Mini Review Hematopoietic Stem Cell Gene Therapy: Progress Toward Therapeutic Targets

Home , CD34, Stem cell, Transdifferentiation

Bone Marrow Transplantation (2003) 32, 1–7 & 2003 Nature Publishing Group All rights reserved 0268-3369/03 $25.00 www.nature.com/bmt Mini review Hematopoietic stem cell gene therapy: progress toward therapeutic targets

JL Vollweiler, SP Zielske, JS Reese and SL Gerson

Division of Hematology–Oncology, Comprehensive Cancer Center at University Hospitals of Cleveland, Case Western Reserve University School of Medicine, USA

Summary: the preclinical and clinical studies, gene therapy studies has also taught us a great deal about stem cell transplant The concept of hematopoietic stem cell gene therapy is as biology. exciting as that of stem cell transplantation itself. The Gene therapy attracted somewhat premature enthusiasm past 20 years of research have led to improved techniques in the mid-1980s, as a targeted and permanent treatment for transferring and expressing genes in hematopoietic for many previously incurable disorders. However, the stem cells and preclinical models now routinely indicate field has met with more clinical disappointment than the ease with which new genes can be expressed in success until very recently. Early results raised more repopulating stem cells of multiple species. Both modified questions than they answered and helped to focus attention murine oncoretroviruses and lentiviruses transmit genes on the elusive prerequisites of identifying an appro into the genome of hematopoietic stem cells and allow priate target cell, establishing efficient transfer of the expression in the host following transplantation. Using therapeutic gene maintaining stable integration and ex- oncoretroviruses, therapeutic genes for severe combined pression of the gene and understanding and explaining immunodeficiency, common variable gamma chain im- the risks involved. munodeficiency, chronic granulomatous disease, Hurler’s and Gaucher’s Disease have all been used clinically with only modest success except for the patients with Bone marrow derived stem cells immunodeficiency in whom a partial T-cell chimerism has been dramatic. Since stem cell selection in vivo Hematopoietic stem cells are arguably the most attractive appears important to the therapeutic success of gene and best-studied target cell population for several reasons. transfer, drug resistance selection, most recently using the They are easily accessible for harvest, and readily delivered MGMT gene, has been developed and appears to be safe. back to the patient by autologous transplant methods that Future trials combining a drug resistance and therapeutic are already familiar and commonplace to transplant gene are planned, as are trials using safety-modified physicians. They are long-lived, and have an enormous lentiviruses. The therapeutic potential of hematopoietic repopulation potential. Furthermore, they are pluripotent stem cell gene therapy, particularly given recent advances and hold the potential to correct defects in all hemato- in stem cell plasticity, remains an exceptionally exciting poietic lineages, and even nonhematopoietic cell lines via area of clinical research. transdifferentiation. Bone Marrow Transplantation (2003) 32, 1–7. doi:10.1038/ Drawbacks of using hematopoietic stem cells as targets sj.bmt.1704081 include historically low rates of successful transduction, Keywords: hematopoietic stem cell; gene therapy; human; given their lack of cell surface receptors and quiescent state. retrovirus; lentivirus; severe combined immune defi- Also, identification of the ideal stem cell has proven elusive. ciency In most clinical gene therapy trials, CD34+ cells are isolated from blood or marrow using immunomagnetic selection and transduced. However, evidence suggests there are CD34À cells that possess the properties of stem cells1,2 and that even a single CD34À cell is capable of long-term 3 Since 1985, when the first studies with stem cell gene repopulation of hematopoiesis. Finally, CD34 has been 4 transfer were reported, we have been tantalized by the shown to be expressed reversibly on murine stem cells. possibilities of gene therapy. It at once captivates us Better identification of the true hematopoietic stem cell, with the promise of curing a vast array of inherited, which appears to be modulated by age, cytokine activation, malignant and even infectious diseases, and challenges us cell cycle state and purification technique, as well as with feasibility, safety and ethical considerations. In both attention to the accessory cells used in the transplantation protocol, may improve transduction efficiency and clinical results in the future. Correspondence: Dr SL Gerson, Case Western Reserve University Other bone marrow-derived cell types of interest include School of Medicine, 10900 Euclid Avenue BRB 3-West, Cleveland, mesenchymal