Rheumatological Diseases in Patients with Inborn Errors of Immunity in the USIDNET Registry

Purpose There is a gap in clinical knowledge regarding associations between specic inborn errors of immunity (IEIs) and rheumatological diseases. In this study, we report the frequency of rheumatological conditions in a large cohort of patients with IEI using the USIDNET (United States Immunodeciency Network) registry. G in various groups. the most common in patients with the in patients most in in patients with mucocutaneous candidiasis(CMC) (7.41%). report was highest in (HIES) and 523 DiGeorge syndrome patients. The rheumatologic condition with highest rate was vasculitis (6.5%) in WAS. Systemic sclerosis and arthritis were reported with lower rates in this group (0.41% and 0.41% respectively). DiGeorge syndrome patients had a lower rate of arthritis (0.19%) and systemic sclerosis (0.19%) and none had vasculitis. It was noted that 1.96% of HIES patients were diagnosed with SLE and 0.98% of HIES patient were reported to have vasculitis.


Introduction
Inborn errors of immunity (IEIs) encompass more than 400 heterogeneous congenital disorders that cause impaired immune function [1]. Although the individual prevalence of individual IEIs are largely unknown, the collective prevalence of IEIs is thought to be 1 in 1000 to 2000 individuals in the US [2]. Since the rst documentation of these disorders in 1952 by Bruton [3], described complications have evolved from only increased susceptibility to infections to the recognition of additional risks for autoimmunity, autoin ammation, atopy and cancer predisposition in many disorders.
A French Primary immunode ciency (PID) registry study showed a signi cant increase (120 times) in the risk of developing autoimmune cytopenia as well as an increase (80 times) in the risk of in ammatory bowel disease (IBD) in children with IEI [4].
The same study showed ten times higher risk of autoimmunity in this population as well. As expected, the presence of autoimmune complications resulted in increased mortality. Interestingly, T cell defects and common variable immunode ciency (CVID) were found to carry the highest risk of autoimmunity. When CVID population in the United States Immunode ciency Network (USIDNET) was evaluated, 5.9% of CVID patients were found to have a physician-diagnosed rheumatologic disease [5]. Furthermore, female gender was identi ed more frequently in patients with rheumatologic conditions and CVID in the US [5]. There are still many unknowns in demographics and biological risk strati cation of patients with IEI when considering rheumatologic diseases.
Although it is shown that primary antibody de ciencies are associated with increased autoimmunity and rheumatologic diseases [6][7][8][9], there is still a large gap in knowledge about speci c disease associations and outcomes. Limited studies were done on speci c IEIs showed interesting ndings. For example, the frequency of arthritis was found to be between 10-30% in patients with X-linked Agammaglobulinemia(XLA) [10]. Similarly, vasculitis was a well-documented complication of Wiskott-Aldrich Syndrome (WAS) and an association was found between systemic lupus erythematosus (SLE) and early complement de ciencies [11,12]. These ndings suggest that similar associations may be present with other IEIs and more work is needed to describe these complications. Strikingly, it is not uncommon to nd rheumatologic complications prior to IEI diagnosis as well [13]. In this study, we aimed to evaluate the frequency of rheumatologic conditions in USIDNET registry. This information can be helpful in guiding clinicians to risk stratify patients with IEI and hopefully reduce morbidity and mortality with early diagnosis and treatment of the rheumatologic diseases.

Methods
We performed a retrospective analysis of data obtained from the USIDNET registry. The U.S. Immunode ciency Network (USIDNET), a program of the Immune De ciency Foundation (IDF), is supported by a cooperative agreement, U24AI86837, from the National Institute of Allergy and Infectious Diseases (NIAID). We obtained de-identi ed patient demographic data including race, gender, age, living status, stem cell transplant status, rheumatologic and IEI diagnoses as well as underlying genetic mutation in patients with rheumatologic diagnoses. Due to a limited number of observations, we chose not to perform comparative statistical analysis.

Clinical data
In ammatory arthritis was described as any form autoimmune/autoin ammatory arthritis including rheumatoid arthritis, juvenile idiopathic arthritis and psoriatic arthritis. In addition to arthritis, our cohort included patients with SLE, in ammatory myopathies (dermatomyositis and polymyositis), vasculitis, Sjogren's syndrome as well as systemic sclerosis. Patients with nonspeci c arthritis and degenerative arthritis were excluded from the cohort. Sjogren's syndrome diagnosis covered both primary and secondary Sjogren's syndrome. Patients with RA and JIA diagnosis at the same time were only included in JIA category. In order to maintain accuracy of data, patients with both SLE and vasculitis were counted only as SLE given vasculitis can be part of the clinical presentation of SLE. Similarly, patients with psoriatic arthritis and RA were counted as only psoriatic arthritis. Our cohort had 2 patients with dermatomyositis and systemic sclerosis which is typically consider an overlap syndrome, these diagnoses were counted twice in total, one in each category.

