Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Aug 15.
Published in final edited form as: J Natl Cancer Inst. 2007 Mar 7;99(5):376–385. doi: 10.1093/jnci/djk065

A Prostate-Specific Antigen–Activated Channel-Forming Toxin as Therapy for Prostatic Disease

Simon A Williams 1, Rosemina F Merchant 1, Elizabeth Garrett-Mayer 1, John T Isaacs 1, J Thomas Buckley 1, Samuel R Denmeade 1
PMCID: PMC4133793  NIHMSID: NIHMS401596  PMID: 17341729

Abstract

Background

Most men will develop prostatic abnormalities, such as benign prostatic hyperplasia (BPH) or prostate cancer, as they age. Prostate-specific antigen (PSA) is a serine protease that is secreted at high levels by the normal and diseased prostate. Therapies that are activated by PSA may prove effective in treating prostatic malignancies.

Methods

We modified proaerolysin (PA), the inactive precursor of a bacterial cytolytic pore-forming protein, to produce a PSA-activated protoxin (PRX302). The viability of the prostate adenocarcinoma cell lines LNCaP, PC-3, CWR22H, and DU145 and the bladder cancer cell line TSU after treatment with PA or PRX302 in the presence or absence of purified PSA was assayed. Mice carrying xenograft tumors derived from LNCaP, CWR22H, or TSU cells were treated with intratumoral injection of PA or PRX302, and tumor size was monitored. To test the safety of PRX302, we administered it into the PSA-secreting prostate glands of cynomolgus monkeys. All statistical tests were two-sided.

Results

Native PA was highly toxic in vitro but had no tumor-specific effects in vitro or in vivo. Picomolar concentrations of PRX302 led to PSA-dependent decreases in cell viability in vitro (PRX302 versus PRX302 + PSA: DU145 cells, mean viability = 78.7% versus mean = 1.6%, difference = 77.1%, 95% confidence interval [CI] = 70.6% to 86.1%; P<.001; TSU cells, mean = 100.2% versus mean = 1.4%, difference = 98.8%, 95% CI = 96.4% to 104.0%; P<.001). Single intratumoral injections of PRX302 produced substantial and often complete regression of PSA-secreting human prostate cancer xenografts (5 µg dose, complete regression in 6 of 26 mice bearing LNCaP or CWR22H xenografts [23%]; 10 µg dose, complete regression in 10 of 26 mice [38.5%]) but not PSA-null bladder cancer xenografts. The prostates of cynomolgus monkeys injected with a single dose of PRX302 displayed extensive but organ-confined damage, with no toxicity to neighboring organs or general morbidity.

Conclusions

Our observations demonstrate the potential safe and effective intraprostatic application of this engineered protoxin.

The prostate gland is often associated with morbidity in the aging male. Approximately 80% of men will have a symptomatic benign overgrowth of the prostate known as benign prostatic hyperplasia (BPH) by age 80 years (1). In addition, the prostate is the most common site of nonskin cancer diagnosed in American men; one in six men develop prostate cancer (2). For men with prostate cancer that is localized within the gland, surgery (3, 4) and radiation therapy (5) are associated with 10-year disease-free survival rates of 40%–90%, depending on the stage and grade of the cancer at diagnosis. Among patients who are treated with radiation therapy, 10%–50% have a local recurrence as the first sign of treatment failure (6). These individuals are more likely to subsequently develop metastatic disease than those with no local recurrence. To effectively treat BPH and/or eliminate locally recurrent cancer within the prostate gland, novel therapies are needed. Optimally, these therapies should not be associated with substantial local or systemic toxicities.

The prostate gland produces unique tissue-specific proteases that include prostate-specific membrane antigen (7) and members of the kallikrein family of serine proteases (8), including prostate-specific antigen (PSA) (9). These proteases are produced by epithelial cells within BPH tissue and by prostate cancer cells. Serum PSA levels are routinely measured in the initial diagnosis of prostate cancer and are also used to follow disease progression and response to therapy (10). PSA is active in normal seminal plasma and in the extracellular fluid surrounding prostate cancers. However, any PSA that leaks into the blood circulation is completely inactivated through the formation of covalent complexes with the abundant serum protease inhibitors α-1-antichymotrypsin and α-2-macroglobulin (11).

The prostate-confined enzymatic activity of PSA has made it a logical target for the development of prostate tissue–specific therapies (12). Previously, we identified a peptide sequence that is efficiently and selectively cleaved by PSA (13). We have coupled this peptide to the cytotoxic agents, doxorubicin and thapsigargin, to generate prodrugs that are activated by PSA within the peritumoral fluid but are not activated within the circulation (12, 14).

Proaerolysin (PA) is a 53-kDa protein secreted by the aquatic Gram-negative pathogen, Aeromonas hydrophila (15, 16). It exists as a water-soluble dimer that binds to glycophosphatidylinositol-anchored proteins found on the surface of most mammalian cells. PA contains a carboxy-terminal inhibitory domain that is cleaved by ubiquitous membrane-bound proteases, such as the furins, producing aerolysin, which rapidly oligomerizes and enters the plasma membrane to form highly stable pores that cause rapid cell death (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Proaerolysin (PA): mechanism of action. Diagram showing the mechanism of receptor-mediated native PA and prostate-specific antigen (PSA)–activated PA (PRX302) toxicity. Native PA, secreted by the Gram-negative bacteria, Aeromonas hydrophila, dimerizes before binding to glycophosphatidylinositol (GPI)-anchored proteins on the target cell surface. Activation of the protein by the cleavage of the C-terminal inhibitory domain by the furin family of proteases allows the formation of heptameric pores that facilitate cell death. PRX302, once activated by PSA cleavage, causes cell death via a mechanism similar to that of its native counterpart.

In this study, we directed PA’s potent cytotoxicity to prostate cell lines and tissues with a dual-targeting strategy. In this approach, targeting was achieved by direct intraprostatic injection of a modified PA toxin that requires the enzymatic activity of PSA for activation.

Materials and Methods

Design and Production of the Prostate-Specific Antigen–Activated Proaerolysin Variant

The furin recognition sequence of the native PA gene (17) (KVRRAR) was subjected to site-directed mutagenesis using polymerase chain reaction to convert it to a PSA-cleavable sequence (13) (HSSKLQ). The mutated gene was then subcloned into the pMMB66HE vector (18) for amplification in Escherichia coli. This construct was then transferred to a protease-deficient strain of A. salmonicida (19) that facilitated the production of large amounts of uncontaminated PRX302. PRX302 was further purified by hydroxyapatite chromatography followed by ion-exchange chromatography and stored at −20 °C (19).

