Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Gene Therapy (2003) 10, 1999–2004 & 2003 Nature Publishing Group All rights reserved 0969-7128/03 $25.00 www.nature.com/gt REVIEW Gene therapy progress and prospects: gene therapy for severe combined immunodeficiency HB Gaspar, S Howe and AJ Thrasher Molecular Immunology Unit, Institute of Child Health, London, UK Severe combined immunodeficiencies have long been targeted as a group of disorders amenable to gene therapy because of their defined molecular biology and pathophysiology, and the prediction that corrected cells would have profound growth and survival advantage. Recently, several clinical studies have shown that conventional gene transfer technology can produce major beneficial therapeutic effects in these patients, but, as for all cellular and pharmacological treatment approaches, with a finite potential for toxicity. Gene Therapy (2003) 10, 1999–2004. doi:10.1038/sj.gt.3302150 Keywords: ; SCID-X1; ADA; PEG-ADA; LMO-2; insertional mutagenesis In brief Progress Sustained correction of X-severe combined immunodeficiency (SCID) has been observed in clinical trials. Cells in patients with adenosine deaminase (ADA) deficiency, treated by lymphocyte or stem cell gene therapy, persist and maintain transgene expression for many years. Withdrawal of PEG-ADA from patients treated by lymphocyte gene therapy for ADA-deficient SCID results in enhanced immunological reconstitution. Successful gene therapy for ADA-deficient SCID can be achieved in the absence of PEG-ADA and in combination with myelosuppression. Animal models of RAG-2- and JAK-3-deficient SCID have been corrected using similar strategies. Insertional mutagenesis has been observed in human studies, reinforcing the need to develop methods for optimization of protocol safety. Sustained correction of X-severe combined immunodeficiency (SCID) has been observed in clinical trials The most severe forms of primary immunodeficiency are known as SCIDs. These are a group of diseases in which Correspondence: Dr AJ Thrasher, Molecular Immunology Unit, Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK Prospects As a group of well-defined disorders, SCIDs are amenable to treatment by gene therapy, and an extended range will enter clinical study over the next few years. The durability of immunological reconstitution will determine the effectiveness and the need for repeated administration. The absolute risk of clinically manifesting mutagenesis using retroviral vectors is at present unknown, and will only be determined by the extended observation of more patients, and by the development of clinically relevant models to test for toxicity in a rigorous way. The characteristics of retroviral and lentiviral integration in human patients will be determined by mapping integration sites. Risks of mutagenesis will be reduced by improved design of vectors that restrict promoter activity to relevant cell types and within the domain of the therapeutic transgene. Development of targeted integration or of stable episomal vector systems will also enhance safety. Gene-repair strategies may have particular efficacy in SCID because of the profound growth and survival advantage conferred to corrected cells. T-lymphocyte development is invariably interrupted, and associated with diverse disorders of development and functionality of B lymphocytes, and natural killer (NK) cells.1 X-linked SCID (SCID-X1) accounts for approximately 50–60% of all SCIDs, and is caused by mutations in the gene encoding the common cytokine receptor gamma chain (gc). This is a subunit of the cytokine receptor complex for interleukins (IL) 2, 4, 7, 9, 15 and 21.2 In the absence of gc signaling, many aspects Gene therapy for severe immunodeficiency HB Gaspar et al 2000 of immune cell development and function are compromised. The classical immunophenotype of SCID-X1 is the absence of T and NK cells, and persistence of dysfunctional B cells (TB þ NKSCID). If a genotypically matched family donor is available, bone marrow transplantation is a highly successful procedure with a long-term survival rate of over 90%. The high survival rates are partly due to the fact that the absence of T and NK cells in SCID-X1 patients allows engraftment in the absence of myelosuppressive conditioning. For the majority of individuals, this is not possible, and the survival from mismatched family (usually parental donors) transplants is less good, and is associated with predictable toxicity. Many incremental advances in gene transfer technology have recently been translated into successful gene therapy for SCID-X1.3 These have included the activation of cells with high concentrations of cytokines (thereby making them susceptible to gamma retrovirus vectormediated gene transfer), and transduction in containers coated with a recombinant fibronectin fragment (RetroNectint) that is believed to facilitate the colocalization of the virus particle and the target cell.4,5 In the first landmark study, a conventional amphotropic