preclinical optimization of mdm2 antagonist scheduling...
TRANSCRIPT
Cancer Therapy: Preclinical
Preclinical Optimization of MDM2 Antagonist Scheduling forCancer Treatment by Using a Model-Based Approach
Brian Higgins1, Kelli Glenn2, Antje Walz4, Christian Tovar1, Zoran Filipovic1, Sazzad Hussain2, Edmund Lee1,Kenneth Kolinsky1, Shahid Tannu1, Violeta Adames3, Rosario Garrido3, Michael Linn3, Christophe Meille4,David Heimbrook1, Lyubomir Vassilev1, and Kathryn Packman1
AbstractPurpose: Antitumor clinical activity has been demonstrated for the MDM2 antagonist RG7112, but
patient tolerability for the necessary daily dosing was poor. Here, utilizing RG7388, a second-generation
nutlin with superior selectivity and potency, we determine the feasibility of intermittent dosing to guide the
selection of initial phase I scheduling regimens.
Experimental Design: A pharmacokinetic–pharmacodynamic (PKPD) model was developed on the
basis of preclinical data todetermine alternative dosing schedule requirements foroptimal RG7388-induced
antitumor activity. This PKPDmodel was used to investigate the pharmacokinetics of RG7388 linked to the
time-course of the antitumor effect in an osteosarcoma xenograft model in mice. These data were used to
prospectively predict intermittent and continuous dosing regimens, resulting in tumor stasis in the same
model system.
Results:RG7388-induced apoptosiswas delayed relative to drug exposurewith continuous treatment not
required. In initial efficacy testing, daily dosing at 30mg/kg and twice a week dosing at 50mg/kg of RG7388
were statistically equivalent in our tumor model. In addition, weekly dosing of 50 mg/kg was equivalent to
10 mg/kg given daily. The implementation of modeling and simulation on these data suggested several
possible intermittent clinical dosing schedules. Further preclinical analyses confirmed these schedules as
viable options.
Conclusion: Besides chronic administration, antitumor activity can be achieved with intermittent
schedules of RG7388, as predicted through modeling and simulation. These alternative regimens may
potentially ameliorate tolerability issues seen with chronic administration of RG7112, while providing
clinical benefit. Thus, both weekly (qw) and daily for five days (5 d on/23 off, qd) schedules were selected
for RG7388 clinical testing. Clin Cancer Res; 20(14); 3742–52. �2014 AACR.
IntroductionThe tumor suppressor p53 plays a key role in preventing
malignant transformation and inhibiting the developmentof cancer. In response to cellular stress, the p53 pathwaytriggers gene transcription, which can result in cell-cyclearrest, apoptosis, and DNA repair and/or senescence (1).
The dysregulation of the p53 pathway is the most frequentalteration detected in a broad range of human cancers, andapproximately half of these cancers have been determinedto carry mutated TP53 (2). Another mechanism by whichp53 function may become lost is the overexpression ofMDM2.Under nonstress conditions, the E3 ubiquitin ligaseMDM2 is a negative regulator that directly binds to andinhibits p53 activity by targeting p53 for ubiquitin-depen-dent degradation (3, 4). Therefore, MDM2 overexpressionresults in reduced p53 levels, inadequate cell growth arrest,and apoptosis. Accordingly, disrupting the binding ofMDM2 to a functional p53 can be expected to restorep53-dependent cellular arrest and apoptosis.
The nutlin family of MDM2 antagonists was designed todisrupt p53–MDM2 binding (5–7), thereby stabilizing andactivating p53. In preclinical studies, the nutlin familymember RG7112 promoted the reactivation of the p53pathway to elicit growth arrest and apoptosis in tumor cellswhereMDM2 gene amplification or other oncogenic drivershave allowed evasion of p53-activated cell death (8). How-ever, during RG7112 clinical testing, there were tolerability
Authors' Affiliations: 1Discovery Oncology; 2Drug Metabolism and Phar-macokinetics; and 3Non-Clinical Safety, PharmaResearchandEarlyDevel-opment, Hoffmann-La Roche, Inc., Nutley, New Jersey; and 4Modeling andSimulation, Pharma Research and Early Development, Hoffmann-LaRoche, Inc., Basel, Switzerland
Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).
B. Higgins, K. Glenn, and A. Walz contributed equally to this article.
Corresponding Author: Brian Higgins, Roche Innovation Center NewYork, 430 East 29th Street, New York, NY 10016. Phone: 646-461-5063;Fax: 646-461-5312; E-mail: [email protected]
doi: 10.1158/1078-0432.CCR-14-0460
�2014 American Association for Cancer Research.
ClinicalCancer
Research
Clin Cancer Res; 20(14) July 15, 20143742
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
challenges with prolonged daily oral administration ofRG7112 (�10 consecutive days in patients), includinggrade 3/4 vomiting, neutropenia, and thrombocytopenia
(9). Therefore, optimization of the schedule of administra-tion, including pursuing intermittent dosing, is an impor-tant consideration for further clinical development ofMDM2 inhibitors.
The pyrrolidine RG7388,which has the samemechanismof action as the imidazoline RG7112, is a more potent andselective second-generation MDM2 inhibitor (8, 10). Inpreclinical studies, RG7388 effectively activated the p53pathway in SJSA1 cells and induced tumor growth inhibi-tion (TGI) in correlative xenografts in nude mice at signif-icantly lower doses and exposures than RG7112 (10, 11).These results suggest the potential of achieving a clinicalbenefit at significantly lower drug concentrations withRG7388 and indicate that it should be possible to achievea decrease in the severe toxicities that limited the less potentRG7112.