Patient Demographics
A total of 5058 patients were reported in the USIDNET registry at the time of data query. Among those, 278 (5.49%) patients had both IEI and at least one rheumatologic disease diagnosis (Table 1). Caucasian ethnicity was the most reported race in our cohort. 62% of patients were female.

Distribution of rheumatologic diseases in USIDNET cohort
We rst investigated the frequency of rheumatologic diseases in the USIDNET population ( Figure 1). In ammatory joint disease (includes RA, JIA and psoriatic arthritis) was the most common rheumatologic condition reported (2.08%) (table 2).
ALPS was reported to have the highest rate of in ammatory arthritis (4.11%) (table 3). Following arthritis, the second most common rheumatologic condition in our cohort was vasculitis (1.01%). As expected, interferonopathies and WAS were the IEIs with the highest rate of vasculitis reported.
In addition to frequently seen rheumatologic conditions, rare rheumatologic diseases were also reported in the USIDNET registry. Among these, SLE and Sjogren's showed similar rates (0.67% and 0.69%, respectively). Interestingly, the rate of SLE was highest in chronic mucocutaneous candidiasis (CMC) (7.41%) followed by complement de ciencies (3.57%). In contrast, Sjogren's was not reported in these conditions. Instead, IgG subclass de ciencies and ALPS had the highest reported rate of Sjogren's (7.41% and 6.85%, respectively). As expected, systemic sclerosis and in ammatory myopathies had the lowest report rate in the cohort (0.19% and 0.24%, respectively). Reported systemic sclerosis rate was highest in familial HLH, whereas in ammatory myopathies had the highest frequency in agammaglobulinemia (2.99% and 1.65%, respectively) (table 3).
After identifying the rates of rheumatologic diseases in our cohort, we next investigated the speci c IEI diagnosis and associated rheumatologic diseases. Interestingly, certain IEIs were noted to have higher rate of speci c rheumatologic conditions.
Distribution of rheumatologic diseases in speci c IEI populations Immunode ciencies affecting cellular and humoral immunity: Overall, rheumatologic disease frequency in this group was noted to be 1.7%. At the time of the query, the USIDNET cohort had 339 severe combined immunode ciency (SCID), 92 nonspeci c combined immunode ciency (CID) and 156 hyper IgM syndrome (HIGM) patients. Among those, 1 SCID (0.29%) patient was reported to have in ammatory arthritis. No other rheumatologic diagnoses were reported in this group. Interestingly, nonspeci c CID patients were reported to have a higher rate of arthritis (3.26%), followed by vasculitis (2.17%), SLE (1.09%) and in ammatory myopathies (1.09%). Rheumatologic disease rate in HIGM patients was 1.28%, evenly distributed between in ammatory arthritis and vasculitis.
Combined immunode ciency with associated or syndromic features The USIDNET cohort included 246 WAS, 102 Hyper IgE syndrome (HIES) and 523 DiGeorge syndrome patients. The rheumatologic condition with highest rate was vasculitis (6.5%) in WAS. Systemic sclerosis and arthritis were reported with lower rates in this group (0.41% and 0.41% respectively). DiGeorge syndrome patients had a lower rate of arthritis (0.19%) and systemic sclerosis (0.19%) and none had vasculitis. It was noted that 1.96% of HIES patients were diagnosed with SLE and 0.98% of HIES patient were reported to have vasculitis.
Predominantly antibody de ciencies: Our query included 82 SAD, 425 agammaglobulinemia, 1739 CVID and 60 speci c IgA de ciency patients (SIgAD). Patients with SIgAD had a lower rate of rheumatologic disease in our cohort with one case (1.67%) of reported arthritis. Interestingly, patients with agammaglobulinemia showed an impressively different pro le with highest rate of in ammatory myopathies in the entire cohort (1.65%). This was followed by arthritis (1.41%), systemic sclerosis (0.47%), vasculitis (0.47%) and SLE (0.24%).

