Introduction

Primary immunodeficiency disorders (PIDs) comprise approximately 400 inherited disorders affecting the development or the function of immune system components [1]. The overall prevalence of PID has been estimated to be more than 1/600 individuals in Western countries [2,3,4]. However, this prediction is likely to be an underestimation in countries with a high rate of consanguineous marriages, since more than 70% of known PID-associated genes have an autosomal recessive pattern of inheritance [1]. PID patients may present with a broad spectrum of clinical manifestations and immunological complications mainly during childhood [5, 6]. Mutations in different PID genes may result in similar presentations; however, they may need also different management and therapeutic measures. Wide variations in the geographical and ethnical prevalence of PID have been reported by means of patients’ registry providing invaluable information for resource allocation and health policy making [3, 4].

Since the establishment of the Iranian Primary Immunodeficiency Registry (IPIDR) in 1999, the diagnosis of the patients was mainly based on clinical parameters, affecting the process of definite diagnosis for targeted therapy and genetic counseling. Before 2014, we had reported 1661 patients in three consequential reports [7,8,9], but none of them included comprehensive information about the patient’s genetic background. During the past 5 years, a growing number of clinical immunologists and recent advances in molecular diagnostic methods have led to an increase in the identification of genetic defects and main disease-causing gene mutations in the cohort, improving individualized therapeutic methods and the performance of prenatal diagnosis [10]. This has also resulted in an earlier diagnosis, a more precise definition of defects, reduced morbidity and mortality rates, and better long-term outcomes [10, 11].

Of note, many of the novel PID gene defects, mainly in autosomal recessive form, have been reported from the Middle East region. Therefore, continuous update on these registries has had a significant impact on our understanding of the pathogenesis of PID and also the function of the immune system [12, 13]. Here, we provide a recent update on newly diagnosed Iranian patients with PIDs and integration of molecular diagnostic outcomes based on the latest classification of PIDs by the International Union of Immunological Societies (IUIS) Expert Committee [1].

Materials and Methods

Iranian Primary Immunodeficiency Registry

This study was conducted as a cohort of patients, prospectively enrolled in the IPIDR from the “National PID Network.” The IPIDR is managed by the Research Centre for Immunodeficiencies (Tehran, Iran) and its main aim is to provide epidemiological, clinical, and molecular data of PID in Iran. By March 2013, 1661 PID patients from different provinces of the country were registered in the IPIDR [7]. By the latest estimation in 2016, Iran has a population of 79,926,270 citizens (with an average annual birth rate of 1,300,000) and according to the age structure, 23.7% are less than 14 years old. This study received approval from the Ethics Committee of the Tehran University of Medical Science. Moreover, written informed consent has been obtained from all patients, their parents, or legal guardians.

Role of Participant Centers

The registry database is located in the Children’s Medical Center (Tehran, Iran) which serves as a referral hospital for suspected or diagnosed cases of PID. In addition, 31 medical centers, affiliated to 26 medical science universities, collaborated in the registry program from the major provinces of the country to form the PID network (Fig. 1). All of the participating centers had access to national guidelines and necessary laboratory equipment for clinical and immunological evaluations. Subsequently, cases with suspected diagnosis were referred and re-evaluated in the Children’s Medical Center for a definitive diagnosis.

Fig. 1
figure 1

Schematic map of Iran indicating the distribution of PID network centers and cumulative incidence of PID (per million) in each province

Full size image

Clinical and Immunologic Diagnoses

The clinical diagnosis of the PID patients was made according to the criteria of the European Society for Immunodeficiencies (ESID, https://esid.org/Working-Parties/Registry/Diagnosis-criteria). A questionnaire surveyed the patients’ demographic information, age of disease onset, age of diagnosis, family history, detailed clinical history that included vaccination history and associated adverse reactions, recurrent infections, physical examination findings, laboratory data, and treatment history. Secondary defects of the immune system, including those caused by human immunodeficiency virus (HIV), were ruled out. Laboratory evaluations were performed in the study group as indicated, including complete blood and differential counts, serum protein profile and immunoglobulin (Ig) levels, serum IgG subclass levels, isohemagglutinin titers, specific antibody responses, disease-specific autoantibody measurements, flow cytometric evaluation of lymphocyte subsets, nitro blue tetrazolium dye/dihydrorhodamine test, granulocyte function and chemotaxis tests, lymphocyte transformation and T cell function tests, radiosensitivity, and measurement of complement component levels and hemolytic complement activity [14]. Microbiological, pathological, and imaging evaluations were performed for clinical diagnosis when required. A computerized database program (new registry section in http://rcid.tums.ac.ir/) was implemented for data entry. After reviewing the cases by the administrator of the system for duplicated cases, patients with incomplete diagnostic criteria were excluded. The online database was updated frequently for approved patients and all follow-up data sent by the end of the study period were included.

