The Oncologist, Vol. 13, No. 10, 1036-1042, October 2008; doi:10.1634/theoncologist.2008-0173 © 2008 AlphaMed Press
Transfer of Drug Resistance Genes into Hematopoietic Stem Cells for Marrow ProtectionThe Cancer Institute of New Jersey, Robert Wood Johnson Medical School, University of Medicine and Dentistry, New Brunswick, New Jersey, USA Correspondence: Joseph R. Bertino, M.D., The Cancer Institute of New Jersey, Robert Wood Johnson Medical School, University of Medicine and Dentistry, New Brunswick, New Jersey, USA. Telephone: 732-235-8510; Fax: 732-235-8181; e-mail: Bertinoj{at}umdnj.edu Received August 7, 2008; accepted for publication September 6, 2008. Disclosure: The content of this article has been reviewed by independent peer reviewers to ensure that it is balanced, objective, and free from commercial bias. No financial relationships relevant to the content of this article have been disclosed by the author, planners, independent peer reviewers, or staff managers.
Some Personal History and Role Models Before I describe some of the research from our laboratory, I thought it would be informative to review my background and those who have had a major impact on my career because it illustrates how important mentors and role models are for young people. I was fortunate to begin my fellowship training with Dr. Clement A. Finch, professor of medicine and head of the Hematology Division at the University of Washington, in 1958. Dr. Finch was a world authority on red cell metabolism and iron metabolism and certainly at that time and for many years was a giant in this field. Of interest was that Clem was one of the first clinical professors to realize that if he was to train fellows to do research as well as clinical work that they should have an opportunity to work with a basic scientist as part of their fellowship training. And as I did not have a great experience in injecting rats with radioactive iron, when I had the opportunity to work with Dr. Frank Huennekens, who at that time was an assistant professor of biochemistry at the University of Washington, I eagerly accepted. Dr. Huennekens was also an outstanding mentor, and with his pioneering work in folate metabolism and with my interest in translating this work into the clinic we made a good team. He has been a friend and an advisor since. During my third year of fellowship, Dr. Arnold Welch, Chairman of Pharmacology at the Yale University School of Medicine visited the University of Washington and was impressed enough with our work to offer me a position in the Department of Pharmacology with a joint appointment in the Department of Medicine. I was delighted to go back east to join this illustrious department—arguably the best pharmacology department in the country, with outstanding colleagues interested in cancer pharmacology. I had a very productive and great time at Yale from 1961 to 1987. In 1976, I took my first sabbatical and it was a great year because I worked with Bob Schimke at Stanford in the Department of Biology and with very talented graduate students, Fred Alt (now a distinguished professor at Harvard) and Randy Kaufman (now a professor of biochemistry at Michigan), and a postdoctoral fellow, Rod Kellems (now a professor of biochemistry at the University of Texas). During that time, gene amplification was discovered as a mechanism of methotrexate (MTX) resistance—a true paradigm shift in that, at that time, DNA was thought to be a stable entity [1]. In 1987, I was recruited to Memorial Sloan- Kettering Cancer Center (MSKCC) by Paul Marks to help develop a program in molecular pharmacology and therapeutics with John Mendelson. A memorable day at MSKCC was October 17, 1988, when George Hitchings, one of my heroes, was scheduled to visit us. It was announced that morning that he had received the Nobel Prize for Medicine with Dr. Gertrude Elion. He was scheduled to fly up and give a seminar that afternoon and of course I was concerned that he would not be able to make it, so I managed to get a call through to ask if he was still planning to come. He responded by saying "Joe, if I can just get my pants on! I've been having so many calls I haven't had time to get dressed, but I will be there and give the lecture," and he did arrive on time. We had extensive coverage by the media and George was able to give his lecture to a full house. In 2002, I moved to the Cancer Institute in New Jersey (CINJ) to join William Hait, one of my former fellows at Yale, who had done a remarkable job in developing the CINJ in New Brunswick at the Robert Wood Johnson Medical School, University of Medicine and Dentistry of New Jersey. At that time, in 2002, after only 9 years in existence, the CINJ received Comprehensive status from the National Cancer Institute and a new building was completed, doubling the space for the cancer center. I have been privileged to work there for the past 6 years, to help recruit new faculty, to see patients with lymphoma, and to continue my research in several areas with the collaboration of faculty members Dr. Debabrata Banerjee, Dr. Emine Abali, and Dr. Tulin Budak-Alpdogan. The projects this division works on are: gene therapy using drug resistance genes for marrow protection, regulation of dihydrofolate reductase, cord blood expansion, development of new drugs for cancer therapy, and MTX and 5-fluorouracil resistance.
