Progress in Cutaneous Cancer Research1

Progress in Cutaneous Cancer Research1 Andrzej Dlugosz, Glenn Merlino,n and Stuart H. Yuspaw Department of Dermatology and Comprehensive Cancer Center, University of Michigan School of Medicine, Ann Arbor, Michigan, U.S.A.; nLaboratory of Molecular Biology and wLaboratory of Cellular Carcinogenesis and Tumor Promotion, Center for Cancer Research, National Cancer Institute, Bethesda, Maryland, U.S.A.

Cutaneous cancers represent a major public health concern due to the very high incidence, associated medical costs, substantial mortality, and cosmetic deformities associated with treatment. Considerable progress in basic research has provided new insights into the underlying genetic basis of the major human cutaneous cancers, malignant melanoma, basal cell carcinoma, and squamous cell carcinoma. In turn, these genetic insights have illuminated biochemical pathways that promise to provide new approaches to the prevention and

treatment of cutaneous neoplasms. This review will detail the evolving genetic information and indicate how this information is being used to re¢ne experimental models that serve to both de¢ne the biochemistry of cancer pathogenesis and test novel approaches to cancer therapy. Combined with preventive measures to reduce exposure to sunlight, these advances are likely to reduce this major public health burden in the coming decade. Key words: melanoma/basal cell carcinoma/carcinogenesis/ skin. JID Symposium Proceedings 7:17 ^26, 2002

P

DNA and cause mutations, its capacity to activate signaling pathways that enhance selection of incipient neoplastic cells, and its activity as an immune suppressant (De Gruijl, 1999). Skin cancer prevention therefore should be achievable through education and lifestyle modi¢cations that reduce exposure to UV radiation. Although both government agencies and dermatologic societies have instituted approaches to educate the population on the dangers of sunlight exposure, societal experience indicates that such programs will have only limited success. Thus, the need for basic and applied research into the causes and potential nondeforming therapies for cutaneous cancers has never been more urgent. Although sun exposure is a major etiologic component of most skin cancers, other exogenous exposures also contribute to this epidemic of human neoplasia. As shown in Table I, lifestyle factors such as smoking and diet, occupational exposures, and certain topical or systemic medicinal therapies contribute to the overall incidence of both melanoma and nonmelanoma cancers (Black et al, 1994; Gallagher et al, 1996; De Hertog et al, 2001). Arsenic exposure, through occupational or environmental contact, also may contribute to the rising skin cancer rate (Schwartz, 1997).

erhaps no other dermatologic condition dominates a major area of human pathology as does cutaneous cancer. More than half of all cancers in North America occur on the skin, and this is likely to be an underestimate. This year more than 1 million nonmelanoma skin cancers will be reported in the U.S.A. Although the mortality from these cancers is relatively low, the magnitude of the incidence is so great that mortality from nonmelanoma skin cancers equals that of Hodgkin’s disease and uterine cancer. Furthermore, the health care costs, morbidity, and cosmetic defects resulting from current treatments make nonmelanoma skin cancers a major public health issue. The statistics for melanomas are even more discouraging. This tumor type is displaying the second largest increase in incidence among cancers in the American population, with over 50,000 cases expected in 2001 and a 7% mortality rate. As exposure to sunlight is the primary etiologic agent for all skin cancers, ultraviolet (UV) radiation must be considered the major carcinogen in the human environment. Ultraviolet radiation is a powerful carcinogenic stimulus by virtue of its ability to damage Manuscript received August 1, 2002; accepted for publication August 20, 2002. Reprint requests to: Dr. Stuart H. Yuspa, Laboratory of Cellular Carcinogenesis and Tumor Promotion, Center for Cancer Research, National Cancer Institute, 37 Convent Drive, MSC- 4255, Building 37, Room 3B25, Bethesda, MD 20892- 4255; Email: [email protected] Abbreviations: AK, actinic keratosis; APAF-1, apoptotic protease activating factor-1; BCC, basal cell carcinoma; CDK4, cyclin-dependent kinase 4; CMM, cutaneous malignant melanoma; EGFR, EGF receptor; HGF/SF, hepatocyte growth factor/scatter factor; LOH, loss of heterozygosity; NBCCS, nevoid basal cell carcinoma syndrome; RTK, receptor tyrosine kinases; TRAIL, TNF-related apoptosis-inducing ligand. 1 We have attempted to adhere to standard nomenclature guidelines (http://www.nature.com/ng/web_specials/nomen/nomen_guidelines.html) through most of the text. Human genes and proteins are indicated in upper case, with only the gene name italicized (e.g., PTCH1 and PTCH1). For mouse homologs, only the ¢rst letter in each is upper case (Ptch1 and Ptch1).

