Accepte

ReprintWACC 4

Surgery

0039-60

� 2012

doi:10.1

Sonic Hedgehog in pancreatic cancer:From bench to bedside, then backto the benchDavid E. Rosow, MD, Andrew S. Liss, PhD, Oliver Strobel, MD, Stefan Fritz, MD, Dirk Bausch, MD,Nakul P. Valsangkar, MD, Janivette Alsina, MD, PhD, Birte Kulemann, MD, Joo Kyung Park, MD,Junpei Yamaguchi, MD, Jennifer LaFemina, MD, and Sarah P. Thayer, MD, PhD, Boston, MA

From the Pancreatic Biology Laboratory, Department of Surgery, Massachusetts General Hospital, Boston, MA

DEVELOPMENTAL GENES are known to regulate cell pro-liferation, migration, and differentiation; thus, itcomes as no surprise that themisregulation of devel-opmental genes plays an important role in the biol-ogy of human cancers. One such pathway that hasreceived an increasing amount of attention for itsfunction in carcinogenesis is the Hedgehog (Hh)pathway. Initially the domain of developmental biol-ogists, the Hh pathway and one of its ligands, SonicHedgehog (Shh), havebeen shown toplay an impor-tant role in body planning and organ development,particularly in the foregut endoderm. Their impor-tance in human disease became known to cancerbiologists when germline mutations that resultedin the unregulated activity of the Hh pathway werefound to cause basal cell carcinoma and medullo-blastoma. Since then,misexpression of theHh path-way has been shown to play an important role inmanyother cancers, including thoseof thepancreas.In many institutions, investigators are targeting mis-expression of the Hh pathway in clinical trials, butthere is still much fundamental knowledge to begained about this pathway that can shape its clinicalutility. This review will outline the evolution of ourunderstandingof this pathway as it relates to thepan-creas, as well as how the Hh pathway came to be ahigh-priority target for treatment.

THE HEDGEHOG PATHWAY---OVERVIEW

Since the identification of the Hh gene inDrosophila nearly 30 years ago, insights into the mo-lecular pathway and its role in development and

d for publication May 11, 2012.

requests: Sarah P. Thayer, MD, PhD, 15 Parkman Street,60, Boston, MA 02114. E-mail: sthayer@partners.org.

2012;152:S19-32.

60/$ - see front matter

Published by Mosby, Inc.

016/j.surg.2012.05.030

tumorigenesis have taken investigation far fromthe bench to the patient’s bedside. The Hh genewas found initially in Drosophila when a large-scalescreening for mutations revealed an altered seg-mentation pattern of larvae, resulting in a short,fat larva covered in a ‘‘lawn’’ of denticles resem-bling a hedgehog; thus the name was born.1 Sincethat time, Hh signaling has been shown to be oneof the most important signal transduction path-ways in animal development, playing importantroles not only in body planning but also in cellfate and stem cell maintenance. Logically, derange-ment of this pathway can alter profoundly theregulation of normal cell growth, and its misregu-lation has now been shown to play an importantrole in human cancers.

Mammals have 3 Hh ligands, Desert Hedgehog(Dhh), Indian Hedgehog (Ihh), and Sonic Hedgehog(Shh). Much of what is known about the Hh path-way has been derived from studies involving Shh,which exerts its effect on pattern formation andcontrol of cell growth through activation of a sig-naling cascade. Shh is translated as a 45-kDa pre-cursor protein that goes through multiple stepsessential for generating active ligands. Shh un-dergoes autocatalytic processing, which yields a19-kDa, N-terminal, Shh signaling domain (Shh-N) and a 25-kDa C-terminal domain (Shh-C).Shh-C has no known signaling activity but becomesactive when its C terminus is modified by choles-terol and its N terminus by palmitate. When mod-ified Shh-N reaches its target cells, it binds to the12-pass transmembrane protein Patched (Ptch)with high affinity at the cell surface. Ptch existsas 2 isoforms, Ptch1 and Ptch2. Shh binds equiva-lently to both of these but is believed to exert itseffect mainly through Ptch1. In the absence ofShh ligand, Ptch1 inhibits the activity of Smooth-ened (Smo), the 7-pass transmembrane protein.When Shh binds to Ptch1, it inhibits Smo repres-sion and results in its activation, which transduces

SURGERY S19

SurgerySeptember 2012

S20 Rosow et al

the Hh signal to the cytoplasm. Although the exactmechanism is still unknown, Smo activation resultsin the formation or release of activating forms ofGli (Gli-1 and Gli-2) while blocking the productionof its repressive forms (Gli-3). The balance be-tween the activating and repressing forms of theGli proteins results in the expression of a varietyof target genes (Fig 1).

In addition to this ‘‘canonical pathway’’ of Hh,recent evidence points toward the existence of anoncanonical pathway, where the mechanism ofactivation deviates from that described previously.To date, noncanonical mechanisms that have beenidentified include Shh signaling independent ofGli-mediated transcription and interaction or acti-vation of the Hh pathway by other pathways inde-pendent of ligand activation. For example, Ptchhas been shown to interact directly with cyclin B1and caspases to inhibit proliferation; this effect isindependent of Gli.2 In several other studies, au-thors further suggest that GLI transcription factorsare regulated not solely by Hh signaling but canalso be regulated by RAS signaling.3,4 Because ofthe context- and gradient-dependent manner bywhich Hh exerts its effects on the cell, spatialand temporal activity of both the canonical andnoncanonical Hh pathways must be regulatedtightly for normal embryonic development to oc-cur. This type of sophisticated regulation of theHh pathway is seen during pancreatic develop-ment, which relies on temporal and spatial regula-tion of several Hh ligands.5

Hh SIGNALING AND PANCREATICDEVELOPMENT

The formation of the mammalian pancreas is acomplex process. Development begins with thealignment of endoderm cells along the verticalbody axis to form the gut tube, which is thendivided into the foregut, midgut, and hindgut. Atthe boundary between foregut and midgut, theseptum transversum generates 2 pancreatic budscalled the dorsal and ventral endoderm. Subse-quently, these 2 epithelial extensions that branchinto the surrounding mesenchyme fuse to formthe mature pancreas, which follows the rotation ofthe stomach.6 The dorsal bud arises first and gen-erates the main part of the pancreas. The ventralbud develops later as part of the hepatic diverticu-lum; thus, it is associated with the bile duct andforms only part of the head and uncinate processof the pancreas.

All 3 mammalian hedgehog genes, Shh, Ihh, andDhh, are expressed differentially throughout

pancreatogenesis, with Shh showing the widestrange of expression.7 Early in development,around embryonic day 8.5 to 9.0 (e8.5–9.0), Shhand Ihh are expressed throughout the endodermalepithelium of the primitive gut but are noticeablyabsent and down-regulated in the area of the fu-ture pancreas. This down-regulation identifies thearea of the endoderm that has been programmedspecifically to become the future pancreas.8 Ihhand Dhh ligands, including the Ptch1/Smo recep-tor complex, are expressed later in the develop-ment of the pancreas (approximately e13.5).9 Ithas been reported that epithelial and b-cell expan-sion are regulated by Hh signaling.10,11

The importance of the regulation of the Hhpathway in pancreatic morphogenesis has beendemonstrated by misexpression experiments. Ec-topic expression of Shh early in developmentresults in the misdirection of the pancreas towarda duodenal fate,12,13 whereas loss-of-function muta-tions result in an expansion of the pancreatic do-main.14 Loss of Ihh has been associated withannular pancreas, a rare condition characterizedby an extension of the pancreatic gland, whichthen encircles the duodenum.13 To date, completeablation of Hh signaling in transgenic mice has notbeen achieved because of the early embryonic le-thality associated with Hh depletion.9 These datasuggest that misregulation of Shh signaling duringembryogenesis results in morphologic and func-tional defects in the developing pancreas, reveal-ing the important role this pathway playsthroughout embryonic pancreatic development.

