quantitative and mechanistic understanding of azd1775 ......cancer therapy: clinical quantitative...

Cancer Therapy: Clinical

Quantitative and Mechanistic Understanding ofAZD1775 Penetration across Human Blood–BrainBarrier in Glioblastoma Patients Using anIVIVE–PBPK Modeling ApproachJing Li1, Jianmei Wu1, Xun Bao1, Norissa Honea2, Youming Xie1, Seongho Kim1,Alex Sparreboom3, and Nader Sanai2

Abstract

Purpose:AZD1775, afirst-in-class, small-molecule inhibitor ofthe Wee1 tyrosine kinase, is under evaluation as a potentialchemo- and radiosensitizer for treating glioblastoma. This studywas to prospectively, quantitatively, and mechanistically investi-gate the penetration of AZD1775 across the human blood–brainbarrier (BBB).

Experimental Design: AZD1775 plasma and tumor pharma-cokinetics were evaluated in 20 patients with glioblastoma. Thedrug metabolism, transcellular passive permeability, and inter-actionswith efflux anduptake transporterswere determined usinghuman derived in vitro systems. A whole-body physiologicallybased pharmacokinetic (PBPK) model integrated with a four-compartment permeability-limited brain model was developedfor predicting the kinetics of AZD1775 BBB penetration andassessing the factors modulating this process.

Results: AZD1775 exhibited good tumor penetration inpatients with glioblastoma, with the unbound tumor-to-plas-ma concentration ratio ranging from 1.3 to 24.4 (median, 3.2).

It was a substrate for ABCB1, ABCG2, and OATP1A2, but notfor OATP2B1 or OAT3. AZD1775 transcellular passive perme-ability and active efflux clearance across MDCKII–ABCB1 orMDCKII–ABCG2 cell monolayers were dependent on the baso-lateral pH. The PBPK model well predicted observed drugplasma and tumor concentrations in patients. The extent andrate of drug BBB penetration were influenced by BBB integrity,efflux and uptake active transporter activity, and drug bindingto brain tissue.

Conclusions: In the relatively acidic tumor microenviron-ment where ABCB1/ABCG2 transporter-mediated efflux clear-ance is reduced, OATP1A2-mediated active uptake becomesdominant, driving AZD1775 penetration into brain tumor.Variations in the brain tumor regional pH, transporter expres-sion/activity, and BBB integrity collectively contribute to theheterogeneity of AZD1775 penetration into brain tumors. ClinCancer Res; 23(24); 7454–66. �2017 AACR.

See related commentary by Peer et al., p. 7437

IntroductionInsufficient penetration of potentially effective chemothera-

peutic agents across the blood–brain barrier (BBB) is a hugehurdle to the successful treatment of brain cancer (1). The BBBis composed of a continuous layer of brain endothelial cellsconnected by tight junctions. The tight junctions limit paracellularpermeability, thus directly restricting the diffusional transport ofhydrophilic substances and the entry of proteins and immune

cells into the brain. Small lipophilic molecules may cross the BBBby transcellular passive diffusion, but the efficiency of this processis dependent ondrug physicochemical properties (2). In addition,the BBB expresses a broad range of efflux and uptake transportersthat actively modulate the transport of substrate drugs out of andinto the brain (3, 4).

The pharmacokinetics of a drug in the brain, including thetarget site (e.g., brain tumor), are determined bymultiple dynam-ic, interactive processes governed by both drug physicochemicalproperties (e.g., molecular weight, lipophilicity, charge, bindingto plasma proteins/brain tissues, and affinity to uptake/effluxtransporters) and brain pathophysiology (e.g., BBB integrity,transporter abundance/activity, cerebral blood flow, and cerebro-spinal fluid bulk flow; refs. 5–7). Mechanistic understanding ofthe role of these interactive processes in drug BBB penetration andearly, accurate prediction of drug exposure in the brain or braintumors are of paramount importance to guide the rational drugdevelopment and treatment for brain cancer. However, the phar-macokinetics of many new and existing anticancer agents inhuman brain or brain tumors remain understudied, underre-ported, and misunderstood because of the difficulty in accessinghuman brain specimens and the lack of in vitro assays or animalmodels that reliably predict humanBBBpermeability. Clearly, it isimperative that novel translational approaches are developed

1Karmanos Cancer Institute, Wayne State University School of Medicine, Detroit,Michigan. 2Barrow Neurological Institute, St. Joseph's Hospital and MedicalCenter, Phoenix, Arizona. 3College of Pharmacy & Comprehensive CancerCenter, Ohio State University, Columbus, Ohio.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Authors: Jing Li, Karmanos Cancer Institute, Wayne StateUniversity, 4100 John R, HWCRC, Room 523, Detroit, MI 48201. Phone: 313-576-8258; Fax: 313-576-8828; E-mail: [email protected]; and Nader Sanai,Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, 350West Thomas Road, Phoenix, AZ 85013. Phone: 602-406-5096; E-mail:[email protected]

doi: 10.1158/1078-0432.CCR-17-0983

�2017 American Association for Cancer Research.

ClinicalCancerResearch

Clin Cancer Res; 23(24) December 15, 20177454

on May 7, 2021. © 2017 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 19, 2017; DOI: 10.1158/1078-0432.CCR-17-0983

http://crossmark.crossref.org/dialog/?doi=10.1158/1078-0432.CCR-17-0983&domain=pdf&date_stamp=2017-11-22

http://clincancerres.aacrjournals.org/

to better understand drug penetration across human BBB. Onesuch approach is the in vitro–in vivo extrapolation (IVIVE), phys-iologically based pharmacokinetic (PBPK) modeling that allowsintegration of drug-specific parameters (e.g., in vitro enzyme/transporter kinetic data) and system-specific parameters (e.g.,brain pathophysiology) into a mechanistic pharmacokineticmodel to predict the systemic and brain pharmacokinetics of adrug. Because PBPK models can delineate the fractional role ofindividual disposition pathways, they also provide an invaluabletool for mechanistically and quantitatively assessing the sourcesof intraindividual, interindividual, or interspecies variability inplasma/brain pharmacokinetics.

AZD1775 is a first-in-class, highly selective, potent, ATP-competitive, small-molecule inhibitor of the Wee1 tyrosinekinase. It shows an enzyme IC50 of 5.2 nmol/L (2.6 ng/mL) forinhibiting Wee1 activity and an IC50 of �80 nmol/L (40 ng/mL)for inducingDNAdamage andG2-checkpoint escape in cell-basedassays (https://ncats.nih.gov/files/AZD1775.pdf). The Wee1kinase is a critical regulator of the S and G2–M checkpoint (8).Wee1 phosphorylates the Tyr 15 residue of the cyclin-dependentkinase 1 (CDK1) and inhibits its activity. Inhibition of Wee1thereby activates CDK1, facilitating cells to enter themitotic phase(9, 10). Additionally, inhibition of Wee1 causes aberrantly highCDK2 activity in S-phase cells, leading to unstable DNA replica-tion structures and, ultimately, DNA damage (11). Thus, Wee1inhibition offers a chemo- and radiosensitizing strategy by forcingcancer cells with damagedDNA to enter unscheduledmitosis andto undergo cell death (12). Because the tumor suppressor proteinp53 regulates the G1 checkpoint, cancer cells harboring abnor-malities in the p53 pathway becomemore dependent onG2 and Scheckpoints in cell-cycle control (9, 13), and, therefore, p53-deficient cancer cells treated with Wee1 inhibitors would beparticularly susceptible to DNA damage induction due to geneticand pharmacologic abrogation of both G1 and G2 checkpoints(14). Given the high frequency of p53 mutations and the pivotal

role of the G2–M checkpoint in preventing the programmed celldeath in glioblastoma (14), AZD1775 is under evaluation as apotential chemo- and radiosensitizer for the treatment of glio-blastoma, themost commonmalignant primary brain cancer thatuniversally develops resistance to radiation and chemotherapyand unrelentingly results in mortality.

As the first step in the clinical development, we prospectivelyassessed the tumor penetration of AZD1775 in patients withglioblastoma in a phase 0 clinical trial. Furthermore, leveragingin vitro studies and clinical data,wedeveloped awhole-bodyPBPKmodel integrated with a four-compartment permeability-limitedbrain (4Brain) model to predict the extent and rate of AZD1775penetration across the human BBB and assess the factors modu-lating this process.

