Targeted Therapy–Based Combination Treatment in Rhabdomyosarcoma Anke E.M

Home , Dasatinib, Mapatumumab, Myosarcoma, Talazoparib, Vorinostat

Review Molecular Cancer Therapeutics Targeted Therapy–based Combination Treatment in Rhabdomyosarcoma Anke E.M. van Erp1, Yvonne M.H. Versleijen-Jonkers1, Winette T.A. van der Graaf1,2, and Emmy D.G. Fleuren3

Abstract

Targeted therapies have revolutionized cancer treatment; in this regard, as this affects multiple hallmarks of cancer at however, progress lags behind in alveolar (ARMS) and embry- once. To determine the most promising and clinically relevant onal rhabdomyosarcoma (ERMS), a soft-tissue sarcoma mainly targeted therapy–based combination treatments for ARMS and occurring at pediatric and young adult age. Insulin-like growth ERMS, we provide an extensive overview of preclinical and factor 1 receptor (IGF1R)-directed targeted therapy is one of the (early) clinical data concerning a variety of targeted therapy– few single-agent treatments with clinical activity in these dis- based combination treatments. We concentrated on the most eases. However, clinical effects only occur in a small subset of common classes of targeted therapies investigated in rhabdo- patients and are often of short duration due to treatment myosarcoma to date, including those directed against receptor resistance. Rational selection of combination treatments of tyrosine kinases and associated downstream signaling path- either multiple targeted therapies or targeted therapies with ways, the Hedgehog signaling pathway, apoptosis pathway, chemotherapy could hypothetically circumvent treatment resis- DNA damage response, cell-cycle regulators, oncogenic fusion tance mechanisms and enhance clinical efﬁcacy. Simultaneous proteins, and epigenetic modiﬁers. Mol Cancer Ther; 17(7); 1365–80. targeting of distinct mechanisms might be of particular interest 2018 AACR.

Introduction treatment including surgery, chemotherapy, and radiotherapy has increased the 5-year overall survival (OS) to approximately Rhabdomyosarcoma is the most common type of soft-tissue 70%–90% for intermediate- and low-risk rhabdomyosarcoma, sarcoma (STS) observed in young patients with the most respectively. However, patients with high-risk rhabdomyosar- frequent subtypes being embryonal (ERMS) and alveolar rhab- coma still have a 5-year OS of <40%. In addition, treatment- domyosarcoma (ARMS). ERMS represents approximately 70% related toxicities severely decrease quality of life (1, 2). In an of childhood rhabdomyosarcoma and is most often observed attempt to increase survival and improve quality of life, the in the head and neck region and genitourinary track. ARMS is field of targeted therapy has gained substantial interest in seen in approximately 30% of rhabdomyosarcoma cases and rhabdomyosarcoma, and its potential is supported by various usually occurs in the deep tissue of the extremities. The majority lines of (pre)clinical research, which are mainly centered on of ARMS tumors are characterized by a fusion between PAX3 or targeted therapies originally developed for other tumor types. PAX7 on chromosome 2 and FOXO1 on chromosome 13 In the clinic, however, intrinsic and acquired resistance (80%). The remaining 20% are fusion negative. Although mechanisms have emerged as common pitfalls in rhabdomyo- generally ARMS have a poorer outcome compared with ERMS, sarcoma treatment. As such, increasing evidence exists that fusion-negative ARMS show a genetic profile similar to ERMS single-agent targeted therapy will not be sufficient to reach and an equally favorable clinical outcome. Multimodality clinical efficacy in patients with rhabdomyosarcoma. The current hypothesis is that combination therapy could enhance clinical efficacy and/or decrease treatment-associated toxicities. In this regard, simultaneous targeting of different mechanisms 1Department of Medical Oncology, Radboud University Medical Center, Nijme- gen, the Netherlands. 2The Institute of Cancer Research, Division of Clinical of action could be more effective as opposed to combining Studies, Clinical and Translational Sarcoma Research and The Royal Marsden inhibitors of similar classes, as the characteristic hallmarks of NHS Foundation Trust, Sutton, United Kingdom. 3The Institute of Cancer cancer illustrate that tumor progression is regulated by a wide Research, Division of Clinical Studies, Clinical and Translational Sarcoma variety of different processes. Research, Sutton, United Kingdom. To determine the most promising and clinically relevant Note: Supplementary data for this article are available at Molecular Cancer combination treatments for rhabdomyosarcoma, we reviewed Therapeutics Online (http://mct.aacrjournals.org/). the preclinical and (early) clinical trial data addressing combi- Corresponding Authors: Winette T.A. van der Graaf, Division of Clinical Studies, nations of targeted therapies or targeted therapy combined with The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, chemotherapy. We focused on the most common classes of 15 Cotswold Road, SM2 5NG, Sutton, United Kingdom. Phone: 4402-0872- targeted therapies investigated in rhabdomyosarcoma to date, 24448; Fax: þ31-24-36-15025; E-mail: [email protected]; and including those directed against receptor tyrosine kinases (RTK) Emmy D.G. Fleuren, emmy.fl[email protected] andassociateddownstreamsignaling pathways, the Hedgehog doi: 10.1158/1535-7163.MCT-17-1131 signaling pathway, apoptosis pathway, DNA damage response, 2018 American Association for Cancer Research. cell-cycle regulators, fusion proteins, and epigenetic modifiers.

www.aacrjournals.org 1365

RTKs and a limited decrease of downstream AKT activity upon RTKs are membrane-bound proteins involved in signal trans- IGF1R antibody treatment. IGF1R antibodies combined with duction to the tumor cell. Activation by extracellular ligand recombinant IGFBP2, the PI3K inhibitor BKM130, or the binding or genetic mutations can lead to constitutive activity. mTOR inhibitor Ku-0063784 resensitized the resistant cell Intracellular signaling pathways, including the PI3K/AKT/mTOR, lines to IGF1R targeting (12). The possible low bioavailability RAS/MEK/ERK, and JAK/STAT3 pathway, are subsequently acti- of recombinant IGFBP2 could affect the clinical potential of vated. Several RTKs have been identified as possible targets for this particular combination treatment. Clinical trials did therapy in rhabdomyosarcoma, including the insulin-like growth show a partial response (PR) in a patient with IGF1R-positive factor 1 receptor (IGF1R), anaplastic lymphoma kinase (ALK), soft-tissue sarcoma and stable disease (SD) in 2 of 11 pediatric, platelet-derived growth factor receptors a and b (PDFGRa/b), adolescent, and young adult (AYA) patients with rhabdomyo- VEGFR, EGFR, and the fibroblast growth factor receptor 4 sarcoma for the combination of IGF1R inhibitors and mTOR (FGFR4; Fig. 1A; ref. 3). Despite the promising preclinical effects, inhibitors (Table 2; refs. 13–16). This shows that only a small clinical efficacy is limited and observed in small subsets of pati- group responded to treatment and one study showed that the ents (4, 5). Combination treatment might enhance the clinical addition of temsirolimus led to increased toxicity without a efficacy of RTK-targeted therapies (Table 1). clear increase in efficacy in most patients (15). Nevertheless, these data do show that combined targeted therapy can over- IGF1R/ALK. One way of optimizing IGF1R treatment might be come primary treatment resistance and suggest that, when its combination with other (R)TK inhibitors. As coexpression of given simultaneously, might delay or prevent treatment resis- the RTKs IGF1R and ALK has been described in rhabdomyo- tance altogether. sarcoma, this may present a rational combination. In vitro, Because chemotherapy remains fundamental in rhabdomyo- combined anti-IGF1R antibody R1507 and ALK inhibitor sarcoma treatment, combinations with chemotherapy have TAE684 treatment showed synergism in ARMS cell lines. In been investigated. In young patients with ARMS and ERMS, ERMS, however, no enhanced effect was observed (6). In ARMS, the combination of the IGF1R antibody cixutumumab with the characteristic PAX3-FOXO1 protein can enhance IGF1R conventional chemotherapy was compared with the combina- and ALK transcription, possibly explaining this difference in tion of temozolomide with conventional chemotherapy. The sensitivity. However, we, among others, could not find intrinsic combination with cixutumumab led to a higher percentage ALK activity in rhabdomyosarcoma cells (7–9). In addition, the of patients reaching an 18-month event-free survival (EFS; antitumor effects observed with the ALK inhibitor ceritinib cixutumumab 68% vs. temozolomide 39%; NCT01055314; were mostly explained by its capacity to inhibit IGF1R signaling Table 2). Of note, the combination with cixutumumab had (9). Combined treatment of ceritinib with the multikinase more reports of high-grade toxicity compared with the combi- inhibitor sorafenib showed synergistic effect in vitro (8). In nation with temozolomide. High-grade toxicities were, how- addition, we observed high activity of the signaling protein Src ever, only observed in a very small group (3/97, 3%), leaving post ceritinib treatment in both subtypes, and combined treat- the combination with cixutumumab preferential to the combi- ment of ceritinib and the Src inhibitor dasatinib was synergistic nation with temozolomide (17). Although preclinical research in vitro (9). Prior to our finding, increased activity of the and limited clinical research suggests that combinations of Src family tyrosine kinase (SFK) YES was observed in IGF1R targeted and cytostatic agents could render success, phase I treatment–resistant cell lines. Combination treatment of IGF1R studies should first investigate the optimal dose and schedule antibodies and SFK inhibitors led to superior in vitro apoptosis of new combinations in clinical practice. induction and in vivo tumor growth reduction compared with the monotherapies (10). One phase I/II trial is recruiting ARMS VEGFR. VEGFsbindtoVEGFRsandinduceangiogenesis.This and ERMS patients to investigate the combined effects of the makes both VEGFs and VEGFRs a possible target for treatment. IGF1R antibody ganitumab and dasatinib (NCT03041701; The camptothecin analogue namitecan was shown to reduce Table 2). No results have yet been reported; however, based angiogenesis in an ERMS mouse model. Moreover, combined on the consistent preclinical findings concerning IGF1R treatment of low-dose namitecan with either the VEGF anti- targeting and an increase in Src signaling, the results are body bevacizumab or the VEGFR inhibitor sunitinib led to an eagerly awaited. enhanced tumor growth reduction compared with the mono- In addition to an increase in Src activity, ERMS showed therapies. Sunitinib was in favor of bevacizumab, possibly due enhanced PDGFRb activity as a resistance mechanism to IGF1R to the effect of camptothecins on VEGF expression or the treatment. The combination of IGF1R- and PDGFRb-targeted multikinase inhibition of sunitinib(18).VEGFcanalsobe therapies increased growth inhibition in IGF1R-resistant ERMS targeted via inhibition of heparanase. Heparanase is an enzyme cell lines. Three PDGFRb inhibitors were tested, and combined necessary to generate heparin sulfate–bound growth factors, therapy with pazopanib (VEGFR, PDGFR, c-KIT) or crenolanib including VEGF. Heparanase activity is increased in rhabdo- (PDGFRa/b)wasthemostbeneficial in vivo. The combinations myosarcoma and combination treatment of the heparanase delayed tumor growth compared with each monotherapy, inhibitor SST0001 and bevacizumab or sunitinib reduced although no complete tumor regressions were achieved (11). angiogenic growth factor expression and decreased cellular Resensitization of IGF1R treatment–resistant tumor cells invasion in vitro. In vivo SST0001 monotherapy effectively was also examined by combinations of IGF1R ligand antago- decreased ARMS and ERMS tumor volumes. No further exam- nists and downstream signaling protein inhibitors. Insulin-like ination of the combination was performed in vivo,precluding growth factor binding proteins (IGFBP) regulate binding of examination of the clinical potential of combinations with insulin-like growth factor 1 and -2 (IGF1/2) to IGF1R. In ARMS, SST0001 (19). Clinical trials have not (yet) shown convincing IGF1R-resistant cell lines showed reduced IGFBP2 expression effects of anti-VEGF(R) combination treatments in young

