Inborn Errors of Immunity Genotype: A Retrospective Study from the United Arab Emirates

Abstract

Inborn errors of immunity (IEI) contribute substantially to morbidity in the Middle East and North Africa (MENA), particularly in the United Arab Emirates (UAE), where high consanguinity increases the prevalence of autosomal-recessive and complex genetic disorders. Regional IEI genomic data remain limited, and population-specific variants are underrepresented in global databases, limiting diagnostic accuracy and precision-medicine implementation. We retrospectively reviewed 2,500 patients assigned immunodeficiency-related ICD codes over ten years at two tertiary centers in Al Ain, UAE. Based on the availability of genetic testing reports for inborn errors of immunity, 232 patients (~ 9%) were included in the analysis. Variants were annotated using American College of Medical Genetics and Genomics guidelines and open-access resources, including Ensembl and ClinVar, and genes were classified according to the International Union of Immunological Societies framework. Allele frequencies were reported based on gnomAD and the Emirati Genome Project. Reported variants were systematically reclassified using Franklin, VarSome, ClinGen, and HGMD, where available. Based on original reports, 245 IEI-related variant observations were identified across 107 genes: 99 (40.4%) pathogenic/likely pathogenic, 68 (27.8%) benign/likely benign, and 78 (31.8%) variants of uncertain significance (VUS). Following reclassification, pathogenic/likely pathogenic variants increased to 107 (43.7%), VUS decreased to 54 (22.0%), and benign/likely benign variants increased to 76 (31.0%). Recurrent genes included ATM, DOCK8, RAG1, RAG2, UNC13D, LRBA, MEFV, and NLRP3, and 7.8% of patients carried multiple IEI-related variants. This study provides one of the largest genomic characterizations of IEI in the UAE and demonstrates that systematic variant reinterpretation improves diagnostic accuracy, underscoring the importance of population-specific genomic resources for precision medicine.

Graphical Abstract

Genomic characterization of a UAE cohort with suspected inborn errors of immunity (IEI). Review of 2,500 immunology-related clinical records identified 232 genetically tested patients, yielding 245 variant observations across 107 genes. Variant reassessment reduced variants of uncertain significance from 78 (31.8%) to 54 (22.0%), highlighting the importance of regional genomic resources, prospective IEI registries, and segregation and functional studies.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s10875-026-02060-8.

Introduction

Inborn errors of immunity (IEI) represent a heterogeneous group of disorders characterized by impaired immune development or function, resulting in recurrent infections, immune dysregulation, autoimmunity, and malignancy. Failure to treat IEIs in a timely manner is associated with significant morbidity and mortality [1]. Recent advances in next-generation sequencing (NGS) technologies have transformed the diagnosis of IEIs, facilitating precision in patient management and prognosis [2]. Notably, this supports effective genetic counseling and cascade testing for at-risk family members, alongside the implementation of targeted therapies, including immunoglobulin replacement, hematopoietic stem cell transplantation, gene-directed treatments, and biologic agents [3]. This is especially important in the Middle East and North Africa (MENA) region, where high rates of consanguinity drive the prevalence of autosomal recessive disorders and increase the burden of IEI [4]. Genetic data in the MENA region, including the United Arab Emirates (UAE), remain underrepresented in global variant databases, limiting IEI diagnostic accuracy and delaying optimal care [5]. To address inherited disorders, the UAE has implemented premarital and newborn screening programs; however, these remain targeted and fragmented [6–8]. In this context, region-specific genetic studies are essential to define the IEI landscape and identify both recurrent and novel pathogenic variants. This study aims to describe the genetic variant profile of individuals with IEI in the UAE. Emphasizing the identification of recurrent, rare, and novel variants in a highly consanguineous population. This will inform future population-specific interpretation and diagnostic strategies such as extension of premarital screening and IEI newborn screening.

Methods

Study Design and Cohort

This retrospective study included patients evaluated at two tertiary care centers, Tawam Hospital and Sheikh Tahnoun Bin Mohammed Medical City, Al Ain, United Arab Emirates. Ethical approval for this study was obtained (Ref: DOH/ADHRTC/2025/1156). This study used de-identified data, and informed consent was waived by the ethics committee.

