A Tale of Two SCIDs

See allHide authors and affiliations

Science Translational Medicine  24 Aug 2011:
Vol. 3, Issue 97, pp. 97ps36
DOI: 10.1126/scitranslmed.3002594


Hematopoietic stem cell (HSC) transplantation may be curative for severe combined immunodeficiency (SCID). However, for a majority of infants with SCID a suitable donor is not available, and even with a matched donor, allogeneic HSC transplantation itself carries potential complications such as graft-versus-host disease as well as side effects from myelosuppressive chemotherapy. In the past decade, substantial advances have been made in the transplantation of gene-modified autologous HSCs, especially for two forms of SCID: X-linked SCID (SCID-X1) and adenosine deaminase (ADA)–deficient SCID. Two new reports in this issue of Science Translational Medicine add to the accumulating findings from gene therapy trials in Italy, France, and the United States that show clinical benefits of this alternative treatment.

It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness, it was the epoch of belief, it was the epoch of incredulity, it was the season of Light, it was the season of Darkness, it was the spring of hope, it was the winter of despair, we had everything before us, we had nothing before us…”

Charles Dickens, A Tale of Two Cities

Treating human genetic diseases at the DNA level is an attractive alternative for those disorders that do not respond (or respond suboptimally) to currently available conventional treatments. As genes for human genetic diseases were identified and cloned, and as the methods for introduction of new genetic material into mammalian cells were improved, gene therapy moved from being a hypothetical technique to a reality in clinical trials. However, initial gene therapy trials performed in the 1990s did not produce the anticipated successes. Now, Thrasher, Gaspar, and colleagues report clinical benefits in extended follow-up results from two clinical trials performed in London for X-linked severe combined immunodeficiency (SCID) and adenosine deaminase (ADA)–deficient SCID (1, 2).

Although rare (1 in 200,000 to 1 in 1,000,000 live births), ADA-deficient SCID was identified as one of the first monogenic diseases for which gene therapy was feasible because normal human ADA complementary DNA (cDNA) had been cloned since 1983 (3), and the disorder had been well characterized biochemically (4). Thus, in 1990, under a protocol developed at NIH, two girls with ADA-deficient SCID were infused with mature T cells that had been transduced with a γ-retroviral vector encoding the ADA cDNA (5). Although there were no complications from the procedures, which included several cycles of leukopheresis, ADA gene transfer, and cell reinfusion, the subjects remained on ADA enzyme replacement therapy (ERT) throughout, so potential clinical benefits could not be determined. Two other trials using peripheral blood T cells were performed, but the results were also inconclusive (68).

It has been known for more than 40 years that a bone marrow transplant (BMT) from a healthy donor (allogeneic) can restore a functional immune system in patients with primary immune deficiency (PID) diseases because of engraftment of the donor’s hematopoietic stem cells (HSCs), which give rise to all types of mature immune cells. The allure of gene therapy for PIDs is that a patient’s own HSCs (autologous) could be corrected and retransplanted, curing the immunodeficiency without the immunological complications, such as graft-versus-host disease (GVHD) (Table 1). Thus, subsequent clinical trials for ADA-deficient SCID (and other PIDs) were focused on transferring the relevant gene into this blood cell progenitor population (often isolated by using the CD34 surface marker). However, initial HSC gene therapy trials for ADA-deficient SCID (6, 9, 10) showed no signs of efficacy, ostensibly because gene transfer was inefficient and resulted in low numbers of gene-containing peripheral blood cells. Additionally, subjects continued on ERT, which may have inhibited the growth advantage conferred to the corrected cells. Also, they did not receive any chemotherapy before reinfusion of the gene-modified HSC; chemotherapy could increase HSC engraftment by eliminating some of the endogenous HSC but increases the risk of toxic side effects.

Table 1.

Benefits and risks of allogeneic HSCT compared with autologous gene therapy HSCT.

