Journal of Surgical Oncology 51:254-258 (1992)

Antitumor Effect of Triphenylethylene Derivative (TAT-59) Against Human Breast Carcinoma Xenografts in Nude Mice JUN-ICHI KOH, MD, TETSURO KUBOTA, MD, FUMlKl ASANUMA, MI), YOSHlNORl YAMADA, MU, Ell1 KAWAMURA, MD, YOlCHlKO HOSODA, MD, MITSUMASA HASHIMOTO, MD, OSAMI YAMAMOTO, MD, SHOJISAKAI, M i l , KOUTARO MAEDA, MD, AND EllCHl SHIINA, MD From the Department of Surgery, Social lnsurance Saitama Chuo Hospital, Urawa, Saitama O.K., Y.H., M.H., O.Y., S.S., K.M., E.S.), Department of Surgery, School of Medicine, Keio University, Tokyo (J.K.), and Department of Surgery, Kitasato Institute Hospital, Tokyo (F.A., Y.Y., E.K.), lapan

The antitumor activity of a newly synthesized triphenylethylene derivative {(E)-4-[ 1- [4-[2-(dimethylamino)ethoxy-phenyl]-2-(4-isopropyl)phenyl-1butenyl] phenyl monophosphate} (TAT-59) was investigated against human breast carcinoma xenografts in nude mice with reference to the changes of hormone receptors. Five strains (MCF-7, Br-10, R-27, ZR75- 1 , and T-61) used for the experiments possessed cytosol estrogen receptor (ER), and their growth was estradiol dependent. Five mg of TAT-59 and tamoxifen citrate (TAM) per kg were administered p.0. daily except Sunday. TAT-59 showed a positive antitumor effect against MCF-7 and R-27, whereas TAM was effective on MCF-7, and their adverse effects detected by mortality rate, body weight loss, and spleen weight loss were similar to each other. The reduction of ER and production of progesterone receptor (PgR) after the treatment with TAT-59 were more potent than after TAM, suggesting that TAT-59 exerts its antitumor effect through binding to ER. These findings suggest that TAT-59 might merit use in clinical trials with breast cancers. 0 1992 Wiley-Liss, Inc. KEY WORDS:antiestrogen, TAT-59, breast cancer, endocrine therapy

INTRODUCTION Tamoxifen citrate (TAM) has been widely used in the endocrine treatment of breast cancer patients. The efficacy rate of TAM against advanced breast cancer was reported to be 3040% [ I ] , and it has been proven that the postoperative adjuvant therapy with TAM significantly reduced the recurrence rate of operable breast cancer patients 121. Although the mechanism of the action of antiestrogen has not been clarified completely, it has been supposed that antiestrogens exert their antitumor activity through competing with estrogen binding to estrogen receptor (ER). However, the binding affinity of TAM to ER was less than 1% of that to estradiol (E2) [3], so the development of new antiestrogens with higher affinity to ER was undertaken [4]. Recently, it was reported that a new triphenylethylene derivative, TAT-59, 0 1992 Wiley-Liss, Inc.

showed a higher affinity to ER than TAM and a superior growth inhibitory effect against MCF-7 in vitro and dimethylbenzanthracene (DMBA)-induced rat mammary carcinoma in vivo [ 5 ] .The present study deals with the antitumor activity of this agent against human breast carcinoma xenografts in nude mice in comparison with TAM.

MATERIALS AND METHODS Mice Female nude mice with a BALBic nuinu genetic background were purchased from CLEA Japan (Tokyo) and Accepted for publication July 23, 1992 Address reprint requests to Dr. Jun-ichi Koh, Department of Surgery, Social Insurance Saitama Chuo Hospital, 4-9-3 Kitaurawa, IJrawa, Saitamn 336, Japan.

