| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2000;70:1865-1871
© 2000 The Society of Thoracic Surgeons
a Thoracic Oncology Research Laboratory, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania, USA
Address reprint requests to Dr Kaiser, General Thoracic Surgery, 6 Silverstein Pavilion, Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104
e-mail: kaiser{at}mail.med.upenn.edu
Presented at the Thirty-sixth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 31–Feb 2, 2000.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Methods. Immunocompetent animals with established abdominal tumor were treated with either one or three (given weekly) intraperitoneal injections of Ad.HSVtk (109 plaque-forming units) followed by daily ganciclovir and monitored for survival. Survival studies were also performed in mice previously immunized with adenovirus.
Results. Animals treated with multiple courses of Ad.HSVtk showed significantly improved survival versus singly injected animals and control animals with some long-term survivors in the multiple injected group. Preexisting neutralizing immunity did not diminish this survival advantage.
Conclusions. Multiple treatments using an adenoviral vector to deliver HSVtk significantly improves survival in a murine intraperitoneal tumor model. The presence of preexisting neutralizing antibodies does not blunt this effect. Repeat Ad.HSVtk is a feasible approach and may be a useful strategy in human cancer gene therapy.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
An important lesson learned from this clinical experience was that to achieve therapeutic efficacy, it will be necessary to increase the amount of gene transfer to tumor cells. One possible way to achieve this goal would be to give multiple courses of Ad.HSVtk/GCV with the goal of creating "onion skin" type of killing; that is, a situation where successive layers of tumor are sequentially eradicated. However, the feasibility of this approach remains uncertain because experiments in rats [5], mice [6], rabbits [7], and primates [8] have shown that administration of adenoviral vectors can induce strong anti-Ad humoral and cellular immune responses that limit transgene expression. Administration of Ad vectors results in the development of neutralizing antibodies [6]. Additionally, virally infected cells stimulate T cell-mediated cytotoxicity that results in limited persistence of transgene [9]. These immune responses, especially neutralization of adenovirus by antibodies, could potentially limit transgene expression after repeated dosing and thus limit therapeutic benefit. On the other hand, there is growing evidence that immunological responses induced by Ad.HSVtk/GCV therapy can result in "cross-priming" with induction of antitumor cell immunological activity [10–12], a response that could augment antitumor effects.
The purpose of this study was to evaluate the potential efficacy of multiple doses of Ad.HSVtk/GCV in an animal model that would mimic potential future clinical trials. To accomplish this, it was necessary to study antitumor effects in a cell line that was both sensitive to Ad.HSVtk/GCV killing and was able to grow intraperitoneally in an immunocompetent host in a manner that resembled mesothelioma. Although some murine models of mesothelioma exist, in our experience, they are unlike most human mesothelioma lines in that they are rather resistant to Ad.HSVtk/GCV treatment. Therefore, as a model system, we utilized EJ-6-2-Bam-6A cells, a ras-transformed murine fibroblast that is quite infectable by adenoviral vectors and that grows well in the peritoneum of immunocompetent BALB/c mice. Tumor growth in these mice is highly reproducible and thus provides an excellent model for localized malignancies such as mesothelioma or ovarian carcinoma.
Using this immunocompetent animal model, we first compared the efficacy of single versus multiple courses of Ad.HSVtk/GCV therapy in animals never previously exposed to adenovirus (naïve animals). In a second series of experiments, we wanted to more closely mimic the clinical scenario of human trials by comparing the efficacy of single- versus multiple-course Ad.HSVtk/GCV therapy in animals that had been previously immunized with adenovirus and who possessed relatively high titers of neutralizing anti-Ad antibodies at the time of initiation of treatment (immunized animals). Our results indicate that multiple-dose Ad.HSVtk/GCV therapy was clearly superior to single-dose therapy in both immunocompetent naïve animals and animals that had been previously immunized with adenovirus.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Cell lines and culture
EJ-6-2-Bam-6A is a BALB/c mouse fibroblast cell line transformed with the H-Ras oncogene derived from a human bladder carcinoma. It was obtained from American Type Culture Collection (Manassas, VA) and maintained in DMEM media with 10% fetal calf serum (FCS), 100 U/mL penicillin G, 100 µg/mL streptomycin, and 2 mmol/L glutamine (Mediatch, Washington, DC). REN is a human mesothelioma cell line isolated in our laboratory from a patient tumor specimen [13]. These cells were cultured and maintained in RPMI 40 (Roswell Park Medical Institute) media supplemented with 10% FCS, 100 U/mL penicillin G, 100 µg/mL streptomycin, and 2 mmol/L glutamine. For all experiments, cells were harvested during exponential growth of the cell culture.
