This subproject is one of many research subprojects utilizing theresources provided by a Center grant funded by NIH/NCRR. The subproject andinvestigator (PI) may have received primary funding from another NIH source,and thus could be represented in other CRISP entries. The institution listed isfor the Center, which is not necessarily the institution for the investigator.To evaluate the safety of escalating doses of 14g2a.zeta chimeric receptor transduced autologous EBV specific cytotoxic T-lymphocytes (EBV-CTL) and 14g2a.zeta transduced autologous peripheral blood T-cells administered to patients with Neuroblastoma who have been lymphodepleted by CD45 monoclonal antibodies (MAbs).Secondary ObjectivesTo determine the differential survival and function of these two infused cell-types in vivo, in particular to determine if chimeric receptor transduced EBV-CTLs survive longer than transduced peripheral-blood T-cells.To determine anti-tumor effects of transduced peripheral blood T-cells and EBV specific CTLs in vivo.3. Background and Rationale and Interpretation3.1 NeuroblastomaNeuroblastoma, the most common extracranial solid tumor of childhood, affects approximately 600 children a year in the United States. When the tumor occurs in infants (children 1 year of age), the prognosis is far worse. Although patients with localized disease may still be cured by conventional therapy, 80% or more of those with disseminated tumor can be expected to relapse within 3 years, and virtually none of this subgroup will become long-term survivors4 Over the past decade, attempts to improve the outcome of advanced neuroblastoma have focused on greater intensification of the induction and consolidation phases of chemo-radiotherapy, with or without stem cell rescue.5-8 Despite improvements in remission rates, long-term survival remains poor. This failure has lead to a resurgence of interest in alternative methods of disease eradication immune modulation as one particular option.9,103.2 Immunotherapy of NeuroblastomaThe concept of exploiting the immune system to eradicate disease has had along currency in neuroblastoma research, supported by many aspects of thetumor's biologic behavior, including spontaneous regression in younger children.3 In addition, because neuroblastoma is derived from embryonic neuroectoderm, it expresses antigens not widely detected in non-embryonic tissues11,12 and may overexpress other cellular antigens.4 Finally, in animals and in man, neuroblastoma cells are susceptible to cytotoxic effector mechanisms both in vitro and in vivo.13-15 Clinical immunotherapy for neuroblastoma has taken many forms which may be loosely classified into vaccine / cytokine approaches or the use of therapeutic monoclonal antibodies.One of the earliest attempts to use immunotherapy was to immunize patientswith disrupted tumor mixed with bacillus Calmette-Guerin (BCG). These crudemixtures produced severe local reactions, but did lead to detectable, albeittransient, tumor responses (1982 protocol developed at St. Jude Children fsResearch Hospital, unpublished results). Recombinant cytokines such as IL-2,which stimulate the immune system,16 or agents such as IFN- , which mayrender neuroblastoma cells more immunogenic to cytotoxic T cells by increasing their expression of class I MHC molecules have also been tried.17 More recently, gene-modified tumor vaccines have been used. This approach is based on murine studies showing that tumor cell lines transfected with cytokine genes show a substantial increase in their immunogenicity.18 These tumor vaccines enhanced the immune response sufficiently to reject the modified tumor cells, as well as nontransduced tumor cells administered concomitantly or shortly thereafter. Our own group has studied the effects of expressing the IL-2 gene alone or in combination with lymphotactin in autologous or allogeneic neuroblasts, which were then injected into children with advanced neuroblastoma.19-21 In patients receiving IL-2 alone, significant tumor responses were seen, including 3 complete remissions among 28 patients treated to date. These patients have also made antibodies that react with autologous and allogeneic neuroblastoma cells in cell surface immunofluorescence studies.19 The most widely used immunotherapy for neuroblastoma consists of murine or human-murine chimeric monoclonal antibodies directed against GD2, a disialoganglioside expressed in tumors of neuroectodermal origin. There have been both Phase I and Phase II studies of these antibodies, and