Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

Rosemary J. Akhurst

Department of Anatomy and Helen Diller Family Comprehensive Center, University of California San Francisco, San Francisco, California 94158-9001 Correspondence: [email protected]

Transforming growth factor bs (TGF-bs) are closely related ligands that have pleiotropic activity on most cell types of the body. They act through common heterotetrameric TGF-b type II and type I transmembrane dual specificity kinase receptor complexes, and the outcome of signaling is context-dependent. In normal tissue, they serve a role in maintaining homeostasis. In many diseased states, particularly fibrosis and cancer, TGF-b ligands are overexpressed and the outcome of signaling is diverted toward disease progression. There has therefore been a concerted effort to develop drugs that block TGF-b signaling for therapeutic benefit. This review will cover the basics of TGF-b signaling and its biological activities relevant to oncology, present a summary of pharmacological TGF-b blockade strategies, and give an update on preclinical and clinical trials for TGF-b blockade in a variety of solid tumor types.

he three transforming growth factor bs titude of TGF-b-induced tumor promoting ef- T(TGF-bs), TGF-b1, -b2, and -b3, are closely fects modulate the tumor cells directly through related ligands that act on most cells types of the enhancement of tumor cell invasion and metas- body and have pleiotropic activities. Expression tasis, and induction and maintenance of cells of TGF-b1 is activated by tissue perturbations with tumor initiating properties, sometimes that induce cellular stress, such as cell prolifera- termed cancer stem-like cells (CSCs). Another tion and inflammation, and the induced ligand major site of protumorigenic TGF-b activity acts to reestablish homeostasis, acting as part is the tumor microenvironment (TME). Here, of a negative feedback circuit (Cui et al. 1995; TGF-b induces extracellular matrix (ECM) Akhurst et al. 1988; Li and Flavell 2008). In many deposition, myofibroblast differentiation, and diseased states, however, including fibrosis and angiogenesis, and suppresses both the innate cancer, TGF-b expression is chronically and ab- and adaptive immune systems. This results in a errantly elevated (Derynck et al. 1987; Fowlis feed-forward circuit of interactions between the et al. 1992; Gorsch et al. 1992; Walker and Dear- tumor and TME, which furthers tumor progres- ing 1992; Bellone et al. 1999, 2001). What is sion and results in aggressive, invasive, and met- more, responses to the ligand are altered toward astatic tumors that can be desmoplastic, with events that promote disease progression (Der- elevated intratumoral tension and high intersti- ynck et al. 2001; Roberts and Wakefield 2003). tial fluid pressure (IFP), all features that may be This is especially true in cancer, in which a mul- ameliorated by TGF-b signaling blockade. Over

Editors: Rik Derynck and Kohei Miyazono Additional Perspectives on The Biology of the TGF-b Family available at www.cshperspectives.org Copyright # 2017 Cold Spring Harbor Laboratory Press; all rights reserved Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301

1 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

the last two decades this signaling pathway has THE TGF-b SIGNALING PATHWAY therefore become a target for drug development, both for fibrosis and for oncology (Akhurst and The TGF-b family is comprised of more than Hata 2012). This review will focus on oncology 30 different homo- and heterodimeric pleiotro- applications, because there has been a rebirth of pic ligands encoded by 33 different genes. The interest in TGF-b blockade for cancer immuno- family includes TGF-bs, bone morphogenetic therapy with the clinical successes in this rapidly proteins (BMPs), GDFs (growth and differenti- expanding field. Drug development for fibrosis ation factors), activins and inhibins, , and applications, other than that related to oncology, anti-Mu¨llerian hormone (AMH) (Schmierer is not addressed because this topic has been re- and Hill 2007). Each of these ligands binds to viewed previously (Akhurst and Hata 2012). and activates signaling through heteromeric Moreover, treatment of chronic fibrotic condi- combinations of dual specificity kinase receptors tions by current anti-TGF-b signaling drugs that phosphorylate and activate downstream may be challenging because of the need to define Smad and non-Smad signaling components in a therapeutic window and dosing regimen be- a receptor kinase-dependent or independent tween efficacy and side effects. This review will manner (Fig. 1) (Derynck and Zhang 2003; Sor- cover the basics of TGF-b signaling and its bio- rentino et al. 2008). There can be considerable logical activities relevant to oncology, present a cross talk between intracellular signaling path- summary of pharmacological TGF-b blockade ways of the different TGF-b subfamilies, both strategies, and give an update on preclinical and downstream from and upstream of their respec- clinical trials for TGF-b blockade in a variety of tive receptors (Ray et al. 2010; Gro¨nroos et al. solid tumor types. 2012; Peterson and O’Connor 2013). In disease

TGF-β

TNF Wnt TGF-α FGF EGF Smad3 Non-Smad pathways PI3K P Ras-Erk MAPK Smad3 Smad4 Rho-JNK Modification of Smad P RhoA-ROCK transcriptional output Smad3 TAK1-p38 MAPK

P Context-dependent TF Smad3 Transcriptional output

Figure 1. Context-dependent transforming growth factor b (TGF-b) signaling outputs arise from pathway interactions. Schematic of TGF-b signaling via the canonical Smad pathway (center) or noncanonical signaling pathways (right). Modification of Smad transcriptional output (left) may be by posttranslational modification of Smads, such as by linker phosphorylation byextracellular signal-related kinase (ERK) mitogen-activated protein kinase (MAPK) or by the presence or activation status of interacting transcription factors that are regulated by other stress and other growth factor signaling pathways. EGF, Epidermal growth factor; FGF, fibroblast growth factor; TNF, tumor necrosis factor; PI3K, phosphoinositide 3-kinase; JNK, c-Jun amino-terminal kinase.

2 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

states, ablation of signaling from one ligand sociated peptide (LAP), is cleaved from the subtype may interfere with the signaling output carboxy-terminal mature peptide by a subtili- from others in either a positive or negative man- sin-like proprotein convertase, furin, but re- ner, with one of the finest examples being found mains noncovalently associated with the mature through human genetics (Hatsell et al. 2015). homodimeric ligand (Annes et al. 2003). Di- Chronic versus acute inhibition of TGF-b sig- meric LAP encapsulates the mature dimeric naling may result in quite different outcomes peptide within a cage-like structure, keeping it (Connolly et al. 2011), because negative feed- in an inactive state that facilitates spatial and back loops have evolved to keep this important temporal regulation of storage, delivery, and ac- signaling pathway in equilibrium within the cell tivation (Shi et al. 2011). On the outer surface of (Fig. 1). this “cage,” one of the two LAP arms also binds The three bona fide TGF-b ligands are dis- covalently to a LTBP (latent TGF-b-binding tinguished from other ligands of this family by protein), which is a large ECM component extensive sequence homology and their unique that anchors latent TGF-b within the ECM ability to bind and signal via the TGF-b receptor rather than permitting free diffusion within type II, TbRII. When the homodimeric ligands the tissue (Munger et al. 1997; Ota et al. 2002; bind to the extracellular domain of TbRII, het- Annes et al. 2003). LTBP noncovalently binds ero-oligomerization of the two canonical recep- fibrillin (Chaudhry et al. 2007), mutations of tors, TbRI and TbRII, causes a conformational which are causative for Marfan syndrome (Rob- change in the receptor complex that results in inson et al. 2006). Dimeric LAP also binds non- phosphorylation, and subsequent activation, of covalently to cell surface integrins, particularly the type I receptors by the type II receptors (Shi integrins b1, b6, and b8, which can bind latent and Massague´ 2003). Activation of TbRI leads TGF-b1 and b3 via an RGD site within the cor- to signal propagation through the well-charac- responding LAP domains (Munger et al. 1999; terized Smad-dependent canonical pathway Annes et al. 2004; Yang et al. 2007; Arnold et al. and/or Smad-independent noncanonical path- 2014). This interaction results in TGF-b activa- way(s) (Derynck and Zhang 2003). Cross talk tion, triggered by the molecular tension result- between Smad and non-Smad signaling path- ing from physical stretch between LAP-integrin ways can directly modify Smad transcriptional binding at the cell surface and LAP–LTBP an- output, as can interaction between TGF-b and chorage to the ECM (Yang et al. 2007; Shi et al. other growth factor signaling pathways. Smads, 2011; Dong et al. 2014; Mi et al. 2015). Proteo- forexample, can be phosphorylated within their lytic cleavage of the amino-terminal motif of central “linker” domain by other kinases, such LAP can also activate TGF-b by releasing the as the extracellular signal-regulated kinase (Erk) entrapped active homodimer from its latent mitogen-activated protein kinase (MAPK) path- protein cage (Dong et al. 2014). On the surface way (Kretzschmar et al. 1999), c-Jun amino- of T cells and platelets, the transmembrane pro- terminal kinase (JNK) MAPK pathway (Engel tein glycoprotein A repetitions predominant et al. 1999), protein kinase C (PKC) (Yakymo- (GARP or LRRC32), a leucine-rich repeat mol- vych et al. 2001), Erk MAPK 1 (MEKK1) ecule, serves a similar and essential role in (Brown et al. 1999), and Ca2þ/calmodulin- activating TGF-b at the cell-platelet surface dependent protein kinase II (CamKII) (Wicks (Stockis et al. 2009; Tran et al. 2009). Thrombin, et al. 2000), which can negatively or positively matrix metalloproteinases (MMP)-2 and -9, affect their activity. Thus, the result of ligand and thrombospondin-1 (THBS1), which can stimulation is highly contextual (Fig. 1). all be enriched in carcinomas, are each capable Each of the three TGF-b ligands is translat- of cleaving LAP to activate TGF-b (Schultz- ed as a large pre-pro-polypeptide. The amino- Cherry et al. 1994). Other means of ligand ac- terminal signal peptide plays a role in classic tivation include acidic pH, reactive oxygen spe- intracellular translocation and secretion. The cies (ROS) and ionizing radiation, all of which prodomain, otherwise termed the latency-as- are relevant to tumor progression and cancer

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 3 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

treatment (Barcellos-Hoff 1993; Barcellos-Hoff homeostatic negative regulator of cellular pro- and Dix 1996; Annes et al. 2004; Jobling et al. liferation is blunted by activation of oncogenes. 2006; Shi et al. 2011). Moreover, TGF-b1 expression is sequentially elevated by multiple mechanisms during the progression of multistage carcinogenesis from LIGAND EXPRESSION AND TGF-b tumor initiation through to metastasis (Oft RESPONSIVENESS IN HUMAN TUMORS et al. 2002)—for example, by oncogene-driven AP-1 transcription factor binding to and activa- Early studies showed that TGF-b expression lev- tion of the TGFB1 gene promoter (Kim et al. els are elevated in both human and murine tu- 1989; Weigert et al. 2000; Davies et al. 2005) mors relative to their respective normal tissues and by latent TGF-b activation within a prote- (Derynck et al. 1987; Fowlis et al. 1992; Gorsch ase-rich TME (Leitlein et al. 2001). Tumor and et al. 1992; Walker and Dearing 1992; Bellone patient are therefore bathed in excess TGF-b, et al. 1999, 2001). Generally, it is thought that contributing to an immunosuppressed state. TGF-b1 and TGF-b2 are the major players in Which cells within the tumor respond to tumor progression, with only a few reports of TGF-b, and how they respond, are highly rele- TGF-b3 involvement (Constam et al. 1992). The vant questions for the design of TGF-b-block- sparsity of reports of TGF-b3 involvement in ade therapy. Parenchymal cells of certain tu- cancer may be because of ascertainment bias mors are unresponsive to TGF-b because of (i.e., few studies have systematically examined genetic inactivation of TGFBR1 or TGFBR2 or the expression of all three ligands during tumor- loss of intracellular signaling pathway compo- igenesis); alternatively, there is disincentive to nents (Akhurst and Derynck 2001). This is par- report negative data (i.e., TGF-b3 may not be ticularly true for of the gastrointestinal expressed). However, the question of which (GI) tract, including colon, pancreas, and gas- tissue and what cell types secrete the various tric cancer. In colon cancer with microsatellite ligands is critical when using ligand-specific instability (MIS), TGFBR2 is a mutational hot- drugs (Bedinger et al. 2016), especially in the spot for genetic inactivation caused by the pos- light of older reports that TGF-b1 and TGF- session of a 10-bp polyadenine repeat within its b3 may have antagonistic effects (Li et al. coding sequence (Parsons et al. 1995). However, 1999; Ask et al. 2008; Laverty et al. 2009). In loss of a single component of the pathway, such breast cancer, which has been most extensively as SMAD4/DPC4 in pancreatic cancer (Iacobu- studied in this respect, reports suggest that both zio-Donahue et al. 2009), does not necessarily TGF-b1 and TGF-b2 are functionally associated ablate all TGF-b responses. Tumor cells are with a more invasive and early onset breast can- highly plastic, and can rewire signaling net- cer, whereas TGF-b3 expression appears higher works independently of Smad4 (Hocevar et al. in more differentiated lobular breast tumors 1999; Giehl et al. 2007; Descargues et al. 2008). (Flanders and Wakefield 2009). Moreover, in GI and other tumor types, such as Circulating plasma levels of TGF-b1 and glioblastoma, prostate cancer, hepatocellular -b2, but not TGF-b3, can be exceedingly high carcinoma, bladder, and breast cancer, muta- in cancer patients, correlate with tumor grade, tional loss of TGF-b signaling components is and decrease significantly following surgical tu- uncommon, whereas transcriptional or epige- mor resection (Kong et al. 1995; Krasagakis et al. netic repression of genes encoding signaling 1998; Shim et al. 1999). In particular, TGF-b1 components can occur (Ogino et al. 2007; Shi- expression is frequently activated in response to pitsin et al. 2007; Yamashita et al. 2008; Dong many forms of cellular stress, such as during et al. 2012). A study of 34 matched primary and epithelial hyperplasia, where it serves a function recurrent breast tumors shows that, despite ap- in reestablishing homeostasis (Akhurst et al. parent lack of TGFBR2 mutations in primary 1988; Cui et al. 1995). However, during tumor tumors, 12% of recurrent breast tumors contain progression, the normal function of TGF-b as a receptor kinase-attenuating point mutations,

4 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

suggesting that TGFBR2 mutation in a minority TGF-b ACTIONS IN TUMORIGENESIS of breast tumors is a late event rather than a Inhibition of Cell Proliferation driver (Lucke et al. 2001). In human cutaneous TGF-b is a most potent epithelial growth in- squamous cell carcinoma (cSCC), attenuated hibitor, as well as a suppressor of endothelial, expression of Smad proteins has been reported hematopoietic, and immune cell proliferation (Hoot et al. 2008), but whether this is owing (Coffey et al. 1988; Derynck et al. 2001). to mutation or transcriptional or epigenetic The proapoptotic and differentiation-inducing dysregulation was not investigated. In head activities of TGF-b on epithelial cells, in the and neck squamous cell carcinoma (HNSCC), context of cancer, result in tumor suppression mutations in TGFBR2 and SMAD4 occur but (Heldin et al. 2009). In normal epithelial cells, at low frequency (1% or less) (TCGA Network TGF-b activates transcription of CDKN1A and 2015). CDKN2A, which encode the cyclin-dependent Examination of Smad activation in archival kinase (CDK) inhibitors, p21CIP1 and p15Ink4b human tumor samples, probed by phospho- respectively, causing cell-cycle arrest at the G Smad (pSmad) immunoreactivity, may be mis- 1 phase (Gomis et al. 2006a). Conversely, TGF-b leading (Xie et al. 2002; Ogino et al. 2007; Hoot acting via a Smad–FoxO complex represses et al. 2008; Harradine et al. 2009), not only be- the transcription of MYC and ID family genes, cause of the common limitations of immuno- which encode transcription factors that control histochemistry (IHC), such as specificity and cell proliferation, cell fate determination, and nonlinear sensitivity, but also because pSmad2 cellular differentiation (Siegel et al. 2003; Kondo activation is heterogeneous and highly dynamic et al. 2004; Tang et al. 2007; Anido et al. 2010; within the tumor. IHC analysis has shown in- James et al. 2010). In carcinomas, many tumors creased levels of activated nuclear pSmad2 at the lose the growth inhibitory response to TGF-b, invasive front of human breast tumors (Kang but still respond to this ligand but in a protu- et al. 2005), and intravital microscopy using a morigenic manner, such as increased migration, fluorescent reporter to track Smad activation in invasion, and epithelial–mesenchymal transition live mouse tumors reveals the dynamic nature of (EMT). Thus, depending on the tumor type and this activity within breast cancer cells in vivo the stage of tumor progression, TGF-b may po- (Giampieri et al. 2010). The latter study shows tently suppress or promote cancer progression that TGF-b signaling promotes single cell mo- through direct actions on the tumor cells (Go- tility and metastasis in a transient manner. Im- mis et al. 2006b; Hannigan et al. 2010), presum- portantly, TGF-b signaling is down-regulated at ably through its control over differential gene the destination site of metastasis, favoring out- expression programs (see below). growth of secondary tumors (Giampieri et al. 2010). Clearly, active TGF-b signaling depends not only on the tumor genetics but also, in a Extracellular Matrix Regulation dynamic fashion, on a cell state that fluctuates between active and inactive signaling. Further- The ECM is the noncellular component of more, the response to TGF-b signaling also de- connective tissue, which supports cells and pends on the status of other signaling and tran- their functions, and is composed of multiple scriptional pathways, which are probably also in proteins, collagen, elastin, fibrillin, fibronectin, dynamic flux (Fig. 1). laminin, and proteoglycans. Fibrosis is charac- Even when the carcinoma cells are resistant terized by the accumulation and activation of to TGF-b signaling by virtue of genetic or epi- fibroblasts to secrete excessive ECM, and this genetic loss of TGFBR2, the cells of the TME is stimulated by TGF-b, making this signaling retain an intact signaling pathway. These effects pathway a therapeutic target in various patho- of TGF-b on tumor stroma, vessels, and im- logical fibroses (Akhurst and Hata 2012). Sever- mune cells may turn out to be the Achilles’ al genes encoding ECM proteins that are known heel in TGF-b blockade therapy (see below). to be important in driving fibrosis are directly