Cell Culture and Viability Assay

The human prostate cancer cell lines LNCaP, PC-3, DU145, CWR22H and the bladder cancer cell line TSU (American Type Culture Collection, Manassas, VA) were grown as monolayers in RPMI-1640 medium supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, and 2 mM l-glutamine (Invitrogen, Carlsbad, CA). Cells were maintained at 37 °C in an atmosphere of air with 5% CO2.

For viability assays, cells were seeded into 96-well plates at 20 000 (LNCaP) or 5000 (PC-3, DU145, and TSU) cells per well and cultured for 2 days before treatment. The medium was then rep laced with serum-free RPMI-1640 medium containing increasing concentrations of PA (19 pM–10 nM) or PRX302 (19 pM–10 nM) supplemented with the synthetic androgen methyltrienolone (R1881, PerkinElmer, Boston, MA) at 10 nM, which was added to sustain endogenous PSA production in the absence of serum. Exogenous PSA (EMD Biosciences, San Diego, CA), when used, was added at 5 µg/mL. Cell viability was assayed after 48 hours of treatment using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT Dye Solution, Promega, Madison, WI) according to the manufacturer’s instructions. Samples were measured at 570 nm using a spectrophotometer (Spectramax Plus, Molecular Devices, Sunnyvale, CA), and readings of treated cells were normalized to those of untreated cells that were also maintained in serum-free conditions. The IC50 was determined as the concentration at which a 50% loss of viability occurred relative to untreated cells. For each cell line, the experiment was performed three times in replicates of eight.

Tumor Xenografts

Mouse care and treatment was approved by and performed in accordance with the guidelines of the Animal Care and Use Committee of The Johns Hopkins University School of Medicine. LNCaP and TSU cells maintained under standard conditions were detached by treatment with 0.25% trypsin–EDTA solution and washed in Hank’s balanced salt solution (HBSS). Cells were then suspended in a 60% mixture of Matrigel Matrix (BD Biosciences, San Jose, CA) in HBSS at a concentration of 5.0 × 106 (LNCaP) or 1 × 106 (TSU) cells per 200 µL of solution. LNCaP and TSU cell suspensions (200 µL) were then injected into the subcutis overlying the rear flanks of 6-week-old male nude mice (LNCaP, n = 21; TSU, n = 30). The CWR22H xenograft is maintained in the subcutis of male nude mice and is not passaged in tissue culture (20). For expansion to experimental conditions, 1 g of CWR22H tumor tissue was surgically removed from maintenance mice and finely minced with scalpels. Twenty micrograms of tumor tissue was suspended in 200 µL Matrigel Matrix and injected into the flank of each of 15 male nude mice (20). Tumors were allowed to grow for 1 week (TSU) or 2 weeks (LNCaP and CWR22H) so that they were palpable before making an initial tumor measurement with calipers. Tumor volume (in cm3) was calculated by the formula 0.5236 × length × width × height (14). Tumor-bearing mice were assigned to one of four groups (n = 5–10 per group) on the basis of initial tumor volumes. One week after assignment, mice were given either single intratumoral injections of PRX302 (0, 1, 5, 10 µg) or multiple intratumoral injections of PA (50 ng in 100 mL of saline), or saline alone, into the center of the tumor with a 26-gauge needle. Tumors were measured on treatment days or twice weekly for as long as 28 days, at which point untreated tumor-bearing mice had become moribund. The mice were killed by CO2 overdose 24 hours after the last treatment. Blood was collected by cardiac puncture and used to determine plasma PSA level. The tumors were surgically removed, weighed, fixed in a 10% formalin solution in phosphate-buffered saline (PBS, Invitrogen), and embedded in paraffin.

In Vivo Analysis of Tumor Cell Death

Tumor sections on slides were deparaffinized and washed in PBS. DNA fragmentation was visualized by staining tissue sections with a commercial kit (DeadEnd fluorometric terminal deoxynucleotidyltransferase-mediated UTP end-labeling [TUNEL] system, Promega) according to the manufacturer’s instructions. Briefly, tissues were incubated in equilibration buffer for 10 minutes before the addition of the reaction cocktail containing terminal dNTP transferase and fluorescein isothiocyanate–conjugated nucleotides. Slides were incubated for 1 hour at 37 °C, and reactions were terminated by incubation in 2× standard saline citrate for 15 minutes. Tissues were washed in PBS and counterstained with a 1 µg/mL solution of 4′,6-diamidino-2-phenylindole (Invitrogen) in PBS for 10 minutes. After three washes in PBS, slides were mounted with Prolong (Invitrogen) onto coverslips. Images were captured with an Axioplan 2 microscope and a Hamamatsu color chilled 3CCD camera using Optimus software (Bioscan, Silver Spring, MD).

125I-PRX302 Biodistribution

In an institutionally approved toxin biodistribution study conducted by ClinTrials BioResearch (Senneville, Quebec), three male Sprague-Dawley rats were anesthetized with isoflurane. An abdominal surgical incision was made to visualize the prostate, and a single 8.7-µg dose (i.e., 5.9 µCi/rat) of 125I-labelled PRX302 was injected into the right ventral lobe in a volume of 10 µL. For whole-body tissue samples, rats were killed by CO2 overdose and frozen whole in a hexane/dry ice slurry for 20 minutes. The right sides of the rats were embedded in 2% carboxymethylcellulose, and 30 µm tissue slices were prepared. For imaging, sections were exposed to imaging plates, which were then read with a scanner and an image reader.

Red Blood Cell Lysis Assay

A solution of red blood cells (RBCs) (Sigma-Aldrich, St Louis, Mo) was mixed with saline or 50% human plasma (Sigma-Aldrich) to a final concentration of 2%. PRX302 at 10 nM ± PSA at 10 µg/mL was added to the mixture. After 1 hour, the remaining intact cells were collected by centrifugation at room temperature for 5 minutes at 100g, and the absorbance of the supernatant was measured at 540 nm. Absorbance was compared with that of a 100% lysis control, in which RBCs were treated with 1% Triton X-100 (Sigma-Aldrich). A single experiment was performed in triplicate.

Intraprostatic Injection of PRX302 Into Cynomolgus Monkeys

In this study, conducted by Charles River Laboratories (Sparks, NV), sexually mature male cynomolgus monkeys (two per condition) were anesthetized, and a small perianal incision was made under sterile conditions to expose the prostate. Monkeys were then injected with varying doses of toxin (0.35, 4.1, and 25.8 µg/g of prostate tissue) in 25 µL of PBS–EDTA vehicle. The total dose was individualized for each monkey based on the weight of its prostate, which was estimated from a nomogram constructed from control monkeys in which body weight was plotted against prostate weight. The final dose delivered was determined through analysis of dosing solutions within the syringe after injection. The monkeys were closely observed following surgery. They were then killed by deep anesthesia and exsanguination 2, 3, and 10 days later, when prostate glands were removed, imaged, and prepared for histologic assessment. Areas of damage were quantified from hematoxylin-eosin (H & E)–stained sections that were subjected to analysis with MetaMorph imaging software (Molecular Devices). The beagle is used as a model for toxicity based on the relative size and anatomic similarity of the dog prostate to the human prostate.

