PROSTATE CANCER GENE THERAPY

0889-8588/01 $15.00

PROSTATE CANCER

+

.OO

PROSTATE CANCER GENE THERAPY Fernando A. Ferrer, MD, and Ronald Rodriguez, MD, PhD

The death of a patient at the University of Pennsylvannia gene therapy trial for ornithine transcarbamylase deficiency led to an outpouring of interest regarding the efficacy and potential hazards of gene therapy. This unfortunate outcome prompted internal and external reviews of the strategies and potential toxicities of adenoviral-based gene therapy as well as gene therapy as a whole. The development of gene therapy treatments largely will profit from this tragedy, however, because these inquiries will strengthen the rigor of product development and risk assessment. The possibility that through genetic manipulation the course of diseases, such as prostate cancer, can be altered has captured the imagination of scientists and the lay public. Regardless of the obstacles and technical difficulties to be overcome, ultimately molecular medicine and gene therapy will be a reality in the treatment of human disease. As a consequence, medical practitioners need to understand the role for gene therapy in the treatment of urologic malignancy so that they may answer patients’ questions and refer interested, eligible patients to clinical trials. EX VlVO GENE THERAPY (IMMUNOTHERAPY)

Tumor Immunology

Gene therapy, broadly defined, refers to the use of genetic materials in the treatment of human disease. Such manipulation of genetic materiFrom The James Buchanan Brady Urological Institute, The Johns Hopkins Hospital, Baltimore, Maryland

HEMATOLOGY/ONCOLOGY CLINICS OF NORTH AMERICA VOLUME 15 NUMBER 3 JUNE 2001

497

498

FERRER & RODNGUEZ

als (RNA or DNA) can occur in vivo (in the patient or animal) or ex vivo (outside of the patient, with the intention of subsequent patient use). The earliest examples of prostate cancer gene therapy relied on the ex vivo model, whereby prostate cancer cells were engineered to be highly stimulatory for a given patient's immune system, in the hopes of establishing an effective prostate cancer vaccine.19 The basic immunologic principles underlying this strategy are outlined in Figure 1, and the currently available clinical trials are summarized in Table 1. Degenerating or dying prostate cells liberate antigens as they are processed by the macrophages and monocytes locally. Certain subsets of these phagocytic cells are highly specialized for antigen processing and antigen presentation to T cells. These professional antigen-presenting cells (APC) include the so-called dendritic cells (DCs), in reference to their cellular appearance as they extend multiple membrane processes outward to sample and process. antigens in their local microenvironment.23DCs arise from a common CD34-positive progenitor in the bone marrow whose expansion and differentiation are influenced by a variety of cytokine growth factors, including stem cell factor, flt3 ligand, interleukin-3 (IL-3), granulocyte-macrophage colony-stimulating factor

Figure 1. Dendritic cell (DC) activation of T-helper and cytotoxic T lymphocytes involved in tumor immunologic response.

W

Liposomal IL-2 gene transfer in situ for advanced prostate cancer Liposomal IL-2 gene transfer as neoadjuvant therapy Autologous dendritic cells pulsed with mRNA encoding PSA Autologous dendritic cells pulsed with PSMA peptides GM-CSF retrovirally transduced allogeneic prostate cancer cells IL-2/interferon-y retrovirally transduced allogeneic prostate cancer cells GM-CSF retroviral gene transfer to allogeneic prostate cancer cells GM-CSF retroviral gene transfer to autologous prostate cells IL-2 liposomal gene transfer to autologous prostate cells Globo-H hexasaccharide, common antigen on prostate cancer cells Recombinant PSA produced through vaccinia vector Recombinant PSA produced through vaccinia vector Recombinant MUC-1/IL-2 produced through vaccinia vector Recombinant PSA produced through vaccinia vector Recombinant PSA produced through vaccinia/fowlpox vector Recombinant PSA produced through vaccinia/fowlpox vector