Patients were initially identified from Cerner electronic medical records (EMRs) between 2015 and 2025 using immunology-related International Classification of Diseases, Tenth Revision (ICD-10) diagnostic codes. The ICD-10 codes used for the initial extraction are provided in Table S1. This search identified approximately 2,500 patients with immunology-related diagnoses. The present genetic analysis was restricted to 232 patients with available genetic testing results performed for inborn errors of immunity during the study period.

Patients were excluded if the available records indicated an alternative primary diagnosis, including inborn errors of metabolism or secondary immunodeficiency, rather than suspected monogenic IEI. Genetic testing was requested as part of routine clinical care at the discretion of the treating physicians (e.g. clinical immunologist, geneticist), based on clinical presentation, immunological abnormalities, family history, consanguinity, disease severity, and/or suspicion of an underlying IEI.

Extracted data included demographic characteristics, including age, sex, and ethnicity, and available genetic findings with indication concordance.

Genotype Review and Classification

Genotype data were obtained from EMRs, and testing included targeted gene panels and next-generation sequencing approaches, such as whole-exome sequencing (WES) and whole-genome sequencing (WGS). Gene testing modalities were selected based on documented indications. These tests were conducted either in-house or through external accredited laboratories. Variants were categorized based on the International Union of Immunological Societies (IUIS) [9]. All variants were annotated utilizing open-access resources, including Ensembl and ClinVar. Variants were classified as pathogenic, likely pathogenic, variants of uncertain significance (VUS), likely benign, or benign according to the American College of Medical Genetics and Genomics/ Association of Molecular Pathology (ACMG/AMP) classification and based on extracted reports. In silico pathogenicity for missense variants was not consistently reported in the extracted data, therefore, prediction scores from multiple validated tools are used for evaluation, including CADD, SIFT, AlphaMissense, REVEL, EVE, ClinPred, and PolyPhen-2 [10, 11]. In addition, the Genome Aggregation Database (gnomAD) allele frequency was obtained from Ensembl. Given the underrepresentation of Arab populations in global genomic reference databases, population-specific allele frequencies (AAF) were reported from the Emirati Genome Program (EGP: ~140,000 individuals) [5, 12].

Variant Reclassification

All reported IEI-related variants identified in the original genetic reports underwent systematic re-evaluation to determine their current pathogenicity classification. Reclassification was performed primarily using Franklin by Genoox and was supplemented with data from VarSome, ClinGen, and the Human Gene Mutation Database (HGMD) where available. All databases were accessed between 10 July 2026 and 12 July 2026 for the purposes of this study.

Evidence incorporated into reclassification included population allele frequency, prior clinical classifications, published disease associations, gene-disease validity, predicted functional consequences, and available computational and functional evidence. Updated classifications were compared with those documented in the original genetic reports to identify changes in variant interpretation over time. Variants without a corresponding pathogenicity classification in the queried databases were recorded as not reported in the reclassification dataset.

Variants were additionally annotated as novel or emerging for descriptive analysis. Novel variants were defined as variants absent from both ClinVar and HGMD at the time of review. Emerging variants were defined as variants absent from ClinVar but present or classified in Franklin and/or other variant interpretation resources. These labels were used descriptively and were not considered evidence of pathogenicity in isolation.

Results

Cohort Characteristics and Overall Genetic Findings

The available genetic results of 232 out of 2,500 patients are reported. The cohort consists of 58% (n = 135) males, with 80% (n = 186) younger than 18 years. The ethnic distribution showed a significant majority Emirati (UAE nationals) of 69% (n = 160) (Table 1).

Table 1.

