Profound Suppression of Plasma Estrogens by Megestrol Acetate in Postmenopausal Breast Cancer Patients’

Home , Aminoglutethimide, Megestrol, Megestrol acetate, Testosterone (patch)

Vol. 2, 1515-1521, Septenber /996 Clinical Cancer Research 1515

Steinar Lundgren, Svein I. Helle, and been shown to induce tumor regression in postmenopausal Per E. L#{248}nning2 women with hormone-sensitive breast cancers ( 1 ). Randomized studies have revealed similar response rates to progestins and Department of Oncology. Trondheim University Hospital, 7006 Trondheim IS. L.J, and Department of Oncology, Haukeland Hospital. other contemporary treatment modalities like tamoxifen (2, 3) University of Bergen. N-502I Bergen IS. I. H., P. E. LI, Norway and the aromatase inhibitor aminoglutethimide (4). Despite the fact that pnogestins are shown to be effective against breast cancer, the exact mechanism(s) of their antitumor ABSTRACT action is not known. Several mechanisms, like down-regulation Twelve postmenopausal women suffering from ad- of the concentration of the estrogen receptor (5), alterations in vanced breast cancer had plasma estrogens, androgens, tissue metabolism of estrogens (6). suppression of plasma es- cortisol, and gonadotropins determined before therapy trogens (7), interactions with growth factors (8), and a direct and during treatment with megestrol acetate (MA) in oral cytostatic influence on tumor cells (9) have all been proposed. doses escalated from 40 to 160 mg. The plasma clearance and production rate of estrone and estrone sulfate were Progestins are known to suppress adrenal steroid synthesis, determined before treatment and after 4 weeks of therapy reducing the secretion of cortisol and also the secretion of with 160 mg MA. Treatment with MA suppressed plasma androstenedione and testosterone ( 10). Estrogen production in levels of dehydroepiandrosterone sulfate, androstenedi- postrnenopausal women occurs by conversion of circulating one, and cortisol in a dose-dependent manner to < 10% of androgens into estrogens in peripheral tissue ( 1 1 ). MPA and MA pretreatment values. Plasma testosterone, estradiol, es- have been reported to suppress plasma E, and E, by 20-40% trone, and estrone sulfate were suppressed to 18-29% of (7, 10, 12). However, in two previous studies, we found MA to pretreatment values, whereas the gonadotropins were suppress plasma E,S to less than 50% of its pretreatment value suppressed to 35-52%. The plasma clearance rates of (10, 13), suggesting MA could have a selective influence on the estrone and estrone sulfate were increased by a mean metabolism or PR ofthis estrogen conjugate. E,S is synthesized value of 23.7% (P < 0.01) and 23.5% (P < 0.025), whereas from circulating E, and E, (14, 15). Plasma levels of E1S are the production rates were reduced by 76.7% (P < 0.0005) about 5- and 20-fold higher than plasma levels of E, and E2, and 76.1 % (P < 0.0005), respectively. Our findings mdi- respectively, in postmenopausal women (16, 17). Although E,S cate that MA causes profound suppression of adrenal is not biologically active on its own, breast cancer tissue con- steroid production but in addition suppresses ovarian tains the enzymes required to convert E1S into the biologically secretion of androgens in postmenopausal breast cancer active E, (18, 19). Because circulating E1S may be an important patients. The reduction in plasma estrogens is comparable estrogen reservoir in postrnenopausal women, suppression of the to values obtained with commonly used aromatase inhib- plasma level of this estrogen conjugate may be of importance to itors and may be responsible for its antitumor effects in the antitumor effect of progestins. breast cancer. The aim of this study was to evaluate the influence of treatment with MA in different doses on the plasma concentra- INTRODUCTION tion, metabolic clearance rate, and PRs of estrogens in post- Endocrine therapy plays a pivotal role in the treatment of menopausal breast cancer patients. Thus, we measured plasma breast cancer. Currently, several treatment options, including levels of E, E2, and E,S together with their precursors, andro- antiestrogens, estrogen deprivation (castration in premenopausal stenedione and testosterone. Because both these androgens have women or aromatase inhibitors in postmenopausal ones), or an adrenal and also an ovarian origin in postrnenopausal women, pnogestins are available. we measured plasma DHEAS, cortisol, and gonadotropins to Synthetic progestins given in high-dose regimens (MPA address the influence of MA on each glandular system. To in a dose of 1000 mg on MA in a dose of 160 mg daily) have evaluate the influence of MA treatment on the pharmacokinetics of E, and E1S, we measured the Cl of E, and E,S and the fraction of E, transferred into E,S after injection of [‘4C]E, and [3H]ES before treatment and after 4 weeks of therapy with MA Received 3/I 5/96: accepted 5/14/96. ( 160 rng o.d.). From these data we calculated the PR of E, and I Supported by grants from the Norwegian Cancer Society and the ES before and during treatment. Cancer Foundation at Trondheim University Hospital. 2 To whom requests for reprints should be addressed. Phone: 47-55- 972010; Fax: 47-55-972()46. 3 The abbreviations used are MPA, medroxyprogesterone acetate; AUC, area under plasma concentration curve: Cl. plasma clearance rate; DHEAS, dehydroepiandrosterone sulfate; E,, estrone; E,, estradiol; E,S, compound (also termed ‘transfer factor”): MA, megestrol acetate; o.d.. estrone sulfate: fni, fraction of a compound metabolized into another once daily; PR, production rate.