retroviral vector encoding a gc cDNA (regulated by Moloney murine leukaemia virus long-terminal repeat sequences) was used to transduce ex vivo autologous CD34 þ cells (separated by conventional magnetic bead technology from a bone marrow harvest). The cells were reinfused into the patients in the absence of preconditioning. The results obtained from the first five patients have recently been reported in the scientific literature.6 To date, 10 infants in total have now been treated, with good immunological reconstitution in all but one, in whom the graft appears to have become sequestered in a pathologically enlarged spleen6,7 In nearly all patients, NK cells appeared between 2 and 4 weeks after infusion of cells, followed by new thymic T-lymphocyte emigrants at 10–12 weeks. With some variation, the number and distribution of these T cells normalized rapidly (more rapidly than observed following haploidentical transplantation). They also appeared to function normally in terms of proliferative response to mitogens, Tcell receptor (TCR), and specific antigen stimulation, and to have a complex phenotypic and molecular diversity of TCR. Functionality of the humoral system was also restored, maybe not quite as effectively, but to a sufficient degree that discontinuation of immunoglobulin therapy was possible. Persistent long-term marking in myeloid cells (between 0.1 and 1%) suggests that long-lived stem or progenitor cells have been successfully transduced. More recently, we have initiated a similar study for the treatment of SCID-X1. The transduction protocols and vector are very similar although we have used a gibbon– ape–leukaemia virus (GALV) pseudotype, and conditions that obviate the requirement for foetal calf serum. Although the follow-up period for four children treated in our study is short, all have cleared viral infections, and immunological reconstitution has followed a similar pattern8 (Thrasher et al, manuscript in preparation). The contribution to the initial burst of thymopoiesis from relatively late T-cell precursors in the original transduced CD34 þ cell population versus that from cells earlier in the haematopoietic differentiation hierarchy, which have engrafted in the bone marrow, has not yet Gene Therapy been determined. This may have important implications for the durability of immunological reconstitution, and for sustained production of new T cells. These issues may be resolved by longitudinal study of naive T-cell production, and by the isolation of common integration sites between myeloid and lymphoid populations. Ultimately, the longevity of functional reconstitution can only be determined by clinical monitoring, but it should also be feasible to repeat gene therapy on multiple occasions. Unknown, however, are the time frames within which this will be clinically effective, particularly bearing in mind that there may be agerelated restrictions to the reinitiation of thymopoiesis in these patients. Cells in patients with ADA deficiency treated by lymphocyte or stem cell gene therapy persist and maintain transgene expression for many years Deficiency of the purine salvage enzyme adenosine deaminase (ADA) accounts for approximately 10–20% of all SCIDs. ADA catalyses the deamination of deoxyadenosine (dAdo) and adenosine to deoxyinosine and inosine, respectively, and the lack of ADA leads to the build of the metabolites deoxyATP (dATP) and dAdo, which have profound effects on lymphocyte development and function through a number of cellular mechanisms. There is variation in the severity of the condition but most ADA patients have very low numbers of T and B lymphocytes. Bone marrow transplantation is highly successful in the genotypically matched setting, but human leucocyte antigen (HLA)mismatched transplants have poor survival outcomes. An alternative modality of treatment is exogenous enzyme replacement with polyethylene glycol-conjugated bovine ADA, which as regular intramuscular injections can result in the correction of metabolic and immunological abnormalities, albeit only partially in a significant number of cases.9 The first human gene therapy studies were conducted on patients with ADA deficiency in the early 1990s. It is generally agreed that these initial studies were unsuccessful in correcting the immune defect in ADA-SCID. This was in part due to the continued use of PEG-ADA enzyme replacement therapy, which in itself improved the immune function but may also have blunted the survival advantage of gene-modified cells (discussed further below). However, a decade on, more detailed analysis of the patients originally treated does provide important information about the longevity and efficacy of gene transfer.10 In the first human clinical gene therapy study, two patients were treated following repeated gammaretroviral vector-mediated ADA gene transfer into stimulated peripheral blood lymphocytes. Patient 1 still shows over 15% of gene-marked cells in peripheral blood mononuclear cells (PBMCs) and ADA activity in PBMCs