Evaluation of Crizotinib AbsoluteBioavailability, the Bioequivalence ofThree Oral Formulations, andthe Effect of Food on CrizotinibPharmacokinetics in Healthy Subjects

Huiping Xu, PhD1, Melissa O’Gorman, BS2, Tanya Boutros, MSc1,Nicoletta Brega, MD3, Constantino Kantaridis, MD4, Weiwei Tan, PhD1,and Akintunde Bello, PhD5

AbstractCrizotinib (Xalkori1) is an orally administered, selective, small-molecule, ATP-competitive inhibitor of the anaplastic lymphoma kinase (ALK) andmesenchymal epithelial transition factor/hepatocyte growth factor receptor tyrosine kinases, and has recently been approved for the treatment ofALK-positive non-small cell lung cancer. The absolute bioavailability of crizotinib, effect of a high-fat meal on crizotinib pharmacokinetics (PK), andbioequivalence of several oral formulations (powder in capsule [PIC], immediate-release tablet [IRT], and commercial formulated capsule [FC]) wereevaluated in two phase I clinical studies involving healthy volunteers who received single doses of crizotinib. PK parameters for crizotinib and itsmetabolite, PF-06260182, were determined using non-compartmental methods. The absolute oral bioavailability of crizotinib was approximately 43%,with a slight decrease in crizotinib exposures (area under the plasma concentration–time profile andmaximum plasma concentration) following a high-fat meal that was not considered clinically meaningful. The FC was bioequivalent to the clinical development IRT and PIC formulations. No seriousadverse events were observed during either study and the majority of adverse events were mild, the most common being diarrhea. Single-dosecrizotinib could be safely administered to healthy subjects.

Keywordscrizotinib, absolute bioavailability, bioequivalence, pharmacokinetics

Lung cancer is the leading cause of cancer deathworldwide, and non-small cell lung cancer (NSCLC) isthe predominant form of lung cancer. Chemotherapy hasbeen the primary NSCLC treatment option for decades,but is associated with only modest activity.1 Theidentification of activating mutations in epidermal growthfactor receptor (EGFR) tyrosine kinase and the subse-quent approval of EGFR inhibitors (gefitinib anderlotinib) heralded a new era of molecular therapy forNSCLC.2–6 Rearrangements in the anaplastic lymphomakinase (ALK) gene, recently described in NSCLC, offer anadditional molecularly defined subgroup of lung cancerpatients in which to explore the benefits of targetedtherapy, compared with standard chemotherapy.7,8 Cri-zotinib (Xalkori1) is a first-in-class, oral, potent,adenosine triphosphate (ATP)-competitive small-mole-cule inhibitor of Met/hepatocyte growth factor receptor(HGFR) and ALK tyrosine kinases and their oncogenicvariants.9 Consistent with these mechanisms of action,crizotinib dose-dependently inhibits phosphorylation andkinase target-dependent functions of c-Met/HGFR, ALK,and selected variants in tumor cells both in vitro andin vivo. In a retrospective analysis, the median 2-year

survival of patients with ALK-positive NSCLC treatedwith crizotinib was shown to be greater than 50%, butonly 12% for the same patient population treated withstandard therapy.10 The robust clinical activity ofcrizotinib in previously treated patients withALK-positiveNSCLC resulted in a priority review for a new drugapplication by the Food and Drug Administration(May 2011) and the subsequent approval of the 250mgtwice daily (BID) regimen in August 2011.10

Powder in capsule (PIC) and immediate-release tablet(IRT)were the formulationsused in early crizotinib clinical

The Journal of Clinical Pharmacology55(1) 104–113© 2014, The American College ofClinical PharmacologyDOI: 10.1002/jcph.356

1Pfizer Inc., La Jolla, CA, USA2Pfizer Inc., Groton, CT, USA3Pfizer Inc., Milan, Italy4Pfizer Inc., Brussels, Belgium5Pfizer Inc., New York, NY, USA

Submitted for publication 28 February 2014; accepted 30 June 2014.

Corresponding Author:Huiping Xu, Clinical Pharmacology, Pfizer Inc., La Jolla, CA 92121,USAEmail: Huiping.xu@pfizer.com

Pharmacokinetics

trials as well as the pivotal efficacy studies. These clinicalformulations were made available as 50- and 100-mgcapsules or tablets with the flexibility for administration atmultiple dose levels. The 250-mg commercial formulatedcapsule (FC)wasdeveloped formarketuseonce theclinicalcrizotinib dose and schedule was established.The absolute oral bioavailability of crizotinib was

determined in a study in healthy volunteers following theadministration of single doses of an experimentalintravenous (IV) formulation and the oral clinical IRTformulation. The objectives of the absolute bioavailabilitystudy were to determine the fraction of crizotinibabsorbed after oral administration and to provideinformation on the distribution and elimination ofcrizotinib. A separate study that included an assessmentof the bioavailability/bioequivalence of the FC relative tothe clinical PIC and IRT formulations (used to establishcrizotinib safety and efficacy) was performed to comparethe commercial FC with those used during clinicaldevelopment (PIC and IRT). This study also included anassessment of the potential effect of high-fat food onsingle-dose crizotinib pharmacokinetics (PK) followingadministration of the IRT formulation in order to provideinformation on how crizotinib should be administered.

