Investigational New Drugs 6:173-178 (1988) �9 Kluwer Academic Publishers - Printed in the Netherlands

Evaluation of cardiotoxicity o f a new anthracycline derivative:

4 ' -deoxy-4 ' - iodo-doxo rubicin

Fabrizio Villani 1, Milena Galimberti 1, Enrica Lanza 2, Annalinda Rozza 2, Luigia Favalli 2 and Paola Poggi 3 1Servizio di Fisiopatologia Cardiorespiratoria, Istituto Nazionale per lo Studio e la Cura dei Tumori, Milan; 2 Cattedra di Saggi e Dosaggi Farmacologici and 3Istituto di Istologia ed Embriologia Generale, Universitit di Pavia, Pavia, Italy

Key words: 4'-deoxy-4'-iodo-doxorubicin, cardiotoxicity, anthracyclines

Summary

The present study in rats was performed to evaluate the cardiotoxic activity of 4' -deoxy-4' -iodo-doxorubicin (4'-deoxy-4'-I-DXR), a new anthracycline derivative with interesting antineoptastic properties and possible lower cardiotoxicity than doxorubicin (DXR). DXR produced ECG alterations consisting of a progressive and irreversible prolongation of SaT and QaT. In contrast, in 4'-deoxy-4'-I-DXR-treated rats the increase in SaT and QaT duration was significantly lower than that observed in DXR-treated rats and slightly in- creased over controls.

DXR produced significant histologic changes in myocardium consisting of myocyte vacuolization and myofibrillar loss. No significant modifications were observed in mitochondria. In contrast, no significant cardiac lesions were observed in 4'-deoxy-4'-I-DXR-treated rats. These results suggest that this new anthra- cycline derivative has a significantly lower degree of cardiotoxic activity than DXR.

Introduction

Dose-related chronic cardiotoxicity represents the major limitation to the use of anthracyclines in long-term treatment. For this reason much effort has been made to identify new analogs possibly with better antitumor activity and lower cardiotoxic activity. 4'-Deoxy-4'-I-DXR is a new analog that has shown interesting antineoplastic activity and, in a preliminary study in mice, lower cardiotoxicity than DXR [1-5]. In vitro, the new derivative is about 2 to 13 times more cytotoxic than DXR against a panel of cultured human cell lines (HeLa, LoVo, HT29, HRT18, MCF-7) [2-4]. In vivo in mice the new anthracycline is 1.5 to 2 times more toxic than DXR, and, moreover, it shows a very interesting activity against P388 and Gross leuke- mias. In this model 4'-deoxy-4'-I-DXR is also ac-

tive after oral treatment at doses similar to those used for intravenous administration [3-5]. The compound given intravenously and orally has shown good antitumor activity against C3H carci- noma, B16 melanoma and Lewis lung carcinoma [1]. In preliminary tests performed in C3H and CD1 mice, the analog was found almost completely devoid of cardiotoxic side effects [4,5].

The aim of the present investigation was to evalu- ate the cardiotoxic activity of 4' -deoxy-4' -I-DXR, compared to that of DXR, in a model of proven delayed myocardial toxicity in rats [6,7]. This ex- perimental model was chosen because it produces 100% cardiac toxicity routinely: in particular all animals sacrificed 5 weeks after the last DXR ad- ministration presented with significant cardiac le- sions more than 0.5 according to the score scale after mentioned and relevant ECG changes.

174

Material and methods

Female Sprague-Dawley rats (Charles River, Calco, Italy), originally weighing 120-130 g, were used. The animals were maintained under standard labo- ratory conditions and in quarantine for 7 days be-

fore being randomized into experimental groups of

at least 7 animals. DXR and 4'-deoxy-4'-I-DXR (kindly supplied

by Farmitalia Carlo Erba, Milan, Italy) were dis- solved in distilled water, and each was administered intravenously at the dose of 3 mg/kg (DXR 5.2 tzM, 4' -deoxy-4' -I-DXR 4.3/zM). Controls were treated i.v. with the same amount of saline. Animals were treated four times at one-week intervals and sacri- ficed five weeks after the last anthracycline admini-

stration. The two drugs were used at doses which were

found to be equiactive on the mouse solid tumor models investigated using a repetitive treatment

schedule [1]. The body weight of the animals was recorded

every week for the first month and then every two weeks, as a sign of general toxicity. Blood samples for biochemical and hematologic analysis were ob-

tained after the sacrifice. A complete blood count and serum determinations of urea nitrogen, creati-

nine, glucose, sodium, potassium, calcium, lactate dehydrogenase, alkaline phosphatase and trans- aminases were performed.