Conditioning of Media From Prostatic Tissue

Pieces of fresh tissue from beagle, cynomolgus monkey, and human prostates (The Johns Hopkins Medical Institute Prostate Tissue Archives) were weighed and incubated in an equal volume of RPMI-1640 in a 15-mL conical tube for 2 hours at room temperature. The conditioned media was separated from the tissue by centrifugation (21 000g) and transferred to another tube. Samples of the conditioned media were subjected to analyses of protein concentration and PSA levels. For the hemolysis assay, the samples from the PSA-secreting tissue were adjusted with RPMI-1640 to a final concentration of 1 µg/mL of PSA or 500 µg/mL of total protein for the dog-conditioned media (an amount comparable to protein levels in all samples). Purified PSA was used at a concentration of 1 µg/mL as the positive control. Two independent experiments were performed, each in triplicate.

Hemotoxylin–Eosin Tissue Staining

H & E staining was performed on sections from xenograft tumors, human prostate, and cynomolgus monkey prostate using standard protocols.

PSA and PRX302 Immunohistochemistry

PSA and PRX302 staining were performed on the treated cynomolgus monkey prostates using similar protocols. Briefly, slides were deparaffinized by melting followed by rehydration as described above. Slides were placed in one change of distilled water with 0.1% Tween 20 and then placed into preheated Target Retrieval Solution (Dako, Glostrup, Denmark). Slides were incubated in this solution for 40 minutes in a steamer (Handy Steamer, Black and Decker, Hunt Valley, MD). Slides were cooled at room temperature for 15 minutes and then rinsed twice in PBS with 0.1% Tween (PBST) for 5 minutes. Samples were stained using the EnVision staining kit for rabbit-derived antibodies (Dako) according to the manufacturer’s protocol. Endogenous peroxidase activity was blocked with 3% H2O2 (provided in the kit) for 5 minutes. Slides were rinsed in PBST. For PSA staining, polyclonal rabbit anti-human PSA (Dako) was used at a dilution of 1 : 4000. For PRX302, polyclonal rabbit anti-PA (Protox Therapeutics, Inc) was used at a dilution of 1 : 1000. Antibodies were diluted using ChemMate diluent (Ventana Medical Systems, Tucson, AZ). Slides were placed in diluted antibody solution for 1 hour at room temperature. The slides were then rinsed with Tris-Buffered Saline with 0.1% Tween (TBST). Slides were then placed in solution of anti-rabbit peroxidase-conjugated secondary antibody bound to polymer from the EnVision kit for 30 minutes and then rinsed twice for 5 minutes with TBST. Diaminobenzidine colorimetric reagent solution from the EnVision kit was applied to slides for 20 minutes, and then the slides were rinsed well in distilled water. Slides were counterstained with hematoxylin (Tombal #15 2, Shandon Inc., Pittsburgh, PA) for 30 seconds, rinsed well with running tap water for 2 minutes, and then placed in Shandon bluing solution for 1 minute. The slides were again rinsed well in tap water and then dehydrated in an ethanol series followed by xylene. Slides were then overlaid with a coverslip and then evaluated under bright-field illumination to assess PSA and PRX302 distribution.

Statistical Analysis

To test for differences in tumor growth in xenograft tumor–bearing mice treated with different doses of PRX302 or saline, random effects linear regression models were fit. To achieve assumptions of linearity, log (tumor volume + 0.05) was used as the outcome variable. Flexible quadratic models were fit including main effects of time (days), time squared, and interactions between conditions (saline, 1 µg dose, and 5 µg dose) and the time variables. For tumors derived from TSU cells, the relationship between time and outcome appeared to be linear based on Wald tests, so quadratic terms were dropped. For tumors derived from LNaCP cells, there was strong evidence that the quadratic terms were important for model fit, and so, they were retained in the model. Fitted regression lines were obtained. Differences between conditions were tested by fitting models to data for two of the conditions. For each pair of conditions, a null model (excluding treatment terms from the model) and a full model (including treatment terms in the model) were estimated and compared using Akaike (AIC) or Bayesian (BIC) information criteria statistics. The conditions in each pair were deemed different if the AIC and/or BIC statistic was smaller in the full model. Fit of regression models was checked using scatter plots and examination of residual plots. For in vitro studies of viability, P values were derived from the Student’s t test. All statistical tests were two-sided, and P value less than .05 was considered to be statistically significant.

Results

Native Aerolysin Antitumor Activity

To confirm that PA can induce death in cancer cells, we treated a panel of cell lines with various doses of the protoxin (LNCaP, PSA positive; PC-3, DU145, and TSU, PSA negative). After 48 hours, cell viability was inhibited by picomolar concentrations of the toxin, independent of cell type or PSA production status (Fig. 2, A). In vivo toxicity studies in mice were performed to identify a nonlethal dose of PA that could be used in antitumor efficacy studies (data not shown). Based on these studies, mice were injected intratumorally with a dose of 50 ng of PA in 100 µL of saline at 72-hour intervals (Fig. 2, B and C). No differences in tumor growth rates were observed between PA-treated and vehicle-treated tumors of either cell type. These results document the lack of therapeutic index with PA because effective antitumor doses of PA could not be delivered intratumorally without death of the host.

Fig. 2.

Fig. 2

Open in a new tab

Effect of native proaerolysin (PA) on cancer cell viability. A) In vitro toxicity. Prostate cancer cell lines LNCaP (squares), PC-3 (diamonds), DU145 (triangles), and bladder cancer cell line TSU (circles) were treated with increasing concentrations of PA for 48 hours and subjected to the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide assay. The viability of treated cells was normalized to that of untreated cells (100%). Means and 95% confidence intervals shown for at least three experiments performed in replicates of eight. B and C) In vivo effect of native PA on prostate-specific antigen–null tumor growth. Nude mice carrying established subcutaneous TSU (B) or LNCaP (C) xenograft tumors were injected intratumorally with 50 ng of PA in 100 µL of saline (diamonds) or saline alone (squares) at 72-hour intervals. Tumor volume measurements were taken on treatment days or twice weekly. Means and 95% confidence intervals from two experiments with 10 mice each are shown.

