---

Background:  Due to limitations of local and systemic therapies for prostate cancer, interest has continued in the development of new treatment modalities.  Gene therapy has emerged as a new approach that may prevent or treat disease by using the therapeutic information encoded in DNA sequences.  Several institutions are actively experimenting with this approach.
Methods:  The authors review the most common genetic alterations in prostate cancer, the principles of gene therapy, and gene delivery including both viral and nonviral vectors.  Treatment strategies for both cytoreductive gene therapy as well as corrective gene therapy are described, and the available protocols to date with gene therapy for the treatment of prostate cancer are presented.
Results and Conclusions:  More than 150 active protocols are ongoing to evaluate gene therapy in the treatment of cancer, with 13 of these open for patients with prostate cancer.  The future of gene therapy as applicable to prostate cancer depends on additional development of vector systems and a better understanding of the genes involved in tumor induction and proliferation.  Although gene therapy is clearly in its infancy, it is in an explosive growth phase and holds tremendous promise as a treatment modality for prostate cancer.

---

Introduction to Gene Therapy

    Although the incidence rate and mortality rate of prostate cancer have decreased slightly in the past few years, prostate cancer remains a major health threat to American men. This disease remains the most commonly diagnosed internal malignancy in men and the second leading cause of cancer death among men in the United States.¹ In 1998, an estimated 185,500 men will be diagnosed with prostate cancer, and 39,200 will die of the disease.² Biologically, prostate cancer represents a heterogeneous disease entity that exhibits varying degrees of aggressiveness, patterns of metastasis, and response to therapy.

    Universal agreement has not been reached as to the best treatment for prostate cancer at any stage. Radical prostatectomy, external-beam radiation therapy, brachytherapy, and cryotherapy can affect local tumor control and are potentially curative in patients with clinically localized disease. In spite of the widespread use of prostate-specific antigen (PSA) in early detection and screening, many cases are not diagnosed until the disease has advanced or metastasized beyond the reach of these local treatment modalities. Hormonal therapy and chemotherapy are the only systemic treatments available at the present time. Unfortunately, progressive disease develops in many patients who undergo these treatments, thus proving them to be noncurative.

    Because of the significant limitations of currently used local and systemic therapies for prostate cancer, interest has continued in the development of new treatment modalities. Gene therapy has emerged as an exciting new treatment that may affect both local and systemic control of prostate cancer.

    The term gene therapy broadly refers to the transfer of genetic material into human cells and the expression of that material in these cells for a therapeutic purpose.³ With respect to cancer, the goal of gene therapy is to prevent or treat disease by using the therapeutic information encoded in the DNA sequences.

    Evidence suggests that tumor formation is caused by the overexpression of oncogenes or by mutations in suppressor genes in the presence or absence of cancer-causing environmental events. In the first human experiment of gene therapy, Blaese and coworkers⁴ successfully transferred the gene for adenosine deaminase into the cells of a 4-year-old girl with severe combined immunodeficiency caused by adenosine deaminase deficiency. The gene therapy dramatically improved her immune system function. Presently, more than 150 active protocols are evaluating gene therapy in the treatment of cancer, with at least 13 of these protocols open for patients with prostate cancer⁵ (Table 1).⁶

