ORIGINAL ARTICLE

Lapatinib, a HER2 tyrosine kinase inhibitor, induces stabilization and

accumulation of HER2 and potentiates trastuzumab-dependent cell

cytotoxicity

M Scaltriti1, C Verma2, M Guzman1, J Jimenez3, JL Parra1, K Pedersen1, DJ Smith2, S Landolfi3,S Ramon y Cajal3, J Arribas1 and J Baselga1,4

1Medical Oncology Department, Vall d’Hebron Research Institute, Vall d’Hebron University Hospital, Barcelona, Spain;2Biomolecular Modeling and Design Group, Bioinformatics Institute, Singapore; 3Pathology Department, Vall d’Hebron UniversityHospital, Barcelona, Spain and 4Universitat Autonoma de Barcelona, Barcelona, Spain

Lapatinib is a human epidermal growth factor receptor 2(HER2) tyrosine kinase inhibitor (TKI) that has clinicalactivity in HER2-amplified breast cancer. In vitro studieshave shown that lapatinib enhances the effects of themonoclonal antibody trastuzumab suggesting partially non-overlapping mechanisms of action. To dissect these mechan-isms, we have studied the effects of lapatinib and trastuzumabon receptor expression and receptor signaling and haveidentified a new potential mechanism for the enhancedantitumor activity of the combination. Lapatinib, given aloneor in combination with trastuzumab to HER2-overexpressingbreast cancer cells SKBR3 and MCF7-HER2, inhibitedHER2 phosphorylation, prevented receptor ubiquitination andresulted in a marked accumulation of inactive receptors at thecell surface. By contrast, trastuzumab alone caused enhancedHER2 phosphorylation, ubiquitination and degradation of thereceptor. By immunoprecipitation and computational proteinmodeling techniques we have shown that the lapatinib-inducedHER2 accumulation at the cell surface also results in thestabilization of inactive HER2 homo- (HER2/HER2) andhetero- (HER2/EGFR and HER2/HER3) dimers. Lapatinib-induced accumulation of HER2 and trastuzumab-mediateddownregulation of HER2 was also observed in vivo, where thecombination of the two agents triggered complete tumorremissions in all cases after 10 days of treatment. Accumula-tion of HER2 at the cell surface by lapatinib enhancedimmune-mediated trastuzumab-dependent cytotoxicity. Wepropose that this is a novel mechanism of action of thecombination that may be clinically relevant and exploitable inthe therapy of patients with HER2-positive tumors.Oncogene (2009) 28, 803–814; doi:10.1038/onc.2008.432;published online 8 December 2008

Keywords: ErbB receptor; trastuzumab; tyrosine kinaseinhibitor; antibody-dependent cell cytotoxicity (ADCC);breast cancer

Introduction

The human epidermal growth factor receptor 2 (HER2)belongs to the HER family of tyrosine kinase receptors,which also includes HER1 (epidermal growth factorreceptor, EGFR), HER3 and HER4. Ligand bindingand/or receptor overexpression induces homo- orheterodimerization of HER receptors, transphosphor-ylation of the kinase domains and subsequent activationof downstream signaling (Tzahar et al., 1996; Yardenand Sliwkowski, 2001; Citri and Yarden, 2006; Serginaet al., 2007). Overexpression/amplification of HER2 isseen in approximately 25–30% of human breast cancersand is associated with a more malignant phenotype anda worse prognosis (Slamon et al., 1987, 1989). Trastu-zumab, a humanized monoclonal antibody directed atthe extracellular domain of HER2, is active in patientswith HER2-overexpressing metastatic breast cancer,reducing relapse-free survival and improving overallsurvival (Baselga et al., 1996; Slamon et al., 2001; Martyet al., 2005). Recently, trastuzumab has also been foundto be efficacious in the adjuvant setting (Piccart-Gebhartet al., 2005; Romond et al., 2005; Slamon et al., 2005).The mechanisms of action of trastuzumab are complexand not fully understood. Described mechanismsinclude receptor downregulation (Sliwkowski et al.,1999; Ozcelik et al., 2002; Diermeier et al., 2005; Yusteet al., 2005), cell cycle arrest (Kim et al., 2003),inhibition of angiogenesis (Izumi et al., 2002) andinduction of antibody-dependent cell-mediated cytotoxi-city (ADCC) (Clynes et al., 2000).

Lapatinib, a dual tyrosine kinase inhibitor (TKI) thattargets both EGFR and HER2 (Wood et al., 2004;Baselga, 2006), inhibits the growth of HER2-over-expressing breast cancer cells in culture and in tumorxenografts (Chu et al., 2005; Konecny et al., 2006). Inthe clinic, lapatinib is active and improves time todisease progression in patients with advanced diseasewho have progressed to trastuzumab (Geyer et al.,2006).

Although trastuzumab and lapatinib provide con-siderable clinical benefit, a large fraction of HER2-positive tumors display primary resistance to theseagents. Even initially sensitive tumors will invariably

Received 14 April 2008; revised 21 October 2008; accepted 5 November2008; published online 8 December 2008

Correspondence: Dr J Baselga, Medical Oncology Department, Valld’Hebron University Hospital, Paseo Vall d’Hebron 119-129,Barcelona 08035, Spain.E-mail: jbaselga@vhebron.net

Oncogene (2009) 28, 803–814& 2009 Macmillan Publishers Limited All rights reserved 0950-9232/09 $32.00

www.nature.com/onc

develop acquired resistance in patients with advanceddisease. Therefore, there is a need to develop newstrategies to decrease primary resistance and to delay theappearance of acquired secondary resistance. One suchapproach would be to give the two agents in combina-tion. In preclinical models, the combination is superiorto single drug treatment and enhanced apoptosishas been proposed as a mechanism (Xia et al., 2005;Konecny et al., 2006). In the clinic, a phase IIIstudy comparing the efficacy of lapatinib versus thecombination of lapatinib and trastuzumab inpatients with advanced trastuzumab-resistant HER2-positive breast cancer has shown improved clinicaloutcome with the combination (O’Shaughnessy et al.,2008). In addition, the combined administration oflapatinib and trastuzumab is being studied in a largeadjuvant study.

Taking into consideration the promising activity ofthe combined treatment with trastuzumab andlapatinib, we aimed at exploring further the potentialdifferences between the mechanisms of action oflapatinib and trastuzumab and to search forpotential explanations for the enhanced activity of thecombination.

Results

Lapatinib induces accumulation of HER2 receptors at thecell surfaceLapatinib treatment of the HER2-overexpressing breastcancer cells SKBR-3 and MCF-7HER2 resulted, asexpected, in inhibition of HER2 phosphorylation (datanot shown) and inhibition of mitogen-activated proteinkinase (MAPK) phosphorylation, a readout of lapatinibinhibition of HER2 signaling ((Scaltriti et al., 2007) andFigure 1a). In terms of total levels of HER2, lapatinibresulted in an accumulation of HER2 when comparedwith untreated cells (Figure 1a). In a time courseexperiment we isolated cell surface membrane proteinsby whole cell biotin labeling and showed that accumula-tion of HER2 observed under lapatinib treatmentoccurred at the plasma membrane, detectable alreadyafter 12 h of treatment (Figure 1b).

