| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cancer Therapy: Preclinical |
Departments of 1 Medical Biophysics and 2 Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre and Toronto Sunnybrook Regional Cancer Centre, University of Toronto; 3 Divisions of Haematology/Oncology and Clinical Pharmacology and Toxicology, The Hospital for Sick Children, Toronto, Ontario, Canada; 4 Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts; and 5 Ipsogen, Marseille, France
Requests for reprints: Yaacov Ben-David, Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, Research Building (S-216), Toronto, Ontario, Canada M4N 3M5. Phone: 416-480-6100; E-mail: bendavid{at}srcl.sunnybrook.utoronto.ca.
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experimental Design: Juvenile erythroleukemic mice (n = 8) received no treatment, celecoxib (1600 mg/kg/d), vinblastine (0.5 µg/g twice weekly), vehicle controls, or celecoxib + vinblastine combination (n = 9) over a 6-month period from time of tumor induction. Overt toxicity was assessed daily and recorded weekly.
Results: Among randomly selected mice from celecoxib treatment groups, plasma concentrations ranged from 2 to 6 µmol/L. As a single agent, celecoxib was not associated with any apparent toxicity. Monotherapy with vinblastine, however, caused early mortality marked by severe diarrhea, lethargy, and weight loss. At the tested doses, neither vinblastine nor celecoxib enhanced survival as monotherapies. Coadministration of these two drugs alleviated the overt toxicity associated with vinblastine and resulted in a significant increase in survival (P < 0.05). Survivors sampled throughout the study showed a trend to decreased weight loss and hematocrit levels among all groups, but significance was evidenced earlier in the vinblastine monotherapy group overall (P < 0.05). Despite similar degree of splenomegaly, histologic analysis revealed preserved splenic mantle architecture from mice given combination therapy compared with those sampled from mice on all other monotherapies, exhibiting a more diffuse burden of blasts and destruction of germinal centers.
Conclusion: We propose that addition of a selective cyclooxygenase-2 inhibitor to a modified low-dose conventional chemotherapeutic regimen protects juvenile mice with Friend disease from succumbing to low-dose cytotoxicity, in part, by neutralizing acute inflammatory responses.
Key Words: celecoxib • combination therapy • erythroleukemia
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A murine model, Friend disease, whereby the stepwise progression of a viral-induced erythroleukemia can be monitored from an early polyclonal stage through to the proliferation of tumorigenic clones leading to blast crisis and death of the organism, is invaluable for testing new therapies in these cancers. Pathophysiology of murine erythroleukemia was first described by Charlotte Friend as a presentation of anemia, thrombocytopenia, leukocytosis, increased liver and spleen volume, and hematopoiesis (6). Subsequent studies over the years have discerned the molecular mechanisms surrounding the growth and survival advantages associated with Friend virus–induced erythroleukemia cells (for review, see refs. 7, 8). The pathology and aggressive nature of Friend disease are akin to the human M6 variant.
The study presented herein is exploring a therapeutic regimen of clinically available drugs. The anticancer activity of cyclooxygenase (COX)-2-selective nonsteroidal anti-inflammatory drugs (NSAID) has now been documented both in vitro and in vivo (for review, see ref. 9). In particular, the selective COX-2 inhibitor celecoxib, which inhibits the conversion of arachidonic acid to prostaglandin-H2, has shown promising results. Human prostate LNCaP and PC-3 cancer cells exhibit sensitivity toward such COX-2 inhibitors through a proapoptotic mechanism mediated via inactivation of the antiapoptotic kinase (Akt; ref. 10). Similar effects are evident against the in vitro proliferation of colorectal cancer cells, in which the induction of apoptosis occurs by a currently undescribed mechanism, albeit independent of COX-2 inhibition (11). From an angiogenesis perspective, an interesting parallel has also been currently established whereby increased expression of vascular endothelial growth factor protein is associated with increased COX-2 activity in head and neck cancers (12). Recent findings from our laboratory (13) and others (14) have implicated a role for COX-2 in the proliferation and differentiation of murine and human leukemia cell lines through the use of COX-2-specific inhibitors; thus we embarked on evaluating the efficacy of COX-2 inhibitors as prevention or intervention therapies in hematologic malignancies.