stem cells and multipotent adult progenitor OH 44106-4937, USA cells, or MAPCs. Mesenchymal stem cells isolated from the Hematopoietic stem cell gene therapy JL Vollweiler et al 2 bone marrow have the ability to differentiate into not only Table 1 Therapeutic targets of corrective stem cell gene therapy hematopoietic support stroma, but osteocytes, chondro- (A) Primary immunodeficiencies cytes, tenocytes, adipocytes and skeletal myocytes. In their ADA-deficient SCID role as support stroma, mesenchymal stem cells have shown X-linked SCID promise in improving the transduction efficiency of Common variable immunodeficiency hematopoietic stem cells, immuno-downmodulation of Chronic granulomatous disease [CGD] X-linked agammaglobulinemia graft-vs-host disease after an allograft and potential Wiskott–aldrich syndrome engraftment.5,6 Mesenchymal stem cells also express glucocerebrosidase, alpha-l-iduronidase, arylsulfatases (B) Hemoglobinopathies and adrenoleukodystrophy protein, raising the possibility Sickle cell anemia of a therapeutic application in lysosomal storage disea- Beta-thalassemia sescharacterized by loss of expression of these proteins, and (C) Other single-gene disorders in other diseases requiring replacement therapy. Although Hurler’s disease allogeneic bone marrow transplantation has failed to show Gaucher’s disease engraftment of donor mesenchymal stem cells,7 preclinical Hemophilia A, B models using autologous mesenchymal stem cells and gene Alpha-1 antitrypsin deficiency 8 transfer have had early success. (D) Stem cell defects MAPCs are a unique, newly described class of stem cells, Fanconi anemia characterized as CD34-, c-kitÀ, HLA-DRÀ, but AC133+, VEGFR2/Flk1/KDR+, and VEGFR1/Flt1+. They have demonstrated a superior proliferative potential with cyto- genetic and telomeric stability, while maintaining multi- been the most widely studied. These vectors are based on potentiality. They can differentiate in vitro into simple type C murine oncoretroviruses such as murine mesenchymal lineage cells, neural cells, hepatocyte-like leukemia virus. An appealing property of retroviral vectors cells and angioblasts.9,10 A potential application involves is that after reverse transcription of their genomic RNA early preclinical work using marrow-derived MAPCs to into proviral DNA, it is integrated into the host cell’s neovascularize infracted myocardium.11 genome at random locations. This means that all progeny derived from a transduced cell will contain the provirus, and therefore potentially express the therapeutic gene. Therapeutic targets Hence, an integrating vector is needed when expression of the transgene is required after many cell divisions, such Diseases of hematopoietic stem cells and their progeny are as when a gene needed in terminally differentiated blood obvious choices for gene therapy. Classically, this has cells is transferred to HSC. The oncoretroviral genome involved providing a therapeutic gene to patients with a is simple in that three genes are encoded. The gag, pol single locus genetic deficiency such as any of the genetic and env genes code for structural, replicative and disorders that make up severe combined immunodeficiency envelope proteins, respectively. Each end of the integrated (SCID), adenosine deaminase (ADA) deficiency, hemoglo- provirus contains long terminal repeat (LTR) sequences binopathies, glycogen storage diseases such as Gaucher’s that have promoter activity for driving the expression and Hurler’s, and disorders of stem cell proliferation such of viral genes. as Fanconi’s anemia (Table 1). More recently, there has Most current viral gene therapy vectors targeting been interest in not just correcting a deficit, but also hematopoietic stem cells are modified in a number of key providing stem cells with new or enhanced properties. This ways to make them safe for clinical use. First, the vectors has typically been investigated in the setting of malignant are ‘gutted’ of the endogenous viral genes necessary for disease. Approaches include chemotherapy resistance viral replication and budding from the cell. Only the genetic genes, and expression of tumor-specific antigens or information necessary for reverse transcription of the viral cytokines to stimulate immune responses against tumors RNA into DNA, its subsequent transport to the nucleus among differentiated progeny such as dendritic cells and and its integration into the genome are retained. Second, lymphocytes. In addition, the ability of hematopoietic stem the vectors are modified to accept the therapeutic gene. cells to promote cardiac muscle healing suggests a possible Third, the vectors are modified to enhance expression role in the delivery of vascular endothelial growth factor, of the new gene in the target cell and its progeny for many VEGF. Finally, genetically manipulated hematopoietic cell generations including many months or years