In vitro cell viability assays
To test the sensitivity of Ad.RSVtk to GCV, subconfluent monolayers of tumor cells were infected at various multiplicities of infection of 0.1, 1, 10, and 100 in serum-free DMEM for 4 hours. After the infection, medium containing 10% FCS was added and the cells were incubated for 24 hours. After the incubation period, infected cells were plated in 96-well plates in triplicate at a density of 2,000 cells/well in 100 µL of DMEM containing 10% FCS. Cells were then incubated for 5 additional days at 37°C with 20 µmol/L GCV (Hoffmann-La Roche, Inc, Nutley, NJ) diluted in media containing 10% FCS. Cell survival was assessed using a colorimetric cell proliferation assay that measures the viable cell dehydrogenase activity (CellTiter 96 Aqueous Non-radioactive MTS Cell Proliferation Assay; Promega Corporation, Madison, WI). All values were compared with control cells that received GCV alone.
Neutralizing antibody assay
Mouse serum was heat inactivated at 56°C for 1 hour and then serially diluted (1:25 to 1:25,600) in serum-free DMEM. The diluted serum was added in equal volumes to Ad.CMV/lacZ (107 plaque-forming units [pfu]/mL) and incubated at room temperature for 1 hour. The mixture (200 µL) was then applied to 5 x 103 REN cells (at 80% confluency) growing in 96-well plates. After 24 hours, cells were stained for beta-galactosidase activity. The neutralizing antibody titer was defined as the highest dilution at which approximately 50% of cells stained blue.
Animal studies
An intraperitoneal model using EJ-6-2-Bam-6A in BALB/c mice was established for all in vivo experimentation. The Animal Use Committees of the Wistar Institute and the University of Pennsylvania in compliance with the Guide for the Care and Use of Laboratory Animals (NIH No. 85-23, revised 1985) approved all animal protocols. Female BALB/c mice (6 to 8 weeks old weighing approximately 25 g) were obtained from Taconic Laboratory (Germantown, NY) and maintained in the animal facility at the Wistar Institute (Philadelphia, PA).
For the naïve animal experiments, intraperitoneal tumor was established in the mice by IP injection of 105 EJ-6-2-Bam-6A cells in 0.5 mL of serum-free DMEM. At 1 week, mice (n = 3) were sacrificed, and macroscopic tumor was confirmed on the bowel mesentery and retroperitoneum. Mice were then randomly divided into five treatment groups (n = 10); (1) mock (media only); (2) Ad.RSVtk/GCV treated singly; (3) Ad.RSVtk/GCV treated with three courses; (4) Ad.CMVlacZ treated with three courses; and (5) GCV only. Treatment animals (10 animals/group) received an IP injection of virus (109 pfu in 500 µL serum-free media) followed by 6 days of IP injection of GCV (50 mg/kg/day per animal). The multiple-treated animals received three cycles of weekly IP injection of virus followed by 6 days of IP injection of GCV (50 mg/kg/day per animal). The mice were followed daily for survival, which was analyzed by the Kaplan-Meyer method. Mice were euthanized when they appeared moribund, which was recorded as the day of death.
For the survival studies in immunized animals, mice were first immunized with Ad.CMV/lacZ by an IP injection of 108 pfu in serum-free media. At 2 and 6 weeks after immunization, serum was obtained from Ad.CMV/lacZ-injected mice (n = 4), analyzed for neutralizing antibodies, and compared against naïve animals (n = 3). After confirmation of the presence of neutralizing antibodies, tumor was established and animals were treated as in the naïve experiments.
