CENTER FOR DRUG EVALUATION AND RESEARCH
`
`APPROVAL PACKAGE FOR:
`
`APPLICATION NUMBER
`
`21-344
`
`Clinical Pharmacology and Biopharmaceutics
`Review
`
`

`

`CLINICAL PHARMACOLOGY AND BIOPHARMACEUTICS NDA
`
`REVIEW
`
`NDA number, type: NDA 21—344, 1S
`
`Brand name: FASLODEX® Injection
`
`Generic name: fulvestrant
`
`Type of dosage form and strength(s): pre-filled syringes, 2.5 ml (125 mg) and 5 ml
`
`(250 mg)
`
`lndication(s): FASLODEX is indicated for the
`
`D K R 11"
`
`Applicant name: AstraZeneca Pharmaceuticals, LP
`
`Submission (letter date):
`
`initial
`
`(March 28, 2001)
`
`BB
`
`(November 13, 2001)
`
`C
`
`(December 3 l, 200])
`
`BB
`
`(January 30, 2002)
`
`OCPB and ORM Division names: Division of Pharmaceutical Evaluation 1 and
`
`Division of Oncologic Drug Products
`
`OCPB Reviewer(s) and Team Leader names: Gene Williams, Ph.D. and N.A.M.
`
`Rahman, PhD.
`
`Type of Submission: New Drug Application (NME)
`
`I.
`
`Executive Summary
`
`A. Recommendations
`
`The Clinical Pharmacology and Biopharmaceutics portion of this NDA is acceptable. No
`new risk management recommendations have resulted from this review.
`
`B. Phase 4 commitments
`
`A single Phase 4 commitments is recommended. We recommend that the Applicant
`perform a study of the effect of ketoconazole on fulvestrant pharmacokinetics. For ease,
`to allow for fewer patients (the IV route has less inter-individual variability than the 1M
`route) and to increase safety during performance of the study, we recommend that this
`study be conducted using the intravenous formulation of fulvestrant.
`
`

`

`II.
`
`Table of Contents
`
`I.
`
`Executive Summary
`
`A. Recommendations
`
`B. Phase 4 Commitments
`
`II.
`
`Table of Contents
`
`Glossary
`
`III.
`
`IV.
`
`Summary of Clinical Pharmacology and Biopharmaceutics
`Findings
`
`Question Based Review
`
`?~ General Attributes
`
`pa General Clinical Pharmacology
`
`TUNED
`
`Intrinsic Factors
`
`Extrinsic Factors
`
`General Biopharmaceutics
`
`Analytical Section
`
`Detailed Labeling Recommendations
`
`VI.
`
`Appendices
`
`A. Proposed Package Insert
`
`B.
`
`Individual Study Review (Applicant’s Study Synopses)
`
`C. Consult Review (including Pharmacometric Reviews)
`
`D. Cover Sheet and OCPB Filing/Review Form
`
`E. Applicant ’5 Individual Analytical Methods Summaries
`
`14
`
`I9
`
`24
`
`25
`
`28
`
`39
`
`58
`
`NA
`
`136
`
`139
`
`

`

`Glossary
`
`l4C — radioactive carbon (molecular weight = 14)
`AUC — area under the concentration versus time curve
`Cl -- clearance
`
`Cmax —— maximum concentration
`
`Cmin — minimum concetration: the concentration just prior to the next dose
`Ctrough - minimum concetration: the concentration just prior to the next dose
`CV -— coefficient of variation
`
`CYP — cytrochrome P450
`ER — estrogen receptor
`FSH — follicle stimulating hormone
`Gmean —— geometric mean
`HDL - high-density lipoprotein
`IM — intra-muscular
`in vitro — not in humans or animals
`
`in viva — in humans or animals
`IV — intra-venous
`
`kg — kilogram
`L — liter
`
`LA — long-acting
`LDL - low-density lipoprotein
`LH - lutenizing hormone
`mg -— milligram
`min - minute(s)
`ml — milliliter
`
`NDA -— New Drug Application
`NME — new molecular entity (a molecule not previously approved as a human drug)
`OCPB — Office of Clinical Pharmacology and Biopharrnaceutics
`P450 — cytochrome P450
`PD — pharmcodynamic(s); a measure of drug effect
`PgR — progesterone receptor
`Phase 4 — the post-approval stage of drug development
`PK — pharmacokinetic(s)
`PK/PD —- relating concentration (PK) to effect (PD)
`POSTHOC — an analysis option within the NONMEM software program
`Q — once every
`SA — short-acting
`Tmax —- time at which maximum concentration (Cmax) is reached or was measured
`V —- volume of distribution
`
`Vd — volume of distribution of the central compartment
`VLDL — very-low-density lipoprotein
`Vss — steady-state volume of distribution
`uM — micromolar or micromoles
`
`

