Tankyrase Inhibition Blocks Wnt/b-Catenin Pathway and ...Daniel Silberschmidt5, Jordi Rodon2,6, Stefania Landolﬁ7, Aleix Prat8, Eloy Espín9, Ramon Charco10, Paolo Nuciforo11, Ana

Cancer Therapy: Preclinical

Tankyrase Inhibition Blocks Wnt/b-CateninPathway and Reverts Resistance to PI3K and AKTInhibitors in the Treatment of Colorectal CancerOriol Arqu�es1, Irene Chicote1, Isabel Puig1, Stephan P. Tenbaum1, Guillem Argil�es2,3,Rodrigo Dienstmann2,3,4, Natalia Fern�andez2,3, Ginevra Carat�u5, Judit Matito5,Daniel Silberschmidt5, Jordi Rodon2,6, Stefania Landolfi7, Aleix Prat8,Eloy Espín9, Ram�on Charco10, Paolo Nuciforo11, Ana Vivancos5,Wenlin Shao12,Josep Tabernero2,3, and H�ector G. Palmer1

Abstract

Purpose: Oncogenic mutations in the KRAS/PI3K/AKT path-way are one of the most frequent alterations in cancer. AlthoughPI3K or AKT inhibitors show promising results in clinical trials,drug resistance frequently emerges. We previously revealedWnt/b-catenin signaling hyperactivation as responsible for suchresistance in colorectal cancer. Here we investigate Wnt-mediatedresistance in patients treated with PI3K or AKT inhibitors inclinical trials and evaluate the efficacy of a new Wnt/tankyraseinhibitor, NVP-TNKS656, to overcome such resistance.

Experimental Design: Colorectal cancer patient-derivedsphere cultures and mouse tumor xenografts were treated withNVP-TNKS656, in combination with PI3K or AKT inhibitors.Weanalyzed progression-free survival of patients treated with differ-ent PI3K/AKT/mTOR inhibitors in correlation with Wnt/b-cate-nin pathway activation, oncogenic mutations, clinicopathologi-cal traits, and gene expression patterns in 40 colorectal cancerbaseline tumors.

Results: Combination with NVP-TNKS656 promoted apopto-sis in PI3K or AKT inhibitor-resistant cells with high nuclearb-catenin content. High FOXO3A activity conferred sensitivity toNVP-TNKS656 treatment. Thirteen of 40 patients presented highnuclear b-catenin content and progressed earlier upon PI3K/AKT/mTOR inhibition. Nuclear b-catenin levels predicted drugresponse, whereas clinicopathologic traits, gene expression pro-files, or frequent mutations (KRAS, TP53, or PIK3CA) did not.

Conclusions: High nuclear b-catenin content independentlypredicts resistance to PI3K and AKT inhibitors. Combined treat-ment with a Wnt/tankyrase inhibitor reduces nuclear b-catenin,reverts such resistance, and represses tumor growth. FOXO3Acontent and activity predicts response to Wnt/b-catenin inhibi-tion and together with b-catenin may be predictive biomarkers ofdrug response providing a rationale to stratify colorectal cancerpatients to be treated with PI3K/AKT/mTOR and Wnt/b-catenininhibitors. Clin Cancer Res; 22(3); 644–56. �2015 AACR.

IntroductionColorectal cancer is a leading cause of death worldwide (1),

mostly because conventional treatments or new target-directeddrugs are ineffective in patients presenting late-stage metastaticdisease (2). It is therefore crucial to unmask the molecularmechanisms responsible for such resistance and to provide newpredictive biomarkers of drug response that could improve theselection of patients sensitive to treatment.

Activating mutations in genes encoding constituents of theKRAS/PI3K/AKT signaling pathway can be considered one of themost frequent cancer-causing genetic alterations in solid tumors,including colorectal cancer (3). Thus, a new generation of drugstargeting PI3K or AKT activity is being tested in numerous clinicaltrials with promising results in some tumor types. Unfortunately,colorectal cancer patients show an enhanced resistance to thesedrugs (4–8).

Active AKT phosphorylates FOXO proteins promoting theirsequestration in the cytoplasm and blocking their capacity toinduce the expression of target genes coding for proteins involvedin cell-cycle arrest and apoptosis (9). Therefore, the efficacy ofPI3K and AKT inhibitors can be mediated in part through nuclear

1Stem Cells and Cancer Laboratory, Vall d'Hebron Institute of Oncol-ogy, Barcelona, Spain. 2Medical Oncology Department,Vall d'HebronUniversity Hospital, Barcelona, Spain. 3Gastrointestinal andEndocrineTumors Group,Vall d'Hebron Institute of Oncology, Barcelona, Spain.4SageBionetworks, FredHutchinsonCancerResearchCentre, Seattle,Washington. 5Cancer Genomics Group, Vall d'Hebron Institute ofOncology, Barcelona, Spain. 6Early Clinical Drug Development Group,Vall d'Hebron Institute of Oncology, Barcelona, Spain. 7Department ofPathology,Vall d'HebronUniversityHospital,UniversitatAut�onomadeBarcelona, Barcelona, Spain. 8Translational Genomics Group, Valld'Hebron Institute of Oncology, Barcelona, Spain. 9General SurgeryService,Vall d'Hebron University Hospital, Barcelona, Spain. 10Depart-ment of HBP Surgery and Transplantation, Vall d'Hebron UniversityHospital, Universitat Aut�onoma de Barcelona, Barcelona, Spain.11Molecular Oncology Group, Vall d'Hebron Institute of Oncology,Barcelona, Spain. 12Novartis Institutes for Biomedical Research, Inc.,Cambridge, Massachusetts.

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

Corresponding Author: H�ector García Palmer, Vall d'Hebron Institute of Oncol-ogy, Passeig Vall d'Hebron 119-129, Barcelona 08035, Spain. Phone: 349-3489-4167; Fax: 349-3274-6708; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-14-3081

�2015 American Association for Cancer Research.

ClinicalCancerResearch

Clin Cancer Res; 22(3) February 1, 2016644

on February 6, 2021. © 2016 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 29, 2015; DOI: 10.1158/1078-0432.CCR-14-3081

http://clincancerres.aacrjournals.org/

relocalization of FOXO proteins and consequent induction ofapoptosis.Wepreviously described thatWnt/b-catenin oncogenicsignaling confers resistance to FOXO3A-dependent apoptosispromoted by PI3K or AKT-inhibitory drugs (10). Such resistancewas driven by nuclear b-catenin that impaired the capacity ofFOXO3A to execute its apoptotic program. Thus,we hypothesizedthat reducing nuclear b-catenin content by Wnt inhibitors wouldovercome the resistance to PI3K or AKT inhibitors and combinedtreatments could be beneficial for treating colorectal cancerpatients.

Abnormal activation of the Wnt/b-catenin pathway by muta-tions in APC, CTNNB1/b-catenin, or AXIN2 is responsible for theinitiation and progression of almost all colorectal cancers (11).These mutations reduce the capacity of the Wnt pathway destruc-tion complex, formed by APC, AXIN, and GSK3b, to commitb-catenin to degradation. As a result, b-catenin accumulates in thenucleus, binds the TCF/LEF transcription factors, and induces theexpression of Wnt target genes that play key roles in tumorprogression (12). We and others have previously shown thatbinding of b-catenin to different transcription factors enhancesthe expression of alternative sets of target genes (13). FOXO3A isone of these transcription factors, for which b-catenin acts as atranscriptional coactivator (14).

As inappropriate activation of the Wnt/b-catenin pathwaywas first linked to colon cancer three decades ago, there hasbeen intense interest in developing effective inhibitors (15, 16).It has been described that tankyrases promote AXIN1/2 parsy-lation and degradation through the proteasome (17). Recently,a new family of tankyrase inhibitors was shown to stabilizeAXIN1/2, enhancing the activity of the destruction complex andreducing free b-catenin. These inhibitors are showing promis-ing preclinical results as Wnt/b-catenin inhibitory drugs for thetreatment of Wnt-addicted tumors (18–20).

