A New Aromatase Inhibitor, in Postmenopausal Women

Home , Atamestane, Estrone sulfate, Fadrozole

(CANCERRESEARCH52, 5933-5939, November1, 1992J Phase I and Endocrine Study of Exemestane (FCE 24304), a New Aromatase Inhibitor, in Postmenopausal Women

T. R. Jeffry Evans,' Enrico Di Salle, Giorgio Ornati, Mercedes Lassus, Margherita Strolin Benedetti, Eio Pianezzola, and R. Charles Coombes Department ofMedical OncoIoij@,St. Geoa@ge'sHospital Medical School, Creamer Terrace, London SWI7 ORE, England fT. R. I. E.J; Departments of Oncology IE. D. S., G. 0., M. Li and Pharmacokinetics and Metabolism [M. S. B., E. P.J, Farmitalia Carlo Erba, Via Carlo Imbonati, Milan, Italy; and Department of Medical Oncology, Charing Cross Hospital, FuThoin Palace Roa@ London W6 8RF, England (R. C. C.]

ABSTRACT aminoglutethimide and fadrozole (CGS 16949A) (7). Objective tumor regression occurred in approximately 21% of patients Aromatase inhibitors are a useful therapeutic option in the manage treated with 4}IAI@@,2witha low incidence of adverse effects; ment of endocrine-dependent advanced breast cancer. A single-dose 4.5% of patients were withdrawn from treatment because of administration of exemestane (FCE 24304; 6-methylenandrosta-l,4-dl ene-3,17-dione), a new Irreversible aromatase inhibitor, was investi side effects. However, 4HAD undergoes extensive metabolism gated in 29 healthy postmenopausal female volunteers. The compound, in the liver to form the inactive glucuronide (8) and conse given at p.o. doses ofO.5, 5, 12.5, 25, 50, 200, 400, and 800 mg(n = 3â€”4), quently it is recommended that it is given i.m. rather than p.o. was found to be a well tolerated, potent, long-lasting, and specific in The use of aminoglutethimide as an aromatase inhibitor is re hibitor of estrogen biosynthesis. The minimal dose which produced the stricted by the frequency of its adverse events (9), that of fadro maximum suppression ofplasma estrogens was 25 mg, reducing plasma zole by its interference with aldosterone biosynthesis (10, 11). estrone, estradlol, and estrone sulfate to 35, 28, and 39%ofbasal values, In an attempt to achieve maximal aromatase suppression with respectively. This maximum suppression, observed at 3 days, persisted a p.o. effective, nontoxic compound, a new irreversible aro for at least 5 days after administration of a single dose. However,there matase inhibitor, exemestane (6-methylenandrosta-l,4-diene was no interference on cortisol, aldosterone, 17-hydroxyprogesterone, or 3,17-dione; FCE 24304; Fig. 1), has been developed. Exemes dehydroeplandrostenedione sulfate plasma levels.Peak plasma exemes tane concentrations of 27, 221, 343, and 414 ag/mI were reached within tune was found to cause a time-dependent inactivation of 2 h after administration of 50, 200, 400, and 800 mg, respectively. human placental aromatase (tÂ½,13.9 mm) and to have a K1of Plasma concentrations declined rapidly and fell under the detection 26 flM, thus showing an affinity for the enzyme to be 2.6-fold limit (10 ng/ml) at 4 (50 mg) or 24 h (200 and 400 mg). No clinically higher than the substrate androstenedione (Km 69 nM) (12). significant adverse events which could be attributed to the drug were Exemestane induced 30 and 73% regression ofestablished 7,12- reported. Apart from transient eosinophilia in 3 patients, all biochem dimethylbenz(a)anthracene-induced mammary tumors in rats ical and hematological laboratory parameters were within 1.25-fold of when given daily at 10 and 50 mg/kg s.c., associated with 86 the normal ranges. and 93% decrease in total ovarian aromatase activity (13, 14). Consequently we considered exemestane worthy of further INTRODUCTION study in a human population, and we have studied the endo One-third of human breast tumors are hormone dependent crime and toxicity effects and the pharmacokinetics of an esca (1) and estrogens are the most important hormones involved in lating single dose of exemestane in healthy postmenopausal supporting growth of these hormone-dependent tumors (2, 3). female volunteers. The extraglandular conversion of androstenedione to estrone catalyzed by the aromatase enzyme is considered to be the most MATERIALS AND METHODS important source of circulating estrogens in the postmeno pausal woman (4). Consequently, inhibition ofestrogen biosyn Subjects. Twenty-nine postmenopausal women attending the breast thesis by means ofselective aromatase inhibitors is a potentially clinic at St. George's Hospital, London, were recruited into the study. All gave written informed consent, and the trial was approved by the useful therapeutic option in hormone-sensitive breast cancer ethical committee of St. George's Hospital Medical School. The study (5). The enzyme system is localized in the endoplasmic reticu was conducted in two parts. In both parts female subjects over the age lum of cells in which it is expressed and consists of two com of45 years, who were postmenopausal as defined by greater than I year ponents (6). The first is a form of cytochrome P-450 known as since last menses (FSH > 50 milliunits/ml; LH, 15â€”40milliunits/ml) or aromatase cytochrome P-450. This hemoprotein is responsible who had had bilateral oophorectomy wereeligible.The mean age of the for binding the C-19 steroid substrate and catalyzing the series subjects recruited was 60 years (range, 48â€”75 years); the mean time of reactions which leads to the formation of a phenolic A-ring. from menopause was 12.3 years (range, 2â€”36years).Subjects with a The second component is a flavoprotein, NADPH cytochrome clinically significant history of allergic disorders, with cardiac, hema P-450 reductase, which is an ubiquitous protein in the endo tological, respiratory, endocrine, renal, hepatic, or gastrointestinal dis plasmic reticulum of most cell types and which transfers reduc orders, or who had hypertension were excluded, as well as subjects who had participated in a trial in the previous 3 months or who had taken ing equivalents from NADPH to cytochrome P-450 (6). In the any medication in the previous 2 weeks.All volunteers were assessed by search for potent inhibitors of aromatase, various analogues of history, examination, and appropriate investigations before entering the aromatase substrate androstenedione have been investigat the study. All subjects who had previously had breast cancer were fully ed. One of these, 4-hydroxyandrostenedione, an irreversible ar investigated to exclude any evidence of recurrent disease. omatase inhibitor, has been extensively investigated in clinical In studypartA, 15subjectswereinvestigated;14hadpreviouslyhad trials as have some nonsteroidal aromatase inhibitors such as breast cancer but with no evidence of recurrent malignant disease on