Interestingly, all Hh ligands are expressed in theadult pancreas, suggesting that Hh may also con-tinue to play an important role in pancreaticmaintenance. The Ptch receptor is identified inthe mesenchyme and Ptch-2 in the epithelium.The expression of the ligand appears to be re-stricted in different compartments throughout thepancreas. For example, Ihh is identified in theadult islet, Dhh in the peripheral nerves, and Shhhas been found in a discrete epithelial compart-ment, the set of pancreatic duct glands (PDG).15

Thus, the Hh pathway and its signaling are impor-tant not only for pancreatic development but alsofor homeostasis and regeneration.

Hh SIGNALING IN PANCREATICREGENERATION

The identification of Hh in the adult pancreasand its up-regulation in response to injury suggestthat this pathway plays a central role in pancreaticregeneration. Evidence from studies of pancreatic

Fig 1. Diagrammatic representation of the hedgehog (Hh) signaling pathway. In the absence of Hh ligand (A),PATCHED protein (PTCH) prevents smoothened (Smo) from accessing the cilium. The zinc finger protein Gli un-dergoes proteosomal cleavage and is converted to its repressor form (GliR). Gli R migrates to the nucleus and silencesHh target genes. In the presence of Hh ligand (B), Hh binds to PTCH and inactivates it via endocytosis and lysosomaldegradation. This process allows Smo to associate with the cilium, and Smo facilitates the activation of Gli. Activated Gli(GliA) migrates to the nucleus and stimulates Hh target gene transcription. Inhibitors of the Hh signaling pathway havebeen indicated (blue boxes) and may act primarily on Hh, Smo, or Gli. Of these, Smo antagonists are being evaluatedcurrently in phase 1 and 2 clinical trials.

SurgeryVolume 152, Number 3S

Rosow et al S21

regeneration after pancreatic parenchymal injuryin rodents supports the hypothesis that the globalregeneration process occurs via 2 mechanisms.The first mechanism is an expansion of the ductalcell population, followed by differentiation of thisexpanded pool into mature exocrine, endocrine,or duct cells. The second mechanism occurs viareplication of existing beta cells in the remnantpancreas.16,17 The specific signaling pathwaysrequired for this regeneration process have beenexamined by the use of molecular approaches inmodel systems, including transgenic mice andzebra fish.

To analyze the mechanisms of regenerative re-pair in the exocrine pancreas after injury, a variety ofpharmacologic and surgical approaches have been

used to simulate pancreatitis. One of the mostcommon pharmacologic approaches is the use ofthe cholecystokinin analogue cerulein. In a well-characterizedmodel of acute pancreatitis, exposureto cerulein via intraperitoneal injection in rodentsleads to the loss of acinar cell mass followed byregeneration. In themouse, cerulein leads to a near-complete loss of acinar cells within 72 hours afterinjection, followed by restoration of the acinar cellcompartment within 7 days of injury.18 After acuteinjury, some acinar cells undergo a process of dedif-ferentiation with loss of expression of acinar cellmarkers, expression of pancreatic progenitorcell markers (nestin, Pdx1), and morphologicchanges that include the formation of a metaplasticepithelium. These steps involve the rapid

SurgerySeptember 2012

S22 Rosow et al

proliferation of these cells after re-entering the cellcycle, followed by their redifferentiation into spe-cific cell types.

Evidence for the role of Hh signaling in regen-eration of the exocrine compartment comes fromstudies in which authors have used a combinationof pharmacologic and genetic approaches to showthat in the absence of an intact Hh pathway, theexocrine compartment is unable to regenerate.During normal regeneration, Hh is up-regulatedafter injury; indeed, levels of Hh are detected inthe regenerating metaplastic epithelium as well asin the acinar cells. After regeneration is complete,Hh levels become undetectable. Inhibition of Hhsignaling does not affect the initial steps that leadto the formation of a metaplastic epithelium afteracute injury or the severity of injury. The maineffect of Hh inhibition is the inability of thededifferentiated acinar cells to redifferentiateinto mature cells after formation of metaplasticlesions and the persistence of these lesions ininjured pancreata.19

These results support the hypothesis that exo-crine regeneration after pancreatic injury requiresthe activation of Hh signaling and up-regulation ofother embryologic markers, including Pdx1 andnestin. This observation suggests that pancreaticexocrine regeneration recapitulates some aspectsof pancreatic development, but, more importantly,that dysregulation of this pathway by chronicinflammation may play an early role in pancreaticmetaplasia and cancer.

THE EARLY AND LATE ROLE OF THE HhPATHWAY IN PANCREATIC CANCER: THEROLE OF Hh IN CHRONIC INFLAMMATIONAND METAPLASIA

Aberrant regulation of developmental pathwayshas been identified as an important property inmany cancers, including pancreatic adenocarci-noma. The same pathways that direct organ pat-terning and growth during embryonic developmentalso govern tissue homeostasis in adult organs byregulation of stem cell proliferation and differenti-ation.20 Through inappropriate activation of thesepathways, chronic tissue injury may result in misdi-rection of tissue repair, ultimately resulting in neo-plasia. Increasing evidence suggests that the Hhpathway plays a key role in the initiation of pancre-atic cancer as well as in its maintenance andprogression.

Aberrant expression of members of the Hhpathway in chronic pancreatitis and pancreatic can-cer was first noted by Kayed et al21 in 2003, although

these authors believed Ihh to be the importantligand in the diseased pancreas. Subsequent re-search showed that the ligand Shh is expressed aber-rantly in pancreatic cancer and its precursor PanINlesions and that Shh functions as a mediator of can-cer initiation and growth.22 Mice with transgenicmisexpression of Shh in the pancreatic endodermdeveloped lesions resembling PanIN, and Hh inhi-bition by cyclopamine induced apoptosis andblocked proliferation of pancreatic cancer cellsin vitro and in vivo. Thus, Hh signaling was identi-fied as an early and late mediator of pancreatic duc-tal adenocarcinoma (PDAC).22 These findings weresupported by a concurrent study in which the au-thors demonstrated that the up-regulation ofHh lig-and is required for the maintenance of severaldigestive tract cancers, including pancreatic can-cer.23 Meanwhile, Hh misexpression has also beendescribed in intraductal papillary mucinous neo-plasms (IPMN), the second precursor of invasivepancreatic cancer.24

In subsequent studies, investigators focused onthe mechanisms of activation of the Hh pathwayand how it contributes to carcinogenesis. Thespecial role that aberrant activation of the Hhpathway plays in pancreatic carcinogenesis is ex-plained by its function in pancreatic development.In the embryonic foregut, Shh promotes develop-ment of the gastrointestinal tube and repressespancreatic development.25 Shh signaling wasthought previously to disappear early in pancreaticdevelopment,25 but work from our group identi-fied Shh expression in a specialized epithelial com-partment which appears as gland-like outpouchesoff larger pancreatic ducts; we termed these pancre-atic duct glands (ie, PDG).15 In response to chronicinjury, PDG proliferate, up-regulate Shh as well asthe Hh receptor PTCH (Fig 2), and undergo a gas-trointestinal mucinous metaplasia.15

It had been noted previously that pancreaticcancer and its precursor lesions, PanIN, frequentlydisplay misexpression of gastrointestinal markers,such as the mucins Muc-5ac and Muc-6,26,27 as wellas Shh.22 In vitro, activation of the Hh pathway bytransfection with the Hh transcription factor Gli-1 resulted in up-regulation of several foregutmarkers.28 Strobel et al15 demonstrated that reca-pitulating its role in development, up-regulationof the ligand Shh is sufficient to misdirect the pan-creatic ductal epithelium toward a gastrointestinalmetaplastic phenotype both in vitro and in vivo(Fig 3). The underlying up-regulation of Shh in re-sponse to chronic injury may be caused by signalsinvolved in tissue repair and expression of other

Fig 2. Pancreatic duct glands (PDG) in the normal pancreas (A), in chronic inflammation (B), and in inflammation-associated severe metaplasia (C). Shh is expressed at a low level in normal PDG and up-regulated in hypertrophied PDGin response to chronic injury (arrowheads). This up-regulation is accompanied by up-regulation of the Hh receptor anddownstream pathway member PTCH2 in the epithelium and PTCH1 in the stroma (modified from Strobel et al15).