PATIENTS, MATERIALS, AND METHODS respectively, and the intra-assay coefficient of variation was Patients. Twelve patients suffering from advanced breast 4.4%, 3.9%, and 5.9%, respectively. cancer who were to receive MA treatment for progressive dis- Plasma androstenedione, testosterone, DHEAS, and cortisol were determined by commercial RIA, and the gonadotropins ease took part in the investigation. The protocol was approved were determined by commercial IRMA kits (Orion Diagnostica, by the regional ethical committee. All patients gave their written Diagnostic Products Corp., and Diagnostic System Laborato- informed consent. Their median age was 67 years (range 60-78 ries). The intra-assay coefficient of variation was <7% for all years), median body weight was 71 kg (range 55-83 kg), and analyses. All samples obtained from each patient were analyzed median height was 1.62 rn (range 1.53-1.67 m). Two of the in the same batch. patients were moderate smokers who did not change their smok- Measurement of the Cl of E1 and E1S by Isotope Injec- ing habits during the last months before the investigation or tions. The Cl of E1 and E1S was determined by administering during the investigation period; the other 10 patients were a mixture of [3H]E1S and [‘4C]E1 as a bolus injection followed nonsmokers. None of the patients received any other forms of by determination of the concentration of radioactive E and E1S endocrine therapy, anticancer treatment, or drugs known to in the plasma after certain time intervals as described previously enhance or inhibit drug-metabolizing enzymes during the inves- (15, 21). Briefly, each patient received 25 iCi of [4-’4C]E1 and tigation period. Previous anticancer therapy was terminated 4 75 iCi of [6,7-3H]E1S dissolved in 20 ml of ethanol:saline 0.9% weeks before commencing treatment with MA. (8:92, v/v) as a 1-mm bolus injection. The midpoint of the Study Protocol. The treatment schedule was as follows: injection was taken as time zero. Blood samples were drawn each patient commenced on MA at a dose of 40 rng o.d. from an indwelling needle in the opposite arm immediately (Megestat#{174} 40 mg tablets; Bristol Arzneimittel Gmbh, Munich, before injection of the tracers and after 5, 10, 15, 22.5, 30, 45, Germany). The dose was subsequently escalated every fourth 60, 90, 120, 150, 180, 2 10, 240, 300, 360, 480, 600, 720, and week to 80 mg o.d., 120 mg o.d., and, finally, 160 mg o.d. After 900 mm. the study period was completed, each patient continued treat- The Cl for E1 and E1S was calculated from the equation: ment with MA 160 mg o.d. until evidence ofdisease progression Cl = Dose/AUC, where AUC is the area under the elimination (none of the patients had progressive disease during the study curve. In previous studies, we have shown that the concentration of radioactive E1S over time after a bolus injection of tracer E1S period). Blood samples for plasma hormone measurement were may not be described by a simple mathematical function (15, obtained before initiating therapy and subsequently at the end of 21), probably due to enterohepatic cycling of estrogens with a substantial amount of the material reabsorbed entering the each 4-week treatment interval. In addition, each patient had Cls plasma as conjugates due to sulfatation or glucuronidation in the of E1 and E1S determined before initiating therapy and after 4 splanchnic area and liver (22, 23). Thus, the area under the weeks of treatment with MA 160 mg o.d. [3H]E1S concentration curve was determined by the trapezoid Materials. [6,7,-3H]E1S (60 Ci/mmol), [2,4,6,7-3H]E1 rule, adding the residual by extrapolating to infinity. (85-105 Ci/mmol), and L4-14CE1] (50-60 Ci/mol) were ob- Plasma concentration of [‘4C]E1 as a function of time was tamed from DuPont New England Nuclear (Dreiech, Germany). tested for goodness of fit to a multicompartmental exponential E2-6-carboxymethyloxime-[2- ‘25ljiodohistamine (-2000 Cu function using Kaleidograph software on an Apple MacIntosh mmol) was obtained from Amersharn International (U.K.). LC computer. Although the elimination curves for E3 were Sephadex LH-20 was obtained from Pharmacia (Uppsala, Swe- found to be well fitted to a three-compartmental model in den), and sulfatase (S-9754) was obtained from Sigma (United several instances, in 8 of I 2 patients the elimination curve for E1 Kingdom). All solvents were of analytical or spectrophotometric did not fit to a three-compartmental model in one or both of the grade and were obtained from Merck (Darmstadt, Germany) test situations (one or more of the parameters gave a negative except for ethanol, which was obtained from A/S Vinmonopolet value). In these patients, the curves were described by a two- (Oslo, Norway). compartmental model in both of the test situations. Therefore, to Plasma Hormone Measurements. Plasma estrogens handle all cases in a uniform way, the AUC for E1 was calcu- (E2, E1, and E1S) were determined as reported elsewhere (16, lated by use of the trapezoid rule in all patients. In addition, the 20). Briefly, [3H]E1S (-400 cpm) was evaporated to dryness in AUC was calculated by integrating the multicompartmental test tubes. Plasma (2 ml) was added, and the samples were function (three-compartmental model in four patients, two-com- allowed to equilibrate overnight. Unconjugated estrogens were partmental in eight patients). extracted with diethyl ether, were purified on a Sephadex LH-20 The fraction of E1 metabolized to E3S (fm) was calculated column using benzene:methanol (90: 10, v/v) as eluent, and were from the equation (24): measured by RIA. E, S was hydrolyzed by sulfatase [5-9754; 1.5

