Journal of the National Cancer Institute Advance Access originally published online on November 13, 2007
JNCI Journal of the National Cancer Institute 2007 99(22):1724-1728; doi:10.1093/jnci/djm202
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author 2007. Published by Oxford University Press.
Effect of Pretreatment With Atenolol and Nifedipine on ZD6126-Induced Cardiac Toxicity in Rats
Affiliation of authors: AstraZeneca, Mereside, Alderley Park, Macclesfield, Cheshire, U.K
Correspondence to: Anderson J. Ryan, PhD, Department of Cancer Bioscience, AstraZeneca, Mereside, Alderley Park, Macclesfield SK10 4TG, Cheshire, U.K. (e-mail: anderson.ryan{at}astrazeneca.com).
 |
ABSTRACT |
|---|
 Top Abstract Context and CaveatsReferencesNotes |
|---|
Antivascular agents that act by destabilizing microtubules, such as ZD6126 (N-acetylcolchinol-O-phosphate), are associated with adverse cardiovascular effects, including transient hypertension, cardiac ischemia, myocardial infarction, and increases in circulating levels of markers of cardiac damage (e.g., troponins). We investigated mechanisms underlying these effects of ZD6126 in rats by continuously monitoring their heart rate and blood pressure and by assessing heart histopathology and plasma troponin T levels. ZD6126 induced acute transient hemodynamic changes (hypertension and delayed tachycardia), which were associated with statistically significant increases in circulating troponin T levels (median level 3 hours after treatment with vehicle or 12.5 mg/kg ZD6126: <9 pg/mL and 563 pg/mL, respectively; P <.001 [two-sided Wilcoxon rank sum test]) and in the incidence of left ventricular myocardial fiber necrosis (incidence 24 hours after treatment with vehicle or 12.5 mg/kg ZD6126: 0/10 rats and 9/10 rats, respectively; P <.001 [two-sided Wilcoxon rank sum test]). Pretreatment of rats with atenolol and nifedipine ameliorated the acute hemodynamic changes and prevented ZD6126-induced increases in both troponin T and myocardial necrosis but did not prevent ZD6126-induced tumor necrosis in an Hras5 tumor xenograft model in nude rats. Our findings suggest that ZD6126-induced acute hemodynamic changes are a prerequisite for cardiac damage in rats.
Prior knowledge Antivascular agents selectively destroy the existing tumor vasculature. However, some vascular disrupting agents in early clinical development that act by destabilizing microtubules, such as ZD6126, are associated with adverse cardiovascular effects. Study design A rat preclinical model was used to examine the underlying mechanisms responsible for the adverse cardiac events associated with ZD6126. Contribution ZD6126 induces acute hemodynamic changes that are a prerequisite for cardiac damage in rats. Implications Clinical investigations of approaches that reduce the acute hemodynamic effects of microtubule-destabilizing vascular disrupting agents are warranted. Limitations Not all patients who receive ZD6126 experience cardiovascular adverse events, suggesting that the underlying biologic mechanisms in patients may differ from those in rats. The direct relevance of this rat model to patients treated with ZD6126 is unclear.
|
Vascular disrupting agents selectively destroy the existing tumor vasculature by inducing morphologic changes in tumor endothelial cells that lead to vessel occlusion and massive central tumor necrosis (1–3). ZD6126, a soluble prodrug of N-acetylcolchinol, is a vascular disrupting agent that acts by inhibiting tubulin polymerization, which destabilizes the microtubule network and induces morphologic changes in immature tumor endothelial cells (4–6). Several phase I studies have been carried out to identify the optimal dosing schedule of ZD6126 and to define its tolerability and pharmacokinetic profile (7–9). Dynamic contrast-enhanced magnetic resonance imaging has demonstrated substantially reduced tumor uptake of gadolinium contrast agent in ZD6126-treated mice and patients, a finding that is consistent with the selective vascular disrupting effects of ZD6126 in tumors (10). However, the phase I studies (7–9) reported that ZD6126 treatment was associated with adverse cardiovascular events that generally occurred within a few hours after ZD6126 was administered and comprised one or more of the following: electrocardiogram abnormalities, an increase in the circulating level of troponin I (a marker of cardiac damage), and/or arm or chest pain. ZD6126 treatment is also associated with a transient increase in blood pressure (9). Similar adverse cardiac events have been reported for other microtubule-destabilizing vascular disrupting agents that are in clinical development (11–13). In this study, we used a rat preclinical model to examine the underlying mechanisms responsible for the adverse cardiac events associated with ZD6126.