Antitumor Effect of TAT-59 Against Breast Carcinomas

255

TABLE 1. Characteristics of Human Breast Carcinoma Strains in Nude Mice Tumor MCF-7 R-27 ZR-75-1 Br-I0 T-6 I

ER"

+ + + + +

Growth in male mice

Growth in female mice

-

-

+

*' i

+ +

Dose of EZh 0.1 mgiwk X 3 0.1 mgiwk x 1 0.1 mg/wk x 1 0.1 mgiwk X I -

Tumor doubling time (days)

12.3 11.2

7.6 7.5 9.5

Estrogen receptor more than 10 fmolimg protein. Minimum dose of estradiol treatment required for the experiments. E, was administered intramuscularly at the start ofexperirnents. When the tumor tissues were inoculated in the untreated female nude mice, no exponential growths were observed, but the tumor tissues were not conipletely rejected as shown in MCF-7 tumor.

'I

maintained under specific pathogen-free conditions using laminar air flow racks in our institutes. Mice 6-7 weeks old, weighing -20 g, were used for the experiments. Each group consisted of four to six mice.

Tumor Five human breast carcinoma strains (MCF-7, R-27, ZR-75-1, Br-10, and T-61), serially transplanted into nude mice, were used. MCF-7 was established as a cultured cell line in 1970 by Soule et al. [6] and was successfully transplanted by the authors into nude mice treated with estrogen and progesterone in 1983 [ 7 ] .R-27 [8] and ZR-75-1 191 were also transplanted into nude mice from the cultured cell line by the same procedure as for MCF-7. Br- 10 was established from cancerous pleural effusion of a 43-year-old female with a common ductal carcinoma in 1974 by Hirohashi et al. [lo] (kindly supplied to our institutes). T-61 was derived from cancerous tissue of a 54-year-old female patient with breast cancer in Frankfurt am Main [ 1 I ] (kindly supplied by Dr. Brunner of Copenhagen University). The characteristics of tumors are listed in Table I. All strains possessed ER, and their growth was enhanced by the extrinsic estradiol (EJ treatment except T-61, whose growth was suppressed by E,. After tumor inoculation into the subcutaneous tissue of the mice, R-27, Br-10, and ZR-75-1 were treated with 0.1 mg of 17-6 estradiol dipropionate (E,) intramuscularly once, whereas MCF-7 was treated with E, once a week three times.

sulfoxide and 0.5% carboxy methylcellulose was administered p.0. except Sundays.

Evaluation The tumors were measured (length and width) with sliding calipers three times weekly by the same observer. According to the method of Geran et al. [ 121, the tumor weight in mg was calculated from the linear measurements using the formula: tumor weight (mg) = length (mm) X [(width(mm)12/2. When the tumors reached 100-300 mg, the tumor-bearing mice were randomized into test groups, each consisting of four to six mice. The relative mean tumor weight (RW) was calculated as RW = WiiWo, where Wi is the mean tumor weight at any given time and Wo is the mean tumor weight at the time of initial treatment. The antitumor effect of the agents was evaluated as the lowest TIC rate (%) during the experiment, where T is the relative mean tumor weight in the treated group and C is the relative mean tumor weight in the control group at any given time. The antitumor effect was evaluated as positive when the lowest T/C was equal to or less than 50%. The growth curve was drawn by plotting RW against days after the initial treatment to assess the antitumor activity pattern of the agents. At the end of the experiments, all mice were killed and spleens were weighed. The toxicity of the agents was assessed in terms of the mortality of mice, the lowest body weight loss during the experiments, and the loss of spleen weight of the tumor-bearing mice. Hormone Receptor Assay

Agents Agents used for the experiments were TAT-59 {(E)-4[ 1-[4-[2-dimetthy lamino) ethoxy]-phenyl]-2-(4-isopropy1)phenyl- 1-butenyl] phenyl monophosphate} and tamoxifen citrate (TAM) (both drugs kindly supplied by Taiho Pharmaceutical Co., Tokyo, and ICI Pharma Co., Osaka). A daily dose of 5 mg of TAT-59 and TAM per kg body weight dissolved in saline containing 10%dimethyl

At the end of experiments, tumors were resected and stored at -70°C for the assay. Cytosol ER and progesterone receptor (PgR) were measured according to the dextran-coated charcoal method as previously described [ 131.

