[CANCER RESEARCH 49. 13064312. March 1. I989]
`
`Dose-related Endocrine Effects and Pharmacokinetics of Oral and Intramuscular
`
`4-Hydroxyandrostenedione in Postmenopausal Breast Cancer Patients‘
`
`M. Dowsett,2 D. C. Cunningham, R. C. Stein, S. Evans, L. Dehennin, A. Hedley, and R. C. Coombes
`Department ofBiochemical Endocrinology. Royal Manden Hospital, Fulham Road, London SW3 6]] IM. 0., S. 5.], and Department ofMedical Oncology, St. George’s
`Hospital, London SW17 007' (D. C. C., R. C. S., A. 11.. R. C. C.], United Kingdom, and Fondation de Recherche en Hormonolagr'e, 66-77 Boulevard Pasteur, 94260
`Fresnes, France [L D.]
`
`ABSTRACT
`
`4-Hydroxyandrostenedione (CGP32349; 4-Ol-IA) is a clinically effec-
`tive treatment for advanced postmenopausal breast cancer by both the
`parenteral and p.o. routes, as a result of its inhibition of aromatase and
`consequent suppression of plasma estrogen levels. Thirty patients were
`randomized to treatment with 250 mg 4-OHA orally once, twice, and 4
`times daily for 2 weeks and 29 of these plus a further 11 patients were
`then randomized to treatment with 250 or 500 mg i.m. every 2 weeks to
`determine the optimal dose for each route according to the suppression
`of serum estradiol levels. There was no significant difference between the
`3 oral doses in their suppression of estradiol levels indicating that the
`maximum required p.o. dose of 4-OHA is probably 250 mg daily.
`Suppression by the parenteral dose of 250 mg every 2 weeks was
`marginally suboptimal but clinical considerations of response and toler-
`ability indicate this as the optimal dose for i.m. injection. 4-Ol-IA had no
`effect on serum levels of androstenedione, testosterone, or Sa-dihydrotes-
`tosterone when given by either route but p.o. treatment with 4 doses of
`250 mg daily reduced sex hormone-binding globulin levels by a mean of
`34%. Serum levels of estrone as measured by gas chromatography-mass
`spectrometry were suppressed to approximately 40% of baseline by
`parenteral treatment. The half-life of 4-0HA p.o. was approximately 3
`h, whereas the apparent half-life of injected drug was between 5 and 10
`days after a more rapid clearance during the first 4 days after injection.
`
`INTRODUCTION
`
`Inhibition of aromatase, the enzyme complex which converts
`androgens to estrogens, was first established as an effective
`measure in the endocrine treatment of advanced breast cancer
`
`sion, whereas a single 500-mg injection led to maximal suppres-
`sion for over 14 days in 6 of 7 patients. Repeated weekly
`injections of 500 mg i.m.
`resulted in persistent estrogen
`suppression but at this and higher doses sterile abscesses and
`lumps at the injection site were noted in some patients (13). It
`was therefore important to determine to what degree the dose
`of 4—OHA could be reduced without loss of efficacy.
`Suppression of estradiol levels has been used to determine
`the optimum dose by the i.m. route (11). This approach is
`utilized in the current study to compare the effectiveness of 250
`and 500 mg 4wOHA i.m. every 2 weeks, and of 250 mg 4-OI-IA
`p.o. once, twice, or four times a day. Pharmacokinetic profiles
`were obtained for both routes of administration.
`
`We previously reported that estrone levels, as measured by
`radioimmunoassay after chromatography, were not signifi-
`cantly affected by i.m. 4-0I-IA treatment (13). To clarify this
`unexpected result, samples from parenterally treated patients
`were measured in this study by the definitive technique of
`GCMS. 4-01-1A has no effect on LH, FSH, or SHBG levels
`when given i.m. at a dose of at least 500 mg weekly (13), but
`no report has been made of the effects of p.o. 4-01-IA on these
`parameters or of the effects on androgen levels of 4-OI-IA
`administration by either route. Both of these effects have been
`documented in this study.
`
`MATERIALS AND METHODS
`
`in postmenopausal women by the use of aminoglutethimide (l—
`3). There are a number of disadvantages to the use of this drug.
`These include its inhibition of some other steroid hydroxylases
`(4, 5), which necessitates its combination with a glucocorticoid
`for maximum effectiveness (6) and possibly therapeutic safety
`(7), and its side effects such as subclinical hypothyroidism in
`some patients (8) and the frequent occurrence of a skin rash
`(9). This has led to efforts on the part of many research groups
`to derive more specific and more potent aromatase inhibitors,
`which lack significant clinical side effects.