Presented here is a study evaluating the efficacy of a COX-2-selective NSAID in conjunction with continuous, low-dose administration of the conventional cytotoxic agent, vinblastine, for the treatment of Friend disease, used here as a clinical surrogate of FAB M6 AML. Vinblastine was chosen based on the previous observations of Vacca et al. (15) and Klement et al. (16, 17), showing that low doses of vinblastine exhibited potent antiangiogenic properties in vitro and in vivo with sustained tumor regression and minimal toxicity. We show that as monotherapies neither celecoxib nor vinblastine provided survival benefits for juvenile erythroleukemic mice, whereas the combination of these two drugs provided significant results. Similar observations have recently been reported with the use of another variant of the COX-2-targeted drugs, rofecoxib, in the clinical management of metastatic melanoma (18). Only when combined with metronomic, low-dose treosulfan was stable disease achieved in these patients. Similar to their findings, the synergistic action of the agents used in our present study cannot be explained through presently known mechanisms of action for either drug. However, celecoxib may provide protection against the toxicity of low-dose, continuous antiangiogenic regimens with vinblastine, particularly evident in vulnerable, juvenile mice with aggressive disease, or it may further enhance lymphoid-activating capacity to effectively control the erythroleukemic blast population.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Viral Lysate Production. Viral lysate of the replication-competent NB-tropic Friend murine leukemia virus (F-MuLV) was prepared through repeated culturing of the fibroblastic clone B cell line in MEM (Life Technologies, Inc., Gaithersburg, MD) supplemented with 10% fetal bovine serum (Life Technologies, Inc.) and penicillin/streptomycin at 1000 units/mL (Life Technologies, Inc.). Collection of viral-containing medium occurred when adherent cells seemed to be 70% confluent. These cells were briefly centrifuged at 14,000 rpm to remove cellular debris and stored at –70°C until time of injection into murine neonates.
Tumor Induction and Drug Administration. Seven-week-old BALB/c female mice (Charles River Laboratories, Saint-Constant, Quebec, Canada) were housed 15 mice per cage to synchronize estrous cycles to ensure sufficient numbers of similarly aged offspring. Neonates were randomly divided into two groups: infected and noninfected controls. Infections were carried out by i.p. administration of 100 µL of F-MuLV viral lysate via a 1-mL U-100 insulin syringe (Becton Dickinson and Company, Mississauga, Ontario, Canada) 1 day post birth. Uninfected neonates (three per group) served as toxicity controls for celecoxib and vinblastine. Weaning of all offspring took place 3 weeks post birth, at which point infected mice were randomly separated into the following groups (n = 9 mice per group for celecoxib + vinblastine treatments; n = 8 mice per group for remaining treatments): nontreated control; celecoxib chow (1600 mg/kg ad libitum); vinblastine (0.5 µg/g i.p., twice weekly); celecoxib-vehicle chow (1600 mg/kg ad libitum); celecoxib + vinblastine coadministration (1600 mg/kg ad libitum + 0.5 µg/g i.p. twice weekly); vehicle-containing chow + vinblastine (1600 mg/kg ad libitum + 0.5 µg/g i.p., twice weekly); celecoxib + vinblastine vehicle (1600 mg/kg ad libitum + equivalent volume of Ringer's solution alone). Uninfected control mice were also given injections of vinblastine vehicle (Ringer's solution) twice weekly. Weights were recorded at regular time intervals.
Determination of Plasma Celecoxib Concentration. Randomly selected, moribund mice were anesthetized using inhalation anesthesia at 1 L O2/min with 2% isoflurane and 1 mL of blood was obtained by terminal bleed using a direct cardiac puncture. Blood was then transferred to Microtainer plasma separator tubes with lithium heparin (Becton Dickinson). Samples were centrifuged at 14,000 x g for 5 minutes and plasma was collected into sterile 1.5-mL Eppendorf tubes. All samples were kept at –70°C until analyzed. Plasma samples were thawed and analyzed for celecoxib by high-pressure liquid chromatography (HPLC) according previously described methods (19). Briefly, 300 µL of plasma was treated with 100 µL of 1.0 N phosphoric acid and SC-751 (Pharmacia Corporation, St. Louis, MO) was added as the internal standard. Celecoxib and SC-751 were extracted using Bond Elute Certify 130-mg solid phase extraction columns (Varian Canada, Mississauga, Ontario, Canada). All columns were preconditioned with 2 x 1 mL of acetonitrile followed by 2 x 2 mL of water. Samples were eluted using 2 mL of 0.6% ammonium hydroxide in methanol. Extracts were evaporated to dryness using nitrogen and the residue was reconstituted in 250 µL of HPLC mobile phase. Separation was achieved by injecting 150 µL of sample onto a 150 x 4.6-mm Prodigy ODS3 column (5 µm) equipped with a C18 SecurityGuard cartridge (Phenomenex, Torrence, CA). Celecoxib and SC-751 were detected using fluorescence detection (Shimadzu Corporation; Kyoto, Japan; excitation = 240 nm, emission = 380 nm). The mobile phase consisted of acetonitrile and 0.01 mol/L sodium phosphate in a 1:1 ratio (v/v) and run at 1 mL/min. An untreated mouse was sampled whose plasma was analyzed in a similar manner as a negative control.