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 5 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

activated by TGF-b-induced Smad and Smad- late the basic biology of the tumor by increasing independent signaling (Stampfer et al. 1993; ECM elaboration and eliciting a tissue contrac- Hocevar et al. 1999, 2005; Piek et al. 2001). tion process, which contributes to further TGF- The fibrotic response to TGF-b is highly rele- b activation (Shi et al. 2011) and to elevation in vant to cancer progression, because a desmo- IFP (Lammerts et al. 2002; Heldin et al. 2004; plastic response is seen in many cancer types, Salnikov et al. 2005). This attenuates the effi- especially pancreatic cancers and mesothelioma ciency of delivery of some drugs to the tumor, (Kano et al. 2007; Margadant and Sonnenberg because, under positive IFP, small molecule 2010; Fujii et al. 2012a,b). Desmoplasia not only drugs do not efficiently penetrate tumor tissue presents a barrier to drug delivery (Kano et al. (Heldin et al. 2004). Conversely, the leaky na- 2007), but can also provide feedback on the tu- ture of tumor vasculature presents an opportu- mor to promote growth and metastasis (Leven- nity for drug delivery of novel nanoparticles to tal et al. 2009; Ng and Brugge 2009). There is the tumor (Kano et al. 2007; Park et al. 2012), reciprocal regulation between TGF-b by ECM. which is enhanced by TbRI blockade (Kano Latent TGF-b bound to ECMcomponents, such et al. 2007). as fibronectin and fibrillin, is inactive until phys- iological or pathological processes initiate its TGF-b Action in Cancer Stem-Cell release (Chaudhry et al. 2007; Shi et al. 2011). Maintenance Moreover, interaction between integrins and ECM can generate tension that feeds back to Accumulating evidence links TGF-b signal- more TGF-b expression and activation, gener- ing to the acquisition and maintenance of a ating a cycle toward cumulative fibrosis. “stem-cell-like” state of carcinoma cells. The elusive tumor-initiating cell or CSC generates diverse progeny that lead to the characteristic Induction of Epithelial–Mesenchymal heterogeneity of the tumor while also maintain- Transition and the Myofibroblast Phenotype ing a self-propagating function that is essential Partial or full EMT can be induced by TGF-b in for primary tumor growth and for seeding of both normal and neoplastic epithelia and has tumors at distant metastases (Shipitsin et al. consequences for disease progression by stimu- 2007; Mani et al. 2008). It is now widely accepted lating invasion, metastasis and seeding poten- that EMT can generate a CSC-like state; indeed, tial of primary tumor cells, as well as contrib- this cell type is reported to occupy a partial uting to a stromal component that is refractile EMT position with characteristics of both cell to drugs (Derynck and Akhurst 2007). Ex- states (Kalluri and Weinberg 2009; Jolly et al. pression of E-cadherin, which can suppress tu- 2015). mor progression, is commonly down-regulated There is enrichment of expression of TGF-b within many carcinomas during EMT (Caulin signaling components in CD44þ/CD242 breast et al. 1995; Portella et al. 1998; Lacher et al. carcinoma cells with CSC-like properties rela- 2006; Fransvea et al. 2008; Hoot et al. 2008). tive to their CD442/CD24þ progeny (Shipitsin The TGF-b-Smad pathway mediates expression et al. 2007). CD44þ cells, but not their progeny, of high-mobility group A2 (HMGA2), which express TGFB1 and TGFBR2 together with a contributes to the induction of the expression “TGF-b gene expression signature.” Indeed, of Snail and Slug, two zinc-finger transcription within the non-CSC compartment, epigenetic factors that repress the E-cadherin gene (Padua modifications were found that silence the and Massague´ 2009). In breast and skin cancer, TGFBR2 gene (Shipitsin et al. 2007). Short- tumor cell EMT contributes to cancer progres- term treatment with LY2109761, a small mole- sion, as cells consequently become more mig- cule inhibitor of the TbRI and TbRII kinases, ratory, invasive, and can ultimately transition to (Sawyer et al. 2003; Yingling et al. 2004) de- a myofibroblastic phenotype (Derynck and creases the expression of markers of “stemness” Akhurst 2007). Myofibroblasts further modu- and induces differentiation to a less aggressive

6 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

carcinoma type, suggesting that TGF-b signal- to chemotherapeutic drug resistance (Singh and ing controls stem-cell maintenance (Shipitsin Settleman 2010; Huang et al. 2012). Some tu- et al. 2007). In support of this postulate, de- mors show intrinsic resistance to creased expression of CSC markers was also de- while others acquire resistance after prolonged tected in tumor cells of a single glioblastoma exposure to drug (Engelman and Settleman patient undergoing the first of the 2008; Sequist et al. 2008; Corcoran et al. small molecule TbRI inhibitor, galunisertib 2010). The unpleasant truth in oncology is (Anido et al. 2010). However, others made con- that in the most aggressive tumor types, such trary findings in gastric carcinoma cells in cul- as , lung, pancreas, and glioblastoma ture and in a breast carcinoma xenograft model multiforme, patients may initially respond well (Tang et al. 2007; Ehata et al. 2011). TbRI– to chemotherapy but then have an exceedingly Smad2 signaling has been implicated in main- high rate of relapse as a consequence of “ac- taining epigenetic silencing that promotes and quired” drug resistance. Similarly, even with sustains EMT and the CSC phenotype (Papa- the newer wave of targeted therapies, such as georgis et al. 2010). Even in hematological herceptin (anti-HER2 mAb) and vemurafinib malignancies, in which TGF-b has a predomi- (small molecule B-Raf inhibitor), drug resis- nantly tumor suppressive role, TGF-b signaling tance develops within months of treatment. supports leukemia-initiating cell maintenance The mechanisms for development of drug resis- in chronic myeloid leukemia (CML) through tance are varied, with some targeted therapies its association with the FOXO pathway and ac- evolving highly specific mechanisms to avoid tivation of Akt signaling (Naka et al. 2010; drug action. With epidermal growth factor re- Miyazono 2012). It has also been shown that ceptor (EGFR) inhibitors in non-small-cell hemangioblasts from BCR-ABL-positive CML lung carcinoma (NSCLC) therapy, 50% to patients express higher levels of TGF-b1 (al- 70% of cases of acquired drug resistance involve though not TGF-b2 or TGF-b3) than those somatic mutation of the gene encoding the drug from control individuals. This was attributed target, namely EGFR. In particular, a common to activation of TGF-b1 by the BCR-ABL onco- T790M “gatekeeper” mutation changes the rel- protein (Zhu et al. 2011). Indeed, the inhibitor ative binding of EGFR to the ATP-mimetic drug of ABL tyrosine kinase, imatinib, suppresses versus its binding to endogenous substrate, ATP the expression of TGF-b1 by CML hemangio- (Yun et al. 2008). Amplification of the MET blasts. Conversely, treatment of CML heman- oncogene is another common cause of drug re- gioblasts with TGF-b1 increases the expression sistance (Engelman et al. 2007). However, in of MMP9, soluble KitL and soluble ICAM-1, 30% to 50% of cases, there is no specific drug- which all contribute to CML progression. More- evading mutation. Instead, global epigenetic over, TGF-b induces ICAM-1 synthesis and changes take place that lead to an altered chro- suppresses the activity of T lymphocytes and matin state (Sharma et al. 2010; Vinogradova natural killer (NK) cells. BCR-ABL-induced et al. 2016). This altered cellular differentiation TGF-b1 not only assists stem-cell maintenance, state is likely reached during EMT, and coin- but produces an immune privileged microenvi- cides with the appearance of chemoresistant ronment for the stem cells (Zhu et al. 2011). cells with stem-cell-like features, coined drug- TGF-b inhibitors might therefore be uniquely tolerant persistor cells (DPTs) (Voulgari and poised to kill or maim CSCs, making this class Pintzas 2009; Sharma et al. 2010). Even in breast of drug even more attractive to oncologists. carcinomas driven by HER2/NEU/ERBB2 am- plification, epithelial-like luminal tumors evolve to enrich for CD44highCD24low cells TGF-b Signaling Promotes Resistance with the characteristics of CSCs, and these tu- to Chemotherapy mor cells are resistant to neoadjuvant chemo- EMTand the resultant CSC-like phenotype not therapy and pharmacological HER2 inhibition only help drive metastasis, but also contribute (Li et al. 2008). Generally, although patients

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 7 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

with basal mesenchymal-like tumors may ini- TGFB3, SMAD2, and SMAD7 to a lesser degree, tially respond to chemotherapy better than and the CSC-related CD44 and ALDHA1 those with luminal breast cancer, individuals mRNAs were also elevated after chemotherapy. with triple negative breast tumors ultimately Moreover, coadministration of paclitaxel with have the worst prognoses, possibly because of TGF-b signaling inhibitors (galunisertib, anti- a more aggressive CSC-like subpopulation that TbRII antibody or siSMAD4) in a nude mouse is drug resistant. xenograft model reduces primary tumor growth The central importance of TGF-b signaling compared with tumors receiving paclitaxel in driving a drug tolerant state has been illus- alone (Bhola et al. 2013). Most important, the trated by the discovery of a suppressor of TbRII number of tumor initiating cells within the pri- expression in a genome-wide RNAi screen to mary tumor, assayed by their ability to generate identify genes that mediate chemoresistance of mammospheres in cell culture, was decreased in breast cancer cells (Huang et al. 2012). Intrigu- mice treated with the TGF-b inhibitor com- ingly, up-regulation and enhanced cell surface pared with paclitaxel alone (Bhola et al. 2013). presentation of TbRII protein is mediated by Because the tumor initiating cells are essential loss of the MED12 gene (encoding a subunit for tumor growth and metastasis, these studies of the mediator complex), and this was the show the essential need for TGF-b signaling for most common cause of drug resistance in these tumor growth and metastatic spread. breast carcinoma cells. Indeed, up-regulation of In a model of cSCC, TGF-b contributes to TbRII expression appears to be a common con- drug resistance by yet another mechanism, inde- tributor to acquired resistance against numer- pendent of its effect on cell cycle or DPTs. Spe- ous drugs, including chemotherapeutics, such cifically, induction of p21Cip1 expression in re- as cisplatin, as well as the molecular targeted sponse to TGF-b activates the expression of the therapies, erlotinib and gefitinib (EGFR in- transcription factorNrf2, whichin turnactivates hibitors), sorafenib (a multikinase RAS/RAF/ genes of the glutathione-S-transferase pathway. ERK, VEGFR, PDGFR inhibitor), PLX4032 (a Activation of this pathway is thought to neutral- BRAF inhibitor), and crizotinib (a MET and ize tumor-damaging ROS, generated conse- ALK tyrosine kinase inhibitor). Moreover, as a quent to chemotherapy (Oshimori et al. 2015). mechanism of acquired drug resistance, up-reg- ulation of TbRII consequent to loss of MED12 TGF-b as a Potent Suppressor of Tumor is observed in a variety of different tumor Immunosurveillance types, including NSCLC, colorectal cancer, melanoma, and hepatocellular carcinoma. The The negative regulation of the increased expression of TbRII results in activa- by TGF-b is complex and context-dependent, tion of both Smad and non-Smad Erk MAPK but was recognized as a target for cancer therapy signaling in response to autocrine TGF-b, and as early as 1988 (Kasid et al. 1988; Kuppner et al. is both sufficient and necessary for acquired 1988; Zuber et al. 1988), and proof-of-principle cancer drug-resistance (Huang et al. 2012). Im- was shown in vivo in 1993 (Arteaga et al. 1993; portantly, the small molecule TbRI inhibitor, Gridley et al. 1993). Under normal conditions, LY2157299(galunisertib), resensitizes drug-tol- TGF-b signaling delicately regulates the tolero- erant cells to anticancer drugs (Huang et al. genic versus immunogenic arms of the immune 2012). Proof of this concept was shown in clin- system to balance an adequate host defense to ical material from paclitaxel-treated primary foreign substances while limiting collateral in- breast cancer patients wherein paclitaxel treat- flammatory tissue damage (Gorelik and Flavell ment in the neoadjuvent setting increased a 2001, 2002; Rubtsov and Rudensky 2007; Flavell TGF-b gene expression signature in the primary et al. 2010). TGF-b has potent growth suppress- tumor (Bhola et al. 2013). TGFBR2, TGFBR3, ing activity on most precursor cells of the im- and SMAD4 mRNA levels were increased mune system, particularly T and B cells of the approximately twofold, TGFBR1, TGFB1, adaptive arm. It suppresses T-cell proliferation

8 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

(Rubtsov and Rudensky 2007), induces B-cell the osteolytic effects of these cancers, and help apoptosis (Ramesh et al. 2009), and suppresses reestablish a homeostatic equilibrium that neu- the effector functions of cytotoxic CD8 T cells tralizes bone loss and encourages normal bone and helper CD4þ T cells (Gorelik and Flavell growth. Therefore, when considering potential 2002). Regulatory T cells (Tregs) that suppress disease indications for anti-TGF-b therapy, the the activity of effector T cells, are stimulated to organ site of neoplastic origin should not be the expand and differentiate by TGF-b, and Treg primary consideration. Rather, it is necessary to functionality is mediated in part by further consider (1) the genetic makeup of the tumor TGF-b secretion (Chen et al. 2005). (e.g., mutations in TGFBR2, TGFBR1, SMAD4), TGF-b signaling also has potent immuno- (2) activation of a TGF-b responsive gene ex- suppressive action via direct effects on innate pression signature, (3) extent of excess TGF-b immune cells, which modulate their phenotype production (TGF-b1 or TGF-b2) by the tumor from tumor cell destroying to tumor cell sup- or its environment, as measured in the circula- porting in response to this cytokine (Mantovani tion or from tumor biopsy, (4) tumor stage and et al. 2002; Fridlender et al. 2009). TGF-b sup- grade, invasive or metastatic, (5) specificity of presses the generation of NK cells in response to metastatic tropism (e.g., to bone that is rich in interferon-g, which is required for NK tumor TGF-b), and (6) host cellular responses to the killing activity, through transcriptional control tumor (e.g., local or systemic levels of Tregs or of the interferon-g promoter by Smad3 (Laouar Th17 cells) or desmoplastic responses. et al. 2005). It also “polarizes” macrophages Based on preclinical studies and clinical tri- (Mantovani et al. 2002) and neutrophils (Frid- als, it is unlikely that TGF-b inhibitors will be lender et al. 2009) from an inflammatory phe- efficacious when used alone, because they are notype that targets and destroys foreign agents, not cytotoxic. Moreover long-term treatment such as cancercells, toward an immunosuppres- regimens, as with any drug, should be avoided sive protumorigenic cell type (Flavell et al. because of potential inflammatory, autoim- 2010). Dendritic cells (DCs) show reduced an- mune, and cardiovascular side effects (Laping tigen presentation capability in the presence of et al. 2007; Anderton et al. 2011). It is likely TGF-b (Yamaguchi et al. 1997), and tumor-de- that TGF-b inhibitors will have their most im- rived TGF-b also alters chemokine receptor ex- portant therapeutic activity in cancer through pression to blunt DC chemotaxis (Sato et al. effects on the TME, particularly, but not only, in 2000), further suppressing immune surveil- neutralizing or reversing immune suppression. lance, which can be relieved by small molecule Indeed, the combination of TGF-b inhibition TbRI kinase inhibitors. together with existing , such as cancer vaccines (Jia et al. 2005; Nemunaitis and Murray 2006; Kim et al. 2008), adoptive T- POTENTIAL ONCOLOGY APPLICATIONS cell transfer (Suzuki et al. 2004; Wallace et al. FOR TGF-b BLOCKADE 2008), chimeric antigen receptor T-cell therapy Many solid tumor types share similar properties, (Gill and June 2015; Wu et al. 2015), and im- regardless of their site of origin. For example, mune checkpoint blockade (Topalian et al. breast cancer, prostate cancer, and melanoma 2012; Lipson et al. 2013), will likely provide metastasize to bone, and multiple myeloma the major basis of their therapeutic usage. also colonizes bone. One would expect a thera- Here again, the major players appear to be peutic benefit from TGF-b therapy in these TGF-b1 and -b2. Augmenting adoptive T-cell bone-tropic cancers (Yin et al. 1999; Kang therapy with short-term acting drugs, rather et al. 2003; Kozlow and Guise 2005; Mohammad than genetic manipulation of tumor homing cy- et al. 2011), and targeting of all three isoforms, totoxic T lymphocytes (CTLs) to reduce their or at least TGF-b1 and TGF-b2, should help sensitivity to TGF-b, may be a particularly at- neutralize the TGF-b-enriched microenviron- tractive application because patients need not be ment of the bone. This approach might reduce exposed to genetically manipulated T cells.

