Original Article

Cancer Gene Therapy (2003) 10, 224–238. doi:10.1038/sj.cgt.7700562

Assessment of p53 gene transfer and biological activities in a clinical study of adenovirus-p53 gene therapy for recurrent ovarian cancer

Shu Fen Wen1, Vikas Mahavni2, Erlinda Quijano1, Jeremy Shinoda1, Michael Grace3, Mary Lynn Musco-Hobkinson3, Tong-Yuan Yang3, Yuetian Chen3, Ingo Runnenbaum4, JoAnn Horowitz3, Dan Maneval1, Beth Hutchins1 and Richard Buller2

  1. 1Canji, Inc., San Diego, California, USA
  2. 2The University of Iowa Hospital and Clinic, Iowa City, Iowa, USA
  3. 3Schering-Plough Research Institute, Kenilworth and Union, New Jersey, USA
  4. 4Universitäts-Frauenklinik, Freiburg, Germany

Correspondence: Dr Shu Fen Wen, Canji, Inc., 3525 John Hopkins Ct., San Diego, CA 92121, USA. E-mail: shufen.wen@canji.com

Received 6 May 2002.

Top

Abstract

A cohort study was designed to evaluate the efficiency of gene transfer and whether biological activity from the expressed therapeutic gene resulted after administration of a recombinant adenovirus containing the human wild-type p53 (p53wt) gene (rAd-p53 SCH 58500). The cohort study was conducted in five trial subjects with recurrent ovarian cancer. Each trial subject received multiple cycles of rAd-p53 SCH 58500, each cycle comprised of doses of 7.5 times 1013 particles on each of five consecutive days. Subjects were treated with rAd-p53 SCH 58500 alone during Cycle 1 and in combination with gemcitabine during the subsequent cycles. Both tumor biopsies and peritoneal aspirates were collected and evaluated for gene transfer and evidence of the biological activities of the expressed p53wt gene. Using quantitative PCR and RT-PCR, and in situ PCR, gene transfer and expression were documented in tumor biopsies (four of five patients) collected from Cycle 1. Furthermore, upregulation of p21/WAF1, bax and mdm-2, and downregulation of survivin were observed in these same tumor biopsy samples, suggesting that intraperitoneal administration of rAd-p53 SCH 58500 leads to detectable p53 biological activity in target tumor tissue. In addition, gene transfer and its expression were observed in cells obtained from peritoneal aspirates. These fluids were mainly comprised of polymorphonuclear neutrophils, indicating that successful gene transfer can be achieved by multiple cycle intraperitoneal administration of recombinant adenovirus.

Keywords:

rAd-p53, ovarian cancer, gene therapy, QPCR, SCH 58500

rAd-p53 SCH 58500 is a replication-deficient adenovirus encoding the p53 tumor-suppressor gene and has been studied in human subjects with melanoma, breast cancer,1 small-cell lung cancers,2,3 bladder cancers,4 liver,5 and ovarian cancers.6,7 These studies established the safety and feasibility of regional injections of rAd-p53 SCH 58500 for cancer. Evidence of p53 gene transfer was also shown for most of the human subjects who received high doses of rAd-p53 SCH 58500. Extensive safety data and gene transfer information have also been reported by other investigators using recombinant adenoviral (rAd) gene transfer in various clinical trials.8,9,10,11,12 While many of these initial protocols have demonstrated the feasibility of using recombinant adenovirus to deliver genes to tumors in humans, further investigation is required to understand the potential for effective gene therapy with this delivery system.

In this study we investigated: (1) whether the gene was delivered to the tumors of human subjects using a recombinant adenovirus vector, (2) whether the delivered gene demonstrated the anticipated transcriptional activity of human p53wt, and (3) whether repeat administration of a recombinant adenovirus will result in significant gene transfer. To this end, we designed a clinical protocol to evaluate the efficiency of gene transfer after multiple cycles of rAd-p53 SCH 58500 and measured the expression of genes known to be transcribed by p53.

It has been reported in clinical studies that a single intraperitoneal administration of a recombinant adenovirus vector carrying the gene encoding thymidine kinase13 or anti-erbB-2 single-chain antibody10 can result in detectable biological activity of the therapeutic gene. However, in these studies only low levels of gene transfer and expression were observed in tumors. The authors of these studies have also suggested that changing the dose regimen to include multiple administrations of the rAd might be a means to enhance the therapeutic benefit. In fact, favorable clinical responses have been reported in several clinical studies where administration of a rAd vector expressing p53wt gene over multiple cycles was employed to achieve a greater degree of gene transfer.6,7,8 Animal studies have shown that high titers of antiadenovirus antibodies detected in animals previously treated with recombinant adenoviruses caused transduction efficiency to be diminished or completely abrogated during subsequent dosing.14,15 It has therefore been the subject of debate as to whether multiple dosing strategies for recombinant adenoviruses are likely to be effective in increasing or maintaining levels of transgene expression in humans. Therefore, we believed that a careful evaluation of gene transfer and expression in human subjects scheduled to receive multiple cycles of a rAd would provide helpful information in evaluating a multiple dosing strategy.

To address this issue, we enrolled five additional subjects into a phase I/II trial with a multiple cycle administration schema for rAd-p53 SCH 58500 administration to human subjects with recurrent ovarian cancer.6,7 In this subcohort of individuals with recurrent ovarian cancer, we focused on the levels of p53 transgene expression and the assessment of p53 transcriptional activity during multiple cycle administration of rAd-p53 SCH 58500.

It has been shown in preclinical studies that the introduction of the p53wt gene into tumor cells lacking functional p53 results in antitumor activity through transcriptional regulation of genes involved in apoptosis and/or cell cycle arrest.16,17 In addition to gene delivery, it is important to know whether the gene delivered to the trial subjects has biological activity. The p53wt gene can activate genes such as p21/WAF1 and bax18,19,20,21 to induce G1 arrest and allow cells to repair DNA prior to proceeding to proliferation,22 or to accelerate cell apoptosis.23 Recently, Mirza et al.32 suggested that p53wt represses survivin expression, a member of the inhibitor of apoptosis (IAP) gene family. A feedback loop regulation between the mdm-2 and p53 genes has also been reported in various studies.24,25 Upregulation of mdm-2 by overexpression of p53wt can result in the repression of p53 transcriptional functions. Therefore, to better understand if rAd-p53 SCH 58500 exhibits antitumor biological activity in patients, it is important to know if the expression of the proapoptotic/apoptotic and cell cycle regulated genes such as p21/WAF1, bax, mdm-2, and survivin are modulated following the administration of rAd-p53 SCH 58500.