The China Connection Twenty-five years later I returned to China and the change was astonishing, with an agricultural country transforming into an industrial economy. Figure 1 shows Dr. Bruce Chabner, me, and our good friend and host Dr. "Sam" Shimasaka, who arranged the lecture tour for us. Several years later I returned to China, again to lecture, with Dr. Robert Pinedo and his lovely daughter.
The studies described below represent the work of many students, postdoctoral fellows, and collaborators and especially the contributions of Tulin Budak-Alpdogan. Our goal is to protect marrow from cytotoxicity resulting from chemotherapeutic agents by transfer of drug resistance genes into marrow stem cells. For a recent more comprehensive treatment of the use of drug resistance genes for marrow protection see the review by Budak-Alpdogan et al. [2].
Hematopoietic stem cells (HSCs) are an attractive target for the gene therapy of inherited diseases and for the protection of marrow from cytotoxic drugs. HSCs can be transduced ex vivo, concentrated via CD34+ or CD133+ selection, and returned to the host following marrow ablation with high-dose chemotherapy with or without irradiation, and possibly with little or no chemotherapy or irradiation. Improvements in vector design, use of a fibronectin fragment, spinoculation, and the use of cytokines that allow stem cell replication without differentiation (Kit ligand, thrombopoietin [TPO], and FLT-3) have improved the efficiency of retroviral gene transfer into HSCs and allowed some successful outcomes. Success has been achieved in patients with certain genetic abnormalities, in particular in children with X-linked severe combined deficiency (X-SCID) (vide infra) using retroviral vectors [3, 4]. The two viral vector classes that have been most useful for gene transfer into HSCs are oncoretroviruses (gamma retrovirus), using the Maloney murine leukemia virus (MMLV) backbone, and lentiviruses such as HIV-1 and HIV-2. Viral constructs derived from HIV-1 and HIV-2 have received more attention lately because of safety concerns with the MMLV [5]. Our understanding of acquired drug resistance mechanisms in cancer cells (Table 1) and the development of efficient and relatively safe self-inactivating retroviral vectors have led to the concept that HSCs can be transduced with a viral vector containing one or more of these drug resistance genes, thus protecting HSCs from drugs that have marrow as their limiting toxicity. This may result in less toxicity to standard drug doses and, perhaps more importantly, an increase in drug dose, which could result in better cure rates in those patients that are drug sensitive. Further, by expression of more than one drug resistance gene, the therapeutic benefit would be further enhanced.
There have been several human trials using gene transfer of drug resistance genes in HSCs that have had mixed results in terms of benefit. Transfer of the multidrug resistance gene (MDR-1) that encodes P-glycoprotein results in resistance to a wide variety of structurally unrelated drugs in widespread clinical use, that is, taxanes, vinca alkaloids, anthracylines, and epipodophyllotoxins [6]. Based on encouraging studies in mouse models, four clinical studies have been reported using a retroviral construct containing MDR-1 cDNA (Table 2). The studies differed in the conditions used, for example, the vector titer, cytokines, and use of the retronectin fragment CH-296, to increase transduction efficiency [7–10]. The study by Abonour et al. [9] used the retronectin fragment CH-296 to increase transduction efficiency, and in a novel clinical trial design, two tandem transplants were administered to 12 patients with germ cell tumors following high-dose chemotherapy, the first with unmanipulated marrow and the second with cytokine-stimulated CD34+ cells (Kit ligand and interleukin-6) in the first 10 patients and Kit ligand, G-CSF, and TPO in the last two patients. Following transplant, patients were treated with etoposide daily for 21 days, orally, for three cycles. A high level of gene transfer was seen in granulocytes (up to 15% in 1 year), and after chemotherapy, evidence for in vivo selection was observed in that there was a three- to ninefold increase in the detection of the transgene in peripheral leukocytes. Importantly, in the eight surviving patients there have been no abnormalities noted in hematopoiesis.