As RG7112 had poor tolerability when given daily, wesought to determine whether RG7388 could be dosedintermittently to circumvent the need for daily administra-tion. To better understand the scheduling requirements foroptimal RG7388 antitumor activity, a model-basedapproach was undertaken (12). First, a dynamic pharma-cokinetic–pharmacodynamic (PKPD) model was devel-oped to investigate how the pharmacokinetics of RG7388are related to the time course of the antitumor effect in SJSAosteosarcoma xenografts in mice (Fig. 1, step 1 and 2).Second, the model was used to prospectively predict inter-mittent and continuous dosing regimens that would result
0
400
800
1,200
1,600
2,000
2,400
2,800
32292623201714118
Days post-tumor cell implant
Mea
n tu
mor
vol
ume
(m
m3 )
± S
EM
Control
qd 1.1 mg/kg
qd 3.3 mg/kg
qd 10 mg/kg
qd 30 mg/kg
1x/week 50 mg/kg
2x/week 50 mg/kg =+
Step 1: Design and conduct PKPD experiment
Step 2: PKPD modeling to integrate knowledge, assumptions, and data
kg
Step 3: Use the PKPD model to explore “what if” scenarios
Step 4: Compare experimental data witha priori predictions
Step 5: Test hypothesis and confirm/revise PKPD model
3,000
2,000
1,000
00 7 14
Time (day)
Tum
or v
olum
e (m
m3 )
21 28
5,500
5,000
4,500
4,000
3,500
3,000
2,500
2,000
1,500
1,000
500
010 15 20 25 30 35
Mea
n tu
mor
vol
ume
(m
m3 )
± S
EM
Days post-tumor cell implant
Tumor
PK
E
T1
tau–1 tau–1 tau–1 tau–1
T2 T3 T4
Figure 1. Strategy to better understand dosing schedule requirements for optimal antitumor activity.
Translational RelevanceNutlins, small-molecule inhibitors of the p53–MDM2
interaction, reactivate p53 and have provided preclinicalproof-of-concept as therapeutics for patients withtumors expressing functional p53. The nutlin familymemberRG7112was thefirst small-molecule antagonistof MDM2 to be tested clinically. However, prolongeddaily administration was poorly tolerated. Here, usingthe next-generation MDM2 antagonist RG7388, wedemonstrate that intermittent regimens could be usedinstead of chronic administration, with the expectationof improving patient tolerability while inducing equiv-alent antitumor activity and efficacy. Modeling andsimulation applied toward foundation data from aninitial efficacy dataset facilitated a prospective evaluationof alternative regimens predicted to achieve tumor stasis.When tested in a follow-up efficacy study, predictionswere in alignmentwithobserved responses. These resultssupported the selection of the intermittent schedulesutilized in the initial phase I RG7388 clinical trial tocircumvent tolerability issues.
Optimization of MDM2 Antagonist RG7388 for Clinical Testing
www.aacrjournals.org Clin Cancer Res; 20(14) July 15, 2014 3743
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
in tumor stasis (Fig. 1, step 3). This allowed for the testing ofdosing hypotheses by comparing a priori predictedresponses to the actual observed tumor growth profile (Fig.1 step 4). Given that both p53 and MDM2 have short half-lives, two hypotheses were that continuous suppression ofthe p53–MDM2 interaction would be required for optimalantitumor activity, or alternatively, that stabilization andactivation of p53 and subsequent apoptosis, trigger a sus-tained antitumor effect. The first hypothesis would beconsistent with the need for continuous dosing andassumes that the PD is directly linked to the PK, whereasthe second hypothesis implies that intermittent dosingshould be sufficient to elicit a robust antitumor response,while potentially reducing the occurrence of severe adverseevents associated with the prolonged continuous adminis-tration of RG7112 (Fig. 1, step 5).
Materials and MethodsIn vitro testing in cancer cell lines
RG7388 was prepared at concentrations of 1 and 10mmol/L in DMSO and stored in aliquots at �20�C. SJSA,RKO, HCT116, H460, A375, SK-MEL-5, SW480, MDA435,and HeLa cells were obtained from the ATCC. Cell lineswere authenticated by short tandem repeat analysis throughPromega authentication services. For in vitro studies, cellswere cultured in their ATCC-designated media. Mediumwas supplemented with 10% FBS and 1% 200 nmol/L L-glutamine. To assess cell viability, cells were seeded atdensities identified for best growth for a 5-day assay in96-well plates in normal growth media. Serial dilutions ofRG7388 (1–3 in fresh media) starting at 300 mmol/L wereapplied towells (1–10) in triplicate for a final concentrationrange of 0.01 to 30 mmol/L and control wells were treatedwith 0.3% DMSO equivalent to DMSO at the highestRG7388 concentration. Cell respiration, as an indicator ofcell viability, was measured by the reduction of MTT toformazan as previously described (5).
Percent apoptosis was determined as described in Tovarand colleagues (8). For Western blot analysis, cells werecultured in T-75 flasks (4 mL total volume at 5 � 105 cells/well) and incubated overnight at 37�C, 5% CO2. Cells weretreatedwith 0.3 or 1.8 mmol/L of RG7388 or 0.1%DMSOascontrol. Treatment duration was 16 hours, and lysates wereprepared before washout and at 4, 8, 24, and 48 hours afterRG7388 washout.
Western blot analysisAntibodies against p53 andMDM2were purchased from
SantaCruzBiotechnology. The anti-p21WAF1was purchasedfrom Calbiochem, Merck KGaA and anti–b-actin fromSigma-Aldrich. Protein was extracted from cells or tumortissue with 1� RIPA buffer (Sigma-Aldrich) containingprotease inhibitors (Roche Diagnostics) by homogeniza-tion or scraping, respectively. Equal amounts of totalprotein were resolved on 4%–12% NuPAGE gradient gel(Invitrogen, Life Technologies) and blotted with antibodiesas indicated. The chemiluminescent signal was generated
with ECL Plus (GE Healthcare Life Sciences) and detectedwith a Fujifilm LAS-4000 imager. The densitometric quan-titation of specific bandswas determined usingMulti GaugeSoftware (Fujifilms).
Pharmacokinetic analysisTo determine RG7388 plasma concentrations, blood
samples were collected from female mice during in vivoantitumor studies. In each treatment group, on the first and/or last dosing day, blood samples (generally n ¼ 2/timepoint) were collected at various predetermined time pointsranging from 0.5 to 24 hours after the last dose. Pharma-cokinetic assessment was performed via noncompartmen-tal analysis using Watson v7.4 (Thermo Fisher Scientific),where parameters were calculated on the basis of the com-posite concentration–time data from each treatment groupand sampling day. Sampling times were reported as nom-inal time, with concentrations below the limit of quantita-tion excluded. Parameters reported include plasma half-life(t1/2),Cmax,Tmax, andareaunder theplasma concentration–time curve from end of dosing to the last bleeding timepoint (AUC0–24 hours). The AUCs were calculated using thelinear trapezoidal rule. TheCmax and Tmax values were takendirectly from the plasma concentration–time profiles with-out extrapolations.