Study cohort demographics

Characteristic	n (%) / mean ± SD
Total patients	232
Gender
Male	135 (58)
Female	97 (42)
Age Distribution
Pediatric (≤ 17 years)	186 (80.2) / 8.7 ± 4.0
Infant (< 2)	3 (1.3)
Toddler (2–5)	44 (19)
Child (6–11)	91 (39.2)
Adolescent (12–17)	48 (20.7)
Adult (≥ 18 years)	46 (19.8) / 37 ± 12.8
Young adult (18–39)	31 (13.4)
Middle-aged adult (40–64)	13 (5.6)
Older adult (65+)	2 (0.9)
Ethnicity
UAE	160 (69)
Other GCC	24 (10.3)
Levant	13 (5.6)
North Africa	8 (3.4)
South Asia	21 (9.1)
Others	6 (2.6)

The table summarizes age distribution, sex, and ethnicity of the 232 genotyped patients

A total of 245 reported IEI-related variant observations across 107 genes were identified. Based on the original reports, 99 variants were classified as pathogenic or likely pathogenic (P/LP; 40.4%), 78 as variants of uncertain significance (VUS; 31.8%), and 68 as benign or likely benign (B/LB; 27.8%) (Fig. 1).

Fig. 1.

Variant pathogenicity classification. This figure shows the distribution of 245 identified genetic variants according to pathogenicity classification (pathogenic, likely pathogenic, VUS, likely benign, and benign)

Original Report Variant Classification

Based on the original genetic report classifications, the 245 identified variants were distributed across nine IUIS disease categories, demonstrating marked heterogeneity in both variant burden and pathogenicity classification (Table 2). The largest number of variants was observed in combined immunodeficiency (CID) with syndromic features (21.2%, n = 52), diseases of immune dysregulation (16.3%, n = 40), and autoinflammatory disorders (16.3%, n = 40). Across multiple categories, pathogenic and likely pathogenic (P/LP) variants represented a substantial fraction of identified variants, including immunodeficiencies affecting cellular and humoral immunity (70.6%, n = 24 of 34), CID with syndromic features (48.1%, n = 25 of 52), and congenital defects of phagocyte number or function (56.7%, n = 17 of 30).

Table 2.

Distribution of genes and variants across IUIS categories based on original genetic report classifications

IUIS category	Genes n	Total variants n	P/LP n (%)	VUS n (%)
Immunodeficiencies affecting cellular and humoral immunity	11	34	24 (70.6)	9 (26.5)
CID with associated or syndromic features	18	52	25 (48.1)	10 (19.2)
Predominantly Antibody Deficiencies	9	10	2 (20.0)	5 (50.0)
Diseases of Immune Dysregulation	17	40	7 (17.5)	19 (47.5)
Congenital defects of phagocyte number or function	12	30	17 (56.7)	8 (26.7)
Defects in Intrinsic and Innate Immunity	11	14	6 (42.9)	6 (42.9)
Autoinflammatory Disorders	16	40	6 (15.0)	9 (22.5)
Complement Deficiencies	4	7	3 (42.9)	3 (42.9)
Bone Marrow Failure	9	18	9 (50.0)	9 (50.0)

This table summarizes the distribution of reported IEI-related variant observations across IUIS disease categories in the genetically tested cohort of 232 patients, based on the classifications documented in the original genetic reports. “Genes, n” indicates the number of distinct genes identified within each IUIS category. “Total variants, n” indicates the number of reported variant observations within each category. Percentages for pathogenic/likely pathogenic (P/LP) variants and variants of uncertain significance (VUS) were calculated using the total number of reported variant observations within each IUIS category as the denominator

In addition, a considerable proportion of these variant allele frequencies (VAF) are absent from global reference databases and remained rare (low allele frequency) or unobserved even within the Emirati Genome Program allele data (local genomic data). These findings highlight the importance of population-specific variant interpretation. Comprehensive gene and variant information, including in silico prediction scores and Emirati allele frequencies, are provided in Supplementary Tables S2–S10.

Variant Reclassification

Systematic re-evaluation of the 245 reported variants shifted the overall distribution of pathogenicity classifications (Fig. 2). Relative to the original genetic reports, the proportion of variants classified as pathogenic or likely pathogenic (P/LP) increased from 40.4% (n = 99) to 43.7% (n = 107). Conversely, the proportion of VUS decreased from 31.8% (n = 78) to 22.0% (n = 54). The proportion of benign or likely benign (B/LB) variants also increased, from 27.8% (n = 68) to 31% (n = 76). The distribution of reclassified variants across IUIS categories is summarized in Table 3.