Here, we present evidence that high nuclear b-catenin con-tent is associated to resistance to PI3K and AKT inhibitors in thecontext of clinical trials, whereas frequent mutations or clini-copathologic traits implicated in colorectal cancer progressiondo not. We demonstrate that combining these drugs withNVP-TNKS656, a new therapeutic small-molecule inhibitor ofthe Wnt/tankyrase pathway that reduces nuclear b-catenin (21),overcomes such resistance and represses tumor growth incolorectal cancer patient–derived xenograft (PDX) models. We

also identified FOXO3A as a determinant of response toWnt/b-catenin inhibitors and FOXO3A/b-catenin target genesas better pharmacodynamic markers than the canonical TCF/b-catenin targets.

Our data indicate that nuclear FOXO3A and b-catenin contentand activity could be valuable predictive biomarkers of drugresponse and we propose an experimental-based rationale tobetter guide the molecular selection of colorectal cancer patientsentering new clinical trials with PI3K/AKT and/or Wnt/b-cateninpathway inhibitors.

Materials and MethodsPatients in clinical trials

Patients were enrolled in clinical trials with PI3K/AKT/mTORinhibitors carried out in the Vall d'Hebron University Hospi-tal (Barcelona, Spain; Clinical trial identifiers: 14-MC-JWAA,NCT01115751; B2151001, NCT00940498; CBEZ235A2101,NCT00620594; BKM120�2101, NCT01723800; CBYL719�2101,NCT01219699; INK1117-001, NCT01449370; PAM4743g,NCT01090960; XL765-00, NCT00485719). Tumor response wasassessed according to RECIST 1.0 or 1.1 (22, 23). We analyzedformalin-fixed paraffin-embedded (FFPE) tumor samples fromcolorectal cancer patients at baseline before entering clinical trialswith PI3K/AKT/mTOR inhibitors.

Patient-derived cellsWritten informed consent was signed by all patients. The

project was approved by the Research Ethics Committee of theVall d'Hebron University Hospital (Barcelona, Spain; ApprovalID: PR(IR)79/2009). Patient-derived cells were obtained as pre-viously described (24). Cells were injected subcutaneously inNOD-SCID mice or were seeded as sphere cultures.

Animals, xenotransplantation, and treatmentsExperiments were conducted following the European Union's

animal care directive (2010/63/EU) and were approved by theEthical Committee of Animal Experimentation of Vall d'HebronInstitute of Research (ID: 40/08 CEEA and 47/08/10 CEEA).Xenografts were obtained as described in ref. (24). API2(1 mg/kg in PBS-2% DMSO; Tocris Bioscience) was adminis-tered by intraperitoneal injection three times per week, NVP-TNKS656 (100 mg/kg) was injected subcutaneously twice daily.

Gene expressionGene expression of 292 selected genes was profiled in base-

line tumors of patients treated with PI3K/ATK/mTOR inhibitorsusing the nCounter platform from Nanostring Technologies.Differentially expressed genes were identified in tumors pre-senting high or low nuclear b-catenin content using PartekGenomics Suite Software. Lists were cut-off at fold change of1.2 and a P value < 0.075 (two-tailed one-way ANOVA test). Formicroarray analyses, we used a genome wide Human Gene1.0 ST Array (Affimetrix). Data were acquired using the Affime-trix GeneChip/GeneTitan platforms. Genes were considereddifferentially expressed in NVP-TNKS656 versus vehicle-treatedtumors at 1.5 fold change and P < 0.05 using a two-tailed one-way ANOVA test. Microarray data are deposited at ArrayExpressdatabase (E-MTAB-2446). To perform qRT-PCR, RNA fromendpoint tumor xenografts was used to synthesize cDNA usingSuperscript-III reverse transcriptase with oligo-dT and random

Translational Relevance

To date, PI3K and AKT inhibitors are showing limitedclinical benefit mostly due to unknown resistance mechan-isms and the lack of predictive biomarkers of drug re-sponse. We demonstrate that Wnt inhibitors can overcomeb-catenin–induced resistance to PI3K and AKT inhibitors incolorectal cancer tumors. We also provide the experimentalevidence for a rational stratification of patients to be treatedwith PI3K/AKT and/or Wnt/b-catenin pathway inhibitorsusing b-catenin and FOXO3A as predictive biomarkers ofdrug response. Such refined molecular selection of patientscould represent a significant improvement in response totreatment and an important step forward in advancingcolorectal cancer therapy.

Therapeutic Wnt Pathway Inhibition in Colorectal Cancer

www.aacrjournals.org Clin Cancer Res; 22(3) February 1, 2016 645




hexamer primers (Life Technologies). A 7900HT qPCR Systemwas used with Power SYBR-green (Applied Biosystems) andspecific primer pairs. Relative gene expression was determinedby the comparative Ct method (25) and significance wasdetermined using an unpaired t test with Welch correction.See details in Supplementary Materials and Methods.

GenotypingTumor samples from colorectal cancer patients in clinical trials

with PI3K/AKT/mTOR inhibitors were genotyped by sequencingthe amplified product of a multiplexed PCR reaction (Ampliconsequencing) as described in Supplementary Materials and Meth-ods. Frequent mutations in 57 oncogenes and tumor suppressorgenes were interrogated (Supplementary Table S1). Microsatelliteinstability was analyzed using the MSI-Analysis System (Pro-mega). Purified patient-derived cells were genotyped straight aftersurgical resection of patients' tumors by Sequenom or Haloplexplatforms. Genotyping by Sequenom (CLIA panel) was per-formed as previously described (10). Haloplex Target EnrichmentSystem (Agilent Technologies) was used to capture the completecoding regions of 388 oncogenes and tumor suppressor genes(Supplementary Table S2).

Three PDX models were genotyped by Exome sequencing.Patients provided written informed consent for somatic andgermline DNA analysis. Mutations were called with VarScan2software, either using the mpileup2snp or somatic commands,depending on the availability of normal tissue (26). Nontumoraltissue was not available for PDX-P2, thus, common SNPs werefiltered according to the 1000 genome catalogue (27). SIFT andPolyphen-2 helped predicting functionality of the identifiedmutations. Complete Exome sequencing data from PDX-P2,P5, and P30 are available at the SRA database at NCBI (BioProjectID: PRJNA242531).

Immunohistochemistry and immunofluorescenceSamples from paraffin-embedded tissues were stained as

described in ref. 24 using the following antibodies (Supple-mentary Table S3). Nuclei were stained with Hoechst 33342(5 mg/mL; Sigma-Aldrich). Pictures of the immunofluorescentsignal were captured using a NIKON C2þ confocal microscopeand analyzed with MBF ImageJ software using criteria previ-ously described (28, 29).

b-Catenin immunohistochemistry was done using the DakoAutostainer Plus Staining System. For visualization, EnVisionFLEX detection system (DAKO) was used. Sections were counter-stained with hematoxylin, dehydrated, cleared, and mountedfor examination. b-Catenin staining was evaluated by a patho-logist as described in Supplementary Materials and Methods.

DLD1F and HT29F cells were seeded on glass coverslipsand treated for 6 hours with 4-hydroxytamoxifen 100 nmol/L(4-OHT, Sigma-Aldrich). Cells were fixed in 4% para-formalde-hyde (PFA) and immunofluorescent staining was performed asdescribed previously (10).

Apoptosis assaysPatient-derived cells were seeded in suspension as sphere

cultures on low attachment multiwell dishes, whereas cell lineswere seeded in adherent multiwell cell culture dishes. Cells werepretreated with NVP-TNKS656 (100 nmol/L, Novartis) or DMSOfor 48 hours and then with API2 (20 mmol/L, Tocris Bioscience)

and/or NVP-BKM120 (2.4 mmol/L, Selleck Chemicals) foranother 48 hours prior apoptosis analysis. Proportions ofapoptotic cells were determined using the Annexin V-eGFP(BioVision) kit. Dead cells were detected as DAPI negative(1 mg/mL, Roche). Cells were analyzed by flow cytometry usinga Navios Flow Cytometer (Beckman Coulter).

To measure apoptosis by immunofluorescence in sphere cul-tures, cells suspended in culture media were mixed 1:1 withMatrigel (BD Biosciences), fixed for 1 hour in 4% PFA, permea-bilized with PBS/1% Triton X-100 at room temperature for3 hours, and blocked overnight at 4�C in PBS/1% TritonX-100/3% BSA. Samples were incubated for 24 hours with pri-mary antibodies (Supplementary Table S3). Secondary anti-bodies and Hoechst 33342 (5 mg/mL) were incubated overnightat room temperature.