Received3/24/9Z accepted8/26/92. The costs ofpublication ofthis article weredefrayedin part by the paymentof 2 The abbreviations used are: 4HAD, 4-hydroxyandrostenedione; DHEAS, pagecharges.Thisarticlemustthereforebe herebymarkedadvertisenaentinaccord dehydroepiandrostenedionesulfate;RIA, radioimmunoassay,FSH, follicle-stimu ance with 18 U.S.C. Section 1734solelyto indicatethis fact. latinghormone;LH, luteinizinghormone;HPLC, highpressureliquidchromatog I To whom requests for reprints should be addressed. raphy. 5933

column chromatography. An aliquot (0.75 ml) of the aqueous phase (containing estrone sulfate) was retained for enzyme hydrolysis with 2 IU (100 Ml)of arylsulfatase (Helix pomalia; Merck) and 0.75 ml of 1Macetate buffer, pH 5, in a 37C water bath. After 18 h of incubation, the free estrone was extracted twicewith diethyl ether and processed as above prior to Celite column chromatography. To determine total (free and unconjugated) urinary estrone and estradiol, the urinary samples (0.5 ml) were hydrolyzed with 2 IU arylsulfatase and 4 IU @-glucu ronidase (H. pomatia; Merck) and 0.5 ml of 1 Macetate buffer, pH 5, in a 37T water bath and processed as above.In the chromatography stage the redissolved sample was applied to the column and successiveelu tions with increasing concentrations of ethyl acetate in isooctane (0,6, 20, and 40%) were collected. The fractions containing estrone (20% ethyl acetate) and estradiol (40% ethyl acetate) were evaporated under EXEMESTANE (FCE 24304) N2 and the dried samples were used for radioimmunoassay. The final recovery rates for the procedures, monitored with [3Hjestrone, [3HJ Fig. I. Structure of exemestane. estradiol, I3Hlestrone sulfate, and [3HJestrone glucuronate are shown in Table 1. Estrone and estradiol assays were performed using the commercially entering the study, and 1 had had benign breast disease. A cohort of 3 available radioimmunoassay kits [3H]estrone (BioMerieux) and [125I]@ subjects was studied at each of 5 dose levels: 25, 50, 200, 400, and 800 estradiol (Baxter). The standard curve of the estradiol kit (supplied in . mg in open uncontrolled conditions. Doses were escalated from an serum) was substituted with a standard curve in buffer. Each purified initial 25 mg to 800 mg or until the first occurrence ofan adverse event dry sample was redissolved in the appropriate incubation buffer and an or a laboratory abnormality related to the study drug. At each dose level aliquot of this suspension was subjected to the specific RIA procedure. assessment, subjects were observed for 2 weeks (first 2 subjects) or >2 The sensitivities of the assays were approximately 2.5 and 0.5 pg/tube days (third subject) before escalation to the next dose level.All subjects for estrone and estradiol, respectively. The intraassay coefficients of were admitted to the hospital the day before drug administration for variation (n = 8) were for estrone 12.2 and 2.7% at 5 and 40 pg/tube, clinical evaluation and a 24-h (day 0) basal urine collection commenced. respectively,and the coefficients for estradiol were 7.6 and 4.8% at 0.5 . On the day ofadministration, clinical evaluation was repeated and basal and 4 pg/tube, respectively. The hormone concentrations were cor blood levels were obtained. A single dose of exemestane was adminis rected for the recovery of the labeled hormones and expressed in terms tered within 15â€”30mmafter a high lipid breakfast and blood samples of free hormones. were taken at 0.5, 1, 2, 4, 8, 24, and 48 h after drug administration. For the plasma estradiol and estrone assays, values measured during Urine samples were collected for 0â€”24h(day 1) and 24â€”48h(day 2). drug administration frequently fell below assay sensitivities, and in the Clinical evaluation and blood samples to detect aberrant values and calculation their values were considered to be the values representing possible toxicity were taken at 8, 24, and 48 h and included full blood the levelofdetectability (Table 1). In contrast, plasma levelsof estrone count, urea, electrolytes, liver