mg dissolved in 2 ml 0.2 M sodium acetate buffer (pH 5)], and fin (AUC[’4C]E,S x C1Es)/AUC[’4C]E, X CIE the unconjugated E1 was extracted and purified on Sephadex LH-20 columns using dichloromethane. E1 was subsequently and the PR for plasma E1 and E1S from the equations: reduced into E2 by sodium borohydride in methanol. The sample PR E, = CI E x Plasma concentration E, was neutralized with sodium acetate, and E2 was extracted with diethyl ether, was purified on a LH-20 column, and was rneas- and ured with RIA. The sensitivity limit of the assay was 2.1

pmol/liter, 6.3 pmol/Iiter, and 2.7 pmol/liter for E2, E1, and E1S, PR ES = CI ES x Plasma concentration E,S

800 40

400 20’

I. 200 a 10’ I 5. 50-j

2.5 25 Control 80mg 160mg Control 40mg 80mg 120mg 160mg

12.81 80

a 40

C E C 20

10 Control 80mg 160mg

3.20

1.60 3.2

1 0.80 a 1.6

L. a :: 0.40 C 0.8

- 0.20’ I 0.4 0.10-’ 0.2

0.05 0.1 Control 40mg 80mg 120mg 160mg Control 80mg 160mg

Fig. I Plasma levels of cortisol (top), androstenedione (middle), and Fig. 2 Plasma levels of luteinizing hormone (top), follicle-stimulating DHEAS (bottom) in patients before and during treatment with MA at hormone (middle), and testosterone (bottom) before and during treat- different doses. Geometrical mean values with 95% confidence intervals ment with MA at different doses. Geometrical mean values with 95% of the mean. confidence intervals of the mean.

Statistics. In a previous study, we reported plasma treatment with 160 rng o.d. MA were compared by the estrogen and androgen levels in postmenopausal breast can- Wilcoxon matched pair sign rank test. cer patients to be best fitted to a lognormal distribution ( 16). Thus, plasma hormone concentrations expressed as absolute RESULTS values and values obtained during treatment with MA ex- Plasma Hormone Levels. Plasma hormone levels before pressed as percentage of pretreatment values are given as and during treatment with MA at different doses are depicted in their geometrical mean values with 95% confidence intervals Figs. 1-3. Values obtained during treatment, expressed as per- of the mean. In addition, plasma hormone levels obtained in centage of pretreatment levels, are shown in Table 1 . Treatment different situations were compared using the Friedman test with MA in doses from 40 mg o.d. to 160 mg o.d. caused a (two-way nonparametnical analysis of variance), and the dose-dependent suppression of plasma androstenedione, clearance rates and PRs for E1 and E,S before and during DHEAS, and cortisol to a mean value of less than 10% of

Table I Plasma hormone values and hormone ratios during treatment with escalating doses of MA given as percentage of pretreatment level. Geometrical mean values with 95% confidence 80 intervals of the mean

a Drug dose

40 Steroid” 40 mg 80 mg 120mg 160mg

C E E2 69(49-96) 47(33-57) 31 (21-47) 29(18-47) C. 20 E1 54(33-88 40(22-70) 19(12-32) 16(10-32) E1S 66(43-100) 50(34-73) 25(17-35) 18(13-25) C 75(53-96) 54(34-86) 26(15-45) 9(6-13) DHEAS 59 (45-78) 39 (25-60) 21 (14-32) 9 (5-14) 10 A 84(79-104) 51 (28-93) 22(10-46) 8(4-13) T 44 (29-67) 20 (14-31) LH 40(27-60) 35(18-66) FSH 51 (39-66) 52 (38-71) E,1E2 78(59-104) 85(65-112) 62(46-84) 63 (45-87) 40 ES/E, 96 (64-144) 107 (76-149) 79 (58-107) 62 (47-81) ES/E 122(81-184) 125(80-194) 126(78-203) 98(64-150) E1/A 64(35-114) 78(32-189) 99(32-309) 237(93-608)

20 a Abbreviations: C. cortisol; A, androstenedione; T, testosterone; LH, luteinizing hormone; FSH, follicle-stimulating hormone.