MethodsClinical pharmacokinetic study

The plasma and tumor pharmacokinetics of AZD1775 wereevaluated in 20patientswithfirst-recurrence, denovo glioblastomain a phase 0 trial (ClinicalTrials.gov identifier: NCT02207010).The protocol was approved by the Institutional Review Board ofBarrow Neurological Institute, St. Joseph's Hospital and MedicalCenter (Phoenix, AZ), and written informed consent wasobtained from each patient. Patient characteristics (shown asthe median and range) include age (59, 28–81 years), weight(79, 54–127 kg), height (172, 154–184 cm), body surface area(1.9, 1.52–2.43 m2), total bilirubin (0.5, 0.2–1.7 mg/dL), aspar-tate aminotransferase (16, 8–62 IU/L), alanine aminotransferase(24, 14–133 IU/L), serum albumin (4.0, 2.1–4.6 mg/dL), andserum creatine (0.8, 0.6–1.0 mg/dL).

AZD1775 was provided by AstraZeneca (Wilmington, DE) as100-mg gelatin capsules. Patients were treated with a single oraldose of AZD1775 at the dose of 100 (4 patients), 200 (4 patients),or 400 mg (12 patients) prior to the surgical tumor resection.Patients receiving a dose of 100 or 200 mg had their tumorresection at 8 to 10 hours following dosing. Patients receivinga dose of 400mg had their tumor resection at 2 to 4, 8 to 10, or 24to 26 hours after dosing. For patients with the tumor resectionscheduled at 2 to 4 hours following dosing, blood samples werecollected at before dosing and at 2 to 4 (at the time of tumorresection), 8, and 24 hours after dosing. For patients with thetumor resection scheduled at 8 to 10 hours following dosing,blood samples were collected at before dosing and at 8 to 10 (atthe timeof tumor resection), 12, and16hours after dosing. For thepatients with the tumor resection scheduled at 24 to 26 hoursfollowing dosing, blood samples were collected at before dosingand at 2, 8, and 24 to 26 hours (at the time of tumor resection)after dosing. The total and unbound concentrations of AZD1775in plasma and tumor samples were determined using a validatedLC/MS-MS method (15).

In vitro studiesTo allow IVIVE in the PBPK modeling, human-derived in vitro

systems were used to determine AZD1775 drug-specific para-meters, including (i) in vitro metabolic intrinsic clearance (CLint)by human liver and intestinal microsomes; (ii) apparent trans-cellular passive permeability in the apical-to-basolateral (Papp,A-B)and basolateral-to-apical (Papp,B-A) directions across MDCKII cellmonolayers; (iii) interactions with efflux transporters ABCB1 andABCG2usingMDCKII cells with overexpression of humanABCB1

Translational Relevance

Early, accurate prediction andmechanistic understanding ofpenetration of anticancer drugs across the humanblood–brainbarrier (BBB) are of paramount importance to the rationaldrug development and therapy for brain cancer. In this study,using an integrated, quantitative clinical pharmacologyapproach that leverages clinical trial, in vitro–in vivo extrapo-lation (IVIVE), and physiologically based pharmacokinetic(PBPK) modeling, we provided the first clinical evidence ofgood tumor penetration of AZD1775 in patients with glio-blastoma, and we further elucidated the underlying mecha-nism. Our study provided quantitative and mechanisticinsights into the heterogeneous brain/brain tumor penetra-tion of AZD1775 to inform decision-making regarding furtherclinical development. The developed mechanistic PBPK mod-el, verified by clinical observed data, can be generalized toother drugs that undergo similar disposition pathways. Thisstudy composes a framework for prospective, quantitative,and mechanistic investigation of the penetration of novelanticancer drugs across the human BBB in early-phase clinicaldevelopment.

PBPK Modeling of AZD1775 Penetration into Brain Tumors

www.aacrjournals.org Clin Cancer Res; 23(24) December 15, 2017 7455



https://ncats.nih.gov/files/AZD1775.pdf


or ABCG2 (named MDCKII-ABCB1 and MDCKII-ABCG2 cells);(iv) interactions with major BBB uptake transporters usingHEK293 cells with overexpression of human OATP1A2,OATP2B1, or OAT3; and (v) fraction unbound in plasma (fu,p)and fraction unbound in brain tissue (fu,br) using patient plasmaandbrain tumor specimens. The details for in vitro experiments arepresented in Supplementary Methods.

PBPK modeling and simulationSimcyp Simulator V16 (Simcyp Ltd.) was used for PBPK

modeling and simulations. Awhole-body PBPKmodel integrated

with a 4Brainmodel as implemented in the Simcyp Simulatorwasdeveloped to predict the plasma and brain concentration-timeprofiles of total and unbound AZD1775. Themodel structure andassumptions are illustrated in Fig. 1 (16). System-specific para-meters were derived based on the existing virtual Caucasianpopulation within the Simcyp Simulator unless stated otherwise.Details on AZD1775 drug-specific parameters for the whole-bodyPBPK model and 4Brain model are presented in Table 1. In brief,physicochemical parameters were obtained from literature. Theoral absorption was predicted using the Simcyp AdvancedDissolution, Absorption, and Metabolism model, where the

Figure 1.

A, Schematic illustration of the BBB,blood–cerebrospinal fluid (CSF) barrier,and brain–CSF barrier in human brain.Drug transport across the BBB (1) wasgoverned by the passive bidirectionalpermeability and transporter-mediatedactive efflux and active uptake. Drugtransport across the blood–CSF barrier(2) was controlled by the passivebidirectional permeability andtransporter-mediated active efflux andactive uptake. Drug transport acrossthe brain–CSF barrier (3) was governedby the passive bidirectionalpermeability only. B, The structure ofthe 4Brain model as implemented inSimcyp Simulator V16. The passivebidirectional permeability across theBBB, blood–CSF barrier, and brain–CSFbarrier are parameterized as PSB, PSC,and PSE, respectively. Active efflux anduptake at the BBB are parameterized asCLefflux,BBB and CLuptake,BBB,respectively. Active efflux anduptake atthe blood–CSF barrier areparameterized as CLefflux,CSF andCLuptake,CSF, respectively. The modelassumes that (i) only unbound andunionized drug can passively passthrough all barriers, whereastransporters act upon unbound drug(including both unionized and ionizedspecies); (ii) fluid balance ismaintainedby the circulation of CSF between spinaland cranial compartments andreabsorbed into the brain blood. Flowrates are described by the CSFsecretion rate (QBCSFB), bulk flow ratefrom brain mass to cranial CSF (Qbulk),CSF flow rate out of cranial and spinalcompartments (Qsink), CSF shuttle flowrate between cranial and spinalcompartments (QSin and QSout), watertransfer rate from the brain blood tobrain mass (QBBB); (iii) the cerebralblood flow rate (QBrain) links the brainmodel to the whole-body PBPK model;(iv) metabolism in brain mass isnegligible; and (v) all compartments arewell stirred with defined volumes.

Li et al.

Clin Cancer Res; 23(24) December 15, 2017 Clinical Cancer Research7456




dissolution rate was estimated using the built-in diffusion layermodel and the human intestinal effective permeability was esti-mated by IVIVE scaling from the passive permeability determinedfromMDCKII cell monolayer. The drug distribution to all organs/tissues except for brainwas perfusion rate limited. The hepatic andintestinal metabolic clearance was estimated by IVIVE scaling ofthe in vitrometabolic intrinsic clearance determined from human

liver and intestinal microsomes, respectively. The renal clearancewas assigned based on sensitivity analysis. Drug-dependent para-meters for the 4Brain model were derived based on experimen-tally determined in vitro data on passive permeability and inter-action with efflux and uptake transporter, as described below:

The passive permeability-surface area product on the BBB (PSB)was predicted using Equation (Eq.) 1 (17), where Papp,A!B is the

Table 1. AZD1775 drug-specific parameters for the whole-body PBPK model integrated with a 4Brain model

Parameter Value Comments/reference

PhysicochemicalMW (g/mol) 500.6 ChEMBL databaseLogP 3.37 ChEMBL databasePKa (monoprotic base) 6.0 Investigator brochureB/P 1.29 Investigator brochurefu,p 0.2 Experimental determined from pooled human plasma