the reported Phase I data revealed an apparently significant response rate, with pain, hypertension, fever and rashes reported as the major adverse effects. Other studies have coupled these antibodies to toxins or to radionuclides with broadly comparable results.22-25 It is our intention to combine these successful clinical cellular and antibodymediated immunotherapies of neuroblastoma by generating cytotoxic T cellsexpressing chimeric single chain antibodies directed against GD-2 anti-tumorimmune receptors.3.3 Chimeric Immune ReceptorsChimeric receptors are generated first by joining the heavy and light chainvariable regions of a monoclonal antibody with a linker to form a single-chain Fv (scFv) molecule. This scFv is then attached to the transmembrane andcytoplasmic portion of T Cell Receptors (TCR) chain via a flexible hinge region. Engagement of the extracellular scFv of the chimeric receptor results in tyrosine phosphorylation of immune-receptor activation motifs present in the cytoplasmic domain, initiating T cell signaling to the nucleus. Human T lymphocytes genetically engineered to express these recombinant receptor genes have exhibited specific lysis via the perforin/granzyme pathways, as well as cytokine secretion upon exposure to tumor cells expressing the cognate target antigen.26 Engagement of single T cell or Fc receptor chains suffices to induce cellular activation and proliferation.27-30 Adoptively transferred chimeric receptortransduced cells were protective in murine tumor models.31-33 Chimeric receptor transduced T cells have numerous advantages over immunotherapies based on monoclonal antibodies or T lymphocytes alone. Since there is no need to select and expand tumor-specific antigens from scanty precursors, large populations of antigen-redirected T lymphocytes can be obtained in a matter of weeks. Moreover, chimeric T cell receptors are MHCunrestricted, so that tumor escape by downregulation of HLA class I molecules or defects in antigen processing are bypassed. Finally, since both CD4+ and CD8+ T cells can express the same chimeric receptor the full network of T cell functionis directed against tumor cells. The presence of chimeric TCR mediated effector function may be more likely to produce tumor cell lysis than humoral immune responses alone. The perforin/granzyme killing mechanism may be effective against cells that are relatively resistant to antibody and complement, while cytokine secretion upon T cell activation by tumor antigen recruits additional components of the immune system, amplifying the antitumor immune response. Furthermore, unlike intact antibodies, T cells can migrate through microvascular walls, extravasate and penetrate the core of solid tumors to exert their cytolytic activity. Finally, a single T lymphocyte can sequentially kill a multiplicity of target cells.3.4 Lack of Persistence of Transduced T Cells As The Major Limitation And Our Solution There is little doubt from clinical studies that activated chimeric T cells rapidlylose their function and may disappear from the circulation. The principal problem appears to be that scFv expressing T cells rapidly return to a resting state. Both in vitro and in vivo studies of T cells stimulated by antigenthrough their conventional receptors has indicated that in the absence of continuing costimulation and of growth factors (as in most tumor directed killing), there is a progressive decrease in proliferation and loss of cytotoxic T cell activity34,35. We propose to overcome this deficit by expressing the tumor-specific chimeric TCR in EBV-specific T cells. These cells will express their native TCR receptor, which targets EBV antigens. Simultaneously they will express the chimeric TCR targeting GD-2 expressed on neuroblastoma. As EBV infection persists lifelong all positive individuals have a background EBV infected cells which are deleted by EBV-specific CTLs. Our proposition is that this background of EBV positive cells will provide a constant activating and proliferating stimulus to the infused transduced CTLs. Transduced cells will kill tumor neuroblastoma targets but will not depend on them for activation and co-stimulation. Clearly this will limit the study to patients who have been previously infected with EBV. Unlike most tumor cells (including malignant neuroblasts), EBV transformed B cells express not just specific antigens but also high levels of many different costimulatory molecules, including CD40L and B7.1 and B7.2. Because they express both Class I and Class II