1366 Mol Cancer Ther; 17(7) July 2018 Molecular Cancer Therapeutics

Figure 1. Overview of the cellular processes used as targets in the targeted therapy–based combination treatments in ARMS and ERMS. A, Membrane-bound growth factor receptors IGF1R, ALK, PDGFR, FGFR4,EGFR,VEGFR,Patched1(PTCH1),SMO,andTRAILR1/2;ligandsIGF1/2,VEGFand(Sonic,Indian, Desert) Hh; intracellular signaling proteins of the PI3K/AKT/mTOR, JAK/STAT3, RAS/MEK/ERK, Hh and apoptosis pathway (4E-BP1, eukaryotic translation initiation factor 4E-binding protein 1; eIF4E, eukaryotic translation initiation factor 4E; JAK, Janus kinase; FADD, Fas-associated protein with death domain; CASP, caspase; RIP1, receptor-interacting serine/threonine-protein kinase 1; BID, BH3 interacting-domain death agonist; BAX, Bcl-2-associated X protein; BAK, Bcl-2 homologous antagonist killer; BCL-2, B-cell lymphoma 2; MCL-1, induced myeloid leukemia cell differentiation protein; BCL-XL,B-cell lymphoma-extra large; Smac, second mitochondria-derived activator of caspases; Cyt C, cytochrome C, IAP, inhibitor of apoptosis protein). B and C, Intranuclear processes: DNA damage response (DDR) (PARP), the epigenome and the cell cycle that are used as therapeutic targets in (pre)clinical combination treatments in ARMS and ERMS.

www.aacrjournals.org Mol Cancer Ther; 17(7) July 2018 1367

Table 1. Preclinical targeted therapy–based combination treatments in ARMS and ERMS Subtypes Effect combination/effect Category Target(s) Combinationa included In vivo monotherapy Ref. RTKs IGF1R IGFR þ ALK R1507 þ TAE684 ARMS No CI 1.0 in ARMS (Rh41, Rh30) cell (6) ERMS lines ALK (IGF1R) þ multi-kinase Ceritinib þ sorafenib ARMS No CI < 1.0 in ERMS (RD) and ARMS (8) ERMS (Rh30) cell lines (ceritinib þ 10 mmol/L sorafenib) ALK (IGF1R) þ Src Ceritinib þ dasatinib ARMS No CI < 1.0 in ARMS (Rh41, Rh30) and (9) ERMS ERMS (RD, Rh18) cell lines IGF1R or IR/IGF1R þ YES MK0646 þ AZD0530 ARMS Yes In vitro: Increased apoptosis (10) BMS754807 þ AZD0530 ERMS In vivo: Increased tumor reduction R1507 þ AZD0530 R1507 þ dasatinibb IGF1R þ PDGFRb R1507 þ pazopanib ARMS Yesc In vitro: Re-sensitization to anti- (11) R1507 þ crenolanibb ERMS IGF1R treatment R1507 þ imatinib In vivo: Slight increased delay tumor growth, no complete regression IGF1R þ PI3K or mTORC1/2 R1507 þ recombinant IGFBP2 ARMS No Re-sensitization to anti-IGF1R (12) R1507 þ buparlisib treatment R1507 þ Ku-0063794 VEGFR VEGF(R) þ chemotherapy Bevacizumab þ namitecan ERMS Yes In vivo: Low dose namitecan (18) Sunitinib þ namitecan enhanced tumor growth reduction Heparanase þ VEGF(R) SST0001 þ bevacizumab ARMS No Reduced angiogenic factor (19) expression SST0001 þ sunitinib ERMS Reduced cellular invasion EGFR EGFR þ chemotherapy Cetuximab þ dactinomycin ARMS No CI < 1.0 in ARMS (Rh30) and ERMS (22) ERMS (RD) cell lines with increased apoptosis EGFR (drug conjugate) EGFR-conjugated immunotoxin ERMS No N.A. (23) (exotoxin A) EGFR (drug conjugate) EGFR-conjugated immunotoxin ERMS No N.A. (chloroquine increased (24) (human granzyme B) potency) (þchloroquine) FGFR FGFR4 þ IGF1R BGJ398 þ AEW541 ARMS No CI < 1.0 in ARMS (RMS13) cell line (29) PDGFRa Multi-kinase þ Src Imatinib þ PP2 (murine) No Imatinib þ PP2: Enhanced cell (31) Sorafenib þ PP2 ARMS viability reduction Sorafenib monotherapy: Most effective in reducing cell viability. Effective in absence of PDGFRa. Sorafenib þ PP2: No added effect Downstream signaling pathways PI3K/AKT/mTOR PI3K þ IGF1R or mTOR or Buparlisib þ AEW541 ARMS No CI < 1.0 in ARMS (SJCRH30) and (36) MEK Buparlisib þ rapamycinc ERMS (RD) cell lines Buparlisib þ trametinibc ERMS PI3K þ MEK PI103 þ U0126 ARMS No CI < 1.0 in ARMS (Rh30, RMS13) and (37) ERMS ERMS (RD) cell lines PI3K þ MEK AZD8055 þ selumetinibb ARMS Yes In vitro: CI < 1.0 in ARMS (Rh30) and (38) ZSTK474 þ selumetinibc ERMS ERMS (RD, RMS-YM) cell lines In vivo: BEZ235 þ selumetinibb,c AZD8055: Enhanced reduction downstream signaling Increased toxicity BEZ235: No added effect RAS/MEK/ERK MEK þ chymotrypsin-like PD98059 þ TPCK ERMS Yesd In vitro: Increased reduction cell (44) serine protease proliferation In vivo: Enhanced reduction tumor growth MEK U0126 þ radiotherapy ARMS No In vitro: ERMS: Enhanced reduction (45) ERMS sphere culture ARMS: No added effect JAK/STAT3 STAT3/GP160 þ MEK LY5 þ doxorubicin ARMS No Increased inhibition STAT3 activity (46) LY5 þ cisplatin Increased reduction of cell LY5 þ selumetinib migration Bazedoxifine þ doxorubicin Increased apoptosis Bazedoxifine þ cisplatin Bazedoxifine þ selumetinib (Continued on the following page)

1368 Mol Cancer Ther; 17(7) July 2018 Molecular Cancer Therapeutics

Table 1. Preclinical targeted therapy–based combination treatments in ARMS and ERMS (Cont'd ) Subtypes Effect combination/effect Category Target(s) Combinationa included In vivo monotherapy Ref. Hedgehog signaling Hedgehog GLI1–2 þ mTOR GANT61 þ temsirolimus ARMS No Induction cell cycle arrest (49) GANT61 þ rapamycin ERMS Increased apoptosis GANT61 þ vincristine GLI1-2 þ PI3K/mTOR GANT61 þ PI103b ARMS Yese In vitro: Enhanced apoptosis with (50) GANT61 þ BEZ235 ERMS reduced downstream signaling GANT61 þ GDC0941 In vivo: Decreased clonogenic GANT61 þ RAD001 survival, sphere formation and GANT61 þ AZD8055 tumor growth GLI1 þ chemotherapy ATO þ vincristine ARMS No CI < 1.0 in ARMS (Rh30) and ERMS (51) ATO þ vinblastine ERMS (RD) cell lines ATO þ eribulin GLI1 þ SMO ATO þ itraconazole ARMS No Limited added effects in spheroid (52) ERMS culture GLI1 þ GSK3 ATO þ lithium chloride ARMS No Limited added effects in spheroid (53) ERMS culture Apoptosis pathway Apoptosis TRAILR1 of TRAILR2 þ IAP mapatumumab þ IAPi ARMS No TRAILR1: No added effect (54) #2f(TRAILR1) ERMS TRAILR2: Increased reduction cell mapatumumab þ IAPi #3f viability lexatumumab þ IAPi #2 (TRAILR2) lexatumumab þ IAPi #3 Survivin þ chemotherapy YM155 þ cisplatin ERMS Yes In vitro/in vivo: Increased cell (55) viability and caspase-3 activity Despite increase caspase 3 activity, no increase in apoptotic cells mTOR þ Bcl-2/BCL-xL/ AZD8055 þ ABT737 ARMS No ABT737:CI<1.0 in ARMS (RMS13, (56) BCL-w or chemotherapy AZD8055 þ doxorubicin ERMS Rh30) and ERMS (RD, TE671) cell AZD8055 þ vincristine lines AZD8055 þ dactinomycin Chemotherapy: No added effects DNA damage response (DDR) PARP1 PARP1 and/or tdp1 siRNA Rucaparib þ irinotecan ARMS No PARP/tdp1 siRNA þ irinotecan: (59) with or without Olaparib þ irinotecan ERMS Increased irinotecan sensitivity chemotherapy Veliparib þ irinotecan Combination tdp1 siRNA: Added tdp1 siRNA þ irinotecan effect on cell viability Rucaparib þ tdp1 siRNA PARP1 þ chemotherapy Olaparib þ carboplatin ARMS No Irinotecan, melphalan, (60) Olaparib þ SN38 ERMS doxorubicin: CI < 1.0 in ARMS Olaparib þ vincristine (SJCRH30) and ERMS (RD) cell Olaparib þ melphalan lines Olaparib þ doxorubicin Carboplatin, vincristine: CI >0.5 and 1.0 in ARMS (SJCRH30) and ERMS (RD) cell lines PARP1 þ chemotherapy Talazoparib þ temozolomideh ARMS Yesk In vitro: (61) Temozolomide: Talazoparib þ topotecan ERMS potentiating anti-tumor effects in ARMS (Rh41, Rh30) and ERMS (RD, Rh18) cell lines Topotecan: No added effects In vivo: 2/3 models showed treatment response 1/3 models showed maintained CR Cell cycle PLK PLK1 catalytic domain þ BI2536 þ vincristineh ARMS Yes In vitro: (65) chemotherapy Vincristine, vinblastine, BI2536 þ vinblastine ERMS vinorelbine: CI < 1.0 in ARMS BI2536 þ vinorelbine (Rh30) and ERMS (RD) cell lines BI2536 þ doxorubicin Doxorubicin, paclitaxel: CI > 1.0 in BI2536 þ paclitaxel ARMS (Rh30) and ERMS (RD) cell Volasertib þ vincristineh lines Volasertib þ vinblastine In vivo: Reduced tumor growth Volasertib þ vinoelbine which remained stable for 56 Volasertib þ doxorubicin days Volasertib þ paclitaxel PLK1 catalytic domain þ Volasertib þ vincristine ERMS No Vincristine:CI 1.0 in ERMS (66) chemotherapy Volasertib þ etoposide (RMS-1) cell line Etoposide:CI> 1.0 in ERMS (RMS-1) cell line (Continued on the following page) www.aacrjournals.org Mol Cancer Ther; 17(7) July 2018 1369