On the contrary, as reported earlier (Cuello et al.,2001; Valabrega et al., 2005; Henson et al., 2006; Tsenget al., 2006; Scaltriti et al., 2007), trastuzumab aloneresulted in overall downregulation of HER2. Asfor the combined treatment with lapatinib and trastu-zumab, the net result was an accumulation of receptor atthe cell surface of a similar magnitude to that oflapatinib alone at each time point for MCF-7HER2cells and starting at 36 h of treatment for SKBR-3 cells.To avoid massive cell death, SKBR-3 cells weretreated with lower concentrations of lapatinib andtrastuzumab compared with MCF-7HER2 cells.This likely explains the less marked effects in term ofHER2 downregulation induced by trastuzumab orHER2 accumulation induced by lapatinib observed inthese cells.

Stabilization of HER2 and HER2 dimers in presence oflapatinibTyrosine receptor endocytosis and degradation isregulated by post-translational modifications such asreceptor phosphorylation and ubiquitination (Marmorand Yarden, 2004). To evaluate the potential role ofreceptor ubiquitination in lapatinib-induced HER2accumulation, we transiently expressed hemagglutinin(HA)-tagged ubiquitin in MCF-7HER2 cells andanalysed HER2 ubiquitination in presence of lapatinib,trastuzumab or the combination of both. In cells treatedwith lapatinib alone or in combination with trastuzu-mab the levels of ubiquitinated receptor were barelydetectable (Figure 2a). To determine the turnover rate ofHER2 in control cells and in cells treated with eitheragent alone or the combination, we performed a timecourse experiment where we metabolically labeledMCF-7HER2 cells with 35S-methionine for 1 h (pulse)and chased the samples at different time points. Cellstreated with lapatinib alone or in combination withtrastuzumab showed marked HER2 stability (reducedreceptor degradation) compared with untreated cells orcells receiving only trastuzumab, with persistence ofhigh levels of HER2 receptor up to 48 h (Figure 2b). Onthe other hand, consistent with previously reported data(Klapper et al., 2000), trastuzumab treatment markedlyincreased HER2 ubiquitination (Figure 2a) and degra-dation (Figure 2b) compared with untreated cells.In addition to the effects on receptor expression,ubiquitination and degradation, we also wanted tostudy the consequences of lapatinib treatment on thedimerization status of HER2.

In a series of immunoprecipitation experiments, wefound that lapatinib enhanced the formation of inactiveHER2-containing homodimers and HER2-EGFRand HER2-HER3 heterodimers in both SKBR-3 andMCF-7HER2 cells (Figure 3a and Supplementary Figure 1).The stabilization of HER2-containing dimers wasconfirmed by cross-linking experiments (SupplementaryFigure 1). Quantification of total levels of HER2 and theratios phospho-tyrosine (pTyr)/HER2 is provided inFigure 3b. Compared with untreated cells, trastuzumabalone resulted in HER2 downregulation and increasedp-Tyr/HER2 ratios whereas lapatinib, alone or incombination with trastuzumab, caused accumulationof the receptor with decreased p-Tyr/HER2 ratios.Consistent with the results shown in Figure 1, theseeffects were more marked in MCF-7HER2 cells. Asabove, all the experiments were repeated three times.

Modeling of lapatinib-binding affinity to HER receptorsLapatinib competes with ATP for binding to the kinasedomain of both HER2 and EGFR. Given the biochem-ical data on lapatinib-dependent HER2 dimer stabiliza-tion, we opted for a structural modeling approach tomeasure the energy gain (degree of stabilization) ofHER2 dimers associated with lapatinib versus HER2dimers associated with ATP. We computed andcompared the affinities of both lapatinib and ATP forthe monomeric and dimeric forms of the kinase domainsof the members of EGFR, HER2 and HER3 (HER3

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

804

Oncogene

only binds ATP). Although structural data is onlyavailable for the kinase domain of EGFR, the closesimilarity in sequences of the other family members (thesequences of HER2 and HER3 catalytic domains are77.7 and 56.7% identical respectively, to EGFR)enabled us to construct reliable structural models forthe other members based on homology. The manner inwhich the domains are thought to dimerize, leading toactivation, (asymmetric dimerization (Zhang et al.,2006)) is shown in Figure 4a. In agreement withexperimental observations (Rusnak et al., 2001), ourcalculations show that: (a) lapatinib has a higher affinity

for HER2 monomers than it does for EGFR monomers(Figure 4b), (b) lapatinib has higher affinity than ATPfor HER2 monomers (Figure 4b) (c) the HER2 dimers(specifically HER2 homodimers and heterodimers withEGFR and HER3) are more stable in the presence oflapatinib (Figure 4c).

Effects of lapatinib and trastuzumab on BT474 xenograftsTo expand our results in vivo, we evaluated tumorgrowth inhibition and HER2 expression in xenograftsderived from BT474 cells in response to lapatinib,

MCF-7HER2

SKBR-3

SKBR-3

MCF-7HER2

C

HER2

HER2

TfR

p-MAPKs

MAPKs

HER2

p-MAPKs

MAPKs

T L T+L C T L T+L

HE

R2/

TfR

(Arb

itra

ry U

nit

s)

3.53

2.52

1.51

0.50

HE

R2/

TfR

(Arb

itra

ry U

nit

s)

3.53

2.52

1.51

0.50

HE

R2/

TfR

(Arb

itra

ry U

nit

s)

3.53

2.52

1.51

0.50

HE

R2/

TfR

(Arb

itra

ry U

nit

s)

3.53

2.52

1.51

0.50

HE

R2/

MA

PK

s(A

rbit

rary

Un

its)

1.81.61.41.2

10.80.60.40.2

0C T L T+L

HE

R2/

MA

PK

s(A

rbit

rary

Un

its)

1.81.61.41.2

10.80.60.40.2

0C T L T+L

C T L T+L

HE

R2/

TfR

(Arb

itra

ry U

nit

s)

3

2.5

2

1.5

1

0.5

0C T L T+L

HE

R2/

TfR

(Arb

itra

ry U

nit

s)

3

2.5

2

1.5

1

0.5

0C T L T+L

HE

R2/

TfR

(Arb

itra

ry U

nit

s)

3

2.5

2

1.5

1

0.5

0C T L T+L

HE

R2/

TfR

(Arb

itra

ry U

nit

s)

3

2.5

2

1.5

1

0.5

0C T L T+L

C T L T+L

C T L T+L C T L T+L C T L T+L

12hC T L T+L

24hC T L T+L

36hC T L T+L

48h

C T L T+L12h

C T L T+L24h

C T L T+L36h

C T L T+L48h

Figure 1 Lapatinib induces HER2 accumulation. (a) Western blot showing HER2, phospho-MAPKs (p-MAPKs) and total MAPKs(loading control) expression in total lysates of both SKBR-3 and MCF-7HER2 cells treated 48 h with trastuzumab (T), lapatinib (L) orthe combination (TþL). Untreated cells served as controls (C). Western blot bands were quantified by Image J (NIH) and HER2levels of treated cells (normalized to MAPKs) from three experiments were expressed as arbitrary units±s.e.m. relative to controls.(b) Western blot to detect membrane-localized HER2 in both SKBR-3 and MCF-7HER2 cells treated for 12, 24, 36 and 48 h withtrastuzumab, lapatinib or the combination. Whole cells were labeled with biotin and membrane bound proteins were pulled down withNeutrAvidin beads. IB, immunoblot. HER2 levels of treated cells (normalized to transferring receptor (TfR)) were quantified andexpressed as described above.

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

805

Oncogene

trastuzumab or the combination. BT474 cells were usedas they are highly tumorigenic and sensitive to bothlapatinib and trastuzumab. Treatments were started onday 13 post injection, when tumors were alreadyestablished. On day 19, we sacrificed two animals pergroup for immunohistochemistry analyses and theexperiment continued on the remaining animals untilday 23.

