Cancer Biology and Translational Studies

Phosphatidylinositol-3-kinase (PI3K)/AktSignaling is Functionally Essential in MyxoidLiposarcomaMarcel Trautmann1,2, Magdalene Cyra1,2, Ilka Isfort1,2, Birte Jeiler1,2, Arne Kr€uger1,2,Inga Gr€unewald1,2, Konrad Steinestel1,3, Bianca Altvater4, Claudia Rossig4,5,Susanne Hafner6, Thomas Simmet6, Jessica Becker7, Pierre Åman8, Eva Wardelmann1,Sebastian Huss1, and Wolfgang Hartmann1,2

Abstract

Myxoid liposarcoma (MLS) is an aggressive soft-tissue tumorcharacterized by a specific reciprocal t(12;16) translocationresulting in expression of the chimeric FUS–DDIT3 fusionprotein, an oncogenic transcription factor. Similar to othertranslocation-associated sarcomas, MLS is characterized by alow frequency of somaticmutations, albeit a subset of MLS haspreviously been shown to be associatedwith activating PIK3CAmutations. This studywas performed to assess the prevalence ofPI3K/Akt signaling alterations inMLSand thepotential ofPI3K-directed therapeutic concepts. In a large cohort of MLS, keycomponents of the PI3K/Akt signaling cascade were evaluatedby next generation seqeuncing (NGS), fluorescence in situhybridization (FISH), and immunohistochemistry (IHC). InthreeMLS cell lines, PI3Kactivitywas inhibitedbyRNAiand thesmall-moleculePI3K inhibitorBKM120(buparlisib) in vitro. An

MLS cell line–based avian chorioallantoic membrane modelwas applied for in vivo confirmation. In total, 26.8% of MLScases displayed activating alterations in PI3K/Akt signalingcomponents, with PIK3CA gain-of-function mutations repre-senting the most prevalent finding (14.2%). IHC suggestedPI3K/Akt activation in a far larger subgroup of MLS, implyingalternative mechanisms of pathway activation. PI3K-directedtherapeutic interference showed thatMLS cell proliferation andviability significantly depended on PI3K-mediated signals invitroand in vivo.Ourpreclinical studyunderlines the elementaryrole of PI3K/Akt signals in MLS tumorigenesis and provides amolecularly based rationale for a PI3K-targeted therapeuticapproach which may be particularly effective in the subgroupof tumors carrying activating genetic alterations in PI3K/Aktsignaling components.

IntroductionMyxoid liposarcoma (MLS) accounts for approximately

5%–10% of all soft-tissue sarcomas, representing about 20% ofall malignant adipocytic tumors (1). MLS constitutes the most

frequent liposarcomasubtype inpatients below theage of 20years.A high rate of local recurrence andparticular propensity to developdistantmetastases hasbeen reported in approximately 40%ofMLSpatients (2). MLS exhibit a morphologic spectrum ranging frommyxoid tumors with low cellularity to hypercellular, round-cellsarcomas associated with an aggressive clinical course (3). Genet-ically, 95% of MLS are characterized by a chromosomal t(12;16)(q13;p11) translocation, juxtaposing the FUS and DDIT3 genes.The remaining 5% display an alternative chromosomal t(12;22)(q13;q12) rearrangement leading to an EWSR1-DDIT3 genefusion (4). The resulting FUS-DDIT3 and EWSR1-DDIT3 fusionproteins have been shown to play an essential role in MLStumorigenesis, acting as pathogenic transcriptional (dys-)regula-tors (5–8). Current therapeutic approaches in high-grade MLScomplement radical surgery and adjuvant radiation and/or con-ventional chemotherapy based on anthracyclines and ifosfamide,recently supplemented by agents such as trabectedin or eribu-lin (9–11). AlthoughMLS displays a higher chemosensitivity thanother liposarcoma subtypes, the substantial frequency of recur-rence andmetastasis in MLS underlines the urgent need for novel,biology-guided therapeutic strategies. In principle, counteractingthe effects of the aberrant FUS–DDIT3 transcription factor repre-sents the most promising strategy to selectively target MLS tumorcells; however, the therapeutic interference with (chimeric) tran-scription factors in vivo represents a particular challenge.

In line with the situation in other translocation-associatedsoft-tissue and bone sarcomas driven by specific gene fusions,

1Gerhard-Domagk-Institute of Pathology, M€unster University Hospital, M€unster,Germany. 2Division of Translational Pathology, Gerhard-Domagk-Institute ofPathology, M€unster University Hospital, M€unster, Germany. 3Institute of Pathol-ogy and Molecular Pathology, Bundeswehrkrankenhaus Ulm, Ulm, Germany.4Department of Pediatric Hematology and Oncology, University Children'sHospital M€unster, M€unster, Germany. 5Cells in Motion Cluster of Excellence(EXC 1003 – CiM), University of M€unster, M€unster, Germany. 6Institute ofPharmacology of Natural Products & Clinical Pharmacology, Ulm University,Ulm, Germany. 7Institute of Human Genetics, School of Medicine & UniversityHospital Bonn, University of Bonn, Bonn, Germany. 8Sahlgrenska Cancer Center,University of Gothenburg, Gothenburg, Sweden.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

M. Trautmann and M. Cyra contributed equally to this article.

Corresponding Authors: Marcel Trautmann and Wolfgang Hartmann. Division ofTranslational Pathology, Gerhard-Domagk-Institute of Pathology, Münster Univer-sity Hospital, Domagkstr. 17, 48149, Germany. Phone: þ49 (0) 251-83-57623 and-58479, Fax:þ49 (0) 251-83-57559. E-mail:marcel.trautmann@ukmuenster.de andwolfgang.hartmann@ukmuenster.de

doi: 10.1158/1535-7163.MCT-18-0763

�2019 American Association for Cancer Research.

MolecularCancerTherapeutics

Mol Cancer Ther; 18(4) April 2019834

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

such as, Ewing sarcoma, synovial sarcoma, or alveolar rhabdo-myosarcoma (12–14), MLS are genomically "stable" tumors withfew mutations beyond the pathognomic gene fusion. However,two recent studies described activating gain-of-function muta-tions in the PIK3CA gene encoding the catalytic PI3K subunit in asubset of MLS. Demicco and colleagues provided evidence ofPTEN losses occurring as further structural alteration of crucialPI3K/Akt pathway components (12, 15). Beyond that, it has beenreported that EGFR-, MET-, VEGFR-, RET-, and PDGFRB signalingsustained by autocrine/paracrine transduction and receptor tyro-sine kinase (RTK) cross-talk may lead to an increase in PI3K-dependent signals (16, 17). ThePI3K/Akt cascade is an elementaryhub in the signal transduction of diverse RTK stimuli involvingdifferent growth-controlling regulators such as GSK-3b as well asthe cell-cycle effectors Cyclin D1 and p21WAF (18–21).

Overall, previously published results indicate a fundamentalrelevance of PI3K/Akt signals with respect to the specific liability ofMLS tumor cells, accomplished either by genetic PIK3CA and/orPTEN alterations or complex signaling networks (8, 12, 15–17, 22)resulting in aberrant activation of PI3K activity. This preclinicalstudywas conducted to explore the functional importance ofPI3K/Akt signals in MLS tumorigenesis, and to establish a molecularlytargeted strategy employing small-molecule PI3K inhibitors.

