Therapeutics Compensatory Insulin Receptor (IR) Activation ......increased insulin signaling is...

Ther

ComInsufor

ElizabMarylJonat

Abst

Intro

The1R) inlisheda di-dthe grcouplAKT sprotequired(3, 4),

AuthorPharmaPharmPharma

Note: SCancer

CorresBioscieFax: 63

doi: 10

©2010

Mol C2652

D

Published OnlineFirst October 5, 2010; DOI: 10.1158/1535-7163.MCT-10-0318

Molecular
Cancer
apeutics

apeutic Discovery

pensatory Insulin Receptor (IR) Activation on Inhibition oflin-Like Growth Factor-1 Receptor (IGF-1R): Rationale

Ther

Cotargeting IGF-1R and IR in Cancer

eth Buck1, Prafulla C. Gokhale3, Susan Koujak1, Eric Brown3, Alexandra Eyzaguirre1, Nianjun Tao4,
and Rosenfeld-Franklin3, Lorena Lerner4, M. Isabel Chiu4, Robert Wild3, David Epstein2, han A. Pachter2, and Mark R. Miglarese1
ractInsu

the phigeneslogicatumorRTKs.tumorIR(A)IR, bidselectiIR, whsmall-relativnot tophosp

for oncand disr

s' Affiliationceuticals, Faceut ica lceuticals, C

upplementTherapeuti

ponding Ance Park D1-845-5671

.1158/1535-

American A

ancer Ther

ownload

lin-like growth factor-1 receptor (IGF-1R) is a receptor tyrosine kinase (RTK) and critical activator ofosphatidylinositol 3-kinase–AKT pathway. IGF-1R is required for oncogenic transformation and tumor-is. These observations have spurred anticancer drug discovery and development efforts for both bio-l and small-molecule IGF-1R inhibitors. The ability for one RTK to compensate for another to maintaincell viability is emerging as a common resistance mechanism to antitumor agents targeting individualAs IGF-1R is structurally and functionally related to the insulin receptor (IR), we asked whether IR isigenic and whether IR-AKT signaling contributes to resistance to IGF-1R inhibition. Both IGF-1R andare tumorigenic in a mouse mammary tumor model. In human tumor cells coexpressing IGF-1R andirectional cross talk was observed following either knockdown of IR expression or treatment with ave anti–IGF-1R antibody, MAB391. MAB391 treatment resulted in a compensatory increase in phospho-ich was associated with resistance to inhibition of IRS1 and AKT. In contrast, treatment with OSI-906, amolecule dual inhibitor of IGF-1R/IR, resulted in enhanced reduction in phospho-IRS1/phospho-AKTe to MAB391. Insulin or IGF-2 activated the IR-AKT pathway and decreased sensitivity to MAB391 butOSI-906. In tumor cells with an autocrine IGF-2 loop, both OSI-906 and an anti–IGF-2 antibody reducedho-IR/phospho-AKT, whereas MAB391 was ineffective. Finally, OSI-906 showed superior efficacy com-with MAB391 in human tumor xenograft models in which both IGF-1R and IR were phosphorylated.
pared
Collectively, these data indicate that cotargeting IGF-1R and IR may provide superior antitumor efficacycompared with targeting IGF-1R alone. Mol Cancer Ther; 9(10); 2652–64. ©2010 AACR.

(5, 6)mor cexpreetiolocancetor ofradiattors oand mbeen

duction

role of insulin-like growth factor-1 receptor (IGF-tumor cell proliferation and survival is well estab-(1). IGF-1R is a receptor tyrosine kinase (RTK) withimeric α2β2 structure and is activated on bindingowth factor ligands IGF-1 and IGF-2 (2). IGF-1Res to the phosphatidylinositol 3-kinase (PI3K)–ignaling pathway via interactions with the adaptorin insulin receptor substrate (IRS). IGF-1R is re-

ogenic transformation and tumorigenesisuption of IGF-1R activity by either genetic

logicahibitooncolfor IGagaintherapexploIGF

acid idoma(A) anIR(A)a regi

s: 1Translational Research and 2Cancer Biology, OSIarmingdale, New York; 3In Vivo Pharmacology, OSIs , Bou lder , Co lo rado; and 4Bio logy , AVEOambridge, Massachusetts

ary material for this article is available at Molecularcs Online (http://mct.aacrjournals.org/).

uthor: Elizabeth Buck, OSI Pharmaceuticals, Inc., 1rive, Farmingdale, NY 11735. Phone: 631-962-0782;. E-mail: [email protected]

7163.MCT-10-0318

ssociation for Cancer Research.

; 9(10) October 2010

on July 31, 2021. © 2010mct.aacrjournals.org ed from

or pharmacologic (7–9) approaches can reduce tu-ell proliferation and promote apoptosis. Increasedssion of IGF-1R and its ligands is associated withgy, progression, and prognosis for many humanr types (10, 11). IGF-1R signaling is a key contribu-resistance to cytotoxic chemotherapeutics, ionizingion, and certain targeted agents, including inhibi-f epidermal growth factor receptor (EGFR), HER2,ammalian target of rapamycin (12–15). IGF-1R hasintensely pursued as a cancer target, and both bio-l and small-molecule tyrosine kinase domain in-rs (TKI) of IGF-1R are under investigation inogy clinical trials (16–19). Given the important roleF-1R signaling as an adaptive survival mechanismst a diverse array of antitumor agents, combinationies centered on IGF-1R inhibitors are being widelyred.-1R is closely related to the IR, sharing 70% aminodentity overall and 84% identity within the catalyticin (20, 21). IR can exist as either of two isoforms [IRd IR(B)] due to alternative splicing of exon 11 (22).
(short form) is a fetally expressed isoform that lackson within exon 11. IGF-1R and IR αβ monomers
American Association for Cancer Research.

http://mct.aacrjournals.org/

homoentialIGF-2potenof IGFterestIR(B)in mehomesurvivcue min migrowiexplo30). ENIH3(31, 3has becells, wing anpancrratesinsuli(33, 34levelsprognmor tcolon,more,treatman incCom

modetargettion oincreadata sIR. IndrivengrowtIGF-1IGF-1lationassocicellulaloss oinsuliupregIGF-1systembetweincreation oAlt

scribegenicicellulabetwe

soughcan mand wsuperof tumbitioncan inthe IRcell mrangethroutowarspecifincreathermand IRcomplines.levelsinhibactiviassocior IGFcreaseof IGFof theblockhypotas OSmAbsmedia

Mate

IGF-1OSI

MAB3were

Cell lCel

ture Cmentaand pCell vment

PrepaLys

ly de(1:200p42/phospSignadilut(40 ng

Rationale for Dual IGF-1R and IR Targeting in Cancer

www.a

D


dimerize or heterodimerize, and dimers are differ-ly activated by the ligands, insulin, IGF-1, and. Insulin is the canonical ligand for IR and mosttly activates IR homodimers. However, the ability-2 to activate IR is also well established (23–25). In-ingly, affinity of IGF-2 for IR(A) is 5-fold tighter thanhomodimers (22, 26–28). In addition to the role of IRtabolic signaling for tissues that regulate glucoseostasis, IR can also promote cell proliferation andal. Increased IGF-2–mediated IR signaling can res-ouse embryonic development to prevent dwarfismce caused by knockout of the IGF1R gene (24). Ang body of data indicates that tumor cells can alsoit IR to promote proliferation and survival (25, 29,ctopic expression of IR oncogenically transformsT3 fibroblasts and 184B5 mammary epithelial cells2). Signaling through the IR(A) isoform specificallyen shown to mediate mitogenic signaling in tumorhich is especially important for tumor cells harbor-IR(A)–IGF-2 autocrine loop (22, 26–28). Ablation ofeatic islet cells in rodent models reduces the growthof implanted xenograft tumors, suggesting thatn-mediated IR signaling can promote tumor growth). Epidemiologic studies have shown that elevatedof insulin and C-peptide are associated with poorosis and accelerated tumor growth for several tu-ypes, including carcinomas of the breast, prostate,endometrium, liver, and ovary (1, 35, 36). Further-clinical studies of an inhaled form of insulin for theent of type I diabetes were recently halted due toreased risk of developing lung cancer (37).pensatory RTK signaling is emerging as a majorof resistance to antitumor agents that selectivelya single RTK in tumor cells. Resistance to inhibi-f EGFR or HER2 can be mediated by an adaptivese in METor IGF-1R activity (38, 39). There are alsohowing reciprocal cross talk between IGF-1R andmouse embryogenesis, compensatory IR signalingby IGF-2 can fully maintain normal embryonic