Disease of immune dysregulation:
Patients with immune dysregulation were reported to have arthritis (3.37%), vasculitis (3.37%), Sjogren's (2.24%) and systemic sclerosis (1.12%). Among 64 patients with HLH, 3 patients (4.68%) were reported to have a rheumatologic disease with 2 having systemic sclerosis and one patient with arthritis. Interestingly, 13.69% of ALPS patients had a rheumatologic condition and the majority of these cases were reported to be Sjogren's syndrome (50%).
Congenital defects of phagocyte number, function or both: Chronic granulomatous disease (CGD) was the second most commonly reported IEI in the USIDNET registry. We found that 2.05% of CGD patients had been diagnosed with a rheumatologic disease. SLE was the most common (1.49%) rheumatologic condition in CGD patients followed by vasculitis (0.37%) and arthritis (0.19%).
There were 57 reported cases of GATA2 defects in USIDNET. Among those, 4 patients (7.02%) were reported to have a rheumatologic diagnosis. Reported conditions were vasculitis (50%), arthritis (25%) and systemic sclerosis (25%). In contrast to CGD, there were no reported cases of SLE in patients with GATA2 defect.

Defects in intrinsic and innate immunity:
The USIDNET cohort included 27 CMC patients. Interestingly, SLE was the only rheumatologic condition reported in patients with CMC (7.41%).

Auto-in ammatory disorders:
Three patients were reported to have interferonopathies in the USIDNET cohort. Interestingly 2 of them (66.67%) were reported to have vasculitis. No other rheumatologic diseases were reported in this group.
Our query did not have patients with a rheumatologic diagnosis from IEI groups of defects in bone marrow failure and phenocopies of IEI.

Discussion
This is the rst USIDNET study to our knowledge to investigate the frequency of rheumatologic diseases among the various IEIs. Although it is widely accepted that IEI has higher frequency of autoimmune diseases in comparison to the general population, it is unknown if certain rheumatologic diagnoses are associated with speci c IEI types. Our results showed that the frequency of rheumatologic diagnoses have a wide spectrum in IEI and differ even in the same IEI category. For example, although in ammatory arthritis was the most commonly reported rheumatologic disease in CVID; in ammatory myopathies had the highest rate in patients with agammaglobulinemia. These differences could re ect the unknown differences in pathophysiology of autoimmunity in individual IEIs. In contrast to previous work on USIDNET CVID population, we did not detect a signi cant female dominance in our cohort [5]. This could be due to a larger sample size with other IEI included.
Frequencies of rheumatologic diseases are limited to small cohorts in the US, however rates of rheumatologic diseases in the USIDNET cohort was found to be higher in all rheumatologic diseases investigated ( Table 2). For example, frequency of rheumatoid arthritis in the US was reported in the range of 0.5-1% [14,15] with much smaller frequencies for JIA(0.086%) [16,17] and psoriatic arthritis(0.1%-0.25%) [18,19]. Whereas, arthritis rate in the USIDNET cohort was found to be 2.08%. Similar pattern was noted with SLE. Reported US frequencies of SLE was between 0.08%-0.15% [20][21][22]. Interestingly, the reported SLE rate in USIDNET was 0.67%. Limited studies showed frequency of systemic sclerosis in the US between 0.01% − 0.02% [23,24]. The rate of systemic sclerosis in our cohort was 0.20%. Similarly, estimated frequency of in ammatory myopathies in the US was reported as 0.01% − 0.02% [25,26], whereas the reported rate of this condition was 0.24% in our cohort. However, these results should be analyzed in the background of many limitations related to USIDNET cohort and comparison of small cohort studies.
There are multiple limitations to this study. Most importantly, although USIDNET is the largest cohort of reported IEI patients, it may not completely re ect the real-life population. More speci cally, it is possible that patients with both IEI and rheumatologic disease were overrepresented in our query considering the majority of IEI cases are followed in tertiary care centers if they have a rheumatologic condition as well. Another limitation is the di culty of comparing the general population with the USIDNET cohort from rheumatologic disease frequency standpoint. There is a signi cant gap in knowledge in the epidemiology of rheumatologic diseases in the general population; therefore, comparison between the USIDNET cohort and the general population cannot be made based on limited studies, especially for vasculitis, Sjogren's and systemic sclerosis. Lastly, this is a retrospective study involving multiple centers across the US which could introduce bias in terms of missing data or variability of clinical practice in diagnosing patients with rheumatological disease. Epidemiology of rheumatologic disease and IEIs remain as an area that require improvement in the eld of immunology.

Conclusions
Our results conclude that although the frequency of rheumatologic disease in IEI remains unknown, the rates of reported rheumatologic diseases in the USIDNET registry demonstrates a susceptibility to rheumatologic diseases in IEI with different rates in individual IEIs. Further studies would provide guidance to clinicians for detection of rheumatologic conditions earlier, therefore improving outcomes.

Declarations
Funding: None

Con icts of interest/Competing interests: None
Availability of data and material: Non-applicable Code availability: Not applicable