, unlike the CD8 alpha/beta heterodimers expressed on cytotoxic T cells.

NK cells can be distinguished from other lymphocytes on electron microscopy by the presence of large cytoplasmic granules containing the enzymes and other molecules used in cytotoxicity (picture 1).

Subsets — The majority of NK cells in the peripheral blood express intermediate levels of CD56 and are referred to as "CD56^dim." A minor population (approximately 6 percent, which trends higher in very young children) has very high expression of CD56 and is referred to as "CD56^bright" [9,11]. However, there is a great diversity of NK cells present in the human body and a number of additional NK cell subsets corresponding specific stages of NK cell development or tissue distribution [11,16]. The two major peripheral blood subsets as reflected by CD56 expression have important functional and developmental differences:

●CD56^dim NK cells are effectors of NK cell cytotoxicity and antibody-dependent cellular cytotoxicity (ADCC). (See 'Mechanisms of killing' below.)

●CD56^bright NK cells are effective at the production of cytokines but do not express the cytolytic machinery and mediate cytotoxicity only weakly. CD56^bright NK cells are considered to be developmentally immature and have been shown to be a precursor to mature CD56^dim NK cells. (See 'Immunostimulation' below.)

Functions — An important function of NK cells in the innate immune system is cellular killing or cytotoxicity. NK cells constantly survey surrounding cells for abnormalities and kill those cells that fail to display various signatures of health, particularly virally infected and malignant cells. Multiple families of inhibitory receptors on the surface of NK cells, including the KIRs, act to restrain the destructive potential of these cells, mostly by recognizing and binding class I MHC on healthy cells. Additional NK cell functions include cytokine and chemokine production and costimulation of other immune cells [17].

Mechanisms of killing — The two primary killing mechanisms effected by NK cells are NK cell cytotoxicity and ADCC. Both mechanisms involve the release of lytic granule contents, which include the pore-forming molecule perforin, as well as cell death-inducing enzymes, such as granzyme serine esterases. NK cell cytotoxicity is the most rapid form of killing and represents an important defensive component of the innate immune system. ADCC involves the NK cell recognizing an immunoglobulin-bound target and, thus, interfaces with the adaptive immune system.

A third distinct type of killing is mediated by interactions between death receptors, such as Fas on a target cell and the cognate ligand expressed on an NK cell (ie, FasL in this example). The ligation of Fas by FasL induces apoptosis in the target cell [18]. The kinetics of this type of NK cell killing are slower than that mediated by perforin.

NK cell cytotoxicity — NK cell cytotoxicity is a rapid process. During an acute viral infection, increased NK cell cytotoxicity is evident within three days [14,19]. By comparison, cytotoxic T cells are not capable of cytotoxicity prior to activation by viral exposure, and, once activated, their numbers and function begin to peak approximately one week later (following clonal expansion of those T cells having a T cell receptor specific for a viral peptide).

NK cells are considered part of the innate immune response because they act through receptors that are encoded in the germline DNA and do not undergo genetic recombination to attain specificity. This distinguishes them from lymphocytes of the adaptive immune system (ie, T, B, and natural killer T [NKT] cells), which recombine DNA following exposure to antigen and generate antigen-specific T cell receptors and immunoglobulins, respectively [20]. In addition, NK cells can mediate the destruction of target cells based entirely upon interactions between ligands expressed by the target cell and cognate receptors present on the NK cell. This activity does not require interaction with the adaptive immune system and is also called "natural" or "spontaneous" cytotoxicity.

NK cells constantly survey surrounding cells for abnormalities and rely upon the detection of various signatures of health to inhibit their potentially lethal functions. Restraining the activity of these cells are multiple families of cell surface inhibitory receptors, of which the KIR family is the best characterized. KIRs bind to the class I MHC molecules found on the surface of most nucleated cells, which identify the cell as "healthy self." NK cell killing is inhibited if class I MHC molecules are present in normal density. However, if class I MHC is deficient or missing, activation receptor signaling overcomes inhibitory receptor signaling, and NK cell functions are triggered. Thus, NK cells detect and can be triggered to destroy cells that are "missing self" [21].

Tumor cells and cells that are infected with certain viruses downregulate surface class I MHC molecules and are therefore recognized by NK cells as nonself and targeted for destruction. However, a variety of viruses have evolved to exploit this mechanism and escape NK cell-mediated defenses [22,23]. For example, some viruses encode for the expression of class I MHC decoy molecules. Others cause selective downregulation of those class I MHC molecules that are most likely to engage cytotoxic T lymphocytes, while having minimal effect on the class I MHC molecules detected by inhibitory KIRs on NK cells, thus allowing evasion of the T cell response and inhibition of NK cell cytotoxicity.

KIRs signal through cytoplasmic immunoreceptor tyrosine-based inhibitory motifs (ITIMs). Upon receptor ligation, tyrosines in these ITIMs are phosphorylated, allowing the recruitment of Src-homology tyrosine phosphatase (SHP). Recruited SHPs dephosphorylate a number of cellular targets to prevent NK cell activation directed at the target cell. The most notable is Vav-1, which is a direct requirement in the signaling leading to NK cell activation and cytotoxicity [24].