function tests (bilirubin, alkaline phos sulfate and urinary levels ofestrone and estradiol were always above the phatase, transaminases, albumin, and total protein), serum lipids (cho levels of detectability. All samples from individual women were ana lesterol and triglycerides), and serum glucose. All subjects remained in lyzed in the same run of assay procedure. the hospital for 48 h. Clinical evaluation and blood samples for toxicity The potential interference of exemestane and its metabolite 17-hy were performed 1 week after discharge, i.e., 9 days after dose adminis droexemestane (@methylenandrosta-l,4-diene-17@9-ol-3-one)on the tration. At the highest dose level (800 mg), blood samples and urine radioimmunoassay of the various hormones is shown in Table 3. A collections were taken daily for 6â€”7daysafter drug administration. relevant cross-reaction was evidencedby exemestane in the androstene Study part B was performed in double-blind conditions after the dione assay (25%) and by 17-hydroexemestane in the testosterone assay results of study part A had been collected. In study part B, 14 subjects (8%). The interference by exemestane in the DHEAS assay (3%) is were recruited, 11 of whom had previously had breast cancer and 3 of considered irrelevant, taking into account the very high plasma levels of whom had had benign breast disease. There was no evidence of recur this hormone (Table 3) versus exemestane peak concentrations (<500 rent malignant disease in any of these patients. Again the same eligi ng/ml at 800 mg). bility criteria were used, and the preliminary evaluation of these pa Exemestane Assay. The levels of exemestane were determined in tients was as for part A. As maximal estrogen inhibition was observed plasma by a HPLC method with UV detection, with a detection limit of in all subjects studied in part A, 4 dose levels (0.5, 5, 12.5, and 25 mg) 10 ng/ml. Briefly plasma (1 ml) was extracted with a methylene chlo were studied in part B; a 2-weekobservation interval betweendoses was ride: isooctane (2:3, v/v) mixture, the organic phase was evaporated to considered unnecessary since no clinically relevant adverse events had dryness, and the residue was dissolved with a mixture of acetonitrile occurred at doses >25 mg. Thus all subjects were randomized in double :water (1:1) and then analyzed by reversed-phase HPLC with a detector blind conditions to 1 of 4 dose levels, with 3 or 4 subjects/level. None ofthe subjects were inpatients. On the first day ofthe trial (day â€”2),all subjects received placebo, followed by the drug administered 2 days later (day 0). Blood samples were collected daily from day â€”2upto day Table 1 Final recoveryand detectionlimit ofestrogen assays in plasma and urine samples 5 and at least twice during the second week. Urine samples were col lected daily from day â€”2upto day 5 postdosing. Clinical examination % of final recovery limit and blood sampling for toxicity were performed on the day of placebo (pg/mI)â€•PlasmaEstroneHormone (labeled hormone)Detection administration and 5 days after exemestane administration. Hormone Assays. All hormone assays were performed at the Farmi ([3Hjestrone)4Estradiol80â€”90 (I3Hlestradiol)2Estrone70â€”75 talia Carlo Erba Laboratories, Milan, Italy. Radioimmunoassays were sulfate)16UrineTotalsulfate 60â€”70((3Hjestrone used to determine plasma and urinary estrogen levels and also for determinations of plasma aldosterone, cortisol, DHEAS, l7-hydrox yprogesterone, testosterone, androstenedione, LH, and FSH. The assay 4@50b40Totalestrone procedures are summarized in Tables 1 and 2. Briefly, for plasma C20a estradiol estrone and estradiol, the sample (1.5 ml) was extracted twice with Expressedaspgoffreehormone. glucuronate.Cb [3HlEstrone sulfate and [3Hjestrone 4.5 ml diethyl ether, the ether phase was dried and dissolved in 0.5 ml totalestroneConjugated labeled estradiol not commercially available(the recoveryof isooctane before proceeding to Celite (Chromatolite A; BioMerieux) was used in the calculation). 5934

