New Strategies in Sarcoma Therapy: Linking Biology and Novel Agents

Published OnlineFirst August 28, 2012; DOI: 10.1158/1078-0432.CCR-12-0875

Clinical Cancer CCR New Strategies Research

Katherine A. Janeway1 and Robert G. Maki2

Abstract The ability to better interrogate the genetic state of a given cancer is giving rise to a new paradigm in cancer therapeutics in which the specific genetic alterations that give rise to the cancer inform the therapeutic decision-making for that specific patient. Sarcomas of soft tissue and bone represent model diseases that underscore this paradigm. However, many barriers prevent linkage of one of the 75 or more different types of sarcoma to novel therapeutic agents. In the present perspective, the authors outline key therapeutic opportunities and hurdles in clinical sarcoma research, focusing on specific examples of sarcomas that are on the verge of new breakthroughs, as well as those in which promise has not lived up to expectations. Focused clinical trial design, ideally with several biomarker or histology-specific arms, is one means to be simultaneously parsimonious and inclusive. Clin Cancer Res; 18(21); 5837–44. 2012 AACR.

Background cancers (Fig. 1). The most common sarcomas that arise in Sarcomas represent a family of rare cancers of bone and children are aneuploid sarcomas, such as osteogenic sarco- soft tissue accounting for less than 1% of cancer in adults ma as well as those with specific translocations, such as and approximately 15% of pediatric cancer (1). They are Ewing sarcoma and alveolar rhabdomyosarcoma (ARMS). biologically heterogeneous, with more than 50 histologic In adults, the most common sarcomas are usually aneu- subtypes identified in soft tissue and more than 20 in bone, ploid, including leiomyosarcoma, and undifferentiated in keeping with the multiple types of connective tissue that pleomorphic sarcoma (UPS, formerly termed MFH, malig- comprise the scaffolding of the human body (2). nant fibrous histiocytoma), whereas other sarcomas that KIT Given their heterogeneity, a number of approaches have have specific genetic changes, such as mutation in SS18-SSX been used to characterize them, including histology (e.g., gastrointestinal stromal tumor (GIST) or the the 3 classes of liposarcoma), genetics (aneuploid sarcomas translocations seen in synovial sarcoma, are collectively vs. those with driver mutations or specific chromosomal somewhat less common. translocations), anatomic primary site, or whether they are In this commentary, we highlight the hurdles that face more prevalent in children or adults. In general, while clinical investigations into sarcomas more common in children who get sarcomas will most typically receive multi- children and adults, as well as new therapeutic options, modality therapy with surgery, chemotherapy, and radia- some of which have been decades in the making. Tantaliz- tion, in adults, the use of chemotherapy in the primary ing recent discoveries place us on the cusp of a new era disease setting is not as widespread, given the more limited linking tumor genetics to specific therapeutics for at least a benefit observed with the use of adjuvant systemic therapy few of the histologies discussed herein, but the ballast in many subtypes (3). otherwise hindering more rapid clinical research may prove The biology of sarcomas is interesting due to the involve- difficult to jettison. ment of particular signaling pathways in specific sarcoma subtypes (4, 5), many of which have parallels in other Sarcomas more common in children The most common histologies of sarcoma treated in children include osteogenic sarcoma, Ewing sarcoma, rhabdomyosarcoma (alveolar and embryonal subtypes, primar- Authors' Affiliations: 1Deparment of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts; and 2Departments of Medicine, ily), and less commonly other sarcomas, such as synovial Orthopaedics, and Jack Martin Fund Division of Pediatric Hematology- sarcoma and desmoplastic small round cell tumor, which as Oncology, Mount Sinai School of Medicine, New York a group are termed nonrhabdomyosarcoma soft tissue Prior presentation: Presented in part at the American Association for sarcomas. New therapeutics stand to impact upon the way Cancer Research Annual Meeting, Chicago, IL, April 2012. we treat each of the most common childhood sarcoma Corresponding Author: Robert G. Maki, Departments of Medicine, Ortho- diagnoses (Table 1). paedics, and Jack Martin Fund Division of Pediatric Hematology-Oncol- ogy, Mount Sinai School of Medicine, 1 Gustave Levy Place, Box 1208, Osteosarcoma. Outcomes for newly diagnosed patients New York 10029. Phone: 212-241-7022; Fax: 212-360-6921; E-mail: with osteosarcoma had not changed in over 20 years (6). [email protected] The lack of progress reflects the dearth of effective newer doi: 10.1158/1078-0432.CCR-12-0875 agents, save for the immunotherapeutic mifamurtide, 2012 American Association for Cancer Research. which is now approved outside the United States and

www.aacrjournals.org 5837

Janeway and Maki

Target Neoplasm Sarcoma subtype VEGF GIST Breast carcinoma RTK Ewing sarcoma Others RAS- Desmoid tumor? RAF Colorectal carcinoma Angiosarcoma MEK VEGF Solitary fibrous tumor? Thyroid carcinoma Alveolar soft part sarcoma? Others Clear cell sarcoma? ERK Autocrine, Akt paracrine loops Dermatofibrosarcoma Multiple myeloma protuberans Golgi Tenosynovial giant cell tumor

TOR TOR PEComa (perivascular Renal carcinoma epithelial cell tumor)

