| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Department of Oncology [S. D., R. D. V., A. J. G.] and Medical Research Council Toxicology Unit [P. G., T. W. G.], University of Leicester, Leicester, LE1 9HN, United Kingdom; Istituto di Ricercha Farmacologiche Mario Negri, 1-20157 Milan, Italy [T. C., Mar. Z., R. F., Mas. Z., M. D.]; Policlinico Universitario A. Gemelli, 1-00168 Rome, Italy [D. M., R. R.]; and PharmaMar SA, E-28770 Colmenar Viejo, Spain [L. L. L., J. J.]
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
We wished to explore experimental strategies aimed at the amelioration of drug-induced hepatotoxicity to aid the additional clinical development of this promising anticancer drug. On the basis of preliminary clinical observations (13) , we reasoned that the steroid dexamethasone might afford protection against yondelis-induced hepatotoxicity. This hypothesis was tested in female rats, the animal species that has been shown to be most susceptible toward the adverse hepatic side effect of yondelis (6 , 7) . Furthermore, the effect on antitumor activity of combining yondelis with dexamethasone was explored in five rodent models. To gain insight into the mechanism by which dexamethasone may affect yondelis-mediated hepatotoxicity, we measured levels of yondelis in the liver and plasma of rats pretreated with dexamethasone.
| MATERIALS AND METHODS |
|---|
|
|
|---|
/
, and CDI
/
strains (used in the antitumor activity studies) were obtained from Charles River Laboratories (Margate, United Kingdom or Rome, Italy). Yondelis formulated for injection originated from the drug manufacturer PharmaMar SA (Colmenar Viejo, Spain). The specific CYP3A substrate 7-benzoyloxyquinolone and its metabolite 7-hydroxyquinolone were obtained from BD Biosciences (Oxford, United Kingdom), all other reagents from Sigma-Aldrich Co. Ltd. (Poole, United Kingdom).
Study of Hepatotoxicity.
Female Wistar rats (230260 g) were pretreated with a single dose of dexamethasone (520 mg/kg, dissolved in glycerol formal, p.o.; Ref. 14
) 24 h before, or simultaneous with, a hepatotoxic dose of yondelis (40 µg/kg, i.v. via the lateral tail vein). Control animals received the vehicles used, i.e., glycerol formal in the case of dexamethasone, and water as the vehicle for yondelis. Each treatment group comprised 5 animals. Hepatic changes were studied by assessment of alterations in plasma levels of bilirubin and liver enzymes ALP and AST, and by conventional histopathological investigation of liver tissue as described previously (7)
. Experiments were conducted as stipulated by Project License 80/1250 granted by the United Kingdom Home Office, and the experimental design was vetted and approved by the Leicester University Ethical Committee for Animal Experimentation.
Study of Antitumor Activity.
The 13762 tumor was propagated in female Fischer rats. Tumor fragments (100200 mg weight) were implanted (s.c.) into the flank of rats (100120 g). Female C57Bl/6J mice (20 ± 2 g body weight) received 106 B16F1 melanoma cells s.c. or 105 M5076 reticulum cell sarcoma cells i.m. Female MCr-
/
mice (22 ± 2 g body weight) received 5 x 106 Igrov/1 ovarian carcinoma cells s.c. Male CDI
/
mice (20 ± 2 g body weight) received 1.5 x 106 TE-671 rhabdomyosarcoma cells s.c. Properties of the tumor models have been described in the following representative references: 13762 (15)
, B16F1 (16)
, M5076 (17)
, Igrov/1 (18)
, and TE 671 (19)
. Rats (8/group) received dexamethasone (10 mg/kg, i.p.) on day 9 after 13762 tumor implantation. Yondelis (30 or 40 µg/kg, i.v.) was administered 24 h after dexamethasone. Thus, the treatment protocol in the rat tumor model mimics faithfully the protocol used in the hepatoprotection studies. In the murine models mice (10/group) received repeated doses of dexamethasone (20 or 40 mg/kg i.p.) daily between days 7 and 24 after tumor implantation. Yondelis (0.15 or 0.2 mg/kg, i.v.) was administered once, twice, or three times between days 10 and 25 after tumor implantation. The specific dosing schedules used in each model are listed in the legend to Table 1
. TW was determined on days 14 and 17 after tumor implantation in rats, and between days 23 and 34 after tumor implantation in mice, as detailed for each model in Table 1
. TW was calculated via tumor diameter, using a Vernier caliper and the formula
![]() |
|
4700 hybridizable mouse ESTs derived from IMAGE clones obtained from Research Genetics (Huntsville, AL) or from the Medical Research Council Human Gene Mapping Project.4
One array was used for each pair of rats with the individual RNAs being labeled with Cy3 or Cy5. The labels were reversed for subsequent hybridizations. Microarray preparation, RNA labeling, and hybridization were performed as described previously (7
, 20)
.
