| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Thoracic Oncology Laboratory, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The incidence of MM, an asbestos-related malignancy, is increasing worldwide. There is no currently available effective therapy for this tumor and mortality remains disturbingly high with median survival of 
6–12 months. Novel therapies are desperately needed to treat this devastating disease (12)
. Human mesothelioma tumors have been shown to overexpress COX-2 (13)
, and high levels of COX-2 protein have been demonstrated to be a prognostic factor indicating poor outcome in this tumor (14)
, suggesting that COX-2 may contribute to the pathogenesis of MM. Blockade of COX-1 and/or COX-2 has some direct antiproliferative effects on human mesothelioma cell lines and has been reported to restore the depressed lymphokine-activated killer cell activity ex vivo (15)
. On the basis of these observations, the first goal of the present study was to evaluate the effect of a COX-2 inhibitor on mesothelioma growth in an animal model.
There are a number of studies suggesting that mesothelioma may be susceptible to immunotherapy (16, 17, 18) . Our group previously reported success in treating small mesothelioma tumors in mice with liposomes containing immunogenic cDNA (19) and even more impressive responses with a single i.p. dose of Ad.IFN-ß, an adenovirus expressing murine IFN-ß (16 , 20) . Both therapies acted through a CD8+ T-cell-dependent immune response associated with increased TILs. As with most immunotherapy approaches, however, these therapies were much less effective in larger tumors. Preliminary data in the Ad.IFN-ß system (see below) suggested that this lack of efficacy in large tumors was not attributable to a failure to produce CTLs but to a failure of existing CTLs to adequately enter and kill cells within the tumor. Given that one of the effects of COX-2 inhibition is postulated to be reduction of the immunosuppressive environment within the tumor, the second goal of this study was to test the hypothesis that COX-2 inhibition would complement Ad.IFN-ß therapy by allowing the CD8+ T cells greater access to the tumor environment and/or allowing improved cytotoxic functions. We therefore investigated the effect of COX-2 blockade combined with Ad.IFN-ß in mice bearing small and large tumors.
In this article, we report three novel findings: (a) a previously unreported antitumor effect of COX-2 inhibition in a mesothelioma model; (b) an observation that the mechanism of this antitumor effect was associated with increased numbers of TILs; and (c) the presence of complementary antitumor effects between COX-2 inhibition and therapy with Ad.IFN-ß when treating tumors.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
20–25 g) or female CB-17 SCID (6–8 weeks of age; weight, 20–25 g) that were obtained from Taconic Laboratory (Germantown, NY). All animals were housed in the animal facility at the Wistar Institute (Philadelphia, PA) and were maintained in a pathogen-free facility for at least 1 week before each experiment. Each experiment used at least 5 mice/condition and was repeated at least once for confirmation of results. The University of Pennsylvania and the Wistar Institute approved all protocols in compliance with the Guide for the Care and Use of Laboratory Animals. The mouse mesothelioma cell line AB12 was originally generated by i.p. implantation of asbestos fibers in BALB/c mice and has been characterized previously (16 , 19) . AB12 cells were cultured and maintained in high glucose DMEM (Mediatech, Washington, DC) supplemented with 10% FBS (Georgia Biotechnology, Atlanta, GA), 100 units/ml penicillin G, 100 µg/ml streptomycin, and 2 mM glutamine.
AB12 Flank Tumor Model.
We injected 5 x 105 AB12 cells in 0.1 ml of serum-free media s.c. Animals were treated 5 days later when 3-mm tumor nodules were visualized. Mice treated with Ad.IFN-ß were injected with a single i.t. dose of 1 x 109 or 1 x 108 pfu as described. We performed measurements of tumors twice weekly, and animals were sacrificed when they met predetermined criteria established for minimizing pain and suffering. Tumor volumes were estimated using the formula (
x long axis x short axis x short axis)/6.
Recombinant Adenovirus Vectors.
Production and characterization of the adenovirus encoding murine IFN ß has been described in detail (16
, 20)
. Vector preparations were shown to be free of wild-type adenovirus. We determined the particle to pfu ratio of each preparation in 293 cells; this ranged from 17:1 to 70:1.
Transduction Efficiency.
It was observed that COX-2 inhibition in combination with Ad.IFN-ß therapy had a more significant antitumor effect than either therapy alone (see "Results"). To determine whether COX-2 inhibition affected levels of viral transduction, BALB/c mice fed Rofecoxib-containing chow or regular chow were injected i.p. with Ad.IFN-ß. Forty-eight h later, mice were sacrificed and peritoneal fluid harvested by a lavage with 5 ml of saline. IFN-ß levels in the peritoneal fluid were assayed by ELISA as described previously (16)
.