MethodsThe study protocols (A8081010 and A8081011) andinformed consent documentation were approved by theEthics Committee of the Erasme Hospital (Brussels,Belgium). The studies were conducted in compliancewiththe ethical principles originating in or derived from theDeclaration of Helsinki and in compliance with allInternational Conference on Harmonization Good Prac-tice Guidelines. In addition, all local regulatory require-ments were followed; in particular, those affordinggreater protection to the safety of study participants.

SubjectsSubjects were healthy males or females of non-childbearing potential between the ages of 18 and 55 yearswith bodymass index of 17.5–30.5 kg/m2 and a total bodyweight of >50 kg (110 lb). All subjects signed aninformed consent document before screening.Any subject with one of the following was to be

excluded: evidence or history of clinically significantdiseases, any condition possibly affecting drug absorp-tion, a positive drug screen, history of excess alcoholconsumption, QTc> 450ms at screening, use of drugs ordietary supplements within 7 days or 5 half-lives prior tothe start of study drug treatment, blood donation ofapproximately 500mL within 56 days prior to dosing, apositive serology for hepatitis B or C, and currentlysmoking.

Study DesignsStudies A8081010 and A8081011 were open-label,randomized, crossover, single-dose phase I studiesperformed at the Pfizer Clinical Research Unit (PCRU)at the Erasme Hospital, Brussels, Belgium to evaluate theabsolute bioavailability of crizotinib (A8081010) and therelative bioavailability/bioequivalence and food effect ofcrizotinib in healthy subjects (A8081011). The overalldesigns of the two studies are summarized in Table 1.Subjects who withdrew were not replaced unless the totalnumber of evaluable subjects fell below the targetedsample size (see below). Evaluable subjects were thosewho completed the study with all required PK samplescollected. In each treatment period, all subjects wereadmitted to the PCRU on day 0, dosed on day 1, anddischarged from the PCRU on day 5, and all returned foran outpatient visit on day 7. If necessary, a follow-up visitwas performed after the last period. There was a washoutperiod of at least 14 days between each crizotinibtreatment.

Table 1. Summary of Study Designs

Study Primary objective(s) Na Design Treatments

A8081010 To determine the absolutebioavailability of crizotinib

14 2-Treatment 1A: a 50-mg single IV dose as 200mL of a 0.25mg/mL IVsolution infused over approximately 2 hours2-Period

2-Sequence 1B: a 250-mg single oral dose in a fasted state as 1� 50-mgIRT and 2� 100-mg IRTs

A8081011 To demonstrate thebioequivalence of FC with PICand IRT

36 4-Treatment 2A (reference 1 for BE): a 250-mg single oral dose in a fastedstate as 1� 50-mg IRT and 2� 100-mg IRTs4-Period

To evaluate the effect of food oncrizotinib PK

4-Sequence 2B (reference 2 for BE): a 250-mg single oral dose in a fastedstate as 1� 50-mg PIC and 2� 100-mg PICs

2C (test for BE and reference for food effect): a 250-mg singleoral dose in a fasted state as 1� 250-mg FC

2D (test for food effect): a 250-mg single oral dose with high-fat meal as 1� 250-mg FC

BE, bioequivalence; FC, formulated capsule; IRT, immediate-release tablet; IV, intravenous; PIC, powder in capsule; PK, pharmacokinetics.aNumber of subjects to be enrolled.

105

All crizotinib doses were administered in the morning. IVdoses of crizotinib were administered as a constant-rateinfusion over approximately 2 hours. All oral doses ofcrizotinib were given with 240mL ambient-temperaturewater. Tablets and capsule(s) were required to beswallowed whole without chewing. In order to test forfood effect (Treatment 2D, Table 1), subjects wererequired to consume a high-fat meal provided by theinvestigator site personnel prior to drug administration.The meal was to be completed within 30minutes, withcrizotinib FC administered 30minutes after the start of themeal. For all other treatments, subjects fasted for 10 hoursbefore crizotinib administration.In order to standardize the conditions on PK sampling

days, all subjects were required to refrain from lying down(except when required for blood pressure, pulse rate, andelectrocardiogram measurements), eating, and drinkingbeverages other than water during the first 4 hours afterdosing. Subjects were not allowed to eat grapefruit orgrapefruit-related citrus fruits (eg, Seville oranges,pomelos) or drink grapefruit juice from 7 days prior tothe first dose of crizotinib until collection of the final PKblood sample. Subjects were to abstain from alcohol andcaffeine-containing products for 24 hours prior to admis-sion to the PCRU and continue abstaining from alcoholuntil collection of the final PK sample of each studyperiod. Subjects were to abstain from the use of tobacco-or nicotine-containing products prior to dosing and untilcompletion of the study. All subjects were to abstain fromstrenuous exercise for at least 48 hours prior to each bloodcollection for clinical laboratory tests per study period.In each study, a randomization number was allocated

to each subject. This number was retained throughout thestudy and corresponded to a treatment schedule deter-mined by a Pfizer-generated randomization code.Crizotinib dosage forms were supplied by Pfizer: IV

solution in 50-mL vials containing a sterile 5.0-mg/mL IVformulation of crizotinib (preparation of individual doseswas performed by PCRU pharmacy personnel), IRT andPIC in 50- and 100-mg tablets or capsules, and FC as 250-mg capsules. All study treatments were administeredunder supervision.