samplesHormone Table 2 Methodologyforthe determinationofvarious hormones in plasma RIA kitaSample preparation Detection limit Aldosterone RSL [3Hlaldosterone Ethyl acetate/hexane extraction (0.6 ml) 10 pg/mI 17-OH-progesterone OHPK-l25 Ether extraction (0.1 ml) 20 pg/mI Cortisol CORT-Cl@K-2 Direct assay(0.05 ml) 5 ng/ml DHEAS DHEA-S-CrK Direct assay (0.05 ml) 30 ng/ml Testosterone RSL [3Hjtestosterone Ether extraction (0.5 ml) 0.1 ng/ml Androstenedione RSL [3Hjandrostenedione Ether extraction (0.5 ml) 0.2 ng/mI LH LH-CFK Direct assay (0.1 ml) 0.25 mlU/ml FSH FSH-CTK Direct assay (0.1 ml) 0.15 mlU/ml aImmunoradiometricassaykitsforLHandFSHassay,kitsforaldosterone,testosterone,andandrostenedioneassayswereobtainedfromICNBiomedicals;allother kits obtainedfrom Sorin Biomedica(Italy).

Table3 Cross-reaction(at50%displacement)ofexemestaneanditsmetabolite1 right knee and the other who had back pain at the start of the 7-hydroexemestaneonhormonesHormone% the PJA ofvarious study complained of increasing lower back pain during the course of the study. of Cross-reaction plasmalevel (ng/ml) In trial part B, headache was again the most frequently noted in1 (range event, with 3 episodes recorded, 1 each at 12.5, 5, and 0.5 mg, postmenopausalHormone 7-Hydro- women)Estrone Exemestane exemestane of which 2 needed analgesic medication. Nausea occurred in 1 0.005â€”0.030Estradiol 0.003 0.0004 subject (at 25 mg) and abdominal discomfort occurred in 1 0.004â€”0.020Aldosterone <0.00002 0.0005 subject (at 5 mg); both cases resolved spontaneously without 0.03â€”0.1l7-Hydroxyprogesterone<0.00015 <0.00015 any medication. One subject (at 25 mg) complained of somno 0.2â€”1.0Cortisol <0.02 <0.02 60â€”240DHEAS <0.7 <0.7 lence on entry into the study which persisted unchanged 400-1400Testosterone 3 <3 throughout the study. With all the events noted in both parts of 0.2â€”1.0Androstenedione 0.08 8 the study there was no dose-dependent relationship, and a mm 25 0.15 0.5â€”3.0 imal toxic dose (i.e., a dose at which one-half of the subjects experience an adverse event) could not be defined. Physical wavelength of 247 nm. A drug-related peak having in the chromato examination did not reveal any new findings on completion of gram the same retentiontime as that of the probablemetabolite17- the study. hydroexemestane was tentatively quantified by using the response fac tor of exemestane. Laboratory Toxicity

RESULTS Hematology. No significant changes compared to pretreat ment values were observed for hemoglobin, platelets, or eryth Subject Reported Effects rocyte sedimentation rate. A decrease of WBC to <4000/mm3 (the lower limit of the normal range in our laboratory) was At preliminary evaluation all subjects had base line plasma observed in four subjects receiving 25, 25, 200, and 400 mg of estradiol and gonadotrophin levels within the expected range exemestane. The lowest count observed was 3300/mm3, and the for their age. Nine volunteers exceeded their ideal body weight decrease was transient, limited to one determination. An in by over 10%, including two who exceeded their ideal body crease in eosinophils above the upper limit of normal was oh weight by over 20%. All subjects were asymptomatic at prelim served in 4 subjects at 8 h (1 case) or 48 h (3 cases) after 200, mary evaluation with the exception of some symptoms such as 200, 400, and 800 mg of exemestane, respectively. Values had headache or nausea which did not exclude them from the study. returned to normal at subsequent measurements. Two of these In addition, one patient complained of long-standing lower four subjects had a history of allergy. back pain due to degenerative joint disease and another patient BlOOd Chemistry and Urinalysis. Throughout the study, the complained of multiple bizarre symptoms of doubtful organic variation was within Â±1.25 of normal intervals for urea, elec origin following a traffic accident 2 years previously. trolytes, liver function tests, uric acid, and cholesterol. Minor In trial part A, the dose escalation was continued to 800 mg fluctuations were observed in triglyceride values, probably re exemestane which was the predetermined maximum, without lated to the timing ofthe sampling, because most blood samples any serious adverse events. However, numerous minor symp were taken at a variable interval following meals. Similarly in toms of mild intensity were recorded by several subjects. The part A in which blood samples were taken frequently and at a most common was headache, recorded on five occasions, in two variable interval after meals, fluctuations in blood glucose levels patients at 200 mg and in one each at 25, 800, and 400 mg; the were observed. However, in part B, where samples were mea latter case was associated with dizziness. In all cases the head sured at the baseline and on the morning of the fifth day, no aches resolved without any sequelae and only one required an fluctuations were recorded. One subject had mild, recurrent algesic medication. Fatigue and malaise were reported by two hemoglobinuria no other abnormalities were observed. subjects on entry into the study and persisted unchanged throughout the study. There were 4 episodes of nausea: 3 re Endocrine Assessment current episodes of nausea of brief duration in 1 subject who had a previous history of irritable bowel syndrome and 1 epi Estrogens. In study part A, all dose levels from 25 to 800 mg sode of nausea and dyspepsia in another subject who had com produced marked suppression of plasma and urinary estrogen planed of these symptoms before entry into the study and who levels (Figs. 2â€”4).Anincrease in plasma estrogens was observed was known to have a hiatus hernia. This subjects' symptoms at 2 h after drug administration at some of the higher dose responded to medication; the other subject's symptoms re levels (200 and 400 mg). Some degree of inhibition was first solved spontaneously without treatment. Two subjects com observed at 8â€”24hafter drug administration, depending on the plained of pain; one had long-standing osteoarthritis of the drug dose and the type of hormone assayed. The degree of 5935