Endoplasmic ALK, BRAF, other ALK(ϩ) non-small cell reticulum intracellular kinases ALK lung cancer, anaplastic Inflammatory large cell lymphoma, myofibroblastic tumor malignant melanoma p53-HDM2 interaction Dedifferentiated Acute leukemias? liposarcoma

HDACs, ?hsp90 Cutaneous T-cell Synovial sarcoma lymphoma p53 Nuclear hormone HDM2 receptors Breast carcinoma Endometrial stromal HDACs Prostate carcinoma sarcoma ER CDKs Dedifferentiated Chronic leukemias? liposarcoma CDK4 © 2012 American Association for Cancer Research

Figure 1. Schematic of clinically useful targets in sarcomas versus other forms of cancer. MEK, mitogen-activated protein/extracellular signal–regulated kinase; ERK, extracellular signal–regulated kinase; RTK, receptor tyrosine kinase.

referenced later. The challenge in identifying new agents RANKL and its receptor RANK are members of the with activity in osteosarcoma is a function of the complex tumor necrosis factor family of receptors and ligands biology of this disease, challenges in detecting radiographic with the physiologic function of regulating bone turn- responses in tumors with calcified stroma and deficiencies over. Osteosarcoma expresses both RANK and RANKL in risk stratification. While there have been no research and in osteosarcoma, RANK–RANKL interaction activates discoveries pointing to a single essential pathway in oste- downstream signaling and modulates gene expression osarcoma, preclinical studies have identified new agents (10, 11). Interfering with RANK–RANKL interaction in with potential clinical activity. vivo decreases osteosarcoma growth and metastatic tumor Two agents of significant interest are antidisialoganglio- burden (12–14). These data will hopefully build on a side (GD2) antibody and receptor activator of NF-kB clinical trial in giant cell tumor of bone (GCTB), in which (ligand; RANKL) antibody denosumab. GD2 is a cell surface interruption of RANKL signaling by virtue of denosumab sialic acid–containing glycosphingolipid expressed normal- led to rapid and sustained radiological changes with ly in central nervous system, peripheral nerves, and mela- clinical benefit (pain relief) in patients with recurrent or nocytes. In neuroblastoma anti-GD2 antibody, Ch14.18 metastatic GCTB (15). significantly improves event free and overall survival (7). Ewing sarcoma. Ewing sarcoma is a small round blue More than 90% of osteosarcomas express GD2 (8). In the cell tumor that arises from bone or soft tissue and is phase I/Ib trial of the murine anti-GD2 antibody, 14.2G2a, characterized by a translocation involving EWSR1 and one plus IL-2, 1 of 2 patients with osteosarcoma treated had a of several genes in the E twenty-six family of transcription complete response persisting for 8 months (9). factors, most often FLI1. The EWSR1-FLI1 translocation is

5838 Clin Cancer Res; 18(21) November 1, 2012 Clinical Cancer Research

Linking Sarcoma Biology to Novel Therapeutics

Table 1. New targets and new agents for pediatric sarcomas

Diagnosis Agent(s) of interest Clinical trials open or pending Osteosarcoma Anti GD2 mAb IGF1R mAb þ TOR inhibitor Denosumab Multitargeted TKI Mifamurtide Ewing sarcoma IGF1R mAb þ chemotherapy IGF1R mAb þ TOR inhibitor Phase II PARP inhibitor or phase I in combinations Rhabdomyosarcoma Sonic hedgehog inhibitor IGF1R mAb þ TOR inhibitor Nonrhabdomyosarcoma soft tissue sarcoma IGF1R mAb Crizotinib (IMT) Lisitinib (GIST)

Abbreviations: mAb, monoclonal antibody; TKI, tyrosine kinase inhibitor; TOR, (mammalian) target of rapamycin.