after stem cells are being pursued as an avenue to provide transplantation. The oncoretroviral vectors have been progeny lymphocytes with resistance to infection, particu- modified for safety and to carry a therapeutic gene larly HIV. by separating viral functions and the creation of stable virus ‘packaging’ cell lines that produce virus. Since the virus carries all proteins needed to complete reverse Vector choice and design transcription and integration within the virion, no viral coding sequences are required to be transferred to the Gene transfer into hematopoietic stem cells has been target cell. To accomplish this, the gag/pol and env genes demonstrated using many viral and nonviral vectors with are separated on plasmids and expressed in trans from a varying efficiencies. However, oncoretroviral vectors have heterologous promoter, while the region between the LTRs,

Bone Marrow Transplantation Hematopoietic stem cell gene therapy JL Vollweiler et al 3 is replaced with the therapeutic gene, also on a plasmid. Table 2 Stem cell gene therapy in the treatment of nongenetic Since the packaging signal is maintained only on the LTR diseases and therapeutic-gene-containing construct, only it will be (I) Cancer packaged, together with the necessary proteins, into virions Tumor-associated antigens to be transferred to the target cell. Virus can be produced Cytokines and chemokines by transiently transfecting a suitable cell line and collecting Chemotherapy resistance genes culture supernatant or preferably, a packaging line is made MDR-1 DHFR by selecting for stably transfected virus-producing cells MGMT using selectable genes contained elsewhere on the transfected plasmids. Packaging lines are advantageous in that (II) Cardiovascular diseases they and the virus produced can be well characterized for VEGF recombination events and consistent virus preparations can (III) Infectious disease resistance easily be obtained. HIV-1 Unfortunately, it has proved quite difficult to transduce human HSC compared to those of animal models, because of the nondividing nature of human HSC and the requirement of division for oncoretroviral vector transduc- lentivirus, despite their improved efficacy in gene delivery tion.12 For this reason, lentiviral vectors based on the in preclinical models, has delayed the test of their clinical human immunodeficiency virus (HIV) and others have been utility. Some of the unknown factors are the capability of gaining momentum because they do not require extensive the modified vectors to recombine in vivo–either with culture and cell proliferation for gene transfer to take place. themselves or with an endogenous virus such as HIV or The most studied HIV-based vector system was originally HTLV - the impact of cell proliferation and differentiation developed in 1996 by Naldini et al,13 while concurrently the on proviral promiscuity, the potential impact of endogen- first demonstration of CD34+ cell transduction using an ous transposons and the insertional mutagenesis of the HIV-based vector was shown by Akkina et al.14 As provirus (Tables 2 and 3). required, this vector has been extensively modified to increase safety. HIV is a complex retrovirus and its genome contains nine genes. Of these, four are necessary for in vivo Methodological advances in vector delivery pathogenicity, but dispensable for in vitro propagation.15 Therefore, safety modifications include the deletion of these Other methodological advances have significantly im- four accessory genes and deletion of the wild-type envelope. proved oncoretroviral gene transfer. One of the primary Packaging, transfer and envelope functions have been limitations of clinical retroviral gene transfer studies has delegated to separate plasmids to minimize the generation been the low gene transfer rates into hematopoietic stem of replication-competent virus. In addition, a deletion has cells, resulting in low frequency and transient detection of been made in the viral LTR eliminating its promoter transduced cells after reinfusion into the patient. Williams activity, thus requiring an internal promoter. This creates a and co-workers20,21 found that transduction efficiency ‘dead’ virus upon integration, since transcription cannot could be significantly increased by use of a specific adhesion begin from the LTR. Production methods still require domain of recombinant fibronectin, CH-296, which colo- transient transfection of plasmids for the highest viral calizes the viral envelope protein and target cell receptors. production because the envelope, VSV-G and certain HIV Another factor that influences the proliferation rate of proteins are toxic to producer cells. Packaging lines with hematopoietic progenitors is the nature and the duration regulated expression have been developed.16 of the cytokines used in the ex vivo culture. Addition of Since cell division is not required for transduction, thrombopoietin to the combination of stem cell factor and targeting the true hematopoietic stem cell is potentially FLT-3 ligand during the transduction period results in an more efficient. This has proven to be the case in several increase in the number of CD34+ progenitors and high studies of human CD34+ cells.17–19 Unfortunately, reluc- levels of engraftment of the transduced human hemato- tance to consider clinical trials with safety-modified poietic progenitors in immunocompromised mice, compared

