¹Urologic Research Laboratories, University of Tennessee, Memphis, Tennessee 38163, USA

²GTx Inc., Memphis, Tennessee 381201, USA

³Department of Comparative Medicine, University of Tennessee, Memphis, Tennessee 38163, USA

Correspondence to: Dr Jeffrey R Gingrich, Department of Urology, 956 Court Avenue, H216, Memphis, TN 38163, USA. E-mail: jgingrich@utmem.edu

Direct transrectal delivery of therapeutic genes utilizing adenoviral vectors for advanced prostate cancer may offer effective treatment at the molecular level. Large animal models to assess feasibility and the intraprostatic and systemic dissemination patterns of these vectors have not been reported. For these studies, a replication-deficient (E1^-/E3^-) recombinant adenovirus (AdRSVlacZ) expressing bacterial -galactosidase (-gal) was delivered under transrectal ultrasound guidance. Two prostate biopsies, followed by concurrent injection of 4.8´10⁹ pfu of the adenoviral vector divided into either 1 or 2 mL of diluent, were performed (n=4). Swabs of the rectum, sputum, and urine were collected and after 72 hours, the animals were sacrificed. Specimens were assayed for the presence of virus and -gal activity. Rectal swabs were transiently positive, whereas urine and sputum samples showed no detectable vector throughout the experiment. -gal activity was observed at the prostate injection sites with detectable activity noted up to 7.5 mm away from the injection site. Systemic dissemination was observed regardless of the injected volume. In conclusion, transrectal prostate biopsy with concurrent prostate injection is a feasible method to deliver therapeutic adenoviral vectors for the treatment of prostate cancer; however, systemic distribution and temporary rectal shedding of virus should be anticipated. Cancer Gene Therapy (2002) 9, 189-196 DOI: 10.1038/sj/cgt/7700425

gene therapy; prostate; adenovirus; -galactosidase

In the year 2001, an estimated 31,500 American men will die from end-stage prostate cancer.¹ Currently available definitive treatments, including radical prostatectomy and radiation therapy, are ineffective for the majority of men diagnosed with locally advanced prostate cancer. Androgen deprivation or manipulation is at best temporarily palliative. Clearly, innovative therapy is greatly needed for those patients who are at high risk at the time of diagnosis for pathologically locally advanced prostate cancer.

The expanding field of molecular biology offers gene therapy as a promising new approach to the treatment of advanced prostate cancer. Numerous authors have studied the potential utilization of adenoviral-mediated transfer of genes to prostate cancer cells primarily in gene replacement or cytotoxic strategies. Kleinerman et al² have explored the application of cell adhesion molecules (CAMs) as tumor suppressors using a CAM-expressing recombinant adenoviral delivery system. Adenoviral-mediated antitumor therapy using the p53 has demonstrated inhibition of prostate tumor growth³ and Shariat et al⁴ have reported that adenovirus-mediated transfer of inducible caspases can trigger apoptosis in prostate cancer cells. Specific enhancer-containing adenoviruses have been shown to be effective against human prostate cancer cell lines producing prostate-specific antigen.⁵ Adenovirus-mediated transduction of the herpes simplex virus thymidine kinase gene in conjunction with ganciclovir has also shown significant tumor suppression.⁶ These and many other studies have established adenoviral-mediated transfer as an effective gene delivery method.

As concentrated efforts to identify one or more critical genes that would be efficacious for therapy are ongoing, the optimal prostate delivery method may be simultaneously or independently investigated. Various unique modes to deliver viral vectors containing therapeutic genes for the treatment of other tumors have been investigated including intravenous, intraarterial, intramuscular, intratumoral, intrathecal, and intravesical injection.^7,8,9 Comparisons of adenoviral gene delivery approaches have shown minimal tissue diffusion of virus with intramuscular injection and wide systemic dissemination of adenovirus with intravenous delivery.¹⁰