In Vivo Depletion of CD4+ and CD8+ T Cells.
To deplete specific effector cell subsets before and during treatment with Rofecoxib chow in the flank model tumor AB12 model, BALB/c mice were injected with 200 µg of purified monoclonal antibodies purified from the anti-CD4+ hybridoma GK1.5 or the anti-CD-8+ hybridoma 53–6.7 [obtained from the American Type Culture Collection (Manassas, VA)]. Injections were administered 3 days and 1 day before inoculation with AB12 cells. Thereafter, a maintenance dose of antibody was injected i.p. every 7 days throughout the entire experimental period to ensure depletion of the targeted cell type. CD4+ and CD8+ T-cell depletion was confirmed by flow cytometry of splenic suspensions at the time of tumor injection and at later time points.
Phenotypic Analysis.
Flow cytometry was carried out splenocytes harvested from tumor-bearing mice as described previously (16)
. Briefly, tumor-bearing mice were treated with regular or Rofecoxib-containing chow. Seven days after initiation of treatment, the mice were sacrificed, and the spleens harvested and lymphocytes isolated. Flow cytometry was performed using fluorescently conjugated monoclonal antibodies for CD4, CD8, DX-5 (to identify NK cells), and CD11c (to identify dendritic cells; Southern Biotech, Birmingham, AL). Samples were analyzed using a Becton-Dickinson FACSCalibur flow cytometer (BD Biosciences, San Jose, CA).
Morphological Analysis.
For immunohistochemistry, we snap-froze tumors in Tissue Tek OCT (Sakura Finitek, Torrance, CA), made 5–7-µm sections on gelatin-coated slides, and fixed them in cold acetone. We performed immunoperoxidase staining by using the avidin-biotin-peroxidase Vecastain Elite ABC kit (Vector Laboratories, Burlingame, CA). Primary antibodies were antimouse CD4 (GK 1.5) and antimouse CD8 (Ly-2) purchased from Southern Biotech. We developed sections with 3,3'-diaminobenzidine tetrahydrochloride and counterstained with 1% hematoxylin.
Tumor Neutralization Test (Winn Assay).
Winn assays were performed as described previously (21)
. Spleen cells were isolated and CD8+ T cells were purified using the MACs system (Miltenyi Biotec, Auburn, CA). Briefly, splenocytes were reacted with anti-CD8+ microbeads and run through a magnet-mounted column (Miltenyi Biotec) that allowed the CD8- cells to pass but bound the bead-labeled CD8+ cells. Removal of the column from the magnet released the CD8+ cells, which were then eluted. This cell population contained >90% CD8 + cells by FACs (data not shown).
The CD8 + T-cell-enriched population from normal, tumor-sensitized or tumor-sensitized and treated mice was admixed with viable tumor cells at a ratio of 1 purified CD8+ splenocyte for each tumor cell and the mixture inoculated s.c. into naïve BALB/c mice. Tumor growth was measured after 15 days and expressed as the mean ± SE of 5 mice/group.
COX-2 Inhibition.
Specific COX inhibition was achieved using the COX-2 inhibitor Rofecoxib (MK-0966) obtained as a gift from Merck Frosst Canada & Co. (Kirkland, Quebec, Canada). This was incorporated into mouse chow by Test Diet (Richmond, IN) at a concentration of 0.0075%. Mice were fed this chow according to experimental design.
Measurement of Rofecoxib Levels in Serum.
Mice were sacrificed and blood harvested by phlebotomy of the left ventricle. Blood was allowed to clot and serum removed by pipette. Plasma Rofecoxib levels were measured by high-performance liquid chromatography as previously described (21)
and detected at a wavelength of 275 nm.
Debulking and Metastatic Foci Experiments.
To test for antitumor effects of COX-2 inhibition or COX-2 inhibition combined with Ad.IFN-ß that were undetectable in mice bearing large tumors, a model using debulking and metastatic foci was devised. In these experiments, 20 BALB/c mice were injected in one flank with 5 x 105 AB12 cells. On day 6, 10 mice were injected with 1 x 109 pfu of Ad.IFN-ß i.t. The other 10 mice were injected i.t. with a similar volume (100 µl) of vehicle. Three days later, all 20 mice had their tumors surgically removed, and 5 x 105 AB12 cells were injected in the opposite flank from the debulking. This injection was designed to create a standard residual or metastatic tumor burden. In the group previously treated with Ad.IFN-ß, 5 mice were fed Rofecoxib chow and 5 were fed control chow. In the vehicle-pretreated mice, 5 mice were fed Rofecoxib chow and 5 were fed control chow. Tumor volumes were measured twice weekly.