In contrast to KIRs, which restrain the killing activity of NK cells, there are numerous other receptors on the surface of NK cells that are capable of recognizing target cell ligands and generating signals that promote NK cell cytotoxicity. They all require surpassing a signaling threshold set by the negative signals arising from appropriately ligated inhibitory receptors. Some activating receptors are potent in their function, while others require the cooperative efforts of multiple receptors to achieve an appropriate signaling threshold. A few of the receptors, such as 2B4 (CD244), NKG2D, and LFA-1 (CD11a/CD18), are involved in activation of NK cells as well as other cell types. Others are specific to NK cells, including the natural cytotoxicity receptor family [25,26]:

●NKp44 (CD336) and NKp46 (CD335), which bind viral hemagglutinin [27]

●NKp30 (CD337), which binds human leukocyte antigen (HLA)-B-associated transcript 3 and H7-B6

These receptors transmit intracellular signals through the immunoreceptor tyrosine-based activation motifs, mobilizing intracellular calcium. Intracellular calcium then enables the fusion and secretion of NK cell lytic granules with the NK cell membrane where it touches the target cell. This contact leads to secretion of lytic granule contents onto the target cell to promote cytotoxicity. A single NK lytic granule is capable of mediating a fatal blow to a target cell, although, more typically, between two and five granules lead to the irreversible death of the target [28]. The cell death induced by an NK cell is a combination of apoptosis and necrosis [29].

The interface between the NK cell and target, as well as the molecular arrangements that occur at that site, are collectively referred to as the NK cell immunologic synapse [30]. The organization, maturation, and function of the NK cell immunologic synapse progresses in a stepwise manner and can be specifically affected by certain genetic diseases of the immune system, most notably the Wiskott-Aldrich syndrome and the congenital hemophagocytic syndromes. (See 'Genetic disorders and other immunodeficiencies' below.)

Antibody-dependent cellular cytotoxicity — A well-characterized function of NK cells is antibody-dependent cellular cytotoxicity (ADCC), which is a variation of NK cell cytotoxicity that cooperates with the humoral immune system to facilitate elimination of targets recognized by specific IgG. CD16 (the Fc-gamma-RIIIa receptor) on the surface of the NK cell binds to the Fc portion of IgG. IgG, in turn, is attached to a cell or organism through its antigen-recognition sites in the Fab portion. This can be through the NK cell recognizing IgG opsonized targets or through multiple NK cell CD16 molecules being pre-loaded with IgG specific for a target. Unopposed, these interactions activate the NK cell, leading to ADCC of the target cell. However, as is the case with other types of cytotoxicity receptors, activation is countered by the presence of inhibitory receptors that keep the cytotoxic functions of the cell in check. ADCC is important to the mechanism of certain biologic therapies, and some cell-targeting monoclonal antibodies utilize NK cell ADCC to eliminate the targeted cell. One example is benralizumab, which targets eosinophils via the interleukin (IL-)5 receptor and uses NK cell ADCC to destroy them [31,32].

Immunostimulation — NK cells, particularly those that are CD56^bright, serve important roles in immune responses and host defense through their ability to produce and secrete various cytokines and chemokines. NK cells are best known for their ability to secrete interferon (IFN)-gamma after activation. This enables them to participate in antiviral responses through the direct and immunostimulatory properties of this cytokine. IFN-gamma produced by NK cells is also useful in activating macrophages and facilitating their respiratory burst.

NK cells also produce tumor necrosis factor and, under certain circumstances, IL-5 and IL-13 (among others). These last two cytokines may facilitate humoral immunity, although they also suggest a possible pathogenic role for NK cells in asthma and atopic conditions, as IL-5 and IL-13 are main drivers of eosinophil development and differentiation.

Relationship to NKT cells — NK cells are distinct from NKT cells. NKT cells are a subset of T lymphocytes. They are CD3+, CD16/CD56+ and rearrange their T cell receptor genes, as do conventional T cells. There is a clinical entity of NKT cell deficiency [33], and NKT cells are also deficient in a variety of other conditions [34]. NKT cells are not discussed further in this review.

Relationship to innate lymphoid cells — NK cells are an innate lymphoid cell (ILC) and are most closely related to ILC1s, although NK cells have distinguishing characteristics of their own. In general, the ILCs participate specifically in particular aspects of immune regulation and targeted host defense. (See "An overview of the innate immune system", section on 'Innate lymphoid cells'.)

CLINICAL FEATURES OF NK DISORDERS — As a group, true NK cell deficiency disorders are rare, with fewer than 100 cases reported in the literature. However, as with all primary immunodeficiencies, now called inborn errors of immunity (IEIs), milder cases are being identified as the ability to make definitive molecular diagnoses improves.

Viral infections — The primary presenting feature in the majority of cases described is severe, recurrent, or atypical infections with herpes viruses and cutaneous wart-causing papilloma viruses.

The most commonly implicated herpes viruses in patients with these disorders include:

●Cytomegalovirus (CMV)

●Varicella-zoster virus (VZV)

●Herpes simplex viruses (HSV) I and II

●Epstein-Barr virus (EBV)

Patients may develop fulminant VZV, CMV, or EBV infection in childhood or adolescence, leading to prolonged hospitalization, critical illness, or even death. Other clinical scenarios include recurrent VZV presenting as repeated primary infections or frequent herpes zoster. For patients with HSV, infections may affect multiple dermatomes, impact internal organs, or require pharmacologic suppression to prevent frequent eruptions. Cutaneous warts caused by papilloma viruses are less commonly seen compared with herpes viruses.

Risk of malignancy — Many IEIs increase the risk of malignancies, and NK cell deficiencies may also have this effect. Patients with GATA2 deficiency develop human papillomavirus-associated cancers [35], but the data in patients with other causes of NK deficiencies are scant, although it held to represent a risk [36].

