Dual Blockade of the PI3K/AKT/mTOR (AZD8055) and RAS/MEK ... · pathway, and cross-talk to...

Cancer Therapy: PreclinicalSee related commentary by Jahangiri and Weiss, p. 5811

Dual Blockade of the PI3K/AKT/mTOR (AZD8055) andRAS/MEK/ERK (AZD6244) Pathways Synergistically InhibitsRhabdomyosarcoma Cell Growth In Vitro and In Vivo

Jane Renshaw1,2, Kathryn R. Taylor1,2, Ryan Bishop1,2, Melanie Valenti2, Alexis De Haven Brandon2,Sharon Gowan2, Suzanne A. Eccles2, Ruth R. Ruddle2, Louise D. Johnson2, Florence I. Raynaud2,Joanna L. Selfe2,3, Khin Thway2,3,4, Torsten Pietsch5, Andrew D. Pearson1, and Janet Shipley2,3

AbstractPurpose: To provide rationale for using phosphoinositide 3-kinase (PI3K) and/or mitogen-activated

protein kinase (MAPK) pathway inhibitors to treat rhabdomyosarcomas, a major cause of pediatric and

adolescent cancer deaths.

Experimental Design: The prevalence of PI3K/MAPK pathway activation in rhabdomyosarcoma clinical

samples was assessed using immunohistochemistry. Compensatory signaling and cross-talk between PI3K/

MAPK pathways was determined in rhabdomyosarcoma cell lines following p110a short hairpin RNA–

mediated depletion. Pharmacologic inhibition of reprogrammed signaling in stable p110a knockdown

lines was used to determine the target-inhibition profile inducing maximal growth inhibition. The in vitro

and in vivo efficacy of inhibitors of TORC1/2 (AZD8055), MEK (AZD6244), and P13K/mTOR (NVP-

BEZ235) was evaluated alone and in pairwise combinations.

Results:PI3Kpathway activationwas seen in82.5% rhabdomyosarcomaswith coactivatedMAPK in36%

and 46% of alveolar and embryonal subtypes, respectively. p110a knockdown in cell lines over the short

and long termwas associatedwith compensatory expression of other p110 isoforms, activation of theMAPK

pathway, and cross-talk to reactivate the PI3K pathway. Combinations of PI3K pathway and MAP–ERK

kinase (MEK) inhibitors synergistically inhibited cell growth in vitro. Treatment of RD cells with AZD8055

plus AZD6244 blocked reciprocal pathway activation, as evidenced by reduced AKT/ERK/S6 phosphory-

lation. In vivo, the synergistic effect on growth and changes in pharmacodynamic biomarkers was

recapitulated using the AZD8055/AZD6244 combination but not NVP-BEZ235/AZD6244. Pharmacoki-

netic analysis provided evidence of drug–drug interaction with both combinations.

Conclusions: Dual PI3K/MAPK pathway activation and compensatory signaling in both rhabdo-

myosarcoma subtypes predict a lack of clinical efficacy for single agents targeting either pathway,

supporting a therapeutic strategy combining a TORC1/2 with a MEK inhibitor. Clin Cancer Res; 19(21);

5940–51. �2013 AACR.

IntroductionRhabdomyosarcomas are the most common soft tissue

sarcomas in children of ages 0 to 14 years, accounting foraround 50% of all cases in this age group (reviewed in

ref. 1). The two major histologic subtypes, alveolar rhab-domyosarcoma and embryonal rhabdomyosarcoma carrydistinctmorphologic and genetic alterations: Themajorityof alveolar rhabdomyosarcomas harbor PAX3-FOXO1 orPAX7-FOXO1 chimeric transcription factors (2), and thepresence of the PAX3-FOXO1 fusion transcript has beenassociated with poor prognosis (3). Embryonal rhabdo-myosarcomas generally have a more favorable outcomeand carry no consistent chromosomal translocations,although allelic imbalances at 11p15.5 are commonlyidentified (1, 2). Dysregulation of the RAS signaling path-way is likely relevant to the pathogenesis of embryonalrhabdomyosarcoma (4, 5), and also insulin-like growthfactor (IGF) signaling to the pathogenesis of all rhabdo-myosarcomas (6). The use of multiagent chemotherapy,surgery, and radiation has improved the outcome forpatients with rhabdomyosarcoma with favorable features,although at a cost of disfigurement and long-term sequelae

Authors' Affiliations: Divisions of 1Clinical Studies, 2Cancer Therapeutics,and 3Molecular Pathology, The Institute of Cancer Research, Sutton,Surrey; 4Histopathology Department, The Royal Marsden Hospital NHSFoundation Trust, London, United Kingdom; and 5Department of Neuro-pathology, University of Bonn, Bonn, Germany

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

CorrespondingAuthors: JaneRenshaw,Division of Cancer Therapeutics,The Institute of Cancer Research, 15 Cotswold Road, Sutton, Surrey SM25NG, United Kingdom. Phone: 44-20-8722-4327; Fax: 44-20-8722-4084;E-mail: [email protected]; and Janet Shipley, [email protected]

doi: 10.1158/1078-0432.CCR-13-0850

�2013 American Association for Cancer Research.

ClinicalCancer

Research

Clin Cancer Res; 19(21) November 1, 20135940

on January 11, 2020. © 2013 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 5, 2013; DOI: 10.1158/1078-0432.CCR-13-0850

http://clincancerres.aacrjournals.org/

(7). Unfortunately, the prognosis for metastatic andrelapsed rhabdomyosarcoma tumors remains poor, andthe need for novel, less toxic therapeutic strategies ispressing.High levels of phosphorylated AKT, reported in rhabdo-

myosarcoma cell lines and a significant proportion ofprimary tumors, are indicative of constitutive activation ofthe PI3K/AKT/mTOR pathway and suggest that rhabdo-myosarcoma may be sensitive to the targeted inhibition ofthis pathway (8, 9). However, previous reports in othertumor types have indicated that activatedmitogen-activatedprotein kinase (MAPK) signaling mediates resistance tophosphoinositide 3-kinase (PI3K) inhibitors (10, 11).Indeed, NRAS mutations, identified in approximately20% of patients with embryonal rhabdomyosarcoma (4),may identify those tumors unlikely to respond to PI3Kinhibitors, as has been shown for KRAS-mutated lungcancers (10). Conversely, intrinsic resistance to MAP–ERKkinase (MEK) inhibitors has been associated with strongPI3K signaling in colorectal and breast cancer cell lines (12,13). Thus, dual activation of both PI3K/AKT/mTOR andRAS/RAF/MEK/ERK pathways is likely to result in resistanceto the targeting of either pathway alone. Encouragingly,coinhibition of both pathways has shown use in reducingtumor growth in a variety of xenograft cancer models(10, 13–15), and clinical trials of such combinations areunder way in adults.Here, we assessed the prevalence of coactivation of the

PI3K and MAPK pathways in a large series of well-charac-

terized rhabdomyosarcoma clinical samples by conductingimmunohistochemistry for phosphorylated markers ofboth pathways, as well as downstream S6. We used shorthairpin RNA (shRNA)–induced p110a knockdown toexplore compensatory signaling and cross-talk between thePI3K/MAPK pathways in rhabdomyosarcoma cell lines andconducted comprehensive preclinical in vitro and in vivostudies of PI3K and/or MAPK pathway–targeted therapies.We provide evidence to support the class of PI3K inhibitorthat is most effective, identify mechanisms likely to beresponsible for treatment failure, and undertook combina-tion studies that address this problem. Our results predict ageneral lack of clinical efficacy of PI3K/AKT/mTOR or RAS/RAF/ERK pathway monotherapy in rhabdomyosarcoma,and we propose a dual-targeted strategy for clinical testing.