SurgeryVolume 152, Number 3S

Rosow et al S23

proteins known to mediate inflammatory cues,such as nuclear factor kappa B.29

Apart from PDG, Shh misexpression can also beidentified inmucinousmetaplastic lesions that formin response to injury in the pancreatic paren-chyma.30 A minority of these lesions have beenfound by lineage tracing to originate from acinarcells by true transdifferentiation, whereas themajor-ity arise most likely from terminal ductal or centroa-cinar cells but not from beta cells.30,31

Chronic injury thus appears to result in a Shh-mediated gastrointestinal metaplasia as an earlyand important step in pancreatic carcinogenesis.This process may induce a common mechanismin the initiation of pancreatic cancer, or perhapsthis observation suggests a common compartmentof origin that results in both precursor lesionsPDAC, PanIN and IPMN. Similar to the role ofBarrett’s metaplasia at the gastroesophageal junc-tion, this Shh-mediated gastrointestinal metaplasiain the pancreas may be the basis of furthermolecular aberrations and mutations that mediatethe progression of precursor lesions to invasivecancer.

Other in vitro and in vivo studies demonstratethat Hh contributes to pancreatic carcinogenesis at

several stages and through several mechanisms.32

Synergistic molecular crosstalk between the Hhpathway and activated K-RAS, a key mediator ofpancreatic carcinogenesis,33 have been clearlydemonstrated.3,32,34,35

Hedgehog is not only an early mediator in-volved in carcinogenesis but is also a late mediatorof pancreatic cancer.22 Analyses of global genomicsequencing have recently identified the Hh path-way as one of the few ‘‘core’’ pathways of centralimportance for the biology of human pancreaticcancer.36 It has meanwhile become evident thatHh signaling contributes to several hallmarks ofthe maintenance and progression of pancreaticcancer, such as cancer stem cells (CSC), interac-tions between cancer epithelium and stroma, angi-ogenesis, and metastases. These different aspectsunderscore the therapeutic potential of Hh path-way inhibition in pancreatic cancer.

Shh AND CSC AND METASTASIS

Cancer can be defined as a disease of unregulatedgrowth and self-renewal. Many conceptual similari-ties exist between stem cells and cancer cells,suggesting the existence of CSC that initiatetumorigenesis and possess indefinite potential for

Fig 3. Effects of the Shh ligand on pancreatic ducts cells in vitro (A) and in vivo (B). (A) Stimulation of human pan-creatic ductal epithelium (HPDE) with recombinant Shh results in a mucinous phenotype with expression of gastroin-testinal mucins. (B) Transgenic misexpression of Shh in the pancreas results in misdirection of ductal epithelialdifferentiation toward a phenotype that resembles closely the gastric pyloric epithelium (modified from Strobel et al15).

SurgerySeptember 2012

S24 Rosow et al

self-renewal and multilineage differentiation.37,38

CSC are very resistant to adverse conditions in the tu-mor microenvironment.39 Interestingly, Shh hasbeen identified recently as being up-regulated inpancreatic CSC, with distinct populations of CSC(CD24+/CD44+/ESA+) shown to have Shh expres-sion 46-fold greater than CD24�/CD44�/ESA�.40

This finding suggests a ubiquitous up-regulation ofthe Shh ligand in the CSC population comparedwith the heterogeneous population of the tumor.The role of Shh in pancreatic cancer stem cellsremains to be determined, but up-regulation ofShh indicates a role in carcinogenesis drivenbyCSC.

It is believed widely that CSC are responsiblefor progression and metastasis of epithelial tu-mors. Hermann et al41 discovered a population ofCD133+/CXCR4+ cells, as compared with CD133+/CXCR4� cells, that is able to set up metastases,which are able to induce tumor growth but donot form distant metastases. The authors suggest

the existence of 2 different CSC populations, mi-gratory and stationary CSC. If true, this conceptopens a new window for potential treatment to pre-vent metastatic spread. The Shh expression levelsof these distinct cells remain to be determined,but they are likely to be up-regulated becausethey are up-regulated in their CD24+/CD44+/ESA+ counterparts. The possibility of treating thesespecial CSC to prevent the lethal event of meta-static spread and thus keeping the tumor localizedmay offer a major insight and advantage for fur-ther investigations in the treatment of pancreaticcancer. Only a few cases of CSC have been isolatedand purified in pancreatic cancer. The ability toidentify and compare those cells of origin thatare able to set up metastases would give us the pos-sibility---in PDAC as well as in other epithelial can-cers---to finally find a molecular target forpotentially promising results in the battle againstthis dismal disease.

SurgeryVolume 152, Number 3S

Rosow et al S25

THE Hh PATHWAY AS A POTENTIAL THERA-PEUTIC TARGET

With a large body of evidence supporting a rolefor Shh in PDAC, a number of preclinical studieshave been performed that allow us to understandfurther the role of the Hh pathway in PDAC as wellas to evaluate the efficacy of inhibition of the Hhpathway for the treatment of this disease. Usingcyclopamine, a naturally occurring Smo inhibitor,our group in collaboration with other authors firstreported the efficacy of inhibiting the Hh pathwayin decreasing the growth and tumorigenesis ofPDAC cell lines.22 In response to cyclopamine,50% of PDAC cell lines examined showed de-creased proliferation and increased apoptosisin vitro relative to control cells. Importantly, thetreatment of mice bearing heterotopic xenograftcancers with cyclopamine resulted in a 50% de-crease in tumor growth in responsive lines. Twostudies in which investigators used an orthotopicmodel of PDAC demonstrated that inhibition ofthe Hh pathway with cyclopamine or IPI-269609,an orally bioavailable inhibitor of Smo, preventeddistant metastases.42,43 In addition to blockingthe activity of Smo, strategies have been used totarget Shh directly. The treatment of mice withorthotopic xenograft cancers with a neutralizingantibody against Shh, 5E1, resulted in a greaterthan 4-fold decrease in tumor volume.44 This treat-ment also decreased the incidence of metastasis tothe lymph nodes, liver, and peritoneum.

Although the use of cells derived from humantumors in these preclinical studies allows one toevaluate the efficacy of inhibiting the Hh pathwayin the context of the diverse genetic makeup ofPDAC, genetically engineered mouse models ofPDAC play an important role in understandinghow inhibition of the Hh pathway affects thenatural course of tumor development. In onesuch study, investigators used a mouse model ofPDAC induced by the pancreas-specific activationof KRAS and inactivation of Ink4a/Arf.45 Treat-ment of these mice with cyclopamine resulted ina small increase in survival. In a second study, Oliveet al46 used a mouse model of PDAC induced bythe activation of KRAS and inactivation of p53.They demonstrated that inhibition of Hh signalingby the Smo inhibitor IPI-926 can enhance the ef-fectiveness of chemotherapeutics. Although thetreatment of mice with this compound alone hadno effect on survival, combining treatment withgemcitabine resulted in a doubling of the survivaltime for these mice. Analysis of these tumorsrevealed that tumors from mice treated with IPI-

926 had a decrease in tumor stroma, which re-sulted in an increased delivery of gemcitabine tothe tumor cells.