Table 1. Preclinical targeted therapy–based combination treatments in ARMS and ERMS (Cont'd ) Subtypes Effect combination/effect Category Target(s) Combinationa included In vivo monotherapy Ref. PLK1 catalytic domain or BI2536 þ eribulinh ARMS Yesj Catalytic domain: (67) polo-box domain þ Poloxin þ vincristine ERMS In vitro: CI < 1.0 in chemotherapy ERMS (RD, TE381.T) CI 0.5-1 and 1 in ARMS (RMS13, Rh30) cell lines. l In vivo: Reduction tumor growth Polo-box domain In vitro: CI 0.5-1.0 and 1.0 in ERMS (RD) cell line Wee1 Wee1 þ proteasome or AZD1775 þ bortezomib ARMS No Cabozantinib and bortezomib: (70) multi-kinase or AZD1775 þ cabozantinib ERMS Most effective in combination chemotherapy AZD1775 þ cyclophosphamide with AZD1775 AZD1775 þ irinotecan AZD1775 þ etoposide AZD1775 þ dactinomycin AZD1775 þ virorelbine Wee1 þ chemotherapy AZD1775 þ irinotecan þ (O-PDX) Yesk Increased tumor response (71) vincristine ARMS compared to monotherapy ERMS AZD1775 or combined chemotherapy CDK CDK þ IGF1R Palbociclib þ BMS754807 RMSp No CI not available, synergistic effects (73) mentioned CDK þ Wee1 or Palbociclib þ AZD1775 ERMS No AZD1775: CI < 1.0 in ERMS (RD) cell (74) chemotherapy Palbociclib þ doxorubicin line Doxorubicin: CI > 1.0 in ERMS (RD) cell line Fusion protein PAX3-FOXO1 RGB-LRP-P3Fo ARMS Yes N.A. (75) Epigenome Histone modification HDAC þ chemotherapy Vorinostat þ doxorubicin ARMS No Increased reduction cell viability (77) Vorinostat þ etoposide ERMS and increased apoptosis Vorinostat þ cyclophosphamide Vorinostat þ vincristine HDAC þ chemotherapy Quisinostat þ doxorubicing ARMS Yesh,j In vitro: Increased reduction cell (78) Quisinostat þ etoposide ERMS viability and increased apoptosis Quisinostat þ vincristine In vivo: Enhanced reduction tumor Quisinostat þ cyclophosphamide growth Quisinostat þ dactinomycin HDAC þ chemotherapy Vorinostat þ cisplatin ERMS No Increased reduction cell viability (79) Valproic acid þ cisplatin and increased apoptosis HDAC þ multi-kinase or Vorinostat þ sorafenib ERMS No Sorafenib: CI < 1.0 in ERMS (RD18) (80) HSP90 or nucleoside Vorinostat þ 17-DMAG cell line reverse transcriptase Vorinostat þ abacavir 17-DMAG/abacavir: CI > 1.0 in (Sorafenib þ 17-DMAG) ERMS (RD18) cell line (Sorafenib þ 17-DMAG:CI< 1.0 in ERMS (RD18) cell line) HDAC þ SMO or mTORC1 þ Valproic acid þ vismodegib ARMS Yesl Etinostat þ docetaxel vs. (81) COX2 or chemotherapy (SMO) þ Atorvastatin PDX etinostat: 65% tumor regression Valproic acid þ metformin vs. 43% tumor regression (mTOR) þ celecoxib (COX2) All other combination:PD Entinostat þ docetaxel LSD þ HDAC GSK690 þ Quisinostat ARMS No CI < 1.0 in ARMS (Rh30) and ERMS (82) GSK690 þ Vorinostat ERMS (RD) cell lines Ex917 þ Quisinostat DNA methylation DNA demethylation þ CTL 50 aza-20-deoxtidine (DAC) þ CTL ARMS No DAC increased sensitivity to CTL (85) activity ERMS cytotoxicity Abbreviations: CI, combination index; CI < 1.0, synergism; CI ¼ 1.0, additive effect; CI > 1.0, antagonism; IR, insulin receptor; IAPi, inhibitor of apoptosis protein inhibitor; N.A., not applicable; SN38, irinotecan active metabolite. aPubChem ID available in Supplementary Table S1. bCombination tested in vivo. cOnly tested ERMS. dZebrafish model. eChicken embryo model. fCompound not further specified. gPubChem ID available in Supplementary Table S1. hCombination tested in vivo. iOnly tested ERMS. jChicken embryo model. kOnly ARMS. lOnly in vivo examination. mLiposome-protamine-siRNA particles specifically directed against PAX3-FOXO1 fusion. nSubtype not specified.

1370 Mol Cancer Ther; 17(7) July 2018 Molecular Cancer Therapeutics

patients with rhabdomyosarcoma. Bevacizumab combined in ARMS and ERMS cell lines. In addition, combined IGF1R and with standard chemotherapy did not significantly increase FGFR4 inhibitor treatment showed synergism in an ARMS cell the median EFS in patients with rhabdomyosarcoma line (29). This suggests that combinations of IGF1R and/or (NCT00643565; Table 2; ref. 20). The combination of bevaci- PI3K/mTOR and FGFR4 inhibitors might be able to prevent zumab with sorafenib and low-dose cyclophosphamide did acquired treatment resistance. The Pediatric MATCH study is an lead to a PR in a patient with rhabdomyosarcoma. However, ongoing trial, in which the pan-FGFR inhibitor JNJ-42756493 only 2 patients with rhabdomyosarcoma were included in the is tested in pediatric patients with relapsed and refractory study, making the clinical efficacy of this combination in advanced solid tumors, including STS (NCT032107140). No patients with rhabdomyosarcoma difficult to determine (21). combination studies with an FGFR4 inhibitor are currently in the clinic for rhabdomyosarcoma. EGFR. Similar to IGF1R and VEGF(R), combined therapy of the anti-EGFR antibody cetuximab and the chemotherapeutic dacti- PDGFRa. Expression of both PDGFRa and PDGF ligands has nomycin was superior to dactinomycin treatment alone in EGFR- been described in rhabdomyosarcoma tumors (30). However, expressing ARMS and ERMS cell lines (22). Despite the fact that single-agent treatment ultimately leads to treatment resistance. EGFR overexpression is present in 37%–76% of ERMS and 16%– Imatinib-resistant murine ARMS cell lines no longer showed a 50% of ARMS (23), no in vivo or clinical trials have been described decrease of PDGFRa activity, and demonstrated increased Src for EGFR-based combination treatment in rhabdomyosarcoma; activity following imatinib (Abl, c-KIT, PDGFR) treatment. Com- hence, no solid conclusions can be drawn regarding the potential bination therapy of imatinib with the SFK inhibitor PP2 enhanced of this combination in the clinic. A novel way of targeting EGFR in cell viability reduction in the resistant cell line. However, this combination with cytotoxic compounds was recently introduced, combined treatment was not more effective than monotherapy by creating antibody–drug conjugations. Niesen and colleagues with the PDGFR/RAF inhibitor sorafenib. Both in absence and used single-chain fragment variables (scFv) with high tumor presence of PDGFR, sorafenib reduced cell viability more effec- penetration capacities in combination with truncated exotoxin tively than imatinib and/or PP2. A similar reduction in cell A to create immunotoxins (IT) based on the anti-EGFR antibodies viability was seen in both na€ve and imatinib-resistant cell lines cetuximab and panitumumab. EGFR binding led to the internal- following sorafenib treatment. This suggests that PDGFRa activity ization of the compound and reduced cell viability in a number of is not as important for cell viability as RAF activity, questioning EGFR-expressing cell lines. Further analysis showed binding of the the importance of PDGFRa targeting in ARMS (31). One clinical IT to EGFR-expressing rhabdomyosarcoma cell lines and reduced trial compared the combination of the multikinase inhibitor cell viability with induction of apoptosis. Moreover, the IT was pazopanib (VEGFR, PDGFR) with the MEK inhibitor trametinib shown to bind to EGFR-expressing rhabdomyosarcoma tumor to pazopanib monotherapy in advanced STS. One sinonasal tissue (23). Niesen and colleagues more recently developed fully ERMS was included, which showed a PR to the combined treat- human cytolytic fusion proteins (hCFP) by conjugating serine ment. However, this patient was diagnosed with a PIK3CA E542K protease granzyme B to EGFR scFvs. hCFPs were shown to be aberration and previously progressed on PI3K inhibitor mono- superior to ITs since the generation of neutralizing antibodies therapy. Resistance to PI3K treatment can be related to an increase against the bacterial toxins was prevented. EGFR-directed hCFPs in MEK activity, the specific target of trametinib; thus, the PR showed similar effects as the IT, including binding to rhabdo- might not necessarily be linked to the combination with pazo- myosarcoma tumor tissue, and the effects were increased when panib and might merely be the result of the administration of combined with potency-enhancing chloroquine (24). Despite the trametinib (32). This again questions the influence of PDGFRa lack of in vivo experiments and clinical trials in patients with inhibition on the observed effect. Nevertheless, the anti-PDGFRa rhabdomyosarcoma, these hCFPs remain an interesting com- antibody olaratumab has gained FDA approval for treatment of pound, especially because in vivo experiments with ITs in breast STS and olaratumab combined with doxorubicin showed an cancer xenografts, and a phase I study with anti-CD22 immuno- enhanced OS compared with doxorubicin alone in a group of toxin moxetumomab pasudotox in childhood acute lymphoblas- mixed STS. No inclusion of rhabdomyosarcoma was mentioned tic leukemia showed good tumor penetration and well-tolerated and PDGFRa expression did not correlate with outcome (33–35). toxicity levels, respectively (25, 26). The ANNOUNCE trials will further examine the use of olaratumab combined with chemotherapeutics in a larger group of STS FGFR4. Similar to IGF1R, FGFR4 is a transcriptional target of (NCT02451943; NCT02659020). Results are not yet available, the PAX3-FOXO1 fusion protein leading to increased protein leaving the clinical efficacy of olaratumab in patients with rhab- expression (27). In rhabdomyosarcoma, FGFR4 targeting domyosarcoma unknown. is expected to be most effective in those tumors expressing activated FGFR4 due to amplification or mutation (28). Downstream signaling pathways Amplification is more likely to occur in ARMS as a result of PI3K/AKT/mTOR. The PI3K/AKT/mTOR pathway shows aber- the PAX3-FOXO1 fusion protein regulating FGFR4 expression. rant activation in rhabdomyosarcoma, either by mutations in Indeed, higher activated FGFR4 expression levels were observed the PIK3CA gene or by high levels of growth factor signaling. in ARMS as compared with ERMS cell lines (7). In contrast, PI3K/AKT/mTOR signaling promotes gene transcription, cell whole-exome/transcriptome sequencing revealed activated growth, metabolism, cell motility, and invasion (Fig. 1A). FGFR4 mutations in 4 of 60 patients with rhabdomyosarcoma, Combination treatment of the PI3K inhibitor buparlisib with all of which were ERMS (28). The only study describing a the IGF1R inhibitor AEW541, the mTOR inhibitor rapamycin, combination treatment with anti-FGFR4 therapy showed that or the MEK inhibitor trametinib showed synergism in vitro (36). FGFR4 activity functions as a compensatory mechanism to the Other MEK inhibitors (including selumetinib) showed similar effects of combined IGF1R and PI3K/mTOR inhibitor treatment synergism with multiple PI3K/mTOR inhibitors (37, 38).