Body Weights, Hematocrit Measurements, and Blood Sampling. Body weight and overt toxicity of all animals were documented twice weekly, that is, at time of treatments. Unless otherwise indicated, mice were sacrificed at term as per institutional guidelines. Peripheral blood was collected on a weekly basis by tail clipping and smear preparations were prepared by wedge slide technique followed by standard Wright's staining. Whole blood (
1 mL) was then collected by cardiac puncture with a 1-mL U-100 insulin syringe (Becton Dickinson) into a K2-EDTA-coated, 4-mL Vacutainer (Becton Dickinson). Automated and manual peripheral differential cell counts were done to confirm presentation of erythroleukemic blasts. As well, both peripheral smears and automated differentials were done to ascertain the presence of malignant erythroblasts. Hematocrits were measured weekly by clipping tails and collecting 50 to 60 µL of blood into heparinized capillary tubes, followed by centrifugation at 14,000 x g in an Ultra microhematocrit centrifuge.
Immunohistochemistry. Splenic size was assessed by weight and vernier caliper measurement before tissue processing. Splenic tissues of healthy and erythroleukemic, moribund mice (n 
3) from the indicated treatment groups were sectioned (Leica CM3050) to a thickness of 7 µmol/L and placed onto ProbeOn Plus microscope slides (FisherBiotech, Mississauga, Ontario, Canada) and fixed with cold acetone and 1.5% hydrogen peroxide/methanol. After blocking nonspecific sites with 10% rabbit serum, purified anti-mouse CD31 [platelet/endothelial cell adhesion molecule 1 (PECAM-1); PharMingen, San Diego, CA] was incubated with a 1:400 dilution in antibody diluting buffer (Dako Diagnostic Canada, Mississauga, Ontario, Canada) overnight at 4°C. Anti-mouse IgG + IgM (H + L, Jackson ImmunoResearch Laboratories, Kirkland, Quebec, Canada) in a dilution of 1:200 served as a negative control. The following day, all specimens were washed with PBS and incubated with 1:200 dilutions of anti-rat IgG (Jackson ImmunoResearch Laboratories) for 30 minutes at room temperature. Parameters set out by streptavidin/biotin labeling via HistoStain SP kit (Zymed Laboratories, San Francisco, CA), using purified anti-mouse CD31 (PECAM-1; PharMingen) were followed.
Survival and Data Statistical Analyses. Survival among all groups was computed and plotted according to the nonparametric Kaplan-Meier analysis. Comparison of monotherapy and combination therapy was made using the nonparametric Mann-Whitney U test with two-sided P values and significant differences reflecting a P < 0.05. Student's t test for one and two populations for body weights and hematocrit values were conducted within a given treatment group and between treatment groups, respectively (P < 0.05).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
12% survival) and this trend was maintained for the duration of the experiment in favor of combination therapy.