Strategy

UCLA UCLA DUMC PNCF JHH MSKCC UCSF JHH DUMC MSKCC DFCI UMAA UCLA NNMC AECM DFCI

A. Belldegrun A. Belldegrun J. Vieweg G. Murphy J. Simons 8.Gansbacher

E. Small J. Simons

D. Paulson s. Slovin D. Kufe M. Sanda R. Filglin A. Chen H. Kaufman J. Elder

Institution*

Primary Investigator

'AECM = Albert Einstein College of Medicine, New York, NY; DFCI = Dana Farber Cancer Institute, Boston, MA; DUMC = Duke University Medical Center, Durham, NC; JHH = Johns Hopkins Hospital, Baltimore, MD; MSKCC = Memorial Sloan-Kettering Cancer Center, New York, NY; NNMC = National Navy Medical Center, Bethesda, MD; PNCF = Pacific Northwest Cancer Foundation, Seattle, WA; UCLA = University of California Los Angeles Medical Center, Los Angeles, CA; UCSF = University of California San Francisco, San Francisco, CA; UMAA = University of Michigan Ann Arbor, Ann Arbor, MI. IL-2 = Interleukin-2; PSA = prostate-specific antigen; PSMA = prostate-specific membrane antigen; GM-CSF = granulocyte-macrophage colony-stimulating factor.

Vaccine-allogeneic Vaccine-autologous Vaccine-autologous Vaccine-carbohydrate Vaccine-vaccinia Vaccine-vaccinia Vaccine-vaccinia Vaccine-vaccinia Vaccine-fowlpox Vaccine-fowlpox

In situ cytokine In situ cytokine Dendritic cell Dendritic cell Vaccine-allogeneic Vaccine-allogeneic

Method

Table 1. PROSTATE CANCER IMMUNOTHERAPY TRIALS

500

FERRER & RODRIGUEZ

(GM-CSF), tumor necrosis factor-a (TNF-a), and transforming growth f a ~ t 0 r - PAfter . ~ ~ antigen internalization by the DCs and complex formation with major histocompatibility complex (MHC) molecules, the cell begins the process of activation and maturation. These activated DCs undergo a series of physiologic changes leading to their terminal differentiation. They down-regulate their antigen-uptake machinery, up-regulate adhesion and costimulatory molecules, and stabilize peptide-MHC complexes on their cell surface. The DCs travel through the lymphatic system to local lymph nodes, where they complex with T cells by extending further their dendritic processes to interact with antigenreactive clones. Ligation of CD40 on the surface of the activated DCs by T cell-derived CD40 ligand results in further up-regulation of CD8O (HLA-B7-1) and CD86 (HLA-B7-2) and high levels of IL-12 secretion by the DCs. Activation of these co-stimulatory factors promotes the development of T helper-1 (Thl) CD4-positive T cells and the maturation of cytotoxic T lymphocytes (CTL). DCs are unique in that they express high levels of these costimulatory molecules on their cell surface, and the degree of immune response stimulation is regulated in large part by the ratio of these costimulatory to coinhibitory molecules!, 23

Strategies of Prostate Cancer lmmunotherapies

Tumor cell vaccines initially were explored in animal models for prostate cancer using cultured cells engineered to express IL-2 or GMCSF. Work using the Dunning rat R3327 Mat-Ly-Lu model showed that treated animals had regression of their primary lesions and immunologic protection from subsequent tumor challenge^.^^ The results of these studies formed the basis for initial cytokine vaccine studies in humans. In the human trials, prostate cancer cells harvested at the time of radical prostatectomy were engineered to become highly antigenic for use as an autologous vaccine. Simons et all9 at the Johns Hopkins Hospital performed a phase I trial in which 8 patients underwent vaccination with autologous, GM-CSF-secreting, irradiated prostate cells. Tumor vaccines were prepared by ex vivo transfection of surgically harvested cells. Vaccine site biopsy specimens showed infiltrates of DC and macrophages. New T-cell and B-cell responses were noted after vaccination in these patients. Both findings suggested a host immune response to the vaccine. Side effects in this group were limited to pruritus, erythema, and swelling of the vaccination site. Follow-up was insufficient to assess outcomes in this small study. The cost and difficulty in obtaining adequate tissue for gene transfer led to the development of an autologous prostate cancer vaccine, using the immortalized human line LNCaP and PC-3, which have been engineered to oversecrete human GM-CSF. Although this trial is still ongoing (G-VAX, Cell Genesys, Inc, Foster City, CA), the preliminary results indicate minimal side effects and modest