Statistics.
Tumor volume data at a given time point was analyzed by comparing the tumor volumes of the control group with the treatment groups for each experiment using ANOVA with appropriate posthoc testing (Wilcoxon test) when significant differences (P < 0.05) were found. Calculations were made using StatView (Cary, NC).
SEs for each set of tumor measurements were calculated and represented as y-axis error bars on each graph. In some instances, the error bars cannot be seen because they are smaller than the size of the graphed symbol.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tumor Growth Is Suppressed in Mice Fed the COX-2 Inhibitor Rofecoxib.
To investigate the effect of COX-2 blockade on the ability of mouse mesothelioma cells to form tumors, we injected AB12 mesothelioma cells into the flanks of mice and fed them control or Rofecoxib chow. Tumor size was measured twice weekly. As shown in Fig. 1A
, injected tumor cells in the control chow animals grew into large tumors over the course of 40 days. In contrast, animals fed Rofecoxib chow showed marked tumor inhibition. On day 21, the tumor size in the control group was 471 mm3 compared with 92 mm3 in the COX-2-inhibited group (P < 0.001). Interestingly, growth inhibition was only maintained 20–25 days after injection when tumor growth resumed at a rate similar to that of control tumors. These data demonstrate that treatment with COX-2 inhibition therapy was able to significantly inhibit the growth of injected mesothelioma tumor cells for 21 days.
|
. Although angiogenesis inhibition may function when T cells are present because of the immune mediated genesis of antiangiogenic factors, this experiment indicates that the effects of COX-2 inhibition in this model are not because of direct antiproliferative effects or angiogenesis inhibition but are dependent on B or T lymphocytes.
To further evaluate the apparent immune-dependent antitumor effects of COX-2 inhibition in these models, we used antibodies to selectively deplete CD8+ T cells from tumor-bearing mice fed Rofecoxib-containing chow (Fig. 2A)
. As above, mice fed Rofecoxib chow have significantly smaller tumors than those fed normal chow. The mean tumor volume at day 24 was 561 mm3 in the control group versus 176 mm3 in the Rofecoxib chow-fed group; COX-2 inhibition resulting in a 70% reduction of size (P < 0.05). CD8+ T-cell depletion in tumor-bearing mice fed normal chow led to markedly accelerated tumor growth. This finding is consistent with our detection of CTL activity in untreated animals (see below). Tumors in CD8+ T-cell-depleted mice fed Rofecoxib also grew faster than control tumors; however, growth was significantly less than in the CD8+ T-cell-depleted mice fed normal chow. The mean tumor volume at day 24 was 1820 mm3 in the CD8+ T-cell-depleted mice fed normal chow group versus 1074 mm3 in the CD8+ T-cell-depleted mice fed Rofecoxib chow group. The effect of COX-2 inhibition was diminished after CD8 depletion resulting in approximately a 40% reduction in size P < 0.05). These data show that COX-2 inhibition exerts only some of its antitumor effects through CD8+ T-cell-dependent mechanisms.
|
. Interestingly, depletion of CD4+ T cells in mice fed control chow resulted in suppression of tumor growth. At day 27, mice fed regular chow and depleted of CD4+ T cells (AB12 anti-CD4) had a mean tumor volume of 462 versus 698 mm3 for controls (P < 0.05). This effect has been observed in other tumors and suggests that there may be suppressor (regulatory) CD4+ T cells operative in this model. However, depletion of CD4 T+ cells did not inhibit the suppressive effects of Rofecoxib chow. Mice fed Rofecoxib-containing chow and depleted of CD4+ T cells had a mean tumor volume of 56 mm3 compared with 276 mm3 in the group depleted of CD4+ T cells and fed normal chow alone, a reduction in tumor volume of 80% (P < 0.05). These data show that COX-2 inhibition does not act exclusively through CD4+ or CD8+ T cells but rather is dependent on both cell types. To determine the effect of COX-2 inhibition on the number of splenic lymphocytes, tumor-bearing mice fed either Rofecoxib or regular chow were sacrificed, the spleens harvested, and the total lymphocyte cell count determined. There were no significant differences (data not shown). To assess whether COX-inhibition altered the relative proportions of lymphocyte subsets, flow cytometric analysis was performed on the splenic lymphocytes from tumor-bearing animals fed either Rofecoxib or regular chow using monoclonal antibodies for CD4, CD8, DX-5 (NK cells), and CD11c (dendritic cells). There were no significant changes in the percentages or absolute numbers of these cells in response to COX-2 inhibition (data not shown).