Intraprostatic injection has been proposed as a minimally invasive delivery technique that could be readily performed by urologists or radiologists. To date, intratumoral injection as a means to deliver gene therapy to solid tumors comprises a majority of the approved gene therapy protocols.¹¹ Due to the generous and variable vascularity of the prostate, the local distribution and systemic dissemination patterns of an adenoviral vector delivered by an intraprostatic technique are difficult to predict. The local distribution and expression pattern of viral vector delivered into the channel created after prostatic needle biopsy has not been previously investigated. Theoretically, local viral distribution may be enhanced through perfusion along the open arterial, venous, and lymphatic channels. In addition, the biopsy tissue obtained could be analyzed to assess response to treatment during a course of therapy that included multiple treatments. Clearly, a better understanding of gene delivery techniques and distribution pharmacokinetics is critical for the evaluation and design of new gene therapy vectors to determine the clinical efficacy of this new treatment modality.

Any feasible delivery technique must provide efficient gene transfer to the target tissues. The polymerase chain reaction (PCR) is a highly sensitive technique to detect the presence of adenoviral vector after gene transfer. Numerous previous studies have evaluated transfer efficiency and expression in vitro and in vivo in a wide variety of tissue types by measuring the levels of bacterial -galactosidase (-gal) expressed in cells following transfection with vectors expressing lacZ.^{12,13,14,15,16,17} Measurement of -gal activity provides quantitative data to evaluate gene transfer delivery and expression. The primary objective of this study was to use these methods to evaluate the technique of prostate biopsy with concurrent gene therapy injection, to explore the effect of diluent volume on gene transfer, and to determine the systemic dissemination of a lacZ-containing adenoviral vector after transrectal intraprostatic delivery.

Methods

Construction of adenovirus lacZ

A replication-deficient (E1^-/E3^-) recombinant adenoviral vector (AdRSVlacZ) expressing bacterial -gal was generated as previously described.¹⁸ Briefly, the vector was generated via in vivo recombination by cotransfection with an adenoviral genome plasmid. AdRSVlacZ -gal expression was directed under the control of a modified Rous sarcoma virus (RSV) promoter.¹⁷

Concurrent biopsy and intraprostatic delivery of adenoviral vector

After review and approval by the University of Tennessee Animal Care and Use Committee, adult (>2 years of age) dogs received a saline enema the morning of the following procedure and a perioperative dose of cefazolin was administered intravenously. Once anesthesized, they were positioned prone and a digital rectal exam performed. The 7.5-MHz, single-plane, transrectal ultrasound probe was then placed into the rectum and the prostate visualized (Advanced Tech Laboratories, Bothell, WA). Prostate biopsies were performed utilizing an 18-gauge biopsy needle with a spring-loaded biopsy gun (Manan Pro-Mag 2.2, Northbrook, IL). After the biopsy, with the biopsy needle still within the prostate, a specially designed metal sheath (Cook Urological, Indianapolis, IN) was advanced over the needle into the biopsy site. The needle was then removed, allowing prostate injection via the sheath into the biopsy site. The sheath was slowly withdrawn during injection over 2-3 seconds, distributing the gene therapy under ultrasound visualization along the entire tract. Animals received two 0.5- or 1.0-mL injections (one biopsy and injection into each side of the prostate) containing a total of 4.8´10⁹ pfu of AdRSVlacZ diluted in 10 mM Tris-HCl, pH 8.0, 2 mM MgCl₂, and 4% sucrose. The control animal received two 1.0-mL injections of viral diluent alone. Hematoxylin and eosin staining of the prostate biopsies was performed to confirm actual intraprostatic injection.