Typical preclinical methods for investigating the expression of p53 and its regulated genes after p53 gene therapy involve assessing the extent of protein phosphorylation or changes in protein levels, or using Northern blots to assess relative mRNA levels. Although a relatively straightforward endeavor when samples are obtained from tumor cell lines or xenografts in mice, human clinical samples are usually available in very limited quantities and with few or no proper controls. Limited human tissue specimens must be subjected to a battery of very sensitive tests to evaluate the activity. For this reason, prior to analyzing subject samples, we developed and validated quantitative PCR (QPCR) and quantitative RT-PCR (QRT-PCR) assays using various xenograft tumor models.26 Since the presence of endogenous human p53 and p21 proteins causes interference in many protein-based assays, and because such methods are relatively insensitive for quantifying specific proteins, we also applied QRT-PCR assay as an alternative tool to quantify p53 gene expression at mRNA level, instead of as protein.

Results from a phase I/II trial of rAd-p53 SCH 58500 gene therapy in recurrent ovarian cancer were recently published.6,7 The primary objective of these studies was to evaluate safety. We were limited in our ability to obtain biological samples, which require invasive procedures, for a more detailed evaluation. To obtain tumor biopsies and peritoneal aspirates during multiple cycle administration, we enrolled an additional five subjects to the trial. Safety and clinical activities of rAd-p53 SCH 58500 have been discussed elsewhere;6,7 therefore, the present study focused on the assessment of p53 biological activity and the levels of p53 transgene expression possible in recurrent ovarian cancer subjects receiving multiple cycles of daily dosing of rAd-p53 SCH 58500.

Top

Materials and methods

Clinical preparations of rAd-p53 SCH 58500

rAd-p53 SCH 58500 is a replication-deficient recombinant human adenovirus type 5 encoding the human p53wt gene under transcriptional control of the cytomegalovirus promoter. The construction, production, and purification via column chromatography of rAd-p53 SCH 58500 have been previously described.27,28 rAd-p53 SCH 58500 was administered on the basis of adenoviral particle number, which was determined via an OD260 nm/SDS method.29

Trial subject characteristics

Five female subjects with peritoneal carcinoma with pathologically confirmed recurrent ovarian cancer were entered into the study in the bioanalytical subcohort. Entry criteria included peritoneal fluid positive for tumor cells and tumor accessible for laparoscopic or percutaneous biopsy. All trial subjects must have tumors confirmed to have a p53 mutation as determined by cDNA sequencing.30

Study design and sample collection

Trial subjects were consented to receive multiple dose cycles of rAd-p53 SCH 58500 via intraperitoneal administration. For trial subjects with pre-existing peritoneal fluid of a clinically significant amount, as much peritoneal fluid as possible was drained prior to administration of the first dose of rAd-p53 SCH 58500. In all subjects, 7.5 times 1013 particles of rAd-p53 SCH 58500 was administered intraperitoneally in 250 mL of saline on each of five consecutive days during each cycle of treatment.7 Each cycle consisted of the five daily administrations of SCH 58500 followed by a 3-week 'rest' interval prior to the next cycle. Subjects were dosed with rAd-p53 SCH 58500 alone during the first cycle and were dosed in combination with gemcitabine in the four subsequent cycles. For cycles 2 and 3, 800 mg/m2 gemcitabine was delivered intravenously on days 1, 8, and 15. Peritoneal aspirate was scheduled for collection via paracentesis prior to rAd-p53 SCH 58500 administration on day 1 of each cycle and on day 5 prior to administration of the fifth dose of each cycle. Cell pellets from peritoneal aspirate were obtained and washed with PBS. Approximately, 1.0 times 106–1.0 times 107 cells from each subject were analyzed by QPCR and QRT-PCR assays, while 2.0 times 104 cells from each subject were analyzed via in situ PCR. Tumor biopsies were collected via laparoscopy or percutaneously predose on day 1 and after the fourth dose of the first and third cycles (prior to the fifth dose) on day 5. Biopsy samples consisted of approximately 10–100 mg of tissue. Samples were snap frozen and kept at -80°C until use. rAd–p53 SCH 58500 was administered on an outpatient basis. The subjects were kept under observation for at least 2 hours after every dose, or until the subject had recovered from any toxicity.

Quantification of expression of rAd-p53 SCH 58500, p21/WAF1, bax, caspase-3, survivin, mdm-2, and viral DNA

QPCR and QRT-PCR were employed to quantify rAd-p53 SCH 58500 viral DNA and gene expression using the Sequence Detector 7700 (Taqman®, Applied Biosystems, Foster City, CA) as reported previously.26 The GAPDH housekeeping gene was used as an internal control to assess the quality of assay samples and to normalize results. No results were reported if GAPDH DNA or RNA was less than 1000 copies per PCR reaction. rAd-p53 SCH 58500 DNA was quantified against a standard curve constructed from viral DNA extracted from purified rAd-p53 SCH 58500 virus (Qiagen, Valencia, CA). Two types of standard curves were used to quantify gene expression in this study. cRNA26 was used to quantify p53, p21/WAF1, bax, mdm-2, and GAPDH gene expression. A standard created from serially diluted RNA extracted from rAd-p53 SCH 58500-infected A549 cells was used to quantify caspase-3 and survivin expression. Gene expression results were expressed as the number of copies of the gene per 1000 copies of GAPDH (eg, when the cRNA standard was applied). In the case of caspase-3 and survivin, expression results were expressed as molecular equivalents (MEQ) per 1000 copies of GAPDH (eg, when the serially diluted total RNA standard was applied). MEQ is an arbitrarily assigned number based on serially diluted total RNA from rAd-p53 SCH 58500-infected cells. The sequences of the oligonucleotide primers and probes are shown in Table 1. The same primer sets and probes were used to perform PCR and RT-PCR for the p53 transgene. This set of primers and probe do not recognize the endogenous p53 gene sequence, and therefore only amplify rAd-p53 SCH 58500.26,27