Genetic Analysis and Diagnoses

Genomic DNA was extracted from whole blood, as previously described [15]. For patients with classical clinical presentations suggestive of a specific PID, Sanger sequencing was performed on the panel of most likely genes (Table S1). Patients with thymic defects were examined by using fluorescent in situ hybridization (FISH) for 22q11.2 deletion and a comparative genomic hybridization array. For patients in whom Sanger sequencing failed or who had a clinical presentation resembling several genetic defects, next-generation sequencing (e.g., targeted gene sequencing [Table S2] and whole exome sequencing) was performed to detect single-nucleotide variants, insertions/deletions, and large deletions using a pipeline described previously [10, 11, 16]. Candidate variants were evaluated by the Combined Annotation-Dependent Depletion (CADD) algorithm and an individual gene cutoff given by using the mutation significance cutoff (MSC) was considered for impact predictions [17]. The Gene Damage Index (GDI) server and the Human Gene Connectome (HGC) were used to make a combined effects prediction [17]. The pathogenicity of all disease attributable gene variants was re-evaluated using the updated guideline for interpretation of molecular sequencing by the American College of Medical Genetics and Genomics criteria [18, 19], considering the allele frequency in the population database, computational data, immunological data, familial segregation and parental data (confirmatory Sanger sequencing for probands and their parents), and clinical phenotyping.

Statistical Analyses

After confirmation of their clinical and genetic diagnosis, patients were classified according to the IUIS updated classification including nine categories of immunodeficiencies affecting cellular and humoral immunity (non-syndromic combined immunodeficiency or CID), combined immunodeficiencies with associated or syndromic features (syndromic CID), predominantly antibody deficiencies (PAD), diseases of immune dysregulation, congenital defects of phagocyte number or function (phagocytosis disorders), autoinflammatory disorders, defects in intrinsic and innate immunity, complement deficiencies, and phenocopies of inborn errors of immunity [1]. A commercially available software package (SPSS Statistics 17.0.0; SPSS, Chicago, IL) was used for statistical analysis and the one-sample Kolmogorov-Smirnov test was applied to estimate whether data distribution was normal. Parametric and nonparametric analyses were performed based on the findings of this evaluation. Linear regression analysis was performed to evaluate the chronological effect of time on the different parameters. Kaplan-Meier curves and log-rank tests were used to compare different survival estimates. A p value of 0.05 or less was considered statistically significant.

Results

Newly Diagnosed Patients and Total Registry Characteristics

A total number of 1395 PID patients (824 male and 571 female) were enrolled in the study in the time period between March 2013 and March 2018. Due to an increased awareness by physicians and improved diagnostic facilities [11], an estimated incidence in this 5-year period was 0.21 cases per 1000 births. Among the newly diagnosed patients, autoinflammatory disorders were the most common group of PID, affecting 438 cases (31.4%), followed by PAD in 310 patients (22.2%), syndromic CID in 256 cases (18.6%), defects in innate immunity in 122 cases (8.7%), phagocytosis disorders in 117 cases (8.4%), non-syndromic CID in 83 cases (5.9%), diseases of immune dysregulation in 36 cases (2.6%), and complement deficiencies in 33 cases (2.3%). The median diagnostic delay was 10 months, ranging from 0 to 9.5 years. Of the 1395 cases, a diagnosis was made in 934 patients (66.9%) within 1 year from the onset of disease and within 2 years from the onset of disease in 1050 cases (75.2%). Diagnostic delay was higher than 5 years in only 153 cases (10.9%). The longest diagnostic delay was observed among patients with complement deficiency (median, 10.3 years, Table S3). Direct association between the age of the patients at the onset of disease and diagnostic delay was observed similar to our previous reports (r = 0.02, p = 0.001). However, comparison of diagnostic delay in newly diagnosed patients and old registered patients showed only a marginal difference during last 5 years (p = 0.08).

Accounting the newly registered cases in IPIDR, epidemiologic indexes of a total number of 3056 PID patients (1852 male and 1204 female) were evaluated. The cumulative incidence of PID in Iran during the past 10 years was estimated to be around 81 cases per 1,000,000 inhabitants. Figure 2 shows the number of registered PID patients based on the IUIS-classified group. Out of these 3056 PID patients, the majority was diagnosed with PAD (903 cases, 29.5%, Table 1). Neither cases with somatic mutations nor autoantibodies-mediated phenocopies of inborn errors of immunity were diagnosed in our cohort. Figure 1 shows the cumulative rate of clinical diagnoses of PID in different regions of the country.