Absorption: Advanced Dissolution, Absorption, and Metabolism (ADAM) modelMDCKII Papp (10�6 cm/s) for AZD1775 20 Experimental determined from MDCKII cell monolayersMDCKII Papp (10�6 cm/s) for midazolam 62 Experimental determined from MDCKII cell monolayersLag time (h) 0 Observedfugut 0.7 Experimental determined from human liver microsomesQgut (L/h) 10.2 Simcyp predictedImmediate release—diffusion layer modelIntrinsic aqueous solubility (mg/mL) 0.034 Simcyp predictedMonodispersed radius (mm) 10 Assigned

Distribution: full PBPKVss (L/kg) 19.2 Predicted by Rodgers and Roland method (Method 2), Kp scalar of 1

Elimination: whole-organ metabolic clearanceHuman liver microsome (HLM):Vmax (pmol/min/mg) 197 Experimental determined from HLMKm (mmol/L) 4.94 Experimental determined from HLMfuinc 0.7 Experimental determined from HLM

Human intestinal microsome (HIM):CLint (mL/min) 17 Experimental determined from HIMfumic 0.7 Experimental determined from HIM

CLR (L/h) 6 Optimized based on sensitivity analysis4Brain modelBBBPSB (L/h) 12.6 IVIVE scaling of MDCKII Papp,A!B using Eq. 1a

fu,br 0.07 Experimental determined from patient brain tumor tissueCLABCB1,vitro (mL/min/mg) 152 and 43 Estimated using Eq. 2 at the basolateral pH 7.4 and 6.5, respectivelyb

CLABCG2,vitro (mL/min/mg) 76 and 33 Estimated using Eq. 2 at the basolateral pH 7.4 and 6.5, respectivelyb

RAF for ABCB1 295 Estimated using Eq. 3c

RAF for ABCG2 651 Estimated using Eq. 3c

CLuptake,vivo (L/h) 44.0 Assigned based on sensitivity analysisBlood–cranial CSF barrierd:PSC (L/h) 6.3 Assumed to be half of PSBfucsf 1 Assigned based on low protein concentration in CSF

Brain–cranial CSF barrier:PSE (L/h) 300 Assigned given the high permeability of this barrier

Abbreviations: B/P, blood-to-plasma partition ratio; CLABCB1,vitro and CLABCB1,vitro, ABCB1- and ABCG2-mediated in vitro efflux clearance, respectively; CLint, in vitrointrinsic metabolic clearance; CLR, renal clearance; CLuptake,vitro, uptake transporter-mediated in vitro clearance; CSF, cerebrospinal fluid; Eq., equation; fu,br, fractionunbound drug in brain tissue; fu,p, fraction of unbound drug in plasma; fucsf, fraction unbound drug inCSF; fugut, fraction of unbounddrug in enterocytes; fumic, fractionof unbound drug in in vitromicrosomal incubation; Km, substrate concentration at which half of Vmax is achieved; logP, logarithm of the neutral species octanol-to-buffer partition ratio; MW,molecule weight; Papp, apparent passive permeability; PKa, acid dissociation constant; PSB, passive permeability-surface area product onthe BBB; Qgut, gut blood flow; RAF, in vivo–in vitro relative activity factor; Vmax, maximum metabolic rate; Vss, volume of distribution at steady-state using tissuevolumes for a population representative of healthy volunteers population.aPSB ¼ Papp,A!B � SA (Eq. 1), where Papp,A!B is the apparent permeability determined from MDCKII cell monolayer (mean, 17.5 � 10�6 cm/s) and SA is the humanbrain microvasculature surface area (mean, 20 m2).bCLefflux;vitro ¼ 2 � ðER � 1Þ � Papp;A�B � SA

Procell(Eq. 2), where CLefflux,vitro (mL/min/mg) is the efflux transporter-mediated in vitro clearance; ER is the efflux ratio determined

from MDCKII-ABCB1 or MDCKII–ABCG2; Papp,A-B is the apparent passive permeability determined from MDCKII; SA is the filter surface area (0.143 cm2) in a 96-welltranswell; and Procell is the protein amount of MDCKII-ABCB1 or –ABCG2 cells in a 96-well transwell.cRAF ¼ Abundance in vivo

Abundance in vitro � BMvPGB� BW (Eq. 3), where BMvPGB is the milligrams of brain microvessels per gram brain; BW is the average human brain weight;

abundance in vivo or in vitro represents the ABCB1/ABCG2 transporter protein expression level in human brain microvessels or in MDCKII-ABCB1 and -ABCG2 cells,respectively.dTransporter-mediated clearance at the blood–cranial CSF barrier was not incorporated in the 4Brainmodel because little information is available on the transportersat this barrier, and also because the drug disposition in CSF was not the focus of this study.






apparent permeability determined from MDCKII cell monolayerand SA is the human brainmicrovasculature surface area (SA; 15–25 m2).

PSB ¼ Papp;A!B � SA ð1ÞThe passive permeability-surface area product on the blood–

cranial cerebrospinal fluid (CSF) barrier (PSC) was assumed to behalf of the PSB (16). The passive permeability-surface area prod-uct on the brain–CSF barrier (PSE) was fixed at 300 L/h, assuminga high permeability of this barrier (16).

In vitro efflux transporter-mediated clearance (CLefflux,vitro)(mL/min/mg) was estimated using Eq. 2 (18), where ER is theefflux ratio determined from MDCKII-ABCB1 or -ABCG2 cells,Papp,A-B is the apparent passive permeability (cm/s) determinedfromMDCKII, SA is the filter surface area (0.143 cm2) in a 96-welltranswell, and Procell is the protein amount (mg) of MDCKII-ABCB1 (mean, 0.019mg/well) or -ABCG2 cells (mean, 0.014mg/well) in a 96-well transwell.

CLefflux;vitro ¼ 2 � ER � 1ð Þ � Papp;A�B � SAProcell

ð2Þ

CLefflux,vitro was scaled to the whole-brain in vivo efflux trans-porter-mediated clearance (CLefflux,vivo; mL/min/mg) by multiply-ing a relative activity factor (RAF; Eq. 3).

CLefflux;vivo ¼ CLefflux;vitro � RAF

¼ CLefflux;vitro � Abundance in vivoAbundance in vitro

� BMvPGB� BW

ð3Þ

where abundance in vivo or in vitro represents the ABCB1/ABCG2transporter protein expression level in brain microvessels (pmol/mg microvessels) or in cellular models (pmol/mg cells), respec-tively; BMvPGB is the milligrams of brain microvessels per grambrain, and BW is the brain weight (gram). Given an averageBMvPGB of 0.244 mg protein/g brain (19), average brain weightof 1400 g, ABCB1 abundance in vivo of 6.06 pmol/mg and ABCG2abundance in vivo of 8.14 pmol/mg in human brain microvessels(19), and ABCB1 abundance in vitro of 7.02 pmol/mg inMDCKII-ABCB1 cells and ABCG2 abundance in vitro of 4.27 pmol/mg inMDCKII-ABCG2 cells (our unpublished data), the RAF was esti-mated to be 294 for ABCB1 and 651 for ABCG2.

In vivo uptake transporter-mediated clearance (CLuptake,vivo) atthe BBB was assigned based on sensitivity analysis. Efflux anduptake transporter-mediated clearance at the blood–CSF barrierwas not incorporated in the 4Brain model given the little infor-mation available on the transporters at this barrier, and alsobecause the drug disposition in CSF was not the focus of thisstudy.

Simulations of 10 trials with 10 subjects in each trial wereperformed in an existing virtual Caucasianpopulation following asingle oral dose of 400 mg. The PBPK model was verified bycomparing the predicted total and unbound plasma and brainconcentration–time profiles with observed data in patients withglioblastoma.

To better understand the sources of the variability in drug brainpharmacokinetics, sensitivity analyses using the developed PBPKmodel were performed to examine the impact of BBB integrity (asassessed by PSB), transporter activity (as assessed by CLuptake,vivoand CLefflux,vivo), and drug binding to brain tissues (fu,br) on the

extent and rate of BBB penetration. The extent of drug BBBpenetration is often assessed by the total or unbound drugbrain-to-plasma partition coefficient (Kp or Kp,uu), which can beestimated as the brain-to-plasma drug concentration ratio at thesteady state (or brain equilibrium) or area under concentration–time curve (AUC) ratio. Given the notion that unbound drugconcentration drives the in vivopharmacologic effect, the use ofKp,

uu as a measure of the extent of brain penetration is morepharmacologically relevant and therefore was used in this study.Analogous to the concept of drug oral absorption, the rate of BBBpenetration was assessed by the time (Tmax,br) to achieve themaximum drug brain concentration (Cmax,br).