Pharmacokinetic Sample Collection and AnalyticMethodBlood samples (3mL for Study A8081010 and 4mL forStudy A8081011) were collected into appropriatelylabeled tubes containing dipotassium ethylenediamine-teraacetic acid at designated times. For IV treatment (1A,Table 1), collection times were at 0 (predose), and 1, 2,2.5, 3, 4, 6, 8, 12, 24, 36, 48, 72, 96, and 144 hours after thestart of crizotinib infusion. For all oral treatments,collection times were at 0 (predose), and 1, 2, 4, 5, 6,8, 10, 12, 24, 36, 48, 72, 96, and 144 hours after oraldosing. Once collected, blood samples were processed

immediately and kept out of direct sunlight due to thelight-sensitive nature of crizotinib. After thorough mixingwith anticoagulant, blood samples were centrifuged atapproximately 1700� g for 10minutes at 4 °C to harvestthe plasma. Finally, the plasma samples were stored inamber vials at –20 °C within 1 hour of blood samplecollection until shipment on dry ice to CovanceBioanalytical Services, LLC (Indianapolis, IN, USA).All plasma samples were stored at –10 to –30 °C in ananalytical laboratory and then assayed within theestablished stability period. Plasma concentrations ofcrizotinib and its metabolite, PF-06260182, were deter-mined using a validated, sensitive, and specific high-performance liquid chromatography tandem mass spec-trometric method. Calibration standard responses werelinear over the range of 0.200–200 ng/mL for crizotiniband PF-06260182 using a weighted (1/concentration)squared linear regression. Those samples with crizotinibconcentrations greater than the upper limit of quantifica-tion were diluted into calibration range. Clinical speci-mens with crizotinib and PF-06260182 concentrationbelow the lower limit of quantification (LLOQ, 0.200 ng/mL) were reported as below LLOQ.For Study A8081010, between-day assay accuracy (%

relative error [%RE]) for quality control (QC) concen-trations ranged from 2.0% to 5.3% for the low, medium,and high QC samples for crizotinib and from 0.7% to4.2% for the low, medium, and high QC samples for PF-06260182. Assay precision (% coefficient of variation [%CV]) was �3.4% (crizotinib) and �2.3% (PF-06260182)for the low,medium, high, and diluted QC concentrations.For Study A8081011, %RE for QC concentrations rangedfrom 0.7% to 3.0% for the low, medium, and high QCsamples for crizotinib and from –0.7% to 2.0% for thelow, medium, and high QC samples for PF-06260182. %CV was �6.6% (crizotinib) and �5.3% (PF-06260182)for the low,medium, high, and diluted QC concentrations.

Pharmacokinetic AnalysisPlasma concentration–time data for crizotinib and PF-06260182 were analyzed to determine PK parameters foreach subject after each treatment using non-compartmen-tal methods via eNCA (v2.2.2, Pfizer internal program,Groton, CT, USA). Standard non-compartmental analysis(NCA) parameters for the given formulations wereproduced for both analytes, as appropriate. The definitionand method of determination for each PK parameter arelisted in Supplemental Table S1. Samples below theLLOQ were set to 0 ng/mL for the PK analysis. Actualsample collection times were used for the PK parameterestimation. Area under the plasma concentration–timeprofile extrapolated to infinity (AUCinf) and plasma half-life (t1/2) were reported for all treatments as the followingcriteria were met: a well-characterized terminal phasedefined as one with at least three data points, an r2� 0.9

106 The Journal of Clinical Pharmacology / Vol 55 No 1 (2015)

(goodness-of-fit statistics for the log-linear regression),and a span ratio �2 (duration over which t1/2 wasassessed, divided by the t1/2 estimate).

Statistical AnalysisSample Size. Sample size calculations for both studies

were based on estimates of within-subject deviations of0.172 for loge AUCinf and 0.186 for loge maximumplasma concentration (Cmax) obtained from a previous PKstudy in healthy volunteers. For Study A8081010, asample size of 12 subjects provided 90% confidenceintervals (CIs) for the difference between treatments of�0.1609 on the natural log scale for AUCinf with 90%coverage probability. A total of 14 subjects were enrolledin this study to obtain at least 12 evaluable subjects. ForStudy A8081011, a sample size of 32 subjects was chosento provide 98.9% power that the 90% CI for the ratio ofTest to Reference treatment for AUCinf would lie withinthe acceptance region of 80–125% and 97.7% power thatthe 90%CI for the ratio of Test to Reference treatment forCmax would lie within the acceptance region of 80–125%.Consequently, this study had at least 90% power overallto demonstrate bioequivalence of the Test treatmentto both References and to rule out a food effect. Theoverall study power was based on the product of theindividual powers of the comparisons and parameters ofinterest. This estimate was also based on the assumptionthat the true ratio between Test and Reference treatmentsfor both AUCinf and Cmax was 1.05. A total of 36 subjectswas enrolled in order to obtain at least 32 evaluablesubjects.

