The Importance of Primary Immune Deficiency Registries

INTRODUCTION

Primary immune deficiencies (PIDs) represent a steadily growing group of more than 400 different disorders caused by defective or absent mutations in genes essential in the immune network. These rare disorders are treatable, in some cases, even curable, once recognized. Without proper recognition and treatment, these defects can be chronic, serious, or even fatal. Because of a general lack of awareness, diagnostic challenges, complexities, and unique differences, the diagnosis of PIDs can be difficult. Even when recognized, the management of PIDs can be complex. Patient registries are a crucial tool of medical science, as they allow one to gather widespread clinical and laboratory data on selected patient populations that can be used to foster advances in scientific research and ultimately improve patient care. Registries are particularly important for rare diseases, as no one medical center can collect sufficient data to define the actual landscape of these illnesses. From such data, one can estimate the numbers of subjects with these defects, the likely complications, and where new efforts are needed.¹ This was early recognized in PID in many countries, due to the increasing number of these defects, coupled with the extensive clinical and laboratory differences noted between the various conditions. To cope with understanding these genetic defects and their clinical and laboratory features, investigators began using standardized data collection instruments to collect information on patients about 3 decades ago. An early development of a European registry started with an effort in 1994 in Sweden, in collaboration with additional centers in England, Italy, and Spain. This early work compiled limited data on 9707 patients from 26 European countries, creating a basis for research on PID, in particular for clinical studies.² Other early efforts in individual countries included Norway,³ Australia,⁴ Spain,⁵ Switzerland,⁶ and Sweden.⁷ The registry started in France in 2005, the Reference Center for PIDs (CEREDIH),⁸ is based on a strong network of all university teaching hospitals, with 130 clinicians, 30 diagnostic immunology laboratories, and a dedicated and highly trained staff.

In the United States, realizing the importance of registries in this field, the Immune Deficiency Foundation (IDF) first began collecting data in 1992 under a contract from the National Institutes of Health (NIH) on patients with chronic granulomatous disease (CGD). Subsequently, the contract was extended, and registries were formed to collect data on patients with X-linked agammaglobulinemia (XLA) and X-linked hyper-immunoglobulin M (X-HIGM), common variable immune deficiency (CVID), severe combined immune deficiency (SCID), DiGeorge syndrome (DGS), leukocyte adhesion deficiency, and Wiskott-Aldrich syndrome (WAS).⁹ With this as a basis, and the growing realization that these diseases were likely to help elucidate the normal functions of the immune system, the National Institute of Allergy and Infectious Disease and the National Institute of Child Health and Human Development of the NIH posted a competitive research opportunity to build and enhance a PID patient registry, along with other research-enhancing resources. To apply for this research funding, the IDF formed a research program, The United States Immunodeficiency Network (USIDNET) in October 2003. According to its official definition,⁹ the USIDNET is “a research consortium established to advance scientific research in the field of PID, by assembling and maintaining a registry of clinical data from patients with PID, providing education and mentoring for young investigators in the field of PI, and acting as a resource for clinical and laboratory research.”¹⁰ At the formation of USIDNET, the importance of this kind of unified network between patient registries and clinical immunologists was highlighted in a 2004 editorial.¹ Initially based on paper submissions, with improvements in data capture, a web-based entry system went live in 2008. However, patients still had to participate in the registry via one of the enrolling sites at this time. New funds were obtained in 2013, from the Patient-Centered Outcomes Research Institute initiative, to allow all interested patients in the United States to enroll, and the IDF formed a new means for self-enrollment, called PI CONNECT. In 2016, again to improve data entry, language capture, and data homogeneity, the data migrated to a secure, user-friendly REDCap platform.

Having been supported by the NIH since 2004, the USIDNET registry today contains extensive clinical and laboratory data on more than 7815 patients, with multiple PID diagnoses (Table 1), 37 immune defects, and 2125 identified gene mutations (Table 2). Two thousand six hundred eighty-six have enrolled via the PI CONNECT link. Using the registry data, the IDF and the USIDNET registry have now generated more than 50 publications and more than 50 abstract presentations at national and international conferences. In this paper, the authors review the literature that has been supported by data from the USIDNET registry and how these data have advanced the study and treatment of these diseases.

Table 1.