Materials and MethodsRhabdomyosarcoma tissue from patients andimmunohistochemistry

A tissue microarray (TMA) of formalin-fixed, paraffin-embedded diagnostic tumor material from 79 patients withrhabdomyosarcoma (25 alveolar and 54 embryonal byhistology), with approval for the study (Local ResearchEthics Committee protocol Nos 1836 and 2015 andMulti-Regional Research Ethics Committee/06/4/71), hasbeen described previously (16). Following antigen retrieval(0.1 mol/L sodium citrate, pH 6.0, microwave 600 W, 30minutes), and blocking (TBS with 5% milk, 2% normalrabbit serum, 30 minutes), TMA sections (4 mm) wereimmunostained using antibodies against phospho-AKT(Ser473), phosphor-ERK1/2 (Thr202/Tyr204), and phos-pho-S6 (Ser235/236; 1:200 dilution; Cell Signaling Tech-nology) with detection using the avidin–biotin complexmethod (DAKO) visualized by 3,30-diaminobenzidine(DAB). Slides were lightly counterstainedwith hematoxylin.Cores were scored blind by a pathologist as negative, weak,moderate, or strong for intensity and were considered to bepositive if at least 10% of cells in the core showed staining.

Cell culture, compounds, and GI50 estimationThe source of the human tumor cell lines RH30, RMS-1,

SCMC, and RH41 derived from alveolar rhabdomyosarco-ma, and RD, RMS-YM, and CT10 derived from embryonalrhabdomyosarcoma is described elsewhere (16). All celllines were grown in Dulbecco’s modified Eagle medium(DMEM; Sigma) supplemented with 10% FBS (Biosera;Labtech International Ltd.) in 5% CO2 in air at 37�C. Cellline identity was validated by analyzing short-tandemrepeats using the Promega Power Plex 1.2 system, accordingto the manufacturers’ instructions, within 6 months of theexperiments described. Results were matched with thoseavailable from repositories which hold the lines (RH30 andRD) or those produced immediately after delivery of the celllines to our laboratory. All compounds were purchasedfromAxonMedChem, exceptNVP-BMK120 (Selleckchem).Compounds were dissolved in the appropriate solvent at 10mmol/L and diluted in tissue culture medium to working

Translational RelevanceMolecular rationale to underpin the therapeutic use

of PI3K/AKT/mTOR and RAS/RAF/MAPK pathway inhi-bitors is critical. Here, we show coactivation of thesepathways in 43% of primary rhabdomyosarcomas withmost of the remainder exhibiting activation of thephosphoinositide 3-kinase (PI3K), but not mitogen-activated protein kinase (MAPK), pathway. The formerpredicts resistance to single pathway–targeted agentsand the latter, potential sensitivity to PI3K pathwayinhibitors. However, we also show extensive compen-satory signaling and cross-talk between the PI3K/MAPKpathways on inhibition of either pathway alone, both invitro and in vivo. Therefore, simultaneous inhibition ofboth pathways is essential for effective treatment ofrhabdomyosarcomas. The in vivo synergistic antitumorresponse achieved with the combination AZD8055/AZD6244, but not NVP-BEZ235/AZD6244, was ref-lected in reduction of the pharmacodynamic biomar-kers pERK/pS6/pser473AKT. Preliminary pharmacoki-netic analyses revealed drug–drug interactions betweenthe PI3K/MAPK inhibitors studied, indicating a require-ment for pharmacokinetic analyses after repeat dosingin future clinical trials of the promising AZD8055/AZD6244 combination.

TORC1/2 plus MEK Inhibition Synergistically Inhibits Rhabdomyosarcoma

www.aacrjournals.org Clin Cancer Res; 19(21) November 1, 2013 5941




concentrations. GI50 values (concentrations causing 50%inhibition of cell proliferation after 72 hours of continuousexposure), were determined in 96-well plates using theMTSassay (Promega). For the construction of isobolograms, theGI50 values of compound A and B were determined alongwith the GI50 values of compound B in the presence ofvarious fixed sub-GI50 concentrations of compound A. TheGI50 values of compound A and B were then plotted on thex- and y-axis along with the GI50 values of compound Bversus the fixed concentration of compound A.

Lentiviral shRNA knockdown of PIK3CA, Westernblotting, and FACS analysis

Lentiviral particles for shRNAknockdownofPIK3CAwereproduced following transfection of the lentiviral packagingvectors pMD2.G (3 mg), pMDLg/RRE (5 mg), and pRSV-Rev(2.5 mg; Addgene), and either the PIK3CA shRNA vector (10mg; TRCN0000039603; Sigma) or the control shRNA vector(10 mg; SHCOO2; Sigma) into HEK293T cells as describedpreviously (16). Control (CONSH) or PIK3CA shRNA len-tiviral particles were transduced into RH30, RMS-1, RD, andRMS-YM cells (in 2� 6-well plates per line) at a multiplicityof infection of 10 in the presence of polybrene (4 mg/mL;Sigma-Aldrich) and transduced cells were selected withpuromycin (2.5 mg/mL) from day 2. On days 2, 5, and 8,viable cells were counted (ViCell series automated CellViability Analyzer; Beckman Coulter), cells were fixed with70% ethanol for fluorescence-activated cell sorting (FACS)analysis, and proteins were extracted for Western blotting.

ForWestern blotting, following quantitation (BCAProteinAssay Kit; Pierce), 10 mg of each lysate was separated usingNuPageBis–Tris or Tris–Acetate gels (Invitrogen), transferredonto Highbond polyvinylidenedifluoride (PVDF) mem-branes (Amersham Biosciences) and probed with 1:1,000dilution of primary antibody (Cell Signaling Technology)overnight at 4�C, and 1:10,000 dilution of horseradishperoxidase secondary antibody (Amersham Biosciences) for2 hours at room temperature. Signal was detected with ECLPlus reagent (Amersham) and visualized and analyzed usinga STORM 860 PhosphorImager and ImageQuant software(AmershamBiosciences).

For FACS analysis, fixed cells were spun, resuspended in800 mL PBS, 100 mL RNase (0.1mg), and 100 mL propidiumiodide (PI; 40 mg), and incubated at 37�C for 30 minutes.Labeled cells were stored at 4�Cuntil cell-cycle analysis on aBDLSRII FACS analyzer (BD) using BDFACSDiVa softwarewithdoublet discrimination. For each sample 10,000 eventswere recorded.

Pharmacodynamic, pharmacokinetic, and xenograftefficacy studies

All animal experiments were carried out in accordancewith United Kingdom Home Office Regulations under theAnimals (Scientific Procedures) Act 1986 and United King-dom Co-coordinating Committee on Cancer Researchguidelines (17). RD cells (5 � 106 cells in 25% Matrigel;BD Biosciences) were subcutaneously implanted bilaterally(for initial pharmacodynamic and pharmacokinetic stud-

ies), or in one flank (for tumor efficacy studies) into female6- to 8-week-old CrTac:Ncr-Fox1(nu; Ncr) athymic mice(Charles River Laboratories). AZD6244 [10 mg/kg in10% dimethyl sulfoxide (DMSO) water], AZD8055 (10 or20 mg/kg in acidified water), and NVP-BEZ235 (25 mg/kgin 10%NMP in 90%PEG300)were given orally (0.1mL/10g bodyweight of vehicle) alone, or in combination. Controlanimals received the equivalent volume of appropriatevehicle(s). For initial pharmacokinetic and pharmacody-namic analysis, mice (n ¼ 3) bearing tumors of sufficientsize were euthanized at various times following dosing,tumors were excised, bisected and snap-frozen, and bloodwas collected, centrifuged, and the plasma was frozen forstorage at �80�C. For tumor-efficacy studies, dosing com-menced when the tumors were well established (�5 mmmean diameter; n ¼ 6/treatment group) and continueddaily for 19 to 22 days. Animals were weighed at regularintervals and observed for adverse effects. Tumors weremeasured across twoperpendicular diameters, and volumeswere calculated using the formula V ¼ 4/3p[(d1þd2)/4]3.On termination of the experiment, mice were bled, tumorswere excised, and weighed, and samples were processed asdescribed earlier.

Quantitative pharmacokinetic analysis was conducted byliquid chromatography/tandem mass spectrometry andmultiple reaction monitoring, using a modification of thepreviously describedmethod (ref. 18; SupplementaryMeth-ods). Tumor pharmacodynamic biomarkers were assessedby a Meso Scale Discovery (MSD) multispot electroche-miluminescence immunoassay system to detect phospho(T/Y:202/204:185/187)ERK1/2/total ERK1/2, phospho(Ser473)AKT/total AKT, and phospho(240/244)S6/totalS6 in 10 mg tumor lysate according to the manufacturers’instructions.

