Treatment of Breast Cancer with Different Antiprogestins: Preclinical and Clinical Studies

TREATMENT OF BREAST CANCER WITH DIFFERENT ANTIPROGESTINS: PRECLINICAL AND CLINICAL STUDIES

G. H. BAKKER, B. SETYONO-HAN,H. PORTENGEN,F. H. DE JONG,j J. A. FOEKENS and J. G. M. KLIJN Division of Endocrine Oncology (Biochemistry and Endocrinology), The Dr Daniel Den Hoed Cancer Center, P.O. Box 5201, 3008 AE Rotterdam and 1Department of Endocrinology, Erasmus University, Medical Faculty, The Netherlands

Summary--Treatment with antiprogestins is a new treatment modality for breast cancer. Previously, in rats with DMBA-induced mammary tumors we observed significant growth inhibitory effects of chronic treatment with the antiprogestin mifepristone (RU486). In addition, in 11 postmenopausal breast cancer patients, we observed one objective response, six instances of short-term stable disease, and four instances of progressive disease. Side-effects appeared mainly due to antiglucocorticoid properties of the drug. Increased plasma estradiol levels were observed which probably resulted from ovarian (rat) and adrenal (patients) steroidogenesis. Combined treatment with an antiestrogen in the rat model caused additive growth inhibitory effects. Tumor inhibition after single treatment with mifepristone or tamoxifen was 90 and 75%, respectively. In contrast, when combined, tumor remission similar to that caused by LHRH-agonist treatment (50%) was observed. Even higher tumor remission was found after combined treatment with mifepristone plus LHRH-agonist (75%). In first studies in the rat model we observed significant tumor growth inhibitory effects with two new antiprogestins of seemingly greater potency which cause less unfavorable endocrine side-effects. In conclusion: combined treatment (antiprogestin plus antiestrogen or LHRH-agonist) may be of value in endocrine threapy of breast cancer.

INTRODUCTION itary-ovarian functions, which resulted in unfavorable endocrine effects (high plasma steroid Antiprogestins form a new category of anti- hormone and prolactin levels) [15-17]. Recently, hormonal agents of potential interest in the we have also reported[18] growth inhibitory treatment of cancer. Recently, mifepristone effects of chronic single treatment with mifepri- (RU486) became available for preclinical stone in postmenopausal patients with research and clinical testing. Besides anti- metastatic breast cancer. These inhibitory progestational properties [1-6], mifepristone has antiglucocorticoid activities as well[7-9]. effects were accompanied by unfavorable side- effects (i.e. increased plasma estradiol concen- Apart from application as a contraceptive trations) due to peripheral conversion of agent [10-14] or as an antiglucocorticoid in the androgens derived from hyperstimulated treatment of conditions related to excess corti- adrenals [18]. costeroid production [9], mifepristone may be used as a growth inhibitor of several types of Therefore, the effects of combined treatment tumor cells. with mifepristone (to block the PgR) and Previously, we reported that different dosages tamoxifen (to block the ER) or LHRH-agonists of the progestin megestrol acetate caused tumor (buserelin or zoladex; to reduce estradiol growth inhibition of 40-50%, whereas the anti- secretion) were investigated. progestin mifepristone gave rise to 80-90% inhibition[15, 16]. This tumor inhibition was MATERIALS AND METHODS observed in spite of stimulation of the pitu- Preclinical studies Hormonal agent solutions. The progestin Proceedings ~f the 2nd International EORTC Symposium on megestrol acetate was obtained from Sigma, "'Hormonal Manipulation of Cancer: Peptides, Growth Factors and New (Anti-)Steroidal Agents", Rotterdam, St Louis, MO. The antiprogestin mifepristone The Netherlands, 9 11 April 1990. (RU486) was generously provided by Roussel- 789 790 G.H. BAKKER et al.