As expected (Baselga et al., 1998; Konecny et al.,2006), both lapatinib and trastuzumab induced tumorregression of BT474 cell-derived xenografts. All the micereceiving the combination of lapatinib and trastuzumabshowed complete tumor remission after 10 days (day 23)of treatment (Figure 5a). In these animals no tumorrelapse was observed after 8 months from the comple-tion of therapy. Tumors derived from the animalsexcluded at day 19 (6 days of treatment) were excisedand subjected to anti-HER2 immunohistochemistryanalysis. There was an increase in HER2 expression intumors treated with lapatinib alone and a decreaseof HER2 expression in trastuzumab-treated tumors ascompared with controls (Figure 5b). The degree ofdecrease of HER2 expression did not reach its peak atthis point as treatment with trastuzumab for 16 daysresulted in a higher degree of HER2 downregulation(data not shown). In the combination group, the effectsof lapatinib on HER2 accumulation were dominant overthose of trastuzumab (Figure 5b). Quantification of the

HER2 membrane staining is expressed as a mean of10 representative fields for each condition (animalswere obtained from three independent experiments,Figure 5c).

Lapatinib-induced accumulation of inactive HER2 leadsto increased ADCC in vitroEngagement of immune-effector systems is one of themain therapeutic mechanisms of anti-HER antibodies(Clynes et al., 2000; Bleeker et al., 2004; Gennari et al.,2004). Thus, we wanted to test whether the accumula-tion of HER2 induced by lapatinib could increasetrastuzumab-dependent cell cytotoxicity in MCF-7HER2 cells by increasing the number of antibodybinding sites at the cell surface. As shown in Figure 6,trastuzumab-dependent cytotoxicity was significantlyhigher in MCF-7HER2 cells treated with lapatinibcompared with untreated cells. To validate this resultin a different model system, we measured trastuzumab-mediated cytotoxicity in cells expressing low levels ofHER2 (MCF-7IRES) and in cells where the effectof lapatinib on HER2 accumulation was mimicked bystable expression of medium levels of kinase inactivereceptor (MCF-7HER2KD). The cells with higherHER2 expression showed significantly higher trastuzu-mab-mediated cytolysis (data not shown). The resultsare expressed as mean of three independent experiments

C CT L

12h

HE

R2

(Arb

itra

ry U

nit

s)

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0C T L T+L

HE

R2

(Arb

itra

ry U

nit

s)

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0C T L T+L

HE

R2

(Arb

itra

ry U

nit

s)

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0C T L T+L

HE

R2

(Arb

itra

ry U

nit

s)

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0C T L T+L

1hpulse T+L T L

24h 36h 48h

T+L C T L T+L C T L T+L

T+LLTC

IP:HER2

IB:HA

IP:HER2

IB:HER2

2

1.5

1

0.5

0C T L T+L

HA

/HE

R2

(Arb

itar

y U

nit

s)

Figure 2 Effects of lapatinib on HER2 ubiquitination and stabilization. (a) Western blot showing both ubiquitinated (HA) and totalHER2 in MCF-7HER2 cells previously transfected with HA-ubiquitin and treated with trastuzumab (T), lapatinib (L) or thecombination (TþL) for 6 h in presence of 10 mM MG-132 proteasome and calpain inhibitor. Untreated cells served as controls (C). HAlevels of treated cells (normalized to total HER2) were quantified and expressed as arbitrary units±s.e.m. relative to controls.IP, immunoprecipitation; IB, immunoblot. (b) Autoradiography detecting 35S-labeled immunoprecipitated HER2 in MCF-7HER2cells metabolically pulsed for 1 h and chased after 12, 24, 36 and 48 h of treatment with trastuzumab (T), lapatinib (L) or thecombination (TþL). The first lane indicates the amount of labeled HER2 after 1 h of pulse. 35S-labeled HER2 levels of treated cellswere quantified and expressed as arbitrary units±s.e.m. relative to controls. The experiments were repeated three times.

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

806

Oncogene

Discussion

We have shown that lapatinib, a small molecule HER2TKI, prevents HER2 ubiquitination and degradation,which in turn results in a substantial accumulation ofinactive HER2 receptors at the cytoplasmic membrane.This lapatinib-induced accumulation of HER2 was alsoobserved in vivo. These effects remain even in the

presence of trastuzumab that has opposite effects onreceptor ubiquitination and degradation when givenalone. The degree to which HER2 receptors areinternalized and downregulated following treatmentwith anti-HER2 antibodies is still a matter of debate.Although some groups have reported receptor down-regulation (Cuello et al., 2001; Valabrega et al., 2005;Henson et al., 2006; Tseng et al., 2006; Scaltriti et al.,

p-MAPKs

MAPKs

IP:HER2

IP:HER2

IB:HER2

IB:EGFR

IP:HER2

IB:p-Tyr

IP:HER2

C T L T+L

SKBR-3 MCF-7HER2

C T L T+L

Tota

l lys

ates

IB:HER3

C T L T+L C T L T+L

1.6

1.4

1.2

1

0.8

0.6

0.4

0.2

0

C T L T+L

1.6

1.4

1.2

1

0.8

0.6

0.4

0.2

0

1.6

1.4

1.2

1

0.8

0.6

0.4

0.2

0

C T L T+L

3

2.5

2

1.5

1

0.5

0

HE

R2/

MA

PK

s (A

rbit

rary

Un

its)

HE

R2/

MA

PK

s (A

rbit

rary

Un

its)

p-T

yr/H

ER

2 (A

rbit

rary

Un

its)

p-T

yr/H

ER

2 (A

rbit

rary

Un

its)

Figure 3 Lapatinib promotes HER2 dimerization. (a) Western blot detecting phospho-tyrosine (p-Tyr), EGFR, HER2 and HER3 inboth SKBR-3 and MCF-7HER2 cells treated with trastuzumab (T), lapatinib (L) or the combination (TþL) for 48 h and subjected toimmunoprecipitation (IP) with an anti-HER2 antibody. Untreated cells served as controls (C). Total lysates were probed for p-MAPKsand total MAPKs (loading control). (b) Western blot quantification of HER2 (normalized to MAPKs) and the ratios (p-Tyr)/HER2 oftreated cells expressed as arbitrary units±s.e.m. relative to controls. The experiments were repeated three times.

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

807

Oncogene

2007), others have not (Austin et al., 2004; Hommel-gaard et al., 2004; Longva et al., 2005). Interestingly, insome of the models that have not observed trastuzumab-mediated receptor downregulation (Austin et al., 2004;Longva et al., 2005), there is also a lack of trastuzumab-induced HER2 phosphorylation. It is therefore plausiblethat kinase activation is a required step for receptorubiquitination and degradation and that, as a conse-quence, lapatinib and other receptor TKIs preventreceptor downregulation.

Our computational modeling and immunoprecipita-tion experiments showed that lapatinib provides stabi-lity to HER2 dimers and prolongs the half-live of theseinactive, dimerized HER2 receptors. Inactive EGFR/HER dimers also occur after therapy with EGFR TKIs(Anido et al., 2003). The binding of these agents to theATP pocket of the receptor perturbs its three-dimen-sional structure, stabilizing interactions among recep-

tors and promoting the accumulation of inactive EGFRdimers (Arteaga et al., 1997; Gan et al., 2007). Thepresence of high levels of EGFR inactive dimers on thecell surface would also act as a ligand trap, being able tobind (and sequester) the ligands without consequentreceptor phosphorylation. Addition of anti-EGFRantibodies would improve the efficacy of TKIs, as itwould keep the receptor inactive once the TKIsdisassociate. This model provides a possible explanationfor the efficacy of the combination of TKIs with anti-EGFR antibodies, especially in conditions when theligands are present in limiting amounts (Arteaga et al.,1997; Lichtner et al., 2001; Gan et al., 2007). Besidesthe enhanced receptor stability, it is possible that theincreased receptor number could also play a role in theobserved enhanced co-immunoprecipitation of EGFRor HER3 with HER2. Taken together, inhibition ofphosphorylation and ubiquitination followed by stabi-lization of inactive HER dimers and the resultingincrease in receptor number may be a general modusoperandi of small molecule TKIs targeting the HERfamily.