`

`111.
`
`Summary of Clinical Pharmacology and Biopharmaceutics Findings
`
`In vivo and in vitro data support the following conclusions:
`
`In clinical use, drug exposure is controlled by the properties of the LA IM injection
`
`the ratio of Cmax to Ctrough for a 5 ml 1M injection and a 28-day
`inter—dose interval is approximately 2.5.
`
`On a Q 28-day regimen, levels approach approximate steady-state after
`3 doses.
`
`the pharmacokinetics of fulvestrant 250 mg were shown to be similar
`when administered as either a single 5-ml or as two 2.5-ml injections.
`
`no clear relationship has been established between efficacy
`measurements (time to progression, objective response) and
`pharmacokinetic parameters such as Cmax, Cmin, AUC, and
`clearance.
`
`The general pharmacokinetics are:
`
`fulvestrant is rapidly distributed following administration by IV
`infusion, with plasma concentrations decreasing rapidly in a
`multiexponential fashion. Estimates of mean terminal elimination
`half-lives range from approximately 14.0 to 18.5 hours.
`
`fiilvestrant is rapidly cleared (>10 ml/min/kg) and renal elimination is
`low (i.e. <1%).
`
`fulvestrant is extensively metabolized.
`
`No meaningful differences in the pharmacokinetics are apparent between male and
`either pre- or postmenopausal female subjects following administration of a single IV
`dose, nor between male and postmenopausal female subjects following IM
`administration (irrespective of age).
`
`Fulvestrant has been shown to be highly bound (99%) to plasma proteins
`(predominantly lipoproteins) and to have a large steady-state volume of distribution
`(approximately 3 to 5 L/kg), which suggests that the distribution of the compound is
`largely extravascular.
`
`Preclinical studies with human cytochrome P450 isoenzymes and results from clinical
`pharmacokinetic trials involving the co-administration of fulvestrant with midazolam
`or rifampin suggest that
`
`

`

`-
`
`-
`
`therapeutic doses of fileestrant have no inhibitory effects on
`cytochrome P450 enzymes
`
`the clinical pharmacokinetics of fulvestrant are unlikely to be affected
`by cytochrome P450 inducers.
`
`0 There was no apparent effect caused by renal insufficiency or mild hepatic
`impairment on the pharrnacokinetics of fulvestrant. Although data is not available,
`clearance may be reduced in patients with moderate or severe hepatic impairment.
`
`0 No differences were seen in fulvestrant clearance among Black, Hispanic, native
`Japanese, or White subjects.
`
`0 The l7-keto and sulfone metabolites of fulvestrant found in human plasma, and
`formed in the rat and the dog (but not in the plasma in these species), show no
`estrogenic activity. Only the l7-keto compound demonstrates a level of antiestrogenic
`activity of the same order of magnitude as fulvestrant and its activity is 45-fold lower
`than that of the parent compound.
`
`0 A variety of pharmacodynamic endpoints were studied. Generally, the results support
`that fulvestrant is an estrogen receptor antagonist that acts primarily peripherally.
`
`1“
`
`The Table below lists the studies that were present in the NDA. Except for the
`phannacokinetic assessments performed in the initial in-patient studies, all of these
`studies contributed to this NDA review.
`
`

`

`
`
`
`absent_
`
`
`____
`
`_-
`
`---_
`
`_ -
`
`:
`
`—-'-__
`__
`—_
`_—-_—
`
` —__-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`’U "U
`
`‘11
`
`-—_
`—-
`
`
`
`
`
`
`
`
`

`

`Question-Based Review
`IV.
`A. General Attributes
`
`1. What are the highlights of the chemistry and physical-chemical properties of the
`drug substance, and the formulation of the drug product?
`
`Fulvestrant is 7-alpha-[9-(4,4,5,5,S-pentafluoropentylsulphinyl) nony1]estra—l,3,5-(10)-
`triene-3, 1 7-beta-dio].
`
`Fulvestrant is a white powder with a molecular weight of x— The solution for
`injection is a clear, colorless to yellow, viscous liquid. Each injection contains as inactive
`ingredients: alcohol, benzyl alcohol, benzyl benzoate as co-solvents, and castor oil as a
`co—solvent and release rate modifier.
`
`What are the proposed mechanism of drug action and therapeutic indications?
`
`Fulvestrant is an antiestrogenic agent that binds estrogen receptor (ER) in a competitive
`manner. Preclinical studies show that fulvestrant is a reversible inhibitor of the growth of
`estrogen-sensitive human breast cancer cells and of tamoxifen-resistant breast cancer
`cells in vitro. In studies with immature female rats, fulvestrant blocks the uterotrophic
`action of estradiol and the estrogenic (partial agonist) effect of tamoxifen.
`
`FASLODEX is indicated for the
`
`il—u
`
`’—
`
`What is the proposed dosage and route of administration?
`
`FASLODEX is supplied in sterile single patient pre-filled syringes containing 50-mg/
`m] fulvestrant either as a single 5 ml or two concurrent 2.5 m] injections to
`deliver the required monthly dose. FASLODEX is administered as an intramuscular
`injection of 250 mg once monthly.
`
`2. What efficacy and safety information (e.g., biomarkers, surrogate endpoints, and
`clinical endpoints) contribute to the assessment of clinical pharmacology and
`biopharmaceutics study data (e.g., if disparate efficacy measurements or adverse
`event reports can be attributed to intrinsic or extrinsic factors that alter drug
`exposure/response relationships in patients)?
`
`No independent variables (covariates) have been identified by either the Applicant or the
`Clinical Division that appear to explain efficacy or safety.
`
`B. General Clinical Pharmacology
`1. What is the basis for selecting the response endpoints, i.e., clinical or surrogate
`endpoints, or biomarkers (also called pharmacodynamics, PD) and how are they
`measured in clinical pharmacology and clinical studies?
`
`