h in IGF-1R−/− mice, whereas double knockouts,R−/− IR−/−, are nonviable (24). In osteoblasts, whereR stimulates growth and differentiation, genetic ab-of IGF1R results in increased IR activation that isated with enhanced insulin-driven AKT and extra-r signal-regulated kinase (ERK) signaling (40). Onf IGF-1R function, osteoblasts shift from IGF- ton-mediated growth and differentiation. Therefore,ulated IR signaling can compensate for loss ofR to maintain cellular function in several biologicals. More recent data have indicated that cross talken IR and IGF-1R may also occur in tumor cells, assed insulin signaling is observed on downregula-f IGF-1R (41).hough mitogenic signaling by IR has been de-d in some tumor cell models, the potential tumori-ty of this receptor has not been shown, nor have
r codependence on IGF-1R and IR and cross talken these receptors been extensively studied. We
5mincells g

acrjournals.org

on July 31, 2021. © 2010mct.aacrjournals.org ownloaded from

t to determine whether IR(A) is tumorigenic andediate resistance to selective inhibition of IGF-1Rhether coinhibition of IR and IGF-1R could provideior inhibition of AKT signaling as well as inhibitionor cell proliferation compared with selective inhi-of IGF-1R. We show that both IR(A) and IGF-1Rdependently promote tumorigenesis and driveS-PI3K-AKT pathway in a mouse mammary tumorodel. IGF-1R and IR are coexpressed in a wideof human tumor cell lines, and ablation of signalinggh either receptor using short hairpin RNA (shRNA)d IR or a neutralizing monoclonal antibody (mAb)ically directed against IGF-1R (MAB391) resulted insed phosphorylation of the reciprocal receptor. Fur-ore, OSI-906, a selective dual inhibitor of IGF-1R, more effectively inhibited the IRS1-AKT pathwayared with MAB391 in several human tumor cellIn xenograft tumors with readily detectable basalof phospho–IGF-1R and phospho-IR, dual-receptorition by OSI-906 resulted in enhanced antitumorty compared with MAB391, where treatment wasated with an increase in phospho-IR. Either insulin-2 was able to activate the IR-AKT pathway and de-the sensitivity of tumor cells to selective inhibition-1R by the anti–IGF-1R mAb. In contrast, activationIR-AKT pathway by insulin or IGF-2 was fully

ed by OSI-906. Collectively, these data support thehesis that drugs cotargeting IGF-1R and IR, suchI-906, may provide superior efficacy compared withselective for IGF-1R by preventing IR/IGF-1R–ted compensatory signaling.

rials and Methods

R/IR inhibitors-906 was synthesized as previously described (9).91, IGFBP3, and the IGF-2–neutralizing antibodyfrom R&D Systems.

inesl lines were obtained from the American Type Cul-ollection or other sources, as indicated in Supple-ry Materials and Methods, banked after receipt,assaged for <6 months before use in experiments.iability was assayed at 72 hours after drug treat-using CellTiter-Glo (Promega Corp.).

ration of protein lysates and Western blottingates forWestern blotting were prepared as previous-scribed (42). Antibodies included IGF-1R and IRdilution; Santa Cruz Biotechnology); phospho-

p44, phospho-AKT(S473), phospho-AKT(T308),ho-S6, and phospho-PRAS40 (1:1,000 dilution; Cellling Technology); and phospho-IRS1Y612 (1:1,000ion; Biosource). Where indicated, IGF-1/IGF-2/mL) or insulin (5 or 50 μIU/mL) was added for
utes before lysis. All other lysateswere collected fromrowing under basal (10% FCS) growing conditions.
Mol Cancer Ther; 9(10) October 2010 2653



AnalyproteRTK

teomecordinon thand M

TaqmGen

(Applthe msets uas prethe fothe 32

PrepacompDir

1R–IGHER2anicalindivdishesagated

IR shGEO

ted wSigmacells wpurom

In vivFem

ies. Toin therandoMAB3dailydetermV = (lwas d[C0/Cmal ×time 0time t0. Meriod ffinedDunnP valu

In vivTo

pharmwith e

orallyor i.pTumotrogen(MOX-100150 mEDTAhibitosodiucentrinatanarray.

Resu

TumoassocsensiWe

withinOSI-9dual iroutebeenIGF-1and iagainotherrepresentialCell l1 μmo(Fig. 1by OSKRASTen otionswerecells (servedtumowas sto theing inactiviExp

measueach gtile ofel. Weexprethe hinificacompsensit

Buck et al.

Mol C2654

D


sis of RTK phosphorylation via aome arrayphosphorylation states were determined by Pro-Profiler arrays (R&D Systems) and processed ac-g to the manufacturer's protocol. RTKs includede array are described in Supplementary Materialsethods.

an assayse expression assays for IGF2, IGF1, IGF-1R, and IRied Biosystems) were conducted as described byanufacturer using 50 ng template. Primer/probesed for IR(A) were specific for this receptor isoformviously described (30). Data were normalized tourth quartile expression for a given gene within–cell line panel.

ration of IGF-1R and IR(A) directlementation tumor cell linesect complementation (DC) tumor cell lines for IGF-F-2 and IR(A)–IGF-2 were derived from inducible-driven breast primary cultures (43). Using mech-chopping and enzymatic treatment (collagenase),

idual tumor cells were isolated, plated in culturein RPMI 1640 + 10% fetal bovine serum, and prop-for up to six passages.

RNAtumor cells grown to 20% confluence were trea-

ith shRNA lentiviral particles (TRCN0000000379,) at a multiplicity of infection of 2. After 48 hours,ere transferred to medium containing 10 μg/mLycin for selection.

o antitumor efficacy studiesale nu/nu CD-1 mice were used for xenograft stud-assess antitumor efficacy, cells were implanted s.c.right flank and established to 200 ± 50 mm3 beforemization into treatment groups. OSI-906 and91 were administered as indicated according toor every third day schedules. Tumor volumes wereined twice weekly from caliper measurements by

ength × width2)/2. Tumor growth inhibition (TGI)etermined by %TGI = {1 − [(Tt/T0)/(Ct/C0)]/1 −t]} × 100, where Tt = tumor volume of treated ani-at time t, T0 = tumor volume of treated animal × at, Ct = median tumor volume of control group at, and C0 = median tumor volume of control at timedian %TGI was calculated for the entire dosing pe-or each group. Significant antitumor activity is de-as median %TGI >50%. Rank ANOVA with

ett's comparison was used to compare groups, andes of ≤0.05 were deemed statistically significant.

o pharmacodynamic analysisassess the ability of OSI-906 or MAB391 to inhibit
acodynamic end points in tumor tissue, animalsstablished tumors of 300 ± 50 mm3 size were dosed
19 OSIGF1

ancer Ther; 9(10) October 2010


with OSI-906 dissolved in 25 mmol/L tartaric acid. with MAB391 diluted in PBS at indicated doses.r samples were collected, snap frozen in liquid ni-, and homogenized in a Precellys-24 homogenizerBioLaboratories) with tumor lysis buffer [1% Triton, 10% glycerol, 50 mmol/L HEPES (pH 7.4),mol/L NaCl, 1.5 mmol/L MgCl2, 1 mmol/Lsupplemented with protease and phosphatase in-r cocktails (Sigma), 10 mmol/L NaF, and 1 mmol/Lm orthovanadate]. Homogenates were clarified byfugation (14,000 × g for 5minutes at 4°C), and super-ts were analyzed by Western blot or phospho-RTK

lts

r cells with elevated expression of genesiated with the IGF-1R/IR signaling axis aretive to OSI-906sought to determine if gene expression or mutationsthe IGF-1R/IR axis were predictive of sensitivity to