MHC molecules they can present viral epitopes that will stimulate both CD8 and CD4 virus specific cells, favoring the cognate interactions between T cell subsets that are critical for optimal and sustained immune responses.36,37 Indeed our past experience with gene marked EBVspecific CTL has shown that they can persist for 6 years or more after infusion, provided the specific lines infused contains a combination of both CD4- and CD8- positive cells.38It is our hypothesis that any decay in anti-tumor CTL function (mediated by thelimited activation capacity of the scFv molecule) will be offset by theimmunostimulatory capacity of EBV+ targets chronically present in any EBVpositive individual, which would engage transduced T cells through theirconventional T cell receptors. Our model predicts that infused chimeric T cellswill only intermittently encounter their tumor targets. In the interim resting state, their activation status decays, and if they encounter tumor cells there is a failure of reactivation. Subsequently resting cells are either lost from the circulation or apoptose. If, however, the chimeric T cells are also EBV-specific, encounters with virally infected targets would lead to reactivation due to adequate expression of costimulators and antigen, recognized by the conventional T cell receptor, coupled with cognate interactions between CD4 and CD8 T cell subsets. The consequence would be expansion/persistence. Once their encounter with EBVexpressing targets has terminated, these activated chimeric T cells will be capable of recognizing tumor-specific targets, since a single T cell can sequentially engage multiple targets. Thus, a major purpose of this study is to confirm that EBV-positive B cells will positively influence the survival and function of chimeric T cells. There is considerable evidence to support this hypothesis. First, clinical studies of more than 70 recipients of CTL have shown that both CD4+ and CD8+ EBV specific T cells persist in activated form and recirculate.39,40 Second, our gene-marking studies have shown that these circulating EBV-specific T cells will respond both in vivo and ex vivo to viral antigen challenge, by marked proliferation and increased activation, even when this challenge occurs 6 years or more from the date of infusion. Finally, multiple animal studies have shown that the continued presentation of antigens associated with class I and class II MHC molecules on effective antigen presenting cells is critical to establish and maintain T cell memory, since it is only in this way that cognate interactions can occur between CD4+ and CD8+ effector cells.36,37 T cells generated in this way will persist and readily recapitulate an immune response on antigenic challenge.3.5 Testing This HypothesisTo fully test this hypothesis we plan to simultaneously administer bothtransduced EBV-CTLs and transduced peripheral blood T-cells. Both cell types will express identical chimeric receptors but to allow differential tracking by real-time PCR a modification has been introduced into the non-coding region of the retroviral vector. Each patient will receive CTLs transduced with one vector and peripheral blood T-cells with the other. The allocation of different vectors will be randomized to avoid bias introduced by unanticipated biological differences between the vectors. Hence vectors are identical except for a short (13 bases) oligonucleotide inserted after the stop codon of the chimeric receptor and before the 3 f LTR. These oligonucleotides will not be translated so the chimeric receptors generated should be identical. Since they are after splice-donors and acceptors and after promotor regions they should not alter expression levels. However they will easily allow differentiation by PCR. By measuring the level of the transgene in peripheral blood, we will be able to estimate the overall kinetics of gene-modified T-cell survival and determine whether expansion or persistence occurs in vivo. Currently, we can regularly detect a single neo-marked cell in 50,000 cells by real-time PCR analysis, which has been sufficient to track T cells for up to 6 years in our ongoing EBVprophylaxis study. Even when EBV- specific CTLs have been given to nonmyeloablated Hodgkin patients, marker signals have persisted for more than 1 year, the limit of the study period.41 Detection of the current vector should be equally sensitive. Importantly we will also be able to estimate the proportion of the 14.G2a cells originating from the infused EBV-CTLs and from the