800 mg). A clear suppression of urinary estrogens was evident in the urinary fraction collected at 24â€”48h (day 2) after drug administration and with no correlation with drug dose in the range of 25 to 400 mg. At this time the urinary excretion of estradiol and estrone ranged between 62 and 33% and 66 and 34% of basal values, respectively (Fig. 3). At the highest dose, 800 mg, urinary estrogens reached maximal suppression 4 days after drug administration (26 and 24% of basal values for es tradiol and estrone, respectively) which still persisted 6 days after drug administration, the last sampling time (Fig. 4). In part B of the study (low-dose phase), the basal estrogen levels were calculated as the mean of two (urine) or three (plasma) baseline measurements. The 0.5-mg dose was ineffective in inhibiting plasma estro gens. At 5 mg, suppression of all plasma estrogens was evident at day 1 (to 67â€”66%of basal values), reached its maximum at day 3 (to 60â€”47%ofbasal values) and thereafter estradiol levels rose, reaching the basal values 5â€”8days after drug administra tion (Fig. 5). At 12.5 and 25 mg the inhibition was more pro nounced and its duration was somewhat longer, particularly at the highest dose. At 25 mg the maximal suppression was reached at 3 days, when plasma estradiol, estrone, and estrone sulfate were suppressed to 28, 35, and 39% of basal values, respectively. A similar degree ofsuppression persisted up to day I 7 and, with the exception of plasma estradiol, disappeared at day 8 (Fig. 5).

175 150

PLASMAESTRONESULPHATE.@ 125 F:: @5G 25

@ - o 2

Timeafterexemestane(hours') Fig. 2. Change in plasma estradiol, estrone, and estrone sulfate in postmeno pausal women receiving a single p.o. administration ofvarious doses (25,50,200, and 400 mg) of exemestane. All values (n = 3) expressed as a percentage of the basal value;bars, SE.

inhibition further increased at 48 h which was the final deter mination apart from the 800-mg dose cohort. At 48 h, the â€˜4-, 0 degree of estrogen suppression was not dose related at doses from 25 to 400 mg and ranged between 60 and 23% of basal values for estradiol, 52 and 38% for estrone, and 38 and 22% for estrone sulfate (Fig. 2). At the highest dose of 800 mg, plasma estrogen levels were further suppressed from day 2 to day 4 and then persisted up to day 7 after drug administration, 0 reaching undetectable levels for both estradiol (<2 pg/ml) and estrone (<4 pg/mI) and 10% of basal values for estrone sulfate Timeafterexemestane(days) (Fig. 4). Fig. 3. Change in daily urinary estradiol and estrone excretion in postmeno An increase in urinary estradiol excretion was noted in the pausal women receiving a single p.o. administration ofvarious doses (25, 50, 200, and 400 mg) of exemestane. All values (n = 3) expressed as a percentage of the urinary fraction collected at 0â€”24h at higher dose levels (200â€” basal value;bars, SE. 5936

I Plasma levels of androstenedione and testosterone were mea â€œV PLASMAESTROGENSsured, by RIA, in subjects treated with 400 mg exemestane. As 800 ig Exeiiestane expected from the very high cross-reaction in the RIA by exe 125 mestane [25% of androstenedione and its metabolite 17-hy C 0 ESTRONE -r@4 A ESTRADIOL droexemestane (8% of testosterone)], marked increases in an 100 0 ESTRONE drostenedione and testosterone plasma levels were observed. U) SULPHATE The maximum increase in androstenedione (from 1.9 ng/ml to @0 75 89 ng eq/mi) and testosterone (from 0.49 ng/ml to 6.3 ng eq/ml) plasma levels was observed at 2 h, in coincidence with peak 4- 0 50 plasma levels of unchanged exemestane, and these increases may be due to the interference in the RIA by exemestane and 25 17-hydroexemestane. However, since no parallelism was oh served between the standard curve of androstenedione and the cross-reaction curve of exemestane in the androstenedione RIA, any correlation between the plasma levels of unchanged -, o i 234567 exemestane, measured by HPLC, and the androstenedione plasma levels is questionable. At 4 h, the androstenedione and 150 testosterone plasma levels had declined and, at least in some @0 Exemestane URINE ESTROGENS subjects, had reached basal levels at 48 h. Definitive interpre . ESTRADIc1@tation of these data will be possible only when the methodology C @1-4 A ESTRONE .-4100