Natural history — Most of the individuals described have died in childhood or adolescence, unless they underwent hematopoietic cell transplantation, demonstrating the importance of NK cells in defense against specific viral infections. However, some have survived without transplantation, suggesting that there may be a period in the patient's life when NK cell function is most critical. In addition, even within families sharing the same variants, severity can vary dramatically [37]. (See "NK cell deficiency syndromes: Treatment".)

CLASSIFICATION — Isolated NK cell deficiencies are disorders in which the NK cell number or function is either the only or the predominant defect resulting in the clinical immunodeficiency. This distinction is important because the list of identified inborn errors of immunity (IEIs) is expanding rapidly, and some of these disorders include impaired NK cell number or function. In these other disorders, a deficit of NK cells does not account for the patient's clinical presentation or the deficit seems secondary to other immune defects that explain the clinical picture more completely. In a 2022 listing by the International Union of Immunological Societies expert committee, over 75 of the 400+ IEIs identified impacted NK cells [38].

NK cell deficiency syndromes can be divided into two types (table 1) [39]:

Classical NK cell deficiency — Patients with classical NK cell deficiency (cNKD) usually present with severe, recurrent, or atypical herpes infections and papilloma viruses (table 1).

cNKD results from interruption of normal NK cell development or survival. cNKD is characterized by absent or profoundly decreased (defined as ≤1 percent of peripheral lymphocytes) CD3-CD56+ NK cell numbers or a demonstrable developmental abnormality of NK cells [10]. Developmental abnormalities are usually detected as an excess of CD56^bright cells, although there can be aberrations in the expression of other markers relevant to development, such as CD117, CD94, CD16, and the killer cell Ig-like receptors. Patients with low numbers of NK cells may be classified as having cNKD if function is also absent, particularly if the few cells present are all CD56^bright (the type of NK cell that is not involved in cytotoxicity).

Functional NK cell deficiency — Functional NK cell deficiency (fNKD) represents a more diverse diagnosis in which NK cells are present (ie, >1 percent of peripheral lymphocytes) but are persistently deficient in at least one of their functional activities. Patients with fNKD present with recurrent or atypical herpes infections and papilloma viruses, similar to cNKD, although usually less severe (table 1). Because the overall number or percentage of NK cells is not abnormal, fNKD is more challenging to diagnose.

PATHOGENESIS

Classical NK cell deficiency — A small number of variants have been identified in patients with classical NK cell deficiency (cNKD). It is not known what percentage of cNKD cases is attributable to each variant.

Autosomal dominant GATA2 deficiency — An autosomal dominant variant in the gene GATA2, which encodes a widely expressed transcription factor required for maintenance of the stem cell pool, has been identified in some patients with cNKD type 1 (cNKD1). GATA2 deficiency is the only autosomal dominant variant known to cause cNKD.

The most cited case is that of an adolescent female who presented with disseminated and life-threatening varicella-zoster virus (VZV) infection [40]. She had a prior history of recurrent otitis media and intermittent leukopenia but was otherwise well. In late adolescence, she had life-threatening systemic cytomegalovirus (CMV) infection followed by disseminated herpes simplex virus infection. The patient had evidence of normal specific antibody production and lymphocyte responsiveness, as she had high titer IgG against VZV in her serum, and her peripheral blood lymphocytes produced interferon (IFN)-gamma after stimulation and demonstrated proliferation in the presence of VZV. She completely lacked CD56+ or CD16+ lymphocytes and NK cell cytotoxicity, even when her lymphocytes were stimulated in vitro with type I IFN or interleukin (IL-)2. She died during the course of hematopoietic cell transplantation and was posthumously found to have GATA2 haploinsufficiency [41].

A cohort of 18 patients was described a decade later with what appeared to be cNKD in the context of low numbers of monocytes, especially B cells [42]. Many of these patients had essentially no CD56+ NK cells and very low numbers of monocytes (which also express CD56, albeit at low levels). Some of these cases were familial. Severe human papillomavirus (HPV) infections, mycobacterial infections, and fungal infections were reported in 78, 78, and 28 percent of cases, respectively. Variants in the GATA2 gene were subsequently identified [41,43,44]. Some cases do have larger numbers of NK cells present, but they share the unusual characteristic of being deficient in the CD56^bright NK cell subset [41].

However, GATA2 deficiency is a complex disorder that can present in different ways and affect multiple organs, as well as B cells and dendritic cells. All presentations of GATA2 deficiency are grouped together under the classification OMIM#614172. In some patients, GATA2 deficiency is associated with primary lymphedema with myelodysplasia, myelodysplastic syndrome, and acute myeloid leukemia, and these syndromes are not classified as cNKD [45-48]. Patients with these other presentations may have low or normal NK cell numbers [10]. The mechanism linking the presumably pervasive GATA2 transcription factor to the various phenotypes is unclear [36]. In a report of 57 patients with GATA2 variants, 82 percent had NK cell lymphopenia, and 70 percent had severe viral infections at early ages [35].

In summary, cNKD is one of the ways in which GATA2 deficiency can present clinically and immunologically. Other manifestations of GATA2 deficiency are discussed separately. (See "Mendelian susceptibility to mycobacterial diseases: Specific defects", section on 'GATA2 deficiency (MonoMAC syndrome)' and "Familial disorders of acute leukemia and myelodysplastic syndromes", section on 'Familial MDS/AML with mutated GATA2'.)