The results of this study highlight a recent shift inthe understanding of Hh signaling in PDAC with agreater focus on its role in the tumor microenvi-ronment. Although early work in the field focusedon an autocrine role for Shh in the development ofPDAC, recent experiments using cell lines treatedwith Hh pathway inhibitors have demonstrated thatgreater levels of inhibitors were required to exerttheir biologic effect than were needed to inhibit Hhsignaling.47 Although this study suggested that theobserved biologic effects may be caused by off-target effects, the results of at least one other studysuggest that some of the biologic effects are indeedattributable to the inhibition of the Hh pathway.48

Consistent with other studies, the authors demon-strated that the treatment of PDAC cell lines with cy-clopamine resulted in decreased proliferation andincreased apoptosis. Importantly, when the Smo ag-onist purmorphamine was added, cell proliferationwas restored, but there was no change in apoptosis,indicating that at least some of the biologic effectsof cyclopamine are dependent on Hh signaling.Given the wide variation in the sensitivity of PDACcell lines to Hh inhibitors, the disparity in theamounts of compounds used in various studies,and recent discoveries that Gli can be activated bySmo-independent pathways (such as KRAS andTGF-beta), more rigorous analysis will be neededto define a role for autocrine Shh signaling inPDAC.35

Despite the disparate reports of autocrine signal-ing in PDAC, a growing body of evidence supportsthe importance of ligand-dependent activation ofHh signaling in the tumor stroma. In human tumorxenografts, expression of Shh by tumor cells corre-lated with increased expression of GLI-1 andPTCH1 in the stromal compartment. Pathway inhi-bition affected only stromal GLI-1 and PTCH1expression and resulted in decreased tumor growthexclusively in Hh ligand-expressing tumors.47 Con-sistent with these findings, Tian et al49 demon-strated that the expression of an oncogenicSmoothened (SmoM2) in mouse pancreata neitheractivates Hh signaling in epithelial cells nor pro-motes their neoplastic transformation. In murinepancreatic cancer models as well as in human pan-creatic cancer specimens, activation of the Hh path-way was observed only in stromal cells surroundingHh ligand-expressing tumor cells. Not only is activa-tion of the Hh pathway observed in the stroma, butShh appears to play a role in promoting growth of

SurgerySeptember 2012

S26 Rosow et al

the stroma. Employing an orthotopic xenograftmodel of PDAC, it was demonstrated that Capan-2cells overexpressing Shh induced amore robust des-moplastic reaction than that observed in tumors de-rived from control cells.50

The aforementioned work by Olive et al46 pro-vided further insight into the mechanism by whichinhibition of the Hh pathway affects tumor growthin pancreatic cancer. In a genetically engineeredmouse model of PDAC, Olive et al found that inhi-bition of the Hh pathway resulted in depletion ofthe stromal compartment together with a transientincrease in vascular density. This study is especiallyimportant because it was the first to demonstratethe potential importance of the combined treat-ment of the cancer and stroma, a potentially keytherapeutic strategy in treating this disease.39,46

In stark contrast to the findings by Olive et al,46

Hh signaling itself is known to promote angiogen-esis both indirectly and directly in nontumor models.In recent studies authors also identified Hh path-way activation as indirectly proangiogenic in tumormodels. Several groups have reported an indirect ef-fect of Hh on angiogenesis through the inductionof proangiogenic cytokines in mesenchymal cells.Pola et al51 found that the activation of Hh signal-ing improved blood flow and limb salvage in amouse model of hind-limb ischemia by inducingexpression of vascular endothelial growth factor(VEGF) and angiopoietins-1 and -2 (ANG1, 2) inmesenchymal cells. Asai et al52 identified activeHh signaling during wound repair and foundthat Hh signaling accelerates healing by increasingvascularity. Topical gene therapy with SHH in-creased the expression of proangiogenic cytokinesin stromal fibroblasts and enhanced the recruit-ment of endothelial progenitor cells. Using animalmodels of retinal and choroidal neovasculariza-tion, Surace et al53 demonstrated a requirementfor Hh signaling during physiologic and patho-logic retinal vascularization through increasedexpression of VEGF in mesenchymal cells. In 2more recent studies investigators delineate anadditional, direct proangiogenic effect of Hh signal-ing on endothelial cells.54,55 Both found that theHh ligand does not induce its target genes GLI-1and PTCH1 in endothelial cells but ratheractivates the small GTPase RhoA in a SMO-dependent manner. Activation of RhoA leads tocytoskeletal rearrangement and enhanced migra-tion, and ultimately promotes angiogenesis.

More recently, angiogenesis promoted by theactivation of Hh signaling was also identified incancer. Nakamura et al56 reported that inhibition ofthe Hh pathway results indirectly in a decrease of

tumor vasculature by decreased homing of bonemarrow-derived cells to PDAC xenografts togetherwith a decrease in expression of stromal Ang-1 andIGF-1 within the tumors. Chen et al57 found thatHh signaling increases indirectly the tumor vascular-ity in PDAC by the induction of VEGF expression instromal perivascular cells; however, the tumor cellsthemselves are commonly thought to secrete themajority of the VEGF found within cancers.58,59

Further studies are, therefore, needed to clarify:(1) how Hh signaling was found to promotevascularization in PDAC in some studies,56,57

whereas other groups found that its inhibitionleads to increased tumor vascularity46; (2) whyHh inhibition results in a marked decrease in tu-mor vascularity,56,57 even though the amount ofVEGF contributed by mesenchymal cells (and reg-ulated by Hh signaling) is relatively small com-pared to the amount secreted by the tumor cellsthemselves in most cancers58,59; and (3) whetherHh signaling also promotes angiogenesis directlyin cancer by acting on endothelial cells. What isclear is that the Hh pathway plays key roles inthe maintenance, growth, and migration ofPDAC, making it an ideal clinical target.

BEDSIDE: CLINICAL TRIALS USING THE HhPATHWAY

The development of Hh pathway inhibitors be-gan with cyclopamine, which binds and inactivatesSmo.60 In mouse models, cyclopamine and theclosely related compound jervine were associatedwith low affinity, poor bioavailability, chemical insta-bility, and rapid clearance,61 making cyclopamine apoor candidate for clinical use. Hh pathway inhibi-tors appropriate for the clinics were identified withthe use of standard medicinal chemistry and smallmolecule screens. One of the first drugs developedvia the use of small molecule screening, CUR-61414(Curis),62 induced apoptosis in preclinical mousemodels of basal cell carcinoma. This drug was a top-ical agent, limiting its use in other solid tumors.Since that time, a number of Smo antagonistshave been identified. Although most of the currentdrug development targets Smo, attempts to targetmembers of the Hh pathway both upstream(PTCH) and downstream of Smo (GLI) are underway but are still in preclinical testing.