C E P < 0.0025 and P < 0.025, respectively). Considering the E1:A 0. 5 ratio, the 95% confidence interval for the values obtained during treatment expressed as percentage of pretreatment values spanned the 100% value at treatment with all doses of MA. 2.5 ontrol 40mg 80mg 120mg 160mg However, a Friedman test revealed a significant difference between the different test situations (P < 0.025), caused mainly by an increase in the ratio during treatment with 160 mg o.d. MA. Plasma Pharmacokinetics of E1 and E1S. The Cl of E1

1600 and E1S, the fraction of E1 converted into E1S, and the PR of E1 and E1S before treatment and after 4 weeks of therapy with MA (160 mg o.d.) are given in Table 2. Figure 4 shows the plasma 800 concentration curves of radioactive E1 and E1S in a representative patient. Treatment with MA increased the Cl ofE1 and E1S 400 1 a by 23.7% (P < 0.01) and 23.5’7c (P < 0.025), respectively. A small decrease in the transfer of E1 to E1S was seen (mean 200 C decrease of 6%, P < 0.05). Contrariwise, treatment with MA E (160 mg o.d.) caused a pronounced suppression of the PRs of 0. 100 plasma E1 and E,S from mean values of 1.89 nmol/h and 2.02 nrnol/h before treatment to 0.44 nmol/h and 0.48 nmol/h, re- 50 spectively, during treatment with MA (mean suppression of Control 40mg 80mg 120mg 160mg 76.7% and 76. 1%, respectively; P < 0.0005 for both). Fig. 3 Plasma levels of E1 (top), E2 (middle), and E1S (bottom) before Calculating the AUC for E1 by integrating the area under and during treatment with MA at different doses. Geometrical mean the exponential curve provided values for AUC that were some- values with 95% confidence intervals of the mean. what lower than what was achieved with the trapezoidal rule (mean difference of 31.3%). The main reason for this discrepancy is most probably that a two-compartmental model under- pretreatment levels, whereas plasma testosterone was sup- estimates the terminal pant of the AUC. When the percentage pressed to 20% of its pretreatment level. Similarly, plasma change in the clearance rate for E1 caused by treatment with MA levels of E,, E1, and E1S were suppressed in a dose-dependent was calculated using AUC values obtained by integration of the manner to 29%, 16%, and 18% of pretreatment levels, respec- exponential curves, this revealed a mean increase in the clear- tively. Although treatment with MA suppressed plasma levels of ance rate of 24.7% (95% confidence interval, 1 .9-52.6%; P < luteinizing hormone and follicle-stimulating hormone by 48- 0.05). 65%, this suppression was fully developed at a dose of MA 80 mg o.d. DISCUSSION Treatment with MA was found to have no influence on the Plasma levels of androgens, estrogens. and gonadotropins ratio of plasma E1S:E1. Contrariwise, the E1S:E2 and the E1:E2 before treatment were in the same range as reported previously ratios were both reduced during treatment with MA (Friedman for postmenopausal women from our group as well as by others

Table 2 Cl and PR for E1 and E1S and the fraction of E1 converted into E15 before and during treatment with MA (160 mg o.d.). Geometrical mean values (with 95% confidence intervals).

Before treatment During treatment Percentage change P

PR E 1.89 (1.15-3.1 1) 0.44 (0.33-0.59) -76.7 (-87.4-56.8) <0.0005 PR E,S#{176} 2.02 (1.32-3.08) 0.48 (0.37-0.62) -76.1 (-83.5-65.3) <0.0005 Cl E1” 24.6 (22.0-27.5) 30.5 (26.3-35.2) +23.7 (+8.5+1.1) <0.01 Cl EISb 3.8 (3.0-4.8) 4.7 (3.6-6.1) +23.5 (+7.5+41.9) <0.025 fm’ 76.9 (65.2-90.6) 72.5 (59.9-87.8) -6.0 (- 10.4-0.6) <0.05 (I nmol/h. bliterTh

C %

previous studies compared to the results of this investigation. 1 100 a Similar to our findings, a recent investigation by Pommier et al. (3 1 ) found MA ( I 60 mg daily) to suppress plasma levels of #{149}0a 10 DHEA and gonadotrophins. In contrast to our findings, these a1 authors reported MA to increase plasma levels of E2 by 91%. C 0 However, the RIA used to determine plasma E2 in that investi- 1 0 E 0 0 gation had a sensitivity limit of about 80 pmol/liter, which is 0 A #{163} a about 4-fold the mean pretreatment concentration of E2 meas- A C ured in this study. Earlier studies by us 13) and others ( 12, 32, #{149}0 ( .1 . #{149}5 C- .. S 33) have shown treatment with MA and MPA to suppress C plasma E1 and E2 by 20-40%, although one study reported