present in 85% of Ewing sarcoma. The insulin-like growth cations between the DNA-binding domain of either PAX3 or factor (IGF) pathway has been shown to be important in PAX7 and the transactivation domain of FOXO1. ARMS has Ewing sarcoma. High expression of insulin-like growth a poorer prognosis and more commonly occurs in older factor 1 (IGF-1) and insulin-like growth factor 1 receptor children, adolescents, and young adults. As is seen in Ewing (IGF1R) is seen in Ewing sarcoma. The EWS-FLI1 translo- sarcoma, preclinical and clinical data implicated IGF signal- cation is associated with decreased IGFBP-3 expression, ing as essential in rhabdomyosarcoma. Both alveolar and resulting in increased IGF1R signaling. embryonal rhabdomyosarcoma (ERMS) have high expres- Inhibiting the IGF pathway decreases Ewing sarcoma sion of IGF-II and IGF1R through diverse mechanisms. Loss proliferation both in vitro and in vivo (16). In a phase II of imprinting at IGF2 locus is present in ERMS. The fusion study of 106 patients with recurrent or refractory Ewing transcription factor PAX3-FOXO1 targets the IGF1R pro- sarcoma, anti-IGF1R antibody figitumumab produced a moter (16). A phase II trial evaluating the feasibility of 14% response rate (17). In a phase II study of R1507, a adding cixutumumab, an IGF1R antibody, with or without different IGF1R monoclonal antibody, 115 patients were temsirolimus, to intensive chemotherapy is underway (24). enrolled and the response rate was 10% (18). The low Inflammatory myofibroblastic tumor. Inflammatory response rates exemplify the need to identify a biomarker myofibroblastic tumor (IMT) is a rare, locally aggressive that is associated with disease responsiveness and to exam- tumor. It is biologically heterogeneous but was found to ine the signaling feedback loops that may promote resis- harbor translocations involving ALK in 50% of patients, tance. There is a great interest in further studying the activity particularly younger patients (25, 26). Crizotinib is a of IGF1R antibodies in Ewing sarcoma, highlighted by the small-molecule inhibitor of anaplastic lymphoma kinase success of the R1507 trial to bring together pediatric oncol- (ALK). In ALK-translocation positive non–small cell lung ogists and medical oncologists on one common study, cancer, crizotinib has substantial activity and is being eval- something that has been difficult to achieve in practice. uated in phase III trials. In the case of non–small cell lung However, planning additional IGF1R inhibitor–based trials cancer, even though only 5% have ALK mutations the total will be challenging, because many pharmaceutical compa- number of patients per year in the United States with non– nies canceled their IGF1R antibody programs when trials in small cell lung cancer is 7,500, making a number sufficient carcinomas failed to show activity (19). for randomized studies. Conversely, IMT affects perhaps 50 In 2012, 2 articles highlighted the possible use of PARP patients in the United States annually, a number too small to inhibitors in Ewing sarcoma (20, 21), recalling a basic examine in randomized trials. Nonetheless, activity of crizo- science finding from more than a decade earlier (22). While tinib was shown by allowing patients with IMT to be accrued a phase III trial has been completed in breast cancer, PARP to phase I studies in adults (27), and children (28, 29). inhibitors have not been systematically examined in sarcomas. To begin to remedy this deficiency, a phase II trial of Sarcomas more common in adults olaparib in patients with relapsed and recurrent Ewing Scope of the problem. Sarcomas are more uncommon in sarcoma is enrolling patients (23), and further clinical trials adults relative to other cancers, and are more heteroge- are anticipated. neous. Sarcoma subtypes are in part a function of anatomic Rhabdomyosarcoma. Rhabdomyosarcomas are malig- primary site and patient age. These data alone can direct one nancies of skeletal muscle most commonly affecting chil- to a specific differential diagnosis. GIST, UPS, leiomyosar- dren under age of 5 years but also occurring in adolescents coma, and the 3 forms of liposarcoma are the most com- and young adults. There are 2 major histologic subtypes, mon subtypes in adults, although versions of sarcomas alveolar and embryonal. ARMS is characterized by translo- more common in children are also seen, with interesting

www.aacrjournals.org Clin Cancer Res; 18(21) November 1, 2012 5839

Janeway and Maki

clinical variations. For example, Ewing sarcoma is much manuscript, which will further help define who may best more common in soft tissue in adults than in bone, the most benefit from longer courses of imatinib. common site in children, and rhabdomyosarcoma is far An interesting metabolic pathway genetic alteration can more commonly the pleomorphic subtype than in children, lead to a form of GIST that occurs predominantly in chil- in which embryonal and alveolar subtypes predominate. dren and young adults. In this "SDH-negative" form of Despite their heterogeneity, over the last several years a GIST, loss of succinate dehydrogenase (SDH) expression is variety of novel agents have been found to be active in observed, and may ultimately provide new means to treat specific sarcoma subtypes. These are outlined in Table 2; both these so-called "pediatric," syndromic GISTs, and some are described in greater detail later. The same issues much more common KIT or PDGFRA mutant GIST (32, with respect to pediatric sarcomas apply, although for some 33). Lisitinib will undergo evaluation for SDH-negative more common sarcomas there are sufficient numbers of GISTs; in a similar way, crenalonib is being studied in the patients to complete randomized clinical trials. relatively rare D842V PDGFRA mutant GIST. Gastrointestinal stromal tumor. GIST remains the best Synovial sarcoma. Synovial sarcoma has been under example of a sarcoma in which the use of a kinase-directed close study to understand how its SS18-SSX translocation agent led to impressive clinical results (30). This topic is well product drives the phenotype of this cancer. The cell of addressed elsewhere, although notably for the first time a origin, hitherto unknown, was suggested to be the satellite survival advantage was shown for use of imatinib in the cell of skeletal muscle, based on the context-dependent adjuvant setting (31). Patients who received 3 years of tumor growth seen when the same transgene was intro- imatinib had improved 5-year survival as compared with duced into different cell types (34). Work by Ladanyi and patients who received only 1 year of therapy. It will be colleagues highlighted how monophasic and biphasic vari- interesting to see if there is any relationship between geno- eties of the cancer develop based on the SS18-SSX subtype type and outcome, not thoroughly discussed in the initial (35).