remains at B25% of the normal. The level of gene marking (0.1% of PBMCs) and ADA activity (o5% of normal) is considerably less in patient 2, which may reflect the smaller number of gene-transduced cells initially infused, but may also be due to the development of an immune response against the retroviral envelope Gene therapy for severe immunodeficiency HB Gaspar et al and lipoprotein components of the foetal calf serum used for culturing the cells. However, these findings clearly demonstrate that human T cells have a lifespan greater than 10 years in the peripheral circulation, and also that transgenes regulated by gammaretroviral sequences continue to express in peripheral T cells and resist in vivo silencing. Molecular analysis of TCR diversity, combined with transgene integration analysis, reveals that within individual Vb family clones, each cell contains multiple unique integration sites. This suggests that, in the initial phase, cells were repeatedly sampled and transduced. In a later study, umbilical cord blood CD34 þ cells were harvested from antenatally diagnosed ADA-SCID patients, transduced and reinfused in the first week of life. In these patients, little clinical benefit was seen, and the level of gene marking was low (1–10% of T lymphocytes) in all the three children. Recent clonal integration analyses demonstrate that transgene-containing T lymphocytes are monoclonal or oligoclonal (with 1–5 different integration sites) more than 8 years after gene therapy, and that single prelymphoid clones contribute between 25 and 100% of genetically corrected lymphocytes.11 Marking in other lineages is consistently less than 1%. Again, the continuation of PEG-ADA throughout these early studies almost certainly compromised the efficient engraftment of transduced cells. Withdrawal of PEG-ADA from patients treated by lymphocyte gene therapy for ADA-deficient SCID results in enhanced immunological reconstitution Matched sibling donor transplants for ADA-SCID performed without conditioning result in rapid engraftment and persistence of donor T lymphocytes, strongly suggesting that T lymphocytes expressing ADA have a powerful proliferative and survival advantage. The failure of initial gene therapy studies to demonstrate the proliferation and expansion of gene-modified cells seemed to question this premise, although the continued administration of PEG-ADA in all these patients may have abrogated this advantage. Recently, one patient participating in another study demonstrated convincingly that a survival advantage for gene-transduced cells did exist in the absence of PEG-ADA. In this individual, who had received multiple infusions of autologous transduced peripheral blood lymphocytes (PBLs), and who had reached a plateau of 1–3% genetransduced T cells, PEG-ADA-associated immune dysregulatory problems led to a gradual discontinuation of enzyme replacement. As the dose was decreased and finally stopped, the percentage of transduced T cells as assessed by PCR quantification increased, eventually reaching nearly 100% of all T lymphocytes.12 Absolute CD3 þ T-cell counts also increased and stabilized at levels higher than prediscontinuation values. T-cell proliferative responses were also restored, and analysis of ADA metabolites showed a rise in intracellular PBL ADA activity. The blunting effect of PEG-ADA may also be responsible for the disappointing observations made in a more recent study that employed enhanced vectors, and transduction conditions similar to those used for SCID-X1 trials.13 As before, transduced CD34 þ cells were returned to the patients without conditioning and patients continued PEG-ADA treatment. ADA gene marking was only seen at low levels (in a range from 0.0001 to 3.6%). On the basis of previous observations, the investigators plan to withdraw PEG-ADA 1 year after treatment. 2001 Successful gene therapy for ADA-deficient SCID can be achieved in the absence of PEG-ADA and in combination with myelosuppression Results of a new study have clearly demonstrated that ADA-SCID can be successfully treated by gene therapy.14 In this protocol, CD34 þ cells were transduced with an amphotropic gammaretroviral vector (originally used in PBL transduction studies) under current optimal conditions. Two important changes were incorporated into the protocol. Firstly, for economic reasons, patients were not commenced on PEG-ADA and, secondly, patients received a mild dose of conditioning (4 mg/kg of busulphan as 2 mg/kg on two successive days) prior to the return of gene-modified cells. A neutrophil and platelet nadir were seen at approximately 3 weeks, but neither child required blood product support. One patient is now 20 months. Over 2 years, postgene transfer one patient has normal numbers of peripheral T, B and NK cells. This patient has normal immunoglobulin production and is not receiving any prophylactic therapy. There has also been impressive correction of the metabolic defects, with dATP levels falling to 10% of that at diagnosis (comparable to