Fig. 2
figure 2

Distribution of the frequencies of PIDs according to IUIS categories in 3065 Iranian patients in the study

Full size image
Table 1 Comparison of expected versus observed clinical diagnoses in different types of PID in Iran and the rate of molecular diagnosis in the registered patients
Full size table

Demographic Data and Follow-up Outcome

The cohort that underwent registration was assigned to three main ethnicities according to the reference population for the Greater Middle East (GME, http://igm.ucsd.edu/gme/) including Central Asia (80.3%) and Turkish (17.1%) and Arabian (2.6%) Peninsulas. A family history of PID was found in 641 cases (20.9%). Family history of recurrent infections without a known diagnosis of PID was positive in 389 cases (12.7%). In 737 cases (24.1%), a positive family history of death at an early age was documented. A history of diagnosed cancer was reported in the family of 322 patients (10.5%), and 208 individuals (6.8%) revealed a positive family history of autoimmune disorders. Consanguineous marriage was observed in the parents of 1837 cases (60.1%). There was a significant difference between the consanguinity rates of various types of PID, with the highest rate in patients with immune dysregulation (72 cases, 94.7%) and the lowest rate in patients with complement deficiencies (25 cases, 40.3%).

From 3056 patients, 1318 cases (43.0%) were confirmed to be dead (827 male and 491 female) and 528 patients (17.2%) could not be located during the last 6 months of the study period. Beside the specific disorders with a frequency of less than 10 cases, the mortality rate was highest in non-syndromic CID (157 severe CID cases and 57 less profound CID) with a mortality rate of 57.2% (214 out of 374 cases) and lost to follow-up rate of 9.3% (35 out of 374 cases). Cardiopulmonary failure due to severe pneumonia and sepsis was the main cause of death in 70.3% of the cases. Figure 3 compares the mortality rate in different PID categories presenting the lowest survival rate in patients with a CID (mainly due to severe CID) during the first 5 years of life (< 50% survival, significantly different compared to complement deficiencies [p = 0.02], immune dysregulations [p = 0.03], and autoinflammatory diseases [p = 0.045]).

Fig. 3
figure 3

Kaplan-Meier curve showing overall survival in the 3065 studied patients within the 8 categories of PIDs

Full size image

Genetic Diagnosis and Diagnostic Yield

According to the estimated prevalence of PIDs (1/600), the expected prevalence of PIDs in Iran would be more than 130,000 individuals. To date, 3056 clinically diagnosed patients of PIDs (2.3% of expected patients) have been diagnosed, and a definite diagnosis, defined by mutation analysis, was made in 1014 individuals (among 1457 evaluated patients, 69.5% diagnostic yield). As a result, 33.1% of the total number of registered patients has been diagnosed at a molecular level. Table 1 shows the frequency of patients in each category of PID who are genetically diagnosed in Iran. Experimental data and the results of functional assays on 921 patients with genetic diagnosis have been published previously (Table 2). Of note, the cohort has had a great impact on novel PID gene discovery including finding mutations in the HAX1, G6PC3, ELA2, JAGN1, CARD9, IFNGR2, DOCK8, STK4, LRBA, and CD70 genes. The majority of the patients (n = 805, 79.3%) had an autosomal recessive disorder (75.9% homozygous and 3.4% compound heterozygous). X-linked disorders comprised 12.8% of the genetic diagnoses (n = 130) and 7.8% of patients had an autosomal dominant disease (n = 79), where 76 mutations were de novo based on segregation analysis and Sanger sequencing of the parents (Fig. 4).

Table 2 Distribution of different genetic defects among patients with primary immunodeficiencies in Iran
Full size table
Fig. 4
figure 4

Trend of the frequency of different PIDs (a) and the inheritance pattern in patients with definite molecular diagnosis (b–c) in the IPIDR registry during 2002–2018 [7,8,9]

Full size image

The probable prevalence of non-syndromic CID has been reported to be 1/50,000 using both clinically diagnosed patients and newborn screening methods [20, 21]. Considering the total population of Iran, the expected frequency of patients would be 1600 individuals. Until now, there have been 374 patients with a clinical diagnosis of non-syndromic CID and among them, 122 individuals have had a definite diagnosis with known mutations. As a result, 23.3% of the expected patients have been recognized and 32.4% of them have been molecularly diagnosed. Based on the estimated prevalence of syndromic CID (1/100,000) [20], 800 individuals would be expected to be diagnosed in this PID group. Until now, 529 patients have been clinically diagnosed with syndromic CID (66.1% of the expected frequency) and 117 patients (22.1%) were diagnosed using mutation analysis. The estimated prevalence of PAD is > 1/650 [20, 22], which results in an expected frequency of more than 123,000 individuals considering the total population of Iran. It should be noted that approximately 70% of the estimated individuals are expected to be asymptomatic and have no significant clinical complications as a cause of selective IgA deficiency or other specific antibody defects. Only 903 PAD patients, 0.7% of the total expected frequency, have been clinically diagnosed. Among them, 111 patients (12.2%) have had a definite diagnosis by molecular techniques.