Seven small-molecule inhibitors of Smo arebeing evaluated currently in phase 1 and 2 trials.These include GDC-0449 (Genentech), IPI-926(Infinity Pharmaceuticals), BMS-833923 (BMS/Ex-elixis), TAK-441 (Millennium), LDE225, LEQ506(Novartis), and PF-04449913 (Pfizer). Table listsclinical trials in which investigators are using these

Table. Clinical trials investigating inhibitors of SHH in pancreatic cancer

Hedgehoginhibitor

Otherinterventions Study title Phase Trial identifier Status

GDC-0449 Gemcitabine andNab-Paclitaxel

A Phase II Study of Gemcitabine and Nab-Paclitaxel in Combination WithGDC-0449 (Hedgehog Inhibitor) in Patients With Previously UntreatedMetastatic Adenocarcinoma of the Pancreas

hase 2 NCT01088815 Recruiting

GDC-0449 None Proof of Mechanism Study of an Oral Hedgehog Inhibitor (GDC-0449) inPatients With Resectable Pancreatic Ductal Adenocarcinoma in thePre-operative Window Period

hase 2 NCT01096732 Recruiting

GDC-0449 Gemcitabine Cancer Stem Cells and Inhibition of Hedgehog Pathway Signaling inAdvanced Pancreas Cancer: A Pilot Study of GDC-0449 in CombinationWith Gemcitabine

hase 2 NCT01195415 Recruiting

GDC-0449 Gemcitabine A Multi-Center, Double Blind, Placebo-Controlled, Randomized Phase II ofGemcitabine Plus GDC-0449 (NSC 747691), a Hh Pathway Inhibitor, inPatients With Metastatic Pancreatic Cancer (10052747)

hase 2 NCT01064622 Recruiting

GDC-0449 Erlotinib andGemcitabine

Phase I Trial of the Combination of GDC-0449 and Erlotinib+/�Gemcitabine

hase 1 NCT00878163 Unknown*

IPI-926 None A Phase 1 Study of IPI-926 in Patients With Advanced and/or MetastaticSolid Tumor Malignancies

hase 1 NCT00761696 Active, notrecruiting

IPI-926 Gemcitabine A Phase 1b/2 Study Evaluating IPI-926 in Combination With Gemcitabinein Patients With Metastatic Pancreatic Cancer

hase 1/2 NCT01130142 Recruiting

IPI-926 FOLFIRINOX A Phase I Study of FOLFIRINOX Plus IPI-926 for Advanced PancreaticAdenocarcinoma

hase 1 NCT01383538 Recruiting

LDE-225 Gemcitabine Phase 1/2 Safety and Feasibility of Gemcitabine in Combination WithLDE-225 as Neoadjuvant Therapy in Patients With Borderline ResectablePancreatic Adenocarcinoma

hase 1/2 NCT01431794 Not yetrecruiting

PF-04449913 None A Phase 1 Study To Evaluate The Safety, Pharmacokinetics, andPharmacodynamics of PF-04449913, an Oral Hedgehog Inhibitor,Administered as Single Agent in Select Solid Tumors

hase 1 NCT01286467 Recruiting

BMS-833923 None Phase 1 Multiple Ascending Dose Study of BMS-833923 (XL139) in SubjectsWith Solid Tumors

hase 1 NCT01413906 Not yetrecruiting

TAK- 441 None A Multicenter, Open-Label, Dose-Escalation, Phase 1 Study of TAK-441, anOral Hedgehog Signaling Pathway Inhibitor, in Adult Patients WithAdvanced Nonhematologic Malignancies

hase 1 NCT01204073 Recruiting

LEQ506 None A Dose Finding and Safety Study of Oral LEQ506 in Patients With AdvancedSolid Tumors

hase 1 NCT01106508 Recruiting

*Indicates status has not been verified in more than 2 years.ClinicalTrials.gov September 2011.

Surgery

Volu

me152,Number

3S

Rosow

etal

S27

P

P

P

P

P

P

P

P

P

P

P

P

P

SurgerySeptember 2012

S28 Rosow et al

drugs for pancreatic cancer and other solid tu-mors. The Smo inhibitors that are furthest alongin their development are GDC-0449 and IPI-926.The first of these, GDC-0449, is a synthetic, highlypotent small-molecule inhibitor of Smo that isavailable as an oral formulation.63

Results from 2 early-phase clinical trials showedencouraging results with respect to drug safety andtolerance. The first, a phase 1 trial for patients withrefractory, locally advanced or metastatic solidtumors, was conducted by LoRusso et al64 with 68patients (8 of whom had pancreatic cancer). Inthis trial, the overall response rate for the mostcommon diagnosis, basal cell carcinoma, was58%; one patient with PDAC experienced stabledisease as the best response while being on studyfor 3 months. This drug was well tolerated. Grade4 events were observed in 9% of the patients inthis study, including hyponatremia, fatigue, andpyelonephritis. On the basis of the relationship be-tween dose increments and the peak observedsteady-state plasma concentration, the authors rec-ommended a dose of 150 mg/day for future phase2 studies. In the second study, Palmer et al65 evalu-ated interaction between antagonists of the epider-mal growth factor receptor and Hh signalingpathways in a phase 1 trial. They compared a com-bination of erlotinib (epidermal growth factor re-ceptor tyrosine kinase inhibitor) and GDC-0449as cohort I versus erlotinib + GDC-0449 combinedwith gemcitabine as cohort II for unresectablesolid tumors. Preliminary results from 29 patientsindicated that, in cohort I, the only dose-limitingtoxicity was rash; in cohort II, the dose-limiting tox-icities seen were nausea, visual disturbances, andthrombocytopenia. The authors recommendedboth erlotinib and GDC-0449 at a dose of 150mg/day for phase 2 evaluation. In the aforemen-tioned clinical trials GDC-0449 was well toleratedwhen used alone or in combination with other che-motherapeutic drugs, and both studies deemedGDC-0449 suitable for phase 2 trials.

In addition to GDC-0449, the other small-molecule inhibitor of Smo undergoing phase1 evaluation currently is IPI-926 (Infinity Pharma-ceuticals). Preliminary data from 2 ongoing trialswere presented recently at the 2011 meeting of theAmerican Society for Clinical Oncology.66 Rudinet al67 administered IPI-926 in patients with locallyadvanced or metastatic solid tumors to determineits safety, antitumor activity, and pharmacokineticproperties. In the study population, there were 7 pa-tients with pancreatic cancer. With regard to phar-macodynamic effect, a decrease in Gli-1 expressionin normal skin was reported for 72% of the patients.

Themost common adverse events (grade 3) were fa-tigue (3%), alanine aminotransferase elevation(8%), and aspartate aminotransferase elevation(4%). There were no grade 4 or 5 events in thisstudy; the maximally tolerated dose was 160 mg/day. In the second trial focusing on PDAC, IPI-926was combined with gemcitabine and used as first-line treatment for patients with metastatic pancre-atic cancer.68Of the 15 patients enrolled in this trial,9 underwent postbaseline imaging, and 3 of theseshowed radiographic partial responses. The mostcommon adverse events included fatigue, nausea(both, overall 40%; grade 3, 0%) and aspartate ami-notransferase/alanine aminotransferase elevation(grade 3, 7%). There were no adverse events ofgrade 4. In addition, there was no adverse interac-tion between IPI-926 and gemcitabine. On the basisof these encouraging results, the authors havemoved this compound into a phase 2 trial.

Mechanisms of resistance also are4 being iden-tified in these early trials. A patient with metastaticmedulloblastoma who responded initially to Hhpathway inhibition subsequently showed a com-plete loss of therapeutic effect after 3 months oftherapy.69 Further investigation revealed a muta-tion of Smo involving an amino acid substitutionof aspartic acid (Asp to His on codon 473) thathampered the ability of GDC-0449 to bind toSmo.70 Further mechanisms of resistance may bethose relating to the noncanonical activation ofthe Hh pathway downstream of Smo. The develop-ment of second-generation Smo inhibitors andanti-Gli molecules, such as GANT58, GANT61,and further understanding of key signaling path-ways that interact with the Hh pathway, may allowus to anticipate and circumvent these mechanismsof resistance. The overall perception from clinicaltrials involving GDC-0449 and IPI-926 is that theseagents can be used safely in patients. These drugsare well tolerated, have minimal side effects, andcan be used without adverse interaction alongwith other commonly used cytotoxic therapies.Further subset analysis and phase 2 trials are underway to assess the clinical efficacy of Smo inhibitors.The subset of PDAC patients who would derive themaximum benefit from this therapy, and the com-plete side effect profile, remain unknown.