Statistical analysisComparisons of the cellular uptake of AZD1775 between the

cell line overexpressing a particular uptake transporter and itsvector control or between the presence and absence of a trans-porter inhibitor were performed using two-sided independentsamples t test. A P value < 0.05 was regarded as statisticallysignificant.

ResultsAZD1775 tumor penetration in patients with glioblastoma

Figure 2 shows the brain tumor concentrations of total andunbound AZD1775 as well as the extent of drug tumor penetra-tion (as assessed by the Kp and Kp,uu) in patients with glioblas-toma receiving a single oral dose of AZD1775 (100, 200, or 400mg). The unbound (pharmacologically active) AZD1775 tumorconcentrations varied from6 to315ng/g (median, 24ng/g) acrossthree dose levels and three sampling time points, which exceededthe enzyme IC50 (2.6 ng/mL) for inhibition of Wee1 activity. Theextent of AZD1775 tumor penetration (Kp or Kp,uu) appearedindependent of the dose and sampling times. There was largeinterindividual variability in AZD1775 tumor penetration inpatients with glioblastoma. Overall, the Kp varied 10.6-fold(median, 9.1; range, 3.8–40.4) and the Kp,uu varied 19-fold(median, 3.2; range, 1.3–24.4) in 20 patients.

In vitro studiesIn vitrometabolism. The overallmetabolic profiles of AZD1775 inhuman liver and intestinal microsomes were well described byMichaelis–Menten kinetics (Supplementary Fig. S1). The estimat-ed in vitrometabolic kinetic parameterswere used for predictionofwhole-organ metabolic clearance of the liver and intestine in thePBPK model (Table 1).

Passive transcellular permeability and interaction with efflux trans-porters. The apparent permeability and efflux ratios for AZD1775and the positive controls, loperamide (a typical substrate ofABCB1) and gefitinib (a typical substrate of ABCG2), onMDCKII,MDCKII-ABCB1, and MDCKII-ABCG2 cell monolayers are sum-marized in Fig. 3A. The positive controls with loperamide andgefitinib confirmed the functional expression of ABCB1 andABCG2 inMDCKII-ABCB1 andMDCKII-ABCG2 cell lines, respec-tively. The mass recovery for AZD1775, loperamide, and gefitinibfrom all permeability experiments was within 85% to 115%.AZD1775 was highly permeable, with a mean passive permeabil-ity of 17.5� 10�6 cm/s in the apical to basolateral direction acrossMDCKII monolayer. It exhibited an efflux ratio of 10.6 and 4.5across MDCKII-ABCB1 and MDCKII-ABCG2 monolayers,

Li et al.





respectively; the presence of an inhibitor of ABCB1 (cyclosporineA, 10 mmol/L) or ABCG2 (Ko143, 1 mmol/L) abrogated the effluxeffect (Fig. 3A). These data collectively indicated that AZD1775was a substrate of ABCB1 and ABCG2.

Notably, AZD1775 exhibited pH-dependent passive perme-ability and active efflux across MDCKII, MDCKII-ABCB1, andMDCKII-ABCG2 cell monolayers (Fig. 3B). At a fixed pH (7.4) inthe apical chamber (mimicking blood circulation) and withvarying pH (7.4, 7.0, 6.5, or 6.0) in the basolateral chamber(mimicking brain interstitium), the passive permeability ofAZD1775 across MDCKII decreased in the basolateral to apicaldirection while increasing in the apical to basolateral direction,thereby resulting in anet influx at a relatively acidic basolateral pH(e.g., efflux ratio of 0.8 at pH 6.5; Fig. 3B). This observation was inaccordance with the pH partition theory, that is, for weak base oracid drugs, the extent to which they are ionized is exponentiallyrelated to the pH of their milieu. As ionization of a moleculesubstantially decreases its lipophilicity, slight differences in pHmay markedly influence the ability of the drug to traverse acrossbiological membranes. At a relative acidic basolateral pH, the

ionization of AZD1775, a weak base drug, was increased, and as aresult, the basolateral-to-apical permeability decreased and thedrug was trapped in the basolateral compartment. Similar find-ings were also observed on MDCKII-ABCB1 or MDCKII-ABCG2monolayers, and of particular note, efflux ratios were markedlyreduced at a relatively acidic basolateral pH (Fig. 3B). For exam-ple, as thebasolateral pHdecreased from7.4 to 6.5, the efflux ratiowas reduced by 60% (from 9.2 to 3.7) onMDCKII-ABCB1 and by45% (from 4.7 to 2.6) onMDCKII-ABCG2. Because the transportof AZD1775 across MDCKII-ABCB1 andMDCKII-ABCG2mono-layers occurred via both passive diffusion and active efflux, theobserved reduction in efflux ratios at a relatively acidic basolateralpHwas likely attributable to the decrease of both passive outwarddiffusion rate and active efflux rate. Since the transporter expres-sion level or activity of ABCB1 and ABCG2 was not regulated bypH changes (20–22), it is plausible that the decreased active effluxratewas due to less drug remainingwithin the apicalmembrane ofcell monolayer to interact with efflux transporters at a relativelyacidic basolateral pH, a condition favorable to trapping a weakbase drug into the basolateral compartment.

Figure 2.

Observed brain tumor pharmacokinetics of AZD1775 in 20 patients with glioblastoma following a single oral dose (100, 200, or 400 mg). A, Total AZD1775tumor concentrations. B, Unbound AZD1775 tumor concentrations. C, Total AZD1775 tumor-to-plasma concentration ratios. D, Unbound AZD1775 tumor-to-plasmaconcentration ratios. Symbols represent observed data from individual patients, and lines represent median values.






Interaction with uptake transporters. The cellular uptake ofAZD1775 and the positive control, estrone-3-sulfate (a typicalsubstrate for OATP1A2, OATP2B1, and OAT3), in HEK293 cellswith overexpression of OATP1A2, OATP2B1, or OAT3 and theirrespective vector control cells is summarized in Fig. 4. The cellularuptake of estrone-3-sulfate was significantly higher in HEK293cells overexpressing OATP1A2, OATP2B1, or OAT3 comparedwith their respective vector control (P < 0.001), and the presenceof a specific transporter inhibitor (i.e., 100 mmol/L sulfobro-mophthalein for OATP1A2, 10 mmol/L fluvastatin A forOATP2B1, or 50 mmol/L probenecid for OAT3) reducedestrone-3-sulfate uptake to a level similar to that in the vectorcontrol (Fig. 4). These data confirmed the functional activity of thestudied transporters in the present in vitro systems. The cellularuptake of AZD1775 in cells overexpressing OATP1A2 was signif-icantly higher than that in the vector control (P¼ 0.001), and thepresence of an OATP1A2 inhibitor (i.e., 100 mmol/L sulfobro-mophthalein) reduced the drug uptake to the level similar to thatin the vector control (P ¼ 0.005; Fig. 4A), collectively suggestingthat AZD1775 was a substrate of OATP1A2. On the contrary, the

cellular uptake of AZD1775 was similar between the cells over-expressing OATP2B1 or OAT3 and their respective vector controlor between the absence and presence of a specific transporterinhibitor (Fig. 4B and C), suggesting that AZD1775 was not asubstrate of OATP2B1 or OAT3.

Binding to plasma proteins and brain tissue. In 20 patients withglioblastoma, AZD1775 exhibited a higher binding to braintissues (fu,br: 0.067 � 0.013; range, 0.042–0.10) than to plasmaproteins (fu,p: 0.182 � 0.024; range, 0.15–0.24).

PBPK modeling and simulationPBPK modeling. A whole-body PBPK model integrated with a4Brain model (Fig. 1) was developed to predict the kinetics ofAZD1775 penetration across the BBB.Drug penetration across theBBB was controlled by the passive permeability, active efflux, andactive uptake at the BBB. Given an average passive permeability of17.5 � 10�6 cm/s determined from MDCKII cell monolayerand an average human brain microvascular surface area of20 m2, the passive diffusion clearance (PSB) at the BBB was

Figure 3.