Uclaf (Dr R. Deraedt), Romainville, France. tumor load at the end of treatment (in mm-') and Megestrol acetate and micronized mifepristone the average tumor load of the group as a whole were suspended in olive oil and administered by (in mm 2) at the start of treatment, and multi- subcutaneous injection (final dosages, see plied by 100. After treatment rats were killed by Fig. 1). Tamoxifen-citrate was kindly donated decapitation, and mammary tumors, specific by ICI (Pharmaceutical Division), Macclesfield, organs and blood were collected, as de- Cheshire, U.K. Tamoxifen-citrate was dissolved scribed[15, 16]. Steroid receptor and plasma in physiological saline (I mg/ml) and was hormone assays were performed as de- administered subcutaneously (final dosage: scribed [15, 16]. 400#g/kg/day). The LHRH-agonist buserelin Statistical evaluation of results. The signifi- was generously provided by Hoechst A.G., cance of differences between the results obtained Frankfiirt am Main, F.R.G., in a stock solution in the various treatment groups were calculated of 1 mg/ml, or as a 3.3 mg depot-preparation. using Wilcoxon's test. For subcutaneous injections, a work solution was made by dilution of the stock solution Clinical studies with physiological saline (final dosage: Patients, materials and methods. This study 40 #g/kg/day). In addition, the LHRH-agonist was started after approval by a local Human zoladex was kindly provided by ICI (Pharma- Investigations Committee and by the Dutch ceutical Division) as a 3.6 mg implant. Control Cancer Society (Protocol KWF-CKVO 86-09). animals received daily injections of saline. Eleven postmenopausal patients (mean age, Treatment of rats. Mammary tumor induc- 63 yr; range, 46-75 yr) gave informed consent. tion was realized by intragastric injection of They were treated daily during 3-34 weeks with dimethylbenzanthracene (DMBA) as previously 200-400 mg of mifepristone (RU486, Roussel described [15, 16]. Tumor load of each rat was Uclaf, France) p.o. as a second-line single treat- determined [15, 16]. The results of a 3-wk treatment after first-line treatment with tamoxifen, ment are expressed as relative (%) tumor load irrespective of the response to tamoxifen. In values. The relative tumor load value of a rat in four patients, the receptor status of the primary a certain group is defined as the ratio of its tumor was unknown; three patients had an [ER +, PgR +] tumor, and four patients an [ER +,

°.Zo PgR-] tumor. I00 200 300 Hematological, biochemical and endocrine T I' ' START I I END P-value parameters were studied in 10 patients before