HEREDITARY CANCER SYNDROMES

Considerable insight into the genetic basis of sporadic skin cancers has come from the elucidation of speci¢c genes or genetic loci that de¢ne hereditary skin tumor syndromes (Table II) (Halpern and Altman, 1999). Perhaps the best de¢ned and most broadly relevant are the DNA repair genes that comprise the complementation groups of skin cancer prone xeroderma pigmentosum families (van Steeg and Kraemer, 1999). At least six independent genes, on distinct chromosomal loci, de¢ne proteins involved in nucleotide excision repair. Among these are proteins that recognize and bind to sites of DNA damage (XPA, XPC), helicases (XPB, XPD), and endonuclease components (XPG, XPF), defects in any of which give a skin cancer prone phenotype. Potential polymorphisms with functional conse-

0022-202X/02/$15.00 Copyright r 2002 by The Society for Investigative Dermatology, Inc.


17

18

DLUGOSZ ET AL

JID SYMPOSIUM PROCEEDINGS

Table I. Environmental agents associated with the development of human skin cancera Agent

Individual at risk

Route of exposure

Tumor types

UVA/UVB (sunlight) Cigarette smoke Soot Coat tar, pitch Petroleum oils Arsenic 4,40 -Bipridyl PCB Dry cleaning reagents Fiberglass Psoralen (PUVA) Nitrogen mustard Immunosuppressants Therapeutic x-ray

All Smokers Chimney sweep Coker of coal, steel worker Machinest, textile worker Agriculture worker, etc. Pesticide manufacturer Petrochemical worker Dry cleaners Insulators Psoriasis patient CTCL patient Transplant recipients, etc Cutaneous infections

T T or S T T T and S S and/or T T T or S T or S T T and S T S T

AK, BD, SCC, BCC, M SCC, BCC SCC SCC SCC BD, SCC, BCC SCC, BD M BCC BCC SCC, BCC, M SCC SCC, BCC BCC

a T, topical; S, systemic; SCC, squamous cell carcinoma; BD, Bowen’s disease; BCC, basal cell carcinoma; AK, actinic keratosis; PCB, polychlorinated biphenyl; M, melanoma, CTCL, cutaneous T-cell lymphoma.

Table II. Gene targets for mutations in hereditary and sporadic cutaneous cancers Gene

Function

Locus

Tumor type

Syndrome

Spontaneous

p53

DNA repair, apoptosis, cell cycle regulation DNA repair

17 p13.1

BCC, SCC

yes

3p25, 2q21 9q22.3, 19q13.3 16p13.3 -13, 13q22 9q22.3 7q31-32 9p21 7q34 3p22-p21.3

BCC, SCC, melanoma

Li Fraumeni (but no increase in skin cancers) xeroderma pigmentosum

possible

BCC, trichoepithelioma BCC melanoma, SCC, trichoepithelioma melanoma pilomatricoma

nevoid basal cell carcinoma ? dysplastic nevus ? ?

yes yes yes yes yes

cylindroma trichilemmoma sebaceous gland carcinoma

multiple cylindroma Cowden’s Muir-Torre

yes unknown unknown

trichoepithelioma BCC keratoacanthoma

multiple trichoepithelioma Bazex Ferguson-Smith

unknown unknown unknown

XPA, XPB XPC, XPD XPF, XPG PTCH1 SMO p16INK4a BRAF CTNNB (b-catenin) CYLD1 PTEN MSH2 MLH1 ? ? ?

Sonic hedgehog receptor Sonic hedgehog e¡ector cyclin inhibitor kinase cell^cell adhesion, transcription factor unknown phosphatase mismatch repair ? ? ?

16q12-13 10q23.3 2p22-p21 3p21.3 9p21 Xq24 -q27 9q31

quences in these and other DNA repair genes may contribute to susceptibility states in the general population as well (Wei et al, 1994). Chromosomal mapping studies in the basal cell nevus syndrome, coupled with genetic and functional studies of Drosophila development, revealed the Sonic hedgehog pathway, and speci¢cally mutations in the PTCH1 gene, as the basis for hereditary and many sporadic basal cell cancers (Bale and Yu, 2001). Likewise the mapping of the inheritance pattern of the dysplastic nevus syndrome focused attention on the INK4a locus and speci¢cally mutations in the p16INK4a gene in the etiology of heredity-prone and sporadic melanoma (Hussussian et al, 1994). Subsequently, defects in p16INK4a or other components of the cyclin-CDK signaling pathway have been associated with both melanoma and nonmelanoma skin cancer (Sou¢r et al, 1999). Detection of speci¢c mutations in Cowden’s syndrome (PTEN), Muir^Torre syndrome (MSH2, MLH1), pilomatricoma (CTNNB [b-catenin]), and trichoepithelioma (PTCH1, p16INK4aA) has illuminated the underlying pathways associated with adnexal tumors. The delineation of the speci¢c genes mutated in other syndromes where locus mapping is con¢rmed should reveal even more insight into the broader spectrum of skin neoplasms. CUTANEOUS MALIGNANT MELANOMA Etiology and genetics Cutaneous malignant melanoma (CMM) typically presents as a pigmented lesion that evolves

