Gene Therapy for the Treatment of X-Linked Severe Combined Immunodeficiency
Hannah Wollenzien
Severe combined immunodeficiency (SCID) is a group of diseases characterized by T-cell counts under 1,500 cells per mm3, which is a phenomenon known as T-cell lymphopenia (TCL) [1, 2]. This inherited primary immunodeficiency leads to the absence of T cells, however B cells or natural killer (NK) cells may be present but are not always functional [4]. The lack of functional immune cells leads to recurrent infections in infancy and childhood and the disease is fatal in the first two years of life if it is not treated [1]. The exact number of babies born each year with SCID is not well known, as different countries have different inclusion practices but it is estimated to occur in 1:50,000 to 1:100,000 births, with most ethnicities seemingly affected equally [2]. The most common form of SCID is known as X-linked SCID (X-SCID) and is caused by mutations in the IL2RG gene on the X chromosome. IL2RG codes for the common chain (c) on cytokine receptors IL-2, IL-4, IL-7, IL-9, and IL-15. When cytokines bind to one of these receptors, JAK1 and JAK3 are activated, which phosphorylate tyrosine residues, including STAT5, which translocates to the nucleus and acts as a transcription factor (Figure 1)[3, 4]. The c chain plays an important role in T and NK cell development and may play a role in the function of B cells [5, 6]. The majority of pathogenic variants in the IL2RG gene are null mutations and cause no c chain to be expressed [2]. However, if the mutation is loss of function or missense, it can cause minimal c expression and function, and a patient may have atypical X-SCID, characterized by some mature T cells present, but with minimal function [2, 7].
X-SCID is a recessive X-linked disease, as the IL2RG gene is located on the X chromosome. The disease is seen exclusively in males, as they will only inherit one X chromosome. Mothers of one affected son are considered likely carriers and will be considered obligate carriers if more than one son is affected. A mother with only one affected son is not automatically considered a carrier because a progeny may have a de novo mutation. In fact, over half of X-SCID patients have no family history of immune disorders and are considered to be de novo cases. Having an affected child does not mean that fathers are carriers, as they will never pass their X chromosome to their son. Affected fathers will always pass the mutation on to their daughters, making them carriers. Traditionally, if a mother is a carrier of an X-linked disease, they have a 50% chance of passing along the mutation to any progeny. However, in X-SCID, there is germline mosaicism of the X chromosome in mothers, so they may not appear to be carriers by a blood sample genetic test but are in fact carriers of the mutation. Germline mosaicism makes it difficult to screen potential parents for carrier status, as the mothers will not always appear to be carriers. Additionally, a son affected because of germline mosaicism may appear to have a de novo mutation, leading to the conclusion that siblings are not at an increased risk of inheriting the disorder when they in fact are. As a consequence of these difficulties in determining carrier status, females are generally not tested for carrier status unless they have a son who is affected, and in general, genetic testing is most helpful if the son’s pathogenic variant in IL2RG has been identified [2].
Clinically, patients with X-SCID will appear normal at birth because of the presence of maternal antibodies and immunoglobulin that get passed along during development and birth. However, patients will start to develop recurrent infections as maternal immunoglobulin levels decrease a few months after birth. The most common infections are infections of the respiratory and gastrointestinal tracts, candidiasis, ear infections, streptococci, staphylococcus, and pneumonia. Patients will generally exhibit a failure to thrive and slow growth as well as development of food intolerances and the absence of lymphoid tissue [4]. As a consequence of not forming T cells, patients with X-SCID will have decreased thymic function and will often not have a thymus detectible on an X-ray [2].
Screening for primary immune disorders is mandatory at birth in a number of states and is gaining popularity abroad [1]. Newborn screening is done via quantification of T cell receptor excision circles (TREC) from a dried blood drop. TRECs are a DNA byproduct of T cell receptor recombination, which should occur as T cells are developed in the thymus. While this test is effective in identifying patients who have low counts of T cells, it is not able to determine the type of SCID a patient might have. It also is not effective in premature infants, as their thymus will not be producing T cells yet at birth [1]. Patients with a low number of TRECs will have their exact counts of B, NK, and T cells quantified using a blood smear to help identify the type of primary immunodeficiency they may have. If levels of T and NK cells are low, with seemingly ineffective B cells, a physician may suspect a case of X-SCID [4]. Table 1 outlines the numbers of immune cells for both affected and unaffected patients.