06, a small-molecule, reversible, ATP-competitive,nhibitor of IGF-1R and IR (Fig. 1A). The syntheticas well as the mechanism of action for OSI-906 havedescribed (44, 45). OSI-906 selectively inhibits bothR (IC50 = 35 nmol/L) and IR (IC50 = 75 nmol/L)s far less potent (<50% inhibition at 1 μmol/L)st a broad panel (n = 116) of additional RTKs andprotein kinases (45). A panel of 32 tumor cell linesenting 10 tumor types was selected based on differ-sensitivity to OSI-906 in cell proliferation assays.ines were categorized as either sensitive (EC50 <l/L) or insensitive (EC50 > 10 μmol/L) to OSI-906B). For sensitive tumor cell lines, growth inhibitionI-906 was dose dependent (Fig. 1C). Mutations inor BRAF did not preclude sensitivity to OSI-906.

f 19 OSI-906–sensitive tumor cells harbored muta-in either KRAS or BRAF, whereas these mutationsless frequent (3 of 13) in OSI-906–insensitive tumorFig. 1B). In contrast, mutations in PIK3CAwere ob-in nearly half (6 of 13) of the OSI-906–insensitive

r cell lines but did not occur in any cell line thatensitive to OSI-906. IGF-1R and IR couple stronglyPI3K-AKT pathway, and therefore, mutations result-constitutive downstream signaling may mitigate thety of IGF-1R/IR RTK inhibitors.ression of IGF1, IGF2, IGF-1R, and IR mRNAs wasred by quantitative reverse transcription-PCR. Forene, expression was normalized to the fourth quar-expression for that gene within the 32–cell line pan-then ranked the cell lines according to collective

ssion of ligands and receptors. Cell lines exhibitingghest expression of genes in the IGF axis were sig-ntly more sensitive to OSI-906 (P = 0.0004, whenaring IGF axis gene expression in sensitive and in-ive cell lines by two-tailed t test; Fig. 1D). Among
I-906–sensitive cell lines, 14 exhibited expression ofor IGF2 mRNAs at levels that fell within the top
Molecular Cancer Therapeutics



quartiInterenearlymRNAderivand 8SKEStile) f(GEOthe paquarti906–scoexpIGF2)IGF1Rand re906–inmodepairsautocOSI-9

BothBot

ing emnotyphowely estpotenobserto OSof IR(mammHER2(46),clineencodbinatiexprepleme(A) D

Figureto OSI-(EC50 <wild-typfive senfor this receptor isoform as previously described (1). Gene expression was normalized to the fourth quartile expression for a given gene within the 32–cellline panel.


www.a

D


le of expression across the entire 32–cell line panel.stingly, expression of IGF1 and IGF2 mRNAs wasmutually exclusive, with elevated autocrine IGF1expression (top quartile) frequent in tumor cells

ed from hematologic malignancies (U266, H929,22) or sarcomatoid tumor types (A673, RDES, and), and elevated IGF2 mRNA expression (top quar-requent in tumor cells of epithelial derivation, HT-29, MDAH-2774, DU4475, and H322). Withinnel, only GEO tumor cells exhibited elevated (tople) expression of both IGF1 and IGF2. In 9 of 19 OSI-ensitive cell lines, we observed high (top quartile)ression of mRNAs encoding ligand (either IGF1 oralong with mRNAs encoding receptor (eitheror IR). In contrast, elevated coexpression of ligandceptor mRNAs was not observed for any of 13 OSI-sensitive cell lines evaluated. These data support al in which elevated coexpression of receptor-ligandin the IGF-1R/IR axis, consistent with tumor cell
rine signaling, may be predictive for response to06.
mediaThese

acrjournals.org


IGF-1R and IR(A) are tumorigenich IGF-1R and IR can provide cell growth cues dur-bryogenesis and can promote a transformed phe-

e in vitro (23, 31). IGF-1R is also tumorigenic in vivo;ver, the tumorigenicity for IR has not yet been firm-ablished. Because data suggest that IR(A) is moretly activated by IGF-2 than IR(B), and because weved expression of IR(A) in tumor cell lines sensitiveI-906, we sought to assess the tumorigenic potentialA) compared with IGF-1R. We used a DC mouseary tumor model driven by an inducible humanoncogene under doxycycline-directed expression

where repression of HER2 expression by doxycy-withdrawal was followed by introduction of genesing either human IGF-1R or human IR(A), in com-on with human IGF-2 (43). Both IR(A) and IGF-1R,ssed in combination with IGF-2, were able to com-nt tumor growth (Fig. 2A, top). The IGF-1R and IRC tumors exhibited strong penetrance (>70%) and

1. Elevated expression of IGF receptor-ligand pairs is observed among tumor cell lines sensitive to OSI-906. A, structure of OSI-906. B, sensitivity906 for a panel of 32 tumor cell lines derived from 10 tumor types, expressed as EC50 values. Cell lines were categorized as either sensitive1 μmol/L) or insensitive (EC50 > 10 μmol/L) to OSI-906. Mutational status for KRAS, BRAF, and PIK3CA is indicated (black = mutant, white =e, gray = unreported), as indicated by Sanger Wellcome database. C, effect of varying concentrations of OSI-906 on cell growth for a panel ofsitive tumor cell lines. D, expression of IGF1, IGF2, IGF1R, IR, and IR(A) mRNA by quantitative PCR. Primer/probe sets used for IR(A) were specific

n tumor-free latency of 7 and 20 days, respectively.data show that IR(A) is tumorigenic, albeit perhaps




slight1R anwheraccomtumormor cwe fosignaimparFig. Scell liphosption aphospa contreatmAKTdepencate t

potenpendeAKT

InhibcompPho

neousleft).and IstreamOSI-9mor cof IGFmediaand pprope

Figureof eithetreatmephosphorylation of IRS1 , AKT , and PRAS40 for IR and IGF-1R DC cell lines. Phosphorylation of IR and IGF-1R was determined by RTK capture array(top), a spho-IR

Buck et al.

Mol C2656

D


ly less potent than IGF-1R. The growth of both IGF-d IR(A) DC tumors was inhibited by OSI-906,e OSI-906 achieved >100% TGI in each modelpanied by 42% (IR(A) DC) and 68% (IGF-1R DC)regression (Fig. 2A, middle and bottom). Using tu-ells derived from the IR(A) and IGF-1R DC tumors,und that IGF-1R or IR(A) could independently driveling through the IRS1-PI3K-AKT pathway andt sensitivity to OSI-906 (Fig. 2B; Supplementary1). Both the IGF-1R–DC and IR(A)-DC tumornes had detectable levels of phospho-IRS1Y612 andho-AKT and were sensitive to OSI-906 in prolifera-ssays (Fig. 2B; Supplementary Fig. S1). Importantly,ho-IRS1Y612 is a docking site for p85-PI3K, acting asduit to the AKT pathway. Accordingly, OSI-906ent reduced both phospho-IRS1Y612 and phospho-in these cell lines (Fig. 2B), and effects were dose

s described in Materials and Methods, or by a dual phospho–IGF-1R/pho

dent (Supplementary Fig. S1). These results indi-hat both IGF-1R and IR(A) have tumorigenic

datesnaling



tial and both receptors have the capacity to inde-ntly drive tumor cell growth through the IRS1-pathway.

ition of IGF-1R is associated with aensatory increase in IR signalingspho–IGF-1R and phospho-IR are often simulta-ly detectable in human tumor cell lines (Fig. 3A,We determined whether coinhibition of IGF-1RR was required for maximal inhibition of down-signaling through IRS1 and AKT. We compared

06 to the selective anti–IGF-1R mAb MAB391 in tu-ell signaling assays measuring the phosphorylation-1R and IR as well as cytoplasmic signaling inter-tes, including phospho-IRS1Y612, phospho-AKT,hospho-ERK. MAB391 exhibits pharmacologicrties similar to many anti–IGF-1R mAb drug candi-

antibody (Epitomix).