In vivo activity studies in xenograft tumor modelsAthymic female nude mice (Crl:NU-Foxn1nu) were
obtained from Charles River Laboratories. The health ofall animals was monitored daily by gross observation andanalyses of blood samples of sentinel animals. All animalexperiments were performed in accordance with protocolsapproved by the Institutional Animal Care and Use Com-mittee in our Association for Assessment and Accreditationof Laboratory Animal Care-accredited facility.
At 10 to 12 weeks of age, mice were implanted with a 1:1mixture of human SJSA osteosarcoma cells (ATCC) sus-pended in phenol-free Matrigel and PBS. Mice wereimplanted in the right flank at a concentration of 5 �106 cells in 0.2 mL total volume. At approximately day10, animals were randomized according to tumor volume,so that all groups of 10 randomized mice had similarstarting mean tumor volumes of 100 to 250 mm3.
Tumor measurements and weights were taken two tothree times per week. TGI was calculated from percentchange in mean tumor volume compared with the controlgroup. Average percent weight change was used as a sur-rogate endpoint for tolerability in the experiment. Animalsin each group were continuously followed beyond the lastday of treatment to see whether tumor regrowth wouldoccur. In this second phase of analysis, survival was calcu-lated using a cutoff individual tumor volume of 1,500mm3
as a surrogate for mortality. The increase in lifespan (ILS)was calculated as a percentage using the formula: [(medianday of death in treated tumor-bearingmice)� (median dayof death in control tumor-bearing mice)]/median day ofdeath in control tumor-bearing mice � 100. Statisticalanalysis was performed as previously described (8).
Higgins et al.
Clin Cancer Res; 20(14) July 15, 2014 Clinical Cancer Research3744
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
RG7388 was administered as an amorphous solid dis-persionmicrobulk precipitate powder containing 30%drugsubstance and 70% hydroxypropyl methylcellulose acetatesuccinate polymer that was reconstituted immediatelybefore administration as a suspension in Klucel/Tween,and remaining suspension was discarded after dosing.
Modeling and simulationThe PKPD model structure.For PKPD modeling, plasma
PK was fitted to a 1-compartmental model assuming non-linear bioavailability and a first-order absorption and elim-ination rate constant. The differential equations and initialconditions for the PK are as follows:
dA=dt ¼ D� 1� Dhill
Dhill þD50hill
� �� ka�D A0 ¼ 0
dQc=dt ¼ ka�D� ke�Qc Qc0 ¼ 0
Cp ¼ Qc
V
where D corresponds to dose (mg/kg) and A representsunabsorbed drug. Qc (mg) corresponds to the amount ofdrug in central compartment, ka (hour-1) to the first-orderabsorption rate constant, ke (hour-1) to the first-orderelimination rate constant, V (L/kg), and Cp (mg/mL) to,respectively, the volume and drug concentration in centralcompartment.The full PKPD model assumes that the drug effects of
RG7338 are delayed relative to the observed antitumoreffect. This was described by using a signal distributionmodel based on themodel proposed by Lobo and Balthasar(13).The model structure is shown in Supplementary Fig. S1,
and the differential equations and initial conditions for thePKPD are as follows:
dR=dt ¼ kg�R� T4�R; Rð0Þ ¼ w0
dT1=dt ¼ tau�1� E� T1ð Þ; Tð1Þ ¼ 0
dT2=dt ¼ tau�1� T1 � T2ð Þ; Tð2Þ ¼ 0
dT3=dt ¼ tau�1� T2 � T3ð Þ; Tð3Þ ¼ 0
dT4=dt ¼ tau�1�ðT3 � T4Þ; Tð4Þ ¼ 0
E ¼ k2�Cp;
Where R represents the tumor volume, kg the exponentialtumor growth rate, andw0 initial tumor volume (mm3), andtau (hour) represents the mean transit time equal for eachcompartment. Drug effect E was described by a linear drugeffect model.PKPD parameter estimation. PKPD parameter estima-
tion was done using a nonlinear mixed effect modelingsoftware Monolix, 4.3.1 (Lixoft). For the PK, residual errorswere assumed to be proportional to the predicted concen-trations and a proportional error model was selected. Forthe PD, an additional þ proportional error model wasselected.Diagnostic plotswere inspected to select the appro-priate residual error model. The criteria for evaluatingmodel performance was based on visual inspection, good-ness of fit plots, including observed versus predicted plots,
scatter plots of observed data overlay with predicted data,precision of the parameter estimates, and changes in theobjective functions.
Simulations were conducted using software BerkeleyMadonna, version 8.3.18, Copyright 1993-2001 Robert I.Macey & George F. Oster.
Bioluminescent imagingFor each bioluminescent imaging (BLI) session, mice
received 100 mg/kg D-luciferin (Caliper Life Sciences/Per-kinElmer) via intraperitoneal injection and were imaged at20 minutes after luciferin injection at either a 5- or 10-second exposure time. Images were captured by the IVISSpectrum system and data were collected and analyzedwithLiving Image 4.2.0 software (Caliper Life Sciences). Totalphoton fluxes representing luciferase activity within eachfixed region of interest covering whole tumors of individualmice were determined.
IHC and image analysisFormalin-fixed, paraffin-embedded histologic samples
from xenografts were used for IHC analysis using theVentana Discovery XT platform (Ventana Medical System).Briefly, 5 micron tissue sections were stained using rabbitmonoclonal antibodies anticleaved PARP-1 (Cell SignalingTechnology) and anti-Ki67 (Thermo Fisher Scientific). Tis-sues were pretreated using heat-induced epitope retrieval,and a 3-step biotin-streptavidin-horseradish peroxidasedetection method followed by diaminobenzidine (DAB)chromogen (Ventana Medical System) for antibody detec-tion. Sections were counterstained with hematoxylin, dehy-drated, and mounted with Permount mounting media. Forimage analysis, glass slides were scanned at �20 magnifi-cation using a Zeiss-Mirax digital slide scanner (Carl ZeissImaging), andDefiniens Tissue Studio software (Definiens)was used to build custom-made image analysis algorithms.Generated algorithms were applied to the digital slides, toautomatically detect and quantify viable xenograft areas,and areas of DAB deposition (positive antibody detection)within those viable regions. Numerical results wereexpressed as percentage of positively labeled area withinthe viable xenograft regions (percentage area of DAB label-ing/xenograft viable area).