CONTROL and after 4, 8, 12 and 16 wk of treatment (for more details, see [18]). Skin and Lymph node i I i MEGESTROL ACETATE: metastases were measured every 4 wk. X-rays of n=14 ,t ~ I 2.5 mg/kg n.S, specific bone and lung lesions were performed I I o=13 ' ~'I I |0 mg/kg n.S. every 3 months. Bone scans were performed I ...... J ...... before and a half year after start of treatment. i I MIFEPRISTONE: Scoring of response occurred according to Inter- I 2.5 mg/kg <0.005 national Union against Cancer criteria. For I I n=46 I 4 I 10 mg/kg <0.005 statistical methods, we used a Student's paired : I t-test and the non-parametric method of n=15 I --I , 40 mg/kg <0.005 J Wilcoxon. I q I I TAMO×IFEN <0,007 RESULTS 400 ug/kg ...... j I Preclinical studies , I MIFEPRISTONE (IO mg/kg) steroid hormone levels considerable inhibition of tumor growth of were more pronounced after treatment with approx. 75 and 90% respectively, but not in LHRH-agonist implants than after its injection tumor remission. Combined antiprogestational (E2, P < 0.02; Pg, P < 0.005). In combination and antiestrogenic treatment with mifepristone with mifepristone, the effects of treatments on and tamoxifen gave rise to a remission of tumor plasma hormone concentrations were compar- growth by 50%, indicating additive inhibitory able to those caused by the LHRH-agonist. effects (Fig. 1). The same tumor regression was Effects on organ weights and steroid receptor also observed after single treatment with contents of the mammary tumors. For details on LHRH-agonist administered by injection these effects of treatments we refer to [19]. Cyto- (buserelin). Single LHRH-agonist treatment solic tumor ER contents were decreased by all using implants (buserelin or zoladex) was treatment modalities. Tumor PgR contents significantly more efficacious in causing tumor tended to increase after tamoxifen treatment remission than by treatment with daily injec- and were significantly decreased after mifepris- tions of buserelin (remission by 70 vs 50%, tone treatment. Mifepristone treatment made respectively, P <0.03; [19]). Finally, combi- the tumors nearly PgR-negative. Treatment nation treatment with mifepristone and daily with LHRH-agonist implants caused lower ster- LHRH-agonist injections (buserelin) resulted in oid hormone receptor contents than with daily significantly greater tumor growth inhibitory injections of LHRH-agonist. Cytosolic tumor effects than single treatment with LHRH- PgR level was virtually undetectable after com- agonist injections (approx. 75 vs 50% tumor bined treatment with mifepristone and injected remission, respectively, P < 0.002; [19]). When LHRH-agonist. mifepristone was combined with LHRH-agonist Effects of new antiprogestins on tumor growth. implants, such a combination did not result A comparison of the antitumor effects of in a significantly greater antitumor effect mifepristone with those of two new anti- than single treatment with LHRH-agonist im- progestins ORG31710 and ORG31806 is shown plants. in Fig. 2. It was found that treatment with the Effects on mean plasma hormone concen- same dosage of either of these three trations. We have previously described that antiprogestins gave rise to a considerable mifepristone treatment caused increased plasma inhibition of tumor growth (mifepristone, levels of LH, estradiol and progesterone, ORG31806) and even to a slight tumor remis- whereas FSH levels were not significantly sion (ORG31710). However, the differences be- affected[15, 16]. In subsequent studies FSH tween the antitumor effects of the antiprogestins levels were significantly suppressed by 37% were not statistically significant. (P <0.005; [19]), while plasma concentrations of progesterone were not significantly different Clinical studies from those in control animals. After tamoxifen Antitumor effects. One of 11 patients showed treatment all plasma hormone levels studied a partial response of her lymph node metastases were decreased. Combined treatment with (58% decrease of tumor size) lasting 5 months; mifepristone and tamoxifen resulted in plasma six patients showed stable disease of 3-8 month hormone concentrations which were in between duration; and four, progressive disease from those obtained after treatment with only start of treatment (Table l). Anorexia and a mifepristone or tamoxifen: estradiol and pro- slight degree of nausea (8 times); tiredness (6 gesterone levels were unchanged, and LH levels times), not feeling well and dizzziness (4 times) were increased compared to control. However, and somnolence (4 times) were relatively fre- this increase in plasma LH level was signifi- quent complaints (Table 1). cantly less pronounced (P < 0.0035) than after Side-effects generally did not occur in the first single treatment with mifepristone [19]. week of treatment, but sometimes after weeks or LHRH-agonists were administered by subcu- months of treatment, suggesting that certain taneous injections or implants[19]. LHRH- plasma levels of mifepristone have to be reached agonist treatment by injections resulted in before side-effects occur. Ten patients showed a increased levels of LH (30-fold) and FSH (2-3- decrease in body weight (mean, 3 kg; range, fold), whereas treatment with LHRH-analog 0.9-7.5 kg), while patient I l had an initial body 792 G.H. BAKKERet al.