2. Both IGF-1R and IR(A) are tumorigenic and activate the IRS1-AKT pathway in a mouse mammary tumor model. A, top, effect of introductionr IGF-1R or IR (+IGF2) on tumor growth, following doxycycline withdrawal, in a mouse mammary tumor model; middle and bottom, effect of OSI-906nt on the growth of IGF-1R or IR DC tumors. QD, daily. B, effect of OSI-906 (3 μmol/L) on the phosphorylation of IR or IGF-1R or downstream

Y612 S473

currently in clinical development by inhibiting sig-from both IGF-1R homodimers and IGF-1R/IR




heterodecreby >5similaonly mnineMAB3phosping aThe

AKTto deSK-Nbut n

the aphospcell linIGF-1phosptrast,phospin theits abexpretion),phosp

Figureenhanclines unpanel orealizedcells (E(3 μmofour tumH322 cMAB39stable I


www.a

D


dimers but not from IR/IR homodimers. OSI-906ased phospho–IGF-1R by >90% and phospho-IR0% in each cell line tested (Fig. 3A). MAB391 wasrly effective at decreasing phospho–IGF-1R butoderately inhibited (50%) phospho-IR in one of

tumor cell lines tested (Colo205). Interestingly,91 treatment resulted in a substantial increase inho-IR in seven of nine cell lines evaluated, support-model of compensatory IGF-1R/IR signaling.ability of IGF-1R inhibitors to block downstream

and ERK signaling is associated with their abilitycrease tumor cell proliferation and survival. In
-AS (neuroblastoma) tumor cells, phospho–IGF-1R tainin
1 for H295R cells. Results shown are typical of three or more independent experiR-KD on the phosphorylation of IGF-1R and IR in GEO tumor cells. The control fo

acrjournals.org


bility of either OSI-906 or MAB391 to decreaseho-AKT levels (Fig. 3B). However, in three of foures with detectable basal phospho-IR and phospho–R (H322, H295R, and A673), OSI-906 decreasedho-AKT or phospho-ERK levels by >80%. In con-MAB391 had a minimal effect on phospho-ERK orho-AKT levels (<10%). This was especially strikingH295R advanced colorectal cancer cell line. Despiteility to promote a 70% to 90% decrease in IGF-1Rssion (presumably by internalization and degrada-MAB391 was still unable to maximally decreaseho-AKT. These data support a role for IR in main-
g downstream signaling when IGF-1R is selectively
ot phospho-IR was detectable and associated with inhibited.

3. The IGF-1R–neutralizing antibody MAB391 confers a compensatory increase in IR phosphorylation, and cotargeting IGF-1R and IR achievesed inhibition of the IRS1-AKT pathway for select tumor cells. A, top left, phosphorylation of IR and IGF-1R for a group of eight human tumor cellder basal growing conditions (10% FCS); right, effect of OSI-906 (3 μmol/L) or MAB391 (3 μg/mL) on phosphorylation of IR and IGF-1R for af nine OSI-906–sensitive tumor cell lines. Data are captured 16 h after dosing and expressed as percentage of basal phosphorylation. Data wereby RTK capture array as described in Materials and Methods. A set of representative array images is shown for A673 Ewing's sarcoma tumor

wS). NSCLC, non–small cell lung cancer; CRC, colorectal cancer; ACC, advanced colorectal cancer. B, effect of 16-h treatment with OSI-906l/L) or MAB391 (3 μg/mL) on phosphorylation of IR or IGF-1R, total IGF-1R expression, and phospho-AKTS437 or phospho-ERK for a panel ofor cell lines (H322, SK-N-AS, H295R, and A673). C, effect of OSI-906 (3 μmol/L) or MAB391 (3 μg/mL) on phospho-IRS1Y612 for H295R, A673, andells. Also shown are phospho-AKTS473, phospho-PRAS40, and total IGF-1R and IR levels under basal conditions or on treatment with OSI-906 or

ments. D, effect of either treatment with MAB391 or generation ofr IR-KD was derived from cells treated with a nonspecific shRNA.




Inhsite, isand OmodeOSI-9IRS1Y

ed aconsiand dhad ncreaseof phbut nphospestingMAB3phospcontrimor cTo

and iftional(KD)or IR(eratiotreatman incing inin phindicacells treceppathw

DualenhanDu

in vivAS. BmRNAexprereadilnot pof bot4A). Iin sig(P < 0(med0.009)or M(Fig.phospservewithTGI oMAB3TGI odrugs

weighreflecwith(Fig. 4effectlatedtreatmcreaseOSI-9(>50%MAB3first 4phospcontroour inresultfore,resulcorrespharmSK-Nand Iwhere

OSI-9Ele

severaat 50μinsuliAKT,cells (pho–Icells tspondinsulsignifminimall coIGFBbut noto thothosethat esurvivactivi

IGF-2Inc

severof imwe ascrinetumorto OSOSI-9IGF-1

Buck et al.

Mol C2658

D


ibition of phospho-IRS1Y612, the p85-PI3K dockingassociated with activity of IGF-1R inhibitors (12),SI-906 inhibited phospho-IRS1Y612 in the DCls. In A673, H322, and H295R tumor cell lines,06, but not MAB391, strongly inhibited phospho-612 (98%; Fig. 3C). In H295R cells, MAB391 promot-70% decrease in total IGF-1R expression levels,stent with its ability to promote internalizationegradation of this receptor; however, MAB391o effect on total levels of IR and resulted in in-d phospho-IR (Fig. 3C). For H295R cells, inhibitionospho-IRS1 and phospho-AKT by OSI-906 (94%),ot MAB391 (6%), was associated with decreasedho-PRAS40 (98%), a direct substrate of AKT. Inter-ly, the induced activation of phospho-IR by91 was not accompanied by further activation ofho-AKT, suggesting that both IGF-1R and IR likelybute to activation of downstream AKTsignals in tu-ells at the level of IRS1.determine if IR was important for cellular growththe cross talk between IGF-1R and IR was bidirec-, we generated GEO cells with stable IR knockdownusing shRNA that would allow KD of either IR(A)B). IR-KD was accompanied by a decrease in prolif-n rate of ∼50% (data not shown), and althoughent of parental GEO cells with MAB391 promotedrease in IR phosphorylation, ablation of IR signal-GEO IR-KD cells was accompanied by an increaseospho–IGF-1R (Fig. 3D). Collectively, these datate that bidirectional cross talk can occur in tumorhat express both IGF-1R and IR, and targeting bothtors results in greater inhibition of the IRS1-AKTay than specifically targeting IGF-1R alone.