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 9 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

Moreover, patients may avoid potential side ef- in assessing their potential for use in the clinic. fects of long-term systemic drug exposure as Parameters to consider are affinity and specific- only short-term exposure to the drug may be ity for drug target, drug stability, drug clearance required for efficacy. and bioavailability in vivo, as well as mode of drug delivery (e.g., oral vs. intravenous). Generally, the small molecule kinase inhib- DRUGS THAT BLOCK TGF-b SIGNALING itors (SMIs) lack absolute specificity, and, at certain doses, also target activin, nodal, and Drugs that target many different components of possibly the signaling pathways. the canonical TGF-b signaling pathway have Moreover, because these inhibitors block TbRI been developed (Fig. 2; Table 1), including a kinase activity, they will not prevent signaling pyrrole-imidazole polyamide drug that blocks independent of the kinase activity, such as transcription of the TGFB1 target gene (Yao TRAF6-p38 MAPK signaling (Hocevar et al. et al. 2009; Chen et al. 2010; Matsuda et al. 2005; Gudey et al. 2014; Sundar et al. 2015). 2011; Washio et al. 2011; Igarashi et al. 2015), SMIs have poor pharmacokinetics and are antisense RNAs that target TGFB1 or TGFB2 generally cleared from the body with a t1/2 of mRNAs for degradation (Hau et al. 2009; Val- 2–3 h, whereas pharmacodynamic markers, lieres 2009; Schlingensiepen et al. 2011), anti- such as inhibition of Smad2 phosphorylation, bodies against TGF-b ligands or receptors that persist for up to 8 h postdosing (Bueno et al. block ligand–receptor engagement, and many 2008; Gueorguieva et al. 2014). These negative small molecule ATP-mimetic TbRI kinase in- features of SMIs are balanced by the ease of drug hibitors (Fig. 2). Each drug has distinct advan- administration through the oral route. In some tages and disadvantages that have to be balanced cases, the short pharmacokinetic/pharmacody-

Ligand activation blockade, e.g., α-GARP, α-integrin Active TGF-β LTBP Ligand-blocking antibodies, e.g., fresolimumab

Latent LAP Receptor-blocking antibodies, e.g., LY3022859 TGF-β

Small molecule TβRI kinase inhibitors, Smad3 e.g., galunisertib

AAAA P AAAA AAAA Smad3 AAAA Smad4 5′ 5′ P 5′ Smad3 5′ mRNA disruption by antisense, e.g., ISTH0047 Transcription disruption, e.g., pyrrole-imidazole polyamide

P TF Smad3

Figure 2. Drug targets on the TGF-b signaling pathway. The figure indicates molecular targets during the synthesis, activation, and signaling of TGF-b to which drugs have been raised. Molecular targets and processes are shown in boxes.

10 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

Table 1. Summary of TGF-b signaling blockade drugs currently in development as clinical leads Drug class Drug name Target Application References Small molecule TEW-7197 TbRI Oncology Jin et al. 2014; Son et al. 2014; kinase inhibitors Naka et al. 2016 Galunisertib TbRI Oncology Fujiwara et al. 2015; Kovacs (LY2157299) et al. 2015; Rodo´n et al. 2015b; Brandes et al. 2016 Anti-TGF-b ligand Fresolimumab TGF-b3.TGF- Oncology, fibrosis Morris et al. 2014 antibodies b2.TGF-b1 Rice et al. 2015 XPA681 Pan-TGFb Eye diseases Bedinger et al. 2016 (b1.b2.b3) XPA089 TGF-b1.TGF- Oncology b2 no reactivity to TGF-b3 LY2382770 TGF-b1-specific Fibrosis, oncology Cohn et al. 2014 Anti-TbR receptor LY3022859 TbRII Oncology Zhong et al. 2010 antibodies Antisense ISTH0036, TGFB2 RNA Eye diseases Bogdahn et al. 2011; oligonucleotides ISTH0047 Preclinical oncology Chamberlain 2011; Wick and Weber 2011 Pyrrole-imidazole TGFB1 gene Fibrosis, scarring, Yao et al. 2009; Chen et al. polyamides promoter vascular repair 2010; Matsuda et al. 2011; Washio et al. 2011; Igarashi et al. 2015

namic (PK/PD) window may, in fact, be advan- tumors (all cancers, NCT02160106) (Jin et al. tageous to minimize side effects. “Off-target” 2014; Son et al. 2014; Naka et al. 2016). TEW- inhibition of activin, nodal, or even p38 7197 has been shown to cause Smad4 degrada- MAPK signaling may also enhance drug effica- tion in cytotoxic T cells, resulting in enhanced cy, albeit not with the specificity intended. It has cytotoxic T-cell activity (Yoon et al. 2013), as been postulated that SMIs penetrate tumor tis- well as reduced breast tumor metastasis to lung sue more easily than antibodies, and this may be in mice (Son et al. 2014). true in highly dysplastic tumors, such as pan- The other drug class that has received con- creatic carcinoma. However, the anti-TGF-b siderable attention in the clinic (Table 2) is antibody, fresolimumab, was shown to be effi- the TGF-b ligand- and receptor-blocking anti- caciously delivered to glioblastoma cells in a bodies, including the pan-TGF-b1/2/3 block- clinical setting, presumably owing to break- ing antibody, fresolimumab (Genzyme/Sanofi) down of the blood brain barrier in this neoplas- (Morris et al. 2014), the TGF-b1-specific block- tic tissue (Den Hollander et al. 2015). ing antibody LY2382770 (Eli Lilly) (Cohn Many preclinical studies have been under- et al. 2014; Tampe and Zeisberg 2014), and taken, thus evaluating the entire spectrum of the TbRII blocking antibody LY3022859, also TbRI inhibitor drugs (Akhurst and Hata called IMC-TR1 (Eli Lilly, NCT01646203) 2012), but the main SMI drug to proceed to (Zhong et al. 2010). Other companies followed the clinic has been galunisertib or LY2157299 their lead with the development of alternative (Eli Lilly) (Table2). More recently, a first human TGF-b ligand blocking antibodies, such as dose (FHD) study of a new TbRI kinase SMI, an anti-TGF-b1/2 antibody that is not cross TEW-7197 (MedPacto), was started as mono- reactive with TGF-b3, developed by Xoma/ therapy in subjects with advanced stage solid Novartis (Bedinger et al. 2016).

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 11 Downloaded from http://cshperspectives.cshlp.org/ ..Akhurst R.J. 12 Table 2. Interventional clinical oncology trials of TGF-b signaling blockade Clinical trial Drug Target disease Organ site Status number References Galunisertib Phase Ib/II study with gemcitabine and LY2157299 for All metastatic tumors and Completed NCT02154646 Rodo´n et al. dacdOln ril.Ct hsatceas article this Cite Article. Online Advanced patients with metastatic cancer (phase Ib) and unresectable pancreatic Ongoing NCT01373164 2015 advanced or metastatic unresectable pancreatic cancer (phase II) Galunisertib Phase I dose-escalation study of LY2157299 All solid tumors Completed NCT01722825 Fujiwara et al.

monotherapy in patients with solid tumors (Japan) 2015 onSeptember25,2021-PublishedbyColdSpringHarborLaboratoryPress Galunisertib Randomized phase II study LY2157299 monotherapy HCC Ongoing NCT02178358 versus sorafenib monotherapy versus combination of NCT01246986 both drugs in patients with advanced hepatocellular carcinoma Galunisertib Phase II/III study of LY2157299 monotherapy in very MDS Ongoing NCT02008318 low-, low-, and intermediate-risk patients with myelodysplastic syndromes odSrn abPrpc Biol Perspect Harb Spring Cold Galunisertib Phase Ib/IIa study combining LY2157299with standard Glioma Ongoing NCT01220271 Sepulveda- temozolomide-based radiochemotherapy in patients Sa´nchez with newly diagnosed malignant glioma et al. 2015 Galunisertib Phase I/II LY2157299 monotherapy or lomustine Glioma Ongoing phase I NCT01682187 Brandes et al. monotherapy versus combination of both drugs in Ongoing phase II NCT01582269 2016 recurrent glioma/glioblastoma Galunisertib LY2157299and radiotherapy in metastatic breast cancer Metastatic breast Recruiting NCT02538471 Galunisertib Phase 1b trial of LY2157299 with paclitaxel in patients Triple negative metastatic Recruiting NCT02672475

o:10.1101 doi: with triple negative metastatic breast cancer breast Galunisertib Phase Ib dose-escalation and cohort-expansion study Metastatic pancreas cancer Not yet recruiting NCT02734160 galunisertib in combination with the anti-PDL-1 antibody (medi4736) in recurrent or / cshperspect.a022301 refractory metastatic pancreatic cancer Galunisertib Phase Ib/II galunisertib with anti-PD-1 () in Refractory solid tumors; Recruiting NCT02423343 advanced refractory solid tumors (phase Ib) and recurrent NSCLC, HCC, recurrent or refractory non-small-cell lung cancer, and glioblastoma hepatocellular carcinoma, or glioblastoma Continued Downloaded from http://cshperspectives.cshlp.org/ dacdOln ril.Ct hsatceas article this Cite Article. Online Advanced

Table 2. Continued Clinical trial Drug Target disease Organ site Status number References Galunisertib Study of LY2157299 immunomodulatory activity in Recruiting NCT02304419 patients with solid tumors Galunisertib LY2157299 and enzalutamide in metastatic castration- Prostate cancer Recruiting NCT02452008 resistant prostate cancer TEW-7197 First-in-human dose-escalation study of TEW-7197 Advanced stage solid tumors Recruiting NCT02160106 onSeptember25,2021-PublishedbyColdSpringHarborLaboratoryPress monotherapy in subjects with advanced stage solid odSrn abPrpc Biol Perspect Harb Spring Cold tumors Galunisertib Single-dose study to evaluate exposure–response Healthy patients Recruiting NCT02752919 relationship between LY2157299 and cardiac qt interval in healthy Japanese and non-Japanese subjects Fresolimumab Phase I study of fresolimumab to treat advanced Advanced melanoma and Completed NCT00923169 Morris et al. melanoma and renal cell carcinoma RCC 2008, 2014

Fresolimumab Guiding fresolimumab treatment of primary brain Gliomas Completed NCT01472731 Den Hollander TGF- Targeting tumors by 89Zr-fresolimumab PET imaging et al. 2015 o:10.1101 doi: Fresolimumab A phase II trial of fresolimumab in relapsed malignant Relapsed malignant pleural Completed NCT01112293 pleural mesothelioma mesothelioma Fresolimumab Safety and efficacy of fresolimumab and localized Metastatic breast cancer Active—not NCT01401062

/ radiotherapy in metastatic breast cancer recruiting b cshperspect.a022301

Fresolimumab Stereotactic ablative radiotherapy and fresolimumab in NSCLC Not yet recruiting NCT02581787 Gain Therapeutic for Signaling phase I/II trial for patients with stage IA-IB NSCLC IMC-TR1 (anti- Phase I study of anti-TbRII Advanced solid tumors that Completed NCT01646203 Cohn et al. TbRII antibody, LY3022859 in patients with advanced solid tumors failed other therapies 2014 LY3022859) that failed standard therapy NIS793 Phase I/Ib study of NIS793 in combination with Breast, lung, Not yet recruiting NCT02947165 PDR001 patients with advanced malignancies hepatocellular, prostate, renal cancer HCC, Hepatocellular carcinoma; MDS, myelodysplastic syndrome; NSCLC, non-small-cell lung cancer; RCC, renal cell carcinoma. 13 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

MONITORING AND PREVENTION OF TGF-b had no other signs of cardiovascular disease. BLOCKADE–INDUCED DRUG TOXICITIES Galunisertib was therefore concluded to be relatively safe in humans, especially compared On moving TGF-b blockade drugs into the with cardiotoxicities of other cancer drugs clinic, researchers used considerable caution be- (Benvenuto et al. 2003). On the basis of these cause of the early findings of the duplicitous findings, clinical trials were resumed incorpo- activity of TGF-b as either tumor promoter or rating an intermittent dosing schedule (2 weeks tumor suppressor during neoplastic progres- on/2 weeks off ). sion. Moreover, the finding of severe vascular Many have feared the de novo appearance and/or inflammatory defects in mice with si- of unrelated neoplasms or the enhanced out- lenced TGF-b1 expression (Shull et al. 1992; growth of the primary tumor in response to Kulkarni et al. 1993; Dickson et al. 1995) raised TGF-b signaling blockade. Epithelial hyperpla- additional concerns about the potential for un- sia has been seen in some animal models, in- desirable, possibly even life-threatening, side cluding tumor-promoting effects of LY2109761 effects of targeting TGF-b signaling in humans. in renal cell carcinoma (RCC) (Laping et al. For this reason, the transition into clinical trials 2007). There have also been preclinical reports and the patient accrual for trials proceeded of the ability of TGF-b blockade drugs to slowly. Even once commenced, the clinical tri- awaken dormant metastatic breast tumor cells als had long-term holds because of disquieting within the bone marrow, with the supposition findings of hemorrhagic lesions within the that TGF-b2 plays a major role in breast tumor heart valves, as well as aortic aneurysms in cell dormancy (Bragado et al. 2013). During rats and dogs after long-term continuous expo- a clinical trial of the pan anti-TGF-b anti- sure to the highest levels of TGF-b blockade body, fresolimumab, for the treatment of end- (Frazier et al. 2007; Laping et al. 2007; Ander- stage drug-refractile metastatic melanoma and ton et al. 2011). During hold of the clinical RCC, grade 1 or 2 skin rashes that improved or trial, considerable effort was invested in devel- resolved by the end of study were reported in 10 oping biomarkers of drug response and under- of 29 patients, and nonmalignant keratoacan- taking PK/PD modeling to define a better thomas (KA) appeared de novo in four of these therapeutic window for drug response (Gueor- patients (Morris et al. 2014), whereas a low- guieva et al. 2014). Phase I and II studies for grade SCC appeared in another individual galunisertib subsequently incorporated careful (Lacouture et al. 2015). Notably, KAs are com- monitoring of possible adverse cardiac events monly seen in response to other targeted ther- using circulating biomarkers, such as troponin apies such as consequent to treatment with the I, brain natriuretic protein (BNP) and high- multikinase inhibitor, sorafenib, or the B-Raf sensitivity C-reactive protein (hs-CRP), echo- enzyme inhibitor, vemurafinib (Arnault et al. cardiography Doppler imaging for possible mi- 2012; Lacouture et al. 2012), and these KAs are tral and tricuspid valve regurgitation, and considered a manageable side effect of cancer screening for potential aneurysms of the as- therapy. As yet, there have been no published cending aorta and aortic arch by computer to- clinical reports of promotion of other cancer mography (CT) or magnetic resonance imag- types, although this might be because clinical ing (MRI) (Fujiwara et al. 2015; Kovacs et al. studies to date have focused on severely ill 2015; Rodo´n et al. 2015b; Brandes et al. 2016). patients who did not survive long enough for In a clinical safety study (Kovacs et al. 2015), such an event to become apparent. We now only one of 79 patients treated showed an know that the TGF-b signaling pathway is increase in cardiopathologic grade, and then not the only Janus-faced drug target, because only from a baseline of normal to a mild pa- many oncogenic proteins and tumor suppres- thology, as assessed by Doppler. This patient sor gene products, including c-Myc, Ras, and was one of 13 receiving continuous administra- SnoN, have both pro- or antitumor activity tion of the highest drug dose for .6 mo, and (Zhang et al. 2001; Dang et al. 2005; Lamouille

14 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

and Derynck 2009). Furthermore, as cancer the drug is currently not under further clinical survivorship improves, it is becoming increas- development. Antisense TGF-b2 oligonucleo- ingly apparent that standard of care radiother- tides have also been under clinical development apy and chemotherapy can lead to later sec- for the treatment of glioblastoma (Isarna Ther- ondary neoplasms unrelated to the primary apeutics). Although early trials looked promis- tumor (Koo et al. 2015). The goal of eradicat- ing (Bogdahn et al. 2011), as with all antisense ing a primary or metastatic tumor, without approaches to date, systemic drug delivery immediately killing the host or causing longer has been an issue, with the need for localized term negative outcomes remains a major chal- continuous delivery directly into the brain lenge in oncology. tumor by catheter and pump. A next-generation TGF-b2-selective antisense oligonucleotide, ISTH0047, has entered clinical trial for ophthal- UPDATE ON INTERVENTIONAL CLINICAL mology applications (NCT02406833), and the ONCOLOGY TRIALS WITH TGF-b oncology pipeline remains in preclinical stages. BLOCKADE

Antisense RNA Approaches Therapeutic Antibodies The first clinical trials targeting the TGF-b Two phase I trials of fresolimumab monotherapy pathway for cancer treatment were designed to have been completed, one focusing on metastatic enhance the immune system of patients with melanoma (Morris et al. 2014) and another NSCLC using a genetically modified tumor vac- on high-grade glioblastoma patients (Den Hol- cine. The irradiated, and therefore nonprolifer- lander et al. 2015). Fresolimumab is a high- ative vaccine, belagenpumatucel-L (Lucanix), is affinity fully humanized monoclonal antibody composed of four different allogeneic human that neutralizes the active forms of human TGF- NSCLC cell lines transfected with a gene encod- b1, TGF-b2, and TGF-b3(KDs of 2.3 nM, ing antisense TGFB2. When injected into the 2.8 nM, and 1.3 nM, respectively) (Rice et al. patient, Lucanix promotes an antihost NSCLC 2015). The drug has anti-TGF-b activity in hu- cytotoxic T cell response, and vaccine immuno- mans, as assessed by biomarker analysis (Rice genicity is postulated to be potentiated by re- et al. 2015), and was found to be safe, albeit duced TGF-b2 production. Phase I and II trials there was de novo appearance of cutaneous be- went well, the drug was found to be safe, and the nign KAs and one SCC in melanoma patients, phase II results suggested some therapeutic when used at the highest drug dosing level benefit (Nemunaitis et al. 2006, 2009). Howev- (15 mg/kg) (Morris et al. 2014). A phase I trial er, a phase III trial to examine benefit as a of fresolimumab for treatment of systemic NSCLC maintenance therapy following prima- sclerosis using much lower doses than for on- ry treatment (270 Lucanix-treated vs. 262 pla- cology studies (1–5 mg/kg), showed unprece- cebo-treated NSCLC patients) did not meet its dented resolution of skin disease, although primary endpoint criterion of increased overall most patients developed anemia consequent survival (OS) (Giaccone et al. 2015). Stratifica- to GI bleeding (not considered a serious adverse tion of the data revealed a marginal but signifi- effect), which was most likely specific to com- cant therapeutic benefit for NSCLC patients plications of their disease rather than the drug who started Lucanix within 12 weeks of their alone (Rice et al. 2015). Clinical studies of initial chemotherapy (166 drug-treated vs. 149 glioblastoma uptake of 89Zr-labeled fresolimu- control NSCLC patients), and in those receiv- mab after intravenous delivery indicated effi- ing prior radiation therapy (RT), which might cient delivery of the drug into this brain tumor, boost tumor antigenicity (median survival which appeared to be TGF-b-dependent, be- 28.4 mo for RTwith belagenpumatucel-L treat- cause all glioblastomas showing uptake had el- ment vs. 16.0 mo for RT plus placebo; HR 0.61, evated pSmad2 by IHC (Den Hollander et al. p ¼ 0.032) (Giaccone et al. 2015). Nevertheless, 2015). This was the first study to show therapeu-

