|
|
Background We investigated the long-term outcome of gene therapy for severe combined immunodeficiency (SCID) due to the lack of adenosine deaminase (ADA), a fatal disorder of purine metabolism and immunodeficiency.
Methods We infused autologous CD34+ bone marrow cells transduced with a retroviral vector containing the ADA gene into 10 children with SCID due to ADA deficiency who lacked an HLA-identical sibling donor, after nonmyeloablative conditioning with busulfan. Enzyme-replacement therapy was not given after infusion of the cells.
Results All patients are alive after a median follow-up of 4.0 years (range, 1.8 to 8.0). Transduced hematopoietic stem cells have stably engrafted and differentiated into myeloid cells containing ADA (mean range at 1 year in bone marrow lineages, 3.5 to 8.9%) and lymphoid cells (mean range in peripheral blood, 52.4 to 88.0%). Eight patients do not require enzyme-replacement therapy, their blood cells continue to express ADA, and they have no signs of defective detoxification of purine metabolites. Nine patients had immune reconstitution with increases in T-cell counts (median count at 3 years, 1.07x109 per liter) and normalization of T-cell function. In the five patients in whom intravenous immune globulin replacement was discontinued, antigen-specific antibody responses were elicited after exposure to vaccines or viral antigens. Effective protection against infections and improvement in physical development made a normal lifestyle possible. Serious adverse events included prolonged neutropenia (in two patients), hypertension (in one), central-venous-catheter–related infections (in two), Epstein–Barr virus reactivation (in one), and autoimmune hepatitis (in one).
Conclusions Gene therapy, combined with reduced-intensity conditioning, is a safe and effective treatment for SCID in patients with ADA deficiency. (ClinicalTrials.gov numbers, NCT00598481
[ClinicalTrials.gov]
and NCT00599781
[ClinicalTrials.gov]
.)
Gene therapy is effective in patients with X-linked SCID,12 but its use has been hampered by the development of T-cell leukemia due to insertional mutagenesis caused by the retroviral vector.13,14 Pilot trials have shown the safety and feasibility of gene therapy in patients with SCID due to ADA deficiency,15,16,17 but all patients required maintenance with PEG-ADA, and the ADA-transduced stem cells were unable to reconstitute the recipient's immune system. We previously described two patients with ADA deficiency in whom nonmyeloablative conditioning allowed for substantial correction of the metabolic and immune defects 1 year after gene therapy.18 Here, we describe the long-term outcome of these two children and results in eight additional patients who were treated with nonmyeloablative conditioning followed by infusion of autologous CD34+ cells from bone marrow that had been transduced with a viral vector carrying the ADA gene.
Methods
Patients
Patients were enrolled from July 2000 through September 2006 in one of three phase 1–2 clinical protocols: one approved by the Hadassah University Hospital Ethics Committee and Israeli National Regulatory Authorities, and two approved by the San Raffaele Scientific Institute's Ethics Committee and the Italian National Regulatory Authorities (Table 1). Children with SCID due to ADA deficiency who lacked a healthy HLA-identical sibling were eligible for enrollment. In addition, patients who had been treated with PEG-ADA for at least 6 months were eligible in case of inefficacy, defined by immunologic measurements or as intolerance, allergic reaction, or autoimmunity.
|
Gene Therapy
Before gene therapy, a central venous catheter was implanted, and bone marrow specimens were obtained and cryopreserved for possible later use. On day 4 before gene therapy, autologous bone marrow specimens were again harvested under general anesthesia; mononuclear cells were isolated by means of density gradients, and CD34+ cells were purified with the use of immunomagnetic beads (CliniMACS, Miltenyi). CD34+ cells were stimulated with cytokines (fms-related tyrosine kinase 3 ligand, KIT ligand, thrombopoietin, and interleukin-3) and transduced with the retroviral vector (GIADAl) based on the Moloney murine leukemia virus carrying the human ADA gene.18 Supernatant production, cell isolation, and transduction were performed at MolMed according to current Good Manufacturing Practices. Nonmyeloablative conditioning involving the intravenous (or oral, in Patient 2) administration of 2 mg per kilogram per day of busulfan (Busilvex, Pierre Fabre) was performed on days 3 and 2 before gene therapy. Gene therapy consisted of the infusion of CD34+ marrow cells that had been transduced with the ADA-containing vector.18
Laboratory Studies
Blood and marrow samples were obtained from patients with SCID due to ADA deficiency, and blood samples were obtained from healthy children and adults as controls, with approval from the San Raffaele Scientific Institute's Ethics Committee and the Hadassah University Hospital Ethics Committee, according to standard ethical procedures. The Supplementary Appendix (available with the full text of this article at NEJM.org) describes measurements of cell subgroups, frequencies of transduced cells, results of flow cytometry, in vitro T-cell responses, antibodies generated after immunization,19 and ADA activity in cell lysates.20
Safety
Adverse events were recorded and reported according to Good Clinical Practice, and were updated as of August 31, 2008. Patients were monitored through clinical examination, imaging, and hematologic, immunologic, biochemical, and molecular tests, which included testing for replication-competent retrovirus.