inhibition of IR and IGF-1R is associated withced antitumor activity in vivo

al inhibition of IR and IGF-1R was investigatedo in two xenograft tumor models: GEO and SK-N-oth express IGF2mRNA and similar levels of IGF1R. However, GEO cells, but not SK-N-AS cells, also

ss IR(A) mRNA (Fig. 4A, left). SK-N-AS cells havey detectable levels of basal phospho–IGF-1R, buthospho-IR, whereas GEO cells contain high levelsh phospho–IGF-1R and phospho-IR (Figs. 3B andn SK-N-AS tumors, treatment with OSI-906 resultednificant median TGI of 71% over the dosing period.001). MAB391 was also efficacious in this modelian %TGI of 67% over the dosing period; P <. Treatment with a single dose of either OSI-906AB391 resulted in decreased phospho-AKT4A; Supplementary Fig. S2). Similar effects onho-PRAS40, a substrate of AKT, were also ob-d (data not shown). In GEO tumors, treatmentOSI-906 resulted in significant TGI (median %f 80% over the dosing period; P < 0.004), whereas91 was inactive, corresponding to a median %
f only 8% over the dosing period (Fig. 4A). Bothwere well tolerated, with minimal (<10%) body
not IGylatio



t loss. The efficacy of OSI-906 in GEO tumors wasted by decreased phospho-AKT, whereas treatmentMAB391 did not result in decreased phospho-AKTA; data not shown). For GEO tumors, differentials of OSI-906 and MAB391 on phospho-AKT corre-with their effects on phospho-IR (Fig. 4B). Althoughent with either OSI-906 or MAB391 resulted in de-d phospho–IGF-1R (>50%), only treatment with06 resulted in a significant decrease in phospho-IRfor at least 16 hours). In contrast, treatment with91 had no significant effect on phospho-IR for the8 hours after dosing, and by 72 hours after dosing,ho-IR levels increased by >2-fold compared withl tumors (Fig. 4B). These data are consistent withvitro observations, where treatment with MAB391

ed in a compensatory increase in phospho-IR. There-in GEO tumors, cotargeting of IGF-1R and IRted in enhanced inhibition of phospho-AKT,ponding with improved TGI. Taken together, theacodynamic and efficacy studies in the GEO and

-AS tumors indicate that inhibition of both IGF-1RR may be required for optimal efficacy in cancersboth receptors are present and activated.

06 inhibits insulin-driven AKT signalingvated insulin is associated with poor prognosis inl tumor types (1, 35, 36). We confirmed that insulinIU/mL, a level corresponding tomild fasting hyper-nemia, increased both phospho-IR and phospho-but not phospho–IGF-1R, in HT-29 colorectal cancerFig. 5A and B). Only OSI-906 fully inhibited phos-GF-1R, phospho-IR, and phospho-AKT in HT-29reated with either 5 or 50 μIU/mL insulin, corre-ing to normal fasting insulin levels and mild hyper-emic levels, respectively. In contrast, MAB391icantly reduced phospho–IGF-1R in HT-29 but hadal effects on phospho-IR and phospho-AKT undernditions tested (Fig. 5A and B). Treatment withP3, which can neutralize IGF-1 or IGF-2 ligands,t insulin, resulted in effects on phospho-AKTsimilarse observed forMAB391 and far less significant thancaused by OSI-906 (Fig. 5B). These data indicateven mild increases in insulin levels may provideal signals to tumor cells, which may mitigate thety of IGF-1R–selective therapies.

can drive IR-AKT signalingreased expression of IGF-2 has been observed inal tumor types, caused in some instances by lossprinting (47–53). Because IGF-2 can activate IR,ked whether it also signals through AKT in an auto-loop independently of IGF-1R. MDAH-2774 OvCacells use an IGF-2 autocrine loop and are sensitiveI-906 in vitro. We treated MDAH-2774 cells with06 or MAB391 alone or in the presence of insulin,, or IGF-2. Insulin (50 μIU/mL) activated IR, but
F-1R, as reflected by increased receptor phosphor-n (Fig. 6A). Treatment with 40 ng/mL IGF-1 or



IGF-2presuIGF-1creaseIR hethibite

thougall conwerecondi

Figurephosphquantitprimer/IR(A), a(1 mg/msingle-dMAB391 on the phosphorylation states for IR and IGF-1R in vivo for GEO tumors over the dosing period (i.e. 24 h for OSI-906 or 72 h for MAB391).Repres 1 or O


www.a

D


increased IR and IGF-1R phosphorylation. IGF-1mably increased phospho-IR within the context ofR/IR heterodimers, whereas IGF-2 presumably in-d phospho-IR within the context of either IGF-1R/

entative images from the RTK array are shown. Bottom, effects of MAB39

erodimers or IR/IR homodimers. OSI-906 fully in-d IGF-1R and IR phosphorylation in all cases. Al-

Insulphosp

acrjournals.org


h MAB391 also inhibited phospho–IGF-1R underditions, it had varied effects on phospho-IR, whichdependent on the stimulating ligand. Under basaltions, MAB391 activated phospho-IR by ∼2-fold.

SI-906 on tumor phospho-AKTS473 over the dosing period.

4. Xenograft tumors coexpressing phospho–IGF-1R and phospho-IR are sensitive to OSI-906 but not MAB391, whereas tumors expressingo–IGF-1R and not phospho-IR are sensitive to both OSI-906 and MAB391. A, left, expression of IGF1, IGF2, IGF-1R, and IR(A) as determined byative PCR and expression of phospho-IR and phospho–IGF-1R as determined by capture array. Quantitative PCR for IR was conducted usingprobe sets that could detect either IR(A) or IR(B), whereas PCR for IR(A) was conducted using primer/probe sets that would specifically detect onlys previously described (1). Middle, mice bearing SK-N-AS or GEO tumors were treated with either OSI-906 (50 mg/kg daily, orally) or MAB391ouse every third day, i.p.), as indicated, and tumor growth was determined over a 14-d period. Right, time course pharmacodynamic effect ofose OSI-906 or MAB391 on phosphorylation of AKT for GEO and SK-N-AS tumors. B, top, time course pharmacodynamic effect of OSI-906 or

in (50 μIU/mL) promoted a 7-fold increase inho-IR, and this was potentiated to >12-fold when




cells wpromMABIGF-12. BotMAB3AKT (IGF-2data iagent

thosethe lemoralof IGFIGF-1both oTo

evalu

FigureHT-29 tbeforephospho-AKT for HT-29 cells under basal conditions or following stimulation with 5 μIU/mL (30 pmol/L) or 50 μIU/mL (300 pmol/L) insulin. Resultsshown are typical of three or more independent experiments, and a representative experiment along with quantitation is shown.

Buck et al.

Mol C2660

D


ere cotreated with MAB391. Both IGF-1 and IGF-2oted increased phospho-IR; however, although391 completely inhibited phospho-IR driven by, it did not fully inhibit phospho-IR driven by IGF-h ligands promoted downstream AKT signaling.91 fully inhibited IGF-1 stimulation of phospho-Fig. 6B). However, in cells pretreated with MAB391,could partially rescue AKT phosphorylation. These
ndicate that the potential for differential efficacy fors that specifically inhibit IGF-1R, compared with
decredition



that coinhibit IGF-1R and IR, may be affected byvels of various ligands available within the intratu-compartment. Elevated intratumoral concentrations-2 and/or insulin may indicate that cotargeting ofR and IR is required for maximal efficacy becausef these ligands can activate IR homodimers.further validate IGF-2–driven IR-AKT signaling, weated the ability of an IGF-2–neutralizing antibody to

5. Insulin activation of tumor cell IR-AKT signaling is inhibited by OSI-906 but not MAB391. A, effects on phosphorylation of IGF-1R and IR forumor cells treated with either OSI-906 (3 μmol/L) or MAB391 (3 μg/mL) for 16 h followed by stimulation with 50 μIU/mL (300 pmol/L) insulin for 5 mincell lysis. Receptor phosphorylation is expressed as percentage of untreated control cells. B, effect of OSI-906, MAB391, or IGFBP3 on

S473

ase phospho-IR and phospho-AKT. Under basal con-s, MAB391 activated IR in a compensatory manner.