Table 1. -- Current Prostate Cancer Gene Therapy Protocols in the United States
Institution	Principal Investigator(s)	Gene	Vector	Treatment Approach	Delivery	Target Population
Johns Hopkins	Simons	GM-CSF	Retrovirus	Immunotherapy injection	Subcutaneous	Metastatic PC
Vanderbilt University Medical Center	Steiner	Antisense c-myc	Retrovirus	Oncogene inhibition	Intraprostatic injection	Advanced PC
National Naval Medical Center	Chen	PSA gene virus	Vaccinia	Immunotherapy	Intradermal injection	--
Duke University Center	Paulson	IL-2	Cationic liposome complex	Immunotherapy	Intradermal injection	Locally advanced Medical or metastatic PC
Baylor College of Medicine	Scardino	HSV-tk	Adenovirus	Suicide gene (prodrug)	Intraprostatic injection	Local recurrence of PC
Dana-Farber Cancer Institute, Boston	Kufe	PSA gene	Vaccinia virus	Immunotherapy	Intradermal injection	--
University of Michigan	Sanda	PSA gene	Vaccinia virus	Immunotherapy	Intradermal injection	Serologic recurrence of PC following radical prostatectomy
University of California	Belldegrun	IL-2	Cationic liposome complex	Immunotherapy	Intratumoral injection	--
Mount Sinai School of Medicine	Hall	HSV-tk	Adenovirus	Suicide gene (prodrug)	Intratumoral injection	Neoadjuvant treatment for patients with clinically localized PC
UCLA School Medicine	Belldegrun	p53	Adenovirus	Suppressor gene	Intratumoral injection	Locally advanced of or recurrent adenocarcinoma
Johns Hopkins	Simons	GM-CSF	Retrovirus	Immunotherapy	Subcutaneous injection	--
MD Anderson Cancer Center and University of Tennessee	Logothetis, Steiner	p53	Adenovirus	Tumor suppressor gene	Intraprostatic injection	Locally advanced PC
Baylor	Kadmon	HSV-tk	Adenovirus	Suicide gene (prodrug)	Intratumoral injection	Neoadjuvant prior to prostatectomy
Johns Hopkins*	Simons	PSA gene	Adenovirus	Vector-directed cell lysis	Intratumoral injection	Recurrent locally advanced PC
Vanderbilt University and University of Tennessee	Smith, Steiner	BRCA1	Retrovirus	Oncogene	Intraprostatic inhibition	Advanced PC injection
*Protocol under review, not open yet.
GM-CSF = granulocyte macrophage-colony stimulating factor
PSA = prostate-specific antigen
IL-2 = interleukin-2
HSV-tk = herpes simplex virus-thymidine kinase
PC = prostate cancer

Genetic Alterations in Prostate Cancer

    Tumorigenesis is a multistep process that involves initiation, proliferation, loss of contact inhibition, invasion, and metastasis of the cancer cell. It is likely that the pathway leading to malignancy undoubtedly involves many complex genetic and epigenetic influences, such as cell-cycle regulation, angiogenesis, immunoreactivity, and cell adhesion. In fact, no single gene defect has been consistently implicated in the development of any type of cancer; rather, this process has been shown to involve many different mutations.⁷

    A number of genetic changes have been documented in prostate cancer, including allelic loss, point mutations, and changes in DNA methylation pattern. The most consistent changes are allelic loss events, with the majority of tumors examined showing loss of alleles from at least one chromosomal arm. The short arm of chromosome 8 and the long arm of chromosome 16 appear to be the most frequent regions of loss, suggesting the presence of novel tumor suppressor genes.⁸ A summary of the genetic alterations occurring in prostate cancer is presented in Table 2.

Table 2. -- Common Genetic Alterations in Prostate Cancer
Gene	Effect on Tumorigenesis	Normal Gene Function	Comments
ras	Oncogene	Regulates DNA replication through cell cycle	Mutated in only 2%-5% of prostate cancers in US men
bcl-2	Oncogene	Interfaces with the p53 pathway and inhibits apoptosis	Elevated in most androgen-independent and advanced prostate cancers
c-myc	Oncogene	DNA regulatory protein involved in DNA repair and cell proliferation	May be involved in prostate cancer progression
GST-P1	Suppressor gene	Detoxifies potential carcinogens	Most common genomic alteration in prostate cancer; found in over 98% of cases
p53	Suppressor gene	G1 checkpoint in response to DNA damage; also involved in apoptosis	Mutation associated with androgen-independent progression
p21	Suppressor gene	Cell-cycle control and apoptosis pathway	Suppresses prostate cancer cell lines in vitro
p16	Suppressor gene	Binds to and inhibits cdk4, cell-cycle regulation	Inhibits growth of prostate cancer cell lines
Retinoblastoma	Suppressor gene	Cell-cycle regulation	Allelic loss in ~27% of prostate cancers
E-cadherin	Suppressor gene	Cell adhesion molecule	Reduced or absent in about half of prostate cancers; normal expression is a marker of locally confined disease
C-CAM1	Suppressor gene	Cell adhesion molecule	Downregulated in some prostate cancers; inhibits prostate cancer growth in animals
CAM1 = cell adhesion molecules 1
GST = glutathione-s-transferase gene P1