Fig. 2.

Change in variant pathogenicity classification after reclassification. Bar graph comparing the distribution of reported variant observations according to their original genetic report classification and updated classification after systematic re-evaluation using Franklin by Genoox, VarSome, and ClinGen where available. Bars show the proportion of variants classified as pathogenic/likely pathogenic (P/LP), variants of uncertain significance (VUS), and benign/likely benign (B/LB), with percentage changes indicated

Table 3.

Distribution of reclassified variant observations across IUIS categories

IUIS category	Genes n	Total variants n	P/LP n (%)*	VUS n (%)*	Novel n	Genes identified
Immunodeficiencies affecting cellular and humoral immunity	11	34	22 (64.7)	9 (26.5)	5	IL2RG, JAK3, CD3D, RAG1/2, DCLRE1C, ADA, CD40LG, ZAP70, DOCK8, MAP3K14
CID with associated or syndromic features	18	52	23 (44.2)	12 (23.1)	6	WAS, ATM, BLM, PMS2, RECQL4, DiGeorge, CHD7, STAT3, SPINK5, IKBKB, STIM1, PNP, SP110, BCL11B, NFE2L2, STAT5B, KMT2A, KMT2D
Predominantly Antibody Deficiencies	9	10	3 (30.0)	2 (20.0)	1	BTK, PIK3CD, PTEN, TRNT1, NFKB2, IKZF1, AICDA, MSH6, CARD11
Diseases of Immune Dysregulation	17	40	12 (30.0)	12 (30.0)	6	PRF1, UNC13D, STX11, STXP2, LYST, RAB27A, CTLA4, LRBA, DEF6, STAT3, IKZF1, PEPD, TLR7, NCKAP1L, SH2D1A, XIAP, MAGT1
Congenital defects of phagocyte number or function	12	30	20 (66.7)	3 (10.0)	4	ELANE, GFI1, VPS13B, CSF3R, SMARCD2, FERMT3, ACTB, CFTR, CYBB, NCF1/4, G6PD
Defects in Intrinsic and Innate Immunity	11	14	6 (42.9)	5 (35.7)	3	IFNGR1, CYBB, IRF8/9, TYK2, RORC, STAT1, POLR3A, MYD88, TLR7, GATA2
Autoinflammatory Disorders	16	40	8 (20.0)	5 (12.5)	2	ADA2, TREX1, RNASEH2A, DNASE1L3, RELA, MEFV, MVK, NLRP1/3/12, LPIN2, TNFRSF1A, NOD2, CARD14, PSMB8, TNFAIP3
Complement Deficiencies	4	7	4 (57.1)	1 (14.3)	0	C3, C8B, SERPING1, CFD
Bone Marrow Failure	9	18	9 (50.0)	5 (27.8)	4	BRCA1/2, FANCG, FANCI, PALB2, SLX4, TERT, RTEL1, PARN, TP53

This table summarizes reported IEI-related variant observations across IUIS disease categories after systematic reclassification. Gene counts indicate the number of distinct genes per category, while variant counts indicate reported variant observations. Percentages for P/LP and VUS variants were calculated using the total number of variants in each category as the denominator. Novel variants were defined as variants absent from both ClinVar and HGMD

An additional 8 variants (3.3%) lacked a corresponding pathogenicity classification reported in the databases used for re-evaluation and were accordingly categorized as “not reported”. For visual comparison in Fig. 2, these variants were retained under their original genetic report classification rather than plotted as a separate category. Detailed variant-level information, including original classification, updated classification, and IUIS category, is provided in Supplementary Tables S2–S10.