www.aacrjournals.org Mol Cancer Ther; 17(7) July 2018 1371

In vivo, the most effective treatment was the mTOR inhibitor Hh signaling is involved in embryogenesis and ligand binding AZD8055 combined with selumetinib, showing an enhanced leads to GLI transcription factor (GLI) activation and subse- reduction in downstream protein activity. This combination quent target gene transcription. GLIs can also be activated by the was hence suggested for further clinical evaluation (38). PI3K/AKT/mTOR pathway, and a crosstalk between both path- Clinical trials did find an increased efficacy of dual PI3K and ways has been identified in several tumor types. The subsequent MEK inhibition compared with the monotherapies in patients GLI-1 and -2 activity has been shown to have oncogenic con- with advanced cancer; however, this was at the cost of increased sequences (Fig. 1A; ref. 47). toxicity (39). Combination treatment based on inhibiting Hh signaling can be constitutively active in rhabdomyosar- these two main downstream signaling pathways might there- coma and the GLI1/2 inhibitor GANT-61 significantly reduced forenotbefeasibleinpatients with rhabdomyosarcoma. The cell growth in rhabdomyosarcoma xenograft models, although combination of mTOR inhibitors with chemotherapy might no complete responses (CR) were achieved. Combined treat- have more potential, as combinations with liposomal doxoru- ment of GANT-61 with temsirolimus, rapamycin, or vincristine bicin, irinotecan, temozolomide, vinblastine, or cyclophospha- increased these effects, with a preference for temsirolimus over mide and topotecan were well tolerated in pediatric, AYA rapamycin (48, 49). Similar effects were observed in primary patients (40–43). Two clinical studies reported a response in rhabdomyosarcoma cells, in which multiple PI3K and/or a small subset of patients with rhabdomyosarcoma (41, 42). mTOR inhibitors combined with GANT-61 led to enhanced apoptosis. In addition, combination treatment of GANT-61 RAS/MEK/ERK pathway. Similar to PIK3CA, RAS mutations are with the PI3K inhibitor PI103 showed a decreased clonogenic present in a subset of ERMS leading to higher activity of the survival and growth in vivo. The combination was tested in a pathway (28). Both tyrosine and serine residues are phosphory- chicken embryo model, making additional in vivo examination lated and monotherapy with the MEK inhibitor PD98059 and the in a fully formed organism necessary before a translation to the chymotrypsin-like serine protease inhibitor TPCK significantly cliniccanbemade(50).Arsenictrioxide(ATO),anactive delayed tumor growth in a KRAS-mutated rhabdomyosarcoma component of Chinese medicine with anti-GLI1/2 effects, in zebrafish model without affecting normal behavior and growth combination with vincristine, vinblastine, and eribulin also (44). The monotherapies reduced proliferation in an NRAS- showed synergism in both rhabdomyosarcoma subtypes in vitro mutated ERMS cell line, but did not induce apoptosis. Combi- (51). In addition, the dual inhibition of GLI1/2 and Hh-related nation of suboptimal concentrations significantly reduced in vitro protein smoothened (SMO) or glycogen synthase kinase 3 cell proliferation and in vivo tumor growth, indicating that this (GSK3) led to a significant reduction in colony formation in combination could be of interest for RAS-mutated rhabdomyo- both ARMS and ERMS cells. However, 3D-spheroid culture, as a sarcoma. However, with the above-mentioned downstream pro- better representation of the in vivo situation, showed limited tein inhibitor combination in mind, the pharmacokinetics and additive effects of these combinations (52, 53). The current treatment-related toxicities should be closely monitored. in vitro and in vivo models do not allow for direct translation of The inhibition of the RAS/MEK/ERK pathway was also tested in these combinations to a clinical setting. However, with addi- combination with radiotherapy. Cancer stem cells (CSC) are tional in vivo studies in more complex (mouse) models, the involved in self-renewal and migration of the tumor, and high combination of Hh inhibitors with either chemotherapy or expression of CD133 is a CSC marker in ERMS. CD133-positive PI3K/AKT/mTOR inhibitors could have potential for rhabdo- ERMS spheres treated with the MEK inhibitor U0126 showed myosarcoma treatment. reduced sphere formation and combined treatment with radiotherapy enhanced antitumor effects. This suggests that MEK is Apoptosis pathway involved in ERMS CSCs and that CSC may be vulnerable to Most targeted therapies are capable of inducing apoptosis. combined radiotherapy and MEK treatment (45). These are, However, some treatments specifically activate the apoptosis however, preliminary data; thus, further research is necessary pathway. Apoptosis is triggered via an extrinsic death receptor before any potential clinical implications can be made. pathway and the intrinsic mitochondrial pathway. TNF-related apoptosis-inducing ligands (TRAIL) activate membrane-bound JAK/STAT3 pathway. Persistent STAT3 activity has been reported TRAIL receptors (TRAILR) and subsequently activate a caspase in rhabdomyosarcoma and monotherapy with the STAT3-specific cascade or induce intrinsic mitochondrial pathway activation. LY5 or the upstream STAT3 GP130 inhibitor bazedoxifene led to Upon activation of the intrinsic pathway, mitochondrial cyto- decreased cell migration and induced apoptosis in pSTAT3-pos- chrome c, and Smac are released into the cytosol. Cytochrome c itive ARMS in vitro. The effects of both compounds could be subsequently induces the activation of caspase-9, whereas Smac enhanced by the addition of doxorubicin, cisplatin, or the MEK antagonizes the inhibitor of apoptosis (IAP) proteins (including inhibitor AZD6244 (46). These results emphasize again that survivin; Fig. 1A; ref. 54). combinations of downstream protein inhibitors with either che- TRAILR1- and TRAILR2-specific agonistic antibodies, either motherapy or other downstream inhibitors can be very capable of alone or in combination with IAP inhibitors, were examined in improving antitumor effects as compared with monotherapies. both rhabdomyosarcoma subtypes. TRAILR1 and IAP inhibitor These are encouraging findings, although STAT3 inhibition com- monotherapies were not effective. TRAILR2 therapy did show a bined with doxorubicin or cisplatin requires further in vivo eval- dose-dependent cell viability reduction, although this was inde- uation to determine whether clinical evaluation is worthwhile. pendent of the level of TRAILR2 present in the cell line. The TRAILR2 effects could be enhanced by addition of IAP inhibitors Hedgehog signaling in both ARMS and ERMS (54). Several well-known developmental pathways are involved in Aspecific survivin inhibitor, YM155, was recently tested tumorigenesis, including the Hedgehog (Hh) signaling pathway. as a monotherapy and in combination with cisplatin in

1372 Mol Cancer Ther; 17(7) July 2018 Molecular Cancer Therapeutics

ERMS. YM155 reduced survivin levels and cell viability in vitro. A phase Ib trial investigating the effects of olaparib combined Both in vitro and in vivo, combination treatment with cisplatin withtrabectedininSTSreportedaPRin18%andSDin23%of led to an enhanced effect on cell viability and caspase levels, but patients (NCT02398058; Table 2; ref. 62). A phase I trial in apoptosis only slightly increased as compared with cisplatin pediatric and AYA solid tumors is investigating the efficacy monotherapy. No examination of YM155 in the ARMS subtype of combined treatment of talazoparib with irinotecan, with or was performed, leaving its use in ARMS to be evaluated (55). without temozolomide. Preliminary data encouragingly Members of the antiapoptotic Bcl-2 family are overexpressed in showed response in a subset of patients for the combination rhabdomyosarcoma (Fig. 1A). Combined targeting of the Bcl-2 of talazoparib and irinotecan (NCT02392793; Table 2; ref. 63). family inhibitor ABT737 and the mTOR inhibitor AZD8055 None of these studies have yet mentioned a response in showed synergy in rhabdomyosarcoma cell lines. The added rhabdomyosarcoma. One other trial is examining effects of effects of AZD8055 with ABT737 were dependent on the inhibi- olaparib combined with concomitant radiotherapy in locally tion of both mTOR complexes and a decrease in MCL-1 levels. Of advanced STS (NCT02787642; Table 2). note, combination of AZD8055 with chemotherapy was not synergistic (56). Cell cycle All of the abovementioned studies suggest targeting of apo- In addition to the DDR, the cell cycle can be inhibited to ptosis pathway–associated proteins to be a potential therapeutic affect cell viability. The cell cycle is a strictly regulated cellular option. However, only one study examined in vivo effects and process and multiple kinases are involved in its regulation, further research is necessary. Nonetheless, the combination of including the cyclin-dependent kinases (CDK), polo-like kinase anti-BcL-2 and mTOR inhibitors does give a good example of the 1 (PLK1), and Wee1 kinase (64). CDKs play a crucial role potential of a combination treatment that targets multiple pro- throughout the whole cell cycle and act at different stages of cesses in the tumor cell at once. Through the inhibition of cell-cycle progression. PLK1 actively regulates the transition multiple, distinct cellular processes, we might be able to generate from G2–M phase by phosphorylating Wee1, triggering Wee1 a more robust antitumor effect without having to resort to degradation. Inhibition of PLK1 can induce a mitotic arrest combinations with systemic chemotherapy. One clinical trial is leading to cell death. Wee1 negatively regulates entry into comparing the combination of a TRAILR2 agonist and chemo- mitosis by inducing an inhibitory phosphorylation of CDK1, therapy with TRAILR2 and an anti-IGF1R antibody in adult solid leading to a G2–M arrest necessary for DNA repair. Inhibition is tumors, including sarcomas (NCT01327612; Table 2). This trial thought to prevent the G2–M arrest and subsequent DNA repair might give more insights into the clinical efficacy of targeting resulting in a premature entry into mitosis and induction of multiple cellular processes simultaneously and whether it is cell death (Fig. 1C). preferential to combinations with systemic chemotherapy. PLK1. Rhabdomyosarcoma has higher levels of PLK1 compared The abovementioned studies all describe targets consisting of with healthy tissue (65, 66). Multiple preclinical studies have membrane-bound growth factor receptors and associated intra- shown ARMS to be more sensitive to PLK1 inhibitors compared cellular signaling proteins. However, other intracellular processes with ERMS (65–67). This sensitivity might be related to the role of such as DNA damage repair, cell-cycle regulation, and gene PLK1 in the activation and expression of PAX3-FOXO1 (68). All expression regulation are likewise targets for treatment (Table 1; studies showed synergistic effects for the combination of PLK1 Fig. 1B and C). inhibitors and antimicrotubule agents (65–67). No synergism was observed for the combination with etoposide, doxorubicin, DNA damage response (DDR): PARP1 or paclitaxel (65, 66). The working mechanism of etoposide, The DNA damage response (DDR) plays a crucial role in doxorubicin, and paclitaxel is either before (etoposide) or after the defense against the deleterious effects of DNA damage (doxorubicin, paclitaxel) the mitotic phase, which might explain (57, 58). Key to the DDR is PARP1. PARP1 is involved in the observed ineffectiveness of these combinations (66). In con- single-strand break repair, where its binding to damaged DNA trast to PLK1-catalytic domain inhibitors, polo-box domain inhi- leads to poly (ADP-ribose) (pADPr) chain synthesis and recruit- bitors in combination with vincristine were less effective; ment of repair proteins. pADPr chains are also involved in PARP1 although it still induced apoptosis to a larger extend than respec- release from the DNA to ensure access of repair proteins to the tive monotherapies (67). One phase I study with monotherapy damaged site (Fig. 1B). In rhabdomyosarcoma, monotherapy of volasertib in pediatric solid tumors has been concluded the PARP inhibitor olaparib showed intermediate effects on cell (NCT01971476); however, results are not yet available. Despite viability. Combination of PARP inhibitors with multiple che- these promising preclinical results, there are no clinical trials motherapeutics showed enhanced in vitro effects for combined examining combination treatments with PLK1 inhibitors in treatment with irinotecan, melphalan, doxorubicin, and temo- (pediatric) patients with rhabdomyosarcoma. However, based zolomide (59–61). Lower synergism was seen for the combina- on the preclinical data and the tolerable toxicity of the PLK1 tion with carboplatin or vincristine, and addition of topotecan did inhibitor NMS-1286937 in adult patients with solid tumor, such not enhance effects (60, 61). Moreover, the PARP inhibitor trials would be of interest (69). talazoparib combined with temozolomide showed response in 2 of 3 ARMS models in vivo, and one model showed a maintained Wee1. An in vitro drug screen identified compounds with CR until the end of the study (61). In addition, PARP and DNA potential antitumor effects in rhabdomyosarcoma. The mono- repair enzyme tyrosyl-DNA phosphodiesterase (tdp1) can form a therapeutic and combined effects of cyclophosphamide, irino- DNA repair complex. Genetic knockdown of tdp1 combined with tecan, etoposide, dactinomycin, virorelbine, the Wee1 inhibitor irinotecan or the PARP1 inhibitor rucaparib led to enhanced AZD1775, the multi-RTK inhibitor cabozantinib, and the pro- antitumor effects in vitro (59). tease inhibitor bortezomib were tested. The results showed that