Apparent passive permeability in the apical-to-basolateral (Papp,A-B) and basolateral-to-apical (Papp,B-A) directions and efflux ratio (ER) across MDCKII,MDCKII-ABCB1, and MDCKII-ABCG2 cell monolayers. A, The Papp,A-B, Papp,B-A, and ER of AZD1775 (5 mmol/L) and the positive controls, loperamide (a typicalsubstrate of ABCB1, 5 mmol/L) and gefitinib (a typical substrate of ABCG2, 1 mmol/L), in the absence or presence of an inhibitor of ABCB1 (cyclosporine A, 10 mmol/L)or ABCG2 (Ko143, 1 mmol/L). B, The Papp,A-B, Papp,B-A, and ER of AZD1775 (5 mmol/L) at a fixed apical pH (7.4) and varying basolateral pH (7.4, 7.0, 6.5,and 6.0). Data are expressed as the mean from three independent experiments (triplicate in each experiment), with the relative SD <20%.

Li et al.





estimated to be 12.6 L/h (Eq. 1 and Table 1). Notably, AZD1775exhibited pH-dependent efflux across MDCKII-ABCB1 orMDCKII-ABCG2 cell monolayers (Fig. 3B). As the basolateral pHvaried from 7.4 to 6.5, ABCB1-mediated in vitro clearancedecreased from 152 to 43 mL/min/mg, and similarly ABCG2-mediated in vitro clearance decreased from 76 to 33 mL/min/mg.Therefore, using the IVIVE scaling strategy based on the relativein vivo–in vitro activity factor of 295 for ABCB1 and 651 forABCG2 (Eq. 3), it was estimated that ABCB1- and ABCG2-mediated efflux clearance at the BBB would reduce from 5.7 to2.1 L/h as the brain interstitial pH decreased from 7.4 to 6.5.

By incorporating a passive diffusion clearance of 12.6 L/h andactive efflux clearance of 5.7 or 2.1 L/h at the BBB, the PBPKmodelpredicted the Kp,uu to be 0.78 and 0.92, respectively, at the braininterstitial pH of 7.4 and 6.5. Further sensitivity analysis withvarying passive diffusion clearance (1.26–126 L/h) and activeefflux clearance (0.57–57 L/h) suggested that the Kp,uu rangedfrom 0.11 to 1.03 at two extreme scenarios. These results sug-gested that active uptakemust be involved in the BBB penetrationof AZD1775 to achieve the observed unbound tumor-to-plasmaratios (median, 3.2; range, 1.6–24.4) in patients with glioblasto-ma. Subsequent sensitivity analysis with fixed passive diffusionclearance (12.6 L/h) and active efflux clearance (2.1 L/h) indicatedtheKp,uu increased from1.2 to 8.0 as the active uptake clearance attheBBB increased from2.1 to105L/h. Therefore, by incorporatinga passive clearance of 12.6 L/h, active efflux clearance of 2.1 L/h,and active uptake clearance of 44.0 L/h at the BBB, the final PBPKmodel predicted theKp,uu to be 3.2, which was in good agreementwith clinically observed data.

The developed PBPK model adequately predicted the plasmaand brain concentration–time profiles of total and unboundAZD1775 in patients with glioblastoma. As shown in Fig. 5, themodel-predicted total and unbound AZD1775 plasma andbrain concentration–time profiles following a single oraldose of 400 mg in a virtual Caucasian population (n ¼ 100)well recovered observed concentration data in patients withglioblastoma, with �97% of observed data falling between the5th and 95th percentile of the predicted mean concentration–time profile.

Factors modulating drug BBB penetration. The impact of BBBintegrity (assessed by passive diffusion clearance), transporteractivity (assessed by active efflux/uptake clearance), and drugbinding to brain tissue on the extent and rate of drug BBBpenetration is illustrated in Fig. 6 and Supplementary Fig. S2.The impact of BBB integrity on the extent of BBB penetration wasmodulated by active transport process. When insignificant activeefflux or uptake transport was present, variations in BBB integrityhad a negligible influence on the extent of penetration, as indi-cated by a constant Kp,uu (�1.0) when the passive diffusionclearance varied 100-fold (1.26–126 L/h; Fig. 6A and B; Supple-mentary Fig. S2A and S2B). When active efflux was dominant,compromised BBB integrity (i.e., increasing passive diffusionclearance) led to a progressive increase in BBB penetration (Fig.6A and Supplementary Fig. S2A), whereas when active uptakewasdominant, compromised BBB integrity led to a progressivedecrease in BBB penetration (Fig. 6B and Supplementary Fig.S2B). However, irrespective of whether active efflux or uptake ispresent, compromised BBB integrity generally resulted inquicker BBB penetration (i.e., shorter time to reach themaximum

Figure 4.

Cellular uptake of AZD1775 and the positive control, estrone-3-sulfate (E3Sul; atypical substrate for OATP1A2, OATP2B1, andOAT3), in the absence or presenceof a specific transporter inhibitor [i.e., 100 mmol/L sulfobromophthalein (Sul) forOATP1A2, 10 mmol/L fluvastatin A (Flu) for OATP2B1, or 50 mmol/L probenecid(Pro) for OAT3] in HEK293 cells with overexpression of (A) OATP1A2, (B)OATP2B1, or (C) OAT3 and their respective vector control cells. Data areexpressed as the mean � SD from three independent experiments, withtriplicate in each experiment. �� , P < 0.01; �� , P < 0.001.






drug concentration in the brain; Fig. 6C and D; SupplementaryFig. S2C and S2D).

Increasing active efflux clearance led to a decrease in the extentof BBB penetration, whereas the opposite was observed as activeuptake clearance increased (Fig. 6A and B). Of note, the degree ofthe impact of active transporter activity on drug BBB penetrationwas dependent on the magnitude of passive permeability. Spe-cifically, the extent of BBB penetration (i.e., Kp,uu) was moresensitive to the change of active transporter activity when thepassive permeability was relatively lower (Fig. 6A and B; Supple-mentary Fig. S2E and S2F). The rate of BBB penetration wasinsensitive to the change of active uptake, whereas quicker BBBpenetrationwas expected as active efflux increased (Fig. 6C andD;Supplementary Fig. S2G and S2H).

Irrespective of BBB integrity or transporter activity, variations indrugbinding tobrain tissue influenced the rate of BBBpenetrationand total drug brain exposure but did not influence the unbounddrug brain exposure. Specifically, the decrease of drug binding tobrain tissue resulted in quicker BBB penetration while having noinfluence on the extent of BBB penetration (i.e., Kp,uu; Supple-mentary Fig. S2I and S2J).

DiscussionThis study shows that the Wee1 kinase inhibitor AZD1775

extensively accumulates in human glioblastoma tumor tissue andachieves pharmacologically active and potentially therapeuticconcentrations in patients. This is unexpected for a dual substrateof ABCB1 and ABCG2, two main efflux transporters expressed athuman BBB to restrict the penetration of substrate drugs into thebrain (4). Our in vitro cellular studies suggest that the ABCB1/ABCG2–mediated efflux clearance of AZD1775 is significantlyreduced at a relatively acidic basolateral pH (whichmimics acidicbrain tumor microenvironment). In addition, AZD1775 is asubstrate of OATP1A2, an uptake transporter expressed at humanBBB to facilitate the transport of substrate drugs from the blood tobrain (3, 24). Our mechanistic PBPKmodel further indicates thatthe extent of AZD1775 penetration into brain tumors is mainlydependent on the relative transporting efficiency of ABCB1/ABCG2 and OATP1A2. In acidic tumor microenvironment whereABCB1/ABCG2–mediated active efflux clearance is reduced,OATP1A2-mediated active uptake becomes dominant to driveAZD1775 penetration into brain tumors. These findings not only

Figure 5.

