`& 2004 Cancer Research UK All rights reserved 0007 – 0920/04 $30.00
`
`www.bjcancer.com
`
`Hormonal impact of the 17a-hydroxylase/C17,20-lyase inhibitor
`abiraterone acetate (CB7630) in patients with prostate cancer
`
`Clinical
`
`A O’Donnell1,2, I Judson*,1,2, M Dowsett3, F Raynaud2, D Dearnaley1,8, M Mason4, S Harland5, A Robbins6,
`G Halbert7, B Nutley2 and M Jarman2
`1Royal Marsden NHS Trust, Sutton, Surrey SM2 5PT, UK; 2CR UK Centre for Cancer Therapeutics, Institute of Cancer Research, Sutton, Surrey SM2 5NG,
`UK; 3Academic Department of Biochemistry, Royal Marsden Hospital, Fulham Road, London SW3 6JJ, UK; 4Department of Clinical Oncology, Velindre
`Hospital, Whitchurch, Cardiff CF4 7XL, UK; 5Department of Oncology, University College of London, The Middlesex Hospital Mortimer St, London W1N
`8AA, UK; 6Drug Development Off ice, Cancer Research UK, PO Box 123, London WC2A 3PX, UK; 7Cancer Research UK Formulation Unit, University of
`Strathclyde, Glasgow G1 1XW, UK; 8Institute of Cancer Research, Sutton, Surrey SM2 5NG, UK
`
`A series of three dose escalating studies were conducted to investigate the ability of the 17a-hydroxylase/C17,20-lyase inhibitor
`abiraterone acetate, to cause maximum suppression of testosterone synthesis when delivered to castrate and noncastrate males with
`prostate cancer. Study A was a single dose study in castrate males. Study B was a single dose study in noncastrate males and study C
`was a multiple dose study in noncastrate males. The drug was given orally in a once-daily dose and blood samples taken to assess
`pharmacokinetic (PK) parameters and hormone levels in all patients. The study drug was well tolerated with some variability in PKs.
` 1 was seen in four out of six castrate males treated with a single dose of 500 mg.
`Suppression of testosterone levels to o0.14 nmol l
`At 800 mg given days 1 – 12 in noncastrate males, target suppression was achieved in three out of three patients, but a two- to three-
`fold increase of Luteinising Hormone (LH) levels in two out of three patients overcame suppression within 3 days. All patients in the
`multiple dose study developed an abnormal response to a short Synacthen test by day 11, although baseline cortisol levels remained
`normal. This is the first report of the use of a specific 17a-hydroxylase/17,20-lyase inhibitor in humans. Repeated treatment of men
`with intact gonadal function with abiraterone acetate at a dose of 800 mg can successfully suppress testosterone levels to the castrate
`range. However, this level of suppression may not be sustained in all patients due to compensatory hypersecretion of LH. The
`enhanced testosterone suppression achieved in castrate men merits further clinical study as a second-line hormonal treatment for
`prostate cancer. Adrenocortical suppression may necessitate concomitant administration of replacement glucocorticoid.
`British Journal of Cancer (2004) 90, 2317 – 2325. doi:10.1038/sj.bjc.6601879 www.bjcancer.com
`Published online 18 May 2004
`& 2004 Cancer Research UK
`
`Keywords: prostate cancer; hormonal therapy; pharmacokinetics; clinical study
`
`
`
`
`
`Prostate cancer continues to present an enormous challenge in the
`UK, where it is the second most common cause of cancer death in
`men, causing over 9000 deaths per year (Cancer Research UK,
`2002).
`The beneficial effect of androgen ablation on metastatic prostate
`cancer was realised in the 1940s, when Huggins and Hodges
`observed an antitumour response, as measured by a reduction in
`serum acid phosphatase, in patients treated by surgical or medical
`castration (Denmeade and Isaacs, 2002). In general, androgen
`deprivation will give a response of varying duration in 80 – 90% of
`men with advanced disease (Denis and Murphy, 1993). Although
`the multistep pathway of androgen production resulting in
`testosterone (Figure 1) is present only in its entirety within the
`testes, the adrenal gland is also capable of releasing DHEA and
`androstenedione. Enzymes for the subsequent conversion of
`DHEA and androstenedione to testosterone are present
`in a
`
`*Correspondence: Professor I Judson, Clinical Pharmacology, Institute of
`Cancer Research, E-Block, 15 Cotswold Road, Sutton, Surrey SM2 5NG,
`UK; E-Mail: Ian.Judson@icr.ac.uk
`Received 27 November 2003; revised 18 March 2004; accepted 22
`March 2004; published online 18 May 2004
`
`variety of peripheral tissues as well as the prostate (Griffin and
`Wilson, 1998).