Table 2. Status of newer agents in different sarcoma subtypes

Target Agent(s) Sarcoma subtype Activity shown in clinical trials or case series KIT, PDGFR Imatinib, sunitinib, GIST sorafenib, regorafenib, dasatinib Autocrine signaling loop(?): Imatinib Dermatoﬁbrosarcoma protuberans, PDGF, CSF1 tenosynovial giant cell tumor/ pigmented villonodular synovitis RANK–RANKL Denosumab GCTB VEGF and/or VEGFR Bevacizumab, sorafenib, Angiosarcoma, solitary ﬁbrous tumor/ 1, 2, and/or 3 sunitinib, pazopanib, hemangiopericytoma, alveolar soft cediranib, others part sarcoma, clear cell sarcoma IGF1R IGF1R monoclonal antibodies Ewing sarcoma, desmoplastic small round cell tumor ER Tamoxifen Endometrial stromal sarcoma mTOR (TORC1) Sirolimus, other TOR inhibitors Perivascular epithelial cell tumor family (PEComa) ALK Crizotinib IMT Unknown Trabectedin Myxoid-round cell liposarcoma HDM2 RG7112 WD–DD LS CDK4 PD0332991 WD–DD LS Unknown Sorafenib, imatinib, other Desmoid tumor multitargeted kinase inhibitors NY-ESO-1 Autologous NY-ESO-1 T cells Synovial sarcoma

Activity suggested in individual case reports or translational science articles EGFR family Geﬁtinib, cetuximab Chordoma (Histone) deacetylase Vorinostat Synovial sarcoma 26S Proteasome Bortezomib GIST

Abbreviations: CSF1, colony stimulating factor 1; EGFR, EGF receptor; ER, estrogen receptor; PDGF(R), platelet-derived growth factor (receptors); TORC1, (mammalian) target of rapamycin complex 1.

5840 Clin Cancer Res; 18(21) November 1, 2012 Clinical Cancer Research

Linking Sarcoma Biology to Novel Therapeutics

After several years of work leading to the demonstration Ewing sarcoma will be conducted in adults because none of that hsp90 inhibitors and histone deacetylase (HDAC) the PARP inhibitors have yet been studied in phase I in inhibitors could be a useful option for this sarcoma subtype children, preventing an inclusive trial with children and (36–39), Nielsen and colleagues begun to decipher the adults until the appropriate pediatric phase I has been mechanism of this approach in a publication on the bio- conducted. chemistry of the SYT-SSX protein and how it links tran- Because most sarcomas in children are treated with multi- scription factors ATF2 and TLE1 (40). Gene knockdown or modality therapy including chemotherapy, an additional an HDAC inhibitor decreases synovial sarcoma growth and challenge is faced after a novel agent has been shown to have causes apoptosis. These studies provide an obvious inroad single-agent efficacy in sarcoma. To be used in newly diag- to clinical trials using HDAC inhibitors in synovial sarcoma. nosed patients, the novel agent must be integrated with The relative lack of overlapping toxicity of HDAC inhibitors existing therapy, which may produce unanticipated toxicity. with cytotoxic agents or kinase-directed agents may allow This difficulty has been addressed with at least 2 for their combination. Such combination studies are under- approaches, including studying combinations in poor prog- way, although as noted earlier none are presently sarcoma- nostic populations, or studying the agent in a maintenance specific. fashion. The Children’s Oncology Group (COG) study Well differentiated-dedifferentiated liposarcoma. A dif- ARST08P1 will combine cixutumumab or temozolomide ferent angle on deacetylase inhibitors, as well as other or both in patients with high-risk rhabdomyosarcoma, novel-targeted agents, arises in a discussion of well differ- looking at toxicity to see if it will be possible to incorporate entiated–dedifferentiated liposarcoma (WD–DD LS), one either agent in the next phase III study in rhabdomyosar- of the most common forms of sarcoma. WD–DD LS is a coma (45). frustrating diagnosis that often occurs in the abdomen/ Like many sarcomas, IMT also highlights a third issue retroperitoneum and recurs again and again, with deaths common with pediatric and adult sarcomas, specifically the typically from local disease progression rather than overt lack of an appropriate model system to serve as a surrogate metastases. A careful analysis from Taylor and colleagues for patients. Preclinical data are helpful, even essential, to identified 8% of DD LS with mutations in HDAC1 (41), spur clinical trials, as has happened in pediatric and adult suggesting there will be other epigenetic mechanisms by sarcomas alike. In an effective model system, there is also which WD–DD LS requires to survive with what are hun- ideally a biomarker of drug activity that can be used in the dreds or thousands of copies of the same sequencing encod- disease of interest. With lack of an effective model, we are ing HDM2, CDK4, and neighboring genes on chromosome reduced to paraffin tumor samples or frozen tissue, from 12q (5). which limited data can be gleaned. Fortunately, a number of These data underscore the well-recognized amplification models of sarcomas have been developed in recent years of chromosome 12q, the mechanism of which is still (34, 46, 47) and are currently being used for preclinical somewhat mysterious (42). This characteristic amplifica- research. tion brings into focus the use of human homologue of Key elements of trial design, such as biomarker-driven murine double minute 2 (HDM2) or cyclin-dependent studies, and the collaboration of investigators between kinase 4 (CDK4) inhibitors in this form of liposarcoma centers or between disciplines, for example, pediatric and (43). Phase I–II data have been collected for patients with medical oncology, are equally important in linking a diag- WD–DD LS treated with HDM2 inhibitor RG7112 before nosis with a new therapeutic. The COG is the only National surgery, and CDK4 inhibitor PD0332991 has at least minor Cancer Institute (NCI)–funded pediatric cancer cooperative activity in phase I of WD–DD LS in a recent clinical trial group; COG and its predecessors have served as the conduit (44). While no agent has yielded the dramatic results shown for a number of phase I–II–III cancer clinical trials in with kinase-directed therapeutics in GIST, these agents sarcoma (24) and routinely permits adult patients with represent measurable advances upon which future investi- appropriate diagnoses to participate in its studies. Sarcoma gations can build. Alliance for Research through Collaboration’s study of IGF1R antibody R1507 enrolled both children and adults, Challenges and opportunities in identifying novel and is a good example of cooperation between pediatric and agents for sarcomas medical oncologists (18). Another example of cooperation A number of challenges face clinical investigators inter- between pediatric and medical oncologists in preclinical ested in studying novel agents in sarcomas. One such research is the identification and characterization of SDH- challenge is access to new agents. In the case of IGF1R deficient GIST. Finally, the use of a biomarker to identify antibodies, academic researchers and cooperative groups patients for priority enrollment on early-phase trials were interested in designing further trials of IGF1R anti- allowed the study of crizotinib in children with IMT (29). bodies in Ewing sarcoma, but many pharmaceutical com- panies ceased developing IGF1R antibodies because of disappointing trial results in carcinoma. Furthermore, com- On the Horizon plicating access to drugs for clinical trials in sarcomas is the The most common sarcomas, aneuploid tumors in adults issue of early phase trials of novel agents in pediatric more than 50 years of age, have shown little responsiveness patients. The initial phase II trial of PARP inhibitors in to kinase-directed agents studied in the past decade, and