Sample collection

At time 0 (prior to injection) and then 15 minutes, 1 hour, 4 hours, 24 hours, 48 hours, and 72 hours after injection, cotton swabs of the rectum and sputum were taken and placed into viral media. In addition, suprapubic aspiration or catheterized urine specimens were obtained for analysis. The animals were sacrificed at 72 hours after injection. At necropsy, to minimize the possibility of significant viral contamination during specimen collection, tissue samples were procured in the following order: liver, spleen, kidney, left ventricle, right ventricle, lung, testicle, vas deferens, and mesenteric lymph nodes. At that point, a cystoprostatectomy was performed. Specimens were then taken from the pelvic lymph nodes, hypogastric vein, rectum, and brain. Specimens of apical urethra, posterior periprostatic tissue, and bladder were then taken. Finally, the prostate was sectioned transversely at 5-mm intervals and inspected to identify the injection sites in the left and right lobes. Prostate tissue cores were obtained utilizing a disposable 5-mm biopsy punch from the apical (A) and basilar (B) aspects of the left (L) and right (R) injection tracks and labeled LA1, LB1, RA1, and RB1, respectively (Fig 1). Subsequently, utilizing a new punch, two additional samples were obtained adjacent to the respective injection site and identified as LA2, LA3, LB2, LB3, RA2, RA3, RB2, and RB3 (Fig 1). Each core was divided and snap-frozen in liquid nitrogen for later DNA or protein isolation.

Analysis of the rectal, sputum, and urine samples

Rectal, sputum, and urine swabs were diluted in 1.0 mL of RPMI+10% heat-inactivated fetal bovine serum (HI FBS, complete medium), which was then filtered through a 0.22-M syringe top filter. Ten microliters of each sample was added to 90 L of complete medium and subsequently overlaid on 5´10³ 293 cells/well of a 96-well plate. The plate was spun at 1000´g for 90 minutes and then incubated at 37°C with 5% CO₂.¹⁹ On days 5 and 8, 100 L of RPMI+20% HI FBS was added to each well. Each well was observed on days 7, 8, 9, and 10 for cytopathic effects (CPE) defined as evidence of rounded cells detaching from the culture plate in grape-like clusters as an indication of adenovirus presence within the sample.

Cytotoxicity observed within 24 hours of application was deemed too immediate to be attributed to the effects of adenoviral replication alone. To confirm the presence of viral DNA, CPE-positive cell lysates were prepared for PCR using the High Pure PCR Template Preparation Kit (Roche, Indianapolis, IN) according to the manufacturer's protocol. Five-microliter aliquots from each sample were used in PCR reactions with 2 M vector-specific primers (Ad5 primer A 5'-GGGCCGCGGGGACTTTGACCG-3' and reverse primer B 5'-GTGGCCAGGGCCGGAGGTGC-3'). PCR was performed in a total volume of 12.5 L containing 200 M of each of the four deoxynucleotide triphosphates (Promega, Madison, WI), 1.5 mM MgCl₂ and 0.5 U of Taq polymerase. Samples were amplified after denaturation at 94 °C for 4minutes, followed by 35-40 cycles of 94°C for 1 minute, 65°C for 1 minute, and 72°C for 1 minute. A final extension at 72°C was performed for 10 minutes. After amplification, the entire volume of sample was electrophoresed through a 2% agarose gel. The gels were subsequently stained with ethidium bromide and viewed with an ultraviolet transilluminator.

PCR on DNA from tissue samples

DNA was isolated from tissue samples using phenol:chloroform purification methods and the above primers used to detect the presence of adenoviral vector. PCR was performed under the following conditions: a total volume of 50 L containing 500 ng of DNA, 1.5 mM MgCl₂, and 2 M of each primer. The PCR reaction was performed at 94°C for 4 minutes and subsequently for 30 cycles at 94°C for 1minute, 70°C for 1 minute, and 72°C for 1 minute, followed by 72°C for 10 minutes in a Biometra TRIO thermocycler (Tampa, FL). Ten microliters of the PCR reaction was run on a 2% agarose gel containing ethidium bromide and subsequently transferred to nylon membrane (Hybond-N⁺; Amersham, Arlington Heights, IL) using a modified Southern blotting procedure. The PCR product of the control plasmid AdRSVlacZ was electrophoresed, isolated, purified, and randomly labeled as the probe for all hybridizations. The blots were prehybridized at 65°C for 30minutes with Rapid Hybe buffer (Amersham) and the hybridization was performed overnight at 65°C. The blots were washed and then exposed to X-ray film overnight.