For analyses of ZD6126-induced hemodynamic changes, anesthesia was induced in female Wistar-derived Alpk:ApfSD rats (175–200 g; n = 9; AstraZeneca, Macclesfield, Cheshire, U.K.) with 3%–4% isoflurane (Abbott Laboratories, Chicago, IL) in medical-grade air and was maintained with 1.5%–2.5% isoflurane throughout the procedure. The abdominal cavity of each rat was opened, and the descending abdominal aorta was located and cleared of connective tissue. A tie placed underneath the vessel was lifted to occlude the vessel while the catheter of a DSI TA11PA-C40 radiotelemetry implant (Data Sciences International, St Paul, MN) was inserted into the aorta along the bevel of a bent 21-gauge needle. The catheter was held in place by suturing it to the inside of the abdominal wall. The surgical abdominal wound was then closed with dissolving sutures, and the outer skin was closed with autoclips. The rats were allowed to recover from surgery and acclimatized to the handling procedures required for dosing, and they were randomly assigned to one of three groups (n = 3 rats per group). Group 1 (the control group) received 1% polysorbate 80 by mouth at 150 minutes and, 1 hour later (i.e., at 210 minutes), 0.9% sodium chloride solution (physiological saline) by intravenous injection. Time zero (0 minutes) was arbitrarily defined as 6:00 AM. Group 2 received 1% polysorbate 80 by mouth at 150 minutes and ZD6126 (12.5 mg in physiological saline/kg) by intravenous injection at 210 minutes. Group 3 received atenolol (6 mg/kg) and nifedipine (10 mg/kg) (Sigma–Aldrich, Poole, U.K.), both by mouth at 150 minutes, each in 1% polysorbate 80, and ZD6126 (AstraZeneca) (12.5 mg/kg) by intravenous injection at 210 minutes. Blood pressure and heart rate data were collected via the radiotelemetry implant every 5 minutes beginning at 0 minutes for 24 hours. For each rat, data were collected as a moving point average over a 30-second period within each 5-minute interval.
Administration of ZD6126 to rats produced a rapid increase in diastolic blood pressure that lasted for 120–180 minutes, which was accompanied by a more prolonged and complex effect on heart rate (Fig. 1). ZD6126 induced tachycardia that was delayed with respect to the effects on diastolic blood pressure, possibly due, in part, to a reflex bradycardia resulting from normal homeostatic mechanisms acting to reduce the severity of the concurrent hypertension. When the acute hypertensive response abated 120–180 minutes after ZD6126 treatment, tachycardia became apparent and persisted for at least an additional 240 minutes. Preliminary experiments showed that direct vasodilation induced by an oral dose of the calcium channel-blocking agent, nifedipine, was effective in abolishing ZD6126-induced hypertension, whereas the 
-adrenoceptor antagonist atenolol was effective in limiting ZD6126-induced tachycardia (data not shown). Figure 1 shows that the combination of atenolol and nifedipine, given 1 hour before ZD6126 administration, was effective in abolishing both the hypertensive and tachycardic effects of ZD6126.