Table 1
Cell Type Affected Average Affected Range % of Affected Individuals Control Average Control Range
Total Lymphocytes <2,000 79% 5,400 3,400-7,600
T Cells 200 0-800 90%-95% 3,680 2,500-5,500
B Cells 1,399 44-3,000 95% 730 300-2,000
NK Cells <100 88% 420 170-1,100
Table 1: Numbers of immune cells in patients with X-SCID and non-affected controls. Adapted from [2]
The first step in diagnosis of X-SCID is ruling out infection with human immunodeficiency virus (HIV), as the clinical symptoms will often mimic HIV infection [4]. If clinical symptoms and immune cell counts are in line with what is suspected from X-SCID, the IL2RG gene will be sequenced. A mutation in the IL2RG gene will indicate that the patient has X-SCID and IL2RG sequencing is 99% effective in identifying patients with X-SCID [2]. It is vital to sequence the IL2RG gene to determine X-SCID diagnosis as mutations in JACK3 or IL-7R mimic the pathology of X-SCID but are different diseases [2]. On occasion, RT-qPCR or flow cytometry are used to quantify levels of immune cells [1, 4].
After a diagnosis of X-SCID is made, patients will immediately be placed in isolation to minimize the chance for infection. Any current infections will be managed and the parents will be counseled on how to manage symptoms in the future using respiratory therapists, nutritionists, and gastrointestinal specialists [4]. The currently accepted first choice treatment is a hematopoietic stem cell transplant (HSCT), which is often referred to as a bone marrow transplant. Patients who are candidates for a HSCT will have to undergo conditioning chemotherapy prior to their transplant to partially ablate the patient’s own bone marrow in an effort to avoid rejection [2]. Patients who undergo a HSCT have a five year overall survival rate of 74%, which drops to only 56% in the patients who required a second transplant. 80% of the deaths after transplant are due to infections and pulmonary complications. Five year survival rates are better for patients who received a transplant at a young age; 94% of patients who received their transplant under 3.5 months old survived to 5 years. The lowest survival rates are among patients who were older than 3.5 months and actively fighting infection at the time of their transplant. Only 50% of these patients survive for 5 years [8]. Sources of HSCT are varied and may come from bone marrow, peripheral blood, or umbilical cord blood [4]. Donations from human leukocyte antigen (HLA) matched siblings are the most effective, with 90% of patients surviving after one of these transplantations [2, 4]. 72% of patients who receive a donation from an HLA match have T cells present in their blood after just one transplantation, while B and NK cell counts still remain low [8]. HSCTs can come from non-HLA matched parent bone marrow but this will require killing the parental T cells to avoid the risk of graft-versus-host disease (GVHD) [2]. If GVHD occurs, the thymus can be damaged, which will completely ablate any future hope of future T cell production [9]. GVHD is a serious problem in patients who receive HSCT with an occurrence of 15% in two years after transplant [8]. Other concerns with HSCT include T cell senescence, donor rejection, issues with chemotherapy conditioning, and the need for long term infusion immunoglobulin therapy [2, 4].
In the last few years, clinical trials for gene therapy for the treatment of X-SCID have started to emerge. The gene therapy trials have thus far been targeted towards patients who cannot get HSCT or have failed a transplant already [2, 10]. The goal of gene therapy for X-SCID is to lead to the development of T cells in the thymus. This process is normally dependent on cytokine receptors and T cell receptors, which do not exist in patients with X-SCID. X-SCID is a good experimental model for gene therapy as T cell division happens fairly rapidly, T cells are long-lived, and newly created T cells will undergo positive selection in the thymus, allowing for the pruning of cells that do not properly undergo gene therapy [9]. Gene therapy for X-SCID works by taking the patient’s own blood marrow and extracting the CD34+ T cells, which are the multipotent progenitor cells capable of giving rise to a wide variety of immune cells [11]. The CD34+ T cells are transfected with a vector containing the IL2RG gene coding for the c chain, grown, and infused back in to patients [12]. Figure 2 presents an overview of the workflow of gene therapy for X-SCID. The appeal to gene therapy for X-SCID is the absence of the requirement for priming with chemotherapy, which decreases the side effects and increases the quality of life for patients. Additionally, the patients undergoing gene therapy will not be at risk for GVHD because their transplant is their own cells. Ideally, the presence of the IL2RG gene will allow for reconstruction of T and NK cells, and improve the functionality of any B cells present, decreasing the need for intravenous immunoglobulin therapy [12].
