RAS/PI3K Crosstalk and Cetuximab Resistance in...

Cancer Therapy: Preclinical

RAS/PI3K Crosstalk and Cetuximab Resistance in Head andNeck Squamous Cell Carcinoma

T. Rampias1, A. Giagini2, S. Siolos2, H. Matsuzaki1, C. Sasaki1, A. Scorilas3, and A. Psyrri1,2

AbstractPurpose: Cetuximab, an antibody directed against the EGF receptor, is an effective clinical therapy for

patients with head and neck squamous cell cancer (HNSCC). Despite great clinical promise, intrinsic or

acquired cetuximab resistance hinders successful treatment outcomes but little is known about the

underlying mechanism.

Experimental Design: To study the role of oncogenic HRAS in cetuximab resistance in HNSCC, the

frequency of oncogenicHRASmutations was determined in a cohort of 180 genomic DNAs from head and

neck cancer specimens.We also used a combination of cetuximab-resistant cell lines and a transgenicmouse

model of RAS-driven oral cancer to identify an oncogenic RAS-specific gene expression signature that

promotes cetuximab resistance.

Results:Here, we show that activation of RAS signaling leads to persistent extracellular signal–regulated

kinase 1/2 signaling and consequently to cetuximab resistance. HRAS depletion in cells containing

oncogenic HRAS or PIK3CA restored cetuximab sensitivity. In our study, the gene expression signature

of c-MYC, BCL-2, BCL-XL, and cyclin D1 upon activation of MAPK signaling was not altered by cetuximab

treatment, suggesting that this signature may have a pivotal role in cetuximab resistance of RAS-activated

HNSCC. Finally, a subset of patients with head and neck cancer with oncogenic HRASmutations was found

to exhibit de novo resistance to cetuximab-based therapy.

Conclusions:Collectively, these findings identify a distinct cetuximab resistancemechanism.Oncogenic

HRAS in HNSCC promotes activation of ERK signaling, which in turn mediates cetuximab resistance

through a specific gene expression signature. Clin Cancer Res; 20(11); 1–14. �2014 AACR.

IntroductionTheEGF receptor (EGFR) signaling pathway is commonly

activated in head and neck squamous cell carcinoma(HNSCC) and represents a validated target for therapy.Grandis and colleagues first demonstrated that EGFR over-expression is very commonmolecular alteration in HNSCC(1), whereas further work revealed that the intensity of itsexpression is linked to reduced survival (2, 3). EGFR acti-vation triggers signal transduction cascade that includesactivation of the Ras/Raf/mitogen-activated protein kinase(MAPK) and phosphoinositide 3-kinase (PI3K)–Akt signal-ing pathway (4–6).

Cetuximab is chimeric IgG1-human antibody directedagainst the extracellular domain of EGFR, blocking ligandbinding to receptor and has been approved in the UnitedStates for treatment of HNSCC. Cetuximab, however, hasobjective response rate of only 13% when used as a singleagent. There is therefore clear need for biomarkers predic-tive of response to cetuximab to maximize likelihood ofresponse while minimizing toxicities.

One of the mechanisms of cetuximab resistance may bethe presence of mutations that result in constitutive activa-tion of EGFR-mediated signaling. In colon cancer, muta-tions that constitutively activate key signaling mediatorsdownstream of EGFR, particularly KRAS, have been asso-ciated with cetuximab resistance.

Activating point mutations in genes encoding the Rassubfamily of small GTP-binding proteins contribute to theformation of a large proportion of human tumors. Theidentification ofRas-related resistancemechanisms to EGFRinhibitors remains critical to the clinical management ofpatients with head and neck cancer. Although KRAS muta-tions are rare (approximately 1%) in HNSCC, HRASmuta-tions seem more common, whereas the reverse is true forseveral other malignancies (7, 8).

Understanding the mechanisms of cetuximab resistancemay help delineate the subgroup of patients with headand neck cancer that can truly benefit from cetuximab. We

Authors' Affiliations: 1Department of Surgery, Yale University School ofMedicine, New Haven, Connecticut; 2Department of Medicine, AttikonHospital,; and 3Department of Biochemistry andMolecular Biology, Facultyof Biology, National Kapodistrian University of Athens, University ofAthens, Greece

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

Corresponding Author: Amanda Psyrri, Yale Cancer Center, P.O. Box208032, New Haven, CT 06520. Phone 203-737-2476; Fax 203-785-7531;E-mails: [email protected] or [email protected]

doi: 10.1158/1078-0432.CCR-13-2721

�2014 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org OF1

Cancer Research. on September 18, 2018. © 2014 American Association forclincancerres.aacrjournals.org Downloaded from

Published OnlineFirst April 2, 2014; DOI: 10.1158/1078-0432.CCR-13-2721

http://clincancerres.aacrjournals.org/

hypothesized, therefore, that HNSCC with constitutivelyactivated downstream signaling mediators of the EGFRpathway, particularly HRAS, would be refractory to inhibi-tion of the pathway at the receptor level. To definemechan-isms of de novo cetuximab resistance, we studied a series ofcetuximab-resistant head and neck squamous cell carcino-ma cell lines, including cell lines bearing HRAS or PIK3CAmutations and the transgenic mouse model of RAS-drivenoral tumor development. We combined our findings withstudies of tumor specimens from patients with cetuximab-treated HNSCC.

Materials and MethodsPatient population

One hundred eighty tumor specimens were obtainedfrom Attikon Hospital archives after institutional reviewboard approval. All patients provided written informedconsent. The clinopathologic data of patients are providedin Supplementary Table S7.

Cell lines, reagents, and MTT viability assaysThe HRAS-mutated squamous head and neck cell line (p.

Q61L, homozygous) BB49 (9), the PIK3CA-mutated squa-mous head and neck cell line (p.H1047R, heterozygous)Cal-33 (10), and the A-431 epidermoid carcinoma cell linewere obtained from the American Type Tissue Collection.The head and neck cancer cell lines UM-SCC1, UM-SCC25,UM-SCC6, UM-SCC11 that harbor wild-type HRAS, andPI3KCA were obtained from the University of Michigan(11). All cell lines were authenticated by DNA typing andusedwithin 6months of receipt (12). Themutation status ofHRAS and PIK3CA genes was verified by sanger sequencingin all cell lines. Details about cell culture, reagents, andMTTviability assays are described in the Supplementary Materi-als and Methods.

GenomicDNA extraction andHRASmutationdetectionA total of 180 genomic DNAs from head and neck cancer

samples were analyzed in this study. Genomic DNA wasextracted from 10-mm paraffin-embedded sections of thetumor samples. Slides were microscopically examined andtumor areas were marked and carefully dissected undermicroscopic observation. DNA extraction from dissectedmaterial was performed using the EX-WAX DNA ExtractionKit (Millipore) according to the manufacturer’s tissue pro-tocol. The details ofHRASmutation detection are providedin Supplementary Materials.

Lentivirus-mediated short hairpin RNA silencing ofHRAS gene

The lentiviral PLKO.1 puro vectors (MISSION shRNAplasmids) encoding the short hairpin RNA #64(TRCN0000033264) and #65 (TRCN0000033265) thattarget the HRAS coding sequence were purchased fromSigma-Aldrich. The MISSION pLKO.1 puro-non-targetshort hairpin RNA (shRNA) plasmid (SHC016-1EA)served as negative control. The specific shRNA sequencesand further details are provided in the SupplementaryMaterials and Methods.

Lentivirus-mediated overexpression of G12V HRASThe lentivirus vector pLenti CMV RasV12 Neo (w108-

1) encoding the G12V HRAS (plasmid 22259, created byProf. E. Campeau, Addgene) or control empty vector werecotransfected with the packaging plasmids pMD2.G andpsPAX2 (Addgene) to 293T cells. Titrations and infectionswere performed as previously described (13). Furtherdetails are provided in the Supplementary Materials andMethods.