List of primary immune deficiency diagnosis in the USIDNET registry and their frequency (as of January 2020)

Diagnosis	n
Agammaglobulinemia	449
Ataxia telangiectasia	28
Autoimmune lymphoproliferative syndrome (ALPS)	64
Autoinflammatory disease	12
CHARGE syndrome	5
Chronic granulomatous disease	560
Combined immune deficiency	98
Common variable immune deficiency (CVID)	1805
Complement deficiency	26
DiGeorge syndrome	559
Dyskeratosis congenita	2
Ectodermal dysplasia with immunodeficiency (NEMO and others)	30
HLH, including XLP and pigmentary disorders	69
Hyper-IgE syndrome	104
Hyper-IgM syndrome	158
Hypogammaglobulinemia	213
IgA deficiency	72
IgG subclass deficiency	27
Immune deficiency with syndromic features (not otherwise listed)	9
Immune dysregulation	97
Immunodeficiency unknown cause	24
Immunodeficiency with myelodysplasia (GATA2 and others)	68
Interferonopathy (Aicardi–Goutieres syndrome and others)	9
Leukocyte adhesion deficiency	11
Mucocutaneous candidiasis	53
Neutropenia	5
NK cell defect	4
Omenn syndrome	8
Other immune deficiency—known cause	63
Other T-cell problems	8
Predisposition to severe viral infections	30
SCID	355
Specific antibody deficiency with normal Ig concentration and normal number of B cells	99
Susceptibility to mycobacteria (MSMD)	11
TLR pathway abnormality	2
Transient hypogammaglobulinemia of infancy with normal number of B cells	14
Wiskott-Aldrich syndrome	250

Abbreviations: MSMD, Mendelian susceptibility to mycobacterial diseases; NK, natural killer.

Data from The United States Immunodeficiency Network (USIDNET). Registry-Reported Statistics. Available at: https://usidnet.org/registry-data/stats-registry-enrollment/. Accessed Jan 2020.

Table 2.

Frequency of primary immune deficiency diagnostic gene mutation variants in the USIDNET registry (as of January 2020)

Gene Defect	n
11–22 trans/dup at 22q11	1
15q13.3	1
ACAT9	1
ADA	66
AICDA (AID)	3
AIRE	68
AK2 (Hermansky–Pudlak syndrome)	4
ARTEMIS (DCLRE1C)	28
ATM	10
BCL11B	1
BTK	422
C2	14
C3	2
C8	1
CARD11	4
CARD11 (GOF)	3
CARD9	4
CASP10	1
CASP8	2
CBS	1
CD16	2
CD18	1
CD19	3
CD21	1
CD3D	2
CD40 (TNFRSF5)	6
CD40LG	107
CD79a	1
CD86	1
CECR1 (ADA2)	7
Cernunnos (XLF)	2
CFTR	1
CHD7	2
CR2	1
CREBBP	1
CTLA4 (ALPSV)	22
CXCR4	21
CYBA (p22-phox)	31
CYBB (gp91-phox)	357
DAO	1
DEL16q22	1
DEL22q11.2	478
DGCR2	1
DNMT3B (ICF1)	1
DOCK8	27
DQ	1
DUP15q21.1	1
ELANE	1
EPG5	1
F5	1
FOXN1	3
FOXP3 (IPEX)	13
GATA2	66
ICOS	1
IFIH1	1
IFNGR1	9
IFNGR2	1
IFT140	1
IGHM (m heavy chain), PTPRC (CD45)	1
IKBA (NFKIAB)	3
IKBKB	3
IL10	1
IL10RA	2
IL12B (IL-12p40 deficiency)	1
IL12RB1 (IL-12 and IL-23 receptor b1 chain)	3
IL2RA (CD25)	1
IL2RG (gc, CD132)	83
IL7RA	15
IRAK4	1
IRF2BP2	1
ITGB2 (LAD1)	9
JAK3	13
KRAS	1
LIG4	3
LPIN2, RFXAP, TMC8 (EVER2)	1
LRBA	4
LYST (Chediak-Higashi syndrome)	12
MAGT1 (XMEN)	6
MEFV (familial Mediterranean fever)	4
MHC class II	2
MTHFD1	1
MTHFR	2
NCF1 (p47-phox)	76
NCF2 (p67-phox)	14
NEMO (IKBKG)	25
NFKB1	5
NFKB2	5
NLRP3 (NALP3, CIAS1, PYPAF1)	6
OMENN	4
PARN	1
PGM3	2
PIK3CD (PI3 Kinase, GOF)	22
PIK3R1 (PI3 Kinase)	14
PNP	4
PRF1 (FHL2)	2
PRKCD	1
PTPRC (CD45 deficiency)	1
R229 & G319X	1
RAB27A	1
RAD50	1
RAG1	20
RAG2	10
RMRP (CHH)	17
RTEL1	1
SBDS (Shwachman-Diamond syndrome)	1
SERPING1 (C1 inhibitor deficiency)	5
SH2D1A (XLP1)	8
SLC7A7	1
SMARCAL1 (Schimke Immuno-osseous Dysplasia)	2
SMC3	1
SPINK5 (Comel-Netherton syndrome)	4
STAT1	13
STAT1 (GOF)	7
STAT3	50
STAT3 (GOF)	2
STS, WDR1	1
STXBP2 (Munc18–2 deficiency, FHL5)	8
TBX1	1
TLR3	1
TLR9	1
TNFRSF13B (TACI)	32
TNFRSF1A (TRAPS)	1
TNFRSF6 (ALPS-FAS)	9
TRAC (TCR-alpha deficiency)	1
TRNT1	2
TTC37	1
TTC7A	1
TWEAK (TNFSF12)	2
UNC13D (Munc13–4, FHL3)	12
UNG	1
Unspecified	2921
WAS	71
XIAP (BIRC4, XLP2)	15
ZAP70	5