U, @0 lbI â€˜4- ESTRAOIOL75,50-25-0 @- -- 050 125- 100-Â£xisistsnsPLASMA 25 A o i 23456 Timeafterexemestane(days')

Fig. 4. Change in plasma and urinary estrogens in postmenopausal women 1 2 34 5 6 7 8 9 10 11 receivinga singlep.o. administration of 800 mg ofexemestane. Allvalues(n = 3) expressedas a percentageof the basal value;bars, SE. PLASMAESTR@C I, 125- C ;:@10E No effect on urinary estrogens was observed at 0.5 mg. A a, clear suppression of urinary estrone, but not estradiol, was cv â€˜I, ident at 5 mg, reaching 45% of basal values at days 3â€”4.A .0 dose-related decrease in both urinary estradiol and estrone was @5e- observed at doses of 12.5 and 25 mg. At the highest dose the effect peaked at days 3â€”4,with urinary estradiol and estrone reaching 23 and 20% of basal values, respectively, and was still 0- 0 1 2 3 4 5 6 7 8 9 1011 marked at day 5 (34 and 30% of basal values; Fig. 6). @ Other Hormone Levels. In study part A, plasma levels of l@ PLA9@AESTROIC @ cortisol, aldosterone, LH, FSH, . DHEAS, and 17-hydrox .....9. yprogesterone were measured only in the plasma samples of bC .., those subjects treated with 800 mg exemestane. No significant changes were observed on plasma levels of LH, FSH, DHEAS, 75, and 17-hydroxyprogesterone. There was a progressive decrease in the plasma cortisol and aldosterone levels up to the 8-h sample (to 54% of basal cortisol values and to 66% of basal aldosterone values) but these changes are compatible with the B changes observed due to the diurnal variation of plasma levels 0 1 2 3 4 5 6 7 8 9 1011 of these hormones: the basal levels were taken in the morning, Tue after exe.estane (days@l and the 8-h sample was taken in the evening. All subsequent Fig. 5. Change in plasma estradiol, estrone, and estrone sulfate in postmeno samples, which were taken in the morning, were similar to the pausal women receiving a single p.o. administration ofvarious doses (0.5, 5, 12.5, basal values (Table 4). Because no significant changes were and 25 mg)of exemestane.All values(n = 3â€”4)expressedas a percentageof the basalvalue;bars,SE.The basallevelwascalculatedas the meanofthe three values observed at the highest dose, samples taken from subjects re determined in the 3 days preceding exemestane administration. 0, 0.5 mg; , 5 ceiving lower doses in study part A were not analyzed. mg;i@,12.5mg;â€¢,25mg. 5937

â€œA,' The mean area under the curve values at 0â€”8hobtained from ukI@sTRADI0L 0 0.5 . 5mg the plasma level curve of exemestane were 566 Â±104, 907 Â± Exemestane M2.5mg 222, and 1081 Â±105 ng/hlml after 200, 400, and 800 mg, â€¢25 .g respectively. Plasma levels of the probable metabolite 17-hydroexemes tune were found to be less than one-tenth of the corresponding unchanged drug levels in all but one subject who received 800 mg. This subject had peak plasma level of 17-hydroexemestane of 296 ng/ml at 0.5 h, about one-half of the peak plasma level of the unchanged drug. There appears to be a dose-related in 50 crease in plasma exemestane levels following p.o. doses up to 400 mg. At the highest dose (800 mg) the increase is no more evident.

