|
| |||||||||||||||||||||||||||||||||||||||||||||
Background We tested the safety of a nonviral somatic-cell gene-therapy system in patients with severe hemophilia A.
Methods An open-label, phase 1 trial was conducted in six patients with severe hemophilia A. Dermal fibroblasts obtained from each patient by skin biopsy were grown in culture and transfected with a plasmid containing sequences of the gene that encodes factor VIII. Cells that produced factor VIII were selected, cloned, and propagated in vitro. The cloned cells were then harvested and administered to the patients by laparoscopic injection into the omentum. The patients were followed for 12 months after the implantation of the genetically altered cells. An interim analysis was performed.
Results There were no serious adverse events related to the use of factor VIIIproducing fibroblasts or the implantation procedure. No long-term complications developed, and no inhibitors of factor VIII were detected. In four of the six patients, plasma levels of factor VIII activity rose above the levels observed before the procedure. The increase in factor VIII activity coincided with a decrease in bleeding, a reduction in the use of exogenous factor VIII, or both. In the patient with the highest level of factor VIII activity, the clinical changes lasted approximately 10 months.
Conclusions Implantation of genetically altered fibroblasts that produce factor VIII is safe and well tolerated. This form of gene therapy is feasible in patients with severe hemophilia A.
Replacement with intravenously administered concentrates of factor VIII controls bleeding in patients with hemophilia A. In patients with severe disease, spontaneous hemorrhage and progression of arthropathy can be prevented by prophylactic administration of exogenous factor VIII to maintain factor VIII activity levels above 1.0 percent of normal.3 Prophylaxis with factor VIII is costly,1 however, and does not guarantee the prevention of hemorrhage.4 Despite considerable improvements in the safety of plasma-derived and recombinant factor VIII concentrates, concern remains about contamination with infectious prions5,6 and small nonenveloped viruses, such as hepatitis A virus7,8 and parvovirus.9,10,11 Hemophilia A is a suitable candidate disease for gene therapy12 for several reasons: factor VIII production is not regulated in response to bleeding; the broad therapeutic index of factor VIII minimizes the risk of overdoses; delivery of factor VIII into the bloodstream does not require expression of the gene by a specific organ; and even low levels of the protein can be beneficial.
We have developed a nonviral gene-delivery system termed "transkaryotic implantation." It entails the isolation of somatic cells from a patient, the stable introduction of a therapeutic gene into these cells, the isolation and clonal propagation of a single engineered cell, and the implantation of the clonal cells into the patient.13,14,15 Mice that received fibroblast implants carrying DNA sequences of the human factor VIII gene produced the human protein; the procedure was safe and resulted in factor VIII activity levels that exceeded 5.0 percent of normal a level that would be considered therapeutic in patients with severe hemophilia A for more than one year (unpublished data). These results led us to study the safety of the transkaryotic-implantation system in patients with severe hemophilia A.
Methods
Patients
The study was designed as a single-institution, open-label, phase 1 trial in which autologous fibroblasts that produced human factor VIII were administered to patients with severe hemophilia A. Patients were eligible to be included in the study if they had hemophilia A with a level of factor VIII activity below 2.0 percent of normal; were at least 15 years old; had received at least 50 days of factor VIII therapy before study entry; had six or more bleeding episodes per year; had normal factor VIII clearance; had a hemoglobin level above 12 g per deciliter (7.4 mmol per liter); and had a platelet count above 100,000 per cubic millimeter. Patients with any of the following characteristics were excluded: the presence of a factor VIII inhibitor, as determined by the Bethesda inhibitor assay (which detects antibodies that neutralize factor VIII activity) at the time of enrollment; a history of inducible factor VIII inhibitor; an anticipated requirement for fixed-dose prophylaxis with factor VIII infusions during the study; a history of opportunistic infection or cancer related to the acquired immunodeficiency syndrome; use of investigational therapy for hemophilia within 30 days before enrollment in the study; or clinical evidence of abdominal adhesions or portal hypertension (which would increase the risk associated with the laparoscopic procedure).
The clinical protocol and the informed-consent document were approved by the Center for Biologics Evaluation and Research of the Food and Drug Administration and by the Office of Biotechnology Activities of the National Institutes of Health. Local review and approval were obtained from the Beth Israel Deaconess Medical Center institutional review board, the Harvard University Biosafety Committee, and the Harvard University Human Gene Therapy Advisory Committee. Permission to treat up to 12 patients and to monitor them for up to two years was given. All the patients provided written informed consent.