`sources of
`these extratesticular
`Studies have shown that
`testosterone represent an important alternative source of androgen
`stimulation in a significant proportion of patients with prostate
`cancer. As much as 10% of baseline circulating testosterone
`remains in castrated men, due to peripheral conversion of adrenal
`steroids to testosterone (Hellerstedt and Pienta, 2002). First-line
`treatment of prostate cancer by androgen deprivation is generally
`achieved by medical or surgical testicular castration. This leaves
`the testosterone derived from adrenal sources intact.
`It is recognised that the development of androgen-independent
`prostate cancer is caused, in part, by changes in androgen receptor
`regulation, activation of the androgen receptor being mitogenic in
`this malignancy. The sensitivity of
`the androgen receptor is
`increased by the overexpression of
`two nuclear coactivators:
`transcriptional intermediary factor 2 and steroid receptor coacti-
`vator 1. Transactivation of the androgen receptor is thus enhanced
`at lower concentrations of testosterone (Gregory et al, 2001a, b).
`Furthermore, amplification of the number of androgen receptors
`can be shown in hormone refractory tumours, when compared
`to both benign prostatic hyperplasia and primary prostatic
`
`MYLAN PHARMS. INC. EXHIBIT 1003 PAGE 1
`
`
`
`Abiraterone acetate – hormonal and PK study
`A O’Donnell et al
`
`Cholesterol
`
`Desmolase
`
`Pregnenolone
`
`17α OH-pregnenolone
`
`DHEA
`
`17αα hydroxylase
`
`c17−20 lyase
`
`Progesterone
`
`17α OH-progesterone
`
`Androstenedione
`
`11β hydroxylase
`
`21 hydroxylase
`
`Oestrone
`
`Testosterone
`
`5αred
`
`Corticosterone
`
`Cortisol
`
`Oestradiol
`
`DHT
`
`18 hydroxylase
`
`Aldosterone
`Figure 1 Steroid synthesis pathway. DHEA ¼ dehydroepiandrosterone; 5ared¼ 5areductase; DHT¼ dihydrotestosterone.
`
`malignancy, using both reverse transcription PCR and fluorescence
`in situ hybridisation (Bubendorf et al, 1999; Linja et al, 2001).
`The imidazole derivative ketoconazole and the aromatase
`inhibitor aminoglutethimide have been evaluated as possible
`agents with which to achieve decreased production of adrenal
`steroids. Ketoconazole is relatively unselective, inhibiting both
`cholesterol side chain cleavage and 11b-hydroxylation (Loose et al,
`1983). A direct antitumour effect of ketoconazole in vitro has also
`been demonstrated (Eichenberger and Trachtenberg, 1988). The
`action of aminoglutethimide is primarily to block the formation of
`pregnenolone from cholesterol but it is also recognised to inhibit
`11b-hydroxylase and peripheral aromatase (Schwimmer and
`Parker, 1996). In clinical trials, both agents have shown some
`activity as second-line agents (measured by clinical benefit as well
`as reduction in PSA), supporting the concept of a more selective
`inhibitor of the 17a-hydroxylase/C17,20-lyase enzyme (Oh, 2002).
`The novel 17a-hydroxylase/C17,20-lyase inhibitor abiraterone
`acetate (Figure 2) was developed as a mechanism-based steroidal
`inhibitor following observations that nonsteroidal 3-pyridyl esters
`had improved selectivity for inhibition (Rowlands et al, 1995).
`Abiraterone acetate is the 3-acetate and a prodrug form of CB7598
`(17-(3-pyridyl)androsta-5,16-dien-3b-ol, abiraterone), a potent
`inhibitor of the enzyme with a Kiapp of 0.5 nM (Barrie et al, 1997;
`Potter et al, 1995).
`Using a rodent model, following intraperitoneal administration,
`abiraterone acetate showed rapid deacetylation. Levels of deace-
`tylated drug reached 41 mM at 6 h, with persistence of relatively
`high levels of compound (approximately 0.3 mM) for at least 24 h.
`Although this finding may be indicative of depot characteristics of
`the mode of administration used, it may also indicate a degree of
`enterohepatic recirculation that could prove favourable in the
`clinical setting, providing sustained target enzyme inhibition.