@@@@ V :i 4 DISCUSSION 4 @ , Jv Exemestane URINE ESTROPE 0 0.5 Exemestane administered as a single p.o. dose to postmeno a 5mg pausal healthy volunteers was found to be a well tolerated, G) 125 M2.5 mg potent, long-lasting, and specific inhibitor of estrogen biosyn â€¢25 mg C thesis. A dose-related inhibitory effect of estrogen biosynthesis .,-4 â€˜â€”4100 was demonstrated at doses of0.5â€”25mg (low-dose phase of the study). In fact, the lowest single p.o. dose at which some inhi U) bition of estrogen biosynthesis was already evident was 5 mg, @0 which reduced plasma estrogens and urinary estrone to approx 4- imately 50% of basal values 3 days after drug administration. 050 Plasma estrone and estradiol were further reduced to 41 and 25 Table 4 Effectsingle p.o. dose of800 mg ofexemestane on plasma levels 3)Time ofcortisol and androstenedione (n = plasma concentration o i 2345 Â±SEAldosterone Timeafterexemestane(days') (pg/ml)Basal(h)Mean Cortisol (ng/ml) Fig. 6. Change in daily urinary estradiol and estrone excretion in postmeno 58Â±36 pausal women receiving a single p.o. administration ofvarious doses (0.5, 5, 12.5, 2 147Â±49 45Â±14 and 25 mg) of exemestane. All values (n = 3â€”4)expressed as a percentage of the 4 143Â±24 56Â±19 basal value; bars, SE. The basal level was calculated as the mean ofthe two values 8 89Â±36 27Â±12 determined in the 2 days preceding exemestane administration. 24 166Â±29 66Â±14 48 125Â±11 59Â±22 72 138Â±22 75Â±39 96 184Â±33 56Â±22 120 184Â±17 73Â±26 to separate exemestane and 17-hydroexemestane from andros 144 159Â±26 62Â±23 tenedione and testosterone before RIA is available. 168155Â±35 149Â±59 85Â±12 In study part B, plasma levels of cortisol, aldosterone, DHEAS, FSH, LH, and 17-hydroxyprogesterone were mea sured in subjects treated with 25 mg. No significant changes -I were observed up to 5 days after drug administration, including E for cortisol and aldosterone as all samples were taken in the morning throughout the study. In view ofthese results, samples from subjects treated with lower doses of drug were not ana lyzed. In view of the artifacts and the inadequate analyses for > androstenedione and testosterone, these hormones were not â€˜-4 assayed in this part of the study. Pharmacokinetics. At all the doses tested exemestane was rapidly absorbed (Fig. 7) and peak plasma exemestane concen â€˜-4 tration levels of27 Â±6,221 Â±33, 343 Â±83,and 414 Â±77ng/ml 0. (mean values Â±SE) were reached within 2 h after administra C tion of 50-, 200-, 400-, and 800-mg doses, respectively. Plasma -lJ U) concentrations declined rapidly thereafter and fell under the detection limit of the analytical method at 24 h at the higher x doses tested, but at the 50-mg dose, nonmeasurable plasma LU levels were already found 4 h after administration. Because of Time (hours) the low levels of exemestane found in plasma after the 50-mg Fig. 7. Plasma concentrations of exemestane after single p.o. dose in post dose, the plasma levels of the subjects having received 25 mg menopausal women. Points, mean Â±SE (bars) of 3 subjects. The detection limit were not assayed. is 10 ng/ml. 5938