`We have previously reported that 4-OHA3 (Ciba-Geigy, CGP
`32349) is such an inhibitor and is the first selective inhibitor to
`be studied clinically (10, 11). The compound has been charac-
`terized as a suicide inhibitor in vivo (12) and has been shown
`to be clinically effective by both the i.m. and p.o. routes (10,
`13, 14). Serum estradiol levels are consistently and markedly
`suppressed by both p.o. and parenteral 4-0HA (11). For the
`p.o. route daily administration is required to maintain suppres-
`
`Patients. All patients were either postmenopausal (at least 2 years of
`amenorrhea) or surgime ovariectomized women and all had histolog-
`ically or cytologically proven progressive metastatic breast cancer. No
`patient had received endocrine or cytotoxic chemotherapy within 4
`weeks of starting treatment. Informed consent was obtained from all
`patients, and the study was approved by the Ethics Committee at St.
`George’s Hospital. Patients were free to withdraw at any time. Disease
`response was assessed according to standard Union Internationale
`Contre le Cancer criteria (15).
`Drug. 4-OHA was provided by Ciba-Geigy Pharmaceuticals as a
`sterile microcrystalline formulation (CGP 32349) in ampuls and was
`stored at 4'C. The powder was suspended in physiological saline (125
`mg/ml) immediately prior to i.m. injection and in water or saline (50
`mg/ml) for p.o. administration.
`Oral Study. Thirty patients were randomized from random number
`tables to receive 250 mg 4-OHA p.o. once, twice, or four times daily
`for 2 weeks. Four patients were withdrawn from the study because of
`rapidly progressive disease and they were replaced by four others who
`were assigned in a random manner to the appropriate groups. The
`mean ages were 60.2 i 10.4 (SD), 63.4 :t 10.4, and 68.4 1 8.4 years
`and the mean weights were 59.3 1 6.3, 68.6 1 20.3, and 65.9 :t: 11.9
`Reeeived 6/29/88; revised 11/10/88; accepted 11/14/88.
`kg for the once, twice. and four times daily treatment groups, respec-
`The costs of publication of this article were defrayed in part by the payment
`tively, and were not statistically significantly different between the
`of page charges. This article must therefore be hereby marked advertisement in
`groups. No treatment was given for at least 7 days after the last p.o.
`accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
`dose of 4-OHA. At least one clotted blood sample was drawn from
`'Some of these data were the subject of a presentation at “Hormones and
`Cancer, 3" Hamburg, September 1987.
`each patient on the day before starting treatment and a further pretreat-
`’ To whom requests for reprints should be addressed.
`ment sample was obtained immediately prior to the first dose of 4-
`’ The abbreviations used are: 4—0HA, 4-hydroxyandrostenedione; GCMS. gas
`OHA. On-treatment blood samples were collected 5, 10, 15, 20, 30,
`chromatography-mass spectrometry; LH, luteinizing hormone; FSH, follicle-
`45, 60, and 90 min; 3, 4, 6, 8, 10, 12, and 24 h; and 2, 4, 7, and 14
`stimulating hormone; SHBG, sex hormone-binding globulin; SaDHT, Sa—dihy-
`drotestosterone; 4-OHT, Lhydmxytestosterone; 4-OHA-G, 4-OHA-glucuronide.
`days after starting treatment. A further sample was taken 7 days after
`1306
`
`Downloaded from canc-erres.aacrjournalsorg on May 1, 2017. © 1989 American Association for Cancer Research.
`
`AstraZeneca Exhibit 2153 p. 1
`InnoPharma Licensing LLC V. Astracheca AB IPR2017-00904
`Frescnius-Kabi USA LLC V. AstraZencca AB IPR2017-01910
`
`

`

`#HYDROXYANDROSTENEDIONE IN BREAST CANCER
`
`ether/dichloromethane (4/1) to remove the unconjugated 4-OHA and
`was then treated with 0.5 mg bovine liver fl-glueuronidase (type B-3;
`Sigma, Poole, United Kingdom) in 500 pl acetate buffer, pH 4.5, for
`24 h at 37°C. Experiments were conducted which established that these
`hydrolytic conditions achieved >90% of 4-OI-IA-G. After hydrolysis
`[3H]4—0HA (donated by Dr. R. Wade, Ciba-Geigy, Horsham, United
`Kingdom) was added (about 1000 cpm) as a recovery control and the
`mixture was extracted with 2 doses of 3 ml diethyl ether/dichlorometh-
`ane. Thereafter the technique was identical to that for unconjugated 4.
`OHA.
`Statistical Tests. Within- and between-dose comparisons were made
`applying analysis of variance to the logarithms of the original values.