ResultsDual activation of the PI3K and MAPK pathways isprevalent in primary rhabdomyosarcomas

Thirty-six percent (8 of 22) of alveolar rhabdomyosarco-ma and 46% (19 of 41) of embryonal rhabdomyosarcomasamples stained positively for both pAKT and pERK. Ofthese, 50% (4 of 8) of alveolar rhabdomyosarcoma and58% (11 of 19) of embryonal rhabdomyosarcoma samplesalso stained positively for pS6, indicating strong activationofmTOR signaling. Fifty-nine percent (13 of 22) of alveolarrhabdomyosarcoma but only 29% (12 of 41) of embryonalrhabdomyosarcoma samples stained positively for pAKTbut not pERK, suggesting that patients with embryonalrhabdomyosarcoma may be less likely than patients withalveolar rhabdomyosarcoma to respond to single-agentPI3K pathway inhibition. Notably, 93% (27 of 29) samplesstaining positively for pERK also stained positively for pAKTwhich, in addition to samples with no detectable pERKstaining and high levels of pAKT could predict resistance toMAPK pathway inhibitors in the majority of rhabdomyo-sarcoma. Assessment of correlation between staining usingx2 tests for trend showed a significantly positive correlation

Renshaw et al.

Clin Cancer Res; 19(21) November 1, 2013 Clinical Cancer Research5942




in all pairwise analyses: pAKT and pERK trend statistic:4.0912 (P ¼ 0.0431), pAKT and pS6 trend statistic: 8.5413(P ¼ 0.0035), pERK and pS6 trend statistic: 11.1429 (P ¼0.0008; Supplementary Table S1A–S1C).

PIK3CAknockdown induceswidespread compensatorysignaling and reciprocal cross-talk between the MAPKand PI3K pathwaysConsistentwithprevious reports that thePI3Kp110a and

b isoforms are expressed ubiquitously (reviewed in ref. 19),p110awas found to be expressed at similar levels in a panelof seven rhabdomyosarcoma cell lines, whereas p110bexpression was more variable (Supplementary Fig. S1).Unexpectedly, p110d, expressed predominantly in leuko-cytes (19), was expressed at high level in two (RH30 andRH41) of four alveolar rhabdomyosarcoma but none of theembryonal rhabdomyosarcoma lines studied here. PI3Kp110a (encoded by PIK3CA) has been suggested to be theisoform that is predominantly involved in insulin, IGFsignaling, and cell growth in many cell types (reviewed inref. 19). To assess the importance of PI3K p110a to thegrowth and survival of rhabdomyosarcoma cell lines, twoalveolar rhabdomyosarcoma lines (RH30 and RMS-1) andtwo embryonal rhabdomyosarcoma lines (RD and RMS-YM) were transduced with lentiviral shRNA particles tar-geted to PIK3CA. None of these lines carry an oncogenicPIK3CA mutation. Only the alveolar rhabdomyosarcomaline, RMS-1, was growth inhibited in response p110aknockdown (Fig. 1A). Growth inhibition did not involvecell-cycle arrest as evidenced by FACS analysis (Supplemen-tary Fig. S2), nor did there seem to be increased apoptosis(PARP cleavage) or autophagy (LC3BII production; Fig. 1B)indicating cytostasis.Knockdown of p110a induced varying degrees of com-

pensatory upregulation of p110b expression in all four celllines (Fig. 1B), whereas p110d expression was minimallyaffected. Downstream, AKT phosphorylation was not con-sistently inhibited following knockdown of p110a (Fig.1C). Of note, increased pThr308Akt to above control levels(not mirrored by an equivalent increase in pSer473Akt) wasseen on day 5 following lentiviral transduction of RH30cells, with gross inhibition of AKT phosphorylation beingdelayed until day 8. In contrast, both pThr308Akt and pSer473

Akt were reduced to below control levels in RMS-1, RD, andRMS-YM cells by day 5 with recovery toward control levelsin the embryonal rhabdomyosarcoma lines by day 8. Inaddition, PTEN expression was downregulated in RH30cells but increased in the remaining three cell lines (quan-titation shown in Supplementary Fig. S3). Inhibition ofdownstream S6 phosphorylation mirrored inhibition ofAKT phosphorylation in all cell lines except RH30, whereinhibition of S6 and AKT phosphorylation seemed discon-nected. The lattermay reflect signaling fromp110a throughonly one of the three AKT isoforms to mTOR, leading todecreased S6 phosphorylation despite an apparent lack ofinhibition of total AKT phosphorylation.Classical relief of IRS2 (but not IRS1; data not shown),

feedback repression on inhibition of mTOR signaling (20),

was seen in three of the cell lines (exception, RH30), as wellas evidence of reciprocal cross-talk between the PI3K/AKT/mTOR and RAS/RAF/ERK pathways. Varying degrees ofupregulation of pERK1/2 by day 5 following lentiviral trans-duction were observed in the three p110a knockdown–resistant cell lines, but not until day 8 in the sensitiveRMS-1 cells. Evidence of transient activation of AMPK con-comitant with upregulated pERK (21) was observed in threecell lines (exceptionRD,where pAMPK levelswere reduced).Quantitation of the pERK1/2 and total ERK1/2Western blotanalyses confirmed both a higher basal activation of extra-cellular signal-regulated kinase (ERK) signaling in theembryonal rhabdomyosarcoma compared with the alveolarrhabdomyosarcoma lines and the stimulation of ERK phos-phorylation following PIK3CA knockdown (Fig. 1D).

Pharmacologic inhibition of reprogrammed signalingpathways in PIK3CA knockdown stable cell linesestablishes the rationale for dual blockade of thePI3K/MAPK pathways in rhabdomyosarcoma

To determine the long-term effects of reduced PIK3CAexpression on PI3K and MAPK signaling, stable p110aknockdown, and control (CONSH) cell lines were derivedfrom the virally transduced cell lines over a periodof around4weeks. The derived cell lines were then used to screen PI3Kpathway inhibitors with a variety of inhibition profilesagainst the class 1 p110 isoforms and/or mTOR.

Both embryonal rhabdomyosarcoma stable knockdownlines showed increased expression of p110d, a novel obser-vation consistent with p110d being downstream of growthfactor signaling (19), and functional redundancy betweenthe p110a and d isoforms (22). No differences in sensitivityto the pan-PI3K inhibitor BMK120 [modestly selective forp110a versus the other isoforms (Supplementary Fig. S4A)]was observed in the stable p110a knockdown lines andtheir CONSH counterparts (Supplementary Fig. S4B). Inaddition, only the RMS-1 stable knockdown cell line, withminimal p110a and d expression, showed increased sensi-tivity to the p110b inhibitor, TGX221 (78-fold selective forp110b vs. p110a, less so for p110d; Supplementary Fig. S4Aand S4C). Taken together, these data show that all expressedisoforms must be inhibited for maximum inhibition of cellgrowth.