ResultsRG7388-induced apoptosis in SJSA osteosarcoma cellsis delayed relative to drug exposure but does notrequire continuous treatment
Recent data demonstrated that RG7388 induces the p53pathway activation and in vivo activity at much lower con-centrations (�25%) and exposures (�7%-10%) thanRG7112 (8, 10). Therefore, we initiated our experimentsat these lower doses.
To investigate the dosing schedule requirements forRG7388-induced apoptosis of the MDM2-amplified SJSA1osteosarcoma line, cells were treated with a single dose ofRG7388 (300 nmol/L or 1.8 mmol/L) for 16 hours asestablished in previous continuous-dose experiments
Optimization of MDM2 Antagonist RG7388 for Clinical Testing
www.aacrjournals.org Clin Cancer Res; 20(14) July 15, 2014 3745
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
(10). These concentrations were chosen on the basis ofantiproliferative activity as measured by an MTT assay andrepresent the average IC90 and six times the average IC90 forthe P53 wild-type lines tested. Moreover, these concentra-tions are well below both the IC50 and IC90 levels observedfor the 3P53-mutant lines assayed, thus ensuring a sufficientconcentration to elicit a functional p53-dependentresponse (10)whilemaintaining an insignificant likelihoodof an off-target effect (Supplementary Table S1). At 24 and48 hours after washout of RG7388, apoptosis was deter-mined by positivity for annexin V. A significant apoptoticresponse over the control was induced with both the single300 nmol/L and 1.8 mmol/L concentrations of RG7388 (P <0.005), with the 1.8 mmol/L concentration providing opti-mal response (Fig. 2A). Induction of apoptosis was strongerat 48 hours post-washout compared with 24 hours post-washout. As comparable levels of apoptosis were achievedwith 48 hours of continuous RG7338 dosing at the sameconcentrations (10), these in vitro results suggest that oncethe apoptotic component of the p53 pathway is activated intumor cells by the MDM2 antagonist RG7388, continuousexposure to RG7388 is not required to sustain thisactivation.
Western blot analyses of p53, MDM2, and p21 proteinlevels (Fig. 2B) at the optimal apoptosis response level of1.8 mmol/L further supported the idea that continuousdosing was not required for sustained RG7388 activity.P53 protein levels were increased 16 hours after a singletreatment with RG7388. When p53 levels increase, thetranscription of its targets should increase. Accordingly, theprotein levels of MDM2 and p21 were also elevated.Although stabilized p53 protein did decrease 4 hours afterwashout, p53 levels then persisted and remained elevatedcompared with the control for at least 48 hours post-washout. In contrast, MDM2 and p21 protein levelincreases were lost by 24 and 48 hours, respectively,post-washout. These patterns are consistent with previousobservations made while treating cells continuously withRG7112 (8), and indicated that continuous dosing forRG7388 should not be required given its mechanism ofaction. Collectively these data along with efficacy data wereutilized in the design of our modeling and simulation.
PKPD modeling to investigate the relationshipbetween RG7388 exposure and time course ofantitumor effect
The in vitro data suggest that the drug effect is delayed andthen maintained relative to the drug exposure even afterwashout. To further understand how the pharmacokineticof RG3788 is linked to the antitumor effect, we conducted aPKPD study in SJSA osteosarcoma xenograft-bearing mice.In this study, the tumor growth profiles and the respectiveplasma PK profiles weremonitored over a 21-day period forinclusion in the model.
Animals were given oral doses of RG7388 at 1.11, 3.33,10, or 30 mg/kg once daily, 50 mg/kg once weekly, or50 mg/kg twice weekly (Monday/Wednesday). The high-est doses investigated were of predetermined tolerability
based on pilot experiments (data not shown). Althoughthe once-daily dose of 30 mg/kg of RG7388 (total 210 mg/kg/wk) seemed to be the most effective dose at inhibitingtumor growth (>100% TGI with 9 partial regressions), the50 mg/kg of RG7388 given twice a week (total 100 mg/kg/wk) was equivalent by statistical analysis (Fig. 2C; 96% TGIwith 3 partial regressions; P > 0.05). In addition, 50 mg/kgof RG7388 given once a week (50 mg/kg/wk) elicited an
A
0
10
20
30
40
50
60
Control
Ann
exin
posi
tive
(%)
1.8 mmol/L300 nmol/L
24 hours after washout
48 hours after washout
*, ± SD of 3 independent experiments.
B
MDM2
p53
p21
Actin
CTLPost washout (h)
4 8 24 484 8 24 4816 hPost washout (h)
RG7388 300 nmol/L
16 h
RG7388 1.8 mmol/L
C
TGIRG7388 Regimen a PR, (%) b ILS, (%)
---0---Vehicle qd00111.11 mg/kg qd250503.33 mg/kg qd2517710 mg/kg qd
1259>10030 mg/kg qd3217950 mg/kg qw5739650 mg/kg 2x/wk
a ,P < 0.05.b ,N = 10 per group
Days post-tumor cells implant
Mea
n tu
mor
vol
ume
± SE
M (m
m3 )
Figure 2. Continuous treatment is not required for RG7388-inducedapoptosis in SJSA osteosarcoma cells in vitro or for RG7388-mediatedinhibition in SJSA xenografts. Sustained activity of RG7388 afterwashout. Exponentially growing SJSA cells treated with 0.3 or1.8 mmol/L RG7388 for 16 hours and then washed thoroughly in freshmedia.A, percent annexinVpositivity at 24and48hours afterwashout.B,Western blot analysis shows persistent expression of stabilizedp53 protein after washout, as well as the p53 downstream targets MDM2and p21. C, efficacy produced by daily dosing versus once or twiceweekly dosing (a, P ¼ < 0.05; b, n ¼ 10 mice per group). PR,partial regression.
Higgins et al.
Clin Cancer Res; 20(14) July 15, 2014 Clinical Cancer Research3746
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
equivalent TGI as 10 mg/kg of RG7388 given once daily(70 mg/kg/wk; 79% TGI vs. 77% TGI, respectively; P >0.05), despite being a lower total weekly dose. Therefore,intermittent dosing was equivalent to continuous admin-istration for RG7388-mediated TGI in this xenograftmodel.