Cell cultureCell lines were cultured under standard conditions. DLD1F

cells are DLD1 derivates expressing pcDNA-FOXO3A(3A)ER(30). HT29F cells express pLHCX-HA-FOXO3A(3A):ER. Allparental cell lines were originally obtained from ATCC. Celllines were authenticated by short-tandem repeat analysis bythe cell bank.

Western blot analysisThe detailed protocol for protein extraction is described in

SupplementaryMaterials andMethods.Western blot analysis wasperformed as described in ref. (10) using specific antibodies(Supplementary Table S3).

TCF/LEF1 reporter assaysDLD1 cell line was stably transfected with a vector (7TGP,

obtained at Addgene) expressing eGFP controlled by a pro-moter containing seven TCF/LEF transcription factor–bindingsites (7xTOP; ref. 31). Cells were treated with NVP-TNKS656100 nmol/L (Novartis) for 7 days and eGFP accumulation wasmeasured by flow cytometry using a Navios Flow Cytometer(Beckman Coulter).

Statistical analysisWe analyzed progression-free survival (PFS) of patients by

the Kaplan–Meier method and compared the curves using alog-rank (Mantel–Cox) test. We used Pearson correlation test tocompare time on previous line of treatment versus time ontreatment with PI3K/AKT/mTOR inhibitors of patients ana-lyzed for nuclear b-catenin content and to correlate apoptosisversus nuclear b-catenin or FOXO3A content in sphere cellcultures treated with NVP-TNKS656, API2, and NVP-BKM120.Pearson correlation test was also used to compare nuclearFOXO3A content in patients and corresponding PDX tumors,or in primary tumors versus liver metastases or to compare withSLC2A3 mRNA levels.

Differences in apoptosis of treated sphere cell cultures; levelsof AXIN1, b-catenin, phosphoS6, Ki67, and cleaved caspase-3expression in tumor xenografts; mRNA expression of NVP-TNKS656 target genes by qRT-PCR; and apoptosis of DLD1F orHT29F cell lines, were analyzed by an unpaired t test comparingthe means of two groups of values. Fisher exact test served toanalyze the differences in response among patients regardingnuclear b-catenin content, mutations affecting PIK3CA, KRAS,

Arqu�es et al.

Clin Cancer Res; 22(3) February 1, 2016 Clinical Cancer Research646




TP53, APC, or tumor histologic TNM status. P values lower than0.05 were considered significant in all tests.

ResultsEffective pharmacologic inhibition of Wnt/b-catenin andPI3K/AKT pathways reduces tumor growth

We hypothesized that reducing nuclear b-catenin contentcould be sufficient to sensitize colorectal cancer tumors to thetreatment with PI3K or AKT inhibitors. Consequently, weblocked the Wnt/b-catenin signaling using the tankyrase inhib-itor NVP-TNKS656 (21) in colorectal cancer PDX models. Weselected five PDXmodels with high (P2, P7, P19, P22, and P30)and five with low nuclear b-catenin content (P5, P6, P31, P33,and P34; Fig. 1A and B). Sphere cell cultures derived fromxenograft tumors of each model were treated with API2 or NVP-BKM120, inhibiting AKT or PI3K activity, respectively, alone orin combination with NVP-TNKS656 (Fig. 1C; SupplementaryTable S4). API2 or NVP-BKM120 induced significantly lessapoptosis in cells with high rather than low nuclear b-catenincontent by measuring the proportion of Annexin V-positivecells. Combination with NVP-TNKS656 significantly increasesapoptosis in cells with high as opposed to low nuclear b-cate-nin content. Similar results were observed by using NVP-XAV939, another inhibitor of tankyrase activity (Supplemen-tary Fig. S1). Alternative measurement of apoptosis by cleavedcaspase-3 showed equivalent results (Supplementary Fig. S2).Furthermore, apoptosis induced by API2 or NVP-BKM120treatment showed a significant inverse correlation with nuclearb-catenin content (Fig. 1D). Such correlation was lost whenAPI2 or NVP-BKM120 was combined with NVP-TNKS656.Mutations in PIK3CA, KRAS, or TP53 genes or tumor site didnot condition a differential response of sphere cell cultures toPIK3 or AKT inhibition (Supplementary Fig. S3). Only one outof the 10 PDX models presented microsatellite instability(MSI), preventing the possibility to evaluate its impact on drugresponse (Supplementary Table S5).

Our data indicate that nuclear b-catenin content conditionsdrug response in patient-derived sphere cell cultures, whereasfrequentmutations in colorectal cancer do not (Fig. 1C andD andSupplementary Fig. S3; Supplementary Table S5).

We further investigated these results in vivo. Cells from threePDX models with known nuclear b-catenin and FOXO3A statusand with limited response to AKT or PI3K inhibition in vitro(P2, P5, and P30), were injected subcutaneously into NOD-SCIDmice (Figs. 1B and D and 2A). Treatment with NVP-TNKS656caused a systemic reduction of nuclear b-catenin content andfunction in skin and intestine, tissues where the Wnt pathwaytightly controls homeostasis, but showed no major negative sideeffects (Supplementary Fig. S4).

API2 alone or in combination with NVP-TNKS656 did notrepress tumor growth in PDX-P5, with lowbasal nuclear b-cateninand FOXO3A content, probably representing a case of a colorectalcancer tumor resistant to AKT inhibition due to mechanismsindependent of nuclear b-catenin accumulation (Fig. 2A andB). Exome sequencing revealed a mutation in the AKT2 gene thatcould explain the lack of reduction of phosphor-S6 or tumorgrowth upon API2 treatment in this model (Supplementary Fig.S5; Supplementary Table S6).

PDX-P2 presented high nuclear b-catenin and low FOXO3Aamounts, was resistant to API2 alone, and yet tumor growth rate

was reduced upon combination with NVP-TNKS656 (Fig. 2Aand B). Tankyrase inhibition significantly reduced nuclear b-cate-nin, API2 decreased phosphor-S6 content, and both diminishedproliferation (Fig. 2C, Supplementary Fig. S5).

PDX-P30, derived from a liver metastasis, presented highamounts of both nuclear b-catenin and FOXO3A (Fig. 2A).It was also resistant to API2, but NVP-TNKS656 treatmentalone reduced tumor growth rate equally to the drug combi-nation (Fig. 2B). Tankyrase but not AKT inhibition promotedapoptosis, whereas the number of proliferative cells was notaffected (Supplementary Fig. S5). This was the only modelwhere NVP-TNKS656 alone showed an effect on tumorgrowth, probably due to high endogenous amounts ofFOXO3A that might have induced apoptosis when nuclearb-catenin was reduced by NVP-TNKS656. NVP-TNKS656increased AXIN1 protein levels in all subcutaneous tumorsconfirming its activity as a Wnt/tankyrase pathway inhibitor(ref. 21; Fig. 2D).

FOXO3A/b-catenin target genes are pharmacodynamicmarkers of response to Wnt/tankyrase-inhibitory drugs

Although NVP-TNKS656 reduced the high nuclear b-catenincontent observed in both PDX-P2 and PDX-P30, treatmentonly reduced tumor growth rate in the latter model (Fig. 2 andSupplementary Fig. S5B). Contrarily, NVP-TNKS656 did notaffect tumor growth in PDX-P5 model. We studied whetherthe high FOXO3A content observed in the metastatic PDX-P30model (Fig. 2A) could determine the repression by NVP-TNKS656 of a distinct set of Wnt/b-catenin target genes andits enhanced sensitivity to treatment. RNA from tumors ofPDX-P2, PDX-P30, and PDX-P5 models was analyzed at theendpoint of the in vivo experiments (Fig. 2B). The three modelsshowed a distinct gene expression pattern that was modifiedby NVP-TNKS656 treatment (Fig. 3A; Supplementary TablesS7–S9). The prometastatic S100A4 gene was repressed inboth PDX-P2 and PDX-P30 models (Fig. 3B). Four NVP-TNKS656–repressed genes in PDX-P2, two in PDX-P5, and yetnone in PDX-P30 were direct TCF/b-catenin targets (Supple-mentary Table S10). Instead, two genes in PDX-P2, one inPDX-P5, and 9 in PDX-P30 were FOXO3A/b-catenin targets(Supplementary Table S11), many of them formally associatedwith metastasis (10, 32–34). The regulation of some of theseNVP-TNKS656–repressed genes was further confirmed by qRT-PCR in the same tumor xenograft samples (Fig. 3C). Any geneevaluated in PDX-P5 model was significantly regulated byNVP-TNKS656, a result in line with its low nuclear b-catenincontent and its lack of tumor growth response (Fig. 2B). Ourdata suggest that FOXO3A/b-catenin targets could be betterpharmacodynamic markers than TCF/b-catenin target genesfor evaluating therapeutic Wnt/tankyrase pathway inhibition.