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 15 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

tic antibody delivery to a human brain tumor, LY2382770 has a t1/2 of only 9 d (Cohn et al. reflecting antibody leakage through a damaged 2014). Metastatic melanoma was not included blood brain barrier. The metastatic melanoma in the LY2382770 study, so it is possible that trial showed some evidence of clinical benefit discordant results with fresolimumab (Morris (even though this was a phase I trial), with six et al. 2014) relate to the distinct tumor types of 29 patients achieving stable disease, including investigated between the two studies, with mel- those with mixed tumor responses as assessed anoma being particularly sensitive to TGF-b by MRI, and one patient with multiple cutane- inhibition. A TbRII antibody (IMC-TR1, also ous lesions showing a 89% partial response known as LY3022859) has also been developed, that lasted 44 wk (Morris et al. 2014). This was and the murinized derivative showed an excel- despite the fact that these patients had not re- lent response in mouse models of breast and sponded to other therapies (Morris et al. 2014). colon cancer (Zhong et al. 2010). This drug is Fresolimumab oncology trials were halted for in phase I trial for patients with advanced solid administrative reasons to prioritize efforts on tumors who have failed standard therapies antifibrosis applications. However, with the (NCT01646203). recent resurgence of interest in TGF-b blockade to enhance immunotherapies; this decision may Small Molecule Kinase Inhibitors be reconsidered, especially in light of the fact that some clinical benefit was observed in the In the field of SMIs, some of the first clinical first phase I trial for metastatic melanoma. trials for TbRI inhibitors have recently been An 18-patient dose escalation study of published (Fujiwara et al. 2015; Kovacs et al. LY2382770, a TGF-b1-specific antiligand IgG4 2015; Rodo´n et al. 2015a,b; Sepulveda-Sa´nchez mAb evaluated by Eli Lilly, dosed once monthly et al. 2015; Brandes et al. 2016), with more in the over a range of 20 to 240 mg total antibody per pipeline. So far, all published clinical studies patient per month, was generally found to be used the TbRI SMI drug, galunisertib, while safe (the primary endpoint), with fatigue, nau- an additional twelve interventional trials are on- sea, and diarrhea as most common side effects going with this drug, and another trial uses the (17% of 18 patients) (Cohn et al. 2014). How- TbRI SMI TEW-7197 (Table 2) (Jin et al. 2014; ever, most patients discontinued from the pro- Son et al. 2014; Naka et al. 2016). Ongoing tocol after only two cycles of treatment owing to studies with galunisertib include its use as progressive disease with no effect on tumor bur- monotherapy with or without standard of den. As designed, the major endpoint of the care, such as with the alkylating agents, lomus- study was PD analysis of a 37-gene signature tine or temazolamide in radiochemotherapy for of TGF-b expression in circulating mononucle- glioblastoma (NCT01682187, NCT01582269, ar cells. Despite some marginal pre- to post- NCT01220271), its use with the antimetabolite, treatment changes, the gene expression changes gemcitabine, for metastatic pancreatic cancer were not consistent or significant, which might (NCT01373164), and with or without sorafenib account for lack of any clinical response (Cohn for hepatocellular carcinoma (NCT01246986, et al. 2014). Notably, phase I trials are not de- NCT02178358). Other trials include com- signed to test efficacy and lack the statistical bination or not with radiotherapy for breast power to do so. The apparent lack of concor- (NCT02538471) or rectal cancer (NCT02688 dance between the insignificant clinical out- 712), and with the antiandrogen drug, enzalu- come data (Cohn et al. 2014) and results using tamide versus galunisertib versus drug com- fresolimumab in metastatic melanoma (Morris bination in metastatic castration-resistant et al. 2014) might be explained by the need prostate cancer (NCT02452008). A trial of ga- to additionally target TGF-b2 (and possibly lunisertib in combination with the checkpoint TGF-b3) to elicit a robust response and/or the inhibitor, nivolumab (anti-PD-1, Bristol-Myers insufficient antibody concentration and/or Squibb) is recruiting patients with recurrent or dosing regimen of once per month, because advanced NSCLC (NCT02423343).

16 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

As discussed earlier, the discovery of serious blood volume and tumor perfusion, as seen in cardiac toxicity in dogs and rats after continu- response to the anti-VEGF antibody bevaci- ous administration of any TbRI SMI (Laping zumab, the galunisertib effect may have been et al. 2007; Anderton et al. 2011) necessitates associated with effects on the vasculature (Se- the establishment of a therapeutic window pulveda-Sa´nchez et al. 2015). However, this late between disease response and cardiotoxicity. response phenomenon is also a more general These studies were undertaken based on PK/ feature of cancer immunotherapies (Tuma PD modeling in mouse, rat and dog, integrating 2008; Hodi et al. 2010; Hoos et al. 2010), impli- plasma drug levels and magnitude of inhibi- cating a likely effect of galunisertib on potenti- tion of Smad2 phosphorylation in circulating ating tumor immunity as well. mononuclear cells, relative to onset of tumor A phase II randomized study of 158 recur- responses versus valvulopathy (Gueorguieva rent glioblastoma patients treated with galuni- et al. 2014). The therapeutic window turned sertib with or without lomustine compared out to be narrow, between 160 mg and 300 mg with lomustine alone, found that the drug com- per day (administered as two doses of 80– bination showed no improvement in OS com- 150 mg), with dosing in cycles of 2 weeks on pared with lomustine plus placebo. However, (bi-daily treatment) and 2 weeks off drug. galunisertib monotherapy was equally effective The FHD study of galunisertib monother- as lomustine and conferred a clinical advantage apy with or without lomustine chemotherapy in showing less serious hematologic side effects. assessed 58 glioblastoma patients, all of whom Tantalizingly, in the FHD study (Rodo´n et al. had relapsed or progressed on prior effective 2015b), the tumor drug delivery study (Sepul- therapies. Most patients were only able to re- veda-Sa´nchez et al. 2015), and the phase II trial ceive two cycles of galunisertib before tumor (Brandes et al. 2016), galunisertib monotherapy progression, but clinical benefit was seen in showed a trend toward better outcomes com- .20% of patients in terms of complete re- pared with both the lomustine monotherapy sponse (CR), partial response (PR), or stable and drug combination arms. Not only was disease (SD) for more than four drug cycles this seen in the trend toward increased median (.5 mo), with a 6.6% complete response rate. OS, but possibly more important, the censoring Galunisertib monotherapy had better clinical rate (patient survival at study termination) outcomes than combination therapy with lo- within the 2-year span of the phase II study mustine, although this differential is not signif- was 23% for galunisterib monotherapy versus icant (Rodo´n et al. 2015b). Intriguingly, the pa- only 10% for its combination with lomustine, tients who did show a complete or partial and 15% for lomustine alone (Brandes et al. response first went through a phase of stable 2016). disease, with one patient having a complete re- In the phase II study of galunisertib (Bran- sponse as late as drug cycle 28 (Rodo´n et al. des et al. 2016), a number of baseline character- 2015b). This delayed response was also seen in istics were measured to screen for prognostic a smaller study designed to examine clinical re- markers. High-baseline levels of circulating sponses to a galunisertib–lomustine combi- CD3þ and Foxp3þ cells, CD4þCD25þCD1272 nation therapy, assessed by dynamic contrast FOXP3LO cells and eosinophils associated with enhanced MRI to quantify vascular brain per- higher circulating levels of macrophage-derived fusion and permeability (Sepulveda-Sa´nchez chemokine (MDC or CCL22). Moreover, high- et al. 2015). In the latter study, two of 12 glio- median MDC was associated with longer OS blastoma patients showed clinical benefit, but (11 mo) than low-median MDC (4.7 mo). responses did not begin until the fourth to sixth This correlation was seen across all experimen- cycle of galunisterib therapy, contrasting with tal arms, and has previously been reported as a the normally immediate response (,1 mo) to prognostic factor for glioblastoma in general lomustine chemotherapy. Because one of two (Zhou et al. 2015). Circulating TGF-b levels responding patients showed reduced cerebral did not significantly associate with OS, al-

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 17 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

though nine patients with TGF-b1 levels greater vascular stability and/or immune-mediated than the median, who were treated with galuni- tumor rejection. The nonadditive nature of sertib monotherapy, had a median OS of 11 mo, the galunisertib and lomustine responses and compared with an OS of 4.7 mo for the other 30 the suggestive trend toward reduced efficacy of patients in the same treatment arm whose galunisertib when combined with lomustine, pretreatment TGF-b1 levels were less than the might suggest a predominantly immune-medi- median (Brandes et al. 2016). A tentative con- ated mechanism of action, impaired by the al- clusion from all three trials is that low-grade or kylating effects of lomustine on immune cells. secondary gliomas respond better than high- Galunisertib may indeed be most effective with grade tumors, especially those with an IDH agents that stimulate rather than deplete the gene mutation; four out of five (Rodo´n et al. immune system (Garrison et al. 2012; Triplett 2015a,b) and three out of seven (Brandes et al. et al. 2015). 2016) IDH mutation carriers had CR/PR or SD to galunisterib compared with 12%–17% TGF-b BLOCKADE IN COMBINATION WITH overall. CHEMO- OR RADIATION THERAPY In the phase II study of galunisertib, patients with disease progression were offered bevacizu- Because TGF-b inhibitors are not directly cyto- mab as an alternative therapy, which may have toxic, it has been postulated that their combined impacted longer-term outcomes. In fact, there use with cytotoxic chemotherapeutics may be was only one complete response to galunisertib highly efficacious. Several papers have shown during the 2-year study period (Brandes et al. the preclinical efficacy of targeting tumors 2016), despite earlier and smaller studies pro- with combinations of cytotoxic drugs and viding apparently better outcomes. In the FHD TGF-b blockade. Synergism between the DNA study, 16.6% (5/30) and 7.7% (2/26) of glioma intercalating agent, doxorubicin, and TGF-b patients showed CR or PR, and five patients had inhibition has been shown to suppress breast SD for more than six cycles of treatment, with cancer growth and metastasis in a preclinical two patients surviving for at least 3 years follow- model (Bandyopadhyay et al. 2010), and TGF- ing complete response (Rodo´n et al. 2015b). b inhibitors potentiate the cytotoxic effects of Overall, the results of the galunisertib phase II the alkylating agent, melphalan, and the corti- study appear disappointing as they did not costeroid, dexamethasone, in a model of multi- reach the endpoint of enhanced efficacy in com- ple myeloma (Takeuchi et al. 2010). Exposure of bination with lomustine. However, the data may multiple myeloma cells to differentiated versus be consistent with the possibility that the two immature MC3T3-E1 pro-osteoblastic cells in drugs antagonized each other. It is, however, cell culture potentiates chemotherapy-induced encouraging that 235 cancer patients have now myeloma cell death. Because TGF-b inhibition been treated with galunisertib, yet show no re- promotes osteoblastic differentiation within the markable cardiovascular toxicities despite, in bone microenvironment (Yin et al. 1999; Take- some cases, treatment over 3 years (Rodo´n uchi et al. 2010), this combinatorial approach et al. 2015a). Moreover, outcomes in glioblasto- holds promise for treatment of other bone tro- ma are at least as good as standard of care lo- pic metastatic cancers. As mentioned earlier, mustine, and manifest less severe side effects combining galunisertib with the microtubule than the latter. The clinical studies have given disruptor, paclitaxel, gives a better outcome in credence to a role for galunisertib in depleting a mouse model of breast cancer than paclitaxel glioblastoma stem cells, albeit in only one pa- alone, and the TbRI SMI, Ki26894, has an ad- tient (Anido et al. 2010), and in action on the ditive effect with a fluorouracil analog in reduc- TME, particularly vascular disruption (Sepul- ing tumor growth in a mouse model of scir- veda-Sa´nchez et al. 2015). The delayed response rhous gastric cell carcinoma (Shinto et al. to galunisertib therapy may be consistent with 2010). However, galunisertib showed no clinical changes in tumor cell plasticity, angiogenesis/ benefit when combined with standard of care

18 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

lomustine, an alkylating chemotherapy (Bran- TGF-b SIGNALING BLOCKADE IN des et al. 2016). There has been considerable interest in com- bining TGF-b pathway inhibition and radio- Although most clinical trials of TGF-b blockade therapy (Anscher et al. 2008; Bouquet et al. have not shown dramatic clinical benefit by 2011; Zhang et al. 2011). Radiation physically classical RECIST (response evaluation criteria activates the latent TGF-b complex in cell cul- in solid tumors) (Therasse et al. 2000), it is still ture and in vivo (Barcellos-Hoff 1993; Barcellos- early days in TGF-b blockade trials, with only Hoff et al. 1994). In cells in culture, radiation some drugs tested so far in phase I dose escala- induces biological release of TGF-b as part tion studies for assessment of safety. Placing the of the radiation-induced stress response. TGF- first phase I and II clinical data into context, one b plays an important role in supporting the has to consider that even a blockbuster immu- DNA damage repair pathway, particularly notherapy agent such as (anti- through activation of p53 and phosphorylation CTLA-4) showed similar deficiencies in its first of ataxia telangiectasia mutated (ATM), a serine/ clinical trials. An early clinical trial of ipilimu- threonine kinase that is recruited and activated mab in metastatic melanoma showed an objec- by DNA double-strand breaks. LY2109761, tive clinical response in only five out of 46 pa- which targets the kinase activities of TbRII tients (11%), at the expense of 35% of patients and TbRI (Sawyer et al. 2003; Yingling et al. having grade three or four autoimmune toxici- 2004), and the antipan TGF-b antibody, 1D11 ties (Maker et al. 2006), which is not so different (Morris et al. 2014), attenuate radiation-in- from the metastatic melanoma response to fre- duced activation of p53 and ATM in breast solimumab (six of 29 patients achieving stable cancer cells in culture and in vivo, suggesting disease, one for 44 wk) (Morris et al. 2014). It that TGF-b signaling may potentiate thera- is now generally accepted that the classical RE- peutic killing of tumors by preventing DNA SIST or World Health Organization response repair and accentuating the cytotoxic effect criteria developed for patients on cytotoxic of radiation (Bouquet et al. 2011). Further- drugs are not appropriate for assessing immu- more, TGF-b blockade with the 1D11 anti- notherapies (Tuma 2008; Hoos et al. 2010). For body significantly stimulates the anticancer example, responses to checkpoint inhibitors immune response against implanted mamma- such as ipilimumab or nivolumab are frequent- ry tumors (4T1 cells) following RT, resulting ly considerably delayed, with a protracted peri- in tumor clearance when combined with RT od of months of stable disease or even tumor (Vanpouille-Box et al. 2015). Even short-term progression before antitumor activity becomes dosing with TGF-b inhibitors might provide a apparent. Comparison of median OS may not considerable therapeutic window in potentiat- be an adequate parameter when only a small ing radiotherapy. Moreover, RT can cause fraction of patients shows a response, even disfiguring and painful fibrosis, itself a target though the response may be complete, robust, for anti-TGF-b therapy, possibly providing an and durable, possibly lasting for years. A more added benefit of TGF-b inhibition (Anscher significant measure should therefore be assess- et al. 2008). Thus, TGF-b blockade may have ment of the fraction of patients with long-term a triad of benefits, stimulating radiation- durable responses, which was not an endpoint induced tumor cell killing (Bouquet et al. in the TGF-b blockade trials. The fraction of 2011), augmenting immune rejection (Van- glioblastoma or metastatic melanoma patients pouille-Box et al. 2015), while limiting radia- achieving long-term durable responses after ga- tion-induced fibrosis (Rabbani et al. 2003; lunisertib or fresolimumab therapy (Morris Anscher et al. 2008). This approach is part et al. 2014; Brandes et al. 2016), may become of an active clinical trial of fresolimumab in significant if combined with other drugs. Bear- combination with radiotherapy for metastatic ing in mind the safety of TGF-b signaling in- breast cancer (NCT01401062). hibitors at the doses used, the outlook for use of