    Alterations in gene function and expression also are involved in the emergence of androgen-independent disease. Androgen deprivation causes the loss of prostate cancer cells by means of the cell-death process referred to as apoptosis. Some 70% to 80% of patients experience at least partial remission after hormonal therapy, but treatment failure and tumor recurrence are almost inevitable.⁹ Several months or years after treatment, prostate cancer eventually progresses to hormone-refractory status. Several potential mechanisms for the development of androgen-independent disease have been described, including activation of epidermal, fibroblast, or other growth-factor pathways; alteration of genes such as ras; mxi1 (a negative regulator of the c-myc proto-oncogene); mutations; overexpression of the bcl-2 oncoprotein (suppressor of apoptosis); amplification of c-myc; loss or mutation of tumor suppressor genes Rb and p53; and loss of metastasis-suppressor genes such as E-cad (cadherin) and kai1 in prostate cancer cell lines.^10-13 After failing hormonal therapy, most prostate cancers remain incurable with conventional chemotherapy.

Principles of Gene Therapy

    When considering gene therapy as a treatment approach for prostate cancer, the treatment team must decide what genes to insert, how to deliver the genes, and how to express the therapeutic genes at the site of cancer. Two major categories of gene therapy are cytoreductive gene therapy and corrective gene therapy. Cytoreductive gene therapy includes treatment strategies designed to selectively destroy malignant cells either directly (eg, toxic genes) or indirectly (eg, genes that stimulate immune responses). Corrective gene therapy involves replacing or inactivating defective genes in preneoplastic or neoplastic cells with genes that can slow or reverse the loss of growth-control mechanisms (eg, tumor suppressor genes). These therapeutic genetic modifications can be performed either ex vivo or in vivo, depending on the strategy.

Gene Delivery

    One of the rate-limiting factors affecting gene therapy is the development of a safe, reliable vector that can insert the desired gene into the target. Vectors are engineered DNA or RNA sequences into which a therapeutic gene can be inserted. The therapeutic gene generally is positioned adjacent to a promoter sequence for RNA polymerase. From this position, messenger RNA (mRNA) of the therapeutic gene is expressed within the cell. Promoter sequences regulate gene expression downstream and thus can serve as critical pharmacologic targets to modulate gene function. Most viral promoter sequences are, in fact, the endogenous long-terminal-repeat sequences. The ability of a vector to successfully deliver the gene of interest into a large number of target cells is defined as gene transfer efficiency.¹⁴ For instance, if 1 in 3 treated cells takes up the vector successfully, then the gene transfer efficiency is 33%. The overall efficacy of the vector is also determined by the degree of gene transduction achieved. Delivery methods are categorized as viral (eg, retrovirus, adenovirus, and adeno-associated virus [AAV]) and nonviral (plasmid DNA, liposome DNA). The advantages and disadvantages of specific types of vectors are summarized in Table 3.

Table 3. -- Vectors for Gene Therapy
Vector	Advantage	Disadvantage
Viral retrovirus	Easy to produce, efficient transfer, small genome, biology well understood, nontoxic to host cells, high-efficiency genomic integration, stable expression	Targets only dividing cells, risk of replication, carries small DNA sequences only, low transduction efficiency, integration with potential oncogenesis, poor in vivo delivery
Adenovirus	Highly efficient transfer, targets nondividing cells, nontoxic to host cell, high transduction efficiency, immunogenicity	Possible host immune reaction risk of replication, carries small DNA sequences only, low potential oncogenesis, no integration, transient expression
Adeno-associated virus	Less likely to produce immune reactions, targets nondividing cells, nonpathogenic in humans, efficient transfer, good in vivo delivery, integrates into genome	Small capacity, immunogenic, not well studied, risk of replication
Vaccinia virus	High titer, large insert size	Antivector immunity, toxicity
Herpes simplex virus	Large insert size	Toxicity
Nonviral:
Plasmid DNA	No size limitation	Low efficiency
Liposome	Easy to produce, safety features, less likely to produce immune response, no limitation on size and type of nucleic acid	Low efficiency

Nonviral Vectors

    Nonviral vectors offer several advantages over viral vectors with respect to safety and ease of production. Rigorous tests are not required to validate the absence of replication-competent viruses, which saves both time and money. Another advantage is that nonviral vectors can deliver larger pieces of DNA than viral vectors.¹⁵ However, there are some limitations to traditional methods of plasmid transfection. Several limitations are specific to the use of plasmid transfection in prostate cancer.