Regarding diseases of immune dysregulation and their expected prevalence (< 1/1,000,000 [20]), 95% of these patients have been registered by clinical diagnosis (76 of 80 expected patients) and 68 patients (89.4%) were genetically diagnosed by identifying mutated genes. Since the country is located in the eastern Mediterranean region, the probable prevalence of autoinflammatory disorders is more than 1/10,000 [20], and the expected frequency of patients with these disorders would be approximately 8000 individuals. To date, only 476 (5.9%) defined patients, mostly with a clinical diagnosis of familial Mediterranean fever, have been reported and 418 individuals (87.8%) have been definitely diagnosed by molecular techniques. Complement deficiencies were expected with a prevalence of 1/50,000 [20]; however, only 62 patients (3.8% of the expected frequency) and 2 patients (3.2% of registered patients) were diagnosed clinically and genetically, respectively.

Interestingly, the observed prevalence of phagocytosis disorders and defects in innate immunity was higher than their expected rate (1/250,000 and < 1/1,000,000, respectively [20]). There have been 507 clinically diagnosed patients with phagocytosis disorders (58% over than the estimated frequency) and 134 patients with defects in innate immunity (67% over than the estimated frequency). Among these patients, genetic mutations were identified in 117 (23.0%) from the former and 59 (44.0%) from the latter disease groups. Table 2 summarizes the number of cases with different mutations associated to PIDs reported in the cohort.

Discussion

Despite improved diagnostic techniques and molecular characterization of different types of PID, the majority of patients in worldwide registries are still without a definite genetic diagnosis. According to the findings of the updated analysis in our registry and the estimated prevalence rates for different forms of PID to date, only 2.3% of the expected (asymptomatic and symptomatic) patients have been diagnosed clinically, and a molecular diagnosis has only been identified in 33.1% of the registered cases. The rate of diagnosis at the molecular level is comparable with well-known registries around the world. Modell et al. at 2018 registered 102,097 patients from 86 countries spanning six continents, documented in the Jeffrey Modell Foundation Network registry. The most prevalent PID group was predominantly antibody deficiencies, accounting for almost half of the globally diagnosed patients. The most prevalent symptomatic PID entity was IgA deficiency, followed by common variable immunodeficiency (CVID) and DiGeorge anomaly [23]. Although this network was involved in several gene discoveries in the field of PID, less than 35% of the reported patient had a confirmed genetic defect [23].