Bioassay for adenovirus serum neutralizing factors

HEK 293 cells were plated onto collagen-I-coated 96-well plates (BioCoat, Fort Washington, PA) and incubated at 37oC in a 5% CO2 incubator for 3 hours prior to sample addition. Sera for analysis were serially diluted two-fold and placed onto the analysis plate(s). rAd-p53 SCH 58500 was then added so that there were 25 virus particles per cell. Assay plates were incubated for 72 hours. The plated cells were then fixed using a 1:1 (v/v) acetone:methanol solution. Fixed cells were stained for the presence of intracellular adenovirus hexon protein using a rabbit polyclonal anti-hexon antibody that had been raised against rAd-p53 SCH 58500 and a secondary FITC-labeled anti-rabbit Ig antibody (Chemicon, Temecula, CA). The fluorescence in each well was read using a CytoFluor II multiwell plate reader (Waters, Milford, MA). For each sample, the dilution factor was plotted versus the fluorescence intensity. Titer was then calculated by determining the inverse of the dilution factor that resulted in 50% of the maximum fluorescence intensity. The final result, the neutralizing anti-rAd-p53 SCH 58500 antibody titer, was expressed as the number of adenoviral particles neutralized per 1 mL of serum, and was calculated from a positive control curve created using rAd-p53 SCH 58500.

In situ PCR for rAd-p53 SCH 58500 DNA in peritoneal aspirate and tumor biopsies

To assess gene delivery to tumor and surrounding tissues, formalin-fixed paraffin-embedded tissue sections from each subject's tumor biopsy samples were analyzed using the in situ PCR assay as described previously.4,7 To assess gene delivery into cells found in the peritoneal aspirates, cells were pelleted from each subject's peritoneal fluid at 1000 rpm (approximately 300 g) for 5 minutes. After depletion of erythrocytes, approximately 2.0 times 104 cells were centrifuged onto in situ PCR slides using a Cytospin 3 (Shandon, Pittsburgh, PA) at 600 rpm for 10 minutes. After air-drying, the slides were fixed in 10% (v/v) formalin for 10 minutes and then dehydrated through a series of graded alcohol incubations. The slides were then processed as previously described with minor modifications.4 The Proteinase K digestion step was not performed on cells isolated from peritoneal fluid. Dinitrophenyl (DNP) labeled rAd-p53 SCH 58500 specific primers were used to amplify specifically rAd-p53 SCH 58500 but not endogenous p53. The sequence for the forward primer was

5'-CCACTGCTTACTGGCTTATCGAAAT-3'

The sequence for the reverse primer was

5'-CGTGTCACCGTCGTGGA-3'

A rabbit anti-DNP primary antibody (Zymed, San Francisco, CA) and an anti-rabbit IgG antibody conjugated with peroxidase (Vector, Burlingame, CA) were used to detect the PCR product. Positive rAd-p53 SCH 58500 viral DNA was visualized using a 3-amino-9-ethyl carbazole (AEC) substrate (Dako Corporation, Carpinteria, CA), after which samples were counterstained with hematoxylin.

Apoptosis analysis using laser scanning cytometry

The laser scanning cytometry (LSC) method has been previously described.31 Briefly, blocks were cut into 5 mum sections and mounted on glass slides. The tissues were stained for apoptosis using a modified fluorescence TUNEL assay and stained for nuclear identification using 0.001% (w/v) propidium iodide. Slides were scanned using a CompuCyte brand LSCTM. The fluorescein-dUTP incorporated into nicked DNA ends was excited by a 488-nm argon laser interrogation. A total of 12 sections were taken from each tumor block for laser scanning cytometry analysis. These sections were analyzed on six different analysis days (two sections per analysis day) to account for both sample variability and instrument variability. Qualification studies for the assay method had previously determined that the single greatest source of variability in the method, after sampling, was inter-day scanning (analysis day). On each analysis day for each sample, one slide was used for apoptosis staining (TUNEL reaction) and one as the negative control. The negative control consisted of the fluorescein-dUTP reaction performed in the absence of the labeling enzyme TdT. On each analysis day, the percentage of apoptotic nuclei per total nuclei scanned was calculated. Daily percentages were averaged across the 6 days to determine the mean percentage of apoptosis within the 'tumor'. Standard error of the mean was calculated based on the per cent daily variation.

Top

Results

Sample collection

Peritoneal fluid samples successfully collected for analysis are listed in Table 2. The frequency of laparoscopic/percutaneous or fine needle aspiration biopsies was limited because of ethical concerns based upon the invasive nature of these procedures and based on the trial subject health conditions at the time of collection. Pre- and postdose biopsies were collected from all trial subjects from Cycle 1. However, a Cycle 3 biopsy sample was only able to be collected from one trial subject, subject 53. Peritoneal aspirate was collected on Day 1, prior to the first dose in a cycle, and on Day 5th, approximately 24 hours after the fourth dose but prior to administration of the fifth dose administration, for each cycle with the exception of two samples. Owing to the health of the subject, Cycle 2 postdose sample from subject 50 was collected on Day 7 (2 days after the fifth dose) and Cycle 4 postdose sample from subject 52 was collected on Day 8 (3 days after the fifth dose).