Statistical Methods. All PK parameters were summa-rized descriptively by treatment and analyte for eachstudy. For each study separately, natural log-transformedexposure parameters (AUC and Cmax) of crizotinib wereanalyzed using a mixed-effect model with sequence,period, and treatment as fixed effects and subject withinsequence as a random effect. Estimates of the adjustedmean differences (Test–Reference) and corresponding90% CIs were obtained from the model. The adjustedmean differences and 90% CIs for the differences wereexponentiated to provide estimates of the ratio of adjustedgeometric means (Test/Reference) and 90% CIs for theratios. All statistics were calculated using SAS(r)Proprietary Software 9.2 (TS1M0; SAS Institute Inc.,Cary, NC).In Study A8081010, IV administration was the

Reference treatment and oral administration was theTest treatment. Crizotinib absolute bioavailability wasestimated as the ratio of dose-normalized AUCinf afteroral treatment compared with IV treatment.In Study A8081011, the FC formulation administered

in the fasted state was the Test treatment for bothbioequivalence assessments. The IRT and PIC formula-tions were the Reference treatments for the bioequiva-

lence assessments. FC administered in the fasted state wasthe Reference treatment and FC administered in the fedstate was the Test treatment for the food effectassessment. Bioequivalence of the FC formulation withthe IRT and PIC formulations was concluded if the 90%CIs for the ratio of adjusted geometric means for bothAUCinf and Cmax for both comparisons fell wholly within80–125%. Lack of an effect of food was concluded if the90% CIs for the ratio of adjusted geometric means forboth AUCinf and Cmax for the food effect assessment fellwholly within 80–125%.All adverse events (AEs) were reported and tabulated.

ResultsFourteen male subjects with seven in each treatmentsequence were enrolled and treated in Study A8081010.All 14 subjects completed the study. Thirty-six subjectswere enrolled and treated in Study A8081011, with ninesubjects in each sequence. Three subjects discontinuedfrom Study A8081011 without completing all treatments:one during treatment with each of the crizotinib 250-mgregimens except the fasted/FC regimen. Of the subjectdiscontinuations, only one discontinuation (crizotinib250mg fasted/IRT) was due to a study drug-related AE(elevated transaminases, see below). The remaining twodiscontinuations were subjects unwilling to remain in thestudy for personal reasons. Demographic and baselinecharacteristics of all subjects are summarized in Table 2.Subject disposition is shown in Supplemental Table S2.

PharmacokineticsAll treated subjects with at least one crizotinib and PF-06260182 concentration parameter were evaluated forPK: all 14 subjects in Study A8081010 and all except onesubject in Study A8081011, who vomited after receivingthe fasted/PIC treatment. Analyzed datasets are summa-rized in Supplemental Table S2. Themean concentration–time profiles of crizotinib after each treatment are shownin Figures 1–3, and for PF-06260182 in Figures 4 and 5.

Table 2. Subject Demographic and Baseline Characteristics

Characteristic A8081010 (N¼ 14) A8081011 (N¼ 36)

Age, mean (SD) (years) 35.7 (11.5) 38.9 (8.1)Sex, M/F, n 14/0 36/0Race, n (%)White 13 (93) 32 (89)Black 0 (0) 3 (8)Other 1 (7) 1 (3)

Weight, mean (SD) (kg) 80.0 (10.5) 79.9 (12.0)BMI, mean (SD) (kg/m2) 24.7 (3.2) 25.2 (3.2)Height (SD) (cm) 180.1 (6.0) 178.0 (6.6)

BMI, body mass index; F, female; M, male; SD, standard deviation. Note: BMIis calculated as weight (kg)/height (m)2.

107

PK parameters for crizotinib and PF-06260182 followingeach treatment are summarized in Tables 3–5.In Study A8081010, crizotinib Cmax was higher after

the 50-mg IV dose than after oral administration of250mg, whereas crizotinib AUCinf was greater with oraladministration (Figure 1). Values for the mean crizotinibclearance (CL) and steady-state volume of distribution(Vss) following IV administration were 46.8 L/h and1772L, respectively (Table 3). If it is assumed thathepatic clearance was the same following IV and oraladministration, the absolute bioavailability of crizotinibafter a 250mg oral dose was approximately 43%(Table 6). The mean ratio of PF-06260182 to crizotinibexposure, as measured by area under the curve to the lastmeasured concentration (AUClast), was markedly lowerafter IV administration of crizotinib than following oral

administration (0.0343 vs. 0.148, respectively; Table 3),suggesting the role of first-pass metabolism in theformation of PF-06260182 following oral administration.In Study A8081011, similar systemic exposures for

both crizotinib and PF-06260182 were observed afteradministration of the IRT, PIC, and FC formulations inthe fasted state (Figures 2 and 5). The FC formulation wasbioequivalent to the IRT and PIC formulations with meanAUCinf ratios of 99.6% and 107%, respectively, and the90% CIs fell within the 80–125% limit, denotingbioequivalence in both cases (Table 6). Plasma crizotinibconcentrations were modestly lower following adminis-tration of crizotinib FC in the fed state relative to that inthe fasted state (Figure 3). Based on the ratios of theadjusted geometric means (fed/fasted), the coadministra-tion of crizotinib as FC with a high-fat meal resulted indecreases in both AUCinf and Cmax of approximately 14%(Table 6).