www.aacrjournals.org Mol Cancer Ther; 17(7) July 2018 1373

Table 2. Clinical trials examining targeted therapy-based combination treatments in patients with (soft tissue) sarcoma Category Target(s) Combinationa Phase Population Response Toxicity NCT#/Ref. RTKs IGF1R IGF1R þ SFK Ganitumab þ dasatinib I/II ARMS and ERMS —— NCT03041701 IGF1R þ mTOR Cixutumumab þ II Bone and STS patients, allocated PR: 1/57 IGF1Rþ STSb Grade 3-4: 8% of adverse events (13) mct.aacrjournals.org temsirolimus based on IGF1R expression (RMS (n ¼ 2546) n ¼ 10) IGF1R þ mTOR Cixutumumab þ I Pediatric and AYA recurrent solid SD:3/39( 3 cycles) No Mostly reversible. Dose-limiting (14) temsirolimus tumors (RMS n ¼ 9) mention of RMS. mucositis, fatigue, hypercholesterolemia, transaminitis observed IGF1R þ mTOR Cixutumumab þ II Pediatric and adolescent sarcoma SD: 2/11 RMS 1 RMS removed due to toxicity. (15) temsirolimus (RMS n ¼ 11) Others: addition of temsirolimus increased toxicity IGF1R þ mTOR Dalotuzumab þ I Pediatric advanced solid tumors PR: 1/24 No mention of RMS Well tolerated (16) on September 25, 2021. © 2018American Association for Cancer Research. ridaforolimus (RMS n ¼ 4) IGF1R þ chemotherapy Cixutumumab or II Pediatric and adult metastatic ARMS Cixutumumab vs. Cixutumumab: Grade 5 SOS: 1/97 NCT01055314/(17) temozolomide þ and ERMS temozolomide: Grade 4 SOS: 2/97 Temozolimide: combination 18-months EFS: 68% and No toxicity mentioned chemotherapy 39%, respectively VEGF(R) VEGF þ chemotherapy Bevacizumab þ II Childhood and AYA metastatic RMS Chemotherapy vs. No increase grade 3/4 toxicity NCT00643565/(20) chemotherapy and non-RMS STS chemotherapy þ compared to mono chemotherapy bevacizumab: Median EFS (months): 14.9 vs. 20.6 OR: 36% vs. 54% VEGF þ multi-RTK þ Bevacizumab þ sorafenib þ I Recurrent/refractory pediatric and PR: 1/2 RMS Well tolerated (21) chemotherapy low-dose YA solid tumors (RMS n ¼ 2) cyclophosphamide PDGFRa Multikinase þ MEK Pazopanib þ trametinib Ib/II Advanced STS (sinonasal ERMS with PR: 1/1 RMS, 59% tumor size 18/25: grade 1/2 toxicity 3/25: grade (32) PI3KCA E542K aberration reduction 3 toxicity 1/25: heart problems due progressed on PI3K monotherapy, to prior treatment n ¼ 1) Downstream signaling pathways mTOR mTOR þ chemotherapy Temsirolimus þ liposomal I/II Recurrent sarcoma (ARMS n ¼ 2, - Well tolerated (40) doxorubicin ERMS n ¼ 2) mTOR þ chemotherapy Temsirolimus þ irinotecan I YA relapsed/refractory solid tumors SD: 1/4 RMS Well tolerated (41) or temozolomide (RMS n ¼ 4) þ þ oeua acrTherapeutics Cancer Molecular mTOR chemotherapy Sirolimus vinblastine I Pediatric recurrent/refractory solid PR: 1/2 RMS (metastatic Well tolerated (42) tumors (RMS n ¼ 2) ARMS) mTOR þ chemotherapy Sirolimus þ I Pediatric and YA relapsed/refractory SD: 6/20 No mention of RMS Well tolerated (43) cyclophosphamide þ solid tumors (RMS n ¼ 3) topotecan Apoptosis pathway TRAILR TRAILR2 þ chemotherapy Conatumumab þ ongoing II Advanced solid tumors, including –– NCT01327612 or IGF1R chemotherapy or sarcomas ganitumab (Continued on the following page) www.aacrjournals.org Downloaded from Table 2. Clinical trials examining targeted therapy-based combination treatments in patients with (soft tissue) sarcoma (Cont'd ) Category Target(s) Combinationa Phase Population Response Toxicity NCT#/Ref. DNA damage response (DDR) PARP PARP1 þ chemotherapy Olaparib þ Trabectedin I Metastatic and advanced adult PR: 4/22 Well tolerated NCT02398058/(62) sarcomas SD: 5/22 No mention RMS PARP1 þ chemotherapy Talazoparib þ irinotecan I Pediatric and AYA refractory or Talazoparib þ irinotecan: Well tolerated NCT02392793/(63) with/without recurrent solid tumors SD: 9/22 No mention RMS temozolomide mct.aacrjournals.org PARP1 þ radiotherapy Olaparib þ concomitant I Locally advanced/unresectable STS –– NCT02787642 radiotherapy Cell cycle Wee1 Wee1 þ chemotherapy AZD1775 þ gemcitabine, I Advanced solid tumors PR: 17/176 Well tolerated NCT00648648/(72) cisplatin or carboplatin Conﬁrmed PR: 7/17 SD: 94/176 No mention RMS Wee1 þ chemotherapy AZD1775 þ irinotecan I/II Pediatric and AYA relapsed and –– NCT02095132 hydrochloride refractory solid tumors, including RMS CDK CDK4/6 þ chemotherapy Ribociclib þ doxorubicin I Adult unresectable STS –– NCT03009201 on September 25, 2021. © 2018American Association for Cancer Research. Epigenome HDAC HDAC þ VEGF Valproic acid þ I/II Advanced STS –– NCT01106872 bevacizumab þ gemcitabine þ doxetacel HDAC þ proteasome Vorinostat þ bortezomib II Recurrent adult STS –– NCT00937495 HDAC þ proteasome Vorinostat þ bortezomib I Pediatric and adolescent patients –– NCT01132911 with recurrent and refractory solid tumors, including sarcoma HDAC þ chemotherapy Vorinostat þ etoposide I/II Children and adolescents with –– NCT01294670 refractory solid tumors, including sarcoma HDAC þ chemotherapy Abexinostat þ doxorubicin I Metastatic sarcoma SD: 5/11 No mention RMS Well manageable (83) HDAC þ chemotherapy Belinostat þ doxorubicin I/II Advanced STS (RMS n ¼ 1) Phase I: Well tolerated NCT00878800/(84)

PR: 2/25 Therapy Targeted SD: 16/25 PD: 7/25 Phase II: PR: 1/16 CR: 1/16 SD: 9/16

PD: 5/16 – ae obnto ramn nRMS in Treatment Combination based No mention RMS Abbreviations: þ, IGF1R-positive; —, not mentioned; OR, objective response; PD, progressive disease; SOS, sinusoidal obstructive syndrome; YA, young adult. o acrTe;1()Jl 2018 July 17(7) Ther; Cancer Mol aPubchem ID available in Supplementary Table S1. bSubtype not speciﬁed. 1375 van Erp et al.