|
We next examined the effects of the three drugs, alone and in combination, on rat heart histopathology and plasma troponin T levels. Female rats (n = 60) were randomly assigned to one of the following six treatment groups (10 rats per group): physiological saline; ZD6126 (5 mg/kg); ZD6126 (12.5 mg/kg); atenolol (6 mg/kg) and nifedipine (10 mg/kg) alone; atenolol (6 mg/kg) and nifedipine (10 mg/kg) with ZD6126 (5 mg/kg or 12.5 mg/kg). When the three drugs were administered to the same rat, ZD6126 was given 1 hour after atenolol and nifedipine, as described previously. Blood samples (0.8 mL in lithium heparin) collected from each rat via the tail at 3, 6, and 24 hours after saline or ZD6126 dosing were subjected to a Troponin T Stat immunoassay (Roche Diagnostics, Lewes, U.K.) to determine the concentration of troponin T. Samples with values at or below the lower limit of detection for the assay (9 pg/mL) were recorded as 9 pg/mL. Data were analyzed for each time point compared with vehicle-treated control rats using a Wilcoxon rank sum test. All rats were subjected to terminal anesthesia by halothane inhalation 24 hours after saline or ZD6126 dosing, and their hearts were removed, fixed by immersion in neutral phosphate-buffered formalin, and bisected so that left and right ventricular and atrial tissues were present in both samples. Both halves of each heart were processed together and embedded in wax blocks. Step sections (5 µm) were taken at 300-µm intervals. Five hematoxylin-eosin–stained sections were prepared from each block for examination by light microscopy. The sections were examined in a blinded fashion by a pathologist (F. R. Westwood) for myocardial necrosis and associated histopathologic changes (i.e., hemorrhage and mixed inflammatory cell infiltration/fibroplasia), which were graded semiquantitatively as follows: minimal (a single or a few small areas); mild (up to 20% of a region of myocardium affected); moderate (more than 20% up to 50% of the myocardium affected); or severe (more than 50% of the myocardium affected). For each rat heart, the most severe change on any of the five step sections for each of the right and left ventricles (10 sections in total) was recorded for each of the histopathologic parameters. For statistical analysis of left ventricle necrosis data, each grade of histopathologic change was assigned a score of 1–4, where no change = 1, minimal change = 2, mild change = 3, and moderate change = 4. (No severe changes were seen in these experiments.) The scores for the hearts from rats within a treatment group were compared with the scores for vehicle-treated control rats using a Wilcoxon rank sum test. All statistical tests were two-sided.
Compared with hearts from vehicle-treated control rats, hearts from rats 24 hours after dosing with ZD6126 at 5 or 12.5 mg/kg showed a statistically significantly higher incidence of mild or moderate left ventricular myocardial necrosis (median scores in control, 5 mg/kg ZD6126, and 12.5 mg/kg ZD6126 groups were 1, 2.5, and 3, respectively; P<.001) and a higher incidence of left ventricular hemorrhage and mixed cell infiltration/fibroplasia (Table 1). Similar histopathologic changes (i.e., necrosis, hemorrhage, and mixed cell infiltration/fibroplasias) were seen in the right ventricle, although these appeared to be less severe. Similar data were obtained for hearts evaluated 72 hours after dosing and for rats given higher doses (up to 50 mg/kg) of ZD6126 (Supplementary Table 1, available online).
|
Histopathologic examination revealed that pretreatment of rats with atenolol and nifedipine before ZD6126 abolished all myocardial damage at both the 5 and 12.5 mg/kg doses of ZD6126 (Table 1). The single occurrence of minimal ventricular myocardial necrosis seen following administration of ZD6126 at 12.5 mg/kg after atenolol and nifedipine could not be clearly attributed to the administration of ZD6126 because it was observed in only one necrotic fiber in only one of 10 sections assessed. By contrast, the minimal necrosis we observed in the heart of one control rat involved multiple fibers.
Rats treated with ZD6126 at 5 or 12.5 mg/kg had statistically significantly higher median troponin T levels at 3 hours after ZD6126 treatment compared with vehicle-treated control rats (control = 9 pg/mL, 5 mg/kg ZD6126 = 274 pg/mL, and 12.5 mg/kg ZD6126 = 563 pg/mL; P<.001). Testing of a broad range of ZD6126 doses (2.5–50 mg/kg) in rats indicated a statistically significant dose-related increase in plasma troponin T levels 3 hours after treatment (Supplementary Table 2, available online). Troponin T levels remained elevated 6 hours after treatment with ZD6126, but by 24 hours after treatment, the levels decreased to those seen in plasma from vehicle-treated control rats (Table 1). However, pretreatment of rats with atenolol and nifedipine fully prevented the increases in troponin T seen at 3 hours after ZD6126 treatment (Table 1).