Assessment of selected elements of the PI3K/AKT/mTORand RAS/RAF/ERK pathways in the stable knockdown lines(Fig. 2A) revealed increased pThr308AKT but not pSer473AKTlevels in three of four stable knockdown lines (exception,RMS-YM), which was accompanied by altered PTEN expres-sion in the RD stable knockdown line. Increased levels ofpThr308AKT correlated with increased phosphorylation ofGSK3b, a downstream AKT substrate, but did not have aconsistent impact on pS6 levels (Supplementary Fig. S4D).IRS2 expression was upregulated in the RMS-1, RD, andRMS-YM stable knockdown lines suggesting that increasedsignaling through IGF-I receptor (IGF-IR) may contributeto increased pThr308AKT in some cell lines. UpregulatedMAPK signaling, as evidenced by increased pERK levels,was observed in all the stable knockdown lines resulting






A B

C D

*

RH30

CON

UTCON

SHd2KD

d2KD

d5

KD

d8

PIK3CA KD RMS-1

KD

d2

KD

d5

KD

d8CON

UT

CON

SHd2

PIK3CA KD

CON

UT

CON

SHd2KD

d2

KD

d5

KD

d8

PIK3CA KDRD RMS-YM

KD

d2

KD

d5

KD

d8CON

UT

CON

SHd2

PIK3CA KD

110α110β110δ

110α110β110δ

PARP

PARP

αTUB

αTUB

LC3B I/II

LC3B I/II

RH30CON

UTCON

SHd2

KD

d2

KD

d5

KD

d8

PIK3CA KD RMS-1

KD

d2

KD

d5

KD

d8CON

UT

CON

SHd2

PIK3CA KD

pAMPKERK1/2

pERK1/2

AMPK

αTUB

pThr308Akt

pSer473Akt

Total Akt

PTEN

pser235/6S6

IRS2

pAMPKERK1/2

pERK1/2

AMPK

αTUB

CON

UT

CON

SHd2KD

d2

KD

d5

KD

d8

PIK3CA KDRD RMS-YMKD

d2

KD

d5

KD

d8CON

UT

CON

SHd2

PIK3CA KD

pThr308AktpSer473Akt

Total Akt

PTEN

pser235/6S6IRS2

RH30 RMS-1

RD RMS-YM

0 2 4 6 80

5

10

15CONUTCONSHPIK3CA KD

RMS-1

Days following LV InfectionA

cc

um

ula

tiv

e c

ell

no

x 1

06

0 2 4 6 80

5

10


RH30

Days following LV Infection

Accu

mu

lati

ve c

ell n

o x

10

6

0 2 4 6 80

10

20


RD


Accu

mu

lati

ve c

ell n

o x

10

6

0 2 4 6 80

10

20

30


RMS-YM


Ac

cu

mu

lati

ve

ce

ll n

o x

10

6

8

6

4

2

0

1.00

1.00

1.001.06 1.01

1.31 1.27

1.100.94

1.23

1.60

1.00 1.120.68 0.80

1.47

CONUT

CONSH

Day

2

Day

5

Day

8

0.87 0.66

2.01

pE

RK

1/2

/to

tal

ER

K1

/2

8

6

4

2

0

pE

RK

1/2

/to

tal

ER

K1

/2

8

6

4

2

0

pE

RK

1/2

/to

tal

ER

K1

/2

8

6

4

2

0

pE

RK

1/2

/to

tal

ER

K1

/2Days following infection

CONUT

CONSH

Day

2

Day

5

Day

8

Days following infection

CONUT

CONSH

Day

2

Day

5

Day

8

Days following infectionCONUT

CONSH

Day

2

Day

5

Day

8

Days following infection

0.66

Figure 1. PIK3CA shRNA-mediated knockdown (KD) induces cell line–specific compensatory signaling and reciprocal cross-talk between theMAPK andPI3Kpathways: A, representative growth curves following PIK3CA knockdown: two alveolar rhabdomyosarcoma (RH30 and RMS-1) and two embryonalrhabdomyosarcoma (RDandRMS-YM) cell lineswere transducedwith control (CONSH) orPIK3CA-targeted shRNA (PIK3CA knockdown) on day 0 or treatedwith polybrene only (CONUT). On day 2, 5, and 8, cells were counted, and, on day 2 and 5, an appropriate dilution was reseeded in fresh plates enablingaccumulative cell counts to be calculated, and plotted for day 5 and 8. Puromycin (2.5 mg/mL) selectionwas introduced on day 2 andmaintained for the entireexperiment, the difference between the CONUT and CONSH growth curves being a reflection of viral transduction efficiency. B, Western immunoblotanalyses of class 1A PI3K isoforms, confirming selective p110a knockdown over 8 days in the PIK3CA shRNA transduced cell lines, along with a marker forapoptosis (PARPcleavage) andautophagy (LCBI/II expression). ThePARPcleavage seenonday5 is a reflection of puromycin selectionbetweendays 2and5.C, Western immunoblot analyses of PI3K and MAPK pathway biomarkers showing cell line–specific disruption of PTEN, pAKT, pS6, IRS2, pERK, andpAMPK in the above cells. �, pS6 levels in RMS-1 cells were below the PhosphorImager detection levels. This image was therefore collected following10 minutes of exposure to X-ray film. D, quantitation of pERK and total ERK immunoblots shown in C using ImageQuant software, expressed as the ratiopERK:total ERK. Numbers above each bar: in each cell line the ratio of pERK:total ERK in the CONUT cells was set as 1.0 and the relative fold ratio changecalculated for each sample. LV, lentiviral.

Renshaw et al.





in cross-talk that substantially increased pAMPK levels and,in the embryonal rhabdomyosarcoma knockdown lines,increased the basal rate of autophagy (LCB3II production).These latter data are consistent with disruption of aspects ofmTOR signaling in these cell lines. Treatment with NVP-BEZ235, which predominantly inhibits mTOR at concen-trations of less than 100 nmol/L (23), and AZD8055, aspecific TORC1/2 inhibitor, showed that the p110a stableknockdown cell lines with deregulated IRS2 expression aremore sensitive to mTOR inhibition than their CONSHcounterparts (Fig. 2B andC). In addition, stable knockdownlines exhibited increased sensitivity to ZSTK474, a pan PI3Kinhibitor with selectivity for p110d, and modest activityagainstmTor (IC50 of 0.377 mmol/L; ref. 24; SupplementaryFig. S4A and S4E). Although all the stable knockdown linesexpressed increased pERK levels, only the RMS-1 knock-

down line (lacking both p110a and d) exhibited signifi-cantly enhanced sensitivity to the MEK inhibitor AZD6244(Fig. 2D). This suggests that p110a and/or p110d cansupport compensatory signaling through PI3K/AKT/mTORon inhibitionof theMAPKpathway,whereas p110b cannot.

Overall, these data suggest that not only should pan-PI3K(or AKT) and mTOR activity be inhibited to achieve max-imal inhibition of the PI3K/AKT/mTOR pathway, butsimultaneous inhibition of the RAS/RAF/ERK pathway toprevent compensatory cross-talk is also necessary to max-imize the antiproliferative effect.

Dual blockade of PI3K/AKT/mTOR and RAS/RAF/ERKsignaling is synergistic in rhabdomyosarcoma cell lines

To determine whether dual blockade of both thePI3K and MAPK pathways yields improved efficacy in

PI3K

KD

CON

SH

RH30

PI3K

KD

CON

SH

RMS-1

PI3K

KD

CON

SH

PI3K

KDCON

SH

RD RMS-YM

pThr308Akt

pSer473Akt

Akt

IRS2

pERK1/2

ERK1/2

PTEN

pAMPK

LK3B I/IIαTUB

110α110β110δ

NVP-BEZ235B

D AZD8055C AZD6244

A

Figure2. Reprogrammingof thePI3KandMAPKsignalingpathways induces increased sensitivity tomTOR inhibition in the stablep110a knockdown (KD) lineswith deregulated IRS2 expression, but increased sensitivity to MEK inhibition only in RMS-1 knockdown cells with reduced p110a and d expression. A,Western immunoblot analyses of class 1A PI3K isoforms confirming p110a knockdown in the stable knockdown lines (PI3K knockdown) along withupregulation of p110d expression in the embryonal rhabdomyosarcoma knockdown stable lines compared with their CONSH counterparts. Also shown areselected elements of the PI3K andMAPK pathways in the stable p110a knockdown lines showing cell line–specific upregulation of pThr308AKT (but not pSer473

AKT) and IRS2, upregulated pERK and pAMPK levels in all the knockdown lines, and increased autophagy (LCBI/II), particularly in the embryonalrhabdomyosarcoma knockdown lines. B and C, representative growth inhibition curves (MTS assays), following treatment of the paired CONSH andknockdown lines with NVP-BEZ235 and AZD8055, respectively, showing increased sensitivity to mTOR inhibition in the knockdown lines with deregulatedIRS2 expression. D, representative growth inhibition curves following treatment with the AZD6244 showing increased sensitivity to MEK inhibition only inRMS-1 cells lacking p110a and d.






rhabdomyosarcoma cells, combination GI50 isobologramswere constructed using the parental alveolar rhabdomyo-sarcoma and embryonal rhabdomyosarcoma cell linesand the combinations AZD8055/AZD6244, ZSTK474/AZD6244, and NVP-BEZ235/AZD6244. The combination