`In these preclinical studies, there was correlative evidence of
`inhibition of 17a hydroxylation as shown, for example, by the
`reduced weight of the ventral prostate (Barrie et al, 1994; Dowsett
`et al, 1988).
`Here, we describe a series of three phase one trials in which
`abiraterone acetate was tested in humans for the first time. This is
`the first report of the effects of a specific 17a-hydroxylase/C17,20-
`
`N
`
`RO
`Figure 2 Abiraterone structure. R¼ H: abiraterone (CB7598); R¼ Ac:
`abiraterone acetate (CB7630).
`
`lyase inhibitor in humans. The studies were conducted to
`determine the dose of abiraterone acetate that will result in
`maximum suppression of
`testosterone and to obtain safety,
`pharmacokinetic (PK) and endocrine data, the latter to determine
`the specificity of inhibition. All three studies involved patients with
`advanced, that is, unresectable, prostate cancer.
`
`METHODS
`
`Patients were recruited to one of the following three studies that
`were conducted in sequence. The study protocols were approved
`by the Research Ethics Committee of all participating institutions,
`and all patients gave written informed consent prior to inclusion.
`All three studies were conducted under the auspices of the Cancer
`Research UK Phase I/II Committee.
`
`Study A
`
`This was a single dose study in males with castrate levels of
` 1) following orchidectomy or
`testosterone (testosterone p2 nmol l
`Gonadotrophin-Releasing Hormone agonist (GnRHa) therapy, to
`
`British Journal of Cancer (2004) 90(12), 2317 – 2325
`
`& 2004 Cancer Research UK
`
`2318
`
`Clinical
`
`MYLAN PHARMS. INC. EXHIBIT 1003 PAGE 2
`
`
`
`Clinical
`
`Abiraterone acetate – hormonal and PK study
`A O’Donnell et al
`
`to study entry all patients must have had an orchidectomy or have
`received (and continued to receive) ongoing treatment with a
`GnRH agonist for at least 2 months. A confirmatory testosterone
` 1 was also required.
`level of between 0.2 and 2.0 nmol l
`
`2319
`
`Study B
`
`This was a single-centre, open-label, phase one, single-dose study
`in which sequential cohorts of three noncastrate patients were to
`be treated at four dose levels: 200, 500, 650 and 800 mg. The
`starting dose of 200 mg was chosen when the results of study A
`were available. It was envisaged that a dose level would be
`expanded to five patients if
` one patient at any dose level experienced XGrade 3 toxicity
` there was a significant suppression of serum cortisol in one
`patient as defined as a greater than 50% reduction in levels of
`cortisol, or a smaller fall associated with hypotension (systolic
`BP less than 90 mmHg) or persistent electrolyte disturbance.
` if the target suppression of testosterone was achieved in three
`patients at any one dose level.
`
`All patients were required to have a normal testosterone level
` 1) as well as normal
`prior to study entry (i.e. X9.0 nmol l
` 1).
`gonadotrophin levels (Luteinising Hormone (LH) p13 IU l
`
`Study C
`
`This was a three-centre, open-label, phase one, multidose study in
`which sequential cohorts of three noncastrate patients were to
`receive treatment with abiraterone daily for 12 days. All patients
` 1)
`were required to have a normal testosterone level (X9.0 nmol l
` 1) prior to
`as well as normal gonadotrophin levels (LHp13 IU l
`study entry.
`The starting dose of 500 mg was based upon the data from the
`single-dose studies. It was planned that a dose level would be
`expanded to six patients if any patient experienced toxicity X
`Grade III or if there was a significant suppression of serum cortisol
`in one patient (defined as per Study B above). All patients were
`followed for 28 days for any sign of toxicity. If any patient
`developed symptoms suggestive of progressive disease, confirmed
`by a rise in PSA during the study period, they would have been
`offered standard treatment with a GnRH agonist.
`In all studies, the capsules were administered in one oral dose at
`0930 following an overnight fast. Free fluids were permitted and
`patients were allowed a light snack 4 h after dosing on the day of
`PK sampling.
`
`PRETREATMENT ASSESSMENT AND FOLLOW-UP
`INVESTIGATIONS
`
`Prior to the first dose of therapy a complete history, physical
`examination and assessment of performance status was performed
`on all patients. Full-blood count, electrolytes and creatinine, liver
`function, urinanalysis, electrocardiograph and chest X-ray were
`obtained from all patients. The ECG was repeated 6 h after dosing.
`On the day of therapy heart rate, blood pressure and temperature
`were recorded every 4 h and then daily at the time of blood sampling.