Statistical analysis
All of the in vitro cell viability data are expressed as the mean ± standard error of the mean (SEM). Differences among groups were compared utilizing analysis of variance (ANOVA). The Kaplan-Meier survival data were examined using Mantel-Cox log-rank analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
|
As shown in Figure 2A, animals (n = 10/group) bearing intraperitoneal EJ-6-2-Bam-6A tumors that were treated with media alone, with GCV alone, or with three courses of Ad.CMV/lacZ all had virtually identical survival rates with a median survival of approximately 25 days. All animals were dead by 32 days. In contrast, animals treated with a single IP injection of Ad.RSVtk (109 pfu) followed by 6 days of GCV had a small, but significantly increased survival (p = 0.0009, Mantel-Cox log-rank) with a median survival of 35 days, with some animals living until 80 days. However, there were no long-term survivors.
|
These studies showed that in naïve, but immunocompetent, animals, multiple doses of Ad.RSVtk/GCV therapy were markedly more effective than a single treatment.
Survival studies in immunized mice
To better model the clinical setting where most patients have been previously exposed to Ad and have measurable anti-Ad antibody titers [4], we evaluated therapy in animals in which we generated preexisting Ad-neutralizing antibodies.
Mice were immunized with an IP injection of Ad.CMVlacZ. After 5 weeks, serum was obtained from mice (n = 4) and the presence of neutralizing antibodies against adenovirus was confirmed (average titer = 1:600 ± 115). Neutralizing antibody levels were below the sensitivity of the assay (1:25) in the naïve mice.
IP EJ-6-2-Bam-6A tumors were then established by IP injection of tumor cells. After confirmation of tumor at 1 week, groups of mice (n = 10) were randomized to receive: (1) media alone; (2) a single injection of Ad.RSVtk followed by 6 days of GCV; and (3) three courses of Ad.RSVtk given every 7 days, with each dose followed by 6 days of GCV.
As in the naïve animal experiments, control animals died with a median survival of 20 to 25 days (Fig 3). Single-dose Ad.RSVtk/GCV therapy again resulted in a small (median survival increased to 35 to 40 days), but significant survival advantage (p < 0.0001, log-rank) when compared with control. Importantly, even in previously immunized animals, there was a significant increase in median (> 55 days) and a 30% long-term survival (p = 0.0198, log-rank) in the multiple-treated animals.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
The efficacy of adenoviral gene therapy in animals that have been previously exposed to adenovirus is an important, but somewhat complicated issue. A large number of studies have clearly shown that injection of Ad vectors leads to both humoral and cellular anti-Ad and antitransgene immune responses. The effect of these immune responses on transgene expression has shown some variability, however. Most studies have demonstrated that repeat administration of Ad vectors in murine models has limited efficacy. For example, systemic reinfusion of an adenoviral vector encoding the LDL receptor into the liver of hypercholesteremic rabbits led to very little transgene expression compared with the initial infusion [7]. Additionally, other studies have demonstrated that in immunocompetent animals, a potent neutralizing antibody response to the vector completely blocks repeat gene transfer from intravenous challenge 1 month after initial inoculum [16]. To combat this problem, a number of immune suppressants have been coadministered with the Ad vector. Some success has been reported with immunosuppression using FK506 [17], deoxyspergualin [18, 19], cyclophosphamide [18, 20], CTLA4Ig [21], and recombinant interleukin-12 [22]. In contrast, some studies have seen prolonged transgene expression and successful reexpression of Ad vectors after multiple treatments [18, 23, 24]. These differences may be explained by the dose of vector used to immunize the animal [18], the strength of the preexisting immune response [18], how the vector is administered (ie, systemic administration may induce stronger immune responses than local instillation) [16, 23], the immunogenicity of the transgene [24], and the strain of mouse or type of animal used in the experiment [6].