FOXO3A determines the response to Wnt/b-catenin pathwayinhibitors

NVP-TNKS656 treatment was particularly effective in atumor with high endogenous nuclear b-catenin and FOXO3Acontent, promoting apoptosis, reducing tumor growth rate(Fig. 2) and preferentially repressing the expression of FOXO3Ainstead of TCF target genes (Fig. 3). The repression of theWnt/b-catenin pathway by NVP-TNKS656 alone did not pro-mote apoptosis in DLD1 or HT29 colon cancer cells but






Figure 1.Colorectal cancer patient–derived cells with high amounts of nuclear b-catenin present high sensitivity to API2 or NVP-BKM120 in combination withNVP-TNKS656. A, representative pictures of immunofluorescence and confocal microscopy of histologic sections of the indicated PDX models withhigh (left) or low (right) nuclear b-catenin content. Inserts show magnification to better visualize b-catenin sub-cellular localization. Dashed linesdelineate nuclei. Nuclei were stained with Hoechst 33342 (blue). Scale bars, 100 mm; magnifications 50 mm. B, column scatter plot showing theamount of nuclear b-catenin measured by immunofluorescence and confocal microscopy in 10 primary tumors and liver metastases from whichsphere cell cultures were derived and used to test drug response. Horizontal lines indicate arithmetic mean values, and error bars show SD. C,column scatter plot showing the apoptosis induced in sphere cell cultures of these 10 patient-derived models treated as indicated. Data, fold change ofapoptotic cells induced by the treatment compared with cells treated with vehicle. Horizontal lines indicate arithmetic mean values, and error barsshow SEM. P values correspond to unpaired t tests. The original percentage of Annexin V–positive cells is shown in Supplementary Table S4. D,scatter plots representing the apoptosis induced by API2 (top left) and NVP-BKM120 (top right) or API2 þ NVP-TNKS656 (bottom left) andNVP-BKM120 þ NVP-TNKS656 (bottom left) in sphere cell cultures of these 10 patient-derived models versus the histologic amount of nuclear b-cateninin the original patient's tumors. Data, fold change of apoptotic cells induced by the treatment compared to cells treated with vehicle. P valuescorrespond to Pearson correlation test. B and D, b-catenin relative units (r.u.) were calculated as described in Materials and Methods.

Arqu�es et al.





enhanced the apoptosis induced by exogenous nuclearFOXO3A-ER (Fig. 4). These data confirm the capacity of nuclearb-catenin to confer resistance to FOXO3A-induced apoptosis(10) and, therefore, the therapeutic value of reducing nuclearb-catenin by tankyrase inhibitors in FOXO3A-active cancercells.

Furthermore, NVP-TNKS656 promoted apoptosis in patient-derived sphere cell cultures proportionally to the amount ofnuclear FOXO3Adetected in their correspondent patients' tumors(Fig. 5A and B). Apoptosis was also proportional to FOXO3Acontent when sphere cultures were treated with a different tan-kyrase inhibitor, NVP-XAV939 (Supplementary Fig. S1). Similarlyto b-catenin (10), PDX models and primary sphere culturesfaithfully preserved the levels of nuclear FOXO3A detected in

their correspondent original patient samples (Supplementary Fig.S6). We also observed that nuclear FOXO3A content positivelycorrelated with the expression of SLC2A3 mRNA, a FOXO3A/b-catenin target gene (10) thatwas repressed uponNVP-TNKS656treatment in vivo (Fig. 3).

Interestingly, we observed that paired primary tumors andliver metastases accumulated similar nuclear FOXO3Aamounts, showing that its activation could be durable andoccur prior progression to metastatic stages (SupplementaryFig. S7).

Finally, by profiling 130 colorectal cancer cases, we identified adistinctive population of patients with tumors presenting highexpression of FOXO3A/b-catenin target genes (Fig. 5C). Theexpression of TCF/b-catenin target genes also identified such a

Figure 2.NVP-TNKS656 stabilizes AXIN1 and reduces both, nuclear b-catenin and tumor growth alone or in combination with the AKT inhibitor API2 incolorectal cancer PDX models. A, representative pictures of double immunofluorescent staining and confocal microscopy to detect b-catenin (red)and FOXO3A (green) in histologic sections of subcutaneous xenografted tumors from patients P2, P5, and P30. Right panels show magnifications tovisualize the amounts of nuclear b-catenin and FOXO3A. Scale bar, 100 mm; magnifications, 20 mm. Nuclei were stained with Hoechst 33342 (blue).Arrowheads point to b-catenin localized exclusively in cell membranes in tumors from P5 model. B, tumor cells derived from the three indicatedpatients were injected subcutaneously in NOD-SCID mice and treated as indicated. A minimum of 5 mice with tumors in both flanks was treated ineach group. The graphs represent the fold change calculated by comparing the tumor volume at each given time point to the volume at the first day oftreatment. Error bars and � SD are shown for tumor volume fold change of all tumors. Unpaired t tests were used to compare the area under thecurve generated for each growing tumor along the experiment and grouped by treatment. Asterisks indicate significant differences (P < 0.05). C,representative pictures showing immunofluorescent staining of b-catenin in tumor xenografts from PDX-P2 model growing in mice treated withvehicle or NVP-TNKS656. Inserts show magnifications to better visualize b-catenin reduction from cytoplasm and nuclei upon tankyrase inhibition.Scale bars, 100 mm; inserts, 50 mm. D, three xenograft tumors per group of treated mice were processed for the analysis of AXIN1 by Western blotanalysis (top). b-Tubulin was used as loading control. Blots were quantified using ImageJ software (bottom). Protein expression levels are represented asrelative units (r.u.) versus vehicle treated samples. Bars indicate SD. P values correspond to unpaired t tests.






population but showed lower signal. Interestingly, FOXO3A/b-catenin and TCF/b-catenin target genes clustered separatelyamong all cases profiled. Samples were also evaluated for theirMSI and mutational status of KRAS, BRAF, and PIK3CA, allmolecular features relevant for colorectal cancer tumors. Anyobvious correlation was observed between them and gene expres-sion signatures distinctive of active FOXO3A/b-catenin transcrip-tion (Fig. 5C; Supplementary Table S12).

These data suggest that patients with tumors that are activefor FOXO3A/b-catenin transcription can be identified bytranscriptional profiling, whereby reduction of nuclear b-cate-nin by Wnt inhibitors could promote FOXO3A-dependentapoptosis.

Comparative analysis of b-catenin as potential biomarker ofresistance to PI3K and AKT-inhibitory drugs

We decided to compare the potency of nuclear b-catenin inpredicting resistance to PI3K or AKT inhibitors with the mostfrequent mutations or histological traits observed in colorectalcancer tumors. Limited availability of samples from clinicaltrials only permitted the study of tumors at baseline from acohort of 40 colorectal cancer patients treated with PI3K/AKT/mTOR pathway inhibitors in several phase I clinical trials(Fig. 6A, top; Supplementary Table S13). We selected tumorsfrom patients treated with half the maximum tolerated dose

aiming to homogenize the study cohort. There was no signif-icant difference in PFS between patients treated with differentPI3K, AKT, or dual PI3K/mTOR drug subtypes (Fig. 6A, bot-tom). We performed a double-blinded evaluation of nuclearb-catenin content by two independent pathologists usingimmunohistochemistry and immunofluorescence on all base-line tumor samples (Fig. 6B and Supplementary Fig. S8). Weclassified tumors into two histologic categories: high or low,depending on the number of cells positive for nuclear b-cateninaccumulation. Out of 40 cases, 13 were high and 27 were lowin content.