`Thereafter,
`full-blood count, electrolytes, creatinine and liver
`function were re-evaluated on Days 2 and 7. Toxicity was recorded
`using the NCI-CTG Expanded Common Toxicity Criteria V1.
`
`ENDOCRINE ASSESSMENT
`
`Study A and B
`
`Serum samples for endocrine analysis were obtained at 0, 2, 4, 8
`and 24 h on a single day in the week prior to treatment
`
`determine the dose of abiraterone acetate that was sufficient to
`cause suppression of testosterone synthesis to undetectable levels
` 1). Significant suppression was defined as either
`(o0.14 nmol l
` 1 in individual patients with a pretreat-
`testosterone o0.14 nmol l
` 1 or a X75% reduction in individual
`ment value of o0.6 nmol l
` 1.
`patients with a pretreatment value of X0.6 nmol l
`
`Study B
`
`This was a single dose study in noncastrate males (testosterone
` 1) to determine the dose of abiraterone acetate
`level X9.0 nmol l
`that was sufficient to cause suppression of testosterone synthesis to
` 1).
`castrate levels (p2.0 nmol l
`
`Study C
`
`This was a multidose study (Days 1 – 12) in noncastrate males to
`determine the dose of abiraterone acetate that was sufficient to
`cause persistent suppression of testosterone synthesis to castrate
`levels. If this level of suppression was achieved, then it was planned
`to escalate the doses still higher to establish whether further
`suppression of testosterone was possible. Complete suppression of
`testosterone
`synthesis
`should result
`in testosterone
`levels
` 1, based upon local data, indicating that the mean
`o0.7 nmol l
`value for testosterone in patients on GnRH agonist therapy is in
` 1.
`the region of 0.7 nmol l
`In all three studies, the secondary objectives were as follows:
` to determine the safety and tolerability of abiraterone acetate in
`the single and multiple dose setting;
` to study the PKs of this compound; and
` to determine any other endocrine effects especially suppression
`of cortisol synthesis.
`
`INVESTIGATIONAL AGENT
`
`Abiraterone acetate was provided by Boehringer Ingelheim as a
`micronised powder and prepared by the Cancer Research UK
`Formulation Unit (Glasgow) as 10, 50, 100 and 200 mg dry-filled
`capsules. The capsules were stored at room temperature.
`
`PATIENT POPULATION
`
`Patients were required to be at least 18 years of age with a WHO
`performance status of p2. No radiotherapy or hormonal therapy
`(with the exception of GnRHa in Study A as described above) was
`allowed within 6 weeks prior to study, although patients were
`allowed to receive concomitant bisphosphonates. Entry was further
` 1, WBC
`restricted to patients with haemoglobin X10.0 g dl
`
` 1, platelet count X100 109 l 1, alkaline phosphatase
`X4.0 109 l
`less than twice the upper limit of normal and a urea, creatinine and
`bilirubin of not more than 25% above the normal range. Patients
`were excluded with coexistent serious nonmalignant disease and
`were not allowed to take concomitant steroids. All patients had
`stable recurrent malignancy. At the time of participation in these
`trials, none of
`the patients was considered to require any
`alternative therapeutic intervention for symptomatic or progres-
`sive disease.
`
`DOSAGE AND ADMINISTRATION
`
`Study A
`
`This was a single-centre, open-label, phase one, single-dose study
`in which sequential cohorts of
`three medically or surgically
`castrate patients were to receive treatment at five dose levels:
`starting at 10 mg and increasing to 30, 100, 200 and 500 mg. Prior
`
`& 2004 Cancer Research UK
`
`British Journal of Cancer (2004) 90(12), 2317 – 2325
`
`MYLAN PHARMS. INC. EXHIBIT 1003 PAGE 3
`
`
`
`also drawn prior to treatment on Days 2, 3, 4, 7, 8, 9, 10, 11 and 14
`with additional samples on Days 21 and 28. Samples (7 ml) were
`collected into vacutainer
`tubes
`(BD, Rutherford, NJ, USA)
`containing EDTA and immediately centrifuged to separate the
`plasma. At least 2 ml of plasma was transferred into polypropylene
`tubes (NUNC) and frozen at – 201 until analysis.
`
`RESULTS
`
`Study A
`
`A total of 16 male patients with histologically confirmed advanced
`adenocarcinoma of the prostate were enrolled. All patients had
`received previous antiandrogen therapy (flutamide or cyproterone
`acetate) and at time of enrolment in this study all were receiving
`treatment with a GnRH agonist;
`leuprorelin or goserelin. All
`patients were evaluable for safety, PK and endocrine assessments.