`Two-sided tests of significance were applied. with a significance level
`of 5%. Whenever possible a repeated-measures analysis ofvariance was
`applied, with the Greenhouse—Geisser correction for within-dose com-
`parisons; otherwise, within- and between-dose comparisons were tested
`for significance, in the event of a statistically significant F test applying
`the Bonferroni correction.
`
`RESULTS
`
`Oral Treatment. The effect of p.o. 4-0HA on serum estradiol
`levels is shown in Table l for all three doses throughout the 2
`weeks of treatment. The effect of doses on estradiol levels on
`
`days 7 and 14 of treatment are compared in Fig. 1. The mean
`levels were significantly reduced within 3 h of starting treat—
`ment. After 4 days the mean levels were less than 50% of
`pretreatment levels in all three groups. During the next 10 days
`of treatment there was little variability in the degree of suppres-
`sion, but within 7 days of stopping p.o. therapy the mean
`estradiol levels returned to within 16% of pretreatment values
`in all groups. The mean pretreatment estradiol level for the
`500—mg group was over 10 pmol/liter higher than those for the
`other two groups. Although this was not statistically significant
`it is likely that this may have some effect on the levels during
`treatment. It is therefore difficult to perform meaningful com-
`parisons with this group. However, comparisons between the
`lowest and highest dose groups showed no significant difference
`in suppression at any point during treatment.
`The concentrations of 4—OHA in serum were measured for
`the first 24 h after the first dose of 4-OHA. 4—0HA was detected
`
`finishing p.o. treatment. The on-treatment blood samples at 6, 12, and
`24 h were taken immediately prior to administration of 4-0HA. Sam-
`ples were unobtainable at one or more time points in some patients, as
`indicated in the figures.
`Parenteral Study. Forty patients were randomized into groups and
`received either 250 mg (n = 22) or 500 mg (n = 18) i.m. injections of
`4-OHA every 2 weeks. The first 29 patients were the same patients that
`completed the pa. study which is detailed above. The mean ages were
`60.2 i 12.3 (SD) and 66.2 i 9.3 years and the mean weights were 66.2
`i 14.5 and 68.6 i 16.7 kg, for the 250- and 500-mg groups, respectively,
`and were not statistically significantly different between the groups.
`The time between finishing the oral study and beginning the parenteral
`study was between 7 and 14 days for all except one patient who had a
`35-day gap. Clotted blood samples were drawn on the day before and
`on the day of the first injection immediately before the injection was
`given. Samples were then taken at 1, 2, 4, 7, 10, and 14 days after the
`first injection and at weekly intervals thereafter. The samples which
`were taken on the same day as the drug was given were drawn imme-
`diately prior to injection. In a number of patients samples were unavail-
`able at some time points.
`Hormone Assays. The assays for estradiol (11), LH, FSH (16), and
`SHBG (17) were performed according to methodologies which have
`been described in detail previously. Androstenedione, testosterone, and
`SaDHT were analyzed by radioimmunoassays after ether extraction
`and thin layer chromatographic purification to avoid cross-reaction
`with the drug or its putative metabolites. The solvent system used was
`benzene/ethyl acetate/methanol, 80/20/2, and the support was silica
`gel on plastic-backed plates. Recovery was monitored and corrected for
`with tracer quantities of the appropriate tritiated steroids. The radio-
`immunoassay system for testosterone was the St. Thomas’s Hospital
`(London, United Kingdom) ether extraction, iodinated tracer kit, which
`uses an antibody raised against a 3-linlted conjugate. The androstene-
`dione antibody was raised to an androstenedione-l9—carboxymethyl-
`bovine serum albumin conjugate and was used with an iodinated ligand
`(both from Radioassay Systems Laboratories, Carson, CA). The anti-
`body to SaDHT was raised to a 3-linked conjugate (Chelsea Hospital
`for Women) and was used with a tritiated tracer. After chromatography
`there was no significant interference in any of the androgen assays
`when serum was spiked with 300 ng/ml 4-0HA or with 30 ng/ml 4-
`OHT. In this study 4-0HA levels were measured in over 700 samples.
`In two samples values were >200 ng/ml, but no values >300 ng/ml
`were found. There is little information on the metabolism of 4-OHA
`in humans, but in rhesus monkeys circulating levels of 4-OHT were
`approximately 15% of those of 4-OHA (18). These data therefore
`indicate that the assays are not subject to significant interference by
`circulating levels of 4-OHA or its potential metabolite, 4-OHT. In all
`assays the within- and between-assays coefficients of variation were less
`than 9 and 15%, respectively.
`Estrone levels were measured by GCMS basically as described pre-
`viously (19). Some minor modifications were made: extraction of 2 ml
`plasma with dichloromethane and liquid chromatography was per—
`formed on Sephadex LII-20 columns (120 x 4 mm) with hexane/
`ethanol/acetic acid (80/20/ 1) as an eluent. The first 4 ml were discarded
`and estrone was eluted in the next 4 ml. After evaporation to dryness,
`estrone was derivatized to the trimethylsilyl ether.