Mouse breeding and adenovirus infectionTo study the expression signature of oncogenic RAS in

an in vivomurine model of head and neck cancer, we usedthe eEF1a1-KrasG12D knock-in mice (Mouse GenomeInformatics ID: MGI: 3837679). In this strain, upon Crerecombination of the loxP sequences, the floxed block isremoved and the mouse Kras cDNA containing a G to Anucleotide substitution in codon 12 (GGT ! GAT) isconditionally expressed under the eukaryotic translationelongation factor 1 alpha 1 (eEF1a1) promoter. Genotyp-ing of the eEF1a1-KrasG12D mice was performed by PCRanalysis of tail biopsy using specific primers (14). Toinduce recombination of the loxP sequences, a recombi-nant adenovirus (rAd) expressing Cre recombinase underthe control of the cytomegalovirus (CMV) promoter (Ad-CMV-Cre; Clontech) was used for injections in oral tissue.Adevovirus expressing GFP under the control of the CMVpromoter (Ad-CMV-GFP; Clontech) was used as a con-trol. Viruses were propagated in 293A cells and titratedusing a plaque assay as previously described (15). Allexperiments with eEF1a1-KrasG12D mice were performedwith approved animal protocols according to the insti-tutional guidelines established by Biomedical ResearchFoundation of the Academy of Athens.

Translational RelevanceCetuximab has been approved for treatment of head

and neck squamous cell carcinoma (HNSCC). Despitegreat clinical promise, however, only 13% of patientsrespond to cetuximab when used as a single agent.Herein, by using a combination of cetuximab-resistantcell lines and a transgenic mouse model of RAS-drivenoral cancer combined with analyses of patients withcetuximab-treated HNSCC, we uncover aberrant RAS/MAPK/ERK signaling as a central mediator of cetuximabresistance. In our study, the gene expression signature ofC-myc, BCL-2, BCL-XL, and cyclinD1 upon activation ofMAPK signalingwas not altered by cetuximab treatment,suggesting that this signature may have a pivotal role incetuximab resistance of RAS activated HNSCC. We fur-ther showed that repression ofHRAS expression restoresthe ability of cetuximab to suppress the growth of headand neck squamous cancer cells containing activatingmutations in HRAS, pointing to a promising resistancefighting approach.

Rampias et al.

Clin Cancer Res; 20(11) June 1, 2014 Clinical Cancer ResearchOF2




Western blot analysis and quantitative reversetranscription (RT)-PCRMethodology is provided in the Supplementary

Materials.

GST-Raf-1 RBD pull-down assayThe levels of active, GTP-HRAS were determined by GST-

Raf-1RBD pull-down assay. The methodology of GST-Raf-1RBD pull-down assay is provided in the SupplementaryMaterials.

Confocal microscopyFor confocal microscopy, cells were fixed with 4% para-

formaldehyde for 15 minutes, permealized with 0.5% Tri-ton X-100 for 10minutes, and incubated with an a-tubulinmousemonoclonal antibody conjugated to Alexa Fluor 488fluorescent dye (DM1A clone; Cell Signaling; 1:200 dilu-tion) Further details are provided in the SupplementaryMaterials.

Matrigel invasion assay and anchorage-independentgrowth assayMethodology is provided in the Supplementary

Materials.

ResultsHRAS is frequently mutated in HNSCCA total of 180Formalin-Fixed, Paraffin-Embedded (FFPE)

HNSCCwere evaluated forHRASmutations.HotspotHRASmutations (codons 12,13, and 61) were found in 11 of 180(6.67%) specimens. In addition, rare heterozygous HRASmutations were detected in codon 14(V14M), 17(S17I), 54(D54G), and 63(E63K) (Supplementary Fig. S2). The over-all frequency of HRAS mutations in our cohort of HNSCCswas 9.44% (Supplementary Results, Table S3), which isrelatively higher to the reported frequency (8%) of HRASmutations in HNSC in COSMIC mutation database (16).Recently, Pickering and colleagues in whole exomesequencing of 38 oral squamous cell carcinomas reportedthat 9% of their samples harbor HRAS alterations (17).

HRAS silencing reduces the proliferation capacity ofhead and neck cancer cellsMTT viability assays showed that cell lines containing

mutations in HRAS (BB49) or PIK3CA (Cal-33) displayedrobust cetuximab-resistant phenotype when challengedwith 50 nmol/L of cetuximab as compared with controltreated cells (Supplementary Fig. S1).To further address the molecular mechanism underlying

the connection between activation of RAS/RAF/MAPK andPI3K signaling pathways and cetuximab resistance, wesilenced HRAS expression in a panel of cetuximab-resistanthead and neck cancer cell lines harboring activating muta-tions in either HRAS (BB49) or PIK3CA genes (Cal-33)along with cetuximab-resistant cell lines containing wild-typeHRAS andPIK3CA genes (UM-SCC11A,UM-SCC6). Asshown in Fig. 1A, silencing experiments revealed that HRASprotein levels were considerably downregulated (>90%) byshRNA#65 oligo whereas,HRAS silencing using shRNA#64

oligo was less efficient.HRAS silencing was associated withan altered morphology, which is characterized by a flat-shaped phenotype and remodeling of tubulin cytoskeleton(Fig. 1B). The phenotype of lentivirus infected cell lines,confirmed 100% efficiency ofHRAS silencing and selectionin our approach and also verified that the molecular profileof HRAS-depleted cell lines is homogeneous. As expected,phospho-ERK1/2 protein levels were substantially reducedin cells followingHRAS silencing, whereas EGFR, phospho-EGFR (Y1068), AKT, and ERK1/2 protein levels were notaffected (Fig. 1C). Interestingly, downregulation of HRASlevels was associated with decreased levels of phosphory-lated AKT, which is consistent with previously reporteddata showing that RAS and PI3K/AKT signaling pathwaysare highly interconnected (18–20). MTT viability assaysshowed that silencing of HRAS in BB49, Cal-33, UM-SCC11A, and UM-SCC6 was associated with a lower rateof proliferation (approximately, 20% lower) comparedwith uninfected or control infected cells (data not shown).However, HRAS-depleted cells were still able to proliferateandwe did not observe any degree of apoptotic cell death orsenescence. The ability of cells harboring low RAS levels toproliferate even with a low rate was also confirmed by softagar assays as shown in Fig. 1D. These results suggest that inHRAS-deficient BB49 and Cal-33 cells, ERK-independentsignaling pathways can still maintain a basal proliferationcapacity.

Cross-talk between RAS/MAPK and PI3K/AKTpathways in HNSCC

Activation of PI3K recruits AKT to plasma membranewhere AKT is phosphorylated at Thr308 and Ser473. Thedownregulation of phospho-AKT protein levels by HRASdepletion in BB49 and Cal-33 cells in our experimentssuggests that HRAS is important regulator of PI3K activityin both cell lines. To investigate whether phosphorylationof AKT is exclusivelymediated by PI3K,we treated BB49 andCal-33 cells with the PI3K inhibitor LY294002. Treatmentwith LY294002 inhibitor diminished the phospho-AKTlevels in both cell lines (Fig. 2A), suggesting that HRASdepletion in BB49 and Cal-33 cells is associated with stronginhibition of PI3K.

To investigate the role of HRAS and EGFR on PI3Kactivation in head and neck cell lines with oncogenic HRASor PI3KCA, we treated the wild type and the low RAS-containing BB49 and Cal-33 cell lines with cetuximab andPI3K activity was assessed by measuring phospho-AKTprotein levels. Interestingly, as shown in Fig. 2B, phosphor-ylation of AKT was substantially decreased in wild-typeBB49 cells treated with cetuximab, compared with cetux-imab untreated cells, suggesting that EGFR blockade leadsto complete PI3K inhibition in these cells despite thepresence of activated HRAS. This finding supports previousdata showing that the ability of oncogenic mutant RAS byitself to drive PI3Kmembrane localization and activation isprobably dependent on the signaling input from EGFreceptor (21). The fact that HRAS depletion in these cellscan cause a complete PI3K inhibition as well, even in the

Cetuximab Resistance and HRAS Oncogenic Mutations in HNSCC

www.aacrjournals.org Clin Cancer Res; 20(11) June 1, 2014 OF3




presence of activated EGFR, provides evidence that theactivation of wild-type PI3K in these cells requires theconcurrent activation of HRAS and EGFR proteins.