33% of basalvaluesby the fourth day after administration of Although more than one-half of the subjects experienced 12.5 mg, while at 25 mg these hormone levels were reduced symptoms during the study, these were mild, nonspecific, and to 35 and 28%, respectively, by day 3, with a concomitant transient with no dose relationship. No clinically relevant tox reduction of urinary elimination to 20 and 25%. Indeed 5 days icities were observed in this study. There were no changes oh after administration of the single dose of 25 mg, plasma estra served on clinical examination and apart from transient eosi diol was still reduced to 38% of basal value, plasma estrone nophilia (in three subjects) all laboratory parameters were to 40%, plasma estrone sulfate to 50%, urinary estradiol to within 1.25-fold of the normal ranges. Therefore exemestane 34%,and urinary estroneto 30%,indicating a very long dura appears to be less toxic than aminoglutethimide where the oc tion ofaction. Consequently, a 25-mg dose was considered to be currence of clinically relevant toxicities is frequent and often the minimally effective dose producing the maximal estrogen limits its use. reduction. In view of the findings of this study, exemestane warrants The duration ofthe inhibitory effect on estrogen biosynthesis further study in phase I/phase II trials in postmenopausal was best evaluated in the high-dose part ofthe study. There was women with advanced breast cancer, and these studies are cur an initial increase in both plasma and urinary estrogen levels rently in progress. with the higher drug doses (200â€”800mg), suggesting the pos sibility that the compound is aromatized to an estrogenic me REFERENCES tabolite to a small extent. This inhibitory effect was first oh 1. Henderson, I. C., and Canellos, G. P. Cancer of the breast: the past decade served at 8â€”24h in plasma and at 24â€”48h in urine. At the (Part 1). N. Engl. J. Med., 302: 17â€”30,1980. highest dose studied (800 mg), maximum suppression persisted 2. SegaloffA. [email protected] McGuire(ed.), Advancesin Researchand Treatment, ExperimentalBiology,pp. 1-22. New at 6â€”7days after administration (the last sampling time) for York: Plenum PublishingCorp., 1978. all the estrogensassayed:plasma estrone sulfate (10% of basal), 3. Kirschner, M. A. The role of hormones in the development of human breast plasma estrone and estradiol (undetectable levels; reduction cancer. In: W. L. McGuire (ed.), Breast Cancer 3: Advances in Research and Treatment, Current Topics, pp. 199â€”226.New York: Plenum Publishing to <32â€”38% of basal), urinary estrone (20% of basal), and Corp.,1979. urinary estradiol (31% of basal). These degrees of estrogen 4. Grodin, J. M., Siiteri, P. K., and McDonald, P. B. Source of estrogen pro suppression are similar or even more marked than those caused duction in postmenopausal women. J. Clin. Endocrinol. Metab., 36: 207â€” 214, 1973. by chronic administration of aminoglutethimide in postmeno 5. Santen, R. J. Potential clinical role ofnew aromatase inhibitors. Steroids, SO: pausal women. 575â€”593,1987. 6. Mendelson, C. R., and Simpson, E. R. Regulation ofestrogen biosynthesis by The long-lasting inhibitory effect of exemestane on estrogen human adipose cells in vitro.MoL Cell. Endocrinol.,52: 169â€”176,1987. synthesis may be due to the irreversible nature of its enzyme 7. Coombes, R. C, and Evans, T.R. J. Aromatase inhibitors 2. In: T. Powles inhibitory property, rather than its pharmacokinetic properties. and I. E. Smith (eds.), The Medical Management of Breast Cancer, pp. 81â€”93.London: Martin Dunitz, 1991. In fact, the measurement of plasma exemestane levels has dem 8. Goss, P. E., Jarman, M., Wilkinson, J. R., and Coombes, R. C. Metabolism onstrated that the drug is rapidly absorbed and reaches peak of the aromatase inhibitor 4-hydroxyandrostenedione in vivo. Identification levels at approximately 2 h after drug administration, after of the glucoronide as a major urinary metabolite in patients and biliary metabolite in the rat. J. Steroid Biochem., 24: 619â€”622,1986. which it rapidly declines, reaching undetectable levels (<10 ng/ 9. Harris,A. L., Powles,T. J., and Smith,I. E. Aminoglutethimideinthe ml) at 24 h, even at the doses of 200 and 400 mg. However, treatment of advanced postmenopausal breast cancer. Cancer Res., 42 (Suppl.): 3405sâ€”3408s,1982. exemestane plasma levels lower than the detection limit of the 10. Santen, R. J., Demers, L. M., Adlercreutz, H., Harvey, H., Santner, S., HPLC assay (10 ng/ml) may still be pharmacologically active, Sander, S., and Lipton, A. Inhibition of aromatase with CGS l6949A in as the K1 of exemestane for the inhibition of human placental postmenopausal women. J. Clin. Endocrinol. Metab., 65: 99â€”106,1989. 11. Demers, L. M., Melby, J. C., Wilson, T. E., Lipton, A., Harvey, H. A., and aromatase is 8 ng/ml (26 nM). The specificity of the aromatase Santen, R. J. The effects ofCGS l6949A, an aromatase inhibitor on adrenal enzyme inhibition of exemestane was shown by the absence of mineralocorticoidbiosynthesis.J. Clin. Endocrinol.Metab., 70: 1162â€”1166, 1990. any effect, even at the highest dose of 800 mg, on plasma levels 12. Giudici, D., Ornati, G., Briatico, G., Buzzetti, F., Lombardi, P., and Di Salle, of other steroid hormones. Indeed, the fact that exemestane E. 6-methylenandrost-l,4-diene-3,l7-dione (FCE 24304): a new irreversible. does not interfere with cortisol or aldosterone biosynthesis aromatase inhibitor. J. Steroid. Biochem., 30: 391â€”394,1988. 13. Di Salle, E., Briatico, G., Giudici, D., Ornati, G., and Zaccheo, T. Aromatase makes it advantageous over aminoglutethimide and fadrozole, inhibition and experimental antitumor activity of FCE 24304, MDL 18962, respectively. Moreover, exemestane is effective p.o., whereas we and SH 489. J. Steroid. Biochem., 34: 431â€”434,1989. now recommend that 4HAD is given by the i.m. route because 14. Zaccheo, T., Giudici, D., Ornati, G., Panzeri, A., and Di Salle, E. Compar ison of the effects of the irreversible aromatase inhibitor, exemestane, with it is less effective when given p.o. as a consequence of its cx atamestane and MDL 18962 in rats with DMBA-induced mammary tumors. tensive hepatic metabolism to the inactive glucuronide. Eur. J. Cancer, 27: 1145â€”1150,1991.

5939

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1992 American Association for Cancer Research. Phase I and Endocrine Study of Exemestane (FCE 24304), a New Aromatase Inhibitor, in Postmenopausal Women

T. R. Jeffry Evans, Enrico Di Salle, Giorgio Ornati, et al.

Cancer Res 1992;52:5933-5939.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/52/21/5933

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://cancerres.aacrjournals.org/content/52/21/5933. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.