either from a pre-existing nevus or arises de novo in normalappearing skin. Compared with benign nevi, CMM frequently exhibit one or more of the following features (the ABCD of melanoma): asymmetry, border irregularity, color variegation, and diameter greater than 6 mm. CMM is notorious for its highly aggressive nature and its resistance to currently existing radiation and chemotherapeutic modalities. In fact, no standard therapy exists for patients with disseminated melanoma, who have a dismal prognosis. Recently, this potentially fatal disease has exhibited an alarming increase in incidence (Rigel et al, 1996), evolving into a considerable health crisis. Epidemiologic evidence now points to a causal role for exposure to the UV spectrum of sunlight in the etiology of CMM (IARC, 1992; Armstrong and Kricker, 1995), although the relative contributions of UVA (320^400 nm) and UVB (280^320 nm) are still disputed. Whereas UVA is thought to incite oxidative DNA damage, UVB can suppress cell-mediated immunity and induce the formation of characteristic pyrimidine dimers (De Gruijl, 2000). Notably, unlike the more common squamous cell carcinoma (SCC), which is associated with cumulative lifetime UV exposure, melanoma appears to be induced by intense, intermittent exposure to UV, particularly during childhood (Holman et al, 1983; Whiteman et al, 2001). In addition to sun exposure, other strong melanoma risk factors include the total number of nevi and dysplastic nevi, skin and hair color, and germline mutations in speci¢c tumor suppressor genes (Goldstein and Tucker, 1993; Marks, 2000).

VOL. 7, NO. 1 DECEMBER 2002

Normal Melanocyte

Dysplastic Nevus

PROGRESS IN CUTANEOUS CANCER RESEARCH

Radial Growth In Situ

Vertical Growth Invasive

Metastatic Melanoma

(del)9p21 p16INK4a (del)10 (del)6q (del)11q (del)1p +7q (+)EGFR (+)HGF (+)c-MET (-)APAF1 mut N-RAS mut BRAF

Figure 1. Genetic changes associated with melanoma progression. The multistage evolution of metastatic melanoma is depicted schematically with frequently associated stage-speci¢c genetic changes detailed below. Trisomy of 7q, detected in invasive vertical growth phase melanoma, would amplify the listed tyrosine kinase receptors (EGFR, c-MET) the ligand HGF, and BRAF, but a causal relation to progression is not proven. The loss of APAF-1 may occur through gene deletion or gene silencing.

Cytogenetic, linkage, and molecular analyses have provided compelling evidence for a strong underlying genetic basis for the genesis and progression of CMM (Chin et al, 1998) (Fig 1). With few exceptions, however, these same studies have failed to identify consistent melanoma-associated genetic alterations. One signi¢cant exception is the INK4a locus at 9p21, which encodes in alternative reading frames two distinct tumor suppressor genes, p16INK4a and p19ARF (Quelle et al, 1995). Germline mutations in the p16INK4a gene, a negative regulator of the cell cycle, and subsequent loss of heterozygosity (LOH) in arising melanoma, are frequently observed in CMM-prone kindreds (Hussussian et al, 1994; Kamb et al, 1994; FitzGerald et al, 1996). In sporadic CMM, INK4a mutations are less frequent, but functional loss of p16INK4a in tumors can also occur through epigenetic mechanisms, such as promoter hypermethylation (Gonzalgo et al, 1997; Funk et al, 1998). The signi¢cance of the pRB pathway, of which p16INK4a is a major regulator in melanomagenesis, can be gleaned from the identi¢cation of germline mutations in two additional melanoma-prone kindreds of cyclin-dependent kinase 4 (CDK4), a promoter of cell cycle progression and a target of p16INK4a inhibition (Zuo et al, 1996; Russo et al, 1998; Sou¢r et al, 1998). Interestingly, p53, situated at the regulatory nexus of cellular growth and apoptosis, and the most common mutational target in human cancer (Cariello et al, 1994), is infrequently mutated in primary melanoma (Zerp et al, 1999; Bardeesy et al, 2001;and references within). It has been postulated that this rarity can be at least partially accounted for by the positioning of p19ARF, often inactivated in melanoma as a component of the INK4a locus, upstream of p53 (Chin et al, 1998). Such a con¢guration would render inactivating mutations in both genes redundant with respect to cell cycle regulation. A second exception where speci¢c genetic alterations have been clearly associated with CMM is in the MAPK signaling pathway. Although activating mutations in members of the RAS proto-oncogene family are among the most common in human cancer, demonstrating a causal role in CMM has proven elusive; however, more recent molecular analysis of primary and metastatic melanoma has implicated N-RAS and H-RAS mutations in CMM progression (Chin et al, 1997). More signi¢cantly, a recent report has identi¢ed activating mutations in BRAF, which acts downstream of Ras, in almost 70% of human CMM (Davies et al, 2002). Point mutations with characteristic UV-associated signatures have been found in genes situated within those molecular pathways described above, including p16INK4a (Kamb et al, 1994; Pollock et al, 1995), N-RAS (van’t Veer et al, 1989; van Elsas et al, 1997; Jiveskog et al, 1998), and p53 (Zerp et al, 1999), but not ARF (Peris et al, 1999). These ¢ndings suggest that such critical genes can be the target of both hereditary and environmental insults.