Prostate-Specific Antigen–Activated Proaerolysin In Vitro Cell Killing of Prostate-Specific Antigen–Secreting and Nonsecreting Cells

To attempt to achieve therapeutic levels of aerolysin intratumorally without toxicity to the host, we produced a site-directed mutant of the furin activation peptide sequence KVRRAR of the PA gene, changing it to the sequence HSSKLQ, which is a preferential substrate for PSA (PRX302) (13) (Fig. 3, A). This modification greatly attenuated aerolysin activation by PSA-null cells (mean IC50 of PRX302 versus PA: PC-3, 430 versus 22 pM, difference = 408 pM, 95% confidence interval [CI] = 400 to 450 pM; DU145, 910 versus 35 pM, difference = 875 pM, 95% CI = 870 to 950 pM; and TSU, 1800 versus 66 pM, difference = 1734 pM, 95% CI = 1725 to 1875 pM) but not PSA-secreting cells (LNCaP, 155 versus 30 pM, difference = 125 pM, 95% CI = 120 to 190 pM). LNCaP cells produce far less total and active PSA than normal human prostate tissue and human prostate cancers (11); however, PRX302 still preferentially killed LNCaP cells compared with PSA-null (PC-3 and DU145) cells. Moreover, the exogenous addition of enzymatically active PSA to the serum-free media of non–PSA-producing DU145 or TSU cells was enough to activate 500 pM of the protoxin to a potency that killed these previously resistant cells (PRX302 versus PRX302 + PSA: DU145 cells, mean = 78.7% versus mean = 1.6%, difference = 77.1%, 95% CI = 70.6% to 86.1%; P<.001; TSU cells, mean = 100.2% versus mean = 1.4%, difference = 98.8%, 95% CI = 96.4% to 104.0%; P<.001; Fig. 3, C).

Fig. 3.

Fig. 3

Open in a new tab

Design of PRX302 and prostate-specific antigen (PSA)–dependent toxicity of PRX302 in vitro. A) Illustration of proaerolysin (PA) structure and modification of activation site to produce PSA-activated PRX302. B) PA and PRX302 toxicity in vitro. Cell lines were treated for 48 hours with PA or PRX302 at various doses in serum-free conditions supplemented with the synthetic androgen R1881 at 10 nM. The MTT assay was used to measure cell viability. The IC50 was defined as a 50% loss of viability compared with untreated conditions. C) PSA-dependent in vitro toxicity. PSA-null DU145 prostate cancer and TSU bladder cancer cells were treated with PRX302 at 500 pM for 48 hours in serum-free media with (gray) or without (black) 5 µg/mL of PSA. Cell viability was measured by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide MTT assay. Viability of treated cells was normalized to that of untreated cells (absence of toxin or without PSA). Means and 95% confidence intervals of three independent experiments performed in replicates of eight (n = 24) are shown.

Selective Antitumor Effect of PRX302 on Prostate-Specific Antigen–Expressing Xenografts

To monitor the effect on established tumors of a single treatment of PRX302, mice bearing xenongrafts of PSA-secreting LNCaP (21) and CWR22H (20) cells and PSA-null TSU cells (22) were treated once intratumorally with 0, 1, 5, or 10 µg of PRX302. Fifteen days after treatment, a pronounced reduction in growth of PSA-secreting tumors was observed at both the 5 and 10 µg doses (Supplementary Table 1, available online). Overall, 6 of 26 (23%) mice that were treated with 5 µg and 10 of 26 (38.5%) mice that were treated with 10 µg of PRX302 achieved complete response (i.e., no palpable tumor) 15 days after treatment. Although there was a modest reduction in the volume of TSU xenografts after treatment with the highest dose of PRX302, none of the TSU tumor–bearing mice achieved a complete response.

Substantial regression of LNCaP xenograft tumors was maintained for 28 days after a single intratumoral treatment with 5 µg of PRX302, whereas no such change was observed for TSU xenograft tumors (Fig. 4). During the 28 days after treatment, differences in the volumes of LNCaP xenograft tumors in the 5 µg of PRX302 and saline treatment groups and in the 5 and 1 µg of PRX302 treatment groups were observed (day 28 tumor volumes: saline, mean = 0.54 cm2, 95% CI = 0.53 to 0.55 cm2; 1 µg PRX302, mean = 0.52 cm2, 95% CI = 0.51 to 0.53 cm2; 5 µg PRX302, mean = 0.08 cm2, 95% CI = 0.077 to 0.082 cm2; P = .007; Fig. 4, A); however, no differences in tumor volumes in the 1 µg and saline groups were observed. For TSU xenografts (Fig. 4, B), no differences in tumor volumes were observed across treatment groups. Twenty-eight days after receiving a single 5 µg intratumoral dose of PRX302, three of seven (43%) mice in the treated group had both no measurable tumors as well as no measurable levels of PSA in the blood (i.e., complete response) (Fig. 5, A and data not shown). Average tumor weights and PSA levels declined by 90% and 95%, respectively, in the 5 µg treatment group compared with the saline control group (tumor weight: saline, mean = 0.48 g, 95% CI = 0.47 to 0.49 g; 5 µg, mean = 0.05 g, 95% CI = 0.048 to 0.052 g; blood PSA mean concentration: saline, mean = 60.5 ng/mL, 95% CI = 59.1 to 61.9 ng/mL; 5 µg, mean = 3.2 ng/mL, 95% CI = 3.02 to 3.38 ng/mL; Fig. 5, A).

Fig. 4.

Fig. 4

Open in a new tab

Prostate-specific antigen (PSA)–dependent toxicity of PRX302 in vivo. For all experiments, tumors were established subcutaneously in nude mice before treatment, which consisted of a single intratumoral injection of PRX302 (0 µg, squares; 1 µg, circles; or 5 µg, diamonds) in 25 µL of saline. Tumor volumes were measured twice weekly for 28 days. A) LNCaP tumor volumes after treatment with saline or with PRX302. B) TSU tumor volumes after treatment with saline or with PRX302. In A and B, means and 95% confidence intervals are shown from two independent experiments performed using seven mice per treatment group. C) Individual growth lines of LNCaP tumors. D) Individual growth lines of TSU tumors. In C and D, trend lines from linear regression analyses are in bold.

Fig. 5.