that achieved following successful BMT). The second patient is now 12 months postgene therapy, but was an older child (approximately 2.5 years) at the time of treatment and also received a lower cell dosage. In this patient, recovery has been slower and the T-cell reconstitution at present is suboptimal, but significantly improved from pretransplant levels. There is evidence of some immunoglobulin production, but the patient remains on replacement therapy. A third child has been treated more recently using the same protocol,15 and is showing a recovery similar to that in patient 1. Molecular analysis of the first two patients shows a diverse TCR repertoire and an increase in TCR excision circle (TREC) levels, indicating the successful engraftment of prethymic progenitor populations. Lineage-specific transgene analysis by quantitative PCR shows high level marking in T, NK and B cells, and the persistence of gene-modified granulocytes, monocytes and megakaryocytes at levels between 5 and 20%, again suggesting that multipotent progenitors have engrafted. Somewhat surprising is the level to which marking has persisted in myeloid cells, particularly at the dosage of conditioning employed. This may suggest that a survival advantage is not restricted to lymphoid cells, and that it also extends to myeloid and haematopoietic multipotent progenitor cells. The results from this study are extremely encouraging. At present, it is difficult to determine whether the success of the procedure is due to the lack of PEGGene Therapy Gene therapy for severe immunodeficiency HB Gaspar et al 2002 ADA or the use of nonmyeloablative conditioning, but it is likely that the combination is important. The key to correction of the metabolic abnormalities in ADA-SCID seems to be the delivery of large amounts of ADA enzyme, whether exogenously in the form of PEG-ADA or intracellularly as gene-modified cells. The use of conditioning may facilitate the initial engraftment of a greater number of gene-modified cells. Certainly, if immune function in these patients is sustained, and further patients show a similar safety profile and immune response, this strategy holds great promise for ADA-SCID and potentially other haematopoietic conditions. Insertional mutagenesis has been observed in human studies, reinforcing the need to develop methods for optimization of protocol safety For retroviruses, which depend on chromosomal integration for the stability of transduction, the most prominent safety concern has been for insertional mutagenicity.19 On the basis of numerous animal studies and over 300 clinical trials in which patients have received retroviral vectors, and from theoretical considerations, the risk of clinically manifesting insertional mutagenesis has been judged to be low. However, in a recent murine HSC retroviral transduction study, insertion of the vector into the oncogene Evi-1 led to development of myeloid leukaemia.20 This has been followed by the reported development of uncontrolled clonal T-cell proliferation in two patients in the Paris SCID-XI clinical trial (Table 1).7,21 Having initially achieved successful immunological reconstitution, both developed lymphoproliferation approximately 3 years after the gene therapy procedure. In both patients, retroviral vector insertion into or near the LMO-2 proto-oncogene resulted in high-level expression of LMO-2 in the clones, almost certainly as a result of retroviral enhancer-mediated activation of transcription. Activation of LMO-2 is known to participate in human leukaemogenesis by chromosomal translocation, and results in the development of T-cell lymphoproliferation and leukaemia in mice, albeit with a long latency. It is therefore likely that other contributing factors are required for these events to manifest. At present, there is no evidence for the contribution of dysregulated gc expression in lymphoid cells, although this remains a possibility and is being studied carefully. Cells with high proliferative potential, such as thymocytes, are also likely to be more susceptible to transformation following an insertional event than quiescent cells, if they acquire additional adverse mutations unrelated to the gene therapy itself. This increased risk cannot yet be quantified. The integration of the vector into LMO-2 in both cases strongly suggests that there is some preference for the survival of these clones or less likely, for integration at this site (Hacein-Bey-Abina et al, submitted for publication). The detailed molecular analysis of insertion events in patients undergoing gene therapy will greatly assist in the delineation of integration points within the genome, but is unlikely to be able to predict potential for Animal models of JAK-3- and RAG-2deficient SCID have been corrected using similar strategies The molecular basis of autosomal recessive TB þ NKSCID is mutation of the receptor tyrosine kinase gene JAK-3. The dependence of gc on signalling through JAK-3 is responsible for a clinical and immunological phenotype