At the onset of these studies, we considered three potential mechanisms for ZD6126-induced myocardial necrosis to be feasible: 1) by direct and selective morphologic and/or cytotoxic effects on cardiac endothelial cells, because the endothelium, at least in tumors, is a proven target of ZD6126; 2) by direct cytotoxic effects on the myocardium; or 3) as a consequence of the induced acute hemodynamic alterations. We ruled out a direct effect of ZD6126 on cardiac endothelial cells because detailed histopathologic evaluation of rat hearts isolated 2 hours after treatment with ZD6126 at a dose of 12.5 mg/kg using either 0.75-µm thick tissue sections stained with toluidine blue or an antibody against the endothelial cell marker CD31 showed no evidence of morphologic or histologic changes in endothelial cells preceding the myocardial necrosis (data not shown). This finding is in sharp contrast to the effect of ZD6126 on tumor vessels, in which endothelial cell damage has been found to precede tumor cell necrosis (6). We also ruled out the possibility that ZD6126 had direct cytotoxic effects on the myocardium because the myocardial necrosis we observed was localized rather than diffuse (data not shown). Because pretreatment with atenolol and nifedipine prevented ZD6126-induced cardiac necrosis and increased plasma troponin T levels but had no marked effects on the plasma levels of ZD6126 phenol (the active form of ZD6126) (Supplementary Table 3, available online), we conclude that the ZD6126-induced myocardial necrosis in the rat is secondary to ZD6126-induced hemodynamic changes.
The cardiac necrosis we observed in ZD6126-treated rats was most commonly seen in the subendocardial to mid-left ventricular wall areas, which include the papillary muscles. This distribution is similar to that induced by hypoxia and isoprenalin, which shortens the diastolic period when ventricular blood perfusion is maximal (14). The cardiac changes induced by ZD6126 could be due to an imbalance between myocardial oxygen delivery and oxygen demand that results in inadequate myocardial perfusion, which is analogous to the situation in which diastolic filling time, during which the blood irrigates myocardial tissue, is shortened due to tachycardia and coronary artery perfusion pressure is reduced due to systemic hypotension (15,16). We hypothesize that acute vasoconstriction that is accompanied by tachycardia, as we have seen in rats following ZD6126 administration, may produce a similar inadequacy in myocardial perfusion, whereby tachycardia exacerbates hemodynamic changes associated with vasoconstriction.
The mechanism of cardiac toxicity suggested by this study (i.e., acute vasoconstriction/hypertension accompanied by tachycardia) is unlikely to be unique to ZD6126 because other tubulin-binding vascular disrupting agents (i.e., combretastatin A4 phosphate and AVE8062) are reported to have similar cardiac effects in the clinic (7–9) and because we have observed a similar pattern of cardiac necrosis after treating rats with combretastatin A4 phosphate (data not shown). Although the underlying mechanisms responsible for ZD6126-induced acute hypertension and tachycardia remain unknown, other microtubule-destabilizing agents (i.e., colchicine, nocodazole, and demecolcine) have been shown to induce vasoconstriction through their effects on vascular smooth muscle tone (17).