`

`The primary efficacy variable upon which approval is based is time to response rate as
`quantified by radiographic imaging. This response is selected based upon the expectation
`that it will predict clinical benefit. Other measured responses were chosen based upon
`fulvestrant’s mechanism of action (the drug is an anti-estrogen and estrogen, or lack of it,
`results in tumor, reproductive tract, endocrine and bone responses).
`
`The clinical pharmacodynamic program comprised included data on the effects of
`fulvestrant on tumor markers (Trials 0002 and 0018), the female reproductive tract
`(including endometn'al growth) (Trials 0003, 0019, and 0036), endocrinology
`(Trials 0003, 0004, 0019, 0020, and 002]), and bone resorption (Trial 0019). The
`endpoints are identified below. None of these endpoints form the basis of approval.
`
`Category
`
`Tumor marker
`
`
`
`Measurement “ Contributionto
`
`PD portion of
`insert
`
`
`
`
`
`
`
`
`Ki67 labeling index Ki67 labeling
`and/or the apoptotic decreased. Al not
`index (Al) in breast
`changed
`tumor tissue
`
`in breast tumor
`tissue
`
`decreased
`
`
`
`
`
`
`
`
`
`
`
`
`
`levelsoftheprotein p82appears -decreased but
`
`statistics not
`oefionned
`
`
`
`
`Female reproductive change in
`reduction in change
`tract -- Endometrium endometrial
`of thickness
`
`
`
` thickness (drug- or
`
`
`disease- induced
`
`
`no effect
`Female reproductive ovarian volume
`
`tract —
`.
`and/or the presence
`
`Hypothalamic-
`of ovarian follicles
`
`
`pituitary-ovarian axis after fulvestrant
`
`
`dosing to
`
`premenopausal
`
`
`women.
`
`
`
`no apparent effect
`Female reproductive 1. Karyopyknotic
`
`tract - Vaginal
`Index, 2. Maturation but statistics not
`
`cytology
`Value: both measure performed
`
`estrogenization of
`the va- inal tract
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Female reproductive uterine fibroid
`tract -- Fibroid
`volume
`volume
`
`no effect
`
`

`

`Endocrinology
`
`
`
`
`
`Contribution to
`Category
`Measurement
`PD portion of
`
`
`insert
`
` statistics not
`
`
`levels of estradlol,
`
`progesterone, FSH,
`performed
`
`
`
`LH, Sex horrnone-
`
`
`binding globulin
`
`SHBG
`
`
`
`
`Bone resorption in
`
`assays of cross-
`not powered for
`
`
`
`premenopausal
`linked N-telopeptide
`non-inferiority to
`
`
`women
`and free
`
`placebo, but does
`
`
`
`deoxypyn'dlnoline
`show less resorption
`
`
`
`than with coserelln
`
`
`2. Are the active moieties in the plasma (or other biological fluid) appropriately
`identified and measured to assess pharmacokinetic parameters and exposure
`response relationships? (if yes, refer to IV. F, Analytical Section; if no, describe the
`reasons)
`
`Fulvestrant exists as a mixture of 2 diastereoisomers which are epimeric at the sulfur
`atom in the side chain. In finalizing the drug substance specification, these 2
`diastereoisomers were named fulvestrant sulfoxide A and fulvestrant sulfoxide B. These
`
`isomers are present in the ratio (AzB) of approximately 45:55.
`
`Preclinical studies have shown no difference between the pharmacokinetic profiles of
`sulfoxide A and sulfoxide B, and the 2 diastereoisomers were shown to be equally
`pharmacologically potent in preclinical in vitro models. It was not therefore anticipated
`that there would be any differences in the pharmacokinetic profiles in man, and the main
`bioanalytical method measured the diastereoisomers as a mixture for pharmacokinetic
`analysis. However, to validate this approach, a specific chiral method was used to
`monitor the concentrations of the 2 diastereoisomers in 3 clinical trials (Trials 0018, 002]
`and 0026). The results of Trial 18 are most relevant as the route of administration (IM)
`and formulation (clinical) was that which is to-be-marketed.
`
`After 1M administration of 50, 125, and 250 mg LA fulvestrant in Trial 0018, the data for
`the 28 days following the injection from 3 patients at each dose level indicated that the 2
`diastereoisomers were present in similar proportions at all time points, i.e., with
`concentration ratios of approximately 50:50. These data suggest that the disposition
`processes following IV and 1M administration are achiral and support the use of total
`fulvestrant measurements for pharmacokinetic analysis.
`
`