Materials and MethodsTumor specimens and tissue microarray

MLS tumor specimens were selected from the archive of theGerhard-Domagk-Institute of Pathology (M€unster UniversityHospital, M€unster, Germany). All diagnoses were reviewed bythree experienced pathologists (E. Wardelmann, S. Huss, W.Hartmann) based on morphologic criteria, DDIT3 break-apartFISH, or RT-PCR analysis in accordance with the current WHOClassification of Tumours of Soft Tissue and Bone (1). Evaluationof the round-cell component (3, 23) and tissuemicroarray (TMA)sampling/construction was performed as described previous-ly (8). In total, 56 MLS tumor specimens were included(24/42.9% female, 32/57.1% male). Median patient age at diag-nosis was 48 years (range 24–73 years of age). Median tumor sizewas 10 cm (range 1.5–29 cm). Clinicopathologic characteristicsof the cohort are summarized in Table 1 and SupplementaryTable S1. The study was approved by the Ethics Review Board ofthe University of Münster (2015-548-f-S), and experiments wereconformed to the principles set out in the World Medical Asso-ciationDeclaration of Helsinki and the United States DepartmentofHealth andHuman Services Belmont Report.Written informedconsent from patients was not requested by the Ethics ReviewBoard of the University of Münster (2015-548-f-S).

Immunohistochemistry (IHC)The following primary antibodies were applied: phospho-Akt

(S473, monoclonal rabbit, D9E, 1:50, catalog no. 4060), Cyclin D1(monoclonal rabbit, 92G2, 1:50, catalogno. 2978), phosphor-GSK-3a/b(S21/9, polyclonal rabbit, 1:50, catalogno. 9331), p21 (mono-clonal rabbit, 12D1, 1:1000, catalog no. 2947; all purchased fromCell Signaling Technology). IHC staining was performed with aBenchMark ULTRA Autostainer (VENTANA/Roche) on 3 mm tumorsections as described previously (8). IHC results were evaluated by asemiquantitative approach (H-score; ranging from 0 to 300) deter-mining the percentage of cells at each staining intensity level, usingthe following formula: [1� (% cells 1þ)þ 2� (% cells 2þ)þ 3�

(% cells 3þ)]. Immunoreactivity was assessed defining thestaining intensity (0, negative; 1þ, weak; 2þ, moderate; 3þ,strong) in the positive control (breast cancer; no special type) asstrong. TMA tissue cores with H-score >50 were considered pos-itive. The discriminatory threshold (positive ¼ semiquantitativeH-score >50) was prespecified without prior analyses of theclinical course. Loss of PTEN protein was analyzed by IHC asdescribed previously (8, 15). The IHC readers were blinded to theoutcome data.

Fluorescence in situ hybridization (FISH)PIK3CA gene amplification was examined by FISH analyses

using the SPEC PIK3CA/CEN 3 Dual Color Probe (ZytoVision).At least 40 cells of each MLS tumor specimen were assessed.PIK3CA gene amplification was defined by a PIK3CA/centromere3 (CEN3) ratio �2.

Next-generation sequencing/Sanger sequencingA customized GeneRead DNAseq Panel (Mix-n-Match V2, Qia-

gen) was applied to assess the complete exonic region of 23 genesthat are frequentlymutated in various cancer types (summarized inSupplementary Table S2). Target enrichment was performed bymeans of the GeneRead DNAseq PCR V2 Kit (Qiagen). All endrepair, A-addition, and DNA barcode ligation steps were processedapplying the GeneRead DNA Library I Core Kit (Qiagen). Ampli-ficationofadapter-ligatedDNAwasperformedusing theGeneReadDNA I Amp Kit (Qiagen). NGS was conducted with the IlluminaMiSeq system.TheQuantitativeMultiplexFFPEReferenceStandard(Horizon Discovery, catalog no. HD200) was applied as isogenicquality control (11 somatic mutations verified at 0.8%–24.5%allelic frequency in genomic DNA) for routine performancemonitoring and evaluation of NGS workflow integrity (preana-lytical DNA extraction, NGS workflow, and postanalytical

Table 1. Clinicopathological characteristics of MLS patients (n ¼ 56)

Age (years)Mean (�SD) 48.1 (�12.2)Median (range) 48 (24–73)<48 26 (46.4%)�48 30 (53.6%)

TypePrimary tumor 35 (62.5%)Metastasis 9 (16.1%)Recurrence 7 (12.5%)ND 5 (8.9%)

MorphologyMyxoid 35 (62.5%)Round cell 21 (37.5%)

Size (cm)Mean (�SD) 10.6 (�5.7)Median (range) 10.0 (1.5–29)<10 18 (32.1%)�10 25 (44.6%)ND 13 (23.2%)

SexFemale 24 (42.9%)Male 32 (57.1%)

FISHDDIT3 break apart positive 56 (100%)

t(12;16) translocation typeFUS–DDIT3 (type 1; exon 7-2) 11 (19.6%)FUS–DDIT3 (type 2; exon 5-2) 26 (46.4%)ND 19 (33.9%)

Abbreviation: ND, not determined.

PI3K/Akt Signaling in Myxoid Liposarcoma

www.aacrjournals.org Mol Cancer Ther; 18(4) April 2019 835

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

bioinformatics). NGS data were analyzed by means of the CLCBiomedical Genomics Workbench Software (CLC bio, Qiagen).Validation by Sanger sequencing was performed using the BigDyeTerminator v3.1 Cycle Sequencing Kit (Life Technologies).

In silico tools to predict the deleterious impact of detected genevariants

The functional context of NGS-detected variants in thecoding regions was predicted using following in silico tools:PolyPhen-2 (v2.2.2r398; ref. 24), Protein Variation EffectAnalyzer (PROVEAN; v1.1.3; ref. 25), Sorting Intolerant FromTolerant (SIFT; Ensembl 66; ref. 26), Mutation Assessor(release 3; ref. 27), and Combined Annotation DependentDepletion (CADD; v.1.3; ref. 28). In brief, the PolyPhen-2method utilizes evolutionary and physical comparative con-siderations to predict amino acid substitutions on proteinstructure and function. Scoring ranges from 0 (neutral) to 1(deleterious) and functional significance is categorized intobenign, possibly damaging, and probably damaging. The web-based PROVEAN algorithm classified each single-nucleotidevariant (SNV) as either neutral or deleterious (cutoff �2.5).SIFT was used with the default settings classifying variantsfrom 0 (damaging) to 1 (tolerated). CADD is a trained methodto differentiate 14.7 million high-frequency human-derivedalleles for objective integration of diverse annotations into asingle measure (C score). Scoring correlates with allelic diver-sity, annotations of pathogenicity, disease severity, experimen-tally measured regulatory effects, and complex trait associa-tions. Calculated C scores rank a pathogenic variant relative toall possible substitutions of the human genome. To increasethe prediction accuracy and the level of confidence, a combi-nation of in silico tools based on evolutionary information,protein structure, and functional parameters was used. Wecombined the individual output from �3 of 5 in silico predic-tion tools and defined a single-consensus outcome summa-rized in Supplementary Table S3.