The United States Immunodeficiency Network (USIDNET) provides advanced scientific research in the field of PID and epidemiologic indexes of patients registered in the largest national database with 6584 participants. They also reported PAD in 31.8% of household members with a specific diagnosis for PID followed by chronic granulomatous disease (8%) and thymic defect (DiGeorge anomaly, 7.8%). Genetic diagnosis also was ascertained in 36.5% of patients mainly with mutations in X-linked genes including BTK, CYBB, and WAS (https://usidnet.org [3]). The ESID registry also contains more than 20,000 patients where PAD represents the most common category (56.8%), where CVID is the main PID entity (21%). The PID genetic cause was known in approximately 36% of all registered patients and mutations in BTK, 22q11.2, and ATM were the most frequent recorded defects in this multinational registry (https://esid.org/Working-Parties/Registry/ESID-Database-Statistics [4]). Analysis of genetic diagnosis data from 4530 patients in the United Kingdom PID (UKPID) registry showed that 27% have a recorded genetic mutation affecting mainly BTK, 22q11.2, and CYBB (http://www.piduk.org [24]). Based on the updated report of the Latin American Society for Immunodeficiencies (LASID), this registry contains 6646 PID patients from 15 countries with an increased frequency of antibody deficiency compared to other registries (62.9%). Even though defects in BTK, 22q11.2, and CYBB are estimated to be the main genetic diagnosis form in this continent registry, the genetic composition remains unknown (http://imunodeficiencia.unicamp.br:8080/estatistica mensal.html [25]). Although PAD was the main clinical diagnosis in our patients (29.5%) with a prominent increasing rate of diagnosis compared to other forms of PID (Fig. 4), its proportion is slightly lower than that in Western countries, a phenomenon which is observed in many countries within the Middle East region and Asia [26]. In contrast to all of the above-mentioned studies, the three main causes of PID in the current study were MEFV, BTK, and ATM, out of which two have an autosomal recessive mode of inheritance, probably owing to the high rate of parental consanguinity in the genetically diagnosed patients. Indeed, many defective genes with autosomal respective inheritance that underlie PIDs were firstly described in the patients originated from the Middle East area with high rate of parental consanguinity. The DiGeorge anomaly is reported as one of the most common genetic defects in PID patients in most Western countries; in contrast, only 13 patients were identified in our study (~ 8% versus 1.2% of genetically diagnosed patients, respectively) [23]. The difference in our study may partly have been due to the obligatory presence of immunologic profile alterations for consideration of patients as PID. Moreover, a lack of referring patients with prominent non-infectious features (e.g., congenital heart disease, hypocalcemia, facial abnormalities) for immunologic evaluation may result in a lower rate of DiGeorge syndrome discovery. Furthermore, the role of genetic ethnicity should be considered as we observed among patients with X-linked PIDs a similar pattern of dominancy of BTK mutations (84.5% in our agammaglobulinemic patients compared to ~ 90% in Western cohorts), while a less frequent proportion of mutations in IL2RG (8.4% of our severe CID patients compared to ~ 60% in Western cohorts) and CYBB (29% of our chronic granulomatous disease patients compared to ~ 70% in Western cohorts) genes [3, 4].

The proportion of patients with a genetically definite diagnosis varied between 89.4% in patients with immune dysregulations and 3.2% in the cases with complement deficiencies. This wide spectrum might be due to the natural history of the disease, unknown underlying genetic defects, or complex disease pathogenesis including multigenetic or epigenetic/non-genetic etiologies. Of note, the patients with a lower genetic diagnostic yield also had the lowest percentage of clinically diagnosed cases signifying mild manifestations or non-infectious presentation. Our recent studies using enhanced techniques of detection of immunological abnormalities, mainly by targeted panel sequencing and whole exome sequencing, have led to earlier and more precise diagnosis of patients in both categories of antibody deficiency (improved from 7.8 to 12.2%) [10] and combined immunodeficiency (from 7.9 to 32%) [11]. A definite diagnosis of a PID allowed genetic screening and counseling in more than half of our patients’ cohort with the molecular diagnosis. Moreover, carrier detection and preimplantation testing in conjunction with in vitro fertilization helped more than 20 families to plan for their next child.

For the first time, we now report the comparative survival analysis on the 20-year follow-up of different categories of PID. Not surprisingly, CID patients had the highest mortality rate compared to other registered patients [11]. Despite an improvement in our national practice for therapeutic and prophylactic antibiotics for infections and the most common treatment options for PID patients including Ig replacement therapy and immune modulators (e.g., interferon-gamma, granulocyte colony-stimulating factor, and monoclonal antibodies), hematopoietic stem cell transplantation (HSCT) which is the mainstay of curative treatment in CID patients is available only for a limited number of patients due to restrictions in expertise and financial resources. Of the patients requiring HSCT, less than 100 patients (~ 3%) are under or awaiting therapy. Although the resources for HSCT in Iran are limited and no national program for newborn screening is ongoing, genetic diagnoses are pivotal for confirming clinical diagnoses, identifying new genetic carriers of CID genes, and diagnosing of pre-symptomatic individuals.

In order to improve the coverage of genetic testing for PID patients, several parameters should be considered including clinical investigation of the probands, and cost of the test, the probability of incidental findings, and the effects on relatives of the index patient. Therefore, a stepwise clinical, immunological, and genetic approach must be designed and used for different entities of PID. Integration of medical and family histories, findings of the physical examination and laboratory data, and confirmatory evidence for pathogenicity of the candidate gene variants would be important for decision making by clinical immunologists and immunogenetics experts. The approximate number of diagnosed PID cases was reported to be 7 per year in the 1980s, 30 per year in the 1990s, 58 per year in 2000–2006, 104 per year in 2007–2013, and 279 per year afterwards until the March of 2018 (Fig. 4). Although the definite molecular diagnostic rate of PID in Iran is almost similar to that in the developed countries, introduction of newborn sequencing for PID, increasing the rate of clinical diagnoses, providing more laboratory facilities and diagnostic modalities, and specialized immunogenetic centers for PID, in addition to improvement of current national PID network, should still be considered.