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 19 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

these drugs as monotherapy or as adjunct to dividuals helps define the outcomes of germline other immunotherapy agents is very good. genetic loss of TGF-b signaling components, Proof-of-principle usage of TGF-b block- and thus might influence both therapeutic ade for enhancement of immunotherapy has drug responses and on-target side effects (Bo- been shown preclinically using LY2109761 in nyadi et al. 1997; Akhurst 2004; Benzinou et al. adoptive T-cell therapy (Wallace et al. 2008), 2012; Chung Kang et al. 2013; Kawasaki et al. and using the normally unstable TbRI SMI, 2014). Taking into account the many activities SB505124, encapsulated in a slow-release bio- of TGF-b and its pleiotropic actions, major ef- degradable polymeric nanoparticle for the forts focus on identifying alternative targets that stimulation of 2 (IL-2) therapy of influence the TGF-b signaling pathway, with a melanoma (Park et al. 2012). A novel approach strong incentive to identify drugs that might has been taken by generating mRNA encoding a block TGF-b’s tumor progressing activities “fusokine,” that is, a fusion protein of the im- while sparing tumor-suppressing activities. mune stimulant, interferon-b, with the ectodo- Many studies identified pivotal intratumoral main of TbRII, which acts as a TGF-b “sink.” players that regulate such a balance—for exam- DCs exposed to this fusokine take up the RNA ple, disabled homolog 2 (DAB2), which is epi- by phagocytosis and express the encoded fusion genetically down-regulated in human SCCs, protein, resulting in enhanced antigen-present- resulting in attenuation of TGF-b tumor sup- ing capacity. The fusokine also attenuates the pressor activity to enable TGF-b tumor-pro- suppressor activity of myeloid-derived suppres- moting activities, including promotion of cell sor cells (MDSCs) after transfection of the en- motility, anchorage-independent growth, and coding plasmid into these cells (Van Der Jeught in vivo tumor growth (Hannigan et al. 2010). et al. 2014). It has been postulated that intra- Another example includes the pivotal role tumoral delivery of this fusokine RNA might played by differential expression of three iso- enhance antigen presentation, and thus tumor forms of C/EBPb that are generated by usage immunity after intratumoral injection in vivo, of distinct translational start sites during tumor but this remains to be tested. Finally, an agonist progression. In normal cells, the major C/EBPb antibody against the immune activating cell isoforms LAP1 and LAP2 bind pSmads to elicit surface receptor, Ox40, which is expressed on a growth suppressive transcriptional response. CD4þ and CD8þ cells, has been shown to syn- However, as tumors progress, there is preferen- ergize with the TbRI SMI, SM16, leading to tial translation of a third truncated isoform, LIP, complete tumor regression in models of colon which lacks the amino-terminal domain and cancer (CT26) and chemically induced sarcoma acts in a dominant-negative manner to suppress (MCA205) (Triplett et al. 2015). the growth inhibitory action of LAP1/2 during cancer progression (Gomis et al. 2006b). Target- ing such intracellular players is however likely FUTURE DIRECTIONS WITH TGF-b destined for failure because of the incredible BLOCKADE plasticity and evolution of tumor cells, which Several clinical oncology trials with galuniser- evolve rapidly to circumnavigate signaling path- tib, fresolimumab, and LY3022859 are listed as way blockade, as exemplified by clinical trials ongoing by ClinicalTrials.gov (Table 2), and the with drugs that target other pathways, such as initial reports suggest that galunisertib is at least HER2 (Johannessen et al. 2010; Nazarian et al. as efficacious as lomustine for glioblastoma and 2010) and B-Raf (Sergina et al. 2007). A more may have fewer side effects. Nevertheless, pro- fruitful approach may be to target proteins with gress through the clinical pipeline has been slow. a more restricted tissue expression profile than It may be that the window between antitumor that of the TGF-b receptors, which might affect efficacy and toxicological effects has been con- the TGF-b activity only in specific cells of the sidered too narrow for general use, especially TME such as in DCs, NK cells, MDSCs and/or considering that genetic variation between in- T cells. Integrins that activate TGF-b have been

20 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

considered as such drug targets (Van Aarsen approved for widespread use, there will be con- et al. 2008; Dutta et al. 2014). Another example siderable variation in both therapeutic and tox- is GARP, which is required for activation of la- icological responses among patients (Bonyadi tent TGF-b on Tregs and platelets (Stockis et al. et al. 1997; Akhurst 2004; Valle et al. 2008; Ben- 2009; Tran et al. 2009). Blocking the interaction zinou et al. 2012; Chung Kang et al. 2013; Ka- of GARP with latent TGF-b at the surface of wasaki et al. 2014). There is therefore a growing Treg cells using antibodies that bind a specific need for development of predictive biomarkers conformational motif of GARP could block of TGF-b blockade (Gueorguieva et al. 2014; TGF-b activation, and, consequently, also the Kawasaki et al. 2014). immunosuppressive activity of Tregs (Cuende et al. 2015). This antibody may therefore im- pose a more specific TGF-b blockade immuno- ACKNOWLEDGMENTS therapy for cancer, probably with fewer side ef- Workin the Akhurst laboratory is funded by the fects than blocking TGF-b receptors or ligands, National Cancer Institute, the National Heart, because of the restricted pattern of GARP ex- Lung, and Blood Institute, and the Helen Diller pression. Other TGF-b modulatory drugs are in Family Comprehensive Cancer Center. R.J.A. development for various disease applications has ongoing consulting relationships or con- including oncology (Van Aarsen et al. 2008; tracts with Xoma Corporation and the Pfizer Henderson et al. 2013; Salvo et al. 2014). Center for Translational Innovation that have, in part, funded work in the Akhurst Laboratory. CONCLUDING REMARKS Compared with many other anticancer drugs REFERENCES in use, TGF-b blockade would, so far, appear Akhurst RJ. 2004. TGF-b signaling in health and disease. relatively safe. For example, ipilimumab (anti- Nat Genet 36: 790–792. CTLA4) has shown major successes as an im- Akhurst RJ, Derynck R. 2001. TGF-b signaling in cancer— mune checkpoint blockade agent for melanoma A double-edged sword. Trends Cell Biol 11: S44–51. Akhurst RJ, Hata A. 2012. Targeting the TGF-b signalling but has considerable toxicity in some patients, pathway in disease. Nat Rev Drug Discov 11: 790–811. causing GI inflammation, asthma, vitiligo, and Akhurst RJ, Fee F,Balmain A. 1988. Localized production of other autoimmune-like diseases (Spain et al. TGF-b mRNA in tumour promoter-stimulated mouse 2016). Despite this major caveat, the drug is epidermis. Nature 331: 363–365. still in considerable demand by patients, but Anderton MJ, Mellor HR, Bell A, Sadler C, Pass M, Powell S, Steele SJ, Roberts RR, Heier A. 2011. Induction of heart requires close monitoring for drug-related au- valve lesions by small-molecule ALK5 inhibitors. Toxicol toimmune toxicities. The next stage in drug de- Pathol 39: 916–924. velopment for TGF-b blockade will be discov- Anido J, Sa´ez-Borderı´as A, Gonza`lez-Junca` A, Rodo´nL, Folch G, Carmona MA, Prieto-Sa´nchez RM, Barba I, ering the most efficacious drug combinations Martı´nez-Sa´ez E, Prudkin L, et al. 2010. TGF-b receptor for each oncological disease application, wheth- inhibitors target the CD44high/Id1high glioma-initiating er this is radiation, chemotherapy, pathway tar- cell population in human glioblastoma. Cancer Cell 18: geted therapies and/or immunotherapy, and 655–668. Annes JP, Munger JS, Rifkin DB. 2003. Making sense of defining the most potent dosing regimens. It latent TGF-b activation. J Cell Sci 116: 217–224. is possible that antiligand or antireceptor drugs Annes JP, Chen Y, Munger JS, Rifkin DB. 2004. Integrin may not provide a large enough therapeutic aVb6-mediated activation of latent TGF-b requires the window between therapeutic and toxicological latent TGF-b binding protein-1. J Cell Biol 165: 723–734. Anscher MS, Thrasher B, Zgonjanin L, Rabbani ZN, Corbley responses. This could be remediated by opti- MJ, Fu K, Sun L, Lee WC, Ling LE, Vujaskovic Z. 2008. mized dosing strategies, but may require the Small molecular inhibitor of transforming growth factor- development of new drugs against tissue-specif- b protects against development of radiation-induced ic TGF-b signaling components that are active lung injury. Int J Radiat Oncol Biol Phys 71: 829–837. Arnault JP, Mateus C, Escudier B, Tomasic G, Wechsler J, only on discrete cell types of the TME, such as Hollville E, Soria JC, Malka D, Sarasin A, Larcher M, et al. GARP. Finally, if TGF-b blockade drugs are 2012. Skin tumors induced by sorafenib; paradoxic RAS-

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 21 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

RAF pathway activation and oncogenic mutations of triple-negative breast cancer. J Clin Invest 123: 1348– HRAS, TP53, and TGFBR1. Clin Cancer Res 18: 263–272. 1358. Arnold T, Niaudet C, Pang MF, Siegenthaler J, Gaengel K, Bogdahn U, Hau P, Stockhammer G, Venkataramana NK, Jung B, Ferrero GM, Mukouyama YS, Fuxe J, Akhurst R, Mahapatra AK, Suri A, Balasubramaniam A, Nair S, et al. 2014. Excessive vascular sprouting underlies cere- Oliushine V, Parfenov V, et al. 2011. Targeted therapy bral hemorrhage in mice lacking integrin aVb8-TGF-b for high-grade glioma with the TGF-b2 inhibitor trabe- signaling in the brain. Development 141: 4489–4499. dersen: Results of a randomized and controlled phase IIb Arteaga CL, Hurd SD, Winnier AR, Johnson MD, Fendly study. Neuro Oncol 13: 132–142. BM, Forbes JT. 1993. Anti-transforming growth factor Bonyadi M, Rusholme SA, Cousins FM, Su HC, Biron CA, (TGF)-b antibodies inhibit breast cancer cell tumorige- Farrall M, Akhurst RJ. 1997. Mapping of a major genetic nicity and increase mouse spleen natural killer cell activ- modifier of embryonic lethality in TGF-b1 knockout ity. Implications for a possible role of tumor cell/host mice. Nat Genet 15: 207–211. TGF-b interactions in human breast cancer progression. Bouquet F, Pal A, Pilones KA, Demaria S, Hann B, Akhurst J Clin Invest 92: 2569–2576. RJ, Babb JS, Lonning SM, DeWyngaert JK, Formenti SC, Ask K, Bonniaud P, Maass K, Eickelberg O, Margetts PJ, et al. 2011. TGF-b1 inhibition increases the radiosensi- Warburton D, Groffen J, Gauldie J, Kolb M. 2008. Pro- tivity of breast cancer cells in vitro and promotes tumor gressive pulmonary fibrosis is mediated by TGF-b iso- control by radiation in vivo. Clin Cancer Res 17: 6754– form 1 but not TGF-b3. Int J Biochem Cell Biol 40: 484– 6765. 495. Bragado P, Estrada Y, Parikh F, Krause S, Capobianco C, Bandyopadhyay A, WangL, Agyin J, TangY,Lin S, YehIT,De Farina HG, Schewe DM, Aguirre-Ghiso JA. 2013. TGF- K, Sun LZ. 2010. Doxorubicin in combination with a b2 dictates disseminated tumour cell fate in target organs small TGF-b inhibitor: A potential novel therapy for through TGF-b–RIII and p38a/b signalling. Nat Cell metastatic breast cancer in mouse models. PLoS ONE 5: Biol 15: 1351–1361. e10365. Brandes AA, Carpentier AF, Kesari S, Sepulveda-Sa´nchez Barcellos-Hoff MH. 1993. Radiation-induced transforming JM, Wheeler HR, Chinot O, Cher L, Steinbach JP,Capper growth factor b and subsequent extracellular matrix re- D, Specenier P,et al. 2016. A phase II randomized study of organization in murine mammary gland. Cancer Res 53: galunisertib monotherapy or galunisertib plus lomustine 3880–3886. compared with lomustine monotherapy in patients with Barcellos-Hoff MH, Dix TA. 1996. Redox-mediated activa- recurrent glioblastoma. Neuro Oncol 18: 1146–1156. tion of latent transforming growth factor-b1. Mol Endo- Brown JD, DiChiara MR, Anderson KR, Gimbrone MA Jr, crinol 10: 1077–1083. Topper JN. 1999. MEKK-1, a component of the stress Barcellos-Hoff MH, Derynck R, Tsang ML, Weatherbee JA. (stress-activated protein kinase/c-Jun N-terminal ki- 1994. Transforming growth factor-b activation in irradi- nase) pathway, can selectively activate Smad2-mediated ated murine mammary gland. J Clin Invest 93: 892–899. transcriptional activation in endothelial cells. J Biol Chem Bedinger D, Lao L, Khan S, Lee S, Takeuchi T, Mirza AM. 274: 8797–8805. 2016. Development and characterization of human Bueno L, De Alwis DP,Pitou C, Yingling J, Lahn M, Glatt S, monoclonal antibodies that neutralize multiple TGFb Troconiz IF. 2008. Semi-mechanistic modelling of the isoforms. mAbs 8: 389–404. tumour growth inhibitory effects of LY2157299, a new Bellone G, Turletti A, Artusio E, Mareschi K, Carbone A, type I receptor TGF-b kinase antagonist, in mice. Eur J Tibaudi D, Robecchi A, Emanuelli G, Rodeck U. 1999. Cancer 44: 142–150. Tumor-associated transforming growth factor-b and in- Caulin C, Scholl FG, Frontelo P,Gamallo C, Quintanilla M. terleukin-10 contribute to a systemic Th2 immune phe- 1995. Chronic exposure of cultured transformed mouse notype in pancreatic carcinoma patients. Am J Pathol 155: epidermal cells to transforming growth factor-b1 induces 537–547. an epithelial-mesenchymal transdifferentiation and a Bellone G, Carbone A, Tibaudi D, Mauri F,Ferrero I, Smirne spindle tumoral phenotype. Cell Growth Diff 6: 1027– C, Suman F,Rivetti C, Migliaretti G, Camandona M, et al. 1035. 2001. Differential expression of transforming growth fac- Chamberlain MC. 2011. Convection-enhanced delivery of a tors-b1, -b2 and -b3 in human colon carcinoma. Eur J transforming growth factor-b2 inhibitor trabedersen for Cancer 37: 224–233. recurrent high-grade gliomas: Efficacy real or imagined? Benvenuto GM, Ometto R, Fontanelli A, Fortunato A, Ruf- Neuro Oncol 13: 558–559. fini PA, Fosser V,Morandi P.2003. Chemotherapy-related Chaudhry SS, Cain SA, Morgan A, Dallas SL, Shuttleworth cardiotoxicity: New diagnostic and preventive strategies. CA, Kielty CM. 2007. Fibrillin-1 regulates the bioavail- Italian Heart J 4: 655–667. ability of TGF-b1. J Cell Biol 176: 355–367. Benzinou M, Clermont FF,Letteboer TG, Kim JH, Espejel S, Chen ML, Pittet MJ, Gorelik L, Flavell RA, WeisslederR, Von Harradine KA, Arbelaez J, Luu MT, Roy R, Quigley D, et Boehmer H, Khazaie K. 2005. Regulatory T cells suppress al. 2012. Mouse and human strategies identify PTPN14 as tumor-specific CD8 T cell cytotoxicity through TGF-b a modifier of angiogenesis and hereditary haemorrhagic signals in vivo. Proc Natl Acad Sci 102: 419–424. telangiectasia. Nat Commun 3: 616. Chen M, Matsuda H, Wang L, Watanabe T, Kimura MT, Bhola NE, Balko JM, Dugger TC, Kuba MG, Sa´nchez V, Igarashi J, Wang X, Sakimoto T, Fukuda N, Sawa M, et Sanders M, Stanford J, Cook RS, Arteaga CL. 2013. al. 2010. Pretranscriptional regulation of TGF-b1byPI TGF-b inhibition enhances chemotherapy action against polyamide prevents scarring and accelerates wound heal-