Mutations in the O6-methylguanine-DNA methyltransferase gene (MGMT), the repair gene that protects certain tumors and normal cells from toxicity to nitrosoureas, have also been used for marrow protection (Table 3). The current strategy is to use an inhibitor to this enzyme, O6-benzylguanine (O6-BG), to inhibit the activity of MGMT in tumor cells, thus sensitizing them to nitrosoureas, and to use an O6-BG-resistant mutant MGMT (mMGMT) for transduction of HSCs, thus affording selective protection to HSCs. Based on studies in mouse models, and following a feasibility study with nonmutant MGMT [11], gene therapy trials with a retroviral construct containing mMGMT, and O6-BG administration have been initiated at Case Western Reserve University and Cincinnati Children's Hospital (Table 3).
Mutations in Dihydrofolate Reductase Associated with MTX Resistance The first mutation in the dihydrofolate reductase gene (DHFR) described was found in 3T6 cells in vitro, made resistant to stepwise, increasing doses of MTX [12]. Initially, resistance was associated with DHFR gene amplification, and with continued selective pressure with higher doses of MTX, a mutation was found in the amplified genes, associated with a high level of resistance. A point mutation was found in the active site of the enzyme that changed the codon for Leu 22 to Arg, markedly decreasing the binding of MTX to DHFR (Table 4). This mutation also resulted in a marked decrease in Kcat, and has not been useful for gene therapy studies. A second example of a mutation, this time in a human cell line from a colon carcinoma, HCT-8, also made resistant to increasing doses of MTX was also found to be associated with gene amplification. In this instance, the mutation was found to be a result of a single base change, T to C, resulting in a Ser residue instead of Phe at position 31 in DHFR cDNA [13]. A similar event occurred in a Chinese hamster lung cell line [14]. A mutation associated with amplification was also found in vivo, in a mouse tumor, L1210, treated with MTX [15]. In this case, the mutation occurred outside the active site, in the "hinge" region of the enzyme, and a C to T base transition at nucleotide 46 resulted in a Gly to Trp substitution at amino acid 15 of the enzyme. The level of resistance was modest, and enzyme activity good (Table 4).
Although the number of samples tested for mutations in DHFR from patients clinically resistant to MTX is relatively small, mutations in DHFR have not yet been described [16, 17]. Rather, low-level gene amplification and decreased transport of MTX have been the major cause of MTX resistance in patients with acute lymphocytic leukemia [18, 19] and osteosarcoma [20]. The knowledge of the crystal structure of DHFR and the knowledge of MTX and substrate binding have allowed the generation of many mutations in the active site of DHFR. In general, codon 22 mutants are generally more resistant to MTX, but are not as catalytically active as codon 31 variants [21, 22] (Table 4). Two variants in particular have properties that have made them useful for marrow protection, Leu 22 to Tyr [23, 24] and the double mutant (Phe22/Ser31) [25, 26]. Both variants, when transfected into marrow cells, cause a high level of resistance to MTX and trimetrexate, and yet have good catalytic activity (Table 4). Using random sequence mutagenesis, Encel et al. [27] generated variants of DHFR as well as thymidylate synthase with multiple mutations, also resulting in a high level of resistance to MTX and 5-fluorodeoxyuridine, respectively. We recently showed that a naturally occurring single nucleotide polymorphism, in the 3' untranslated region of DHFR, 829C to T, resulted in an increase in DHFR mRNA and DHFR [28]. Of interest, DHFR expression was found to be increased because of interference with miR-24 micro-RNA binding to this region. The increase in expression was attributed to an increase in mRNA stability, and led to a four- to fivefold increase in MTX resistance. Based on studies from our laboratory as well as from other investigators, showing that mutant forms of DHFR provide protection from MTX cytotoxicity [23–26, 29–32], we are planning two clinical trials. The first one uses a retroviral construct containing a mutant DHFR (mDHFR) fused to cytidine deaminase (CD) cDNA, which protects cells from the toxic effects