`Accuracy was checked by standard additions to a plasma pool (10—
`200 pg/ml range). Linear regression of added (x 1- SD) on found (y)
`estrone gave the equation
`
`in 9 of 10 patients at 5 min postdosing. For most subjects the
`subsequent serum profiles were irregular, with more than one
`peak or trough in the profile. Maximumlserum concentrations
`ranged from 23 to 235 ng/ml and were reached 0.5 to 4 h
`postdosing. The mean levels are shown in Fig. 2. Where the
`elimination phase could be determined the decline in plasma
`concentrations was adequately described by a monoexponential
`decay with a half-life of about 3 h. Seven of the patients who
`had a pharmacokinetic profile had either overt liver metastases
`or abnormal liver function. The mean level of 4-OHA at 1.5 h
`after the first dose was 65.4 i- 63.5 (SD) ng/ml in these patients
`compared with 55.6 i 40.6 ng/ml in the rest of the group (P,
`not significant).
`Serum levels of 4-OHA-G were measured in the same 24
`patients 90 min after administration of the first dose of 4»OHA.
`The mean level was 2.40 i 1.07 ug/ml and at the same time
`The interassay coefficient of variation was 10.4% at the 30—pg/ml level.
`point the mean level of 4—0HA was 55.2 i 43.6 ng/ml. The
`Samples from the same patient were analyzed in the same batch for
`mean ratio of conjugated to free 4-OHA (determined after
`all assays.
`conversion to molar equivalents) was 36.3 i 19.8 (range, 5-
`4-Hydroxyandrostenedione Assay. Levels of 4—OHA were measured
`96). The coefficients of variation of these three measurements
`with a cross-reacting androstenedione antibody after prior organic
`at 90 min were thus 79% (4-0HA), 44% (4—0HA-G), and 55%
`extraction and chromatographic purification, as previously described
`(4-OI-IA-G/4-OHA). There was no statistically significant dose
`and validated (11), with the minor modification that extraction was
`relationship between the free and conjugated forms when ex-
`performed with 2 doses of 3 ml diethyl ether/dichloromethane (4/l).
`amined by linear regression analysis (r = 0.38, 0.05 < P <
`4—0HA-G levels were measured after enzymic hydrolysis of the
`conjugate. Serum (50 ul) was extracted with 2 doses of 3 ml diethyl
`0.10).
`1307
`
`y = 1.03(¢o.o9)x + 12.4w.»
`
`Downloaded from cancerres.aacrjournals..org on May 1, 2017. © 1989 American Association for Cancer Research.
`
`AstraZeneca Exhibit 2153 p. 2
`
`

`

`4-HYDROXYANDROSTENEDIONE IN BREAST CANCER
`
`Table 1 Eject ofp.o. l—OHA treatment on serum estradiol levels (mean 1 SEM) in postmenopausal women
`Time on treatment
`
`Units
`pmol/liter
`% of pretreatment
`level
`
`pmol/liter
`% of pretreatment
`level
`
`pmol/liter
`96 of pretreatment
`level
`
`12 h
`6 h
`3 h
`Pre
`26.3 1 3.8 20.4 1 2.7 18.6 1 2.4 19.4 1 2.9
`100.0
`85.9 1 6.9 72.2 1 4.9 71.1 1 6.6
`
`1 day
`14.4 1 2.0
`56.0 1 4.3
`
`7 days
`4 days
`2 days
`12.2 11.5 11.111.9
`14.6 1 3.1
`56.0 1 12.3 48.0 1 6.4 43.7 1 4.4
`
`7 days post-
`treatment
`14 days
`11.6 11.9 22.3 1 3.5
`47.0 1 6.6 89.8 1 9.4
`
`37.6 1 7.8 29.7 1 6.4 26.1 1 7.6 28.4 1 8.2
`100.0
`84.8 1 7.4 72.1 1 9.9 78.3 1 14.7
`
`21.0 1 6.3
`69.8 1 15.0
`
`11.8 1 3.8
`41.4 1 7.9
`
`9.6 1 2.0
`10.8 1 2.0
`35.7 1 6.6 28.6 1 4.9
`
`14.0 1 3.0 27.3 1 8.7
`40.9 1 5.1 95.5 1 19.4
`
`16.8 1 2.6
`27.2 1 5.5 22.3 1 5.6 20.5 1 3.1
`100.0
`86.8 1 5.5 85.3 1 8.9 73.0 1 8.8
`
`12.4 1 2.4
`57.2 1 11.0
`
`8.3 1 1.1 22.5 1 4.8
`8.7 1 1.6 11.0 11.9
`12.0 1 2.2
`59.8 1 13.4 35.5 1 8.7 50.4 1 10.7 41.5 1 8.2 84.5 1 14.2
`
`Daily dose
`of 40114
`(rug)
`250
`
`500
`
`1000
`
`xi x2 x4
`
`xi x2 x4
`xl x2 x4
`DAILY DOSE 4—0HA (x250mg)
`
`
`
`
`xl x2 x4
`DAILY DOSE 4-0HA(x250m;)
`Fig. 1. Effect of 2 weeks of p.o. 4—0HA on mean serum estradiol levels in 30
`postmenopausal breast cancer patients. Bars, SEM. Top, estradiol levels expressed
`as pmol/liter; bottom, estradiol levels expressed as percentage of pretreatment
`levels. D, day.