Recent in vitro work has suggested that PGE2 may inhibit dendritic cell function (10
, 11)
. Thus, one potential mechanism for our observation was that COX-2 inhibition resulted in an increase in the number or activity of CTLs. To test this hypothesis, we assayed for CTLs using an in vivo tumor neutralization assay (Winn assay). CD8+ lymphocytes were purified from the spleens of AB12 tumor-bearing mice that had been fed either regular chow Rofecoxib-containing chow. These isolated CD8+ lymphocytes were then mixed ex vivo with AB12 tumor cells at a ratio of 1:1 and injected into the flanks of naïve Balb/C mice. Tumor growth was assessed after 15 days. At this time, tumor cells that had not been mixed with lymphocytes grew to a size of 312 mm3 (Fig. 3
, control/bar 1). Addition of CD8+ T cells from tumor-bearing mice fed with normal chow to injected AB12 cells (Fig. 3
, bar 2) inhibited tumor growth by 50% (Fig. 3
, bar 1 versus bar 2; P = 0.01), indicating the presence of some CTL activity against the tumor. Importantly, tumor-bearing mice fed COX-2 inhibitor had virtually the same level of CTL activity (Fig. 3
, bar 2 versus bar 3).
|
(top panel), mice fed normal chow showed minimal infiltration of T cells into the tumor. In contrast, tumors from mice fed Rofecoxib-containing chow (Fig. 4
, second panels) showed clear increases in CD4+ and CD8+ T-cell infiltration. These results support the hypothesis that one effect of COX-2 inhibition is enhanced accumulation of lymphocytes within tumors.
|
100 mm3 (4–6 days after injection) at which time animals were continued on normal chow or switched the chow containing COX-2 inhibitor. As shown in Fig. 5
, mice fed Rofecoxib-containing chow showed significant tumor growth inhibition compared with controls fed normal chow. The mean tumor volume at day 25 was 1135 mm3 in the control group versus 390 mm3 in the Rofecoxib chow-fed group (P < 0.05). As we observed previously (Fig. 1A)
, after 21–25 days, the rate of growth of the tumor approached that of the control group despite continued ingestion of COX-2 inhibitor.
|
.
|
, bar 4).
We therefore tested the effect of combined therapy in established AB12 flank tumors. When tumors were 100 mm3, four groups were established. Group 1 received normal chow and no additional therapy. Group 2 were fed Rofecoxib-containing chow and no additional therapy. Group 3 received normal chow and was injected with 1 x 108 pfu Ad.IFN-ß, a dose that would inhibit but not eradicate the tumors. Group 4 received combined therapy with Rofecoxib-containing chow and the same dose of Ad.IFN-ß. As shown in Fig. 6
, up until day 24 all three treatments markedly inhibited tumor growth (P < 0.001 versus control for all groups). After 25–30 days, tumors in mice treated with Rofecoxib-containing chow began to grow rapidly. Tumors treated with Ad.IFN-ß alone also began to grow. In contrast, all tumors were completely eradicated in the group 4 mice receiving the combination of Rofecoxib-containing chow and the subtherapeutic dose of Ad.IFN-ß. None of these cured animals demonstrated tumor growth when subsequently injected with 5 x 105 AB12 cells on day 40 after initial treatment with COX-2 inhibition and Ad.IFN-ß (data not shown).
|
1000 mm3. At this time, the animals were started on Rofecoxib-containing chow, injected with 109 pfu Ad.IFN-ß, or received both therapies. As shown in Fig. 7
, COX-2 inhibitor therapy alone had no effect on tumor growth. Injection of Ad.IFN-ß alone also had minimal effects. In contrast, the combination of COX-2 inhibition and Ad.IFN-ß therapy significantly (P < 0.01 versus all three other groups at day 45) inhibited the growth of the tumors for >25 days. A similar growth inhibitory effect was noted for moderate-sized tumor with volumes of 500 mm3 (data not shown).
Rofecoxib Chow (COX-2 Blockade) Can Be Effectively Combined With Neoadjuvant Immunotherapy to Inhibit the Growth of Metastatic Foci after Debulking.