-galactosidase assay

Prostate tissue (approximately 50 mm³) was homogenized in 500 L of 0.25 M Tris, pH 8.0, buffer. After microcentrifugation at 100´g for 5 minutes at 4°C, the supernatant (cell lysate) was collected. The protein concentration of the cell lysate was determined by Coomassie Plus Protein Assay reagent (Pierce, Rockford, IL). The colorimetric -gal assay was performed using a -Gal Assay Kit (Invitrogen, San Diego, CA) according to the manufacturer's protocol. The units of -gal activity in the lysate were calculated and normalized to the amount of total protein. -gal activity was also determined in random tissue samples (left ventricle of heart, kidney, liver) from the control dog to establish baseline activity. The assay was performed for each prostate specimen (Fig 1) and the mean -gal activity of left and right injection site specimens, adjacent site specimens, and control specimens compared utilizing the Student's t test.

Results

To investigate the impact of prostate gene therapy volume injected on local as well as systemic viral distribution, two dogs underwent biopsy and injection of 0.5 mL per prostate lobe (1 mL total) and two received 1.0 mL per prostate lobe (2 mL total). These animals were designated CN1.1 and 1.2 or CN2.1 and 2.2, respectively. The control animal was designated as CNC. Rectal, sputum, and urine specimens were procured during the treatment phase as outlined above. Due to small prostate size, CN1.1 prostate specimens were not specified as apical or basilar. In addition, for animal CN2.1, the left biopsy site could not definitely be identified; therefore, no samples labeled LA1 or LB1 were obtained. Consequently, LA2, LA3, LB2, and LB3 were samples estimated to be near the left lobe injection site.

Detection of adenovirus in rectal, sputum, and urine swabs

A 293 cell-based assay to detect adenoviral replication by observation of CPE was performed on rectal, sputum, and urine specimens to investigate the time period of viral shedding after transrectal injection of adenovirus into the prostate. CPE results for animals CN2.1 and CN2.2 were determined on day 7 instead of day 10 due to the emergence of apparent bacterial contamination. The results of CPE analysis from the rectal, sputum, and urine swabs collected in the interval between injection and necropsy 72 hours later are shown in Table 1. Sputum samples rarely had evidence of CPE and never indicated the presence of AdRSVlacZ based upon PCR. Immediate CPE was observed in some urine and rectal specimens, raising concerns that the specimens were inherently toxic. Therefore, the presence of viral DNA was assayed utilizing PCR methods as described. Although 293-based CPE assays from the rectal samples were generally confirmed to be positive by PCR assays, urine specimen PCR confirmation was almost uniformly negative. In summary, although shedding of virus through saliva and urine was extremely low in all animals, rectal shedding of virus uniformly persisted up to 4 hours after treatment. In addition, the CPE analysis methodology as described did not appear to be a reliable indicator of the presence of viral particles.

Adenoviral vector systemic distribution

The systemic distribution of adenovirus after transrectal ultrasound-guided prostate biopsy and injection has not been previously demonstrated. Figure 2A-E shows the Southern blots for CNC, CN1.1, CN1.2, CN2.1, and CN2.2, respectively. All tissues obtained from the control canine had no detectable adenoviral vector. In contrast, the bladder, rectum, mesenteric nodes, and heart tissue revealed adenoviral vector by PCR in each experiment. Also, PCR bands were detected in one or more other genitourinary tissues (vas deferens, testicle, or urethra) in each animal injected with adenovirus. Adenoviral dissemination to other distant organs such as lung, kidney, liver, or spleen was absent in CN1.1 and CN2.1 but was noted in CN1.2 and CN2.2. Periprostatic tissue and pelvic nodes were positive in two of four animals. No adenoviral PCR products were detected for brain tissue in any animal. Prostate tissue samples had adenoviral PCR bands of varying intensity. In general, as expected, samples procured from the injection sites qualitatively demonstrated increased viral DNA compared to samples from adjacent tissues which had little or no adenoviral vector PCR products.

