COVID-19 and Inborn Errors of Immunity.

Inborn errors of immunity (IEI) are a heterogeneous group of disorders affecting immune host defense and immunoregulation. Considering the predisposition to develop severe and chronic infections, it is crucial to understand the clinical evolution of COVID-19 in IEI patients. This review analyzes clinical outcomes following SARS-CoV-2 infection, as well as response to COVID-19 vaccines in patients with IEI.

Introduction

As of April 2022, COVID-19 has spread worldwide infecting more than 500 million individuals and causing more than 6 million deaths. Elderly subjects and individuals with underlying comorbidities, including specific groups of immunocompromised patients, often suffered from severe disease and had worse clinical outcomes (1, 2). For patients with inborn errors of immunity (IEI), some studies have reported contrasting data on infection fatality rates that were either comparable to those in the general population (3–5) or increased, with a higher rate of hospitalization, younger age at death, and prolonged viral shedding (6–10). These discrepancies are mainly due to an assortment bias due to the differences in the nature of IEI included in the studies. Importantly, studies conducted on patients with IEI and life-threatening COVID-19 have revealed how impairment in selected immune genes may predispose a patient to critical SARS-CoV-2 infection and multisystem inflammatory syndrome (MIS) (11–18). In this context, an anti-SARS- CoV-2 vaccination campaign was implemented to protect patients with IEI. However, cellular and humoral responses after standard immunization doses are often suboptimal in this group of patients (19–23) and additional vaccination boosts and preexposure prophylaxis with monoclonal antibodies (moAbs) against the SARS-CoV-2 spike protein are indicated (20, 22, 23). Here we review clinical outcomes after SARS-CoV-2 infection and response to COVID-19 vaccines in patients with IEI.

Patients with IEI and SARS-CoV-2 Infection

Since the beginning of the pandemic, multiple cases of SARS-CoV-2 infection have been reported in patients with a broad variety of IEI. The majority of the patients presented with clinical features of classic acute COVID-19 infection, including respiratory symptoms, pneumonia, ARDS, fever, headaches, anosmia, fatigue, and gastrointestinal symptoms, while a smaller group developed features consistent with multisystem inflammatory syndrome (MIS) 4–6 wk after exposure to SARS-CoV-2 (4, 7, 15, 16, 24). In the next paragraphs, we summarize outcomes reported up to March 2022 in patients with COVID-19 and various IEI.

Combined Immunodeficiencies

We identified in the literature 132 patients with a diagnosis of either severe combined immunodeficiency (SCID; n = 22) or combined immunodeficiency (CID; n = 110) and COVID-19 (3–11, 25–40). The overall mortality in this group was 12% (n = 16) with a total of 27 patients (20%) admitted to the intensive care unit (ICU). In the SCID group, 14 patients had received stem cell therapy (SCT) before infection with SARS-CoV-2 (3, 4, 6–8, 11, 28, 36, 39, 40) and their survival rate was 93% (13/14), with only one death reported in a patient that was infected on day +40 post-SCT (8). By contrast, out of eight untransplanted SCID patients who were infected with SARS-CoV-2, only four survived, with death being primarily due to respiratory failure and ARDS (26). Recently Al-Saud et al. (11) reported of a previously healthy 5-mo-old boy with critical COVID-19 pneumonia requiring mechanical ventilation that after immune workup was diagnosed with T-B + NK SCID due to JAK3 deficiency and was successfully transplanted from a matched sibling donor. This case underlines the importance of genetic evaluation in critical cases of COVID-19 in healthy pediatric patients. SARS-CoV-2 infection has been reported in 11 patients with hyper-IgE syndromes (HIES): 2 patients with PGM3 deficiency (4), 7 with heterozygous dominant negative signal transducer and activator of transcription 3 (DNSTAT3) mutations (4, 8), and 2 without known genetic diagnosis (9, 26). In this group, three adults (2 DNSTAT3 and 1 without genetic diagnosis) with lung disease before SARS-CoV-2 infection died of respiratory failure due to ARDS and shock, while the other nine patients had milder disease and recovered. Despite preexisting lung damage, most patients with DNSTAT3 variants had a relatively good outcome. Interestingly elevated IL-6 levels have been associated with poor survival in COVID-19 (41, 42); thus it is reasonable to postulate that impaired IL-6/STAT3 signaling could dampen the hyperinflammation characteristic of severe COVID-19 and ameliorate disease outcomes in patients with DNSTAT3 variants.