19

Recent epidemiologic data indicate that INK4a-associated melanoma-prone kindreds in geographic areas with the greatest exposure to sunlight are also at highest risk for melanoma, linking UV irradiation to the INK4a locus (Bishop et al, 2002). Therefore, assessing the consequences of exposure to combinations of such risk factors represents an area of great experimental opportunity; however, at this time functional relationships between candidate genes and environmental factors in melanomagenesis are largely unknown. Other classes of molecules have been implicated in melanomagenesis as well. Receptor tyrosine kinases (RTK), critical modulators of virtually all fundamental cellular behavior, including growth, di¡erentiation, motility, and survival, play signi¢cant roles in normal melanocyte development and function (Bennett, 1993; Halaban, 1996). These include c-Kit (Witte, 1990), c-Met (Halaban et al, 1992; Takayama et al, 1996; Kos et al, 1999), and the platelet-derived growth factor receptor (Stephenson et al, 1991; Soriano, 1997). Multiple reports have either supported or refuted the role of speci¢c RTK in primary and metastatic melanoma cells and their cultured derivatives (Albino, 1992; Shih and Herlyn, 1994; Halaban, 1996). Of central importance to the aspiring melanoma cell is the acquisition of autonomous growth control through the creation of autocrine RTK signaling loops, i.e., a cell is able to manufacture both an RTK and its associated ligand. Examples of such loops include basic ¢broblast growth factor (bFGF)-FGF receptor, considered a hallmark of CMM development, and transforming growth factor a (TGFa)-epidermal growth factor (EGF) receptor. Recently, progress has been made in identifying new candidate therapeutic targets associated with melanoma cell survival. Soengas et al (2001) demonstrated that melanoma cells avoid apoptotic destruction through inactivation of apoptotic protease activating factor-1 (APAF-1), a requisite caspase-9 activator functioning downstream of p53. Common APAF-1 loss of function, which can occur by allelic loss and methylation silencing, also helps account for the relative rarity of p53 mutations in CMM. Signi¢cantly, restoration of APAF-1 pro-apoptotic function by the methylation inhibitor 5aza2dC rendered melanoma cells chemosensitive (Soengas et al, 2001). In a related development, Gri⁄th et al (2000) showed that adenovirus-mediated expression of TNF-related apoptosisinducing ligand (TRAIL), which induces caspase-8-associated apoptosis, can serve as gene therapy in the destruction of melanoma cells. Experimental models of CMM A major impediment to the study of melanoma has been the lack of relevant, tractable experimental animal models. Historically, the animal models studied in most detail have been the Xiphophorus hybrid ¢sh and the South American opossum, as well as a number of rodent models (Kusewitt and Ley, 1996) (Table III). The problem with all such models, however, is that arising melanocytic tumors do not resemble human CMM at the histopathologic level, and extensive genetic manipulation is not possible. The mouse represents an especially attractive system because of the availability of an extensive genetic base upon which to build. Unfortunately, murine melanocytes, unlike those in human skin, are typically con¢ned to hair follicles and are exceedingly resistant to both spontaneous and UV-induced melanoma. Moreover, melanomas that arise do so with low penetrance, long latencies, and poor metastatic capacity. Genetically engineered mouse models have proven to be amenable to the genetic dissection of molecular pathways in tumorigenesis, and a number of melanoma-prone, transgenic mice have recently been described (Satyamoorthy et al, 1999; Tietze and Chin, 2000). Among these, melanoma induction in transgenic mice has been achieved through melanocytic expression of the oncogenes SV40 T-antigen, c-RET and HRAS (Bradl et al, 1991; Klein-Szanto et al, 1994; Kato et al, 1998; Bardeesy et al, 2001). The success of these models can be credited, in part, to earlier seminal work identifying candidate oncogenic