www.aacrjournals.org Clin Cancer Res; 18(21) November 1, 2012 5841

Janeway and Maki

Table 3. A "most wanted" list of sarcoma tive to Institutional Review Boards and clinical trials offices that look to accrual as a metric for keeping a study open. As a diagnoses in need of new agents result, a design incorporating multiple histologies is pre- Leiomyosarcoma ferred over several independent small studies. The choice of UPS clinical trial design is important, given what proved to be a Osteogenic sarcoma large number of patients treated with an ineffective agent in Chondrosarcoma one randomized discontinuation phase II study in a group Desmoplastic small round cell tumor of unselected sarcomas (56) and remains an active topic of Extraskeletal myxoid chondrosarcoma discussion. WD–DD LS Ideally, studies will be conducted with pre- and posttreat- Clear cell sarcoma ment biopsies to gain the maximum data form the smallest number of patients, which can ethically only be done in adult patients, if at all. A less invasive approach employs have proved frustrating in the era of kinase-directed therapy correlative radiological studies, such as 2[18F]fluoro-2- and GIST, although there is equal frustration for a variety of deoxy-D-glucose (FDG)- or fluoro-L-thymidine–positron translocation-associated sarcomas as well (Table 3). emission tomography (FLT-PET) as part of the conduct of There are glimmers of hope that aneuploid sarcomas will these clinical trials, which has proved a useful approach in respond to newer generations of agents, such as some of the assessing the response to therapy in studies of cytotoxic and otherwise chemotherapy refractory diagnoses that are kinase-directed agents alike (57, 58). Small steps do add up, shown to respond to kinase inhibitors. Osteosarcoma ther- and this incremental approach, driven by careful drug/ apy has been changed, at least in Europe, with the demon- histology selections on the basis of biology, will hopefully stration of the activity if mifurmatide in a phase III clinical lead to greater progress in this rare and diverse group of trial (48, 49), work that was propelled by translational cancers. research (50). A fraction of osteosarcomas respond to multitargeted tyrosine kinase inhibitors (51). Furthermore, pazo- Disclosure of Potential Conflicts of Interest panib was approved in 2012 for use in patients with soft R.G. Maki has consulting fees from GlaxoSmithKline, Hoffman-LaRoche, tissue sarcomas other than GIST or liposarcoma that are Lilly, Pfizer, Merck, and Eisai; honoraria from Novartis and Ziopharm; commercial research support from Eisai and Ziopharm; is a consultant/ refractory to other systemic therapeutics (52). While the advisory board member of Eisai, Novartis, Pfizer, Merck, GlaxoSmithKline, mechanism of action of pazopanib in sarcomas in general is Morphotek, 23 & me, n-of-one, Foundation Medicine, and Bayer; and has expert testimony from Pfizer: ODAC Meeting. No potential conflicts of unknown, VEGF receptor (VEGFR) blockade may be impor- interest were disclosed by K.A. Janeway. tant for the observed responses of alveolar soft part sarcoma and clear cell sarcoma to sunitinib and of alveolar soft part Authors' Contributions sarcoma to cediranib (53, 54; Table 2), which may be a Conception and design: K.A. Janeway, R.G. Maki general phenomenon for VEGFR inhibitors as with renal Writing, review, and/or revision of the manuscript: K.A. Janeway, R.G. cell carcinoma. Finally, leiomyosarcomas have upregula- Maki Administrative, technical, or material support (i.e., reporting or orga- tion of kinases and markers of inflammation that may lead nizing data, constructing databases): R.G. Maki to less toxic therapeutic options (55). Grant Support A prescription for efficient sarcoma clinical trials This study is supported by the Timothy O’Brien Fund (to K.A. Janeway), Small studies of 10 to 20 patients with a specific diagnosis the St. Baldrick’s Foundation (to K.A. Janeway), and Hyundai Hope on can rule in or rule out substantial activity (e.g., a 30% Wheels (to R.G. Maki). RECIST response rate) of a new drug. At the same time, Received June 26, 2012; revised August 7, 2012; accepted August 10, 2012; small histology or biomarker-specific studies are unattrac- published OnlineFirst August 28, 2012.