    Researchers have encountered some difficulties when attempting to passage prostate cancer cells in vitro, and these difficulties make the practical application of plasmid transfection virtually impossible to reproduce.¹⁶ Furthermore, when plasmid transfection is carried out, overall transfection efficiency is very low. With systemic administration, naked DNA plasmids are rapidly cleared from the bloodstream, and DNA degradation occurs within 5 minutes.⁷ The stability and transformation efficiency of naked DNA is enhanced by coating plasmids with lipids (liposome-DNA). In this form of nonviral vector delivery, a liposome shell surrounds the plasmid, protecting the DNA from degradation after systemic administration to the host. In addition, the lipid envelope can fuse with tumor-cell membranes, resulting in direct delivery of the therapeutic gene to the cytoplasm of the cell. Plasmid transfection also is not likely to produce immune response and decrease in efficiency because of the presence of blocking antibodies. Because the majority of liposome gene complexes are rapidly cleared from the circulating bloodstream by the liver, systemic therapy cannot be efficiently administered.¹⁷ It is possible that intratumoral injection of liposomes may be able to bypass this intrahepatic clearance of the gene and therefore improve transfection efficiency.

    Liposome vectors show significant promise; however, more studies are needed to determine their safety and efficacy in vivo before widespread use.

Viral Vectors

    Retroviruses and AAVs have an advantage in that they facilitate stable integration and a steady level of expression once the therapeutic gene is transfected and incorporated into the host genome. As the target DNA is replicated, so too is the inserted therapeutic gene embedded in the transferred chromosomal DNA. Thus, transduction via these vectors can produce durable gene expression. In corrective gene therapy, the durability of the replication is essential for maintaining the corrected cell phenotype over the patient’s lifetime.¹⁸ Furthermore, this can be advantageous in tumor vaccine strategies in which a steady level of gene expression may enhance efficacy. In contrast, adenovirus, vaccinia, and liposomal vector transfer are episomal methods: the transferred gene is expressed without actual integration of the gene into the target cell genome. These vectors are not well suited to clinical strategies requiring durable expression of transferred therapeutic genes.

    Retroviral gene expression and dissemination depend on the presence of a rapidly proliferating tumor-cell population. In some malignancies, such as prostate cancer, in which the tumor is inherently slow growing, the retroviral approaches may not be effective.¹⁹ The possibility of pathogenic mutagenesis during chromosomal insertion of the vector and difficulty in isolating high titers of retrovirus for clinical use further limit retroviral transfer.²⁰

    In contrast to retroviruses, adenoviral vectors can facilitate highly efficient transfection of therapeutic genes into cell culture. The adenovirus has several characteristics that make it ideal for use in prostate cancer gene therapy. The adenovirus enters target cells by receptor-mediated endocytosis after binding to integrins or fibronectins, providing significant tropism to cells of epithelial origin.⁹ Unlike the retrovirus, the adenovirus does not depend on cell replication for transfer of expression of its genetic material.²¹ The safety of adenoviral gene therapy vectors already has been shown in human trials involving
cystic fibrosis, ornithine transcarbamylase deficiency and factor IX deficiency.^21,22 Because the adenovirus can be efficiently delivered into replicating or nonreplicating cells of epithelial origin, it may be ideal for in vivo systemic therapy, provided an effective avenue of delivery to sites of metastatic cancer can be devised.

    Genomic integration rates are low; consequently, there is little risk of long-term sequelae with the administration of adenovirus.²¹ Short-term expression has obvious limitations for chronic gene replacement, but it may be considered advantageous for the treatment of neoplasias. For example, transient overexpression of p53 in prostate cancer may be sufficient to activate apoptosis in neoplastic cells without the concerns associated with integration of genetic material into normal cells. Induction of antiviral antibodies and high levels of hepatic deposition of circulating virus would make these vectors inefficient for readministration and systemic administration, although it has been reported that the coadministration of adenovirus and immunosuppressive drugs such as cyclosporin A can drastically increase the duration of transgene expression in hepatocytes and hence suppress cellular immunity against adenovirus.^9,16

    Some AAV vectors appear to allow durable genetic transduction and are stable for potential direct in vivo gene transfer.²⁰ However, integration of AAV genomes into target cell DNA appears to be less efficient and has been associated with tendency toward inactivations via deletions or rearrangement during gene transfer.