Fatal outcomes due to sepsis and respiratory failure have also been reported in a patient with CID due to STIM1 deficiency and Nijmegen Breakage Syndrome (8, 26). In contrast, three patients with RELB deficiency (3) and two with ARPC1B deficiency had only mild disease (4, 6, 24).

Primary Antibody Deficiencies

The majority of patients with IEI infected with SARS-CoV-2 reported in the literature were affected with primary antibody deficiency (PAD; n = 446) (3–9, 26, 28, 30, 32, 33, 35, 37, 38, 43–68), including 290 with common variable immunodeficiency (CVID; 65% of all PAD), 71 with X-linked agammaglobulinemia (XLA; 16%), and 9 with autosomal recessive agammaglobulinemia (2%). The remaining 76 patients had other B-cell defects, including activated PI3Kdelta syndrome (APDS1: n = 6; APDS2: n = 1), nuclear factor-kappa B subunit 1 (NFKB1) deficiency (n = 2), nuclear factor-kappa B subunit 2 (NFKB2) deficiency (n = 4), TRNT1 (CCA-adding transfer RNA nucleotidyl transferase) enzyme deficiency (n = 1), specific polysaccharide antibody deficiency (SPAD; n = 13), isolated IgA deficiency (n = 17), IgG subclasses deficiency (n = 3), or unspecified antibody deficiency (n = 29). CVID and XLA patients displayed similar ICU admission and mortality rates (15% and 8% respectively); however, the latter group had a significantly lower median age as compared with the CVID group (69). These data caution on the validity of reports early in the pandemic that suggested that XLA patients may not be susceptible to severe COVID-19 disease, because BTK deficiency may decrease virus-driven IL-6 production by monocyte and consequently attenuate the cytokine storm typical COVID -19.

The overall ICU admission rate in the PAD group was 13% with a mortality rate of 7%. Deaths were often due to respiratory failure sometimes associated with septic shock and multiorgan failure, especially in patients with preexisting comorbidities. In particular, a meta-analysis of 88 CVID cases showed that patients with chronic lung involvement have an increased risk for severe COVID-19 in comparison to those without lung diseases. Different from the general population, age and metabolic comorbidities did not represent a risk factor for a severe course in this group of patients (70). Among the other causes of PAD, subjects with APDS (n = 7) and IgA deficiency (n = 17) had the best outcomes with no deaths or need for ICU care, while patients with NFKB2 deficiency had variable clinical severity (4, 37, 43, 57).

Primary Immune Regulatory Disorders

Sixty-nine patients with primary immune regulatory disorders (PIRD) and SARS-CoV-2 infection have been reported (3, 4, 7–9, 15, 16, 28, 29, 35, 71–77). The mortality rate was 16% (n = 11) while the rate of ICU admission was 30% (n = 21), similar to what had already been reported in the meta-analysis by Bucciol et al (78). Thirty-one patients were affected by APECED due to biallelic LOF mutations in the autoimmune regulator (AIRE) gene (8, 71–73, 75, 76). In this group, the mortality rate (16%, n = 5) was comparable to the overall PIRD group while the ICU admission rate (42%, n = 13) was higher. This finding is most likely due to the presence of neutralizing autoantibodies directed against type-I interferons (IFNs) and preexisting autoimmune pneumonitis typical of the disease (71, 72, 75, 79). Respiratory failure was the main cause of death in APECED, but early therapeutic intervention with anti-SARS-CoV-2 mAbs may be beneficial (80).

COVID-19 has been reported in 12 patients with either CTLA-4 haploinsufficiency or lipopolysaccharide-responsive and beige-like anchor protein (LRBA) deficiency, with a mortality rate of 17% (n = 2). One LRBA and one CTLA-4 patient were status post-SCT, 4 and 2 yr before SARS-CoV-2 infection, and both survived (3, 4, 8, 10, 30). Five other patients were affected with X-linked lymphoproliferative syndrome (XLP) 1 (n = 1) or XLP2 (n = 4). One previously healthy child was diagnosed with X-linked inhibitor of apoptosis (XIAP) deficiency after presenting with severe MIS-C (15). Two XLP2 patients died after being infected with SARS-CovV- post-SCT: one was infected on day +6 post-SCT and eventually died due to fungal infection in the context of poor engraftment, while the other developed severe graft-versus-host disease (GVHD) and succumbed to sepsis and HLH (4, 7, 15). The remaining two patients with XLP had mild COVID-19 (9).