PROSTATE CANCER GENE THERAPY

501

decreases in serum prostate-specific antigen (PSA) (Simons, J: personal communication, 2000). Subsequent trials using this strategy in combination with chemotherapeutic agents already are under way. Alternative targets for vaccination include carbohydrate moieties such as Globo H, a complex hexasaccharide present on the surface of many prostate cancer cells. When conjugated to an adjuvant such as the keyhole limpet hemocyanin, a prostate-specific vaccination can be generated. Although these types of trials are only in their preliminary stages, the initial results have shown almost no toxicity and modest improvements in the PSA profiles.20 In a unique approach, Kwon et als attempted to augment host immune response by manipulating the costimulatory pathways involved in T-cell activation. Because binding of HLA-B7 complex to CTLA4 receptors on T cells results in a down-regulation of T-cell activation (see Fig. l), Kwon et a18 sought to inhibit CTLA4 activity by the use of blocking antibodies. Experiments performed in mice receiving intraperitoneal injections of anti-CTLA4 antibodies (and now anti-CD40 antibodies) showed substantial delays in tumor growth when compared with controls. In total, 21 of 50 animals experienced complete tumor rejection. In the same report, these authors showed that genetically altering pTCl (TRAMP derived prostate cancer cells) to overexpress HLA-B7 resulted in tumor rejection in a murine model.* Together, these findings emphasize the important role of the costimulatory pathways in tumor immunology and highlight future directions for tumor vaccination strategies. Although cancer vaccines hold a great deal of promise in combating systemic prostate cancer, an alternative approach of directly manipulating DCs ex vivo has evolved. Recognizing the central importance of DCs in the regulation of the immune system, Murphy et al” reported their experience using pulsed DC therapy in patients with advanced hormone-refractory prostate cancer. In a study reported by Salgaller et al,17 monocyte-derived DCs were harvested and exposed to HLA-A2 binding prostate-specific membrane antigen peptides. Fifty-one patients with hormone-refractory prostate cancer (including control groups) were evaluated. Therapy was well tolerated, with transient hypotension being the only major side effect. The concurrent administration of radiotherapy and hormonal therapy makes interpretation of the vaccine’s effects difficult. Nonetheless, 7 of 51 patients had partial responses as defined by a decrease in PSA of greater than 50%. Data on this approach indicate that patients can be retreated safely with escalating vaccine doses and that some patients may have lasting re~p0nse.l~ Multiple immunoevasive behaviors have been identified in prostate cancer that may allow the cells to escape certain aspects of tumor surveillance. In particular, some advanced prostate cancer cells have defective MHC I processing.16, Because activated killer T cells localize to prostate cancer cells, which express particular tumor antigens in the context of MHC I expression, this lack of expression hinders CTL activity. There is some evidence to suggest that some advanced prostate cancer

502

FERRER & RODRIGUEZ

cells may secrete soluble factors that interfere with T-cell receptor function, resulting in aberrant delta chain expression and a further diminished CTL re~ponse.~ Such mechanisms, along with inadequate costimulatory molecule expression, may be responsible for the phenomenon of prostate cancer immunotolerance. Methods of circumventing some of these immunoevasive features are being developed, including the use of genetically modified tumor cells that express certain costimulatory molecules (e.g., HLA-B7) normally found on antigen-presenting cells.8 Improvements in DC harvesting technology as well as advances in understanding of the regulatory components of the immune system likely will result in further progress in the immunotherapy of prostate cancer. IN VlVO GENE THERAPY