`

`In vitro protein binding measurements, using fresh plasma samples from human, obtained
`with the equilibrium dialysis technique and using [”C]-fiilvestrant, provided
`limited information because of the very low aqueous solubility of fulvestrant. However,
`as at very high concentrations (10 mg/ml) fulvestrant was highly bound (at least 99%) to
`plasma proteins.
`
`The ex vivo binding distribution of fulvestrant and its metabolites was determined in
`human plasma obtained at 12, 24, and 96 hours after a single IM dose of [MC]-fiilvestrant
`(samples collected from Trial 0029). O“? was used to separate plasma
`albumin, alpha-acid glycoprotein, high-density lipoprotein (HDL), low-density
`lipoprotein (LDL), very-low—density lipoprotein (VLDL), and chylomicrons. Results
`showed that lipoproteins appeared to be major binding components in human plasma:
`LDL 29%, VLDL 27%, HDL 17% at 12 hours. Given the much larger proportion of HDL
`in plasma, this indicated that HDL was of lesser importance in the binding of firlvestrant
`or its metabolites. The role of sex hormone-binding globulin in fulvestrant binding, if
`any, could not be determined because of the extreme instability of reference material in
`the test system.
`
`No studies were conducted on drug—drug competitive protein binding interactions. No
`mutual displacement interactions appear to have been reported for other hydrophobic
`drugs binding to lipoproteins, such as cyclosporin and amphotericin B (Wasan and
`Cassidy 1998).
`
`Following 1V administration of l4C—fulvestrant, 51% (n = 8, range = 37 — 64%) of the
`AUC of plasma 14C was accounted for by parent drug. Restated, 49% of the circulating
`radioactivity was in moieties other than fulvestrant. The identity of this radioactivity was
`not determined. Chromatograms of feces showed more than a dozen peaks with no single
`component contributing more than approximately 10% of the total area under the sum of
`the chromatographic areas.
`
`Since less than 1% of the dosed radioactivity had been recovered in the urine after 7 days
`of collection, metabolite profiling was not carried out on urine samples.
`
`In Trial 0036, plasma levels of the 17-ketone and sulfone metabolites were determined in
`3 volunteers after single injections of LA 1M fulvestrant 125 or 250 mg. The results
`showed that plasma concentrations of both metabolites were low in comparison with the
`parent drug; all samples analyzed were close to or below the LOQ bung/ml and 1 ng/ml
`for the l7-ketone and sulfone metabolites, respectively).
`‘
`Similarly, in Trial 0021 afier single and multiple injections of LA IM fulvestrant, the
`plasma concentrations of the 17-ketone and sulfone metabolites of fulvestrant were low
`and the majority of the samples analyzed fell below the LOQ. Low concentrations of the
`l7-ketone (typically 2 to 3 ng/ml) were found in some samples after multiple dosing.
`These results are consistent with the single-dose information generated in Trial 0036 and
`confirm that the circulating levels of these metabolites do not appear to alter after
`repeated administration of fulvestmnt.
`
`lO
`
`