.01 plasma E2 to be suppressed by 65-70% (33). This differs from 0 3 6 9 12 15 the findings ofthis study, in which MA (160 mg o.d.) was found Time (Hours) to suppress all plasma estrogens by 71-82%. Accordingly, the results presented here show treatment with MA 160 mg o.d. to Fig. 4 Elimination curves for [3H]E1S (0). [‘4C]E1S (s), and l’4CIE (#{149})after administration of [‘4C]E, and [3H]E1S to a representative be as effective as many aromatase inhibitors in suppressing patient before treatment with MA (160 mg o.d.). plasma estrogen levels in postmenopausal women (30). Treatment with MA suppressed plasma levels of androstenedione and testosterone to less than 10% and 20% of pretreatment levels, respectively. Progestins in high doses are (16, 25, 26). The Cls for E1 and E1S were in the low range known to express glucocorticoid agonistic effects (34), and the compared to values obtained by us in previous studies (15, 21). observation that all adrenal steroids were suppressed to a similar Despite well-documented clinical effects, the mechanism extent suggests that MA suppresses adrenal steroid synthesis by causing the antitumor effects of progestins in postmenopausal a glucocorticoid-agonistic suppression of adrenocorticotropic breast cancer patients is not known. However, the observation hormone secretion (35). that the expression of estrogen and progesterone receptors pre- The discrepancy between the suppression of testosterone dicts a response to treatment with progestin therapy (27) sug- and steroids of adrenal origin (androstenedione, cortisol, and gests endocrine mechanisms to be involved. Estradiol is a most DHEAS) could be due to a significant contribution from ovarian potent mitogen to estrogen receptor-positive breast cancer cells. synthesis of testosterone (28, 29). The finding that MA (160 mg Although ovarian estrogen synthesis ceases at menopause, o.d.) suppresses plasma levels of gonadotropins by 48-65% estrogens are synthesized by peripheral aromatization of circu- indicates that treatment with MA suppresses ovarian steroid lating androgens. The major pathway is aromatization of andro- synthesis, albeit to a somewhat smaller extent compared to stenedione into E1 , with a minor contribution from aromatiza- suppression of the adrenals. tion of testosterone into E, ( I I ). The main source of these For unexplained reasons, treatment with MA caused a more androgens is the adrenal gland, but the postmenopausal ovary pronounced suppression of plasma androgens than estrogens. contributes significantly to circulating testosterone (28, 29). This occurred despite the fact that MA treatment enhanced the Whereas plasma concentrations of estrogens are considerably Cl of E1. However, this difference occurred during treatment lower in postrnenopausal women compared to premenopausal with MA 160 mg o.d. only, as doses of 40-120 mg o.d. ones, estrogen ablation (through surgical procedures as adrena- suppressed plasma levels of E1 and E1S by the same percentage lectomy or hypophysectomy or medical treatment with aro- as the adrenal steroids. Progestins are known to enhance aromatase inhibitors) is well documented to cause tumor regression matase activity in the endometriurn (36). Although we are not in postmenopausal patients (30). Thus, any influence of pro- aware of any studies evaluating the influence of treatment with gestins on plasma estrogen concentrations may be of importance MA on in vivo aromatization, treatment with MPA (500 mg to the antitumor effects of these drugs. i.miday for 2 weeks) was found to have no influence on total Improvement of the RIAs may explain the difference in body aromatization (37). Glucocorticoids are known to stirnu- plasma androgen and estrogen suppression by MPA and MA in late the aromatase activity in fibroblasts in vitro (38), but short-