Cardiotoxicity was evaluated by means of elec-

trocardiography and morphologic evaluation of the myocardium. Electrocardiograms were recorded at the beginning of the treatment and then at two- week intervals. The six standard electrocardio- graphic leads were recorded by means of a Car- dioline instrument in rats kept under light ether anesthesia with needle electrodes inserted subcu- taneously in the four limbs. The signals were also displayed on a Philips storage oscilloscope and sub- sequently recorded. For each ECG, heart rate, QaT and SaT duration were measured. These parame- ters were chosen because a good correlation was found to exist between them and the degree of mor- phologic lesions [7]. Heart rate was calculated from R-R intervals. QaT was calculated as the interval between the Q wave and the apex of the T wave, and

SaT as the interval between the S wave and the apex of the T wave. QaT and SaT were not corrected for changes in heart rate, according to the findings of some authors [8-10].

Five weeks after the last DXR administration, after recording the last ECG, the animals were killed by cervical dislocation. Immediately after the sacrifice, left ventricle fragments were fixed in 2.5% glutaraldehyde-paraformaldehyde in 0.1 M Na-cacodylate buffer (pH 7.5) for 4 h at 4~ [11]. The specimens were then washed in Na-cacodylate buffer with saccarose and postfixed in OsO4-colli- dine for 1.5 h at 4~ dehydrated and embedded in

epoxy resin. Semithin sections (0.2-2 t~m) were stained with toluidine blue [12] for light microsco-

py. Ultrathin sections (about 80-100 nm), pre- pared and stained with uranyl acetate and lead

citrate [13], were studied with a Zeiss EM109 elec- tron microscope. DXR cardiotoxicity was identi- fied by examination of transverse and longitudinal sections of the whole ventricle, and the degree of the lesions was quantitated according to a scale of

0 to 2 on the basis of the number of muscle cells that showed myofibrillar loss and particularly cytoplas- mic vacuolization: 0, no damage; 0.2, involvement

of very few (less than 5) myocytes; 0.5, involvement of more than 10 myocytes; 1, small clusters of in- volved myocardial cells; 1.5, multiple clusters of

vacuolated myocytes; 2, diffuse myocyte damage. The sections were evaluated without prior knowl- edge of the treatment given to the animals.

Statistical analys&

Differences in body weight, SaT and QaT duration throughout the observation period were assayed by means of a factorial analysis of variance. Morpho- logic data were evaluated by means of Wilcoxon's

nonparametric test.

Results

General toxicity, body weight, clinical chemistry and hematologic determinations

During the study no death was observed in the three

175

g

240

220

I o_ 200

o m 180

160

CONTROLS DOXORUBIC[N

o---o 4'-DEOXY-4'-I-DOXORU BICIN

, / / / /

/ /

0 1 2 3 4 6 8 WEEKS

Fig. 1. T i m e c o u r s e o f b o d y w e i g h t in d o x o r u b i c i n - a n d 4 ' -

d e o x y - 4 ' - I - d o x o r u b i c i n - t r e a t e d rats ( m e a n s __+_ SE) .

SaT

40

30

20

m SeC

CONTROLS = = DOXORUBICIN T

i ~ - o 4 ' - D E O X Y - 4 ' - I - D O X O R U B I C I ~

10

11 L I ~ i I i i ~ I 0 2 4 6 8

WEEKS

Fig. 2. T i m e c o u r s e o f S a T d u r a t i o n in d o x o r u b i c i n - a n d 4 ' - r d e o x y - 4 - I - d o x o r u b i c i n - t r e a t e d rats ( m e a n s + SE) .

experimental groups. The time course of body weight in the treatment groups is reported in Fig. l. 4'-deoxy-4'-I-DXR caused in the first four weeks a delay in body weight growth slightly less than that

QaT

rn sec

CONTROLS 55 H DOXORUBICIN

o---o 4 ' -DEOXY-4'- I -DOXORUBICIN

5O

45

, 0 { / #

/ /

35 "

. I T

30

25 i i i i ] 0 2 4 6 8

WEEKS

Fig. 3. T i m e c o u r s e o f Q a T d u r a t i o n in d o x o r u b i c i n - a n d 4 ' -

d e o x y - 4 ' - I - d o x o r u b i c i n - t r e a t e d rats ( m e a n s + SE) .

observed in DXR treated animals but the difference was not found to be statistically significant. How- ever, after this period body weight progressively in- creased, so that at the end of the study there was no statistically significant difference between controls and 4'-deoxy-4'-I-DXR-treated animals. In con- trast, DXR significantly hindered body weight growth: at the end of the observation period, DXR- treated rats showed a statistically significant lower body weight than controls and 4'-deoxy-4'-I- DXR-treated animals.