Cytoreductive Gene Therapy

    According to the immune surveillance theory of Burnet, the immune system is responsible for eliminating newly transformed cells; therefore, the emergence of a tumor signals the failure of the immune system.⁵ One gene therapy approach involves the activation of antitumor immune responses via T-cell killing of cancer cells.

    The most thoroughly evaluated form of cytoreductive gene therapy for cancer so far involves stimulation of an antitumor immune response against a malignancy by vaccinating affected patients with genetically modified tumor cells.¹⁸ In the ex vivo vaccine approach, tumor cells are removed at surgery from the patient, grown in cell culture, and transfected with cytokine genes that stimulate an immune response to antigens present on the tumor cell vaccine. The gene-modified tumor vaccine is then irradiated to prevent subsequent tumor growth and reinjected into the patient in an attempt to generate either a local or systemic immune response against the remaining tumor burden in the patient. These immune effector cells activated at the vaccination site may include T cells, B cells, natural killer cells, and antigen-presenting cells such as dendritic cells and macrophages.²³ In theory, the B-cell or T-cell arm activated by vaccination can then circulate systemically and eradicate or slow the growth of distant micrometastatic cells that share antigens with the genetically engineered vaccine cell.

    Several therapeutic cytokine genes have been studied for use in gene therapy. Sanda et al²⁴ and Blades et al²⁵ showed a defect in cell-surface expression of class I major histocompatibility complex (MHC) in prostate cancer. Cytokines that are dependent on MHC class I processing for immunostimulatory effects, eg, interferon (IFN)-8 , interleukin (IL)-4, and IL-6, are not prime candidates for cytokine gene-therapy approaches. On the other hand, cytokines not dependent on MHC class I antigen processing (IL-2 and granulocyte macrophage-colony stimulating factor [GM-CSF]) may be more suitable for prostate cancer gene therapy. Sanda and coworkers²⁶ found that therapy using gene-modified, irradiated vaccine cells genetically transduced to secrete GM-CSF prolonged survival in animals with prostate cancer; thus, this approach is feasible and has potential for wide application as a treatment strategy for human prostate cancer.

    There are, however, limitations to the ex vivo approach. The efficacy of this tumor vaccine approach depends on establishing reliable, high levels of cytokine expression within the tumor. Transfection of tumor-cell vaccine often requires surgical removal of the primary tumor. Furthermore, the harvesting, in vitro culture, and transfection of autologous cells makes the ex vivo vaccine approach labor intensive, time consuming, and expensive. Another cytokine therapy approach involves the introduction of cytokine genes directly into viruses or packaged segments of DNA that can deliver the cytokine gene directly into the tumor cells. Transfection and genomic incorporation result in the chronic local production of cytokines from within the tumor itself. Both viral vectors and liposomecytokine gene complexes have been shown to be effective in animal and human studies.⁷

    A second form of cytoreductive gene therapy under clinical development is the transfer of drug-susceptible or "suicide" genes. In this strategy, a gene is transfected into tumor cells that encodes the active site of an enzyme, which converts a nontoxic prodrug form of an antineoplastic drug into a cytotoxic one in transfected tumor cells. The prodrug agent is given intravenously after gene transfer. Herpes simplex virus thymidine kinase (HSV-tk) is a classic suicide gene. Others, such as varicella zoster and E-coli cytosine deaminase, have been used in other models.⁵

    HSV-tk converts nontoxic nucleoside analogues such as ganciclovir (GCV) into phosphorylated compounds that act as chain terminators of DNA synthesis.⁹ GCV is safe when given systemically as an chemotherapeutic antivirus and is a high-affinity substrate for HSV-tk. The introduction of the HSV-tk gene has been achieved via adenoviral or retroviral vectors. One of the potential advantages of HSV-tk therapy is that it demonstrates the "bystander effect," which is the percentage of cells killed exceeds the percentage of cells originally transduced. There are several explanations for this phenomenon, including the possibility that toxic metabolites are transferred between juxtaposed cells and the possibility that systemic immune response is induced.²⁷ To date, close cell–cell contact appears necessary for the "bystander" effect to work. Hall and colleagues²⁸ demonstrated that adenovirus-mediated HSV-tk/GCV therapy leads to systemic activity against spontaneous and induced metastasis in an orthotopic mouse model of prostate cancer. Eastham et al²⁹ showed that HSV-tk/GCV cytotoxic gene therapy can inhibit the growth of mouse and human prostate cancer cells in vitro and can interrupt tumor growth of an aggressive mouse prostate cancer cell line in vivo.