`The group had a median age of 73.5 years (Range 63 – 77 years) and
`all were performance status 0, 1.
`
`Endocrine
`Sequential cohorts of three patients were treated at 10,
`30 and 100 mg. At these doses no consistent effect on testosterone
`was observed and the plasma concentrations of abiraterone were
`below the level of detection. Patient 4 (30 mg dose level) was
`observed to have noncastrate levels of testosterone during the
`study period despite a satisfactory screening testosterone level of
` 1. On further questioning, it was discovered that there
`1.3 nmol l
`had been suboptimal compliance with goserelin therapy and he
`was deemed ineligible.
`A dose escalation to 500 mg was considered necessary as a result
`of the absence of a pharmacodynamic effect at doses up to 100 mg
`after one patient had already consented to and had received therapy
`at 200 mg. A 75% reduction in testosterone was observed in this
`patient within the first 24 h after treatment with abiraterone. In all
`three patients treated at 500 mg, a reduction in testosterone to the
` 1 or X75% reduction in
`target level was seen (o0.14 nmol l
` 1). The duration of the
`baseline level testosterone X0.6 mmol l
`suppression was variable. In two of the three patients, suppression
`was sustained from Days 2 to 5 post-therapy. Three additional
`patients then received treatment at 500 mg. Target testosterone
`suppression was seen in one of these patients. The same level of
`suppression was not observed in the remaining two apparently due
`to incorrect prior dosing with goserelin and therefore escape of
`testosterone levels to noncastrate levels during the study period.
`These results are illustrated in Figure 3. At the 500 mg dose level, a
`reduction in mean androstenedione levels was parallel to the
`reduction in mean testosterone levels, occurring around Day 2. It
`was notable that there was no corresponding reduction in the levels
`of 17-hydroxyprogesterone. A reduction in serum cortisol levels
` 1,
`was seen in one patient treated at 500 mg, baseline 409 nmol l
` 1 Day 1. However, as this reduction was
`falling to 81 nmol l
`apparent at the first time point on Day 1 it was felt to be
`inconsistent with suppression due to abiraterone. On questioning
`this patient denied the concomitant use of glucocorticoids.
`
`Study B
`
`Four male patients with histologically confirmed advanced
`adenocarcinoma of prostate were recruited. All patients had
`received prior antiandrogen therapy and previous therapy with a
`GnRH agonist but at
`the time of study entry had a serum
` 1. All patients were evaluable for safety,
`testosterone of 49 nmol l
`PK and endocrine assessments. The group had a median age of
`71.5 years (range 60 – 77 years), and had performance status 0.
`
`Endocrine The first patient received treatment with abiraterone
`at 200 mg. No testosterone suppression was observed and three
`further patients were then treated at 500 mg. In all three patients a
`
`Abiraterone acetate – hormonal and PK study
`A O’Donnell et al
`
`commencement and then on Days 1, 2, 3, 4 and 7. As the duration
`of testosterone suppression was longer than originally anticipated,
`additional samples were added on Days 10, 14 and 21. The serum
`samples were analysed for testosterone, cortisol, 17 a-hydroxy-
`progesterone (17HP), androstenedione, LH and follicle-stimulating
`hormone (FSH) using commercially available kits. However, the
`DPC Coat-a-count kit for testosterone was sensitised using a larger
`volume of sample/standard and extension of the standard curve.
`Prior to study initiation,
`this was demonstrated to have no
`significant effect on the values of testosterone measured but to
` 1. All endocrine
`provide sensitivity to a level of 0.05 nmol l
`analyses were conducted by radioimmunoassay except for LH
`and FSH, which were by enzyme immunoassay.
`
`Study C
`
`The schedule sampling differed slightly with samples removed at
`0930 and 1730 on the day of treatment and thereafter in the
`morning on Days 2, 3, 4, 7, 8, 9, 10, 11, 14, 21 and 28. In Study C, a
`short Synacthen test was also performed prior to therapy and again
`around Day 11.