`
`The effect of p.o. 4011A treatment in LH, FSH, SHBG,
`testosterone, 5aDHT, and androstenedione levels is shown in
`Table 2. There were no significant differences between the
`groups for any of the analytes before treatment. LH, testoster-
`one, SaDHT, and androstenedione levels were unaffected by
`[1.0. 4-01-1A treatment. For FSH there was a minor significant
`increase in levels (by a mean 16% above pretreatment levels, P
`< 0.05) in patients treated with 250 mg twice daily. SHBG
`levels were significantly suppressed at both the lowest and
`highest doses, by means of 9.5% (P < 0.05) and 34.0% (P <
`0.01), respectively.
`Parenteral Treatment. The mean serum levels of estradiol
`before and during treatment with i.m. 4-OHA are shown for
`each dose in Fig. 3. Levels were significantly suppressed to
`approximately 60% of baseline levels for both doses after 24 h.
`On day 7 the mean (1SEM) levels were 9.2 1 1.5 pmol/liter
`(26.5 1 3.0% of baseline) and 8.7 1 0.9 pmol/liter (42.4 1
`7.2% of baseline) for the lower and higher doses, respectively.
`Thereafter,
`levels were maintained below 50% of baseline
`throughout the next 9 weeks in both groups. There was no
`statistically significant difference in estradiol levels between the
`two doses at any point during treatment. However, the suppres-
`sion was more variable between time points with 250 mg than
`with 500 mg. For the lower dose the estradiol levels were
`significantly higher on day 14 than on day 7 and on day 28
`than on day 21. This phenomenon of recovery just prior to the
`next injection also appeared to occur prior to the fifth and sixth
`injections but this was not statistically significant.
`It can be seen from Fig. 4, in which individual patient values
`are plotted during the first 4 weeks of treatment, that the
`recovery of estradiol levels in the 250-mg group was most
`apparent in those patients with pretreatment values >35 pmol/
`liter. However, of the 6 patients treated with 250 mg 4—OHA
`every 2 weeks who showed an objective clinical response to
`therapy, 3 had pretreatment estradiol levels >35 pmol/liter.
`Only 1 of these 3 responders had on-treatment estradiol levels
`which were maintained below 20 pmol/liter. There were 5
`patients in the 500—mg group with pretreatment estradiol levels
`>35 pmol/liter, but there was little evidence of a recovery
`phenomenon even within that subgroup.
`A mathematical estimate of the quantitative importance of
`the recovery phenomenon in the patients in the 250-mg group
`with pretreatment estradiol values >35 pmol/liter is made in
`Fig. 5. The day 7 and day 21 estradiol values are taken as
`minimal on—treatment levels after the first and second injections
`and the lines connecting these points with the day 14 and day
`28 values, respectively, are taken to represent the intermediate
`estrogen levels. The nonshaded areas then indicate the degree
`1308
`
` 0
`
`
`[2
`
`IO
`
`30609034
`Minutes
`
`6
`
`8
`Hours
`Tm after Mimi oi 4-(MA
`
`Fig. 2. Mean serum levels of ALOHA afier a single 250-mg p.o. dose. Bars.
`SEM. The numbers of observations made at each time point are indicated under
`bars.
`
`Downloaded from cancerres.aacrjournals..org on May 1, 2017. © 1989 American Association for Cancer Research.
`
`Astracheca Exhibit 2153 p. 3
`
`

`

`Table 2 Efl'ecr ofp.a. l-OHA treatment on mean (155M) serum level ofLH, FSH, SHBG, testosterone (term), 5aDHT, and andmrtenedioue (AM) in postmenopausal
`women
`
`4—HYDROXYANDROSTENEDIONE 1N BREAST CANCER
`
`The numbers in parentheses indicate the number of observations. Pre- and on-treatment samples were analyzed for the same patients in all cases.