MCM4 deficiency — Variants in the minichromosome maintenance complex component 4 (MCM4), a DNA helicase that is involved in the unwinding and polymerization of chromosomal DNA, were identified in one large consanguineous family of Irish nomadic descent. The MCM helicase consists of a core set of MCM proteins (MCM2-7) and is required for DNA duplication to occur at replication forks, although how variants in this complex lead to the observed clinical abnormalities is not known [49]. Patients with MCM4 deficiency had short stature, developmental delay, and adrenal insufficiency. One affected family member had Epstein-Barr virus (EBV) lymphoproliferation, and two others had recurrent viral infections [50]. Linkage analysis performed using multiple family members defined 8p11.23-8q11.21 as a susceptibility locus. Subsequently, the genetic variant in these individuals was defined, and there are a total of three consanguineous cohorts [49,51,52]. These individuals lack NK cells but have a small number of CD56^bright cells. They also have a profound somatic phenotype of adrenal insufficiency and growth retardation [49,51]. This disorder is listed in the Online Mendelian Inheritance in Man database (OMIM#609981) and has been termed "cNKD type 2" (cNKD2).

RTEL1 deficiency — A previously healthy 23-month-old girl who presented with severe disseminated VZV infection that resulted in her death was found to have a deficiency of regulator of telomere elongation helicase 1 (RTEL1) [53,54]. She had normal proportions of B and T cells, as well as a proliferative response to T and B cell mitogens and normal immunoglobulin levels. However, NK cytotoxicity was markedly decreased and could not be restored with preincubation with IL-2. RTEL1 deficiency is also a reported cause of Hoyeraal-Hreidarsson syndrome, which presents with dyskeratosis congenita, bone marrow failure, and immunodeficiency [55]. However, the patients with cNKD did not have these clinical disorders. (See "Dyskeratosis congenita and other telomere biology disorders", section on 'Hoyeraal-Hreidarsson syndrome'.)

IRF8 deficiency — Biallelic (ie, affecting both alleles of the gene but not necessarily with the identical variants) variants in the gene for interferon regulatory factor 8 (IRF8) were identified in a family with severe EBV infections, absent NK cell function, and increased CD56^bright NK cells [37,56]. IRFs are transcription factors that help shape the inflammatory response, particularly to viral infections. The clinical disorder was originally noted in three siblings who developed severe and prolonged EBV infection requiring intensive hospitalization [56]. One male died during the acute illness, one female survived but died years later from pulmonary complications, and a third male survived into his fifth decade but with continued viral respiratory infections. This individual had a paucity of CD56^dim NK cells, with a relative expanded presence of CD56^bright NK cells [37]. A fourth sibling was unaffected and had a normal clinical course after EBV infection. Through whole exome sequencing (WES), compound heterozygous variants in IRF8 were identified with appropriate segregation through the family, with the unaffected sibling having only one aberrant allele. Children of the affected surviving male with single aberrant IRF8 alleles also had normal NK cell numbers and function. Through biologic investigations, it was determined that IRF8 is required for normal NK cell maturation and that it activates transcriptional programs central to NK cell development. Individuals with select single damaging variants in IRF8 can have a mild dendritic cell subset abnormality, predisposing them to mycobacterial infections [57]. Through that work, two additional families having biallelic IRF8 variants were identified (one had mycobacterial infection and the other recurrent respiratory infections), and both had decreased total NK cells along with elevated CD56^bright NK cells [37]. In contrast, both of the IRF8 variants in the original family were in a specific domain of the protein (the IRF domain) and did not have obvious dendritic cell abnormalities, suggesting that the lack of dendritic cell abnormalities might be a feature of having IRF domain variants.

GINS1 deficiency — One of the original families clinically and phenotypically reported as having NK cell deficiency was defined as having compound heterozygous variants in the Go-Ichi-Ni-San-1 (GINS1) gene [58]. GINS1 is part of what is known as the CDC45-MCM10-GINS complex, which is required for the function of the MCM helicase. The original clinical description of individuals affected by GINS1 deficiency was a nonconsanguineous French family, in which two children were affected and the older died from severe CMV infection [59]. They had severe growth retardation and facial abnormalities, suggesting a syndromic immunodeficiency. They also had neutropenia and abnormalities in lymphocyte survival, as well as decreased numbers of natural killer T cells, mucosal-associated invariant T cells, and innate lymphoid cells (ILC2s and ILC3s). The surviving sister was previously shown to have impaired lymphoid cell survival in response to IL-2 and IL-15 [60].

Ongoing studies and evaluation through WES identified compound heterozygous GINS1 variants in the proband and also identified three unrelated families having the same variant, also in conjunction with NK cell deficiency. All had viral susceptibility, and most had growth retardation and facial dysmorphism. All patients had very low NK cells, with a selective depletion of the CD56^dim subset. While presumably mechanistically linked to MCM4 deficiency, biologic studies demonstrated that the presence of biallelic GINS1 variants resulted in replication instability and cell cycle abnormalities. Patients with the variants had about 1/10 the GINS1 function of normal individuals. However, the specific role of GINS1 in NK cells remains unclear.