Figure 1 shows the steps involved in the study. First, the patients underwent a clinical evaluation in which a complete medical history was obtained and a physical examination, blood and urine tests, electrocardiography, and chest radiography were performed. Patients received diary forms for use at home to record and describe all bleeding episodes and all infusions of factor VIII during the study. The date, time, and site of bleeding were recorded for each episode, as were the number of units of factor VIII administered with each infusion. The reason for factor VIII infusion was assigned to one of the following categories: spontaneous bleeding, injury-related bleeding, prophylaxis, and other reasons (including use for the skin biopsy, the laparoscopy, and other procedures).
|
A full-thickness, elliptical skin-biopsy specimen (1.6 by 0.4 cm) was obtained from the medial surface of the upper arm after the patient received local anesthesia. Dermal fibroblasts were isolated for cell culture and subsequent plasmid transfection. All the patients received an infusion of factor VIII before the biopsy to increase their plasma factor VIII activity level to 100 percent of normal. Factor VIII was administered for an additional three or four days to maintain normal hemostasis.
Laparoscopy
Approximately seven weeks after the skin biopsy, the patients returned to the hospital for another clinical assessment, which was performed the day before cell implantation. Factor VIII was infused just before surgery to increase plasma factor VIII activity levels to approximately 100 percent of normal, and the infusions were continued two or three times daily for one week after surgery. The laparoscopic procedure was performed while the patients were under general anesthesia. A 10-mm trocar was inserted into the peritoneal cavity through an infraumbilical incision with the use of an open technique. After carbon dioxide insufflation, a laparoscope was introduced at this site. Under direct vision, a single, 5-mm operating trocar was inserted into the midabdomen, and through this trocar, a laparoscopic grasper was used to manipulate the greater omentum. A 9-cm (3.5-in.), 20-gauge spinal needle was inserted directly through the abdominal wall under direct vision, and approximately 0.5 ml of the cell suspension was injected through the needle into the greater omentum at each of several sites. After surgery, the patients were observed overnight in the clinical research center. All six patients were discharged home on the first postoperative day.
Clinical assessments at the patients' homes were performed daily by the study nurse on days 2 through 6 after surgery. The patients were scheduled to undergo clinical assessments at the hospital during weeks 1, 2, 3, 4, 6, 8, 12, and 18 and months 6, 9, and 12 after cell implantation, with additional monitoring through year 2.
Preparation of Autologous Fibroblasts Expressing Factor VIII
A plasmid that contained the gene encoding human factor VIII from which the B domain had been deleted and that contained the human fibronectin promoter was introduced by electroporation into dermal fibroblasts that had been isolated from the skin-biopsy specimen. The B-domain coding sequence was not incorporated into the plasmid, because it is not required for the coagulant activity of factor VIII or the interaction of factor VIII with von Willebrand factor16,17; moreover, the presence of the B domain reduces expression of factor VIII in transfected mammalian cells. Fibronectin, an extracellular-matrix protein, is expressed in fibroblasts; its promoter efficiently directs the transcription of B-domaindeleted factor VIII in primary human dermal fibroblasts.
Stably transfected fibroblast clones that had incorporated the plasmid containing B-domaindeleted factor VIII were selected in G418-containing medium, isolated, expanded in nonselective medium, and characterized. Characterization included measurement of factor VIII expression, assessment of cell-growth properties (including soft-agar tumorigenicity assay in vitro), microbial safety, and Southern blotting. Cells from the clone that was designated for implantation were harvested the day before implantation, extensively washed, and introduced into a syringe. Either 100 million or 400 million cloned cells were then administered. During preparation, the fibroblast cultures were aseptically processed in class 100 conditions.18
Factor VIII Assays
Standard one-stage factor VIII clotting-activity assays and Bethesda inhibitor assays based on measurement of the activated partial-thromboplastin time were performed (Automated APTT reagent, Organon Teknika) on an MDA instrument with the use of a standard curve prepared with a normal human pooled plasma calibration standard (Precision Biologic), which was calibrated against a World Health Organization standard. Abnormal controls (Dade and Behring) and normal controls (Precision Biologic), both derived from human plasma, were used to validate the assays. Serum samples from the patients were analyzed for antibodies against factor VIII with two specific enzyme-linked immunosorbent assays. Full-length human factor VIII derived from Chinese-hamster-ovary cells (Recombinate, HylandBaxter) or human plasma-derived factor VIII (Hemofil-M, HylandBaxter) was used to coat the wells of microtiter plates before the addition of patients' serum. The enzyme-linked immunosorbent assay was developed with a horseradish peroxidaseconjugated goat antihuman antibody. The assay was validated with the use of normal human plasma and abnormal human plasma containing various amounts of factor VIII antibodies (George King Bio-Medical). All specialized blood-coagulation testing was performed at Esoterix Coagulation Laboratories with use of frozen plasma or serum samples that were collected from the patients at all scheduled visits; however, at month 9 only factor VIII activity was tested, and only in Patients 5 and 6.