for both rhabdomyosarcoma subtypes, AZD1775 combined protein, the developments in nanomedicine are of interest, with cabozantinib or bortezomib was most effective (70). although the current data do not suggest fast implementation In line with these findings, a multi-drug screen of pediatric in the clinic. However, once the antitumor effects and bioavail- orthotopic patient-derived xenograft (O-PDX) models also ability are optimized for clinical use, this could be a novel and, showed high sensitivity of ARMS and ERMS for AZD1775. likely, highly effective treatment for patients with ARMS. Moreover, the combination of AZD1775 with irinotecan and vincristine improved tumor response (71). In line with the Epigenome preclinical findings, the combined effects of AZD1775 and The epigenome, alongside PAX3/7-FOXO1 gene transcription, irinotecan hydrochloride will be examined in pediatric and is responsible for the expression profile in rhabdomyosarcoma. AYA recurrent solid tumors, including rhabdomyosarcoma Epigenetic changes affect chromatin, resulting in a more open or (NCT02095132; Table 2). In addition, AZD1775 combined more closed conformation. Open conformations allow for gene with chemotherapy will be examined in adult solid tumor transcription while closed conformations repress gene expression patients (NCT00648648; Table 2; ref. 72). (Fig. 1B). In tumors, epigenetic alterations lead to the repression or induction of cancer-related genes. Several processes, including CDK. CDK inhibitors could also be of interest for combination DNA methylation and histone modification, are involved in treatment in rhabdomyosarcoma. This could particularly be the epigenetic regulation (76). case in ARMS, because activation of cyclin D1/CDK4 leads to PAX3-FOXO1 activation. Preliminary data described synergism Histone deacetylase. Histone deacetylases(HDAC)regulatethe for the combined treatment of CDK inhibitors with IGF1R or structure of chromatin around histone proteins by deacetylat- Wee1 inhibitors in rhabdomyosarcoma cell lines (73, 74). In ing lysine residues, leading to a compact structure. In many contrast, combined treatment of the CDK4/6 inhibitor palboci- tumor types high levels of HDACs were shown to repress tumor clib with doxorubicin showed antagonism in an ERMS cell line. A suppressor gene expression (76). In rhabdomyosarcoma, possible biomarker for response could be expression of the cell- monotherapy with the HDAC inhibitor vorinostat had limited cycle–associated retinoblastoma protein (Rb). High Rb expres- effects in vivo, which was similar to the effects observed in a sion correlated with a higher response to the combination of phase I study in children with recurrent solid tumors. Never- palbociclib with either doxorubicin or AZD1775, whereas knock- theless, HDAC inhibitors combined with multiple chemother- down of Rb led to antagonistic effects (74). One phase I study will apeutics were shown to have added effects in rhabdomyosar- examine the combined effects of the CDK4/6 inhibitor ribociclib coma cell lines (77–79). The combination with doxorubicin and doxorubicin in adult unresectable STS and this might give showed the highest synergism and reduced rhabdomyosarcoma more insight into the clinical potential of this combination in tumor growth in vivo (78). rhabdomyosarcoma (NCT03009201; Table 2). Dual and triple combinations, including vorinostat, were examined in ERMS. Multiple targeted therapies (vorinostat, PAX3-FOXO1 17-DMAG, sorafenib, abacavir) were combined with each Compared with the abovementioned processes, targeting of other and/or with doxorubicin. Dual combinations of vorino- the PAX3-FOXO1 fusion in ARMS could lead to better results stat with the multi-RTK inhibitor sorafenib, and sorafenib with with a lower risk of treatment resistance. Direct targeting of this the HSP90 inhibitor 17-DMAG were synergistic. In contrast, fusion protein has, however, proven difficult, hence multiple triple combinations with doxorubicin, 17-DMAG, and vorino- studies aimed at targeting pathways or proteins regulated by stat or sorafenib did not show a clear increase in antitumor this oncogenic driver. However, as the fusion orchestrates effects (80). One study examined multiple combinations in an multiple cellular processes and influences expression of many ARMS PDX model based on the genomic and proteomic profile þ target genes, which are difficult to target simultaneously, robust of the tumor tissue, showing among others high Sirt1 (NAD antitumor responses in ARMS are difficult to achieve in this HDAC) levels. In vivo combination treatments were designed way. In this light, direct silencing of the fusion with PAX3- and the combination of the HDAC inhibitor entinostat and FOXO1-silencing RNA (siRNA) could be a more effective the chemotherapeutic docetaxel was most effective, with 65% approach, as it would instantly deprive ARMS of its driver. tumor regression. Monotherapy entinostat led to a 43% tumor Delivery of the siRNA to the tumor cell remains, however, an regression (81). obstacle. A promising way of delivery could be a liposome- In addition to HDAC, lysine-specific demethylase 1 (LSD1) protamine-siRNA particle (LRP) that carries siRNA over the is overexpressed in rhabdomyosarcoma. Combination treat- lipid barrier of the tumor cell. It has already been shown that ment of LSD1 and HDAC inhibitors showed synergism in ARMScanbetargetedwithLRPswhenconjungatedtocyclic ARMS and ERMS cell lines (82). HDAC inhibitors in combi- RGD (arginine-glycine-aspartate) peptides, targeting the over- nation with either targeted therapies or chemotherapy are in expressed aBb3 integrin receptor. RGD-LRP particles loaded clinical trials (NCT01106872, NCT00937495, NCT01132911, with anti-PAX3-FOXO1 siRNA (RGD-P3F-LRP) were effectively NCT01294670; Table 2) and a subset of patients with delivered to ARMS cell lines and led to reduction of both advanced STS responded to monotherapy and combination PAX3-FOXO1 and PAX3-FOXO1-target gene expression. Some treatments (NCT00878800; Table 2; refs. 83, 84). Both the effect on cell viability was seen, but no clear induction of preclinical and early clinical trial data show potential for apoptosis. In vivo, tumor outgrowth was delayed compared epigenetic inhibitors combined with chemotherapy. It would with the control, but effects on established tumors were lim- be interesting to see whether epigenetic inhibitors alter the ited, possibly due to the incomplete reduction of PAX3-FOXO1 expression of repressed genes as this could lead to new com- expression following treatment (75). Because no other treat- bination treatments, possibly enhancing the antitumor effects ments are currently capable of directly targeting this fusion of epigenetic inhibitors.

1376 Mol Cancer Ther; 17(7) July 2018 Molecular Cancer Therapeutics

Epigenetic modification and immunotherapy. In line with this research into this field is eagerly awaited. In addition to the hypothesis, demethylating 5-aza-20-deoxytidine (DAC) enhanced targeted therapies currently used in combination treatments, the expression of tumor antigens in rhabdomyosarcoma cell lines. other targeted therapies, such as those directed against checkpoint DAC enhanced mRNA expression of immunogenic cancer-testis kinase 1 (CHK1) and enhancer of zeste homolog 2 (EZH2), could antigens MAGE-A1, MAGE-A3 and NY-ESO1. Of note, mRNA be of interest for future combination regimens. Preclinical data expression did not necessarily correlate with protein expression, have shown promising single-agent efficacy and combination as only MAGE-A1 and –A3 were expressed in ARMS, whereas treatments might be able to reduce treatment resistance and/or mRNA levels increased in both subtypes. The recognition-related enhance the antitumor effects (86, 87). proteins MHC class I–II molecules and costimulatory ICAM-1 This review also highlights some recent developments in nano- were also increased, as was the reactivity of cancer-testis antigen- medicine and drug conjugates. Even though the current com- specific CTLs. This indicated that, despite a low cancer testis- pounds are not yet ready for clinical implementation, these could antigen increase, treatment of demethylating agents can enhance elicit high antitumor effects or even eliminate the ARMS driver. the T-cell response against the tumor (85). No in vivo examination Further optimization of these compounds is therefore needed and has been performed. However, these findings do suggest that examination of their potential in rhabdomyosarcoma treatment targeting of the epigenome can alter protein expression and make should be continued. tumor cells more vulnerable to other treatments, such as Of note, most studies, both preclinical and clinical, showed immunotherapy. their effects to be present in a subset of cell lines, xenografts or patients, underlining the heterogeneous response these combination treatments can generate. Discovery of predictive biomar- Discussion kers could decrease unnecessary treatment of patients and might As can be appreciated from the abovementioned studies, the prevent unnecessary toxicities. Identification of biomarkers in potential of combination therapy has been examined in many rhabdomyosarcoma remains difficult, however, and more different fields of targeted therapy. Recent studies show encour- research is necessary. Worthy of comment are the recent insights aging data, suggesting that a variety of targeted therapy–based from a variety of genomic, epigenomic, and (phospho)proteomic combinations are capable of increasing antitumor effects in projects. Although the majority of patients with rhabdomyosar- rhabdomyosarcoma. The majority of these studies focused on coma have a low mutational burden, a recent report did reveal a the potential of combinations with standard-of-care chemother- number of activating mutations and epigenetic alterations that apeutics. In this regard, targeting of RTKs, downstream proteins, could have an effect on pathway activity (28). Also a phospho- Hedgehog signaling, DDR, cell-cycle proteins and the epigenome proteomics screen in rhabdomyosarcoma cell lines, and a com- shows promise, especially as in many cases synergy was achieved bined genomic and morphoproteomic screening of an ARMS with low-dose chemotherapy, possibly lowering adverse events in sample identified and validated adequate treatment strategies patients. Because chemotherapy is and will for the foreseeable (7, 81). This exemplifies that molecular genomics, epigenomics, future remain a vital part of rhabdomyosarcoma treatment, these and (phospho)proteomics might have a place in rhabdomyosar- drug combinations are of importance. coma diagnosis to provide patients with the best, most person- However, as chemotherapy-based treatments remain toxic, alized treatment available. noncytotoxic-based targeted combination regimens have been Altogether, a variety of targeted therapy–based combina- explored as well, although to a lesser extent. Some combinations tion treatment regimens show promise for patients with targeting associated pathways, such as combined IGF1R and rhabdomyosarcoma. Although more research is required into mTOR inhibition, show some efficacy in rhabdomyosarcoma. the most suitable combinations accompanied with studies Simultaneous targeting of multiple cancer hallmarks that are identifying predictive biomarkers, adequate implementation tumor-specific and have complementary roles in tumor cell of such combined regimens in future clinical trials could survival could hypothetically lead to more robust antitumor improve outcome and reduce side effects of patients with effects with lesser adverse effects on healthy tissue. Indeed, com- rhabdomyosarcoma. binations targeting multiple distinct signaling pathways or different cellular processes show potential for rhabdomyosarcoma Disclosure of Potential Conflicts of Interest treatment, including combined Hedgehog and mTOR inhibition, No potential conflicts of interest were disclosed. and HDAC inhibitors combined with multi-RTK inhibitors. The combination of certain epigenetic modifiers with immunomo- Received November 15, 2017; revised February 27, 2018; accepted May 1, dulating drugs are also worthy of further assessment. More 2018; published first July 2, 2018.

References 1. van der Graaf WT, Orbach D, Judson IR, Ferrari A. Soft tissue 4. Pappo AS, Vassal G, Crowley JJ, Bolejack V, Hogendoorn PC, Chugh R, et al. sarcomas in adolescents and young adults: a comparison with A phase 2 trial of R1507, a monoclonal antibody to the insulin-like their paediatric and adult counterparts. Lancet Oncol 2017;18: growth factor-1 receptor (IGF-1R), in patients with recurrent or refractory e166–e75. rhabdomyosarcoma, osteosarcoma, synovial sarcoma, and other soft 2. El Demellawy D, McGowan-Jordan J, de Nanassy J, Chernetsova E, Nasr A. tissue sarcomas: results of a sarcoma alliance for research through collab- Update on molecular ﬁndings in rhabdomyosarcoma. Pathology oration study. Cancer 2014;120:2448–56. 2017;49:238–46. 5. Kim A, Widemann BC, Krailo M, Jayaprakash N, Fox E, Weigel B, et al. Phase 3. Kashi VP, Hatley ME, Galindo RL. Probing for a deeper understanding of 2 trial of sorafenib in children and young adults with refractory solid rhabdomyosarcoma: insights from complementary model systems. Nat tumors: a report from the children's oncology group. Pediatr Blood Cancer Rev Cancer 2015;15:426–39. 2015;62:1562–6.