Statistical Analysis
Mean or median values are reported, as appropriate. Clinical follow-up (including safety) data were updated as of August 31, 2008; analyses regarding molecular, biochemical, and immunologic variables were performed on data as of November 2007. Comparisons between values at various time points and between values for different variables at the same time point were performed by means of the Wilcoxon signed-rank test for paired data (two-tailed tests). The rates of infection (events per person-month of observation) and days of hospitalization were evaluated before gene therapy (from birth) and after therapy (from 4 months after gene therapy onward, to exclude an initial period of procedure-related hospitalization). The degree of correlation was expressed by means of Pearson's correlation coefficient. The data for Patient 2 were evaluated for efficacy until PEG-ADA was introduced at 4.5 years after gene therapy, whereas the data for Patient 8 were evaluated for safety only because PEG-ADA was reintroduced 0.4 year after gene therapy.
Results
Ten patients with SCID due to ADA deficiency who had early-onset manifestations (median, 2 months of age) underwent ADA gene therapy at a median age of 1.7 years (range, 0.6 to 5.6) (Table 1). This disorder was diagnosed at birth in one patient, a bone marrow transplant from a mismatched related donor had failed in four patients, and six patients had received PEG-ADA for more than 6 months, with an inadequate response. PEG-ADA was discontinued 3 weeks before gene therapy, to favor the growth of ADA-transduced cells. Patients underwent nonmyeloablative preconditioning with busulfan (total dose, 4 mg per kilogram of body weight) followed by the infusion of CD34+ marrow cells that had been transduced with the ADA-containing vector (mean dose, 8.2x106 CD34+ cells per kilogram, with an average of 28.6% transduced colony-forming units) (Table 1). In seven patients, absolute neutrophil counts were less than 0.5x109 per liter for more than 1 day after receipt of busulfan, and the duration of neutropenia was related to the area under the concentration–time curve of busulfan (r=0.65) (Table 1 and Figure 1 in the Supplementary Appendix). In two patients, neutropenia lasted over 30 days (Table 2); Patient 3 had neutropenia and thrombocytopenia and received platelet transfusions and infusion of the autologous marrow held in reserve (1.4x106 CD34+ cells per kilogram) at day 30 after gene therapy. The neutropenia in Patient 8 resolved after administration of granulocyte colony-stimulating factor. Patients were discharged after a median period of hospitalization of 42 days (range, 34 to 110), and follow-up continued according to the protocol and the guidelines of the Italian regulatory agency. Of the 10 patients, 8 did not require PEG-ADA during the follow-up period (Table 2). None of the patients received an allogeneic transplant after gene therapy.
|
There were no adverse events that could be attributed to the ADA-transduced cells. A mild and transient increase in liver enzyme levels was detected in four patients within 2 to 3 weeks after treatment. At present, data from a median duration of follow-up of 4.0 years are available for the 10 patients, and no events suggestive of leukemic transformation have been seen. Moreover, no abnormal expansion or clonal outgrowth was detected in immunologic and molecular studies (Figure 2A in the Supplementary Appendix).21 Serious adverse events included two cases of prolonged neutropenia, one of hypertension, three of central-venous-catheter–related infection, one case of Epstein–Barr virus reactivation, which resolved after preemptive therapy with one dose of anti-CD20 monoclonal antibody, and one case of autoimmune hepatitis (Table 2). Patient 8, who had recurrent autoimmune hemolytic anemia and the macrophage activation syndrome and had received corticosteroids for 2 years before gene therapy, had three episodes of autoimmune thrombocytopenia, requiring long-term corticosteroid administration, reintroduction of PEG-ADA approximately 5 months after gene therapy, and treatment with anti-CD20.
