Delineating the Clinical and Immunologic Characteristics: A Comparative Study of Inborn Errors of Immunity in Adult versus Pediatric Diagnosed

Inborn errors of immunity (IEIs) are rare, heterogeneous genetic disorders commonly diagnosed in childhood. The clinical presentation of IEI is highly variable due to the underlying genetic defects that impact the immune system’s development and/or function. Nearly 485 genetic mutations of the immune system were defined to cause IEI [1, 2]. While IEI has traditionally been diagnosed commonly in pediatric populations, there is an increasing recognition of cases diagnosed in adulthood, attributable to both delayed onset and improved diagnostic awareness. Notably, both germline mutations exhibiting reduced penetrance and acquired somatic mutations have been implicated in the adult onset of IEI. Crucially, the rise in diagnoses is associated with screening for immunodeficiency prompted by recurring and severe infections. In addition to infections, early-onset and resistant autoimmunity, malignancy, and immune dysregulation serve as other essential indicators of IEI [3]. Common variable immunodeficiency (CVID) is the mostly diagnosed immunodeficiency in adulthood. Nevertheless, even within this population, an average diagnostic delay of 10 years has been observed in patients with CVID [4]. Additionally, a heightened risk of mortality was noted in CVID patients with delayed diagnosis [5]. Consequently, delay in the diagnosis of CVID not only poses a significant clinical challenge but also contributes to an increased economic disease burden, estimated at USD 6,500 annually [6].

IEIs are increasingly diagnosed in adults, with over 50% of new IEI cases worldwide being identified in those over the age of 25 years [7]. It is approximated that CVID accounts for 35% of primary immunodeficiency diagnoses, IgA deficiency for 26%, agammaglobulinemia for 13%, and severe combined immunodeficiency, IgG deficiency, and chronic granulomatous disease each for 9% [8]. Despite the current data offering insights into the distribution of these diseases in adulthood, a comprehensive evaluation of childhood and adulthood onsets and their impact on the course of diseases still needs to be explored.

Herein, we aimed to investigate adult IEI patients through clinical and laboratory assessment and compare clinical features between adulthood- and childhood-diagnosed patients to identify disease-associated complications and pursue the outcomes. We also evaluated organ-specific involvements and the immunologic characteristics of adult patients with IEI diagnosed in childhood or adulthood.

Data on IEI patients were collected from hospital medical records between September 2019 and June 2023. Patients with ages older than 17 and diagnosed with IEI either in adulthood or in childhood were included. At the same time, secondary immunodeficiencies comprising HIV infection, malnutrition, malignancies, chronic renal and liver diseases, drug chemical-induced diseases, and metabolic diseases were excluded. The diagnosis and classification were made according to the ESID registry and Middle East Guideline Diagnostic criteria [1, 9]. For any patients diagnosed with IEI, the following data were gathered: age, gender, age at symptoms onset and diagnosis, past family history, referring symptoms, history of treatment, and current treatment modalities. The diagnostic delay was defined as the age at symptoms onset to the date of IEI diagnosis.

The latest screening images were also recorded, such as ultrasound, computed tomography (CT), and magnetic resonance imaging results. We recruited patients’ past endoscopies and all kinds of biopsies. Splenomegaly was defined as an increased spleen size greater than or equal to 12 cm for adults. Lymphadenopathy was detected on examination, ultrasound, and CT or magnetic resonance imaging. Enteropathy includes all cases of noninflammatory bowel disease and biopsy-proven infection and hyperlymphocytosis, defined as lymphocytic colitis. Malignancy encompasses hematological and solid tumor malignancies. Genetic analysis was performed using next-generation sequencing through a target panel or whole-exome sequencing (WES). The identified mutations were confirmed with Sanger sequencing.