Congenital Defects of Phagocytes

Among 34 patients with phagocyte defects reported, 22 had chronic granulomatous disease (CGD) (3, 4, 6, 7, 10, 15, 28, 30, 33, 38, 81). Two patients with CGD required ICU admission and eventually died due to Bulkolderia sepsis and uncontrolled MIS-C, respectively (4, 6). Among the remaining patients with congenital defects of phagocytes, two patients were affected by GATA2 haploinsufficiency (4, 10): both survived but one was hospitalized. One patient had DNAJC21 deficiency with early childhood bone marrow failure and mental disability and was hospitalized due to incomplete HLH but recovered (4).

Innate Immune Defects

Fifty-eight patients with innate immune defects and COVID-19 have been reported (4, 5, 7, 12, 18, 29, 30, 82–88). More than 50% of the subjects in this cohort were admitted to the ICU with a mortality rate of 10% (n = 6). Four patients succumbed to respiratory failure, one patient died of septic shock, and another developed sinus thrombosis, seizures, and multiorgan failure in association with a hyperinflammatory status (87). Worse outcomes in this group of IEI underline the fundamental role of innate immunity in early antiviral response against coronavirus; however, a strong selection bias must be taken into account given that many of the subjects were identified in studies focused on identifying innate immune defects in critically ill COVID-19 patients (12, 18, 88). Zhang et al. (18) identified that around 3.5% of severe COVID-19 cases resulted from IEI in genes involved in the type I IFN signaling pathway (i.e., TLR3, UNC93B1, TBK1, IRF3). Subsequently, TLR7 LOF variants were identified in 21 male subjects with critical COVID-19 with a case fatality rate of 15% (n = 3). Virtually all patients required ICU care despite having few comorbidities before SARS-CoV-2 infection. An impaired signaling in response to specific TLR7 ligands, as well as SARS-CoV-2, leading to reduced production of type I IFN by B cells, myeloid cells, and plasmacytoid dendritic cells was identified as a major pathogenetic factor in this group (12, 88). Of note, two patients with TLR3 and IRF3 deficiency benefit from the administration of subcutaneous pegylated IFNα2 (85). Consistent with a crucial role of the TLR signaling pathway in early responses to SARS-CoV-2, several cases of severe infection have been reported in patients with MyD88 (n = 5) and IRAK4 (n = 3) deficiency, thus supporting the role of these molecules in the control of viral replication in lung tissue as previously shown in a murine model of SARS-CoV-1. Among the five MyD88 cases described, three required ICU care (5, 7, 29).

Six patients with STAT1 GOF were identified (4, 8, 9, 29, 30, 84): five experienced mild disease course, and only one was hospitalized but recovered. This favorable outcome may support the concept that STAT1 GOF mutations may to some degree protect from SARS-CoV-2 infection/severe COVID-19 via the preemptive overactivation of IFN-I signaling, despite their otherwise immune deficient status (89).

Autoinflammatory Disorders

Ninety-one patients with a previous diagnosis of autoinflammatory disease were infected with SARS-CoV-2. The majority of these patients had familial Mediterranean fever (FMF; n = 73) (4, 6, 28, 90–93). In the FMF group, the rate of ICU admission was 5% (n = 4) and the infection fatality rate was 4% (n = 3), lower compared to the overall group infection fatality rate of 6% (n = 5) and ICU admission rate of 7% (n = 6). Of the remaining patients, seven had type I interferon-mediated diseases (IFNopathy): two SAVI (STING-associated vasculopathy with onset in infancy), four CANDLE (chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature) or PRAAS (proteasome-associated autoinflammatory syndrome), and one SAMD9L-associated autoinflammatory disease (SAAD). All patients with IFNopathy had mild disease (94) except for the patient with SAAD and interstitial lung disease that developed severe pneumonia, and systemic hyperinflammation and required mechanical ventilation for 8 days before recovery. Critical disease was also observed in a child with unspecified autoinflammatory disease that died of MIS-C and vasculitis and in another patient with IL-1 receptor antagonist deficiency syndrome that succumbed to respiratory failure (6, 28). In contrast, one child with IL-18-mediated disease, two with neonatal-onset multisystem inflammatory disease (NOMID), and five with Aicardi-Goutieres syndrome (2 treated with JAK inhibitors) experienced mild COVID-19 infection and did not even require hospitalization (4, 5, 10, 94). Finally, one subject with DADA2 deficiency was hospitalized but recovered (8).