Fig. 5

Open in a new tab

Effects of PRX302 on death of prostate-specific antigen (PSA)–secreting tumor cells. A) Images (left), tumor weights, and blood PSA levels (right) of mice with LNCaP xenograft tumors 28 days after intratumoral injection of PRX302 (n = 7 mice per experimental condition). Means and 95% confidence intervals (CIs) of weights and PSA levels are shown. B) Immunohistochemical analysis of LNCaP xenografts. Upper panels: hematoxylin–eosin staining of LNCaP xenograft tumors 24 hours after treatment with single injection of 1 µg PRX302 in 25-µL volume or saline control. Purple staining showing nuclei, with vasculature shown in red. Lower panels: Fluorescence staining of dead cells using DeadEnd fluorometric terminal deoxynucleotidyltransferase-mediated UTP end-labeling system. Blue staining shows nuclei and green staining shows the fragmented DNA of dead cells.

To evaluate the extent of the acute effect of PRX302 on tumor cell morphology, LNCaP tumors were surgically removed 24 hours after a single treatment with 1 µg of PRX302 (Fig. 5, B). Marked changes in tissue cellular structure were observed by H & E staining (Fig. 5, B, upper panels), and the induction of extensive cell death was evident by TUNEL staining (Fig. 5, B, lower panels).

PRX302 Distribution and Activation in Plasma and Prostate Homogenate

The results from xenograft tumor models suggested the potential for intraprostatic delivery of PRX302 as treatment for proliferative disease of the prostate. To carry out an initial evaluation of the feasibility of this approach, we examined whole-body distribution of PRX302 following intraprostatic injection. For this study 125I-PRX302 was prepared and injected into the right prostatic ventral lobe of the prostates of 21 Sprague-Dawley rats (Fig. 6, A). The rats were then killed at various times, and the concentration of radioactivity was determined on a gram-equivalent basis (eq/g) for a variety of tissues, including the ventral and dorsal lobes of the prostate, urinary bladder, liver, kidney, lung, pancreas, and adrenal gland (data not shown). The thyroid gland was also evaluated for non specific uptake of liberated 125I. A radiogram of a sectioned whole rat shows the distribution of radioactivity 24 hours after 125I-PRX302 was injected into the right ventral prostate (Fig. 6, A). At this time point, approximately 65% of the total radioactivity was found in the prostate, with the majority in the right ventral lobe. The remaining activity was found in the thyroid gland. No other tissue had a substantial amount of 125I (i.e., >0.1 µg eq/g), and plasma levels remained low at all time points (data not shown).

Fig. 6.

Fig. 6

Open in a new tab

PRX302 tissue distribution and tissue-specific activity. A) Whole-body autoradiograph of 125I-PRX302–treated rat. A single 8.7-µg dose (i.e., 5.9 µCi/rat) of 125I-labelled PRX302 was injected into the right ventral lobe of the prostate gland in a volume of 10 µL. Sagittal sections were taken 24 hours after treatment. One of seven rats is shown. B) Prostate-specific antigen (PSA)–induced activation of PRX302 is inhibited by plasma. Red blood cells incubated for 90 minutes in 10 nM PRX302 in plasma and PSA remain intact but are lysed in the absence of plasma. Means and 95% confidence intervals from a single experiment performed in triplicate are shown.

Although these results were encouraging in that they indicate tissue localization, intraprostatic injection of the PRX302 toxin to the prostate could still potentially result in leakage of low levels of the toxin to the systemic circulation or into tissues adjacent to the prostate such as bladder, urethra, and rectum. To further evaluate the potential for nonspecific toxicity from PRX302 leakage into adjacent tissue, the beagle dog model was used. The dog is a commonly used model for the study of the prostate due to its organ size, which is comparable to that of humans, and incidence of spontaneous BPH, prostatic intraepithelial neoplasia, and frank adenocarcinoma (23). Like rodents, dogs do not have the PSA gene, nor do they express a protease with activity homologous to PSA. Therefore, this study was designed to assess non-specific activation of PRX302 by other proteases present in the prostate and in adjacent tissue in the event of leakage from the injection site. In this study, individual beagle dogs (n = 1 per group) were injected intraprostatically with 50, 100, 200, or 400 µg of PRX302. Dogs were killed 7 days later, and gross tissue toxicity and organ weights were assessed. There were no deaths and no treatment-related clinical signs in the treated dogs. Although some discoloration of the prostate was observed in the dog that was treated with the 400 µg dose, there was no change in prostatic weight, and no macroscopic abnormalities or weight changes were observed in any of the adjacent tissues analyzed (data not shown).

To determine whether PRX302 could be activated by plasma proteases or by PSA in the plasma should the protoxin enter the circulation, PRX302 (10 nM final concentration) was incubated in a 2% RBC solution with 50% human plasma with or without 10 000 ng/mL of enzymatically active purified human PSA. As a positive control, PRX302 was incubated with purified PSA and 2% RBC solution in serum-free buffer. After incubation of PRX302 for 1 hour in human plasma containing 10 000 ng/mL of PSA, only a small percentage (less than 5%) of RBCs were lysed (Fig. 6, B). In contrast, after incubating PRX302 with 10 000 ng/mL of PSA in serum-free buffer for 1 hour, nearly 90% of the RBCs were lysed (Fig. 6, B).

Previous studies have documented that the PSA gene is present only in humans, apes, and old-world monkeys (24, 25). To identify an appropriate model for further intraprostatic toxicologic studies, we evaluated serum from three different macaque species: rhesus, cynomolgus, and pigtail monkeys. The average blood PSA levels were 0.74 ng/mL in rhesus monkeys (n = 6), 0.86 ng/mL in cynomolgus monkeys (n = 12), and <0.01 ng/mL in pigtail monkeys (n = 12). On this basis, cynomolgus monkeys were selected for further studies. Prostate tissue glandular morphology from these monkeys appeared similar to human prostate tissue on H&E staining. Immunohistochemistry with polyclonal anti-human PSA antibody revealed a similar magnitude and distribution of PSA within the monkey prostate glandular acini (Fig. 7, A). To determine whether PSA from these cynomolgus monkey prostates was enzymatically active and capable of hydrolyzing PRX302, media was conditioned with dog (PSA-null), monkey, and human prostate homogenates and assayed for PRX302 activation by the hemolysis assay. The lack of activation of PRX302 in the PSA-null dog prostate homogenate confirmed that activation of PRX302 within prostate tissue required the presence of active PSA in the tissue (Fig. 7, B). In contrast, activation of PRX302 was observed in the PSA-positive monkey and human prostate homogenates (Fig. 7, B).

Fig. 7.

Fig. 7

Open in a new tab

Validation of the cynomolgus monkey for protoxin studies. A) Hematoxylin–eosin (left panel) and PSA (right panel) staining of human (upper panel) and cynomolgus monkey (lower panels) prostate tissues. B) Conditioned media from dog (DPCM), cymomolgus monkey (MPCM), and human (HPCM) prostates were tested for their ability to activate PRX302 in a hemolysis assay. Means and 95% confidence intervals from three independent experiments performed in triplicate are shown. RBC = red blood cells; PSA = prostate-specific antigen.