Physiologically based pharmacokinetic modeling. A and B, Model-predicted and clinically observed AZD1775 total and unbound plasma concentration–timeprofiles. C and D, Model-predicted and clinically observed AZD1775 total and unbound brain tumor concentration–time profiles. Simulations of 10 virtual trialswith 10 subjects in each were performed in a virtual Caucasian population following a single oral dose of 400 mg. Observed data were obtained from12 Caucasian patients with glioblastoma receiving a single oral dose of 400 mg. The symbols represent observed data. The thick black line represents overallmean concentration–time profile for the virtual population (n ¼ 100); dotted and dashed lines represent the 5th and 95th percentiles of the mean concentration,respectively.

Li et al.





have direct clinical relevance for the treatment of glioblastomaswith AZD1775 but also emphasize the need to consider theunique tumor microenvironment and transporter-mediatedactive uptake mechanisms in the evaluation of brain/tumorpharmacokinetics of agents in the treatment of brain cancer. Inaddition, the PBPKmodel we developed offers amechanistic toolfor the prediction of plasma and brain tumor pharmacokinetics,and provides the basis for the design of future prospective inves-tigations for evaluating alternative and improveddosing regimensof AZD1775 and related agents.

Although the present data support further clinical developmentof AZD1775 for treating glioblastoma, it is noteworthy that ourclinical observations are at odds with preclinical findings in anorthotopic glioblastoma xenograft mouse model, which indicat-ed a limited distribution of AZD1775 inmouse normal brain andtumor tissues (23). Although the exact underlying mechanism isyet to be determined, the observed interspecies variability in

AZD1775 BBB penetration underscores the critical value of earlyprospective evaluation of drug penetration across human BBB toinform decision-making regarding future clinical developmentand trial design.

AZD1775 belongs to the Biopharmaceutics Classification Sys-tem (BCS) class II compound characterized by high permeabilityand low solubility. It is a good substrate of ABCB1 and ABCG2(Fig. 3), twomajor efflux transporters that are highly expressed inthe apical side of the BBB responsible for extruding substrate drugsfrom the brain to blood (4). Surprisingly, the Kp,uu of AZD1775ranges from 1.3 to 24.4 (median, 3.2) in patients with glioblas-toma. The PBPKmodeling and simulation suggest that to achievethis degree of brain tumor penetration (i.e.,Kp,uu >1), a dominantactive drug uptake has to present at the BBB. Several drug uptaketransporters includingOATP1A2,OATP2B1, andOAT3havebeenidentified in human brain microvessels (3). In particular,OATP1A2 is present at high abundance in the luminal membrane

Figure 6.

Sensitivity analyses using the developed PBPK model. A, Impact of BBB integrity, as assessed by passive diffusion clearance (PSB) and active effluxclearance at the BBB (CLefflux,BBB) on the extent of BBB penetration (assessed by Kp,uu). B, Impact of BBB integrity (PSB) and active uptake clearance atthe BBB (CLuptake,BBB) on the extent of BBB penetration (Kp,uu). C, Impact of BBB integrity (PSB) and active efflux clearance at the BBB (CLefflux,BBB) onthe rate of BBB penetration, as assessed by the time to achieve the maximum drug brain concentration (Tmax,br). D, Impact of BBB integrity (PSB) andactive uptake clearance at the BBB (CLuptake,BBB) on the rate of BBB penetration (Tmax,br). Different colors of surface areas represent different values ofthe Kp,uu or Tmax,br.






of capillary endothelial cells of human BBB (3, 24), implicating acritical role of this transporter in facilitating the entry of substratedrugs from the blood to brain. Our cellular uptake experimentssuggest that AZD1775 is a substrate of OATP1A2, but not forOATP2B1 or OAT3 (Fig. 4), supporting the premise thatOATP1A2-mediated uptake is likely involved in AZD1775 brainpenetration. Notably, AZD1775 and other OATP1A2 substratedrugs (e.g., imatinib, methotrexate, paclitaxel, fentanyl, and pen-tazocine) are also the substrates of ABCB1 and/or ABCG2 (25–29). Similar to AZD1775, fentanyl and pentazocine show goodBBB penetration (30, 31). It is known that the Kp,uu is determinedby the ratio of net influx andnet efflux clearances that canoccur viaboth passive diffusion and transporter-mediated active transport(5). Active efflux increases the net efflux clearance, whereas activeinflux increases the net influx clearance. For a dual substrate ofefflux and uptake transporters, it is plausible that under certaincircumstances, active efflux and/or outward passive diffusion atthe BBB is reduced, and, consequently, net efflux clearance isdiminished and net influx clearance becomes dominant, therebyleading to increased drug brain penetration.

One circumstance that could lead to diminished net effluxclearance at the BBB is the acidic tumor microenvironment. Thereis firm evidence that the interstitial pH of many solid tumors,including glioblastoma, ismore acidic (pH, 5.8–7.2) than normaltissue interstitial pH (7.0–7.8; refs. 32–34). The relatively acidictumor microenvironment not only plays an important role intumor pathophysiology but also can influence the tumor pene-tration and distribution of weak base or weak acid chemothera-peutic agents (32, 33). As supported by the pH partition theoryand demonstrated by our in vitro cellular permeability experi-ments, at a relatively acidic basolateral pH, AZD1775 permeabil-ity decreased in the basolateral-to-apical direction while increas-ing in the apical-to-basolateral direction, thus resulting in apronounced reduction in the efflux ratios on MDCKII-ABCB1andMDCKII-ABCG2monolayers (Fig. 3B). These in vitro findingscan be extrapolated to an in vivo situation. By IVIVE scaling of thepassive permeability determined from MDCKII cells (Eq. 1) andefflux ratios determined from MDCKII-ABCB1 and -ABCG2 cells(Eq. 2), it is predicted that when brain tumor interstitial pH variesfrom7.4 to 6.5, the outward passive diffusion clearance at the BBBwould decrease by �26% (from 14.4 to 10.6 L/h), whereas therewould be no apparent change in the inward passive diffusionclearance (from 12.2 to 12.9 L/h); meanwhile, the active effluxclearance would decrease by 63% (from 5.66 to 2.07 L/h). Hence,assuming the uptake transporter-mediated clearance remains thesame (44 L/h), the overall decrease of the net efflux clearancewould lead to a 1.6-fold increase in the Kp,uu (from 2.8 to 4.5).Collectively, these data provide mechanistic and quantitativeevidence supporting the notion that the relative acidic tumorenvironment is favorable to the penetration and trapping of aweak base drug that is a substrate of ABCB1/ABCG2 into braintumors due to reduced net efflux clearance at the BBB. These dataalso suggest that variations in the brain tumor regional pH cancontribute to the heterogeneous penetration of weak base drugssuch as AZD1775 into brain tumors.

In addition, the variation in BBB transporter expression/activityis an important factor accounting for the interspecies, inter-individual, and intraindividual variability in drug penetrationinto the brain or brain tumors. The impact of efflux or uptaketransporter activity on the extent of BBB penetration (Kp,uu) isillustrated by the PBPK simulation and sensitivity analyses. Of

note, this impact is more significant for drugs with low passivepermeability compared to those with high passive permeability(Fig. 6A and B; Supplementary Fig. S2E and S2F). The transporterfunction of ABCB1, ABCG2, or OATP1A2 can be impaired due togenetic polymorphisms or pharmacologic inhibition. A recentstudy using positron emission tomography (PET) clearly dem-onstrated that in peoplewith a commonnon-synonymous SNPofthe ABCG2 gene (421C>A), which occurs in �20% of the pop-ulation, pharmacologic inhibition of ABCB1 caused a significantincrease in brain distribution of dual ABCB1/ABCG2 substrates(35). Three common synonymous SNPs of the ABCB1 gene(3435C>T, 2677G>T/A, and 1236C>T) have been associated withthe susceptibility to fentanyl-induced respiration suppression,indicating the impact of ABCB1 polymorphisms on the brainpenetration of the ABCB1 substrate fentanyl, a synthetic opioidreceptor agonist (28). A number of SNPs have been identified inthe OATP1A2 gene, and a few functional variants have beencharacterized in vitro (24, 36, 37). For example, the 516A>C,559G>A, and833A>—deletion variants have been associatedwitha reduced uptake function of OATP1A2 (24, 38), whereas the38T>C variant increases the uptake of estrone sulfate and meth-otrexate (38). In light of the key role of ABCB1, ABCG2, andOATP1A2 in the BBB penetration of AZD1775, it is plausible thatfunctional genetic polymorphisms of these transporters mayaccount for the interindividual variability in the brain pharma-cokinetics of AZD1775. Further investigation in a larger patientpopulation is needed to test this hypothesis.