In our studies we have not analysed the potentialmechanism responsible for HER2 ubiquination anddegradation. A key regulator of HER receptor degrada-tion is the E3 ubiquitin ligase c-Cbl (Marmor andYarden, 2004). Although E3 ubiquitin ligase c-Cblshows only a marginal effect in ligand-induced HER2ubiquitination (Wang et al., 1999; Hommelgaard et al.,2004), it does play a role in receptor degradation whenc-Cbl is overexpressed (Li et al., 2007) or recruitedfollowing treatment with anti-HER2 antibodies (Klap-per et al., 2000; Wolpoe et al., 2003). However, wecannot rule out that the internalization of HER2following trastuzumab treatment could be mediated bya kinase-dependent activation of other ubiquitin ligases.

We have identified an alternative potential mechanismfor the enhanced effect of combined therapy with a TKIand an anti-ErbB antibody. The accumulation ofinactive HER2 receptor at the cell surface may lead toenhanced or prolonged trastuzumab binding/activity,which in turn could explain the observed increase intrastuzumab-mediated ADCC. ADCC is dependent onboth antibody affinity and expression levels of the targetreceptor; target tumor cells with higher antigen expres-sion are more susceptible to antibody therapy due toenhanced immune effects (Mimura et al., 2005; Tanget al., 2007). In our experiments, this was found to be thecase as there was a good correlation between HER2levels and ADCC, both in HER2-overexpressing cellstreated with lapatinib and in cells transfected with akinase dead HER receptor (MCF7-HER2KD). We arenow studying the role of ADCC in vivo where thetherapy with the combination resulted in a remarkableand rapid complete regression of well-establishedxenografts in all treated animals.

In the clinic, there is also growing evidence thattrastuzumab’s antitumor activity may be partiallymediated by ADCC. For example, in a pilot presurgicaltrastuzumab study, patients who achieved either apartial or a complete response to trastuzumab were

Figure 4 Effect of lapatinib binding on HER2 dimer stabilization.(a) Example of HER2 dimerization modeling. Asymmetric mode ofdimerization of the HER family kinase domains structurallymodeled in this study. The dimer between HER2 (in red andcomplexed with lapatinib) and HER3 (in magenta and complexedwith ATP and associated magnesium and water molecules) isshown as an example. (b) Computed energy differences between thebinding of lapatinib and of ATP respectively to HER monomers.Values reflect the degree of stabilization of HER monomers boundto lapatinib (left column) versus HER monomers bound to ATP(right column). LAP, lapatinib; ATP, Adenosine 50-triphosphate;i, inactive conformation; a, active conformation. (c) Computedenergy differences between the binding of lapatinib and of ATPrespectively to HER2 dimers. Values reflect the degree ofstabilization of HER2 dimers bound to lapatinib (left column)versus HER2 dimers bound to ATP (right column). *Lapatinibdoes not bind HER3. LAP, lapatinib; ATP, Adenosine 50-tri-phosphate; i, inactive conformation; a, active conformation.

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

808

Oncogene

found to have a higher in situ infiltration of leukocytesand a higher capability to mediate in vitro ADCCactivity (Gennari et al., 2004). There is also a suggestionthat patients with certain polymorphisms of theirFcgRIIIA receptors, which are activating antibodyreceptors present on the effector cells responsible fortrastuzumab and other antibody-mediated ADCC, mayhave an enhanced response to trastuzumab (Musolinoet al., 2008). The addition of lapatinib to trastuzumabcould be therefore particularly active in patients withgiven FcgRIIIA genotypes such as FcgRIIIA-158 V/V.

It is also conceivable that prolonged trastuzumabadministration results in a decrease in the total levels ofsurface HER2 in breast tumors in a similar fashion as itoccurs in preclinical models. An interesting studyshowed that in tumors trastuzumab caused a decrease

in HER2 expression while maintaining the levels of geneamplification by FISH, indicating that the phenomenonwas due to true protein downregulation rather thanselective elimination of HER2-positive cells (Milellaet al., 2004). In this regard, it has been recently shownthat increased receptor ubiquitination and downregula-tion plays a role in acquired resistance to antibody-based antireceptor therapy (Lu et al., 2007). Lapatinibcould delay/counteract this occurrence by increasingHER2 expression levels and, as a consequence, preventor delay trastuzumab resistance due to lower HER2expression.

Finally, the concept of combining an antireceptormonoclonal antibody and a receptor-stabilizing TKIcould be expanded to other members of the HERreceptor family. At least three independent groups,

Figure 5 Antitumor activity of lapatinib and trastuzumab on BT474 xenografts. (a) Tumor growth inhibition in response totrastuzumab, lapatinib or the combination of the two agents. Treatments started at day 13. Student’s t-test was used to compare tumorsizes between the groups and data are expressed as mean±s.d. *Po0.05, **Po0.01 versus control; #Po0.05 versus trastuzumab;##Po0.01 versus both lapatinib and trastuzumab. These differences remained statistically significant for the entire duration of theexperiment. The experiment was performed three times with similar results. (b) Representative immunohistochemistry showing HER2expression in tumors xenografts treated as indicated and sacrificed at day 19 (see panel a). (c) Quantification of the median intensity ofthe completely stained cells expressed as mean of 10 representative fields for each condition. Student’s t-test: *Po0.05, **Po0.01versus control; #Po0.05 versus trastuzumab.

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

809

Oncogene

including ours, have shown additional or synergisticantitumor effects using the combination of differentanti-EGFR antibodies with TKIs in targeting EGFR-positive cells (Johns et al., 2003; Huang et al., 2004;Matar et al., 2004; Perera et al., 2005). In furthersupport of this approach we have observed promisingclinical activity of the combination of an EGFR TKIand cetuximab, a monoclonal antibody directed at theextracellular domain of the EGFR (Baselga et al., 2006).

In conclusion, our results provide a new explanationfor the enhanced effects of the combination of lapatiniband trastuzumab. Lapatinib reduces HER2 ubiquina-tion, prevents HER2 degradation, and induces theformation of inactive HER2 dimers at the cell surface,which in turn provides an increase in trastuzumabbinding and a greater trastuzumab-mediated immuneresponse (Figure 7). This is a therapeutically exploitablemechanism of action that deserves further study inpatients.