Data from The United States Immunodeficiency Network (USIDNET). Registry-Reported Statistics. Available at: https://usidnet.org/registry-data/stats-registry-enrollment/. Accessed Jan 2020.

INSIGHTS FROM THE REGISTRY

One of the first and prominent uses of the registry has been to outline the clinical manifestations, laboratory evaluations, treatments, genetic causes, and outcomes of the individual diseases. Over time, these have resulted in definitive reports on the major disease classifications that have been cited hundreds of times in other publications, including CGD, XLA, HIGM, WAS, SCID, and CVID. These data have also been used to contrast the US patient populations with data registries collected in other countries to determine essential differences in disease ascertainment, microbial differences, treatments, and outcomes. A second main use of the registry has been to answer questions posed by physicians who care for patients. The following outlines the highlights of insights obtained from these many studies.

Chronic Granulomatous Disease

The first data collection formed by the IDF, spearheaded by Dr Jerry Winkelstein, was on subjects with CGD supported by a contract from the NIH. This included 368 patients, the majority (259) with the most common X-linked recessive form, and the rest had autosomal recessive or unknown inheritance forms.¹¹ This publication was the first to estimate the incidence of CGD in the United States, and it described pneumonia as the most common clinical manifestation (79% of the patients), with Aspergillus infections as the most frequent culprit, followed by suppurative adenitis, subcutaneous and liver abscesses, osteomyelitis, and sepsis. The publication demonstrated the increased mortality rate in the X-linked form, estimated at 5% per year, compared with 2% per year for the recessive form. One goal was to determine the health of these patients: a question brought to the forefront because interferon-gamma (IFN-γ) was an emerging therapy.¹² The high morbidity revealed that better treatments were required. Not surprisingly, these data were referenced the following year by Horwitz and colleagues,¹³ who presented the feasibility of nonmyeloablative conditioning followed by T-cell–depleted hematopoietic stem cell allograft transplant as a therapeutic option for patients with CGD with recurrent life-threatening infections and an HLA-identical donor in the family. The data concerning morbidity¹¹ were referenced by Gallin and colleagues,¹⁴ who designed a randomized, double-blind, placebo-controlled study for fungal infection prophylaxis with itraconazole in patients with CGD. This intended to address the severity of fungal infections that the registry had reported; it demonstrated a clear reduction of frequency of fungal infections in the itraconazole group. Similarly, the morbidity of the original report was cited by Marciano and colleagues¹⁵ when discussing the long-term results of IFN-γ therapy in patients with CGD.