term treatment with dexamethasone has been reported not to ACKNOWLEDGMENTS influence aromatization in s’ivo (39). It is noteworthy that in this The technical assistance of D. Ekse and the secretarial work of H. study treatment with MA 40 or 80 mg o.d. both caused a Hjortung and Y. Hornsleth are highly appreciated. nonsignificant reduction in the ratio of plasma E1 to androstenedione, whereas 160 mg caused a nonsignificant increase in this ratio. Although the 95% confidence interval spanned the REFERENCES 100% level in every test situation, a Friedman test revealed a I. Sedlacek, S. M., and Horwitz, K. B. The role of progestins and significant difference in the ratio of plasma E1 to androstenedi- progesterone receptors in the treatment of breast cancer. Steroids, 44: 467-484, 1984. one in the different test situations (P < 0.025). Thus, the 2. Ingle, J. N., Ahmann, D. L., Green, S. J., Edmonson, J. H., Creagan, possibility exists that treatment with MA at a dose of 160 mg E. T., Hahn, R. G., and Rubin, J. Randomized clinical trial of megestrol in vivo o.d. may enhance aromatization. acetate versus tamoxifen in paramenopausal or castrated women with In contrast to our previous investigations (10, 13), we advanced breast cancer. Am. J. Clin. Oncol., 5: 155-160, 1982. found no evidence for a selective suppression of plasma E1S 3. Muss, H. B., Wells, H. B., Paschold, E. H., Black, W. R., Cooper, compared to E1. On the other hand, the finding that plasma E2 M. R., Capizzi, R. L., Christian, R., Cruz, J. M., Jackson, D. V., Powell, was suppressed to a smaller extent compared with plasma E1 B. L., Richards, F. I., White, D. R., Zekan, P. J., Spurr, C. L., Pope, E., and E1S is interesting. Plasma E, in postmenopausal women is Case, D., and Morgan, T. M. Megestrol acetate t’ersus tamoxifen in advanced breast cancer: 5-year analysis-a Phase III trial of the Piedmont partly synthesized by aromatization of testosterone but also Oncology Association. J. Clin. Oncol., 6: 1098-1 106, 1988. arises by the reduction of plasma E1 (11). In this study, we 4. Lundgren, S., Gundersen, S., Klepp, R., L#{248}nning, P. E., Lund, E., and found plasma testosterone to be suppressed to a smaller extent Kvinnsland, S. Megestrol acetate versus aminoglutethimide for meta- than plasma androstenedione, and sustained plasma levels of E, static breast cancer. Breast Cancer Res. Treat., 14: 201-206, 1989. may reflect a higher concentration of its precursor. An alterna- 5. Tseng, L., and Gurpide, E. Effect of progestins on estradiol receptor tive explanation could be a differential effect of MA treatment levels in human endometnium. J. Clin. Endocninol. & Metab., 41: 402-404, 1975. on the Cl of E1 and E2. Although we found MA 160 mg o.d. to 6. Tseng, L., and Gurpide, E. Induction of human endometrial estradiol enhance the clearance rate of E1 and E1S by 23-24%, the dehydrogenase by progestins. Endocrinology, 97: 825-833, 1975. possibility exists that MA may not have a similar influence on 7. Veelen, H. v., Willemse, P. H. B., Sleeker, D. T., Ploeg, E. v. d., the clearance rate of E2. Sluiter, W. J., and Doorenbos, H. Mechanism of adrenal suppression by The finding that treatment with MA 160 mg o.d. suppresses high-dose medroxyprogesterone acetate in breast cancer patients. Can- plasma estrogens with an efficacy comparable to most aro- cer Chemother. Pharmacol., 15: 167-170, 1985. matase inhibitors suggests MA may exert its antitumor effect 8. Ewing, T. M., Murphy, L. J., Ng. M-L., Pang, G. Y. N., Lee, C. S. L., through estrogen suppression. An objection to such a hypothesis Watts, C. K. W., and Sutherland, R. L. Regulation of epidermal growth factor receptor by progestins and glucocorticoids in human breast cancer would be the lack of cross-resistance to treatment with pro- cell lines. Int. J. Cancer, 44: 744-752, 1989. gestins and aromatase inhibitors. Although some authors have 9. Allegra, J. C., and Kiefer, S. M. Mechanism of action of progesta- reported a low response rate to aminoglutethimide in patients tional agents. Semin. Oncol., 12: 3-5, 1985. previously exposed to progestins (40), others (4 1 , 42) have 10. Lundgren, S., L#{248}nning, P. E., Utaaker, E., Aakvaag, A., and shown that a limited number of patients may respond to both Kvinnsland, S. Influence of progestins on serum hormone levels in treatment options in sequence. On the other hand, recent data postmenopausal women with advanced breast cancer. I. General findings. J. Steroid Biochem., 36: 99-104, 1990. suggest a lack of cross-resistance between different aromatase II . L#{248}nning, P. E., Dowsett, M., and Powles, T. J. Postmenopausal inhibitors (43), possibly caused by a differential effect on intra- estrogen synthesis and metabolism: alterations caused by aromatase tumor estrogen synthesis. Accordingly, a lack of cross-resis- inhibitors used for the treatment of breast cancer. J. Steroid Biochem. tance to an aromatase inhibitor and progestins does not refute a 35: 355-366, 1990. hypothesis that progestins may act by suppressing plasma es- 12. Alexieva-Figusch, J., Blankenstein, M. A., Hop, W. C. J., Klijn, trogen levels. J. G. M., Lamberts, S. W. J., Jong, F. H. d., Docter, R., Adlercreutz, H., An interesting option for future studies could be to and Gilse, H. A. v. Treatment of metastatic breast cancer patients with different dosages of megestrol acetate: dose relations, metabolic and combine treatment with a potent arornatase inhibitor with endocrine effects. Eur. J. Cancer & Clin. Oncol., 20: 33-40, 1984. MA to achieve maximal estrogen suppression. Although 13. Lundgren, S., and L#{248}nning, P. E. Influence of progestins on serum studies have reported administration of progestmns in concert hormone levels in postmenopausal women with advanced breast cancer with the aromatase inhibitor arninoglutethirnide (44) as well II: a differential effect of megestrol acetate on serum estrone sulfate and as progestins in concert with aminoglutethimide and tamox- sex hormone binding globulin. J. Steroid Biochem., 36: 105-109, 1990. ifen (45) not to improve response rates compared to single- 14. Ruder, H. J., Loniaux, L., and Lipsett, M. B. Estrone-sulfate: production rate and metabolism in man. J. Clin. Invest., 5/: 1020-1033, drug therapy, these studies are confounded by severe drug 1972. interactions, inasmuch as arninoglutethimide significantly en- 15. L#{248}nning, P. E., Johannessen, D. C., and Thorsen. T. Alterations in hances the metabolism of progestins (46, 47) as well as the production rate and the metabolism of oestrone and oestrone sul- tarnoxifen (48). In addition, arninoglutethirnide inhibits the phate in breast cancer patients treated with aminoglutethimide. Br. J. I I -hydroxylase in the adrenal gland, causing substrate ac- Cancer, 60: 107-1 1 1, 1989. cumulation and enhanced secretion of androstenedione (49). 16. L#{248}nning, P. E., Helle, S. I., Johannessen, D. C., Adlercreutz, H., With the introduction of selective aromatase inhibitors, an Lien, E. A., Tally, M., Ekse, D., Fotsis, T., Anker, G. B., and Hall, K. Relations between sex hormones, sex hormone binding globulin, insu- interesting option is to evaluate the endocrine and clinical lin-like growth factor-I, and insulin-like growth factor binding protein- 1 effects of such drugs and MA in concert in patients suffering in postmenopausal breast cancer patients. Clin. Endocninol., 42: 23-30, from advanced breast cancer. 1995.