Safety and TolerabilityIn Study A8081010, there were no deaths or other seriousadverse events (SAEs), and no permanent or temporarydiscontinuations or dose reductions due to AEs. A total of19 treatment-emergent all-causality AEs (eleven fromeight subjects during IV treatment and eight from fivesubjects during oral treatment) were reported (Table 7).Eleven AEs (five during IV treatment and six during oraltreatment) were considered treatment-related. The mostfrequently reported AE was diarrhea (consistentlyconsidered treatment-related), which was mild in severityand experienced by six subjects (two after IV treatmentand four after oral treatment). The only other AE thatoccurred in more than one subject during either treatment

Figure 1. Mean plasma crizotinib concentration–time profiles afteradministration of single doses of crizotinib 50mg intravenously (IV) or250mg orally in Study A8081010. SD, standard deviation.

Figure 2. Mean plasma crizotinib concentration–time profiles after single oral 250-mg doses under fasted conditions in Study A8081011. FC,formulated capsule; IRT, immediate-release tablet; PIC, powder in capsule; SD, standard deviation.

108 The Journal of Clinical Pharmacology / Vol 55 No 1 (2015)

was catheter site pain, which was experienced by twosubjects during IV treatment. All AEs were mild inseverity, except for one moderate influenza-like illnessreported during IV treatment. This AE occurred on day 13after IV administration and was still present at the time ofdata collection, and was considered related to a viralinfection and not to study treatment.In Study A8081011, one subject permanently discon-

tinued treatment with the IRT formulation in Period 3 dueto clinically relevant liver enzymes elevation (alanineaminotransferase >5� upper limit of normal [ULN] andaspartate aminotransferase >2.5�ULN) not associatedwith total bilirubin elevation. This event resolved 21 daysafter the last crizotinib dose. There were no other AEs

Figure 3. Mean plasma crizotinib concentration–time profiles after single oral 250-mg doses of crizotinib formulated capsule (FC) with or withoutfood in Study A8081011. SD, standard deviation.

Figure 4. Mean plasma PF-06260182 concentration–time profiles aftersingle-dose administration of crizotinib 50mg intravenously (IV) or250mg orally in Study A8081010. SD, standard deviation.

Figure 5. Mean plasma PF-06260182 concentration–time profiles after single oral 250-mg doses of crizotinib given under fasted or fed conditions inStudy A8081011. FC, formulated capsule; IRT, immediate-release tablet; PIC, powder in capsule; SD, standard deviation.

109

related to laboratory abnormalities. The most frequentlyreported AE was diarrhea, which was mild in severity andexperienced by nine subjects during treatment with the250mg crizotinib IRT/fasted regimen and by 13 subjectsduring treatment with each of the other three regimens(Table 7). All instances of diarrhea in all treatments wereconsidered related to study treatment. The other mostfrequently reported all-causality and treatment-relatedAEs (five or more subjects) were nausea, headache, andfatigue. The majority of AEs (regardless of causality)were mild in severity, with the exception of the followingmoderate AEs: IRT/fasted treatment, headache (n¼ 2)and elevated liver enzymes (n¼ 1); PIC/fasted treatment,nausea (n¼ 4), vomiting (n¼ 2), headache (n¼ 2), andnasopharyngitis (n¼ 1); FC/fasted treatment, gingivitisand headache (n¼ 1 each); FC/fed treatment, headache,tooth abscess, and urinary tract infection (n¼ 1 each).

DiscussionThe systemic plasma clearance of crizotinib (CLiv) was46.8 L/h in healthy subjects following IV administrationof a single 50-mg dose. Results of a mass balance study inhumans indicated that the renal excretion of unchanged14C-labeled crizotinib was minimal (Pfizer Inc., data onfile). Hence, hepatic clearance of crizotinib can beapproximated by CLiv. Moreover, the blood to plasmapartitioning ratio of crizotinib was estimated to be 1.0.11

Given that hepatic blood flow is 1450mL/min for a 70-kghealthy subject, the hepatic extraction ratio (EH) isestimated to be 53.8%, indicating that crizotinib is amoderate hepatic extraction ratio drug after a single 50-mg IV dose.The observed mean absolute oral bioavailability of

approximately 43% for crizotinib was supportive of theselection of the oral route of administration. The absoluteoral bioavailability of crizotinib was similar to the valueestimated for its hepatic availability (46.2%; as deter-mined by 1 � EH), indicating that approximately 93% ofthe dose (fraction absorbed [Fa]) entered the liver afteroral administration, assuming that clearance after oral andIV administration are the same. However, a significant