The first X-SCID gene therapy trial used a moloney murine viral vector with the c transgene in two patients with X-SCID. One patient had a missense mutation in IL2RG which resulted in a receptor in the cellular membrane but without a cytoplasmic tail and therefore not functional. The other patient had a frameshift IL2RG mutation which deleted exon 6 and resulted in a truncated receptor that was not expressed on the membrane. Their CD34+ T cells were removed and transfected for three days, at which point they were infused back in to the patient. Around 40% of the infused cells contained the transgene. 15 days after transfusion, circulating T cells containing the transgene were able to be detected circulating in both patients by PCR. The transgene was expressed on the surface of both patient’s cells and expression determined by immunostaining was close to the expression of control patients. At 10 months after transfusion, the majority of T cells present were naïve. The proliferation of the X-SCID patient’s T cells in response to pathogens was in line with age-matched control T cell proliferation. The gene therapy patients had a normal response to tetanus and poliovirus after vaccination. A few months after gene therapy, the patients stopped receiving immunoglobulin therapy and had normal levels of IgG, IgA, and IgM. The T cells had the most profound response to gene therapy, although a small amount of NK cells recovered as well. A low number of B cells were found to contain the transgene but their function was not restored to that of patients without X-SCID. The patients had profound clinical improvement, preexisting infections cleared, and the patients were able to live normal lives at home with normal growth and psychomotor development [13].
At 30 months after initiation of the moloney murine virus trial, a total of five patients had been enrolled. Four of five patients had vast clinical improvement including the clearance of lung infections and diarrhea, the disappearance of GVHD syndromes, and improved nutrition. These four patients were able to leave the sterile environment and live at home. Their T cell levels were in line with normal controls and it appeared that almost 100% of the circulating T and NK cells had taken up the transgene. Their production of new T cells was normal and they had levels of CD4+ and CD8+ T cells consistent with levels of age-matched controls, although most were naïve (Figure 3). These four patients contained normal levels of antibodies against diseases they were vaccinated for including poliovirus, tetanus, diphtheria, and influenza. The fifth patient had no response to the gene therapy treatment. After 30 months, he had no detectable T cells in the blood and required an emergency HSCT. Overall, most patients that underwent gene therapy showed a functional immune system, improved clinical outcome, and the development of a thymus at 30 months after transfusion. The four successful patients saw appearance and number of T cells faster and higher than patients who had received HSCT. Considering that most of the T and NK cells contained the transgene, the researchers concluded that the expression of the c chain confers a survival and growth advantage [12].
At ten years after the initiation of the moloney murine virus study, a total of ten patients had been enrolled. All patients had normal or very slightly low T cell counts after a few months, which remained at ten years. CD3+, CD4+, and CD8+ T cell levels were relatively normal in all patients and appeared to be dependent on the number of cells transfused in to the patients, with TREC tests being equal to normal control patients. All patients were producing naïve T cells capable of maturing in to any T cell the body might need. In the four patients tested for NK cell levels, they were normal. All patients had normal proliferative response to pathogens and the appropriate number of diverse T cell receptors. Early on in the study, patients had a population of B cells that was roughly 1% of their immune cells but this fell to less than 0.1% after gene therapy. This is likely to do with the fact that the c chain has nothing to do with the development of B cells. However, only one patient required immunoglobulin infusion therapy. The transgene inserted randomly in to the CD34+ T cells and there was a wide variety of integration sites both within and between patients. Unfortunately, four of the ten patients developed T cell acute lymphoblastic leukemia (T-ALL). One patient died after several rounds of chemotherapy and an emergency bone marrow transplantation [14]. The other three T-ALL patients seemed to initially respond to gene therapy as the others did, seeing a clearing of their preexisting infections, and then rapidly deteriorated and developed T-ALL. They responded to chemotherapy and regained the initial gene therapy benefit observed [15]. Overall, the X-SCID patients treated with the c transgene in a moloney murine virus had a sustained reconstruction of their T cell pool and adequate protection from pathogens. This came with the risk of T-ALL. Additionally, even with the death from T-ALL, patients treated with the moloney murine viral gene therapy had a better survival rate than patients with HSCT [14].