Materials and methods

Cell lines and treatmentsMCF-7 HER2 (overexpressing HER2), MCF-7HER2KD(KD: Kinase Dead; expressing kinase inactive HER2) andMCF-7IRES (mock transfected) cells were obtained asdescribed earlier (Scaltriti et al., 2007). SKBR-3 (HER2amplified) and MDA-MB-468 (HER2 negative) cellswere obtained from the American Type Culture Collection(Rockville, MD, USA). Cells were maintained in Dulbecco’smodified Eagle’s medium/Ham F12 1:1 (DMEM/F12) supple-mented with 10% fetal bovine serum and 2mM L-glutamine(Life Technologies Inc. Ltd., Paisley, UK) at 37 1C in 5% CO2.MCF-7 HER2, MCF-7HERKD and MCF-7IRES cellswere maintained in the same medium containing 30 mg/mlhygromycin B (Life Technologies Inc.).Trastuzumab (Herceptin; kindly provided by F Hoffmann-

La Roche, Basel, Switzerland) and Cetuximab (Erbitux; kindlyprovided by Merck KGaA, Darmstadt, Germany) were

dissolved in sterile apyrogen water and stored at 4 1C.Lapatinib (Tykerb; kindly provided by GlaxoSmithKline,Research Triangle Park, NJ, USA) was dissolved in dimethylsulfoxide (dimethyl sulfoxide as a stock solution at 10mM) andstored at �20 1C. MCF-7 HER2 cells were treated withtrastuzumab and lapatinib at a final concentration of 100 nMand 1mM in the culture media, respectively. SKBR-3 cells weretreated with trastuzumab and lapatinib at a final concentrationof 20 and 100 nM in the culture media, respectively. Dimethylsulfoxide (equal volume to that of treated cells) was added toculture media of the control cells.

Biotin pull down, protein immunoprecipitation, proteincross-linking and western blotFor biotin pull down assays, cells were grown in 60mm dishesand treated with either trastuzumab, lapatinib or thecombination for the indicated times. Cells were incubatedwith EZ-LINK Sulfo-Biotin (Pierce, Rockford, IL, USA) for2 h at 4 1C with gentle rotation. The reaction was stopped bywashing twice with 25 nM Tris-Hcl (pH 7.5) in PBS (phos-phate-buffered saline) and cells were scraped into ice-cold lysisbuffer (50mmol/l HEPES, pH 7.0, 10% glycerol, 1% TritonX-100, 5mmol/l EDTA (ethylenediaminetetraacetic acid),1mmol/l MgCl2, 25mmol/l NaF, 50 mg/ml leupeptin,50 mg/ml aprotinin, 0.5mmol/l orthovanadate, and 1mmol/lphenylmethylsulfonyl fluoride). Lysates were centrifuged at15 000 g for 20min at 4 1C, and supernatants were removedand assayed for protein concentration using the Dc Proteinassay (Bio-Rad, CA, USA). A volume of 500ml of lysis buffercontaining equal amount of proteins was incubatedwith UltraLink Immobilized NeutrAvidin protein (PierceRockford, IL, USA) 2 h at 4 1C with gentle rotation andwashed three times with lysis buffer before suspension in SDS(sodium dodecyl sulfate)-loading buffer.For immunoprecipitation experiments, cells were grown in

100mm dishes and treated with either trastuzumab, lapatinibor the combination for 48 h. A volume of 500ml of lysis buffercontaining equal amount of proteins was incubated with 10mgtrastuzumab for HER2 precipitation overnight at 4 1C withgentle rotation. Protein A sepharose beads (AmershamBiosciences, Uppsala, Sweden) were added for 2 h and washedthree times with lysis buffer before suspension in SDS-loadingbuffer. For cross-linking experiments, cells were grown in100mm dishes and treated with trastuzumab, lapatinib or thecombination for 48 h. Cells were detached using 10mM EDTAin PBS and gentle scraping, and incubated in 5mM bis(sulfo-succinimidyl) suberate (BS3) for 30min at room temperaturewith gentle rotation. Cross-linking reaction was stopped byincubating cells in 25mM Tris-HCl for 15min at roomtemperature with gentle rotation. Cells were thenprocessed for immunoprecipitation with 10mg trastuzumab asdescribed above.For immunoblots, total lysates, biotin pull down and

immunoprecipitation extracts were resolved by SDS–PAGE(polyacrylamide gel electrophoresis) on either 8% (forphosphotyrosine HER2 and HER3 detection) or 12% (forphospho-MAPKs (p-MAPKs) and total MAPKs detection)acrylamide, and electrophoretically transferred to nitrocellu-lose membranes. For cross-linking experiments, precastgradient 4–15% Tris-HCl gels (READY GEL Bio-Rad, CA,USA) were used. Membranes were hybridized with thefollowing primary antibodies: mouse monoclonal anti-p-Tyr(clone 4G10, cat: 05-321) and mouse monoclonal anti-totalHER3 (clone 2F12, cat: 05-390; Upstate Lake Placid,NY, USA), rabbit polyclonal anti-total EGFR (Abcam,Cambridge, UK), mouse monoclonal anti-total HER2

30

25

20

15

10

5

0

% c

yto

toxi

city

Untreated Lapatinib

*

Figure 6 Trastuzumab-dependent cell-mediated cytotoxicity.ADCC mediated by trastuzumab in MCF-7HER2 cells treated48 h with 1 mM lapatinib compared with untreated cells. Theexperiment was repeated three times. Student’s t-test: *Po0.05.

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

810

Oncogene

(CB11, Biogenex, San Ramon, CA, USA), mouse monoclonalanti-transferrin receptor (Zymed Laboratories, San Francisco,CA, USA), rabbit polyclonal phospho-p44/42 MAPK

(Thr202/Tyr204) and rabbit polyclonal total MAPKs (CellSignaling Technology, Beverly, MA, USA). Anti-p-Tyr, anti-EGFR, anti-HER2 and anti-HER3 antibodies were incubated

Figure 7 Proposed alternative mechanism of action of lapatinib based on HER receptor accumulation. In undisturbed conditions,upon ligand binding, the HER receptors form dimers and are phosphorylated (P) by their kinase domains (K). Once phosphorylated,HER2 dimers initiate signaling and undergo ubiquitination (Ub) and lysosomal degradation. Trastuzumab promotes receptorubiquitination and degradation as well. Lapatinib counteracts receptor phosphorylation, ubiquitination and degradation resulting inHER2 dimer accumulation at the plasma membrane and rendering the cells more susceptible to the immune-mediated action of theanti-HER antibodies (mAbs).

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

811

Oncogene

in Tris-buffered saline-Tween buffer (T-TBS, 50mM Tris-HClpH 7.5, 150mM NaCl, 0.1% Tween-20)/5% non-fat dry milk.Anti-p-MAPKs and anti-total MAPKs were incubated inT-TBS/5% bovine serum albumin. Protein–antibody complexeswere detected by chemiluminescence with the SuperSignalWest Dura Extended Duration Substrate (Pierce, Rockford,IL, USA), and images were captured with a FUJIFILM LAS-3000 camera system. Densitometric analyses for proteinquantification were done using Image J 1.38x software(http://rsbweb.nih.gov/ij/index.html). The experiments wererepeated at least three times.

Ubiquitination assayMCF-7HER2 cells were transfected with HA-ubiquitin vector(gift from Dr Jose Gonzales Castano) using the non-liposomalFuGENE 6 reagent (Roche, Indianapolis, IN, USA) accordingto the manufacturer’s protocol. Briefly, 60mm dishes (at 50%density) were transfected with 4 mg of plasmid and treated,after 24 h, with trastuzumab, lapatinib or the combination for6 h in the presence of 10mM MG-132 proteasome and calpaininhibitor (Sigma, St Louis, MI, USA). A volume of 500 ml oflysis buffer containing equal amount of proteins was incubatedwith trastuzumab for HER2 immunoprecipitation. Sampleswere resolved and electrophoretically transferred to nitrocel-lulose membranes as described above and blotted with anti-HA antibody (anti-HA hybridome, 1:100, Babco, Richmond,CA, USA) overnight at 4 1C.