Results
Characteristics of the Patients and Factor VIII Production by the Implanted Fibroblasts
Table 1 shows the characteristics of the six patients. As shown in Table 2, Patients 1, 2, and 3 each received 100 million cells, and Patients 4, 5, and 6 each received 400 million cells. The total factor VIII production of each of the clones that was implanted is also shown.
|
|
The skin biopsy was well tolerated by all six patients, and there were no episodes of bleeding or infectious complications. Laparoscopic implantation of the fibroblasts was also well tolerated. The mean (±SD) duration of surgery was 65±7 minutes. The mean time required for the injection of the fibroblasts was 15±3 minutes. There were no complications related to anesthesia. After the procedure, all six patients had minor abdominal discomfort at the incision site. Other adverse events during the postoperative period were pain referred to the shoulder in five patients, abdominal ecchymoses in two, and low-grade postoperative fever (which did not require treatment) in one. There were no long-term complications associated with surgery.
All the Bethesda inhibitor assays and enzyme-linked immunosorbent assays for antibodies against factor VIII were negative. Preliminary assays for cytotoxic T lymphocytes at selected times through month 12 in Patients 1, 2, and 3 and through month 6 in Patients 4, 5, and 6, in which the patients' lymphocytes were used as the effector cells and transfected autologous fibroblasts expressing B domaindeleted factor VIII were used as the target, were negative; this finding indicates that there was no cellular immune response to the cultured transfected fibroblasts.
One patient with a long-standing history of recurrent paroxysmal atrial tachycardia had two episodes of tachycardia, 7.5 months and 12 months after the implantation procedure. These two serious adverse events were classified as unrelated to the use of the genetically modified fibroblasts or the implantation. There were no other serious adverse events. All six patients had adverse events, most of which consisted of bleeding episodes or pain associated with bleeding. No clinically significant laboratory abnormalities related to the gene therapy were detected.
Clinical Effects
In addition to evaluating the safety of the use of the genetically engineered fibroblasts and of the surgical procedures (skin biopsy and laparoscopy), we appraised several subjective and objective measures of efficacy: bleeding events related to hemophilia, use of exogenous factor VIII for treatment before and during the study, and factor VIII activity levels. Figure 2 shows bleeding events and the use of exogenous factor VIII in Patients 1, 3, and 6, all of whom maintained diaries of bleeding and factor VIII use before participation in the study. Patients 2, 4, and 5 did not keep such diaries before entering the study.
|
|
Table 4 shows the average monthly use of factor VIII before and after cell implantation in all six patients. Treatment with infusions of exogenous factor VIII before enrollment in the study was documented by the patients' diaries (Patients 1, 3, and 6) or by pharmacy records (Patients 2, 4, and 5). Use of exogenous factor VIII by Patients 1 and 2 was unchanged after treatment. Use by Patients 3, 4, and 5 was moderately decreased. In Patient 6 there was a considerable reduction in the use of exogenous factor VIII after gene therapy; in the 20 months before therapy, his total use of factor VIII each month ranged from 12,339 to 30,621 U, whereas his use ranged from 2040 to 7098 U per month at months 4, 6, 7, 8, and 10 after gene therapy (Table 4).
|
Discussion
We studied six patients with severe hemophilia A who received autologous fibroblasts that carried the gene that encodes factor VIII. The implantation procedure was safe, and the cells have been well tolerated, a finding that has continued in all patients during a second year of monitoring. In no patient did evidence of factor VIII inhibitor appear at any time during the study. This is important, because inhibitors have the potential to compromise the effectiveness of standard factor VIIIreplacement therapy.19
The potential for clinical efficacy of this type of gene therapy is suggested by the increased levels of factor VIII activity in four of the six patients. Before they entered the study, the factor VIII activity levels in all six patients were less than 0.8 percent of normal a fact that made it possible to detect relatively small increases in factor VIII in plasma after the implantation of the genetically engineered fibroblasts. There was a general correlation between these increases and clinical improvement, such as a decreased frequency of spontaneous-bleeding episodes or decreased use of exogenous factor VIII.