20

DLUGOSZ ET AL

JID SYMPOSIUM PROCEEDINGS

Table III. Experimental models for melanoma induction Exogenous inducers Agent

Species

Comments

DMBA/TPA DMBA UVR (280- 400 nm) DMBA/UVR

Syrian golden hamster guinea pig monodelphus domestica human skin grafted to nude mice

requires transplantation frequent metastasis 100 week exposure carcinoma and melanoma after repeated exposure

Heritable models Species

Enhanders

Comments

Xiphorphorus Sinclair swine

UVR 2 loci

overexpression of RTK-ONC-Xmrk often regress with vitiligo Genetic modi¢ed mouse models

Modi¢cations

Enhanders

Comments

Tyr-Tag TP-ras MT-RET MT-HGF Tyr-RAS Ink4an/ARF D2,3 Cdk4 þ/ R24C

UVR UVR or DMBA required

mostly ocular, cutaneous requires UVR metastatic melanosis; metastatic melanosis; metastatic non-metastatic metastatic knock in mutant; invasive

UVR INK4a/, p53/ DMBA DMBA and TPA

pathways in human CMM: T-antigen disrupts pRB and p53 function in a fashion analogous to loss of both p16INK4a and p19ARF at the INK4a locus; c-RET, although not normally expressed in melanocytes, is a potent RTK whose presence would unbalance multiple kinase signaling pathways; and the potential role of activated RAS genes has already been noted. In an elegant set of genetic experiments, the activated H-RAS transgene was placed on a background de¢cient in Ink4a resulting in the e⁄cient development of spontaneous nonmetastatic melanoma (Chin et al, 1997). Moreover, when the H-RAS transgene was con¢gured with a tetracycline-regulatable promoter, it could be demonstrated that activated H-RAS was required for melanoma maintenance in this model (Chin et al, 1999). Recently, mice bearing p16Ink4a-speci¢c inactivating mutations, or the Cdk4 R24C mutation described in patients with familial CMM, were reported to be susceptible to DMBAinduced melanoma (Krimpenfort et al, 2001; Sharpless et al, 2001; Sotillo et al, 2001; Rane et al, 2002), strongly supporting a role for the p16INK4a/Cdk4/Rb pathway in melanomagenesis. As in other animal models, however, murine melanocytic neoplasms arising in these various transgenic mice do not closely resemble human lesions, in that they arise within the dermis and lack the epidermal component characteristic of conventional human CMM. In melanoma-prone transgenic mice overexpressing hepatocyte growth factor/scatter factor (HGF/SF), the c-Met ligand, metastasis occurs in approximately 15% of tumor-bearing animals (Takayama et al, 1996; 1997; Otsuka et al, 1998). Unlike other genetically engineered animals, melanocytes in HGF/SF transgenic mice demonstrate extra-follicular survival, residing within the epidermis, dermis, and epidermal^dermal junction. The resulting ‘‘humanized’’ skin has the potential to yield melanomas with a histopathologic pro¢le reminiscent of human CMM (Noonan et al, 2001). As an alternative to making mouse skin morphologically resemble human, human/mouse chimeras have been created in which human skin is grafted onto immunode¢cient mice, and then subjected to experimental analysis (Satyamoorthy et al, 1999). Such chimeras provide an orthotopic milieu in which to study spontaneous and UVinduced melanomagenesis in vivo. In fact, UVB exposure was capable of inducing melanocytic lesions, including a nodular melanoma, in xenografted human skin in concert with either

DMBA treatment or adenovirus-mediated bFGF overexpression (Atillasoy et al, 1998; Berking et al, 2001). The ability to evaluate the consequences of UV irradiation on melanoma development is a paramount feature in any animal model, as exposure to sunlight is thought to be a causal agent in up to 80% of CMM. With respect to genetically tractable transgenic mouse models of melanoma, some demonstrate UV sensitivity, although responses have been relatively ine⁄cient (Kelsall and Mintz, 1998; Broome et al, 1999). The extra-follicular melanocytes in the HGF/SF transgenic skin, however, appear to be highly susceptible to neonatal UV irradiation (Noonan et al, 2000; 2001), linking both the power of genetic manipulation and the relevance of environmental challenge within the same model system. Taken together, these advances brighten future prospects of creating relevant mouse models to rigorously assess genetic and environmental melanoma risk factors, facilitate development of e⁄cacious sun protection strategies, and establish e¡ectual antimelanoma therapeutics. BASAL CELL CARCINOMA (BCC) Etiology and genetics BCC is a very common, slow-growing, locally invasive tumor that typically presents as a pink or pearly papule with super¢cial telangiectasia and occasional ulceration. There are three clinical variants: nodular (the most common), super¢cial, and sclerosing. In contrast to cutaneous SCC, BCC precursor lesions have not been identi¢ed, there is no evidence of neoplastic progression, and metastases are exceedingly rare (Miller, 1995). BCC are probably derived from hair follicles, and analogous to hair follicle epithelium, BCC growth is dependent on proper signaling between neoplastic keratinocytes and surrounding mesenchymal cells. This may account, in part, for the low incidence of BCC metastases, as well as the di⁄culty in establishing and maintaining BCC as xenografts in vivo or as immortalized cell lines in vitro. Chromosomal losses involving 9q have been found in both sporadic and inherited BCC. The latter tumors occur in patients with the autosomal dominant disorder nevoid basal cell carcinoma syndrome (NBCCS), which is characterized by a predisposition to BCC (frequently multiple and appearing at an early age), an increased incidence of several other tumors, and a