Cell culture and cell linesThe human MLS cell lines MLS402-91 (FUS–DDIT3 exon 7-2;

type 1), MLS2645-94 (FUS–DDIT3 exon 5-2; type 2), andMLS1765-92 (FUS-DDIT3 exon 13-2; type 8) were describedbefore (29). For the purpose of cell line authentication, presenceof the MLS specific FUS–DDIT3 fusion gene transcripts wasconfirmed by RT-PCR and Sanger sequencing. All monolayer cellcultures were maintained in Roswell Park Memorial InstituteMedium 1640 (RPMI), supplemented with 10% FBS (Life Tech-nologies), and incubated under standard conditions (humidifiedatmosphere, 5%CO2, 37�C). Cells were passaged for amaximumof 20–30 culturing cycles and Mycoplasma testing was performedquarterly by standardized PCR. To study the biological effects oftreatment with the class I PI3K inhibitor BKM120 (buparlisib)(30, 31) and the prototypic pan-PI3K inhibitor LY294002 (32,33), MLS cells were cultured in medium supplemented with 2%FBS. For experimental PI3K modulation, MLS402-91 cells weretransfected with pCMV2-Tag 2A plasmids encoding either themutant DH1047R PIK3CA (Addgene, catalog no. 16639) or wild-type PIK3CA (Addgene, catalog no. 16643) coding sequence (34).

ImmunoblottingFollowing primary antibodies were used: Akt (monoclonal

rabbit, C67E7, 1:1,000, catalog no. 4691), phospho-Akt (S473,

monoclonal rabbit, D9E, 1:1,000, catalog no. 4060), Cyclin D1(monoclonal rabbit, 92G2, 1:1,000, catalog no. 2978), GSK-3b(monoclonal rabbit, 27C10, 1:1,000, catalog no. 9315), phos-pho-GSK-3b (S9, polyclonal rabbit, 1:1,000, catalog no. 9336),PI3K p110a (rabbit, 1:1,000, catalog no. 4255), p21 (mono-clonal rabbit, 12D1, 1:1,000, catalog no. 2947; all purchasedfrom Cell Signaling Technology), and b-actin (loading control;monoclonal mouse, AC15, 1:10.000, catalog no. A5441 Sigma-Aldrich). Secondary antibody labeling (Bio-Rad Laboratories)and immunoblot development was performed using anEnhanced Chemiluminescence Detection Kit (SignalFire ECLReagent; Cell Signaling Technology) and the Molecular ImagerChemiDoc System (Image Lab Software; Bio-Rad Laboratories).

Cell viability assay (MTT)MTT assays (Roche) were performed as described previous-

ly (35–37). In brief, 2.5 � 103 MLS402-91, MLS2645-94, andMLS1765-92 cells were seeded in 96-well cell culture plates (100mL of medium supplemented with 2% FBS) and exposed toincreasing concentrations of the class I PI3K inhibitor BKM120(buparlisib) (1.25-5 mmol/L) and the prototypic pan-PI3K inhib-itor LY294002 (3.125-50 mmol/L) for 72 hours. DMSO wasincluded as vehicle control.

In vivo efficacy of BKM120 and LY294002 in MLS cell linebased chick chorioallantoic membrane (CAM) studies

For in vivo confirmation, we used the chick chorioallantoicmembrane (CAM) model as previously reported and validatedfor anticancer agents (8, 38, 39). Because of the presence ofvascular supply and the absence of an immune response fromthe graft, the CAM enables the transplantation of human cancercells and the subsequent development of solid tumor xeno-grafts in a three-dimensional in vivo microenvironment. TheCAM model matches the 3R recommendations to reduce mam-malian animal experiments and is regarded as reproducible,reliable, and effective (40).

Seven days after fertilization, MLS402-91 and MLS1765-92cells (1–1.5 � 106 cells/egg; dissolved in medium/Matrigel1:1, v/v) were xenografted onto the CAM and incubated with60% relative humidity at 37�C. Topical treatment with the class IPI3K inhibitor BKM120 (buparlisib) (1–2 mmol/L), the proto-typic pan-PI3K inhibitor LY294002 (10 mmol/L), or vehiclecontrol (0.2% DMSO in NaCl 0.9%) was initiated on day 8 andrecapitulated for two consecutive days. Three days after treatmentinitiation, tumor-bearing CAM xenografts were imaged,explanted, and fixed (5% PFA). Tumor volume (mm3) wascalculated according to the formula: TV ¼ length (mm)� width2

(mm)� p/6. The in vivo studies were approved by an InstitutionalAnimal Care and Use Committee (84-02-04-2016-A195) andconducted in accordance with recognized standards of theNational and European Union guidelines.

CompoundsThe class I PI3K inhibitor BKM120 (buparlisib/NVP-BKM120;

C18H21F3N6O2; CAS#: 944396-07-0; refs. 30, 31) was purchasedfrom Selleck Chemicals and dissolved in DMSO (Sigma-Aldrich).The prototypic pan-PI3K inhibitor LY294002 (C19H17NO3;CAS#: 154447-36-6; refs. 32, 33) was purchased from Cell Sig-naling Technology and dissolved in DMSO. Final DMSO con-centration did not exceed 0.2% (v/v) for all in vitro and in vivoexperiments.

Trautmann et al.

Mol Cancer Ther; 18(4) April 2019 Molecular Cancer Therapeutics836

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

ResultsExpression of PI3K/Akt signaling components in human MLStumor specimens and MLS cell lines

To determine the involvement of PI3K/Akt signaling in MLStumorigenesis, expression of phospho-Akt (S473), phospho-GSK-3a/b (S21/9), p21, and Cyclin D1 was analyzed in a set of56 MLS tumor specimens by IHC (Fig. 1A and B; SupplementaryTable S1). On the basis of the calculated H-score, 62.0% of MLStumor specimens were positive for phosphorylated Akt (S473)

and 38.5% for phosphorylated GSK-3a/b (S21/9). No significantAkt and/or GSK-3b phosphorylation levels were detected in23.2% of cases. 32.7% of MLS specimens were positive forp21, while expression of Cyclin D1 was shown in 48.1%(Fig. 1A and B). Loss of PTEN protein expression was demon-strated in 5 of 56 (8.9%) MLS tumor specimens (SupplementaryTable S1). Phosphorylation and expression levels of PI3K/Aktsignaling components did not correlate with tumor size, FUS-DDIT3 translocation subtype, patients' age, and/or gender. Inaccordance with the IHC results inMLS tumor specimens, protein

Figure 1.

Activation of PI3K/Akt signaling in a comprehensive cohort of MLS tumor specimens (n¼ 56) and MLS cell lines. A, IHC staining shows strong expression ofphosphorylated of Akt (S473) and GSK-3a/b (S21/9) as well as increased expression levels of p21 and Cyclin D1 in a representative case of MLS (originalmagnification:�20, inset�40). B, IHC spectrum of MLS tumor specimens summarized as positive IHC reactivity (H-score >50). C, Immunoblotting resultsdemonstrate expression and phosphorylation levels of PI3K/Akt signaling components in total protein extracts of three different MLS cell lines. Detection of FUS-DDIT3 fusion gene transcripts in MLS402-91 (type 1; exon 7-2), MLS2645-94 (type 2; exon 5-2), and MLS1765-92 (type 8; exon 13-2) cells by RT-PCR (28S rRNAemployed as loading reference; bottom).

PI3K/Akt Signaling in Myxoid Liposarcoma

www.aacrjournals.org Mol Cancer Ther; 18(4) April 2019 837

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

expression (PI3K p110a, p21, and Cyclin D1) and phosphory-lation levels (phospho-Akt S473 and phospho-GSK-3b S9) weredemonstrated in total protein extracts of MLS402-91 (FUS–DDIT3 type 1), MLS2645-94 (FUS–DDIT3 type 2), andMLS1765-92 (FUS-DDIT3 type 8) cells (Fig. 1C, top). Expressionof the specific FUS–DDIT3 fusion transcript was confirmed in allthree MLS cell lines by RT-PCR (Fig. 1C, bottom). These findingsprovide evidence that activation of PI3K/Akt signaling is a com-mon feature of MLS.