    A modification of cytoreductive gene therapy involves tissue-specific expression of drug-susceptible genes. Cell toxin vectors are being constructed that contain tissue-specific promoters to restrict expression of the transferred cytotoxic gene. In prostate cancer, this approach has additional promise for the creation of tissue-specific gene therapy because of the recent discovery of a bipartite, PSA enhancer-promoter sequence. This DNA sequence results in high levels of androgen-sensitive, prostate tissue-specific gene expression. In more than 95% of the cases of metastatic prostate cancers, PSA promoter is used to express detectable levels of PSA protein. When the PSA enhancer-promoter sequence is combined with the HSV-tk, there is potential for prostate cell-specific expression of the transfected gene, with cytotoxic effects limited only to the target tissue.⁷

    An entirely new class of cytoreductive gene therapy vectors can be generated as a consequence of the identification of specific transcription elements: oncolytic, replication-restricted viruses. Minimal enhancer-promoter DNA sequences for PSA have been put into the adenovirus genome to drive the control of viral replication genes.¹⁸ The human adenovirus E1B gene encodes a 55-kd protein (E1B 55K) that binds and inactivates p53. Bischoff et al³⁰ showed that a mutant adenovirus that does not produce this viral protein can replicate in and lyse p53-deficient human tumor cells but not cells with functional p53.

    An alternative strategy for cytoreductive gene therapy focuses on bone metastasis of prostate cancer. Osteoblastic response to prostate cancer is the hallmark of progression at this metastatic site. Osteocalcin (OC), a noncollagenous bone matrix protein, is expressed in high levels by osteoblasts. Ko et al³¹ constructed a recombinant adenovirus, AD-OC-tk, which contains the OC promoter that drives the expression of HSV-tk as suicide gene. They showed that the OC promoter mediated high levels of expression in osteoblast cell lines. Treatment with the AD-OC-tk plus prodrug has potential to eradicate osteoblastic cells that may be required to maintain the survival of osseous metastatic tumors in prostate cancer.

Corrective Gene Therapy

    Many different genetic alterations have been identified in prostate cancer (Table 2). Most of the lesions represent either overexpression of an oncogene or inactivation of a tumor-suppressor gene. Tumor-suppressor genes and antisense oncogenes are used chiefly as reagents for corrective gene therapy.

    In metabolic diseases in which a single gene defect has been identified as the cause of the disease state, such as in cystic fibrosis, replacing the defective gene product is a promising treatment approach. However, the main problem in cancer is that there is no single oncogene or tumor-suppressor gene defect.⁷

    Tumor-suppressor genes are a diverse group of genes present in the normal genome. Their inactivation may result in the initiation or the progression of a cancer. The p53 gene is the most commonly mutated gene in human cancer. The p53 gene replacement is a particularly attractive therapeutic strategy because in vitro restoration of wild-type p53 in many tumor cell lines causes growth arrest or apoptosis. The most common genomic alteration in prostate cancer is inactivation of the glutathione-S-transferase P1 gene (GST-P1).³² This inactivation may occur as early as prostatic intraepithelial neoplasia (PIN), and it is identified in over 98% of cases of clinically detected prostate cancer. GST-P1 is an attractive target gene for corrective gene therapy research. Many other suppressor genes also have been identified as potential targets for prostatic cancer corrective gene therapy, including p21,³³ CAMS, and kai1. The University of Texas M.D. Anderson Cancer Center with the University of Tennessee and the University of California at Los Angeles have begun phase I clinical trials of p53 gene replacement therapy for prostate cancer patients (Table 1).⁶

    Antisense oligodeoxynucleotides are short synthetic nucleotide sequences formulated to be complementary to specific DNA or RNA sequences.³ They may be delivered to target cells with annealing of the strands and thus can potentially disrupt transcription or translation of target oncogenes.⁵ The bcl-2 oncoprotein suppresses apoptosis, and when overexpressed in prostate cancer cells, bcl-2 makes these cells resistant to a variety of therapeutic agents, including hormonal ablation drugs. Dorai et al¹¹ have synthesized a hammerhead ribozyme against bcl-2 mRNA and demonstrated efficient cleavage in vitro. Antisense bcl-2 is only one example of many antisense oligodeoxynucleotides that could be used in corrective gene therapy.