Engraftment of Vector in Multiple Cell Lineages
ADA-transduced CD34+ cells and their progeny were found in purified marrow (Figure 1) and blood (Figure 2) specimens. One year after gene therapy, the mean proportion of bone marrow cells carrying the retroviral vector was 5.1% of CD34+ cells, 3.5% of granulocytic cells (CD15+), 8.9% of megakaryocytic cells (CD61+), 3.8% of erythroid cells (glycophorin A+), and 8.0% of B cells (CD19+) (Figure 1). In peripheral-blood specimens at 1 year, the mean frequencies of transduced T cells, B cells, and natural killer cells were 88.0%, 52.4%, and 59.2%, respectively (P=0.004 for each comparison with granulocytes) (Figure 2A through 2D). In the B-cell lineage, the proportion of vector-positive B cells was significantly higher in the blood than in the bone marrow (P=0.004). The frequency of vector-positive CD34+ cells at 1 year after gene therapy correlated with the proportion of transduced colony-forming units (r=0.60) and vector copy number in CD34+ cells (r=0.75). Moreover, the patients who received a dose of more than 8x106 of CD34+ cells per kilogram (Table 1) or had neutropenia for more than 15 days (Table 1 in the Supplementary Appendix), as compared with the remaining patients, had higher mean percentages of transduced CD34+ marrow cells (6.3% vs. 0.7%) and CD15+ marrow cells (4.3% vs. 0.6%) at 1 year. ADA-transduced cells persisted in all hematopoietic lineages, including mature granulocytes, through the last evaluation (Figure 1 and Figure 2).
|
|
The presence of ADA was documented through detection of its enzymatic activity in blood mononuclear cells (Figure 2E), marrow mononuclear cells (data not shown), T cells,23 and red cells and was confirmed through flow cytometry of T and B cells, and monocytes (Figure 2B in the Supplementary Appendix). The median ADA activity in blood mononuclear cells and red cells was significantly higher at 1 year than at baseline (mononuclear cells, 497 vs. 65 nmol per hour per milligram; red cells, 0.35 vs. 0.06 µmol per hour per milliliter; P=0.004 for both tests), which corresponded to 33.6% and 1.9% of median levels in controls, respectively (Figure 2E and data not shown). The ADA activity in white cells and red cells resulted in a significant reduction of toxic levels of purine metabolites (deoxyadenosine nucleotides) in red cells at 1 year as compared with levels at diagnosis in the same patients (P=0.004) (Figure 2F). With the exception of Patient 2, deoxyadenosine nucleotide levels remained low throughout the follow-up period in all patients who were not receiving PEG-ADA.
Immune Reconstitution
After administration of the ADA vector, there was a progressive increase in T-cell counts, which reached maximum levels at 1 to 3 years (Figure 3A). In children who had been treated with PEG-ADA, the increase occurred after a transient reduction following the discontinuation of enzyme-replacement therapy (Figure 3A). As compared with values before gene therapy, the median CD3+ T-cell count was 0.51x109 per liter at 1 year of follow-up and 1.07x109 per liter at 3 years (P=0.004 and P=0.03, respectively) (Figure 3B). Median levels of CD4+ T cells and CD8+ T cells 3 years after gene therapy were 0.52x109 per liter and 0.47x109 per liter, respectively (P=0.03 for both comparisons with baseline values).
|
|
The T-cell–receptor repertoire was polyclonal, as ascertained by means of flow cytometry (Figure 2A in the Supplementary Appendix). In Patients 1 through 4, a mean (±SD) of 92±8% of the T-cell–receptor Vβ families after gene therapy were polyclonal, according to spectratyping analyses. In three patients previously given PEG-ADA, the proportion of Vβ T-cell receptors displaying a polyclonal profile increased from 18±10% to 79±9% after gene therapy.
B-cell counts increased progressively after gene therapy (Figure 3B), and as of the most recent follow-up visit, the counts were normal in four patients (Table 3). The proportion of CD27+ memory B cells was similar in patients who underwent gene therapy and age-matched controls, with a polyclonal immunoglobulin-gene rearrangement in both groups (data not shown). Serum levels of IgA and IgM reached normal values in the majority of patients (Table 3), and serum IgG levels were within the normal range in five patients after discontinuation of intravenous immune globulin supplementation. In these five patients, antibodies against toxoid, conjugated or bacterial polysaccharide antigens, or measles, and rubella were detectable after immunization with these antigens (Table 3).