RG7388 exposure levels after multiple oral doses infemale miceRG7388 exposure levels were determined in the mice
after the last dose of RG7388 (i.e., day 21). The exposures
increased greater than dose proportionally between 1.11mg/kg and 3.33 mg/kg, and less than dose proportionallybetween 3.33 mg/kg and 50 mg/kg (Fig. 3A and Supple-mentary Fig. S2). The concentration–time profiles were welldescribed by a one-compartmental PK model (Supplemen-tary Figs. S2 and S3).
The dose–exposure response relationship was character-ized quantitatively and a mechanism-based PKPD modelwas fitted simultaneously to the observed individual PKand PD data of all groups (Fig. 3A; PKPD parameter esti-mates Supplementary Table S2). The model separates
Figure 3. Predicted and observed tumor growth data. A, observed and model-predicted tumor growth profiles. Modeling and simulation of PK/PD from initialexperiment was applied to predict intermittent doses/regimens that would elicit tumor stasis. B, intermittent dosing regimens expected to result in 100%TGI.
Optimization of MDM2 Antagonist RG7388 for Clinical Testing
www.aacrjournals.org Clin Cancer Res; 20(14) July 15, 2014 3747
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
system-related properties (tumor size and tumor growthrate) from drug-related properties (potency, drug-inducedkill rate) and incorporates the following assumptions: (i)MDM2 is amplified, resulting in the inhibition of p53-induced apoptosis in the SJSA xenograft tumors, (ii) pro-liferation of SJSA cells thereby becomes the predominantprocess, resulting in unperturbed tumor growth, (iii) afteradministration of RG7388, MDM2 is sufficiently blockedfrom negatively regulating p53, which in turn inducescellular growth arrest and apoptosis, and (iv) the delay intumor response to drug effect is due to signal transductionprocesses intervening between drug–receptor interactionand the killing event with the killing effect occurringthrough a signal transduction cascade (14). Similar model-ing approaches have been reported to assess antitumoreffect in xenograft mice (15, 16). These models allow forthe estimation of a concentration resulting in tumor stasis.The drug effect of RG7388 is assumed to depend on itsplasma concentration [C(t)] and the tumor volume. Thetumor volume predictions for different doses of RG7388versus experimental data are shown (Fig. 3A).
Model-based approach to select intermittent dosingregimens targeting tumor stasis
This model was then used to select intermittent dosingregimens that are predicted to result in tumor stasis. ThePKPD model quantitatively characterized how the phar-macokinetics of RG7388 is linked to the antitumor effectin SJSA osteosarcoma. This in turn enabled prospectiveprediction of intermittent and continuous dosing sche-dules targeting tumor stasis (Fig. 3B). A variety of dosingregimens were selected on the basis of the initial clinicalplan of a 28-day schedule, ranging from daily dosingutilized in the RG7112 clinical program, to once/twiceper week, to five daily doses followed by a 23-day drugholiday.
All of the intermittent dosing schedules suggested bymodeling and simulation (M&S) were tested experimental-ly and confirmed profound reduction in mean tumor vol-ume compared with vehicle control (P < 0.05 all vs.control; Fig. 4). In addition, survival was also examined,and all of the dosing schedules resulted in increased lifespan compared with vehicle control (P < 0.05 all vs.control; Fig. 4). Several of the intermittent dosing regimensdemonstrating dramatic TGI and increased lifespan werestatistically equivalent (P > 0.05). These regimens includedan 80 mg/kg dose given on a 5 days-on and 23 days-offschedule (>100% TGI with 6 partial and 2 complete regres-sions; 127% ILS), 100 mg/kg given on a 2 days-on and 5days-off schedule (>100% TGI with 9 partial regressionsand a 188% ILS), 200mg/kg (two 100mg/kg doses admin-istered 8 hours apart) given once a week (>100%TGI with 7partial regressions and a 162% ILS), and chronic adminis-tration at 30mg/kg given daily for 28 days (>100%TGIwith7 partial regressions and a 127% ILS). The tumor volumechanges predicted by M&S were in alignment with thegenerated data, demonstrating the robustness of themodel-ing techniques.
Induction of apoptosis and antiproliferation in vivowith single dose and short schedule (5-day) RG7388administration
The pharmacodynamic effects of the highest RG7388dose based on traditional tolerability measures (i.e., weightloss) andpredicted for intermittent dosing (5daysof 80mg/kg) by the M&S was then compared with both the highestfeasible single dose of 200 mg/kg (based on viscositylimitations of the suspension) and the 80 mg/kg dose inSJSA xenograft tumors.
Effects on apoptosis were examined by BLI of activatedcaspases-3 and -7. Activated caspases-3 and -7 are consid-ered to be an early sign of apoptosis, as their presence can bedetected before plasma membrane blebbing and DNAfragmentation. Bioluminescent detection of activated cas-pases-3 and -7 was monitored in SJSA1-luc2 tumor-bearingmice using Z-DEVD-aminoluciferin as a substrate for acti-vated caspases-3 and -7. When the DEVD peptide sequenceis cleaved by the activated caspases, aminoluciferin isreleased and becomes a substrate for the luciferase enzymeproduced by SJSA-luc2 cells. Therefore, the luminescentsignal is emitted only in apoptotic cells. An additionalmarker of apoptosis, PARP, was examined. PARP is a sub-strate of caspase-3 and is cleaved to cPARP-1 during
Days post-tumor cell implant
Mea
n tu
mor
vol
ume
±SE
M, (
mm
3 )
0500
1,0001,5002,0002,5003,0003,5004,0004,5005,0005,500
353025201510
P < 0.001 for all groups versus vehicle control.
Note: %TGI assessed at d21 of dosing.
ILS (%)PRTGI (%)RG7388 Regimen
---dq elciheV 0 ---100 mg/kg qw 7719450 mg/kg 2 d on/ 5 off ᵡ 4, qd 11939980 mg/kg 5 d on/ 23 off, qd 1276>100100 mg/kg 2 d on/ 5 off ᵡ 4, qd 1889>10020 mg/kg 20 d on/ 8 off, qd 77598100 mg/kg 2 d on/ 12 off ᵡ 2, qd 1197>10050 mg/kg 4 d on/ 10 off ᵡ 2, qd 112699200 mg/kg (2 100mg/kg 8 hours apart) qw 1627>10030 mg/kg qd 1277>100
Days8040200
0
20
40
60
80
100
Per
cent
sur
viva
l
60
Figure 4. Efficacy data. A, percent TGI and (B) survival curves-schedulespredicted to produce stasis based on modeling (n ¼ 10 mice/group).
Higgins et al.