Although the earliest gene therapy approaches involved genetic manipulation ex vivo, the largest research effort in prostate cancer was in direct injection of the gene therapy constructs in vivo. Effective in vivo gene therapy relies on the successful transfer of genetic information to living cells in situ to achieve a therapeutic benefit; this is accomplished by delivering the DNA or RNA in a vehicle that allows efficient gene transfer. Such vehicles include liposomes, viruses, DNA, and ternary complexes of DNA and viruses. Table 2 lists the more commonly applied vehicles for gene transfer and their advantages and disadvantages. In general, the choice of vehicle depends in large part on the goal of the therapy and the size of the gene to be delivered.12Transient expression may be adequate for eliciting cell death or an intense immune response but is suboptimal for correcting cellular deficiencies. Liposomes and naked DNA approaches can facilitate larger segments of DNA but are less efficient at gene transfer than viral methods.l0 Even within viral vectors, there are significant differences in the method and temporal expression of gene transfer. Retroviruses integrate into the host genome and have longer term expression; however, such integration raises the possibility of incidental activation of neighboring proto-oncogenes or disruption of tumor-suppressor genes-either of which in theory could result in the formation of de novo tumors.3,22 This theoretic risk has not been realized in any clinical trial, however. Adenoviral-based vectors bypass this concern by maintaining their DNA completely in an episoma1 form (i.e., no integration), but as a result the gene expression is transient in nature. Most viruses are highly immunogenic, limiting efficiency of gene transfer and e x p r e s s i ~ nThis . ~ ~ immunogenicity is thought to have played a role in the gene therapy-related death of a University of Pennsylvania patient and underscores the potential pitfalls in current gene transfer methodology. Corrective gene therapy strategies rely on attempting to reconstitute a lost function in the tumor cells (e.g., c-myc, p26, p53, or RB). KO et a17

cn 0 w

Advantage

As above, less immunogenic

Large DNA segments transferred Noninfectious Large DNA segments transferred, noninfectious

Easy-to-render replication deficient, high-efficiency gene transfer, infects dividing and nondividing cells All cells susceptible, stable expression Allows for large inserts, predilection for neural tissue Infects dividing and nondividing cells, stable expression, can be attenuated Attenuated because avian strain, allows large DNA inserts High-efficiency gene transfer, large DNA inserts, immunogenicity may be helpful for vaccine Nonimmunogenic, stable expression

Stable expression, integrates into host DNA

HIV = Human immunodeficiency virus.

Stealth liposome

Simian virus Nonviral Electroporation Gene gun Liposomes

Vaccinia virus

Canary pox virus

Lentivirus (HIV)

Adeno-associated virus Herpes simplex virus

Adenovirus

Viral Retrovirus

Vector

Table 2. CURRENT GENE TRANSFER VECTORS

Ex vivo use only, transient expression Transient expression, ? in-vivo utility Transient expression, immunogenic, poor transfer efficiency As above

Small DNA insert size

Transient expression, immunogenic

Transient expression, immunogenic

Small DNA insert size, immunogenic Cytotoxic, immunogenic, difficult to purify, low titers Theoretic risk HIV, insertional oncogenesis

Lower titers, requires dividing cells, ? insertional oncogenesis Transient expression, immunogenic