Gene transfer and expression from rAd-p53 SCH 58500 in tumor biopsy samples

Gene transfer and expression were assessed using QPCR and QRT-PCR as described. We detected rAd-p53 SCH 58500 DNA in five of five (Fig 1a) and RNA in four of five (Fig 1b) subject Cycle 1/Day 5 biopsy samples. Neither rAd-p53 SCH 58500 DNA nor RNA was detected in any of the predose biopsy samples (Cycle 1/Day 1; data not shown). Samples collected from subject 53 showed no detectable p53 transgene RNA in either the Cycle 1/Day 5 postdose or the Cycle 3/Day 5 postdose tumor samples. Although rAd-p53 SCH 58500 DNA was detected below the limit of quantification (10 copies per PCR reaction) for subject 53's Cycle 1/Day 5 postdose and Cycle 3/Day 1 predose samples, the levels of rAd-p53 SCH 58500 DNA were found to be elevated in the Cycle 3/Day 5 biopsy sample (Fig 1a).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Quantification of rAd-p53 SCH 58500 gene delivery and expression in tumor samples. Panel a: rAd-p53 SCH 58500 gene transfer DNA levels. Panel b: rAd-p53 SCH 58500 gene expression RNA levels. Cycle 1/Day 5, closed bar, Cycle 3/Day 5, open bar. a: below quantification.

Full figure and legend (42K)

Gene transfer and expression of rAd-p53 SCH 58500 in peritoneal aspirates

rAd-p53 SCH 58500 DNA (Fig 2a) and RNA (Fig 2b) from postdose peritoneal aspirates for each cycle were measured using QPCR and QRT-PCR. As shown in Figure 2b, subjects 53 and 54 had elevated levels of rAd-p53 SCH 58500 RNA in samples collected from Cycle 3/Day 5, as compared to that from the previous cycles. In subject 52, we observed an increase in the amount of rAd-p53 SCH 58500 RNA in Cycle 2 with the RNA amount gradually decreasing over Cycles 3 and 4. By contrast, decreasing amounts of RNA were observed in these dose cycles in samples from subjects 49 and 50. All postdose peritoneal aspirates we examined contained detectable levels of rAd-p53 SCH 58500 RNA with the exception of two samples. Interestingly, these two samples were not collected on Day 5 as per the protocol, but instead, they were collected on Cycle 2/Day 8 for subject 50 and on Cycle 4/Day 7 in the case of subject 52. rAd-p53 SCH 58500 DNA was consistently detected at high levels in cells collected from all cycles (Fig 2a). In addition, rAd-p53 SCH 58500 DNA remained detectable in cells collected from peritoneal aspirates between dosing cycles, whereas RNA levels dropped below our limit of quantification (10 copies per PCR reaction) in peritoneal cells collected on Day 1 immediately before the next cycle of dosing. A representative example of this observation is shown in Figure 3 and contains data from subject 54. These data may imply that the duration of RNA is much shorter than for DNA. No rAd-p53 SCH 58500 DNA or RNA was detected in any of the samples collected at the predose Cycle 1 time point (data not shown).

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

rAd-p53 SCH 58500 DNA (a) and RNA (b) levels in postdose peritoneal aspirates. C1: Cycle 1; C2: Cycle 2; C3: Cycle 3; C4: Cycle 4. Black bar: C1/D5; gray bar: C2/D5; white bar: C3/D5; light gray bar: C4/D7. a: below quantification level.

Full figure and legend (46K)

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

rAd-p53 SCH 58500 DNA and RNA in Various Pre- and postdose peritoneal aspirates collected from subject 54. filled square: DNA. filled circle: RNA.

Full figure and legend (24K)

Serum anti-adenovirus neutralizing factors

Positive serum antiadenovirus neutralization activity (SNF) was observed in all subjects prior to treatment; the baseline level of neutralization capacity ranged between 1.5 times 108 and 7.0 times 109 virus particles neutralized per milliliter of serum (Table 3). Neutralization capacity increased in subjects 49, 50, 52, and 53 after each dosing cycle; however, SNF titers dropped to near baseline during the 3 weeks between dosing for all subjects, except in the case of subject 54. Subject 54 had the lowest SNF titers prior to dosing in Cycle 1, and SNF titers increased over the course of the three dosing cycles without dropping back to the predosing baseline value.


p53 biological activity

To understand whether the level of gene transfer delivered in these subjects was sufficient to induce p53 transcriptional activities, we investigated the expression levels of the p53-mediated genes, p21/WAF1, bax, and mdm-2, and of genes involved in the apoptotic pathway, caspase-3, and survivin, using QRT-PCR. rAd-p53 SCH 58500 mediated p21/WAF1 upregulation was observed in the Cycle 1/Day 5 tumor samples from all subjects who received four doses of rAd-p53 SCH 58500 (Table 4), with the exception of the Cycle 1/Day 5 tumor sample from subject 53. We observed a greater than 10-fold increase in p21/WAF1 expression levels in tumor samples from subjects who had detectable rAd-p53 SCH 58500 RNA. Consistent with the high levels of rAd-p53 SCH 58500 RNA detected in tumor samples from subjects 52 and 54, we observed a greater increase in the expression level of p21/WAF1 in these same two subjects as well. The Cycle 1/Day 5 tumor samples from subjects 52 and 54 had a greater than 100- and 30-fold increase in p21/WAF1 expression over the Cycle 1/Day 1 predose tumor samples, respectively.


We also investigated the change in proapoptotic gene expression after rAd-p53 SCH 58500 administration ( Table 4). rAd-p53 SCH 58500 mediated upregulation of bax expression was observed in four of five subject Cycle 1/Day 5 tumor samples, and ranged from a three- to a 27-fold increase. The Cycle 1/Day 5 tumor samples from subjects 49 and 50 showed no increases in caspase-3 expression. Subjects 52 and 54 were both observed to have a greater than four-fold increase in caspase-3 expression levels in their Cycle 1/Day 5 tumor samples ( Table 4). Consistent with the lack of detectable p53 transgene expression (Fig 1b), we observed no changes in p21/WAF1, bax, or caspase-3 expression in subject 53's Cycle 1/Day 5 tumor biopsy sample. Survivin expression decreased in three of five subject tumor samples, ranging from three- to nine-fold. Upregulation of mdm-2 was observed in two out of five Cycle 1/Day 5 tumor samples from subjects with detectable p53 transgene expression ( Table 4). Interestingly, no detectable mdm-2 expression was observed in subject 52's predose tumor biopsy sample.