Metabolic labeling (pulse chase)Dishes of MCF-7HER2 cells (60mm) were preincubated 4 h inserum-free Dulbecco’s modified Eagle’s medium deprived ofMet and Cys and metabolically labeled for 1 h with the samemedium containing 20 mCi/dish of 35S-Translabel (MP Bio-medicals, Irvine, CA, USA). Treatments with trastuzumab,lapatinib or the combination were carried out in 10%serum containing DMEM-F12 medium. After lysis andHER2 immunoprecipitation with trastuzumab, samples wereanalysed by SDS–PAGE and autoradiography.

Tumor xenografts in nude miceMice (Charles Rivers Laboratories, Paris, France) weremaintained and treated as described earlier (Scaltriti et al.,2007). A 17b-estradiol pellet (Innovative Research of America,Sarasota, FL, USA) was inserted subcutaneously to eachmouse 1 day before cell injection. BT474 VH2 cells wereobtained from in vitro explants of BT474-derived xenografts(Baselga et al., 1998). A total of 2� 107 cells were injected intothe right flanks of 48 mice (12 for each experimentalcondition), and treatment began when tumors reached anaverage size of >600mm3 (13 days after injection). Trastuzu-mab (10mg/kg in sterile PBS) or sterile PBS (control) wasgiven intraperitoneally twice weekly. Lapatinib (100mg/kg)was administered daily by oral gavage in 0.5% hydroxypropylmethylcellulose, 0.1% Tween 80. Tumor xenografts weremeasured with calipers three times a week, and tumor volumewas determined using the formula: (length�width2)� (p/6).After 10 days of treatment the animals were anesthetized with1.5% isoflurane–air mixture and killed by cervical dislocation.Results are presented as mean±s.d. The experiments wererepeated three times.

ImmunohistochemistryXenografts samples were prepared as described earlier (Serraet al., 2008). Primary antibody was anti-HER2 (CB11,Biogenex) and secondary antibody was from Amersham. Asa negative control, primary antibody was omitted. Slides were

scanned with ScanScope CS system (Aperio, Vista, CA, USA)and HER2 staining intensity was quantified by PATHIAM-RUO software (BioImagene Inc, San Mateo, CA, USA).

Antibody-dependent cell-mediated cytotoxicity assayAntibody-dependent cell-mediated cytotoxicity was measuredwith the CytoTox 96 non-radioactive cytotoxicity assay(Promega, Madison, WI, USA) according to manufacturer’sinstructions. Briefly, MCF-7HER2, MCF-7IRES and MCF-7HER2KD cells were used as target cells. Peripheral bloodmononuclear cells obtained from a healthy donor were used aseffector cells. In all, 4� 103 MCF-7HER2 cells were seeded intriplicate for each condition in a 96-well plate, treated 48 hwith 1mM lapatinib and, in the presence or absence of 8� 103

viable peripheral blood mononuclear cells, incubated withtrastuzumab (100 nM) for 4 h. MCF-7IRES and MCF-7HER2KD cells were not previously treated with lapatinib.Viability of peripheral blood mononuclear cells was assessedby Guava PCA using Guava ViaCount reagents (GuavaTechnologies, Hayward, CA, USA). The percentage ofcytotoxicity was calculated after correcting for backgroundabsorbance values according to the following formula:

%Cytotoxicity ¼

Experimental�Effector spontaneous�Target spontaneous

Targetmaximum�Target spontaneous�100

Specificity of trastuzumab in causing immune-mediatedcytolysis was ensured performing the same assays withcetuximab, an anti-EGFR therapeutic antibody. In all theconditions, cetuximab-dependent cytotoxicity was lower than5%. MDA-MB-468 cells (HER2 negative) served as negativecontrol for trastuzumab ADCC. Results are presented asmeans±s.d. Each experiment was repeated three times.

Computational protein modelingThe EGFR-ATP complex model was constructed guided bythe structure of cAMP-dependent kinase (PDB code 1ATP(Zheng et al., 1993)) and of EGFR (PDB code 1M14 (Zhanget al., 2006)) using QUANTA (Accelrys, San Diego, CA,USA). The activated EGFR asymmetric and symmetric dimers(Zhang et al., 2006) were generated from the monomer usingcrystallographic symmetry operators. The HER2 and HER3sequences were aligned with that of EGFR and, using thestructure of the monomeric EGFR-ATP complex as atemplate, the structures of HER2 and HER3 were built usingthe MODELLER program (Sali and Blundell, 1993). Homo-and heterodimeric models of EGFR, HER2 and HER3 in theiractive and inactive states were generated by superposition ofthe modeled monomers against the EGFR dimer. Complexeswith lapatinib were constructed based on the EGFR–lapatinibcomplex (PDB code 1XKK (Wood et al., 2004)). All modelswere optimized using CHARMM (Brooks et al., 1983) andminimized until the gradient of potential energy was smallerthan 10�2 kcal/mol/A.

Statistical AnalysisFor in vitro assays and nude mice experiments, comparisonsbetween groups were made using a two-tailed Student’s t-test.Differences for which P was less than 0.05 were consideredstatistically significant.

Acknowledgements

This work was supported in full by a grant of the BreastCancer Research Foundation.

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

812

Oncogene

References

Anido J, Matar P, Albanell J, Guzman M, Rojo F, Arribas J et al.(2003). ZD1839, a specific epidermal growth factor receptor (EGFR)tyrosine kinase inhibitor, induces the formation of inactive EGFR/HER2 and EGFR/HER3 heterodimers and prevents heregulinsignaling in HER2-overexpressing breast cancer cells. Clin Cancer

Res 9: 1274–1283.Arteaga CL, Ramsey TT, Shawver LK, Guyer CA. (1997). Unligandedepidermal growth factor receptor dimerization induced by directinteraction of quinazolines with the ATP binding site. J Biol Chem

272: 23247–23254.Austin CD, De Maziere AM, Pisacane PI, van Dijk SM, Eigenbrot C,Sliwkowski MX et al. (2004). Endocytosis and sorting of ErbB2 andthe site of action of cancer therapeutics trastuzumab and geldana-mycin. Mol Biol Cell 15: 5268–5282.

Baselga J. (2006). Targeting tyrosine kinases in cancer: the secondwave. Science 312: 1175–1178.

Baselga J, Norton L, Albanell J, Kim YM, Mendelsohn J. (1998).Recombinant humanized anti-HER2 antibody (Herceptin) enhancesthe antitumor activity of paclitaxel and doxorubicin against HER2/neu overexpressing human breast cancer xenografts. Cancer Res 58:2825–2831.

Baselga J, Schoffski F, Rojo F, Dumez H, Ramos FL, Macarulla Tet al. (2006). A phase I pharmacokinetic (PK) and molecularpharmacodynamic (PD) study of the combination of two anti-EGFR therapies, the monoclonal antibody (MAb) cetuximab (C)and the tyrosine kinase inhibitor (TKI) gefitinib (G), in patients (pts)with advanced colorectal (CRC), head and neck (HNC) and non-small cell lung cancer (NSCLC). Journal of Clinical Oncology, 2006ASCO Annual Meeting Proceedings Part I. Vol 24, No. 18S (June 20Supplement), 2006: 3006.

Baselga J, Tripathy D, Mendelsohn J, Baughman S, Benz CC, DantisL et al. (1996). Phase II study of weekly intravenous recombinanthumanized anti-p185HER2 monoclonal antibody in patients withHER2/neu-overexpressing metastatic breast cancer. J Clin Oncol 14:737–744.

Bleeker WK, Lammerts van Bueren JJ, van Ojik HH, Gerritsen AF,Pluyter M, Houtkamp M et al. (2004). Dual mode of action of ahuman anti-epidermal growth factor receptor monoclonal antibodyfor cancer therapy. J Immunol 173: 4699–4707.