`

`Metabolites resulting from conversions at the 3- and l7-positions of the steroid nucleus to
`form ketone(s) or sulfate have been synthesized and tested for estrogenic and
`antiestrogenic activity. None showed any estrogenic activity and only the 17-keto
`compound demonstrated a level of antiestrogenic activity of the same order of magnitude
`as firlvestrant: its activity was 45-fold less than that of the parent compound. Based on
`these data, the Applicant concluded that the metabolites of the steroidal part of the
`fiilvestrant molecule are unlikely to contribute in a significant manner to drug activity.
`
`The effect of metabolism of the 17 B-hydroxy to 17-keto, conjugation of the 3 B- and 17
`B-hydroxy groups and oxidation of the side chain sulphoxide to sulphone, was tested in
`the immature rat uterotropic/antiuterotropic assay. The putative sulphone metabolite had
`no estrogenic activity but had antiestrogenic activity comparable with that of the parent
`drug. None of the putative metabolites at the steroid 3- and 17-positions had any
`estrogenic and only the 17-keto compound demonstrated a level of antiestrogenic activity
`of the same order of magnitude as ICl 182,780, approximately 45-fold less than that of
`the parent drug. The Applicant concluded that oxidation of the side-chain sulphoxide to
`sulphone could contribute to drug activity, and that the metabolites of the steroidal part of
`the ICI 182,780 molecule are unlikely to contribute in a significant manner to drug
`activity.
`
`3. What are the characteristics of the exposure-response relationships (dose-
`response, concentration-response) for efficacy and safety?
`a) based on PK parameters, what is the degree of linearity or nonlinearity in the
`dose-concentration relationship?
`
`In studies using different dose levels, maximum plasma concentrations (Cmax) and
`exposure (AUC(0-28 d)) appeared to increase in a dose-related manner. Formal statistical
`analyses (ANCOVA) were performed on the AUC summary data from Trial 0018 to
`assess dose proportionality. This result of this analysis of covariance (ANCOVA)
`indicated that, across the dose range of 50 — 250 mg, exposure was approximately
`proportional to dose, i.e., the proportionality coefficient was not statistically different
`from 1 (at the 5% level).
`
`APPLICANT’S TABLES
`
`Table 8 Guru-uric m- ((T\"/.) JAIN)”. d, values at different doses «ILA im
`fulvmtrnl in prune-opus! patients (Trial mm)
`box (my
`AUL'tmn d:
`
`(ugd'ml)
`SU (Ia-153
`32.90133)
`turn-16)
`69.! (51.0)
`
`230 (rt-20) 116 F2511)
`
`Table!
`
`Dose [impartial-By “Ibis mam: mm mm
`
`PV‘dlm.‘
`95% (‘l
`hqwrtnunahly “(Ix-mt
`Para-mic
`male
`
`
`we“, a,
`urn-33
`03335.1.0281
`u. l r 12
`
`

`

`b) do PK parameters change with time following chronic dosing?
`
`Based upon trough plasma concentrations (see Figure below), accumulation occurs with
`Q 28 days dosing, but accumulation-independent PK parameters (clearance, volume of
`distribution) appear not to change.
`
`APPLICANT’S FIGURE
`
`Funk 9 W nun (SD) M “win-I aluminu- (Trial)!!! and (’10)
`
`.
`g
`2
`
`:
`s
`E
`i
`
`a
`E
`
`i
`
`.
`t
`g
`i
`:-
`i
`
`'
`
`5—:—
`
`-
`.4:
`.f
`‘
`’35-
`:31
`z
`2?.
`
`2
`
`5""
`a
`I‘llllfll milihlhma 7"!!!
`
`I
`i.
`l.
`i2
`.1-
`
`3'
`
`g
`..
`45:?
`:«g
`5
`z;
`i?
`i’f
`i":
`
`-g!
`
`3”
`
`c) how long is the time to the onset and offset of the pharmacological response or
`clinical endpoint?
`
`Presumably onset of estrogen receptor binding occurs very shortly after drug
`administration. The primary clinical efficacy endpoint is response rate (tumor shrinkage)
`— not a time—related measure.
`
`d) are the dose and dosing regimen consistent with the known relationship between
`dose-concentration-response, and are there any unresolved dosing or
`administration issues?
`
`The 250 mg dose appears appropriate. Higher doses cannot be used due to volume
`restrictions (the current dose is a 5 ml 1M injection). Lower doses are unlikely to be
`sufficiently efficacious as the 250 mg dose failed superiority analyses and is being
`
`12
`
`