Table 3 Vector characteristics

Vector Target cells Integration Efﬁciency Comments Oncoretrovirus Dividing cells Stable Low High efﬁciency seen in murine models not seen in large animals or humans Lentivirus Dividing and Nondividing cells Stable High Ability to transduce under minimal stimulation may prevent differentiation and allow better preservation of stem cell phenotype Adenovirus Dividing and Nondividing cells No High Lack of integration or self-replication make this a poor vector for gene therapy Adeno-associated Virus Dividing and Nondividing cells Stable Varied Most success seen in transduction of nonhematopoietic cells Nonpathogenic Foamy Virus Dividing and Nondividing cells Stable (?) High Nonpathogenic in mammals

Bone Marrow Transplantation Hematopoietic stem cell gene therapy JL Vollweiler et al 4 to previously used combinations.22–24 The incorporation of aimed at transferring a marker gene, rather than a both fibronectin fragment CH-296 and thrombopoietin therapeutic gene of interest. In these genetic marking have proved successful in clinical protocols. experiments, a gene with an easily detectable sequence or gene product is transferred to autologous cells of patients receiving a therapeutic autograft for an unrelated indica- Safety issues tion, typically a malignancy. Many of these studies used the neomycin resistance gene, Neo, as the marker,29–31 and With recent reports in the lay press about severe toxicity typically showed detectable but disappointingly low levels and one death in human gene therapy trials, safety of transduction, of less than 5%. considerations have been scrutinized again. Perhaps the The first clinical trial of therapeutic gene transfer was largest theoretical concern has been the development of undertaken in 1990 at the NIH, and enrolled two patients replication-competent viruses. A lymphoma associated with ADA-deficient SCID.32 This trial actually targeted with a replication competent retrovirus in a primate model peripheral T lymphocytes rather than hematopoietic stem was reported in 1992.25 The FDA recommendations for cells. Cells were collected by apheresis, transduced and monitoring recombinant retrovirus (RCR) in patients reinfused several times. One patient demonstrated ADA include 3 and 6 months, 1 year after treatment and yearly activity levels up to 25% normal in circulating T cells, thereafter. If any positive samples are detected, further which was stable for up to 5 years. The second patient had analysis and extensive patient follow-up is warranted less encouraging results. A third patient was treated by the (FDA/CBER Guidance for Industry: Supplemental gui- same protocol, and again achieved significant levels of dance on testing for RCR in retroviral vector-based gene ADA activity.33 Of note, all these patients have continued therapy products and during follow-up of patients in on PEG-ADA enzyme-replacement; therapy, so the true clinical trials using retroviral vectors, 1999). There have not clinical impact of this success is not known. been any additional reports of RCR since the isolated In 1992, a study was undertaken in ADA-deficient primate case in 1992. Insertional mutagenesis is believed to patients, targeting peripheral lymphocytes with one vector, account for the abnormal proliferative process seen in a and bone marrow cells with a slightly different vector.34 murine model of MDR gene transfer.26 Two cases of Over time, the dominant detectable vector in peripheral T insertional mutagenesis at a locus known to be involved in cells switched from the one used to transduce lymphocytes the pathogenesis of a monoclonal T-cell disorder, similar to to the one used to transduce marrow cells. Transduced cells an acute lymphocytic leukemia, has been found in children have persisted for several years, and PEG-ADA has been receiving corrective CD34 cell gene therapy for combined slowly tapered. A study from the Netherlands using variable immunodeficiency (see below), establishing the CD34+ bone marrow cells reported low efficiency of possibility of this occurring in other stem cell gene therapy ADA gene transfer and quite short-lived presence of settings as well. circulating genetically modified cells.35 Kohn’s group in Other concerns include immunologic reactions to the Los Angeles has treated three neonates with their umbilical transgene product or other components in the