`
`2320
`
`Clinical
`
`PKS – ANALYTIC METHOD, ASSAYS AND SAMPLING
`
`Sample extracts were analysed by a fully validated liquid
`chromatography mass spectrometry method. The instrument
`consisted of Wisp Model 717 autosampler including a Model
`600MS system controller with a quaternary U6K LC pump. A
`Finnigan MAT TSQ 700 triple quadrupole mass spectrometer was
`used as the detection system, together with Finnigan MAT ICIS
`and ICL software for data capture and processing (ThermoQuest
`Ltd, San Jose, CA, USA). The separation of analytes was performed
`on a Supelcosil LC-ABZ (5 mm, 250 4.6 mm) analytical column
`protected by a guard column (Supelco, Bellefonte, PA, USA). The
`mobile phase consisted of 570 ml of 20 mM ammonium acetate
`solution, 100 ml tetrahydrofuran and 1330 ml acetonitrile and was
` 1 throughout the system.
`delivered at a flow rate of 1 ml min
`Column eluant was subjected to electrospray ionisation and
`monitored by selected ion monitoring (SIM) of protonated
`pseudo-molecular ions of authentic standards of abiraterone
`acetate and abiraterone, and GP488 (an analogue of abiraterone
`used as internal standard). For SIM, the scan width was 0.25 and
`the
`total
`scan time was
`2.99 s. Also, heated capillary
`temperature¼ 2501C,
`voltage¼ 4.5 kV,
`spray
`collision
`off-
`set¼ 49.9 V and electron multiplier voltage¼ 1200 eV.
`Samples were extracted as follows: 50 ml of acetonitrile and 40 ml
`of 50 mM GP488 internal standard were added to 500 ml of patient
`plasma. After the addition of 3 ml hexane : butanol (98 : 2, v/v) and
`2 min vortexing, 2 ml aliquots of the organic layer were transferred
`for drying in vacuo for 2 h. The dried residue was reconstituted by
`vortexing in 150 ml of acetonitrile and transferred into autosampler
`vials. Aliquots (100 ml) of these samples were injected onto the LC
`column. Calibration curves were obtained by plotting peak area
`ratios for abiraterone acetates or abiraterone to internal standard
`vs the nominal analyte concentrations using linear regression by
`Microsoft Excel version 5.0 (Microsoft, Redmond, WA, USA).
`Calibrations curves were produced at the levels of 500 and 1000 nM
`for abiraterone acetate and 6.25, 12.5, 25, 50, 100 and 500 nM for
`abiraterone. Quality controls were included at the level of 8, 40 and
`400 nM for abiraterone and 500 nM for abiraterone acetate.
`A comprehensive PK profile (Cmax, Tmax, T1/2a T1/2 and Kabs) was
`determined for each patient. Pharmacokinetic parameters were
`evaluated using WinNonLin Softwares and were conducted at
`The Institute of Cancer Research (Sutton).
`Blood was sampled for analysis of abiraterone concentration
`prior to drug administration and at the following times on Day 1:
`30 min, 1, 2, 4, 6, 8 and 12 h. In the multidose study, samples were
`
`British Journal of Cancer (2004) 90(12), 2317 – 2325
`
`& 2004 Cancer Research UK
`
`MYLAN PHARMS. INC. EXHIBIT 1003 PAGE 4
`
`
`
`Abiraterone acetate – hormonal and PK study
`A O’Donnell et al
`
`Testosterone levels
`
`2321
`
`200 mg
`500 mg − pt.1
`500 mg − pt.2
`500 mg − pt.3
`
`Clinical
`
`200 mg
`500 mg − pt.1
`500 mg − pt.2
`500 mg − pt.3
`
`0
`
`3
`
`15 18 21
`12
`69
`Days postdose
`
`Luteinising hormone levels
`
`40
`
`30
`
`20
`
`10
`
`0
`
`−1)
`
`Testosterone (nmol L
`
`25
`
`20
`
`15
`
`10
`
`5
`
`0
`
`−1)
`
`Luteinising hormone (IU L
`
`0
`
`2
`
`4
`
`6
`8 10 12 14 16 18 20
`Days postdose
`
`Figure 4 Serial hormone levels in noncastrate patients treated with a
`single oral dose of abiraterone acetate.
`
`While serum cortisol levels remained within normal limits all three
`patients had an abnormal response to Synacthen by Day 11. The
`mean change
`in cortisol
`in response
`to Synacthen was
` 1 (i.e. þ 77%) at baseline in the patients treated with
`294.3 nmol l
` 1 (þ 10%) by Day 11. A further
`500 mg, falling to only 42 nmol l
`cohort of three patients was then treated at 800 mg to investigate
` 1) could be
`whether target testosterone suppression (p0.7 nmol l
`reached. In the first patient, target suppression was obtained on
`Day 1, sustained for 3 days and then reversed in association with
`rising LH (three-fold increase)
`from Day 3. Despite this
` 1 for the duration of
`testosterone levels remained p2.0 nmol l
`treatment. In the second, target suppression was reached on Day 4,
` 1 on Day 7 but otherwise remained
`testosterone rose to 0.77 nmol l
`below the target level for the duration of treatment. In the final
` 1 by Day 2 but then rose
`patient, testosterone fell to 1.7 nmol l
` 1 from Day 4. A concomitant two-fold rise in
`again to 42.0 nmol l
`LH was seen from Day 3 in this patient.