MCM10 deficiency — Compound heterozygous variants in minichromosomal maintenance complex member 10 (MCM10) have been described in a single patient, a 16-month-old with progressive CMV and essentially no NK cells in peripheral blood [61]. One variant was a truncation preventing nuclear translocation, and the second was a missense that reduced protein function. Neither parent had abnormalities, and the patient's other immunological parameters were either normal or near normal. Interestingly, the few NK cells found in the peripheral blood were mostly of the CD56^bright immature variety, mirroring findings in MCM4 and GINS1 deficiencies. This child with MCM10 deficiency did not have adrenal insufficiency or intrauterine growth delay, potentially distinguishing MCM10 deficiency from MCM4 and GINS1 deficiencies. While a variety of clinical interventions were attempted in this case, including hematopoietic stem cell transplantation, the child did not survive. It remains unclear why abnormalities in the MCM complex result in a specific NK cell abnormality, although a variety of studies performed related to this case demonstrated that intact MCM10 function was required for NK cell terminal maturation and resulting function [61]. Separately, the patient's variants were shown to affect telomere maintenance, which could represent a common mechanism among some of these MCM-related NKDs [62].

GINS4 deficiency — Siblings with compound heterozygous variants in the MCM complex member GINS4 were identified after the proband was evaluated for severe CMV, VZV, and recurrent HSV. The proband also had recurrent infections and a history of intrauterine growth restriction. The proband's sister had recurrent localized HSV. Both siblings had low NK cell counts and some neutropenia, but normal T and B cell counts and subsets. The GINS4 variants found in these patients both decreased expression of the protein and led to abnormal protein function. Normal function was experimentally found to be required for NK cell cycle progression and NK cell differentiation. Like other NKD patients with MCM complex abnormalities, there was a relative overrepresentation of CD56^bright immature NK cells among NK cells found in peripheral blood.

Unknown etiology — In a referral program maintained by the author, and in others shared with the author, most cases fitting the diagnosis of cNKD do not have one of the identified variants listed above. When possible, specimens from patients are retested as new variants are described, and unbiased research-level sequencing approaches have been employed. There are also several case reports in the literature in which a specific defect has not been identified [63-66].

Functional NK cell deficiency

FCGR3A defects — In a few patients, a detrimental functional variant in the gene coding for CD16 (Fc-gamma-RIIIa), FCGR3A, has been identified. This is the only known single gene variation to give rise to an isolated functional NK cell deficiency (fNKD). This is informative because it connects a specific cell activity to the expression of a single gene that is relevant mainly to NK cells. CD16 is expressed on the vast majority of NK cells, as well as some phagocytes. The function of CD16 on the surface of phagocytes is reviewed separately. (See "The adaptive humoral immune response", section on 'Opsonization'.)

Several individuals have been reported who are homozygous for a T to A substitution at position 230 in FCGR3A, resulting in the substitution of a histidine at position 66 in the CD16 molecule (referred to as the "L66H alteration") [67-69]. All had recurrent, progressive, or severe herpesvirus infections.

The L66H alteration disrupts the interaction between the distal extracellular domain of CD16 and other NK cell receptors (CD2 in particular). In three patients from different families who were homozygous for the L66H alteration, NK cell cytotoxicity was impaired but antibody-dependent cellular cytotoxicity was normal, suggesting that CD16 plays a role in IgG-independent NK cell function [67,68,70]. The frequency of the CD16 L66H allele appears to be rare in modern genomic databases [67], although it may be quite high in certain populations (eg, White Scandinavian patients) [70].

Unknown etiology — Other individuals appear to have fNKD without an identifiable genetic etiology, as no NK cell-specific defect has been found. These patients also suffered from recurrent viral infections and severe or invasive herpesvirus infections [71-73]. All had otherwise normal immunologic phenotypes and normal demonstrable T cell and B cell function. None had other immunologic disorders that could potentially interfere with NK cell development, survival, or activity. In all cases, the NK cell defect was persistent in serial assessments, and the author is aware of others that fit this description via the author’s referral program.

REFERRAL — Patients with clear evidence of severe, recurrent, or atypical herpes infections should be referred to an expert in either allergy/immunology or infectious diseases and with experience in evaluating the immune system. The initial evaluation includes flow cytometric tests to quantitate NK cells and other lymphocyte subsets, which are widely available, but advanced testing often requires laboratories specializing in immunodeficiency. Referral to a center of expertise should occur when the patient's presentation has been determined to be sufficiently abnormal to an experienced clinician in allergy/immunology or infectious diseases. Several centers have expertise in NK cell deficiencies [74].

EVALUATION AND DIAGNOSIS

When to suspect an NK cell disorder — Evaluation for an NK cell deficiency syndrome should be considered in patients with severe, recurrent, or atypical herpes infections (table 1). In most cases, the possibility of an NK cell defect is raised only after a severe, debilitating, chronic, or potentially irreversible infectious disease. (See 'Viral infections' above.)

Assessing current and past infections — Pathogens of the herpesvirus family that have caused significant current or past infections are most relevant to consider. Current infections are usually identified by culture, polymerase chain reaction (PCR), or antigenic or serologic tests. Methods of evaluating past infections include postexposure titers, in vitro lymphocyte proliferation or cytokine production in response to pathogen or pathogen-derived antigen exposure, and cytotoxic T lymphocyte-mediated cytotoxicity against allogeneic cells infected with the suspected organism. These tests may be available at some advanced centers. Assessing the past exposure history of a potentially NK cell-deficient patient is also helpful in planning future management. (See "NK cell deficiency syndromes: Treatment".)

Initial testing — A complete blood count (CBC) with differential should be performed initially, although this does not provide information about lymphocyte subsets. CBC is usually normal in patients with classical NK cell deficiency (cNKD) or functional NK cell deficiency (fNKD), although some may have low lymphocytes.