Mutational profiling of MLSAs pathogenic PIK3CA mutations can be responsible for con-

stitutive activation of the PI3K/Akt signaling cascade, we exam-ined the entirePIK3CA coding regionby targetedNGS followedbySanger sequencing for validation purposes. In 12 of 56 (21.4%)MLS cases, genetic alterations in PIK3CA (most frequentlyH1047R) could be detected (Fig. 2A; Supplementary Table S1).Overall, 55.6% (5/9) of the observed variants in the PIK3CAcoding region (N345K, C378Y, C901F, H1047R, and N1068fs)were predicted to have a deleterious impact by �3 independentin silico tools (summarized in Fig. 2B and SupplementaryTable S3). No PIK3CA gene mutations were identified in allthree MLS cell lines. PIK3CA gene amplification was examinedby FISH and could be excluded for all analyzed MLS cases. Inaddition, the coding sequence of AKT1 and PTEN was analyzed.Few deleterious variants were identified in both genes: AKT1(E17K and W80R) and PTEN (V119F and I253fs). Notably,round-cell MLS (3/6; 50.0%) were more likely than myxoid MLStumors (5/13; 38.5%) to harbor a deleterious mutation in thePIK3CA gene. In contrast, all deleterious mutations in AKT1(E17K andW80R) and PTEN (V119F and I253fs) were exclusivelyidentified in myxoid tumors. In Kaplan–Meier correlations, MLSpatientswith round-cellmorphology (Fig. 2C, left; ��,P¼0.0014)and activating alterations in PI3K/Akt signaling components(Fig. 2C, right; ��, P ¼ 0.0071) displayed a significantly reducedoverall/event–free survival. In addition, all coding exons of 20further cancer-associated genes were analyzed with another 11actionable nonsynonymous mutations detected in 10 MLS cases(summarized in Supplementary Tables S2 and S3): CTNNB1(P321H), ERBB2 (R138W), FGFR3 (T311M), GNAQ (M59L),GNAS (P533R), and TP53 (V272G; allelic frequency: 10.5%–

90.3%). In summary, 28 nonsynonymous somatic variants wereidentified in 25 of 56 (44.6%)MLS tumor specimens. The overallconcordance between all 5 in silicoprediction toolswas 52.2%andindependent from the deleterious/damaging or neutral/toleratedstatus. At least 4 of 5 in silico predictions were in agreement in34.8%of variants and inonly 13.0%of variants thepredictionwasbased on 3 of 5 in silico tools (summarized in SupplementaryTable S3). Correlating the (global) presence of gene mutationswithMLS tumormorphology, no significant trend concerning themutational burden could be determined comparing the roundcell to the myxoid MLS subtype, which has been previouslyestablished as a prognostically relevant histologic parameter (1,3, 23). In total, gene mutations were detected in 9 of 21(42.8%) round-cell MLS versus 16 of 35 (45.7%) myxoidtumors. Overall, 85.7% of MLS patients with alterations inPIK3CA, AKT1, or PTEN were positive for phosphorylated Akt(S473) and/or GSK-3a/b (S21/9) as examined by IHC and thusmay be regarded as indicators of an activation of PI3K/Aktsignaling. In the subset of patients with PIK3CA/AKT1/PTENwild-type MLS, the fraction of tumors immunohistochemically

displaying either phosphorylation of Akt and/or GSK-3b wassubstantially lower (69.4%).

RNA interference–mediated depletion of PIK3CA affectsPI3K/Akt signaling activity and cell viability in MLS cells

To confirm the biological importance and to analyze thefunctional role of PI3K/Akt signaling by a nonpharmacologicapproach, three MLS cell lines (MLS402-91, MLS2645-94, andMLS1765-92) were transfected with two different prevalidatedsiRNAs (to exclude unspecific off-target effects) directedagainst human PIK3CA. Consistently, specific knockdown ofPIK3CA led to reduced phosphorylation levels of Akt (S473)and GSK-3b (S9), combined with diminished Cyclin D1downstream target expression (Fig. 3A). In MTT assays, allMLS cell lines displayed a significant reduction of cellviability (���, P < 0.001) in comparison with nontargetingcontrol siRNA (Fig. 3B). These results imply that PI3K/Akt–mediated signals may be of major functional relevance in thepathogenic context of MLS.

BKM120 reduces MLS cell viability in vitroOur results suggested that aberrant PI3K/Akt signaling might

be a therapeutic target in MLS. We therefore evaluated thebiological effects of pharmacologic PI3K inhibition. In MTTassays, three MLS cell lines were exposed to increasing concen-trations (1.25–5 mmol/L) of the class I PI3K inhibitor BKM120(buparlisib). Cell viability and growth were significantly sup-pressed in all three MLS cell lines with IC50 values ranging from1.01 to 1.84 mmol/L (BKM120, buparlisib), indicating a dose-dependent mode of action (Fig. 3C; Table 2). Analogue experi-ments with the prototypic pan-PI3K inhibitor LY294002 (3.125–50 mmol/L) showed comparable effects with IC50 values rangingfrom 9.57 to 12.02 mmol/L (Supplementary Fig. S1A)

BKM120 inhibits PI3K/Akt signal transduction activity in vitroTo examine the functional basis of PI3K/Akt suppression, we

analyzed the specific effects of treatment with the class I PI3Kinhibitor BKM120 (buparlisib) on the activation status of PI3K/Akt signaling components by immunoblotting. The growth-inhibitory effects were associated with a significant dose-depen-dent reduction of phosphorylation levels for Akt (S473) andGSK-3b (S9) in all three MLS cell lines (Fig. 3D). Analogue effects wereobserved with the prototypic pan-PI3K inhibitor LY294002 (Sup-plementary Fig. S1B).

BKM120 reduces MLS cell viability by induction of apoptosisPerforming flow cytometric analyses, poly-adenosine diphos-

phate (ADP)-ribose polymerase (cPARP; Asp214) cleavage wasemployed as a marker of apoptosis. After treatment with increas-ing concentrations of BKM120 (buparlisib) (1–2 mmol/L; 72hours), all three MLS cell lines showed a significantly anddose-dependently increased rate of apoptosis (Fig. 3E and F).

In vivo efficacy of PI3K inhibition in CAM models of MLSAntitumor efficacy of the class I PI3K inhibitor BKM120 (bupar-

lisib) on tumor growth was tested in a previously established invivo model of human MLS (8). We xenografted MLS402-91 andMLS1765-92 cells onto chick CAMs to initiate MLS tumor for-mation (n� 5 eggs/group). A significant reduction of MLS tumorvolume through topical administration of BKM120 (buparlisib)(1–2 mmol/L) compared with the DMSO vehicle-treated control

Trautmann et al.

Mol Cancer Ther; 18(4) April 2019 Molecular Cancer Therapeutics838

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

Figure 2.

Mutational profile of MLS. A, Detected variants in the coding regions of 23 cancer-associated genes. Mutations in different genes (rows) are indicated foreach MLS case (columns). In silico predictions are illustrated as: green (neutral/tolerated) or red (deleterious/damaging). Clinicopathologic information issummarized according to the legend. B,Distribution of detected PIK3CAmutations annotated in the PI3K p110 catalytic subunit alpha protein domains (red:deleterious/damaging). C, In Kaplan–Meier correlations, overall/event–free survival is significantly reduced in MLS tumors with a round-cell content >5%compared with the myxoid subtype (�� , P¼ 0.0014, left) and in tumors carrying activating alterations in PI3K/Akt signaling components (�� , P¼ 0.0071, right).