We used IBM SPSS version 22.0 (IBM Corp., USA) and GraphPad Prism 8 (GraphPad Software Inc., San Diego, CA, USA) for statistical evaluations. Median and interquartile range (IQR) values for continuous variables and the frequency and percentage for the categorical variables were provided. Differences between ordinal data were evaluated with the Mann-Whitney U test and the Kruskal-Wallis test. Categorical variables were evaluated with the 2-tailed χ² or Fisher exact tests. Correlations were assessed with the Spearman correlation test. A p value <0.05 was considered the significance level for differences.

In the study period, between 2019 and 2023, 122 adult patients were followed as IEI at a single tertiary center in Istanbul. Of the patients, 89 (72.9%) were diagnosed in adulthood, aged over 17 years. The median current age was 29 (IQR: 23–40) years. The median age at the onset of the first warning symptom was 9.5 (IQR: 3–25) years, but the median age at diagnosis was 25 (IQR: 13–36) years, with a median diagnostic delay of 96 (IQR: 24–180) months (Table 1). The most clinical manifestation of IEI at onset was recurrent infection (46.7%). Autoimmunity was referred to as a rate of 18%. Interestingly, 20% of patients were diagnosed with screening for positive family history, severe asthma and allergy, and other reasons besides immunologic warning signs. In total, predominantly antibody deficiency (PAD) was the most commonly diagnosed immunodeficiency, and immunodysregulation (ID) was followed with a rate of 26.2%.

Genetic testing, either WES or targeted panel, was performed in 122 patients. Among these patients, 33 (27%) were undergone in childhood, whereas 89 (73%) were in adulthood. A genetic mutation consistent with clinical findings was detected in 62 out of 122 patients (50.8%). Among the 33 children, 17 had genetic mutations aligning with their clinical presentation, while 43 of the 89 adults had genetic mutations consistent with their clinical diagnosis. Figure 1 illustrates the genetic mutations identified in patients.

According to age at diagnosis, the median diagnostic delay was higher in adult-diagnosed patients (132 [IQR: 24–216] versus 24 [IQR: 12–90] months, p < 0.001) (Table 2). Among patients whose diagnoses and symptom onset occurred in childhood, otitis, and bronchiectasis were observed at higher rates. Bronchiectasis affected 49.1% of patients, and the tubular type was the most commonly seen in IEI patients. Figure 2a displays the diagnosis of IEI according to the IUIS classification in the whole cohort, and Figure 2b and c show the diagnosis stratified by age at the onset of symptoms in childhood or adulthood. Figure 3 illustrates the presenting symptoms of patients according to their age at symptom onset.

Malignancy was found to be significantly more prevalent in patients with adulthood onset compared to those with childhood-onset (8 [18.1%] vs. 4 [5.2%] patients, p = 0.028). Clinical and demographic features of patients according to age at the onset of symptoms are shown in Table 3.

Overall, 123 biopsies were recorded during the data collection. The results showed 15 different malignancies in 13 individuals (10.6%). The diagnosis encompassed 8 lymphomas (5 in lymph nodes [diffuse large B-cell lymphoma], 1 in testicular tissue [high-grade B-cell lymphoma], 1 in the ileal region [diffuse large B cell], 1 in the central nervous system), 2 gastric cancers, 2 thymomas, and 1 each of skin, breast, and colonic neuroendocrine cancer (Table 4). Table 5 summarizes clinical and radiologic findings in all cohorts.

Of 91 patients under IgG replacement treatment (IgRT), 64 (70%) were treated intravenously, while 27 (30%) underwent subcutaneous IgRT (SCIG). Approximately 44.6% of the patients were on a regimen of regular antibiotic therapy. Four patients diagnosed with CTLA-4 haploinsufficiency and LRBA deficiency received targeted therapy with abatacept.