The finding that COX-2 blockade augmented the ability of i.t. Ad.IFN-ß therapy to eradicate small tumors suggested that a potentially effective clinical use of this combination would be to use this treatment in a setting where an i.t. injection could be followed by debulking leaving only minimal disease behind. The use of debulking in mesothelioma is well accepted as a way to improve symptoms, although it does not increase overall survival (23
, 24)
. It is possible that antitumor immune therapies administered to patients bearing large tumors might stimulate responses that would eradicate small tumors but would be inhibited or ineffective against the large tumors. Thus, the presence of a large tumor at the time of therapy might alter the response to the therapy. We modeled this clinical scenario by treating mice bearing large tumors (500 mm3), waiting 3 days, then surgically debulking the tumors and simultaneously injecting the contralateral flank with a known tumor burden of 5 x 105 AB12 cells. In this setting, the tumors cells in the injected flank of untreated mice grow rapidly ("None" in Fig. 8
). This model differs from a single injection of tumor without surgery in that the immune response to the small metastatic focus has been conditioned by the presence of the large excised tumor but is observable against only a small tumor. Evidence for this conditioning lies in the demonstration of antitumor activity (by Winn assay) of CD8+ T cells from treated mice bearing large tumors despite a lack of growth inhibition in treated large tumors themselves (data not shown).
|
500 mm3, groups 1 and 2 were injected i.t. with saline and groups 3 and 4 injected i.t. with 109 pfu of Ad.IFN-ß. Three days later, all groups had their tumors surgically removed (debulked) followed by injection of 5 x 105 AB12 cells into the contra lateral flank. Group 1 (None) received no additional treatment. Group 2 (Cox) was fed Rofecoxib chow. Group 3 received normal chow (ß), and group 4 was fed Rofecoxib chow (ß and Cox). The growth of the metastatic focus was followed by measuring tumor size twice weekly. As shown in Fig. 8
, both COX-2 blockade and Ad.IFN-ß therapy given as monotherapy before debulking, significantly slowed the growth of the contralateral tumor versus debulking alone (P < 0.05 and P < 0.0001, respectively). However, the combination of both therapies resulted in virtually complete suppression of tumor growth on the contralateral flank (P < 0.01 versus all three other groups).
Mechanism Of Rofecoxib Augmentation of Ad.IFN-ß Therapy.
One potential explanation for why Rofecoxib augmented the effect of Ad.IFN-ß therapy was that COX-2 inhibition increased the level of vector transduction or production of IFN-ß. To test this possibility, mice fed Rofecoxib-containing chow or regular chow and were injected with Ad.IFN-ß in the peritoneum. After 48 h, the mice were sacrificed and peritoneal fluid assayed for IFN-ß levels. COX-2 inhibition had no effect on the level of IFN-ß in the peritoneal fluid compared with fluid from mice that were not treated with COX-2 inhibition (data not shown).
We also determined whether the combination of COX-2 inhibitor plus Ad.IFN-ß resulted in increased activity of CTLs using the Winn assay. As shown in Fig. 2
, bar 5, addition of Rofecoxib did not increase the amount of CTL activity over that seen in Ad.IFN-ß-treated animals treated with regular chow (Fig. 3
, bar 4).
On the basis of our previous experiments, we examined the hypothesis that the efficacy of the combination of COX-2 inhibition and Ad.IFN-ß was because of an augmented ability of CTLs to accumulate in the tumors. Mice were injected with 5 x 105 AB12 cells and tumors allowed to grow to 500 mm3. At this time, the tumors were injected with saline or Ad.IFN-ß (109 pfu). Rofecoxib-containing chow was then initiated in one group. All animals were sacrificed 6 days later, and tumors were examined by histology for TILs. As shown in Fig. 4
(third set of panels), mice treated with Ad.IFN-ß showed a moderate increase in CD4+ and CD8+ lymphocytes compared with controls. In contrast, mice treated with both Rofecoxib chow and Ad.IFN-ß showed even larger numbers of CD4+ and CD8+ lymphocytes within the tumors.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The antitumor effects of COX-2 inhibition appear attributable to immunological effects because inhibition was completely lost in immunodeficient SCID mice (Fig. 1B)
. It still remains possible that angiogenesis inhibition could be operative if mediated through CD4+ and CD8+ T cells (i.e., through alterations in chemokine production). To further define the immunological mechanisms responsible for COX-2 inhibitor-induced tumor suppression, we conducted additional studies using antibody depletion of CD4+ or CD8+ cells. The results are interesting but interpretation is complicated by the effects of antibody depletion in mice fed normal chow. In this model, CD4+ T-cell depletion (in mice fed normal chow) inhibited tumor growth (Fig. 2B)
, suggesting the presence of a population of suppressor (regulatory) CD4+ cells. However, when Rofecoxib was fed to the CD4+ T-cell-depleted animals, tumor growth was inhibited by 80%. Given SCID mouse results, these data indicate that much of the COX-2 effect is CD8+ T-cell dependent. However, not all of the effect appears to be attributable to CD8+ T cells. CD8+ T-cell depletion led to more rapid growth of the tumors (Fig. 2A)
. Depletion of CD8+ T cells did not completely eliminate the effect of Rofecoxib; COX-2 inhibition blocked tumor growth by 30%. These depletion data show that COX-2 inhibition does not act exclusively through CD4+ or CD8+ T cells but rather is dependent on both cell types with CD8+ T cells appearing to be more important. These results are quite consistent with the immunohistology showing enhanced infiltration of both CD4+ and CD8+ T cells (Fig. 4)
.