The p53 Gene -- An Appropriate Gene for Prostate Cancer Gene Therapy

    The p53 tumor suppressor gene is a 393-amino acid nuclear phosphoprotein that acts as a transcription factor to control expression of proteins involved in the cell cycle.³⁴ The p53 gene maps to the P arm of chromosome 17 with loss of heterozygosity resulting in expression of a mutant allele.³⁵ It is continuously synthesized and degraded; its levels are low under physiological conditions. When p53 levels rise, the result is G₁ arrest of cell cycle or apoptosis. The levels of p53 increase after irradiation and other types of cell damage. These increased levels of p53 in the nucleus arise from an increase in its half-life. Increased levels of p53 downregulate bcl-2 (which inhibits apoptosis) and upregulate the expression of specific genes including p21 and Bax.³⁶ Bax is a potent promoter of the apoptotic cell death pathway, and p21 inhibits cyclin-dependent kinases. Because of its functions, p53 has been called the "guardian of the genome," and its loss has been implicated in tumor progression.³⁴

    There is some evidence that p53 might exert some of its antitumor activity through inhibition of angiogenesis as a consequence of an altered production of thrombospondin (bystander effect).³⁷ In addition, the presence of the wild-type p53 gene is speculated to be useful in accelerating induction of apoptosis caused by cytotoxic drugs such as cis-platinum.¹⁸ Thus, corrective gene therapy with p53 is viewed by some as having the potential to improve the therapeutic index of available chemotherapeutic drugs.

    The p53 gene is the most commonly mutated gene in human cancer.¹⁸ Early studies have shown that approximately 60% of prostate cancer cell lines have mutations in the p53 gene.⁷ Although primary prostate tumors have few mutations in p53 gene, specimens from advanced stages of the disease and metastases as well as their cell lines frequently display mutations or deletions at both alleles of the p53 gene.³⁶ Transfection of wild-type p53 into prostate cancer cell lines expressing mutant alleles results in loss of tumorigenic capability.³⁸ Gotoh et al³⁹ demonstrated that p53 was able to suppress tumorigenicity regardless of the background p53 status of the tumor cells.⁴⁰

Ad-p53 - Intraprostatic Gene Therapy (INGN 201)

    Gene therapy strategies that attack the underlying genetic mechanism of this disease offer promise in improving outcomes. In the protocol described here, investigators at M.D. Anderson Cancer Center and the University of Tennessee propose to study the effect of intraprostatic AD-p53 (INGN 201) injection (Table 1). INGN 201 is a replication-defective adenoviral vector that encodes a wild-type p53 gene driven by a cytomegalovirus (CMV) promoter.

    Eligible patients will have clinical stage T1c or T2a with high-grade disease (Gleason grade 8-10) on initial biopsy or clinical stage T2b-T2c with Gleason grade 7 and PSA >10, or clinical stage T3.

    Patients with locally advanced prostate cancer who are enrolled in the study undergo baseline magnetic resonance imaging (MRI) and ultrasound of the prostate. One course of p53 therapy is defined as three separate injection procedures separated by two weeks, with reevaluation using transrectal ultrasound and MRI. If these show reduction in the size of measurable lesion, then a second course of p53 gene therapy is performed. If there is no change after the first course from the baseline MRI and transrectal studies, then patients are treated with radical prostatectomy. Those patients completing a second course of p53 gene therapy are again reevaluated and proceed to radical prostatectomy.

The Technique of AD-p53 Injection

    Using a specially designed needle guide that fits to the transrectal ultrasound probe and positions a needle at a measurable distance above the ultrasound transducer head, 17-gauge needles with stylettes are inserted transperineally into the prostate at the positions shown in Fig 1. The needles are advanced to the apex of the prostate and then the stylettes are withdrawn, which allows an 18-gauge core biopsy to be obtained longitudinally along the axis of the prostate. The stylette is then advanced back through the needle, and the needle is advanced along the tract of the biopsy longitudinally from the apex to the base of the prostate. This allows a longitudinal core biopsy to be obtained at the site of injection. The p53 gene in the adenoviral vector is preloaded in 1-mL plastic "TB" syringes that are then hooked onto the 17-gauge needles and injected into the prostate as shown in Fig 2.