gene therapy cord blood transduced with the ADA gene. Again, process. The FDA has also cautioned the use of thrombo- persistent but low levels of corrected cells were seen. When poietin because of the potential for induction of immune enzyme-replacement therapy was tapered, the percentages thrombocytopenia. While this is unlikely since the throm- of marked cells rose to 10%, presumably due to a survival bopoietin remaining in the infusion product is undetect- advantage of the corrected cells.36 able, platelet counts should be monitored weekly during the The most dramatic success in gene therapy trials has first 6 weeks after infusion for safety reasons. Patients come in the treatment of X-linked SCID. Fischer’s group in receiving cells transduced in the presence of the fibronectin France reported initial success in two infants in 200037 and fragment should also be monitored monthly, up to 24 recently expanded upon their findings, reporting on five months for formation of antibodies to CH-296. In one patients with up to 2.5 years’ follow-up.38 They collected recent retroviral gene transfer study,27 only one of 12 bone marrow, selected CD34+ cells, performed a super- patients had evidence of weakly positive ELISA for natant transduction with a retroviral vector containing the antibodies to the protein. common gamma chain gene, and reinfused cells without The regulatory oversight for gene therapy trials involves prior chemoablation. In four of the five patients, trans- two separate but parallel systems. These are the public duced T, B and NK cells were persistently detected in the review process of the NIH Recombinant DNA Advisory peripheral circulation. T-cell levels recovered from near Committee and the regulatory oversight of the Food and undetectable to within normal limits, with sustained levels Drug Administration. The role of these bodies is to ensure for the duration of follow-up. Furthermore, this translated that each study is conducted with clear objectives, rationale into a significant clinical benefit with resolution of and research design, including risks and benefits and infectious complications, and other chronic symptoms detailed methods. Preclinical studies are necessary, includ- from the SCID. ing toxicology studies in at least one animal species.28 Unfortunately, in October 2002, the FDA released information indicating that one, then a second of the 10 Clinical success children treated, who have now been treated on this trial, had developed a T-cell monoclonal lymphoproliferative All clinical trials to date have used murine retroviral vectors disorder culminating in acute lymphocytic leukemia. This for transduction. Historically, studies of gene transfer were process was genotyped and the leukemic cells were found to

Bone Marrow Transplantation Hematopoietic stem cell gene therapy JL Vollweiler et al 5 contain the provirus clonally inserted into the LMO2 site Table 4 Key technological advances in hematopoietic stem cell on chromosome 11, commonly associated with a transloca- gene transfer tion activation of LMO2 in T cell ALL (t 11; 14). Since the Improved onco-retroviral backbones classical translocation did not occur, it remains unclear MFG whether the insertion was causative in the leukemia. Well- MSCV founded concerns with all other stem cell gene therapy MND trials is warranted and all trials on immunodeficient Improved viral proteins for transduction of human cells children have been asked to more carefully monitor for GALV in PG-13 cells monoclonality by the FDA.39 Whether insertional muta- RD-114 genesis with clinical consequence is a realistic possibility in VSVG other stem cell trials in which cells are infused into hosts Improved adherence of stem cells during transduction with a competent immune system and without the intense CH-296 fibronectin fragment pressure for T-cell expansion will unfortunately be an Improved cytokines to stimulate retroviral gene integration observation that derives from ongoing clinical trials. Stem cell factor In chronic granulomatous disease, five patients received Thromobopoietin autologous cells transduced with the p47phox gene.40 FLT-3 ligand Oxidase-positive cells were detected at around 0.1% of circulating neutrophils for up to 12 weeks. However, the effect appeared transient, and it is unclear if stem cells were doxorubicin or vinblastine, all known substrates of the p- actually transduced. glycoprotein efflux pump encoded by MDR. By semiquan- After repeated infusions of glucocerebrosidase-trans- titative PCR, three patients appeared to have stable duced peripheral blood stem cells in patients with detection of both genes, while three patients showed Gaucher’s disease, transduced circulating lymphocytes as enrichment for the MDR gene relative to the neo gene well as enzymatic activity were detected for up to several with chemotherapy. This demonstrates for the first time in