transduced PBT cells. By using the discriminating oligos the real-time signal strength for each of the two vectors can be compared at each of the time points of study. Hence, 30 mL of patient blood will be taken preinfusion, and at various intervals subsequently. The proportion of 14g2a-expressing cells in the CD4+ and CD8+ subsets will be determined by immunofluorescence studies. This will be compared with the level of transduced cells using quantitative real-time PCR for retroviral integrants. This will allow us to determine the persistence of the transduced cells and establish their continued ability to express the transgene. Because gene persistence and expression need not correlate with antitumor effector function, we will also determine the functionality of the peripheral bloodT cells by coculturing them with GD2 positive neuroblastoma cell lines andmeasuring IFN-gamma release. Even if CTL-derived signal does persist, it is possible that the transgene-positive cells we detect will have lost one or other of their specificities for EBV and neuroblastoma. Antigen-specific CD8+ T lymphocytes can be isolated and identified42 using tetrameric HLA peptide complexes which bind stably and specifically to appropriate MHC peptide-specific T cells. Fluorochrome-conjugated tetramers can be used to sort peptide-specific T cells from peripheral blood and from CTL lines. We have obtained HLA A2 and HLA A24 and B8 and B44 tetramers complexed to well characterized immunodominant peptides derived from LMP2a and EBNA3a/3c from the NIAID tetramer facility. We have shown that these tetramers can identify CD8+ EBV-specific T cells from peripheral bloodand CTL lines from seropositive donors. Hence, in any patient with one or more of above HLA polymorphisms (which should be 75% of our study population), it will be possible to demonstrate expression of both the 14.G2a sc receptor and the EBV-specific receptor on the same cell. Such analyses will be performed before the cells are infused, and circulating cells will be measured at the intervals described above. This will allow us to detect variations in the proportion of cells that are doubly positive for both receptors, as compared to those that express only one or other of the receptor.We will also be able to use FACS sorting of tetramer positive cells to discoverwhether or not they retain their bifunctional activity against malignant and EBVinfected target cells.43 We will coculture the tetramer/14.G2a-positive T cells with either EBV-LCLs or GD2-positive neuroblasts and measure their response in cytotoxic and cytokine release assays. This approach will only be possible in patients with HLA types for which EBV tetramers exist.3.6 Safety of Targeting The GD-2 Antigen In Patients With NeuroblastomaGD2 (Diasialoganglioside II3a(NeuAc)2GgOse3Cer] ) is a disialogangliosideexpressed in tumors of neuroectodermal origin. Gangliosides are sialic-acid(neuraminic acid) containing glycosphingolipids. They consist of a sialatedpolysaccharide chain linked through a s-glycosidic linkage to a ceramide moiety, which anchors them into the plasma membrane. They are major constituents of neuronal cell membranes and endoplasmic reticulum. The principal gangliosides of normal mammalian brain are GM1, GD1a, GD1b and GT1a.The GD2 tumor associated antigen seems to be ideal for Ab-mediated therapy of neuroblastoma, since it is expressed at high density on almost all neuroblastoma cells, is poorly expressed or absent from most normal tissue, and does not modulate its expression upon mAb binding. It is expressed on cerebellum and peripheral nerves. Many clinical studies have been performed using a variety of monoclonal antibodies targeting GD-2. Published studies are summarized in table 3.6.1 Table 3.6.1Clone Effects ReferenceL72 8 patients with melanoma were treated. One patientdeveloped complete regression. Side effects were limitedto mild erythema. 44 3F8 9 patients with melanoma and 8 with neuroblastoma. Four major responses noted. Complement depletion,hypertension and urticaria were noted. 45 14g2a 12 patients with melanoma. Reversable peripheral motor neuropathy. Delayed pain sydrome. High HAMA titers in all patients. SIADH also occurred. 46 14g2a 5 neuroblastoma patients . 2 partial remissions. 47 Ch14.18 9 patients with stage IV neuroblastoma. Disease response in 6/9. Abdominal and joint pain, pruritis, urticaria and optic atrophy occurred. 48 Ch14.18 16 patients with metastatic melanoma. No clinical benefit noted. Abdominal pain, hypotension, dysasthesias, atrial fibrillation. 