Severe Combined Immunodeficiency, X-Linked

A number sign (#) is used with this entry because T-, B+, NK- X-linked severe combined immunodeficiency (SCID) is caused by mutation in the gene encoding the gamma subunit of the interleukin-2 receptor (IL2RG; 308380). See also X-linked combined immunodeficiency (312863), a less severe form of the disorder that is also caused by mutation in the IL2RG gene.

An autosomal recessive form of T-, B+, NK- SCID (600802) is caused by mutation in the JAK3 gene (600173) on chromosome 19p13. For a general phenotypic description and a discussion of genetic heterogeneity of autosomal recessive SCID, see 601457.

Clinical Features

Severe combined immunodeficiency differs from the Bruton type (300755) of agammaglobulinemia by the additional presence of lymphocytopenia ('alymphocytosis'), earlier age at death, vulnerability to viral and fungal as well as bacterial infections, lack of delayed hypersensitivity, atrophy of the thymus, and lack of benefit from gamma globulin administration. Severe combined immunodeficiency, originally termed 'Swiss type agammaglobulinemia' to distinguish it from Bruton agammaglobulinemia, was first described in Switzerland by Hitzig and Willi (1961). Those cases showed autosomal recessive inheritance (see 601457).

Rosen et al. (1966) reported 3 families with SCID inherited in an X-linked recessive pattern: all patients were male, and 1 kindred had 9 affected males in 5 sibships spanning 3 generations connected through females. Gitlin and Craig (1963) reported 15 boys with hypogammaglobulinemia and noted that they could be divided into 2 groups of almost equal size based on their clinical course. The first group had onset of infections early in life, often before 3 months of age, followed by lymphopenia and persistent pneumonitis, moniliasis, and frequent rashes. This disorder was uniformly fatal in infancy even in children treated with gammaglobulin. Autopsy showed an abnormally small thymus with thymic alymphoplasia. The second group of patients had onset of infections somewhat later, usually between 6 and 18 months of age. Infection was intermittent rather than persistent, and gamma globulin was clinically useful. These patients did not have lymphopenia, and in those who died, the thymus was not found to be small, although lymph nodes lacked germinal follicles and plasma cells. About half the patients in each group had a family history of severe infections in male relatives. The first group would be known now to have X-linked severe combined immunodeficiency and the second group X-linked agammaglobulinemia of Bruton.

Miller and Schieken (1967) suggested that one form of 'thymic dysplasia' is X-linked. Thymic dysplasia is seen in SCID (Nezelof, 1992). An impressive pedigree with 6 affected males in 3 generations was published by Dooren et al. (1968), who, following the recommendations of a workshop on immunologic deficiency diseases in man (Sanibel Island, Fort Myers, Fla., Feb. 1-5, 1967), called the condition 'thymic epithelial hypoplasia.' In the same workshop, Rosen et al. (1968) noted that X-linked SCID had less profound lymphocytopenia than autosomal recessive SCID.

Yount et al. (1978) studied a child with X-linked SCID. Adenosine deaminase (ADA; 608958) and nucleoside phosphorylase (PNP; 164050) levels were normal. The patient had virtual absence of lymphocytes capable of rosetting with sheep red blood cells, absence of reactive skin tests, and lack of in vitro responses to mitogens, antigens or allogeneic cells. He had profound humoral immunodeficiency despite a plethora of B lymphocytes. The authors suggested that B cells were unable to undergo terminal differentiation into plasma cells capable of synthesizing and secreting immunoglobulins. A brother of the patient they studied died at age 10 months of Pneumocystis carinii pneumonia complicated by disseminated influenza infection (Hong Kong strain). Autopsy showed a hypoplastic thymus without epithelial corpuscles and absence of germinal centers in lymph nodes and bowel lamina propria.

In 2 unrelated males with SCID and thymic alymphoplasia, Conley et al. (1984) found that T cells demonstrated a typical XX female karyotype and were probably of maternal origin, whereas the B cells had an XY male karyotype. The authors suggested that there was maternal lymphoid engraftment and that the SCID in these patients was the result of graft-versus-host disease (GVHD; see 614395). Since this would presumably affect only males, repetition in the family would simulate X-linked recessive inheritance.