`250 mg in 1 dose
`250 mg twice daily
`250 mg 4 times daily
`Pre
`Pre
`Pre
`On
`
`Hormone
`
`Units
`
`On
`
`0n
`
`LH
`FSH
`SHBG
`Testo
`5aDHT
`A4A
`‘ P < 0.05.
`" r < 0.01.
`
`1U/liter
`1U/liter
`nmol/liter
`nmol/liter
`nmol/liter
`nmoI/liter
`
`27.8 1 3.3 (10)
`36.8 1 3.0 (10)
`81.5 1 12.2 (10)
`1.08 1 0.30 (8)
`0.40 1 0.06 (10)
`1.73 1 0.28 (10)
`
`27.9 1 4.1
`36.1 1 3.5
`74.5 1 12.1‘
`1.18 1 0.28
`0.41 1 0.10
`1.76 1 0.31
`
`33.] 1 6.5 (9)
`37.9 1 7.6 (9)
`60.0 1 9.5 (9)
`1.15 1 0.30 (6)
`0.33 1 0.07 (7)
`1.62 1 0.38 (7)
`
`35.7 1 6.3
`42.7 1 8.2‘
`55.8 1 8.7
`1.24 1 0.25
`0.31 1 0.06
`2.08 1 0.22
`
`40.2 1 10.5 (10)
`40.6 1 9.5 (10)
`77.0 1 12.4 (10)
`0.89 1 0.20 (5)
`0.40 1 0.07 (8)
`1.94 1 0.43 (8)
`
`35.0 1 8.2
`39.4 1 9.2
`46.4 1 5.5”
`1.40 1 0.24
`0.42 1 0.07
`2.99 1 0.59
`
`'02s) M
`‘1l
`
`80
`
`70
`
`60
`
`\ E
`
`3 50
`.0
`
`E g
`
`la]
`
`30
`20
`
`IO
`
`PRE
`
`|
`
`2
`
` 20
`
`ll -
`I)
`14
`D
`I!
`II
`II
`II
`1 .
`I
`u
`I:
`u
`II
`13
`I!
`I1
`
`14
`21
`O
`7
`28
`35
`42
`49
`56
`63
`70
`DAYS W TREATMENT
`
`$0”
`“to.
`
`Fig. 3. Efiect of i.m. 4—OHA every 2 weeks on mean serum estradiol levels in
`40 postmenopausal breast cancer patients. Bars, SEM. O, 250 mg; 0, 500 mg.
`Arrows, time of injection.
`
`of suppression during the second and fourth weeks as being
`84.5 and 83.7% maximal. An equivalent estimate for all patients
`in the 250—mg group gives values of 85.5 and 87.6% maximal
`suppression, respectively. The subgroup with the higher pre-
`treatment estradiol levels is thus little different from the whole
`group in this respect.
`Twenty-nine of the patients that received p.o. therapy were
`also included in the parenteral study. A comparison of the
`estradiol suppression achieved by the two routes is made in Fig.
`6. It can be seen that there was little difference in any of the six
`subgroups after the change from p.o. to i.m. treatment. The
`subgroups are too small to allow meaningful statistical analysis,
`but an overall comparison of p.o. versus parenteral treatment
`showed no statistically significant difference (P = 0.18).
`After the first injection serum levels of 4—OHA were consist-
`ently lower in the 250-mg group than in the 500-mg group
`[approximately one—half (Fig. 7)]. Peak levels occurred between
`1 and 2 days after the first injection, but by day 4 levels had
`fallen to less than one-half of the peak values. Thereafter the
`fall in semm 4-0HA levels was slower and approximately log-
`linear with an apparent half-life of between 5 and 10 days. After
`the second and subsequent injections serum drug levels were
`measured at weekly intervals and were found to show an in-
`verted V pattern with lowest
`levels just prior to the next
`injection (Fig. 8). For the 250—mg group the mean levels (1
`SEM) at the midpoint between injections became progressively
`higher (day 7, 2.4 1 0.2 ng/ml; day 63, 4.5 1 0.7 ng/ml). The
`10 patients who completed 10 weeks of treatment on 250 mg
`had similar semm 4—OHA levels during the first month of
`therapy as the whole group. There was therefore no evidence
`that a patient selection might be responsible for the progres-
`sively higher serum levels.
`
`4O
`
`3_\.
`E 30
`
`i 20
`T-1.11
`10
`
`PE
`
`|
`
`2
`
`VEEKS W TREATENT
`Fig. 4. Individual senrm estradiol levels in patients during the first 4 weeks of
`treatment with 250 mg (1:11) or 500 mg (right) 4—0HA i.m. every 2 weeks. Top
`and bottom, levels in patients with pretreatment serum estradiol levels >35 pmol/
`liter and 65 pmol/Iiter. respectively.