300.00 R~LATIVI~ TUMOR b.OAD (%)

200.00- IIIIIlll A CONTRCL

B 0~331710

177/7/7~ C OP~3mO6

o F~_~86 1(30.00- [ 0.000 A C D 4O 8 7 # Fig. 2. Relative effects on mammary tumor load after different endocrine treatments for 3 wk with 2 x 1 mg/day of ORG31710, ORG31806 or RU486. Results are means + SEM. Below the figure are indicated the number of rats and the statistical significance vs control: *P < 0.05, 8,p < 0.01, *P < 0.005. The initial tumor load = 100%. weight increase of 5 kg which was followed by without physical complaints. These obser- a decrease of 2 kg without other side-effects. We vations suggest that this complication was did not observe significant effects on blood initiated by mifepristone. In view of the long pressure. However, one patient using anti- half-life of mifepristone of about 20h, the hypertensive drugs before treatment no longer improvement might be related to both excretion required antihypertensive treatment after the of mifepristone and treatment with high-dose start of treatment with mifepristone. dexamethasone that overcame the antigluco- Two patients had to stop treatment with corticoid activity of the gradually decreasing mifepristone because of side-effects (in the plasma levels of mifepristone. absence of tumor progression). One of these Endocrine effects. In addition to plasma patients was hospitalized because of a grand mal ACTH, plasma cortisol concentrations showed seizure and subcoma under suspicion of cerebral a highly significant increase during treatment, metastases. Treatment with dexamethasone was indicating stimulation of the pituitary-adrenal instituted and mifepristone treatment was axis[18]. In contrast to a normal increase of stopped. However, on the computed tomogra- plasma cortisol concentrations by stimulation phy scan of the cerebrum, metastases were not with synacthen before treatment, there was no found, and the patient improved remarkably significant change of the increased basal cortisol after 3 days and was dismissed from the hospital levels by synacthen during mifepristone treatment. Moreover, the increased basal corti- Table 1. Antitumor and side-effects of second-line endocrine treat- sol concentrations could not be suppressed by ment with mifepristone of II postmenopausal patients with l mg of dexamethasone during mifepristone metastatic breast cancer treatment [18]. Surprisingly, plasma androstene- Antitumor effects Objective response (CR + PR) = 1 x dione and especially plasma estradiol levels Stable disease = 6 × (3-8 months) increased significantly during treatment. We Progressive disease = 4 x found no significant effects on basal and Side-effects Anorexia = 8 × stimulated gonadotropin concentrations, while Nausea = 8 x plasma SHBG concentrations slightly de- Dizziness = 4 x Tiredness = 6 x creased. Plasma estradiol concentrations were Sopor = 4 x positively correlated with plasma androstene- Epileptic insult = ! x Reduction in body weight = mean 3 kg (range 0.9-7.5 kg) dione (P < 0.05) and cortisol concentrations Blood pressure = no effect (P < 0.01). Basal prolactin levels did not change Treatment of breast cancer with different antiprogestins 793 during mifepristone treatment, while thyro- ORG31806) at even lower dosages than the tropin-releasing hormone-stimulated plasma 10mg/kg/day previously used, resulted in a prolactin concentrations tended to increase, slight tumor remission. The balance of anti- possibly in relation to the increased estradiol tumor effects and unfavorable endocrine effects levels. appeared improved on the basis of increased antitumor activities and seemingly reduced DISCUSSION endocrine side-effects (not shown) of these new antiprogestins. The antitumor activity and the We have previously demonstrated in exper- mechanism of action of other antiprogestins, imental breast cancer studies that mifepristone such as onapristone, have been described by is a more potent tumor growth inhibitor than Schneider et al. [21] in studies of the DMBA- or the progestin megestrol acetate in rats bearing MNU-induced mammary tumors in the rat. DMBA-induced mammary tumors[15, 16]. In It is concluded that in treatment with