Study of CYP3A Function.
Livers of control and dexamethasone-pretreated rats were homogenized in Tris-KCl buffer (50 mM) containing sucrose (0.25 M; pH 7.4). For isolation of liver microsomes the homogenate was centrifuged (10,000 x g, 20 min, 4°C), and the supernatant was then removed and spun at 100,000 x g for 60 min at 4°C. The microsomal pellet was resuspended in fresh buffer and recentrifuged (100,000 x g) for an additional60 min. The resulting pellet was suspended in 0.25 M phosphate buffer containing 30% glycerol, stored at -80°C, and thawed before analysis. Microsomal protein concentration was determined with the Bradford assay. The ability of the microsomes to metabolize the CYP3A model substrate 7-benzoyloxyquinolone to 7-hydroxyquinolone was tested as described by Renwick et al. (21)
.
Study of ET-743 Levels in Liver and Plasma.
Blood samples and liver tissue were collected from rats before and up to 72 h after administration of yondelis (40 µg/kg i.v.) with or without pretreatment with dexamethasone (10 mg/kg p.o.) given 24 h before yondelis. Blood samples were taken by cardiac puncture and placed in heparinized tubes. Plasma was obtained by centrifugation, and an aliquot (0.3 ml) was mixed with 0.7 ml ammonium acetate buffer (0.2 M; pH 5.0). The mixture was kept on ice (30 min), centrifuged (1000 x g, 10 min) and subjected to solid phase extraction (Bond Elut nonend-capped CN cartridge). Methanolic hydrochloric acid (0.1 M, 2.5 ml) served as eluant. The elution solvent was evaporated, the residue dissolved in mobile phase (200 µl), and an aliquot (10 µl) was injected onto the HPLC column. Liver tissue was homogenized (1:1) in water. An aliquot (0.5 ml) of the homogenate was extracted with 10 parts of hydrochloric acid (0.1 M) in methanol. After shaking for 30 min, samples were centrifuged (1500 x g, 4°C, 10 min). The supernatant was dried under nitrogen, and the residue was dissolved in mobile phase (200 µl, acetic acid 0.1% in water:acetonitrile 90:10). Reconstituted liver extract was filtered through a PTFE filter (Alltech, Deerfield, IL), and an aliquot (10 µl) was injected onto the HPLC column. ET-729 served as internal standard in all of the analytical assays, and in orientation experiments ET-729 was not detected as a metabolite of yondelis in any of the tissues studied. Yondelis was measured by HPLC coupled to electrospray ionization tandem mass spectrometry, as described by Rosing et al. (22)
using an API 3000 triple quadrupole mass-spectrometer (Applied Biosystems-Sciex, Toronto, Ontario, Canada) operating in positive ion mode (standard TurboIonSpray source). Separation were achieved on a Luna C18 column (Phenomenex, Torrance, CA), and a mobile phase of acetic acid (0.1%) in water (solvent A) and acetonitrile (solvent B), with gradient elution of B from 10 to 70% in 10 min, at the flow rate of 200 µl/min. Calibration curves of standard solutions of yondelis (0.0525 ng/g) were prepared using plasma or liver homogenate from untreated mice. The correlation between concentration and peak response in these extracts analyzed on three different days was characterized by r2 > 0.999. Quality control samples of plasma (spiked with 0.1, 0.5, and 2 ng yondelis/ml) or liver homogenate (from livers spiked with 1, 5, and 20 ng yondelis/g tissue) afforded limits of quantitation of 0.05 ng/ml for plasma and 0.5 ng/g for liver; the precision between runs was reflected by correlation coefficients of ±5% (plasma) and ±10% (liver), and the accuracy was in the range of 97105% (plasma) and 90110% (liver).
The AUC for yondelis was calculated by the trapezoideal rule, between time point 0 and 6 h, the last time point at which yondelis was quantifiable in the plasma.
| RESULTS |
|---|
|
|
|---|
|
|
Effect of Simultaneous Treatment with Dexamethasone on Yondelis-induced Liver Changes.