Overall, reclassification reduced the proportion of VUS and increased the proportion of variants assigned to more definitive pathogenic or benign categories, demonstrating the impact of updated variant interpretation on the classifications of genetic findings in this cohort. Recurrent IEI-associated genes with ≥ 3 variants per gene and their corresponding novel variant burden are summarized in Fig. 3.

Fig. 3.

Recurrent IEI-associated genes and novel variant burden by IUIS category. Horizontal stacked bar chart showing recurrent IEI-associated genes with ≥ 3 variants. Bar colors indicate the corresponding IUIS disease category. The light-blue segment represents the number of novel variants, defined as variants absent from both ClinVar and HGMD at the time of review. The remaining-colored segment represents previously reported variants. Total bar length indicates the total number of variants identified for each gene

VUS and Novel Variants

Of the reclassified variants, 54 (22%) remained categorized as VUS. The highest proportions are observed in diseases of immune dysregulation (22%, n = 12), combined immunodeficiencies with associated or syndromic features (22%, n = 12), followed by immunodeficiencies affecting cellular and humoral immunity (16.7%, n = 9). Detailed variant-level data stratified by IUIS category are provided in Supplementary Tables S2–S10.

Among the reclassified dataset, 31 variants were considered novel because they were absent from both ClinVar and HGMD at the time of review (Table 3). These novel variants were distributed across IUIS categories and included genes with established relevance to immune function. Several were rare or absent in available population datasets. Relevant examples included: DOCK8: c.2632 C > T, p.Arg878Trp (AAF = NA, gnomADg_MID = 0 ), ATM: c.112_122delTTAA, p.Ile40Asnfs*3 (AAF = NA, gnomADg_MID = NA), and LRBA: c.974 C > T, p.Ala325Val (AAF = NA, gnomADg_MID = NA). While these findings do not establish pathogenicity, they highlight the contribution of underreported regional variation to the unresolved genetic landscape of IEI in this cohort.

In addition, we also identified a subset of emerging variants that were absent from ClinVar but had interpretive evidence in Franklin and/or other variant interpretation resources. These variants included IL2RG c.365T > C, p.Ile122Thr, CHD7 c.3169 C > T, p.Gln1057*, and ELANE c.199T > C, p.Ser67Pro. This pattern suggests that some variants may have emerging evidence outside ClinVar, potentially recent literature or laboratory submissions. Their identification emphasizes the importance of using multiple complementary variant interpretation resources, particularly in underrepresented populations where database coverage remains incomplete.

Patients with Multiple Variants

A subset of patients (n = 18 of 232) exhibited variants in multiple immunodeficiency-related genes (Fig. 4). These individuals frequently harbored a combination of P/LP variants and VUS, reflecting increased genetic complexity. The most prevalent IUIS categories included autoinflammatory disorders, combined immunodeficiency with syndromic features, predominantly antibody deficiencies, defects of intrinsic and innate immunity, and bone marrow failure. Although some patients carried multiple variants within a single IUIS category, the majority (n = 14) had variants spanning multiple categories, suggesting involvement of distinct immune pathways that may contribute to heterogeneous clinical phenotypes. Moreover, given the highly consanguineous nature of the population, the presence of additional non-immunological variants may modify the clinical phenotype. These variants were not assessed, as the analysis in this paper was focused exclusively on IEI-related genes. Detailed genotype information is provided in Supplementary Table S11.

Fig. 4.

Patients with multiple IEI variants. Sankey plot illustrating patient–gene–IUIS category relationships among patients with more than one reported IEI-related variant. Each flow links an anonymized patient identifier to the affected gene and corresponding IUIS disease category. Flow width reflects the number of reported variants: on the patient side, wider flows indicate patients with a greater number of reported variants, while on the category side, wider flows indicate a greater number of variants mapping to that IUIS category