Seven patients had a history of splenectomy. All patients underwent splenectomy before their immunodeficiency diagnosis and subsequent genetic testing. The indication for splenectomy was Evans syndrome in 4 patients and splenomegaly for 3 patients. Of these, 1 patient with LRBA mutation died due to colonic perforation after HSCT, while 5 were undergoing IgRT treatment. Notably, 1 patient with GIMAP5 mutation was not on IgRT or prophylaxis and had only 1 severe jaw bone osteomyelitis after splenectomy. The remaining patients were diagnosed with CTLA4 haploinsufficiency (n = 2), LRBA deficiency (n = 1), PI3KCD (n = 1), and autoimmune lymphoproliferative syndrome (ALPS) with no mutation (n = 1). After splenectomy, 2 patients with CTLA-4 haploinsufficiency had 2 severe infections that required intensive care unit admission and intubation before IgRT. No hospitalization or severe infection was observed after commencing IgRT.

In our cohort, 1 patient with Good syndrome died due to CMV infection, hydrocephaly, and shunt infection. Another patient with Good syndrome was under regular IgRT but had recalcitrant diarrhea. A patient with MAGT1 deficiency died due to lymphoma and polyneuropathy. One patient with unknown genetic mutations died due to CMV infection and disseminated intravascular coagulation.

The CD3⁺ T-cell counts were lower in the patients with hepatomegaly and splenomegaly (1,084/mm³ [IQR: 590–1,685] vs. 1,521/mm³ [IQR: 1,029–1,986], p = 0.007 and 1,094/mm³ [IQR: 607–1,539] vs. 1,550/mm³ [IQR: 1,029–1,971], p = 0.017, respectively). Likewise, in patients with non-lymphoproliferation, the CD3⁺ T-cell counts were higher (1,532/mm³ [IQR: 685–1,685] vs. 1,124/mm³ [IQR: 1,040–1,969], p = 0.016). A moderate negative correlation was observed between CD3⁺ T-cell counts and spleen size (r = −0.4221, p < 0.001).

Higher CD4⁺ T-cell counts were detected in females compared to males (732.5/mm³ [IQR: 511–1,125] vs. 596/mm³ [IQR: 306–871], p = 0.013). There was a significant moderate negative correlation between spleen size and CD4⁺ and CD8⁺ T-cell counts, respectively (r = −0.3956, p < 0.001, r = −0.3304, p = 0.002). In patients with hepatomegaly and splenomegaly, the CD4⁺ T-cell counts were significantly reduced than those without (529/mm³ [IQR: 261–788] vs. 726/mm³ [IQR: 484–997], p = 0.004, 537/mm³ [IQR: 272–775] vs. 717/mm³ [IQR: 468–1,110], p = 0.010, respectively). A similar result was observed in CD8⁺ T cells, where diminished absolute numbers were observed in patients with splenomegaly (479/mm³ [IQR: 320–698] vs. 670/mm³ [IQR: 467–914], p = 0.009).

A weak negative correlation was identified between CD19⁺ B-cell counts and liver size (r = −0.2521, p = 0.015). The CD19⁺ B-cell counts were significantly lower in the patients with pneumonia and bronchiectasis (102/mm³ [IQR: 33–207] vs. 200/mm³ [IQR: 121–337], p < 0.001, 102/mm³ [IQR: 33–218] vs. 168/mm³ [63–294], p = 0.040, respectively). Interestingly, the patients with a history of recurrent pneumonia displayed considerably reduced CD16⁺56⁺ NK-cell counts (151/mm³ [IQR: 67–231] vs. 72.5/mm³ [IQR: 37–163], p = 0.031). Further, decreased CD16 ⁺ 56⁺ NK-cell counts were also detected in patients with bronchiectasis (119/mm³ [IQR: 58–235] vs. 72.5/mm³ [IQR: 34–159], p = 0.023).