Clin Cancer Res; 20(14) July 15, 2014 Clinical Cancer Research3748
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
apoptosis. cPARP-1 was detected using IHC. Antiprolifera-tion was also examined via IHC using Ki67, which is acellular marker specific for proliferating cells.At 48 hours after a single dose of RG7388, tumors from
treated mice produced a maximal statistically significantinduction in luciferase signal (Fig. 5A), indicative of apo-ptosis via activated caspases-3 and -7, at the 80 mg/kg and200 mg/kg dose levels as compared with vehicle controls(P < 0.05). Representative bioluminescent images frommice bearing vehicle and RG7388-treated tumors show thedose-dependent increase in luminescence and, therefore,apoptosis (Fig. 5A). The apoptotic response to RG7388 wasalso examined by measuring cPARP-1 via IHC. Similar toactivated caspase-3 and -7, the largest increase in cPARP-1levels versus vehicle control was seen at 48 hours after asingle dose of 200 mg/kg (P < 0.05; Fig. 5B). In addition,IHC examination of the proliferation marker Ki67 revealed
themost significant decrease in positive staining at 48 hoursafter a single 200mg/kg dose comparedwith vehicle control(P < 0.05; Fig. 5C).
When RG7388 was administered for 5 days at 80 mg/kg, the maximal effect on apoptosis via activated caspase-3 and -7 levels versus vehicle control was observed on day3 (P < 0.05; Fig. 6A). Moreover, cPARP-1 levels werehighest on day 3 versus vehicle control after 5 days ofRG7388 administration (Fig. 6B). Finally, with 5 daysof administration of RG7388, the maximal decrease inKi67 versus vehicle control also occurred on day 3 (P <0.05; Fig. 6C). Therefore, apoptotic and antiproliferativepharmacodynamic effects were comparable with thecontinuous (80 mg/kg daily) and with the single dose(200 mg/kg) dosing regimen. This supports our model-based hypothesis that a continuous dosing regimen is notneeded.
*,P < 0.05, Kruskal–Wallis one-way ANOVA with Dunn post-test.
B
Vehicle 200 mg/kg80 mg/kg
200 mg/kg80 mg/kgVehicle
cPARP-1 (IHC): 48 h after dose
A
Tot
al fl
ux (
p/s)
cPA
RP
-1/ v
iabl
e ar
ea (
%)
Ki67 (IHC): 48 h after dose
Vehicle 80 mg/kg 200 mg/kg
C
Ki6
7-P
ositi
ve n
ucle
i (%
)
Cleaved caspases 3 and 7:48 hours after dose2.5×109
2.0×109
1.5×109
1.0×109
5.0×108
0
20
15
10
5
0
60
40
20
0
Figure 5. RG7388-mediatedapoptosis is observed with dailyand intermittent schedules in vivo(dose and time dependent). A,dose-dependent apoptosis withsingle-dose administration BLI ofcleaved caspases 3 and 7. B, IHCanalysis of apoptosis markercPARP and (C) proliferation markerKi67 48 hours after single dose ofRG7388.
Optimization of MDM2 Antagonist RG7388 for Clinical Testing
www.aacrjournals.org Clin Cancer Res; 20(14) July 15, 2014 3749
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
DiscussionMDM2 inhibitors can be used in MDM2 gene amplified
tumors to restore functional p53 activity. In preclinicalmodels, RG7388, a second-generation MDM2 inhibitor,
has demonstrated the ability to activate apoptosis andinduce tumor stasis with continuous treatment (10). Here,we provide evidence that continuous dosing of RG7388 isnot required for sustained activity. Although in vitro apo-ptotic effects were delayed relative to noncontinuous treat-ment, once triggered, the p53 pathway activation of apo-ptosiswas irreversible in SJSA1 cells even after drug removal.Interestingly, Western blot analysis showed that p53 levelsremained elevated significantly longer thanMDM2andp21after RG7388 washout. The loss of MDM2 and p21 proteinlevels could be the result of changes in protein half-life onp53 activation as evidenced in published results whereaccelerated MDM2 autodegradation was reported duringp53-mediated DNA damage response (17). However, p53function and stability may also be tightly controlled by itssubcellular localization (18, 19). Therefore, it is possiblethat in a cell triggered into apoptosis, nuclear or mitochon-drial localized p53 remains fairly stable following RG7388washout, whereas MDM2 and p21 are accessible to betargeted by the cytoplasmic ubiquitin proteasome pathway.From a molecular stand point, the continued presence ofp53 after RG7388 removal may be critical for sustainedactivity. These observations were also made with RG7112but necessitated much higher drug concentrations than themore potent RG7388 to retain elevated p53 levels onwashout (data not shown). Collectively, our in vitro resultssupport the hypothesis that RG7388 can induce sustainedantitumor activity when given on an intermittent dosingschedule. However, as previous studies in cell lines haveshown differential apoptotic response (6, 8), we mustconcede that additional investigation is warranted in otherMDM2 antagonist sensitive xenograft models. Further-more, p53 persistence as seen in the SJSA1 after washoutmust be confirmed in other cell lines where significantdifferences in growth characteristics and phenotypes willperhaps necessitate optimized-type specific dosing sche-dules. For the studies outlined in this report, the SJSA1represents an ideal proof-of-concept model to investigatethe feasibility of this approach.
An intermittent dosing schedule is expected to reduce thepotential for and/or severity of adverse events in clinicaltrials such as hematologic toxicities because it allows thesystem to recover. Drug-induced thrombocytopenia andneutropenia observed with RG7112 is assumed to act onprogenitor cells in bone marrow. Therefore, a decrease ofplatelets and/or neutrophils in circulation occurs with a lagtime relative to drug treatment. Washout of the drug is thenfollowed by a rebound and return to baseline of the bloodcells after replenishment and completed maturation phaseof the progenitor cells.