|
|
|
Conserved Splenic Architecture in Erythroleukemic Mice Given Celecoxib + Vinblastine Therapy
Splenomegaly is a hallmark presentation of murine erythroleukemia, resulting in splenic rupture and death in a large proportion of mice. When compared with healthy controls histologic inspection of spleens from vinblastine monotherapy revealed relatively homogeneous infiltration of blasts (Fig. 4A and B). Celecoxib-treated mice exhibited similar pathology, with dispersed, poorly defined germinal centers (Fig. 4C), but spleens from mice on combination therapy indicated considerable maintenance of healthy architecture, as depicted by the appearance of well-defined germinal centers with a prominent vasculature lining the mantle regions (Fig. 4 D). Despite these observations in the combination therapy group, the extent of splenomegaly was similar across all groups (data not shown).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although the concept of low-dose metronomic chemotherapy has become better understood among clinicians and scientists worldwide, the doses applicable for the treatment of leukemias are yet to be established. Given the recent works conducted on solid tumors with respect to effective concentrations against endothelial cells, but not tumor cells, we decided to adopt a dosing regimen of vinblastine previously shown to be effective in the sustained regression of solid tumors (15–17). We found that in young mice a low-dose regimen of vinblastine, if given as a single agent, resulted in significant toxicity. The early mortality in this group was not due to leukemic burden but rather to gastrointestinal and possibly other organ damage. Age-matched mice of the remaining treatment groups, particularly combination therapy, did not exhibit these side effects or significant weight loss when compared with healthy controls over the duration of the experiment. It is therefore quite likely that in young mice the use of antiangiogenic therapy may be associated with increased morbidity, or that in the early postnatal period the doses necessary to achieve antiangiogenic effects may be even lower than anticipated (21). We have observed that BALB/c neonates infected with F-MuLV experience a high rate of mortality at approximately 9 weeks post infection. Similarly, we have observed the greatest rate of mortality at 10 weeks among all treatment groups (median survival time of 68 days), with the exception of the combination-treated mice. Coadministration of celecoxib with low-dose vinblastine alleviated all such toxic side effects and resulted in a significant overall increase in survival. In subsequent studies we have observed that these toxicities are age dependent and do not occur if treatment commences once mice have reached a mean weight of 14 g.7 In this study, however, treatments began at time of weaning (21 days) and resulted in a mean weight loss of 8 g (20% of total body weight) across all groups. Our study supports the advantages of coadministering a selective COX-2 inhibitor to a modified dosing protocol using a conventional chemotherapy, at least in juvenile mice with Friend disease, as it seems to increase tolerance toward potential vinblastine toxicity. Whether this may be the case with other conventional chemotherapeutics in the treatment of hematologic malignancies remains to be tested.
It has been shown that the median survival time after induction of chemotherapy for erythroleukemia patients is approximately 4 months (22). A past case report made reference to the monitoring of a patient with erythroleukemia who was given indomethacin postchemotherapeutic intervention who eventually acquired stable disease (9). In our model of Friend disease we found that celecoxib monotherapy was relatively ineffective in extending survival, and only when coadministered with low-dose vinblastine was survival enhanced. It is possible that similar to our juvenile mice, the patient within the case report experienced the protective effects of COX-2 inhibition. Because these agents inhibit both cyclooxygenase isoforms, we cannot rule out the possibility that COX-1 may also be implicated, such that perhaps the use of combinations of nonspecific inhibitors with low-dose vinblastine could have further enhanced the survival profile observed.
In addition to the above, alternative targets for COX-2-specific NSAIDs have come into question recently (23), and we too have observed that the growth kinetics of murine erythroleukemic blasts overexpressing COX-2 is significantly inhibited when celecoxib is added to the growth medium (13). However, the inhibition of COX-2 at lower concentrations of drug did negatively affect the growth of cells in vitro; rather, the disruption of signaling pathways downstream of Epo and c-Kit receptors at higher concentrations of celecoxib was responsible for these inhibitory observations (13). In the past, pathologic assessment and mortality from erythroleukemia have been documented by way of correlating survival with age, sex, hepatomegaly, lymphodenopathy, infection or hemorrhaging, and splenomegaly (2). Interestingly, based on the results of a retrospective study of 134 patients, the spleen remains the most highly investigated organ in the pathology of erythroleukemia, and the degree of splenomegaly serves as a reliable prognostic factor in some, but not all, hematologic diseases. The spleen is a reflection of tumor burden in hematologic malignancies, but whether this is due primarily to its distensible nature or to a favorable microenvironment and extramedullary hematopoiesis remains elusive. In Friend disease, splenectomized mice did not develop leukemia despite viral infection (24), suggesting a principal role of splenic environment in this disease. Our staining for PECAM suggests a more preserved splenic architecture with combination treatment over that of either vinblastine or celecoxib alone. Similarly to the healthy controls, splenic mantle regions from mice given both celecoxib and low-dose vinblastine show a peripheral vasculature with a highly cellular core. This is in contrast to the homogeneous replacement of the splenic parenchyma by infiltrating blasts from mice given either vinblastine or celecoxib alone.