After the first clinical diagnosis of disease progression bycomputerized axial tomography (CT) scan, all 13 patientspresenting tumors with high nuclear b-catenin content hadprogressed despite treatment with PI3K/AKT/mTOR pathwayinhibitors. Contrarily, 9 of 27 patients with tumors presentinglow b-catenin continued treatment after CT scan evaluation,showing some degree of stabilized disease (Fig. 6C, top). Suchlonger lasting response to PI3K/AKT/mTOR pathway inhibitorsdid not correlate with better response to the previous line oftreatment, ruling out the possibility that those 9 patients wereeither more sensitive to antitumoral drugs in general or hadslower tumor growth independently of treatment [Figs. 6C(bottom) and D]. Contrary to nuclear b-catenin content, rele-vant oncogenic mutations in PIK3CA, KRAS, APC, or TP53

Figure 3.Reduction of nuclear b-catenin content by NVP-TNKS656 regulates gene expression in colorectal cancer PDX models. A, three tumor xenografts frommice injected with PDX-P2, PDX-P30, or PDX-P5 treated with vehicle or NVP-TNKS656 were analyzed for gene expression at the end point of theexperiments. Gene clustering diagrams shown were calculated using robust multiarray normalized expression values from genome-wide microarrayanalysis. Triplicates of each treatment are indicated as colored rectangles (R1, R2, and R3). Significance was calculated using two-tailed ANOVAwith a significance cutoff of P � 0.05. Gene expression scaling was indicated from low (blue) to high (red). B, Venn diagrams showing the number ofgenes downregulated by NVP-TNKS656 in all PDX models and genes that belong to the TCF/b-catenin (44) or FOXO3A/b-catenin (10) transcriptionalprograms. Common genes are indicated on the right and enumerated in brackets. Asterisks indicate genes involved in metastasis. C, relative mRNAexpression levels of selected genes was confirmed by qRT-PCR in the same xenograft tumors from the three PDX models treated with vehicle (blue bars)or NVP-TNKS656 (red bars). Expression values are shown as normalized DCt. Error bars � SD of triplicate values obtained in three independentmeasurements are shown. P values correspond to unpaired t tests.

Arqu�es et al.





genes, the site of tumor samples (primary tumor or metastasis)or TNM stage (T3 or T4) at the time of diagnosis did notcorrelate with any significant difference in response to PI3K/AKT/mTOR pathway inhibitors (Fig. 6C and D and Supple-mentary Fig. S9A–S9F; Supplementary Tables S13 and S14).Only 2 of 28 samples analyzed presented MSI, preventing thepossibility to evaluate its impact on drug response (Supple-mentary Table S13).

Equivalent analyses were performed separately in patientstreated with PKI-587 (n¼ 11), BEZ235 (n¼ 9), or NVP-BKM120(n ¼ 6). We observed that cases with high nuclear b-cateninpresented a shorter PFS upon PKI-587 or BEZ235 treatment

(Supplementary Fig. S9G–S9I). However, results were not statis-tically significant as expected from the small number of casesavailable for the analyses.

To confirm the higher risk of progression in the b-cateninhigh–group, we performed a Cox Proportional Hazards analysison a subset of our cohort (n ¼ 27) with complete annotation forvariables potentially linked to outcome: age, gender, therapy(AKT, PI3Ka, pan-PI3K, PI3K-mTOR inhibitor), number ofprior treatment lines, presence of liver metastasis, and molecularprofile (PIK3CA, KRAS, TP53, APC mutations, PTEN loss). Evenafter multivariate adjustment, patients whose tumors had highnuclear b-catenin content still displayed a significantly worse

Figure 4.NVP-TNKS656 sensitizes colorectal cancer cell lines to the apoptosis induced by FOXO3A. A, immunofluorescence and confocal microscopy to detectthe translocation of exogenous FOXO3A(A3):ER from cytoplasm to nucleus of DLD1F and HT29F colon cancer cells upon 4-OHT treatment. Scale bar,10 mm. B, Western blots showing the amount of exogenous FOXO3A(A3):ER in the nucleus and stabilization of AXIN1 in the cytoplasm of DLD1F andHT29F cells upon indicated treatments. The amounts of b-tubulin and Lamin A/C were evaluated to confirm the purity of nuclear and cytoplasmicextracts and as loading control. C, Representative pictures of DLD1F cells expressing fluorescent green protein (eGFP) under the control of sevenupstream TCF/b-catenin–binding sites (7xTOP-eGFP) upon 5 days of NVP-TNKS656 treatment (left). Scale bar, 200 mm. 7xTOP-eGFP activity aftertreating DLD1F cells with vehicle or NVP-TNKS656 was assessed by flow cytometry (right). Bars indicate SD of experimental triplicates. D, apoptosiswas measured in DLD1F and HT29F cells upon indicated treatments. Data, fold change of apoptotic cells induced by the treatment compared withcells treated with vehicle. Bars show � SD of five replicates in six independent experiments. P values correspond to unpaired t tests.






Figure 5.High FOXO3A activity in colorectal cancer patient-derived cells determines the apoptosis induced by NVP-TNKS656. A, representative pictures of doubleimmunofluorescence and confocal microscopy to detect b-catenin (red) and FOXO3A (green) in models PDX-P33 and PDX-P34. Arrowheads point tocancer cells in PDX-P33 presenting high nuclear FOXO3A and b-catenin accumulation. PDX-P33 presented high and PDX-P34 low nuclear FOXO3A andb-catenin content. Bottom panels show magnifications to better visualize b-catenin and FOXO3A subcellular localization. White dotted lines delineatenuclei stained with Hoechst 33342 (blue). Scale bars, 100 mm; magnifications, 50 mm. B, Scatter plot comparing apoptosis induced by NVP-TNKS656 inxenograft-derived sphere cell cultures versus the histological amount of nuclear FOXO3A in the corresponding PDX models. (Continued on the following page.)

Arqu�es et al.





outcome (PFSHR3.96, 95%confidence interval, 1.03–15.27; log-rank P 0.0158; Supplementary Table S15).

Concerning gene expression patterns, we could analyze 31 ofthe initial 40 baseline samples and observed that tumors respond-ing to PI3K/AKT/mTOR pathway inhibitors clustered together

(Supplementary Fig. S10A). The oncogenic mutations detecteddid not correlate with any of the gene expression clustersobserved. We finally observed that the expression profile of areduced set of genes could also help identifying tumors presentinghigh or low nuclear b-catenin content, cross-validating the initial

Figure 6.High nuclear b-catenin content is associated with resistance to PI3K/AKT/mTOR inhibitors in colorectal cancer patients. A, Table summarizing moleculartargets, drugs, and number of patients (40 in total) in each clinical trial whose baseline tumors were profiled. Kaplan–Meier analysis representingPFS of patients separated by drug subtype administered: dual PI3K/mTOR, PI3K or AKT inhibitors. Number of patients is shown for each cohort (n). P valuewas calculated by log-rank (Mantel–Cox) test. B, representative pictures of double immunofluorescence staining and confocal microscopy to detect b-catenin(red) and a-catenin (green) in colorectal cancer tumor sections of patients treated with PI3K or AKT inhibitors. Tumors present low (top image) orhigh (bottom image) nuclear b-catenin content (Scale bar, 100 mm). Panels on the right show magnifications with fluorescent channels split to bettervisualize b-catenin subcellular localization (Scale bar, 20 mm). All sections were counterstained with Hoechst 33342 (blue) to detect nuclei. Dashed linesdelineate nuclei. C, chart representing the time that each patient received a PI3K/mTOR, PI3K or AKT inhibitor and the time on the correspondingprevious line of treatment (top). The 40 patients studied are split in those with tumors presenting high and those with low nuclear b-catenin accumulation.Pink vertical bar indicates the latest period of time when disease progression was evaluated by computerized axial tomography (CAT) scan. Greenhorizontal bars correspond to patients who had progressed to treatments with PI3K/mTOR, PI3K or AKT inhibitors at the time of first CAT scan afterinitiating the clinical trial. Red bars show the time on treatment for those patients who had not progressed to treatment at the time of first CAT scan. Bottom,scatter plot representing the time on treatment with PI3K/mTOR, PI3K, or AKT inhibitors versus the time on previous line of treatment for all 40colorectal cancer patients. Their level of nuclear b-catenin is indicated. P value corresponds to Pearson correlation test. D, Kaplan–Meier curvesshowing progression-free survival (PFS) of patients presenting tumors with high or low nuclear b-catenin content and treated with PI3K/mTOR, PI3K orAKT inhibitors (left) or the correspondent previous line of therapy (right). Number of patients (n) is shown for each cohort. P value, HR and 95%confidence interval (CI) are shown and calculated by the log-rank (Mantel–Cox) test.