View this table:

Improvements in techniques for the identification, isolation, gene modification, and culture of human HSCs, including producing γ-retroviral vectors at higher titers (11) and identifying cell adhesion and growth factors (12, 13)—which improved efficiency of gene transfer and promoted HSC survival during ex vivo cell culture, respectively—led to a second round of clinical gene therapy trials in the late 1990s. Starting in 1999, a gene therapy trial began at Necker-Enfants Malades Hospital in Paris, France, for SCID-X1, which is caused by mutations in the cytokine receptor common γ chain (γc), a critical component of the receptors for interleukin-2 (IL-2), IL-4, IL-7, IL-9, IL-15, and IL-21 (14). Deficiency of γc results in absence of T cells and natural killer (NK) cells, although pre-B cells are present but nonfunctional. SCID-X1 was seen as a model disease for HSC gene therapy because of the potential proliferative advantage of corrected cells: Corrected cytokine receptors could transmit survival and proliferative signals to leukocyte progenitors (15). Indeed, a spontaneous reversion of a mutant γc gene in a SCID-X1 patient led to partial and sustained correction of T cell deficiency (16, 17).

Within a year (in 2000), a similarly designed trial began at the San Raffaele Telethon Institute for Gene Therapy in Milan, Italy, to treat ADA-deficient SCID using a γ-retroviral vector for gene transfer to autologous bone marrow–derived HSCs. In contrast to previous ADA-deficient SCID gene therapy trials that used HSCs, the first two subjects treated in this study did not have access to ERT and, therefore, did not receive ERT before and after the gene transfer procedure. Additionally, before reinfusing the gene-modified HSCs the investigators administered a moderate dose of chemotherapy (busulfan) to condition the bone marrow and enhance cellular engraftment.

The initial published results of the first two subjects in each trial were remarkable (18, 19). All four subjects demonstrated an increase in their T cell counts, which were immunologically competent as assessed by proliferation to lymphocyte mitogens and to specific antigen stimulation, marking the first cases in which gene therapy conferred clinical benefits. Additionally, immunoglobulin levels normalized, and amelioration of immune deficient–related complications was noted with good health, growth, and development. In the past decade, similar trials for both forms of SCID have also been under way in England and the United States. The initial results of four subjects in the SCID-X1 trial (20) and one subject in the ADA-deficient SCID trial (21) in England and extended follow-up results from the French and Italian studies (22, 23) have been published. In all the studies, gene therapy was efficacious for a majority of the subjects.

Gaspar and colleagues now report extended follow-up results of 10 subjects on the SCID-X1 trial and six subjects on the ADA-deficient SCID trial performed in England (1, 2). In the trial for ADA-deficient SCID, six children between 6 months and 3.25 years of age were treated. Four subjects achieved immune reconstitution, with T lymphocytes rising to >500 to 1000/mm3 over 6 to 24 months. B cell reconstitution has been sufficient to allow intravenous γ-globulin replacement to be stopped in three subjects. However, the gene therapy was not efficacious in two subjects: One received a very low dosage of CD34+ cells (<0.5 × 106/kg), and the other had poor gene transfer to his stem cells.

In the trial for SCID-X1, 10 children between 10 months and 4 years of age were treated. Substantial reconstitution of T cell immunity occurred in all subjects, with nine subjects achieving normal T cell numbers at the latest follow-up. One subject developed leukemia on this protocol (24) and was treated with chemotherapy (see below). The subject is now in remission and has retained a highly polyclonal T cell receptor β (TCRβ) repertoire. Although B cell numbers were normal in nine of the 10 subjects, functionality of the B cells was not completely restored, as evidenced by the low levels of immunoglobulin production, suboptimal serological responses to antigens, and some infectious complications. Additionally, although there was an initial increase in NK cells after engraftment, these levels were not maintained. Despite these deficiencies, the investigators report that all subjects attend school and have normal development.

These new extended follow-up results from London add to the growing knowledge of HSC gene therapy for SCID-X1 and ADA-deficient SCID. Results from multiple clinical centers confirm the ability of autologous gene therapy to restore immunity in the majority of patients with these forms of SCID. The ADA-deficient SCID patients were treated with pretransplant chemotherapy, whereas the SCID-X1 subjects were not, but there were not more complications in the ADA-deficient patients. In fact, there may be better engraftment of gene-transduced HSC after the chemotherapy, which improves production of B cells and NK cells.