Kellermayer et al. (2006) reported an infant boy with X-linked SCID confirmed by genetic analysis. Detailed cellular studies showed a subset of 46,XX CD4+ T cells in the patient's peripheral blood, indicating a chimeric lymphocyte population presumably derived from transplacental maternal T lymphocytes. The patient exhibited a mild to moderate recurrent eczematous rash consistent with spontaneous graft-versus-host disease from recognition of these maternal cells, and was scheduled for bone marrow transplant. Kellermayer et al. (2006) noted that although transplacentally acquired maternal T lymphocytes are present in 40% of SCID patients, untreated cases may still be fatal.

Speckmann et al. (2008) reported a boy with a relatively mild form of X-linked SCID diagnosed by molecular analysis at age 5 years (308380.0013). The main clinical symptom was recurrent bronchitis. Immunologic investigations showed decreased circulating T and NK cells, and normal numbers of B cells. Genetic analysis of peripheral blood cells showed a dual signal, with the wildtype IL2RG gene in T cells and a mutant IL2RG gene in B cells, NK cells, and granulocytes. His unaffected mother was a carrier of the mutation. The findings were consistent with reversion of the mutation within a common T-cell precursor in the patient. In vitro functional analysis showed normal T-cell function, despite low levels of T cells, and impaired B cell antibody response. A similar patient with reversion of mutation in a T-cell progenitor was reported by Stephan et al. (1996) (see 308380.0010). However, Speckmann et al. (2008) noted that the patient reported by Stephan et al. (1996) ultimately showed a deteriorating course and required bone marrow stem cell transplantation at almost 7 years of age. The findings indicated that close immunologic surveillance is still needed in patients with mutation reversion.

Other Features

X Inactivation

By examining a differential pattern of methylation (Vogelstein et al., 1987), Goodship et al. (1988) showed nonrandom X-chromosome inactivation in T cells of 2 obligate XSCID carriers. The method was used to distinguish autosomal recessive and X-linked forms of the disease and to demonstrate carrier status in the mother of a sporadic case.

Conley et al. (1988) analyzed patterns of X-chromosome inactivation in B cells from 9 obligate XSCID carriers. Using somatic cell hybrids to distinguish between active and nonactive X chromosomes, the authors found that all obligate carriers showed preferential use of the nonmutant X chromosome in B cells. The small number of B-cell hybrids that contained the mutant X were derived from an immature subset of B cells. The results indicated that the XSCID gene product was required for B-cell maturation.

Puck et al. (1986, 1987) showed that carriers for X-linked SCID could be detected based on analysis of X-inactivation patterns. In a control group of noncarrier women, Puck et al. (1992) found a wide range of X-inactivation ratios; 20 to 86% of T cells had the paternal X chromosome active, indicating random X-inactivation. Maximum likelihood analysis suggested that mature human T cells were derived from a pool of only about 10 randomly inactivated stem cells. X inactivation in XSCID carriers was markedly skewed, favoring the nonmutant chromosome. The authors developed a maximum-likelihood odds-ratio test which enabled prediction of carrier status in XSCID pedigrees.

Conley et al. (1990) studied X-chromosome inactivation patterns in T cells from 16 women who had sons with sporadic SCID. By analysis of human/hamster hybrids that selectively retained the active human X chromosome and use of an X-linked RFLP for which the woman in question was heterozygous, they showed exclusive use of a single nonmutant X as the active X in T-cell hybrids from 7 of the 16 women, identifying these as carriers of the disorder. Studies on additional family members confirmed the mutant nature of the inactive X and showed the source of the new mutation in 3 of the families. The most consistent finding in 21 patients with X-linked SCID was an elevated proportion of B cells.

By the study of X-chromosome inactivation patterns, Goodship et al. (1991) demonstrated that the mutation is expressed in B lymphocytes and in granulocytes as well as in T lymphocytes. They concluded that this disorder is not in a T-lymphocyte differentiation gene but rather in a metabolic pathway as in ADA deficiency (102700) and PNP deficiency (613179).

De Saint-Basile et al. (1992) reported 6 individuals in 2 sibships of a French family with severe infections. The propositus, a 5-year-old boy, had severe and progressive T- and B-cell functional immunodeficiency. The mother and 1 sister showed nonrandom X chromosome inactivation of T cells and, partially, of B cells but not of polymorphonuclear leukocytes, a pattern similar to that observed in X-linked SCID carriers. RFLP studies identified a haplotype segregating with the abnormal locus that may be localized in the proximal part of the long arm of the X chromosome. The authors suggested that the disorder may represent either a new X-linked immunodeficiency or an 'attenuated phenotype' of X-linked SCID.