The use of the CGD registry continues: Sacco and colleagues¹⁶ in 2018 examined the prevalence of end-stage renal disease among the patients with CGD in the registry, finding 14 patients with renal disease (2.7% of 516 patients in the database), with 10 of those 14 listed as having renal failure, more than previously described. The current role of allogeneic hematopoietic stem cell transplant (HSCT) among patients with CGD in the registry was also recently analyzed by Yonkof and colleagues,¹⁷ who reported 50 out of 507 patients were submitted to transplantation. Patients transplanted before or at the age of 14 years had improved survival.

X-Linked Agammaglobulinemia

A second patient population on which the IDF had early focused was XLA. As early as 1985, in a multicenter retrospective survey of 96 patients, Dr Winkelstein and collaborators had shown the 2 major causes of death in XLA were chronic pulmonary disease leading to cardiac failure and disseminated viral infections, which produced a dermatomyositis-like syndrome, hepatitis, pneumonitis, or meningoencephalitis. Ten years later, with the formation of USIDNET and increased XLA enrollment, Winkelstein and colleagues¹⁸ reported on 201 patients with XLA. The most common clinical presentation was infection (85% of patients); 11% had had neutropenia, which was a known occurrence, but the incidence had not been clarified. There was an average age of diagnosis of 2.6 years in those with an XLA family history (about half of the patients), whereas in its absence, the average age at diagnosis was older at 5.4 years. However, surprisingly, even in those with a known family history, only 34% were diagnosed before developing any symptoms. In this study, over a 4.25-year follow-up, the death rate was 3.75%. One of the known consequences of XLA (and some subjects with CVID) is that live poliovirus formerly used in immunizations could lead to chronic excretion of poliovirus and risk of complications such as chronic meningoencephalitis. In a study that used registry contacts and data, a search for chronic excretors was conducted among 306 patients in the United States, Mexico, Brazil, and the United Kingdom.¹⁹ Fortunately, no excretors were found among this set of studied patients.

What about adults with XLA? Winkelstein used the registry again 2 years later,²⁰ reporting on 25 patients, and described an increased frequency of hospitalization and missed work or school. However, reassuringly, the patients studied in this report showed a quality of life comparable to the general US population, suggested as possibly due to a higher income and level of education than the general population. This report concluded that, despite the disease impact, patients with XLA could do well.

The loss of immune globulin is compensated in XLA by adequate Ig therapy; for this reason, patients with this immune defect are generally considered to have a low risk of autoimmune or inflammatory disease. However, taking another look at XLA, another recent report using USIDNET data challenges this view. For 149 patients with XLA (average age 17.6 years, range 1–50), the majority (69%) reported having at least one inflammatory symptom. Of these, 28% had been formally diagnosed with an inflammatory condition. 21% reported symptoms of chronic diarrhea, and 17% reported abdominal pain; however, only 4% had been diagnosed with Crohn disease.²¹ Perhaps similarly, Barmettler and colleagues²² examined the frequency of gastrointestinal disease in the patients with XLA in the registry, finding diarrhea to be present intermittently in 38% of the patients, whereas 26% had chronic persistent diarrhea. Abdominal pain was present in 22%, and 9.5% of the USIDNET XLA population had a specific diagnosis of inflammatory bowel disease (IBD) and/or enteritis.

The USIDNET XLA registry data have also been used to elucidate a more global perspective on this immune defect–integrating USIDNET data with registries from other countries. A recent report on 783 patients from 40 centers around the world illustrates both the successes and the challenges for patients and physicians alike.²³ Here, acute and chronic lung diseases accounted for 41% of the deaths. Unusual complications such as IBD and large granular lymphocyte disease, among others, were enumerated. They were individually uncommon but collectively seen in 20.3% of patients. Survival greater than 20 years of age reached greater than 70% in Australia, Europe, and the Americas and was lowest in Africa (22%).