Disadvantage

504

FERRER & RODRIGUEZ

reported that reinsertion of wild-type p53 into mouse models of human prostate cancer resulted in high levels of p53 expression and aborted tumor growth. Currently, there are several p53 adenoviral-based gene transfer clinical trials in progress as monotherapy and as neoadjuvant therapy. Antisense strategies for prostate cancer are being evaluated. Steiner et alZ1looked at neutralizing the c-myc proto-oncogene in a mouse model of DU-145 tumors. c-myc overexpression leads to cell transformation, proliferation, and inhibition of apoptosis. Delivery of a retroviral driven c-myc antisense construct resulted in a 94% reduction in tumor size. As shown in Table 3, a clinical trial based on this study currently is in progress. Another approach to in vivo prostate cancer gene therapy is to kill the cells by means of a suicide-type gene transfer. The most commonly used suicide gene is the herpes simplex thymidine kinase gene, which phosphorylates the otherwise inert substrate ganciclovir (an acyclic nucleoside analog of 2'-deoxyguanosine) to its monophosphate form. This ganciclovir monophosphate is a substrate for endogenous kinases and subsequent phosphorylation into the toxic ganciclovir triphosphate, which, if incorporated into the elongating DNA during DNA synthesis, leads to premature chain termination. For a rapidly proliferating cell, poisoning DNA synthesis is a lethal process. Because the toxic ganciclovir monophosphates, diphosphates, and triphosphates are small molecules, they can diffuse across gap junctions and poison bystander cells, which are not even expressing thymidine kinase. The effect can be multiplied over a larger number of cells than can be achieved by gene transfer alone.6 The group at Baylor also reported early data from a thymidine kinase/ganciclovir-based phase I clinical trial for human prostate cancer.6An adenoviral vector-driven construct was used in conjunction with direct intraprostatic injections into the prostate guided by ultrasound. Eighteen patients were enrolled, and minimal toxicity was seen. The most serious side effect occurred in 1 patient receiving the highest dose level and consisted of thrombocytopenia and hepatotoxcity, which improved with cessation of ganciclovir. Objective response (decrease in PSA >50%) was seen in 3 patients, which was sustained for 6 weeks to 1 year, but no patient was cured. Other similar suicide genes include cytosine deaminase, purine nucleoside phosphorylase, and diphtheria toxin.', 9, l3 DNA synthesis poisons traditionally have not been effective in prostate cancers, when approached as standard chemotherapeutic agents. Part of the reason for such a resistance to these chemotherapeutic agents may be that less than 3% of prostate cancer cells are dividing actively at any time. In theory, suicide gene therapy strategies based on active DNA replication may not be as effective as cell cycle-independent strategies. Nonetheless, these hsv-TK adenoviral strategies have resulted in clearly demonstrable cell kill on prostate cancers either excised surgically or as assessed by PSA progression. Concomitant with these modest effects has been a

Adenovirus

Antisense Corrective Corrective Oncolytic Suicide Suicide Suicide Suicide

Suicide

hsv-TK

Inhibit c-myc by expressing antisense mRNA Intratumoral injection, replace nonfunctioning p53 Intratumoral injection, neoadjuvant PSA promoter directing replication of adenovirus Intratumoral injection, constitutively expressing hsv-TK Neoadjuvant, intratumoral injection Neoadjuvant, intratumoral injection Osteocalcin-directed expression of thymidine kinase, for bone metastasis Neoadjuvant radiation treatment

Strategy

BCM

B. Butler

Institution’

UT/WMC UCLA UTMDA JHH BCM/MSKCC MSSM UTMDA WHSC M. Steiner A. Belldegrun C. Logathetis J. Simons P. Scardino S. Hall C. Logathetis T. Gardner

Primary Investigator

= Baylor College of Medicine, Houston, TX; JHH = Johns Hopkins Hospital, Baltimore, MD; MSKCC = Memorial Sloan-Kettering Cancer Center, New York, = Mount Sinai School of Medicine, New York, NY;UCLA = University of California Los Angeles Medical Center, Los Angeles, CA; UTMDA = University of Anderson, Houston, TX; UT/VUMC = University of Tennessee/Vanderbilt University Medical Center; UVHSC = University of Virginia Health Science Center,

Texas-M.D. Charlottesville, VA

NY;MSSM

‘BCM

c-myc

Retrovirus Adenovirus Adenovirus Adenovirus Adenovirus Adenovirus Adenovirus Adenovirus