17. Roberts, K. D., Rochefort, J. G., Bleau, G., and Chapdelaine, A. 36. Tseng. L., Mazella, J., and Sun, B. Modulation ofaromatase activity Plasma estrone sulfate levels in postmenopausal women. Steroids, 35: in human endometnial stromal cells by steroids, tamoxifen, and RU 486. 179-187, 1980. Endocrinology. 118: 1312-1318. 1986. 18. Santner, S. J., Leszcynski, D., Wright, C., Manni. A., Feil, P. D., 37. Purohit, A., Lai, L. C.. Singh. A.. Coldham, N. G.. Ghilchik, M. W.. and Santen, R. J. Estrone sulfate: a potential source of estradiol in Reed, M. J., and James, V. H. T. The effect of medroxyprogesterone human breast cancer tissue. Breast Cancer Res. Treat., 7: 35-44, 1986. acetate on aromatase and DNA polymerase a activities in breast tu- 19. Beranek, P. A., Folkerd, E. J., Ghilchik, M. W., and James, V. H. T. mours. J. Steroid Biochem., 34: 443-446, 1989. l7-3 hydroxysteroid dehydrogenase and aromatase activity in breast fat 38. Simpson. E. R.. Ackerman, G. E.. Smith. M. E.. and Mendelson. from women with benign and malignant breast tumours. Clin. Endocni- C. R. Estrogen formation in stromal cells of adipose tissue in women: nol., 20: 205-212, 1984. induction by glucocortico-steroids. Proc. NatI. Acad. Sci. USA. 78: 20. L#{248}nning, P. E., and Ekse. D. A sensitive assay for measurement of 5690-5694, 1981. plasma oestrone sulphate in patients on treatment with aromatase inhib- 39. Reed, M. J., Beranek, P. A., Franks, S., and James, V. H. T. The itors. J. Steroid Biochem. Mol. Biol., 55: 409-412, 1995. effect of glucocorticoids on the in vito conversion of androstenedione to 21. L#{248}nning, P. E., Kvinnsland, S., Thorsen, T., and Ueland, P. M. oestrone. Horm. Metab. Res., 18: 635-637, 1986. Alterations in the metabolism of oestrogens during treatment with aminoglutethimide in breast cancer patients: preliminary findings. Clin. 40. Alberto, P., Mermillod, B., Kaplan, E., Goldhirsch, A.. Obrecht, Pharmacokinet., 13: 393-406, 1987. J-P., Jungi, F., Martz, G., Barrelet, L., and Cavalli, F. A clinical trial of aminoglutethimide in advanced postmenopausal breast carcinoma: low 22. Longcope, C., Gorbach, S., Goldin, B., Woods, M., Dwyer, J., and Warram, J. The metabolism of estradiol; oral compared to intravenous response in patients previously treated with medroxyprogesterone. Eur. administration. J. Steroid Biochem., 23: 1065-1070, 1985. J. Cancer & CIin. Oncol., 21: 423-428, 1985. 23. Longcope, C., Yesair, D. W., Williams, K. I. H., Callahan, M. M., 41. Nemoto, T., Rosner, D., Patel, J. K., and Dao, T. L. Aminoglute- Bourget, C., Brown, S. K., Carraher, M. S., Flood, C., and Rachwall, thimide in patients with metastatic breast cancer. Cancer. 63: 1673- P. C. Comparison of the metabolism in dogs of estradioI-l7 following 1675, 1989. its intravenous and oral administration. J. Steroid Biochem.. 13: 1047- 42. Garcia-Giralt, E., Ayme, Y., Carton, M., Daban, A., Delozier, T.. 1055, 1980. Fargeot, P., Fumoleau, P., Gonins, A.. Guenin, D., Guenin, R., Maillant. 24. Rowland, M., and Tozer, T. N. Clinical Pharmacokinetics: Con- P., Mauriac. L., May-Levin. F., Metz, R.. Namer. M.. Olivier. J. P.. cepts and Applications. Philadelphia: Lea & Febiger. 1980. Pommatau. E.. Pouillart, P., Pujade-Lauraine. E.. Rouesse. J.. Serrou, 25. Dowsett, M., Stein, R. C., Mehta, A., and Coombes, R. C. Potency B.. Vitse. M., and Zylberait. D. Second- and third-line hormontherapy in and selectivity of the nonsteroidal aromatase inhibitor CGS I 6949A in advanced postmenopausal breast cancer: a multicenter randomized trial postmenopausal breast cancer patients. Clin. Endocrinol., 32: 623-634, comparing medrozyprogesterone acetate with aminoglutethimide in pa- 1990. tients who have become resistant to tamoxifen. Breast Cancer Res. 26. Demers, L. M. Effects of Fadrozole (CGS l6949A) and Letrozole Treat., 24: 139-145, 1992. (CGS 20267) on the inhibition of aromatase activity in breast cancer 43. Murray, R., and Pitt, P. Aromatase inhibition with 4-OHAndro- patients. Breast Cancer Res. Treat., 30: 95-102, 1994. stenedione after prior aromatase inhibition with aminoglutethimide in 27. Lundgren, S. Progestins in breast cancer treatment. Acta Oncol.. 