No substantial changes were found at the end of the observation period in serum concentrations of glucose, urea nitrogen, creatinine, lactate dehydro- genase, alkaline phosphatase, or transaminases. No statistically significant difference was found in WBC counts, RBC counts or hemoglobin concen- tration between any of the groups. A slight increase in serum potassium and a decrease in serum calcium was observed in DXR-treated animals, but the dif-

176

Table 1. Morphologic evaluation of myocardial toxicity of 4'-deoxy-4'-I-doxorubicin in comparison to doxorubicin

Treatment No. of Cardiomyopathy score Mean

preparations 0 0.I 0.2 0.5 1 1.5 2

Preparations with

cardiomyopathy score >_ 0.5

No./ total %

Saline (controls) 8

Doxorubicin 3 mg (5.2 ~M)/kg x 4 7

4' -Deoxy-4' -I-doxorubicin

3 mg (4.3 /zM)/kgx4 9

5 2 1 0 - - - 0.05 0/7

- 2 2 2 1 1 .07" 7 / 7 100

5 3 1 . . . . 0.05 0/9

* P < 0.05 vs. controls.

ference compared to controls was not statistically

significant.

ECG alterations

No significant alterations in heart rate, QRS dura-

tion or voltage were observed in the three experi-

mental groups. SaT duration significantly increas-

ed in DXR-treated animals. The phenomenon was

particularly evident from the fourth week. In con- trast, in 4 ' -deoxy-4 ' -I-DXR-treated rats the in-

crease in SaT duration was significantly lower than

that observed in DXR-treated rats and slightly in-

creased over that of controls. Consequently, QaT

duration was significantly increased by DXR. 4 ' - deoxy-4 '-I-DXR produced a slight increase com-

pared to controls but a significantly lower increase

than that observed in DXR-treated animals.

Myocardial pathology

The most prominent finding in histologic prepara-

tions of left ventricles excised for DXR-treated animals consisted of myocyte vacuolization and myofibrillar loss. No significant modifications in mitochondria were observed. The affected cells were not restricted to any particular layer of the my- ocardium. These alterations were found in the heart of all seven rats treated with DXR: the severity of

lesions in these hearts ranged from 0.5 to 2 (aver- age 1.05). In contrast, no cardiac lesions were ob- served in five of the nine rats given 4 ' -deoxy-4 ' - I- DXR. The remaining four animals in this group

showed minimal alterations (0.1-0.2) as also ob- served in some animals given saline. The difference in severity and incidence of lesions between DXR- treated animals and those given 4 ' -deoxy-4 ' -I- DXR or saline was highly significant (P < 0.001).

Discussion

In the present study, rats received a cumulative dose of DXR of 12 mg/kg over a four-week period and were examined five weeks after the last DXR ad- ministration, as previously described [6,7]. At this time, all animals had significant cardiac lesions: four mild (score 0 .5-1) and three severe (score 1.5-2). Myocardial lesions mainly consisted of cytoplasmic vacuolization and myofibrillar loss, lesions which are similar to those observed in pa- tients treated with DXR and daunorubicin [14].

Moreover, this treatment schedule produced cha- racteristic ECG changes and particularly a progres- sive and irreversible prolongation of SaT and QaT duration, without significant modifications in QRS duration or voltage as previously described [6,7, 10]. The phenomenon was particularly evident af- ter the fourth week and persisted with time, even though DXR and its metabolites are no longer de- tectable in heart tissue three weeks after the last ad-

ministration [15]. These results indicate that in rats, DXR-induced myocardial damage may progress in the absence of continued drug administration to severe cardiomyopathy. QaT and SaT prolongation has been attributed to a marked prolongation of action potential duration [10]. The ionic mecha- nism underlying this phenomenon remains to be clarified. The possibility that alterations in serum electrolyte levels may have contributed to ECG changes can be excluded since no significant modi- fications in serum electrolytes were observed.

4'-Deoxy-4'-I-DXR at the tested dose produced only slight prolongation of QaT and SaT compared to controls but a less significant prolongation com- pared to DXR-treated rats. Morphologic evalua- tion of left ventricles showed that there were no significant alterations in 4'-deoxy-4'-I-DXR- treated animals. Severity and incidence of lesions were not significantly different from those ob- served in control animals. These results indicate that this new analog up to the tested dose is almost devoid of cardiotoxic activity.

The pharmacologic bases of this reduced cardio- toxicity remain to be ascertained. DXR produces a wide range of biological effects that potentially have toxic consequences for heart cells. Three prominent action mechanisms have been ascribed to anthracyclines: a) DNA intercalation, which produces a blockade

of DNA, RNA and protein synthesis, fragmenta- tion of DNA, and inhibition of repair;

b) binding to membranes with consequent altera- tion of membrane function;

c) triggering of the formation of oxygen-free radi- cals with consequent lipid peroxidation [16]. The effects of 4'-deoxy-4'-I-DXR on the men-

tioned target processes have not yet been inves- tigated. It cannot be excluded that the pharma- cokinetic behavior of the new analog could at least partially explain the reduced myocardial toxicity. In fact, preliminary data indicated that, at least in mice, the area under the curve for heart of this new analog is three times lower than that of DXR because of a more rapid clearance of the drug from the myocardium [17]. In conclusion, the in vivo cardiac effects of 4'-deoxy-4'-I-DXR sug- gest that this new anthracycline derivative has a

177

significantly lower degree of cardiotoxic activity than DXR.