Statistical Analysis All statistical analyses were performed by Student's t-test.

256

Koh et al. TABLE 11. Antitumor Activity of TAT-59 and Tamoxifen (TAM) Against Human Breast Carcinoma Strains in Nude Mice Agent TAT-59

0

7

14

21

28

Days after initial treatment

Fig. I . Antitumor activity of TAT-59 and TAM against MCF-7: 0 , control; 0 , tamoxilen; A , TAT-59. TAT-59 and TAM suppressed the growth of MCF-7 and the effects of both agents were evaluated as positive .

R-27

Br-10

T-61

c ( 6 ) '

m ( 6 ) 58.4 (4) 51.2

69.5 (6) 96.8 ( 5 ) 83.2

77.0(6)

57.9 (6) 58.8 (4) 89.6(5) 68.8

77.1 (4) 82.5 ( 5 )

88.0 ( 6 ) 73.2 (5)

74.0 (3) 65.9 (4)

79.8

80.6

70.0

42.3 (6)

Mean'

I

MCF-7

42.5

TAM

30.4 (6)

Mean'

56.1 (5) 67.4 ( 5 ) 51.3

ZK-75-1 (4)

100

"Data were represented by the lowest TIC ratio of the relative mean tumor weight during the experiments. Underscore indicates positive antitumor effect. Number of mice. 'Mean value of the lowest TIC ratio in the repeated experiments

TABLE 111. Toxicity of TAT-59 and Tamoxifen (TAM) Against Tumor-Bearing Nude Mice

Tumor

Spleen weight''

Body weight lossh

Agents

(%)

(%)

Death

TAT-59 TAM TAT-59 TAM TAT-59 TAM TAT-59 TAM TAT-59 TAM

107.7 91.3 87.0 98. I 123.0 79.6 75.4 96.5 70.0 82.2

3.9 8.4 4. I 4.8 11.2 5.8 0 0 16.5 9.4

216' 016 2164 015 214 016 016 114' 013

~

MCF-7 R-27 Br-10 T-6 I Days after initial treatment

116'

ZR-75- I Fig. 2. Antitumor activity of TAT-59 and TAM against R-27: 0, control; 0 , tamoxifen; A , TAT-59. TAT-59 showed positive antitumor effect with the lowest TIC ratio of 43.9%. although the effect of "Percent of the mean actual spleen weight in the treated group relative to control at the end of experiments. TAM was evaluated as negative with the lowest T/C ratio of 57.9%. 'Maximum body weight loss during the experiments in percent relative to the start of experiments. 'Death due to cystic stones. dAccidental death. RESULTS

The antitumor activity of TAT-59 and TAM against MCF-7 is shown in Figure 1. The growths of MCF-7 treated with TAT-59 and TAM were suppressed resulting in TIC ratios of 41.2% for TAT-59 and 30.4% for TAM 25 days after the initial treatment. The effect of agents against R-27 is shown in Figure 2. Both agents suppressed the growth of R-27, whereas the antitumor effect of TAT-59 was superior to that of TAM with the lowest T/C ratio by TAT-59 at 43.9%. The antitumor effect of TAT-59 against 5 strains is summarized in Table 11, comparing it with TAM in terms of the lowest TIC value of RW. Although TAM showed a positive antitumor effect only once on MCF-7, the antitumor activity of TAT-59 was found to be positive on MCF-7 and R-27. The mean values of T/C ratio for TAT-59 on MCF-7 and R-27 were about 10% lower than that of TAM. Br-10, T-61, and ZR-75-1 were insensitive to both agents. The toxicity of TAT-59 as compared with TAM was assessed in terms of the mortality of mice, the maximum body weight loss