Dexamethasone in the 0.010.03 mg/kg dose range is often administered in the clinic at the same time as cytotoxic drugs to prevent drug-induced emesis. Therefore, coadministration of yondelis with high-dose dexamethasone as a potential hepatoprotectant and antiemetic would clearly be clinically expedient. However, when coadministered with yondelis, dexamethasone (10 mg/kg) failed to protect rat livers against yondelis toxicity. In animals that received the combination, plasma ALP and bilirubin levels were as elevated as those seen in rats on yondelis only. Moreover, plasma AST activity in rats on the combination was 4-fold higher than that measured in rats treated with yondelis alone; AST activities (in units per liter) were 85 ± 12 in control rats, 155 ± 19 in rats on yondelis alone, and 615 ± 116 in rats on the combination (mean ± SD; n = 4). Likewise, simultaneous dosing of dexamethasone with yondelis failed to abrogate the histopathological changes induced by yondelis alone. Swelling of the portal tract was greater in 2 of 4 rats that received the combination, and the livers of these rats showed many more foci of hemorrhagic hepatocellular necrosis than did livers of rats on yondelis alone.
Effect of Pretreatment with Dexamethasone on Antitumor Activity of Yondelis.
To justify exploration of the dexamethasone hepatoprotection strategy in the clinic, it is of paramount importance to demonstrate that dexamethasone does not adversely affect the antitumor activity of yondelis. This hypothesis was tested initially in female Fischer rats bearing the 13762 mammary carcinoma. Fig. 3
shows that both dexamethasone (10 mg/kg, i.p.) and yondelis (40 µg/kg, i.v.) individually exhibited activity in this model, and that the combination was not inferior to that of the single agents. In the case of tumor evaluation on day 14, the percentage TW inhibition of the combination was 68 or 36% higher than the values observed with yondelis or dexamethasone, respectively, alone. Very similar results as those shown in Fig. 3
were obtained when the dose of yondelis was reduced to 30 µg/kg. To explore the generic nature of the antitumor results, we investigated the activity of the combination of yondelis with dexamethasone pretreatment also in murine tumor models of melanoma (B-16), reticulum cell sarcoma (M 5076), ovarian carcinoma (IGROV-1), and rhabdomyosarcoma (TE-671). A variety of dosing regimes, and dose levels of yondelis and dexamethasone was used to account for differential sensitivities of the individual models and to allow discrimination between efficacies of the individual constituents (Table 1)
. Dexamethasone exhibited significant antineoplastic activity on its own in the murine B16 melanoma and IGROV-1 ovarian tumor models. In none of the rodent models investigated was the combination less efficacious than yondelis administered on its own. Moreover, in three of the models investigated, the 13762 (Fig. 2)
, B16, and TE-671 models (Table 1)
, the combination was clearly more efficacious than yondelis alone.
|
|
Effect of Dexamethasone on Hepatic Cytochrome P450 3A Enzymes.
In a series of additional experiments biochemical changes in the liver elicited by dexamethasone, which may contribute to the mechanism by which it protects the rat liver against yondelis-inflicted damage, were explored. Hepatic gene expression was analyzed using cDNA microarray and the ESTs described above. Livers were obtained from pairs of untreated rats and rats that had received dexamethasone (10 mg/kg) 1 day before RNA isolation, so that the analysis was performed exactly at the time point at which yondelis would have been injected in the experiments in which the combination was studied. Dexamethasone up-regulated the expression of eight genes by 2.5-fold or more, and these included genes involved in fatty acid synthesis, gluconeogenesis, and carbohydrate metabolism. Most dramatic induction was observed for the cytochrome P450 genes cyp3a11 (23.2-fold induction), cyp3a16 (18.1-fold), cyp3a25 (13.2-fold), and cyp3a13 (10.2-fold). To validate at the functional level the induction of CYP3A gene expression by dexamethasone, liver microsomes of dexamethasone-pretreated rats were obtained, and their ability to metabolize the model substrate 7-benzoyloxyquinolone to 7-hydroxyquinolone was tested. At 10 mg/kg dexamethasone increased the rate of oxidative debenzoylation of the model substrate 6.4-fold, from 0.021 ± 0.004 to 0.134 0 ± 0.010 µmol/min/mg microsomal protein (mean ± SD, n = 3). This result suggests that the increase in cyp3a gene expression elicited by dexamethasone in the rat liver predicts elevated CYP3A enzyme activity. To additionally support the role of CYP3A induction in the dexamethasone-mediated hepatoprotection, another experiment was conducted in which the metabolic removal of yondelis by hepatic microsomes was investigated in vitro. Whereas microsomes from control rats removed 46% of yondelis substrate (1 µM) within 20 min, microsomes from dexamethasone-pretreated rats metabolized 92% of the substrate under these conditions. Therefore, it is conceivable that dexamethasone-mediated induction of CYP3A contributes to the protection by the glucocorticoid against the hepatotoxicity of yondelis in rats.