47 14g2a 33 pediatric patients with neuroblastoma and GD-2positive osteosarcoma. Response was noted in 9 patients. Side-effects included fever, thrombocytopenia, neutropenia, diarrhea, bronchospasm, hypotension and angioedema. 25 Ch14.18 10 pediatric patients with neuroblastoma. Response noted in 5. Toxicities included pain, tachycardia,hypertension, fever, rashes, nephropathy, arthritis and notably mild peripheral paresthesias. 22 3F8 coupled to I-131 24 previously untreated patients with neuroblastoma, 75% remain progression-free with a median follow-up of19 months 49 Disease responses were noted. The majority of toxicities seemed consequences of infusing murine antibodies into human subjects. Although we will be using variable regions derived from one of these monoclonals (14g2a) these will be integrated into the chimeric T-cell receptor and expressed on T-cells. Disease responses were noted. The majority of toxicities seemed consequences of infusing murine antibodies into human subjects. Although we will be using variable regions derived from one of these monoclonals (14g2a) these will be integrated into the chimeric T-cell receptor and expressed on T-cells. 3.7 Enhancing the activity of infused CTL.Unfortunately, treatment of the first five patients on this protocol hasrevealed no evidence for in vivo cellular expansion, even of the chimericEBV-CTL, and has also shown a failure of persistence. As anticipatedfrom this failure of biologic activity, there have been no adverse eventsand no tumor responses. In this protocol modification we will build onobservations made by other investigators and by our own group toattempt to promote in vivo cellular expansion and persistence.Following stem cell transplantation there is recovery of the T lymphocytecompartment. This proliferative environment favors expansion of infusedT lymphocytes. Our studies with gene marked EBV- CTL indicated morethan 1000 fold expansion of the infused cells within 7 days.50,51 In mostpatients with cancer, including those with neuroblastoma, the immunesystem is closer to a steady state, and we and others have consistentlyfailed to observe an equivalent expansion of the infused cells. Since thepatients we treat have advanced malignancies, with upwards of 1012tumor cells, and since we plan to infuse at most 2 x 108 chimeric CTL and2 x 108 chimeric T cells, it is axiomatic that an effective anti-tumorresponse will require significant in vivo expansion and persistence offunctional effector cells. Recently, Rosenberg fs group described how aproliferative environment could be artificially induced, by administration ofFludaribine and cyclophosphamide.52 Patients with advanced melanomareceived lymphoreductive doses of these cytotoxic drugs and were theninfused with autologous tumor infiltrating lymphocytes (TIL). In 6 patientsthere was marked expansion of the infused cells associated with tumorresponses; in two patients the tumor responses were complete, and theinfused TIL came to dominate the lymphoid compartment, suggesting arelationship between cell expansion and anti-tumor activity.52 In thismodification to our original treatment plan, we will investigate whether thesame benefits can be seen in patients with advanced relapsedneuroblastoma who are receiving T cells or EBV-CTL that have beengenetically modified to express chimeric receptors specific for theneuroblastoma-associated antigen GD2. We are making this modification because the first 5 patients showed no (4/5) or minimal presence of infused cells in the peripheral blood (95% depletion of peripheral bloodlymphocytes, predominantly T cells and NK cells. It reduced neutrophilcounts by 90%, but as anticipated, from the murine study, marrowsampling showed retention of CD34+ progenitor cells and this wasfollowed by a partial recovery of the neutrophil count by 48hrs, at whichpoint the ablative conditioning was administered.54 Subsequently theCD45 MAbs have been used in 3 patients as a subablative preparativeregimen. There was persistent lymphopenia but rapid recovery ofrecipient neutrophils until completion of donor engraftment (unpublishedinformation).A tolerated dose of these MAbs, 400 g/kg/day x 4 days, produced apattern of response identical to that seen in the mice, with markedreductions in circulating lymphoid and myeloid cells while largely sparingmarrow progenitors. Because the marrow progenitor compartment wasspared, there was rapid regeneration of CD45 positive cells, whichremoved free antibody from the circulation, to