22 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

ing of the cornea after exposure to alkali. Mol Therapy 18: Derynck R, Zhang YE. 2003. Smad-dependent and Smad- 519–527. independent pathways in TGF-b family signalling. Na- Chung Kang H, Quigley DA, Kim IJ, Wakabayashi Y, Fergu- ture 425: 577–584. son-Smith MA, D’Alessandro M, Birgitte Lane E, Derynck R, Goeddel DV,Ullrich A, Gutterman JU, Williams Akhurst RJ, Goudie DR, Balmain A. 2013. Multiple RD, Bringman TS, Berger WH. 1987. Synthesis of mes- self-healing squamous epithelioma (MSSE): Rare vari- senger RNAs for transforming growth factors a and b ants in an adjacent region of chromosome 9q22.3 to and the epidermal growth factor receptor by human tu- known TGFBR1 mutations suggest a digenic or multi- mors. Cancer Res 47: 707–712. locus etiology. J Invest Derm 133: 1907–1910. Derynck R, Akhurst RJ, Balmain A. 2001. TGF-b signaling Coffey RJ Jr, Sipes NJ, Bascom CC, Graves-Deal R, Penning- in tumor suppression and cancer progression. Nat Genet ton CY, Weissman BE, Moses HL. 1988. Growth modu- 29: 117–129. lation of mouse keratinocytes by transforming growth Descargues P,Sil AK, Sano Y,Korchynskyi O, Han G, Owens factors. Cancer Res 48: 1596–1602. P, Wang XJ, Karin M. 2008. IKKa is a critical coregula- Cohn A, Lahn MM, Williams KE, Cleverly AL, Pitou C, tor of a Smad4-independent TGFb-Smad2/3 signaling Kadam SK, Farmen MW, Desaiah D, Raju R, Conkling pathway that controls keratinocyte differentiation. Proc P,et al. 2014. A phase I dose-escalation study to a prede- Natl Acad Sci 105: 2487–92. fined dose of a transforming growth factor-b1 monoclo- Dickson MC, Martin JS, Cousins FM, Kulkarni AB, Karlsson nal antibody (TbM1) in patients with metastatic cancer. S, Akhurst RJ. 1995. Defective haematopoiesis and vas- Int J Oncol 45: 2221–2231. culogenesis in transforming growth factor-b1 knock out mice. Development 121: 1845–1854. Connolly EC, Saunier EF, Quigley D, Luu MT, De Sapio A, Hann B, Yingling JM, Akhurst RJ. 2011. Outgrowth of Dong Z, Guo W,Guo Y,Kuang G, YangZ. 2012. Concordant drug-resistant carcinomas expressing markers of tumor promoter methylation of transforming growth factor-b aggression after long term TbRI/II kinase inhibition with receptor types I and II occurs early in esophageal squa- LY2109761. Cancer Res 71: 1–11. mous cell carcinoma. Am J Med Sci 343: 375–381. Dong X, Hudson NE, Lu C, Springer TA. 2014. Structural Constam DB, Philipp J, Malipiero UV, ten Dijke P, Schach- determinants of integrin b-subunit specificity for latent ner M, Fontana A. 1992. Differential expression of trans- TGF-b. Nat Struct Mol Biol 21: 1091–1096. forming growth factor-b1, -b2, and -b3 by glioblastoma cells, astrocytes, and microglia. J Immunol 148: 1404– Dutta A, Li J, Lu H, Akech J, Pratap J, Wang T, Zerlanko BJ, 1410. FitzGerald TJ, Jiang Z, Birbe R, et al. 2014. Integrin avb6 promotes an osteolytic program in cancer cells by upre- Corcoran RB, Dias-Santagata D, Bergethon K, Lafrate AJ, gulating MMP2. Cancer Res 74: 1598–1608. Settleman J, Engelman JA. 2010. BRAF gene amplifica- tion can promote acquired resistance to MEK inhibitors Ehata S, Johansson E, Katayama R, Koike S, Watanabe A, Hoshino Y,Katsuno Y,Komuro A, Koinuma D, Kano MR, in cancer cells harboring the BRAF V600E mutation. Sci et al. 2011. Transforming growth factor-b decreases the Signal 3: ra84. cancer-initiating cell population within diffuse-type gas- Cuende J, Lie´nart S, Dedobbeleer O, van der Woning B, De tric carcinoma cells. Oncogene 30: 1693–1705. Boeck G, Stockis J, Huygens C, Colau D, Somja J, Del- Engel ME, Mcdonnell MA, Law BK, Moses HL. 1999. Inter- venne P, et al. 2015. Monoclonal antibodies against dependent SMAD and JNK signaling in transforming GARP/TGF-b1 complexes inhibit the immunosuppres- growth factor-b-mediated transcription. J Biol Chem sive activity of human regulatory T cells in vivo. Sci Transl 274: 37413–37420. Med 7: 284ra256. Engelman JA, Settleman J. 2008. Acquired resistance to ty- Cui W, Fowlis DJ, Cousins FM, Duffie E, Bryson S, Balmain rosine kinase inhibitors during cancer therapy. Curr Opin A, Akhurst RJ. 1995. Concerted action of TGF-b1 and its Genet Dev 18: 73–79. type II receptor in control of epidermal homeostasis in Engelman JA, Zejnullahu K, Mitsudomi T, Song Y, Hyland transgenic mice. Genes Dev 9: 945–955. C, Park JO, Lindeman N, Gale CM, Zhao X, Christensen Dang CV, O’Donnell KA, Juopperi T. 2005. The great MYC J, et al. 2007. MET amplification leads to gefitinib resis- escape in tumorigenesis. Cancer Cell 8: 177–178. tance in lung cancer by activating ERBB3 signaling. Sci- Davies M, Robinson M, Smith E, Huntley S, Prime S, Pat- ence 316: 1039–1043. erson I. 2005. Induction of an epithelial to mesenchymal Flanders KC, Wakefield LM. 2009. Transforming growth transition in human immortal and malignant keratino- factor-bs and mammary gland involution; Functional cytes by TGF-b1 involves MAPK, Smad and AP-1 signal- roles and implications for cancer progression. J Mamma- ling pathways. J Cell Biochem 95: 918–931. ry Gland Biol Neoplasia 14: 131–144. den Hollander MW,Bensch F,Glaudemans AW,Oude Mun- Flavell RA, Sanjabi S, Wrzesinski SH, Licona-Limon P.2010. nink TH, Enting RH, den Dunnen WF, Heesters MA, The polarization of immune cells in the tumour environ- Kruyt FA, Lub-de Hooge MN, Cees de Groot J, et al. ment by TGF-b. Nat Rev Immunol 10: 554–567. 2015. TGF-b antibody uptake in recurrent high-grade Fowlis DJ, Flanders KC, Duffie E, Balmain A, Akhurst RJ. glioma imaged with 89Zr-Fresolimumab PET. JNuc 1992. Discordant transforming growth factor-b1RNA Med 56: 1310–1314. and protein localization during chemical carcinogenesis Derynck R, Akhurst RJ. 2007. Differentiation plasticity reg- of the skin. Cell Growth Diff 3: 81–91. ulated by TGF-b family proteins in development and Fransvea E, Angelotti U, Antonaci S, Giannelli G. 2008. disease. Nat Cell Biol 9: 1000–1004. Blocking transforming growth factor-b up-regulates E-

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 23 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

cadherin and reduces migration and invasion of hepato- Gorsch SM, Memoli VA, Stukel TA, Gold LI, Arrick BA. cellular carcinoma cells. Hepatology 47: 1557–1566. 1992. Immunohistochemical staining for transforming Frazier K, Thomas R, Scicchitano M, Mirabile R, Boyce R, growth factor b1 associates with disease progression in Zimmerman D, Grygielko E, Nold J, DeGouville AC, human breast cancer. Cancer Res 52: 6949–6952. Huet S, et al. 2007. Inhibition of ALK5 signaling induces Gridley DS, Sura SS, Uhm JR, Lin CH, Kettering JD. 1993. physeal dysplasia in rats. Toxicol Pathol 35: 284–295. Effects of anti-transforming growth factor-b antibody Fridlender ZG, Sun J, Kim S, Kapoor V, Cheng G, Ling L, and interleukin-2 in tumor-bearing mice. Cancer Biother Worthen GS, Albelda SM. 2009. Polarization of tumor- 8: 159–170. associated neutrophil phenotype by TGF-b: “N1” versus Gro¨nroos E, Kingston IJ, Ramachandran A, Randall RA, “N2” TAN. Cancer Cell 16: 183–194. Vizan P, Hill CS. 2012. Transforming growth factor b Fujii M, Nakanishi H, Toyoda T, Tanaka I, Kondo Y, Osada inhibits bone morphogenetic protein-induced transcrip- H, Sekido Y. 2012a. Convergent signaling in the regula- tion through novel phosphorylated Smad1/5-Smad3 tion of connective tissue growth factor in malignant me- complexes. Mol Cell Biol 32: 2904–2916. sothelioma: TGF-b signaling and defects in the Hippo Gudey SK, Sundar R, Mu Y, Wallenius A, Zang G, Bergh A, signaling cascade. Cell Cycle 11: 3373–3379. Heldin CH, Landstro¨m M. 2014. TRAF6 stimulates the b Fujii M, Toyoda T,Nakanishi H, Yatabe Y,Sato A, Matsudaira tumor-promoting effects of TGF type I receptor Y, Ito H, Murakami H, Kondo Y, Kondo E, et al. 2012b. through polyubiquitination and activation of presenilin 1. Sci Signal 7: ra2. TGF-b synergizes with defects in the Hippo pathway to stimulate human malignant mesothelioma growth. J Exp Gueorguieva I, Cleverly AL, Stauber A, Sada Pillay N, Rodon Med 209: 479–494. JA, Miles CP, Yingling JM, Lahn MM. 2014. Defining a therapeutic window for the novel TGF-b inhibitor Fujiwara Y,Nokihara H, YamadaY,YamamotoN, Sunami K, LY2157299 monohydrate based on a pharmacokinetic/ Utsumi H, Asou H, TakahashI O, Ogasawara K, Gueor- pharmacodynamic model. Br J Clin Pharmacol 77: 796– guieva I, et al. 2015. Phase 1 study of galunisertib, a TGF- 807. b receptor I kinase inhibitor, in Japanese patients with advanced solid tumors. Cancer Chemo Pharm 76: 1143– Hannigan A, Smith P, Kalna G, Lo Nigro C, Orange C, 1152. O’Brien DI, Shah R, Syed N, Spender LC, Herrera B, et al. 2010. Epigenetic downregulation of human disabled Garrison K, Hahn T, Lee WC, Ling LE, Weinberg AD, Ak- homolog 2 switches TGF-b from a tumor suppressor to a b poriaye ET. 2012. The small molecule TGF- signaling tumor promoter. J Clin Invest 120: 2842–2857. inhibitor SM16 synergizes with agonistic OX40 antibody to suppress established mammary tumors and reduce Harradine KA, Ridd K, Saunier EF, Clermont FF, Perez-Lo- spontaneous metastasis. Cancer Immunol Immunother sada J, Moore DH, Epstein EH Jr, Bastian BC, Akhurst RJ. 61: 511–521. 2009. Elevated cutaneous Smad activation associates with enhanced skin tumor susceptibility in organ transplant Giaccone G, Bazhenova LA, Nemunaitis J, Tan M, Juha´sz E, recipients. Clin Cancer Res 15: 5101–5107. Ramlau R, van den Heuvel MM, Lal R, Kloecker GH, Hatsell SJ, Idone V,Wolken DM, Huang L, Kim HJ, Wang L, Eaton KD, et al. 2015. A phase III study of belagenpuma- Wen X, Nannuru KC, Jimenez J, Xie L, et al. 2015. tucel-L, an allogeneic tumour cell vaccine, as mainte- ACVR1R206H receptor mutation causes fibrodysplasia nance therapy for non-small cell lung cancer. Eur J Cancer ossificans progressiva by imparting responsiveness to ac- Res 51: 2321–2329. tivin A. Sci Transl Med 7: 303ra137. Giampieri S, Pinner S, Sahai E. 2010. Intravital imaging Hau P, Jachimczak P, Bogdahn U. 2009. Treatment of ma- illuminates transforming growth factor b signaling lignant gliomas with TGF-b2 antisense oligonucleotides. switches during metastasis. Cancer Res 70: 3435–3439. Expert Rev Anticancer Ther 9: 1663–1674. Giehl K, Imamichi Y, Menke A. 2007. Smad4-independent Heldin CH, Rubin K, Pietras K, O¨ stman A. 2004. High TGF-b signaling in tumor cell migration. Cells Tissues interstitial fluid pressure—An obstacle in cancer therapy. Organs 185: 123–130. Nat Rev Cancer 4: 806–813. Gill S, June CH. 2015. Going viral: Chimeric antigen recep- Heldin CH, Landstro¨m M, Moustakas A. 2009. Mechanism tor T-cell therapy for hematological malignancies. Im- of TGF-b signaling to growth arrest, apoptosis, and ep- munol Rev 263: 68–89. ithelial–mesenchymal transition. Curr Opin Cell Biol 21: Gomis RR, Alarcon C, He W, Wang Q, Seoane J, Lash A, 166–176. Massague´ J. 2006a. A FoxO-Smad synexpression group in Henderson NC, Arnold TD, Katamura Y, Giacomini MM, human keratinocytes. Proc Natl Acad Sci 103: 12747– Rodriguez JD, McCarty JH, Pellicoro A, Raschperger E, 12752. Betsholtz C, Ruminski PG, et al. 2013. Targeting of av Gomis RR, Alarcon C, Nadal C, Van Poznak C, Massague´ J. integrin identifies a core molecular pathway that regu- 2006b. C/EBP b at the core of the TGFb cytostatic re- lates fibrosis in several organs. Nat Med 19: 1617–1624. sponse and its evasion in metastatic breast cancer cells. Hocevar BA, Brown TL, Howe PH. 1999. TGF-b induces Cancer Cell 10: 203–214. fibronectin synthesis through a c-Jun N-terminal kinase- Gorelik L, Flavell RA. 2001. Immune-mediated eradication dependent, Smad4-independent pathway. EMBO J 18: of tumors through the blockade of transforming growth 1345–1356. factor-b signaling in T cells. Nat Med 7: 1118–1122. Hocevar BA, Prunier C, Howe PH. 2005. Disabled-2 (Dab2) Gorelik L, Flavell RA. 2002. Transforming growth factor-b in mediates transforming growth factor b (TGF-b)-stimu- T-cell biology. Nat Rev Immunol 2: 46–53. lated fibronectin synthesis through TGF-b-activated ki-

24 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

nase 1 and activation of the JNK pathway. J Biol Chem stemness: A tunable “stemness window” model. Oncotar- 280: 25920–25927. get 6: 25161–25174. Hodi FS, O’Day SJ, McDermott DF, Weber RW, Sosman JA, Kalluri R, WeinbergRA. 2009. The basics of epithelial–mes- Haanen JB, Gonzalez R, Robert C, Schadendorf D, Hassel enchymal transition. J Clin Invest 119: 1420–1428. JC, et al. 2010. Improved survival with ipilimumab in Kang Y, Siegel PM, Shu W, Drobnjak M, Kakonen SM, Cor- patients with metastatic melanoma. N Eng J Med 363: do´n-Cardo C, Guise TA, Massague´ J. 2003. A multigenic 711–723. program mediating breast cancer metastasis to bone. Hoos A, Eggermont AM, Janetzki S, Hodi FS, Ibrahim R, Cancer Cell 3: 537–549. Anderson A, Humphrey R, Blumenstein B, Old L, Wol- Kang Y, He W, Tulley S, Gupta GP, Serganova I, Chen CR, chok J. 2010. Improved endpoints for cancer immuno- Manova-Todorova K, Blasberg R, Gerald WL, Massague´ therapy trials. J Natl Can Inst 102: 1388–1397. J. 2005. Breast cancer bone metastasis mediated by the Hoot KE, Lighthall J, Han G, Lu SL, Li A, Ju W, Kulesz- Smad tumor suppressor pathway. Proc Natl Acad Sci 102: Martin M, Bottinger E, Wang XJ. 2008. Keratinocyte- 13909–13914. specific Smad2 ablation results in increased epithelial- Kano MR, Bae Y, Iwata C, Morishita Y, Yashiro M, Oka M, mesenchymal transition during skin cancer formation Fujii T, Komuro A, Kiyono K, Kaminishi M, et al. 2007. and progression. J Clin Invest 118: 2722–2732. Improvement of cancer-targeting therapy, using nano- Huang S, Ho¨lzel M, Knijnenburg T,Schlicker A, Roepman P, carriers for intractable solid tumors by inhibition of McDermott U, Garnett M, Grernrum W, Sun C, Prahal- TGF-b signaling. Proc Natl Acad Sci 104: 3460–3465. lad A, et al. 2012. MED12 controls the response to mul- Kasid A, Bell GI, Director EP. 1988. Effects of transforming tiple cancer drugs through regulation of TGF-b receptor growth factor-b on human lymphokine-activated killer signaling. Cell 151: 937–950. cell precursors. Autocrine inhibition of cellular prolifer- Iacobuzio-Donahue CA, Fu B, Yachida S, Luo M, Abe H, ation and differentiation to immune killer cells. J Immu- Henderson CM, Vilardell F, Wang Z, Keller JW, Banerjee nol 141: 690–698. P,et al. 2009. DPC4 gene status of the primary carcinoma Kawasaki K, Freimuth J, Meyer DS, Lee MM, Tochimoto- correlates with patterns of failure in patients with pan- Okamoto A, Benzinou M, Clermont FF, Wu G, Roy R, creatic cancer. J Clin Oncol 27: 1806–1813. Letteboer TG, et al. 2014. Genetic variants of Adam17 Igarashi J, Fukuda N, Inoue T, Nakai S, Saito K, Fujiwara K, differentially regulate TGF-b signaling to modify vascu- Matsuda H, Ueno T, Matsumoto Y, Watanabe T, et al. lar pathology in mice and humans. Proc Natl Acad Sci 2015. Preclinical study of novel gene silencer pyrrole-im- 111: 7723–7728. idazole polyamide targeting human TGF-b1 promoter Kim SJ, Denhez F, Kim KY,Holt JT, Sporn MB, Roberts AB. for hypertrophic scars in a common marmoset primate 1989. Activation of the second promoter of the trans- model. PLoS ONE 10: e0125295. forming growth factor-b1 gene by transforming growth James D, Nam HS, Seandel M, Nolan D, Janovitz T, Tomi- factor-b1 and phorbol ester occurs through the same shima M, Studer L, Lee G, Lyden D, Benezra R, et al. 2010. target sequences. J Biol Chem 264: 19373–19378. Expansion and maintenance of human embryonic stem Kim S, Buchlis G, Fridlender ZG, Sun J, Kapoor V,Cheng G, cell-derived endothelial cells by TGFb inhibition is Id1- Haas A, Cheung HK, Zhang X, Corbley M, et al. 2008. dependent. Nat Biotech 28: 161–166. Systemic blockade of transforming growth factor-b sig- Jia ZC, Zou LY, Ni B, Wan Y, Zhou W, Lv YB, Geng M, Wu naling augments the efficacy of immunogene therapy. YZ. 2005. Effective induction of antitumor immunity by Cancer Res 68: 10247–10256. immunization with plasmid DNA encoding TRP-2 plus Kondo M, Cubillo E, Tobiume K, Shirakihara T, Fukuda N, neutralization of TGF-b. Cancer Immunol Immunother Suzuki H, Shimizu K, Takehara K, Cano A, Saitoh M, et 54: 446–452. al. 2004. A role for Id in the regulation of TGF-b-induced Jin CH, Krishnaiah M, Sreenu D. 2014. Discovery of N- epithelial-mesenchymal transdifferentiation. Cell Death ((4-([1,2,4]triazolo[1,5-a]pyridin-6-yl)-5-(6-methylpyri- Differ 11: 1092–1101. din-2-yl)-1H-imidazol-2 -yl)methyl)-2-fluoroaniline (EW- Kong FM, Anscher MS, Murase T, Abbott BD, Iglehart JD, 7197): A highly potent, selective, and orally bioavailable Jirtle RL. 1995. Elevated plasma transforming growth inhibitor of TGF-b type I receptor kinase as cancer im- factor-b1 levels in breast cancer patients decrease after munotherapeutic/antifibrotic agent. J Med Chem 57: surgical removal of the tumor. Ann Surg 222: 155–162. 4213–4238. Koo E, Henderson MA, Dwyer M, Skandarajah AR. 2015. Jobling MF, Mott JD, Finnegan MT, Jurukovski V, Erickson Management and prevention of breast cancer after radi- AC, Walian PJ, Taylor SE, Ledbetter S, Lawrence CM, ation to the chest for childhood, adolescent, and young Rifkin DB, et al. 2006. Isoform-specific activation of la- adulthood malignancy. Ann Surg Oncol 22: 545–551. tent transforming growth factor b (LTGF-b) by reactive Kovacs RJ, Maldonado G, Azaro A, Ferna´ndez MS, Romero oxygen species. Radiat Res 166: 839–848. FL, Sepulveda-Sa´nchez JM, Corretti M, Carducci M, Do- Johannessen CM, Boehm JS, Kim SY,Thomas SR, Wardwell lan M, Gueorguieva I, et al. 2015. Cardiac safety of TGF-b L, Johnson LA, Emery CM, Stransky N, Cogdill AP,Bar- receptor I kinase inhibitor LY2157299 monohydrate in retina J, et al. 2010. COT drives resistance to RAF inhibi- cancer patients in a first-in-human dose study. Cardiovas tion through MAP kinase pathway reactivation. Nature Toxicol 15: 309–323. 468: 968–972. Kozlow W,Guise TA.2005. Breast cancer metastasis to bone: Jolly MK, Jia D, Boareto M, Mani SA, Pienta KJ, Ben-Jacob Mechanisms of osteolysis and implications for therapy. J E, Levine H. 2015. Coupling the modules of EMT and Mammary Gland Biol Neoplasia 10: 169–180.