AZD8055/AZD6244 was synergistic in the RH30, RD, andRMS-YM cell lines, but nearer additive in the RMS-1 line (themost sensitive line to single-agent PI3K pathway inh-ibition; Fig. 3A). The embryonal rhabdomyosarcoma celllines were used to confirm synergy with the combinations

AZD8055

0.5 x GI50 ≈≈ 25.0 nmol/L

AZD6244

0.5 x GI50 ≈ 15 µmol/L

AZD8055/6244

0.5 x GI50 in combination

AZD8055

1.0 x GI50 ≈ 50.0 nmol/L

AZD6244

1.0 x GI50 ≈ 30.0 µmol/L

AZD8055/6244

1.0 x GI50 in combination

D

RMS-1 RH30

A B

CRD RMS-YM RD

RD RMS-YM

RMS-YM

AZD8055 + AZD6244 ZSTK474 + AZD6244

NVP-BEZ235 + AZD6244

Figure 3. Dual blockade of both the PI3K andMAPKpathways is synergistic in rhabdomyosarcomacell lines in vitro. A–C, combination isobolograms using thecombinations: AZD8055/AZD6244, ZSTK474/AZD6244, andNVP-BEZ235/AZD6244, respectively. TheGI50 values of compoundAandBareplotted on the x-and y-axis along with the GI50 values of compound B obtained in the presence of various fixed concentrations of compound A. The diagonal linedrawn between the GI50 values for the two compounds on the y- and x-axis is the theoretical line of additivity. All GI50 values to the left of this line indicatesynergy. D, levels of pERK1/2, pSer473AKT, and pSer240/244S6 in RDcells following treatment with AZD8055, AZD6244, and AZD8055/AZD6244 in combinationat 0.5�, and 1.0� GI50 concentrations. Proteins from control- and drug-treated cells were extracted at the indicated time points following treatment.Control and treated samples from each time point were loaded side by side for Western immunoblot analyses and quantitated using ImageQuant software.Drug-treated/control phosphorylated biomarker levels at each time point are expressed as percentage of levels at time 0. NB, levels of pERK in controlsamples were below consistently quantifiable levels at 48 hours, when the cells were reaching confluence.

Renshaw et al.





ZSTK474/AZD6244 and NVP-BEZ235/AZD6244 (Fig. 3Band C).RD cells that have an NRAS mutation (5) and are rela-

tively resistant to both AZD8055 and AZD6244 were selec-ted to investigate the effect of single-agent and combinationtreatment with AZD8055 and AZD6244 on the biomarkersof PI3K/mTOR pathway (pSer473AKT and pS6) and MAPKpathway (pERK) activity over a 48-hour time course. Tomore easily assess any synergistic effects on the pathwaybiomarkers following combination treatment, cells weretreated with 0.5�, and 1.0�GI50 concentrations. AZD8055increased pERK levels, and AZD6244 increased pSer473AKTlevels after 16 hours of treatment, confirming reciprocalcompensatory cross-talk between the PI3K andMAPK path-ways following inhibition of each pathway individually(Fig. 3D). The extent of inhibition of S6, but not AKTphosphorylation, was concentration dependent followingAZD8055 treatment, indicating that at low concentra-tions, and with activating cross-talk from the MAPK path-way, TORC1 was more effectively inhibited than TORC2.AZD6244 treatment ablated pERK levels at all time points,and both concentrations and the activation/inhibitionprofiles of both AKT and S6 phosphorylation, respective-ly, were similar at both concentrations. These data indi-cate that growth inhibition following AZD6244 does notcorrelate with MEK inhibition (IC50 concentration forMEK1/2 of 14 nmol/L; ref. 25), and may be due to offtarget effects.AZD6244 in combination with AZD8055 reduced pERK

levels to the same extent as AZD6244 alone, whereas pS6levels were reduced earlier, to a greater extent, and for alonger duration than with either treatment alone (Fig. 3D).AKT phosphorylation was reduced to the same extent astreatment with AZD8055 alone but did not recover tocontrol levels at later time points. These data imply thatsignificant and protracted inhibition of mTOR activity isessential for maximal antitumor activity and that coinhibi-tion of ERK and AKT signaling is necessary to achieve this.Equivalent alterations to biomarkers of the PI3K and

MAPK pathways occurred in vivo 6 hours following tre-atment of mice bearing RD xenografts with AZD8055 (10mg/kg per os) and AZD6244 (10mg/kg per os) alone, or incombination (Supplementary Fig. S5A). However, levels ofall three biomarkers recovered to control levels or above by16 hours following treatment and correlated with plasmaclearance of both drugs to close to, or below, the limit ofdetection at this time point (Supplementary Fig. S5B). Thisand further pharmacokinetic analysis (Supplementary Fig.S6) revealed an interaction between both drugs when givenin combination. Peak plasma concentrations of AZD6244were reduced and the elimination phase extended whengiven in combination with AZD8055, whereas higher plas-ma and tumor levels of AZD8055were achievedwhen givenin combination with AZD6244. These changes becamemore pronounced on repeat dosing. The possibility ofinteraction at the level of metabolism by P450 enzymeswas eliminated using a drug–drug interaction assay (Sup-plementary Methods; data not shown).

Significantly enhanced antitumor efficacy followingtreatment of RD xenograft tumors with AZD8055, butnot NVP-BEZ235, in combination with AZD6244

A preliminary dose-finding therapeutic study defined aonce daily treatment schedule of AZD8055, 20mg/kg per os,and AZD6244, 10 mg/kg per os, as well tolerated and activewhen given in combination (Supplementary Fig. S7A–S7D).Previous studies have shown that NVP-BEZ235 treatmentof Her2-overexpressing breast cancer cells resulted in com-pensatory activation of ERK signaling, and that the combi-nationofNVP-BEZ235 (25mg/kg) andAZD6244 (8mg/kg)was synergistic in vivo (26). NVP-BEZ235 is a dual PI3K/mTORC1/2 inhibitor at the concentrations (above 500nmol/L) likely to be achieved in vivo (23), and had the pot-ential to inhibit the compensatory increase in pAKT levels onMEK inhibition more effectively than the specific TORC1/2inhibitor, AZD8055. Therefore, these two combinationswere compared in a head-to-head therapeutic study.

When given alone, NVP-BEZ235 inhibited tumor growthto a greater extent than AZD8055. However, synergism wasshown only when using AZD6244 in combination withAZD8055 and not when used in combination with NVP-BEZ235 (Fig. 4A and B). Pharmacokinetic analysis of allthree drugs showed that when given in combination, NVP-BEZ235 reduced plasma concentrations of AZD6244 to asimilar extent as AZD8055 (Fig. 4C). The plasma concen-trations of AZD8055 and NVP-BEZ235 were in the samerange when the agents were administered alone, and theconcentration range of both drugs was increased whengiven in combination with AZD6244 (Fig. 4C).

Assessment of the pharmacodynamic biomarkers, pERK,pSer473AKT, and pS6 in treated tumors confirmed reciprocalcompensatory signaling on inhibition of either the PI3Kpathway by AZD8055 or NVP-BEZ235, or the MAPK path-way by AZD6244 in the in vivo therapeutic setting (Fig. 4D).Individually, AZD8055 and NVP-BEZ235 were equally effi-cient in reducing AKT and S6 phosphorylation althoughAZD8055 increased pERK levels to a greater extent thanNVP-BEZ235. However, while treatment with the combi-nation AZD8055/AZD6244 reduced all three biomarkersto less than 30% of control levels, the combination NVP-BEZ235/AZD6244 failed to reduce pAKT below controllevels and there was a corresponding reduction in theinhibition of S6 phosphorylation. Thus, the inhibitorypotency of NVP-BEZ235 against PI3K and TORC2 wasinsufficient to block the compensatory upregulation of AKTphosphorylation induced by MEK inhibition in this tumortype in vivo. AZD8055/AZD6244 is therefore the combina-tion indicated to take forward to the clinic.