Table 3. Summary of Plasma Crizotinib and PF-06260182Pharmacokinetic Parameters Following Single Administration of anIntravenous or Oral Dose of Crizotinib in Study A8081010a

Parameter (units)Crizotinib50mg IV

Crizotinib250mg oral

Crizotinib n¼ 14 n¼ 14AUCinf (ng h/mL) 1067 (18) 2321 (34)AUClast (ng h/mL) 1007 (18) 2250 (35)Cmax (ng/mL) 155 (19) 99.6 (28)Tmax (h) 1.92 (1.00–1.95) 5.00 (4.00–6.00)t1/2 (h) 38.9 (16) 29.0 (10)CL/F (L/h) NC 108 (32)Vz/F (L) NC 4478 (35)CL (L/h) 46.8 (18) NCVss (L) 1772 (18) NCAUCinf (dn)(ng h/mL/mg)

21.4 (18) 9.28 (34)

AUClast (dn)(ng h/mL/mg)

20.1 (18) 9.00 (35)

Cmax (dn) (ng/mL/mg) 3.10 (19) 0.398 (28)PF-06260182 n¼ 14 n¼ 13AUCinf (ng h/mL) NC 360 (31)AUClast (ng h/mL) 35.6 (45) 343 (32)Cmax (ng/mL) 3.01 (45) 26.5 (24)Tmax (h) 2.50 (1.92–4.02) 5.00 (5.00–6.07)MRAUCinf NC 0.144 (13)MRAUClast 0.0343 (31) 0.148 (14)MRCmax 0.0188 (36) 0.258 (17)

AUCinf, area under the plasma concentration–time profile from time zeroextrapolated to infinite time; AUClast, area under the plasma concentration–time profile from time zero to the time of the last quantifiable concentration(Clast); CL, systemic clearance; CL/F, apparent oral clearance after oral dose;Cmax, maximum observed concentration; CV, coefficient of variation; dn,dose-normalized; IV, intravenous; MR, metabolite to parent ratio; NC, notcalculated; t1/2, terminal half-life; Tmax, time for Cmax; Vss, volume ofdistribution; Vz/F, apparent volume of distribution.aData are presented as geometric mean (%CV) for all parameters except forTmax, which is presented as median (range), and t1/2, which is presented asarithmetic mean (%CV).

Table 4. Summary of Plasma Crizotinib and PF-06260182Pharmacokinetic Parameters Following Single Administration ofThree Different Oral Formulations of Crizotinib in the Fasted Statein Study A8081011a

Parameter (units)CrizotinibIRT

CrizotinibPIC

CrizotinibFC

Crizotinib n¼ 35 n¼ 35 n¼ 35AUCinf (ng h/mL) 2890 (34) 2665 (41) 2887 (36)AUClast (ng h/mL) 2763 (34) 2531 (42) 2761 (36)Cmax (ng/mL) 126 (28) 119 (39) 135 (33)Tmax (h) 5.00

(1.00–8.00)5.00

(2.00–6.00)5.00

(2.00–6.00)t1/2 (h) 34.6 (12) 35.3 (18) 34.9 (14)CL/F (L/h) 86.5 (33) 93.8 (48) 86.6 (56)Vz/F (L) 4290 (37) 4703 (53) 4313 (77)

PF-06260182 n¼ 35 n¼ 35 n¼ 35AUCinf (ng h/mL) 442 (46) 402 (55) 447 (49)AUClast (ng h/mL) 432 (46) 391 (56) 437 (50)Cmax (ng/mL) 32.2 (36) 29.7 (46) 33.0 (34)Tmax (h) 6.00

(4.00–8.02)6.00

(4.00–8.00)5.00

(4.00–10.0)

AUCinf, area under the plasma concentration–time profile from time zeroextrapolated to infinite time; AUClast, area under the plasma concentration–time profile from time zero to the time of the last quantifiable concentration(Clast); CL/F, apparent oral clearance after oral dose; Cmax, maximumobserved concentration; CV, coefficient of variation; FC, formulated capsule;IRT, immediate-release tablet; PIC, powder in capsule; t1/2, terminal half-life;Tmax, time for Cmax; Vz/F, apparent volume of distribution.aData are presented as geometric mean (%CV) for all parameters except forTmax, which is presented as median (range), and t1/2, which is presented asarithmetic mean (%CV).