There are a number of mechanisms by which COX-2 may have potentially important effects on the immune system. Recent studies have shown that COX-2 inhibition or decreased expression of the PGE2 EP2 receptor can prevent tumor-induced PGE2-dependent suppression of dendritic cell activity in vitro (10
, 11)
. However, in our animal model (using a semiquantitative Winn assay), we did not observe increased activity of CTLs in animals receiving COX-2 inhibition. Another mechanism proposed (in lung cancer models) is that COX-2 inhibition may exert its antitumor effect by altering the cytokine ratios of IL-10 and IL-12 within the tumor, thus restoring cell-mediated immunity (3)
. It has also recently been demonstrated that lung cancer cells do not express IL-10 
receptors and thus are not able to down-regulate COX-2 expression in response to increased IL-10 levels (25)
. Furthermore, the COX-2 product, PGE2, can down-regulate IL-2, IL-6, and IFN-
, thus preventing activation and trafficking of CTLs necessary for tumor regression (26)
. Our data showing that COX-2 inhibition did not increase the number or activity of splenic CTLs (Fig. 3)
but did increase the number of CD4+ and CD8+ T cells that were visible within the tumors (Fig. 4)
is more consistent with the hypothesis that COX-2 inhibition alters the immunosuppressive environment found within the tumor; an environment that both inhibits T-cell migration and inactivates or induces apoptosis in those T cells that do gain entry into the tumor. This hypothesis is being evaluated by additional studies using real-time PCR to profile cytokines such as IL-6, IL-10, transforming growth factor ß, IFN-, and chemokines such as monokine induced by interferon gamma (MIG) and gamma interferon inducible protein 10 (IP-10) in treated versus nontreated tumors are under way. In addition, phenotypic studies of dendritic cells in tumor-bearing mice undergoing COX-2 blockade are ongoing to determine the activation level of dendritic cells in vivo in response to this therapy.
An important limitation of this Rofecoxib therapy was the observation that only small tumors could be effectively treated. As with most immunotherapies described in animal models or in patients (27)
, therapeutic efficacy was almost completely lost when mice with moderate and large tumors (Fig. 7)
were treated. These results suggest that COX-2 inhibition in patients with MM would thus only be effective in those patients with minimal disease. Because most patients present with extensive tumors, the most likely therapeutic scenario where administration of COX-2 inhibitor would be helpful would be administration as adjuvant therapy after surgical debulking has been performed. This hypothesis is supported by data from our model of residual disease after debulking (Fig. 8)
where residual tumor growth was significantly slower in debulked animals fed Rofecoxib. Although somewhat controversial, surgical debulking for MM has a number of advocates and is not uncommonly performed (23
, 24) . Addition of COX-2 inhibitors may useful in this setting. Preliminary data from our laboratory also suggests that COX-2 inhibition could be useful in lung cancer. Thus, it may be reasonable to suggest a trial of COX-2 inhibition in patients surgically cured of lung cancer but at high risk of recurrence (i.e., stage IIb patients).