Hyper-IgM Syndrome

Another early data registry collection by the IDF was on the HIGM syndrome. In Italy, data collection on 56 patients with HIGM revealed many severe conditions, including neutropenia in 67%, diarrhea in 55%, sclerosing cholangitis in 19%, neurologic disease and/or unusual and aggressive tumors in 12.5%. All of these were more concerning, as they were more common than anyone center had appreciated.²⁴ These studies led to work on the inciting causes, cellular mechanisms, and the role of infectious agents and also suggested that HSCT should be considered for treatment.²⁵ In the United States, Winkelstein and colleagues²⁶ reported in 2003 on the clinical and immunologic features of 79 patients from 60 unrelated families, all presenting with significant IgG deficiency and most with IgA deficiency, but only about half of the patients with elevated IgM levels showing that, in many ways, the commonly used term “hyper-IgM” was a misnomer. The most common presenting history was infection with pneumonia in 81%, followed by upper respiratory infections (with agents including encapsulated bacteria but also opportunistic organisms such as Pneumocystis jirovecii). In contrast to the Italian study, sclerosing cholangitis occurred in 5 (8%) patients and in 4 of these was associated with Cryptosporidium infection. The registry of so many patients with this rare condition allowed immunologists to estimate an incidence of this defect in the United States of 1/1,030,000 live births. Based on the reported overall morbidly, Jain and colleagues,²⁷ seeking new therapies, used these data in their work, seeking a new therapeutic, recombinant CD40 ligand that could lead to partial immune reconstitution in patients with X-HIGM.

In 2016, Leven and colleagues²⁸ compiled an updated view of patients with HIGM in the registry, then containing 145 patients, almost double the prior reports.²⁶ Mutations were identified for 85 patients; 82 had a CD40 ligand mutation. The prevalence and types of infections were similar to the 2003 paper, as well as the prevalence of other symptoms, such as diarrhea. The data for the USIDNET cohort suggested a lower incidence of both sclerosing cholangitis and Cryptosporidium infection in the US as compared with the European study, possibly due to more recent increased emphasis on Cryptosporidium prevention measures such as the use of bottled water.

What about HSCT in HIGM? De la Morena and colleagues²⁹ examined the long-term outcomes of 176 patients with X-HIGM from several different national and international registries, including USIDNET. These data showed 38% of the patients had been treated with HSCT; however, there was no difference in overall survival between transplant-treated and nontreated patients. Despite that, HSCT patients had an improved quality of life/performance.

Severe Combined Immune Deficiency and Combined Defects

Since the establishment of the registry, one of the central patient populations registered were infants with SCID, as a national resource for infants with these rare conditions was needed for assessment, genetics, and treatment practices. One of the first subsets of patients included were infants with adenosine deaminase (ADA) deficiency. In 2016, 60 patients registered with ADA-deficient SCID, born after 1981, with 85% survival (but none survived with a birth year earlier than 1990). Twenty-three were transplanted; none of these had undergone gene therapy at that time. Using the USIDNET registry data as a resource,³⁰ Jennifer Puck and the SCID Newborn Screening (NBS) Working Group presented the rationale and strategy for the implementation of a population-based universal NBS for SCID, then implemented with NBS, beginning 2 years later. The USIDNET registry has become a data repository and resource for all infants now discovered in the United States; it will become a crucial tool in the assessment of how newborn discovery will improve survival, with hematopoietic transplant being an early option.³¹

USIDNET data have also been used to compare different countries. Differences in SCID detection in the United States, Middle East, China, Southeast Asia, and Australia were apparent in a 2013 USIDNET report.³² More recently, Al-Herz and colleagues³³ compared different variables regarding SCID (and combined immune deficiencies) in the USIDNET registry to infants in the Kuwait National PID registry (KNPIDR). The 69 patients with SCID in the KNPIDR were compared with 98 patients from USIDNET, noting the latter presented and were diagnosed at an earlier age. Although based only on the registry data, this study calculated the incidence of SCID/CID in Kuwait to likely be around 8 times higher than in USIDNET. The explanation may reside in the fact that near 70% of these patients in KNIPDR had a family history of PID compared with around 25% in USIDNET. Parental consanguinity greater than 90% was noted, compared with about 10% in the USIDNET data set. HSCT occurred at similar frequencies in both registries, but notably, 51% were matched related in KNIPDR versus 8% in USIDNET.

Common Variable Immune Deficiency

One of the largest collections of data in USIDNET concerns one of the more common and yet diverse of the immune defects, CVID. This resource has been steadily used for many publications over time, reflecting the range of complications that are characteristic of this immune defect. An early report of 700 patients showed that 56% were women; age at diagnosis for men was 25 years and 36 years for women. Complications at that time included infections (88%), autoimmune phenomena (22%; 11% of ITP and 6% of AIHA), and enteropathy (19%). Malignancy was a diagnosis in 13% of the patients, with 6% having a diagnosis of lymphoma. Other findings included bronchiectasis (7%), other chronic lung diseases (11%), splenomegaly (9%), lymphadenopathy (7%), and granulomatous disease (7%).³⁴ From the same group, Mehra and colleagues³⁵ used an algorithm to identify 296 chronically ill hospitalized patients with clinical features of immune defects.