Effect of Pretreatment with Dexamethasone on Levels of Yondelis in Liver and Plasma.
The experiments described above hint at the possibility that dexamethasone-induced CYP3A enzymes in rat liver oxidize yondelis to a nonhepatotoxic species at a faster rate than livers of rats that have not been pretreated with dexamethasone. If this hypothesis was correct, one may expect that increased metabolism of yondelis after pretreatment with dexamethasone at hepatoprotective doses leads to decreased yondelis levels in the liver compared with levels observed after yondelis alone. Levels of yondelis were measured in liver and plasma of rats that had been pretreated with an optimally hepatoprotective dose of dexamethasone (10 mg/kg), and these levels were compared with those in animals that received yondelis alone. Fig. 5
shows that levels of yondelis in livers of animals pretreated with dexamethasone were dramatically lower than those observed in livers from animals that did not receive the corticosteroid. Whereas liver levels in rats 1 h after administration of yondelis alone were as high as 37 pmol/g of tissue, such levels in dexamethasone-pretreated animals did never exceed 5 pmol/g of tissue. So dexamethasone pretreatment reduced hepatic yondelis levels to 14% of those in unpretreated rats 1 h after dosing. The mean values for the AUCliver (between 0 and 72 h after dosing) were 501 and 108 pmol/g x h (n = 4) for animals that received yondelis without pretreatment or after dexamethasone, respectively. In contrast, dexamethasone pretreatment neither changed the shape of the yondelis plasma concentration versus time curve (data not shown) nor decreased systemic exposure to yondelis, as adjudged by the AUCplasma (between 0 and 6 h after dosing); mean AUC values were 2.2 and 2.3 pmol/ml x h (n = 4) in animals that received yondelis alone or after dexamethasone pretreatment, respectively. In a confirmatory experiment the effect on yondelis plasma disposition of a higher dose of dexamethasone (50 mg/kg) given on 3 consecutive days preceding yondelis was investigated. Under these conditions plasma AUC values of yondelis in unpretreated and dexamethasone-pretreated rats were 1.3 and 2.2 pmol/ml x h (n = 4), respectively. Thus, yondelis levels in dexamethasone-pretreated rats were not lower than those obtained in animals on yondelis alone.
|
| DISCUSSION |
|---|
|
|
|---|
Dexamethasone possesses pleiotropic pharmacological activities including activation of many transcription factors and anti-inflammatory stimuli, many of which could conceivably contribute to its ability to protect rat livers from the adverse effect of yondelis. The findings presented above permit a preliminary insight into the potential mechanism by which dexamethasone protects the rat liver against yondelis. Firstly, pretreatment with the corticosteroid engendered a dramatic suppression of hepatic yondelis levels and, thus, effectively decreased hepatic exposure to the drug. Secondly, this suppression may have been the corollary of the propensity of dexamethasone to induce CYP3A enzymes and, thus, to increase the metabolic removal of yondelis. Consistent with a hepatoprotective role of CYP3A induction are results obtained after pretreatment of rats with phenobarbital or indole-3-carbinol, the former an inducer of several CYPs including CYP3A, the latter a specific CYP3A inducer. Whereas phenobarbital did not protect livers from the deleterious effect of yondelis, indole-3-carbinol (0.1 or 0.5% in the diet) was, like dexamethasone, hepatoprotective.5 Yondelis is a substrate of CYP3A4 in vitro (23 , 24) , although rigorous data to support the relevance of this enzyme in the disposition and/or antitumor activity of yondelis in vivo is still lacking. Human cell microsomes that express CYP3A4 have been shown to metabolize yondelis to three species, N-desmethylyondelis (ET-729) and two molecules generated by oxidative degradation of the drug (24) . It is conceivable that the accelerated generation of some of these species, or of other nontoxic metabolites, is responsible for the hepatic depletion of yondelis in rats that have been pretreated with dexamethasone. In our study yondelis metabolites were not investigated; therefore, we cannot discard alternative explanations, such as that dexamethasone exerts its effect, at least in part, via increased bile flow and consequent elevated rate of yondelis secretion. Nevertheless it is important to stress that dexamethasone pretreatment failed to decrease yondelis plasma levels, compared with those observed after administration of yondelis alone. This finding suggests that the systemic availability of yondelis is not confounded by high-dose dexamethasone, which, in turn, is consistent with the uncompromised antitumor activity of the combination compared with that of yondelis alone. Furthermore, the findings taken together suggest that, whereas dexamethasone pretreatment decreases hepatic levels of yondelis effectively, it does not interfere markedly with the step(s) that determine(s) the rate at which yondelis is eliminated from the systemic circulation. It is conceivable that yondelis levels in the liver represent such a relatively small fraction of the total dose of yondelis in the body, that an increase in hepatic clearance caused by dexamethasone failed to affect its plasma AUC markedly.