LacZ gene expression in prostate

More quantitatively, -gal activity (U/mg protein) as a marker for the presence of adenovirus expressing the lacZ gene was determined for each animal prostate as shown in Table 2. The average activity for samples taken at the injection sites or adjacent to the injection sites for each dog is shown. The lowest activity was noted in the control dog (1.00±0.06 U/mg protein at injection sites and 0.90±0.10 U/mg protein in adjacent tissue). The highest activity was measured in the experimental animals in prostate tissue at the injection site. -gal activity in tissue adjacent to the injection sites was intermediate between control levels and levels obtained at the actual injection site. Except for adjacent tissue specimens of CN1.1, all treated animals demonstrated -gal activity significantly greater than control tissues (Table 2). Expression levels were not significantly higher in animals receiving a 2 mL total injection compared to 1 mL injection volume. -gal activities measured in heart, kidney, and liver samples from CNC were 0.034, 0.106, and 0.285 U/mg protein, respectively, well below endogenous levels of the prostate. Prostate cells have been shown to express low levels of endogenous -gal activity¹⁷ and therefore the control prostate specimens demonstrated an activity level intermediate between prostates injected with the adenovirus and other control tissues as expected.

Discussion

The concept of gene therapy, which was once only a futuristic possibility, has now reached the stage of clinical investigation. Numerous Phase I and Phase I/II trials have been initiated or completed.^{20,21,22,23,24,25,26,27} These studies have demonstrated that prostatic gene therapy appears to be safe and quite well tolerated. However, preclinical studies to investigate the biodistribution of adenovirus throughout the prostate, to contiguous genitourinary organs and systemically to distant sites after transrectal biopsy with injection of variable volumes, have been limited. In addition, the optimal volume of injection, the optimal number of injections, and the ability of transrectal injection to reach periprostatic tissue and therefore treat locally advanced disease have not been determined.

Similar to previous experiments,¹⁷ these studies continue to explore questions of optimal local delivery to the prostate gland. Lu et al compared intravenous, intraarterial, and direct intraprostatic injections of adenovirus. The current studies utilized a technique of concurrent biopsy followed by prostatic injection as conducted in previous clinical trials.^21,22 Table 3 delineates the advantages and disadvantages of the various delivery approaches utilized by Lu et al and the current study. The technique of concurrent biopsy and prostatic injection provides prostate biopsy tissue for potential monitoring of response and for other analyses during treatment cycles. In addition, local viral distribution may be enhanced through perfusion along the open arterial, venous, and lymphatic channels created by the biopsy needle compared to simple injection. Alternatively, however, the volume injected into the biopsy space might not permeate the tissue effectively, but simply regress into the rectum. Similar to previous direct prostatic injection of 1.0 mL containing 4.8´10⁹ pfu of AdRSVlacZ, these studies demonstrated a maximum of approximately 5 U/mg protein of -gal activity along the injection tract. This suggests that this technique does not significantly diminish transfer efficiency to the tissue directly surrounding the injections site and in addition provides tissue for analysis.

The local prostate viral distribution analysis by PCR and Southern blot demonstrated a gradient of adenovirus distribution through the prostate tissue with intense bands noted from tissue at the direct injection sites with generally less intense band intensity from tissue samples from the adjacent sites. Analysis of prostatic adenoviral distribution was more objectively quantified with the -gal assay. Consistent with previous experience, the prostate demonstrated inherent -gal activity which, in the present study, was evidenced by the low -gal activity in the control animal tissue.¹⁷ This inherent activity of approximately 1 U/mg protein is relatively significant compared to -gal baseline levels of organs from the control dog ranging from 0.034 to 0.285 U/mg protein for heart and liver, respectively. The activity observed in the experimental animals was 3.5- to 5-fold greater at the injection sites and 1.5- to 2.5-fold greater in the adjacent prostate tissue compared to that measured in the control animal prostate. As expected, the highest levels of -gal activity were noted at the injection sites. Similarly, high levels of activity were observed at both the apical and basilar aspects of each injection site, suggesting equal distribution along the injection tract.