PI3K/Akt Signaling in Myxoid Liposarcoma

www.aacrjournals.org Mol Cancer Ther; 18(4) April 2019 839

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

Figure 3.

In vitro and in vivo evaluation of PI3K suppression (BKM120, buparlisib) in MLS cell lines. A, In three MLS cell lines, RNAi-mediated knockdown of PIK3CA reducedphosphorylation levels of Akt (S473) and GSK-3b (S9). Expression of p21 was induced and Cyclin D1 downstream target expression diminished. Efficient RNAi-mediated depletion was confirmed by PI3K p110a immunoblotting for two different siRNAmolecules. B, Significant reduction of cell viability (MTT assay) uponRNAi-mediated depletion of PIK3CA in two MLS cell lines (��� , P < 0.001). To exclude unspecific off-target effects, a second siRNA was tested in parallel.Experiments were performed in quintuplicates and results are expressed as meanþ SEM. C, In MTT assays, viability of MLS402-91, MLS2645-94, and MLS1765-92cells was significantly reduced by treatment with increasing concentrations of the class I PI3K inhibitor BKM120 (buparlisib) (1.25-5 mmol/L). D, BKM120(buparlisib) (0.5–1 mmol/L) suppressed phosphorylation levels of Akt (S473) and GSK-3b (S9) in all three MLS cell lines. Changes in the PI3K p110 catalytic subunitalpha protein levels were determined by immunoblotting. E and F, In flow cytometric analyses, significantly increased rates of apoptosis (cleaved PARP) weredetected upon treatment with BKM120 (buparlisib) (1-2 mmol/L; DMSO employed as vehicle control). G,MLS402-91 and MLS1765-92 cells were xenografted onthe CAM of fertilized chick eggs. Tumor-bearing eggs were randomized and topically treated with BKM120 (buparlisib) (n� 5) or vehicle control (0.2% DMSO inNaCl 0.9%; n� 5). Significantly reduced tumor volumesþ SEM in BKM120 (buparlisib)-treated (1-2 mmol/L) versus the DMSO vehicle control group andrepresentative CAM explants of MLS xenografts are shown (scale bar: 1 mm, ��, P < 0.01; � , P < 0.05).

Trautmann et al.

Mol Cancer Ther; 18(4) April 2019 Molecular Cancer Therapeutics840

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

group (Fig. 3G)was demonstrated for two differentMLS cell lines.Analogue experiments with the prototypic pan-PI3K inhibitorLY294002 (10 mmol/L) resulted in comparable effects (Supple-mentary Fig. S1C). Collectively, these results showed that inhi-bition of the PI3K/Akt signaling cascade impairs themaintenanceof MLS tumors in vivo, further supporting the concept that PI3K/Akt pathway activation represents a novel, molecularly basedtarget for therapeutic interventions in patients with advancedhigh-grade MLS.

Mutational PIK3CA status in MLS cells alters the apoptoticeffects in response to BKM120 treatment

To assess the biological effects ofDH1047R-mutated PIK3CA inresponse to BKM120 (buparlisib) treatment, MLS402-91 cellstransfected with pCMV2-Tag 2A plasmid DNA encoding differentPI3K p110 catalytic subunit alpha variants were exposed toBKM120 (buparlisib) (2 mmol/L; DMSOwas employed as vehiclecontrol). Significantly increased rates of apoptosis were detectedapplying two different markers of apoptosis: (i) caspase 3/7activity (Fig. 4A) and (ii) PARP cleavage (cPARP; Asp214;Fig. 4B). Consistently, MLS402-91 cells expressing DH1047R-mutated PIK3CA showed increased rates of apoptosis upon treat-ment with BKM120 (buparlisib) compared with cells expressingwild-type PIK3CA.

DiscussionMLS are malignant tumors of lipogenic differentiation with a

high rate of local recurrence and particular propensity to developdistant metastases. High histologic grade, defined as round-cellcomponent >5% is the primary predictor of an unfavorableclinical outcome (3, 23). Although there is an established rolefor conventional cytotoxic and radiation-based therapies inMLS (10), molecularly targeted therapies are not implementedso far. Almost all MLS are characterized by a FUS-DDIT3 genefusion, encoding an oncogenic transcription factor with thepotential to transform mesenchymal stem cells to form MLS inmice (5). As in other sarcomas driven by specific gene fusions,such as, Ewing sarcoma, synovial sarcoma, or alveolar rhabdo-myosarcoma (12–14), MLS are genomically "stable" tumors witha low mutational burden. However, two recent studies describedrecurrent PIK3CA gene mutations in a subset of 14%–18%MLS cases beyond the pathognomonic FUS–DDIT3 genefusion (12, 15). In addition, PTEN loss and IGF-IR overexpressionhave been described, which both represent alternative mechan-isms of PI3K/Akt signal transduction activation (8, 15, 41).Beyond that, it has been reported that EGFR-, MET-, VEGFR-,RET-, and PDGFRB signaling sustained by autocrine/paracrinetransduction and RTK cross-talk involving the extensive tumorvasculature ofMLSmay lead to an additional increase in PI3K/Akt

signaling pathway activity (16, 17). Both, IGF-IR overexpressionandPIK3CAmutations appear tobeoverrepresented in the round-cell variant of MLS (15, 41). Given the fact that (i) round cellphenotype, (ii) presence of an activating PIK3CA mutation, and(iii) IGF-IR overexpression are associated with an unfavorableclinical course, overactivation of PI3K/Akt signalsmay represent acrucial factor for amore aggressive biology of MLS (3, 12, 15, 41).Unfortunately, the limited size of our MLS tumor set and theretrospective nature of our analysis prevented a definitive evalu-ationof the contribution of eachof these factors toMLSprognosis;however, this interesting issuemight beworth to be systematicallyanalyzed in future prospective trials.

We previously provided substantial evidence that therapeuticinterference with a cell-autonomous stimulation loop involvingIGF-IR-dependent signals may represent a highly effectiveapproach associatedwith therapeutic benefit, which is specificallyfound in the tumor's biology involving FUS-DDIT3-mediatedoverexpression of IGF-II (8). However, this therapeutic strategyapplies only to those tumors in which PI3K/Akt signaling is notactivated through mutations of central components as PIK3CA,AKT1, and/or PTEN. We therefore hypothesized that centralpathway interference through PI3K inhibition might constitutea rational therapeutic alternative since it covers differentmodes ofpathogenic PI3K/Akt activation.