The ratio of males to females in childhood-diagnosed patients was approximately 2:3, contrasting with an almost equal ratio observed in adult-diagnosed individuals. The evaluation of the clinical findings regarding diagnostic ages showed higher immune thrombocytopenia in adult-diagnosed patients than in child-diagnosed patients (18 vs. 1, p = 0.02). Likewise, hepatomegaly (n = 28 vs. n = 18, p = 0.02) and enteropathy (n = 18 vs. n = 1, p = 0.02) were more prevalent in adult-diagnosed patients. On the other hand, pneumonia and otitis are more common in child-diagnosed compared to adult-diagnosed (28 [84.8%] vs. 56 [62.9%], p = 0.020 for pneumonia; 22 [66.6%] vs. 24 [26.9%], p < 0.001 for otitis). In those diagnosed during childhood, bronchiectasis is observed at a significantly higher rate than in those diagnosed during adulthood (35 [75.7%] vs. 25 [39.3%], p < 0.001).

Immune cytopenias were observed in patients with certain genetic defects, including LRBAdeficiencies and CTLA-4haploinsufficiency, Bloom syndrome, MAGT1, GIMAP5, IKZF-1, TACI, FAS deficiencies, and PI3KCD gain of function. Malignancies were detected more frequently in patients with symptom onset in adulthood patients (8 [18.1%] vs. 4 [5.1%], p = 0.028). All malignancies were diagnosed at over 17 years. Lymphomas were diagnosed in 8 (6.5%) patients in our cohort (testicular, ileal, central nervous system, and lymph nodes). Among these, 4 patients identified genetic mutations in MAGT1, PI3KCD, LRBA, and BLM genes. The remaining 4 patients underwent WES, and none exhibited genetic mutations associated with IEI. Two patients had been diagnosed with Good syndrome after thymectomy due to thymomas. Two patients with a genetic mutation in LRBA had gastric cancers and good survival. One of LRBA patients had recovered from lymphoma before gastric cancer. One patient with CARMIL-2 deficiency had neuroendocrine cancer of the colon.

In patients diagnosed in adulthood, CD19⁺ B-cell counts were higher compared to those diagnosed in childhood (167 [IQR: 67–268] vs. 86 [IQR: 35–196], p = 0.040). In comparison, naive B cells were observed to be higher in adult-diagnosed patients (73.5% [IQR: 49–85] vs. 56.6% [IQR: 14–72], p = 0.002). Additional immunologic assessments, encompassing lymphocyte subsets and serum immunoglobulin levels analyses, revealed no significant differences between diagnosed adults and children.

In our cohort, 1 patient with Good syndrome died due to CMV infection, hydrocephaly, and shunt infection. Another patient with Good syndrome was under regular IgRT but had recalcitrant diarrhea. A patient with MAGT1deficiency died due to lymphoma and polyneuropathy. One patient with unknown genetic mutations died due to CMV infection and disseminated intravascular coagulation.

This study showed that the diagnostic delay remains an essential obstacle to improving outcomes for adult patients with IEI. The predominant clinical manifestations in all patients were infections followed by autoimmunity. Notably, 20% of patients were diagnosed through screening for positive family history, drug allergy, severe asthma, and allergic rhinitis beyond the 10 warning signs. PAD and ID were the most commonly diagnosed immunodeficiencies in the entire cohort, as well as in both child and adult-onset cases. Adulthood-diagnosed patients showed more autoimmune findings and lymphoproliferation. Malignancies occurred more frequently among patients whose symptoms initiated in adulthood. Our study provides important hits for the significance of early diagnosis in IEI patients. Additionally, it offers comprehensive information about the potential complications encountered during delayed diagnosis.

Infections are the most common warning signs of IEI. However, autoimmunity, malignancy, and allergic manifestations also increased the diagnosis of IEI with the awareness of immunologists. Despite recurrent acute infections in children with antibody deficiency, adults are usually referred due to chronic sinusitis or bronchitis [10]. According to the ESID register analysis findings, a significant majority (68%) of patients initially presented with infections alone, 9% exhibited immune dysregulation as their sole symptom, and another 9% presented with a combination of both. Early onset of IEI occurred in two-thirds of patients before age 6, whereas a notable proportion (25%) did not manifest symptoms until adulthood. Immune dysregulation typically surfaced as the initial clinical feature of IEI between ages 6 and 25, with a notable male predominance up to age 10, shifting to a female predominance beyond age 40. Conversely, infections were predominantly the first sign in patients presenting after age 30 [11].