Wild-type Cal-33 cells harbor wild-type HRAS and theactivated mutation H1047R in PIK3CA gene. Interestingly,in cetuximabuntreatedCal-33 cells,HRASdepletion causeda complete inhibition of PI3K in the presence of activatedEGFR, suggesting that oncogenic PI3K (H1047R) requiresthe direct interactionwithHRAS to exhibit aberrant activity.

When parental Cal-33 cells were treated with cetuximab,PI3K was still able to promote the phosphorylation of AKTto some extent (p-AKT levels were less than 30% comparedwith the levels in untreated cells). These data suggest thatoncogenic PI3Kmaintains a low level of activity in presenceof HRAS even if EGFR is not activated. As expected, no

phosphorylated AKT was detected when we treated HRAS-depleted Cal-33 cells with cetuximab (Fig. 2B).

HRAS silencing in cetuximab-resistant head and neckcancer cell lines containing oncogenic HRAS or PI3K,restores their sensitivity to cetuximab

Cetuximab treatment was associated with a completegrowth inhibition of HRAS-depleted BB49 and Cal-33 celllines comparedwith cetuximab treated control infected andparental BB49 and Cal-33 cell lines, suggesting that EGFRblockade, in addition to AKT/MAPK abrogation followingHRAS depletion, is sufficient to suppress the growth ofHNSCC (Fig. 3A and B). To the contrary, cetuximab treat-ment of HRAS-depleted UM-SCC11A and UM-SCC6 celllines was not associated with a growth inhibition compared

unA

B

C D

scr #64 #65 un scr #64 #65 un scr #64 #65 un scr #64 #65

HRAS

β-Actin

BB49

BB49 LV_scr BB49 LV_#65 BB49 BB49 LV_scr BB49 LV_#65

Cal-33 LV_scr

un scr #64 #65 un scr #64 #65

Cal-33 LV_#65 Cal-33 Cal-33 LV_scr Cal-33 LV_#65

Cal-33 UM-SCC6 UM-SCC11A

EGFR

p-EGFR (Y1068)

p-AKT (Ser 473)

p-ERK1/2 (Thr202/Tyr204)

ERK1/2

HRAS

BB49 Cal-33

BB49 Cal-33

β-Actin

AKT 1 Wk

2 Wk

Figure 1. HRAS depletion affects cytoskeletal dynamics and cell proliferation. A, efficiency of HRAS silencing in BB49, Cal-33, UM-SCC6, and UM-SCC11Acells. Western blot analysis of HRAS levels was performed from protein extracts that were made from uninfected (un) cells and cells that were infectedwith lentivirus-expressing shRNA#64 (#64), shRNA#65 (#65), or the scrambled sequence control shRNA (scr). B, HRAS depletion affects cytoskeletaldynamics. Representative phase-contrast images of BB49 and Cal-33 cells that were infected with lentivirus-expressing shRNA#65 (LV65shRNA), or thescrambled sequence control shRNA (LVscr) and immunofluorescene analysis of a-tubulin fibers. C, effect of HRAS depletion on MAPK, AKT, andEGFR signaling in head and neck cell lines BB49 and Cal-33. Protein extracts were made from uninfected (un) cells and from cells that were infectedwith lentivirus expressing shRNA#64 (#64), shRNA#65 (#65), or the scrambled sequence control shRNA (scr). D, anchorage-independent growth ofBB49 and Cal-33 cells following knockdown of HRAS. Representative phase-contrast photomicrographs (10�magnification) depicting the colony growth insoft agar for BB49 and Cal-33 cells that were infected with lentivirus expressing shRNA#65, 7 and 14 days after seeding.

Rampias et al.





with the control infected and parental UM-SCC11A andUM-SCC6 cell lines, suggesting that cetuximab resistancemechanisms that are HRAS/PI3K independent cannot beovercomebyAKT/MAPKabrogation followingHRASdeple-tion or inhibition.Biochemical analysis of HRAS-depleted BB49 and Cal-33

cells and their control infected cells, revealed that theproliferation capacity of HRAS-depleted cells is mainlymaintained fromEGFR-dependent signaling pathways suchas the STAT3 pathway (Fig. 3C). Phospho-STAT3 proteinlevels were not affected byHRAS silencing in both cell lines,whereas these were diminished upon treatment of Cal-33and BB49 cells with 50 nmol/L cetuximab. Inactivation ofthe STAT3 pathway by cetuximab treatment was associatedwith growth suppression of BB49 and Cal-33 cells, onlywhen combined with AKT/MAPK abrogation, following

HRAS silencing, suggesting that the proliferation capacityin these cells is regulated by a fine balance between EGFR,RAS/MAPK, and PI3K/AKT signaling pathways.

The effect of cetuximab in viability of HRAS-depletedCal-33 cells was also confirmed by colony-formationassay. As shown in Fig. 3D, cetuximab treatment did notinhibit the anchorage-independent growth of Cal-33 cellsand had no effect on the number of soft agar colonies.However, the size of colonies in cetuximab-treated cellswas substantially reduced compared with untreated,which is in accordance to the finding that EGFR inhibi-tion decreases by approximately 10% the proliferationrate of Cal-33 cells on MTT viability assay. As expected,HRAS depletion, in combination with cetuximab treat-ment caused a complete inhibition of anchorage-inde-pendent growth of Cal-33 cells.

0 5 10 0 5 10 (h) 20 μmol/L LY294002

p-AKT (S473)

AKT

BB49 Cal-33Un LV

_scr

LV_s

hRNA

#65

Un LV_s

crLV

_shR

NA#6

5Un LV

_scr

LV_s

hRNA

#65

Un LV_s

crLV

_shR

NA#6

5

Cetuximab:

A

B

(50 nmol/L)EGFR

p-EGFR

p-AKT (Ser 473)

AKT

p110α (PI3K)

HRAS-GTP

CAL-33BB49

β-Actin

– – – + + + – – – + + +

Figure 2. HRAS expression levels affect PI3K activity and phosphorylation of AKT. A, sustained PI3K inhibition decreases AKT phosphorylation in BB49 andCal-33 cells. BB49 and Cal-33 cells were starved for 24 hours, treated with 20 mmol/L LY294002 for 0, 5, and 10 hours and lysates were evaluated forphospho and total AKT expression by immunoblot. B, HRAS depletion is associated with downregulation of phospho-AKT levels in BB49 and Cal-33 cells.Wild-type (un) and lentivirus infected BB49 and Cal-33 cells expressing either a scambled shRNA (LVscrshRNA) or the shRNA #65 oligo (LV65shRNA)were either treated with 50 nmol/L cetuximab (þ) or not (�). The phospho and total levels of EGFR and AKT were measured by immunoblot analysis of wholeprotein lysates. The protein expression levels of PI3K were measured in the same immunoblot analysis using a specific antibody against the p110a domain.Actin was used as loading control. H-Ras-GTP protein levels were purified using theGST-Raf-RBD pull-down assay. In brief, 107 cells from each cell line werelysed in ice-coldRBD lysis buffer and500mgofwhole protein lysatewere incubatedwith 20mgGST-Raf-RBDglutathione agarose beads for 1 hour at 4�C. TheHRAS-GTP protein levels were measured by immunoblot analysis using a specific anti-HRAS antibody (c-20; Santa Cruz).