-gal activity measured in prostate tissue adjacent to the injection site was intermediate between baseline prostate levels and actual injection site levels relatively independent of the 0.5- or 1.0-mL volume of viral diluent injected. The nature of this data suggests that injection of even a relatively low volume results in moderate diffusion and perhaps even perfusion of adenoviral vector through the prostate tissue, exposing a zone of prostate tissue surrounding the injection site. Further studies are necessary to determine if the biopsy with concurrent injection method provides improved distribution of virus compared to simple injection. Future studies are also necessary to define the actual size of these distribution areas in order to estimate the number and proximity of injections required to expose the entire prostate to injected gene therapy. The presence of adenovirus and -gal activity in the tissue samples adjacent to the injection site (e.g., LA2 and LA3) suggests that therapeutic viral activity reaches a volume of prostate up to 1.5 cm in diameter. Although quantitative PCR was not performed, the band intensity on the Southern blots qualitatively complemented the quantitative -gal activity measurements. Therefore, it is conceivable that a course of six to eight injections could adequately treat the average human prostate volume. Vector delivery and expression may be optimized, further utilizing various matrices such as a gelatin sponge matrix as recently described.²⁸

All animals injected with adenovirus showed evidence of regional spread to genitourinary organs. The involvement of the bladder and urethra was anticipated due to their anatomic relationship with the prostate. Viral detection in the vas deferens and testicle may have resulted from direct reflux of adenoviral media during injection or during subsequent urination and may be a relative obstacle of this gene delivery approach. One possible solution to prevent direct viral spread to the epididymis and testicle would be to perform a vasectomy before gene delivery. Fortunately, recent evidence suggests that germ line incorporation and transmission of viral vectors would be extremely limited, if at all possible.^29,30 The consistent detection of adenoviral vector in rectal tissue samples and rectal swabs suggests that significant retrograde efflux through the biopsy tract occurred as would generally be anticipated. In addition, adenoviral vector uptake by the lymphatic system was also observed as PCR bands were present in nodal tissue draining the prostate and rectum. The inconsistent detection of periprostatic bands noted in two of four dogs suggests that extraprostatic distribution of virus may be quite good, but is variable. This variability may possibly be attributed to the fact that the canine prostate is not as fixed to the pubis via puboprostatic ligaments as in humans and is therefore very mobile. This increased mobility occasionally necessitates repositioning of the biopsy needle, which may result in variable distribution or retrograde efflux of gene therapy from the tract into the periprostatic space. The availability of a biplanar ultrasound probe may result in a more consistent biopsy/injection technique in the clinical setting although the cost of this equipment for the experimental setting may be prohibitive.

In addition to local distribution, subsequent regional and systemic biodistribution was also addressed. Local spread of virus through prostate tissue, to adjacent genitourinary organs and to contiguous pelvic organs, was demonstrated by utilizing PCR and Southern blotting. A mild degree of systemic dissemination was observed in all dogs injected with adenoviral vector. Significant systemic dissemination was observed in two of four experimental dogs (CN1.2 injected with 1 mL and CN2.2 injected with 2 mL). The heart (left and right ventricles), lung, liver, and kidney (organs that receive high cardiac output) contained detectable adenoviral vector and the pattern of organ involvement was consistent with other reports of adenoviral vector after intravenous injection.^8,18,31 It is likely that, due to the prostate's rich vascular supply, vascular dissemination of the vector is responsible for gene transfer to these distant organs that receive a high percentage of cardiac output. Doubling the relatively low volume of viral diluent injected did not consistently improve the degree of local distribution or result in greater systemic dissemination, suggesting that variables other than the volume injected such as needle placement were of more importance. The increased volume may simply have regressed into the rectum. Finally, systemic distribution of adenovirus even at low injected volumes may be greater than previously appreciated due to vascular or lymphatic spread dissemination.

In conclusion, transrectal ultrasound-guided intraprostatic injection resulted in a wide intraprostatic distribution of adenoviral vector activity utilizing either 0.5 or 1.0 mL of volume at each injection site. Local and systemic viral dissemination was observed at both volumes, the clinical significance of which is unclear at this time. In the event that shedding is ultimately clinically significant, further development of prostate-specific promoter-driven expression of the therapeutic gene may be utilized to limit possible local and systemic toxicity. Although further optimization experiments with a larger number of animals are warranted, these studies suggest that an ultrasound-guided transrectal treatment regimen including six to eight biopsy with injections of 0.5-1.0 mL of gene therapy containing 4.8´10⁹ pfu may result in sufficient gene transfer to induce gene expression throughout the human prostate.