A notable meta-analysis focusing on patients with CVID revealed that manifestations other than infections were observed in 25% of pediatric cases and 62% of adult cases. Interstitial lung disease and inflammatory bowel disease were notably more prevalent among adults [12]. According to our results, the initial clinical manifestation of IEI was predominantly recurrent infections (46.7%). In this cohort, pneumonia was observed at a higher rate in both adult-onset and child-onset patients. However, otitis was significantly more prevalent in individuals with symptoms onset during childhood. Autoimmunity was reported at a rate of 18%. Notably, 20% of patients were diagnosed through screening for a positive family history, severe asthma, allergies, and other reasons aside from immunologic warning signs. PAD emerged as the most frequently diagnosed immunodeficiency, followed by ID at a rate of 26.2%. Our findings showed that nearly half of patients with IEI may be admitted to clinics other than infections.

The European registry comprising 7,615 entries has documented 4,265 individuals aged 15 years. Late-onset IEI manifesting in adulthood in this cohort accounts for 42.8% B-cell deficiency, 4.8% T-cell deficiency, 2.7% phagocytosis defects, 4.9% late-onset complement deficiency, and only 0.9% other categories of IEIs [13]. It is noteworthy that the prevalence of IEI in countries with a low rate of parental consanguinity is higher in adults compared to children; however, this prevalence experiences a significant decrease after the age of 40 years [13]. Within this cohort, among patients with adult-onset manifestations, the diagnosis of PAD exhibited a higher prevalence at 59%, while ID accounted for 22% of diagnoses. CID was diagnosed at a higher rate of 6.8%. However, phagocytosis defects and phenocopies were observed at a rate of 4.5%. Our results are similar to those of the ESID registry. Nevertheless, it is crucial to systematically exclude secondary immunodeficiencies and certain anatomical defects when assessing adult patients with suspected primary immunodeficiency due to recurrent infections. These conditions are notably more prevalent than IEIs within this age group [14].

Specific physical examination clues are few in adults, except for disorders associated with autoimmunity, in particular autoimmune cytopenia, which are more frequent than in early childhood. Autoimmune cytopenia may be related to adult-onset adenosine deaminase deficiency, CVID, ALPS, autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy, Good syndrome, and idiopathic CD4⁺ lymphocytopenia [15‒17]. Recognition of IEIs in adult patients with autoimmune cytopenias is essential for the therapeutic decision. Clinically, high-dose immunoglobulin therapy, rituximab, and other immunosuppressants can be provided to CVID patients with idiopathic cytopenias. A retrospective study demonstrated that splenectomy can be effective in treating autoimmune cytopenia, but it also highlighted significant risks. Post splenectomy, diligent monitoring, and ongoing IgRT are essential to prevent severe infections, as splenectomy can aggravate existing immunodeficiency. The study found that mortality and morbidity rates after splenectomy were affected by factors such as other comorbidities, the patient’s overall health, and the success of IgRT [18]. Consequently, a careful risk-benefit assessment is necessary for each patient when considering splenectomy. Recent genetic advances allowed the identification of different monogenic immunodeficiency disorders and clinical phenotypes accompanied by pathogenic and clinical crossroads between immunodeficiency and autoimmunity. This advancement facilitates expedited diagnosis and offers the potential for more precise therapeutic approaches. These targeted treatments may ultimately reduce the necessity for splenectomy, although this potential benefit has yet to be conclusively established [19].

In contrast, in patients with ALPS, immunosuppressants such as mycophenolate mofetil or sirolimus should be used if corticosteroid therapy fails [16, 20]. In our patient cohort, autoimmunity was observed in 18% of cases, with a higher prevalence noted in adult-onset diseases at 22.7%. Immune thrombocytopenia constituted the most frequent autoimmune finding, followed by autoimmune hemolytic anemia, with rates of 15.5% and 11.4%, respectively. Other organ-specific autoimmunity, especially thyroiditis and diabetes, are also common in this cohort, which can be helpful for early diagnosis.