Our genomic study, identifying activating gain-of-functionPIK3CA mutations in 14.2% of MLS concordantly confirmspublished results (12, 15), while genomic amplifications ofPIK3CA could be excluded. In line with data presented by Barre-tina and colleagues (12), we identified PIK3CA mutations to beassociated with a negative prognostic impact. Notably, round-cellMLS (50.0%)weremore likely thanmyxoidMLS tumors (38.5%)to harbor a deleterious mutation in the PIK3CA gene. In addition,we detected deleterious AKT1 (E17K and W80R) and PTEN(V119F and I253fs) alterations, which were bioinformaticallypredicted to be of functional relevance; these were exclusivelydetected in myxoid tumors. In agreement with Demicco andcolleagues (15), we found 8.9% of the analyzed cases to benegative for PTEN, pointing either to a genetic loss or epigeneticsilencing through promotor hypermethylation (8, 42, 43). Alldeleterious mutations found in PIK3CA, AKT1, and PTEN weremutually exclusive, resulting in a total fraction of 26.8% of MLStumors with evidence of an activated PI3K/Akt signaling pathwaycaused by genetic/epigenetic alterations. The few additionalmutations in classic oncogenic drivers detected in our genomicscreen were randomly spread among all MLS cases and appearto be secondary passenger alterations rather than crucial onco-genic drivers in MLS tumorigenesis. Given further kinase sig-naling input in MLS, for example, via the IGF-IR system as wellas the effector loop discussed by Negri and colleagues (8, 17), itis not surprising that the fraction of tumors displaying eitherphosphorylation of Akt and/or GSK-3b (both regarded asindicators of an activation of PI3K/Akt signaling) in IHCanalysis was substantially higher than 26.8%. With regard tothe prognostic impact of PIK3CA mutations as shown in ourstudy and the previous works published by Barretina andDemicco (12, 15) it appears improbable that these alterationsrepresent mere passenger mutations; however, it remains to beelucidated in which way aberrant PI3K signaling contributes toMLS pathogenesis, which essentially depends on the FUS–DDIT3 gene fusion. Given our findings, it appeared rationalto extend the study to preclinically evaluate the therapeutic

Table 2. IC50 values for the class I PI3K inhibitor BKM120 and the prototypicpan-PI3K inhibitor LY294002 analyzed in three MLS cell lines

IC50 (mmol/L)Compound MLS402-91 MLS2645-94 MLS1765-92

BKM120 (buparlisib) 1.46 � 0.08 1.01 � 0.13 1.84 � 0.12LY294002 9.57 � 0.35 12.02 � 1.86 11.96 � 1.72

NOTE: Cytotoxic effects on MLS (MLS402-91, MLS2645-94, and MLS1765-92)cell viability were assessed in MTT assays (72 hours). Results are represented asmean � SEM of at least three independent experiments performed inquintuplicates.

PI3K/Akt Signaling in Myxoid Liposarcoma

www.aacrjournals.org Mol Cancer Ther; 18(4) April 2019 841

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

potential of a PI3K/Akt–directed approach, which up to nowhas not been assessed in the context of MLS.

Considering the therapeutic accessibility of PI3K/Akt signaling,PI3K enzymes themselves represent the central hub of the signal-ing cascade as they serve as integrators and (enhancing) distri-butors of diverse signaling cues, and the development ofinhibitory molecules is certainly most advanced for PI3K astherapeutic target. Combining pharmacologic studies and RNAinterference (RNAi)-based approaches, our study convincinglyshows in vitro that proliferation and viability of MLS cells signif-icantly depend on PI3K-mediated signals. Importantly, we wereable to show that expression of the DH1047R-mutated PIK3CAvariant inMLS402-91 was associated with an increased sensitivityto BKM120 (buparlisib). Consistent with our in vitro findings,administration of BKM120 (buparlisib) and LY294002 to xeno-grafted MLS402-91 or MLS1765-92 cells led to a significantsuppression ofMLS tumor growth in vivo. Although these findingsare consistent and imply a potential therapeutic role of PI3Kinhibitory approaches in MLS, we are aware of the limitations ofthe employed avianCAMmodel as an in vivo tool in the preclinicaltransfer. However, apart from a published patient-derived xeno-graft (PDX) model which was not available to us (44), establish-ment and propagation of cell line–based MLS xenotransplantshave not been successful to our best knowledge, which led to ourchoice of the (artificial) CAM system. It would, of course, berational to try to transfer our findings to a mammalian in vivosystem and to elaborate on combinations of diverse therapeuticagents including PI3K inhibitors. The results presented by Qi andcolleagues showing efficacy of the PI3K inhibitor PF-04691502 ina PIK3CA-mutatedMLS PDXmodel (44) support the relevance ofour findings.

Given its particular dependence on kinase signals, MLS some-how resembles a genetically different mesenchymal neoplasia asgastrointestinal stromal tumors or dermatofibrosarcoma protu-

berans. These tumors are either driven by activating mutations inthe receptor tyrosine kinases KIT/PDGFRA or by translocation-dependent overexpression of the PDGFR ligand PDGFB (45, 46).However, PI3K-directed therapeutic approaches were not estab-lished in these tumors as effective inhibition of the pathogenicsignaling pathway could be achieved by interference with theactivated KIT and PDGFR RTKs employing the tyrosine kinaseinhibitor imatinib, resulting in significant clinical benefit (46–48). In the context of MLS, the situation is more complex, as theprimary tumor-driving oncogene is the chimeric FUS-DDIT3transcription factor, and only a subset of about 30% MLS casesis characterized by additional activating alterations in centralcomponents of the PI3K/Akt signaling pathway. However, givenour previously reported finding of a FUS–DDIT3–driven cell-autonomous stimulation of MLS cells involving an IGF-IR/-(PI3K/Akt) transactivation loop, a much larger fraction of MLSappear to in fact represent a PI3K/Akt–dependent neoplasia.Given the diversity of activation modes of PI3K/Akt signal trans-duction during MLS tumorigenesis and regarding the fact thatIGF-IR–driven tumors involve signaling cascades in parallelto PI3K/Akt with higher probability, the identification of appro-priate predictive biomarkers once more becomes the crucialissue in the establishment of molecularly targeted therapeuticapproaches. InMLS, PIK3CA, AKT1, and PTENmutational testingas well as PTEN IHC/FISHmay help to identify those patients thatmost probably may clinically benefit from a PI3K-directed ther-apeutic approach. In patients without alterations in these com-ponents a (combined) IGF-IR–directed therapeutic approachmight be more beneficial as documented IGF-IR-driven signalsalso involve PI3K/Akt–independent intracellular signaling path-ways as the MEK/ERK cascade (8, 41). Despite promising pre-clinical data, clinical trials testing diverse PI3K inhibitoryapproaches in different solid tumors showed a limited efficacyof monotherapy inhibition (49, 50). This may be due to

Figure 4.

In vitro evaluation of response to BKM120 (buparlisib) depending on PIK3CAmutational status in MLS cells. A, Apoptotic rates of MLS402-91 cells expressingwild-type or DH1047R-mutated PIK3CA after treatment with the class I PI3K inhibitor BKM120 (buparlisib) for 24 hours. In ApoTox-Glo analyses, significantlyincreased rates of apoptosis (caspase 3/7 activity; ��� , P < 0.001) were detected upon treatment with BKM120 (buparlisib). B, Employing cleaved PARP (cPARP)as a second marker for apoptosis in flow cytometric analyses, the increase in apoptotic rate upon treatment with BKM120 (buparlisib) was more marked inMLS402-91 cells transfected with DH1047R-mutated PIK3CA compared with the wild-type PIK3CA control.

Trautmann et al.

Mol Cancer Ther; 18(4) April 2019 Molecular Cancer Therapeutics842

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

insufficient target inhibition, which may be increased by opti-mized small-molecule structures, but also due to the fact that inmany solid tumors, activating mutations in components of thePI3K/Akt signaling cascade occur as one genetic alterationamong many others, entering the discussion on driver and pas-senger mutations. In MLS, the situation differs from the vastmajority of cancers in so far as activating mutations in compo-nents of the PI3K/Akt signaling cascade appear to represent theoutstanding alteration apart from the FUS-DDIT3 gene fusionagainst the background of genetic stability. This characteristicfeature of MLS underlines the potential of PI3K inhibitoryapproaches and should promote systematic biomarker-guidedclinical trials.