Toxicity of PRX302 in the Cynomolgus Prostate

Cynomolgus monkeys produce PSA in the prostate, and their prostate gland is anatomically and morphologically similar to the human prostate gland (Fig. 7, A). This makes these primates the best preclinical model to investigate the safety of the toxin in nondiseased conditions. Monkeys were treated with a single intraprostatic injection into each of the two prostate lobes with vehicle, or with 0.35, 4.1, or 25.8 µg/g gland weight of PRX302 in a 25 µL volume. Freshly excised prostatic tissue was imaged (Fig. 8, A, left panel) and prepared for histochemical analysis. H&E-stained tissue sections of injected prostates that were harvested 3 days after injection revealed damaged areas with complete ablation of the columnar epithelial cells, producing individual prostate acinar “ghosts” surrounded by an influx of RBCs (Fig. 8, A). The areas of damage, although extensive at the highest doses, were organ confined. Staining of prostate tissue for PRX302 revealed a distribution that overlapped the areas of toxin-induced damage (Fig. 8, A). Damaged areas were quantified using image analysis, and results indicated that a single injection of 25 µL of PRX302 could kill up to 50% of cells in the injected prostate (Fig. 8, B).

Fig. 8.

Fig. 8

Open in a new tab

PRX302 treatment of cynomolgus monkey prostate glands. Prostate glands from monkeys treated intraprostatically with saline or with 0.35 µg/g or 4.1 µg/g of PRX302 were removed for analysis at day 3. A) Gross pathology (left panels), hematoxylin–eosin (H & E) staining (center panels), and proaerolysin (PA) staining (right panels) of prostate glands of cynomolgus monkeys after PRX302 treatment. A representative image from the single set of experiments is shown. B) PRX302-induced damage to the monkey prostate is dose dependent. Mean area and percentage of cells damaged in the entire region and 95% confidence intervals are shown. C) Blood prostate-specific antigen (PSA) levels following treatment with single dose of PRX302 into each lobe of monkey prostate. Means and 95% confidence intervals from the experiment performed in triplicate are shown.

Injection of PRX302 resulted in a dose-responsive spike in serum PSA levels in treated monkeys that peaked on day 2 after injection and returned to baseline levels by day 10 (Fig. 8, C). Extensive gross and histopathologic evaluation of tissues, including those of the genitourinary system, was performed on treated monkeys (Supplementary Table 2, available online). No substantial toxicity was observed in periprostatic tissues, including the lateral pelvic fascia (data not shown). In addition, minimal toxicity was observed in tissues, including the anal sphincter, adjacent to the prostate gland compared with control prostates injected with vehicle alone. No gross injury to the urethra, urinary bladder, or rectum was observed at any dose. Histopathologic changes were observed in the prostatic urethra of both control and treated monkeys, but none appeared severe enough to have altered normal urethral function.

Discussion

The studies presented were designed to test the selectivity, efficacy, and safety of a PSA-activated PA toxin before anticipated clinical testing as intraprostatic therapy for men with locally recurrent prostate cancer. Selectivity and efficacy were demonstrated in vivo by comparing the effects of a single intratumoral injection of PRX302 on non–PSA-producing TSU human bladder cancer xenografts and on PSA-producing and -secreting LNCaP and CWR22H human prostate cancer xenografts. In these studies, minimal antitumor effect was seen in the TSU xenografts, whereas complete remission was observed in three independent experiments with two different PSA-expressing xenograft models. In addition, a single injection of PRX302 into PSA-producing monkey prostates caused extensive damage to the prostate, killing up to 50% of cells in the injected gland. In contrast, little damage was observed following an injection of high-dose PRX302 into the dog prostate, which does not produce PSA (Fig. 7, B). Finally, safety was demonstrated in the models used, including the monkey prostate. No toxicity was observed in organs adjacent to the prostate (i.e., bladder, urethra, rectum) following intraprostatic injection. Similarly, no toxicity was observed in mouse tissues following injection into PSA-producing xenografts.

This study has several limitations. One is the lack of appropriate models of PSA biology. Although there are a few PSA-secreting prostate cancer cell lines available, of which the LNCaP line is the best known, PSA-secreting isogenic clones of PSA-null cell lines do not generate enzymatically active protein (26). A homologue to PSA is not a part of the murine genome, and recent attempts at targeting the human gene to the mouse prostate in a transgenic system appear to also generate inactive protein (unpublished data), thus limiting our ability to test our agent orthotopically against a PSA-secreting model of prostate cancer.

The PSA-activated PA strategy has a number of inherent advantages as a therapeutic strategy for locally recurrent prostate cancer and BPH. First, it involves the administration of a highly potent agent whose potential nonspecific toxicity is limited by two targeting methodologies: direct injection into the target tissue and selective activation by PSA produced only by the target tissue. Studies with 125I-labeled protoxin injected into the rat prostate clearly document the advantage of direct tissue injection because almost no attributable accumulation of the protoxin was detectable outside of the prostate gland. In addition, data from dog and monkey models showed that PSA was required to activate the protoxin within the prostate, and data from the monkey model demonstrated no toxicity in adjacent non–PSA-producing tissue. PRX302 was also not activated in plasma containing excessively high concentrations of PSA. A final advantage of the strategy is that the PRX302 protoxin is activated by PSA, a secreted protease that is present in high amounts in the prostatic ductal network and in the peritumoral fluid. Because the toxin is activated in the extracellular milieu, not every cell needs to make PSA to be killed by the PRX302 protoxin due to a potent local “bystander” effect, in which cancer cells, stromal cells, and endothelial cells that do not make PSA can also be killed by PSA-activated toxin.

The development of novel therapies to treat prostate cancer could be of substantial benefit to large numbers of men. Current therapies for these local recurrences include salvage prostatectomy (27) and cryosurgery (28), techniques that are associated with substantial morbidity, including increased rates of incontinence and impotence. The potential advantage of the PRX302 toxin approach over these therapies relates to the requirement for PSA-specific activation of the toxin as this limits nonspecific injury to normal structures associated with the prostate, such as the urethra and neurovascular bundles. On this basis, an investigational new drug application was recently approved by the Food and Drug Administration for testing this PRX302 toxin as intraprostatic treatment for locally recurrent prostate cancer following definitive radiation therapy. A phase I trial is now underway to assess toxicity of PRX302 administered as multiple injections using a modified brachytherapy template in patients with rising PSA levels after prior definitive radiation therapy.