mifepri- addition, it was found that the antitumor effect stone, the additional blockade of the tumor ER of mifepristone was not different from that of by tamoxifen or the decrease in plasma steroid the antiestrogen tamoxifen[19]. The tumor hormone levels caused by LHRH-agonists, pre- growth inhibitory effects of mifepristone oc- vented the unfavorable endocrine effects re- curred in spite of increased plasma concen- sulting from single treatment with mifepristone. trations of estradiol (and prolactin; cf. [15]) as a The combined effects result in a more pro- consequence of stimulation of the pituitary- nounced tumor growth inhibition during com- ovarian axis[15, 16]. Also, in postmenopausal bined endocrine treatment. Moreover, new women with metastatic breast cancer, chronic antiprogestins (and their combinations with single second- or third-line treatment with other hormonal agents) will be tested for the mifepristone caused tumor growth inhibitory appreciation of their therapeutic values in the effects [18, 19]. In our group of postmenopausal treatment of human breast cancer. patients, and similar to the results in the rat, Acknowledgements--We gratefully acknowledge receiving these inhibitory effects were observed in spite the hormonal agents used in this study, obtained from of increased estradiol concentrations [18]. The Roussel-Uclaf (Dr R. Deraedt, mifepristone), ICI (tamoxi- latter was caused presumably by stimulation of fen citrate, zoladex implants), Hoechst (Dr J. Sandow, buserelin solution and implants) and Organon International the pituitary-adrenal axis, followed by periph- B.V. (ORG31710, ORG31806). In addition, we would like eral conversion of adrenal androgens into estro- to thank the Dutch Cancer Society (kwf) for support (grant gens[18]. These observations, together with DDHK ~g-6). those of direct growth inhibition of cultured mammary tumor cells [1-3, 16], indicate that the REFERENCES main mechanism of action of antiprogestational 1. Bardon S., Vignon F., Chalbos D. and Rochefort H.: treatment is a direct growth inhibitory effect at RU486, a progestin and glucocorticoid antagonist, in- the level of the mammary tumor cells. The hibits the growth of breast cancer cells via the progester- effects of combined treatment with mifepristone one receptor. J. Clin. Endocrinol. Metab. 60 (1985) 692-697. (to block the PgR) and tamoxifen (aiming also 2. Bardon S., Vignon E., Montcourrier P. and Rochefort at blockade of the ER), or with mifepristone H.: Steroid receptor-mediated cytotoxicity of an anti- and an LHRH-agonist (to decrease estrogen estrogen and an antiprogestin in breast cancer cells. Cancer Res. 47 (1987) 1441-1448. secretion) indicate that combined antiproges- 3. Gill P. G., Vignon F., Bardon S., Derocq D. and tational-antiestrogenic treatment with mifepris- Rochefort H.: Difference between R5020 and the anti- tone plus tamoxifen was superior to treatment progestin RU486 in anti-proliferative effects on human breast cancer cells. Breast Cancer Res. Treat. 10 (1987) with either drug alone, i.e. combined treatment 37-45. resulted in strong tumor remission compared to 4. Herrmann W., Wyss R., Riondel A., Philibert D., only tumor growth inhibition, but no remission, Teutsch G., Sakiz E. and Baulieu E. E.: The effects of an antiprogesterone steroid in women: interruption of after treatment with either agent[19]. Pre- the menstrual cycle and of early pregnancy. Comptes viously, additive growth inhibitory effects on Rendus Acad. Sci. Paris 294 (1982) 933-938. breast cancer cells were observed in vitro with 5. Horwitz K. B.: Progestins, antiprogestins and progesterone receptors in breast cancer: mechanisms and biologi- certain concentrations of mifepristone and cal actions. J. Steroid Biochem. 19 (1983) 106S (abstr. tamoxifen [2]. 319). Interestingly, treatment of rats bearing 6. Schaison G., George M., Lestrat N., Reinberg A. and Baulier E. E.: Effects of the antiprogesterone steroid DMBA-induced mammary tumors with two RU486 during the mid-luteal phase in normal women. new antiprogestins (i.e. ORG31710 and J. Clin. Endocrinol. Metab. 61 (1985) 484-489. 794 G.H. BAKKER et al.