DiscussionPrevious reports have indicated that dual activation of the

PI3K/AKT/mTORandRAS/RAF/MEK/ERKpathways is likelyto result in innate resistance to the targeting of either path-way alone (10–13). Although activation of both pathways,individually, has been documented in rhabdomyosarcoma(4, 5, 8, 9), dual activation status has not been assessed.






Using immunohistochemical analysis of TMAs, we haveshown dual activation in 43% of primary rhabdomyosar-coma samples, and of these, 55% showed strong activationof mTOR signaling. A higher proportion of alveolar rhab-domyosarcoma than embryonal rhabdomyosarcoma sam-ples stained positively for pAKT in the absence of pERKstaining (59% and 29%, respectively). Thus, while theoret-ically, some patients with alveolar rhabdomyosarcomamight benefit from targeted inhibition of the PI3K pathway,it is less likely to be the case for patients with embryonalrhabdomyosarcoma. Importantly, we provide evidence forcompensatory upregulation of theMAPKpathway followingPI3Kpathway inhibition in rhabdomyosarcoma,ashas beenshown in other tumor types (15). This has the potential tocircumvent the antiproliferative effect of PI3K pathway inhi-bitors and result in treatment failure.

Small-molecule inhibitors of PI3Kmainly target the class1 PI3Ks, namely p110a, b, d, and g , of which only p110a ismutated in cancer (19). PI3K p110a is also the isoform thatis predominantly involved in IGF-IR signaling (commonlyactivated in rhabdomyosarcoma; ref. 6) and cell growth.Using lentiviral shRNA particles targeted to PIK3CA, weshowed that only one of four rhabdomyosarcoma cell lineswas growth inhibited following PI3K p110a knockdown,suggesting that a p110a-specific inhibitor would not havegeneral use in the rhabdomyosarcoma clinic. Examinationof basal class 1 PI3K isoform expression profiles in rhab-domyosarcoma cell lines revealed uniform expression ofp110a, more variable expression of p110b, and no detect-able expression of p110g . Unexpectedly, previously unre-ported expression of p110d, generally restricted to cells ofhematopoietic origin, was also seen in two of four alveolar

pERK pAKT pS6

A B

C Plasma AZD6244

Plasma AZD8055 & BEZ235

D

Figure 4. Significantly enhanced antitumor efficacy following treatment of RD xenografts with AZD8055, but not NVP-BEZ235, in combination with AZD6244.AandB,head-to-headtherapeutic studyofRDxenografts treatedwithAZD6244andeitherAZD8055orNVP-BEZ235, respectively, at the indicateddosesalone,and in combination. Tumor volumes are expressed as a percentage volume of each tumor on day 0. Final tumor weights (g) show significantly increasedefficacy of the combination AZD8055/AZD6244 compared with AZD6244 alone (��, P¼ 0.015) or AZD8055 alone (�, P¼ 0.038) and no significant difference inefficacy of the combination NVP-BEZ235/AZD6244 compared with AZD6244 alone (P ¼ 0.13) or BEZ235 alone (P ¼ 0.82; Mann–Whitney t test). C, plasmaAZD6244 concentrations 3 hours following the final dose in mice treated with AZD6244 alone or in combination with AZD8055 or NVP-BEZ235 andplasma AZD8055 and NVP-BEZ235 concentrations from the same mice as above. D, tumor pharmacodynamic biomarkers pERK: phospho(T/Y:202/204:185/187)ERK/total ERK1/2, pAKT: phospho(Ser473)AKT/total AKT and pS6: phospho(240/244)S6/total S6, as determined by Meso Scale Discovery (MSD)immunoassay. p.o., per os.

Renshaw et al.





rhabdomyosarcoma lines. Consistent with this, mining ofgene expression profiling data of rhabdomyosarcoma pri-mary tumor samples (3) revealed higher levels of PI3KCDmRNA compared with skeletal muscle (data not shown).High-level p110d expression has also been reported inbreast cancer and neuroblastoma cells (27, 28). However,in rhabdomyosarcoma cell lines, neither the expressionof p110d nor the overall isoform expression profiles wasassociated with sensitivity/resistance to p110a knockdown.Compensatory upregulation and reprogramming of

alternative signaling pathways in the short and long termoffers clues to the mechanisms of innate and acquiredresistance to targeted inhibitors. Here, we have shown thatcell line–specific, widespread compensatory and adaptivesignaling occurs following PIK3CA knockdown, even in theabsence of any overt phenotypic effect. Using various PI3Kpathway inhibitors and a MEK inhibitor to inhibit thesebypass mechanisms in stable p110a knockdown lines, wehave determined which elements of the PI3K and MAPKpathways must be inhibited to achieve a maximum anti-proliferative effect.In the short term, p110a knockdown induced compen-

satory upregulation of p110b although this was not main-tained in the p110a knockdown stable lines. However,significant upregulation of p110d was seen in the embry-onal rhabdomyosarcoma stable knockdown lines, whichhighlights the plasticity of PI3K isoform expression andsupports functional redundancy, particularly between thep110a and d isoforms. Nodifference in the sensitivity to thepan-PI3K inhibitor, BMK120, was seen in the stable p110aknockdown lines compared with their CONSH counter-parts, and only the RMS-1 knockdown line, with minimalp110a and d expression, showed increased sensitivity to theselective p110b inhibitor, TGX221. These data suggest thatpan-PI3K inhibitors should possess potent activity againstall class 1A isoforms to ensuremaximum growth inhibitionin rhabdomyosarcoma.AKT phosphorylation was not consistently inhibited fol-

lowing p110a knockdown despite early inhibition of S6phosphorylation in all four cell lines. Of note, we showedincreasedpThr308AKT, delayed gross inhibitionof pAKT, anddownregulation of the phosphatase, PTEN, in RH30 cellsfollowing PIK3CA knockdown, a novel observation. Incontrast, early inhibition of AKT phosphorylation andupregulation of PTEN expression was seen in the otherthree lines. Transcriptional regulation of PTEN has beenshown previously in various settings, but the functionalroles and mechanisms responsible remain poorly under-stood. For example, decreased PTEN expression on inhibi-tion of mTOR has been reported in cells with loss orinactivation of TSC2 (29), whereas upregulation of PTENin response to activation of c-jun-NH2 kinase (JNK; ref. 30),and repression of PTEN expression by activated NFkBsignaling (31), have also been reported.Compensatory upregulation of pERK levels following

PIK3CA knockdown was observed by day 5 in the threecell lines resistant to the effects of p110a knockdown, butthis was delayed in the sensitive RMS-1 cells. However,

increased levels of ERK phosphorylation were seen in allfour stable knockdown lines. A lack of early stimulationof ERK signaling, together with extremely low levels ofbasal mTOR activity, may underpin the initial sensitivityof RMS-1 cells to PIK3CA knockdown, whereas the subse-quent upregulation of ERK signaling may have contributedto their ultimate recovery and survival. Although previousreports have suggested that activated MAPK signaling med-iates resistance to PI3K inhibitors, these studies indicate thatit is not necessarily the basal rate of MAPK activity thatdictates resistance toPI3Kpathway inhibitors, but rather thedegree and kinetics of compensatory cross-talk. Thus, whileall four stable PIK3CA knockdown lines exhibited higherlevels of pERK they were not more resistant to any of thePI3K pathway inhibitors than their CONSH counterparts.Similarly, increased activation of MAPK signaling did notalter the sensitivity of three of the four stable knockdownlines to MEK inhibition. Only the RMS-1 knockdown line,expressing 110b but not p110a or d, exhibited increasedsensitivity to theMEK inhibitor AZD6244. This suggests thatp110b does not support compensatory activation of thePI3K pathway following MEK inhibition, whereas p110aand d do.