In contrast to biopsy samples, there were no consistent changes in the RNA levels of bax, caspase-3, mdm-2, or survivin in the peritoneal aspirates (data not shown). We did observe consistent upregulation of p21/WAF1 gene expression in all but one postdose peritoneal aspirate. Since we were more consistently able to collect peritoneal aspirates, we were able to evaluate whether rAd-p53 SCH 58500 could upregulate p21/WAF1 in a multiple dosing regimen using data from the pre- and postdose peritoneal aspirates collected in each cycle. As shown in Figure 4, upregulation of p21/WAF1 was observed not only in the Cycle 1 peritoneal aspirates, but also in peritoneal aspirates from subsequent cycles.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Changes in p21/WAF1 gene expression in peritoneal aspirates. Pre- and postdose samples collected from the same cycle are summarized for comparison. square: C1/D1, Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author: C1/D5, Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author: C2/D1, Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author: C2/D5, Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author: C3/D1, filled square: C3/D5.

Full figure and legend (27K)

Gene delivery localization

In situ PCR was utilized to localize rAd-p53 SCH 58500 viral DNA in cells collected from peritoneal aspirates and tumor biopsy samples. Positive rAd–p53 SCH 58500 DNA signal was mainly found in peritoneal polymorphonuclear neutrophils (PMNs) (Fig 5, panels c and d). Our observations also indicate that PMN numbers increased significantly in postdose peritoneal aspirates (Fig 5, panels a and b). Tumor cells were very difficult to identify in the peritoneal aspirates. Predose peritoneal aspirates (Fig 5, panel e) did not contain any positive signal for rAd-p53 SCH 58500. rAd-p53 SCH 58500 DNA sequence was detected in tumor biopsy samples following rAd-p53 SCH 58500 dosing (Fig 6, panel b). Positive rAd-p53 SCH 58500 DNA signal was detected in epithelial cells, stromal cells, and tumor cells. No rAd-p53 SCH 58500 DNA signal was observed in any of the predose tumor samples (Fig 6a).

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

In situ PCR Localization of rAd-p53 SCH 58500 DNA in peritoneal aspirates. cells from peritoneal aspirates were collected on Day 1 (prior to the first dose) and Day 5 (prior to the fifth dose) of each cycle. H&E staining (panels a and b) shows that PMNs with multilobed nuclear characteristics (arrow head) significantly increased in numbers in postdose peritoneal aspirates (panel a, Cycle 1/Day 1, times 100, panel b, Cycle 1/Day 5, times 100). Panels c and d show the results of the in situ PCR assay for rAd-p53 SCH 58500: DNP-labeled primers were used to amplify rAd-p53 SCH 58500 DNA in situ. Brown positive rAd-p53 SCH 58500 DNA staining was observed in the Day 5 postdose samples from Cycle 1 (panel c, times 1000) and Cycle 2 (panel d, times 100) but not in the Cycle 1/Day 1 predose sample (panel e, times 100). Arrows indicate positive rAd-p53 SCH 58500 DNA signal (brown color). Dotted arrows indicate negative rAd-p53 SCH 58500 DNA signal (blue color).

Full figure and legend (42K)

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

In situ PCR localization of rAd-p53 SCH 58500 DNA in tumor biopsy samples. Tumor biopsies from subject 50 prior to SCH 58500 administration, Cycle 1/Day 1 (a, times 100), and at the end of the first cycle of SCH 58500 dosing, Cycle 1/Day 5(b, times 400). E: epithelial cells; S: stromal cells; T: tumor cells. SCH 58500 DNA was visualized using NBT/BCIP (dark blue color), and counter-stained with fast red (pink color).

Full figure and legend (104K)

Apoptosis analysis

Pre- and postdose tumor biopsies from subjects 49, 50, and 52 were made available for the LSC/TUNEL analysis. This analysis was quantified over six different tissue sections, with an average total of 50,000 nuclei per tumor section having been scanned. We observed a trend in samples from three subjects of increased apoptosis from pre- to postdose; however, the difference could not be resolved at P<.05. For example, subject 50 changed from an apoptosis level of 13.5plusminus5.9% on Cycle 1/Day 1 to 15.5plusminus4.6% apoptosis positive signal on Cycle 1/Day 5. Subject 52 biopsy samples were found to have an apoptosis measurement of 8.5plusminus3.2% on Cycle 1/Day1, and that increased to 11.0plusminus3.0% on Cycle 1/Day 5. The mean apoptosis value was determined by averaging six different sections within each tumor as described in Materials and methods. A false-color bit-map analysis of a representative set of scans for subject 52 shows that apoptosis was present on the periphery of the section at Cycle 1/Day 1, and that after 4 days (Cycle 1/Day 5), apoptosis was observed within the interior of tumor section (Fig 7). Furthermore, intermittent check sections were stained using conventional ApopTagTM immunohistochemistry staining; these sections confirmed the levels of apoptosis observed in the LSC scans (data not shown).

Figure 7.
Figure 7 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Laser scanning cytometry analysis of apoptosis (TUNEL). Four representative 5 mum tissue sections from subject 52 are shown. Tissues were scanned as described in the Materials and methods. Black dots represent TUNEL-negative nuclei; red dots represent TUNEL-positive nuclei. panels a and c show scans of tissue from Cycle 1/Day 1 and from Cycle 1/Day 5 post-treatment, respectively. TdT enzyme was included in the TUNEL reaction. The tissue sections were cut at different depths in the tumor biopsy. Panels b and d contain scans of tissue sections that were sequentially cut relative to the sections in panels a and c, respectively. In these panels, the TdT enzyme was omitted from the TUNEL reaction to detect nonspecific binding of the fluorescein-dUTP (panels b and d).

Full figure and legend (108K)

Top

Discussion

In this study, we applied several bioanalytical methodologies to analyze biopsy and ascitic samples with the aim of assessing gene transfer efficiency in a multiple dose regimen of rAd-p53 SCH58500. We also sought to investigate what biological activity resulted from p53 gene expression after rAd-p53 SCH 58500 treatment. One of the major hurdles we encountered was the difficulty in collecting sufficient amounts and numbers of biopsy samples from subjects at an appropriate time point for each specific evaluation. This limits the amount of useful information available to understand the impact of cancer gene therapy in humans. However, we were able to conclude that (1) gene transfer was not abrogated by the presence of antiadenovirus neutralizing factors, and (2) the levels of rAd-p53 SCH 58500 delivered into tumor in human subjects were sufficient to induce p53-mediated transcriptional activity.