HoweinhibiFurthwas cparedenhanway iinsulitain s

Discu

ThedrivediscovHoweplasti

Figure(3 μmo(40 ng/OSI-90MDAH-signalin


www.a

D


ver, neutralization of IGF-2 achieved near-completetion of IGF-1R and IR phosphorylation (Fig. 6C).ermore, greater inhibition of phospho-PRAS40aused by the IGF-2–neutralizing antibody com-with MAB391. These data indicate that the

ced activity for OSI-906 against the IR-AKT path-s specific, and indicate that IGF-2, in addition to
n, can activate IR signaling in tumor cells to main- the p
2774 cells following 16-h treatment. Results shown are typical of three or moreg through IR that can occur on specific inhibition of IGF-1R.

acrjournals.org


ssion

observation that a range of RTKs can function totumorigenesis has revolutionized anticancer drugery and development efforts in recent decades.ver, tumor cells exhibit a high degree of signalingcity, which can contribute to adaptive survival in
resence of RTK inhibitors, and identifying the
urvival signaling. mechanisms of acquired resistance for these agents is

6. MAB391 inhibits IGF-1–mediated, but not IGF-2– or insulin-mediated, stimulation of phospho-IR and phospho–IGF-1R. A, effect of OSI-906l/L) or MAB391 (3 μg/mL) on phospho-IR and phospho–IGF-1R for control cells or cells treated with insulin (50 μIU/mL, 300 pmol/L), IGF-1mL), or IGF-2 (40 ng/mL) for 5 min before lysis. B, effect of OSI-906 or MAB391 on phospho-AKTS473 in the presence of IGF-1 or IGF-2. C, effect of6 (3 μmol/L), MAB391 (3 μg/mL), or an IGF-2–neutralizing antibody (10 μg/mL) on phosphorylation of IGF-1R or IR (left) or phospho-PRAS40 (right) for

independent experiments. D, cartoon illustrating compensatory




a majoMultisingletalk badaptindivprocittion oEGFRThe

tumor1R ortransffirst dwas foIGF-1studiewhensignifIR(A)tumor1R ormoteMechpathwand Ilines,anothactivaspecifour obselectpensaMABbodieit prodimerprom55). Uthat cbidireincreaIn c

showesultedpathwachievpressiwas ewhichtectab1R anand/osuch atumor1R. Awith Oto chr

tion ining anbeen1R–spincludstudiand Oprofilwith iHy

risk ahypotthroumenttion otargethypertion otreatmphospMAB3IR into 12-candiprovofore, tto anenhanIR,

MAB3IGF-1activasphorheteronalinMABAKT sactivation ogetedwere eCol

getingantituthroutancecreaseTKIs1R anactivalevera

Discl

EmpS. Kou

Buck et al.

Mol C2662

D


r goal toward individualizing their use in the clinic.ple RTKs can be activated simultaneously within acell, and cross talk can exist between them. Cross

etween EGFR and either IGF-1R or METcan provideive survival for tumor cells when EGFR is targetedidually (38, 39). Preclinical data highlighting reci-y for these receptor pairs have spurred the evalua-f combinatorial RTK targeting in the clinic forinhibitors.re is growing support for IR as amitogenic driver forcells, and there are several examples in which IGF-IR can compensate for inhibition of the other in non-ormed cells. Indeed, the activity of IGF-2 on IR wasiscovered in studying mouse development, where itund that IR, activated by IGF-2, can compensate forR disruption to rescue embryonic growth (24). Others have described enhanced signaling by insulinIGF-1R is disrupted in tumor cells (41). Of particularicance is the observation that IGF-2 activation of thefetal isoform can stimulate mitogenic signaling incells (22, 26–28). We showed here that either IGF-IR(A), expressed along with IGF-2 ligand, can pro-tumor growth in a mouse mammary tumor model.anistically, both receptors can activate the IRS1-AKTay. Elevated phosphorylation of both IGF-1R

R has been observed in many human tumor celland we have shown that IGF-1R/IR cross talk iser means exploited by tumor cells to maintaintion of cell survival pathways when IGF-1R isically targeted (Fig. 6D). Of particular relevance isservation that treatment of tumor cell lines with aive anti–IGF-1R mAb, MAB391, promoted a com-tory increase in phospho-IR in select cell lines.391 functions similarly to IGF-1R–specific anti-s that are currently in clinical development becausemotes the downregulation of either IGF-1R homo-s or heterodimers with IR; however, it does notote the downregulation of IR homodimers (7, 54,sing shRNAdirected toward IR,we further showedompensatory signaling between IGF-1R and IR isctional, as ablation of IR was accompanied by anse in the phosphorylation state of IGF-1R.ontrast to observations with the IGF-1R mAb, wed that cotargeting IGF-1R and IR with OSI-906 re-in enhanced inhibition of the IRS1-AKT signalingay. Finally, whereas both OSI-906 and MAB391ed efficacy in a human tumor xenograft model ex-ng only detectable phospho–IGF-1R, only OSI-906fficacious in a human tumor xenograft model inboth phospho-IR and phospho–IGF-1R were de-le. We speculate that in such a setting, both IGF-d IR are required in tumor cells to mediate growthr survival signals, and a dual IGF-1R/IR inhibitors OSI-906may have enhanced efficacy against selects compared with an inhibitor that targets only IGF-lthough hyperglycemia is evident in mice treated
SI-906, at maximum tolerated dose, OSI-906 is able
onically suppress both IGF-1R and IR phosphoryla-

D. EpsPharmapatents



tumor tissue, translating to effects on AKT signal-d TGI (44, 45). In the clinic, hyperinsulinemia hasobserved in select patients treated with either IGF-ecific antibodies or small-molecule IGF-1R/IR TKIsing OSI-906 (56–58). However, in phase I clinicales, hyperglycemia was transient and reversible,SI-906 was well tolerated with an acceptable safetye and plasma pharmacokinetic profile that correlatesnhibition of IGF-1R and IR targets in blood.perinsulinemia has been implicated as an increasednd poor-prognosis factor for certain cancers, and onehesis is that insulin is driving tumor cell survivalgh IR-AKTsignaling.We have determined that treat-with either insulin or IGF-2 could maintain activa-f the AKT pathway when IGF-1R was selectivelyed. Insulin concentrations corresponding to mildinsulinemia promoted an increase in phosphoryla-f IR and AKT, independent of IGF-1R, and insulinent promoted resistance toward inhibition ofho-AKT by MAB391. Under basal conditions,91 promoted a compensatory increase in phospho-tumor cells by ∼2-fold, which was increased furtherfold by addition of insulin. IGF-1R–selective drugdates in clinical development have been shown toke an increase in systemic insulin levels (59); there-he compensatory increase in phospho-IR in responseanti–IGF-1R antibody in tumor cells may be furtherced by increased supplies of endocrine insulin.in addition to IGF-1R, can also be activated by IGF-2.91 inhibited IGF-1– or IGF-2–stimulated phospho–R. However, although MAB391 inhibited IR whented by IGF-1, presumably mediated by transpho-ylation by IGF-1R within the context of IGF-1R/IRdimers, it had little effect on IGF-2–activated IR sig-g. Furthermore, for tumor cells pretreated with391, IGF-2, but not IGF-1, could partially rescueignaling. These data indicate that IGF-2–mediatedtion of IR homodimers may compensate for activa-f the AKT pathway when IGF-1R is individually tar-. Finally, tumor cell lines with IGF-2 autocrine loopsspecially sensitive to OSI-906 compared toMAB391.lectively, these data support the approach of cotar-IGF-1R and IR to deliver enhanced and sustainedmor activity for cancers that rely on signalinggh both of these receptors. Moreover, because resis-to IGF-1R–specific antibodies may emerge via in-d IR signaling, dual targeting of IGF-1R and IR bymay be efficacious following failure of an anti–IGF-tibody. Identifying biomarkers associated with thetion of IGF-1R and IRwill be important for optimallyging emerging therapeutic agents like OSI-906.

osure of Potential Conflicts of Interest

loyment by OSI Pharmaceuticals for: E. Buck, P.C. Gokhale,jak, E. Brown, A. Eyzaguirre, M. Rosenfeld-Franklin, R. Wild,
tein, J.A. Pachter, and M. Miglarese. Employment by AVEOceuticals for: N. Tao, L. Lerner, and M.I. Chiu. Inventorship onfor: E. Buck, D. Epstein, J.A. Pachter, and M. Miglarese.