To determine why the four patients in the moloney murine gene therapy study developed T-ALL, the integration of the transgene was evaluated. In the eight X-SCID patients included in the integration study, there were 9,767 unique sites of transgene integration. They found an association between the number of cells successfully transfected and the number of unique integration sites and T cell diversity in a single patient. Since the vectors contained a short, noncoding unique barcode sequence, they were able to trace the lineage of individual mature T cells to determine how many transfected cells gave rise to circulating T cells. It was found that fewer than one in 100 transfected cells gave rise to functional, circulating T cells. The majority of the transgenes inserted near histone marks for promoters and transcription start sites (such as H3K4me1, H3K9me1, H3K27me1, H2AZ, bound RNA polymerase II). Integration was also found to be likely to occur at gene promoter regions such as CpG islands, DNAase1 sites, and actively transcribed gene regions. The repressive H3K9me3 mark was negatively associated with integration. Most sites of integration were in the 5’ end of a gene, near the first intron or upstream of the transcription start site near genes for leukocyte activation, lymphocyte differentiation, apoptosis, phosphorylation, transcriptional control, cell cycle progression, and tumor necrosis factor signaling. In general, in each patient, clustering of integration sites was observed, with the clustering increasing over time, inferring that clustered integration sites confer a survival advantage [16]. Patients who developed T-ALL had more integration sites near the promoter of cancer causing genes like CCND2, SPAG6, LMO2, BMI, STIL, TAL, NOTCH1, MYB, and CDKN2A (Figure 4) although healthy patients had integrations at CCND2 and SPAG6 as well, so integration at just this locus is not enough to induce T-ALL [15, 16].
The four patients with T-ALL had 617 unique integration sites including integration at known oncogenes [15]. Figure 4 maps the genes affected in the patients affected with T-ALL. The long terminal repeat promoters included in the transgene likely increased transcription of oncogenes, leading to the cancer phenotype [15, 17]. is unlikely that the cases of T-ALL were caused by increased expression of IL2RG as its expression on cancer cells was normal. Additionally, downstream of cytokine signaling is the JAK3 pathway, which was not overexpressed in the T-ALL patients, further excluding IL2RG overexpression as a driver of oncogenesis [15, 17].
The likely gene driving the development of T-ALL was found to be LMO2, a transcription factor that is a central regulator of hematopoiesis. The patients who had a large number of transgene insertions near the promoter for LMO2 had more effective and rapid expansion of their T cells immediately after transfusion. However, the counts of T cells skyrocketed from 80,000 cells per mm3 to over 300,000 cells per mm3 in 34 months (Figure 5 A, B). Before the development of T-ALL, these patients had a variety of transgene integration sites but after the diagnosis of T-ALL, the patients only had T cells that had transgene integration upstream of the LMO2 promoter. In addition to having increased expression of LMO2, these patients had other chromosomal abnormalities including a partial chromosome 6/chromosome 13 translocation (Figure 5 D), a trisomy 10, and a SIL-TAL1 fusion protein. It is most likely that T cells that took up the transgene at the promoter of LMO2 (Figure 6) had a growth advantage over other insertion sites, causing them to proliferate and take over, eventually causing T-ALL [17].