A second potential application of the PRX302 toxin is in the treatment of BPH, a condition that is clinically evident in 50% of men by age 50 years and in 80% by age 80 years. Currently, surgical treatment by transurethral resection of the prostate is the standard treatment for patients with symptomatic BPH (renal insufficiency or recurrent bladder infections) failing pharmacologic therapy with α-blockers or 5-α-reductase inhibitors (29). This treatment produces improvement of symptoms in 70%–85% of patients but is associated with several complications, including acute urinary retention, stricture, urinary incontinence, and erectile dysfunction (30). Many patients choose treatment with less invasive procedures, such as transurethral needle ablation (31) and transurethral microwave therapy (32), which, although effective, can also result in urinary retention and infection and often have to be repeated. Injection of large volumes of absolute ethanol remains an experimental cytotoxic approach (33). In canine models, large areas of normal prostate (25%–50%) can be destroyed following injection of 10–25 mL of absolute ethanol without disruption of the prostate capsule or injury to bladder or urethra (34). In contrast, in this study, a single injection of 4.1 µg of PRX302 in only 25-µL volume (an approximately 4 µM solution) resulted in complete destruction of up to 50% of the monkey prostate gland with no substantial toxicity to normal urinary bladder or urethra. These results support the further study of intraprostatic PRX302 as potential treatment for BPH and suggest that clinical trials are warranted.

Supplementary Material

Suppl 1-
Suppl 2

CONTEXT AND CAVEATS.

Prior knowledge

PSA is a serine protease that is secreted by the prostate gland.

Study Design

PRX302, a PSA-activated protoxin, was synthesized and tested in prostate cancer cell lines and in several animal models.

Contributions

PRX302 treatment led to PSA-dependent decreases in cell viability and tumor growth. No toxicity to non-PSA expressing cells and tissues, neighboring organs, or general morbidity was observed.

Implications

PRX302 may be effective and safe for the treatment of locally recurrent prostate cancer and BPH.

Limitations

Although PSA-expressing cells were used in mouse models of prostate cancer, mice do not express PSA. Long-term effects of PRX302 treatment are unknown.

Acknowledgments

We thank Marc Rosen and Rebecca Ricklis for their excellent technical assistance.

This project was supported in part by funding from the National Cancer Institute Prostate SPORE (P50CA58236, S. R. Denmeade) and the US Department of Defense (PC040107, S. A. Williams).

R. F. Merchant holds stocks in Protox Therapeutics, Inc, the maker of PRX302, and is an officer of the company. J. T. Isaacs holds stocks in Protox Therapeutics, Inc. J. T. Buckley is the founder and Chief Scientific Officer of Protox Therapeutics, Inc. S. R. Denmeade holds stocks in Protox Therapeutics, Inc, and is a consultant for the company and a member of their scientific advisory board.