Hendriks et al. (1992) raised the possibility of 2 distinct XSCID defects. They determined the pattern of X-chromosome inactivation in 14 females, including 6 obligate carriers, from 3 unrelated pedigrees with XSCID. All 6 obligate carriers showed nonrandom X-inactivation of the mutant chromosome in T cells. Four obligate carriers had nonrandom X-inactivation in B cells, and 4 did not, consistent with the observation that B cells with the XSCID mutation exhibit a relative maturation disadvantage rather than an absolute arrest in differentiation. In carriers from 1 pedigree, granulocytes had complete inactivation of the mutated X chromosome, whereas granulocytes from carriers from the other 2 pedigrees showed a random X-chromosome inactivation. The authors concluded that an XSCID phenotype with involvement of granulocytes represented an XSCID variant.

Wengler et al. (1993) demonstrated that all 4 lymphoid cell populations studied, NK cells, B cells, CD4+ T cells, and CD8+ T cells, from 3 heterozygous women exhibited exclusive use of a single X as the active X, whereas both X chromosomes were used as the active X in neutrophils and monocytes. The study was done by means of a PCR technique based on 2 observations: that active and inactive X chromosomes differ in methylation and that throughout the genome there are highly polymorphic sites consisting of sequences of 2-to-5 nucleotides that are repeated a variable number of times.

Clinical Management

Shortly after the discovery of the HLA system (Amos and Bach, 1968), Gatti et al. (1968) restored immune function in an infant with SCID by transplantation of bone marrow from his HLA-identical sister. Over the following decade, however, lethal GVHD was a major problem when bone marrow from HLA-mismatched donors was transplanted. In the late 1970s, studies in rats and mice demonstrated that allogeneic marrow or spleen cells that were depleted of T cells rescued the recipient from lethal irradiation without causing fatal GVHD, despite differences in MHC antigens between the donor and the host. Techniques developed in the early 1980s to deplete human marrow of T cells made it possible to restore immune function by marrow transplantation in patients with any form of SCID.

Borzy et al. (1984) reported a patient with SCID who had maternally derived peripheral blood lymphocytes identified by chromosomal heteromorphisms defined by the quinacrine banding technique. These markers were also used to monitor the successful engraftment of lymphocytes from a sister after bone marrow transplantation.

Flake et al. (1996) reported the successful treatment of a fetus with X-linked SCID by the in utero transplantation of paternal bone marrow that was enriched with hematopoietic cell progenitors. The mother had lost a previous son at 7 months of age to this disease. Studies of that child's DNA identified a splice site mutation in the IL2RG gene (308380).

Buckley et al. (1999) reported on the outcome of hematopoietic stem cell transplantation in 89 consecutive infants with SCID at Duke University Medical Center over the previous 16.5 years and the extent of immune reconstitution in the 72 surviving patients. Patients with X-linked SCID represented the largest category with 43 patients, of whom 34 (79%) survived. Other patients treated by Buckley et al. (1999) included 6 cases of JAK3 deficiency (600802), 2 cases of interleukin-7 receptor alpha deficiency (IL7R; 608971), and 13 cases of adenosine deaminase deficiency (102700). Twenty-one of the patients had autosomal recessive SCID of unknown cause. At the time of latest evaluation, Buckley et al. (1999) found that all but 4 of the 72 survivors had normal T-cell function, and all the T cells in their blood were of donor origin; however, B-cell function remained abnormal in many of the recipients of haploidentical marrow. Forty-five of the 72 children were receiving intravenous immune globulin. A striking finding of the study was that all but 1 of the patients who were younger than 3.5 months of age when they received a bone marrow graft had survived. The results emphasized the necessity of early diagnosis of the disorder, which should be considered a pediatric emergency. Whereas the absence of T cells prevented GVHD, mild GVHD occurred most often in patients in whom maternal T-cell engraftment, which occurred during pregnancy, was detected. This finding strongly suggested that most of the transient graft-versus-host reactions were actually graft-versus-graft reactions: T cells in the graft vs maternal T cells.