identical to that of SCID-X1, and the rationale for gene therapy is therefore similar. Correction of a murine model of JAK-3-deficient SCID has been achieved using both myelosuppresssive, and more relevant to clinical studies, conditioning-free protocols.16 Patients with mutations of the recombinaseactivating genes RAG-1 and RAG-2 characteristically present with the absence of both B and T cells. Moloneybased gammaretroviral vectors have recently been shown to effectively reconstitute RAG-2-deficient mice, and in the absence of detectable toxicity, even though gene expression was not tightly regulated.17 One way to obviate the toxicity arising from dysregulated gene expression in any condition, and to achieve physiological activity, is to correct genetic mutations by gene repair or homologous recombination. Recently, it has been shown that RAG2/ mutant murine embryonic stem (ES) cells, repaired by standard homologous recombination technology, can be grown in vitro to provide sufficient haematopoietic progenitors for engraftment and correction of RAG-2 mutant mice.18 This is the first example of gene therapy combined with a therapeutic cloning strategy and, clearly, has important implications for future treatment of many genetic disorders. Table 1 Current clinical trials of gene therapy for SCID Disease Gene Retroviral vector Envelope ADA-D ADA-D ADA ADA A A ADA-D SCID-X1 SCID-X1 AR-SCID ADA gc gc JAK-3 GIADA1 GCsap-M-ADA MND-ADA SFFV-ADA-WPRE MFG MFG MSCV G A G G Cell type treated *BM CD34+ BM CD34+ UCB CD34+ *BM CD34+ BM CD34+ BM CD34+ BM CD34+ Reference/PI 14 11,13 AJ Thrasher/HB Gaspar 3,6,7 AJ Thrasher BP Sorrentino Summary of clinical trials for SCID using gammaretroviral vectors ongoing from 2000. A, amphotropic envelope; G, gibbon–ape–leukaemia virus envelope; ADA, adenosine deaminase deficiency; SCID-X1, X-linked SCID; BM, bone marrow; UCB, umbilical cord blood; gc, common cytokine receptor gamma chain. *Denotes the use of preconditioning. Gene Therapy Gene therapy for severe immunodeficiency HB Gaspar et al mutagenesis unless recurrent hotspots associated with clinical disease become evident.22–24 In combination with conventional monitoring of lymphocyte numbers and distributions, longitudinal monitoring of integration sites will provide an important way of monitoring for pathological clonal expansions. The applicability of any novel therapy, including gene therapy, ultimately depends on the balance of risks against those of alternative treatments. The accurate characterization of adverse events, the utilization of protocols to test toxicity in a rigorous way, and the development of methods to minimize risks are therefore essential. Future prospects for SCID Recent clinical trials have shown that at least some forms of SCID can be effectively treated by gene therapy. Much, however, can be done to improve the efficiency and safety of current protocols. The design of vectors used for gene delivery is clearly important and modifications may be possible, which will limit the risks of mutagenesis, for example by the incorporation of DNA and RNA insulator sequences in integrating vectors, by the use of self-inactivating vectors or by targeting safe regions in the genome.25,26 Alternative vectors based on lentiviruses or foamy viruses that obviate prolonged ex vivo culture may allow the preservation of larger numbers of multipotential progenitor cells, but at the same time may produce higher numbers of insertion events in each cell.27–31 Methods to minimize the number of integration events per cell, and to limit the number of engrafting clones, for example by more stringent purification of stem cell (or defined target cell) populations, may therefore also be beneficial. Probably, the most straightforward way to improve safety is to dispense with the powerful viral enhancer sequences that can dysregulate gene expression over large chromatin domains. Lentiviral vectors, in particular, provide greater capacity for the incorporation of more complex and physiological regulatory sequences. The relative risk for each type of vector modification needs to be determined in clinically relevant animal-model systems, and the effectiveness of these models to predict side effects in humans will have to be validated. The development of homologous recombination or gene repair to correct mutations, or the construction of mitotically stable extrachromosomal vectors, would obviate many of these problems, but current technologies are inefficient.32 Once again, SCID may be a perfect initial target for this strategy, as even limited efficiency will be sufficient to provide clinical benefit. The future for gene therapy of SCID is exciting, but has been clouded by the occurrence of toxicity. As for all novel therapeutic modalities, increased understanding of mechanisms, and increased sophistication of technology will translate into even more effective and safe application. There is no better paradigm for this process than allogeneic bone marrow transplantation. References 1 Fischer A. Primary immunodeficiency diseases: an experimental model for molecular medicine. Lancet 2001; 357: 1863–1869. 