120

100

80

60

40

20

0

A

B

C

D

120

100

80

60

40

20

0

120

100

80

60

40

20

0

un+cet

scr+cet

LV_65+cet

BB49 CAL-33 UM-SCC6 UM-SCC11A

Cel

l via

bilit

y (%

of c

ontr

ol)

Cel

l via

bilit

y (%

of c

ontr

ol)

Cel

l via

bilit

y (%

of c

ontr

ol)

0 10 20 30 40 50 60

Cetuximab (nmol/L)

Cetuximab:(50 nmol/L)

Cetuximab:(50 nmol/L)

– + + +

0 10 20 30 40 50 60

BB49 LV_ scr shRNA

BB49 LV_shRNA#65 Cal-33 LV_shRNA#65

Cal-33 LV_scr shRNA

p-STAT3 (Tyr705)

STAT3

CAL-33BB49

β-Actin

Un

LV_S

crLV

_shR

NA#

65U

nLV

_Scr

LV_s

hRN

A#65

Un

LV_S

crLV

_shR

NA#

65U

nLV

_Scr

LV_s

hRN

A#65

– – – + + + – – – + + +

25

20

15

10

5

0Un Un LV_Scr LV_shRNA#65

Un Un LV_Scr

CAL-33

LV_shRNA#65

Col

ony

num

ber

Cetuximab (nmol/L)

Figure 3. HRAS silencing incombination with cetuximabtreatment suppress the growth ofBB49 and Cal-33 cells. A,cetuximab growth response ofHRAS-depleted BB49 and Cal-33,UM-SCC6, and UM-SCC11A celllines. Uninfected (un) andlentivirus-infected BB49 andCal-33 cells expressing either ascambled shRNA (scr) or theshRNA #65 oligo (LV_65) weretreated with 50 nmol/L cetuximabfor 72 hours. Growth wasmeasured using the MTT viabilityassay and plotted as a percentageof growth relative to the uninfected(un)cells of each cell line. Datapoints are represented as mean �SEM (n ¼ 3). B, cetuximabsensitivity in HRAS-depleted headand neck cell lines. Lentivirus-infected BB49 and Cal-33 cellsexpressing either a scrambledshRNA (scr) or the shRNA#65 oligowere treated with cetuximab at theindicated concentrations for72 hours, and viable cells weremeasured by MTT assay andplotted (mean � SD) as apercentage of growth relative tountreated controls. C, EGFRsignaling promotes STAT3activation inBB49andCal-33 cells.Uninfected (un) and lentivirusinfected BB49 and Cal-33 cellsexpressing either a scambledshRNA (scr) or the shRNA#65 oligo(LV_65) were either treated with50 nmol/L cetuximab (þ) for72 hours or not (�). Phospho andtotal levels of STAT3 weremeasured by immunoblot analysis.Actin was used as loading control.D, HRAS depletion in combinationwith cetuximab treatment inhibitsthe colony formation ofCal-33 cellsin soft agar assay. Uninfected (un)and lentivirus-infected Cal-33 cellsexpressing either a scrambledshRNA (scr) or the shRNA#65 oligo(LV_65) were either treated with50 nmol/L cetuximab (þ) or not (�).After 2 weeks of growth, thecolonies were stained with crystalviolet and quantified by lightmicroscopy. The relative growth ofcolonies for each cell line isprovided by representative phase-contrast photomicrographs (10�magnification) 10 days afterseeding.

Rampias et al.





Oncogenic HRAS activates MAPK and AKT signalingpathways in HNSCCAs described above, HRAS depletion in cetuximab-resis-

tant cell lineswith oncogenicmutations inHRAS orPIK3CArestored their sensitivity to cetuximab. In a next step, weasked whether overexpression of oncogenic G12VHRAS incetuximab-sensitive cell lines can lead to specific molecularsignature that is associated with cetuximab resistance.Among HNSCC cell lines tested for cetuximab resistance,UM-SCC1 and UM-SCC25 cell lines were found to becetuximab sensitive and their growth was suppressed bymore than 50% in the presence of 50 nmol/L cetuximab inMTT viability assays. Sequencing analysis of HRAS andPIK3CA genes revealed that both cell lines harbor wild-typeHRAS and PIK3CA genes (data not shown).In this direction, the cetuximab-sensitive cell lines UM-

SCC1 and UM-SCC25 were either mock infected, orinfected with a lentivirus expressing the oncogenic G12VHRAS protein. G12V HRAS–expressing cell lines exhibit amore rounded shape; however, the cellular morphologicchanges were less marked than those observed in HRAS-depleted head and neck cancer cells (Fig. 4A). As shownin Fig. 4B, a robust overexpression of HRAS protein wasdetected in cell lines infected by lentivirus expressing theoncogenic G12V HRAS, compared with wild-type or mock-infected cell lines.As expected, overexpression of activated HRAS was asso-

ciated with activation of MAPK and AKT signaling path-ways. As shown in Fig. 4B, we observed amarked activationof p42Erk2 and a marked upregulation on the levels ofphospho AKT in UM-SCC1 and UM-SCC25 cancer cellswith oncogenic HRAS overexpression.

The gene expression signature upon activation of theMAPK pathway by oncogenic RAS in HNSCCUpon activation, MAPKs phosphorylate and control the

activity of key nuclear proteins, which in turn can regulategene expression. To identify the group of downstreamtranscriptional factors that their expressionwas upregulatedbecause of the G12V HRAS–driven activation of the MAPKpathway, we compared the expression profile of knownnuclear transcriptional targets of the MAPK pathwaybetween the G12V HRAS–expressing UM-SCC1 and UM-SCC25 cells and their mock infected controls by Westernblot and qRT-PCR analyses.Among the transcription factor targets of the MAPK

pathway included in our expression analysis, phospho-CREB and phospho-Elk1 were found to be significantlyupregulated inG12VHRAS–expressing cells comparedwithcontrols, whereas total levels of CREB and Elk1 were unaf-fectedbyG12VHRAS expression (Fig. 4C). Elk1 is oneof thebest-studied targets of the ERK cascade as ERKs phosphor-ylate Elk1 in several sites increasing its transactivationpotential mostly by allowing increased interaction withother cofactors such as the histone acetyl-transferases (HAT)CBP/p300 and the mediator coactivator complex (22, 23).Phosphorylation of CREB is regulated by ERK5, a MAPK

distantly related to ERK1/2 that is also activated by activated

HRAS and has been implicated as important for cellularsurvival in cultured cells (24–26). CREB within the cell isbelieved to bind the DNA regulatory element CRE, either asa homodimer or as a heterodimer, withATF-1. Interestingly,phosphorylated ATF-1 protein levels were significantlyreduced upon G12V HRAS overexpression, suggesting thatnuclear ATF-1/CREBheterodimers are probably replaced byCREB homodimers on the CRE binding transcriptionalactivator complexes.

Another direct target of MAPK signaling is the product ofproto-oncogene v-Myc Myelocytomatosis Viral OncogeneHomolog (c-myc). Previous studies have shown that phos-phorylation of Myc at Ser 58 and Ser 62 by MAPK stabilizesMyc, allowing Myc to activate transcription as a heterodi-meric partner with Myc-associated factor X (Max; refs. 27and 28). It is worth noting that in our expression analysis,we observed a robust upregulation of phospho and total c-myc protein levels (Fig. 4C).

Cyclin D1 proto-oncogene associates with its bindingpartners cyclin-dependent kinase 4 and 6 (CDK4 andCDK6), and forms active complexes that promote cell-cycleprogression by phosphorylating and inactivating the reti-noblastomaprotein (29). InHNSCC,high cyclinD1expres-sion is correlated with chemotherapy failure and poorprognosis (30–32). In response to G12V HRAS–drivenactivation of the MAPK pathway, we observed a robustupregulation of cyclin D1 protein levels in UM-SCC1 cells,whereas cyclin D1 levels were only slightly upregulated inUM-SCC25 cells (Fig. 4C).

Previously, it has been shown that expression of anactivated formof KRAS (KRASG12D) inmicemodels initiatessquamous tumour formation in the oral epithelium as wellas in the skin (33–36). To validate the specific role of c-mycand cyclin D1 overexpression in cell proliferation of RAS-driven squamous cell carcinoma in head and neck cancer,we used an eEF1a1 KrasG12D mouse model. In this mouse,the highly expressed eEF1a1 locus, serves as recipient site forknock-in of the sequence encoding the active oncogenicKrasG12D cDNA. The tissue-specific and inducible control ofoncogene expression is achieved by the Cre/LoxP technol-ogy (14).