`

`approved on the basis of non-inferiority. Further, a 125 mg dose was studied early in
`Phase 3 but was eliminated because efficacy results were not sufficiently promising.
`
`4. How does the PK of the drug and its major active metabolites in healthy
`volunteers compare to that in patients?
`a) what are the basic PK parameters?
`
`Following IV administration of 10 mg to healthy postmenopausal women (tn'al 0038,
`n=6) the following values were obtained (mean (standard deviation»:
`
`Cl (ml/min/k)
`half-life (h) Vss (L/k)
`“MB.
`
`No study directly compared pharmacokinetics between healthy postmenopausal women
`and postmenopausal breast cancer patients. However, across study comparisons can be
`made between studies using the to-be-marketed long-acting intramuscular formulation in
`healthy post-menopausal women (2 studies, one in Japanese women) and in breast cancer
`patients (6 studies). These data show no definitive difference between these populations:
`
`APPLICANT’S TABLE
`
`Table 7
`
`Trill
`
`hamuokimic pan-meter: following a sting}: 25.413 dose of LA in
`lulwur-nl in bulliy [amine-open“! Vial-ulcers, postmenopausal hmsl
`ulcer patients. and pmnnopannl patients with benign p-mdogir disuse
`
`ALK}“3 a, mgr] ml)
`Gum ((‘V'n
`
`C“,| Lng'mh
`Gwen IFV’k)
`
`(“ml-33ml)
`Om» ICV‘J»)
`
`In“ “by”
`Malian
`
`26125.2)
`17”.“
`
`3.] (LI)
`2.) 123.] i‘
`mouth
`19 I23.“
`13 rim)
`2.! 04!.0)
`
`3.3 (4.2)
`
`6.0
`1.0
`
`7.0
`7.0
`79'
`8}
`to
`7.0f
`
`6.9
`
`"A [44.33
`menu:
`
`105.17)
`71 (23,0)
`Isms.
`4.! (6.“)
`GAMER;
`6.3013”f
`
`Funnel-pin! vol-Mun
`”6135.5!
`(1)36 ur-IOI
`".8163!”
`DIS-ll Iii-5)I
`Pod-cup”! hum ulcer plint-
`wNin-Hn
`“our“;
`DOIS urn-22]
`116125.0l‘
`comm—is.”
`mus.”
`002! m-U‘
`88.4 (17.1]
`09219. I x 5m! trID)
`mums;
`[019.2nliulln-lll
`l05l59.3l
`Imp-In] k".- pm‘bp' putt-II
`ll (41!)
`(x119 iii-7)!
`in mm 11
`l
`.-
`' or n-
`m
`.m m
`*rmimxuuainhcusmmamgmpm.
`”Trdwzluxddrlxlfi-nfldoxingmfim
`"hliernwithma'i-efihmidx
`‘II-ZO.
`’n-I7.
`£rl].
`hluld.
`‘n-li
`’I-5.
`CVCod'Ec'unuIr-iflm
`
`b)
`
`is this a high extraction ratio or a low extraction ratio drug?
`
`No experiments to directly assess clearance across an individual organ were performed.
`Based upon clearance approximating hepatic blood flow, and low concentration
`following oral administration (possibly attributable to low oral bioavailability possibly
`
`13
`
`

`

`due to high first pass effect), it can be speculated that fulvestrant is a high extraction
`drug.
`
`c) does mass balance study suggest renal or hepatic the major route of elimination?
`
`Following IV dosing of ”C, less than 1% of the I"C was recovered in urine.
`Approximately 80% of the MC dose was recovered in feces, of which approximately 8%
`was fulvestrant. It appears that metabolism is the primary route of elimination and that
`feces is the primary roue of excretion of drug-derived material.
`
`5. What is the inter- and intra-subject variability of PK parameters in volunteers
`and patients, and what are the major causes of variability?
`
`Intra-subject variability has not been formally assessed. Inter-subject variability is high,
`as evidenced by the table below which is excerpted from Table 6 (Section 4.2) of the
`Applicant’s Study Report for Trial 0039.
`
`EXCERPTED from APPLICANT’S TABLE
`
`AUC(o-t.28d)
`
`
`
`
`
`_
`
`2x2.5mlinjections,
`n-— 17a
`
`(n: Iml)
`(n: Iml)
`(n d/ml
`Gmean (CV %) Gmean (CV %) Median (range) Gmean
`(CV %)
`6.98 (3.0to9. 1) 2.13 (410)
`
`105.5(59. 3)
`
`6. 17(67. 3)
`
`
`
`
`
`
`AUC(O-t, 28d) —— Area under the plasma concentration-time curve from time zero to 28
`days afier injection.
`Cmax -- Maximum plasma concentration.
`Cmin -- Plasma concentration at 28 days after dosing.
`Tmax -- Time to maximum plasma concentration.
`an = 18 for Cmax and Tmax.
`
`Attempts to attribute intersubject variability to patients characteristics will be discussed
`below in the sections “C. Intrinsic Factors” and “D. Extrinsic Factors.”
`
`C. Intrinsic Factors
`
`1. What intrinsic factors (age, gender, race, weight, height, disease, genetic
`polymorphism, pregnancy, and organ dysfunction) influence exposure and/or
`response and what is the impact of any differences in exposure on the
`pharmacodynamics?
`
`The dependency of fulvestrant pharrnacokinetics on the effects of various disease states
`and demographic factors was investigated by collectively analyzing data from the Phase
`III efficacy trials (Trials 0021 and 0020). Data from a total of 294 subjects after single
`and multiple doses were analyzed (73 administered 125 mg and 221 administered 250 mg
`
`14
`
`