Complement Deficiencies

Among 42 reported patients with complement deficiency, 41 had a diagnosis of hereditary angioedema with or without C1inhibitor deficiency (HEA) and 1 patient had C3 deficiency. All subjects suffered from either asymptomatic or mild COVID-19 with no ICU admissions or deaths reported (7, 9, 95). Forty percent of HAE patients developed edema attacks (mainly extremities and abdominal edema, rare laryngeal edema) including one subject that had not suffered from angioedema attacks in the previous 14 yr (95). Sustained complement activation has been associated with negative COVID-19 outcomes (96, 97). In this regard, it is surprising that patients with HAE that present with bradykinin overproduction (which had been proposed as a possible mediator involved in the respiratory complications of COVID-19 infection) display only mild COVID-19 clinical manifestations.

Good Syndrome

Thirteen cases of SARS-CoV-2 infection have been reported in patients with Good syndrome (thymoma with hypogammaglobulinemia) (5–7, 9, 59, 65, 98–100). Most subjects had severe and critical disease, with an ICU admission rate of 62% (n = 8) and a mortality rate of 31% (n = 4). Higher median age in this group of patients should be considered as a confounder. Hyperinflammatory status, septic shock, and respiratory and cardiac failure were the cause of death (6, 7). Neutralizing autoantibodies to type I IFN are common in Good syndrome and likely represent a risk factor for severe COVID-19; however, 2 of the 13 patients lacked these autoantibodies and yet suffered from critical or severe disease (65).

MIS-C and MIS-A

MIS in children (MIS-C) or adults (MIS-A) occurs 4–6 wk after exposure to SARS-CoV-2 (101–105) and manifests with multiorgan disease, with the majority of cases displaying cardiovascular involvement, associated myocarditis (88%), and shock in half of the patients. Gastrointestinal, cutaneous, respiratory, and neurologic involvement is also common. Older age at onset and the severe hyperinflammatory status with overlapping features with HLH represent distinctive features of MIS-C compared to Kawasaki disease (102–105). At least 15 patients with IEI and MIS-C have been described (4, 6, 7, 15, 16) with fatal outcomes in 3 and one-third of the patients requiring intensive care unit care for cardiovascular and respiratory support. These patients were between 1 and 17 yr of age, and 85% of them were males. The genetic diagnosis was known in nine patients (1 IRAK4 deficiency, 1 SOCS1 deficiency, 2 XIAIP, 1 CD40L, 1 Chediak-Higashi, and 3 CGD) while five patients carried a diagnosis of autoinflammatory conditions (6), innate immune defects, and disorders of immune regulation without a specific genetic diagnosis.

Risk Factors and Therapeutic Strategies of SARS-CoV-2 Inborn Errors of Immunity

Specific risk factors among patients with IEI have been associated with worse prognoses. Shields et al. (10) reported that higher age, use of prophylactic antibiotics, and comorbidities such as diabetes, cardiovascular disease, chronic kidney, and lung disease correlate with increased hospitalization and mortality rates. In line with these findings in the study by Meyts et al. (4), all adult patients that died of COVID19 (n = 7) carried significant preexisting multiple organ impairment. By contrast, another multicenter study (9) did not confirm that increased age and body mass index have a negative impact on COVID-19 outcomes; nonetheless, this and other groups convincingly showed increased mortality in patients with low lymphocyte counts (T, B, and NK cells) and low levels of IgG, IgM, and IgA (9, 57, 69).