A second clinical trial evaluated the use of a pMFG gammaretroviral vector with a gibbon ape leukemia virus envelope for gene therapy of patients with X-SCID. The ten patients enrolled in this trial did not have HLA matches for a HSCT and did not require chemotherapy priming prior to gene therapy. At the end of 8-9 years of follow-up, all patients had CD34+ T cells that seemed to be in proportion to the number of T cells transfused. All patients had detectable CD3+ T cells and 60% of these patients had normal levels within six months after therapy. 90% of patients had CD4+ T cells within normal range, 60% had CD8+ T cell populations within the normal range. There was an initial increase in the level of NK cells and B cell recombination but this was not sustained to the end of the study. All patients had clearance of their preexisting viral infections, normal growth and development, and were discharged from the hospital to live normal lives at home. The treated patients had a normal T cell proliferation to pathogens and response to vaccines, as well as expected levels of IgG and IgM. All patients except for three were able to cease intravenous immunoglobulin therapy, and the continuation of immunoglobulin therapy was temporary in one of these three. 60% of the patients had a normal amount of variation in their T cell receptors and all patients had at least some T cell receptor variation, which indicates that the transgene was successfully transfected in to the CD34+ T cells. There was wide variation of integration sites and sites did not appear to be associated with physiologically relevant loci, such as LMO2. Unfortunately, one patient developed T-ALL but responded favorably to treatment and is now in remission [18, 19].
An ongoing trial involves a variation on the moloney murine virus vector, that is self-inactivating. In mouse models, this vector completely restored the number and function of T, B, and NK cells, without causing tumors or inserting at LMO2 [20]. In a two year study of the vector in humans, eight of the nine enrolled patients are still alive. The single fatality was due to a previously existing adenovirus infection, and was not due to the gene therapy. One patient required an umbilical cord transfusion and was excluded from the data analysis. Of the seven remaining patients, all had clearance of previous infections. Six of the seven patients had normal levels of CD3+, CD4+, and CD8+ T cells. Normal TRECs were observed and the development of the thymus was clear on X-rays. All patients had normal T cell proliferation to antigens and immunizations. While all patients demonstrated some amount of T cells, NK cells were present in only some of the patients, there was no B cell maturity in any patients, and all patients are still receiving immunoglobulin therapy. There has not been any cases of T-ALL yet in these patients. However, the latest age of T-ALL development among patients in previous gene therapy studies was 5 years after therapy, so the possibility of T-ALL development cannot be ruled out. The integration sites of the transgene do not appear to be any different than in the original moloney murine virus study, although there were fewer integration clumps around oncogenes in the current study (Figure 7). These patients will continue to be monitored for the next several years for adverse effects [21].
Gene therapy has thus far been shown to be an extremely effective treatment for X-SCID. In patients who are not healthy enough for a HSCT or do not have a HLA matched transplant, gene therapy to replace the c gene, IL2RG, allowing them some hope of a functional immune system. The possibilities of gene therapy for X-SCID are growing rapidly, with many clinical trials currently enrolling patients in phase I and II studies. Approaches include self-inactivating gammaretroviral vectors (ClinicalTrials.gov identifier: NCT01129544), lentiviral gene transfer (NCT01512888), self-inactivating lentiviral vector (NCT03217617), lentivirus with bisulfan conditioning (NCT03311503), and two studies of lentiviral vectors in children older than two years of age (NCT01306019 and NCT03315078). As effective as gene therapy for the treatment of X-SCID is, it comes with the risk for the development of T-ALL. Including the single death from T-ALL, 95% of the 20 patients who have gone through the long term trials have survived [22]. When compared to the 74% overall survival rate in HSCT [8], it is evident that gene therapy may be a more effective therapy. Current and future trials will need to weigh the risks and benefits of gene therapy for all diseases, and carefully consider the implications of altering the human genome, especially in clinical trials for childhood diseases such as X-SCID [22]. Despite the risk for adverse events such as T-ALL, the initial data supports gene therapy as a viable, and potentially safer and more effective substitute for treatment of X-SCID.
1. van der Spek, J., et al., TREC Based Newborn Screening for Severe Combined Immunodeficiency Disease: A Systematic Review. J Clin Immunol, 2015. 35(4): p. 416-30.