BACK TO THE BENCH

Since its initial discovery as a segment polaritygene in Drosophila and its mammalian orthologs,there has been an explosion of information aboutthe Hh pathway and the wide variety of functionsregulated by the Hh pathway during pancreatic em-bryogenesis and adult pancreatic homeostasis, and

Fig 4. The functions of Sonic Hedgehog in cancer.

SurgeryVolume 152, Number 3S

Rosow et al S29

more importantly, how aberrations in this pathwaycontribute to the formation and maintenance ofpancreatic cancer. During the past decade, it has be-come evident that the Hh pathway contributes tothe maintenance and progression of pancreaticcancer by maintaining the CSC and the tumor mi-croenvironment, altering angiogenesis, and regu-lating metastatic spread (Fig 4). Given the vastamount of preclinical studies, academia and indus-try have been working together to develop novelcompounds against this pathway. To date, sevencompounds are in clinical trials throughout theUnited States. Although encouraging interval datahave been reported on other solid tumors (basalcell cancers) known to have mutations in the Hhpathway, nothing is yet known about the efficacyof these drugs against pancreatic cancer.

Although Hh research has leaped from thebench to the bedside, many important questionsremain unanswered. For example, the molecularinteraction between Ptch and Smo, as well as theregulation of activating and repressive forms of Gli,remain to be determined. It is also unclear how Hhregulates the tumor microenvironment and tumorangiogenesis. Most recently, the identification ofShh in an epithelial compartment (PDG) that has

progenitor-like function and undergoes a muci-nous metaplasia resembling both pancreatic pre-cursor lesions (PanIN and IPMN) suggests that thiscompartment may be a common compartment oforigin for pancreatic cancer. This observation,added to the knowledge that CSC express highlevels of Shh, may suggest that cancer may originatefrom Shh-positive stem cells within the PDG. Butmuch work needs to be done before such specula-tion is proven to be true, so it’s back to the bench.

SPECIAL ACKNOWLEDGMENT TO DR WAR-SHAW BY DR SARAH THAYER

History of the pancreatic biology laboratory---the Warshaw laboratory. There was a fair amountof serendipity involved in the establishment of thePancreatic Biology Laboratory (PBL) by Dr An-drew Warshaw at the Massachusetts General Hos-pital, beginning with his decision as a surgicalresident in 1965 to spend 2 years as a fellow at theNational Institutes of Health. He saw this as analternative to being drafted as a general medicalofficer just prior to major U.S. involvement inVietnam, and spending 2 years on an army base.Dr Warshaw joined the National Institute of Ar-thritis and Metabolic Diseases to work in the

SurgerySeptember 2012

S30 Rosow et al

Gastroenterology Section and was the first surgeonto join what was essentially a medical hepatologyunit. Over time, however, the interests of thisgroup came to include malabsorption and pancre-atic insufficiency. By the time he returned toclinical service as a resident in 1967, he was seeinga wide range of pancreatic pathology and publish-ing his findings.

When Dr Warshaw finished his chief residencyand founded the PBL in 1972, rigorous research byAmerican surgeons into pancreatic physiology andbiology was still in its relative infancy. The PancreasClub had just been founded in 1966 and was still aninformal gathering of physicians and researcherswho discussed primarily acute pancreatitis. The or-ganization even disbanded for a year in 1974, beforeDr Warshaw and several other members pushed torestart it. Likewise, the precursors to the AmericanPancreatic Association began having meetings in1969, but these too were quite informal, with writtenminutes beginning only in 1971. It was not until 1979that the APA was formally incorporated.

In this environment,DrWarshaw is fondof sayingthat he chose his ultimate career path because thepancreas was ‘‘the last organ that had not beentaken.’’ He had already had 3 articles accepted bythe New England Journal of Medicine and despite justfinishing his residency, he was already gaining a rep-utation as an authority on the pancreas. Still, he wasgiven only a single bench and a technician to startout, with no funds or protected time for research.

Research and education. The initial efforts ofthe PBL were like much pancreatic research doneby surgeons at the time---focused primarily onpancreatitis. Dr Warshaw built on his clinicalexperience with pancreatitis and developed aresearch interest in amylase isoenzymes, publish-ing several papers on the subject in the 1970s.Dozens of his articles in this decade offer insightinto serum markers, anatomic studies, and clinicalobservations. For many years, the focus of the PBLwas strictly on pancreatic physiology and markersof pancreatitis, but this focus gradually shifted overtime to the study of pancreatic tumor biology.

Dr Warshaw was assisted in this transition by theimplementation of a fellowship program, begin-ning in 1986. Given his role as an attendingsurgeon at Harvard Medical School’s largest andoldest teaching hospital, Dr Warshaw felt a com-mitment to teaching and training the next gener-ations of surgeons and researchers; indeed, he hastrained 29 surgeon-scientists from 4 different con-tinents. Of these fellows, 5 have gone on tobecome chairs of their respective departments,

and many others have developed highly distin-guished academic careers.

As one of Dr Warshaw’s surgical prot�eg�es, I amhonored to have been a small part of his greatachievements and hope to continue his legacy ofscientific contribution and teaching. In 2001, whenDrWarshaw askedme to become a coinvestigator inthePBL, I began towork on thedevelopmental geneShh, known to inhibit pancreatic development. Ulti-mately, this work led to investigation into theHedge-hog pathway as a possible mechanism forunderstanding pancreatic tumorigenesis and re-sulted in a seminal report in Nature in 2003. Now,as director of thePBL, I have thepleasureofworkingwith and overseeing great new surgeon–scientists ofthe future. Many of these talented residents and sci-entists from this laboratory have contributed greatlyto our present-day insights into theHedgehog path-way. Thus this review was written as a collaborativegroup effort.

I am indebted to Dr Warshaw’s mentoring,encouragement and promotion, as well as theopportunity to train and work with the gifted fellowsof the PBL. ---Sarah Thayer

REFERENCES

1. Nusslein-Volhard C, Wieschaus E. Mutations affecting seg-ment number and polarity in Drosophila. Nature 1980;287:795-801.

2. Jenkins D. Hedgehog signalling: emerging evidence fornon-canonical pathways. Cell Signal 2009;21:1023-34.

3. Pasca di Magliano M, Sekine S, Ermilov A, Ferris J, DlugoszAA, Hebrok M. Hedgehog/Ras interactions regulate earlystages of pancreatic cancer. Genes Dev 2006;20:3161-73.

4. Fern�andez-Zapico ME. Primers on molecular pathwaysGLI: more than just Hedgehog? Pancreatology 2008;8:227-9.

5. Kayed H, Kleeff J, Osman T, Keleg S, B€uchler MW, Friess H.Hedgehog signaling in the normal and diseased pancreas.Pancreas 2006;32:119-29.

6. Murtaugh LC. Pancreas and beta-cell development: fromthe actual to the possible. Development 2007;134:427-38.

7. Hebrok M. Hedgehog signaling in pancreas development.Mech Dev 2003;120:45-57.

8. Aubin J, D�ery U, Lemieux M, Chailler P, Jeannotte L.Stomach regional specification requires Hoxa5-driven me-senchymal-epithelial signaling. Development 2002;129:4075-87.