Activation of oncogenic KRAS expressionwas achieved byremoving the floxed block using recombinant adenovirus-expressing cre recombinase (Ad-CMV-Cre).More specifical-ly, we delivered Cre with intra-oral submucosal injectionsof 2.5 � 107 particles of adenovirus in head and necktissue (tongue, oral cavity, larynx) of anesthetized eEF1a1KrasG12D mice. Four weeks after injection of recombinantadenovirus, oral verrucous squamous carcinoma wasdetected in 20% (4/10 mice) of injected mice (Fig. 4D).Microscopically, all examined cases of RAS-induced carci-nomas were histologically homogeneous and consistedof large masses of well-differentiated epithelium with ker-atinized SCC with clear evidence of keratinization. Immu-nostaining analysis showed a strong expression of phos-phor-ERK, cyclin D1, and c-myc, suggesting that theseproteins play apivotal role in cell proliferationofRas-drivencancer.






UM-SCC25LV_EV

A

C

B

D

UM-SCC25LV_G12VHRAS

UM-SCC1LV_EV

UM-SCC1LV_G12VHRAS

UN

UN

LV_G

12VH

ras

LV_G

12VH

ras

LV_E

V

LV_E

V

UN

LV_G

12VH

RAS

LV_E

V

UN

LV_G

12VH

RAS

LV_E

V

Cyclin D1

p-CREB (Ser133)

CREB

p-ATF1(Ser63)

ATF1

p-cMyc (Thr58/Ser62)

c-Myc

p-Elk1 (Ser383)

Elk1

UM-SCC1UM-SCC25

UM-SCC1UM-SCC25

β-Actin

HRAS

ERK1/2

p-ERK1/2 (Thr202/Tyr204)

AKT

p-AKT (Ser 473)

p-EGFR(Y1068)

EGFR

β-Actin

eEF1a1 KrasG12D

oral tumor H&N Pankeratin

phospho-ERK Cyclin D1 c-Myc

Figure 4. Effect of oncogenic RAS expression in cell proliferation. A, bright field microscopic images of UM-SCC1 and UM-SCC25 cells that were either mockinfected (empty vector, LV_EV) or infectedwith a lentivirus expressingG12VHRAS (LV_G12VHRAS). Scale bar represents 100mmol/L. B, effect ofG12VHRASexpression in MAPK and AKT signaling pathways. Protein lysates from UM-SCC1 and UM-SCC25 cells that were uninfected (un), mock infected (LV_EV), orinfected with a lentivirus expressing G12V HRAS (LV_G12VHRAS) were used for Western blot analysis of the indicated proteins. Actin expression served asloading control. C, G12V HRAS signature in expression profile of transcription factors. Protein lysates from UM-SCC1 and UM-SCC25 cells that wereuninfected (un), mock infected (LV_EV), or infected with a lentivirus-expressing G12V HRAS (LV_G12VHRAS) were used for Western blot analysis of theindicatedproteins. Actin expression servedas loading control. D, gross appearanceof oral tumorsdeveloped in eEF1a1KrasG12Dmice4weeksafter Ad-CMV-Cre injection. Hematoxylin and eosin staining and immunohistochemistry analysis of oral tumors for pankeratin, phospho-ERK, cyclin D1, and c-Myc.

Rampias et al.





G12V HRAS expression profile in HNSCC is associatedwith an invasive phenotype and with resistance tocetuximabActivated HRAS in cooperation with TGF-b has been

implicated in tumor progression, invasion and metastasis,inducing an EMT phenotype in squamous cell carcinoma(37). Therefore, we evaluated the effect of G12V HRASsignaling on cell invasion, using matrigel-coated transwellculture chambers. After 48 hours of incubation, invadingcells, attached to the membrane were stained and counted.We observed that forced expression of G12V HRAS in bothUM-SCC1 and UM-SCC25 cell lines did indeed stimulatecell invasion compared with their respective mock-infectedcells (Fig. 5A). To investigate whether the G12V HRAS–driven invasive phenotype of UM-SCC1 and UM-SCC25cells is also cetuximab resistant, the invasion assays wererepeated in presence of 50 nmol/L cetuximab. As a result,themock-infected cells fromUM-SCC1 andUM-SCC25 celllines failed to survive or invade to the matrigel matrix,consistent with our data that these cell lines are cetuximabsensitive. To the contrary, cetuximab-treated UM-SCC1 andUM-SCC25 cells expressing G12V HRAS were cetuximabresistant and capable of invading to the matrigel matrix. Asshown in Fig. 5A, the number of membrane-attached cellsfrom UM-SCC1 and UM-SCC25 cell lines expressing G12VHRAS is not affected by cetuximab treatment, suggestingthat oncogenic RAS signaling can maintain the prolifera-tion, survival, and invasion potential in HNSCC. About theexpression of biomarkers that are associated with cell–cellor cell–matrix interactions and survival, weobserved a slightreduction of E-cadherin and a marked overexpression ofCD44v6 levels in G12V HRAS cells (Fig. 5B). We alsoobserved that the levels of antiapoptotic proteins BCL-2and BCL-XL were significantly increased in G12V HRAScells, whereas the expression of proapoptotic Bad proteinwere unaffected by overexpression of oncogenic HRAS(Fig. 5B).To further study the cetuximab resistance of UM-SCC1

and UM-SCC25 cells expressing G12V HRAS, we comparedthe growth between G12V HRAS–overexpressing cells andtheir relative mock-infected controls in different cetuximabconcentrations byMTT viability assay. In these experiments,when cetuximab was not added in the culture medium, thegrowth of G12V HRAS–expressing UM-SCC25 and UM-SCC1 cells was only 2% and 5% increased, respectively,compared with their mock-infected control cells. MTT via-bility assays with increasing concentrations of cetuximab(0–50 nmol/L) in themedium did not cause any significantgrowth suppression of G12V HRAS–expressing UM-SCC25and UM-SCC1 cells. To the contrary, we observed markedgrowth suppression in mock-infected cells, consistent withour observation that the parental cell lines were cetuximabsensitive. As shown in Fig. 5C, treatment with 25 nmol/Lcetuximab caused a 55% and 50% growth suppression tothe mock-infected UM-SCC1 and UM-SCC25 cells, respec-tively, compared with their G12V HRAS–expressing cellsOverall, these findings indicate that head and neck cancer

cells with oncogenic RAS signaling exhibit an aggressive

phenotype, which is characterized by cetuximab resistanceand enhanced invasion potential.

The gene expression signature of G12V HRAS about C-myc, BCL-XL, and cyclin D1 in HNSCC is not altered bycetuximab treatment

In a next step, we asked whether the expression profile ofG12V RAS signaling in HNSCC is altered by cetuximabtreatment. More specifically, we investigated whether theprotein and mRNA expression of c-myc and cyclin D1, askey drivers of proliferation, and of BCL-XL, as key driver ofantiapoptosis, and survival in G12V HRAS–expressing cellsis affected by EGFR signaling inhibition. As shown in Fig.6A, treatment of UM-SCC1 and UM-SCC25 cells expressingG12VHRASwith 50nmol/L cetuximab for 72hours did notalter the protein expression profile of c-myc, cyclin D1, andBCL-XL. Quantitative real time PCR analysis confirmed thatthe gene expression profile of c-myc, cyclin D1, and BCL-XLwas unaffected by cetuximab treatment in cells with onco-genic HRAS expression and also revealed that c-myc upre-gulation is mediated by both posttranslational modifica-tions (phosphorylation on threonine 58 and serine 62) andtranscriptional activation (Fig. 6B). Taken together, thesedata suggest that the expression signature of G12V HRASabout c-myc, BCL-XL and cyclinD1 inHNSCC is not alteredby cetuximab treatment.