of gemcitabine and cytarabine as well as MTX [30, 32]. The second trial uses a construct containing mDHFR fused to a mutant thymidylate synthase (TS) cDNA, with an internal ribosome entry site (IRES) separating this fusion gene from CD cDNA. This latter construct protects cells from the toxic effects of pemetrexed and gemcitabine [31]. Our first clinical trial will use the mDHFR/CD vector construct in patients with non-Hodgkin's lymphoma who are at high risk for recurrence, following an autologous transplant. MTX and cytarabine have synergistic antilymphoma activity, and are used to treat patients with large-cell and high-grade lymphomas [33, 34]. Escalating doses of MTX and cytarabine will be administered post-transplant to eradicate minimal residual disease and possibly improve the cure rate. The rationale for the second proposed clinical trial, using the construct containing the mDHFR cDNA fused to the mutant TS cDNA, is that pemetrexed plus gemcitabine is an effective combination for the treatment of non-small cell lung cancer, but is not curative [35, 36]. This viral construct, by protecting marrow from pemetrexed and gemcitabine toxicity, will allow treatment with higher doses of these drugs post-transplant and possibly will allow the addition of other drugs that could not be given safely because of the marrow toxicity of gemcitabine and pemetrexed.
The good news and bad news of gene therapy in 2007 relates to development of leukemia in patients with SCID treated with gene therapy. A retroviral construct containing a cDNA that encoded a T-cell receptor was used to transfer this gene to marrow stem cells and was subsequently expressed in T cells. Some of these patients are now out 6 years and have been cured [3, 4], and, more recently, two patients with chronic granulomatous disease were also treated in a similar fashion, replacing the defective gene via retroviral gene transfer to marrow stem cells, and appear to be doing well [37]. The bad news is that three patients of the first 17 SCID children, successfully treated, have developed T-cell proliferative disease, some 30–34 months after the transplant [38, 39]. What has been learned by looking at the vector integration sites using linear amplification-mediated polymerase chain reaction is that vector integration was nonrandom and clustered, and integration occurred preferentially in gene-coding regions, skewed toward transcriptional start sites [40]. While leukemia has not been reported in other patients who have had marrow stem cells transfected with retroviral constructs, these events, and the propensity of oncoretroviral constructs to integrate into active promoter regions, have stimulated the search for safer viral vectors. HIV-1 and HIV-2 vectors have the advantage that, unlike oncoretroviral vectors, they can infect cells that are not in cycle [41]. It also may be that they are safer, in that integration may not be targeted preferentially to active sites of transcription [42]. Other viruses, in particular the foamy leukemia virus, are also being investigated to determine their safety.
What are the obstacles to the successful use of gene transfer for the purpose of marrow protection? These clinical studies require a large team effort involving pathologists, hematologists, molecular biologists, and supporting staff. Important regulatory hurdles and expenses associated with this type of project need to be dealt with. Gene therapy requires cellular products to be generated under a highly regulated environment, in a facility that applies current Good Manufacturing Practice (GMP) guidelines. We are currently validating the procedures in the newly established Production Facility for Gene and Cellular Therapy (GMP facility) located in the CINJ. We have generated a clinical-grade producer cell line and have optimized vector production and human CD34+ cell transduction conditions.
I am indebted to my many colleagues who have contributed to this effort, in particular, Tulin Budak-Alpdogan, Debu Banerjee, and Emine Abali. I thank Bob Pinedo for his many years of friendship and collegiality and his enormous contributions toward the eradication of cancer. This Pinedo Prize Lecture was delivered at the Medical Knowledge Institute, Amsterdam, The Netherlands, October 18, 2007 (Fig. 2).
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||