NIH therapeutic guidelines for immunocompromised patients including IEI include antivirals (in particular, Paxlovid and Remdesivir) and anti-SARS-CoV-2 monoclonal antibodies as primary or secondary prevention of COVID-19. In addition, immunomodulators such as IL-1 inhibitors (e.g., Anakinra), IL-6 inhibitors (e.g., Tocilizumab), and kinase inhibitors (JAK inhibitors and BTK inhibitors) are currently under investigation in clinical trials of patients with severe COVID-19 (106–109). Early treatment with type I interferons to promptly limit viral penetration has been attempted despite the timing of administration being a challenge (85, 110). Convalescent plasma appeared to be safe and effective in multiple patients with IEI; however, caution is required due to the presence of anti-type I interferon antibodies in the plasma preparations (25, 36, 49, 59, 66). Furthermore, adequate IgG levels during immunoglobulin replacement therapy and/or effective immunosuppressive therapies, including monoclonals, that target specific hyperactivated immune pathways dampening underlying inflammation before SARS-CoV-2 infection (i.e., tocilizumab, sirolimus, JAK inhibitors) have been associated with mild disease course (57, 84, 90, 109). This evidence is not definitive given the paucity of data, and the clinical immunologist should be aware that guidelines for patients with rheumatological diseases treated with immunomodulatory medication recommend withholding the immunosuppression during symptomatic COVID-19 up to 2 wk after resolution of symptoms (111). However, in patients with IEI, immunodeficiency and immune dysregulation often coexist; therefore, genetic diagnosis and the patient’s clinical and immunological features should be taken into account to identify the best treatment during SARS-CoV-2 infection.

Anti-SARS-CoV-2 Vaccination in Inborn Errors of Immunity

Although inactivated or nonviable vaccines are generally considered safe in patients with primary immunodeficiencies, the efficacy of these vaccines is variable, mainly depending on the specific IEI and the associated defect in the immune response (112). Of note, many patients with humoral immunodeficiencies can mount a protective T-cell-dependent antibody response to protein-conjugated vaccines (112).

Measured response to SARS-CoV-2 infection in otherwise healthy subjects has revealed a strong interplay between the cellular and humoral adaptative immune systems, with a prominent CD4+ T-cell response, which in turn helps induce antibodies against the Spike (S) and Nucleocapsid (N) proteins of SARS-CoV-2 (113). Similarly, immune response to the available messenger RNA (mRNA) COVID-19 vaccines has been found to induce a significant T-cell and neutralizing anti-S antibody response (114).

As for the studies evaluating the effect of SARS-CoV-2 infection in IEI, the majority of reports focus on COVID-19 vaccination in CVID patients (19–23, 115–124). Table 1 summarizes the data on seroconversion in CVID patients following the COVID-19 vaccine. Of note, CVID patients showed a relative ability to mount a specific humoral response after two vaccine doses, although variable across the studies. Combining the seroconversion data reported in Table 1, anti-Spike antibodies were detected in 196/329 (60%) CVID patients after two vaccine doses. However, these results need to be interpreted with some caveats. All the studies that included vaccinated healthy controls (HC) (19, 22, 23, 116–118) demonstrated a lower magnitude of antibody response in CVID patients compared to HC. In addition, the development of the humoral immune response was slower in CVID than in HC, since most CVID patients needed two vaccine doses to produce virus-specific antibodies. Moreover, the studies that assessed the antibody neutralizing activity (21, 117) showed that only a limited number of CVID patients can develop virus-neutralizing antibodies. Overall, these data suggest that both the quantity and quality of the humoral immune response to COVID-19 vaccination may be suboptimal in CVID subjects. In this regard, the evaluation of SARS-CoV-2-specific B- and T-cell immunity seems to be critically important to analyze vaccine-induced protection in these patients. Recent studies have shown that, in CVID patients, immunization with two doses of mRNA vaccine did not generate Spike-specific classical memory B cells (mB) but atypical mB with low binding capacity to Spike protein (22, 23, 118). Interestingly, the B-cell response in convalescent CVID patients was consistent with that observed in immunized immunocompetent individuals, who generated Spike-specific mB (22), suggesting that natural infection may induce more robust germinal center responses than the mRNA vaccine. Moreover, immunization after SARS-CoV-2 infection generated Spike-specific classical mB with low binding capacity to Spike protein and Spike-specific antibodies in CVID patients (22). This observation that the immune response elicited by SARS-CoV-2 natural infection was enhanced by subsequent immunization underlies the need to immunize convalescent COVID-19 CVID patients after recovery (22). Spike-specific T-cell responses were induced in immunized CVID patients, although with a variable frequency (22, 23, 116, 118) and less efficiently than observed for specific T cells generated after influenza virus immunization (125). For CVID patients, it has been postulated that while the influenza virus vaccine induces specific T cells after multiple exposures to viral antigen, the limited Spike-specific T-cell response may be a consequence of a limited antigenic stimulation by a new pathogen, which has never been encountered before (126). In this context, CVID patients might require additional vaccine doses or combinations of different SARS-CoV-2 vaccines to obtain possible protection. Recently, Gernez et al. (119) assessed the efficacy of a third dose of the COVID-19 mRNA vaccine on the humoral responses in CVID patients. All 10 patients with CVID mounted a virus-specific antibody response, and one-half (n = 5) had functional antibodies that had 50% or greater ACE2 blocking activity, thus showing that a third “booster” dose is required to achieve a better immune response against COVID-19. However, further studies are needed to evaluate the effect of the third vaccine dose on the efficacy and persistence of humoral and cellular responses to SARS-CoV-2.