Our observations in this intracavitary tumor model suggest that repeat administration of adenovirus to deliver thymidine kinase in combination with GCV is a feasible strategy even in the presence of preexisting humoral immunity. The success of this approach may be due to a number of factors. First, overcoming neutralizing antibodies may be much easier after injection of large doses of Ad into a localized cavity than after intravenous administration. That is, the relative antibody titer that is present with intracavitary or intratumor vector administration may differ markedly from that seen after intravenous administration due to differences in locoregional cytokines, inflammatory cell recruitment, and splenic or lymphatic clearance. Second, injection of repeated doses of vector within a relatively short time period (2 weeks) may avoid administration of vector at times when an amnestic anti-Ad antibody may be peaking. Third, although not specifically examined in this study, the immune responses elicited by Ad.HSVtk may have actually augmented the antitumor effects. A growing body of literature supports the hypothesis that in addition to the direct cytotoxic effects of phosphorylated GCV, the generation of antitumor immunological activity is as an important aspect of HSVtk therapy [1, 10–12]. As examples, Vile and associates observed a mononuclear cell infiltrate and intratumoral cytokine production after HSV-tk-therapy [25]. Kianmanesh and associates saw regression in remote nontransduced tumors after local tumor treatment [11]. This immunological mechanism may play a role in the observed improvement of efficacy seen in the multiple-treated animals. That is, multiple Ad.HSVtk injections may produce a more intense immunostimulatory environment that translates into improved therapeutic efficacy and improved survival even in the presence of decreased transgene expression.
Interestingly, there are a number of human gene therapy trials that are using repeated dosing of Ad vectors. One example is the ongoing Ad.p53 gene therapy trial, where large doses of Ad.p53 are administered intratumorally for advanced non-small cell lung cancer [26]. In fact, despite increased levels of anti-adenoviral antibody after treatment, transgene expression was seen even after three cycles of treatment [27]. Another example is the set of trials using the replication-selective Ad virus, Onyx 015, where vector is repeatedly injected over months. Preliminary reports seem to indicate gene transfer occurs after these multiple injections. In our previous clinical trial, we observed the presence of a significant titer (>1:100) of preexisting anti-Ad neutralizing antibodies in 8 of 21 patients, however, this level of antibody did not seem to preclude gene transfer. The levels of anti-Ad neutralizing antibodies in our animals (ie, 1:600) were somewhat higher than we saw in our patient population, suggesting that effects in human subjects might be similar. It should be noted, however, that much higher titers of antibodies (up to 1:4,000) were induced by Ad.HSVtk therapy approximately 14 to 21 days after Ad instillation [4]. These higher levels of neutralizing antibodies may be more inhibitory. An administration schedule that incorporates a multiple course of therapy over a relatively short period of time (ie, 2 to 3 weeks) would seem to be supported by our data.
In conclusion, we have demonstrated that Ad.HSVtk/GCV therapy provides a survival advantage in a murine intraperitoneal tumor model. The efficacy can be improved by repeat infections of Ad.HSVtk/GCV in both naive and immune animals. The mechanism for tumor killing potentially involves not only the direct cytotoxic effects of phosphorylated GCV metabolites, but also an antitumor immune response. An understanding of antiadenoviral vector immunity, as well as the antitumor immunological response induced by Ad.HSVtk/GCV therapy, will better define means to optimize this adenoviral-based suicide gene therapy. Our observations in this model, however, suggest that repeat treatments are a feasible strategy for future clinical trials in such diseases as malignant mesothelioma, brain tumor, and ovarian carcinoma.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
DR LAMBRIGHT: I will address your second question. We have not performed that study yet. However, most of the observations we have seen with the expression of the thymidine kinase is that it is maximized at about 48 to 72 hours, and by 1 week to 10 days, there is minimal at least immunohistochemical evidence of persistence of the transgene. So we do have some evidence to suggest that even at 1 week, a repeated dosing of ganciclovir would probably not be efficacious at that point in time.
A systemic administration of our adenoviral vector is a reasonable strategy; however, we have recently seen some of the significant toxicities that are associated with intraarterial as well as intravenous therapy of adenovirus. So our main strategy has been a localized therapy for this tumor, but intravenous therapy is indeed a potential strategy.