www.aacrjournals.org Mol Cancer Ther; 17(7) July 2018 1377

6. van Gaal JC, Roeffen MH, Flucke UE, van der Laak JA, van der Heijden G, de 23. Niesen J, Brehm H, Stein C, Berges N, Pardo A, Fischer R, et al. In vitro effects Bont ES, et al. Simultaneous targeting of insulin-like growth factor-1 and ex vivo binding of an EGFR-specific immunotoxin on rhabdomyo- receptor and anaplastic lymphoma kinase in embryonal and alveolar sarcoma cells. J Cancer Res Clin Oncol 2015;141:1049–61. rhabdomyosarcoma: a rational choice. Eur J Cancer 2013;49:3462–70. 24. Niesen J, Hehmann-Titt G, Woitok M, Fendel R, Barth S, Fischer R, et al. A 7. Fleuren EDG, Vlenterie M, van der Graaf WTA, Hillebrandt-Roeffen MHS, novel fully-human cytolytic fusion protein based on granzyme B shows in Blackburn J, Ma X, et al. Phosphoproteomic profiling reveals ALK and MET vitro cytotoxicity and ex vivo binding to solid tumors overexpressing as novel actionable targets across synovial sarcoma subtypes. Cancer Res the epidermal growth factor receptor. Cancer Lett 2016;374:229–40. 2017;77:4279–92. 25. Amoury M, Kolberg K, Pham AT, Hristodorov D, Mladenov R, Di Fiore S, 8. Dolgikh N, Fulda S. Rhabdomyosarcoma cells are susceptible to cell death et al. Granzyme B-based cytolytic fusion protein targeting EpCAM specif- by LDK378 alone or in combination with sorafenib independently of ically kills triple negative breast cancer cells in vitro and inhibits tumor anaplastic lymphoma kinase status. Anticancer Drugs 2017;28:1118–25. growth in a subcutaneous mouse tumor model. Cancer Lett 2016; 9. van Erp AEM, Hillebrandt-Roeffen MHS, van Houdt L, Fleuren EDG, 372:201–9. van der Graaf WTA, Versleijen-Jonkers YMH. Targeting anaplastic 26. Wayne AS, Shah NN, Bhojwani D, Silverman LB, Whitlock JA, Stetler- lymphoma kinase (ALK) in rhabdomyosarcoma (RMS) with the second- Stevenson M, et al. Phase 1 study of the anti-CD22 immunotoxin mox- generation ALK inhibitor ceritinib. Targeted oncology 2017:1–12. etumomab pasudotox for childhood acute lymphoblastic leukemia. Blood 10. Wan X, Yeung C, Heske C, Mendoza A, Helman LJ. IGF-1R inhibition 2017;130:1620–7. activates a YES/SFK bypass resistance pathway: rational basis for Co- 27. Cao L, Yu Y, Bilke S, Walker RL, Mayeenuddin LH, Azorsa DO, et al. Targeting IGF-1R and Yes/SFK kinase in rhabdomyosarcoma. Neoplasia Genome-wide identification of PAX3-FKHR binding sites in rhabdomyo- 2015;17:358–66. sarcoma reveals candidate target genes important for development and 11. Heske CM, Yeung C, Mendoza A, Baumgart JT, Edessa LD, Wan X, et al. cancer. Cancer Res 2010;70:6497–508. The Role of PDGFR-beta activation in acquired resistance to IGF-1R 28. Seki M, Nishimura R, Yoshida K, Shimamura T, Shiraishi Y, Sato Y, et al. blockade in preclinical models of rhabdomyosarcoma. Translat Oncol Integrated genetic and epigenetic analysis defines novel molecular sub- 2016;9:540–7. groups in rhabdomyosarcoma. Nat Commun 2015;6:7557. 12. Kang Z, Yu Y, Zhu YJ, Davis S, Walker R, Meltzer PS, et al. Downregulation 29. Wachtel M, Rakic J, Okoniewski M, Bode P, Niggli F, Schafer BW. FGFR4 of IGFBP2 is associated with resistance to IGF1R therapy in rhabdomyo- signaling couples to Bim and not Bmf to discriminate subsets of alveolar sarcoma. Oncogene 2014;33:5697–705. rhabdomyosarcoma cells. Int J Cancer 2014;135:1543–52. 13. Schwartz GK, Tap WD, Qin L-X, Livingston MB, Undevia SD, Chmielowski 30. Ehnman M, Missiaglia E, Folestad E, Selfe J, Strell C, Thway K, et al. Distinct B, et al. Cixutumumab and temsirolimus for patients with bone and soft- effects of ligand-induced PDGFRalpha and PDGFRbeta signaling in the tissue sarcoma: a multicentre, open-label, phase 2 trial. Lancet Oncol human rhabdomyosarcoma tumor cell and stroma cell compartments. 2013;14:371–82. Cancer Res 2013;73:2139–49. 14. Fouladi M, Perentesis JP, Wagner LM, Vinks AA, Reid JM, Ahern C, et al. 31. Abraham J, Chua YX, Glover JM, Tyner JW, Loriaux MM, Kilcoyne A, et al. A phase I study of cixutumumab (IMC-A12) in combination with An adaptive Src-PDGFRA-Raf axis in rhabdomyosarcoma. Biochem Bio- temsirolimus (CCI-779) in children with recurrent solid tumors: a phys Res Commun 2012;426:363–8. children's oncology group phase I consortium report. Clin Cancer Res 32. Subbiah V, Meyer C, Zinner R, Meric-Bernstam F, Zahurak ML, O'Connor A, 2015;21:1558–65. et al. Phase Ib/II study of safety & efficacy of combination therapy with 15. Wagner LM, Fouladi M, Ahmed A, Krailo MD, Weigel B, DuBois SG, et al. multikinase VEGF inhibitor pazopanib and MEK inhibitor trametinib in Phase II study of cixutumumab in combination with temsirolimus in advanced soft tissue sarcoma. Clin Cancer Res 2017;23:4027–34. pediatric patients and young adults with recurrent or refractory sarcoma: a 33. Tap WD, Jones RL, Van Tine BA, Chmielowski B, Elias AD, Adkins D, et al. report from the Children's Oncology Group. Pediatr Blood Cancer Olaratumab and doxorubicin versus doxorubicin alone for treatment of 2015;62:440–4. soft-tissue sarcoma: an open-label phase 1b and randomised phase 2 trial. 16. Frappaz D, Federico SM, Pearson AD, Gore L, Macy ME, DuBois SG, Lancet 2016;388:488–97. et al. Phase 1 study of dalotuzumab monotherapy and ridaforolimus- 34. van der Graaf WT. Olaratumab in soft-tissue sarcomas. Lancet 2016; dalotuzumab combination therapy in paediatric patients with advanced 388:442–4. solid tumours. Eur J Cancer 2016;62:9–17. 35. Judson I, van der Graaf WT. Sarcoma: Olaratumab - really a breakthrough 17. Malempati S, Weigel B, Anderson JR, Parham D, Teot LA, Rodeberg DA, for soft-tissue sarcomas? Nat Rev Clin Oncol 2016;13:534–6. et al. Early results from Children's Oncology Group (COG) ARST08P1: 36. Anderson JL, Park A, Akiyama R, Tap WD, Denny CT, Federman N. pilot studies of cixutumumab or temozolomide with intensive multiagent Evaluation of in vitro activity of the class I PI3K inhibitor buparlisib chemotherapy for patients with metastatic rhabdomyosarcoma (RMS). (BKM120) in pediatric bone and soft tissue sarcomas. PLoS One 2015; J Clin Oncol 2015;33:10015-. 10:e0133610. 18. Cassinelli G, Zuco V, Petrangolini G, De Cesare M, Tortoreto M, Lanzi C, 37. Guenther MK, Graab U, Fulda S. Synthetic lethal interaction between PI3K/ et al. The curative efficacy of namitecan (ST1968) in preclinical models of Akt/mTOR and Ras/MEK/ERK pathway inhibition in rhabdomyosarcoma. pediatric sarcoma is associated with antiangiogenic effects. Biochem Phar- Cancer Lett 2013;337:200–9. macol 2012;84:163–71. 38. Renshaw J, Taylor KR, Bishop R, Valenti M, De Haven Brandon A, Gowan S, 19. Cassinelli G, Lanzi C, Tortoreto M, Cominetti D, Petrangolini G, et al. Dual blockade of the PI3K/AKT/mTOR (AZD8055) and RAS/MEK/ Favini E, et al. Antitumor efficacy of the heparanase inhibitor ERK (AZD6244) pathways synergistically inhibits rhabdomyosarcoma SST0001 alone and in combination with antiangiogenic agents in the cell growth in vitro and in vivo. Clin Cancer Res 2013;19:5940–51. treatment of human pediatric sarcoma models. Biochem Pharmacol 39. Shimizu T, Tolcher AW, Papadopoulos KP, Beeram M, Rasco DW, Smith LS, 2013;85:1424–32. et al. The clinical effect of the dual-targeting strategy involving PI3K/AKT/ 20. ChisholmJC,MerksJHM,CasanovaM,BisognoG,OrbachD,Gentet mTOR and RAS/MEK/ERK pathways in patients with advanced cancer. JC, et al. Open-label, multicentre, randomised, phase II study of Clin Cancer Res 2012;18:2316–25. the EpSSG and the ITCC evaluating the addition of bevacizumab 40. Thornton KA, Chen AR, Trucco MM, Shah P, Wilky BA, Gul N, et al. A dose to chemotherapy in childhood and adolescent patients with meta- finding study of temsirolimus and liposomal doxorubicin for patients static soft tissue sarcoma (the BERNIE study). Eur J Cancer 2017; with recurrent and refractory bone and soft tissue sarcoma. Int J Cancer 83:177–84. 2013;133:997–1005. 21. Navid F, Baker SD, McCarville MB, Stewart CF, Billups CA, Wu J, et al. Phase 41. Bagatell R, Norris R, Ingle AM, Ahern C, Voss S, Fox E, et al. Phase 1 trial of I and clinical pharmacology study of bevacizumab, sorafenib, and low- temsirolimus in combination with irinotecan and temozolomide in chil- dose cyclophosphamide in children and young adults with refractory/ dren, adolescents and young adults with relapsed or refractory solid recurrent solid tumors. Clin Cancer Res 2013;19:236–46. tumors: a Children's Oncology Group Study. Pediatr Blood Cancer 22. Yamamoto Y, Fukuda K, Fuchimoto Y, Matsuzaki Y, Saikawa Y, Kitagawa Y, 2014;61:833–9. et al. Cetuximab promotes anticancer drug toxicity in rhabdomyosarcomas 42. Morgenstern DA, Marzouki M, Bartels U, Irwin MS, Sholler GL, Gam- with EGFR amplification in vitro. Oncol Rep 2013;30:1081–6. mon J, et al. Phase I study of vinblastine and sirolimus in pediatric