during the simultaneous experiments, and the spleen weight at the end of simultaneous experiments (Table 111). The loss of spleen weight was observed in the experiments of R-27, T-61, and ZR-75-1. Although the maximum loss of spleen weight was 30.0% in ZR-75-1-bearing mice treated with TAT-59, no significant difference was observed between the spleen weights treated with TAT-59 and TAM. The loss of body weight treated with TAT-59 ranged from 3.9-1 1.296, which was similar to that of TAM. The death of mice bearing MCF-7 was due to bladder stones caused by the extrinsic estradiol, and the deaths of mice bearing R-27 and Br-10 treated with TAM were accidental. The binding sites of cytosol ER and progesterone receptor (PgR) treated with TAT-59 and TAM compared with their control are shown in Table IV. Data of R-27 and MCF-7 are shown as one determination consisting of

Antituimor Effect of TAT-59 Against Breast Carcinomas TABLE IV. Effects of TAT-59 and Tamoxifen (TAM) Against the Hormone Receptors of Human Breast Carcinoma Strains

Tumor T'6 I

ZR-75- I R-27

MCF-7

Group

ER" (fmolimg protein)

PgRb (fmolimg protein)

Control TAM TAT-59 Control TAM TAT-59 Control TAM TAT-59 Control TAM TAT-59

32.5 2 10.5' 44.9 4 0.5 4.5 7.8* 134.4 2 53.1 5.6 0.4** 5.8 2 1.2** 623.9 2 102.9 307.9 2 163.8** 198.9' 42.8 15.8 18.8' 36.2'

181.6 108.7 657.6 -+ 72.9** 1036.8 k 193.3* U.D.d 5.5 ? 2.6 16.1 ? 1.5*** U.D. 13.1 2 18.5 U.D. 63.6 2 13.7 27.4' 33.9'

* *

*

*

N

4 3 4 4 6 4 3 3 I 3 2 1

"Binding sites of estrogen receptor. hBinding sites of progesterone receptor. Data were shown as M -+ SD. dundetectable; 0 fmol/mg protein. "Data were shown as one determination consisting of four to five tumors each. * P < 0.05 to control and TAM; **P < 0.05 to control; * * * P < 0.001 to TAM.

four to six tumors each because each tumor was too small for ER assay. Binding sites of ER treated with TAT-59 were reduced in all the strains compared with control, whereas reduction of ER treated with TAM was observed in three of four strains except T-61. The differences between ERs of TAT-59 and control were statistically significant in T-6 1 and ZR-75- 1. A significant production of PgR was observed on T-61 and ZR-75-1 treated with TAT-59.

DISCUSSION The present study was conducted to confirm the antitumor activity of TAT-59 in vivo using human breast carcinoma xenografts in nude mice. It has been reported that these strains used for the experiments were reliable models to assess nonsteroidal antiestrogens, because these strains possessed ER and their growth was estradiol-dependent to various degrees [ 141. It is essential to adjust the dose of antiestrogens for the experiments, since the exogenous estradiol required for their exponential growth might inhibit the antitumor activity of antiestrogens. In our previous report concerning the antitumor activity of TAM, the exogenous estradiol was eliminated from the serum at 2 weeks after the last administration, and 5 mg of TAM per kg body weight was enough to exert its antitumor effect under this condition [ 151. When the same dose of TAT-59 with TAM was used, TAT-59 showed the positive antitumor effect against MCF-7 and R-27, whereas TAM showed the positive effect against MCF-7 only, suggesting that the effect of TAT-59 was