VOL. 7, NO. 1 DECEMBER 2002

variety of developmental anomalies (Gorlin, 1987). These clinical features suggested that the gene involved in BCC formation is also important during embryogenesis, and this notion was con¢rmed with the discovery in NBCCS patients of germline mutations of PTCH1 (Hahn et al, 1996; Johnson et al, 1996), a homolog of the Drosophila ptc gene involved in embryonic development. BCC from NBCCS patients had lost the remaining normal PTCH1 allele, which was also found to be de¢cient in spontaneous BCC, suggesting that PTCH1 is a tumor suppressor. A second PTCH gene has subsequently been identi¢ed (PTCH2) (Zaphiropoulos et al, 1999), but its role in BCC development is not yet known. Ptch1 is a 12-pass transmembrane molecule that functions as a receptor for Sonic hedgehog (Shh), a secreted ligand that regulates proliferation and patterning of multiple tissues and organs during embryogenesis (Chuang and Kornberg, 2000). According to the current model (Fig 2), Ptch1 normally antagonizes signaling activity by repressing Smoothened (Smo), a cell-surface molecule with homology to G protein-coupled receptors. Shh initiates signaling in responsive cell types by inhibiting Ptch, resulting in derepression of Smo and, ultimately, the activation of Shh target genes. Two ‘‘universal’’ Shh target genes are Gli1 (which encodes a transcription factor) and Ptch1, and the expression level of these transcripts has proven to be a reliable indicator of physiologic and pathologic Shh signaling. Under normal conditions, activation of this pathway is dependent on Shh, whose expression is tightly regulated both in space and in time. In human BCC, however, loss of PTCH1 results in constitutive signaling that is irreversible and independent of SHH. In addition to loss-of-function PTCH1 mutations, gain-of-function (oncogenic) SMO mutations have been found in some BCC where PTCH1 appears to be normal (Xie et al, 1998; Lam et al, 1999). Together, PTCH1 or SMO mutations have been identi¢ed in less than 75% of BCC whereas hedgehog target genes are upregulated in essentially all tumors examined (Dahmane et al, 1997), indicating that additional mechanisms must exist for uncontrolled activation of this pathway. The current data suggest that constitutive Shh signaling, regardless of how this is brought about, plays a central role in the genesis of BCC. PTCH1 gene alterations have also been found in trichoepitheliomas (Vorechovsky et al, 1997) and nevus sebaceous of Jadassohn (Xin et al, 1999), suggesting that deregulation of Shh signaling plays a role in the genesis of other follicle-derived tumors. Although mutations in p53 have been found in up to 50% of BCC, their involvement in the development or maintenance of these tumors is not known, particularly as most BCC fail to exhibit the genomic instability associated with other cancers where p53 function is compromised. What is the normal function of Shh signaling in skin? Analysis of genetically engineered mutant mice has revealed a crucial requirement for Shh during hair follicle growth and morphogenesis, but not terminal di¡erentiation (St Jacques et al, 1998; Chiang et al, 1999). In addition, both loss-of-function (Wang et al, 2000) and gain-of-function (Sato et al, 1999) studies implicate Shh signaling in the regulation of hair follicle growth during the anagen phase of the hair cycle. Thus, whereas physiologic Shh signaling in skin may govern growth of follicle keratinocytes at appropriate times, uncontrolled Shh signaling due to PTCH1 or SMO mutations may cause sustained proliferation, resulting in the development of follicle-derived tumors. How does stimulation of Shh signaling alter cell function? Transcriptional responses in the highly homologous Drosophila hedgehog pathway are mediated by ci (reviewed in Aza-Blanc and Kornberg, 1999), which belongs to the Gli family of zinc ¢ngercontaining transcription factors. Vertebrate ci homologs include Gli1, Gli2, and Gli3, all of which bind to the consensus DNA sequence GACCACCCA (Matize and Joyner, 1999). In contrast to Gli3, which appears to function primarily as a transcriptional repressor, both Gli1 and Gli2 are transcriptional activators. Major

PROGRESS IN CUTANEOUS CANCER RESEARCH

Physiologic Activation in Hair Follicles (Reversible)