To investigate whether the expression of p53 affected the expression of p53-regulated genes after rAd-p53 SCH 58500 administration, one would prefer to assess the extent of protein phosphorylation or changes in protein levels. However, the interference from endogenous proteins (eg, p21) and the limitation in the amount of clinical sample made it difficult to attribute the changes to exogenously introduced p53 gene. In addition, the absence of an adenovirus control made it more difficult to assess p53 specific activities independent of any effects because of the adenovirus vector itself. For these reasons, we had previously conducted preclinical studies with appropriate treatment control groups to demonstrate that QRT-PCR and PCR assays are feasible alternatives to evaluate the changes in the expression of p53-mediated genes.26

To examine if delivered p53wt gene could result in transactivating its downstream genes, we investigated p21, bax, mdm2, caspase-3, and survivin gene expression levels and their changes with dosing. The decreased levels of survivin observed in three of five subjects' tumor biopsies are inconsistent with a previous study32 that demonstrated when p53wt was introduced into 2774qw1 cells, survivin expression was repressed while p21/WAF1 expression increased. Although the significance of this observation is not clear, it has been suggested that survivin may play a role in the p53-mediated apoptotic pathway.32 We found that the only sample that did not show elevation of p21/WAF1 and bax expression was tumor from subject 53. However, subject 53's tumor did not have detectable p53 transgene expression.

Together these observations strongly suggest that the upregulation of p21/WAF1 and bax expression ( Table 4) were mediated by rAd-p53 SCH 58500. Indeed, rAd-p53 SCH 58500-mediated upregulation of p21/WAF1 has been documented in trials conducted in small-cell lung cancer and bladder cancer.4,33 Since no subjects were treated with a rAd-control vector, we could not rule out the possibility of a vector effect. Notably, we did not observe upregulation of p21/WAF1 or bax gene expression in a control group of animals treated with a rAd vector containing no transgene in a pilot study we performed.26 Importantly, no chemotherapeutic agent was administered during the cycle. This indicates that this observation was not a result of chemotherapy. Since biopsies were collected on day 5, 4 days after the first dose and 24 hours after the fourth dose of rAd-p53 SCH 58500, PMN had significantly increased in peritoneal fluid by this time point. Therefore, we cannot rule out the possibility that the changes in gene expression are because of infiltrating cells that have migrated to tumor sites in response to the rAd-p53 SCH 58500 administration. In peritoneal aspirates, consistent upregulation of p21 was observed; however, changes in the expression levels of the apoptotic-related genes, bax, caspase-3, and survivin were inconsistent among subjects.

We were unable to identify tumor cells in the peritoneal aspirates from the Day 5 time points (Fig 5a and b). The failure to identify tumor cells in peritoneal aspirate may also be because of the timing of samples. The Day 5 time point is long enough to allow tumor cells to complete the apoptotic process if triggered to do so by expression of p53 from the administered rAd-p53 SCH 58500. Therefore, the Day 5 results likely represent the expression levels primarily in PMN cells. In order to evaluate peritoneal tumor cell responses to rAd-p53 SCH 58500 treatment, we may need to develop a different method so as to isolate the relatively few tumor cells from the more abundant nontumor cells in any future studies.

Although the roles of bax, caspase-3, and survivin genes in apoptosis are well established, we cannot confirm that the changes in bax, caspase-3, and survivin gene expression levels we observed would in fact result in a favorable apoptotic response in the tumor cells of our trial subjects. We have previously shown in preclinical studies that LSC can be used to resolve rAd-p53 SCH 58500-induced apoptosis in human xenograft tumors.31 Using the same methodology, we wanted to see if we could observe a significant increase in rAd-p53 SCH 58500-mediated apoptosis in tumor biopsies derived from subjects on study. Although some investigators have reported an increase in apoptotic tumor cells after treatment with rAd-p53 and cisplatin8 in patients with nonsmall-cell lung cancer, we were unable to establish that the increases seen were statistically significant. We did observe a clear trend of increasing apoptosis in the tumor biopsy sections for subjects between the predose and postfourth dose time points in a cycle. The lack of statistical significance may be because of the small number of trial subject tumor tissue blocks available, and/or the variation between trial subjects. Sample timing may also provide one plausible explanation for the inability to observe a statistical increase in apoptosis. It has been suggested that the apoptotic process is complete within several hours.34 In this trial, samples were collected 24 hours after the fourth of the five consecutive daily administrations within a cycle, a time that may have caused us to underestimate the degree of the apoptosis induced by rAd-p53 SCH 58500.

In addition to evaluating p53 transcriptional activity, another major goal of this study was to investigate the impact of multiple administration cycles on gene transfer and expression. QPCR and QRT-PCR results from the peritoneal aspirates and a limited number of tumor biopsies indicate that it is possible to deliver the p53 gene into cells even in later dosing cycles. Indeed, the elevation of rAd-p53 SCH 58500 DNA (Fig 1a and Fig 2a) and RNA (Fig 2b) detected in the later cycles indicates that the higher levels of transgene expression can be achieved with repeated intraperitoneal dosing even with increased antiadenovirus neutralizing factor (SNF) in all trial subjects. As expected, SNF titers increased in all subjects during the course of treatment; however, with the exception of one subject, there was a rise and fall pattern between dosing cycles. SNF titers had dropped to baseline on the day of initiation of a subsequent cycle of rAd-p53 SCH 58500. Although anti-adenovirus neutralizing factors found in the peritoneal aspirates were not measured in this study, it has been documented in a previous study35 that similar titers were observed in both serum and peritoneal fluid from ovarian cancer patients. Although all subjects in this study had pre-existing anti-adenovirus SNF, gene transfer was detected in all subjects and there was no correlation between the level of p53 gene transfer and SNF titer. Similar results have been reported for a mesothelioma phase I clinical trial where a rAd5 carrying a suicide gene was administered to trial subjects intrapleurally.36 One possible explanation for this is the dose level (3.0 times 1014 particles) given to the subjects prior to sample collection. The dose amount may exceed the antiadenovirus SNF neutralizing capacity.