A model-based approach was thereby applied to guidethe selection of intermittent dosing regimen achievingtumor stasis. As a first step, a PKPD study was conductedto relate the pharmacokinetics to the time course of theanticancer effects by modeling. A mechanism-based PKPDmodel was developed assuming a delayed response relativeto the drug concentration motivated by the in vitro findingsshowing a delayed apoptotic response relative to RG7388
Caspase-3/7 imaging (BLI)A
B
C
Cleaved PARP-1 (IHC)
Ki67 (IHC)
2.0×109
1.5×109
1.0×109
5.0×108
0
2.0×109
1.5×109
1.0×109
5.0×108
0
Sin
gle
dose
tota
l flu
x (p
/s)
2.0×109
1.5×109
1.0×109
5.0×108
0
Tota
l flu
x (p
/s)
Vehi
cle
Vehi
cle
qb ×
1
qb ×
3
qb ×
5
Vehi
cle
qb ×
1
qb ×
3
qb ×
5
Vehi
cle
qb ×
1
qb ×
3
qb ×
5
16 h
24 h
48 h
Vehi
cle
16 h
24 h
48 h
Vehi
cle
16 h
24 h
48 h
72 h
120
h
Vehi
cle
16 h
24 h
48 h
72 h
120
h
Vehi
cle
16 h
24 h
48 h
Vehi
cle
16 h
24 h
48 h
72 h
120
h
Mul
tiple
dos
eto
tal f
lux
(p/s
)S
ingl
e do
secP
AR
P-1
/via
ble
area
(%
)M
ultip
le d
ose
cPA
RP
-1/v
iabl
e ar
ea (
%)
cPA
RP
-1/v
iabl
e ar
ea (
%)
80 mg/kg
80 mg/kg
80 mg/kg
80 mg/kg
200 mg/kg
80 mg/kg
80 mg/kg
200 mg/kg
200 mg/kg
6
4
2
0
20
15
10
5
0
10
8
6
4
2
0
80
60
40
20
0
80
60
40
20
0
Sin
gle
dose
Ki6
7-po
sitiv
e nu
clei
(%
)M
ultip
le d
ose
Ki6
7-po
sitiv
e nu
clei
(%
)
Ki6
7-po
sitiv
e nu
clei
(%
)
60
40
20
0
* ,P < 0.05, Kruskal-Wallis one-way ANOVA with Dunn post-test.
Figure 6. A, caspase-3/7 imaging (BLI)-single dose versus multiple dailydoses. B, cleaved PARP-1 (IHC)-single dose versusmultiple daily doses.C, RG7388-mediated antiproliferative effects observed with daily andintermittent schedules in vivo.
Higgins et al.
Clin Cancer Res; 20(14) July 15, 2014 Clinical Cancer Research3750
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
exposure. Similar modeling approaches have been reportedto assess antitumor effect in xenograft mice (15, 16). Thesemodels allow for the estimation of a concentration resultingin tumor stasis. Rocchetti and colleagues (20) demonstratedthat this estimated concentration correlated with typicaltherapeutic doses for select chemotherapeutic agents,emphasizing the clinical translational potential of thisapproach. The tumor stasis concentration can be regardedas a reference concentration to be achieved in vivo forattaining significant anticancer activity. In addition, Luo andcolleagues (21) demonstrated, for cetuximab, that the activeserum concentration determined in a preclinical xenograftmodel (average concentration at steady state) correlatedwellwith the active drug concentration achieved in patients withcancer. As with any model, there are associated limitations.Our model assumes the tumor mass to consist solely oftumor cells and does not consider the interaction betweenstroma and tumor compartment. Moreover, drug-inducedresistance, which can certainly influence outcome, is exclud-ed. Nonetheless, our semimechanistic model described thePK and time course of tumor growth across all doses rea-sonably well. Therefore, based on these modeling results,simulations were conducted to identify intermittent dosingregimens achieving tumor stasis. A follow-up study wasconducted to evaluate the observed tumor growth curvescompared with the prospectively predicted profiles.Both daily and intermittent dosing with RG7388 was
effective at achieving tumor stasis as predicted by the PKPDmodel. Therefore, despite the inherent short half-lives of p53and MDM2, continuous suppression of the p53–MDM2interaction was not required for optimal antitumor activity.Pharmacodynamic effects of single versus short-term
(5 days) dosing were examined in vivo using the MDM2-amplified SJSA1 osteosarcoma model. In vivo analysis ofapoptosis and antiproliferative effects revealed that thesingle highest feasible dose of RG7388 (200 mg/kg) wassufficient to activate apoptosis and decrease proliferation,with amaximal effect at 48 hours. Likewise, a dose predictedas optimal via M&S on a 5-day schedule (80 mg/kg) alsoactivated apoptosis and decreased proliferation, with amaximal effect after 3 days.RG7112, a cis-imidazoline, and RG7388, a pyrroli-
dine compound, both bind to MDM2 and prevent theassociation of MDM2 with p53. Although these com-pounds have a shared mechanism of action and both
reversibly bind to MDM2, RG7388 can be used at signifi-cantly lower doses due to superior potency and specificity(8, 10). These superior properties of RG7388 may alsoallow RG7388 to be dosed intermittently, which shouldcircumvent toxicities associatedwith prolonged continuoustreatment with RG7112 while maintaining clinical benefit.The preclinical data presented herein provide proof-of-principle that intermittent RG3788 dosing can provide thesame activity as daily dosing.
Although RECIST responses were achieved in previousRG7112 clinical trials, patients had difficulty toleratingRG7112 given on the requisite daily schedule (9). On thebasis of those results and on thework presented here, a onceweekly � 3 followed by 13 days of rest, 28-day cycleschedule, and a 5 consecutive days of a 28-day cycle sched-ule were utilized for the initial phase I RG7388 clinical trial(trial registration identifier: NCT01462175). The ability toadminister RG7388 intermittently should provide bettertolerability for patients, which may in turn result in morerobust clinical responses.
Disclosure of Potential Conflicts of InterestB. Higgins is an employee of Roche. No potential conflicts of interest
were disclosed by the other authors.
Authors' ContributionsConception and design: B. Higgins, G. Kelli, A. Walz, C. Tovar, E. Lee,K. Kolinsky, D. Heimbrook, L.T. Vassilev, K. PackmanDevelopment of methodology: B. Higgins, A. Walz, Z. Filipovic, S. Hus-sain, E. Lee, K. Kolinsky, S. Tannu, V. Adames, R. Garrido, K. PackmanAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): B. Higgins, C. Tovar, Z. Filipovic, S. Hussain,E. Lee, S. Tannu, R. Garrido, M. LinnAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): B. Higgins, G. Kelli, A. Walz, Z. Filipovic,E. Lee, K. Kolinsky, S. Tannu, R. Garrido, M. Linn, K. PackmanWriting, review, and or revision of the manuscript: B. Higgins, A. Walz,C. Tovar, M. Linn, C. Meille, K. PackmanAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): B. Higgins, A. Walz, C. Tovar,Z. Filipovic, K. Kolinsky, S. Tannu, C. Meille, K. PackmanStudy supervision: B. Higgins, Z. Filipovic, E. Lee, S. Tannu, L.T. Vassilev,K. Packman
Grant SupportThis study was supported by Roche.The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisementin accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received February 24, 2014; revised April 11, 2014; accepted April 21,2014; published OnlineFirst May 8, 2014.