TABLE 1.

Antibody responses to COVID-19 vaccine in patients with CVID

Country	Seroconversion Data	Factors Associated with Increased Risk of Failure	Major Adverse Events	Reference
USA	6/6 (100%)	NA	None	118
USA	7/8 (87.5%)	NA	None	20
USA	12/15 (80%)*	NA	None (see Table 2)	21
USA	10/10 (100%) after 3 doses†	NA	None	113
Israel	10/12 (83.3%)	Lower response in older pt	None	114
Israel	11/15 (73.3%)	B ≤1% or B <6% and smB ≤2% of B	NA	109
Italy	3/4 (75%)	B <1%	None	116
Italy	11/33 (33%)‡	Low RBD-specific smB	NA	112
Italy	8/34 (23.5%)‡	All CVID patients lacked mB and activated mB with high binding capacity	NA	23
Italy	14/38 (36.8%)‡	smB ≤2% of B; low IgA, IgM	NA	22
Italy	13/14 (92%)‡	NA	None	110
Sweden	28/41 (68.3%)‡	NA	None	19
Wales	43/60 (71.7%)	Low IgA + IgM, low B, ChAdOx1-S recipients	NA	115
Spain	15/18 (83%),‡ 9/18 (50%) neutralizing Ab	B cell lymphopenia; autoimmune/lymphoproliferation	None	111
Austria	15/31 (48.4%)	Low IgG pre-IVIG, low B, low smB, reduced Tmem activation	NA	117

Ab, antibody; B, B cells; CVID, common variable immunodeficiency; NA, not available; IVIG, intravenous immunoglobulin; pt; patients; RBD, receptor-binding domain; smB, switched memory B cells; Tmem, vaccine antigen-specific T-memory cells. *Only 8.3% with neutralizing activity (SARS-CoV-2 ACE2 blocking activity) after 2 doses. †50% with neutralizing activity (SARS-CoV-2 ACE2 blocking activity) after 3 doses. ‡Low magnitude of antibody response compared to healthy controls.

The recently reported ‘‘breakthrough’’ COVID-19 cases in vaccinated individuals, which are the result of both the emergence of new viral variants and waning immunity over time, represent a current health challenge. A recent study has demonstrated that patients with primary immunodeficiencies, together with other immunocompromised patients as organ transplant recipients, have a higher incidence rate of breakthrough infections after two mRNA vaccine doses compared to healthy vaccinated individuals (127).

Another important aspect regarding the studies that assessed the antibody responses to the COVID-19 vaccine in patients with CVID is the definition of the clinical and immunological factors associated with increased risk of seroconversion failure after two vaccine doses. These risk factors are specified in Table 1 and included older patients’ age (120), history of autoimmunity (117), B-cell lymphopenia (including B cells ≤1%; switched memory B cells ≤2% of the total B cells; low SARS-CoV-2-specific memory B cells) (20, 22, 23, 115, 117, 118, 121–123), low levels of IgA and IgM (22, 121), and T-cell lymphopenia (20) with reduced T-memory activation (123).