Activated ERK signaling has been shown to activatemTOR by phosphorylation of both TSC2 and Raptor(32), and also to activate AMPK resulting in increasedretinoblastoma (Rb) phosphorylation and stimulation ofcell growth (21). In addition, activated AMPK acts as asurvival factor protecting cells from hypoxia and nutrientdeprivation through increased glucose uptake and a shift toanaerobic respiration with altered intermediary metabo-lism (theWarburg effect), often accompanied by inhibitionof protein synthesis and cell growth through suppression ofmTOR signaling (reviewed in ref. 33). Activation of AMPKalong with deregulated IRS2 expression, and an increasedbasal rate of autophagy (the latter prominent in the embry-onal rhabdomyosarcoma lines), suggests that mTOR sig-naling is partially compromised in the stable p110a knock-down lines. Importantly, these cells are more sensitive toTORC1/TORC2 inhibitors, suggesting that mTOR is a cen-tral node for integrating cross-talk between the PI3K andMAPK pathways and an important target element for max-imal PI3K pathway inhibition.

Dual blockade of both the PI3K and MAPK pathwayswas shown to be synergistic in vitro and was shown to bedue to the reciprocal inhibition of the compensatory acti-vation of the alternate pathway seen following inhibition ofeach pathway individually. As predicted, treatment of theNRAS-mutated RD tumor xenografts in vivo with theTORC1/TORC2 inhibitor AZD8055 or the MEK inhibitorAZD6244, showed no therapeutic benefit, whereas the dualPI3K/mTOR inhibitor NVP-BEZ235 was more active. How-ever, in combination, AZD8055/AZD6244 resulted in sig-nificant inhibition of tumor growth, whereasNVP-BEZ235/AZD6244 resulted in no additional benefit over treatmentwith NVP-BEZ235 alone. This was shown to be due to theinability of NVP-BEZ235 to inhibit the compensatory acti-vation of AKT induced by MEK inhibition. Therefore, the






biomarkers of effective activity are the simultaneous reduc-tion of pAKT, pS6, and pERK.

Using data collected from phase I clinical trials, a recentstudy has evaluated the clinical outcome of dual targetingboth the PI3K and MAPK signaling pathways comparedwith targeting either pathway alone (34). The resultsshowed increased efficacy in many tumor types but at thecost of additional toxicity. Although the in vivo combinationdoses used here were well tolerated, pharmacokinetic anal-ysis revealed an interaction between the two classes ofcompounds, resulting in lower plasma and tumor levels ofAZD6244 but higher levels of the PI3K pathway inhibitors,both effects becoming more pronounced with repeat dos-ing. However, the concentrations of AZD6244 achieved invivowere still in excess of levels required forMEK inhibition.Further pharmacokinetic/pharmacodynamic–driven pre-clinical studies to identify the lowest AZD6244 doserequired for MEK inhibition, followed by escalation ofAZD8055 to achieve therapeutic efficacy, will help informearly clinical trials. In addition, careful monitoring of plas-ma pharmacokinetics of both drugs following repeateddosing will be required to assess the extent of drug–druginteractions in patients.

In summary, in the preclinical proof-of-principle rhab-domyosarcoma xenograft studies presented here, the com-bination of AZD8055 and AZD6244 showed significantlyincreased therapeutic benefit over the combination NVP-BEZ235/AZD6244.We show that the three phosphorylatedbiomarkers of ERK, S6, and AKT must be reduced forsynergistic activity. These studies confirm that dual inhibi-tion of both the PI3K and MAPK pathways offers a wayforward for the treatment of those tumor types such asrhabdomyosarcoma that are predicted to be resistant toblockade of either pathway alone. Even in innately sensitivetumors, addition of aMEK inhibitor to a PI3K inhibitormayforestall the emergence of resistance.

Disclosure of Potential Conflicts of InterestS.A. Eccles is an employee of ICR, which has a commercial interest in the

discovery and development of anticancer drugs, including PI3K and AKTinhibitors, and operates a Rewards to Inventors scheme. No potentialconflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: J. Renshaw, K.R. Taylor, S.A. Eccles, A.D. Pearson,J. ShipleyDevelopment of methodology: J. Renshaw, K.R. Taylor, R.R. Ruddle, F.I.Raynaud, T. PietschAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): J. Renshaw, K.R. Taylor, R. Bishop, M. Valenti, A.De Haven Brandon, S. Gowan, S.A. Eccles, R.R. Ruddle, K. Thway, T. PietschAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): J. Renshaw, K.R. Taylor, R. Bishop, S.Gowan, S.A. Eccles, R.R. Ruddle, L.D. Johnson, F.I. Raynaud, J.L. Selfe, K.Thway, T. Pietsch, J. ShipleyWriting, review, and/or revision of the manuscript: J. Renshaw, S.A.Eccles, R.R. Ruddle, F.I. Raynaud, T. Pietsch, A.D. Pearson, J. ShipleyAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): M. Valenti, A. De Haven Brandon,T. PietschStudy supervision: J. Renshaw, S.A. Eccles, J. Shipley

AcknowledgmentsThe authors thank Chris Jones, Susanne Gatz, Paul Clark, and Paul

Workman for their review of the article and helpful comments. We aregrateful for help with tumor collection and annotation from the Children’sCancer and Leukemia Group.

Grant SupportThis work was supported by NHS funding to the NIHR Biomedical

Research Centre (to J. Renshaw), the Royal Marsden Hospital Charitablefunds (to K.R. Taylor and R. Bishop), Cancer Research UK (grant referencesC309/A11566 and C5066/A10399; to M. Valenti, A. De Haven Brandon, S.Gowan, S.A. Eccles, R.R. Ruddle, L.D. Johnson, F.I. Raynaud, and J. Shipley),and the Chris Lucas Trust (to J.L. Selfe).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received March 28, 2013; revised June 21, 2013; accepted July 30, 2013;published OnlineFirst August 5, 2013.

References1. De Giovanni C, Landuzzi L, Nicoletti G, Lollini PL, Nanni P. Molecular

and cellular biology of rhabdomyosarcoma. Future Oncol 2009;5:1449–75.

2. Xia SJ, Pressey JG, Barr FG. Molecular pathogenesis of rhabdomyo-sarcoma. Cancer Biol Ther 2002;1:97–104.

3. Missiaglia E, Williamson D, Chisholm J, Wirapati P, Pierron G, Petel F,et al. PAX3/FOXO1 fusion gene status is the key prognostic molecularmarker in rhabdomyosarcoma and significantly improves current riskstratification. J Clin Oncol 2012;30:1670–7.

4. Martinelli S, McDowell HP, Vigne SD, Kokai G, Uccini S, Tartaglia M,et al. RAS signaling dysregulation in human embryonal rhabdomyo-sarcoma. Genes Chromosomes Cancer 2009;48:975–82.

5. Shukla N, Ameur N, Yilmaz I, Nafa K, Lau CY, Marchetti A, et al.Oncogene mutation profiling of pediatric solid tumors reveals signif-icant subsets of embryonal rhabdomyosarcoma and neuroblastomawith mutated genes in growth signaling pathways. Clin Cancer Res2012;18:748–57.

6. Martins AS, Olmos D, Missiaglia E, Shipley J. Targeting the insulin-likegrowth factor pathway in rhabdomyosarcomas: rationale and futureperspectives. Sarcoma 2011;2011:209736.

7. StevensMC. Treatment for childhood rhabdomyosarcoma: the cost ofcure. Lancet Oncol 2005;6:77–84.

8. Cen L, Hsieh FC, Lin HJ, Chen CS, Qualman SJ, Lin J. PDK-1/AKTpathway as a novel therapeutic target in rhabdomyosarcoma cellsusing OSU-03012 compound. Br J Cancer 2007;97:785–91.

9. Petricoin EF III, Espina V, Araujo RP, Midura B, Yeung C, Wan X, et al.Phosphoprotein pathway mapping: Akt/mammalian target of rapamy-cin activation is negatively associated with childhood rhabdomyosar-coma survival. Cancer Res 2007;67:3431–40.

10. Engelman JA, Chen L, Tan X, Crosby K, Guimaraes AR, Upadhyay R,et al. Effective use of PI3K and MEK inhibitors to treat mutant KrasG12D and PIK3CA H1047R murine lung cancers. Nature Med 2008;14:1351–6.

11. Weigelt B, Downward J. Genomic determinants of PI3K pathwayinhibitor response in cancer. Front Oncol 2012;2:109.

12. Balmanno K, Chell SD, Gillings AS, Hayat S, Cook SJ. Intrinsicresistance to the MEK1/2 inhibitor AZD6244 (ARRY-142886) isassociated with weak ERK1/2 signalling and/or strong PI3K sig-nalling in colorectal cancer cell lines. Int J Cancer 2009;125:2332–41.