1 Greenlee RT, Hill-Harmon MB, Murray T et al. Cancer statistics. CA Cancer J Clin 2001; 51: 15-36. MEDLINE

2 Kleinerman DI, Zhang WW, Lin SH et al. Application of a tumor suppressor (C-CAM1)-expressing recombinant adenovirus in androgen-independent human prostate cancer therapy: a preclinical study. Cancer Res 1995; 55: 2831-2836. MEDLINE

3 Yang C, Cirielli C, Capogrossi MC et al. Adenovirus-mediated wild-type p53 expression induces apoptosis and suppresses tumorigenesis of prostatic tumor cells. Cancer Res 1995; 55: 4210-4213. MEDLINE

4 Shariat SF, Desai S, Song W et al. Adenovirus-mediated transfer of inducible caspases: a novel "death switch" gene therapeutic approach to prostate cancer. Cancer Res 2001; 61: 2562-2571. MEDLINE

5 Rodriguez R, Schuur ER, Lim HY et al. Prostate-attenuated replication-competent adenovirus (ARCA) CN706: a selective cytotoxic for prostate-specific antigen-positive prostate cancer cells. Cancer Res 1997; 57: 2559-2563. MEDLINE

6 Hall SJ, Mutchnik SE, Chen SH et al. Adenovirus-mediated herpes simplex virus thymidine kinase gene and ganciclovir therapy leads to systemic activity against spontaneous and induced metastasis in an orthotopic mouse model of prostate cancer. Int J Cancer 1997; 70: 183-187. Article MEDLINE

7 Siegfried W. Perspectives in gene therapy with recombinant adenoviruses. Exp Clin Endocrinol 1993; 101: 7-11. MEDLINE

8 Wood M, Perrotte P, Onishi E et al. Biodistribution of an adenoviral vector after intravesical administration in a mouse model. Cancer Gene Ther 1997; 4: S34.

9 Culver KW. Methods for gene transfer and repair. In: Culver KW, editor. Gene Therapy: A Primer for Physicians 2nd ed Larchmont, NY: Mary Ann Lievert, 1996, 19-37.

10 Stratford-Perricaudet LD, Makeh I, Perricaudet M et al. Widespread long-term gene transfer to mouse skeletal muscles and heart. J Clin Invest 1992; 90: 626-630. MEDLINE

11 Steiner MS, Gingrich JR. Gene therapy for prostate cancer: where are we now? J Urol 2000; 164: 1121-1136. MEDLINE

12 Mitchell MF, Hamada K, Sastry KJ et al. Transgene expression in the Rhesus cervix mediated by an adenovirus expressing beta-galactosidase. Am J Obstet Gynecol 1996; 174: 1094-1101. MEDLINE

13 Hashimoto M, Aruga J, Hosoya Y et al. A neural cell-type-specific expression system using recombinant adenovirus vectors. Hum Gene Ther 1996; 7: 149-158. MEDLINE

14 Quantin B, Perricaudet LD, Tajbakhsh S et al. Adenovirus as an expression vector in muscle cells in vivo. Proc Natl Acad Sci USA 1992; 89: 2581-2584. MEDLINE

15 Arbuthnot PB, Bralet MP, Le Jossic C et al. In vitro and in vivo hepatoma cell-specific expression of a gene transferred with an adenoviral vector. Hum Gene Ther 1996; 7: 1503-1514. MEDLINE

16 Byrnes AP, Rusby JE, Wood MJ et al. Adenovirus gene transfer causes inflammation in the brain. Neuroscience 1995; 66: 1015-1024. Article MEDLINE

17 Lu Y, Carraher J, Zhang Y et al. Delivery of adenoviral vectors to the prostate for gene therapy. Cancer Gene Ther 1999; 6: 64-72. MEDLINE

18 Ko S-C, Gotoh A, Thalmann GN et al. Molecular therapy with recombinant p53 adenovirus in an androgen-independent, metastatic human prostate cancer model. Hum Gene Ther 1996; 7: 1683-1691. MEDLINE