To investigate the effects of pretreatment with atenolol and nifedipine on the antitumor efficacy of ZD6126, we measured tumor necrosis in a tumor xenograft model in nude rats. Athymic female rats (HsdHan:Rnu-rnu; 100–160 g; n = 24; AstraZeneca, Alderley Park, U.K.) were implanted subcutaneously with Hras-transformed mouse 3T3 fibroblasts (Hras5 cells; 2 x 105 cells per rat) (6). When Hras5 xenograft tumors reached a size of approximately 0.5 cm3, the rats were randomly assigned to one of six treatment groups (n = 4 rats per group): physiological saline and 1% polysorbate 80; atenolol (6 mg/kg) and nifedipine (10 mg/kg); ZD6126 (5 or 12.5 mg/kg); atenolol (6 mg/kg) and nifedipine (10 mg/kg) plus ZD6126 (5 or 12.5 mg/kg). When all three agents were administered to the same rat, atenolol and nifedipine were given 1 hour before ZD6126. Rats were killed 24 hours after ZD6126 or vehicle dosing and their tumors excised, fixed in 10% neutral-buffered formalin, and processed for embedding in wax blocks. Necrosis was assessed by light microscopy in sections (4-µm thick) that were stained with hematoxylin–eosin. For each tumor, necrosis was assessed in a single section cut from the widest part of the tumor with the use of operator-guided image analysis software (KS400 version 3.0 image analysis system; Carl Zeiss Vission GmbH, Hallbergmoos, Germany). For each treatment group, the mean percentage of necrotic cells and the 95% confidence interval (CI) was calculated.
ZD6126 induced extensive tumor necrosis in Hras5 xenografts 24 hours after dosing (necrotic fraction of cross-sectional tumor area at 5 and 12.5 mg/kg = 32% and 87%, respectively) (Table 2), consistent with results from a previous study (4). Pretreatment with atenolol and nifedipine did not prevent tumor necrosis induced 24 hours after treatment with ZD6126 at 5 mg/kg (proportion of necrotic cells with versus without atenolol/nifedipine: 65.8% versus 32.7%, difference = 33.1%, 95% CI = 3.9% to 69.9%, P = .07 [two-sided t test]) or 12.5 mg/kg (proportion of necrotic cells with versus without atenolol/nifedipine: 94.0% versus 86.9%, difference = 7.1%, 95% CI = 5.2% to 19.5%, P = .208 [two-sided t test]) (Table 2) even though the same pretreatment schedule with atenolol and nifedipine ameliorated the cardiotoxic effects of ZD6126.
|
This small study has several limitations. In the clinic, only a minority of patients experience cardiovascular adverse events following ZD6126 treatment. By contrast, in the rat models used here the response to ZD6126 dosing was more homogeneous, and almost all animals displayed both increases in plasma troponin T level and histopathologically confirmed cardiac damage at ZD6126 dose levels of 5.0 mg/kg or higher. The factors that differentiate patients at risk for cardiovascular adverse events following ZD6126 dosing are not known, but extending the current studies to different strains of rat may help identify the underlying biologic mechanisms. In addition, the mechanism of ZD6126-induced tachycardia seen in rats is unclear but is likely to be indirect because no effect of ZD6126 on heart rate has been observed in isolated perfused rat hearts (data not shown). Further work is required to understand whether tachycardia observed in this rat model is relevant to patients treated with ZD6126.
In conclusion, our data in rats suggest that the adverse cardiac events observed in the clinic following ZD6126 administration may result, at least in part, from acute hemodynamic changes. This conclusion has potential therapeutic implications because our studies also suggest that these acute hemodynamic changes, and the subsequent cardiac side effects, may be ameliorated by antihypertensive therapy without overtly affecting the antitumor efficacy of ZD6126.
|
NOTES |
|---|
|
TopAbstractContext and CaveatsReferencesNotes |
|---|
The authors are employees of AstraZeneca, which funded this work. No external funding was received.
|
REFERENCES |
|---|
|
TopAbstractContext and CaveatsReferencesNotes |
|---|
(1) Thorpe PE, Chaplin DJ, Blakey DC. The first international conference on vascular targeting: meeting overview. Cancer Res (2003) 63:1144–7.
(2) Thorpe PE. Vascular targeting agents as cancer therapeutics. Clin Cancer Res (2004) 10:415–27.
(3) Siemann DW, Bibby MC, Dark GG, Dicker AP, Eskens FALM, Horsman MR, et al. Differentiation and definition of vascular-targeted therapies. Clin Cancer Res (2005) 11:416–20.