Brooks BR, Bruccoleri RE, Olafson BD, States DJ, Swaminathan S,Karplus M. (1983). CHARMM: a program for macromolecularenergy, minimization, and dynamics calculations. J Comp Chem 4:187–217.

Citri A, Yarden Y. (2006). EGF-ERBB signalling: towards the systemslevel. Nat Rev Mol Cell Biol 7: 505–516.

Clynes RA, Towers TL, Presta LG, Ravetch JV. (2000). Inhibitory Fcreceptors modulate in vivo cytotoxicity against tumor targets. Nat

Med 6: 443–446.Cuello M, Ettenberg SA, Clark AS, Keane MM, Posner RH,Nau MM et al. (2001). Down-regulation of the erbB-2 receptor bytrastuzumab (herceptin) enhances tumor necrosis factor-related apoptosis-inducing ligand-mediated apoptosis in breast andovarian cancer cell lines that overexpress erbB-2. Cancer Res 61:4892–4900.

Chu I, Blackwell K, Chen S, Slingerland J. (2005). The dual ErbB1/ErbB2 inhibitor, lapatinib (GW572016), cooperates with tamoxifento inhibit both cell proliferation- and estrogen-dependent geneexpression in antiestrogen-resistant breast cancer. Cancer Res 65:18–25.

Diermeier S, Horvath G, Knuechel-Clarke R, Hofstaedter F, SzollosiJ, Brockhoff G. (2005). Epidermal growth factor receptor coexpres-sion modulates susceptibility to Herceptin in HER2/neu over-expressing breast cancer cells via specific erbB-receptor interactionand activation. Exp Cell Res 304: 604–619.

Gan HK, Walker F, Burgess AW, Rigopoulos A, Scott AM,Johns TG. (2007). The epidermal growth factor receptor (EGFR)tyrosine kinase inhibitor AG1478 increases the formation ofinactive untethered EGFR dimers. Implications for combination

therapy with monoclonal antibody 806. J Biol Chem 282:2840–2850.

Gennari R, Menard S, Fagnoni F, Ponchio L, Scelsi M, Tagliabue Eet al. (2004). Pilot study of the mechanism of action of preoperativetrastuzumab in patients with primary operable breast tumorsoverexpressing HER2. Clin Cancer Res 10: 5650–5655.

Geyer CE, Forster J, Lindquist D, Chan S, Romieu CG, Pienkowski Tet al. (2006). Lapatinib plus capecitabine for HER2-positiveadvanced breast cancer. N Engl J Med 355: 2733–2743.

Henson ES, Hu X, Gibson SB. (2006). Herceptin sensitizes ErbB2-overexpressing cells to apoptosis by reducing antiapoptotic Mcl-1expression. Clin Cancer Res 12: 845–853.

Hommelgaard AM, Lerdrup M, van Deurs B. (2004). Association withmembrane protrusions makes ErbB2 an internalization-resistantreceptor. Mol Biol Cell 15: 1557–1567.

Huang S, Armstrong EA, Benavente S, Chinnaiyan P, Harari PM.(2004). Dual-agent molecular targeting of the epidermal growthfactor receptor (EGFR): combining anti-EGFR antibody withtyrosine kinase inhibitor. Cancer Res 64: 5355–5362.

Izumi Y, Xu L, di Tomaso E, Fukumura D, Jain RK. (2002). Tumourbiology: herceptin acts as an anti-angiogenic cocktail. Nature 416:279–280.

Johns TG, Luwor RB, Murone C, Walker F, Weinstock J, Vitali AAet al. (2003). Antitumor efficacy of cytotoxic drugs and themonoclonal antibody 806 is enhanced by the EGF receptor inhibitorAG1478. Proc Natl Acad Sci USA 100: 15871–15876.

Kim JY, Sun Q, Oglesbee M, Yoon SO. (2003). The role of ErbB2signaling in the onset of terminal differentiation of oligodendrocytesin vivo. J Neurosci 23: 5561–5571.

Klapper LN, Waterman H, Sela M, Yarden Y. (2000).Tumor-inhibitory antibodies to HER-2/ErbB-2 may act by recruit-ing c-Cbl and enhancing ubiquitination of HER-2. Cancer Res 60:3384–3388.

Konecny GE, Pegram MD, Venkatesan N, Finn R, Yang G, RahmehM et al. (2006). Activity of the dual kinase inhibitor lapatinib(GW572016) against HER-2-overexpressing and trastuzumab-trea-ted breast cancer cells. Cancer Res 66: 1630–1639.

Li X, Shen L, Zhang J, Su J, Liu X, Han H et al. (2007). Degradationof HER2 by Cbl-based chimeric ubiquitin ligases. Cancer Res 67:8716–8724.

Lichtner RB, Menrad A, Sommer A, Klar U, Schneider MR. (2001).Signaling-inactive epidermal growth factor receptor/ligand com-plexes in intact carcinoma cells by quinazoline tyrosine kinaseinhibitors. Cancer Res 61: 5790–5795.

Longva KE, Pedersen NM, Haslekas C, Stang E, Madshus IH. (2005).Herceptin-induced inhibition of ErbB2 signaling involves reducedphosphorylation of Akt but not endocytic down-regulation ofErbB2. Int J Cancer 116: 359–367.

Lu Y, Li X, Liang K, Luwor R, Siddik ZH, Mills GB et al. (2007).Epidermal growth factor receptor (EGFR) ubiquitination as amechanism of acquired resistance escaping treatment by theanti-EGFR monoclonal antibody cetuximab. Cancer Res 67:8240–8247.

Marmor MD, Yarden Y. (2004). Role of protein ubiquitylation inregulating endocytosis of receptor tyrosine kinases. Oncogene 23:2057–2070.

Marty M, Cognetti F, Maraninchi D, Snyder R, Mauriac L, Tubiana-Hulin M et al. (2005). Randomized phase II trial of the efficacy andsafety of trastuzumab combined with docetaxel in patients withhuman epidermal growth factor receptor 2-positive metastatic breastcancer administered as first-line treatment: the M77001 study group.J Clin Oncol 23: 4265–4274.

Matar P, Rojo F, Cassia R, Moreno-Bueno G, Di Cosimo S,Tabernero J et al. (2004). Combined epidermal growth factorreceptor targeting with the tyrosine kinase inhibitor gefitinib(ZD1839) and the monoclonal antibody cetuximab (IMC-C225):superiority over single-agent receptor targeting. Clin Cancer Res 10:6487–6501.

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

813

Oncogene

Milella M, Trisciuoglio D, Bruno T, Ciuffreda L, Mottolese M,Cianciulli A et al. (2004). Trastuzumab down-regulates Bcl-2expression and potentiates apoptosis induction by Bcl-2/Bcl-XLbispecific antisense oligonucleotides in HER-2 gene—amplifiedbreast cancer cells. Clin Cancer Res 10: 7747–7756.

Mimura K, Kono K, Hanawa M, Kanzaki M, Nakao A, Ooi A et al.(2005). Trastuzumab-mediated antibody-dependent cellular cyto-toxicity against esophageal squamous cell carcinoma. Clin Cancer

Res 11: 4898–4904.Musolino A, Naldi N, Bortesi B, Pezzuolo D, Capelletti M, Missale Get al. (2008). Immunoglobulin G fragment C receptor polymorph-isms and clinical efficacy of trastuzumab-based therapy in patientswith HER-2/neu-positive metastatic breast cancer. J Clin Oncol 26:1789–1796.