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 25 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

Krasagakis K, Tholke D, Farthmann B, Eberle J, Mansmann Li MO, Flavell RA. 2008. TGF-b: A master of all T cell trades. U, Orfanos CE. 1998. Elevated plasma levels of trans- Cell 134: 392–404. forming growth factor (TGF)-b1 and TGF-b2 in patients Li J, Foitzik K, Calautti E, Baden H, Doetschman T, Dotto with disseminated malignant melanoma. Br J Cancer 77: GP. 1999. TGF-b3, but not TGF-b1, protects keratino- 1492–1494. cytes against 12-O-tetradecanoylphorbol-13-acetate-in- Kretzschmar M, Doody J, Timokhina I, Massague´ J. 1999. A duced cell death in vitro and in vivo. J Biol Chem 274: mechanism of repression of TGF-b/Smad signaling by 4213–4219. oncogenic Ras. Genes Dev. 13: 804–816. Li X, Lewis MT,Huang J, Gutierrez C, Osborne CK, Wu MF, Kulkarni AB, Huh CG, Becker D, Geiser A, Lyght M, Flan- Hilsenbeck SG, Pavlick A, Zhang X, Chamness GC, et al. ders KC, Roberts AB, Sporn MB, Ward JM, Karlsson S. 2008. Intrinsic resistance of tumorigenic breast cancer 1993. Transforming growth factor b1 null mutation in cells to chemotherapy. J Natl Cancer Inst 100: 672–679. mice causes excessive inflammatory response and early Lipson EJ, Sharfman WH, Drake CG, Wollner I, Taube JM, death. Proc Natl Acad Sci 90: 770–774. Anders RA, Xu H, Yao S, Pons A, Chen L, et al. 2013. Kuppner MC, Hamou MF,Bodmer S, Fontana A, De Tribo- Durable cancer regression off-treatment and effective re- let N. 1988. The glioblastoma-derived T-cell suppressor induction therapy with an anti-PD-1 antibody. Clin Can- factor/transforming growth factor b2 inhibits the gener- cer Res 19: 462–468. ation of lymphokine-activated killer (LAK) cells. Int J Lucke CD, Philpott A, Metcalfe JC, Thompson AM, Cancer 42: 562–567. Hughes-Davies L, Kemp PR, Hesketh R. 2001. Inhibiting Lacher MD, Tiirikainen MI, Saunier EF,Christian C, Anders mutations in the transforming growth factor b type 2 M, Oft M, Balmain A, Akhurst RJ, Korn WM. 2006. receptor in recurrent human breast cancer. Cancer Res Transforming growth factor-b receptor inhibition en- 61: 482–485. hances adenoviral infectability of carcinoma cells via Maker AV, Yang JC, Sherry RM, Topalian SL, Kammula US, up-regulation of coxsackie and adenovirus receptor in Royal RE, Hughes M, Yellin MJ, Haworth LR, Levy C, et conjunction with reversal of epithelial–mesenchymal al. 2006. Intrapatient dose escalation of anti-CTLA-4 an- transition. Cancer Res 66: 1648–1657. tibody in patients with metastatic melanoma. J Immun- Lacouture ME, O’Reilly K, Rosen N, Solit DB. 2012. Induc- other 29: 455–463. tion of cutaneous squamous cell carcinomas by RAF in- Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, hibitors: Cause for concern? J Clin Onc 30: 329–330. Brooks M, Reinhard F, Zhang CC, Shipitsin M, et al. Lacouture ME, Morris JC, Lawrence DP, Tan AR, Olencki 2008. The epithelial–mesenchymal transition generates TE, Shapiro GI, Dezube BJ, Berzofsky JA, Hsu FJ, Guitart cells with properties of stem cells. Cell 133: 704–715. J. 2015. Cutaneous keratoacanthomas/squamous cell Mantovani A, Sozzani S, Locati M, Allavena P,Sica A. 2002. carcinomas associated with neutralization of transform- Macrophage polarization: Tumor-associated macro- ing growth factor b by the monoclonal antibody fresoli- phages as a paradigm for polarized M2 mononuclear mumab (GC1008). Cancer Immunol Immunother 64: phagocytes. Trends Immunol 23: 549–555. 437–446. Margadant C, Sonnenberg A. 2010. Integrin-TGF-b cross- Lammerts E, Roswall P,Sundberg C, Gotwals PJ, Koteliansky talk in fibrosis, cancer and wound healing. EMBO Rep 11: VE, Reed RK, Heldin NE, Rubin K. 2002. Interference 97–105. with TGF-b1 and -b3 in tumor stroma lowers tumor Matsuda H, Fukuda N, Ueno T, Katakawa M, Wang X, Wa- interstitial fluid pressure independently of growth in ex- tanabe T, Matsui S, Aoyama T, Saito K, Bando T, et al. perimental carcinoma. Int J Cancer 102: 453–462. 2011. Transcriptional inhibition of progressive renal dis- Lamouille S, Derynck R. 2009. Oncogene and tumour sup- ease by gene silencing pyrrole-imidazole polyamide tar- pressor: The two faces of SnoN. EMBO J 28: 3459–3460. geting of the transforming growth factor-b1 promoter. Laouar Y, Sutterwala FS, Gorelik L, Flavell RA. 2005. Trans- Kidney Int 79: 46–56. forming growth factor-b controls T helper type 1 cell Mi LZ, Brown CT,Gao Y,Tian Y,Le VQ, WalzT,Springer TA. development through regulation of natural killer cell in- 2015. Structure of bone morphogenetic protein 9 pro- terferon-gamma. Nat Immunol 6: 600–607. complex. Proc Natl Acad Sci 112: 3710–15. Laping NJ, Everitt JI, Frazier KS, Burgert M, Portis MJ, Miyazono K. 2012. Tumour promoting functions of TGF-b Cadacio C, Gold LI, Walker CL. 2007. Tumor-specific in CML-initiating cells. J Biochem 152: 383–385. efficacy of transforming growth factor-b RI inhibition Mohammad KS, Javelaud D, Fournier PG, Niewolna M, in Eker rats. Clin Cancer Res 13: 3087–3099. McKenna CR, Peng XH, Duong V, Dunn LK, Mauviel Laverty HG, Wakefield LM, Occleston NL, O’Kane S, Fergu- A, Guise TA. 2011. TGF-b-RI kinase inhibitor SD-208 son MW. 2009. TGF-b3 and cancer: A review. Cytokine reduces the development and progression of melanoma Growth Factor Rev 20: 305–317. bone metastases. Cancer Res 71: 175–184. Leitlein J, Aulwurm S, Waltereit R, Naumann U, Wagen- Morris JC, Shapiro GI, Tan AR, Lawrence DP, Olencki TE, knecht B, Garten W, Weller M, Platten M. 2001. Process- Dezube BJ, et al., eds. 2008. Phase I/II study of GC1008: ing of immunosuppressive pro-TGF-b1,2 by human A human anti-transforming growth factor-beta (TGFb) glioblastoma cells involves cytoplasmic and secreted monoclonal antibody in patients with advanced malig- furin-like proteases. J Immunol 166: 7238–7243. nant melanoma (MM) or renal cell carcinoma (RCC). Levental KR, Yu H, Kass L, Lakins JN, Egeblad M, Erler JT, ASCO Annual Meeting. Fong SF, Csiszar K, Giaccia A, Weninger W, et al. 2009. Morris JC, Tan AR, Olencki TE, Shapiro GI, Dezube BJ, Matrix crosslinking forces tumor progression by enhanc- Reiss M, Hsu FJ, Berzofsky JA, Lawrence DP.2014. Phase ing integrin signaling. Cell 139: 891–906. I study of GC1008 (fresolimumab): A human anti-trans-

26 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

forming growth factor-b (TGFb) monoclonal antibody Papageorgis P,Lambert AW,Ozturk S, Gao F, Pan H, Manne in patients with advanced malignant melanoma or renal U, Alekseyev YO, Thiagalingam A, Abdolmaleky HM, cell carcinoma. PLoS ONE 9: e90353. Lenburg M, et al. 2010. Smad signaling is required to Munger JS, Harpel JG, Gleizes PE, Mazzieri R, Nunes I, maintain epigenetic silencing during breast cancer pro- Rifkin DB. 1997. Latent transforming growth factor-b: gression. Cancer Res 70: 968–978. Structural features and mechanisms of activation. Kidney Park J, Wrzesinski SH, Stern E, Look M, Criscione J, Ragheb Int 51: 1376–1382. R, Jay SM, Demento SL, Agawu A, Licona Limon P,et al. Munger JS, Huang X, Kawakatsu H, Griffiths MJ, Dalton SL, 2012. Combination delivery of TGF-b inhibitor and IL-2 Wu J, Pittet JF, Kaminski N, Garat C, Matthay MA, et al. by nanoscale liposomal polymeric gels enhances tumour 1999. The integrin avb6 binds and activates latent TGF- immunotherapy. Nat Mat 11: 895–905. b1: A mechanism for regulating pulmonary inflamma- Parsons R, Myeroff LL, Liu B, Willson JK, Markowitz SD, tion and fibrosis. Cell 96: 319–328. Kinzler KW, Vogelstein B. 1995. Microsatellite instability Naka K, Hoshii T, Muraguchi T, Tadokoro Y, Ooshio T, and mutations of the transforming growth factor-b type Kondo Y, Nakao S, Motoyama N, Hirao A. 2010. TGF- II receptor gene in colorectal cancer. Cancer Res 55: b-FOXO signalling maintains leukaemia-initiating cells 5548–5550. in chronic myeloid leukaemia. Nature 463: 676–680. Peterson AJ, O’Connor MB. 2013. inhibi- Naka K, Ishihara K, Jomen Y, Jin CH, Kim DH, Gu YK, tion by Smad2 regulates Drosophila wing disc patterning Jeong ES, Li S, Krause DS, Kim DW, et al. 2016. Novel through BMP-response elements. Development 140: oral transforming growth factor-b signaling inhibitor 649–659. EW-7197 eradicates CML-initiating cells. Cancer Sci Piek E, Ju WJ, Heyer J, Escalante-Alcalde D, Stewart CL, 107: 140–148. Weinstein M, Deng C, Kucherlapati R, Bottinger EP,Rob- Nazarian R, Shi H, Wang Q, Kong X, Koya RC, Lee H, Chen erts AB. 2001. Functional characterization of transform- b Z, Lee MK, Attar N, Sazegar H, et al. 2010. ing growth factor signaling in Smad2- and Smad3-de- ficient fibroblasts. J Biol Chem 276: 19945–19953. acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature 468: 973–977. Portella G, Cumming SA, Liddell J, Cui W, Ireland H, Akhurst RJ, Balmain A. 1998. Transforming growth fac- Nemunaitis J, Murray N. 2006. Immune-modulating vac- tor b is essential for spindle cell conversion of mouse skin cines in non-small cell lung cancer. J Thoracic Oncol 1: carcinoma in vivo: Implications for tumor invasion. Cell 756–761. Growth Diff 9: 393–404 Nemunaitis J, Dillman RO, Schwarzenberger PO, Senzer N, Rabbani ZN, Anscher MS, Zhang X, Chen L, Samulski TV,Li Cunningham C, Cutler J, Tong A, Kumar P, Pappen B, CY, Vujaskovic Z. 2003. Soluble TGF-b type II receptor Hamilton C, et al. 2006. Phase II study of belagenpuma- gene therapy ameliorates acute radiation-induced pul- tucel-L, a transforming growth factor b-2 antisense gene- monary injury in rats. Int J Radiat Oncol Biol Phys 57: modified allogeneic tumor cell vaccine in non-small-cell 563–572. lung cancer. J Clin Oncol 24: 4721–4730. Ramesh S, Wildey GM, Howe PH. 2009. Transforming Nemunaitis J, Nemunaitis M, Senzer N, Snitz P, Bedell C, growth factor b (TGF-b)-induced apoptosis: The rise Kumar P, Pappen B, Maples PB, Shawler D, Fakhrai H. and fall of Bim. Cell Cycle 8: 11–17. 2009. Phase II trial of belagenpumatucel-L, a TGF-b2 antisense gene modified allogeneic tumor vaccine in ad- Ray BN, Lee NY, How T, Blobe GC. 2010. ALK5 phosphor- vanced non small cell lung cancer (NSCLC) patients. ylation of the cytoplasmic domain regulates Cancer Gene Ther 16: 620–624. Smad1/5/8 signaling and endothelial cell migration. Carcinogenesis 31: 435–441. Ng MR, Brugge JS. 2009. A stiff blow from the stroma: Collagen crosslinking drives tumor progression. Cancer Rice LM, Padilla CM, McLaughlin SR, Mathes A, Ziemek J, Cell 16: 455–457. Goummih S, Nakerakanti S, York M, Farina G, Whitfield ML, et al. 2015. Fresolimumab treatment decreases bio- Oft M, Akhurst RJ, Balmain A. 2002. Metastasis is driven by markers and improves clinical symptoms in systemic sequential elevation of H-ras and Smad2 levels. Nat Cell sclerosis patients. J Clin Invest 125: 2795–2807. Biol 4: 487–494. Roberts AB, Wakefield LM. 2003. The two faces of trans- Ogino S, Kawasaki T, Ogawa A, Kirkner GJ, Loda M, Fuchs forming growth factor b in carcinogenesis. Proc Natl CS. 2007. TGFBR2 mutation is correlated with CpG is- Acad Sci 100: 8621–8623. land methylator phenotype in microsatellite instability- Robinson PN, Arteaga-Solis E, Baldock C, Collod-Be´roud high colorectal cancer. Human Pathol 38: 614–620. G, Booms P,De Paepe A, Dietz HC, Guo G, Handford PA, Oshimori N, Oristian D, Fuchs E. 2015. TGF-b promotes Judge DP, et al. 2006. The molecular genetics of Marfan heterogeneity and drug resistance in squamous cell car- syndrome and related disorders. J Med Genet 43: 769– cinoma. Cell 160: 963–976. 787. Ota T, Fujii M, Sugizaki T, Ishii M, Miyazawa K, Aburatani Rodo´n J, Carducci M, Sepulveda-Sa´nchez JM, Azaro A, H, Miyazono K. 2002. Targets of transcriptional regula- Calvo E, Seoane J, Bran˜a I, Sicart E, Gueorguieva I, Clev- tion by two distinct type I receptors for transforming erly A, et al. 2015a. Pharmacokinetic, pharmacodynamic growth factor-b in human umbilical vein endothelial and biomarker evaluation of transforming growth factor- cells. J Cell Physiol 193: 299–318. b receptor I kinase inhibitor, galunisertib, in phase 1 Padua D, Massague´ J. 2009. Roles of TGF-b in metastasis. study in patients with advanced cancer. Invest New Drugs Cell Res 19: 89–102. 33: 357–370.