Resting State

SHH SHH

PTCH1

X

SMO repressed

Pathologic Activation in BCC (Irreversible) PTCH1 or loss-of-function

PTCH1 SMO repressed active GLI1,2

21

SMO active GLI1,2

SMO gain-of-function

PTCH1

SMO active

GLI1,2

Figure 2. Proposed model depicting the molecular basis of BCC. In contrast to the multistep evolution of SCC and melanoma, deregulation of the SHH signaling pathway may be su⁄cient for BCC development. During physiologic activation in responsive cell types, SHH binds and inhibits PTCH1, which normally represses SMO. Derepression of SMO results in transient activation of SHH target genes, including PTCH1, GLI1, and cell type-speci¢c genes, which are likely to play an important role in growth control. In BCC, mutations involving PTCH1 or SMO result in uncontrolled signaling and constitutive expression of SHH target genes.

emphasis has been placed on Gli1 as a potential e¡ector of normal and constitutive Shh signaling for several reasons: it is consistently upregulated in cells where the Shh pathway is active, both during embryogenesis and in neoplasms; ectopic Gli1 expression can mimic responses to Shh in certain settings; and when overexpressed in frog skin, GLI1 can give rise to primitive skin tumors (Dahmane et al, 1997). Despite these ¢ndings Gli1 knockout mice are phenotypically normal (Morris and Potten, 1999), arguing against an essential role for this Gli protein in physiologic Shh signaling, whereas Gli2 knockout mice exhibit abnormalities in multiple organs whose development is dependent on Shh. In particular, loss of Gli2 function results in severe impairment of hair follicle growth (Grachtchouk et al, 2000a). Experimental models of BCC Rats exposed to chemical carcinogens (MCA, DMBA) or ionizing radiation preferentially develop BCC rather than squamous tumors, and frequently at a high incidence (reviewed in Zackheim, 1985;Table IV). Rat BCC appear to arise in the outer root sheath of hair follicles and have a slow growth rate, similar to human BCC. The molecular basis for experimentally induced BCC in rats is not yet known, but is expected to result in Shh pathway activation if the current model is accurate. In striking contrast to their strong predisposition to squamous tumor development, mice appear to be remarkably resistant to BCC induction. One notable exception is the chemical carcinogen dehydroretronecine, which is e¡ective at inducing mouse BCC (and other tumors) following subcutaneous or topical administration (Johnson et al, 1978). The discovery of inactivating PTCH1 mutations in BCC fueled the development of a number of transgenic and knockout mouse models exploring the potential role of deregulated Shh signaling in BCC tumorigenesis. Overexpression of SHH using a K14 promoter resulted in upregulation of Shh target genes and development of basal cell-like proliferations in newborn mouse skin (Oro et al, 1997), with similar results obtained using the K5 promoter to drive expression of a gain-of-function SMO mutant, M2SMO (Xie et al, 1998). Overexpression of SHH in human keratinocytes followed by grafting onto SCID mice resulted in development of BCC-like changes as well (Fan et al, 1997). These studies supported the hypothesis that constitutive activation of Shh signaling in keratinocytes is su⁄cient for BCC development, but analysis of tumor phenotypes in adult transgenic mice could not be performed due to impaired viability of these animals. When K14-SHH mouse skin was transplanted onto SCID mouse hosts, BCC-like proliferations were apparently replaced by well-di¡erentiated hair-follicle-like structures (Oro et al, 1997), supporting the proposed requirement for an appropriate tumor stroma to maintain BCC growth. Mouse models have also been developed in which Ptch gene

22

DLUGOSZ ET AL

JID SYMPOSIUM PROCEEDINGS

Table IV. Experimental models for basal cell carcinoma Exogenous inducers Agent

Species

Comments

MCA DMBA

Rat Rat

other skin tumors SCC 4 BCC; lower DMBA doses may increase BCC incidence other skin tumors BCC 4 SCC, also internal tumors

Ionizing radiation Rat Dehydroretronecine Mouse

Genetic models Modi¢cation

Ehancers

Comments

K14-SHH LTR-SHH K5-M2SMO Ptch þ / K5-GLI1 K5-Gli2

? ? ? UV, IR, Arsenic ? ?