months.41 Other investigators have been unable to dupli- a human trial that a chemotherapy resistance gene is a cate these results.42 In addition, Gaucher’s disease does not selectable and enrichable marker. show the classic cross-correction demonstrated by other At our center, a trial is currently underway investigating lysosomal storage diseases, where a small number of the MGMT chemotherapy resistance gene, which protects normal hematopoietic cells can serve as enzyme donors from alkylation damage. This trial takes advantage of the for many cells of differing lineages.43 Therefore, it is demonstration that mutations within MGMT render it doubtful that low levels of gene transfer will translate into active for repair of DNA damage but resistant to a any significant clinical benefit. clinically used inhibitor, O6-benzylguanine.50 In preclinical Similarly, when patients with Fanconi anemia comple- models, selection at the stem cell level in primary and mentation group C received transduced autologous cells, secondary recipient mice is in excess of 1000-fold when the corrected circulating leukocytes were detectable for several primary recipient does not undergo myeloablation.51 High months, but disappeared by 1 year.44 levels of expression are sufficient to protect the transduced Another active area of investigation has been transduc- hematopoietic progenitors from repeated doses of BCNU tion of chemotherapy resistance genes, most notably MDR- or temozolomide, rendering the marrow resistant to the 1. Several groups have demonstrated the feasibility of chemotherapy. Thus, instead of experiencing cumulative transferring the MDR-1 gene, typically in the setting of a myelosuppression, the marrow is enriched with transduced myeloablative autologous transplant.45–47 Levels of gene drug-resistant cells. Should this selection strategy be marking have typically been less than 5%. More recently, effective in humans, it might be reasonable to consider transduction levels of 5–15% of bone marrow progenitor using a mutant, O6-benzylguanine-resistant MGMT cells, persisting up to 1 year, have been shown using the coupled with another gene to transduce hematopoietic CD-296 fibronectin fragment during transduction stem cells with therapeutic genes, such as those outlined in (Table 4).27 Table 1. Transduction has also been shown to translate into a clinical benefit, albeit small. Four patients were transplanted with cells transduced with the MDR gene, then Conclusions treated with six cycles of paclitaxel. Three demonstrated engraftment of the genetically modified cells. While no Progress has been slow, and gene therapy has not proven to enrichment was shown with chemotherapy, there was a be the cure-all magic bullet promised by the press in early correlation between the per cent of granulocytes containing 1980s (Table 5). However, progress is being made. the transgene and the granulocyte nadirs following Significant advances have occurred in vector design, chemotherapy.48 especially with lentiviral vectors, opening up whole new In another novel study, the Cowan group at the NIH avenues of possible clinical investigation. Modest improve- infused autologous cells into breast cancer patients after ments in transduction efficiency are made possible with the ablative chemotherapy, one half of which was incubated advent of new cytokines and combinations, and the use of with a Neo vector supernatant, and the other half of which fibronectin. Remaining issues involve optimal expression, was incubated with a MDR vector supernatant.49 Patients stem cell delivery, concerns with insertional mutagenesis then received chemotherapy with paclitaxel followed by and silencing over time that could limit the efficacy of long-

Bone Marrow Transplantation Hematopoietic stem cell gene therapy JL Vollweiler et al 6 Table 5 9 Reyes M, Verfaillie CM. Characterization of multipotent adult progenitor cells a subpopulation of mesenchymal stem cells. The future of hematopoietic stem cell gene transfer directions Ann NY Acad Sci 2001; 938: 231–233 discussion 233–235. Use of safety-modified lentiviruses in human clinical trials Gene transfer of hematopoietic stem cells with transdifferentiation 10 Reyes M, Dudek A, Jahagirdar B et al. Origin of endothelial Dual gene transfer with therapeutic genes and drug selection progenitors in human postnatal bone marrow. J Clin Invest 2002; 109: 337–346. Unresolved issues 11 Kocher AA, Schuster MD, Szabolcs MJ et al. Neovasculariza- Insertional mutagenesis tion of ischemic myocardium by human bone-marrow-derived Multiple random integrations angioblasts prevents cardiomyocyte apoptosis, reduces remo- Poor gene expression and silencing especially of large genes (beta globin) deling and improves cardiac function. Nat Med 2001; 7: 430– Targeting repopulating stem cells 436. 12 Roe T, Reynolds TC, Yu G et al. Integration of murine leukemia virus DNA depends on mitosis. EMBO J 1993; 12: 2099–2108. 