produce a serum half-life of6-9hrs.54 In other words, within 36 hrs of terminating infusion, nodetectable MAb was present in the circulation. These dual features ofpotent T cell destruction coupled with a short half-life make this MAb anappropriate choice for T depletion in these patients, and it hassubsequently safely been used as a component of a subablativeconditioning regimen for high risk allogeneic stem cell transplantrecipients.55Finally, we have used these MAbs on 9 patients who received EBV- orLMP2-specific CTL for the treatment of EBV+ Hodgkin disease orrelapsed EBV+ Nasopharyngeal carcinoma. The infusions were welltolerated, and substantial and sustained rises in EBV/LMP2-specific CTLwere observed. In patients who had previously received EBV-CTL withoutMAb, analysis showed that precursor frequencies were an order ofmagnitude or more greater with infusions given after CD45, compared toinfusions before treatment. The responses also persisted longer (Figs 1-3)3.9 Potential ToxicitiesPotential toxicities may be categorized as those related to CD45 Mab, thoserelated to autologous T cell and EBV-CTLs infusion, and those caused bytransduction and cross-reactivity with normal tissues.3.9.1 CD45 MAb The dose limiting toxicity of this antibody relates to its ability to cause complement fixation, which is part of the mechanism of T cell depletion. At doses in excess of 1200-1600ug/kg/day patients may experience urticaria and wheezing. These symptoms resolve on stopping the MAb infusion and resolution is hastened by administration of steroids and antihistamines. At current dose levels (800ug400ug/kg/day times 24) complement mediated toxicities have been modest and the antibody is well tolerated. Hypersensitivity to the rat proteins has not been seen in any patients to date Nonetheless full precautions are observed during infusion as outlined in section 5 Treatment Plan.3.9.2 Infusion of autologous T-cellsCytotoxic T-lymphocytes will be autologous. Hence the risks of administration will be minimal as patient reactive cells will be absent. All EBV-CTLs will be tested for their ability to kill patient-derived PHA stimulated lymphoblasts, which would indicate autoreactivity. Lines with such activity will be eliminated from the study. Infusion of EBV transformed B cells which have been co-cultured with the CTL during generation of T-cell lines is unlikely to constitute an additional risk to the patient because the lymphoblastoid cells have been irradiated with 40Gy and cocultured with virus specific CTLs. The risk of introduction of infectious EBV into the patient is eliminated by the culture of LCLs for at least 14 days with Acyclovir prior to use. We have already shown that this abrogates the release of infectious virus from LCLs during the period of coculture. Finally, we will monitor levels of EBV-DNA in peripheral blood by PCR before and after CTL infusions. Over fifty patients have been treated on our previous CTL protocols in which allogeneic EBV specific CTLs were administered to treat EBV-positive tumors. We have not observed either of these two eventualities or other immediate complications that could be attributed to the CTL infusions. The main complication has been delayed inflammatory reaction at tumor sites in patients with disease at the time of CTL administration. Inflammation at sites of neuroblastoma is possible and patients with tumor at critical locations (for example potentially causing airway obstruction) will be excluded from the trial at the discretion of the principle investigator. If tumor inflammation has a lifethreateningmanifestation administration of corticosteroids should result in rapidresolution of inflammation.In this study we will also be administering activated transduced autologousperipheral blood T-cells. Previous clinical studies have reported administration of higher doses of activated transduced syngeneic56 and allogeneic57 T-cells with not serious adverse effects.3.9.3 Retroviral TransductionRetroviral transduction results in new random integration in host cell DNA, which rarely may cause abnormal or uncontrolled proliferation58. This effect is much more common with replication-competent retrovirus where each cell receives multiple integrants 59. We will test the producer line and all batches of supernatant with biological assays of RCR to exclude this possibility.Recently, there has been concern that even a single retroviral integration cancontribute to oncogenesis. Alain Fischer replaced the missing common gamma chain ( c) in