(Continued.) FOXO3A relative units (r.u.) were calculated as described in methods. Apoptosis is represented as fold change of apoptotic cells inducedby NVP-TNKS656 compared with the cells treated with vehicle. P values correspond to Pearson correlation test. Nuclear b-catenin content for each model isalso indicated. C, Nanostring platform was used to analyze the expression of 31 FOXO3A/b-catenin and 21 TCF/b-catenin target genes in FFPE tumorsamples from 130 colorectal cancer patients. Samples and genes were ordered by hierarchical clustering using uncentered Pearson correlation distanceand complete linkage. MSI and mutational status of KRAS, PIK3CA and BRAF were evaluated as indicated in methods. Patients showing highest expression ofFOXO3A/b-catenin and TCF/b-catenin target genes are highlighted with a yellow box. Gene expression scaling is shown from low (blue) to high (red).






cut-off selected to catalog tumors histologically (Fig. 6B andSupplementary Figs. S8 and S10B).

DiscussionThe efficacy of several PI3K and AKT inhibitors is being tested

in multiple clinical trials worldwide. Although initial resultswere promising in some tumor types, clinical responses are nullor limited in most colorectal cancer patients (4–8). A reason-able hypothesis for such limited responses is the absence of auniversal biomarker to select drug-sensitive patients or discardresistant cases. As activating mutations in PIK3CA gene confersensitivity to PI3K/AKT pathway inhibitors in preclinical assays,they have been commonly used as inclusion criteria in clinicaltrials. However, PIK3CA mutations showed conflicting resultsin predicting response to single-agent PI3K or AKT inhibitors inearly-phase clinical trials (7, 35–37). Similarly, we show herethat mutations in genes frequently altered in colorectal cancer,including KRAS, TP53, or PIK3CA, do not predict response toPI3K/AKT pathway inhibitors. Interestingly, multivariate anal-ysis indicates that APC mutations are a risk factor for patientstreated with these inhibitors (Supplementary Table S15). Thiswould suggest that oncogenic activation of the Wnt/b-cateninpathway could be a mechanism of resistance to PI3K and AKTinhibitors.

Indeed, we previously described that nuclear accumulation ofb-catenin conferred resistance to PI3K andAKT inhibitors in coloncancer cells (10). In the current study, we observed that colorectalcancer patients with high nuclear b-catenin present a shorter PFSwhen treated with PI3K/AKT/mTOR pathway inhibitory drugs inthe context of clinical trials. Such differential response contrastswith the fact that all colorectal cancer patients in our cohortprogressed to previous lines of treatment irrespective of theirnuclear b-catenin content. These results on clinical samples indi-cate that nuclear b-catenin accumulation could be an indepen-dent predictive biomarker of resistance to PI3K and AKT inhibi-tors beyond other molecular alterations frequent in colorectalcancer.

In accordance, we explored here the therapeutic potential ofovercoming such resistance by reducing nuclear b-catenin contentwith a new Wnt/tankyrase inhibitor, NVP-TNKS656. We showthat it sensitizes patient-derived cells to PI3K or AKT inhibitors invitro and in vivo, especially those with high accumulation ofnuclear b-catenin and, thus, high oncogenic Wnt/b-catenin path-way activity.

Our results on Wnt/tankyrase inhibitors reveal a new ther-apeutic opportunity for the treatment of advanced colorectalcancer patients beyond their combination with PI3K or AKTinhibitors. Indeed, we previously showed that cancer cellswith high proapoptotic FOXO3A transcriptional activityrequire high levels of nuclear b-catenin to preserve a viablebalance between survival and apoptosis (10). Here, we ob-served that NVP-TNKS656 treatment was particularly effectivein cells and tumors with high endogenous nuclear b-cateninand FOXO3A content, where the pharmacologic reduction ofnuclear b-catenin promoted FOXO3A-dependent apoptosis.Therefore, it is expected that cancer cell survival in such endog-enous FOXO3A-active scenario, more frequent in advancedmetastatic colorectal cancer tumors (10), would rely onWnt/b-catenin pathway activity. This could represent an excep-tional opportunity to compromise cancer cell viability by

treating patients with tankyrase inhibitors. Indeed, we identifyby gene expression profiling a population of patients present-ing tumors with high FOXO3A/b-catenin activity whocould potentially respond best to the treatment with theseWnt/b-catenin pathway inhibitors.

However, very little is known about the determinants ofnuclear FOXO3A activation in colorectal cancer. It is welldescribed that oxidative stress promotes FOXO3A transloca-tion from the cytoplasm to the nucleus and the consequentinduction of genes involved in cell-cycle arrest, survival, apo-ptosis, and metastasis (9). Future investigations should revealthe precise contribution of oxidative stress and other stimulioriginated in the surrounding tumor stroma on determiningthe final levels of activated FOXO3A. Our results suggest thatdescribing FOXO3A-activating factors would be of particularinterest to understand the response to Wnt/b-catenin inhibi-tors. The use of PDX models could be pivotal in these inves-tigations since they preserve equivalent levels of FOXO3Athan the original patients' tumors. In fact, the efficacy ofWnt/tankyrase inhibitors observed in cancer cell lines or geneticmouse models has been modest (18–20), as they may notrecapitulate tumoral FOXO3A activity as consistently as PDX.Indeed, PDX are generally considered the best preclinical mod-els to evaluate drug response as they faithfully recapitulatepatient's disease preserving their molecular alterations andhistopathologic traits (24, 38, 39).

Other alterations frequent in colorectal cancer could also berelevant determinants of response to Wnt/b-catenin inhibitors.For instance, truncating mutations in APC gene are present inmore than 80% of colorectal cancer patients and constitutivelyactivate the oncogenic Wnt/b-catenin pathway (3). Here weshow that NVP-TNKS656 can reduce nuclear b-catenin contentand repress tumor growth even in APC-mutant PDX models.These data suggest that the therapeutic potential of tankyraseinhibitors could be extended to the majority of colorectalcancer tumors. This wide spectrum would contrast with thelack of activity in APC-mutated tumors expected from thetreatment with NVP-LGK974. This is the first Wnt/b-cateninpathway inhibitor tested in clinical trials (NCT01351103),which blocks porcupine activity and the maturation of Wntligands upstream in the oncogenic signaling. As tankyrasesparsylate and commit other proteins to degradation in additionto AXIN1 and 2 (40, 41), it could be of interest to evaluate towhat extent the antitumoral capacity of tankyrase inhibitorsrely on affecting any other cell processes beyond Wnt/b-cateninsignaling.

Aiming to define clinically useful biomarkers to predict theresponse to PI3K/AKT and Wnt/b-catenin pathway inhibitors,we investigated the potential of gene expression signatures. Ourinitial results suggest that specific gene expression profilescould identify colorectal cancer tumors with high FOXO3Aand b-catenin content and transcriptional activity. In particular,we observed that a particular set of genes was overexpressed incolorectal cancer tumors with high nuclear b-catenin content.Our drug treatment experiments indicate that such Wnt/b-cate-nin activated tumors could be resistant to PI3K and AKTinhibitors. Furthermore, we observed a significant correlationbetween nuclear FOXO3A accumulation and mRNA expressionof one of its target genes, SLC2A3 (10). These results suggestthat nuclear accumulation of FOXO3A and b-catenin observedby histology reflects their activation as transcription factors

Arqu�es et al.





inducing the expression of their corresponding target genes.Finally, we could identify, using Nanostring nCounter plat-form, a population of colorectal cancer patients with tumorspresenting high expression of FOXO3A/b-catenin and TCF/b-catenin target genes. Our preclinical data indicate that thesepatients could benefit from the treatment with Wnt/tankyraseinhibitors.