Discussion

This study describes the spectrum of IEI in a consanguineous cohort from the United Arab Emirates, highlighting population-specific features with important implications for diagnosis, therapy, and public health. We identified recurrent involvement of genes including RAG1, RAG2, DOCK8, ATM, PMS2, STAT3, UNC13D, STXBP2, LRBA, CFTR, G6PD, and MEFV, based on variant frequency (more than 3 variants per gene). These findings reinforce the essential role of genetic testing, consistent with previous studies demonstrating diagnostic yields up to 64% using next-generation sequencing approaches [13, 14]. Notably, many of these genes are associated with disorders for which targeted therapies are available, including PI3Kδ inhibition, abatacept for CTLA4 or LRBA deficiency, JAK inhibition for STAT3-related disease, and hematopoietic stem cell transplantation or gene therapy for severe combined immunodeficiency and chronic granulomatous disease [15–17]. However, clinical actionability in individual patients requires genotype–phenotype concordance, inheritance assessment, and clinical confirmation.

A key finding of this study was that systematic variant reclassification altered the overall distribution of pathogenicity categories, reducing the proportion of variants remaining as VUS and increasing the proportion assigned to more definitive pathogenic/likely pathogenic or benign/likely benign categories. This highlights the evolving nature of variant interpretation, particularly in underrepresented populations where allele-frequency data, ClinVar submissions, and gene–disease evidence remain incomplete. In line with ACMG/AMP variant interpretation guidance and recommendations on genomic result re-evaluation, periodic reassessment allows emerging genetic and clinical evidence to be integrated into variant interpretation, increasing the likelihood of resolving previously uncertain findings and supporting more informed clinical management and genetic counseling [18, 19]. However, updated classifications should be interpreted alongside zygosity, inheritance pattern, phenotype concordance, segregation data, and functional evidence before being considered definitive molecular diagnoses [18].

Despite reclassification, unresolved VUS and variants absent or inconsistently represented in major clinical databases remained important findings in this cohort. We distinguished novel variants, defined as absent from both ClinVar and HGMD, from emerging variants, defined as absent from ClinVar but reported or interpreted in Franklin or other variant resources. Several of these variants occurred in established IEI-associated genes and were rare or absent in available population datasets. The use of EGP allele-frequency data further emphasized the importance of population-specific reference datasets, as several variants remained uncommon or absent in both global and local datasets. These findings should be interpreted in light of the evolving scope of national population-scale sequencing efforts and the continued need for expanded regional genomic resources to improve variant interpretation, diagnostic accuracy, and precision-based care [5, 12, 20, 21]. Importantly, absence from clinical databases or discordance between databases does not establish pathogenicity. These variants should instead be prioritized for phenotype correlation, segregation analysis, and functional validation before their contribution to immune disease can be inferred.

A defining characteristic of this cohort was the prevalence of patients harboring multiple variants, occasionally involving different IEI-associated genes or IUIS categories. In the context of a highly consanguineous population, such findings highlight the genetic complexity encountered during IEI evaluation. However, co-occurrence of multiple variants, particularly when including VUS or benign/likely benign variants, does not establish oligogenic inheritance. These observations should therefore be interpreted cautiously and require segregation analysis, phase determination, phenotype concordance, and functional validation before any contribution to disease can be assumed. This is particularly relevant in consanguineous populations, where complex genetic backgrounds may contribute to variable expressivity and atypical phenotypes [20].

Globally, initiatives such as the J Project, the Jeffrey Modell Foundation, and curated resources like the GenIA database demonstrate how integration of genetic variant data enhances IEI diagnosis and clinical care [21, 22]. Prospective, large-scale sequencing studies have shown that early genomic testing achieves definitive molecular diagnoses in approximately 25% of pediatric and 9% of adult patients, leading to changes in clinical management in more than 75% of diagnosed cases. In patients with complex immune phenotypes, diagnostic yields exceed 30% and frequently result in clinically actionable interventions [23, 24]. The higher diagnostic yield observed in individuals from low- and middle-income countries underscores the impact of global genomic initiatives in advancing healthcare equity [25]. This impact extends to population-level prevention through genomic newborn screening, exemplified by well-established severe combined immunodeficiency (SCID) screening using T-cell receptor excision circles, which has increased 5-year survival to approximately 87% in the USA and Canada [26–29].