`The first and third patients treated at 500 mg had higher LH
`levels at baseline than all patients treated at 800 mg. This may have
`contributed to the difficulty in achieving suppression of testoster-
`one at the lower dose level. As in those treated at 500 mg, the
`cortisol response to the short Synacthen test in all three patients
`treated at 800 mg was abnormal on Day 11. The mean change in
` 1 (120%) at
`cortisol levels in response to Synacthen was 385 nmol l
`baseline in this cohort of patients, falling to an increment of
` 1
`(23%) by Day 11. Serum cortisol
`levels were
`65.3 nmol l
`themselves reduced by the evening of Day 1 in three patients but
`all other assessments remained within normal
`limits. Evening
`cortisol falling by 60, 71 and 69%, respectively, from baseline
`evening cortisol in these three patients.
`
`PHARMACOKINETICS
`
`The PK parameters all show considerable variability between
`patients and are presented in Table 1. The plasma concentration of
`abiraterone at the first three dose levels in study A was below the
`level of detection of the assay precluding analysis of the PK
`
`10 mg
`
`30 mg
`
`100 mg
`200 mg
`
`500 mg
`
`10 mg
`
`30 mg
`
`100 mg
`200 mg
`
`500 mg
`
`10 mg
`30 mg
`100 mg
`200 mg
`500 mg
`
`Mean testosterone levels
`
`0
`
`1
`
`5
`4
`3
`2
`Days postdose
`
`6
`
`7
`
`Mean androstenedione levels
`
`0
`
`1
`
`5
`4
`3
`2
`Days postdose
`
`6
`
`7
`
`Mean 17- hydroxyprogesterone levels
`
`0
`
`1
`
`5
`4
`3
`2
`Days postdose
`
`6
`
`7
`
`1234
`
`0
`
`012345678
`
`3.5
`
`3
`
`2.5
`
`2
`
`1.5
`
`1
`
`0.5
`
`0
`
`−1)
`
`Testosterone(nmol L
`
`Androstenedione
`
`−1)
`
`(nmol L
`
`17-hydroxyprogesterone
`
`−1)
`
`(nmol L
`
`Figure 3 Selected mean hormone levels in castrate men receiving a
`single dose of abiraterone acetate.
`
`reduction in testosterone level of more than 50% from baseline was
`seen. The testosterone nadir was observed on the second day after
`therapy with recovery to pretreatment levels 6 – 9 days later. A
`corresponding rise in LH levels was seen (47 – 75%) maximal on
`Day 3 with recovery to pretreatment levels by Day 10 (Figure 4).
`No change in cortisol level was seen.
`
`Study C
`
`Six male patients with histologically confirmed advanced adeno-
`carcinoma of the prostate were accrued. Five of the six had
`received prior antiandrogen therapy and the same five of six had
`received and completed prior therapy with a GnRH agonist. At the
`time of study entry, all patients had a testosterone level of
` 1. The group had a median age of 68.5 years (Range
`49 nmol l
`62 – 80 years) and had performance status 0 or 1.
`
`Endocrine An initial cohort of three patients received treatment
`at 500 mg. Although a reduction in testosterone level
`to
` 1 was seen in all three patients, this did not reach
`p2.0 nmol l
` 1. The pattern of suppression was
`the target level of p0.7 nmol l
`variable with maximal suppression occurring Days 1 – 3 and
`substantial suppression sustained for up to 9 days (Figure 5).