Ackn

Weassistan

Refe1. Po

Na2. De

cep769

3. PeinsMo

4. SevirufibrNa

5. ChbregrosisCa

6. ReGrsenCa

7. CohaninsCli

8. Hatuminh

9. Ji QselinsinsMo

10. LeRcan

11. Rucel311

12. Buchde83

13. GutyrIGF

14. Nagrohecan

15. O'Rrec200

16. BaOn

17. Yufac20

18. Gurecdir

19. Ro


www.a

D


Thepaymenadvertis

owledgments

ritt, and Jennifer Workman for

this fac

eptor inhibitors in oncology: early clinical trial results and futureections. Oncogene 2009;28:3009–21.don J, DeSantos V, Ferry RJ, Jr., Kurzrock R. Early drug develop-

meles

20. Laits69

21. Wastr20

22. DeisoMe

23. Bagro73

24. LoIGme

25. MostiAc

26. BeIGF

27. Beisobri

28. Fraly rtal

29. Vene

30. KaFuce20

31. Giocetra

32. Friexliga66

33. Hegro6:3

34. HetheinRe

35. MaC-96

36. JenanPro20

37. Kli38. Be

witres10

39. En

acrjournals.org


costs of publication of this article were defrayed in part by thet of page charges. This article must therefore be hereby markedement in accordance with 18 U.S.C. Section 1734 solely to indicate
t.
thank Darla Landfair, Carrie Pirce with the animal studies.
Received 04/09/2010; revised 07/29/2010; accepted 08/08/2010.
rencesllak M. Insulin and insulin-like growth factor signalling in neoplasia.t Rev Cancer 2008;8:915–28.Meyts P, Whittaker J. Structural biology of insulin and IGF1 re-tors: implications for drug design. Nat Rev Drug Discov 2002;1:–83.ruzzi F, Prisco M, Dews M, et al. Multiple signaling pathways of theulin-like growth factor 1 receptor in protection from apoptosis.l Cell Biol 1999;19:7203–15.ll C, Rubini M, Rubin R, Liu JP, Efstratiadis A, Baserga R. Simians 40 large tumor antigen is unable to transform mouse embryonicoblasts lacking type 1 insulin-like growth factor receptor. Proctl Acad Sci U S A 1993;90:11217–21.ernicky CL, Yi L, Tan H, Gan SU, Ilan J. Treatment of humanast cancer cells with antisense RNA to the type I insulin-likewth factor receptor inhibits cell growth, suppresses tumorigene-, alters the metastatic potential, and prolongs survival in vivo.ncer Gene Ther 2000;7:384–95.snicoff M, Coppola D, Sell C, Rubin R, Ferrone S, Baserga R.owth inhibition of human melanoma cells in nude mice by anti-se strategies to the type 1 insulin-like growth factor receptor.ncer Res 1994;54:4848–50.hen BD, Baker DA, Soderstrom C, et al. Combination therapy en-ces the inhibition of tumor growth with the fully human anti-type 1ulin-like growth factor receptor monoclonal antibody CP-751,871.n Cancer Res 2005;11:2063–73.luska P, Carboni JM, Loegering DA, et al. In vitro and in vivo anti-or effects of the dual insulin-like growth factor-I/insulin receptoribitor, BMS-554417. Cancer Res 2006;66:362–71.S, Mulvihill MJ, Rosenfeld-Franklin M, et al. A novel, potent, and

ective insulin-like growth factor-I receptor kinase inhibitor blocksulin-like growth factor-I receptor signaling in vitro and inhibitsulin-like growth factor-I receptor dependent tumor growth in vivo.l Cancer Ther 2007;6:2158–67.oith D, Roberts CT, Jr. The insulin-like growth factor system andcer. Cancer Lett 2003;195:127–37.bin R, Baserga R. Insulin-like growth factor-I receptor. Its role inl proliferation, apoptosis, and tumorigenicity. Lab Invest 1995;73:–31.ck E, Eyzaguirre A, Rosenfeld-Franklin M, et al. Feedback me-anisms promote cooperativity for small molecule inhibitors of epi-rmal and insulin-like growth factor receptors. Cancer Res 2008;68:22–32.ix M, Faber AC, Wang SE, et al. Acquired resistance to EGFRosine kinase inhibitors in cancer cells is mediated by loss of-binding proteins. J Clin Invest 2008;118:2389–92.hta R, Yuan LX, Zhang B, Kobayashi R, Esteva FJ. Insulin-likewth factor-I receptor/human epidermal growth factor receptor 2terodimerization contributes to trastuzumab resistance of breastcer cells. Cancer Res 2005;65:11118–28.eilly KE, Rojo F, She QB, et al. mTOR inhibition induces upstreameptor tyrosine kinase signaling and activates Akt. Cancer Res6;66:1500–8.serga R. Customizing the targeting of IGF-1 receptor. Futurecol 2009;5:43–50.en JS, Macaulay VM. Targeting the type 1 insulin-like growthtor receptor as a treatment for cancer. Expert Opin Ther Targets08;12:589–603.alberto A, Pollak M. Emerging role of insulin-like growth factor

nt of inhibitors of the insulin-like growth factor-I receptor pathway:sons from the first clinical trials. Mol Cancer Ther 2008;7:2575–88.wrence MC, McKern NM, Ward CW. Insulin receptor structure andimplications for the IGF-1 receptor. Curr Opin Struct Biol 2007;17:9–705.rd CW, Garrett TP, McKern NM, et al. The three dimensionalucture of the type I insulin-like growth factor receptor. Mol Pathol01;54:125–32.nley A, Wallace JC, Cosgrove LJ, Forbes BE. The insulin receptorform exon 11- (IR-A) in cancer and other diseases: a review. Hormtab Res 2003;35:778–85.ker J, Liu JP, Robertson EJ, Efstratiadis A. Role of insulin-likewth factors in embryonic and postnatal growth. Cell 1993;75:–82.uvi A, Accili D, Efstratiadis A. Growth-promoting interaction ofF-II with the insulin receptor during mouse embryonic develop-nt. Dev Biol 1997;189:33–48.rrione A, Valentinis B, Xu SQ, et al. Insulin-like growth factor IImulates cell proliferation through the insulin receptor. Proc Natlad Sci U S A 1997;94:3777–82.lfiore A. The role of insulin receptor isoforms and hybrid insulin/-I receptors in human cancer. Curr Pharm Des 2007;13:671–86.lfiore A, Frasca F, Pandini G, Sciacca L, Vigneri R. Insulin receptorforms and insulin receptor/insulin-like growth factor receptor hy-ds in physiology and disease. Endocr Rev 2009;30:586–623.sca F, Pandini G, Scalia P, et al. Insulin receptor isoform A, a new-ecognized, high-affinity insulin-like growth factor II receptor in fe-and cancer cells. Mol Cell Biol 1999;19:3278–88.lla V, Sciacca L, Pandini G, et al. The IGF system in thyroid cancer:w concepts. Mol Pathol 2001;54:121–4.lli KR, Falowo OI, Bale LK, Zschunke MA, Roche PC, Conover CA.nctional insulin receptors on human epithelial ovarian carcinomalls: implications for IGF-II mitogenic signaling. Endocrinology02;143:3259–67.rgino F, Belfiore A, Milazzo G, et al. Overexpression of insulin re-ptors in fibroblast and ovary cells induces a ligand-mediatednsformed phenotype. Mol Endocrinol 1991;5:452–9.ttitta L, Vigneri R, Stampfer MR, Goldfine ID. Insulin receptor over-pression in 184B5 human mammary epithelial cells induces and-dependent transformed phenotype. J Cell Biochem 1995;57:6–9.uson JC, Legros N. Effect of insulin and of alloxan diabetes onwth of the rat mammary carcinoma in vivo. Eur J Cancer 1970;49–51.uson JC, Legros N. Influence of insulin deprivation on growth of7,12-dimethylbenz(a)anthracene-induced mammary carcinoma