19 Nyberg-Hoffman C, Shabram P, Li W et al. Sensitivity and reproducibility in adenoviral infectious titer determination. Nat Med 1997; 3: 808-811. MEDLINE

20 Herman JR, Adler HL, Aguilar-Cordova E et al. In situ gene therapy for adenocarcinoma of the prostate: a phase I clinical trial. Hum Gene Ther 1999; 10: 1239-1249. Article MEDLINE

21 Logothetis CJ, Hossan E, Pettaway CA et al. AD-p53 intraprostatic gene therapy preceding radical prostatectomy (RP): an in vivo model for targeted therapy development. Paper presented at AUA 94th Annual Meeting, Dallas, TX. 1999.

22 Steiner MS, Lerner J, Greenberger M et al. Clinical phase I gene therapy trial using BRCA1 retrovirus is safe. J Urol 1998; 159: 133. MEDLINE

23 Patel B, Naitoh J, Stiles A et al. Effect of interleukin-2 (IL-2) gene therapy on prostate-specific antigen in patients undergoing neoadjuvant or adjuvant gene therapy for prostate cancer. Paper presented at AUA 94th Annual Meeting, Dallas, TX. 1999.

24 Simons JW, Jaffee EM, Weber CE et al. Bioactivity of autologous irradiated renal cell carcinoma vaccines generated by ex vivo granulocyte-macrophage colony-stimulating factor gene transfer. Cancer Res 1997; 57: 1537-1546. MEDLINE

25 Sanda MG, Smith DC, Charles LG et al. Recombinant vaccinia-PSA (PROSTVAC) can induce a prostate-specific immune response in androgen-modulated human prostate cancer. Urology 1999; 53: 260-266. Article MEDLINE

26 Chen AP, Bastian A, Dahut W et al. A phase I study of recombinant vaccinia virus that expresses prostate-specific antigen (PSA) in adult patients with adenocarcinoma of the prostate. Paper presented at Proc Am Soc Clin Oncol, Los Angeles. 1998.

27 Eder JJP, Kantoff PW et al. A phase I trial of recombinant vaccinia virus, PROSTVAC that expresses prostate-specific antigen (rV-PSA) as a vaccine in men with advanced prostate cancer. Paper presented at Proc Am Soc Clin Oncol, Los Angeles. 1998.

28 Siemens DR, Austin JC, Hedican SP et al. Viral vector delivery in solid-state vehicles: gene expression in a murine prostate cancer model. J Natl Cancer Inst 2000; 92: 403-412. MEDLINE

29 Hall SJ, Bar-Chama N, Ta S et al. Direct exposure of mouse spermatogenic cells to high doses of adenovirus gene therapy vector does not result in germ cell transduction. Hum Gene Ther 2000; 11: 1705-1712. Article MEDLINE

30 Paielli DL, Wing MS, Rogulski KR et al. Evaluation of the biodistribution, persistence, toxicity, and potential of germ-line transmission of a replication-competent human adenovirus following intraprostatic administration in the mouse. Mol Ther 2000; 1: 263-274. Article MEDLINE

31 Herz J, Gerard RD. Adenovirus-mediated transfer of low density lipoprotein receptor gene acutely accelerates cholesterol clearance in normal mice. Proc Natl Acad Sci USA 1993; 90: 2812-2816. MEDLINE

Figure 1 Schematic of prostate tissue acquisition and labeling. Lxx=left lobe; Rxx=right lobe; xAx=apical; xBx=basilar; xx1=at injection site; xx2 or xx3=adjacent injection site.

Figure 2 PCR Southern blot for detection of adenoviral vectors in various tissues. (a) CNC (canine control), (b) CN1.1 (canine injected with one ml viral diluent), (c) CN1.2 (1 mL of viral diluent), (d) CN2.1 (2 mL of viral diluent), (e) CN2.2 (2 mL of viral diluent).

Table 1 CPE and PCR results of rectum, sputum, and urine swab samples

Table 2 -Gal activity in control and treated prostate tissue

Table 3 Advantages and disadvantages of various gene therapy delivery approaches