Another brain pathophysiology factor that may contribute toheterogeneous drug penetration into brain tumors is the variationin BBB integrity (39, 40). The magnitude of brain tumor vascularpermeability varies both spatially and temporally, with the great-est permeability elevation in tumor core anda largely or complete-ly intact BBB at the proliferating tumor edge or infiltrating tumorregions (41).One conventionalway tomeasure BBB integrity is bygadolinium-enhancedmagnetic resonance imaging (MRI; ref. 42).However, it is important to note that given the complex physi-ology of the BBB permeability affected by active transporters,molecule charge, and binding to plasma protein or brain tissue,contrast enhancement by water-soluble, non-ionic gadoliniumagents may not accurately reflect the BBB penetration of lipophil-ic, charged, efflux/uptake transporter substrate drugs (40). Theinterplay of BBB integrity and active transporter activity on theextent of drug BBB penetration (i.e., Kp,uu) can be illustrated byPBPK simulations (Fig. 6 and Supplementary Fig. S2). Wheninsignificant active transport is present, variation in BBB integrityhas negligible impact on the Kp,uu (Fig. 6A and B; SupplementaryFig. S2A and S2B). This may explain, at least in part, the obser-vation that permeability can differ by more than 100-fold amongcompounds capable of penetrating into the brain (43). Whenactive efflux is dominant, compromised BBB integrity results inincreased Kp,uu, whereas an opposite change is observed whenactive uptake is dominant (Fig. 6A and B; Supplementary Fig. S2Aand S2B). Hence, given the fact that plasma protein-bound drugsgenerally do not cross the intact or partially impaired BBB, thegeneral belief that the extent of drug penetration into contrast-enhanced tumor area is higher than that in non-enhanced areamay be valid only for the drugs that undergo dominant activeefflux transport at the BBB (e.g., topotecan andmethotrexate; refs.44, 45) or those that are BSC class III compounds characterizedwith high water solubility and low passive permeability (e.g.,gadolinium imaging agents) but may not be applied to the drugs

Li et al.





that undergo dominant active uptake at the BBB (e.g., AZD1775)or those that are not the substrates of efflux/uptake transporters.The PBPK simulations suggest that the penetration of AZD1775into the invasive edge or infiltrative region of glioblastoma wherethe BBB remains largely intact would be better than, or at leastsimilar to, that in bulky tumor areas where the BBB is "leaky,"assuming other factors (e.g., regional pH and transporter activity)remain the same in these regions. Further studies are needed toverify this prediction by comparisons of the Kp,uu measured frompaired tumor samples (enhanced vs. non-enhanced region) ofeach individual patient. Because invasive, infiltrative glioblasto-ma regions are often unresectable and give rise to recurrentdisease, a drug with good penetration into these regions wouldprovide a huge therapeutic advantage.

In conclusion, using an integrated quantitative, clinical phar-macology approach that leverages clinical trial, IVIVE, and PBPKmodeling, we provided the first clinical evidence of good tumorpenetration of AZD1775 in patients with glioblastoma, and wefurther elucidated the underlying mechanism. Specifically, in therelatively acidic tumor microenvironment where ABCB1/ABCG2–mediated efflux clearance is reduced, OATP1A2-mediat-ed active uptake becomes dominant to drive AZD1775 penetra-tion intobrain tumors. Variations in thebrain/tumor regional pH,transporter expression/activity, and BBB integrity collectivelyattribute to the heterogeneity of drug penetration into braintumors. Our study not only provides quantitative and mechanis-tic insights into the penetration of AZD1775 into brain tumors toinform decision-making regarding further clinical developmentof this novel agent for the treatment of glioblastoma but alsocomposes a framework for prospective, quantitative, and mech-

anistic investigating the penetration of novel anticancer drugsacross human BBB in general.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: J. Li, A. Sparreboom, N. SanaiDevelopment of methodology: J. Li, J. Wu, Y. XieAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): J. Li, J. Wu, X. Bao, N. Honea, Y. Xie, A. SparreboomAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): J. Li, J. Wu, S. KimWriting, review, and/or revision of the manuscript: J. Li, J. Wu, N. Honea,Y. Xie, A. Sparreboom, N. SanaiAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): J. Li, J. Wu, X. Bao, N. HoneaStudy supervision: J. Li, N. Sanai

AcknowledgmentsThe authors thank AstraZeneca for providing the drug (AZD1775) for this

study. The authors particularly thank the patients enrolled in the study.This study was supported by the United States Public Health Service Cancer

Center Support Grant P30 CA022453, the American Society of Clinical Oncol-ogy Career Development Award, and the Ben and Catherine Ivy Foundation.

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received April 4, 2017; revised May 12, 2017; accepted September 12, 2017;published OnlineFirst September 19, 2017.

References1. Henderson JT, Piquette-Miller M. Blood-brain barrier: an impediment to

neuropharmaceuticals. Clin Pharmacol Ther 2015;97:308–13.2. PardridgeWM.Drug transport across theblood-brainbarrier. J CerebBlood

Flow Metab 2012;32:1959–72.3. Bronger H, Konig J, Kopplow K, Steiner HH, Ahmadi R, Herold-Mende C,

et al. ABCC drug efflux pumps and organic anion uptake transporters inhuman gliomas and the blood-tumor barrier. Cancer Res 2005;65:11419–28.

4. Miller DS. Regulation of ABC transporters at the blood-brain barrier. ClinPharmacol Ther 2015;97:395–403.

5. Hammarlund-Udenaes M, Friden M, Syvanen S, Gupta A. On the rate andextent of drug delivery to the brain. Pharm Res 2008;25:1737–50.

6. de Lange EC. Themastermind approach to CNS drug therapy: translationalprediction of humanbrain distribution, target site kinetics, and therapeuticeffects. Fluids Barriers CNS 2013;10:12.

7. Yamamoto Y, Danhof M, de Lange EC. Microdialysis: the key to physio-logically basedmodel prediction of human CNS target site concentrations.AAPS J 2017;19:891–909.

8. Hirai H, Iwasawa Y, Okada M, Arai T, Nishibata T, Kobayashi M, et al.Small-molecule inhibition of Wee1 kinase by MK-1775 selectively sensi-tizes p53-deficient tumor cells to DNA-damaging agents. Mol Cancer Ther2009;8:2992–3000.

9. Parker LL, Piwnica-WormsH. Inactivation of the p34cdc2-cyclin B complexby the human WEE1 tyrosine kinase. Science 1992;257:1955–7.

10. Watanabe N, Broome M, Hunter T. Regulation of the human WEE1HuCDK tyrosine 15-kinase during the cell cycle. EMBO J 1995;14:1878–91.

11. Beck H, Nahse V, Larsen MS, Groth P, Clancy T, Lees M, et al. Regulators ofcyclin-dependent kinases are crucial for maintaining genome integrity in Sphase. J Cell Biol 2010;188:629–38.

12. DeWitt Hamer PC, Mir SE, Noske D, Van Noorden CJ, Wurdinger T. WEE1kinase targeting combined with DNA-damaging cancer therapy catalyzesmitotic catastrophe. Clin Cancer Res 2011;17:4200–7.

13. Kuerbitz SJ, Plunkett BS,WalshWV, KastanMB.Wild-type p53 is a cell cyclecheckpoint determinant following irradiation. Proc Natl Acad Sci U S A1992;89:7491–5.

14. Leijen S, Beijnen JH, Schellens JH. Abrogation of the G2 checkpoint byinhibition of Wee-1 kinase results in sensitization of p53-deficient tumorcells to DNA-damaging agents. Curr Clin Pharmacol 2010;5:186–91.

15. Wu J, Sanai N, Bao X, LoRusso P, Li J. An aqueous normal-phase chro-matography coupled with tandem mass spectrometry method for deter-mining unbound brain-to-plasma concentration ratio of AZD1775, aWee1 kinase inhibitor, in patients with glioblastoma. J Chromatogr BAnal Technol Biomed Life Sci 2016;1028:25–32.

16. Gaohua L, Neuhoff S, Johnson TN, Rostami-Hodjegan A, Jamei M. Devel-opment of a permeability-limited model of the human brain and cere-brospinal fluid (CSF) to integrate known physiological and biologicalknowledge: estimating time varying CSF drug concentrations and theirvariability using in vitro data. Drug Metab Pharmacokinet 2016;31:224–33.