In situ PCR analysis revealed that we were able to deliver rAd-p53 SCH 58500 to tumor cells (Fig 6) in a variety of cell types. In cells collected from peritoneal aspirates, we found most of the positive rAd-p53 SCH 58500 signal in PMN cells with multilobed nuclear characteristics (Fig 5b). Taken together, the observation of high levels of p53 transgene expression (RNA) and upregulation of p21/WAF1 in peritoneal aspirates, we conclude that these cells were infected with rAd-p53 SCH 58500. Adenoviral vector-mediated gene transfer and expression in peritoneal cells via intraperitoneal injection was also evident in an ovarian cancer trial using an anti-erbB-2 single-chain antibody encoding adenovirus.10

In this study, we were able to demonstrate p53 transgene expression in both peritoneal aspirates and in tumor biopsy samples after multiple dose administration cycles of rAd-p53 SCH 58500. These data provide strong evidence that gene transfer was not abrogated by the presence of elevated levels of anti-adenovirus neutralizing factors. Furthermore, the increases in p21/WAF1, mdm-2, and bax gene expression suggest that the levels of rAd-p53 SCH 58500 delivered into tumors were sufficient to induce p53-mediated transcriptional activity in humans. It remains unclear whether the level of p53 expression achieved is sufficient for effective antitumor activity that would be evidenced by clinical responses. To this end, one of the trial participants, subject 52, had an interesting outcome. The bioanalytical analyses of her tumor biopsies showed both high p53 transgene expression and consistent and favorable p53 downstream gene responses. Subject 52's results reflected one of the best situations one might expect for a rAd-p53 gene therapy. We also observed that this subject had a positive clinical response to the rAd-53 SCH 58500 trial regimen. The results of her CT scan revealed that prior to treatment, subject 52 had malignant peritoneal fluid, an adrenal mass, a soft tissue mass at the ostomy site, and pulmonary nodules. The soft tissue ostomy site mass, pulmonary nodules, and malignant peritoneal fluid completely resolved by the end of Cycle 3. The size of the adrenal mass was reduced from 2.0 times 2.5 cm2 in size to 1.0 times 2.0 cm2 by the end of Cycle 3 (data not shown). Although this anecdotal response may not be attributable to rAd-p53 SCH 58500 regimen alone, subject 53 had previously failed three prior chemotherapy regimens. This suggests that rAd-p53 SCH 58500 was at least partially responsible for the response.