31: women with advanced breast cancer. Breast Cancer Res. Treat.. 35: 709-722, 1992. 249-253. 1995. 28. Vermeulen, A. The hormonal activity of the postmenopausal ovary. 44. Nagel. G. A.. Wander. H-E., and Blossey. H-C. Phase II study of J. Clin. Endocrinol. & Metab., 42: 247-253, 1976. aminoglutethimide and medroxyprogesterone acetate in the treatment of 29. Sluijmer, A. V., Heineman, M. A., Jong, F. H. d., and Evers, J. L. H. patients with advanced breast cancer. Cancer Res.. 42 (Suppl.): 3442s- Endocrine activity of the postmenopausal ovary: the effects of pituitary 3444s, 1982. down-regulation and oophorectomy. J. Clin. Endocrinol. & Metab., 80: 45. Hardy, J. R., Powles, T. J., Judson, I. R., Sinnett, H. D., Ashley, 2163-2167, 1995. S. E., Coombes, R. C., and Ellin, C. L. Combination of tamoxifen, 30. L#{248}nning, P. E., and Lien. E. Mechanisms of action of endocrine aminoglutethimide, danazol, and medroxyprogesterone acetate in ad- treatment in breast cancer. Cnit. Rev. Oncol. Hematol., 2/: 158-193, vanced breast cancer. Eur. J. Cancer, 26: 824-827. 1990. 1995. 46. Deijk. W. A. v., Blijham, G. H., Mellink, W. A. M.. and Meulen- 3 1 . Pommier, R. F., Woltening, E. A., and Fletcher, W. S. Changes in berg. P. M. M. Influence of aminoglutethimide on plasma levels of serum sex steroid levels during megestrol acetate therapy. Sung. Oncol., medroxyprogesterone acetate: its correlation with serum cortisol. Can- 3: 351-359, 1994. cer Treat. Rep., 69: 85, 1988. 32. Veelen, H. v., Willemse, P. H. B., Sleijfer, D. T., Sluiter, W. J., and Doorenbos, H. Endocrine effects of medroxyprogesterone acetate: rela- 47. Lundgren. S., L#{248}nning,P. E.. Aakvaag. A., and Kvinnsland, S. tion between plasma levels and suppression of adrenal steroids in Influence of aminoglutethimide on the metabolism of medroxyproges- patients with breast cancer. Cancer Treat. Rep., 69: 977-983, 1985. terone acetate and megestrol acetate in postmenopausal patients with 33. Dowsett, M., Lal, A., Smith, I. E., and Jeffcoate, S. L. The effects advanced breast cancer. Cancer Chemother. Pharmacol.. 27: 101-105, of low and high dose medroxyprogesterone acetate on sex steroids and 1990. sex hormone binding globulin in postmenopausal breast cancer patients. 48. Lien, E. A., Anker, G., L#{248}nning,P. E., Solheim, E., and Ueland, Br. J. Cancer, 55: 31 1-313, 1987. P. M. Decreased serum concentrations of tamoxifen and its metab- 34. Briggs, N. H.. and Briggs, M. Glucocorticoid properties of proges- olites induced by aminoglutethimide. Cancer Res., 50: 5851-5857. togens. Steroids, 22: 555-559, 1973. 1990. 35. Leinung, M. C., Liporace, R., and Miller, C. H. Induction of adrenal 49. Lonning. P. E., and Kvinnsland, S. Mechanisms of action of ami- suppression by megestrol acetate in patients with AIDS. Ann. Intern. noglutethimide as endocrine therapy of breast cancer. Drugs. 35: 685- Med., 122: 843-845, 1995. 710. 1988.

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 1996 American Association for Cancer Research. Profound suppression of plasma estrogens by megestrol acetate in postmenopausal breast cancer patients.

S Lundgren, S I Helle and P E Lonning

Clin Cancer Res 1996;2:1515-1521.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/2/9/1515

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/2/9/1515. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.