DR MALCOLM M. DeCAMP (Cleveland, OH): One of the questions I have is on one of the last points that you made about whether you are inducing an immunoreactivity to the tumor in addition to the ganciclovir/tk-mediated response. When you look histologically at these animals in the single injection versus multiple injections, are you seeing more lymphocytic infiltrates? Are you seeing more either mononuclear or even polymorphonuclear infiltrates associated with tumor in surviving animals?
DR LAMBRIGHT: We have not specifically quantitated the cellular infiltrates in the single- or the multiple-injected groups. That is to say that other groups have seen very significant infiltrates of mononuclear cells after local administration of the Ad-tk therapy. But in our experimentations, we have not seen this sort of specific change in cellular infiltrates in the single versus the multiple-injected groups.
DR LARRY R. KAISER (Philadelphia, PA): I think one of the very nice points of this study as Dr Lambright has presented it is that there does seem to be a fairly striking immune response directed against the adenoviral vector no matter what the transgene is. If you just take adenovirus and inject it, whether intraperitoneally or intrapleurally or maybe even intravenously, there is a fairly profound immune response against the adenoviral vector. One of the hopes had been that if you deliver an adenoviral vector with some transgene to tumor cells, maybe you can get an immune response directed against tumor cells as well, and, in fact, we have seen some preliminary evidence that that may in fact be the case. But one of the major questions in this study had to do with the very significant issue of the fact that most humans have seen adenovirus. So most of us, in fact, if not all of us, have preexisting anti-adenoviral antibodies, and one of the big questions has been whether you can apply adenoviral gene therapy in somebody who has already been immunized and who has preexisting antibodies, and especially the issue with repeat injections, and I think what this model shows very nicely is that using a syngeneic model where we have an immune response that we can in fact infect these animals. Even with the preexisting immunity, we still see the antitumor response here with the adenoviral vector and the ganciclovir. So the further extension of that is whether we can demonstrate some immune response directed against the tumor itself, and I think this is a model that we looked at and worked on for a long time to be able to develop a syngeneic model where adenovirus would work, and I think this is a model that has some real promise to it, and I think you can see with multiple injections, we may be on to something, and it is certainly feasible to do that in humans. Jack Roth’s group has done that with adenoviral p53 in intratumoral injection in humans as well. So it is feasible to do and hopefully will allow us another strategy.
DR LAMBRIGHT: Thank you very much.
DR MARK I. BLOCK (San Francisco, CA): I would like to follow up on the question of the immune response and its relevance to your model. The hypothesis is that gene therapy might not be as effective in mice that have been immunized against the adeno-virus because the vector is being destroyed by the host immune system. In the two experiments in which you demonstrated very nicely that survival was roughly equivalent between immunized and nonimmunized mice, maybe a little worse in the immunized mice, did you look at target gene expression in those two groups? In other words, do the immunized mice have the same or lower levels of target gene expression compared with the nonimmunized mice? I think that would be an important end point to evaluate. Your control group of vector alone addresses this question indirectly, but the definitive experiment would be to look at target gene expression in those two groups.
DR LAMBRIGHT: We have started to do that, and at this point in time we do not have any specific definitive conclusions, but we do have some immunohistochemical evidence that, indeed in the immunized group, there is transgene expression even after the third treatment. So we do have some preliminary evidence to suggest that we can see comparable transgene expression.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Kurdow, A. S. Boehle, S. Haye, L. Boenicke, B. Schniewind, P. Dohrmann, and H. Kalthoff Ganciclovir prodrug therapy is effective in a murine xenotransplant model of human lung cancer Ann. Thorac. Surg., March 1, 2002; 73(3): 905 - 910. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Margetts, S. Gyorffy, M. Kolb, L. Yu, C. M. Hoff, C. J. Holmes, and J. Gauldie Antiangiogenic and Antifibrotic Gene Therapy in a Chronic Infusion Model of Peritoneal Dialysis in Rats J. Am. Soc. Nephrol., March 1, 2002; 13(3): 721 - 728. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