110 The Journal of Clinical Pharmacology / Vol 55 No 1 (2015)

proportion of unchanged crizotinib (53% of the dose),likely representing unabsorbed parent drug, was recov-ered in the feces after a single 250-mg oral dose of 14C-labeled crizotinib in a mass balance study (Pfizer Inc.,data on file). The apparent overestimation of Fa suggests

that crizotinib PK is dose-/concentration-dependentfollowing IV and oral administration of single doses.PF-06260182 (with activity against the ALK target) wasidentified as the most abundant circulating crizotinibmetabolite in the mass balance study. The notably lowerAUClast metabolite to crizotinib ratio following IVadministration (0.0343) compared with oral dosing(0.148) suggests that first-pass metabolism (in the GItract and/or liver) may play a primary role in theformation of PF-06260182 after the oral administration ofcrizotinib.The finding that the FC formulation was bioequivalent

to both the IRT and PIC suggests similar performancecharacteristics and that the marketed FC formulation isexpected to perform in a similar manner to the PIC andIRT formulations used during clinical development. Theco-administration of crizotinib with a high-fat mealresulted in a slight (14%) reduction in crizotinib exposure,with no change in other PK parameters (eg, Tmax or t1/2).The observed decrease in crizotinib Cmax and AUC withfood is not considered to be clinically meaningful whenviewed in the context of the overall inter-subjectvariability of crizotinib systemic exposure seen in thetwo studies following oral administration (coefficient ofvariation values for Cmax and AUC of 28%–42%).In conclusion, crizotinib has an absolute oral bioavail-

ability of approximately 43% following oral administra-tion of a single 250-mg dose compared with IVadministration of a 50-mg dose. However, since crizotinibdemonstrates non-linear time-dependent PK followingmultiple dosing,12 the potential effects of repeated dosingon oral bioavailability are not known. The formulationsused in clinical development (PIC and IRT) arebioequivalent to the marketed FC form. In addition, these

Table 6. Statistical Summary of Treatment Comparison for Crizotinib

Parameter (units)

Adjusted geometric meansRatio (Test/Reference)of adjusted meansa

90% CIfor ratioTest Reference

A8081010, crizotinib 250mg oral (Test) vs. crizotinib 50mg IV (Reference)AUCinf (dn) (ng h/mL/mg) 9.28 21.4 43.4 39.7, 47.6

A8081011, crizotinib FC fasted (Test) vs. crizotinib IRT fasted (Reference)AUCinf (ng h/mL) 2886 2899 99.6 91.5, 108Cmax (ng/mL) 135 126 107 96.6, 119

A8081011, crizotinib FC fasted (Test) vs. crizotinib PIC fasted (Reference)AUCinf (ng h/mL) 2886 2699 107 98.3, 116Cmax (ng/mL) 135 121 111 100, 123

A8081011, crizotinib FC fed (Test) vs. crizotinib FC fasted (Reference)AUCinf (ng h/mL) 2475 2886 85.8 78.9, 93.3Cmax (ng/mL) 116 135 86.2 77.9, 95.4

AUCinf, area under the plasma concentration–time profile from time zero extrapolated to infinite time; CI, confidence interval; dn, dose-normalized; Cmax,maximum observed concentration; FC, formulated capsule; IRT, immediate-release tablet; IV, intravenous; PIC, powder-in-capsule.aExpressed as percentages.

Table 5. Summary of Plasma Crizotinib and PF-06260182Pharmacokinetic Parameters Following Single Administration of theFormulated Capsule in the Fed or Fasted State in Study A8081011a

Crizotinib, fasted Crizotinib, fed

Crizotinib n¼ 35 n¼ 36AUCinf (ng h/mL) 2887 (36) 2475 (39)AUClast (ng h/mL) 2761 (36) 2359 (40)Cmax (ng/mL) 135 (33) 116 (37)Tmax (h) 5.00 (2.00–6.00) 5.00 (1.00–8.00)t1/2 (h) 34.9 (14) 35.4 (16)CL/F (L/h) 86.6 (56) 101 (36)Vz/F (L) 4313 (77) 5096 (41)

PF-06260182 n¼ 35 n¼ 36AUCinf (ng h/mL) 447 (49) 342 (51)b

AUClast (ng h/mL) 437 (50) 325 (53)Cmax (ng/mL) 33.0 (34) 23.6 (42)Tmax (h) 5.00 (4.00–10.0) 6.00 (2.00–10.0)

AUCinf, area under the plasma concentration–time profile from time zeroextrapolated to infinite time; AUClast, area under the plasma concentration–time profile from time zero to the time of the last quantifiable concentration(Clast); CL/F, apparent oral clearance after oral dose; Cmax, maximumobserved concentration; CV, coefficient of variation; t1/2, terminal half-life;Tmax, time for Cmax; Vz/F, apparent volume of distribution.aData are presented as geometric mean (%CV) for all parameters except forTmax, which is presented as median (range), and t1/2, which is presented asarithmetic mean (%CV).b35 subjects contributed to the geometric mean.

111

findings indicated that the three formulations could beused interchangeably in the clinical research setting.Although a high-fat meal resulted in a slight decrease incrizotinib exposure, this was not clinically meaningfuland crizotinib can be administered without regard to food.

Acknowledgments

Studies A8081010 and A8081011 were sponsored by Pfizer Inc.Editorial support was provided by Martin Quinn at ACUMED1

(Tytherington, UK) with funding from Pfizer Inc. We thankGrace Ni, Suzanne Phillips, and other operational colleagues atPfizer for their contribution to the conduct of the two studies,Cathie Leister for providing statistical support, and RobertLaBadie at Pfizer for a critical statistical review of themanuscript.