Acknowledgements

The study was supported in part by contract no. 86.02610.44, Special Project Oncology, from the Consiglio Nazionale delle Ricerche, Rome. The authors thank Ms. B. Johnston for editing and preparing the manuscript.

References

1. Barbieri B, Giuliani FC, Bordoni T, Casazza AM, Geroni C, Bellini O, Suarato A, Gioia B, Penco S, Arcamone F: Chemical and biological characterization of 4 '- iodo-4'- deoxydoxorubicin. Cancer Res 47:4001-4006, 1987

2. Bordoni T, Barbieri B, Geroni C, Marsiglio A, Bellini O, Giuliani FC: Preclinical evaluation of 4'-deoxy-4'-I-doxo- rubicin (FCE 21954), a new anthracycline derivative (Abstr). Proc. XIV Int. Cancer Congress, Budapest, 21-27 August 1986

3. Barbieri B, Grandi M, Bordoni T, Cassinelli G, Giuliani FC: In vitro evaluation of idarubicin and 4'-I-doxorubicin and their 13-OH-metabolites on LoVo/Dx (Abstr). Proc. 5th NCI-EORTS Symp. New Drugs in Cancer Therapy, Amsterdam, 22-24 October 1986

4. Giuliani FC, Bordoni T, Barbieri B, Geroni C, Bellini O: Biological characterization of 4'-deoxy-4'-I-doxorubicin (Abstr). Proc. 5th NCI-EORTS Syrup. New Drugs in Can- cer Therapy, Amsterdam, 22-24 October 1986

5. Barbieri B, Suarato A, Penco S, Geroni C, Bellini O, Fumagalli A, Casazza AM, Giuliani FC: Biological activity of 4'-halo-anthracyclines (Abstr). Proc Am Assoc Cancer Res 25:305, 1984

6. Villani F, Monti E, Favalli L, Lanza E, Poggi P: A model of doxorubicin-induced cardiomyopathy in the rat. IRCS Med Sci 14:282-283, 1986

7. Villani F, Favalli L, Lanza E, Poggi P: Relazione tra modi- ficazioni elettrocardiografiche ed alterazioni indotte da an- tracicline. Cardiologia 32:585-590, 1987

8. Beinfield WH, Lehr D: QRS-T variations in the rat elec- trocardiogram. Am J Physiol 214:197-204, 1968

9. Buschmann G, Schumaker R, Budden R, Ktihe U: Evalua- tion of the effect of dopamine and other catecholamines on the electrocardiogram and blood pressure of rats by means of on-line biosignal processing. J Cardiovasc Pharmacol 2:777-795, 1980

10. Jensen RA, Acton EM: Doxorubicin cardiotoxicity in the rat: comparison of ECG, transmembrane potential and structural effects. J Cardiovasc Pharmacol 6:186-200, 1984

178

11. Karnovsky M J: A formaldehyde fixative of high osmolarity for use in electron microscopy. J Cell Biol 27:137a, 1965

12. Richardson KC: Embedding in epoxy resin for ultrathin sec- toring microscopy. Stain Technol 35:313, 1960

13. Venables JH, Caggeshall R: A simplified citrate stain for use in electron microscopy. J Cell Biol 25:407-408, 1965

14. Ferrans V J: Overview of cardiac pathology in relation to an- thracycline cardiotoxicity. Cancer Treat Rep 62:955-961, 1978

15. Monti E, Piccinini F, Villani F, Favalli L: Myocardial con- tractility and heart pharmacokinetics of adriamycin follow- ing a single administration in rats. Cancer Chemother Pharmacol 18:289-297, 1986

16. Young RC, Ozols RF, Myers CE: The anthracycline an- tineoplastic drugs. N Engl J Med 305:139-153, 1981

17. Formelli F, Carsana R, Pollini C: Pharmacokinetics and metabolism of 4'-deoxy-4'-I-doxorubicin (4'-IDX) in tumor-bearing mice: comparison with doxorubicin (DX) (Abstr). Proc 5th NCI-EORTS Symp. New Drugs in Cancer Therapy, Amsterdam, 22-24 October 1986

Address for offprints: F. Villani, Fisiopatologia Cardiorespira- toria, Istituto Nazionale Tumori, Via Venezian 1, 20133 Milan, Italy