References

  • 1.Napalkov P, Maisonneuve P, Boyle P. Worldwide patterns of prevalence and mortality from benign prostatic hyperplasia. Urology. 1995;46(Suppl A):41–46. doi: 10.1016/s0090-4295(99)80249-0. [DOI] [PubMed] [Google Scholar]
  • 2.Jemal A, Siegel R, Ward E, Murray T, Xu J, Smigal C, et al. Cancer statistics. CA Cancer J Clin. 2006;56:106–130. doi: 10.3322/canjclin.56.2.106. [DOI] [PubMed] [Google Scholar]
  • 3.Roehl KA, Han M, Ramos CG, Antenor JA, Catalona WJ. Cancer progression and survival rates following anatomical radical retropubic prostatectomy in 3,478 consecutive patients: long-term results. J Urol. 2004;172:910–914. doi: 10.1097/01.ju.0000134888.22332.bb. [DOI] [PubMed] [Google Scholar]
  • 4.Han M, Partin AW, Zahurak M, Piantadosi S, Epstein JI, Walsh PC. Biochemical (prostate specific antigen) recurrence probability following radical prostatectomy for clinically localized prostate cancer. J Urol. 2003;169:517–523. doi: 10.1097/01.ju.0000045749.90353.c7. [DOI] [PubMed] [Google Scholar]
  • 5.Hanks GE. Radiotherapy or surgery for prostate cancer? Ten and fifteen-year results of external beam therapy. Acta Oncol. 1991;30:231–237. doi: 10.3109/02841869109092359. [DOI] [PubMed] [Google Scholar]
  • 6.Eastham JA, DiBlasio CJ, Scardino PT. Salvage radical prostatectomy for recurrence of prostate cancer after radiation therapy. Curr Urol Rep. 2003;4:211–215. doi: 10.1007/s11934-003-0071-6. [DOI] [PubMed] [Google Scholar]
  • 7.Ghosh A, Heston WD. Tumor target prostate specific membrane antigen (PSMA) and its regulation in prostate cancer. J Cell Biochem. 2004;91:528–539. doi: 10.1002/jcb.10661. [DOI] [PubMed] [Google Scholar]
  • 8.Borgono CA, Diamandis EP. The emerging roles of human tissue kallikreins in cancer. Nat Rev Cancer. 2004;4:876–890. doi: 10.1038/nrc1474. [DOI] [PubMed] [Google Scholar]
  • 9.Lilja H. Biology of prostate-specific antigen. Urology. 2003;62(Suppl 1):27–33. doi: 10.1016/s0090-4295(03)00775-1. [DOI] [PubMed] [Google Scholar]
  • 10.Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD, Walsh PC. Natural history of progression after PSA elevation following radical prostatectomy. JAMA. 1999;281:1591–1597. doi: 10.1001/jama.281.17.1591. [DOI] [PubMed] [Google Scholar]
  • 11.Denmeade SR, Sokoll LJ, Chan DW, Khan SR, Isaacs JT. Concentration of enzymatically active prostate-specific antigen (PSA) in the extracellular fluid of primary human prostate cancers and human prostate cancer xenograft models. Prostate. 2001;48:1–6. doi: 10.1002/pros.1075. [DOI] [PubMed] [Google Scholar]
  • 12.Denmeade SR, Isaacs JT. Enzymatic activation of prodrugs by prostate-specific antigen: targeted therapy for metastatic prostate cancer. Cancer J Sci Am. 1998;4(Suppl 1):S15–S21. [PubMed] [Google Scholar]
  • 13.Denmeade SR, Lou W, Lovgren J, Malm J, Lilja H, Isaacs JT. Specific and efficient peptide substrates for assaying the proteolytic activity of prostate-specific antigen. Cancer Res. 1997;57:4924–4930. [PMC free article] [PubMed] [Google Scholar]
  • 14.Denmeade SR, Jakobsen CM, Janssen S, Khan SR, Garrett ES, Lilja H, et al. Prostate-specific antigen-activated thapsigargin prodrug as targeted therapy for prostate cancer. J Natl Cancer Inst. 2003;95:990–1000. doi: 10.1093/jnci/95.13.990. [DOI] [PubMed] [Google Scholar]
  • 15.Abrami L, Fivaz M, van der Goot FG. Adventures of a pore-forming toxin at the target cell surface. Trends Microbiol. 2000;8:168–172. doi: 10.1016/s0966-842x(00)01722-4. [DOI] [PubMed] [Google Scholar]
  • 16.Rossjohn J, Feil SC, McKinstry WJ, Tsernoglou D, van der Goot G, Buckley JT, et al. Aerolysin — a paradigm for membrane insertion of beta-sheet protein toxins? J Struct Biol. 1998;121:92–100. doi: 10.1006/jsbi.1997.3947. [DOI] [PubMed] [Google Scholar]
  • 17.Abrami L, Fivaz M, Decroly E, Seidah NG, Jean F, Thomas G, et al. The pore-forming toxin proaerolysin is activated by furin. J Biol Chem. 1998;273:32656–32661. doi: 10.1074/jbc.273.49.32656. [DOI] [PubMed] [Google Scholar]
  • 18.Nakatsu CH, Wyndham RC. Cloning and expression of the transposable chlorobenzoate-3,4-dioxygenase genes of Alcaligenes sp. strain BR60. Appl Environ Microbiol. 1993;59:3625–3633. doi: 10.1128/aem.59.11.3625-3633.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ellis AE, do Vale A, Bowden TJ, Thompson K, Hastings TS. In vivo production of A-protein, lipopolysaccharide, iron-regulated outer membrane proteins and 70-kDa serine protease by Aeromonas salmonicida subsp. salmonicida. FEMS Microbiol Lett. 1997;149:157–163. doi: 10.1111/j.1574-6968.1997.tb10323.x. [DOI] [PubMed] [Google Scholar]
  • 20.Pretlow TG, Wolman SR, Micale MA, Pelley RJ, Kursh ED, Resnick MI, et al. Xenografts of primary human prostatic carcinoma. J Natl Cancer Inst. 1993;85:394–398. doi: 10.1093/jnci/85.5.394. [DOI] [PubMed] [Google Scholar]
  • 21.Horoszewicz JS, Leong SS, Chu TM, Wajsman ZL, Friedman M, Papsidero L, et al. The LNCaP cell line — a new model for studies on human prostatic carcinoma. Prog Clin Biol Res. 1980;37:115–132. [PubMed] [Google Scholar]
  • 22.van Bokhoven A, Varella-Garcia M, Korch C, Miller GJ. TSU-Pr1 and JCA-1 cells are derivatives of T24 bladder carcinoma cells and are not of prostatic origin. Cancer Res. 2001;61:6340–6344. [PubMed] [Google Scholar]
  • 23.Johnston SD, Kamolpatana K, Root-Kustritz MV, Johnston GR. Prostatic disorders in the dog. Anim Reprod Sci. 2000;60–61:405–415. doi: 10.1016/s0378-4320(00)00101-9. [DOI] [PubMed] [Google Scholar]
  • 24.Neal DE, Jr, Clejan S, Sarma D, Moon TD. Prostate specific antigen and prostatitis. I. Effect of prostatitis on serum PSA in the human and nonhuman primate. Prostate. 1992;20:105–111. doi: 10.1002/pros.2990200205. [DOI] [PubMed] [Google Scholar]
  • 25.Karr JF, Kantor JA, Hand PH, Eggensperger DL, Schlom J. The presence of prostate-specific antigen-related genes in primates and the expression of recombinant human prostate-specific antigen in a transfected murine cell line. Cancer Res. 1995;55:2455–2462. [PubMed] [Google Scholar]
  • 26.Denmeade SR, Litvinov I, Sokoll LJ, Lilja H, Isaacs JT. Prostate-specific antigen (PSA) protein does not affect growth of cancer cells in vitro or prostate cancer xenografts in vivo. Prostate. 2003;56:45–53. doi: 10.1002/pros.10213. [DOI] [PubMed] [Google Scholar]
  • 27.Stephenson AJ, Scardino PT, Bianco FJ, Jr, Eastham JA. Salvage therapy for locally recurrent prostate cancer after external beam radiotherapy. Curr Treat Options Oncol. 2004;5:357–365. doi: 10.1007/s11864-004-0026-2. [DOI] [PubMed] [Google Scholar]
  • 28.Han KR, Belldegrun AS. Third-generation cryosurgery for primary and recurrent prostate cancer. BJU Int. 2004;93:14–18. doi: 10.1111/j.1464-410x.2004.04547.x. [DOI] [PubMed] [Google Scholar]
  • 29.Tarter TH, Vaughan ED., Jr Inhibitors of 5alpha-reductase in the treatment of benign prostatic hyperplasia. Curr Pharm Des. 2006;12:775–783. doi: 10.2174/138161206776056010. [DOI] [PubMed] [Google Scholar]
  • 30.Reich O, Gratzke C, Stief C. Techniques and long-term results of surgical procedures for BPH. Eur Urol. 2006;49:970–978. doi: 10.1016/j.eururo.2005.12.072. discussion 978. [DOI] [PubMed] [Google Scholar]
  • 31.Braun M, Mathers M, Bondarenko B, Engelmann U. Treatment of benign prostatic hyperplasia through transurethral needle ablation (TUNA). Review of the literature and six years of clinical experience. Urol Int. 2004;72:32–39. doi: 10.1159/000075270. [DOI] [PubMed] [Google Scholar]
  • 32.Hoffman RM, MacDonald R, Monga M, Wilt TJ. Transurethral microwave thermotherapy vs transurethral resection for treating benign prostatic hyperplasia: a systematic review. BJU Int. 2004;94:1031–1036. doi: 10.1111/j.1464-410X.2004.05099.x. [DOI] [PubMed] [Google Scholar]
  • 33.Plante MK, Folsom JB, Zvara P. Prostatic tissue ablation by injection: a literature review. J Urol. 2004;172:20–26. doi: 10.1097/01.ju.0000121690.37499.1c. [DOI] [PubMed] [Google Scholar]
  • 34.Zvara P, Karpman E, Stoppacher R, Esenler AC, Plante MK. Ablation of canine prostate using transurethral intraprostatic absolute ethanol injection. Urology. 1999;54:411–415. doi: 10.1016/s0090-4295(99)00206-x. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Suppl 1-
NIHMS401596-supplement-Suppl_1-.pdf (41.7KB, pdf)
Suppl 2
NIHMS401596-supplement-Suppl_2.pdf (107.5KB, pdf)

RESOURCES