9. Hebrok M, Kim SK, St Jacques B, McMahon AP, Melton DA.Regulation of pancreas development by hedgehog signal-ing. Development 2000;127:4905-13.

10. KingPJ,GuastiL,LauferE.Hedgehog signalling inendocrinedevelopment and disease. J Endocrinol 2008;198:439-50.

11. Lau J, Hebrok M. Hedgehog signaling in pancreas epithe-lium regulates embryonic organ formation and adult beta-cell function. Diabetes 2010;59:1211-21.

12. Apelqvist A, Ahlgren U, Edlund H. Sonic hedgehog directsspecialised mesoderm differentiation in the intestine andpancreas. Curr Biol 1997;7:801-4.

SurgeryVolume 152, Number 3S

Rosow et al S31

13. Lau J, Kawahira H, Hebrok M. Hedgehog signaling in pan-creas development and disease. Cell Mol Life Sci 2006;63:642-52.

14. Kim SK, Hebrok M, Melton DA. Notochord to endodermsignaling is required for pancreas development. Develop-ment 1997;124:4243-52.

15. Strobel O, Rosow DE, Rakhlin EY, Lauwers GY, TrainorAG, Alsina J, et al. Pancreatic duct glands are distinctductal compartments that react to chronic injury and me-diate Shh-induced metaplasia. Gastroenterology 2010;138:1166-77.

16. Bonner-Weir S, Li W-C, Ouziel-Yahalom L, Guo L, Weir GC,Sharma A. b-cell growth and regeneration: replication isonly part of the story. Diabetes 2010;59:2340-8.

17. Dor Y, Brown J, Martinez OI, Melton DA. Adult pancreaticbeta-cells are formed by self-duplication rather than stem-cell differentiation. Nature 2004;429:41-6.

18. Jensen JN, Cameron E, Garay MV, Starkey TW, Gianani R,Jensen J. Recapitulation of elements of embryonic develop-ment in adult mouse pancreatic regeneration. Gastroenter-ology 2005;128:728-41.

19. Fendrich V, Esni F, Garay MV, Feldmann G, Habbe N, Jen-sen JN, et al. Hedgehog signaling is required for effectiveregeneration of exocrine pancreas. Gastroenterology2008;135:621-31.

20. Taipale J, Beachy PA. The Hedgehog and Wnt signallingpathways in cancer. Nature 2001;411:349-54.

21. Kayed H, Kleeff J, Keleg S, Buchler MW, Friess H. Distribu-tion of Indian hedgehog and its receptors patched andsmoothened in human chronic pancreatitis. J Endocrinol2003;178:467-78.

22. Thayer SP, Pasca di Magliano MP, Heiser PW, Nielsen CM,Roberts DJ, Lauwers GY, et al. Hedgehog is an early andlate mediator of pancreatic cancer tumorigenesis. Nature2003;425:851-6.

23. Berman DM, Karhadkar SS, Maitra A, Montes DO, Gersten-blith MR, Briggs K, et al. Widespread requirement forHedgehog ligand stimulation in growth of digestive tract tu-mours. Nature 2003;425:846-51.

24. Ohuchida K, Mizumoto K, Fujita H, Yamaguchi H, KonomiH, Nagai E, et al. Sonic hedgehog is an early developmentalmarker of intraductal papillary mucinous neoplasms: clini-cal implications of mRNA levels in pancreatic juice.J Pathol 2006;210:42-8.

25. Hebrok M, Kim SK, Melton DA. Notochord repression ofendodermal Sonic hedgehog permits pancreas develop-ment. Genes Dev 1998;12:1705-13.

26. Kim GE, Bae HI, Park HU, Kuan SF, Crawley SC, Ho JJ, et al.Aberrant expression of MUC5AC and MUC6 gastric mucinsand sialyl Tn antigen in intraepithelial neoplasms of thepancreas. Gastroenterology 2002;123:1052-60.

27. Sessa F, Bonato M, Frigerio B, Capella C, Solcia E, Prat M,et al. Ductal cancers of the pancreas frequently expressmarkers of gastrointestinal epithelial cells. Gastroenterol-ogy 1990;98:1655-65.

28. Prasad NB, Biankin AV, Fukushima N, Maitra A, Dhara S, El-kahloun AG, et al. Gene expression profiles in pancreaticintraepithelial neoplasia reflect the effects of Hedgehog sig-naling on pancreatic ductal epithelial cells. Cancer Res2005;65:1619-26.

29. Nakashima H, Nakamura M, Yamaguchi H, Yamanaka N,Akiyoshi T, Koga K, et al. Nuclear factor-kappaB contributesto hedgehog signaling pathway activation through sonichedgehog induction in pancreatic cancer. Cancer Res2006;66:7041-9.

30. Strobel O, Dor Y, Alsina J, Stirman A, Lauwers G, Trainor A,et al. In vivo lineage tracing defines the role of acinar-to-ductal transdifferentiation in inflammatory ductal metapla-sia. Gastroenterology 2007;133:1999-2009.

31. Strobel O, Dor Y, Stirman A, Trainor A, Fern�andez-Del Cas-tillo C, Warshaw AL, et al. Beta cell transdifferentiation doesnot contribute to preneoplastic/metaplastic ductal lesionsof the pancreas by genetic lineage tracing in vivo. ProcNatl Acad Sci USA 2007;104:4419-24.

32. Morton JP, Mongeau ME, Klimstra DS, Morris JP, Lee YC,Kawaguchi Y, et al. Sonic hedgehog acts at multiple stagesduring pancreatic tumorigenesis. Proc Natl Acad Sci USA2007;104:5103-8.

33. Hingorani SR, Petricoin EF, Maitra A, Rajapakse V, King C,Jacobetz MA, et al. Preinvasive and invasive ductal pancre-atic cancer and its early detection in the mouse. CancerCell 2003;4:437-50.

34. Lauth M, Bergstrom A, Shimokawa T, Tostar U, Jin Q, Fen-drich V, et al. DYRK1B-dependent autocrine-to-paracrineshift of Hedgehog signaling by mutant RAS. Nat StructMol Biol 2010;17:718-25.

35. Nolan-Stevaux O, Lau J, Truitt ML, Chu GC, Hebrok M,Fern�andez-Zapico ME, et al. GLI1 is regulated throughSmoothened-independent mechanisms in neoplastic pan-creatic ducts and mediates PDAC cell survival and transfor-mation. Genes Dev 2009;23:24-36.

36. Jones S, Zhang X, Parsons DW, Lin JC, Leary RJ, AngenendtP, et al. Core signaling pathways in human pancreatic can-cers revealed by global genomic analyses. Science 2008;321:1801-6.

37. Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem cells,cancer, and cancer stem cells. Nature 2001;414:105-11.

38. Lee CJ, Dosch J, Simeone DM. Pancreatic cancer stem cells.J Clin Oncol 2008;26:2806-12.

39. Bissell MJ, Labarge MA. Context, tissue plasticity, and can-cer: are tumor stem cells also regulated by the microenvi-ronment? Cancer Cell 2005;7:17-23.

40. Li C, Heidt DG, Dalerba P, Burant CF, Zhang L, Adsay V,et al. Identification of pancreatic cancer stem cells. CancerRes 2007;67:1030-7.

41. Hermann PC, Huber SL, Herrler T, Aicher A, Ellwart JW,Guba M, et al. Distinct populations of cancer stem cells de-termine tumor growth and metastatic activity in humanpancreatic cancer. Cell Stem Cell 2007;1:313-23.