However, these positive results have been severely marred by the development of an unforeseen complication: a leukemia-like T lymphoproliferation occurring in five of the 20 SCID-X1 subjects between the Paris and London studies as a result of insertional oncogenesis from the integrating retroviral vector (24, 25). A major unresolved issue is why the complication of T lymphoproliferative disease developed in 25% of the SCID-X1 subjects but not in any of more than 30 ADA-deficient SCID subjects. A plethora of explanations have been considered to explain the discrepancy. One possible explanation is that the specific transgenes required to treat the two defects are notably different: The γc gene encodes a cytokine receptor chain capable of providing proliferative signals to cells, whereas the ADA gene encodes a housekeeping enzyme of purine metabolism needed for lymphocyte survival but that does not confer any activation signals. Additionally, subtle overexpression of γc and cellular protooncogenes, such as LMO2, that are transactivated by adjacent vector integration, may cooperate to deregulate cell growth. Also, the time courses for immune reconstitution are clearly distinct between the ADA-deficient SCID and SCID-X1 subjects, as reported in this issue as well as in the other trials. ADA-deficient SCID takes a slower course for T cell recovery (>18 months), whereas the SCID-X1 subjects have rapid increases in T cell numbers (within a few months). The more rapid cell proliferation in SCID-X1 may predispose the cells to acquiring additional mutations. Notably, the γc defect is intrinsic to T cells, such that only T cells expressing a functional γc transgene will develop and survive. However, ADA deficiency is amenable to cross-correction of uncorrected lymphocytes, resulting in less selective proliferative pressure on the transduced cells. Moreover, thymus damage that may occur in ADA deficiency may slow immune recovery (26). There may also be differences in the hematopoietic progenitors present in the marrow of SCID-X1 subjects compared with ADA-deficient SCID subjects: SCID-X1 children may have a larger population of progenitors poised to proliferate once their growth block is alleviated by adding the γc receptor. There has even been evidence that the γc-deficient background itself predisposes to transformation (27).

The hard-learned lessons about the risks from the gene therapy in these trials temper the optimism from the clear-cut clinical benefits realized. The first-generation γ-retroviral vectors used in the trials have retroviral long-terminal repeats (LTRs) to drive transgene expression, and these contain potent transcriptional enhancers, which can trans-activate adjacent cellular genes (Fig. 1). Understanding these events has led to the development of a next generation of γ-retroviral and lentiviral vector designed for greater safety. These new vectors are of the so-called self-inactivating (SIN) design, in which the enhancers of the LTR become deleted during the viral reverse transcription process. In SIN vectors, the transgenes are driven from internal promoters derived from cellular genes that lack strong enhancer activity (such as elongation factor α-1 and phosphoglycerate kinase) (Fig. 1). Vectors containing insulators—DNA sequences that can shield the surrounding chromosomal genes from the enhancer effects—are also being evaluated (28). Clinical trials using these putatively safer vectors are commencing in Europe and the United States for ADA-deficient SCID and SCID-X1. On the basis of the seminal observations with SCID, a broad array of blood cell diseases are being approached with gene therapy, including Wiskott-Aldrich syndrome, chronic granulomatous disease, X-adrenoleukodystrophy, metachromatic leukodystrophy, Hurler’s syndrome, and β-thalassemia. Gene therapy using autologous HSC transplantation (HSCT) could supplant allogeneic HSCT as the treatment of choice for these conditions if it can yield similar benefits with less morbidity and continue in its “spring of hope.”

Fig. 1.

First- and second-generation γ-retroviral and lentiviral vectors. First-generation γ-retroviral vectors used the enhancer and promoters (E/P) of the LTRs from the retroviruses to drive transcription of the exogenous gene. The strong enhancers are capable of trans-activating cellular proto-oncogenes that are adjacent to the sites of vector integration into cellular chromosomes (dashed lines). Second-generation γ-retroviral and lentiviral vectors have the enhancers of the LTR deleted and drive transcription of the exogenous gene using internal promoters (prom) that lack strong enhancer activity. These types of vectors have shown minimal potential to trans-activate cellular genes in preclinical models.



    References and Notes

    • Citation: K. L. Shaw, D. B. Kohn, A Tale of Two SCIDs. Sci. Transl. Med. 3, 97ps36 (2011).

    View Abstract

    Stay Connected to Science Translational Medicine

    Navigate This Article