`

`fulvestrant). Relationships presented below were generated from NONMEM using
`Bayesian based methodology (POSTHOC). Although the parameter shown in the Figures
`below is clearance, similar results were obtained for modeled Cmax, Cmin, single dose
`AUC and steady-state AUC.
`
`Hepatic and renal impairment
`
`The Applicant sought, and received from FDA, prior commitment that lack of a study in
`patients with severe hepatic impairment would not be a filing issue. The Applicant’s
`rationale for not performing such a study is that IV drug administration is accomplished
`via apheresis and that the majority of cirrhotic patients do not meet the apheresis
`requirements for hematologic factors and negative infectious disease (e.g., hepatitis C).
`
`There were several patients with hepatic or renal impairment at entry to Trials 0021 or
`0020. For the purposes of this analysis, mild hepatic impairment was defined as an
`alanine aminotransferase concentration (at any visit) greater than the upper limit of the
`normal reference range (ULN) but less than twice the ULN, or if any 2 of the following 3
`parameters were between 1- and 2-times the ULN: aspartate aminotransferase, alkaline
`phosphatase, or total bilirubin. Two hundred sixty-one patients were classified as having
`normal liver function while 24 had mild impairment. Categorical renal impairment was
`not defined but fulvestrant kinetics were assessable relative to creatinine clearance in 280
`
`patients.
`
`There was no clear relationship between fulvestrant clearance and hepatic impairment
`(see Figure, excerpted from the Applicant’s Pharmacokinetics Summary, below). The
`kinetics of the LA IM formulation are controlled by absorption and hence this lack of
`effect is not surprising.
`
`APPLICANT’S FIGURE
`
`Figure ll Smut-r plnl Mildiu‘dlul dean-n u a function oflim fuminn (categorical
`Mme-t) [Trials ”III and “III. 1:288)
`
`”1
`£51.
`- m‘
`5 351
`E3025]
`in;
`101s
`
`30
`
`oJ—-~_ _
`
`l
`s
`
`8
`
`.
`
`
`
`Wig-Ema
`
`Hum
`
`Although there were several patients with low creatinine clearance (<30 ml/min), there
`was no clear relationship between this parameter and fulvestrant clearance (see Figure,
`
`15
`
`

`

`excerpted from the Applicant’s Pharmacokinetics Summary, below). This is consistent
`with the fact that fulvestrant is eliminated almost entirely by metabolism/biliary
`excretion. This suggested that clearance was relatively stable in these groups and may be
`due to the slow release of the compound from the injection site and the capacity of an
`impaired organ to metabolize in excess of this rate of release.
`
`APPLICANT’S FIGURE
`
`Figure 11 Scatter plot Nildividull dun-cc as a III-dial den-lining- darn-cc
`(Trish .021 Ind 0020. F294)
`
`507
`U.
`K)
`v.4 ,.
`:5.
`
`..9..,.
`Inc
`
`
`
`rum-dundnnlm(Mn)
`
`|)
`
`IL
`
`JD
`
`H)
`
`33
`
`I00
`
`33)
`
`I“!
`
`IfI'J
`
`Patan m allnlm- clelra-oe tmt/nin)
`
`Age
`
`As most of the clinical phannacokinetic data were obtained in postmenopausal female
`volunteers or patients, a separate trial to evaluate the pharrnacokinetics of fulvestrant in
`the elderly was not conducted. To examine the relationship between fulvestrant
`concentrations and age, data obtained from the Phase III efficacy trials (Trials 0021 and
`0020) were evaluated using a population pharmacokinetic model of data from 294
`patients with ages ranging from 33 to 89 years. No clear relationships could be identified
`between fulvestrant clearance and age (see Figure, excerpted from the Applicant’s
`Pharmacokinetics Summary, below).
`
`APPEARS THIS WAY
`0N ORIGINAL
`
`l6
`
`

`

`APPLICANT’S FIGURE
`
`Figure 13 Scam-r plot oli-dividnll delta-re as I function dam (Trials ”III and .02..
`F294)
`
`n-u
`u,atu»-..
`-Ii
`
`
`
`
`
`Din-"muthan.”(I‘ll)
`
`u
`
`u
`
`y)
`
`n
`
`M
`
`u
`
`M
`
`09
`
`m
`
`r».
`
`5|:
`
`:5
`
`v.
`
`v5
`
`hut-nil age or)
`
`Ethnicity
`
`The patient population in Trials 0021 and 0020 included 257 white, 23 black, 13
`Hispanic, and 1 Asian subject. The Figure below (excerpted from the Applicant’s
`Pharmacokinetics Summary, below) illustrates that there was no relationship between
`fulvestrant clearance and race.
`
`APPLICANT’S FIGURE
`
`Figure H Scatter plot Mildiridual dun-cc u a function a! rate (Trish M2! and “Ill.
`F294)
`
`E
`8
`
`8
`g
`
`D
`
`o
`
`.....
`(autumn
`
`
`
`..-#_,____-__
`mrk
`"Spit":
`Patient rate (ultgofinl)
`
`_
`__ A...
`0th?!
`
`,,,.. ._._ , .4
`
`60
`55
`50 _
`_ u .
`.
`a
`'7:
`“I
`I
`.
`_
`s
`y.
`‘E
`:‘L‘
`i
`:5 -
`
`2D -
`I.‘
`JI‘
`
`5u
`
`3
`“I-
`
`In addition to the demographic analysis from Trials 0021 and 0020, phannacokinetics for
`5 female Japanese at the 250-mg dose (Tn'al 0-15-1 l) were similar to those obtained
`from Trial 0004, which was conducted in a predominantly white population in the United
`Kingdom (18 white and 1 black). The mean plasma fulvestrant profiles for Japanese and
`UK data are illustrated in the Figure below.
`
`17
`
`