(4) Blakey DC, Ashton SE, Westwood FR, Walker M, Ryan AJ. ZD6126: a novel small molecule vascular targeting agent. Int J Radiat Oncol Biol Phys (2002) 54:1497–502.[CrossRef][Web of Science][Medline]
(5) Davis PD, Dougherty GJ, Blakey DC, Galbraith SM, Tozer GM, Holder AL, et al. ZD6126: a novel vascular-targeting agent that causes selective destruction of tumor vasculature. Cancer Res (2002) 62:7247–53.
(6) Blakey DC, Westwood FR, Walker M, Hughes GD, Davis PD, Ashton SE, et al. Antitumor activity of the novel vascular targeting agent ZD6126 in a panel of tumor models. Clin Cancer Res (2002) 8:1974–83.
(7) Beerepoot LV, Radema SA, Witteveen EO, Thomas T, Wheeler C, Kempin S, et al. Phase I clinical evaluation of weekly administration of the novel vascular-targeting agent, ZD6126, in patients with solid tumors. J Clin Oncol (2006) 24:1491–8.
(8) Budd GT, Evelhoch J, Langmuir P, Veiero J, Shepherdson K, LoRusso P. A phase I dose-escalation trial of ZD6126 administered as 5 daily dose every 3 weeks to patients with cancer refractory to other treatments. Eur J Cancer (2004) (1 Suppl):S165.
(9) Gadgeel SM, LoRusso PM, Wozniak AJ, Wheeler C. A dose-escalation study of the novel vascular-targeting agent, ZD6126, in patients with solid tumors. Proc Am Soc Clin Oncol (2002) 21. Abstract 438.
(10) Evelhoch JL, LoRusso PM, He Z, DelProposto Z, Polin L, Corbett TH, et al. Magnetic resonance imaging measurement of the response of murine and human tumors to the vascular targeting agent ZD6126. Clin Cancer Res (2004) 10:3650–7.
(11) Cooney MM, Radivoyevitch T, Dowlati A, Overmoyer B, Levitan N, Robertson K, et al. Cardiovascular safety profile of combretastatin A4 phosphate in a single-dose phase I study in patients with advanced cancer. Clin Cancer Res (2004) 10:96–100.
(12) van Heeckeren WJ, Bhakta S, Ortiz J, Duerk J, Cooney MM, Dowlati A, et al. Promise of new vascular-disrupting agents balanced with cardiac toxicity: is it time for oncologists to get to know their cardiologists? J Clin Oncol (2006) 24:1485–8.
(13) Pilat MJ, LoRusso PM. Vascular disrupting agents. J Cell Biochem (2006) 99:1021–39.[CrossRef][Web of Science][Medline]
(14) Woolf N, Davies MJ, Shaw MJ, Trickey RJ. Experiences with isoprenaline induced myocardial necrosis in the rat. J Pathol (1976) 120:65–73.[CrossRef][Web of Science][Medline]
(15) Belin V, Hodge T, Picaut P, Jordan R, Algate C, Gosselin S, et al. The myocardial lesions produced by the potassium channel opener aprikalim in monkeys and rats are prevented by blockade of cardiac beta-adrenoceptors. Fundam Appl Toxicol (1996) 31:259–67.[CrossRef][Web of Science][Medline]
(16) Balazs T, Ferrans VJ, El Hage A, Ehrreich SJ, Johnson GL, Herman EH, et al. Study of the mechanism of hydralazine-induced myocardial necrosis in the rat. Toxicol Appl Pharmacol (1981) 59:524–34.[CrossRef][Web of Science][Medline]
(17) Platts SH, Falcone JC, Holton WT, Hill MA, Meininger GA. Alteration of microtubule polymerization modulates arteriolar vasomotor tone. Am J Physiol Heart Circ Physiol (1999) 277:100–6.
Manuscript received January 16, 2007; revised August 29, 2007; accepted September 11, 2007.
Related Article in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2007 99: 1653.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||