Top

References

  1. Dummer R, Bergh J, Karlsson Y, et al. Biological activity and safety of adenoviral vector-expressed wild-type p53 after intratumoral injection in melanoma and breast cancer patients with p53-overexpressing tumors. Cancer Gene Ther. 2000;7:1069–1076. | Article | PubMed | ISI | ChemPort |
  2. Schuler M, Rochlitz C, Horowitz JA, et al. A phase I study of adenovirus mediated wild type p53 gene therapy in patients with advanced non-small cell lung cancer. Hum Gene Ther. 1998;15:1750–1758.
  3. Schuler M, Herrmann R, De Greve JL, et al. Adenovirus-mediated wild-type p53 gene transfer in patients receiving chemotherapy for advanced non-small-cell lung cancer: results of a multi-center phase II study. J Clin Oncol. 2001;19:1750–1758. | PubMed | ISI | ChemPort |
  4. Kuball J, Wen SF, Leissner J, et al. Successful adenovirus –mediated wild-type p53 gene transfer in patients with bladder cancer by intravesical vector instillation. J Clin Oncol. 2002;20:957–965. | Article | PubMed | ISI | ChemPort |
  5. Ring E, Nonaka-Wong S, Horowitz JA, et al. Gene therapy of colorectal liver metastases using a recombinant adenovirus encoding wt p53 (SCH 58500) via hepatic artery infusion: a phase I study. Abstract. ASCO, 1998.
  6. Buller RE, Shahin MS, Horowitz J, et al. Long term follow-up of patients with recurrent ovarian cancer after Ad p53 gene replacement with SCH 58500. Cancer Gene Ther. 2002;9:567–572. | Article | PubMed | ISI | ChemPort |
  7. Buller RE, Runnebaum IB, Karlan B, Horowitz JA, et al. A phase I/II trial of rAd-p53 (SCH 58500) gene replacement in recurrent ovarian cancer. Cancer Gene Ther. 2002;9:553–566. | Article | PubMed | ISI | ChemPort |
  8. Nemunaitis J, Swisher SG, Timmons T, et al. Adenovirus-mediated p53 gene transfer in sequence with cisplatin to tumors of patients with non-small-cell lung cancer. J Clin Oncol. 2000;18:609–622. | PubMed | ISI | ChemPort |
  9. Albelda SM, Wiewrodt R, Sterman DH. Gene therapy for lung neoplasms. Clin Chest Med. 2002;23:265–277.  | Article | PubMed | ISI |
  10. Alvarez RD, Barnes MN, Gomez-Navarro H, et al. A cancer gene therapy approach utilizing an anti-erbB-2 single-chain antibody-encoding adenovirus (AD21): a phase I trial. Clin Cancer Res. 2000;6:3081–3087.  | PubMed | ISI | ChemPort |
  11. Crystal RG, Mcelvaney NG, Rosenfeld MA, et al. Administration of an adenovirus containing the human CFTR cDNA to the respiratory tract of individuals with cystic fibrosis. Nat Genet. 1994;8:42–51. | Article | PubMed | ISI | ChemPort |
  12. Sterman DH, Treat J, Litzky LA, et al. Adenovirus-mediated herpes simplex virus thymidine kinase/ganciclovir gene therapy in patients with localized malignancy: results of a phase I clinical trial in malignant mesothelioma. Hum Gene Ther. 1998;9:1083–1092. | PubMed | ISI | ChemPort |
  13. Hasenburg A, Tong XW, Fischer DC, et al. Adenovirus-mediated thymidine kinase gene therapy in combination with Topotecan for patients with recurrent ovarian cancer: 2.5-year follow-up. Gynecol Oncol. 2001;83:549–554. | Article | PubMed | ISI | ChemPort |
  14. Benihoud K, Saggio I, Opolon P, et al. Efficient repeated adenovirus-mediated gene transfer in mice lacking both tumor necrosis factor alpha and lymphotoxin alpha. J Virol. 1998;72:9514–9525. | PubMed | ISI | ChemPort |
  15. Dong JY, Wang D, Van Ginkel FW, et al. Systematic analysis of repeated gene delivery into animal lungs with a recombinant adenovirus vector. Hum Gene Ther. 1996;7:319–331. | PubMed | ISI | ChemPort |
  16. Ozbun MA, Butel JS. Tumor suppressor p53 mutations and breast cancer: a critical analysis. Adv Cancer Res. 1995;66:71–141. | PubMed | ISI | ChemPort |
  17. Selter H, Montenarh M. The emerging picture of p53. Int J Biochem. 1994;26:145–154. | Article | PubMed |
  18. Miyashita T, Krajewski S, Krajewska M, et al. Tumor suppressor p53 is a regulator of bcl-2 and bax gene expression in vitro and in vivo. Oncogene. 1994;9:1799–1805. | PubMed | ISI | ChemPort |
  19. Miyashita T, Reed JC. Tumor suppressor p53 is a direct transcriptional activator of the human bax gene. Cell 1995;80:293–299. | Article | PubMed | ISI | ChemPort |
  20. Somasundaram K, El-Deiry WS. Tumor suppressor p53: regulation and function. Frontiers Biosci 2000;5:d424–d437.
  21. El-Deiry WS, Harper JW, O'connor PM, et al. WAF1/CIP1 is induced in p53-mediated G1 arrest and apoptosis. Cancer Res. 1994;54:1169–1174. | PubMed | ChemPort |
  22. El-Deiry WS, Tokino T, Velculescu VE, et al. WAF1, a potential mediator of p53 tumor suppression. Cell 1993;75:817–825. | Article | PubMed | ChemPort |
  23. Kitanaka C, Namiki T, Noguchi K, et al. Caspase-dependent apoptosis of COS-7 cells induced by Bax overexpression: differential effects of Bcl2 and Bcl-xL on Bax-induced caspase activation and apoptosis. Oncogene. 1997;15:1763–1772. | Article | PubMed | ChemPort |
  24. Chen Cy, Oliner Jd, Zhan Q, et al. Interactions between p53 and MDM2 in a mammalian cell cycle checkpoint pathway. Proc Natl Acad Sci. 1994;91:2684–2688. | PubMed | ChemPort |
  25. Oliner JD, Pietenpol JA, Thiagalingam S, et al. Oncoprotein MDM2 conceals the activation domain of tumor suppressor p53. Nature 1993;362:857–860. | Article | PubMed | ISI | ChemPort |
  26. Wen SF, Xie L, Mcdonald M, et al. Development and validation of sensitive assays to quantitate gene expression after p53 gene therapy and paclitaxel chemotherapy using in vivo dosing in tumor xenograft models. Cancer Gene Ther. 2000;7:1469–1480. | Article | PubMed | ISI | ChemPort |
  27. Wills KN, Maneval DC, Menzel P, et al. Development and characterization of recombinant adenoviruses encoding human p53 for gene therapy of cancer. Hum Gene Ther. 1994;5:1079–1088. | PubMed | ISI | ChemPort |
  28. Huyghe BG, Liu X, Sutjipto S, et al. Purification of a type 5 recombinant adenovirus encoding human p53 by column chromatography. Hum Gene Ther. 1995;6:1403–1416. | PubMed | ISI | ChemPort |
  29. Maizel JV Jr, White DO, Scharff MD. The polypeptides of adenovirus. I. Evidence for multiple protein components in the virion and a comparison of types 2, 7A, and 12. Virology 1968;36:115–125. | Article | PubMed | ISI | ChemPort |
  30. Shahin MS, Hughes JH, Sood AK, et al. The prognostic significance of p53 tumor suppressor gene alterations in ovarian carcinoma. Cancer 2000;89:2006–2017. | Article | PubMed | ISI | ChemPort |
  31. Grace M, Xie L, Musco ML, et al. The use of laser scanning cytometry to assess depth of penetration of adenovirus p53 gene therapy in human xenograft biopsies. Am J Pathol. 1999;155:1869–1877. | PubMed | ISI | ChemPort |
  32. Mirza A, McGuirk M, Hockenberry T, et al. Human survivin is negatively regulated by wild-type p53 and participates in p53-dependent apoptotic pathway. Oncogene 2002;21:2613–2622. | Article | PubMed | ISI | ChemPort |
  33. Boylay JL, Perruchoud AP, Reuter J, et al. p21 gene expression as an indicator for the activity of adenovirus-p53 gene therapy in non-small cell lung cancer patients. Cancer Gene Ther. 2000;7:1215–1219. | Article | PubMed |
  34. Bursch W, Kleine L, Tenniswood M. The biochemistry of cell death by apoptosis. Biochem Cell Biol. 1990;68:1071–1074. | PubMed | ISI | ChemPort |
  35. Stallwood Y, Fisher KD, Gallimore PH, Mautner V. Neutralisation of adenovirus infectivity by ascitic fluid from ovarian cancer patients. Gene Ther. 2000;7:637–643. | Article | PubMed | ChemPort |
  36. Molan-Kimber KL, Sterman DH, Chang M, et al. Impact of preexisting and induced humoral and cellular immune responses in an adenovirus-based gene therapy phase I clinical trial for localized mesothelioma. Hum Gene Ther. 1998;14:2121–2133.
Top

Acknowledgements

We thank the patients and their families for their participation in this bioanalytical cohort of the trial. We thank Michelle Kerin and Melanie Hattermann for the tremendous effort they put in coordinating study samples. We would also like to thank Dr. Suxing Liu for her helpful discussion of the survivin data.

Extra navigation

.
ADVERTISEMENT