    A total of 30 patients will be enrolled in this study. Because the study is ongoing and the results are preliminary, it is not possible to comment on the efficacy of the treatment at this time.

The Future of Gene Therapy

    The future of gene therapy approaches to prostate cancer or other cancers will depend on further development of vector systems at the basic science level as well as a better understanding of the genes involved in tumor induction and proliferation. In the long term, we will look at the construction of artificial chromosomes that can carry whole clusters of genes with their natural control elements into cells. With this new technology also comes new ethical responsibilities to ensure that these strategies are safe for patients and staff.³

References

1. Parker SL, Tong T, Bolden S, et al. Cancer statistics, 1996. CA Cancer J Clin. 1996;46:5-27.

2. Cancer facts and figures - 1998. Atlanta, Ga: American Cancer Society; 1998.

3. Sikora K, Pandha H. Gene therapy for prostate cancer. Br J Urol. 1997;79:64-68.

4. Blaese RM, Culver K, Anderson WF. The ADA human gene therapy clinical protocol. Hum Gene Ther. 1990;1:331-362.

5. Hrouda D, Dalgleish AG. Gene therapy for prostate cancer. Gene Ther. 1996;3:845-852.

6. ORDA/NIH Internet. Human gene therapy protocols. Site: www.nih.gov/od/orda/protocol.htm update, 4/7/98.

7. Naitoh J, Belldegrun A. Gene therapy -- the future is here: a guide to the practicing urologist. Urology. 1998;51:367-380.

8. Freireich EJ, Stass SA, eds. Molecular Basis of Oncology. Cambridge, Mass: Blackwell Science; 1995:341-377.

9. Dannull J, Belldegrun AS. Development of gene therapy for prostate cancer using a novel promoter of prostate-specific antigen. Br J Urol. 1997;79:97-103.

10. Thompson TC. Growth factors and oncogenes in prostate cancer. Cancer Cells. 1990;2:345-354.

11. Dorai T, Olsson CA, Katz AE, et al. Development of a hammerhead ribozyme against bcl-2, I: preliminary evaluation of a potential gene therapeutic agent for hormone-refractory human prostate cancer. Prostate. 1997;32:246-258.

12. Umbas R, Isaacs WB, Bringuier PP, et al. Decreased E-cadherin expression is associated with poor prognosis in patients with prostate cancer. Cancer Res. 1994;54:3929-3933.

13. Dong JT, Lamb PW, Rinker-Schaeffer CW, et al. KAI1, a metastasis suppressor gene for prostate cancer on human chromosome 11p11.2. Science. 1995;268:884-886.

14. Sanda MG, Simons JW. Gene therapy for urologic cancer. Urology. 1994;44:617-624.

15. Harrison GS, Glode LM. Current challenges of gene therapy for prostate cancer. Oncology. 1997;11:845-856.

16. Taneja SS, Pang S, Cohan P, et al. Gene therapy: principles and potential. Cancer Surv. 1995;23:247-266.

17. Mahato RI, Kawabata K, Nomura T, et al. Physicochemical and pharmacologic characteristics of plasmid DNA/cationic liposome complexes. J Pharm Sci. 1995;84:1267-1271.

18. Simons JW, Marshall FF. The future of gene therapy in the treatment of urologic malignancies. Urol Clin North Am. 1998; 25:23-38.

19. Ali M, Lemoine NR, Ring CJ. The use of DNA viruses as vectors for gene therapy. Gene Ther. 1994;1:367-384.

20. Mulligan RC. The basic science of gene therapy. Science. 1993;260:926-932.

21. Yei S, Bachurski CJ, Weaver TE, et al. Adenoviral-mediated gene transfer of human surfactant protein B to respiratory epithelial cells. Am J Respir Cell Mol Biol. 1994;11:329-336.

22. Smith TA, Mehaffey MG, Kayda DB, et al. Adenovirus mediated expression of therapeutic levels of human factor 9 in mice. Nat Genet. 1993;5:397-402.

23. Anderson WF. Human gene therapy. Science. 1992;256:808-813.