Clinical Outcome
All 10 patients are alive. The nine patients who could be evaluated for efficacy (all but Patient 8) are well, with the duration of follow-up ranging from 1.8 to 8.0 years (Table 2). Patients 1 through 6 go to school regularly or, if of preschool age, have normal social relationships with other children and parents. All but two patients (Patients 1 and 9) were below the reference, the 5th percentile, in height and weight before gene therapy; at 1 year after gene therapy, their median weight increased from the 1.7th to the 13.6th percentile and the median height increased from the 3.4th to the 10.9th percentile (Figure 4 in the Supplementary Appendix).
The rate of severe infections, expressed as the number of events per 10 person-months of observation, decreased from 0.93 before gene therapy to 0.13 after gene therapy. The median number of hospitalization days dropped from 45 before gene therapy to 2 after therapy. Three of the six infections were due to catheter-related bacteremia (Table 2); no life-threatening opportunistic infections have occurred. Most patients had abnormalities in neuropsychomotor development at onset that improved during the follow-up period. Three patients had hearing deficits before gene therapy that persisted afterward.
Discussion
We found that treatment of SCID due to ADA deficiency by means of nonmyeloablative chemotherapy followed by an infusion of autologous hematopoietic stem cells that had been transduced with a retroviral vector bearing the ADA gene is not associated with adverse events during a median follow-up period of 4.0 years. The treatment supplies the patient with hematopoietic stem cells that pass a functional ADA gene to all their progeny. Of the 10 patients with SCID due to ADA deficiency who were treated in this manner, there was restoration of immune function and protection against severe infection in 9. The sustained expression of ADA in multiple hematopoietic-cell lineages allowed for the detoxification of purine metabolites and improvement in the patients' physical development.
The mortality rates in ADA-deficient patients who receive transplants from unrelated and haploidentical donors after cytoreduction are 37% and 70%, respectively.6,25 Transplantation without conditioning, involving marrow from a parent, is successful for most types of SCID,7,26 but only half of ADA-deficient patients with SCID have sustained donor engraftment. Our protocol, in contrast, affords excellent survival without serious complications, such as graft-versus-host disease.1,6,8 In addition, gene therapy is suitable for older children with SCID due to ADA deficiency, who have a higher risk of failure and complications after transplantation.6,26 Patients from whom a small number of bone marrow cells were harvested or who have preexisting chromosomal alterations in the marrow27 may not be candidates for gene therapy.
Enzyme-replacement therapy is effective in most patients with ADA deficiency2 but often fails to sustain lymphocyte counts and T-cell function.10,11,25 We found that ADA gene therapy improves immune function in patients who had insufficient immune reconstitution during PEG-ADA therapy. Taken together, these results indicate that the intracellular expression of ADA after gene transfer is superior to extracellular detoxification by PEG-ADA in permitting the maturation and survival of functional lymphocytes.
The use of nonmyeloablative conditioning18,28 and withdrawal of PEG-ADA were crucial factors in the successful outcome of our trial. Earlier gene-therapy trials for SCID due to ADA deficiency,15,16,17 which did not include conditioning regimens, were hampered by limited engraftment and immune reconstitution. Our data show that nonmyeloablative conditioning allows for the engraftment of transduced stem cells. Conditioning with busulfan was also used (at a dose of 8 mg per kilogram) in a gene-therapy protocol for treating chronic granulomatous disease,29 resulting in the engraftment of 10 to 15% of transduced granulocytes. Another gene-therapy trial for SCID due to ADA deficiency used melphalan for conditioning and achieved metabolic and T-cell reconstitution, but only 0.1% of granulocytes carried the ADA gene.22 In our trial, the number of infused CD34+ cells and the efficiency of in vitro gene transfer were also critical.
Previous gene-therapy studies with mature lymphocytes30 or hematopoietic stem cells17,18,22 indicated that enzyme-replacement therapy inhibited the outgrowth of ADA-transduced cells. Our study supports the notion that a toxic environment caused by high levels of purine metabolites at the time of stem-cell engraftment is advantageous in supporting the differential expansion of gene-corrected cells, especially in lymphoid lineages.