It needs to be emphasized that the results described here have been obtained in the female rat, the animal model with the highest sensitivity toward yondelis-induced liver changes. Manifestations of damage induced by yondelis in human liver are less severe than those reported in the female rat (7) , and hepatic changes are generally reversible in patients. Therefore, it is conceivable that details of the mechanism by which yondelis induces hepatic damage differ between rats and humans, and that such differences in turn engender discrepant susceptibilities toward amelioration by dexamethasone. For that reason the conclusions drawn from the results obtained in rats should be extrapolated to the clinic only with extreme prudence. High-dose dexamethasone therapy is feasible, as it is used clinically in conditions in which immunosuppression and anti-inflammation are desired, such as systemic lupus erythematosus, renal transplantation, steroid-resistant nephrotic syndrome, and crescentic glomerulonephritis. Long-term use of dexamethasone can lead to diabetes mellitus, especially in older patients, and other adverse effects of high-dose dexamethasone include hypertension, arrhythmias, hypokalemia, psychosis, and susceptibility to infection, but these effects are often manageable (25 , 26) .
In conclusion, the results of the experiments described above emphatically support the notion that pretreatment with a single high dose of dexamethasone prevents the deleterious effects of yondelis on function, structure, and gene expression in the rat liver without compromising its experimental antitumor activity. The timing of the pretreatment seems to be crucial for optimal protective potency, with a 24-h pretreatment interval being apparently most propitious, and simultaneous dosing probably enhancing toxicity. These results provide a powerful rationale to evaluate carefully the potential clinical usefulness of pretreatment with high-dose dexamethasone in cancer patients who receive yondelis, with the aim to suppress its unwanted hepatic effects.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 Supported in part by Medical Research Council Toxicology Unit, University of Leicester, core grant funding and a grant from PharmaMar SA, Colmenar Viejo, Spain. The project is a collaboration under the auspices of the European Organization for Research and Treatment of Cancer Pharmacology and Molecular Mechanisms (PAMM) Group. R. F. was supported by a fellowship from Federazione Italiana Ricerca Cancro. ![]()
2 To whom requests for reprints should be addressed, at Department of Oncology, Robert Kilpatrick Clinical Sciences Building, Leicester Royal Infirmary, University of Leicester, Leicester, LE2 7LX, United Kingdom. Phone: 44-116-223-1856; Fax: 44-116-223-1855; E-mail: ag15{at}le.ac.uk ![]()
3 The abbreviations used are: ALP, alkaline phosphatase; AST, aspartate aminotransferase; TW, tumor weight; EST, expressed sequence tag; HPLC, high-performance liquid chromatography; AUC, area under the plasma concentration-versus-time curve. ![]()
4 Internet address: http://www.le.ac.uk/mrctox/microarray_lab/. ![]()
5 Donald, Verschoyle, Greaves, Orr, Jimeno, Gescher, unpublished observations. ![]()
Received 1/24/03. Revised 6/19/03. Accepted 6/30/03.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. A.B. van Waterschoot, R. M. Eman, E. Wagenaar, C. M.M. van der Kruijssen, H. Rosing, J. H. Beijnen, and A. H. Schinkel ABCC2, ABCC3, and ABCB1, but not CYP3A, Protect against Trabectedin-Mediated Hepatotoxicity Clin. Cancer Res., December 15, 2009; 15(24): 7616 - 7623. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Fayette, I. R. Coquard, L. Alberti, D. Ranchere, H. Boyle, and J.-Y. Blay ET-743: A Novel Agent with Activity in Soft Tissue Sarcomas Oncologist, November 1, 2005; 10(10): 827 - 832. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Sessa, F. De Braud, A. Perotti, J. Bauer, G. Curigliano, C. Noberasco, F. Zanaboni, L. Gianni, S. Marsoni, J. Jimeno, et al. Trabectedin for Women With Ovarian Carcinoma After Treatment With Platinum and Taxanes Fails J. Clin. Oncol., March 20, 2005; 23(9): 1867 - 1874. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Lau, J. G. Supko, S. Blaney, L. Hershon, N. Seibel, M. Krailo, W. Qu, D. Malkin, J. Jimeno, M. Bernstein, et al. A Phase I and Pharmacokinetic Study of Ecteinascidin-743 (Yondelis) in Children with Refractory Solid Tumors. A Children's Oncology Group Study Clin. Cancer Res., January 15, 2005; 11(2): 672 - 677. [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 |