rats subjected to alloxan diabetes and food restriction. Cancers 1972;32:226–32.J, Giovannucci E, Pollak M, et al. A prospective study of plasma

peptide and colorectal cancer risk in men. J Natl Cancer Inst 2004;:546–53.ab M, Riboli E, Cleveland RJ, et al. Serum C-peptide, IGFBP-1

d IGFBP-2 and risk of colon and rectal cancers in the Europeanspective Investigation into Cancer and Nutrition. Int J Cancer07;121:368–76.ng J. Inhaled insulin's last gasp? Nat Biotechnol 2008;26:479–80.an J, Brennan C, Shih JY, et al. MET amplification occurs with orhout T790M mutations in EGFR mutant lung tumors with acquired
istance to gefitinib or erlotinib. Proc Natl Acad Sci U S A 2007;4:20932–7.gelman JA, Zejnullahu K, Mitsudomi T, et al. MET amplification



leanal

40. FuDisbla25

41. Zhinscer

42. BuInaerltumTh

43. LerdritheCaDe

44. MuinhRS

45. MuOS1 r

46. Zhdiffde

47. ChtheRe

48. Cupo

49. Fogroge

50. GiproCli

51. NoinsRe

52. Waexdia

53. Baca

54. Wafachu12

55. Poon

56. Cadogrowitab

57. Evstuinh(IR15

58. Lincorecsoab

59. Ha

Buck et al.

Mol C2664

D


ds to gefitinib resistance in lung cancer by activating ERBB3 sig-ing. Science 2007;316:1039–43.lzele K, DiGirolamo DJ, Liu Z, Xu J, Messina JL, Clemens TL.ruption of the insulin-like growth factor type 1 receptor in osteo-sts enhances insulin signaling and action. J Biol Chem 2007;282:649–58.ang H, Pelzer AM, Kiang DT, Yee D. Down-regulation of type Iulin-like growth factor receptor increases sensitivity of breast can-cells to insulin. Cancer Res 2007;67:391–7.

ck E, Eyzaguirre A, Haley JD, Gibson NW, Cagnoni P, Iwata KK.ctivation of Akt by the epidermal growth factor receptor inhibitorotinib is mediated by HER-3 in pancreatic and colorectalor cell lines and contributes to erlotinib sensitivity. Mol Cancer

er 2006;5:2051–9.ner L, Liu Q, Yang J, et al. Generation of in vivo tumor modelsven by insulin-like growth factor receptor IGF1R and their use indevelopment of OSI-906, a selective IGF1R inhibitor. Molecularncer Therapeutics 8 (Meeting Abstract Supplement), A233,cember 10, 2009.lvihill MJ, Buck E. The discovery of OSI-906, a small-moleculeibitor of the insulin-like growth factor-1 and insulin receptors.C Drug Discovery Series 2011, No. 2. In press.lvihill MJ, Cooke A, Rosenfeld-Franklin M, et al. Discovery ofI-906: a selective and orally efficacious dual inhibitor of the IGF-eceptor and insulin receptor. Future Med Chem 2009;1:1153–71.ou Y, Rideout WM, Zi T, et al. Chimeric mouse tumor models revealerences in pathway activation between ERBB family- and KRAS-pendent lung adenocarcinomas. Nat Biotechnol 2009;28:71–8.en CL, Ip SM, Cheng D, Wong LC, Ngan HY. Loss of imprinting ofIGF-II and H19 genes in epithelial ovarian cancer. Clin Cancer

s 2000;6:474–9.i H, Cruz-Correa M, Giardiello FM, et al. Loss of IGF2 imprinting: atential marker of colorectal cancer risk. Science 2003;299:1753–5.
ttner C, Hoeflich A, Wolf E, Weber MM. Role of the insulin-likewth factor system in adrenocortical growth control and carcino-nesis. Horm Metab Res 2004;36:397–405.
ofCP20



ordano TJ, Kuick R, Else T, et al. Molecular classification andgnostication of adrenocortical tumors by transcriptome profiling.n Cancer Res 2009;15:668–76.nomura N, Nishimura K, Miki T, et al. Loss of imprinting of theulin-like growth factor II gene in renal cell carcinoma. Cancers 1997;57:2575–7.ng Z, Ruan YB, Guan Y, Liu SH. Expression of IGF-II in earlyperimental hepatocellular carcinomas and its significance in earlygnosis. World J Gastroenterol 2003;9:267–70.rlaskar FM, Hammer GD. The molecular genetics of adrenocorticalrcinoma. Rev Endocr Metab Disord 2007;8:343–8.ng Y, Hailey J, Williams D, et al. Inhibition of insulin-like growthtor-I receptor (IGF-IR) signaling and tumor cell growth by a fullyman neutralizing anti-IGF-IR antibody. Mol Cancer Ther 2005;4:14–21.llak M. Targeting insulin and insulin-like growth factor signalling incology. Curr Opin Pharmacol 2008;8:384–92.rden CP, Kim ES, Jones RL, et al. Phase 1 study of intermittentsing of OSI-906, a dual tyrosine kinase inhibitor of insulin-likewth factor-1 receptor (IGF-1R) and insulin receptor (IR) in patientsh advanced solid tumors. J Clin Oncol 2010;28:15S (suppl;stract 2530).ans TRJ, Linsdsay CR, Chan E, et al. Phase 1 dose escalationdy of continuous oral dosing of OSI-906, a dual tyrosine kinaseibitor of insulin-like growth factor (IGF-1R) and insulin receptor), in patients with advanced solid tumors. J Clin Oncol 2010;28:S (suppl; abstract 2531).dsay CR, Chan E, Evans TRJ, et al. Phase I dose escalation ofntinuous oral dosing of OSI-906, an insulin-like growth factor-1eptor (IGF-1R) tyrosine kinase inhibitor, in patients with advancedlid tumors [abstract 2559]. J Clin Oncol 2009;27:15S (suppl;stract 2559).luska P, Shaw HM, Batzel GN, et al. Phase I dose escalation studythe anti insulin-like growth factor-I receptor monoclonal antibody
-751,871 in patients with refractory solid tumors. Clin Cancer Res07;13:5834–40.



2010;9:2652-2664. Published OnlineFirst October 5, 2010.Mol Cancer Ther Elizabeth Buck, Prafulla C. Gokhale, Susan Koujak, et al. Cotargeting IGF-1R and IR in CancerInsulin-Like Growth Factor-1 Receptor (IGF-1R): Rationale for Compensatory Insulin Receptor (IR) Activation on Inhibition of

Updated version

10.1158/1535-7163.MCT-10-0318doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2010/10/07/1535-7163.MCT-10-0318.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/9/10/2652.full#ref-list-1

This article cites 57 articles, 26 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/9/10/2652.full#related-urls

This article has been cited by 25 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://mct.aacrjournals.org/content/9/10/2652To request permission to re-use all or part of this article, use this link

on July 31, 2021. © 2010 American Association for Cancer Research. mct.aacrjournals.org Downloaded from


http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-10-0318

http://mct.aacrjournals.org/content/suppl/2010/10/07/1535-7163.MCT-10-0318.DC1

http://mct.aacrjournals.org/content/9/10/2652.full#ref-list-1

http://mct.aacrjournals.org/content/9/10/2652.full#related-urls

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

mailto:[email protected]

http://mct.aacrjournals.org/content/9/10/2652


Therapeutics Compensatory Insulin Receptor (IR) Activation ......increased insulin signaling is...

Documents

Transcript of Therapeutics Compensatory Insulin Receptor (IR) Activation ......increased insulin signaling is...