Gene therapy restored normal immune function in five patients and resulted in significant improvement in lymphocyte counts and functions in the other five patients, leading to protection from infectious complications. The reconstitution of lymphocyte levels was considerably slower in comparison to recovery after a bone marrow transplant including T cells from an HLA-identical donor. This difference is most likely due to the time required for the differentiation of T cells from purified, vector-containing stem cells. We believe that early intervention with gene therapy in patients with SCID due to ADA deficiency can reduce the risk of thymic involution and that optimization of the conditioning procedure will improve engraftment and immune recovery.
Gene therapy has been shown to benefit patients with X-linked SCID or chronic granulomatous disease, but the results were seriously limited by the development of leukemic proliferation (in 5 of 19 patients with X-linked SCID31,32) and clonal expansion of myeloid cells (in 2 patients with chronic granulomatous disease29). These complications were associated with retroviral-vector insertions near cellular proto-oncogenes. Our long-term follow-up and the experience in other trials of patients with SCID due to ADA deficiency17,22,33,34 did not reveal such complications. This is consistent with the polyclonal pattern of vector integration and T-cell repertoire, and the lack of in vivo skewing for potentially dangerous insertions.21 (See the Supplementary Appendix for further discussion of this complication of gene therapy in SCID.)
In conclusion, gene therapy with nonmyeloablative conditioning is an option to be considered for all patients with SCID due to ADA deficiency who lack an HLA-identical sibling donor. Our study suggests that gene therapy in combination with appropriate conditioning regimens could be successfully extended to the treatment of other congenital diseases involving the hematopoietic system.
Supported by grants from the Italian Telethon Foundation (HSR-TIGET), from Association Française contre les Myopathies–Telethon (GAT0205), the independent drug research program of the Italian Medicines Agency (AIFA) (FARM5JRXRM), and the European Commission (Concerted Safety and Efficiency Evaluation of Retroviral Transgenesis in Gene Therapy of Inherited Diseases [CONSERT] LSBH-CT-2004-005242 and Clinigene LSHB-CT2006-018933).
Dr. Bordignon reports being the chief of the board and chief executive officer (CEO) of MolMed, a drug company authorized to produce and release gene-therapy–based medicinal products for human use. MolMed manufactured the vector and engineered cells under Good Manufacturing Practices as a service to Telethon. Dr. Bordignon left the clinical study when he became CEO of MolMed in 2006. No other potential conflict of interest relevant to this article was reported.
We thank all the physicians and nurses of the Pediatric Clinical Research Unit (HSR-TIGET) and the Pediatric Immunohematology and Bone Marrow Transplantation Unit (San Raffaele Scientific Institute) for care of the patients, Alessio Palini for cell sorting, Miriam Casiraghi for coordination of patients' care and data management, Dr. Vivian Hernandez-Tujillo and William Blouin for samples and data, and Dr. Michael S. Hershfield for the anti–ADA monoclonal antibody.
Source Information
From the San Raffaele Telethon Institute for Gene Therapy (HSR-TIGET) (A.A., F.C., U.B., B.C., L.C., S. Scaramuzza, G.A., M.M., I.B., S.M., M.-G.R.), University of Milan-Bicocca (S.G., M.G.V.), Ospedale San Giuseppe (M.B.), San Raffaele Scientific Institute (F.C.), Università Vita–Salute San Raffaele (C.B., M.-G.R.), and MolMed (C.B.) — all in Milan; Tor Vergata University (A.A., P.R.) and Children's Hospital Bambino Gesù (P.R.) — both in Rome; University of Siena, Siena (A.T., F.C.); and University of Turin, Turin (R.M.) — all in Italy; Immunologische Tagesklinik, Vienna, Austria (M.E.); Hadassah University Hospital, Jerusalem, Israel (M.A., S. Slavin); King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia (H.A.-M., A.A.G.); Hôpital Universitaire des Enfants Reine Fabiola–Université Libre de Bruxelles, Brussels (A.F.); University Children's Hospital, Bern, Switzerland (A.D.); Children's Hospital, Harvard Medical School, Boston (L.N.); Universitäts-Kinderklinik München, Munich, Germany (U.W.); and Duke University Medical Center, Durham, NC (R.H.B.).
Address reprint requests to Dr. Roncarolo at HSR-TIGET, Via Olgettina 58, 20132 Milan, Italy, or at m.roncarolo{at}hsr.it.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