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 27 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

Rodo´n J, Carducci MA, Sepulveda-Sa´nchez JM, Azaro A, Shi Y, Massague´ J. 2003. Mechanisms of TGF-b signaling Calvo E, Seoane J, Bran˜a I, Sicart E, Gueorguieva I, Clev- from cell membrane to the nucleus. Cell 113: 685–700. erly AL, et al. 2015b. First-in-human dose study of the Shi M, Zhu J, Wang R, Chen X, Mi L, Walz T, Springer TA. novel transforming growth factor-b receptor I kinase in- 2011. Latent TGF-b structure and activation. Nature 474: hibitor LY2157299 monohydrate in patients with ad- 343–349. vanced cancer and glioma. Clin Cancer Res 21: 553–560. Shim KS, Kim KH, Han WS, Park EB. 1999. Elevated serum Rubtsov YP,Rudensky AY.2007. TGF-b signalling in control levels of transforming growth factor-b1 in patients with of T-cell-mediated self-reactivity. Nat Rev Immunol 7: colorectal carcinoma: Its association with tumor progres- 443–453. sion and its significant decrease after curative surgical Salnikov AV,Roswall P,Sundberg C, Gardner H, Heldin NE, resection. Cancer 85: 554–561. Rubin K. 2005. Inhibition of TGF-b modulates macro- Shinto O, Yashiro M, Kawajiri H, Shimizu K, Shimizu T, phages and vessel maturation in parallel to a lowering of Miwa A, Hirakawa K. 2010. Combination effect of a interstitial fluid pressure in experimental carcinoma. Lab TGF-b receptor kinase inhibitor with 5-FU analog S1 Invest 85: 512–521. on lymph node metastasis of scirrhous gastric cancer in Salvo E, Garasa S, Dotor J, Morales X, Pelaez R, Altevogt P, mice. Cancer Sci 101: 1846–1852. Rouzaut A. 2014. Combined targeting of TGF-b1 and Shipitsin M, Campbell LL, Argani P,Weremowicz S, Bloush- integrin b3 impairs lymph node metastasis in a mouse tain-Qimron N, YaoJ, Nikolskaya T,Serebryiskaya T,Ber- model of non-small-cell lung cancer. Mol Cancer 13: 112. oukhim R, Hu M, et al. 2007. Molecular definition of Sato K, Kawasaki H, Nagayama H. 2000. TGF-b1 recipro- breast tumor heterogeneity. Cancer Cell 11: 259–273. cally controls chemotaxis of human peripheral blood Shull MM, Ormsby I, Kier AB, Pawlowski S, Diebold RJ, Yin monocyte-derived dendritic cells via chemokine recep- M, Allen R, Sidman C, Proetzel G, Calvin D, et al. 1992. tors. J Immunol 164: 2285–2295. Targeted disruption of the mouse transforming growth factor-b1 gene results in multifocal inflammatory dis- Sawyer JS, Anderson BD, Beight DW, Campbell RM, Jones ease. Nature 359: 693–699. ML, Herron DK, Lampe JW, McCowan JR, McMillen WT, Mort N, et al. 2003. Synthesis and activity of new Siegel PM, Shu W, Massague´ J. 2003. Mad upregulation and aryl- and heteroaryl-substituted pyrazole inhibitors of Id2 repression accompany transforming growth factor the transforming growth factor-b type I receptor kinase (TGF)-b -mediated epithelial cell growth suppression. J domain. J Med Chem 46: 3953–3956. Biol Chem 278: 35444–35450. Singh A, Settleman J. 2010. EMT,cancer stem cells and drug Schlingensiepen KH, Jaschinski F,Lang SA, Moser C, Geiss- resistance: An emerging axis of evil in the war on cancer. ler EK, Schlitt HJ, Kielmanowicz M, Schneider A. 2011. Oncogene 29: 4741–4751. Transforming growth factor-b2 gene silencing with tra- bedersen (AP 12009) in pancreatic cancer. Cancer Sci Son JY, Park SY, Kim SJ, Lee SJ, Park SA, Kim MJ, Kim SW, 102: 1193–1200. Kim DK, Nam JS, Sheen YY. 2014. EW-7197, a novel ALK-5 kinase inhibitor, potently inhibits breast to lung b Schmierer B, Hill CS. 2007. TGF- –SMAD signal transduc- metastasis. Mol Cancer Therap 13: 1704–1716. tion: Molecular specificity and functional flexibility. Nat Rev Mol Cell Biol 8: 970–982. Sorrentino A, Thakur N, Grimsby S, Marcusson A, von Bulow V, Schuster N, Zhang S, Heldin CH, Landstro¨m Schultz-Cherry S, Ribeiro S, Gentry L, Murphy-Ullrich JE. M. 2008. The type I TGF-b receptor engages TRAF6 to 1994. Thrombospondin binds and activates the small and activate TAK1 in a receptor kinase-independent manner. large forms of latent transforming growth factor-b in a Nat Cell Biol 10: 1199–1207. chemically defined system. J Biol Chem 269: 26775– Spain L, Diem S, Larkin J. 2016. Management of toxicities of 26782. immune checkpoint inhibitors. Cancer Treat Rev 44: 51– Sepulveda-Sa´nchez J, Ramos A, Hilario A, DE Velasco G, 60. Castellano D, Garcia DE LATorre M, Rodon J, Lahn MF. Stampfer MR, Yaswen P,Alhadeff M, Hosoda J. 1993. TGF b 2015. Brain perfusion and permeability in patients with induction of extracellular matrix associated proteins in advanced, refractory glioblastoma treated with lomustine normal and transformed human mammary epithelial and the transforming growth factor-b receptor I kinase cells in culture is independent of growth effects. J Cell inhibitor LY2157299 monohydrate. Oncol Lett 9: 2442– Physiol 155: 210–221. 2448. Stockis J, Colau D, Coulie PG, Lucas S. 2009. Membrane Sequist LV, Martins RG, Spigel D, Grunberg SM, Spira A, protein GARP is a receptor for latent TGF-b on the sur- Ja¨nne PA, Joshi VA,McCollum D, Evans TL, Muzikansky face of activated human Treg. Eur J Immunol 39: 3315– A, et al. 2008. First-line gefitinib in patients with ad- 3322. vanced non-small-cell lung cancer harboring somatic Sundar R, Gudey SK, Heldin CH, Landstro¨m M. 2015. EGFR mutations. J Clin Oncol 26: 2442–2449. TRAF6 promotes TGFb-induced invasion and cell-cycle Sergina NV,Rausch M, WangD, Blair J, Hann B, Shokat KM, regulation via Lys63-linked polyubiquitination of Lys178 Moasser MM. 2007. Escape from HER-family tyrosine in TGFb type I receptor. Cell Cycle 14: 554–565. kinase inhibitor therapy by the kinase-inactive HER3. Suzuki E, Kapoor V, Cheung HK, Ling LE, Delong PA, Nature 445: 437–441. Kaiser LR, Albelda SM. 2004. Soluble type II transform- Sharma SV, Lee DY, Li B, Quinlan MP, Takahashi F, Mahes- ing growth factor-b receptor inhibits established murine waran S, McDermott U, Azizian N, Zou L, Fischbach MA, malignant mesothelioma tumor growth by augmenting et al. 2010. A chromatin-mediated reversible drug-toler- host antitumor immunity. Clin Cancer Res 10: 5907– ant state in cancer cell subpopulations. Cell 141: 69–80. 5918.

28 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-b Signaling for Therapeutic Gain

Takeuchi K, Abe M, Hiasa M, Oda A, Amou H, Kido S, induced antitumor immunity. Cancer Res 75: 2232– Harada T, Tanaka O, Miki H, Nakamura S, et al. 2010. 2242. TGF-b inhibition restores terminal osteoblast differen- Vinogradova M, Gehling VS, Gustafson A, Arora S, Tindell tiation to suppress myeloma growth. PLoS ONE 5: CA, Wilson C, Williamson KE, Guler GD, Gangurde P, e9870. Manieri W, et al. 2016. An inhibitor of KDM5 demethy- Tampe D, Zeisberg M. 2014. Potential approaches to reverse lases reduces survival of drug-tolerant cancer cells. Nat or repair renal fibrosis. Nat Rev Nephrol 10: 226–237. Chem Biol 12: 531–538. Tang B, Yoo N, Vu M, Mamura M, Nam JS, Ooshima A, Du Voulgari A, Pintzas A. 2009. Epithelial–mesenchymal tran- Z, Desprez PY, Anver MR, Michalowska AM, et al. 2007. sition in cancer metastasis: Mechanisms, markers and Transforming growth factor-b can suppress tumorigen- strategies to overcome drug resistance in the clinic. Bio- esis through effects on the putative cancer stem or early chim Biophys Acta 1796: 75–90. progenitor cell and committed progeny in a breast cancer Walker RA, Dearing SJ. 1992. Transforming growth factor xenograft model. Cancer Res 67: 8643–8652. b1 in ductal carcinoma in situ and invasive carcinomas of TCGA Network T. 2015. Comprehensive genomic charac- the breast. Eur J Cancer 28: 641–644. terization of head and neck squamous cell carcinomas. Wallace A, Kapoor V, Sun J, Mrass P, Weninger W, Heitjan Nature 517: 576–582. DF,June C, Kaiser LR, Ling LE, Albelda SM. 2008. Trans- Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan forming growth factor-b receptor blockade augments the RS, Rubinstein L, Verweij J, Van Glabbeke M, van Oos- effectiveness of adoptive T-cell therapy of established sol- terom AT, Christian MC, et al. 2000. New guidelines to id cancers. Clin Cancer Res 14: 3966–3974. evaluate the response to treatment in solid tumors. Eu- Washio H, Fukuda N, Matsuda H, Nagase H, Watanabe T, ropean Organization for Research and Treatment of Can- Matsumoto Y,TeruiT.2011. Transcriptional inhibition of cer, National Cancer Institute of the United States, Na- hypertrophic scars by a gene silencer, pyrrole-imidazole tional Cancer Institute of Canada. J Natl Cancer Inst 92: polyamide, targeting the TGF-b1 promoter. J Invest 205–216. Derm 131: 1987–1995. Topalian SL, Drake CG, Pardoll DM. 2012. Targeting the Weigert C, Sauer U, Brodbeck K, Pfeiffer A, Haring HU, PD-1/B7-H1(PD-L1) pathway to activate anti-tumor Schleicher ED. 2000. AP-1 proteins mediate hyperglyce- immunity. Curr Opin Immunol 24: 207–212. mia-induced activation of the human TGF-b1 promoter Tran DQ, Andersson J, Wang R, Ramsey H, Unutmaz D, in mesangial cells. J Am Soc Nephrol 11: 2007–2016. Shevach EM. 2009. GARP (LRRC32) is essential for the Wick W,WellerM. 2011. Trabedersen to target transforming surface expression of latent TGF-b on platelets and acti- growth factor-b: When the journey is not the reward. vated FOXP3þ regulatory T cells. Proc Natl Acad Sci 106: Neuro Oncol 13: 559–560. 13445–13450. Wicks SJ, Lui S, Abdel-Wahab N, Mason RM, Chantry A. Triplett TA, Tucker CG, Triplett KC, Alderman Z, Sun L, 2000. Inactivation of Smad-transforming growth factor b Ling LE, Akporiaye ET, Weinberg AD. 2015. STAT3 sig- signaling by Ca2þ-calmodulin-dependent protein kinase naling is required for optimal regression of large estab- II. Mol Cell Biol 20: 8103–8111. lished tumors in mice treated with anti-OX40 and TGFb Wu CY, Roybal KT, Puchner EM, Onuffer J, Lim WA. receptor blockade. Cancer Immunol Res 3: 526–535. 2015. Remote control of therapeutic T cells through Tuma RS. 2008. New response criteria proposed for immu- a small molecule-gated chimeric receptor. Science 350: notherapies. J Natl Cancer Inst 100: 1280–1281. aab4077. Valle L, Serena-Acedo T, Liyanarachchi S, Hampel H, Xie W, Mertens JC, Reiss DJ, Rimm DL, Camp RL, Haffty Comeras I, Li Z, Zeng Q, Zhang HT,Pennison MJ, Sadim BG, Reiss M. 2002. Alterations of Smad signaling in hu- M, et al. 2008. Germline allele-specific expression of man breast carcinoma are associated with poor outcome: TGFBR1 confers an increased risk of colorectal cancer. A tissue microarray study. Cancer Res 62: 497–505. Science 321: 1361–1365. Yakymovych I, ten Dijke P, Heldin CH, Souchelnytskyi S. Vallieres L. 2009. Trabedersen, a TGFb2-specific antisense 2001. Regulation of Smad signaling by protein kinase C. oligonucleotide for the treatment of malignant gliomas FASEB J 15: 553–555. and other tumors overexpressing TGFb2. IDrugs 12: Yamaguchi Y, Tsumura H, Miwa M, Inaba K. 1997. Con- 445–453. trasting effects of TGF-b1 and TNF-a on the develop- Van Aarsen LA, Leone DR, Ho S, et al. 2008. Antibody- ment of dendritic cells from progenitors in mouse bone mediated blockade of integrin avb6 inhibits tumor pro- marrow. Stem Cells 15: 144–153. gression in vivo by a transforming growth factor-b-reg- Yamashita S, Takahashi S, McDonell N, Watanabe N, Niwa ulated mechanism. Cancer Res 68: 561–570. T, Hosoya K, Tsujino Y,Shirai T, Ushijima T. 2008. Meth- Van Der Jeught K, Joe PT, Bialkowski L, Heirman C, Dasz- ylation silencing of transforming growth factor-b recep- kiewicz L, Liechtenstein T, Escors D, Thielemans K, tor type II in rat prostate cancers. Cancer Res 68: 2112– Breckpot K. 2014. Intratumoral administration of 2121. mRNA encoding a fusokine consisting of IFN-b and Yang Z, Mu Z, Dabovic B, Jurukovski V,Yu D, Sung J, Xiong the ectodomain of the TGF-b receptor II potentiates an- X, Munger JS. 2007. Absence of integrin-mediated titumor immunity. Oncotarget 5: 10100–10113. TGFb1 activation in vivo recapitulates the phenotype Vanpouille-Box C, Diamond JM, Pilones KA, Zavadil J, of TGFb1-null mice. J Cell Biol 176: 787–793. Babb JS, Formenti SC, Barcellos-Hoff MH, Demaria S. Yao EH, Fukuda N, Ueno T, Matsuda H, Nagase H, Mat- 2015. TGFb is a master regulator of radiation therapy- sumoto Y, Sugiyama H, Matsumoto K. 2009. A pyrrole-

Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 29 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

R.J. Akhurst

imidazole polyamide targeting transforming growth Zhang M, Kleber S, Ro¨hrich M, Timke C, Han N, Tuetten- factor-b1 inhibits restenosis and preserves endothelial- berg J, Martin-Villalba A, Debus J, Peschke P,Wirkner U, ization in the injured artery. Cardiovasc Res 81: 797– et al. 2011. Blockade of TGF-b signaling by the TGFbR-I 804. kinase inhibitor LY2109761 enhances radiation response Yin JJ, Selander K, Chirgwin JM, Dallas M, Grubbs BG, and prolongs survival in glioblastoma. Cancer Res 71: Wieser R, Massague´ J, Mundy GR, Guise TA. 1999. 7155–7167. TGF-b signaling blockade inhibits PTHrP secretion by Zhong Z, Carroll KD, Policarpio D, Osborn C, Gregory M, breast cancer cells and bone metastases development. J Bassi R, Jimenez X, Prewett M, Liebisch G, Persaud K, et Clin Invest 103: 197–206. al. 2010. Anti-transforming growth factor b receptor II Yingling JM, Blanchard KL, Sawyer JS. 2004. Development antibody has therapeutic efficacy against primary tumor of TGF-b signalling inhibitors for cancer therapy. Nat growth and metastasis through multieffects on cancer, Rev Drug Discov 3: 1011–1022. stroma, and immune cells. Clin Cancer Res 16: 1191– Yoon JH, Jung SM, Park SH, Kato M, Yamashita T, Lee IK, 1205. Sudo K, Nakae S, Han JS, Kim OH, et al. 2013. Activin Zhou M, Bracci PM, Mccoy LS, Hsuang G, Wiemels JL, Rice receptor-like kinase 5 inhibition suppresses mouse mel- T, Zheng S, Kelsey KT, Wrensch MR, Wiencke JK. 2015. anoma by ubiquitin degradation of Smad4, thereby de- Serum macrophage-derived chemokine/CCL22 levels repressing eomesodermin in cytotoxic T lymphocytes. are associated with glioma risk, CD4 T cell lymphopenia EMBO Mol Med 5: 1720–1739. and survival time. Int J Cancer 137: 826–836. Yun CH, Mengwasser KE, Toms AV, Woo MS, Greulich H, Zhu X, Wang L, Zhang B, Li J, Dou X, Zhao RC. 2011. TGF- Wong KK, Meyerson M, Eck MJ. 2008. The T790M mu- b1-induced PI3K/Akt/NFkB/MMP9 signalling path- tation in EGFR kinase causes drug resistance by increas- way is activated in Philadelphia chromosome-positive ing the affinity for ATP. Proc Natl Acad Sci 105: 2070– chronic myeloid leukaemia hemangioblasts. J Biochem 2075. 149: 405–414. Zhang Z, Wang Y, Vikis HG, Johnson L, Liu G, Li J, Ander- Zuber P, Kuppner MC, De Tribolet N. 1988. Transforming son MW, Sills RC, Hong HL, Devereux TR, et al. 2001. growth factor- b2 down-regulates HLA-DR antigen ex- Wildtype Kras2 can inhibit lung carcinogenesis in mice. pression on human malignant glioma cells. Eur J Immu- Nat Genet 29: 25–33. nol 18: 1623–1626.

30 Advanced Online Article. Cite this article as Cold Spring Harb Perspect Biol doi: 10.1101/cshperspect.a022301 Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

Targeting TGF-β Signaling for Therapeutic Gain

Rosemary J. Akhurst

Cold Spring Harb Perspect Biol published online February 28, 2017

Subject Collection The Biology of the TGF-β Family

TGF-β Family Signaling in Early Vertebrate TGF-β Family Signaling in Mesenchymal Development Differentiation Joseph Zinski, Benjamin Tajer and Mary C. Mullins Ingo Grafe, Stefanie Alexander, Jonathan R. Peterson, et al. Bone Morphogenetic Protein−Based Therapeutic TGF-β1 Signaling and Tissue Fibrosis Approaches Kevin K. Kim, Dean Sheppard and Harold A. Jonathan W. Lowery and Vicki Rosen Chapman TGF-β Family Signaling in Ductal Differentiation Bone Morphogenetic Proteins in Vascular and Branching Morphogenesis Homeostasis and Disease Kaoru Kahata, Varun Maturi and Aristidis Marie-José Goumans, An Zwijsen, Peter ten Dijke, Moustakas et al. TGF-β Signaling in Control of Cardiovascular TGF-β Family Signaling in Epithelial Function Differentiation and Epithelial−Mesenchymal Marie-José Goumans and Peter ten Dijke Transition Kaoru Kahata, Mahsa Shahidi Dadras and Aristidis Moustakas TGF-β Family Signaling in Tumor Suppression TGF-β Family Signaling in Connective Tissue and and Cancer Progression Skeletal Diseases Joan Seoane and Roger R. Gomis Elena Gallo MacFarlane, Julia Haupt, Harry C. Dietz, et al. Targeting TGF-β Signaling for Therapeutic Gain The TGF-β Family in the Reproductive Tract Rosemary J. Akhurst Diana Monsivais, Martin M. Matzuk and Stephanie A. Pangas Regulation of Hematopoiesis and Hematological TGF-β Family Signaling in Drosophila Disease by TGF- β Family Signaling Molecules Ambuj Upadhyay, Lindsay Moss-Taylor, Myung-Jun Kazuhito Naka and Atsushi Hirao Kim, et al. TGF-β Family Signaling in Neural and Neuronal Signaling Cross Talk between TGF-β/Smad and Differentiation, Development, and Function Other Signaling Pathways Emily A. Meyers and John A. Kessler Kunxin Luo For additional articles in this collection, see http://cshperspectives.cshlp.org/cgi/collection/

Copyright © 2017 Cold Spring Harbor Laboratory Press; all rights reserved Downloaded from http://cshperspectives.cshlp.org/ on September 25, 2021 - Published by Cold Spring Harbor Laboratory Press

For additional articles in this collection, see http://cshperspectives.cshlp.org/cgi/collection/

Copyright © 2017 Cold Spring Harbor Laboratory Press; all rights reserved