References1. Vogelstein B, Lane D, Levine AJ. Surfing the p53 network. Nature
2000;408:307–10.2. Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 mutations in
human cancers. Science 1991;253:49–53.3. Momand J, Zambetti GP, Olson DC, George D, Levine AJ. The mdm-2
oncogene product forms a complex with the p53 protein and inhibitsp53-mediated transactivation. Cell 1992;69:1237–45.
4. Michael D, Oren M. The p53-Mdm2 module and the ubiquitin system.Semin Cancer Biol 2003;13:49–58.
5. Vassilev LT, Vu BT, Graves B, Carvajal D, Podlaski F, Filipovic Z, et al.In vivo activation of the p53 pathway by small-molecule antagonistsof MDM2. Science 2004;303:844–8.
6. Tovar C, Rosinski J, Filipovic Z, Higgins B, Kolinsky K, Hilton H, et al.Small-molecule MDM2 antagonists reveal aberrant p53 signaling incancer: implications for therapy. Proc Natl Acad Sci U S A 2006;103:1888–93.
7. Shangary S, Wang S. Small-molecule inhibitors of the MDM2-p53protein-protein interaction to reactivate p53 function: a novelapproach for cancer therapy. Annu Rev Pharmacol Toxicol 2009;49:223–41.
8. Tovar C, Graves B, Packman K, Filipovic Z, Higgins B, Xia M, et al.MDM2small-molecule antagonistRG7112activatesp53 signaling andregresses human tumors in preclinical cancer models. Cancer Res2013;73:2587–97.
www.aacrjournals.org Clin Cancer Res; 20(14) July 15, 2014 3751
Optimization of MDM2 Antagonist RG7388 for Clinical Testing
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
9. Ray-Coquard I, Blay JY, Italiano A, Le Cesne A, Penel N, Zhi J, et al.Effect of theMDM2antagonist RG7112 on the P53 pathway in patientswith MDM2-amplified, well-differentiated or dedifferentiated liposar-coma: an exploratory proof-of-mechanism study. Lancet Oncol2012;13:1133–40.
10. Ding Q, Zhang Z, Liu JJ, Jiang N, Zhang J, Ross TM, et al. Discovery ofRG7388, a potent and selective p53-MDM2 inhibitor in clinical devel-opment. J Med Chem 2013;56:5979–83.
11. Higgins B, Tovar C, Glenn K, Walz A, Filipovic Z, Zhang YE, et al.Antitumour activity of the MDM2 antagonist RG7388 [abstract]. In:Proceedings of the AACR-NCI-EORTC International Conference:Molecular Targets and Cancer Therapeutics; 2013 Oct 19–23; Boston,MA. Philadelphia, PA: AACR; 2013. Abstract nr B55.
12. Lalonde RL, Kowalski KG, Hutmacher MM, Ewy W, Nichols DJ, Milli-gan PA, et al. Model-based drug development. Clin Pharmacol Ther2007;82:21–32.
13. Lobo ED, Balthasar JP. Pharmacodynamic modeling of chemothera-peutic effects: application of a transit compartment model to charac-terize methotrexate effects in vitro. AAPS Pharm Sci 2002;4:E42.
14. Sun YN, Jusko WJ. Transit compartments versus gamma distributionfunction to model signal transduction processes in pharmacodynam-ics. J Pharm.Sci 1998;87:732–7.
15. SimeoniM,Magni P, CammiaC, DeNicolaoG,Croci V, Pesenti E, et al.Predictive pharmacokinetic-pharmacodynamic modeling of tumorgrowth kinetics in xenograft models after administration of anticanceragents. Cancer Res 2004;64:1094–1101.
16. Yang J, Mager DE, Straubinger RM. Comparison of two pharmaco-dynamic transduction models for the analysis of tumor therapeuticresponses in model systems. AAPS J 2010;12:1–10.
17. Stommel JM,Wahl GM.AcceleratedMDM2auto-degradation inducedby DNA-damage kinases is required for p53 activation. EMBO J2004;23:1547–56.
18. O'Brate A, Giannakakou P. The importance of p53 location: nuclear orcytoplasmic zip code? Drug Resist Updat 2003;6:313–22.
19. Davis JR, Mossalam M, Lim CS. Controlled access of p53 to thenucleus regulates its proteasomal degradation by MDM2. Mol Pharm2013;10:1340–9.
20. Rocchetti M, SimeoniM, Pesenti E, DeNicolaoG, Poggesi I. Predictingthe active doses in humans from animal studies: a novel approach inoncology. Eur J Cancer 2007;43:1862–8.
21. Luo FR, Yang Z, Dong H, Camuso A, McGlinchey K, Fager K, et al.Correlation of pharmacokinetics with the antitumor activity of cetux-imab in nudemice bearing theGEOhuman colon carcinomaxenograft.Cancer Chemother Pharmacol 2005;56:455–64.
Clin Cancer Res; 20(14) July 15, 2014 Clinical Cancer Research3752
Higgins et al.
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460
2014;20:3742-3752. Published OnlineFirst May 8, 2014.Clin Cancer Res Brian Higgins, Kelli Glenn, Antje Walz, et al. Treatment by Using a Model-Based ApproachPreclinical Optimization of MDM2 Antagonist Scheduling for Cancer
Updated version
10.1158/1078-0432.CCR-14-0460doi:
Access the most recent version of this article at:
Material
Supplementary
http://clincancerres.aacrjournals.org/content/suppl/2014/05/08/1078-0432.CCR-14-0460.DC1
Access the most recent supplemental material at:
Cited articles
http://clincancerres.aacrjournals.org/content/20/14/3742.full#ref-list-1
This article cites 20 articles, 6 of which you can access for free at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
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/20/14/3742To request permission to re-use all or part of this article, use this link
on July 1, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from
Published OnlineFirst May 8, 2014; DOI: 10.1158/1078-0432.CCR-14-0460