42. Feldmann G, Fendrich V, McGovern K, Bedja D, Bisht S, Al-varez H, et al. An orally bioavailable small-molecule inhibi-tor of Hedgehog signaling inhibits tumor initiation andmetastasis in pancreatic cancer. Mol Cancer Ther 2008;7:2725-35.

43. Feldmann G, Dhara S, Fendrich V, Bedja D, Beaty R, Mul-lendore M, et al. Blockade of hedgehog signaling inhibitspancreatic cancer invasion and metastases: a new paradigmfor combination therapy in solid cancers. Cancer Res 2007;65:2187-96.

44. Bailey JM, Mohr AM, Hollingsworth MA. Sonic hedgehogparacrine signaling regulates metastasis and lymphangio-genesis in pancreatic cancer. Oncogene 2009;28:3513-25.

45. Feldmann G, Habbe N, Dhara S, Bisht S, Alvarez H, Fen-drich V, et al. Hedgehog inhibition prolongs survival in agenetically engineered mouse model of pancreatic cancer.Gut 2008;57:1420-30.

46. Olive KP, Jacobetz MA, Davidson CJ, Gopinathan A, McIn-tyre D, Honess D, et al. Inhibition of hedgehog signalingenhances delivery of chemotherapy in a mouse model ofpancreatic cancer. Science 2009;324:1457-61.

SurgerySeptember 2012

S32 Rosow et al

47. Yauch RL, Gould SE, Scales SJ, Tang T, Tian H, Ahn CP,et al. A paracrine requirement for hedgehog signalling incancer. Nature 2008;455:406-10.

48. Xu XF, Guo CY, Liu J, Yang WJ, Xia YJ, Xu L, et al. Gli1maintains cell survival by up-regulating IGFBP6 and Bcl-2through promoter regions in parallel manner in pancreaticcancer cells. J Carcinog 2009;8:13.

49. Tian H, Callahan CA, DuPree KJ, Darbonne WC, Ahn CP,Scales SJ, et al. Hedgehog signaling is restricted to the stro-mal compartment during pancreatic carcinogenesis. ProcNatl Acad Sci USA 2009;106:4254-9.

50. Bailey JM, Swanson BJ, Hamada T, Eggers JP, Singh PK, Caff-ery T, et al. Sonic hedgehog promotes desmoplasia in pan-creatic cancer. Clin Cancer Res 2008;14:5995-6004.

51. Pola R, Ling LE, Silver M, Corbley MJ, Kearney M, Blake Pe-pinsky R, et al. The morphogen Sonic hedgehog is an indi-rect angiogenic agent upregulating two families ofangiogenic growth factors. Nat Med 2001;7:706-11.

52. Asai J, Takenaka H, Kusano KF, Ii M, Luedemann C, CurryC, et al. Topical sonic hedgehog gene therapy accelerateswound healing in diabetes by enhancing endothelial pro-genitor cell-mediated microvascular remodeling. Circula-tion 2006;113:2413-24.

53. Surace EM, Balaggan KS, Tessitore A, Mussolino C, Co-tugno G, Bonetti C, et al. Inhibition of ocular neovas-cularization by hedgehog blockade. Mol Ther 2006;13:573-9.

54. Chinchilla P, Xiao L, Kazanietz MG, Riobo NA. Hedgehogproteins activate pro-angiogenic responses in endothelialcells through non-canonical signaling pathways. Cell Cycle2010;9:570-9.

55. Renault MA, Roncalli J, Tongers J, Thorne T, Klyachko E,Misener S, et al. Sonic hedgehog induces angiogenesis viaRho kinase-dependent signaling in endothelial cells. J MolCell Cardiol 2010;49:490-8.

56. Nakamura K, Sasajima J, Mizukami Y, Sugiyama Y, YamazakiM, Fujii R, et al. Hedgehog promotes neovascularization inpancreatic cancers by regulating Ang-1 and IGF-1 expressionin bone-marrow derived pro-angiogenic cells. PLoS One2010;5:e8824.

57. Chen W, Tang T, Eastham-Anderson J, Dunlap D, Alicke B,Nannini N, et al. Canonical hedgehog signaling augmentstumor angiogenesis by induction of VEGF-A in stromal per-ivascular cells. Proc Natl Acad Sci USA 2011;108:9589-94.

58. Brown LF, Berse B, Jackman RW, Tognazzi K, Manseau EJ,Senger DR, et al. Expression of vascular permeability factor(vascular endothelial growth factor) and its receptors in ad-enocarcinomas of the gastrointestinal tract. Cancer Res1993;53:4727-35.

59. Itakura J, Ishiwata T, Friess H, Fujii H, Matsumoto Y, Buch-ler MW, et al. Enhanced expression of vascular endothelialgrowth factor in human pancreatic cancer correlates withlocal disease progression. Clin Cancer Res 1997;3:1309-16.

60. Binns W, James LF, Shupe JL, Everett G. A congenitalcyclopian-type malformation in lambs induced by maternalingestion of a range plant Veratrum californicum. Am J VetRes 1963;24:1164-75.

61. Lipinski RJ, Hutson PR, Hannam PW, Nydza RJ, Washing-ton IM, Moore RW, et al. Dose- and route-dependent terato-genicity, toxicity, and pharmacokinetic profiles of thehedgehog signaling antagonist cyclopamine in the mouse.Toxicol Scii 2008;104:189-97.

62. Mas C, Ruiz i Altaba A. Small molecule modulation of HH-GLI signaling: current leads, trials and tribulations. Bio-chem Pharmacol 2010;80:712-23.

63. Scales SJ, de Sauvage FJ. Mechanisms of Hedgehog pathwayactivation in cancer and implications for therapy. TrendsPharmacol Sci 2009;30:303-12.

64. LoRusso PM, Rudin CM, Reddy JC, Tibes R, Weiss GJ, BoradMJ, et al. Phase I trial of hedgehog pathway inhibitor vismo-degib (GDC-0449) in patients with refractory, locally ad-vanced or metastatic solid tumors. Clin Cancer Res 2011;17:2502-11.

65. Palmer SR, Erlichman C, Fern�andez-Zapico M, Qi Y, Al-mada L, McCleary-Wheeler A, et al. Phase I trial erlotinib,gemcitabine, and the hedgehog inhibitor, GDC-0449 (ab-stract). J Clin Oncol 2011;29(15 [May 20 Suppl]):3092.

66. Cheng H, Merika E, Syrigos KN, Saif MW. Novel agents forthe treatment of pancreatic adenocarcinoma. Highlightsfrom the ‘‘2011 ASCO Annual Meeting,’’ Chicago, IL,USA; June 3-7, 2011 (review). JOP 2011;12:334-8.

67. Rudin CM, Jimen A, Miller WH, Eigl BJ, Gettinger SN,Chang ALS, et al. A phase I study of IPI-926, a novelhedgehog pathway inhibitor, in patients (pts) with ad-vanced or metastatic solid tumors. J Clin Oncol 2011;29(15 Suppl):3014.

68. Stephenson J, Richards DA, Wolpin BM, Becerra C, HammJT, Messersmith WA, et al. The safety of IPI-926, a novelhedgehog pathway inhibitor, in combination with gemcita-bine in patients (pts) with metastatic pancreatic cancer.J Clin Oncol 2011;29(15 Suppl):4114.

69. RudinCM,HannCL, Laterra J, YauchRL, CallahanCA, FuL,et al. Treatment of medulloblastoma with hedgehog pathwayinhibitor GDC-0449. N Engl J Med 2009;361:1173-8.

70. Yauch RL, Dijkgraaf GJP, Alicke B, Januario T, Ahn CP, Hol-comb T, et al. Smoothened mutation confers resistance to aHedgehog pathway inhibitor in medulloblastoma. Science2009;326:572-4.