In conclusion, the results of this study imply that (apart fromthe pathognomonic FUS-DDIT3 gene fusion) activation of PI3K/Akt signaling is an essential pattern inMLS tumorigenesis which isrealized, at least in part by genetic alterations in PIK3CA, AKT1, orPTEN. Interference with PI3K-mediated signals via small-mole-cule compounds is an effective therapeutic strategy for advancedhigh-grade MLS that could be exploited for clinical benefit. Thecurrent preclinical testing of a PI3K-targeted therapeutic approachindicates potent effects both in vitro and in vivo, qualifying thePI3K/Akt signaling pathway as a molecularly based target in MLScancer therapy.

Disclosure of Potential Conflicts of InterestK. Steinestel has received speakers bureau honoraria from Boehringer Ingel-

heim and is a consultant/advisory board member for Novartis. E. Wardelmannhas provided expert testimony for Novartis Oncology, Milestone, Menarini,PharmaMar, Roche, Nanobiotix, Bayer, and Lilly. No potential conflicts ofinterest were disclosed by the other authors.

Authors' ContributionsConception and design: M. Trautmann, S. Huss, W. HartmannDevelopment of methodology: M. Trautmann, M. Cyra, I. Isfort, B. Jeiler,A. Kr€uger, B. Altvater, W. HartmannAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): I. Gr€unewald, B. Altvater, C. Rossig, S. Hafner,E. Wardelmann, S. Huss, W. HartmannAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): M. Trautmann, M. Cyra, I. Isfort, B. Jeiler, A. Kr€uger,K. Steinestel, B. Altvater, S. Hafner, J. Becker, S. Huss, W. HartmannWriting, review, and/or revision of the manuscript: M. Trautmann, M. Cyra,I. Isfort, I. Gr€unewald, K. Steinestel, B. Altvater, C. Rossig, S. Hafner, J. Becker,P. Åman, E. Wardelmann, S. Huss, W. HartmannAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): K. Steinestel, P. Åman, E. Wardelmann,W. HartmannStudy supervision: M. Trautmann, W. HartmannOthers (designed and supervised CAM experiments): T. Simmet

AcknowledgmentsThe authors thank Charlotte Sohlbach, Inka Buchroth and Christian Bertling

for excellent technical support. The study was supported by grants from theDeutsche Forschungsgemeinschaft (DFG,HA4441/2-1, toW.Hartmann andM.Trautmann;), the Wilhelm-Sander-Stiftung (2016.099.1, to W. Hartmann, M.Trautmann, and K. Steinestel;), and the "Innovative Medical Research" fundingprogram of the University of M€unster Medical School (IMF; TR121716 and I-TR221611, to M. Trautmann; I-HU121421, to M. Trautmann and S. Huss).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received July 10, 2018; revised December 13, 2018; accepted January 28,2019; published first February 20, 2019.

References1. Antonescu CR, Ladanyi M. WHO Classification of Tumours of Soft Tissue

and Bone. Fourth edition: Lyon, France: IARC Press; 2013.2. Dei Tos AP. Liposarcomas: diagnostic pitfalls and new insights. Histopa-

thology 2014;64:38–52.3. Antonescu CR, Tschernyavsky SJ, Decuseara R, Leung DH, Woodruff JM,

Brennan MF, et al. Prognostic impact of P53 status, TLS-CHOP fusiontranscript structure, and histological grade in myxoid liposarcoma: amolecular and clinicopathologic study of 82 cases. Clin Cancer Res2001;7:3977–87.

4. Panagopoulos I, HoglundM, Mertens F, Mandahl N, Mitelman F, Aman P.Fusion of the EWS and CHOP genes in myxoid liposarcoma. Oncogene1996;12:489–94.

5. Kuroda M, Ishida T, Takanashi M, Satoh M, Machinami R, Watanabe T.Oncogenic transformation and inhibition of adipocytic conversion ofpreadipocytes by TLS/FUS-CHOP type II chimeric protein. Am J Pathol1997;151:735–44.

6. Perez-Losada J, Pintado B, Gutierrez-Adan A, Flores T, Banares-Gonzalez B,del Campo JC, et al. The chimeric FUS/TLS-CHOP fusion protein specif-ically induces liposarcomas in transgenic mice. Oncogene 2000;19:2413–22.

7. Riggi N, Cironi L, Provero P, SuvaML, Stehle JC, Baumer K, et al. Expressionof the FUS-CHOP fusion protein in primarymesenchymal progenitor cellsgives rise to amodel ofmyxoid liposarcoma.Cancer Res 2006;66:7016–23.

8. TrautmannM,Menzel J, Bertling C, CyraM, Isfort I, Steinestel K, et al. FUS-DDIT3 fusion protein-driven IGF-IR signaling is a therapeutic target inmyxoid liposarcoma. Clin Cancer Res 2017;23:6227–38.

9. Grosso F, Jones RL, Demetri GD, Judson IR, Blay JY, Le Cesne A, et al.Efficacy of trabectedin (ecteinascidin-743) in advanced pretreated myxoidliposarcomas: a retrospective study. Lancet Oncol 2007;8:595–602.

10. Ratan R, Patel SR. Chemotherapy for soft tissue sarcoma. Cancer 2016;122:2952–60.

11. Schoffski P, Chawla S, Maki RG, Italiano A, Gelderblom H, Choy E, et al.Eribulin versus dacarbazine in previously treated patients with advancedliposarcoma or leiomyosarcoma: a randomised, open-label, multicentre,phase 3 trial. Lancet 2016;387:1629–37.

12. Barretina J, Taylor BS, Banerji S, Ramos AH, Lagos-Quintana M, DecarolisPL, et al. Subtype-specific genomic alterations define new targets for soft-tissue sarcoma therapy. Nat Genet 2010;42:715–21.

13. Crompton BD, Stewart C, Taylor-Weiner A, Alexe G, Kurek KC, CalicchioML, et al. The genomic landscape of pediatric Ewing sarcoma.Cancer Discov 2014;4:1326–41.

14. Shern JF, Chen L, Chmielecki J, Wei JS, Patidar R, Rosenberg M, et al.Comprehensive genomic analysis of rhabdomyosarcoma reveals a land-scape of alterations affecting a common genetic axis in fusion-positive andfusion-negative tumors. Cancer Discov 2014;4:216–31.

15. Demicco EG, Torres KE, Ghadimi MP, Colombo C, Bolshakov S, HoffmanA, et al. Involvement of the PI3K/Akt pathway in myxoid/round cellliposarcoma. Mod Pathol 2012;25:212–21.

16. Andersson MK, Goransson M, Olofsson A, Andersson C, Aman P. Nuclearexpression of FLT1 and its ligand PGF in FUS-DDIT3 carrying myxoidliposarcomas suggests the existence of an intracrine signaling loop.BMC Cancer 2010;10:249.

17. Negri T, Virdis E, Brich S, Bozzi F, Tamborini E, Tarantino E, et al.Functional mapping of receptor tyrosine kinases in myxoid liposarcoma.Clin Cancer Res 2010;16:3581–93.

18. Sherr CJ, Roberts JM. Inhibitors of mammalian G1 cyclin-dependentkinases. Genes Dev 1995;9:1149–63.

19. Diehl JA, Cheng M, Roussel MF, Sherr CJ. Glycogen synthase kinase-3betaregulates cyclin D1 proteolysis and subcellular localization. Genes Dev1998;12:3499–511.