7. Lamberts S. W. J., Uitterlinden P., Bons E. G. and experimental breast cancer in rats. In Hormonal Verleun T.: Comparison of the actions of RU38486 and Manipulation of Cancer: Peptides, Growth Factors and megestrol acetate in the model of a transplantable New (Anti)steroidal Agents (Edited by J. G. M. Klijn, adrenocorticotropin- and prolactin-secreting rat pitu- R. Paridaens and J. A. Foekens), EORTC Monograph itary tumor. Cancer Res. 45 (1985) 1015-1019. Series. Raven Press, New York, Vol. 18 (1987) 8. Moguilewsky M. and Philibert D.: RU38486: potent pp. 39-45. antiglucocorticoid activity correlated with strong bind- 16. Bakker G. H., Setyono-Han B., Henkelman M. S., De ing to the cytosolic glucocorticoid receptor followed by Jong F. H., Lamberts S. W. J., Van der Schoot P. and an impaired activation. J. Steroid Biochem. 20 (1984) Klijn J. G. M.: Comparison of the actions of the 271-276. antiprogestin mifepristone (RU486), the progestin 9. Nieman L. K., Chrousos G. P., Kellner C., Spitz I. M., megestrol acetate, the LHRH-analog buserelin, and Nisula B. C., Cutler G. B., Merriam G. R., Bardin ovariectomy in treatment of rat mammary tumors. C. W. and Loriaux D. L,: Successful treatment Cancer Treat. Rep. 71 (1987) 1021 1027. of Cushing's syndrome with the glucocorticoid antag- 17. Van der Schoot P., Bakker G. H. and Klijn J. G. M.: onist RU486. J. Clin. Endocrinol. Metab. 61 (1985) Effects of the progesterone antagonist RU486 (mifepri- 536-540. stone) on ovarian activity in the rat. Endocrinology 121 10. Baulieu E.: Fertility control in women: results with (1987) 1375-1382. RU486 by the end of 1985. J. SteroidBiochem. 25 (1986) 18. Klijn J. G. M., De Jong F. H., Bakker G. H., Lamberts 847-851. S. W. J., Rodenburg C. J. and Alexieva-Figusch J.: 11. Couzinet B., Lestrat N., Ulmann A., Baulieu E. E. Antiprogestins: a new form of endocrine therapy for and Schaison G.: Termination of early pregnancy by breast cancer. Cancer Res. 49 (1989) 2851-2856. the progesterone antagonist RU486 (mifepristone). 19. Bakker G. H., Setyono-Han B., Portengen H., De Jong N. Engl. J. Med. 315 (1986) 1565-1570. F. H., Foekens J. A. and Klijn J. G. M.: Endocrine 12. Das C. and Cart K. J.: Antifertility actions of the and antitumor effects of combined treatment with an progesterone antagonist RU486 include direct inhi- antiprogestin and antiestrogen or luteinizing hormone- bition of placental hormone secretion. The Lancet 2,i releasing hormone agonist in female rats bearing mam- (1987) 599-601. mary tumors. Endocrinology 125 (1989) 1593 1598. 13. Ulmann A.: The antiprogestins: a recent advance in 20. Romieu G., Maudelonde T., Ulmann A., Pujol H., fertility regulation. J. Steroid Biochem. 27 (1987) Grenier J., Cavalie G., Khalaf S. and Rochefort H.: The 1009--1012. antiprogestin RU486 in advanced breast cancer: pre- 14. Grimes D. A., Mishell Jr D. R., Shoupe D. and Lacarra liminary clinical trial. Bull. Cancer 74 (1987) 455-459. M.: Early abortion with a single dose of the anti- 21. Schneider M. R., Michna H., Nishino Y. and El Etreby progestin RU486. Am. J. Obstet. Gynecol. 158 (1988) M. F.: Antitumor activity and mechanism of action of 1307-1312. different antiprogestins in experimental breast cancer 15. Bakker G. H., Setyono-Han 8, De Jong F. H. models. J. Steroid Biochem. Molec. Biol. 37 (1990) and Klijn J. G. M.: Mifepristone in treatment of 783 787.