`

`APPLICANT’S FIGURE
`
`Figure IS (fmpar‘nnl n! the mnSD [flann- coltulntinl llmflIt-s I‘m flhutml LA
`250 lag in Tn'ab (HS-ll II-‘-‘5) and mm(n=15)
`
`l0)
`
`(Mm-l) 6.1
`Plainum.
`
`l0
`
`1;:
`
`I
`
`II
`
`H
`
`it
`
`The sun dole (days)
`+6154] . .1) -leum
`
`Gender
`
`It should be noted that the Indication is gender specific: “...treatment of postmenopausal
`women...”.
`
`The pharmacokinetics of fulvestrant following IV injection were characterized in healthy
`men and women in Trials 0012 (postmenopausal women only) and 0038 (both pre-and
`postmenopausal women). Comparison of the data shows little difference between men
`and postmenopausal women in terms of the phannacokinetics of fulvestrant. Statistical
`comparison (ANOVA) of the effect of gender indicated that AUCos values obtained for
`men and postmenopausal women were almost identical, giving a ratio close to 1 (see
`Table below). There does appear to be little difference in AUCO-8 values between men
`and premenopausal women. However, the sample sizes are small.
`
`APPLICANT’S TABLE
`
`Table 20 Statistical ”unis oi I’Ihwtnnl expowm in mm and pre- Ild plume-opal“!
`var-en after a single lo-lg iv dose (his! M38)
`
`Common
`AUCg. luplnnl 1
`Emu-n: oi mm 9m. sour-lance
`61.5mm
`min
`intent]
`him Wm
`
`Mun-x WMWIn-Sl
`25!
`NP
`0.96
`031101.10
`n
`.
`Lien Vs Em roux. (Ir-B)
`2.“
`192
`1.20
`I‘m In l.36
`
`18
`
`

`

`Body weight
`
`No relationship was identified between fulvestrant clearance and body weight in 285
`patients ranging from 40 to 127 kg (Figure below).
`
`APPLICANT’S FIGURE
`
`“gun 16 Scatter pint dhdividual dean-cc u a [union ol'bndy weight (Trials .02]
`and 0010, F294)
`
`Il
`u
`
`alum-lean."It.“
`
`Duh
`
`-—-uuumI:eu.I’‘'u..-,
`
`anI.
`;.4C.
`‘lL'! *——1
`-:
`40
`50
`V.)
`70
`in
`9:)
`{CG
`
`~77
`
`77777177
`
`7
`
`1 I I
`
`lZL'
`
`Patient weight (M!)
`
`2. Based upon what is known about exposure-response relationships and their
`variability, and the groups studied (volunteers vs. patients); what dosage
`regimen adjustments, if any, are recommended for each of these subgroups
`(examples shown below)? If dosage regimen adjustments are not based upon
`exposure-response relationships, describe the alternative basis for the
`recommendation.
`
`No dosage adjustments are recommended. Consistent with the indication (post-
`menopausal) no pregnancy and lactation use information is in the application.
`
`D. Extrinsic Factors
`
`1. What extrinsic factors (drugs, herbal products, diet, smoking, and alcohol use)
`influence exposure and/or response and what is the impact of any differences in
`exposure on pharmacodynamics?
`
`With the exception of drugs, which will be discussed below, no extrinsic factors were
`studied.
`
`2. Based upon what is known about exposure-response relationships and their
`variability, what dosage regimen adjustments, if any, do you recommend for
`each of these factors? If dosage regimen adjustments across factors are not
`based on the exposure-response relationships, describe the basis for the
`recommendation.
`
`No dosage adjustments for extrinsic factors are recommended.
`
`19
`
`

`

`3. Drug-Drug Interactions
`a)
`is there an in vitro basis to suspect in vivo drug-drug interactions?
`
`Yes. In human liver microsomes fulvestrant appears to be metabolized by CYP 3A4.
`
`b)
`
`is the drug a substrate of CYP enzymes?
`
`To investigate