With these data as an early baseline, the number of enrolled CVID subjects is now 1805, and it has been used to reexamine the same data points in much greater detail. For example, autoimmune cytopenias were recently reported in 10.2% of patients, with thrombocytopenia (7.4%), anemia (4.5%), and neutropenia (1%). Notably, patients with CVID with any cytopenia were more likely to have additional complications, such as lymphoproliferative disease (including lymphomas), granulomatous disease, liver disease, enteropathy, and interstitial lung disease, as shown, among others, by Feuille and colleagues,³⁶ confirming the view that immune dysregulation was a common feature with different outcomes. Separately, the rheumatologic manifestations of CVID were characterized by Gutierrez and colleagues.³⁷ Here, women and nonwhite patients with CVID were more likely to have a rheumatological disease. Lung disease continues to be a significant issue in CVID, and this has been studied in several publications. Weinberger and colleagues³⁸ compared the pulmonary manifestations between patients with CVID and XLA, finding more asthma, bronchiectasis, interstitial lung disease (ILD), pneumonia, and respiratory infections in the CVID group. Patients with ILD were more likely to have autoimmunity and bronchiectasis. For the patients with CVID, lower levels of T and B cells were observed. This possible association of cellular defects with chronic lung disease in the patients with CVID in the registry was also evaluated by Kellner and colleagues,³⁹ examining 1518 patients with CVID, comparing 138 (9.1%) who had ILD and 147 (9.7%) who had bronchiectasis, with the 1233 (81.2%) other patients with no reported chronic lung disease. Here also, patients with ILD had lower CD3, CD4, and CD8 cell counts; these patients also had an increased frequency of pneumonia, fungal, and herpes virus infections.

Another way to use USIDNET data is to compare these national data with that of individual medical centers that also collect substantial numbers of patients with PID. Farmer and colleagues⁴⁰ have used the resource in this manner, examining noninfectious disease “endotypes” of patients with CVID, by using a method of unbiased network clustering applied to both the USIDNET registry and the Partners data collection in Boston. From that analysis, unique patterns were defined for lymphoproliferative disease (2 clusters, with low class-switched memory B cells and low serum C3 in the total lymphoproliferative cluster), autoimmune disease (2 clusters), and atopic disease (1 cluster, with high serum IgE).

DiGeorge Syndrome

Patel and colleagues⁴¹ published a study in 2012 evaluating the scope of immunoglobulin deficiency in DGS patients from different patient registries, with data from 21 countries, including 1023 patients, 662 of which obtained from the USIDNET Registry. The study concluded an association of DGS with humoral immune deficiency, describing hypogammaglobulinemia in 6% of the patients with DGS in the group who were 3 years old and older, with 3% of the patients in the group requiring intravenous immunoglobulin (IVIG).

Autosomal Dominant Hyper-IgE Syndrome

Gernez and colleagues⁴² compiled the data on 85 patients with autosomal dominant hyper-IgE syndrome (AD-HIES) from the USIDNET registry; 45.9% had a family history of AD-HIES. The most common complications were cutaneous abscesses (74.4%), eczema (57.7%), and retained primary teeth (41.4%). The IgE mean level for the group was 8383.7 kU/mL, with eosinophilia present in 49.4%. Prophylactic antimicrobials were used in 56% of the cases for trimethoprim-sulfamethoxazole and 26.6% for antifungal agents; 30.6% of the patients had been placed on immunoglobulin replacement.

ANSWERING QUESTIONS

Physicians who care for subjects with CVID are often asked novel questions about this immune defect that the registry can answer. Since its inception, one of the main goals was to provide a place where such questions could be asked—this is done by using a standard “query” form available on the Internet (https://usidnet.org/registry-data/query-form/). Some of the many questions that have been asked and answered are summarized in the following section.

What Happened to the Patients Infected by Hepatitis C by Contaminated Immunoglobulin Products in 1993 to 1994?