The number of NK (CD3-, CD56/16+) cells, as well as T and B cells can be determined with a standard flow cytometry panel performed on peripheral blood.

Other testing that is usually performed before more advanced NK cell studies include lymphocyte responses to mitogens and antigens, immunoglobulin levels (IgG, IgA, IgM, and IgE), and response to protein and polysaccharide vaccines. Abnormalities in any of these studies should prompt further advanced testing. (See "Laboratory evaluation of the immune system".)

Cytotoxicity testing — Cytotoxicity testing of peripheral blood lymphocytes is used to assess NK function. A chromium release assay (using ⁵¹Cr) is the classical and standard method, although a variety of flow cytometry-based target cell death assays can accomplish the same objective. This test assesses overall killing ability by lytic granule secretion (ie, perforin- and granzyme-mediated mechanisms). This testing is available through many commercial laboratories as well as several academic centers [75]. If cytotoxicity testing is abnormal, it should be repeated on three separate occasions, separated in time by at least one month because there are many variables that can compromise the result (eg, damage during shipping of samples, active illness, medications, physiologic stress). External causes of impaired NK function are reviewed below. (See 'Differential diagnosis' below.)

Most cell-based function assays available clinically utilize the K562 erythroleukemia target cell, with the exception of those testing antibody-dependent cellular cytotoxicity, which use the otherwise NK cell-resistant RAJI B cell line opsonized with rituximab.

CD107a upregulation testing is available clinically as a useful surrogate for NK cell cytotoxicity. In this assay, peripheral blood lymphocytes are stimulated with either K562 target cells or a stimulant like PMA and ionomycin and then evaluated for the upregulation of the CD107a protein via flow cytometry. CD107a (also known as LAMP1) is contained inside of NK cell lytic granules and is upregulated on the NK cell surface as the lytic granules are released. Thus, the CD107a assay is a measure of NK cell degranulation, which in most cases is an excellent surrogate of cytotoxicity [76]. However, there are situations in which degranulation (and CD107a upregulation) can be normal but cytotoxicity can be deficient (an example is perforin deficiency, which causes hematophagocytic lymphohistiocytosis), and thus caution is needed so as not to overinterpret.

Algorithmic evaluation of NK cell number and function — An algorithmic approach to the evaluation of a patient with a suspected NK cell disorder is provided (algorithm 1) and additional conceptual detail is found elsewhere [77]. It is important to consider other causes for depressed NK numbers or function at each step, as true NK cell deficiencies are rare, even among IEIs. (See 'Differential diagnosis' below.)

●The number of NK (CD3-, CD56/16+) cells is determined using flow cytometry (table 2). Standard laboratory panels include CD4- and CD8-positive T cells, B cells, and NK cells. To exclude a transient deficiency, at least two repeat assessments should be performed, ideally when the patient is not experiencing major stress and is free of any infections, drugs, and toxins that can affect NK cells. Note that active herpes virus infections can suppress NK cell numbers but not to the very low levels seen in cNKD. (See 'Differential diagnosis' below.)

●Some hospital and commercial laboratories can provide information about NK subsets (CD56^bright to CD56^dim ratio) also. The normal ratio of bright to dim is 1:10. A ratio higher than 1:5 (ie, more than 20 percent bright cells) should be considered abnormal.

●NK cell function is evaluated via cytotoxicity testing, preferentially using a chromium release, or a target cell death-based assay. If cytotoxicity testing is abnormal, it should be repeated on three separate occasions, separated in time by at least one month.

Based on the results of these above tests, the following deficiencies can be diagnosed, which may lead to further analysis (algorithm 1):

●If NK cells were found to be ≤1 percent of lymphocytes and NK cytotoxicity is severely depressed or absent, the patient is diagnosed as having cNKD. Variant analysis of the GATA2, MCM4, RTEL1, IRF8, GINS1, MCM10, and GINS4 genes is appropriate in such patients.

●If NK cells were present but NK cytotoxicity was severely depressed or absent, then fNKD is present. Testing for abnormal CD16 should then be performed. (See 'Testing for abnormal CD16 as a cause of fNKD' below.)

To further assess function, NK cells should be stimulated with interleukin (IL-)2, IL-12, interferon (IFN) alfa (each individually), and then reassayed for cytotoxic activity. This helps determine whether the cytolytic defect is specific to "natural cytotoxicity" or affects activation-induced cytolytic functions. Variant analysis of the FCGR3A gene should also be pursued.

●If NK cells were found to be low or absent but NK cytotoxicity is present, then the patient may have a phenotypically abnormal but functional population. This is occasionally seen in some of the NK cell lymphocytoses, which are considered premalignant conditions. (See "Natural killer (NK) cell large granular lymphocyte leukemia", section on 'Chronic NK cell lymphocytosis'.)

Testing for abnormal CD16 as a cause of fNKD — If NK cell numbers are normal but cytotoxicity is depressed, the patient may have an fNKD. In this setting, an evaluation for the presence of abnormal CD16 should be performed. Patients with the L66H alteration have an altered CD16 protein with an abnormal structure that can be detected by certain anti-CD16 antibodies. The clinician should first ascertain which monoclonal antibody reagents were used to identify the NK cells. Most commercial laboratories in the United States use a combination of the 3G8 anti-CD16 antibody and an anti-CD56 antibody. 3G8 will stain the abnormal CD16. If 3G8 was used to detect CD16, the clinician should request that flow cytometry be repeated with two different anti-CD16 antibodies: B73.1 (which will not stain the abnormal CD16) and 3G8 (or others, if available), in separate fluorescent channels (table 2). If the cells stain B73.1 negative and 3G8 positive, the abnormal variant of CD16 is more likely present and gene-sequence confirmation should be obtained. As discussed previously, most patients with an fNKD do not have the abnormal CD16 protein, and pathogenesis in these patients is unknown. (See 'Functional NK cell deficiency' above.)