At the relatively low levels of factor VIII activity we observed, several factors may have influenced subjective measures of the clinical efficacy of the gene-therapy procedure we tested. One of these is the presence or absence of active joint disease: treatment of severe joint disease requires persistent, relatively high levels of factor VIII activity. The patients' investment in the therapeutic outcome of the experimental treatment may also have influenced subjective measures of efficacy that suggested therapeutic benefit. However, the correlation between objective and subjective signs of improvement indicates that a prolonged placebo effect was unlikely.
The minimal levels of factor VIII that are required to protect against spontaneous or post-traumatic bleeding are unknown. Protection against bleeding related to trauma or surgery may require high factor VIII levels, in the range of 30 to 100 percent of normal, depending on the nature of the injury.20 In contrast, the clinical features that distinguish between severe and moderate hemophilia suggest that factor VIII activity levels as low as 1.0 to 2.0 percent of normal can protect against spontaneous bleeding. A recent study of prophylactic treatment in patients with severe hemophilia indicated that even levels below 1.0 percent of normal may be sufficient.21 For these reasons, the clinical responses we observed in our patients, who had low plasma levels of factor VIII, are encouraging.
The nonviral ex vivo gene-therapy system we used has several potential advantages over gene therapy based on viral vectors. Patients received a homogeneous clonal population of cells containing a single genetic modification, thereby minimizing the risk of insertional mutagenesis that might occur with viral vectors. In addition, viral vectors can evoke immune responses that may attenuate the effectiveness of subsequent treatments.22 Viral vectors may become infectious by mutation or other mechanisms or may modify the patient's germ line, but these risks are obviated with the use of the nonviral approach. Despite these advantages, our system has noteworthy disadvantages. The implantation procedure is moderately invasive, and the factor VIIIproducing autologous fibroblasts must be prepared individually for each patient. Fortunately, there is no technological obstacle to the production of such patient-specific fibroblasts, so the procedure could be feasible if the therapeutic benefit of this approach can be proved in subsequent studies.
Supported by Transkaryotic Therapies.
Drs. Treco and Selden are employees of and own stock in Transkaryotic Therapies.
We are indebted to the patients with hemophilia and their families for their commitment to this investigation and their continuous support and trust; to members of the Departments of Medicine, Surgery, Anesthesia, and Nursing and of the General Clinical Research Center at the Beth Israel Deaconess Medical Center for their contributions; to members of the basic-science and clinical-research teams at Transkaryotic Therapies for their help; and to Dr. Paula Fraenkel of Beth Israel Deaconess Medical Center for her assistance with patient care.
Source Information
From the Center for Hemostasis and Thrombosis Research, Department of Medicine (D.A.R., J.M.O.), and the Department of Surgery (N.E.T.), Beth Israel Deaconess Medical Center and Harvard Medical School, Boston; and Transkaryotic Therapies, Cambridge, Mass. (D.A.T., R.FS.).
Address reprint requests to Dr. Roth at Beth Israel Deaconess Medical Center, Center for Hemostasis and Thrombosis Research, RE-302, 41 Ave. Louis Pasteur, Boston, MA 02115, or at droth{at}caregroup.harvard.edu.
References
kiewicz M, Russell PS, Goodman HM. Regulation of insulin-gene expression: implications for gene therapy. N Engl J Med 1987;317:1067-1076. [Web of Science][Medline]
Other members of the Factor VIII Transkaryotic Therapy Study Group were as follows (asterisks indicate a financial interest in Transkaryotic Therapies): J.D. Levine, J. Proper, and B. Furie* (Beth Israel Deaconess Medical Center and Harvard Medical School, Boston) and V.A. Roman,* Z.M. Sabine,* C.W. Phillips,* A.M. Zuliani,* N.A. Savioli,* D. Fisher,* J. Harrington,* M. Borowski,* M.W. Heartlein,* J.C. Lamsa,* E.M. Morrel,* and K.C. Gunter* (Transkaryotic Therapies, Cambridge, Mass.).
| |||||||||||||||||||||||||||||||||||||||||||||
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. |