`
`Nine patients had either overt liver metastases or abnormal
`liver function. The mean level of 4-OHA in the five treated
`with 250 mg was 2.9 1 0.9 (SD) ng/ml after 7 days and in the
`four treated with 500 mg was 6.2 1 6.6 ng/ml. The comparable
`values of the other patients in the two groups were 2.3 1 0.9
`and 5.6 1 2.5 ng/ml, respectively. There was no significant
`difference in the levels according to the presence or absence of
`liver disease.
`
`The data were examined to determine whether there was any
`correlation on days 1, 7, 14, 21, and 28 between 4-0HA
`concentrations and the suppression of estradiol levels, when the
`latter were expressed as absolute concentrations or as percent-
`ages of pretreatment. No significant relationship was found for
`either dose on any of these days.
`Serum levels of estrone were measured by GCMS in 9 pa-
`tients before and during treatment with either 250 mg (n = 5)
`1309
`
`Downloaded from cancerres.aacrjournalsnorg on May 1, 2017. © 1989 American Association for Cancer Research.
`
`AstraZencca Exhibit 2153 p. 4
`
`

`

`4—HYDROXYANDROSTENEDIONE IN BREAST CANCER
`
`20
`
`4M(liq/ml)
`
`DAYS
`
`(N TREATMENT
`
`Fig. 7. Mean serum levels of 4-OHA after a single i.m. injection of 250 mg
`(----) or 500 mg (—) 4-0HA. Bars, SEM. The numbers of observations made
`at eact time point are indicated alongside the bars.
`
`14
`
`
`
`12345678910
`WEEKS
`01 THEATER
`
`Fig. 8. Mean serum levels of 4-OHA 7 and 14 days after repeated 2 weekly
`i.m. injections with 250 or 500 mg 4—0HA. Bars, SEM. The numbers of obser-
`vations made at eact time point are indicated alongside the bars.
`
`125
`
`75
`
`Estrone
`pool/L
`
`.\
`
`PRE
`
`0N
`
`Fig. 9. Effect of mm i.m. every 2 weeks on serum estrone as measured by
`GCMS. O, 250 mg; 0, 500 mg. I test, P= 0.001.
`
`Table 3 Efl‘ect ofi.m. l—OHA treatment on mean (1SEM) serum level: of
`testosterone. ondrosteuedione (AM), and SaDHT (wool/liter)
`On-treatment levels were measured on day 21 of therapy. The numbers in
`parentheses indicate the number of observations.
`250 mg
`
`500 mg
`
`Hormone
`
`Pre
`
`On
`
`Pre
`
`0n
`
`1.09 1 0.17
`Testosterone 0.78 1 0.09 (10) 0.75 1 0.07 1.35 1 0.29 (9)
`5aDHT
`0.38 1 0.06 (10) 0.38 1 0.04 0.41 1 0.05 (9) 0.41 1 0.07
`A4A
`1.67 1 0.26 (10)
`1.97 1 0.40 1.87 1 0.35 (9)
`1.59 1 0.27
`
`DISCUSSION
`
`as a result of the low toxicity of the medical regimens now
`available. Aromatase
`inhibition with
`aminoglutethimide
`achieves a similar profile in terms of clinical efiicacy to the
`antiestrogen tamoxifen (20, 21), which has become the bench-
`mark for new endocrine treatments, but aminoglutethimide has
`Endocrine treatment is the favored first line therapy for breast
`significant side effects (8, 9) and requires concurrent glucocor-
`ticoid administration for maximal effectiveness (6). Our earlier
`cancer in tumors identified as estrogen receptor positive, largely
`1310
`
`Downloaded from cancerresr.aacrjournals..org on May 1, 2017. © 1989 American Association for Cancer Research.
`
`Astrachcca Exhibit 2153 p. 5
`
`WW W
`
`mum
`we"
`
`”MIL
`
`O
`
`7
`
`D
`DAYS
`
`2|
`14
`(N TREATIENT
`
`28
`
`§'
`
`§ 3
`
`Fig. 5. Estimation of mean quantitative importance of recovery phenomenon
`in patients treated with 250 mg 4-OHA i.m. every 2 weeks and pretreatment
`estradiol levels >35 pmol/liter. The points at days 7 and 21 are taken as being
`the mean minimal on-treatment estradiol level after the first and second injection.
`I, degree by which suppression is submaximal during the second week after
`injection, quantified as a percentage of the oblong: enclosed by the dashed liner.
`
`__25°M~'m m
`
`m4
`
`at
`
`I
`
`o.
`
`ms
`
`20
`I
`
`Pu pn i.m.
`
`PM no. Lin
`
`n-5
`
`n-5
`
`250M xgfl
`
`Pro an on
`
`m4
`
`I
`
`.