Rosen (2002) reported that the infant boy with X-linked SCID who received a successful bone marrow transplant from his HLA-identical sister in 1968 (Gatti et al., 1968) was in robust health 34 years later.

Ting et al. (1999) showed that DNA from hair roots was particularly useful for the diagnosis of X-linked SCID in children who had been subjected to bone marrow transplantation where no pretransplant blood had been stored. They performed mutation analysis in 13 unrelated boys who had had bone marrow transplantation. Five boys had an affected male relative. Mutations were found in 11 cases, 6 of which were sporadic, and maternal mosaicism was found in 1 family. Three mothers of the 6 sporadic cases were identified as carriers.

Gene Therapy

After preclinical studies, Cavazzano-Calvo et al. (2000) initiated gene therapy trials for X-linked SCID based on the use of cDNA containing a defective gamma-c Moloney retrovirus-derived vector and ex vivo infection of CD34+ hematopoietic stem cells. After a 10-month follow-up, gamma-c transgene (IL2RG)-expressing T and NK cells were detected in 2 patients. T, B, and NK cell counts and function, including antigen-specific responses, were comparable to those of age-matched controls. that

Cavazzano-Calvo (2002) noted that gene therapy for SCID is indicated only for those patients for whom a satisfactory HLA match is not available. Given an HLA match, bone marrow transplantation is the treatment of choice. In the absence of T cells in an affected son, T cells from the mother may persist in the affected son, resulting in graph-versus-host manifestations such as dermatitis and enteritis. After gene therapy with the patient's cells carrying a gamma-c transgene, the maternal T cells (marked by the XX chromosomes) decline in a reciprocal arrangement with the rise in T cells with the XY sex chromosome constitution.

Hacein-Bey-Abina et al. (2002) reported successful treatment of 5 SCIDX patients with autologous CD34+ bone marrow cells that had been transduced in vivo with a defective retroviral vector carrying the IL2RG gene (308380). Integration and expression of the transgene and development of lymphocyte subgroups and their functions were sequentially analyzed over a period of up to 2.5 years after gene transfer. No adverse effects resulted from the procedure. Transduced T cells and natural killer cells appeared in the blood of 4 of the 5 patients within 4 months. The numbers and phenotypes of T cells, the repertoire of T-cell receptors, and the in vitro proliferative responses of T cells to several antigens after immunization were nearly normal up to 2 years after treatment. Thymopoiesis was documented by the presence of naive T cells and T-cell antigen-receptor episomes and the development of a normal-sized thymus gland. The frequency of transduced B cells was low, but serum immunoglobulin levels and antibody production after immunization were sufficient to avoid the need for intravenous immunoglobulin. Correction of the immunodeficiency eradicated established infections and allowed patients to have a normal life.

Hacein-Bey-Abina et al. (2003) stated the results of their earlier studies (Hacein-Bey-Abina et al., 2002) had been confirmed in 4 additional patients with typical X-linked SCID who were treated by the same ex vivo, retrovirally-mediated transfer of the IL2RG gene into CD34+ cells. Of the first 4 successfully treated patients, 3 continued to do well up to 3.6 years after gene therapy, whereas a serious adverse event occurred in the fourth patient. At routine checkup 30 months after gene therapy, the patient was found to have integration of the provirus into 1 site on 11p within the LMO2 locus (180385), which had previously been reported as the basis of acute lymphoblastic leukemia arising from T cells with alpha/beta receptors, usually with the chromosomal translocation t(11;14). Between 30 and 34 months after gene therapy, the patient's lymphocyte count rose to 300,000 per cubic millimeter, and hepatosplenomegaly developed. Response to chemotherapy regimen was satisfactory at the time of report.

Marshall (2002, 2003) reported the development of leukemia in 2 children who received gene therapy. Hacein-Bey-Abina et al. (2003) demonstrated that in the 2 patients who developed T-cell leukemia after retrovirus-mediated gene transfer into autologous CD34 cells, the retrovirus vector integration was in proximity to the LMO2 protooncogene promoter, leading to aberrant transcription and expression of LMO2. Hacein-Bey-Abina et al. (2003) speculated that SCIDX1-related features may have contributed to the unexpectedly high rate of leukemia-like syndrome in their gene therapy-treated patients. They speculated that, because of the differentiation block, there were more T-lymphocyte precursors among CD34 cells in SCIDX1 marrow than in marrow of normal controls, thus augmenting the number of cells at risk for vector integration and further proliferation once the gamma-c transgene is expressed.