`
`& 2004 Cancer Research UK
`
`British Journal of Cancer (2004) 90(12), 2317 – 2325
`
`MYLAN PHARMS. INC. EXHIBIT 1003 PAGE 5
`
`
`
`Abiraterone acetate – hormonal and PK study
`A O’Donnell et al
`
`Testosterone levels
`
`Pt.1 − 500 mg
`Pt.2 − 500 mg
`Pt.3 − 500 mg
`Pt.1 − 800 mg
`Pt.2 − 800 mg
`Pt.3 − 800 mg
`
`Pt.1 − 500 mg
`Pt.2 − 500 mg
`Pt.3 − 500 mg
`Pt.1 − 800 mg
`Pt.2 − 800 mg
`Pt.3 − 800 mg
`
`1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
`Days postdose
`
`Lutenising hormone levels
`
`1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
`Days postdose
`
`30
`
`25
`
`20
`
`15
`
`10
`
`5
`
`0
`
`45
`
`40
`
`35
`
`30
`
`25
`
`20
`
`15
`
`10
`
`05
`
`−1)
`
`Testosterone (nmol L
`
`−1)
`
`Lutenising hormone level (IU L
`
`2322
`
`Clinical
`
`Figure 5 Serial hormone levels in noncastrate patients treated with multiple oral doses of abiraterone acetate.
`
`behaviour. At 100 mg, concentrations were low and the terminal
`half-life was unable to be determined confidently. Detectable levels
`were obtained at all doses greater than or equal to 200 mg. Data are
`not available for one patient treated at 500 mg due to an assay that
`failed quality control.
`The mean Tmax was 2.70 h (7s.d. 2.71) with a mean elimination
`half-life of 27.6 h (7s.d. 20.17). A range of up to 10-fold in AUC
`was seen for a given dose.
`Within the patient groups studied, we were unable to identify
`any distinguishing characteristics to explain this further. The level
`of interpatient variability made analysis of dose-dependent PK
`relationships difficult. Combining the data from all three studies,
`while the mean AUC at each dose level increased with dose, it
`appeared that the association between AUC and dose is nonlinear
`(R2¼ 0.34). There was no evidence of
`saturation of drug
`absorption at the dose levels studied (Table 2 and Figure 6).
`
`OVERALL TOXICITY
`
`In all three trials, abiraterone acetate was very well tolerated and
`no serious adverse events attributable to treatment were recorded.
`No haematologic or biochemical effects were observed at any dose
`level or schedule evaluated. No alteration in resting heart rate or
`blood pressure was seen. Systemic effects attributable to abirater-
`one were uncommon. Headache, hot flushes, a mild increase in
`abdominal and testicular pain, and a transient depression in mood
`
`(all grade II) were reported by individual patients but no relation
`to dose or schedule was apparent. There were no grade three or
`four toxic events.
`
`DISCUSSION
`
`These are the first data to describe the systematic assessment of the
`endocrine effects of a specific 17a-hydroxylase/C17,20-lyase in-
`hibitor in humans. The endocrine results that will determine the
`further development of abiraterone acetate are likely to be
`qualitatively representative of other drugs of this class.
`The single dose study in castrate patients demonstrated that
`treatment with abiraterone acetate results in sustained suppression
`of the testosterone/androstenedione axis. The protracted duration
`of this suppression is possibly due to the irreversible nature of the
`drug action. In turn, therefore, one may predict that it may be
`possible to increase the effect with continuous dosing. This single
`dose study showed no effect on 17a-OH-progesterone production.
`This indicates that any inhibition of 17a-hydroxylation that may
`occur as a result of treatment with abiraterone acetate is over-
`ridden by compensatory mechanisms related to cortisol feedback.
`Despite 17a-hydroxylase and C17,20-lyase activities being contained
`in a single enzyme the compensated effect on 17a-hydroxylase
`activity clearly did not prevent an inhibition of C17,20-lyase (as
`evidenced by androgen suppression). Supportive evidence for this
`is provided by the observation that there was no significant effect
`
`British Journal of Cancer (2004) 90(12), 2317 – 2325
`
`& 2004 Cancer Research UK
`
`MYLAN PHARMS. INC. EXHIBIT 1003 PAGE 6
`
`
`
`Table 1 Summary of pharmacokinetic data for abiraterone acetate when given orally in single and multiple dose studies
` 2)
` 1)
`
`Patient
`
`Dose (mg m
`
`AUC (lM h
`
`Cmax (lM)
`
`Abiraterone acetate – hormonal and PK study
`A O’Donnell et al
`
`Tmax (h)
`
`T1/2a (h)
`
`2323
`
`T1/2b (h)
`
`Kabs (h)
`
`Clinical
`
`1
`10
`ND
`ND
`ND
`ND
`ND
`ND
`2
`10
`ND
`0.001
`2
`ND
`ND
`ND
`3
`10
`ND
`0.018
`11
`ND
`ND
`ND
`4
`30
`ND
`ND
`ND
`