Functional assessment following stimulation — If the patient's NK cells are present but functionally deficient, they should be stimulated with IL-2, IL-12, IFN alfa (each individually), and then assayed for cytotoxic activity, to determine whether the cytolytic defect is specific to "natural cytotoxicity" or affects activation-induced cytolytic functions (table 2). In the author's program, IL-2 is used as an initial screen given is lower cost and potent effects. In some patients, the abnormal NK cells can mediate cytotoxicity after stimulation with the various cytokines, while other patient's cells are completely unresponsive despite stimulation. Cells that can mediate killing after cytokine stimulation may be unable to activate normally or complete other cytokine-dependent functions. In addition to providing information about the extent of the underlying defect, testing with IL-2 also helps determine if IL-2 therapy would be a potentially useful therapy for that patient. (See "NK cell deficiency syndromes: Treatment", section on 'Interleukin 2'.)

Determining the mechanism underlying an NK cell deficiency syndrome requires advanced techniques available in specialized and research laboratories, such as activation-induced CD107a expression [78], granule release, conjugate formation, or others (table 2). These tests can better define deficiencies in cytolytic function and differential responses to cytokines. The CD107a assay has been the topic of substantive investigation [79] and is a useful surrogate for cytotoxicity, although it measures degranulation and not killing. Degranulation and killing can be disconnected, such as in the case of perforin deficiency, where CD107a upregulation is normal after activation, but killing is absent.

Further investigations, which are mostly performed on a research basis, can include detailed NK cell phenotype to determine the specific presence or absence of NK cell surface receptors that may be critical for regulation of NK cell functions and functions mediated through specific receptors on NK cells (table 2).

Variant analysis for known genetic defects — If a patient appears to have cNKD, variant analysis of the GATA2, MCM4, RTEL1, IRF8, GINS1, MCM10, and GINS4 genes is appropriate. These are contained on many of the commercially available sequencing panels for immunodeficiency and can also be accomplished using whole exome sequencing (WES), which is commercially available as well. Analysis of FCGR3A should be pursued in patients with an fNKD and is best approached by direct single gene sequencing if the flow-cytometry-based screen is positive.

Panels of tests for various IEIs are commercially available [80,81].

DIFFERENTIAL DIAGNOSIS — The diagnosis of an NK cell disorder is challenging to make with certainty because there are multiple known diseases and drugs that can suppress NK cell numbers or activities. These include acute stress and significant illness, drugs and toxins, other genetic diseases and IEIs, and autoimmune and malignant diseases. These conditions may impair NK cell development, survival, or specific function.

The most common alternative explanations for an apparent NK cell deficiency are the following [82]:

●Incorrect usage or interpretation of laboratory tests. The finding of slightly depressed NK numbers are unlikely to signify a true NK cell deficiency.

●Suppression of NK cell function by prescribed medications. (See 'Drugs and toxins' below.)

●Transient suppression of NK cell function by acute stress or illness [82].

Drugs and toxins — There are a number of medications and toxic exposures that are capable of affecting NK cells [14].

NK cell numbers can be reduced by the following:

●The anti-CD52 monoclonal antibody alemtuzumab (used in B cell chronic lymphocytic leukemia, multiple sclerosis) essentially eliminates NK cells because they express CD52, as do other lymphocytes.

●Systemic glucocorticoids can reduce NK numbers, although not to undetectable levels [83].

NK cell function can be suppressed by several medications. In most instances, cytotoxic activity is inhibited, although the mechanism by which this occurs may be distinct in each case. Examples include:

●Glucocorticoids [83]

●Ifosfamide and other immunoablative drugs [84]

●Alcohol [85]

●Salicylates [86]

●Medications that increase cyclic adenosine monophosphate (such as theophylline) [87]

●Prostaglandins [88]

●Cannabinoids [89]

●Antimalarials [90]

Genetic disorders and other immunodeficiencies — There are a surprisingly large number of genetic diseases and other IEIs that affect NK cells (table 3) [15,49,91-107]. These disorders may impact NK cell development, survival, or specific function. Most of these disorders have recognizable characteristics or immunologic defects of other cell types.

Common variable immunodeficiency (CVID) is the most common IEI that can affect NK cell function. Although only a subset of CVID cases have defects in NK cell function, CVID should be entertained in patients with isolated NK cell deficiency because of its relative frequency in the population [108]. CVID is best distinguished from an NK cell deficiency by the demonstration of hypogammaglobulinemia and impaired specific antibody production. (See "Clinical manifestations, epidemiology, and diagnosis of common variable immunodeficiency in adults".)

In some of these diseases, it is accepted that an NK cell deficiency contributes to the clinical phenotype, such as in familial erythrophagocytic lymphohistiocytosis (a subtype of hemophagocytic lymphohistiocytosis) [109]. In others, such as the radiation hypersensitivity syndromes like Bloom syndrome, the contribution of an NK cell deficiency to the clinical course is unknown.