Together, these results suggest that gene expression profilingcould help to build complex predictive biomarkers of responseto PI3K/AKT and Wnt/b-catenin pathway inhibitors. However,selecting a precise set of genes to build robust signatures wouldrequire evaluating whole gene expression patterns by micro-arrays or RNAseq in prospective studies with fresh-frozentumor samples. These expression profiles should be cross-compared with histologic evaluation of nuclear FOXO3A andb-catenin content and the response to treatment, to finallydefine those gene sets associated to such histologic and clinicaltraits. These biomarker exploratory studies should ideally focuson analyzing tumor biopsies taken from progressive lesions atthe time of inclusion in clinical trials and not from archivalsamples. In the particular case of nuclear FOXO3A, we haveobserved that advanced metastatic tumors present the highestproportion of cases activated for this transcription factor (10).Therefore, validating the use of nuclear FOXO3A accumulationand its associated gene expression signatures as potential bio-marker of sensitivity to Wnt/b-catenin pathway inhibitorswould require taking biopsies preferentially from metastaticlesions that will actually be progressing at the time of patients'inclusion in clinical trials.

Hence, we suggest combining histologic evaluation ofFOXO3A and b-catenin with functional gene expression sig-natures to build complex predictive biomarkers of responseto PI3K/AKT and Wnt/b-catenin pathway inhibitors in colo-rectal cancer, which may help the design of future clinicaltrials with this family of drugs. Similarly, current studies areanalyzing the expression of 50 selected genes by NanostringnCounter platform to facilitate a more precise definition ofbreast cancer subtypes (42, 43), whose differential response totreatment is currently under validation in several clinical trials.In the same line as our proposal, such gene expression sig-natures are being combined with clinical evaluation of hor-mone receptors and HER2 protein levels by histopathologictechniques.

In summary, we propose combining gene expression pro-filing and histology to define nuclear b-catenin and FOXO3Acontent and activity as predictive biomarkers of drug re-sponse (Supplementary Fig. S11). We hypothesize that thismolecular prescreening could establish three groups ofpatients presenting tumors with: (i) low nuclear b-cateninand FOXO3A activity, more suitable for the treatment withPI3K or AKT inhibitors alone, (ii) high b-catenin and FOXO3Aactivity who could benefit from Wnt/b-catenin inhibitorsalone and (iii) high nuclear b-catenin but low FOXO3A

activity who may benefit from combined treatments. Suchmolecular stratification of patients could represent a signifi-cant improvement in response to therapy and an importantstep forward towards reverting the long-stalled scenario ofcolorectal cancer therapy.

Disclosure of Potential Conflicts of InterestJ. Tabernero is a consultant/advisory board member for Novartis. No

potential conflicts of interest were disclosed by the other authors.

DisclaimerThe funders had no role in study design, data collection and analysis,

decision to publish, or preparation of the manuscript.

Authors' ContributionsConception and design: O. Arques, W. Shao, J. Tabernero, H.G. PalmerDevelopment of methodology: O. Arques, I. Chicote, I. Puig, R. Dienstmann,G. Caratu, J. Tabernero, H.G. PalmerAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.):O. Arques, I. Chicote, I. Puig, S.P. Tenbaum,G. Argiles,R. Dienstmann,N. Fernandez, G. Caratu, J. Matito, J. Rodon, S. Landolfi, A. Prat,E. Espin, R. Charco, P. Nuciforo, A. Vivancos, J. TaberneroAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): O. Arques, S.P. Tenbaum, R. Dienstmann,D. Silberschmidt, J. Rodon, A. Prat, P. Nuciforo, A. Vivancos, J. Tabernero,H.G. PalmerWriting, review, and/or revision of the manuscript: O. Arques, G. Argiles,R. Dienstmann, J. Rodon, E. Espin, P. Nuciforo, W. Shao, H.G. PalmerAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): O. ArquesStudy supervision: O. Arques, H.G. Palmer

AcknowledgmentsThe authors thank Dr. Alan Huang (Novartis Institutes for BioMedical

Research, Inc) for providing the PI3K inhibitor NVP-BKM120 and Dr. Paul J.Coffer (Utrecht, Netherlands) for kindly providing the pcDNA3-FOXO3A(A3):ER expression plasmid. The authors also thank Javier Hern�andez Losa forevaluation of MSI status and Amanda Wren for her valuable assistance inthe preparation of the English manuscript.

Grant SupportThis work was supported by Instituto Carlos III - Fondo de Investiga-

ciones Sanitarias - ISCIII (FIS-PI081356, FIS-PI11/00917, FIS-PI14/00103and FIS-PI14/01239). O. Arqu�es was supported by Ag�encia de Gesti�od'Ajuts Universitaris i de Recerca - AGAUR; S.P. Tenbaum was supportedby FERO Fellowship and Red Tem�atica de Investigaci�on Cooperativa delC�ancer -RTICC (RD12/0036/0012), I. Puig was funded by the Fundaci�onCientífica de la Asociaci�on Espa~nola Contra el Cancer (AECC), and H.G.Palmer was supported by the Miguel Servet Program, Instituto de SaludCarlos III-ISCIII.

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

Received December 5, 2014; revised May 18, 2015; accepted July 2, 2015;published OnlineFirst July 29, 2015.

References1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer

statistics. CA Cancer J Clin 2011;61:69–90.2. Janne PA, Mayer RJ. Chemoprevention of colorectal cancer. N Engl J Med

2000;342:1960–8.

3. TCGA. Comprehensive molecular characterization of human colon andrectal cancer. Nature 2012;487:330–7.

4. Courtney KD, Corcoran RB, Engelman JA. The PI3K pathway as drug targetin human cancer. J Clin Oncol 2010;28:1075–83.






5. Markman B, Atzori F, Perez-Garcia J, Tabernero J, Baselga J. Status of PI3Kinhibition and biomarker development in cancer therapeutics. Ann Oncol2010;21:683–91.

6. Bendell JC, Rodon J, Burris HA, de Jonge M, Verweij J, Birle D, et al.Phase I, dose-escalation study of BKM120, an oral pan-class I PI3Kinhibitor, in patients with advanced solid tumors. J Clin Oncol 2012;30:282–90.

7. JuricD, Rodon J,Gonzalez-AnguloAM,BurrisHA, Bendell J, Berlin JD, et al.BYL719, a next generation PI3K alpha specific inhibitor: preliminary safety,PK, and efficacy results from the first-in-human study [Abstract]. In:Proceedings of the 103rd Annual Meeting of the American Associationfor Cancer Research; 2012 Mar 31–Apr 4; Chicago, IL. Philadelphia (PA):AACR; Cancer Res 2012;72(8 Suppl):Abstract nr CT-01. doi:1538-7445.AM2012-CT-01.

8. Burris H Jr, Sharma S, Herbst RS, Tabernero J, Infante JR, Silva A, et al.First-in-human phase I study of the oral PI3K inhibitor BEZ235 inpatients (pts) with advanced solid tumors. J Clin Oncol 2010;28:15s(suppl; abstr 3005).

9. Eijkelenboom A, Burgering BM. FOXOs: signalling integrators for homeo-stasis maintenance. Nat Rev Mol Cell Biol 2013;14:83–97.

10. Tenbaum SP, Ordonez-Moran P, Puig I, Chicote I, Arques O, Landolfi S,et al. beta-catenin confers resistance to PI3K and AKT inhibitors andsubverts FOXO3a to promote metastasis in colon cancer. Nat Med 2012;18:892–901.

11. Clevers H, Nusse R. Wnt/beta-catenin signaling and disease. Cell 2012;149:1192–205.

12. Reya T, Clevers H. Wnt signalling in stem cells and cancer. Nature 2005;434:843–50.

13. Palmer HG, Gonzalez-Sancho JM, Espada J, BercianoMT, Puig I, Baulida J,et al. Vitamin D(3) promotes the differentiation of colon carcinoma cellsby the induction of E-cadherin and the inhibitionof beta-catenin signaling.J Cell Biol 2001;154:369–87.

14. Essers MA, de Vries-Smits LM, Barker N, Polderman PE, Burgering BM,Korswagen HC. Functional interaction between beta-catenin and FOXOin oxidative stress signaling. Science 2005;308:1181–4.