HRASmutation is associated with de novo resistance inpatients with cetuximab-treated HNSCC

On the basis of our in vitro findings demonstrating a rolefor HRAS mutation in causing cetuximab resistance, wesought to determinewhether thismechanism alsomediatesclinical cetuximab resistance. We evaluated the clinicalimpact of de novo HRASmutation in a cohort of 55 patients(HRASwild-typen¼48;HRASmutantn¼7)whohadbeentreated with cetuximab-based chemoradiotherapy. HotspotHRASmutations (codon 12, 13, and 61) were detected in 7of 55 (12.7%) primary HNSCC specimens. There was noevidence for association between HRAS mutation and age(P¼ 0.60), sex (P¼ 0.33), tumor–node–metastasis (TNM)stage (P ¼ 0.99), ETOH (P ¼ 0.69), surgery (P ¼ 0.21),tobacco use (P ¼ 0.99), and subsite (P¼ 0.10; Supplemen-tary Table S4 and Supplementary Results). HRASmutationwas associated with lower likelihood of attaining response(P ¼ 0.046, Fisher exact test) to cetuximab-based therapy.Specifically, the group of "nonmutants" was more likely tohave complete or partial response, as compared withmutants (81.3% vs. 42.9%, respectively). Time to progres-sion (TTP) was significantly longer for patients withoutHRAS mutation (P ¼ 0.053, log rank test). Specifically, themedian TTP for HRAS mutants was 30 months whereasmedian TTP for patients with no HRAS mutation hasnot been reached (Fig. 6C). The median OS was longerfor patients without HRAS mutation (HRAS mutants24 months; HRAS nonmutants, 48 months, P ¼ 0.21).Using the Cox proportional hazards model, we carried outmultivariable analysis to assess the prognostic value ofHRASmutation for TTP andOS.We included the following






Cetuximab :

A

B

C

(50 nmol/L)– + Cetuximab :

(50 nmol/L)– +

LV_EV

LV_G12VHRAS

UM-SCC25UM-SCC1

UN

LV_E

V

LV_G

12VH

RAS

UN

LV_E

V

LV_G

12VH

RAS

BCL-XL

BCL2

Bad

E-cadherin

CD44v6

β-Actin

UM-SCC25 UM-SCC1

Cetuximab (nmol/L)

UM-SCC1 LV_G12VHRAS

UM-SCC1 LV_EV

UM-SCC25 LV_G12VHRAS

Cel

l via

bilit

y (%

of c

ontr

ol)

Cel

l via

bilit

y (%

of c

ontr

ol)

UM-SCC25 LV_EV

Cetuximab (nmol/L)

120

100

80

60

40

20

0

120

100

80

60

40

20

00 10 20 30 40 50 60 0 10 20 30 40 50 60

Rampias et al.





prognostic variables in the regression model: gender, TNMstage, age, tobacco, alcohol use, and surgery. Multiple Coxregression analysis examining the relationship between OSand HRAS mutation did not show a statistically significanteffect (P ¼ 0.178), whereas HRASmutation was associatedwith shorter TTP (P ¼ 0.022) with a HR for HRAS mutanttumors of 5.098 (Supplementary Table S6 and Supplemen-tary Results).Collectively, these clinical studies further support our in

vitro studies and demonstrate that HRAS mutation is asso-ciated with de novo resistance to cetuximab-based therapy inpatients with HNSCC.

DiscussionCetuximab is a validated target for therapy in HNSCC.

Cetuximab has been approved in combination with radio-therapy in locally advanced HNSCC and in recurrent met-astatic setting based on results of pivotal phase III studies l(38, 39). Despite its great clinical promise, the majority ofHNSCC patients do not respond to cetuximab.Findings from studies of drug resistance to EGFR-targeted

therapies have been applied to develop the next generationof clinical trials in lung cancer (40). In contrast, there hasbeen limited exploration of mechanisms of resistance tocetuximab in patients with HNSCC. Herein, by using acombination of cetuximab-resistant cell lines and a trans-genicmousemodel of RAS-driven oral cancer, coupledwithanalyses of patients with cetuximab-treated HNSCC, weuncover aberrant RAS/MAPK/ERK signaling as a centralmediator of cetuximab resistance in HNSCC. In our study,we found a high incidence of HRAS mutations in HNSCC(9.44%). HRAS mutation status was associated with lowerlikelihood of attaining complete response (CR) or partialresponse (PR) to treatment and significantly reduced TTP ina cohort of locally advanced HNSCC treated with cetux-imab-based therapy. In multivariate analysis, HRAS muta-tion status retained its prognostic significance for TTP afteradjustment for well-characterized prognostic indicators.Another interesting observation in our study was the

molecular crosstalk between the RAS/RAF/MAPK andPI3K/AKT pathways. PI3K and Ras are pivotal players insignal transduction pathways that coordinate proliferation,survival, and migration in head and neck cancer cells.Morris and colleagues performed sequencing and high-resolution copy number assessment of core componentsof the PI3K pathway in HNSCC (41). The authors reportedthat 74% of HNSCC contain activating genetic alterations,mainly copynumber alterations, of components of the PI3K

pathway. In our model system, we found that RAS isrequired for PI3K activity, even in the presence of activeEGFR signaling. Our data suggest that constitutive activa-tion of PI3K is not sufficient to promote its oncogenicactivity in the absence of activated RAS in HNSCC cells. Inthe same context, Rodriquez-Viciana and colleagues,showed that overexpression of a dominant-negative Asn17Ras mutant that does not interact with PI3K, partly inhibitsinduction of PIP3 by epidermal and nerve growth factor incultured PC12 cells (42). To the contrary, a recent study byChaussade and colleagues, based on in vitro PI3K assayswith exogenous H1047R PI3K, revealed that recombinantoncogenic RAS (G12V) does not significantly activate theH1047R PI3K (43). Another recent study by Burke andcolleagues, indicated that compared with the wild-typePI3K, mutant H1047R PI3K has higher affinity for lipidsand this may explain its higher basal kinase activity com-pared with the wild-type enzyme (44). It is worth notingthat the strong and clear dependence of H1047R PI3Kactivity on HRAS interaction observed in our study wasbased on endogenous levels of H1047R PI3K and HRASproteins in a cellular environment with activated EGFRsignaling; thus, this may explain the different effects onH1047R PI3K activity compared with previous in vitrostudies. Our findings may have tremendous implicationsfor overcoming resistance to cetuximab in HNSCC. If wepharmacologically modulate the interaction between PI3KandHRAS,wewill be able to restore sensitivity to cetuximabin cancers bearing genetic alterations of components ofPI3K or MAPK/RAF/RAS signaling pathways, which repre-sent the vast majority of HNSCC.

Another important finding of our study is that that theproliferation capacity of HRAS-depleted CAL33 and BB49cells is mainly maintained from EGFR-dependent signalingpathways such as STAT3 and therefore, EGFR blockadecould be combined with RAS inhibition for effective treat-ment of HNSCC containing aberrations in PI3K or RASsignaling pathways.

In our study, we used eEF1a1 KrasG12D transgenic mouseas model for Ras-driven oral tumor development. Activa-tion of oncogenic KrasG12D in oral cavity of this mousepromotes squamous cell carcinoma, which is characterizedby strong expression of phospho-ERK, cyclin D1, and c-myc. Consistent with these data, overexpression of G12VHRAS in UM-SCC1 and UM-SCC25 HNSCC cell lines thatharbor wild-type HRAS and PI3K revealed that overexpres-sion of cyclin D1 and c-myc has pivotal role in head andneck carcinoma with oncogenic RAS signaling. We also

Figure5. Effect ofG12VHRASexpressionpromotes cell survival, invasion, and resistance to cetuximab.A, effect ofG12VHRASexpressionon the cell invasioncapacity. UM-SCC1 and UM-SCC25 cells that were either mock infected (emty vector, LV_EV) or infected with a lentivirus-expressing G12V HRAS(LV_G12VHRAS) allowed to invade through transwell inserts (8 mm) coated with matrigel either in presence of 50 nmol/L cetuximab (þ) or not (�). After24 hours, noninvading cells were removedwith cotton swabs and invading cells on the reverse side of the filter were fixed, stained, and photographed under alight microscope. B, G12V HRAS signature in expression profile of proteins that promote cell survival and invasion. Protein lysates from UM-SCC1 and UM-SCC25 cells that were uninfected (un), mock infected (LV_EV), or infected with a lentivirus-expressing G12V HRAS (LV_G12VHRAS) were used for Westernblot analysis of the indicated proteins. Actin expression served as loading control. C, cetuximab resistance is associated with G12V HRAS expression. UM-SCC1 and UM-SCC25 cells that were mock infected (LV_EV) or infected with a lentivirus-expressing G12V HRAS (LV_G12VHRAS) were treated withcetuximab at the indicated concentrations for 72 hours, and viable cells were measured by MTT assay and plotted (mean � SD) as a percentage of growthrelative to untreated controls.






identified BCL-2 and BCL-XL as antiapoptotic proteins thatare significantly overexpressed upon activation of RAS sig-naling. Interestingly, the gene expression signature of onco-genic RAS about c-MYC, BCL-2, BCL-XL, and cyclin D1 wasnot altered by cetuximab treatment, suggesting that thissignature may have a pivotal role in cetuximab resistance of

RAS-activated HNSCC. One of the limitations of our studywas that we studied a single HNSCC cell line with HRASmutation and one cell line with PIK3CA mutation.