2. Allenspach, E., D.J. Rawlings, and A.M. Scharenberg, X-Linked Severe Combined Immunodeficiency, in GeneReviews(R), M.P. Adam, et al., Editors. 1993: Seattle (WA).
3. Leonard, W.J. and J.X. Lin, Cytokine receptor signaling pathways. J Allergy Clin Immunol, 2000. 105(5): p. 877-88.
4. van der Burg, M. and A.R. Gennery, Educational paper. The expanding clinical and immunological spectrum of severe combined immunodeficiency. Eur J Pediatr, 2011. 170(5): p. 561-71.
5. Gaspar, H.B., K.C. Gilmour, and A.M. Jones, Severe combined immunodeficiency–molecular pathogenesis and diagnosis. Arch Dis Child, 2001. 84(2): p. 169-73.
6. Waickman, A.T., J.Y. Park, and J.H. Park, The common gamma-chain cytokine receptor: tricks-and-treats for T cells. Cell Mol Life Sci, 2016. 73(2): p. 253-69.
7. Okuno, Y., et al., Late-Onset Combined Immunodeficiency with a Novel IL2RG Mutation and Probable Revertant Somatic Mosaicism. J Clin Immunol, 2015. 35(7): p. 610-4.
8. Pai, S.Y., et al., Transplantation outcomes for severe combined immunodeficiency, 2000-2009. N Engl J Med, 2014. 371(5): p. 434-46.
9. Fischer, A., S. Hacein-Bey-Abina, and M. Cavazzana-Calvo, Gene therapy of primary T cell immunodeficiencies. Gene, 2013. 525(2): p. 170-3.
10. Kuo, C.Y. and D.B. Kohn, Gene Therapy for the Treatment of Primary Immune Deficiencies. Curr Allergy Asthma Rep, 2016. 16(5): p. 39.
11. Schmitt, C., et al., CD34-positive early stages of human T-cell differentiation. Leuk Lymphoma, 1995. 17(1-2): p. 43-50.
12. Hacein-Bey-Abina, S., et al., Sustained correction of X-linked severe combined immunodeficiency by ex vivo gene therapy. N Engl J Med, 2002. 346(16): p. 1185-93.
13. Cavazzana-Calvo, M., et al., Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease. Science, 2000. 288(5466): p. 669-72.
14. Hacein-Bey-Abina, S., et al., Efficacy of gene therapy for X-linked severe combined immunodeficiency. N Engl J Med, 2010. 363(4): p. 355-64.
15. Hacein-Bey-Abina, S., et al., Insertional oncogenesis in 4 patients after retrovirus-mediated gene therapy of SCID-X1. J Clin Invest, 2008. 118(9): p. 3132-42.
16. Wang, G.P., et al., Dynamics of gene-modified progenitor cells analyzed by tracking retroviral integration sites in a human SCID-X1 gene therapy trial. Blood, 2010. 115(22): p. 4356-66.
17. Hacein-Bey-Abina, S., et al., LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1. Science, 2003. 302(5644): p. 415-9.
18. Gaspar, H.B., et al., Gene therapy of X-linked severe combined immunodeficiency by use of a pseudotyped gammaretroviral vector. Lancet, 2004. 364(9452): p. 2181-7.
19. Gaspar, H.B., et al., Long-term persistence of a polyclonal T cell repertoire after gene therapy for X-linked severe combined immunodeficiency. Sci Transl Med, 2011. 3(97): p. 97ra79.
20. Zhou, S., et al., A self-inactivating lentiviral vector for SCID-X1 gene therapy that does not activate LMO2 expression in human T cells. Blood, 2010. 116(6): p. 900-8.
21. Hacein-Bey-Abina, S., et al., A modified gamma-retrovirus vector for X-linked severe combined immunodeficiency. N Engl J Med, 2014. 371(15): p. 1407-17.
22. Deakin, C.T., I.E. Alexander, and I. Kerridge, Accepting risk in clinical research: is the gene therapy field becoming too risk-averse? Mol Ther, 2009. 17(11): p. 1842-8.