Studies from Arab and MENA populations have reported enrichment of autosomal-recessive IEI and population-specific pathogenic variants in the context of high consanguinity [4, 20, 30]. Underrepresentation in global genomic databases may contribute to novel or discordantly classified variants, increasing uncertainty in interpretation and potentially delaying diagnosis. Prior studies have shown that over 17% of pathogenic or likely pathogenic variants in Emiratis are absent from gnomAD or ClinVar, a finding that is also observed in our current analysis [31, 32]. Regional biobanks and population-specific databases have improved variant discovery, genotype–phenotype correlations, and the development of tailored diagnostic panels [31–36]. Region-specific diagnostic algorithms and national registries in the MENA region demonstrate clear clinical utility by improving diagnostic yield and reducing unnecessary testing. Evidence from multicenter national experiences in Algeria and Morocco further shows that systematic registry-based clinical–immunological evaluation captures population-specific IEI phenotypes and genetic architecture, enabling identification of founder and novel genes and informing the development of tailored diagnostic and preventive strategies [37–40].

Despite these advances, important gaps remain, including sparse representation of Arab populations in global genetic databases, limited functional validation of variants, incomplete subpopulation coverage, and insufficient de novo sequencing to resolve complex genomic regions [5, 20, 34]. As a result, many rare and novel variants remain uncharacterized and inconsistently interpreted across local and international databases, compounded by uneven access to genomic diagnostics and research infrastructure across the region [20, 30, 38, 41, 42]. Addressing these gaps is essential before population-tailored screening strategies can be evaluated. The current findings support the development of population-informed diagnostic resources, prospective IEI registries, and structured approaches for periodic VUS re-evaluation as new evidence emerges. In the future, large prospective datasets integrating standardized clinical, genomic, segregation, and functional data may help inform evidence-based evaluation of targeted screening strategies for high-prevalence and treatable IEIs.

This study has several limitations. First, the genetically tested subgroup was clinically selected and should not be interpreted as representative of all patients with immunology-related ICD-10 diagnoses or of the wider UAE IEI population. Referral and clinician-selection bias should therefore be considered when interpreting the frequency of genes, variants, and IUIS categories. Second, testing modalities were heterogeneous and included targeted gene panels, WES, and WGS performed across different laboratories, which may have affected gene coverage, reporting thresholds, and variant classification consistency. Although CNVs reported in the original clinical reports were retained, they were not systematically reanalyzed or reclassified, as the present study primarily focused on SNVs and small insertions/deletions. Third, standardized segregation analysis, phase information, and functional validation were not uniformly available. Therefore, P/LP variants should not be interpreted as confirmed molecular diagnoses unless supported by zygosity, inheritance pattern, phenotype concordance, and clinical interpretation. Finally, although reclassification was performed systematically across multiple resources, database coverage, evidence availability, and classification criteria varied across platforms. The updated classifications should therefore be interpreted as research-level harmonization of available evidence rather than a replacement for clinical laboratory-issued diagnostic reports.

Overall, this study provides one of the largest genomic characterizations of inborn errors of immunity in the UAE and demonstrates the value of systematic variant reinterpretation for improving diagnostic accuracy. The persistence of unresolved, novel, and emerging variants underscores the need for prospective registries, population-specific genomic resources, segregation studies, and functional validation to improve variant interpretation, support precision medicine, and strengthen regional contributions to global IEI genomic resources.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

HA and FA contributed to the study conception and design. Data collection and analysis were performed by [MA, FAA, MS and FA]. The first draft of the manuscript was written by [MA, HA and FA] and all authors [MA, FAA, MS, HA and FA] commented on previous versions of the manuscript. All authors [MA, FAA, MS, HA and FA] read and approved the final manuscript.

Funding

The authors declare that financial support was received for this work. This work was supported by the United Arab Emirates University Research Grant (Grant No. 12M238).

Data Availability

The data supporting the findings of this study contain sensitive clinical and genetic information and are protected by ethical and privacy restrictions related to patient confidentiality, as well as institutional data governance policies. The data are therefore not publicly available.

Declarations

Competing Interest

The authors declare no competing interests.