1378 Mol Cancer Ther; 17(7) July 2018 Molecular Cancer Therapeutics

patients with recurrent or refractory solid tumors. Pediatr Blood Cancer bone and soft tissue sarcomas (BSTS): an Italian Sarcoma Group (ISG) 2014;61:128–33. study. J Clin Oncol 2016;34:11018-. 43. Vo KT, Karski EE, Nasholm NM, Allen S, Hollinger F, Gustafson WC, et al. 63. Federico SM, Stewart E, Coleman JL, Bishop MW, Santana VM, Lam C, Phase 1 study of sirolimus in combination with oral cyclophosphamide et al. Phase I study of talazoparib and irinotecan in children and and topotecan in children and young adults with relapsed and refractory young adults with recurrent/refractory solid tumors. J Clin Oncol solid tumors. Oncotarget 2017;8:23851–61. 2017;35:10542-. 44. Le X, Pugach EK, Hettmer S, Storer NY, Liu J, Wills AA, et al. A novel 64. Lapenna S, Giordano A. Cell cycle kinases as therapeutic targets for cancer. chemical screening strategy in zebrafish identifies common pathways in Nat Rev Drug Discov 2009;8:547–66. embryogenesis and rhabdomyosarcoma development. Development 65. Hugle M, Belz K, Fulda S. Identification of synthetic lethality of PLK1 2013;140:2354–64. inhibition and microtubule-destabilizing drugs. Cell Death Differ 45. Ciccarelli C, Vulcano F, Milazzo L, Gravina GL, Marampon F, Macioce G, 2015;22:1946–56. et al. Key role of MEK/ERK pathway in sustaining tumorigenicity and in 66. Abbou S, Lanvers-Kaminsky C, Daudigeos-Dubus E, L LED, Laplace-Builhe vitro radioresistance of embryonal rhabdomyosarcoma stem-like cell C, Molenaar J, et al. Polo-like kinase inhibitor volasertib exhibits antitumor population. Mol Cancer 2016;15:16. activity and synergy with vincristine in pediatric malignancies. Anticancer 46. Wu X, Xiao H, Wang R, Liu L, Li C, Lin J. Persistent GP130/STAT3 signaling Res 2016;36:599–609. contributes to the resistance of doxorubicin, cisplatin, and MEK inhibitor 67. Stehle A, Hugle M, Fulda S. Eribulin synergizes with Polo-like kinase 1 in human rhabdomyosarcoma cells. Curr Cancer Drug Targets 2016; inhibitors to induce apoptosis in rhabdomyosarcoma. Cancer Lett 16:631–8. 2015;365:37–46. 47. Wu F, Zhang Y, Sun B, McMahon AP, Wang Y. Hedgehog signaling: from 68. Thalhammer V, Lopez-Garcia LA, Herrero-Martin D, Hecker R, Laubscher basic biology to cancer therapy. Cell Chem Biol 2017;24:252–80. D, Gierisch ME, et al. PLK1 phosphorylates PAX3-FOXO1, the inhibition of 48. Almazan-Moga A, Zarzosa P, Molist C, Velasco P, Pyczek J, Simon-Keller K, which triggers regression of alveolar Rhabdomyosarcoma. Cancer Res et al. Ligand-dependent Hedgehog pathway activation in Rhabdomyosar- 2015;75:98–110. coma: the oncogenic role of the ligands. Br J Cancer 2017;117:1314–25. 69. Weiss GJ, Jameson G, Von Hoff DD, Valsasina B, Davite C, Di Giulio C, et al. 49. Srivastava RK, Kaylani SZ, Edrees N, Li C, Talwelkar SS, Xu J, et al. GLI Phase I dose escalation study of NMS-1286937, an orally available Polo- inhibitor GANT-61 diminishes embryonal and alveolar rhabdomyosar- Like Kinase 1 inhibitor, in patients with advanced or metastatic solid coma growth by inhibiting Shh/AKT-mTOR axis. Oncotarget 2014;5: tumors. Invest New Drugs 2018;36:85–95. 12151–65. 70. Kahen E, Yu D, Harrison DJ, Clark J, Hingorani P, Cubitt CL, et al. 50. Graab U, Hahn H, Fulda S. Identification of a novel synthetic lethality of Identification of clinically achievable combination therapies in childhood combined inhibition of hedgehog and PI3K signaling in rhabdomyosar- rhabdomyosarcoma. Cancer Chemother Pharmacol 2016;78:313–23. coma. Oncotarget 2015;6:8722–35. 71. Stewart E, Federico SM, Chen X, Shelat AA, Bradley C, Gordon B, et al. 51. Meister MT, Boedicker C, Graab U, Hugle M, Hahn H, Klingebiel T, et al. Orthotopic patient-derived xenografts of paediatric solid tumours. Nature Arsenic trioxide induces Noxa-dependent apoptosis in rhabdomyosarco- 2017;549:96–100. ma cells and synergizes with antimicrotubule drugs. Cancer Lett 2016; 72. Leijen S, Geel RMJMv, Pavlick AC, Tibes R, Rosen L, Razak ARA, et al. Phase I 381:287–95. study evaluating WEE1 inhibitor AZD1775 as monotherapy and in com- 52. Boehme KA, Zaborski JJ, Riester R, Schweiss SK, Hopp U, Traub F, et al. bination with gemcitabine, cisplatin, or carboplatin in patients with Targeting hedgehog signalling by arsenic trioxide reduces cell growth advanced solid tumors. J Clin Oncol 2016;34:4371–80. and induces apoptosis in rhabdomyosarcoma. Int J Oncol 2016;48: 73. Montoya J, Lee DW, Aleem E. Abstract 2349: sensitivity and resistance to 801–12. cell cycle and IGF-1R inhibitors in rhabdomyosarcoma. Cancer Res 53. Schleicher SB, Zaborski JJ, Riester R, Zenkner N, Handgretinger R, Kluba T, 2017;77:2349-. et al. Combined application of arsenic trioxide and lithium chloride 74. Francis AM, Alexander A, Liu Y, Vijayaraghavan S, Low KH, Yang D, et al. augments viability reduction and apoptosis induction in human rhabdo- CDK4/6 inhibitors sensitize Rb-positive sarcoma cells to wee1 kinase myosarcoma cell lines. PLoS One 2017;12:e0178857. inhibition through reversible cell-cycle arrest. Mol Cancer Ther 2017; 54. Basit F, Humphreys R, Fulda S. RIP1 protein-dependent assembly of a 16:1751–64. cytosolic cell death complex is required for inhibitor of apoptosis (IAP) 75. Rengaswamy V, Zimmer D, Suss R, Rossler J. RGD liposome-protamine- inhibitor-mediated sensitization to lexatumumab-induced apoptosis. siRNA (LPR) nanoparticles targeting PAX3-FOXO1 for alveolar rhabdo- J Biol Chem 2012;287:38767–77. myosarcoma therapy. J Control Rel 2016;235:319–27. 55. Ueno T, Uehara S, Nakahata K, Okuyama H. Survivin selective inhibitor 76. Cote GM, Choy E. Role of epigenetic modulation for the treatment of YM155 promotes cisplatininduced apoptosis in embryonal rhabdomyosarcoma. Curr Treat Options Oncol 2013;14:454–64. sarcoma. Int J Oncol 2016;48:1847–54. 77. Heinicke U, Fulda S. Chemosensitization of rhabdomyosarcoma cells by 56. Preuss E, Hugle M, Reimann R, Schlecht M, Fulda S. Pan-mammalian target the histone deacetylase inhibitor SAHA. Cancer Lett 2014;351:50–8. of rapamycin (mTOR) inhibitor AZD8055 primes rhabdomyosarcoma 78. Heinicke U, Kupka J, Fulda S. JNJ-26481585 primes rhabdomyosarcoma cells for ABT-737-induced apoptosis by down-regulating Mcl-1 protein. cells for chemotherapeutics by engaging the mitochondrial pathway of J Biol Chem 2013;288:35287–96. apoptosis. Oncotarget 2015;6:37836–51. 57. Lord CJ, Ashworth A. PARP inhibitors: synthetic lethality in the clinic. 79. Jarzab A, Luszczki JJ, Guz M, Gumbarewicz E, Polberg K, Stepulak A. Science 2017;355:1152–8. Additive interaction of cisplatin and histone deacetylase inhibitors com- 58. Brandsma I, Fleuren EDG, Williamson CT, Lord CJ. Directing the use of bined treatment in rhabdomyosarcoma cells - an isobolographic analysis. DDR kinase inhibitors in cancer treatment. Expert Opin Investig Drugs Anticancer Res 2017;37:1067–74. 2017;26:1341–55. 80. Dumont SN, Yang D, Dumont AG, Reynoso D, Blay JY, Trent JC. Targeted 59. Fam HK, Walton C, Mitra SA, Chowdhury M, Osborne N, Choi K, et al. polytherapy in small cell sarcoma and its association with doxorubicin. TDP1 and PARP1 deficiency are cytotoxic to rhabdomyosarcoma cells. Mol Oncol 2014;8:1458–68. Mol Cancer Res 2013;11:1179–92. 81. Brown RE, Buryanek J, Katz AM, Paz K, Wolff JE. Alveolar rhabdomyosar- 60. Norris RE, Adamson PC, Nguyen VT, Fox E. Preclinical evaluation of the coma: morphoproteomics and personalized tumor graft testing further PARP inhibitor, olaparib, in combination with cytotoxic chemotherapy in define the biology of PAX3-FKHR(FOXO1) subtype and provide targeted pediatric solid tumors. Pediatr Blood Cancer 2014;61:145–50. therapeutic options. Oncotarget 2016;7:46263–72. 61. Smith MA, Reynolds CP, Kang MH, Kolb EA, Gorlick R, Carol H, et al. 82. Haydn T, Metzger E, Schuele R, Fulda S. Concomitant epigenetic targeting Synergistic activity of PARP inhibition by talazoparib (BMN 673) with of LSD1 and HDAC synergistically induces mitochondrial apoptosis in temozolomide in pediatric cancer models in the pediatric preclinical rhabdomyosarcoma cells. Cell Death Dis 2017;8:e2879. testing program. Clin Cancer Res 2015;21:819–32. 83. Choy E, Flamand Y, Balasubramanian S, Butrynski JE, Harmon DC, George 62. Grignani G, D'Ambrosio L, Pignochino Y, Palmerini E, Zucchetti M, S, et al. Phase 1 study of oral abexinostat, a histone deacetylase inhibitor, in Boccone P, et al. A phase 1b trial with the combination of trabectedin combination with doxorubicin in patients with metastatic sarcoma. Cancer and olaparib in relapsed patients (pts) with advanced and unresectable 2015;121:1223–30.

www.aacrjournals.org Mol Cancer Ther; 17(7) July 2018 1379

84. Vitfell-Rasmussen J, Judson I, Safwat A, Jones RL, Rossen PB, Lind-Hansen 86. May CD, Beckmann R, Blosser W, Dowless M, VanWye A, Burke T, et al. M, et al. A phase I/II clinical trial of belinostat (PXD101) in combination Abstract 2458: targeting checkpoint kinase 1 (CHK1) with the small with doxorubicin in patients with soft tissue sarcomas. Sarcoma molecule inhibitor LY2606368 mesylate monohydrate in models of 2016;2016:2090271. high-risk pediatric cancer yields signiﬁcant antitumor effects. Cancer Res 85. Krishnadas DK, Bao L, Bai F, Chencheri SC, Lucas K. Decitabine 2016;76:2458-. facilitates immune recognition of sarcoma cells by upregulating 87. Ciarapica R, Carcarino E, Adesso L, De Salvo M, Bracaglia G, Leoncini PP, CT antigens, MHC molecules, and ICAM-1. Tumour Biol 2014;35: et al. Pharmacological inhibition of EZH2 as a promising differentiation 5753–62. therapy in embryonal RMS. BMC Cancer 2014;14:139.

1380 Mol Cancer Ther; 17(7) July 2018 Molecular Cancer Therapeutics

Anke E.M. van Erp, Yvonne M.H. Versleijen-Jonkers, Winette T.A. van der Graaf, et al.

Mol Cancer Ther 2018;17:1365-1380.

Updated version Access the most recent version of this article at: http://mct.aacrjournals.org/content/17/7/1365

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2021/03/09/17.7.1365.DC1

Cited articles This article cites 86 articles, 19 of which you can access for free at: http://mct.aacrjournals.org/content/17/7/1365.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/17/7/1365.full#related-urls

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/17/7/1365. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.