Besides CVID, data on the immunogenicity of the SARS-CoV-2 vaccine in patients with IEI are few and often limited to anecdotal cases or heterogeneous cohorts. Table 2 summarizes the available literature (19–21, 23, 89, 120, 121, 124, 128). Except for patients with XLA and combined immunodeficiency, most patients with IEI can produce anti-Spike antibodies following two vaccine doses. Moreover, Spike protein-specific T-cell responses evaluated using SARS-CoV-2 IFN-γ release assays or ELISpot assays tested positive in most patients with XLA, demonstrating, as expected, that also these patients can mount SARS-CoV-2-specific T-cell responses. Of note, all the studies showed that the COVID-19 vaccine has a good safety profile with no major adverse events reported in IEI patients. However, because of the increased risk of thrombosis with thrombocytopenia syndrome, the use of the Johnson & Johnson vaccine should be avoided in IEI patients with thrombocytopenia or with anti-phospholipid antibodies (129, 130).

TABLE 2.

Humoral and cellular responses to COVID-19 vaccine in IEI other than CVID*

Country	SARS-CoV-2 Tests Performed	IEI	Evidence of Vaccine Response	Major Adverse Events	Reference
USA	Anti-S Ab; anti-N Ab	XLA (n = 1)	No	None	118
		WAS (n = 1)	Anti-S Ab +
		DiGeorge syndrome (n = 1)	Anti-S Ab +
USA	Anti-S Ab; IGRA	Agammaglobulinemia (n = 2)	IGRA + (2/2, 100%)	Most frequent adverse event was sore arm. No major adverse events (only one patient reported a flare of enteropathy one week after vaccination).	21
		Hypogammaglobulinemia (n = 4)	Anti-S Ab + (2/4, 50%)
			IGRA + (3/4, 75%)
		SpAD (n = 2)	Anti-S Ab + (2/2, 100%)
			IGRA + (2/2, 100%)
		Good syndrome (n = 4)	Anti-S Ab + (0/4)
			IGRA + (1/4, 25%)
		Hyper IgM syndrome (n = 2)	Anti-S Ab + (0/2)
			IGRA + (2/2, 100%)
		CTLA-4 deficiency (n = 1)	IGRA +
		PIK3R1 mutation causing APDS2 (n = 1)	IGRA not performed
		Ataxia telangiectasia (n = 1)	IGRA not performed
		ATP6AP1 gene/immunodeficiency (n = 1)	IGRA +
USA	Anti-S Ab; anti-N Ab	STAT3 DN (n = 26)	Anti-S ab + in	None	20
		APECED (n = 14)
		Other immune regulation disorders (n = 10, including STAT1 GOF, STAT3 GOF, RAG deficiency with autoimmunity, unspecified CID with autoimmunity, recurrent autoimmune cytopenias, MAGT1 deficiency and thymoma)	27/46 patients (58.7%) after 1 dose
		Antibody deficiency (n = 13, including CVID, XLA, and other forms of hypogammaglobulinemia)
		Other forms of IEI (n = 14, including WAS, WHIM syndrome, prolidase deficiency, SASH3 deficiency, CD40L deficiency, RALD, ADA deficiency and FOXN1 deficiency)	63/74 (85.1%) after 2 dose
		Post-HCT patients (n = 6)
USA	Anti-S Ab; anti-N Ab	MAGT1 deficiency (n = 1)	Anti-S Ab + after 2 doses	None	122
Israel	Anti-S Ab; anti-N Ab; neutralizing Ab; specific anti-RBD B cells; ELISpot assay	XLA (n = 4)	Anti-S ab + in 18/26 (70%)	None	114
		Predominantly antibody deficiency (n = 17)
		ALPS-like (n = 2)
		STAT1-GOF (n = 1)	S-peptide–specific T-cell response + in 19/26 (73%)
		STAT3 DN (n = 1)
		C4 deficiency (n = 1)
Sweden	Anti-S Ab	XLA (4)	Anti-S Ab + in 0/4	None	19
		Idiopathic T-cell Lymphopenia (n = 11)	Anti-S Ab + in 10/11 (90.9%)
		Monogenic diseases (n = 9)	Anti-S Ab + in 7/9 (77.8%)
		Other PID (n = 10)	Anti-S Ab + in 10/10 (100%)
Wales	Anti-S Ab	CID without molecular diagnosis (n = 8)	Anti-S Ab + in	NA	115
			0/8
		SpAD (n = 8)	7/8 (87.5%)
		DiGeorge syndrome (n = 4)	4/4
		XLA (n = 3)	0/3
		STAT1 GOF (n = 2)	1/2 (50%)
		APECED (n = 1)	1/1
		CD40L deficiency (n = 1)	0/1
		CGD (n = 1)	1/1
		CTLA- 4 deficiency (n = 1)	1/1
		Complement C2 deficiency (n = 1)	1/1
		ADA2 deficiency (n = 1)	1/1
		IFNGR1 deficiency (n = 1)	1/1
		NEMO deficiency (n = 1)	1/1
		CHH (n = 1)	1/1
		STAT3 DN. Posthematopoietic stem cell transplantation (n = 1)	1/1
		Idiopathic T-cell lymphopenia (n = 1)	0/1
		WAS (n = 1)	1/1
Italy	Anti-S Ab; anti-RBD ab; Spike-specific T-cells	XLA (n = 7)	Anti-S ab + in 0/7	NA	23
			Anti-RBD ab + in 0/7
			Spike-specific T-cells in 5/6†
Czech Republic	Anti-S Ab; anti-N Ab; IGRA	STAT1 GOF (n = 7)	Anti-S Ab + in 6/7	None	89
			IGRA + in 6/6†