References 1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer 6. Mirabello L, Troisi RJ, Savage SA. Osteosarcoma incidence and J Clin 2012;62:10–29. survival rates from 1973 to 2004: data from the Surveillance, Epide- 2. Fletcher CDM, Unni KK, Mertens F, editors. Pathology and genetics of miology, and End Results Program. Cancer 2009;115:1531–43. tumours of soft tissue and bone. Lyon, France: IARC Press; 2002. 7. Yu AL, Gilman AL, Ozkaynak MF, London WB, Kreissman SG, Chen 3. Clark MA, Fisher C, Judson I, Thomas JM. Soft-tissue sarcomas in HX, et al. Anti-GD2 antibody with GM-CSF, interleukin-2, and isotret- adults. N Engl J Med 2005;353:701–11. inoin for neuroblastoma. N Engl J Med 2010;363:1324–34. 4. Taylor BS, Barretina J, Maki RG, Antonescu CR, Singer S, Ladanyi M. 8. Heiner JP, Miraldi F, Kallick S, Makley J, Neely J, Smith-Mensah WH, Advances in sarcoma genomics and new therapeutic targets. Nat Rev et al. Localization of GD2-specific monoclonal antibody 3F8 in human Cancer 2011;11:541–57. osteosarcoma. Cancer Res 1987;47:5377–81. 5. Barretina J, Taylor BS, Banerji S, Ramos AH, Lagos-Quintana M, 9. Frost JD, Hank JA, Reaman GH, Frierdich S, Seeger RC, Gan J, et al. A DeCarolis PL, et al. Subtype-specificgenomicalterationsdefine phase I/IB trial of murine monoclonal anti-GD2 antibody 14.G2a plus new targets for soft-tissue sarcoma therapy. Nat Genet 2010;42: interleukin-2 in children with refractory neuroblastoma: a report of the 715–21. Children's Cancer Group. Cancer 1997;80:317–33.