By searching a database containing the sequences of more than 3,000 retroviral integration sites cloned from mouse retrovirally induced hematopoietic tumors, Dave et al. (2004) identified 2 leukemias with integrations at Lmo2 and 2 leukemias with integrations at Il2rg (308380). One of these leukemias contained integrations at both sites. These integrations were clonal, suggesting that they were acquired early during the establishment of the leukemia. The authors noted that the probability of finding a leukemia with clonal integrations at Lmo2 and Il2rg by random chance was exceedingly small, providing genetic evidence for cooperation between LMO2 and IL2RG. Leukemia 98-031 had a T-cell phenotype and upregulated Lmo2 expression, a finding consistent with that seen in SCIDX1 patient leukemias. Dave et al. (2004) suggested that the results provided a genetic explanation for the high frequency of leukemia in the gene therapy trials. In transplant patients, IL2RG is expressed from the ubiquitous Moloney viral long terminal repeat. Although this was expected to be safe, Dave et al. (2004) concluded that retrovirally expressed IL2RG might be oncogenic due to some subtle effect on growth or differentiation of infected cells. Dave et al. (2004) further concluded that their results boded well for future gene therapy trials, because in most trials the transplanted gene is unlikely to be oncogenic and occurrences of insertional mutagenesis will be low.

Although gene therapy had been shown to be highly effective treatment for infants with typical SCIDX1, the optimal treatment strategy in patients with previous failed allogeneic transplantation and those with attenuated disease who present late in life was unclear. Thrasher et al. (2005) reported the failure of gene therapy in 2 such patients, despite effective gene transfer to bone marrow CD34+ cells, suggesting that there are intrinsic host-dependent restrictions to efficacy. The authors considered it likely that initiation of normal thymopoiesis is time dependent and suggested that gene therapy in such patients should be considered as early as possible.

The low frequency of homologous recombination in human cells was an impediment to permanent modification of the human genome. Urnov et al. (2005) reported a general solution using 2 fundamental biologic processes: DNA recognition by C2H2-zinc finger proteins and homology-directed repair of DNA double-strand breaks. Zinc finger proteins engineered to recognize a unique chromosomal site can be fused to a nuclease domain, and a double-strand break induced by the resulting zinc finger nuclease can create specific sequence alterations by stimulating homologous recombination between the chromosome and an extrachromosomal DNA donor. Urnov et al. (2005) showed that zinc finger nucleases designed against an X-linked SCID mutation in the IL2RG gene yielded more than 18% gene-modified human cells without selection. Remarkably, about 7% of the cells acquired the desired genetic modification on both X chromosomes, with cell genotype accurately reflected at the mRNA and protein levels. Urnov et al. (2005) observed comparably high frequencies in human T cells, raising the possibility of strategies based on zinc finger nucleases for the treatment of disease.

Hacein-Bey-Abina et al. (2010) reported the results of a 9-year follow-up of 9 SCID patients treated with retrovirus-mediated transfer of the IL2RG gene to autologous CD34+ cells. Eight of 9 patients initially had successful correction of the immune dysfunction, but 4 patients developed T-cell acute lymphoblastic leukemia, resulting in death in 1. Transduced T cells were detected for up to 10.7 years after gene therapy. Seven patients, including 3 with leukemia, had sustained immune response; 3 required immunoglobulin replacement therapy. Transduced B cells were not detected in long-term follow-up.

Mapping

De Saint Basile et al. (1987) mapped the X-linked SCID locus to Xq11-q13 by linkage analysis with RFLPs. No recombination was observed with marker DXS159. According to Mensink and Schuurman (1987), J. L. Mandel found close linkage with the DXS159 marker at Xq12-q13 in 6 pedigrees. They also suggested that there may be more than one X-linked SCID locus because there was immunologic heterogeneity.