The patients experienced bad outcomes, but physicians did not have the whole story, as it was a matter of litigation. Of 58 PCR-positive hepatitis C–infected patients, 30 were treated with IFN-α, in combination with ribavirin in 5 cases; 26 other subjects were not treated. Of those who were treated, 11 (37%) resolved the infection and became PCR-negative; of the 26 who were not treated, 5 (19%) resolved the infection. Patients aged 20 years or younger had a significantly better outcome compared with those older than age 20 years (P = .02). Five subjects had had liver transplantation, a sixth had had 2 transplants, and 10 (17%) of the group had died.⁴³

Do Patients with Primary Immune Deficiency on Intravenous Immunoglobulin Have a Risk of Stroke?

Since the first reports of IVIG-associated thromboembolic events, this question is commonly asked about patients with PID who need this therapy continuously. In fact, all Ig products bear a “black box” warning on the package insert. Basta compared the incidence of stroke among 1127 patients with PID on IVIG from the USIDNET with Centers for Disease Control and Prevention (CDC) data on the frequency of stroke in the general population of the United States. The results showed the overall prevalence of stroke in patients with PID to be approximately 4 times lower than in the general public (0.62 vs 2.6%).⁴⁴

What About Cancer in Primary Immune Deficiency? What Forms and Who Is at Risk?

The question of cancer in the patients with PID in the registry was analyzed by Mayor and colleagues,⁴⁵ which showed an overall 1.42-fold increased risk of cancer in these patients when compared with age-adjusted rates from a general population database. Interestingly, women had a similar rate to the general population, with men justifying the increased risk, with a 1.91-fold increase. Most of the increased risk was explained by increases in lymphoma in both genders (10-fold for men, 8.34-fold for women), with common cancers such as lung, colon, breast, and prostate found to have a similar incidence in the registry and the general population.

What About Transplantation Data?

The accumulation of data on the different PID, and the multiple papers originating from it, are a launch pad for future work, with data from different papers arising in registry data, quoted by de la Morena and colleagues³¹ when writing up a global review of the advances in transplantation for PID in 2014. That paper comprehensively reviews the remarkable advances in transplantation for patients with both SCID and non-SCID, noting the crucial role of multinational collaborative studies, such as the USIDNET registries, when trying to advance knowledge in the setting of rare disorders.

Who Has Granulomatous Disease in Primary Immune Deficiency and Why?

Leung and colleagues⁴⁶ estimated the frequency of granulomas in patients with PID through analyzing both data from the USIDNET registry and commercial insurance and Medicaid databases, with frequencies of 4.4% for USIDNET and of 1.2% to1.5% in the other databases, with the investigators defining the 1% to 4% range as a baseline to keep in mind for future studies. A landmark study by Perelygina and colleagues,⁴⁷ on the same topic of granulomatous disease in PID, also used data from the USIDNET registry. This was the unexpected finding that the granulomatous skin lesions of subjects with significant T-cell defects actually contained Rubella virus, not known until this time.

Do Patients with Primary Immune Deficiency Have Atopic Disease? And If So, Which Ones? What About Drug Allergy?

Patients have asked this question, and there has been no answer thus far. A study⁴⁸ found that overall, for 2923 patients, atopic disease and food allergy were less common in PID. Food allergy was reported in CD40 ligand deficiency (7.7%), XLA (7.1%), and hyper-IgE syndrome (HIES) (6.3%) with reactions including anaphylaxis in 20%. Also, atopic dermatitis was most commonly reported in patients with NEMO (62.5%), WAS (41.5%), combined defects (33.3%), selective IgM deficiency (33.3%), and HIES (25%). Patients with CVID, combined defects, and HIES reported both food allergy and atopic dermatitis. Although patients with CVID commonly report this, there is commonly no serum IgE, so is this true? Hartman and colleagues⁴⁹ examined 15 patients with CVID at their institution with self-reported penicillin allergy. After skin testing and graded drug challenge, none of the patients had immediate IgE-mediated hypersensitivity to penicillin, which paired with their nondetectable IgE. However, 35% of the CVID subjects in the USIDNET registry had a self-reported diagnosis of drug allergy (specifically beta-lactam allergy in 7%), as compared with a general population self-reported prevalence of 10%.⁵⁰ The investigators suggested that true penicillin allergy is likely much lower and needs to be addressed. Amplifying this, Lawrence and colleagues⁵¹ showed that undetectable total serum IgE in 75.6% of the patients with CVID in the registry (compared with only 3.3% in the general population) suggested that the absence of IgE was helpful in the diagnosis of CVID.