20. Vivanco I, Sawyers CL. The phosphatidylinositol 3-Kinase AKT pathway inhuman cancer. Nat Rev Cancer 2002;2:489–501.

PI3K/Akt Signaling in Myxoid Liposarcoma

www.aacrjournals.org Mol Cancer Ther; 18(4) April 2019 843

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

21. Osaki M, Oshimura M, Ito H. PI3K-Akt pathway: its functions and altera-tions in human cancer. Apoptosis 2004;9:667–76.

22. Guo S, Lopez-Marquez H, Fan KC, Choy E, Cote G, Harmon D, et al.Synergistic effects of targeted PI3K signaling inhibition and chemotherapyin liposarcoma. PLoS One 2014;9:e93996.

23. Smith TA, Easley KA, Goldblum JR. Myxoid/round cell liposarcomaof the extremities. A clinicopathologic study of 29 cases with particularattention to extent of round cell liposarcoma. Am J Surg Pathol 1996;20:171–80.

24. Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P,et al. A method and server for predicting damaging missense mutations.Nat Methods 2010;7:248–9.

25. Choi Y, Sims GE, Murphy S, Miller JR, Chan AP. Predicting the func-tional effect of amino acid substitutions and indels. PLoS One 2012;7:e46688.

26. Kumar P, Henikoff S, Ng PC. Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm.Nat Protoc 2009;4:1073–81.

27. Reva B, Antipin Y, Sander C. Determinants of protein function revealed bycombinatorial entropy optimization. Genome Biol 2007;8:R232.

28. KircherM,WittenDM, Jain P,O'roak BJ, Cooper GM, Shendure J. A generalframework for estimating the relative pathogenicity of human geneticvariants. Nat Genet 2014;46:310–5.

29. Aman P, Ron D, Mandahl N, Fioretos T, Heim S, Arheden K, et al.Rearrangement of the transcription factor gene CHOP in myxoid liposar-comas with t(12;16)(q13;p11). Genes Chromosomes Cancer 1992;5:278–85.

30. 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 2011;30:282–90.

31. Maira S-M, Pecchi S, Huang A, Burger M, Knapp M, Sterker D, et al.Identification and characterization of NVP-BKM120, an orally availablepan-class I PI3-kinase inhibitor. Mol Cancer Ther 2012;11:317–28.

32. Vlahos CJ, Matter WF, Hui KY, Brown RF. A specific inhibitor of phospha-tidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002). J Biol Chem 1994;269:5241–8.

33. Friedrichs N, Trautmann M, Endl E, Sievers E, Kindler D, Wurst P, et al.Phosphatidylinositol30kinase/AKT signaling is essential in synovial sarco-ma. Int J Cancer 2011;129:1564–75.

34. Samuels Y, Wang Z, Bardelli A, Silliman N, Ptak J, Szabo S, et al. Highfrequency of mutations of the PIK3CA gene in human cancers. Science2004;304:554.

35. Michels S, TrautmannM, Sievers E, Kindler D, Huss S, Renner M, et al. SRCsignaling is crucial in the growth of synovial sarcoma cells. Cancer Res2013;73:2518–28.

36. Trautmann M, Sievers E, Aretz S, Kindler D, Michels S, Friedrichs N, et al.SS18-SSX fusion protein-induced Wnt/b-catenin signaling is a therapeutictarget in synovial sarcoma. Oncogene 2014;33:5006.

37. Sievers E, TrautmannM, Kindler D, Huss S, Gruenewald I, Dirksen U, et al.SRC inhibition represents a potential therapeutic strategy in liposarcoma.Int J Cancer 2015;137:2578–88.

38. Syrovets T, Gschwend JE, B€uchele B, Laumonnier Y, Zugmaier W, Genze F,et al. Inhibition of IkB kinase activity by acetyl-boswellic acids promotesapoptosis in androgen-independent PC-3 prostate cancer cells in vitro andin vivo. J Biol Chem 2005;280:6170–80.

39. Vogler M, Walczak H, Stadel D, Haas TL, Genze F, Jovanovic M, et al.Targeting XIAP bypasses Bcl-2–mediated resistance to TRAIL and coop-erates with TRAIL to suppress pancreatic cancer growth in vitro and in vivo.Cancer Res 2008;68:7956–65.

40. Ribatti D. The chick embryo chorioallantoic membrane as a model fortumor biology. Exp Cell Res 2014;328:314–24.

41. Cheng H, Dodge J, Mehl E, Liu S, Poulin N, van de Rijn M, et al. Validationof immature adipogenic status and identification of prognostic biomarkersin myxoid liposarcoma using tissue microarrays. Hum Pathol 2009;40:1244–51.

42. Hartmann W, Digon-Sontgerath B, Koch A, Waha A, Endl E, Dani I, et al.Phosphatidylinositol 30-kinase/AKT signaling is activated in medulloblas-toma cell proliferation and is associated with reduced expression of PTEN.Clin Cancer Res 2006;12:3019–27.

43. Song MS, Salmena L, Pandolfi PP. The functions and regulation of thePTEN tumour suppressor. Nat Rev Mol Cell Biol 2012;13:283–96.

44. Qi Y, Hu Y, Yang H, Zhuang R, Hou Y, Tong H, et al. Establishing a patient-derived xenograft model of human myxoid and round-cell liposarcoma.Oncotarget 2017;8:54320.

45. Simon MP, Pedeutour F, Sirvent N, Grosgeorge J, Minoletti F, Coindre JM,et al. Deregulation of the platelet-derived growth factor B-chain gene viafusion with collagen gene COL1A1 in dermatofibrosarcoma protuberansand giant-cell fibroblastoma. Nat Genet 1997;15:95–8.

46. Corless CL, Barnett CM, Heinrich MC. Gastrointestinal stromal tumours:origin and molecular oncology. Nat Rev Cancer 2011;11:865–78.

47. Greco A, Roccato E, Miranda C, Cleris L, Formelli F, Pierotti MA. Growth-inhibitory effect of STI571 on cells transformed by the COL1A1/PDGFBrearrangement. Int J Cancer 2001;92:354–60.

48. McArthurGA,Demetri GD, vanOosteromA,HeinrichMC,Debiec-RychterM, Corless CL, et al. Molecular and clinical analysis of locally advanceddermatofibrosarcoma protuberans treated with imatinib: Imatinib TargetExploration Consortium Study B2225. J Clin Oncol 2005;23:866–73.

49. Fruman DA, Rommel C. PI3K and cancer: lessons, challenges and oppor-tunities. Nat Rev Drug Discov 2014;13:140.

50. LoRusso PM. Inhibition of the PI3K/AKT/mTOR pathway in solid tumors.J Clin Oncol 2016;34:3803–15.

Mol Cancer Ther; 18(4) April 2019 Molecular Cancer Therapeutics844

Trautmann et al.

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763

2019;18:834-844. Published OnlineFirst February 20, 2019.Mol Cancer Ther   Marcel Trautmann, Magdalene Cyra, Ilka Isfort, et al.   Essential in Myxoid LiposarcomaPhosphatidylinositol-3-kinase (PI3K)/Akt Signaling is Functionally

  Updated version

  10.1158/1535-7163.MCT-18-0763doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://mct.aacrjournals.org/content/suppl/2019/02/20/1535-7163.MCT-18-0763.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://mct.aacrjournals.org/content/18/4/834.full#ref-list-1

This article cites 49 articles, 16 of which you can access for free at:

   

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

  Subscriptions

Reprints and

  .pubs@aacr.org

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://mct.aacrjournals.org/content/18/4/834To request permission to re-use all or part of this article, use this link

on February 7, 2021. © 2019 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 20, 2019; DOI: 10.1158/1535-7163.MCT-18-0763