Given that COX-2 inhibition had the ability to enhance lymphocyte accumulation in tumors, we postulated that an important new use for COX-2 inhibition therapy could be in combination with immunotherapy. The second goal of this study was to test the hypothesis that COX-2 inhibition would complement Ad.IFN-ß therapy by allowing the CD8+ T cells greater access to the tumor environment. We have previously shown that Ad.IFN-ß increased the number of CTLs in an i.p. model of mesothelioma (16)
. Using the Winn assay, we confirmed that the activity of CTLs present after viral injection into flank tumors was also markedly increased (Fig. 3)
. Although these CTLs were able to markedly slow (Fig. 6)
or eradicate small tumors (data not shown at the higher dose of 109 pfu), they had only minimal effects on growth of larger tumors (Fig. 7)
. The limited number of stainable CD4+ and CD8+ cells within tumors with Ad.IFN-ß-treated animals (Fig. 3
, third row panels) suggests that this failure is likely attributable to an either an inability of the CTLs to effectively enter to the tumor or the rapid apoptosis of these cells within the tumor. The augmented efficacy of COX-2 plus Ad.IFN-ß therapy in both small (Fig. 6)
and very large (Fig. 7)
tumors plus the demonstration of increased numbers of CD4+ and CD8+ T cells within the tumors of animals given combined therapy (Fig. 4
, bottom panels) supports our hypothesis. We believe that COX-2 inhibition is effective in combination therapy because it targets the immune response at a different stage (CTL access and/or function in the tumor microenvironment) than Ad.IFN-ß (generation and/or expansion of CTLs).
Given that combination therapy is also much more effective in smaller tumors, our experiments suggest that COX-2 inhibitors could be especially useful in combination with immunogene therapy and surgical debulking in a clinical setting. Using a model of minimal residual disease after surgery, we found that the combination of COX-2 inhibitor plus Ad.IFN-ß therapy was more effective in preventing recurrences than either agent alone (Fig. 8)
. We are currently conducting a Phase I trial of Ad.IFN-ß in patients with mesothelioma. If this therapy proves safe, we anticipate that design of the Phase II study will be administration of Ad.IFN-ß followed by debulking followed by COX-2 inhibition. A similar approach might be useful for other tumors such as lung cancer.
In summary, we have shown that COX-2 inhibition does block the growth of small mesothelioma tumors through an immunological mechanism that appears to allow more effective accumulation of CTLs in the tumors. This therapeutic effect can be greatly magnified by combining COX-2 inhibition with an immunotherapy (such as Ad.IFN-ß) that works via a different mechanism (increasing the number of CTLs). The use of COX-2 inhibition thus presents a nontoxic, easily available approach that can be potentially be used in conjunction with wide variety of other immunotherapeutic approaches.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
P. D. and T. T. contributed equally to this manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Requests for reprints: Peter DeLong, Thoracic Oncology Research Laboratory, 8th Floor BRB II/III, 421 Curie Boulevard, Philadelphia, PA 19104. Phone: (215) 573-9918; Fax: (215) 573-4469; E-mail: pdelong@mail.med.upenn.edu.
1 The abbreviations used are: COX, cyclooxygenase; PGE2, prostaglandin E2; IL, interleukin; MM, malignant mesothelioma; Ad.IFN-ß, adenovirus-encoding IFN-ß; i.t., intratumoral; NK, natural killer; SCID, severe combined immunodeficiency; FBS, fetal bovine serum; pfu, plaque forming unit; TIL, tumor-infiltrating lymphocyte. 