A potential explanation of the mechanism of action of celecoxib may be in the defined role of the E family of prostaglandins in the suppression of immunogenic responses in favor of TH2 (25–27). It has been observed that the phenotypically suppressed lymphocytes collected from Hodgkin's patients become markedly activated in the presence of indomethacin (28). Although the presence of celecoxib alone in our murine model of erythroleukemia did not result in significant survival advantage, in combination with low-dose vinblastine it was seen to be most effective. Thus, celecoxib could be driving a TH1 cell–mediated immune response by decreasing prostaglandin E, thus increasing antigen presentation by the mutated erythroblast and leading to recognition and eradication of blasts by supra-activated T cells. Alternatively, there may exist a currently uncharacterized synergy between low-dose vinblastine and celecoxib, which controls the growth of malignant leukemic blasts.
In this study we have examined two very important issues related to the treatment of erythroleukemia. First, we have tested the suitability of using a viral-induced hematologic malignancy as a preclinical model of the human AML variant, M6. Second, we have tested the response of this model to a modified clinical regimen incorporating a selective COX-2-inhibiting NSAID with metronomic antiangiogenic dosing of a conventional chemotherapeutic drug. The premise of this work is based on the successes of several recent studies in various solid tumors, whereby the use of each regimen alone has revealed significant antitumor effects by both direct and indirect mechanisms. Coupled with an antiangiogenic dosing regimen of vinblastine we therefore have shown considerable efficacy in the use of such a combination therapy for the treatment of aggressive hematologic malignancies. The toxicity of this regimen in early postnatal period could be counteracted by combination therapy. Recent investigations in colorectal cancer models support celecoxib adjuvant therapy, as it has been shown to exhibit considerable antitumor effects in addition to decreasing the diarrhea side effect associated with topoisomerase inhibitors (26). Similarly, considerable benefits in using COX-2 inhibitors have been reported whereby the response of tumors to both radio- and chemoradiotherapy was enhanced, along with improving tumor growth inhibition (26). Most importantly, decreased damage of normal tissue was reported. Even though erythroleukemia represents only 1% to 3% of all leukemias, its aggressive nature and chemotherapy resistance makes for improved prognosis and alternative therapies a necessity (22, 29–31). As such, the antiangiogenic metronomic low-dose chemotherapy with its lowered toxic profile and improved efficacy provides a valid therapeutic alternative for these patients.
|
FOOTNOTES |
|---|
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.
Note: D. Cervi and G. Klement contributed equally to this work.
6 D. Cervi, Y. Shaked, Y. Ben-David, unpublished material. 
7 D. Cervi, G. Klement, Y. Ben-David, unpublished material.
Received 7/ 6/04; revised 10/21/04; accepted 10/28/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. Baryawno, B. Sveinbjornsson, S. Eksborg, A. Orrego, L. Segerstrom, C. O. Oqvist, S. Holm, B. Gustavsson, B. Kagedal, P. Kogner, et al. Tumor-growth-promoting cyclooxygenase-2 prostaglandin E2 pathway provides medulloblastoma therapeutic targets Neuro-oncol, January 1, 2008; 10(5): 661 - 674. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Lilenbaum, M. A. Socinski, N. K. Altorki, L. L. Hart, R. S. Keresztes, S. Hariharan, M. E. Morrison, R. Fayyad, and P. Bonomi Randomized Phase II Trial of Docetaxel/Irinotecan and Gemcitabine/Irinotecan With or Without Celecoxib in the Second-Line Treatment of Non-Small-Cell Lung Cancer J. Clin. Oncol., October 20, 2006; 24(30): 4825 - 4832. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Shaked, D. Cervi, M. Neuman, L. Chen, G. Klement, C. R. Michaud, M. Haeri, B. J. Pak, R. S. Kerbel, and Y. Ben-David The splenic microenvironment is a source of proangiogenesis/inflammatory mediators accelerating the expansion of murine erythroleukemic cells Blood, June 1, 2005; 105(11): 4500 - 4507. [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 |