Declaration of Conflicting Interests

All authors are employees and stockholders of Pfizer Inc.

References01. Breathnach OS, Freidlin B, Conley B, et al. Twenty-two years of

phase III trials for patients with advanced non-small-cell lungcancer: sobering results. J Clin Oncol. 2001;19(6):1734–1742.

02. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations inthe epidermal growth factor receptor underlying responsivenessof non-small-cell lung cancer to gefitinib. N Engl J Med. 2004;350(21):2129–2139.

03. Paez JG, Jänne PA, Lee JC, et al. EGFR mutations in lung cancer:correlation with clinical response to gefitinib therapy. Science.2004;304(5676):1497–1500.

04. Pao W, Miller V, Zakowski M, et al. EGF receptor gene mutationsare common in lung cancers from “never smokers” and areassociated with sensitivity of tumors to gefitinib and erlotinib. ProcNatl Acad Sci USA. 2004;101(36):13306–13311.

05. Cohen MH,Williams GA, Sridhara R, et al. United States Food andDrug Administration Drug Approval summary: Gefitinib (ZD1839;Iressa) tablets. Clin Cancer Res. 2004;10(4):1212–1218.

06. Johnson JR, Cohen M, Sridhara R, et al. Approval summary forerlotinib for treatment of patients with locally advanced ormetastatic non-small cell lung cancer after failure of at least one

Table 7. Treatment-Emergent All-Causality and Treatment-Related Adverse Events by MedDRA Preferred Term (�2 Subjects for A8081011)a

A8081010

Crizotinib 50mg IV (n¼ 14) Crizotinib 250mg oral (n¼ 14)

AC TR AC TR

AEDiarrhea 2 2 4 4Catheter site pain 2 0 0 0Flatulence 1 1 1 1Somnolence 1 1 1 1Paresthesia 1 0 1 0Rhinorrhea 1 0 1 0Catheter site phlebitis 1 1 0 0Influenza-like illness 1 0 0 0Back pain 1 0 0 0Total preferred term events 11 5 8 6

Crizotinib 250mg BID

Fasted/IRT(n¼ 35)

Fasted/PIC(n¼ 36)

Fasted/FC(n¼ 35)

Fed/FC(n¼ 36)

A8081011 AC TR AC TR AC TR AC TR

AEDiarrhea 9 9 13 13 13 13 13 13Nausea 5 5 10 10 2 2 5 5Headache 4 4 5 5 4 3 3 2Fatigue 3 3 5 5 2 2 1 1Nasopharyngitis 2 0 3 0 0 0 1 0Abdominal pain upper 2 2 1 1 1 1 1 1Epigastric discomfort 2 2 1 1 2 2 0 0Abnormal gastrointestinal sounds 1 1 0 0 1 1 2 2Somnolence 1 1 1 1 1 1 3 3Abdominal pain 0 0 1 1 0 0 2 2Vomiting 0 0 2 2 0 0 0 0

AC, all-causality; AE, adverse event; BID, twice daily; FC, formulated capsule; IRT, immediate-release tablet; IV, intravenous; MedDRA, Medical Dictionary forRegulatory Activities; PIC, powder in capsule; TR, treatment-related.aData are presented as numbers of subjects.

112 The Journal of Clinical Pharmacology / Vol 55 No 1 (2015)

prior chemotherapy regimen. Clin Cancer Res. 2005;11(18):6414–6421.

07. Rikova K, Guo A, Zeng Q, et al. Global survey of phosphotyrosinesignaling identifies oncogenic kinases in lung cancer. Cell.2007;131(6):1190–1203.

08. Soda M, Choi YL, Enomoto M, et al. Identification of thetransforming EML4-ALK fusion gene in non-small-cell lungcancer. Nature. 2007;448(7153):561–566.

09. Christensen JG, Zou HY, Arango ME, et al. Cytoreductiveantitumor activity of PF-2341066, a novel inhibitor of anaplasticlymphoma kinase and c-Met, in experimental models of anaplasticlarge-cell lymphoma. Mol Cancer Ther. 2007;6(12 Pt 1):3314–3322.

10. Scagliotti G, Stahel RA, Rosell R, Thatcher N, Soria JC. ALKtranslocation and crizotinib in non-small cell lung cancer: an

evolving paradigm in oncology drug development. Eur J Cancer.2012;48(7):961–973.

11. Yamazaki S, Skaptason J, Romero D, et al. Prediction of oralpharmacokinetics of cMet kinase inhibitors in humans: physio-logically based pharmacokinetic model versus traditional one-compartment model. Drug Metab Dispos. 2011;39(3):383–393.

12. Li C, Alvey C, Bello A,Wilner KD, TanW. Pharmacokinetics (PK)of crizotinib (PF-02341066) in patients with advanced non-smallcell lung cancer (NSCLC) and other solid tumors. J Clin Oncol.2011;29: (suppl; abstr e13065).

Supporting InformationAdditional supporting information may be found in theonline version of this article at the publisher’s web-site.

113

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具