Our findings are directly relevant to patients withHNSCCthat are resistant to cetuximab-based therapy and may helpguide subsequent treatment. Several agents that target PI3K

HRAS MUTATION

NoYes

Yes-censoredNo-censored

UN

LV_E

VLV

_G12

VHra

sU

NLV

_EV

LV_G

12VH

ras

UN

LV_E

VLV

_G12

VHra

sU

NLV

_EV

LV_G

12VH

ras

Cetuximab :

A

B

C

– – – – – –+ + + + + +

(50 nmol/L) EGFR

p-EGFR(Y1068)

c-Myc

Cyclin D1

Bcl-xl

UM-SCC1UM-SCC25

β-Actin

1.0

0.8

0.6

0.4

0.2

0.0

1.0

0.8

0.6

0.4

0.2

0.0

0.0 10.0 20.0 30.0Time to progression (months)

Cum

. sur

viva

lC

um. s

urvi

val

40.0 50.0 60.0

0.0 10.0 20.0 30.0Overall survival (months)

40.0 50.0 60.0

TTP

OS

2.50

2.00

1.50

1.00

0.50

0.00

2.50

2.00

1.50

1.00

0.50

0.00

UMSCC-1 Empty G12V G12V

vector HRAS

Cyclin D1

Fol

d-ch

ange

exp

ress

ion

Fol

d-ch

ange

exp

ress

ion

c-Myc Bcl-xLHRAS+Cet


vector HRAS HRAS+Cet




vector HRAS

Cyclin D1 c-Myc Bcl-xLHRAS+Cet





Figure 6. G12V HRAS molecular phenotype is associated with cetuximab resistance in HNSCC. A, G12V HRAS expression signature is not altered bycetuximab treatment. UM-SCC1 and UM-SCC25 cells that were uninfected (un), mock infected (LV_EV), or infected with a lentivirus expressingG12V HRAS (LV_G12VHRAS) were either treated with 50 nmol/L cetuximab for 72 hours (þ) or not (�) and analyzed by immunoblotting to detect theprotein levels of c-myc, cyclin D1, and BCL-XL. Actin expression was used as loading control. B, transcriptional activation of c-myc, cyclin D1, andBCL-XL by oncogenic HRAS is not inhibited by EGFR blockage. Gene expression of c-myc, cyclin D1, and BCL-XL was examined with qRT-PCR analysis ofRNA harvested from UM-SCC1 and UM-SCC25 cells that were uninfected (un), mock infected (empty vector), or infected with a lentivirus expressingG12VHRASand either treatedwith 50 nmol/L cetuximab for 72 hours (G12VHRASþcet) or not (G12VHRAS). Gene expressionwas normalized to b-actin, andthe fold-change was calculated with respect to gene expression in uninfected cells. Data are from 3 individual experiments and are expressed as themean�SD. C, Kaplan–Meier survival analyses for TTP andOS. The comparison of TTP between patients with noHRASmutation (blue line, n¼ 48) andHRASmutants (green line, n ¼ 7) shows a statistically significant better prognosis for patients with no HRAS mutation (P ¼ 0.04). The comparison of OS betweenpatients with wild-type HRAS (blue line, n ¼ 48) and HRAS mutants (green line, n ¼ 7) does not show a statistically significant difference in OS (P ¼ 0.21).

Rampias et al.





or RAS signaling are undergoing clinical development.Hence, the findings from this study can be immediatelyused to design potential clinical therapies for patients withHNSCC. Given the retrospective nature of our studies inpatient samples, these findings need further clinical valida-tion. The frequency and the relationship ofHRASmutationin cetuximab-resistant cancers need to be fully assessed inprospective studies.

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

Authors' ContributionsConception and design: A. PsyrriDevelopment of methodology: T. Rampias, A. PsyrriAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): T. Rampias, H. Matsuzaki, A. Scorilas, A. PsyrriAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): T. Rampias, S. Siolos, A. Psyrri

Writing, review, and/or revisionof themanuscript: T. Rampias, S. Siolos,A. PsyrriAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): T. Rampias, S. Siolos, A. PsyrriStudy supervision: A. Psyrri

AcknowledgmentsThe authors thank Dr. C. Kontos for technical assistance.

Grant SupportThis work was supported by the 22430A OHSE CHARLES W 22430A

Science Award, Yale University.The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received October 7, 2013; revised March 6, 2014; accepted March 24,2014; published OnlineFirst April 2, 2014.

References1. Grandis JR, Tweardy DJ. Elevated levels of transforming growth factor

alpha and epidermal growth factor receptor messenger RNA are earlymarkers of carcinogenesis in head and neck cancer. Cancer Res1993;53:3579–84.

2. Psyrri A, Yu Z, Weinberger PM, Sasaki C, Haffty B, Camp R, et al.Quantitative determination of nuclear and cytoplasmic epidermalgrowth factor receptor expression in oropharyngeal squamous cellcancer by using automated quantitative analysis. Clin Cancer Res2005;11:5856–62.

3. Ang KK, Berkey BA, Tu X, Zhang HZ, Katz R, Hammond EH, et al.Impact of epidermal growth factor receptor expression on survival andpattern of relapse in patientswith advanced head and neck carcinoma.Cancer Res 2002;62:7350–6.

4. KalyankrishnaS,Grandis JR. Epidermal growth factor receptor biologyin head and neck cancer. J Clin Oncol 2006;24:2666–72.

5. McCubrey JA, Steelman LS, Chappell WH, Abrams SL, Wong EW,Chang F, et al. Roles of the Raf/MEK/ERK pathway in cell growth,malignant transformation and drug resistance. Biochim Biophys Acta2007;1773:1263–84.

6. Meloche S, Pouyssegur J. The ERK1/2 mitogen-activated proteinkinase pathway as amaster regulator of the G1- to S-phase transition.Oncogene 2007;26:3227–39.

7. SaranathD,ChangSE,Bhoite LT, Panchal RG,Kerr IB,MehtaAR, et al.High frequency mutation in codons 12 and 61 of H-ras oncogene inchewing tobacco-related human oral carcinoma in India. Br J Cancer1991;63:573–8.

8. Anderson JA, Irish JC, McLachlin CM, Ngan BY. H-ras oncogenemutation and human papillomavirus infection in oral carcinomas. ArchOtolaryngol Head Neck Surg 1994;120:755–60.

9. Mandruzzato S, Brasseur F, Andry G, Boon T, van der Bruggen P. ACASP-8 mutation recognized by cytolytic T lymphocytes on a humanhead and neck carcinoma. J Exp Med 1997;186:785–93.

10. Gioanni J, Fischel JL, Lambert JC, Demard F,MazeauC, Zanghellini E,et al. Two new human tumor cell lines derived from squamous cellcarcinomas of the tongue: establishment, characterization andresponse to cytotoxic treatment. Eur J Cancer Clin Oncol 1988;24:1445–55.

11. Brenner JC, GrahamMP, Kumar B, Saunders LM, Kupfer R, Lyons RH,et al. Genotyping of 73 UM-SCC head and neck squamous cellcarcinoma cell lines. Head Neck 2010;32:417–26.