257

slightly superior to that of TAM without increasing toxicity in vivo. This excellent antitumor activity might be due to its higher affinity to ER than that of TAM. In addition, the reduction of ER and the production of PgR treated with TAT-59 were more prominent than those of TAM on T-61 and ZR-75-1, suggesting sufficient processing of the ER system by TAT-59. In contrast, the production of PgR by TAT-59 was not observed in MCF-7 and R-27, and this might be related to the report that the long-term treatment of antiestrogens resulting in tumor suppression caused a reduction of PgR in human breast carcinoma xenografts in nude mice [ 161. From the viewpoint of pharmacokinetics, it was reported that the concentrations of TAT-59 in plasma and tumor tissue in rats bearing DMBA-induced rat mammary carcinoma were higher than those of TAM [ 5 ] .In the same report, TAT-59 was demonstrated to be rapidly metabolized to 4-hydroxy TAT-59 with a high affinity to ER to the same extent as 4-hydroxy TAM, an active metabolite of TAM. And then, the peak concentration of 4-hydroxy TAT-59 in tumor was 1 1-fold higher than that of 4-hydroxy TAM, and prolonged retention of 4-hydroxy TAT-59 in tumor 24 hours after oral administration to tumor-bearing rats was observed. These advantages of the pharmacokinetics of TAT-59 also might be one reason why its antitumor activity was superior to that of TAM. In our previous report, we assessed the antitumor activity of analogous compounds of TAM and revealed that the prediction of endocrine sensitivity of each strain was difficult only by ER status of the xenografts [9,14]. The present results were comparable with these findings, since three strains with positive ERs were insensitive to TAT-59. These three strains have relatively long doubling times compared with the other human tumor xenografts serially transplanted into nude mice in our laboratory. Although their slow growing character can represent the clinical feature of breast cancer, their small growth fractions might be concerned with their low sensitivity to antiestrogens, since the growth rates of human tumor xenografts are partially related to their chemosensitivity [ 171. However, when the chemotherapeutic agents were tested on these human breast carcinoma xenografts, their chemosensitivity was rather higher than that of gastrointestinal carcinomas [IS]. In contrast, the sensitivity of these breast cancer cell lines to antiestrogens is relatively low and complete responses, which are observed clinically, were not observed in the experimental endocrine therapy as shown in the present and our previous study (191. Because the concentration of TAM in the mice serum was almost equivalent to that in the human serum [ 191 and the binding sites of estrogen receptors of breast cancer xenografts are suppressed in the present report, the low sensitivity to antiestrogens of breast cancer xenografts seems not due to the slow grow-

258

Koh et al.

ing rate of these tumors but to the different endocrine circumstances of host nude mice and different estrogendependency of human breast carcinoma xenografts. In addition, an incomplete secretion of negative growth factors might be one of the reasons for the low sensitivity of these tumors to antiestrogens. Recently, it was reported that antiestrogens exert their inhibitory effects against hormone-dependent breast cancers through regulating the secretion of transforming growth factor-f3 (TGF-P), which is one of the negative autocrine growth factors [20]. They reported that a variant line of MCF-7 with ER and resistant to antiestrogens lacks the induction of TGF-P by antiestrogens, and the mechanism of TGF-P induction is mediated through ER systems. Although the complete mechanism of antitumor activity of antiestrogens still remains unsolved, this might be one of the reasons why all of ER positive tumor cells do not respond to endocrine therapy.

5. 6.

7. 8. 9.

10. II .

12.

CONCLUSION TAT-59 would be a promising agent to test in clinical trials against breast cancers that are ER positive and insensitive to conventional TAM treatment because of its superior antitumor activity and higher affinity to ER against human breast carcinoma xenografts in nude mice. ACKNOWLEDGMENTS The authors thank Mr. Tomofusa Migita for his assistance with the experiments. Our thanks also to Taiho Pharmaceutical and ICI Pharma for their kind gifts of TAT-59 and tamoxifen citrate, respectively. REFERENCES I . Nomura Y, Takatani 0, Sugano H, Matsumoto K: Oestrogen and progesterone reccptor and response to endocrine therapy in Japanese breast cancer. J Steroid Biochem 13:565-566, 1980. 2. Nolvadex Adjuvant Trial Organization: Controlled trial of tamoxifen as single adjuvant agent in management of early breast cancer: Analysis at six years. Lancet:83&840, 1985. 3. Furr BJA, Jordan VC: The pharmacology and clinical use of tamoxifen. Pharmac Ther 25: 127-205, 1984. 4. Robinson SP, Jordan VC: New antiestrogen. In Pannuti F (ed):

13.