X-SCID patients with ex-vivo retroviral transduction of autologous stem cells. Two of 11 treated patients have developed T-ALL associated with a single retroviral integrant. In both patients the integration site of the retrovirus was similar. It is likely that correction of common-gamma chain deficiency and related immunodeficiency syndromes will represent special cases. The c is a shared component of IL-2, IL-4, IL-7, IL-9 and IL-1560. Hence it is a crucial component T-cell proliferation and thymogenesis. A proliferative advantage is expected for progeny of stem cells with functional c. In female carriers of XSCID there is a pattern of non-random X-inactivation in T-cells, B-cells and NKcells61. Moreover a patient with X-SCID developed substantial numbers of T-cells following reversion of the mutant allele in a single haemopoietic stem cell 60. High efficiency retroviral transduction of human stem cells is difficult to achieve even with GALV pseudotyping62. It is likely that in X-SCID few truly pluripotent stem cells were transduced and the stem cell pool expressing the highest level of Transgene (due to integration at a transcriptionally active site) then undergo numerous doublings to restore the entire T-cell compartment. Random mutationscaused by this supra-physiological proliferation combined with a retroviralintegration in a transcriptionally active region led to leukemogenesis.Our proposal should have a very different risk profile. The construct transmitsan isolated CD3- signal, which is insufficient to induce proliferation63. While individual transduced EBV-specific CTLs may be expected to proliferate in response to EBV-infected cells, this proliferation will be modest andphyisiological and suffice only to maintain the EBV-specific T-cellcompartment. Retrovirally marked EBV-CTLs infused into humans result inlow-level marking with no evidence of marked proliferation64. Transducedcells will have no particular proliferative advantage over non-transduced EBVCTLs.Human gene-therapy studies of a chimeric receptor composed of theextracellular domain of CD4 and the intracellular domain of TCR- (as in ourstudy) retrovirally transduced into peripheral-blood T-cells to treat HIV havebeen performed 56,65. Large doses of transduced T-cells (orders of 1010 cells)were administered and no leukemogenesis occurred.To date more than 200 patients have received genetically modified cells inclinical trials 66 including patients we have treated on our protocols . usingretrovirally marked autologous marrow 67 or retrovirally marked EBV-CTLs 50.In none of these has malignancy caused by retroviral transduction beenreported. However, there is a possibility that the vector could randomlyintegrate into a site that could lead to leukemogenesis.Current intensive regimens to treat relapsed neuroblastoma describesubsequent secondary malignancy rates in excess of 10%. In light of this, thenatural history and poor prognosis of advanced neuroblastoma, and giventhe entire previous experience with retroviral gene therapy we feel that therisks of retroviral induced leukemogenesis are small and are justified in thispatient group.3.10.3 Cross-reactivityGD-2 is expressed at low levels on cerebellum and peripheral nerves. Clinicalstudies using 14g2a murine monoclonal (donor of variable domains for ourchimeric receptor) have resulted in reversible neurotoxicity. Patients in thesestudies developed high titers of human-anti-mouse antibodies and toxicitywas believed to be caused by immune precipitation of HAMA-14g2acomplexes. Although our chimeric receptor is derived from 14g2a mousehybridoma only the variable regions are retained and hence only anti-idiotype antibodies can be formed. Since the chimeric receptor is membrane-bound anti-idiotype antibodies will lead to clearance of the transduced cells in the spleen rather than immune-precipitation. Lower density of expression of GD2 is present43 on neural cells than on neuroblastoma so killing of neural cells by transduced T-cells is unlikely. Direct neurotoxicity from targeting of GD-2 expressing normal tissues by transduced cells is however still possible. In case of this eventuality a number of therapeutic maneuvers are available. Mostly likely administration of corticosteroids in doses used to treat graft versus host-disease (GvHD) will deplete the majority of circulating transduced cells. If this therapy fails addition of anti-thymocyte globulin (which is also used to treat GvHD) may be effective.
Showing the most recent 10 out of 459 publications