13. Hoeflich KP, O'Brien C, Boyd Z, Cavet G, Guerrero S, Jung K, et al. Invivo antitumor activity of MEK and phosphatidylinositol 3-kinaseinhibitors in basal-like breast cancer models. Clin Cancer Res 2009;15:4649–64.

Renshaw et al.





14. Wang Z, Zhou J, Fan J, Qiu SJ, Yu Y, Huang XW, et al. Effect ofrapamycin alone and in combination with sorafenib in an orthotopicmodel of human hepatocellular carcinoma. Clin Cancer Res 2008;14:5124–30.

15. Holt SV, Logie A, Davies BR, Alferez D, Runswick S, Fenton S, et al.Enhanced apoptosis and tumor growth suppression elicited by com-bination of MEK (selumetinib) and mTOR kinase inhibitors (AZD8055).Cancer Res 2012;72:1804–13.

16. Tonelli R, McIntyre A, Camerin C, Walters ZS, Di Leo K, Selfe J, et al.Antitumor activity of sustained N-myc reduction in rhabdomyosarco-mas and transcriptional block by antigene therapy. Clin Cancer Res2012;18:796–807.

17. Workman P, Aboagye EO, Balkwill F, Balmain A, Bruder G, Chaplin DJ,et al. Guidelines for the welfare and use of animals in cancer research.Br J Cancer 2010;102:1555–77.

18. Raynaud FI, Eccles S, Clarke PA, Hayes A, Nutley B, Alix S, et al.Pharmacologic characterization of a potent inhibitor of class I phos-phatidylinositide 3-kinases. Cancer Res 2007;67:5840–50.

19. Chaussade C, Rewcastle GW, Kendall JD, Denny WA, Cho K, Gron-ning LM, et al. Evidence for functional redundancy of class IA PI3Kisoforms in insulin signalling. Biochem J 2007;404:449–58.

20. Harrington LS, Findlay GM, Gray A, Tolkacheva T, Wigfield S, RebholzH, et al. The TSC1-2 tumor suppressor controls insulin-PI3K signalingvia regulation of IRS proteins. J Cell Biol 2004;166:213–23.

21. RiosM, ForetzM, Viollet B, Prieto A, FragaM, Costoya JA, et al. AMPKactivation by oncogenesis is required to maintain cancer cell prolifer-ation in astrocytic tumors. Cancer Res 2013;73:2628–38.

22. Foukas LC, Berenjeno IM, Gray A, Khwaja A, Vanhaesebroeck B.Activity of any class IA PI3K isoform can sustain cell proliferation andsurvival. Proc Natl Acad Sci USA 2010;107:11381–6.

23. Serra V, Markman B, Scaltriti M, Eichhorn PJ, Valero V, Guzman M,et al. NVP-BEZ235, a dual PI3K/mTOR inhibitor, prevents PI3K sig-naling and inhibits the growth of cancer cells with activating PI3Kmutations. Cancer Res 2008;68:8022–30.

24. Kong D, Dan S, Yamazaki K, Yamori T. Inhibition profiles of phospha-tidylinositol 3-kinase inhibitors against PI3K superfamily and humancancer cell line panel JFCR39. Eur J Cancer 2010;46:1111–21.

25. Yeh TC, Marsh V, Bernat BA, Ballard J, Colwell H, Evans RJ, et al.Biological characterization of ARRY-142886 (AZD6244), a potent,

highly selective mitogen-activated protein kinase kinase 1/2 inhibitor.Clin Cancer Res 2007;13:1576–83.

26. Serra V, Scaltriti M, Prudkin L, EichhornPJ, IbrahimYH,ChandarlapatyS, et al. PI3K inhibition results in enhancedHERsignaling and acquiredERK dependency in HER2-overexpressing breast cancer. Oncogene2011;30:2547–57.

27. Sawyer C, Sturge J, Bennett DC, O'Hare MJ, Allen WE, Bain J, et al.Regulation of breast cancer cell chemotaxis by the phosphoinositide3-kinase p110delta. Cancer Res 2003;63:1667–75.

28. Boller D, Schramm A, Doepfner KT, Shalaby T, von Bueren AO, EggertA, et al. Targeting the phosphoinositide 3-kinase isoform p110deltaimpairs growth and survival in neuroblastoma cells. Clin Cancer Res2008;14:1172–81.

29. Das F, Ghosh-Choudhury N, Dey N, Mandal CC, Mahimainathan L,Kasinath BS, et al. Unrestrained mammalian target of rapamycincomplexes 1 and 2 increase expression of phosphatase and tensinhomolog deleted on chromosome 10 to regulate phosphorylation ofAkt kinase. J Biol Chem 2012;287:3808–22.

30. Redondo-Munoz J, Escobar-Diaz E,HernandezDel CerroM,PandiellaA, Terol MJ, Garcia-Marco JA, et al. Induction of B-chronic lympho-cytic leukemia cell apoptosis by arsenic trioxide involves suppressionof the phosphoinositide 3-kinase/Akt survival pathway via c-jun-NH2terminal kinase activation and PTEN upregulation. Clin Cancer Res2010;16:4382–91.

31. KimS,Domon-Dell C,KangJ,ChungDH, FreundJN,EversBM.Down-regulation of the tumor suppressor PTENby the tumor necrosis factor-alpha/nuclear factor-kappaB (NF-kappaB)-inducing kinase/NF-kap-paB pathway is linked to a default IkappaB-alpha autoregulatory loop.J Biol Chem 2004;279:4285–91.

32. Carriere A, Cargnello M, Julien LA, Gao H, Bonneil E, Thibault P, et al.OncogenicMAPK signaling stimulates mTORC1 activity by promotingRSK-mediated raptor phosphorylation. Curr Biol 2008;18:1269–77.

33. Luo Z, Zang M, Guo W. AMPK as a metabolic tumor suppressor:control of metabolism and cell growth. Future Oncol 2010;6:457–70.

34. Shimizu T, Tolcher AW, Papadopoulos KP, Beeram M, Rasco DW,Smith LS, et al. The clinical effect of the dual-targeting strategyinvolving PI3K/AKT/mTOR and RAS/MEK/ERK pathways in patientswith advanced cancer. Clin Cancer Res 2012;18:2316–25.






2013;19:5940-5951. Published OnlineFirst August 5, 2013.Clin Cancer Res Jane Renshaw, Kathryn R. Taylor, Ryan Bishop, et al.

In Vivo and In VitroRhabdomyosarcoma Cell Growth RAS/MEK/ERK (AZD6244) Pathways Synergistically Inhibits Dual Blockade of the PI3K/AKT/mTOR (AZD8055) and

Updated version

10.1158/1078-0432.CCR-13-0850doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2013/08/05/1078-0432.CCR-13-0850.DC1

http://clincancerres.aacrjournals.org/content/suppl/2013/11/01/1078-0432.CCR-13-0850.DC2Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/19/21/5940.full#ref-list-1

This article cites 34 articles, 20 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/19/21/5940.full#related-urls

This article has been cited by 17 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/19/21/5940To 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-13-0850

http://clincancerres.aacrjournals.org/content/suppl/2013/11/01/1078-0432.CCR-13-0850.DC2 http://clincancerres.aacrjournals.org/content/suppl/2013/08/05/1078-0432.CCR-13-0850.DC1

http://clincancerres.aacrjournals.org/content/suppl/2013/11/01/1078-0432.CCR-13-0850.DC2 http://clincancerres.aacrjournals.org/content/suppl/2013/08/05/1078-0432.CCR-13-0850.DC1

http://clincancerres.aacrjournals.org/content/19/21/5940.full#ref-list-1

http://clincancerres.aacrjournals.org/content/19/21/5940.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/19/21/5940


Dual Blockade of the PI3K/AKT/mTOR (AZD8055) and RAS/MEK ... · pathway, and cross-talk to...

Documents

Transcript of Dual Blockade of the PI3K/AKT/mTOR (AZD8055) and RAS/MEK ... · pathway, and cross-talk to...