`
`Pie in on
`
`n-5
`
`250mga4.p.o.
`
`——
`
`3
`
`d E
`
`-
`g l
`
`d E
`
`‘ an
`
`1% I0
`
`.1
`
`E-
`
`g.
`
`Pro pa in.
`Pro [to [in
`Fig. 6. Comparison of the efi‘ect of p.o. and parenteral 4—01-1A treatment on
`mean senim estradiol levels. Bars, SEM. Top, middle. and bottom. patients who
`received 250 mg in 1 dose, 2 doses or 4 doses p.o. daily, respectively. Left and
`right, those who transferred to 250 or 500 mg i.m. every 2 weeks, respectively.
`
`or 500 mg 4-OHA every 2 weeks. On-treatmcnt levels were
`made after at least 21 days of therapy (range, 21-36 days).
`Levels were suppressed in all patients, the mean fall being from
`82.3 1 10.3 [SEM] to 33.0 1 9.2 pmol/liter, i.e., to a mean
`40.3 1 7.0% of baseline [P = 0.001 (Fig. 9)]. The serum
`estradiol levels in these same patients were suppressed to a
`mean 29.4% of baseline at the same time points. There was no
`statistically significant difference in any of the androgen levels
`between samples taken prior to treatment and those taken on
`day 21 of i.m. therapy with either dose of 4-01-1A (Table 3).
`
`

`

`#HYDROXYANDROSTENEDIONE IN BREAST CANCER
`
`reports (10, l l, 13, 14) indicate that 4-OHA is a potent, specific
`aromatase inhibitor with few significant clinical side effects
`when given by either the p.o. or i.m. route. In this report we
`have continued to use the suppression of serum estradiol levels
`as a marker of the effectiveness of aromatase inhibition with 4-
`
`by p.o. but not parenteral treatment is probably due to the high
`levels of drug presented by p.o. treatment to the liver, the organ
`of SHBG synthesis.
`The levels of androstenedione, testosterone, and SaDHT
`were not significantly affected by either p.o. or parenteral
`therapy. It therefore seems unlikely that the effect of high doses
`of the drug on l7B—hydroxysteroid dehydrogenase and 5a-
`reductase which have been noted in vitro (23, 25) will be of any
`importance to the pharmacological activity of 4-OHA at ther—
`apeutic doses in postmenopausal breast cancer patients.
`Comparison of 250 mg with 500 mg 4-0HA given i.m. every
`2 weeks indicates that the lower dose may be slightly suboptimal
`in terms of estradiol suppression since there is an increase in
`mean estradiol levels prior to the next injection which was
`statistically significant during the first month of treatment but
`also apparent after the fourth and fifth injections. It was also
`clear, however, that this recovery phenomenon was most ap-
`parent in those patients with the higher pretreatment levels of
`estradiol (>35 pmol/liter). It may therefore have been consid-
`ered that these patients might benefit from higher dose treat-
`ment but the observations that 3 of the 6 responders to treat—
`ment with 250 mg every 2 weeks were in the group with
`pretreatment levels >35 pmol/liter and that only 1 of these 3
`responders had estradiol levels which-were suppressed consist-
`ently below 20 pmol/liter throw doubt on the significance of
`the recovery of estradiol levels.
`The estimates of the quantitative importance of recovery in
`the groups with pretreatment levels of estradiol above and below
`35 pmol/liter may provide an explanation for the apparent lack
`of importance of the suboptimal estradiol levels. Firstly, the
`estimates indicate that the suppression is suboptimal by only
`about 16% during the second weeks after injection and over the
`2-week injection period this will amount to about 8%. Indeed,
`this may be a high estimate since the recovery may follow a
`concave exponential function rather than a straight line. Sec-
`ondly, the proportional recovery is similar between the group
`with higher estradiol levels and the group as a whole. In a Phase
`11 study of 250 mg i.m. every 2 weeks the response rate was
`identical (33%) to that in a previous study of higher dose (500
`mg weekly) treatment. The incidence of pain and sterile abscess
`formation at the injection side was, however, substantially
`reduced at the lower dose.‘ In the light of the clinical efficacy
`of the lower dose of 4-OHA, the lower incidence of side effects
`and the marginal nature of the submaximal suppression of
`estradiol, 250 mg i.m. every 2 weeks would appear to be the
`optimal parenteral dose.
`The long half-life of 4~OHA after i.m. administration indi-
`cates that a depot of drug is formed. The release of the com-
`pound from the depot appears to be more rapid during the first
`few days than after 4 days; the levels fall by about 50% between
`days 2 and 4 but thereafter the apparent half-life is approxi-
`mately 5—10 days. The much lower levels of drug during the
`few days prior to the next injection u