5842 Clin Cancer Res; 18(21) November 1, 2012 Clinical Cancer Research

Linking Sarcoma Biology to Novel Therapeutics

10. Mori K, Le Goff B, Berreur M, Riet A, Moreau A, Blanchard F, et al. 30. Joensuu H, DeMatteo RP. The management of gastrointestinal stromal Human osteosarcoma cells express functional receptor activator of tumors: a model for targeted and multidisciplinary therapy of malig- nuclear factor-kappa B. J Pathol 2007;211:555–62. nancy. Annu Rev Med 2012;63:247–58. 11. Mori K, Berreur M, Blanchard F, Chevalier C, Guisle-Marsollier I, Masson 31. Joensuu H, Eriksson M, Sundby Hall K, Hartmann JT, Pink D, M, et al. Receptor activator of nuclear factor-kappaB ligand (RANKL) Schutte J, et al. One vs three years of adjuvant imatinib for operable directly modulates the gene expression profile of RANK-positive Saos-2 gastrointestinal stromal tumor: a randomized trial. JAMA 2012;307: human osteosarcoma cells. Oncol Rep 2007;18:1365–71. 1265–72. 12. Lamoureux F, Picarda G, Rousseau J, Gourden C, Battaglia S, Charrier 32. Janeway KA, Weldon CB. Pediatric gastrointestinal stromal tumor. C, et al. Therapeutic efficacy of soluble receptor activator of nuclear Semin Pediatr Surg 2012;21:31–43. factor-kappa B-Fc delivered by nonviral gene transfer in a mouse 33. Janeway KA, Kim SY, Lodish M, Nose V, Rustin P, Gaal J, et al. Defects model of osteolytic osteosarcoma. Mol Cancer Ther 2008;7:3389–98. in succinate dehydrogenase in gastrointestinal stromal tumors lacking 13. Lamoureux F, Richard P, Wittrant Y, Battaglia S, Pilet P, Trichet V, et al. KIT and PDGFRA mutations. Proc Natl Acad Sci U S A 2011;108: Therapeutic relevance of osteoprotegerin gene therapy in osteosar- 314–8. coma: blockade of the vicious cycle between tumor cell proliferation 34. Haldar M, Hancock JD, Coffin CM, Lessnick SL, Capecchi MR. A and bone resorption. Cancer Res 2007;67:7308–18. conditional mouse model of synovial sarcoma: insights into a myo- 14. Akiyama T, Choong PF, Dass CR. RANK-Fc inhibits malignancy via genic origin. Cancer Cell 2007;11:375–88. inhibiting ERK activation and evoking caspase-3-mediated anoikis in 35. Saito T, Nagai M, Ladanyi M. SYT-SSX1 and SYT-SSX2 interfere with human osteosarcoma cells. Clin Exp Metastasis 2010;27:207–15. repression of E-cadherin by snail and slug: a potential mechanism for 15. Thomas D, Henshaw R, Skubitz K, Chawla S, Staddon A, Blay JY, et al. aberrant mesenchymal to epithelial transition in human synovial sar- Denosumab in patients with giant-cell tumour of bone: an open-label, coma. Cancer Res 2006;66:6919–27. phase 2 study. Lancet Oncol 2010;11:275–80. 36. Nguyen A, Su L, Campbell B, Poulin NM, Nielsen TO. Synergism of heat 16. Kolb EA, Gorlick R. Development of IGF-IR inhibitors in pediatric shock protein 90 and histone deacetylase inhibitors in synovial sar- sarcomas. Curr Oncol Rep 2009;11:307–13. coma. Sarcoma 2009;2009:794901. 17. Juergens H, Daw NC, Geoerger B, Ferrari S, Villarroel M, Aerts I, et al. 37. Terry J, Lubieniecka JM, Kwan W, Liu S, Nielsen TO. Hsp90 inhibitor Preliminary efficacy of the anti–insulin-like growth factor type 1 recep- 17-allylamino-17-demethoxygeldanamycin prevents synovial sarco- tor antibody figitumumab in patients with refractory Ewing sarcoma. J ma proliferation via apoptosis in in vitro models. Clin Cancer Res Clin Oncol 2011;29:4534–40. 2005;11:5631–8. 18. Pappo AS, Patel SR, Crowley J, Reinke DK, Kuenkele KP, Chawla SP, 38. Su L, Cheng H, Sampaio AV, Nielsen TO, Underhill TM. EGR1 reac- et al. R1507, a monoclonal antibody to the insulin-like growth factor 1 tivation by histone deacetylase inhibitors promotes synovial sarcoma receptor, in patients with recurrent or refractory Ewing sarcoma family cell death through the PTEN tumor suppressor. Oncogene 2010;29: of tumors: results of a phase II Sarcoma Alliance for Research through 4352–61. Collaboration study. J Clin Oncol 2011;29:4541–7. 39. Lubieniecka JM, de Bruijn DR, Su L, van Dijk AH, Subramanian S, van 19. Fidler MJ, Shersher DD, Borgia JA, Bonomi P. Targeting the insulin-like de Rijn M, et al. Histone deacetylase inhibitors reverse SS18-SSX- growth factor receptor pathway in lung cancer: problems and pitfalls. mediated polycomb silencing of the tumor suppressor early growth Ther Adv Med Oncol 2012;4:51–60. response 1 in synovial sarcoma. Cancer Res 2008;68:4303–10. 20. Garnett MJ, Edelman EJ, Heidorn SJ, Greenman CD, Dastur A, Lau 40. Su L, Sampaio AV, Jones KB, Pacheco M, Goytain A, Lin S, et al. KW, et al. Systematic identification of genomic markers of drug Deconstruction of the SS18-SSX fusion oncoprotein complex: insights sensitivity in cancer cells. Nature 2012;483:570–5. into disease etiology and therapeutics. Cancer Cell 2012;21:333–47. 21. Brenner JC, Feng FY, Han S, Patel S, Goyal SV, Bou-Maroun LM, et al. 41. Taylor BS, Decarolis PL, Angeles CV, Brenet F, Schultz N, Antonescu PARP-1 inhibition as a targeted strategy to treat Ewing's sarcoma. CR, et al. Frequent alterations and epigenetic silencing of differenti- Cancer Res 2012;72:1608–13. ation pathway genes in structurally rearranged liposarcomas. Cancer 22. Soldatenkov VA, Albor A, Patel BK, Dreszer R, Dritschilo A, Notario V. Discov 2011;1:587–97. Regulation of the human poly(ADP-ribose) polymerase promoter by 42. Garsed DW, Holloway AJ, Thomas DM. Cancer-associated neochro- the ETS transcription factor. Oncogene 1999;18:3954–62. mosomes: a novel mechanism of oncogenesis. Bioessays 2009;31: 23. clinicaltrials.gov. Olaparib in adults with recurrent/metastatic Ewing's 1191–200. sarcoma, NCT01583543. [Accessed 2012 June 1; cited 2012 June 1]. 43. Millard M, Pathania D, Grande F, Xu S, Neamati N. Small-molecule Available from: http://clinicaltrials.gov/ct2/show/NCT01583543. inhibitors of p53-MDM2 interaction: the 2006–2010 update. Curr 24. Malempati S, Hawkins DS. Rhabdomyosarcoma: review of the Chil- Pharm Des 2011;17:536–59. dren's Oncology Group (COG) soft-tissue sarcoma committee expe- 44. Flaherty KT, Lorusso PM, Demichele A, Abramson VG, Courtney R, rience and rationale for current COG studies. Pediatr Blood Cancer Randolph SS, et al. Phase I, dose-escalation trial of the oral cyclin- 2012;59:5–10. dependent kinase 4/6 inhibitor PD 0332991, administered using a 21- 25. Bridge JA, Kanamori M, Ma Z, Pickering D, Hill DA, Lydiatt W, et al. day schedule in patients with advanced cancer. Clin Cancer Res Fusion of the ALK gene to the clathrin heavy chain gene, CLTC, in 2012;18:568–76. inflammatory myofibroblastic tumor. Am J Pathol 2001;159:411–5. 45. clinicaltrials.gov. Temozolomide, cixutumumab, and combination 26. Coffin CM, Patel A, Perkins S, Elenitoba-Johnson KS, Perlman E, chemotherapy in treating patients with metastatic rhabdomyosarco- Griffin CA. ALK1 and p80 expression and chromosomal rearrange- ma, trial NCT01055314. [Accessed 2012 June 2; cited 2012 June 2]. ments involving 2p23 in inflammatory myofibroblastic tumor. Mod Available from: http://clinicaltrials.gov/ct2/show/NCT01055314. Pathol 2001;14:569–76. 46. Walkley CR, Qudsi R, Sankaran VG, Perry JA, Gostissa M, Roth SI, et al. 27. Butrynski JE, D'Adamo DR, Hornick JL, Dal Cin P, Antonescu CR, Conditional mouse osteosarcoma, dependent on p53 loss and poten- Jhanwar SC, et al. Crizotinib in ALK-rearranged inflammatory myofi- tiated by loss of Rb, mimics the human disease. Genes Dev 2008; broblastic tumor. N Engl J Med 2010;363:1727–33. 22:1662–76. 28. clinicaltrials.gov. Crizotinib in treating young patients with relapsed or 47. Keller C, Capecchi MR. New genetic tactics to model alveolar rhab- refractory solid tumors or anaplastic large cell lymphoma, trial domyosarcoma in the mouse. Cancer Res 2005;65:7530–2. NCT00939770. [Accessed 2012 June 2; cited 2012 June 2]. Available 48. Meyers PA, Schwartz CL, Krailo MD, Healey JH, Bernstein ML, Betcher from: http://clinicaltrials.gov/ct2/show/NCT00939770. D, et al. Osteosarcoma: the addition of muramyl tripeptide to chemo- 29. Mosse YP, Balis FM, Lim MS, Laliberte J, Voss SD, Fox E, et al. Efficacy therapy improves overall survival—a report from the Children's Oncol- of crizotinib in children with relapsed/refractory ALK-driven tumors ogy Group. J Clin Oncol 2008;26:633–8. including anaplastic large cell lymphoma and neuroblastoma: a Chil- 49. Ando K, Mori K, Corradini N, Redini F, Heymann D. Mifamurtide for the dren's Oncology Group phase I consortium study. J Clin Oncol 30, treatment of nonmetastatic osteosarcoma. Expert Opin Pharmacother 2012 (suppl; abstr 9500). 2011;12:285–92.