Are Children with Common Variable Immune Deficiency Different from Adults with Common Variable Immune Deficiency?

Sanchez and colleagues⁵² compared registry data on pediatric and adult patients with CVID and found few differences suggesting that this was more of a continuum; however, as one would expect, there was an increased frequency of otitis media and failure to thrive in the children. Looking at this from another standpoint, Yong and colleagues⁵³ assessed the immunologic and phenotypic associations of different levels of switched memory B cells in the pediatric CVID population.

What Are the Quality of Life and Perceived Health in Primary Immune Deficiency? What About Fatigue? What About Being Underweight?

Seeborg and colleagues⁵⁴ published data on perceived health of 1587 patients with PID, respondents to a national survey using the IDF database registry, using data from the USIDNET registry to compare the study demographics with other PID registries, finding a similar average age of 30 years. Patients with immune defects commonly say that they are overly fatigued, and this is a significant issue in their lives. To examine this, Hajjar and colleagues⁵⁵ queried the USIDNET registry data. Although 18% of the patients reported fatigue, as compared with a population prevalence of around 6% to 7.5%, those with primary antibody diseases (particularly CVID) had higher levels of fatigue, with 69% of the patients in this group reporting this. Ruffner and colleagues⁵⁶ looked specifically at underweight patients with PID in the USIDNET registry, finding granulomatous disease and earlier age of CVID diagnosis associated with the presence of underweight in the adult population. At the same time, for the pediatric population, underweight patients with CVID were associated with the presence of lymphopenia. The investigators also looked into the prevalence of obesity in the registry, which was similar to the general population.

What Does the Registry Tell Us About the Use of Immune Globulin Products in the United States?

Stonebraker and colleagues⁵⁷ examined data regarding the dosage of immunoglobulin replacement therapy used by USIDNET registry patients as compared with the medical literature and other databases. The investigators concluded that significantly different doses of immunoglobulin are used in different locations to treat patients with CVID, and concluded also that different doses are used when comparing patients with CVID with patient with XLA. They noted that an estimate of the latent therapeutic demand for each country would help to define what an adequate product supply for subjects with PID is.

SUMMARY

As seen earlier, the human, financial, and time investment placed in the creation of a project such as the USIDNET registry has had a significant impact on scientific advancements and understanding of PID as a whole, as well as of specific diseases. Naturally, the quantity and quality of the data in a registry depend on a constant addition of new data and maintenance/cleaning of the existing data, to guarantee the growth in knowledge does not plateau or, even worse, decrease. Data collection can reveal areas of low-diagnostic rates and potentially the lengths of diagnostic delay, commonly associated with increased morbidity and mortality. In fact, such regional differences were noted in the French registry.⁸ Patient registries provide information helpful to funding agencies and governmental bodies to plan educational programs and estimate treatments and their long-term costs. These data are also crucial to the Pharmaceutical Industry to plan for the supplies of medical products and to seek new areas of research.⁵⁸ Patients are essential stakeholders in this work, and for this reason, the IDF and PICONNECT, providing a means for patient enrollment of valuable and verified data, have been an essential ingredient of USIDNET. As the discovery of specific monogenic diseases and better understanding and definition of polygenic diseases has been growing at a fast pace and will tend to grow exponentially in the next decades, it is crucial that an excellent database of clinical data and a biorepository of patient samples pairs with all the information added by genetic studies. Continuing this process, expanding on it, and refining it will benefit us all as a scientific community, but, even more importantly, it will benefit our patients, reinforcing our goal of continually improving their quantity and quality of life.

KEY POINTS.

• Patient disease registries are essential data tools for many diseases, but particularly crucial for rare disorders, as most of the primary immune deficiencies.

• The United States Immunodeficiency Network is a research consortium, created to advance the field of primary immune deficiency that includes an extensive registry of patients with primary immune deficiency.

• Multiple points of data analysis have emerged from the registry, generating numerous publications and answering several research questions.