13 Naldini L, Blomer U, Gallay P et al. In vivo gene delivery and term correction of genetic disorders. While safety issues stable transduction of nondividing cells by a lentiviral vector. Science remain paramount, clinical experience is beginning to 1996; 272: 263–267. 14 Akkina RK, Walton RM, Chen ML et al. High-efficiency gene accumulate, and thus far hematopoietic stem cell gene transfer into CD34+ cells with a human immunodeficiency therapy has proven fairly safe. The concept of in vivo virus type 1-based retroviral vector pseudotyped with vesicular selective advantage and selectable markers is beginning to stomatitis virus envelope glycoprotein G. J Virol 1996; 70: be fully understood and exploited. 2581–2585. With continued scientific endeavor and innovation and 15 Trono D. HIV accessory proteins: leading roles for the further clinical experience, particularly with further devel- supporting cast. Cell 1995; 82: 189–192. opment of lentiviral vectors, with their high titers, high 16 Farson D, Witt R, McGuinness R et al. A new-generation transgene expression and maintenance of the stem cell stable inducible packaging cell line for lentiviral vectors. Hum phenotype during transduction, further advances are Gene Ther 2001; 12: 981–997. certain. It is our hope that we are only just beginning 17 Zielske SP, Gerson SL. Lentiviral transduction of P140K MGMT into human CD34(+) hematopoietic progenitors at to realize the promise of hematopoietic stem cell gene low multiplicity of infection confers significant resistance to therapy. BG/BCNU and allows selection in vitro. Mol Ther 2002; 5: 381–387. 18 Miyoshi H, Smith KA, Mosier DE et al. Transduction of Acknowledgements human CD34+ cells that mediate long-term engraftment of NOD/SCID mice by HIV vectors. Science 1999; 283: 682–686. This work was supported by Public Health Service Grants 19 Woods NB, Fahlman C, Mikkola H et al. Lentiviral gene RO1CA73062, 2RO1CA63193 and RO1ES06288. transfer into primary and secondary NOD/SCID repopulating cells. Blood 2000; 96: 3725–3733. 20 Hanenberg H, Xiao XL, Dilloo D et al. Colocalization of References retrovirus and target cells on specific fibronectin fragments increases genetic transduction of mammalian cells. Nat Med 1 Goodell MA, Brose K, Paradis G et al. Isolation and 1996; 2: 876–882. functional properties of murine hematopoietic stem cells that 21 Hanenberg H, Hashino K, Konishi H et al. Optimization of are replicating in vivo. J Exp Med 1996; 183: 1797–1806. fibronectin-assisted retroviral gene transfer into human 2 Morel F, Galy A, Chen B et al. Equal distribution of CD34+ hematopoietic cells. Hum Gene Ther 1997; 8: 2193– competitive long-term repopulating stem cells in the CD34+ 2206. and CD34À fractions of Thy-1lowLin-/lowSca-1+ bone 22 Murray LJ, Young JC, Osborne LJ et al. Thrombopoietin, flt3, marrow cells. Exp Hematol 1998; 26: 440–448. and kit ligands together suppress apoptosis of human 3 Osawa M, Hanada K, Hamada H et al. Long-term lympho- mobilized CD34+ cells and recruit primitive CD34+ Thy-1+ hematopoietic reconstitution by a single CD34À low/negative cells into rapid division. Exp Hematol 1999; 27: 1019–1028. hematopoietic stem cell. Science 1996; 273: 242–245. 23 Schilz AJ, Schiedlmeier B, Kuhlcke K et al. MDR1 gene 4 Sato T, Laver JH, Ogawa M. Reversible expression of CD34 expression in NOD/SCID repopulating cells after retroviral by murine hematopoietic stem cells. Blood 1999; 94: 2548– gene transfer under clinically relevant conditions. Mol Ther 2554. 2000; 2: 609–618. 5 Koc ON, Lazarus HM. Mesenchymal stem cells: heading into 24 Schiedlmeier B, Wermann K, Kuhlcke K et al. Human the clinic. Bone Marrow Transplant 2001; 27: 235–239. multidrug resistance-1 gene transfer to long-term repopulating 6 Deans RJ, Moseley AB. Mesenchymal stem cells: biology and human mobilized peripheral blood progenitor cells. Bone potential clinical uses. Exp Hematol 2000; 28: 875–884. Marrow Transplant 2000; 25(Suppl 2): S118-S124. 7 Koc ON, Peters C, Aubourg P et al. Bone marrow-derived 25 Donahue RE, Kessler SW, Bodine D et al. Helper virus mesenchymal stem cells remain host-derived despite successful induced T cell lymphoma in nonhuman primates after hematopoietic engraftment after allogeneic transplantation in retroviral mediated gene transfer. J Exp Med 1992; 176: patients with lysosomal and peroxisomal storage diseases. Exp 1125–1135. Hematol 1999; 27: 1675–1681. 26 Bunting KD, Galipeau J, Topham D et al. Transduction of 8 Baxter MA, Wynn RF, Deakin JA et al. Retrovirally mediated murine bone marrow cells with an MDR1 vector enables ex correction of bone marrow-derived mesenchymal stem cells vivo stem cell expansion, but these expanded grafts cause a from patients with mucopolysaccharidosis type I. Blood 2002; myeloproliferative syndrome in transplanted mice. Blood 1998; 99: 1857–1859. 92: 2269–2279.