P53 P53 Ela hsv-TK hsv-TK hsv-TK hsv-TK

Gene

Vector

Method

Table 3. PROSTATE CANCER IN VlVO GENE THERAPY TRIALS

506

FERRER & RODRIGUEZ

stimulation of cellular immunity, however, and much of the effect seen with the current adenovirus gene therapy constructs may be a result of immunologic stimulation by the adenoviral vehicle, rather than any particular transgene being used. Further evaluation of the degree of immunologic stimulation by in vivo gene transfer is warranted. Cytoreductive gene therapy is not limited to suicide genes, but can occur by a variety of viruses as a part of their life cycle. When these viruses have been engineered to divide preferentially and lyse in cancer cells, they are referred to as oncolytic viruses. Many different oncolytic viruses have been devised, including the Onyx virus, which replicates preferentially in cells with mutant p53 status15;the herpes simplex virus G207,26which replicates in cells abundant in ribonucleotide reductase (a phenotype common to many malignancies); and prostate-specific attenuated replication competent adenoviruses CN706 and CV787, which have been engineered to replicate in cells capable of expressing the PSA gene.14,27 Oncolytic viruses have several advantages over conventional adenoviral gene transfer methods. First, the injected dose is less than the final treatment dose because the virus replicates at the site of injection and multiplies the initial dose by many-fold. Second, toxicity of replication is minimized by using a prostate-specific expression system. Third, immunostimulation by the replicating virus may have an adjuvant effect on stimulating the host immune system against the patient’s cancer. Such immunomodulatory effects are largely theoretical, but given the observations of cellular immunity activation in other adenoviral gene transfer systems, this is still a likely benefit.

SUMMARY

Cancer-specific gene therapy is still in its infancy. Although the first gene therapy trials were initiated in the late 1980s, it was only more recently that the first successful treatment of a genetic disease was r e p ~ r t e d The . ~ current problems with low efficiency of gene transfer coupled with the immunologic difficulties with certain vectors indicate that more effort needs to be directed at the basic science of gene transfer. Ultimately, successful cancer-specific gene therapy will require combinations of the lessons learned from the ex vivo and in vivo paradigms. The next generation of gene therapy trials likely will focus on combination therapy with conventional chemotherapeutic agents, differentiating agents, or radiation therapy. The obstacles to the development of genebased human therapeutics (i.e., molecular medicine) are formidable, but the benefits are so great that eventually the technical issues of gene transfer methodology will be worked out, and ultimately this will become the standard of care, not only for inborn errors of metabolism, but also for cancer.