15. Voronkov A, Krauss S. Wnt/beta-catenin signaling and small moleculeinhibitors. Curr Pharm Des 2013;19:634–64.

16. Anastas JN, Moon RT. WNT signalling pathways as therapeutic targets incancer. Nat Rev Cancer 2013;13:11–26.

17. Huang SM, Mishina YM, Liu S, Cheung A, Stegmeier F, Michaud GA, et al.Tankyrase inhibition stabilizes axin and antagonizes Wnt signalling.Nature 2009;461:614–20.

18. LauT,ChanE,CallowM,Waaler J, Boggs J, BlakeRA, et al. Anovel tankyrasesmall-molecule inhibitor suppresses APC mutation-driven colorectaltumor growth. Cancer Res 2013;73:3132–44.

19. Waaler J, Machon O, Tumova L, Dinh H, Korinek V, Wilson SR, et al. Anovel tankyrase inhibitor decreases canonical Wnt signaling in coloncarcinoma cells and reduces tumor growth in conditional APC mutantmice. Cancer Res 2012;72:2822–32.

20. Morrisey E, James RG, Davidson KC, Bosch KA, Biechele TL, Robin NC,et al. WIKI4, a novel inhibitor of tankyrase and Wnt/b-catenin signaling.PLoS ONE 2012;7:e50457.

21. Shultz MD, Cheung AK, Kirby CA, Firestone B, Fan J, Chen CH-T, et al.Identification of NVP-TNKS656: the use of structure–efficiency relation-ships to generate a highly potent, selective, and orally active tankyraseinhibitor. J Med Chem 2013;56:6495–511.

22. Therasse P, Arbuck SG, Eisenhauer EA,Wanders J, Kaplan RS, Rubinstein L,et al. New guidelines to evaluate the response to treatment in solid tumors.European Organization for Research and Treatment of Cancer, NationalCancer Institute of theUnited States, National Cancer Institute of Canada. JNatl Cancer Inst 2000;92:205–16.

23. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al.New response evaluation criteria in solid tumours: revised RECIST guide-line (version 1.1). Eur J Cancer 2009;45:228–47.

24. Puig I, Chicote I, Tenbaum SP, Arqu�es O, Herance JR, Gispert JD, et al. Apersonalized pre-clinical model to evaluate the metastatic potential ofpatient-derived colon cancer cells. Clin Cancer Res 2013;2013:6787–801.

25. Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparativeC(T) method. Nat Protoc 2008;3:1101–8.

26. Koboldt DC, Zhang Q, Larson DE, Shen D, McLellan MD, Lin L, et al.VarScan 2: somatic mutation and copy number alteration discovery incancer by exome sequencing. Genome Res 2012;22:568–76.

27. AbecasisGR,AutonA, Brooks LD,DePristoMA,DurbinRM,Handsaker RE,et al. An integrated map of genetic variation from 1,092 human genomes.Nature 2012;491:56–65.

28. Arqu�es O, Chicote I, Tenbaum S, Puig I, Palmer HG. Quantitativeprocedure to analyze nuclear b-catenin using immunofluorescencetissue staining. Protoc exch. 2014 Jun 5. pii: 3121. doi:10.1038/protex.2014.018.

29. Arqu�es O, Chicote I, Tenbaum S, Puig I, Palmer HG. Standardized relativequantification of immunofluorescence tissue staining. Protoc exch. 2012Apr 2. pii: 2349. doi:10.1038/protex.2012.008.

30. Dijkers PF, Medema RH, Lammers JW, Koenderman L, Coffer PJ. Expres-sion of the pro-apoptotic Bcl-2 family member Bim is regulated by theforkhead transcription factor FKHR-L1. Curr Biol 2000;10:1201–4.

31. Fuerer C, Nusse R. Lentiviral vectors to probe and manipulate the Wntsignaling pathway. PLoS One 2010;5:e9370.

32. Bartucci M, Dattilo R, Moriconi C, Pagliuca A, Mottolese M, Federici G,et al. TAZ is required for metastatic activity and chemoresistanceof breast cancer stem cells. Oncogene 2015;34:681–90.

33. Wang L, Shi S, Guo Z, Zhang X, Han S, Yang A, et al. Overexpression ofYAP and TAZ is an independent predictor of prognosis in colorectalcancer and related to the proliferation and metastasis of colon cancercells. PLoS ONE 2013;8:e65539.

34. Mishra SK, SiddiqueHR, SaleemM. S100A4 calcium-binding protein is keyplayer in tumor progression and metastasis: preclinical and clinical evi-dence. Cancer Metastasis Rev 2012;31:163–72.

35. Dienstmann R, Serpico D, Rodon J, Saura C, Macarulla T, Elez E, et al.Molecular profiling of patients with colorectal cancer and matchedtargeted therapy in phase I clinical trials. Mol Cancer Ther 2012;11:2062–71.

36. Elkabets M, Vora S, Juric D, Morse N, Mino-KenudsonM, Muranen T, et al.mTORC1 inhibition is required for sensitivity to PI3K p110alpha inhibi-tors in PIK3CA-mutant breast cancer. Sci Transl Med 2013;5:196ra99.

37. Janku F,Wheler JJ,Westin SN,Moulder SL, Naing A, TsimberidouAM, et al.PI3K/AKT/mTOR inhibitors in patients with breast and gynecologicmalig-nancies harboring PIK3CA mutations. J Clin Oncol 2012;30:777–82.

38. Bertotti A, Migliardi G, Galimi F, Sassi F, Torti D, Isella C, et al. Amolecularly annotated platform of patient-derived xenografts ("xenopa-tients") identifies HER2 as an effective therapeutic target in cetuximab-resistant colorectal cancer. Cancer Discov 2011;1:508–23.

39. Julien S, Merino-Trigo A, Lacroix L, Pocard M, Goere D, Mariani P, et al.Characterization of a large panel of patient-derived tumor xenograftsrepresenting the clinical heterogeneity of human colorectal cancer. ClinCancer Res 2012;18:5314–28.

40. Lehtio L, Chi NW, Krauss S. Tankyrases as drug targets. FEBS J 2013;280:3576–93.

41. Riffell JL, Lord CJ, Ashworth A. Tankyrase-targeted therapeutics: expandingopportunities in the PARP family. Nat Rev Drug Discov 2012;11:923–36.

42. Parker JS, Mullins M, Cheang MC, Leung S, Voduc D, Vickery T, et al.Supervised risk predictor of breast cancer based on intrinsic subtypes. J ClinOncol 2009;27:1160–7.

43. Prat A, Adamo B, Fan C, Peg V, Vidal M, Galvan P, et al. Genomic analysesacross six cancer types identify basal-like breast cancer as a unique molec-ular entity. Sci Rep 2013;3:3544.

44. Van der Flier LG, Sabates-Bellver J, Oving I, Haegebarth A, De Palo M, AntiM, et al. The intestinal Wnt/TCF signature. Gastroenterology 2007;132:628–32.


Arqu�es et al.




2016;22:644-656. Published OnlineFirst July 29, 2015.Clin Cancer Res Oriol Arqués, Irene Chicote, Isabel Puig, et al. Colorectal CancerResistance to PI3K and AKT Inhibitors in the Treatment of

-Catenin Pathway and RevertsβTankyrase Inhibition Blocks Wnt/

Updated version

10.1158/1078-0432.CCR-14-3081doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2015/07/27/1078-0432.CCR-14-3081.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/22/3/644.full#ref-list-1

This article cites 41 articles, 13 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/22/3/644.full#related-urls

This article has been cited by 6 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/22/3/644To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-14-3081

http://clincancerres.aacrjournals.org/content/suppl/2015/07/27/1078-0432.CCR-14-3081.DC1

http://clincancerres.aacrjournals.org/content/22/3/644.full#ref-list-1

http://clincancerres.aacrjournals.org/content/22/3/644.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/22/3/644


Documents

Tankyrase Inhibition Blocks Wnt/b-Catenin Pathway and ...Daniel Silberschmidt5, Jordi Rodon2,6, Stefania Landolﬁ7, Aleix Prat8, Eloy Espín9, Ramon Charco10, Paolo Nuciforo11, Ana