Puck et al. (1988) found linkage with loci in Xq12-q21.3, but concluded that the exact localization remained uncertain and that heterogeneity might exist. Puck et al. (1989) performed linkage analysis in 6 kindreds using a random pattern of T-cell X-inactivation to rule out the carrier state in at-risk women. Their findings, combined with analysis of Xq interstitial deletions, allowed assignment of the locus to Xq13.1-q21.1 and defined flanking markers for prenatal diagnosis and carrier testing. Smead et al. (1989) found no recombination among SCID, PGK1 (311800), and DXS72. DXS72 is known to be distal to SCID because males with normal immunity have been described with Xq21 interstitial deletions involving DXS72. DXS159 and DXS3 appeared to be flanking markers for SCID. Goodship et al. (1989) demonstrated no recombination between IMD4 and DXS159, PGK1, or DXS72; the maximum lod score for linkage to PGK1 was 5.03.

Molecular Genetics

In 3 unrelated patients with X-linked SCID, Noguchi et al. (1993) identified 3 different mutations in the IL2RG gene (308380.0001-308380.0003).

Population Genetics

X-linked SCID is the most common form of SCID and has been estimated to account for 46% (Buckley, 2004) to 70% of all SCID cases (Stephan et al., 1993; Fischer et al., 1997).

In a study of 108 patients with SCID, Buckley et al. (1997) found that IL2RG deficiency and JAK3 deficiency accounted for approximately 42% and approximately 6% of cases, respectively.

Nomenclature

X-linked SCID was earlier referred to as 'Swiss-type agammaglobulinemia' or thymic epithelial hypoplasia (Nezelof, 1992).

Leonard (1993) suggested that the common gamma chain of IL2R be designated gamma-c, and that X-linked SCID be termed gamma-c deficiency XSCID. X-linked severe combined immunodeficiency has been known colloquially as 'Bubble Boy disease' because it was the abnormality in a patient who lived in an isolation unit in Houston for a prolonged period.

See review by Leonard et al. (1994).

Animal Model

By somatic cell hybrid analysis and methylation differences, Deschenes et al. (1994) demonstrated that female dogs carrying X-linked SCID have the same lymphocyte-limited skewed X-chromosome inactivation patterns as human carriers. In canine XSCID, Henthorn et al. (1994) demonstrated a 4-bp deletion in the first exon of the IL2RG gene, resulting in a nonfunctional protein.

In addition to XSCID caused by mutations in the common IL2RG gene, an autosomal form of SCID (608971) with T-cell deficiency occurs in patients with a mutation in the IL7R gene (146661). IL7 (146660) is vital for B-cell development in mice, but not in humans. Ozaki et al. (2002) developed a mouse model with a phenotype resembling human XSCID by knocking out the genes for both Il4 (147780) and Il21r (605383). Mice lacking only the Il21r gene had normal B- and T-cell phenotypes and functions, with the exception of lower IgG1 and IgG2b and higher serum IgE levels. After immunization with various antigens and with the parasite Toxoplasma gondii, the normal increase in IgG1 antibodies, as well as antigen-specific IgG2b and IgG3 antibodies, was significantly lower than in wildtype mice, and there was an uncharacteristic marked increase in antigen-specific IgE responses. In contrast, mice lacking both Il4 and Il21r exhibited lower levels of IgG and IgA, but not IgM, analogous to humans with XSCID. After immunization, these double-knockout mice did not upregulate IgE, indicating that this phenomenon is Il4-dependent, nor did they upregulate the IgG subclasses. The double-knockout mice, but not mice lacking only Il4 or Il21r, had disorganized germinal centers. Ozaki et al. (2002) proposed that defective signaling by IL4 and IL21 (605384) might explain the B-cell defect in XSCID.

To investigate the origin of T-cell lymphoma risk in XSCID patients treated with IL2RG gene therapy, Woods et al. (2006) expressed IL2RG inserted into a lentiviral vector in a murine model of XSCID, and followed the fates of mice for up to 1.5 years posttransplantation. Unexpectedly, 15 (33%) of these mice developed T-cell lymphomas that were associated with a gross thymic mass. Lymphomic tissues shared a common lymphomic stem cell, with similar vector-integration sites evident in the DNA of the thymus, bone marrow, and spleen of individual mice; however, no common integration targets were found between mice. Woods et al. (2006) concluded that IL2RG itself may be oncogenic to patients. They further cautioned that any preclinical experimental treatments involving transgenes should include long-term follow-up before they enter clinical trials.