17. Badhan RK, Chenel M, Penny JI. Development of a physiologically-basedpharmacokinetic model of the rat central nervous system. Pharmaceutics2014;6:97–136.

18. Kalvass JC, Pollack GM. Kinetic considerations for the quantitative assess-ment of efflux activity and inhibition: implications for understanding andpredicting the effects of efflux inhibition. Pharm Res 2007;24:265–76.

19. Uchida Y, Ohtsuki S, Katsukura Y, Ikeda C, Suzuki T, Kamiie J, et al.Quantitative targeted absolute proteomics of human blood-brain barriertransporters and receptors. J Neurochem 2011;117:333–45.

20. Neuhoff S, Ungell AL, Zamora I, Artursson P. pH-dependent bidirectionaltransport of weakly basic drugs across Caco-2monolayers: implications fordrug-drug interactions. Pharm Res 2003;20:1141–8.

21. Goda K, Balkay L, Marian T, Tron L, Aszalos A, Szabo G Jr. Intracellular pHdoes not affect drug extrusion by P-glycoprotein. J Photochem Photobiol BBiol 1996;34:177–82.






22. Altenberg GA, Young G, Horton JK, Glass D, Belli JA, Reuss L. Changes inintra- or extracellular pH do not mediate P-glycoprotein-dependent mul-tidrug resistance. Proc Natl Acad Sci U S A 1993;90:9735–8.

23. Pokorny JL, Calligaris D, Gupta SK, Iyekegbe DO Jr, Mueller D, Bakken KK,et al. The efficacy of the Wee1 inhibitor MK-1775 combined with temo-zolomide is limited by heterogeneous distribution across the blood-brainbarrier in glioblastoma. Clin Cancer Res 2015;21:1916–24.

24. Lee W, Glaeser H, Smith LH, Roberts RL, Moeckel GW, Gervasini G, et al.Polymorphisms in human organic anion-transporting polypeptide 1A2(OATP1A2): implications for altered drug disposition and central nervoussystem drug entry. J Biol Chem 2005;280:9610–7.

25. Eechoute K, Franke RM, LoosWJ, Scherkenbach LA, Boere I, Verweij J, et al.Environmental and genetic factors affecting transport of imatinib byOATP1A2. Clin Pharmacol Ther 2011;89:816–20.

26. van de Steeg E, van Esch A, Wagenaar E, Kenworthy KE, Schinkel AH.Influence of human OATP1B1, OATP1B3, and OATP1A2 on the pharma-cokinetics of methotrexate and paclitaxel in humanized transgenic mice.Clin Cancer Res 2013;19:821–32.

27. Shukla S, Ohnuma S, Ambudkar SV. Improving cancer chemotherapy withmodulators of ABC drug transporters. Curr Drug Targets 2011;12:621–30.

28. Park HJ, Shinn HK, Ryu SH, Lee HS, Park CS, Kang JH. Genetic poly-morphisms in the ABCB1 gene and the effects of fentanyl in Koreans. ClinPharmacol Ther 2007;81:539–46.

29. Moriki Y, Suzuki T, Fukami T, Hanano M, Tomono K, Watanabe J.Involvement of P-glycoprotein in blood-brain barrier transport of pentaz-ocine in rats using brain uptake index method. Biol Pharm Bull2004;27:932–5.

30. Henthorn TK, Liu Y, Mahapatro M, Ng KY. Active transport of fentanyl bythe blood-brain barrier. J Pharmacol Exp Ther 1999;289:1084–9.

31. Suzuki T, Moriki Y, Goto H, Tomono K, Hanano M, Watanabe J. Inves-tigation on the influx transport mechanism of pentazocine at the blood-brain barrier in rats using the carotid injection technique. Biol Pharm Bull2002;25:1351–5.

32. Gerweck LE, Seetharaman K. Cellular pH gradient in tumor versus normaltissue: potential exploitation for the treatment of cancer. Cancer Res1996;56:1194–8.

33. Tannock IF, Rotin D. Acid pH in tumors and its potential for therapeuticexploitation. Cancer Res 1989;49:4373–84.

34. MartinGR, Jain RK. Noninvasivemeasurement of interstitial pH profiles innormal and neoplastic tissue using fluorescence ratio imagingmicroscopy.Cancer Res 1994;54:5670–4.

35. Bauer M, Romermann K, Karch R, Wulkersdorfer B, Stanek J, Philippe C,et al. Pilot PET study to assess the functional interplay between ABCB1 andABCG2 at the human blood-brain barrier. Clin Pharmacol Ther2016;100:131–41.

36. Zhou Y, Yuan J, Li Z, Wang Z, Cheng D, Du Y, et al. Genetic polymorph-isms and function of the organic anion-transporting polypeptide 1A2and its clinical relevance in drug disposition. Pharmacology 2015;95:201–8.

37. Iida A, Saito S, Sekine A,Mishima C, Kondo K, Kitamura Y, et al. Catalog of258 single-nucleotide polymorphisms (SNPs) in genes encoding threeorganic anion transporters, three organic anion-transporting polypeptides,and three NADH:ubiquinone oxidoreductase flavoproteins. J Hum Genet2001;46:668–83.

38. Badagnani I, Castro RA, Taylor TR, Brett CM, Huang CC, Stryke D, et al.Interaction of methotrexate with organic-anion transporting polypep-tide 1A2 and its genetic variants. J Pharmacol Exp Ther 2006;318:521–9.

39. Lockman PR, Mittapalli RK, Taskar KS, Rudraraju V, Gril B, Bohn KA, et al.Heterogeneous blood-tumor barrier permeability determines drug efficacyin experimental brain metastases of breast cancer. Clin Cancer Res2010;16:5664–78.

40. Gerstner ER, Fine RL. Increased permeability of the blood-brain barrier tochemotherapy in metastatic brain tumors: establishing a treatment para-digm. J Clin Oncol 2007;25:2306–12.

41. Ewing JR, Brown SL, Lu M, Panda S, Ding G, Knight RA, et al. Modelselection in magnetic resonance imaging measurements of vascular per-meability: Gadomer in a 9L model of rat cerebral tumor. J Cereb BloodFlow Metab 2006;26:310–20.

42. Heye AK, Culling RD, Valdes HernandezMdel C, ThrippletonMJ,WardlawJM. Assessment of blood-brain barrier disruption using dynamic contrast-enhanced MRI. A systematic review. Neuroimage Clin 2014;6:262–74.

43. Liu X, Tu M, Kelly RS, Chen C, Smith BJ. Development of a computationalapproach to predict blood-brain barrier permeability. Drug Metab Dispos2004;32:132–9.

44. Leggas M, Adachi M, Scheffer GL, Sun D, Wielinga P, Du G, et al. Mrp4confers resistance to topotecan and protects the brain from chemotherapy.Mol Cell Biol 2004;24:7612–21.

45. Blakeley JO, Olson J, Grossman SA, He X, Weingart J, Supko JG. Effect ofblood brain barrier permeability in recurrent high grade gliomas on theintratumoral pharmacokinetics of methotrexate: a microdialysis study.J Neurooncol 2009;91:51–8.


Li et al.




2017;23:7454-7466. Published OnlineFirst September 19, 2017.Clin Cancer Res Jing Li, Jianmei Wu, Xun Bao, et al.

PBPK Modeling Approach−Patients Using an IVIVE Brain Barrier in Glioblastoma−Penetration across Human Blood

Quantitative and Mechanistic Understanding of AZD1775

Updated version

10.1158/1078-0432.CCR-17-0983doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2017/09/19/1078-0432.CCR-17-0983.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/23/24/7454.full#ref-list-1

This article cites 45 articles, 19 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/23/24/7454.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/23/24/7454To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-17-0983

http://clincancerres.aacrjournals.org/content/suppl/2017/09/19/1078-0432.CCR-17-0983.DC1

http://clincancerres.aacrjournals.org/content/23/24/7454.full#ref-list-1

http://clincancerres.aacrjournals.org/content/23/24/7454.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/23/24/7454


quantitative and mechanistic understanding of azd1775 ......cancer therapy: clinical quantitative...

Documents