The largest study, analyzing 3,658 IEI patients between 2003 and 2015, reported that approximately 5% of these patients developed cancers, showing a 1.42-fold increased relative risk compared to the age-adjusted population. Male IEI patients exhibited a 1.91-fold higher cancer risk relative to their counterparts, whereas female IEI patients had comparable cancer rates to the age-adjusted general population. Notably, lymphoma rates were significantly elevated, with a 10-fold increase in males and an 8.34-fold increase in females with IEI. However, the incidence of common cancers such as lung, colon, breast, and prostate did not differ significantly from the control population [21].

In a study involving 1,318 IEI patients, 8% were diagnosed with malignancies in whom they had previously been diagnosed with PADs. Patients with CIDs, phagocyte disorders, and severe autoimmunity exhibited a notably higher malignancy risk, with incidences of 13%, 13%, and 12%, respectively, compared to other groups [22]. In our cohort of IEI patients, 8 individuals (6.5%) were diagnosed with lymphomas involving 5 cases in lymph nodes (all diffuse large B-cell lymphoma), 1 case in testicular tissue (high-grade B-cell lymphoma), 1 in the ileal region (diffuse large B-cell lymphoma), and 1 in the central nervous system. A systematic review involving 456 patients with IEI and malignancy identified the prevalence of unspecified non-Hodgkin lymphoma at 37%, diffuse large B-cell lymphoma at 15%, and Hodgkin lymphoma at 13%, with comparable incidence rates between males and females. The most commonly reported disorders were antibody deficiencies and DNA repair disorders, primarily Nijmegen breakage syndrome and ataxia telangiectasia deficiencies. T-cell lymphomas were noted in 74 patients, predominantly males [23]. In our cohort, among patients who were diagnosed with and recovered from malignancy, there is a predominance of males. Among those with LRBA mutations, two developed gastric cancer yet had favorable outcomes; one had previously recovered from lymphoma. Additionally, a patient with CARMIL-2 deficiency was diagnosed with neuroendocrine colon cancer. Two had Good syndrome following thymectomy for thymomas. Genetic mutations in the MAGT1, PI3KCD, LRBA, and BLM genes were identified in patients with lymphoma. Four other patients had WES but showed no mutations linked to IEI.

Our study had a few limitations. This study encompasses patients with heterogeneous disease profiles and varying underlying genetic defects. So, interpreting results for a specific disease is difficult due to a few cases. However, we know IEIs are rare, and the prognosis varies among all patients. Second, the study is limited to patients who sought care at an immunodeficiency center, potentially introducing selection bias. The spectrum of immunodeficiencies in individuals not visiting such centers remains unknown, limiting the generalizability of our findings. Additionally, the timeline of diagnoses is unclear, particularly how many adult cases could have been identified in childhood. Future research should include a broader patient population and consider advancements in diagnostic methods to address these gaps.

In conclusion, this study found significant differences in diagnosis in the prevalence of infectious disease, autoimmunity, or ID between pediatric and adult-onset. Despite all advancements in genetics and laboratory technologies, diagnostic delays persist as a problem.

The authors thank Ahmet Ozen and Elif Aydıner for their collaboration. We thank all patients for their participation.

The study was approved by the Ethics Committee of Marmara University, Faculty of Medicine (ID No. 1353). All patients provided written informed consent to collect and analyze their medical records and participate in the study. Written informed consent was obtained from a parent/legal guardian or next of kin for participants under the age of 18 years to participate in the study. This study was conducted according to the Good Clinical Practice guidelines, the Declaration of Helsinki, and the local ethical and legal guidance.

The authors have no conflicts of interest to declare.

The authors received no funding for this manuscript.

Esra Karabiber and Safa Baris designed the study, collected the data, performed the statistical analysis, and wrote the manuscript.

Edited by: H.-U. Simon, Bern.

All data generated and analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.