www.aacrjournals.org Clin Cancer Res; 18(21) November 1, 2012 5843

Janeway and Maki

50. Kleinerman ES, Maeda M, Jaffe N. Liposome-encapsulated muramyl 55. Edris B, Espinosa I, Muhlenberg T, Mikels A, Lee CH, Steigen SE, et al. tripeptide: a new biologic response modiﬁer for the treatment of ROR2 is a novel prognostic biomarker and a potential therapeutic osteosarcoma. Cancer Treat Res 1993;62:101–7. target in leiomyosarcoma and gastrointestinal stromal tumour. 51. Grignani G, Palmerini E, Dileo P, Asaftei SD, D'Ambrosio L, Pignochino J Pathol 2012;227:223–33. Y, et al. A phase II trial of sorafenib in relapsed and unresectable 56. Schwartz GK, Maki RG, Ratain MJ, Undevia SD, Jones RL, Rudin CM, high-grade osteosarcoma after failure of standard multimodal therapy: et al. Brivanib (BMS-582664) in advanced soft-tissue sarcoma: bio- an Italian Sarcoma Group study. Ann Oncol 2012;23:508–16. marker and subset results of a phase II randomized discontinuation 52. van der Graaf WT, Blay JY, Chawla SP, Kim DW, Bui-Nguyen B, Casali trial. J Clin Oncol 29: 2011 (suppl; abstr 10000). PG, et al. Pazopanib for metastatic soft-tissue sarcoma (PALETTE): a 57. Vernon CB, Eary JF, Rubin BP, Conrad EU III, Schuetze S. FDG randomised, double-blind, placebo-controlled phase 3 trial. Lancet PET imaging guided re-evaluation of histopathologic response in 2012;379:1879–86. a patient with high-grade sarcoma. Skeletal Radiol 2003;32: 53. Gardner K, Judson I, Leahy M, Barquin E, Marotti M, Collins B, et al. 139–42. Activity of cediranib, a highly potent and selective VEGF signaling 58. Choi H, Charnsangavej C, Faria SC, Macapinlac HA, Burgess MA, inhibitor, in alveolar soft part sarcoma. J Clin Oncol 27:15s, 2009 Patel SR, et al. Correlation of computed tomography and positron (suppl; abstr 10523). emission tomography in patients with metastatic gastrointestinal stro- 54. Stacchiotti S, Negri T, Zaffaroni N, Palassini E, Morosi C, Brich S, et al. mal tumor treated at a single institution with imatinib mesylate: pro- Sunitinib in advanced alveolar soft part sarcoma: evidence of a direct posal of new computed tomography response criteria. J Clin Oncol antitumor effect. Ann Oncol 2011;22:1682–90. 2007;25:1753–9.

5844 Clin Cancer Res; 18(21) November 1, 2012 Clinical Cancer Research

New Strategies in Sarcoma Therapy: Linking Biology and Novel Agents

Katherine A. Janeway and Robert G. Maki

Clin Cancer Res 2012;18:5837-5844. Published OnlineFirst August 28, 2012.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-12-0875

Cited articles This article cites 52 articles, 16 of which you can access for free at: http://clincancerres.aacrjournals.org/content/18/21/5837.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/18/21/5837.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://clincancerres.aacrjournals.org/content/18/21/5837. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.