Received 3/26/03. Revised 8/19/03. Accepted 9/ 5/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-based therapy in mesothelioma: assessment in a murine model. J. Interferon Cytokine Res., 15: 213-223, 1995.[Medline]
expression contributes to the maintenance of elevated cyclooxygenase-2 in non-small cell lung cancer cells. Cancer Res., 63: 766-770, 2003.This article has been cited by other articles:
|
|
 |
|
  N. Shinohara, A. Kumagai, K. Kanagawa, S. Maruyama, T. Abe, A. Sazawa, and K. Nonomura Multicenter Phase II Trial of Combination Therapy with Meloxicam, a COX-2 Inhibitor, and Natural Interferon-{alpha} for Metastatic Renal Cell Carcinoma Jpn. J. Clin. Oncol., November 1, 2009; 39(11): 720 - 726. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Eruslanov, S. Kaliberov, I. Daurkin, L. Kaliberova, D. Buchsbaum, J. Vieweg, and S. Kusmartsev Altered Expression of 15-Hydroxyprostaglandin Dehydrogenase in Tumor-Infiltrated CD11b Myeloid Cells: A Mechanism for Immune Evasion in Cancer J. Immunol., June 15, 2009; 182(12): 7548 - 7557. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Talmadge Pathways Mediating the Expansion and Immunosuppressive Activity of Myeloid-Derived Suppressor Cells and Their Relevance to Cancer Therapy Clin. Cancer Res., September 15, 2007; 13(18): 5243 - 5248. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Suzuki, S. Kim, H.-K. Cheung, M. J. Corbley, X. Zhang, L. Sun, F. Shan, J. Singh, W.-C. Lee, S. M. Albelda, et al. A Novel Small-Molecule Inhibitor of Transforming Growth Factor {beta} Type I Receptor Kinase (SM16) Inhibits Murine Mesothelioma Tumor Growth In vivo and Prevents Tumor Recurrence after Surgical Resection Cancer Res., March 1, 2007; 67(5): 2351 - 2359. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Mann, M. G. Backlund, F. G. Buchanan, T. Daikoku, V. R. Holla, D. W. Rosenberg, S. K. Dey, and R. N. DuBois Repression of Prostaglandin Dehydrogenase by Epidermal Growth Factor and Snail Increases Prostaglandin E2 and Promotes Cancer Progression. Cancer Res., July 1, 2006; 66(13): 6649 - 6656. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Ferrandina, F. O. Ranelletti, F. Legge, V. Salutari, E. Martinelli, A. Fattorossi, D. Lorusso, G. Zannoni, V. Vellone, A. Paglia, et al. Celecoxib Up-Regulates the Expression of the {zeta} Chain of T Cell Receptor Complex in Tumor-Infiltrating Lymphocytes in Human Cervical Cancer. Clin. Cancer Res., April 1, 2006; 12(7): 2055 - 2060. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Haas, J. Sun, A. Vachani, A. F. Wallace, M. Silverberg, V. Kapoor, and S. M. Albelda Cycloxygenase-2 Inhibition Augments the Efficacy of a Cancer Vaccine Clin. Cancer Res., January 1, 2006; 12(1): 214 - 222. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Wilderman, J. Sun, A. S. Jassar, V. Kapoor, M. Khan, A. Vachani, E. Suzuki, P. A. Kinniry, D. H. Sterman, L. R. Kaiser, et al. Intrapulmonary IFN-{beta} Gene Therapy Using an Adenoviral Vector Is Highly Effective in a Murine Orthotopic Model of Bronchogenic Adenocarcinoma of the Lung Cancer Res., September 15, 2005; 65(18): 8379 - 8387. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Suzuki, V. Kapoor, A. S. Jassar, L. R. Kaiser, and S. M. Albelda Gemcitabine Selectively Eliminates Splenic Gr-1+/CD11b+ Myeloid Suppressor Cells in Tumor-Bearing Animals and Enhances Antitumor Immune Activity Clin. Cancer Res., September 15, 2005; 11(18): 6713 - 6721. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. G. Backlund, J. R. Mann, V. R. Holla, F. G. Buchanan, H.-H. Tai, E. S. Musiek, G. L. Milne, S. Katkuri, and R. N. DuBois 15-Hydroxyprostaglandin Dehydrogenase Is Down-regulated in Colorectal Cancer J. Biol. Chem., February 4, 2005; 280(5): 3217 - 3223. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. R. Holla, D. Wang, J. R. Brown, J. R. Mann, S. Katkuri, and R. N. DuBois Prostaglandin E2 Regulates the Complement Inhibitor CD55/Decay-accelerating Factor in Colorectal Cancer J. Biol. Chem., January 7, 2005; 280(1): 476 - 483. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Suzuki, V. Kapoor, H.-K. Cheung, L. E. Ling, P. A. DeLong, L. R. Kaiser, and S. M. Albelda Soluble Type II Transforming Growth Factor-{beta} Receptor Inhibits Established Murine Malignant Mesothelioma Tumor Growth by Augmenting Host Antitumor Immunity Clin. Cancer Res., September 1, 2004; 10(17): 5907 - 5918. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. Spaner Amplifying cancer vaccine responses by modifying pathogenic gene programs in tumor cells J. Leukoc. Biol., August 1, 2004; 76(2): 338 - 351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. E. Zeytin, A. C. Patel, C. J. Rogers, D. Canter, S. D. Hursting, J. Schlom, and J. W. Greiner Combination of a Poxvirus-Based Vaccine with a Cyclooxygenase-2 Inhibitor (Celecoxib) Elicits Antitumor Immunity and Long-Term Survival in CEA.Tg/MIN Mice Cancer Res., May 15, 2004; 64(10): 3668 - 3678. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