Bone Marrow Transplantation Hematopoietic stem cell gene therapy JL Vollweiler et al 7 27 Abonour R, Williams DA, Einhorn L et al. Efficient retro- 42 Dunbar CE, Kohn DB, Schiffmann R et al. Retroviral virus-mediated transfer of the multidrug resistance 1 gene into transfer of the glucocerebrosidase gene into CD34+ cells autologous human long-term repopulating hematopoietic stem from patients with Gaucher disease: in vivo detection of cells. Nat Med 2000; 6: 652–658. transduced cells without myeloablation. Hum Gene Ther 1998; 28 Evans ME, Jordan CT, Chang SM et al. Clinical infection 9: 2629–2640. control in gene therapy: a multidisciplinary conference. Infect 43 Wolfe JH, Sands MS, Barker JE et al. Reversal of pathology in Control Hosp Epidemiol 2000; 21: 659–673. murine mucopolysaccharidosis type VII by somatic cell gene 29 Brenner MK, Rill DR, Holladay MS et al. Gene marking to transfer. Nature 1992; 360: 749–753. determine whether autologous marrow infusion restores 44 Walsh CE, Mann MM, Emmons RV et al. Transduction of long-term haemopoiesis in cancer patients. Lancet 1993; 342: CD34-enriched human peripheral and umbilical cord blood 1134–1137. progenitors using a retroviral vector with the Fanconi anemia 30 Dunbar CE, Cottler-Fox M, O’Shaughnessy JA et al. Retro- group C gene. J Investig Med 1995; 43: 379–385. virally marked CD34-enriched peripheral blood and bone 45 O’Shaughnessy JA, Cowan KH, Nienhuis AW et al. Retroviral marrow cells contribute to long-term engraftment after mediated transfer of the human multidrug resistance gene autologous transplantation. Blood 1995; 85: 3048–3057. (MDR-1) into hematopoietic stem cells during autologous 31 Stewart AK, Sutherland DR, Nanji S et al. Engraftment of transplantation after intensive chemotherapy for metastatic gene-marked hematopoietic progenitors in myeloma patients breast cancer. Hum Gene Ther 1994; 5: 891–911. after transplant of autologous long-term marrow cultures. 46 Hanania EG, Giles RE, Kavanagh J et al. Results of Hum Gene Ther 1999; 10: 1953–1964. MDR-1 vector modification trial indicate that granulocyte/ 32 Blaese RM, Culver KW, Miller AD et al. T lymphocyte- macrophage colony-forming unit cells do not contribute directed gene therapy for ADA- SCID: initial trial results after to posttransplant hematopoietic recovery following 4 years. Science 1995; 270: 475–480. intensive systemic therapy. Proc Natl Acad Sci USA 1996; 33 Onodera M, Ariga T, Kawamura N et al. Successful peripheral 93: 15346–15351. T-lymphocyte-directed gene transfer for a patient with severe 47 Hesdorffer C, Ayello J, Ward M et al. Phase I trial of combined immune deficiency caused by adenosine deaminase retroviral-mediated transfer of the human MDR1 gene as deficiency. Blood 1998; 91: 30–36. marrow chemoprotection in patients undergoing high-dose 34 Bordignon C, Notarangelo LD, Nobili N et al. Gene therapy chemotherapy and autologous stem-cell transplantation. J Clin in peripheral blood lymphocytes and bone marrow for ADA- Oncol 1998; 16: 165–172. immunodeficient patients. Science 1995; 270: 470–475. 48 Cowan KH, Moscow JA, Huang H et al. Paclitaxel 35 Hoogerbrugge PM, van Beusechem VW, Fischer A et al. Bone chemotherapy after autologous stem-cell transplantation and marrow gene transfer in three patients with adenosine engraftment of hematopoietic cells transduced with a retro- deaminase deficiency. Gene Therapy 1996; 3: 179–183. virus containing the multidrug resistance complementary DNA 36 Kohn DB, Weinberg KI, Nolta JA et al. Engraftment of gene- (MDR1) in metastatic breast cancer patients. Clin Cancer Res modified umbilical cord blood cells in neonates with adenosine 1999; 5: 1619–1628. deaminase deficiency. Nat Med 1995; 1: 1017–1023. 49 Moscow JA, Huang H, Carter C et al. Engraftment of MDR1 37 Cavazzana-Calvo M, Hacein-Bey S, de Saint Basile G et al. and NeoR gene-transduced hematopoietic cells after breast Gene therapy of human severe combined immunodeficiency cancer chemotherapy. Blood 1999; 94: 52–61. (SCID)-X1 disease. Science 2000; 288: 669–672. 50 Reese J, Davis B, Liu L, Gerson SL. Simultaneous protection 38 Hacein-Bey-Abina S, Le Deist F, Carlier F et al. Sustained of G156A methylguanine DNA methyltransferase gene-trans- correction of X-linked severe combined immunodeficiency by duced hematopoietic progenitors and sensitization of tumor ex vivo gene therapy. N Engl J Med 2002; 346: 1185–1193. cells using O6-benzylguanine and temozolomide. Clin Cancer 39 Bonetta L. Leukemia causes tightening of gene therapy Res 1999; 5:163–169. controls. Nat Med 2002; 8: 1189. 51 Davis BM, Koç ON, Gerson SL. Limiting numbers of G156A 40 Malech HL. Progress in gene therapy for chronic granuloma- O6- methylguanine-DNA methyltransferase-transduced mar- tous disease. J Infect Dis 1999; 179(Suppl 2): S18-S325. row progenitors repopulate nonmyeloablated mice after drug 41 Barranger JA, Rice EO, Swaney WP. Gene transfer ap- selection. Blood 2000; 95: 3078–3084. proaches to the lysosomal storage disorders. Neurochem Res 1999; 24: 601–615.

Bone Marrow Transplantation