O’Shaughnessy J, Blackwell KL, Burstein H, Storniolo AM, Sledge G,Baselga J et al. (2008). A randomized study of lapatinib alone or incombination with trastuzumab in heavily pretreated HER2+metastatic breast cancer progressing on trastuzumab therapy.J Clin Oncol 26(Suppl): abstract 1015.

Ozcelik C, Erdmann B, Pilz B, Wettschureck N, Britsch S, Hubner Net al. (2002). Conditional mutation of the ErbB2 (HER2) receptor incardiomyocytes leads to dilated cardiomyopathy. Proc Natl Acad

Sci USA 99: 8880–8885.Perera RM, Narita Y, Furnari FB, Gan HK, Murone C, Ahlkvist Met al. (2005). Treatment of human tumor xenografts with mono-clonal antibody 806 in combination with a prototypical epidermalgrowth factor receptor-specific antibody generates enhanced anti-tumor activity. Clin Cancer Res 11: 6390–6399.

Piccart-Gebhart MJ, Procter M, Leyland-Jones B, Goldhirsch A,Untch M, Smith I et al. (2005). Trastuzumab after adjuvantchemotherapy in HER2-positive breast cancer. N Engl J Med 353:1659–1672.

Romond EH, Perez EA, Bryant J, Suman VJ, Geyer Jr CE, DavidsonNE et al. (2005). Trastuzumab plus adjuvant chemotherapyfor operable HER2-positive breast cancer. N Engl J Med 353:1673–1684.

Rusnak DW, Lackey K, Affleck K, Wood ER, Alligood KJ, Rhodes Net al. (2001). The effects of the novel, reversible epidermal growthfactor receptor/ErbB-2 tyrosine kinase inhibitor, GW2016, on thegrowth of human normal and tumor-derived cell lines in vitro and invivo. Mol Cancer Ther 1: 85–94.

Sali A, Blundell TL. (1993). Comparative protein modelling bysatisfaction of spatial restraints. J Mol Biol 234: 779–815.

Scaltriti M, Rojo F, Ocana A, Anido J, Guzman M, Cortes J et al.(2007). Expression of p95HER2, a truncated form of the HER2receptor, and response to anti-HER2 therapies in breast cancer.J Natl Cancer Inst 99: 628–638.

Sergina NV, Rausch M, Wang D, Blair J, Hann B, Shokat KM et al.(2007). Escape from HER-family tyrosine kinase inhibitor therapyby the kinase-inactive HER3. Nature 445: 437–441.

Serra V, Markman B, Scaltriti M, Eichhorn PJ, Valero V, Guzman Met al. (2008). NVP-BEZ235, a dual PI3K/mTOR inhibitor, preventsPI3K signaling and inhibits the growth of cancer cells withactivating PI3K mutations. Cancer Res 68: 8022–8030.

Slamon D, Eiermann W, Robert N, Pienkowski T, Martin M,Pawlicki M et al. (2005). Phase III randomized trial comparingdoxorubicin and cyclophosphamide followed by docetaxel (ACT)with doxorubicin and cyclophosphamide followed by docetaxeland trastuzumab (ACTH) with docetaxel, carboplatin and trastu-zumab (TCH) in HER2 positive early breast cancer patients:BCIRG 006 study. 28th Annual San Antonio Breast CancerSymposium abstract 1. San Antonio, TX, USA (abstract 1).

Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuireWL. (1987). Human breast cancer: correlation of relapse andsurvival with amplification of the HER-2/neu oncogene. Science

235: 177–182.Slamon DJ, Godolphin W, Jones LA, Holt JA, Wong SG, Keith DEet al. (1989). Studies of the HER-2/neu proto-oncogene in humanbreast and ovarian cancer. Science 244: 707–712.

Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, BajamondeA et al. (2001). Use of chemotherapy plus a monoclonal antibodyagainst HER2 for metastatic breast cancer that overexpressesHER2. N Engl J Med 344: 783–792.

Sliwkowski MX, Lofgren JA, Lewis GD, Hotaling TE, Fendly BM,Fox JA. (1999). Nonclinical studies addressing the mechanism ofaction of trastuzumab (Herceptin). Semin Oncol 26: 60–70.

Tang Y, Lou J, Alpaugh RK, Robinson MK, Marks JD, Weiner LM.(2007). Regulation of antibody-dependent cellular cytotoxicity byIgG intrinsic and apparent affinity for target antigen. J Immunol

179: 2815–2823.Tseng PH, Wang YC, Weng SC, Weng JR, Chen CS, Brueggemeier RWet al. (2006). Overcoming trastuzumab resistance in HER2-overexpres-sing breast cancer cells by using a novel celecoxib-derived phosphoinosi-tide-dependent kinase-1 inhibitor. Mol Pharmacol 70: 1534–1541.

Tzahar E, Waterman H, Chen X, Levkowitz G, Karunagaran D, LaviS et al. (1996). A hierarchical network of interreceptor interactionsdetermines signal transduction by Neu differentiationfactor/neuregulin and epidermal growth factor. Mol Cell Biol 16:5276–5287.

Valabrega G, Montemurro F, Sarotto I, Petrelli A, Rubini P, TacchettiC et al. (2005). TGFalpha expression impairs Trastuzumab-inducedHER2 downregulation. Oncogene 24: 3002–3010.

Wang Z, Zhang L, Yeung TK, Chen X. (1999). Endocytosis deficiencyof epidermal growth factor (EGF) receptor-ErbB2 heterodimers inresponse to EGF stimulation. Mol Biol Cell 10: 1621–1636.

Wolpoe ME, Lutz ER, Ercolini AM, Murata S, Ivie SE, Garrett ESet al. (2003). HER-2/neu-specific monoclonal antibodies collaboratewith HER-2/neu-targeted granulocyte macrophage colony-stimulat-ing factor secreting whole cell vaccination to augment CD8+ T celleffector function and tumor-free survival in Her-2/neu-transgenicmice. J Immunol 171: 2161–2169.

Wood ER, Truesdale AT, McDonald OB, Yuan D, Hassell A,Dickerson SH et al. (2004). A unique structure for epidermal growthfactor receptor bound to GW572016 (Lapatinib): relationshipsamong protein conformation, inhibitor off-rate, and receptoractivity in tumor cells. Cancer Res 64: 6652–6659.

Xia W, Gerard CM, Liu L, Baudson NM, Ory TL, Spector NL.(2005). Combining lapatinib (GW572016), a small moleculeinhibitor of ErbB1 and ErbB2 tyrosine kinases, with therapeuticanti-ErbB2 antibodies enhances apoptosis of ErbB2-overexpressingbreast cancer cells. Oncogene 24: 6213–6221.

Yarden Y, Sliwkowski MX. (2001). Untangling the ErbB signallingnetwork. Nat Rev Mol Cell Biol 2: 127–137.

Yuste L, Montero JC, Esparis-Ogando A, Pandiella A. (2005).Activation of ErbB2 by overexpression or by transmembraneneuregulin results in differential signaling and sensitivity toherceptin. Cancer Res 65: 6801–6810.

Zhang X, Gureasko J, Shen K, Cole PA, Kuriyan J. (2006). Anallosteric mechanism for activation of the kinase domain ofepidermal growth factor receptor. Cell 125: 1137–1149.

Zheng JH, Trafny EA, Knighton DR, Xuong NH, Taylor SS, TenEyck LF et al. (1993). 2.2-angstrom refined crystal-structure of thecatalytic subunit of cAMP-dependent protein-kinase complexedwith MnATP and a peptide inhibitor. Acta Crys D 49: 362–365.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Lapatinib-induced stabilization and accumulation of HER2M Scaltriti et al

814

Oncogene