PROSTATE CANCER GENE THERAPY

507

References 1. Blackbum RV, Galford SS, Corry PM, et al: Adenoviral transduction of a cytosine deaminase/thymidine kinase fusion gene into prostate carcinoma cells enhances prodrug and radiation sensitivity. Int J Cancer 82293-297, 1999 2. Cavazzana-Calvo M, Haccin Bay S, de Saint Basile G, et al: Gene therapy of human severe combined immunodeficiency (SC1D)-X1 disease. Science 288:669472, 2000 3. Coffin J: Retrovirade and their replication. In Virology. New York, Raven Press, 1990 4. Greenfield EA, Nguyen KA, Kuchroo VK CD28/B7 costimulation: A review. Crit Rev Immunol 18:389418, 1998 5. Healy CG, Adler HL, Aguilar-Cordova E, et a1 Impaired expression and function of signal-transducing zeta chains in peripheral T cells and natural killer cells in patients with prostate cancer. Cytometry 32109-119, 1998 6. Herman JR, et al: In situ gene therapy for adenocarcinoma of the prostate: A phase I clinical trial. Hum Gene Ther 10:1239-1249, 1999 7. KO SC, Gotoh A, Thalmann GN, et al: Molecular therapy with recombinant p53 adenovirus in an androgen-independent, metastatic human prostate cancer model. Hum Gene Ther 71683-1691, 1996 8. Kwon ED, Hurwitz AA, Foster BA, et al: Manipulation of T cell costimulatory and inhibitory signals for immunotherapy of prostate cancer. Proc Natl Acad Sci U S A 94:8099-8103, 1997 9. Martiniello-Wilks R, Daja M, et a1 In vivo gene therapy for prostate cancer: Preclinical evaluation of two different enzyme-directed prodrug therapy systems delivered by identical adenovirus vectors. Hum Gene Ther 9:1617-1626, 1998 10. Matthews KE, Keating A. Gene therapy with physical methods of gene transfer. Transfus Sci 1729-34, 1996 11. Murphy GP, Tjoa B, Simmons SJ, et a1 Higher-dose and less frequent dendritic cell infusions with PSMA peptides in hormone-refractory metastatic prostate cancer patients. Prostate 43:5942, 2000 12. Rodriguez R, Simons JW: Urologic applications of gene therapy. Urology 54:401406, 1999 13. Rodriguez R, Lim HY, Bartowski CM, et a1 Identification of diphtheria toxin via screening as a potent cell cycle and p53-independent cytotoxin for human prostate cancer therapeutics. Prostate 34259-269, 1998 14. Rodriguez R, Schuur ER, Lim HY, et a1 Prostate attenuated replication competent adenovirus (ARCA) CN706: a selective cytotoxic for prostate-specific antigen-positive prostate cancer cells. Cancer Res 572559-2563, 1997 15. Rothmann T, Hengstermann A, Whitaker NJ, et a1 Replication of ONYX-015, a potential anticancer adenovirus, is independent of p53 status in tumor cells. J Virol72947C 9478, 1998 16. Ruiz-Cabello F, Klein E, Garrido F MHC antigens on human tumors. Immunol Lett 29:181-189, 1991 17. Salgaller ML, Tjoa B, Lodge PA et a1 Dendritic cell-based immunotherapy of prostate cancer. Crit Rev Immunol 18:109-119, 1998 18. Sanda MG, Restifo NP, Walch JC, et a1 Molecular characterization of defective antigen processing in human prostate cancer. J Natl Cancer Inst 87280-285, 1995 19. Simons JW, Tjoa B, Rodgers M, et a1 Induction of immunity to prostate cancer antigens: Results of a clinical trial of vaccination with irradiated autologous prostate tumor cells engineered to secrete granulocyte-macrophage colony-stimulating factor using ex vivo gene transfer. Cancer Res 59:5160-5168, 1999 20. Slovin SF, Scher HI: Peptide and carbohydrate vaccines in relapsed prostate cancer: Immunogenicity of synthetic vaccines in man-clinical trials at Memorial SlomKettering Cancer Center. Semin Oncol26448454, 1999 21. Steiner MS, Anthony CT, Lu Y, et al: Antisense c-myc retroviral vector suppresses established human prostate cancer. Hum Gene Ther 9747-755, 1998 22. Temin H: Origin and general nature of retroviruses. In The Retrovirade. New York, Plenum Press, 1992

508

FERRER & RODRIGUEZ

23. Timmerman JM, Levy R Dendritic cell vaccines for cancer immunotherapy. Annu Rev Med 50507-529, 1999 24. Trapnell BC: Gene therapy using adenoviral vectors. Curr Opin Biotechnol 5:617425, 1994 25. Vieweg J, et al: Immunotherapy of prostate cancer in the Dunning rat model Use of cytokine gene modified tumor vaccines. Cancer Res 541760-1765, 1994 26. Walker JR, et a l Local and systemic therapy of human prostate adenocarcinoma with the conditionally replicating herpes simplex virus vector G207. Hum Gene Ther 10:2237-2243,1999 27. Yu DC, et al: The addition of adenovirus type 5 region E3 enables calydon virus 787 to eliminate distant prostate tumor xenografts. Cancer Res 1999. 59:4200-4203, 1999

Address reprint requests to Ronald Rodriguez, MD, PhD Marburg Room 145 600 N. Wolfe Street Brady Urologic Institute Johns Hopkins Hospital Baltimore, MD 21287-2101