12. Dirks WG, Drexler HG. Online verification of human cell line identity bySTR DNA typing. Methods Mol Biol 2011;731:45–55.

13. Rampias T, Sasaki C,Weinberger P, Psyrri A. E6 and e7 gene silencingand transformed phenotype of human papillomavirus 16-positiveoropharyngeal cancer cells. J Natl Cancer Inst 2009;101:412–23.

14. Klinakis A, Szabolcs M, Chen G, Xuan S, Hibshoosh H, Efstratiadis A.Igf1r as a therapeutic target in a mouse model of basal-like breastcancer. Proc Natl Acad Sci U S A 2009;106:2359–64.

15. Graham FL, Prevec L. Manipulation of adenovirus vectors. MethodsMol Biol 1991;7:109–28.

16. Forbes SA, Tang G, Bindal N, Bamford S, Dawson E, Cole C, et al.COSMIC (the Catalogue of Somatic Mutations in Cancer): a resourceto investigate acquired mutations in human cancer. Nucleic Acids Res2010;38:D652–7.

17. Pickering CR, Zhang J, Yoo SY, Bengtsson L, Moorthy S, Neskey DM,et al. Integrative genomic characterization of oral squamous cellcarcinoma identifies frequent somatic drivers. Cancer Discov 2013;3:770–81.

18. Kaur H, Park CS, Lewis JM, Haugh JM. Quantitative model of Ras-phosphoinositide 3-kinase signalling cross-talk basedonco-operativemolecular assembly. Biochem J 2006;393:235–43.

19. Pacold ME, Suire S, Perisic O, Lara-Gonzalez S, Davis CT, Walker EH,et al. Crystal structure and functional analysis of Ras binding to itseffector phosphoinositide 3-kinase gamma. Cell 2000;103:931–43.

20. Suire S, Hawkins P, Stephens L. Activation of phosphoinositide 3-kinase gamma by Ras. Curr Biol 2002;12:1068–75.

21. Schulze A, Lehmann K, Jefferies HB, McMahon M, Downward J.Analysis of the transcriptional program induced by Raf in epithelialcells. Genes Dev 2001;15:981–94.

22. Gille H, Sharrocks AD, Shaw PE. Phosphorylation of transcriptionfactor p62TCF by MAP kinase stimulates ternary complex formationat c-fos promoter. Nature 1992;358:414–7.

23. Cruzalegui FH, Cano E, Treisman R. ERK activation induces phos-phorylation of Elk-1 at multiple S/T-P motifs to high stoichiometry.Oncogene 1999;18:7948–57.

24. Watson FL, Heerssen HM, Bhattacharyya A, Klesse L, Lin MZ, SegalRA. Neurotrophins use the Erk5 pathway to mediate a retrogradesurvival response. Nat Neurosci 2001;4:981–8.

25. Barros JC, Marshall CJ. Activation of either ERK1/2 or ERK5 MAPkinase pathways can lead to disruption of the actin cytoskeleton. JCellSci 2005;118:1663–71.

26. Kamakura S, Moriguchi T, Nishida E. Activation of the protein kinaseERK5/BMK1 by receptor tyrosine kinases. Identification and charac-terization of a signaling pathway to the nucleus. J Biol Chem1999;274:26563–71.

27. Salghetti SE, Kim SY, Tansey WP. Destruction of Myc by ubiquitin-mediated proteolysis: cancer-associated and transforming mutationsstabilize Myc. EMBO J 1999;18:717–26.

28. Sears R, Leone G, DeGregori J, Nevins JR. Ras enhances Myc proteinstability. Mol Cell 1999;3:169–79.






29. Malumbres M, Barbacid M. Cell cycle, CDKs and cancer: a changingparadigm. Nat Rev Cancer 2009;9:153–66.

30. Feng Z, Guo W, Zhang C, Xu Q, Zhang P, Sun J, et al. CCND1 as apredictive biomarker of neoadjuvant chemotherapy in patients withlocally advanced head and neck squamous cell carcinoma. PLoSONE2011;6:e26399.

31. Zhang P, Zhang Z, Zhou X, Qiu W, Chen F, Chen W. Identification ofgenes associated with cisplatin resistance in human oral squamouscell carcinoma cell line. BMC Cancer 2006;6:224.

32. YuZ,Weinberger PM,Haffty BG, Sasaki C, Zerillo C, Joe J, et al. CyclinD1 is a valuable prognostic marker in oropharyngeal squamous cellcarcinoma. Clin Cancer Res 2005;11:1160–6.

33. Vitale-Cross L, Amornphimoltham P, Fisher G, Molinolo AA, GutkindJS. Conditional expression of K-ras in an epithelial compartment thatincludes the stem cells is sufficient to promote squamous cell carci-nogenesis. Cancer Res 2004;64:8804–7.

34. Gilad O, Nabet BY, Ragland RL, Schoppy DW, Smith KD, Durham AC,et al. Combining ATR suppression with oncogenic Ras synergisticallyincreases genomic instability, causing synthetic lethality or tumori-genesis in a dosage-dependent manner. Cancer Res 2010;70:9693–702.

35. Lapouge G, Youssef KK, Vokaer B, Achouri Y, Michaux C, Sotiropou-lou PA, et al. Identifying the cellular origin of squamous skin tumors.Proc Natl Acad Sci U S A 2011;108:7431–6.

36. Scheitz CJ, Lee TS, McDermitt DJ, Tumbar T. Defining a tissue stemcell-driven Runx1/Stat3 signalling axis in epithelial cancer. EMBO J2012;31:4124–39.

37. Oft M, Akhurst RJ, Balmain A. Metastasis is driven by sequentialelevation of H-ras and Smad2 levels. Nat Cell Biol 2002;4:487–94.

38. Bonner JA, Harari PM, Giralt J, Azarnia N, Shin DM, Cohen RB, et al.Radiotherapy plus cetuximab for squamous-cell carcinoma of thehead and neck. N Engl J Med 2006;354:567–78.

39. Vermorken JB, Specenier P. Optimal treatment for recurrent/met-astatic head and neck cancer. Ann Oncol 2010;21 Suppl 7:vii252–61.

40. Crawford J. New targeted therapies for non-small-cell lung cancer: afocus on the epidermal growth factor receptor. J Natl Compr CancNetw 2003;1 Suppl 1:S78–86.

41. Morris LG, Taylor BS, Bivona TG, Gong Y, Eng S, Brennan CW, et al.Genomic dissection of the epidermal growth factor receptor (EGFR)/PI3K pathway reveals frequent deletion of the EGFR phosphatasePTPRS in head and neck cancers. Proc Natl Acad Sci U S A2011;108:19024–9.

42. Rodriguez-Viciana P, Warne PH, Dhand R, Vanhaesebroeck B, Gout I,Fry MJ, et al. Phosphatidylinositol-3-OH kinase as a direct target ofRas. Nature 1994;370:527–32.

43. Chaussade C, Cho K, Mawson C, Rewcastle GW, Shepherd PR.Functional differences between two classes of oncogenic mutationin the PIK3CA gene. Biochem Biophys Res Commun 2009;381:577–81.

44. Hon WC, Berndt A, Williams RL. Regulation of lipid binding underliesthe activation mechanism of class IA PI3-kinases. Oncogene2012;31:3655–66.


Rampias et al.




Published OnlineFirst April 2, 2014.Clin Cancer Res T. Rampias, A. Giagini, S. Siolos, et al. Neck Squamous Cell CarcinomaRAS/PI3K Crosstalk and Cetuximab Resistance in Head and

Updated version

10.1158/1078-0432.CCR-13-2721doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2014/04/09/1078-0432.CCR-13-2721.DC1Access the most recent supplemental material at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/early/2014/05/06/1078-0432.CCR-13-2721To 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-13-2721

http://clincancerres.aacrjournals.org/content/suppl/2014/04/09/1078-0432.CCR-13-2721.DC1

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/early/2014/05/06/1078-0432.CCR-13-2721


RAS/PI3K Crosstalk and Cetuximab Resistance in...

Documents

Transcript of RAS/PI3K Crosstalk and Cetuximab Resistance in...