14.

15. 16.

17. 18.

19.

20.

“Hormone therapy results and perspectives.” Pavia: Edizioni Medico-scientiche 1988, 177-194. Toko T , Sugimoto Y, Matsuo K, et al.: TAT-59, a new triphenylethylene derivative with antitumor activity against hormcine-dependent tumors. Eur J Cancer 26:397404, 1990. Soule HD, Varquets J, Lung A: A human cell line from a pleural effusion derived from a breast carcinoma. J Natl Cancer Inst 5 1 :1409-14 16, 1973. Kubota T , Kubouchi K, Koh J, et al.: Human breast carcinoma (MCF-7) serially transplanted into nude mice. Jpn J Surg 13:38 I384, 1983. Nawata H , Bronzart D, Lippman ME: Isolation and characterization of a tamoxifen-resistant cell line derived from MCF-7 human breast cancer cells. J Biol Chem 256:5016-5021, 1981, Kubota T, Oka S, Utsumi T, et al.: Human breast carcinoma (ZR-75-1 ) serially transplanted into nude mice-with reference to estradiol dependency and sensitivity to tamoxifen. Jpn J Surg 19:44&451, 1989. Hirohashi S, Shimosato Y, Kameya T , er al.: Hormone dependency of a serially transplantable human breast cancer (Br-10) in nude mice. Cancer Res 37:3 184-3 189, 1977. Brunner N , Spang-Thomsen M , Skovgaard S A , et al.: Endocrine sensitivity of receptor positive T-6 1 human breast carcinoma serially grown in nude mice. Int J Cancer 35:SY-64, 1985. Geran RI, Greenberg NH, Macdonald MM, et al.: Protocols for screening chemical and natural products against animal tumors and other biological systems. Cancer Chemother Rep 3:51-61, 1972. Koh J, Shiina E, Hosoda Y, et al.: Changes in the hormone receptors of human breast carcinoma xenografts in nude mice by treatment with cytotoxic agents. Jpn J Surg 2039-96, 1990. Fukutomi T, IkedaT, Nishiumi T , et al.: Effect of hydroxy tamoxifens against human breast carcinomas transplanted into nude mice. J Jpn Cancer Ther 21:5&59, 1986 (in Japanese with English abst). Kohn J: Experimental chemo and endocrine therapy of human breast carcinomas serially transplanted into nude mice. Jpn J Surg 89:91-102, 1988 (in Japanese with English abst). Kubouchi K: Experimental and clinical study on endocrine therapy for breast carcinoma with special reference to nuclear estrogen receptor. Jpn J Surg 85:1405-1417, 1984 (in Japanese with english abst). Kubota T , Nakada M , Tsuyuki K, et al.: Cell kinetics and chemosensitivity of human carcinomas serially transplanted into nude mice. Jpn J Cancer Res 77:502-507, 1986. Kubota T, lshibiki K , Abe 0: The clinical usefulness of human xenografts in nude mice. In Hall C (ed): “Prediction of Response to Cancer Therapy.” New York: Alan R, Liss, 1988, 213-225. Fukutomi T , Kubota T, Ikeda T , et al.: Experimental combined hormone therapy on human breast carcinomas serially transplanted into nude mice. Jpn J Cancer Res 77:92-97, 1986. Knabbe C , Lippman ME, Wakefield LM, et al.: Evidence that transforming growth factor-p is a hormonally regulated negative growth factor in human breast cancer cells. Cell 48:417428, 1987.