NK cell numbers are decreased or absent in disorders affecting NK cell development or survival (table 3). An example is severe combined immunodeficiency (SCID) due to variation of the common gamma chain (IL2RG) [110]. This molecule is required for responsiveness to the cytokines needed for NK cell development, including IL-15. One IL2RG SCID patient was reported to have somatic reversion in his T cells but not NK cells [111]. He had a specific NK cell abnormality and had susceptibility to human papillomavirus (HPV), which was cured following hematopoietic stem cell transplantation and reconstitution of his NK cell defenses. This unusual case of a broader IEI affecting NK cells further illustrates how a specific deficiency of NK cells can present clinically. Other disorders in which NK cell numbers are decreased or absent include certain forms of autosomal recessive SCID (due to variants in JAK3 or ADA), Bloom syndrome, and Fanconi anemia. (See "Mendelian susceptibility to mycobacterial diseases: Specific defects" and "Severe combined immunodeficiency (SCID): An overview".)

In contrast, the diseases in which a gene variant affects NK cell function are characterized by relatively normal populations of NK cells. This category is useful biologically, in that genes can be linked to specific functions in NK cells. An example is familial erythrophagocytic lymphohistiocytosis due to variation of perforin [109]. In this condition, NK cells are unable to kill target cells due to a dysfunctional perforin molecule. (See "Treatment and prognosis of hemophagocytic lymphohistiocytosis".)

The other diseases in which NK cell deficiencies may be seen are Chediak-Higashi syndrome, Griscelli syndrome, X-linked lymphoproliferative syndrome, Wiskott-Aldrich syndrome, IL-12 receptor beta-1 deficiency, and ectodermal dysplasia with immunodeficiency [15,91]. Broadly, the list would include any disease that involves the potential for lymphohistiocytosis (ie, the uncontrolled accumulation of activated T lymphocytes and macrophages in many organs) or results in an increased susceptibility to herpesvirus infection or those with a disproportionate susceptibility to herpesviruses and have a wide array of diverse treatments given the underlying pathophysiology and breadth of immunodeficiency [112]. (See "Treatment and prognosis of hemophagocytic lymphohistiocytosis" and "Primary disorders of phagocyte number and/or function: An overview" and "Syndromic immunodeficiencies" and "Combined immunodeficiencies: An overview" and "Mendelian susceptibility to mycobacterial diseases: Specific defects" and "Wiskott-Aldrich syndrome".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Inborn errors of immunity (previously called primary immunodeficiencies)".)

SUMMARY AND RECOMMENDATIONS

●NK cell identification and interferon production – Natural killer (NK) cells are lymphocytes that are part of the innate immune system. They have large cytoplasmic granules (picture 1) and are CD3-, CD16/56+ upon flow cytometric analysis. NK cells produce interferon-gamma, which activates macrophages and facilitates their respiratory burst. (See 'Phenotypic characteristics' above and 'Immunostimulation' above.)

●Normal functions of NK cells – NK cells constantly survey surrounding cells for abnormalities and kill those cells that fail to display various signatures of health, particularly virally infected and malignant cells. Multiple families of inhibitory receptors on the surface of NK cells, including the killer cell immunoglobulin (Ig)-like receptors, act to restrain the destructive potential of these cells. NK cell cytotoxicity is rapid compared with the responses of other lymphocytes and peaks approximately three days after viral infection. (See 'Functions' above and 'Mechanisms of killing' above.)

●Clinical features of NK cell disorders – NK cell disorders are rare and characterized clinically by susceptibility to severe and/or recurrent infection with herpes viruses (especially varicella-zoster virus, herpes simplex viruses I and II, Epstein-Barr virus, and cytomegalovirus) and papilloma viruses (table 1). Many of the patients reported have died in childhood or adolescence, although hematopoietic cell transplantation has been successful in some. (See 'Clinical features of NK disorders' above.)

●Types of NK cell deficiencies – NK cell deficiency syndromes are disorders in which either the number, function, or both of NK cells is abnormal, in the absence of any other immunodeficiency, medication, or other condition known to affect NK cells. These syndromes can be divided into two types: classical NK cell deficiency (cNKD) and functional NK cell deficiency (fNKD) (table 1). (See 'Classification' above.)

●Evaluation and diagnosis – Evaluation for an NK cell deficiency syndrome should be considered in patients with severe, recurrent, or atypical infections with herpesviruses.

•Flow cytometry should be performed to quantify NK (CD3-, CD56/16+) cells. If NK cells are low or absent (≤1 percent of peripheral lymphocytes), testing should be repeated at least once when the patient is free of infection or major stress, to exclude a transient deficiency. (See 'Evaluation and diagnosis' above.)

•If NK cells are low or absent, then the patient's lymphocytes should be tested for NK cell function (by chromium release assay or flow cytometric cell cytotoxicity assay), which would be expected to be accordingly low or absent, confirming the diagnosis of cNKD (algorithm 1). Functional testing usually requires a specialty laboratory (table 2). If NK cells are present, then cytotoxicity testing should be performed to evaluate their function. Absent or severely decreased function is suggestive of fNKD. (See 'Algorithmic evaluation of NK cell number and function' above.)

●Differential diagnosis – A wide and varied differential diagnosis must also be considered because NK cell deficiencies are rare, and medications, infections, and other genetic diseases are more common reasons for altered NK cell number or function. (See 'Differential diagnosis' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges E Richard Stiehm, MD, who contributed as a Section Editor to an earlier version of this topic review.