Ab, antibody; ALPS, autoimmune lymphoproliferative syndrome; ADA, adenosine deaminase; ADA2, adenosine deaminase 2; AIRE, autoimmune regulator; ALPS, autoimmune lymphoproliferative syndrome; anti-S, anti-Spike; anti-N, anti-Nucleocapsid; APECED, autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy; APDS2, activated PI3K Delta syndrome 2; CD40L, CD40 ligand; CGD, chronic granulomatous disease; CHH, cartilage hair hypoplasia; CID, combined immunodeficiency; CTLA-4, cytotoxic T-Lymphocyte antigen 4; CVID, common variable immunodeficiency; DN, dominant negative; FOXN1, forkhead box N1; GOF, gain-of-function; IFNGR1, interferon-gamma receptor; IGRA, spike protein–specific T-cell responses evaluated using SARS-CoV-2 IFN-γ release assay; MAGT1, magnesium transporter 1; NA, not available; NEMO, NF-kappa B essential modulator; PIK3R1, phosphoinositide-3-kinase regulatory subunit 1; RAG, recombination-activating genes; RALD, ras-associated autoimmune leukoproliferative disorder; RBD, receptor-binding domain; SASH3, SAM and SH3 domain containing 3; SpAD, specific antibody deficiency; STAT, signal transducer and activator of transcription; WAS, Wiskott-Aldrich syndrome; WHIM, warts, hypogammaglobulinemia, infections, and myelokathexis; XLA, X-linked agammaglobulinemia. *CVID have been included in the analysis only for the studies in which disaggregated data were not available. †Not performed in one patient.

Conclusions

The large number of cases of SARS-CoV-2 infection reported in patients with IEI has led to the identification of differences in the risk of developing severe disease among different forms of IEI. In some cases, such as in patients with impaired production of or response to type I IFN, the genetic defect favors active replication of the virus and spreading to the lower respiratory tract. In other forms of IEI (combined immunodeficiencies, antibody defects, patients within 1 yr of HSCT), the inadequacy of adaptive immune responses may impede control of the infection. Finally, hyperinflammation associated with some forms of PIRD may exacerbate some of the features of COVID-19, with an increased risk of multisystem involvement. In this regard, knowledge of the underlying gene defect may be essential to prompt timely and appropriate preventive and therapeutic interventions to counteract the risk of severe disease in patients with IEI. At the same time, in case of severe SARS-CoV-2 infection, especially in children and in adults with no apparent comorbidities, it is fundamental to consider an IEI as an underlying cause and further investigations, including genetic tests and autoantibodies measurement, should be performed (131).

The results reported in the studies published so far indicate that COVID-19 vaccines are efficacious and safe in patients with IEI, supporting the current recommendation that these patients should be vaccinated against COVID-19. However, additional studies are needed to assess the durability and robustness of the immune responses to COVID-19 vaccines in IEI patients. Larger patient cohorts are required to drive definite conclusions on vaccine efficacy in this heterogeneous population. There is an urgent need for further studies evaluating both humoral and cellular responses in different IEI cohorts longitudinally over time. By combining these data with analysis of the frequency and severity of breakthrough cases among patients with IEI and in the general population, it will be possible to obtain a more solid indication of the possible need for additional vaccine boosts in patients with IEI.