2 Leonard WJ. Cytokines and immunodeficiency diseases. Nat Rev Immunol 2001; 3: 200–208. 2003 3 Cavazzana-Calvo M et al. Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease. Science 2000; 288: 669–672. 4 Demaison C et al. A defined window for efficient gene marking of severe combined immunodeficient-repopulating cells using a gibbon ape leukaemia virus-pseudotyped retroviral vector. Hum Gene Ther 2000; 11: 91–100. 5 Hacein-Bey S et al. Optimization of retroviral gene transfer protocol to maintain the lymphoid potential of progenitor cells. Hum Gene Ther 2001; 12: 291–301. 6 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: 1185–1193. 7 Hacein-Bey-Abina S et al. LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1 ((gamma) c deficiency). In press. 8 Thrasher AJ et al. Immune Recovery Following Retroviral Mediated Common Gamma Chain Gene Therapy for X-linked Severe Combined Immunodeficiency. American Society of Gene Therapy’s Sixth Annual Meeting Executive Summaries, Vol. 36; 2003. 9 Hershfield M. ESID 2002. 10 Muul LM et al. Persistence and expression of the adenosine deaminase gene for twelve years and immune reaction to gene transfer components: long-term results of the first clinical gene therapy trial. Blood 2002; 101: 2563–2569. 11 Schmidt M et al. Clonality analysis after retroviral-mediated gene transfer to CD34+ cells from the cord blood of ADAdeficient SCID neonates. Nat Med 2003; 9: 463–468. 12 Aiuti A et al. Immune reconstitution in ADA-SCID after PBL gene therapy and discontinuation of enzyme replacement. Nat Med 2002; 8: 423–425. 13 Candotti F et al. Corrective gene transfer into bone marrow CD34+ cells for adenosine deaminase (ADA) deficiency: results in four patients after one year of follow-up. Mol Ther 2003; 7: S448. 14 Aiuti A et al. Correction of ADA-SCID by stem cell gene therapy combined with nonmyeloablative conditioning. Science 2002; 296: 2410–2413. 15 Aiuti A. Correction of the immune and metabolic defect of ADA-SCID by stem cell gene therapy combined with nonmyeloablative conditioning. Mol Ther 2003; 7: S448. 16 Bunting KD, Lu T, Kelly PF, Sorrentino BP. Self-selection by genetically modified committed lymphocyte precursors reverses the phenotype of JAK3-deficient mice without myeloablation. Hum Gene Ther 2000; 11: 2353–2364. 17 Yates F et al. Gene therapy of RAG-2/ mice: sustained correction of the immunodeficiency. Blood 2002; 100: 3942–3949. 18 Rideout III WM et al. Correction of a genetic defect by nuclear transplantation and combined cell and gene therapy. Cell 2002; 109: 17–27. 19 Baum C et al. Side effects of retroviral gene transfer into hematopoietic stem cells. Blood 2003; 101: 2099–2114. 20 Li Z et al. Murine leukemia induced by retroviral gene marking. Science 2002; 296: 497. 21 Hacein-Bey-Abina S et al. A serious adverse event after successful gene therapy for X-linked severe combined immunodeficiency. N Engl J Med 2003; 348: 255–256. 22 Schroder AR et al. HIV-1 integration in the human genome favors active genes and local hotspots. Cell 2002; 110: 521–529. 23 Laufs S et al. Retroviral vector integration occurs into preferred genomic targets of human bone marrow repopulating cells. Blood 2002; 101: 2191–2198. 24 Wu X et al. Transcription start regions in the human genome are favored targets for MLV integration. Science 2003; 300: 1749–1751. 25 Burgess-Beusse B et al. The insulation of genes from external enhancers and silencing chromatin. Proc Natl Acad Sci USA 2002; 99 (Suppl 4): 16433–16437. Gene Therapy Gene therapy for severe immunodeficiency HB Gaspar et al 2004 26 Olivares EC et al. Site-specific genomic integration produces therapeutic Factor IX levels in mice. Nat Biotechnol 2002; 20: 1124–1128. 27 Follenzi A et al. Gene transfer by lentiviral vectors is limited by nuclear translocation and rescued by HIV-1 pol sequences. Nat Genet 2000; 25: 217–222. 28 Glimm H, Oh IH, Eaves CJ. Human hematopoietic stem cells stimulated to proliferate in vitro lose engraftment potential during their S/G(2)/M transit and do not reenter G(0). Blood 2000; 96: 4185–4193. 29 Demaison C et al. High-level transduction and gene expression in hematopoietic repopulating cells using a human immunodeficiency [correction of immunodeficiency] virus type Gene Therapy 1-based lentiviral vector containing an internal spleen focus forming virus promoter. Hum Gene Ther 2002; 13: 803–813. 30 Josephson NC et al. Transduction of human NOD/SCIDrepopulating cells with both lymphoid and myeloid potential by foamy virus vectors. Proc Natl Acad Sci USA 2002; 99: 8295– 8300. 31 Woods NB et al. Lentiviral vector transduction of NOD/SCID repopulating cells results in multiple vector integrations per transduced cell: risk of insertional mutagenesis. Blood 2002; 101: 1284–1289. 32 Van Craenenbroeck K, Vanhoenacker P, Haegeman G. Episomal vectors for gene expression in mammalian cells. Eur J Biochem 2000; 267: 5665–5678.