lethal phenotype, not transplantable super¢cial BCC impaired survival other tumor types other tumor types exclusively BCC

function has been disrupted (Goodrich et al, 1997; Hahn et al, 1998), and Ptch þ /^ mice have many features in common with NBCCS patients. Detailed analysis of skin from PtchlacZ/ þ mice has revealed microscopic hair-follicle-derived proliferations, with the appearance of a variety of macroscopic skin tumors, including BCC, following exposure to ionizing or UV radiation (Aszterbaum et al, 1999). Taken together, these ¢ndings strongly support the concept that deregulated Shh signaling plays a central role in BCC development. Other transgenic mouse studies have focused on Gli proteins as potential mediators of constitutively activated Shh signaling in BCC. GLI1 and Gli2 have both been overexpressed in mouse skin using the same bovine K5 promoter with intriguingly di¡erent results. K5-GLI1 transgenic mice developed a variety of follicle-derived tumor types with relatively few BCC (Nilsson et al 2000), whereas K5-Gli2 transgenic mice developed only BCC (Grachtchouk et al 2000b). These ¢ndings, coupled with the results of knockout mouse studies described above, strongly implicate Gli2 both in physiologic (hair follicle growth) and pathologic (BCC development) Shh signaling in skin. Despite the compelling phenotypes produced in these mouse models, there is as yet no direct evidence that Gli transcription factors are required for tumorigenesis associated with PTCH1 or SMO mutations. As it appears that deregulation of Shh signaling is su⁄cient for BCC development, these tumors may be uniquely responsive to mechanism-based therapeutic intervention. The steroidal alkaloid cyclopamine has been shown to inhibit Shh signaling by blocking SMO (Taipale et al, 2000), but it is not yet known whether BCC could be successfully treated, or their appearance prevented, using this agent. In September 2001, one smallmolecule inhibitor of Shh signaling had been granted FDA approval for phase I clinical trials in patients with BCC. Although this type of work is just getting under way, it may eventually lead to novel nonsurgical approaches to treating BCC and possibly other cancers associated with deregulated Shh signaling, such as medulloblastomas and a subset of rhabdomyosarcomas. CUTANEOUS SCC Etiology and genetics Cutaneous SCC frequently presents as a ¢rm, pink papule or nodule, with a conspicuous hyperkeratotic surface. Although they represent only about 20% of nonmelanoma skin cancers, SCC are generally more aggressive and occasionally lethal. SCC is more frequent with higher cumulative sunlight exposure, as cancers associated with

Figure 3. Genetic changes associated with human cutaneous SCC. The multistage evolution of invasive SCC is depicted schematically with frequently associated genetic changes detailed below. Single base mutations in early lesions frequently are characteristic of UV light-induced damage, whereas later changes are associated with genomic instability. Increased activity of telomerase (deletion of inhibitor) or EGFR tyrosine kinase (gene ampli¢cation) may also result from epigenetic changes.

occupational exposures, and in immunosuppressed patients. Hereditary syndromes uniquely associated with SCC risk have not been described although DNA repair defects in xeroderma pigmentosum substantially increase the risk for SCC development. Insight into the pathogenesis of cutaneous SCC has come from studies of the frequent precursor lesion, actinic keratosis (AK) (Fig 3). These benign hyperproliferativehyperkeratotic lesions frequently have sunlight-induced clonal p53 mutations suggesting clonal expansion from a single cell carrying a speci¢c p53 mutation. Most frequently, these mutations are in codon 278 or other codons of the DNAbinding domain of p53 that contain dipyrimidine sites (Hussain and Harris, 1998). Of particular interest is the common ¢nding of LOH at particular chromosomal sites such as 13q, 17p, 17q, 9p, and 9q (Rehman et al, 1994). Frequently, multiple sites of allelic loss are detected in the same lesion. As actinic keratoses progress to SCC at an extremely low frequency, the challenge is to determine which of these genetic lesions, if any, contributes to premalignant progression. Analyses of SCC have also revealed frequent LOH and p53 mutations, but a modal allelic loss or speci¢c p53 codon strongly associated with the acquisition of a malignant phenotype is yet to be identi¢ed (Quinn et al, 1994). Other genetic changes have been associated with progression in SCC, and their possible involvement in pathogenesis has been explored in experimental studies. Mutations in the K-RAS and Ha-RAS gene are detected in both AK and SCC, and activation of the RAS pathway through mutation of the gene or growth factor stimulation may be extremely common in squamous tumors, particularly in sites with intense exposure to UV radiation (Pierceall et al, 1991; Kreimer-Erlacher et al, 2001). Inactivating mutations or epigenetic silencing of p16INK4a and activation of telomerase are other pathways associated with SCC development (Taylor et al, 1996; Sou¢r et al, 1999). Constitutive activation of the EGF receptor (EGFR) by ampli¢cation or expression of ligands with the formation of an autocrine loop is a frequent ¢nding in SCC (Moghal and Sternberg, 1999). Although these correlations have provided clues to pathways involved in SCC pathogenesis, de¢nitive causal associations in human SCC have not yet been con¢rmed. For this reason, model systems utilizing human and mouse keratinocytes in culture and animal models in vivo have been developed and utilized to test causal relationships at the molecular, cellular, and organism levels. Experimental models of cutaneous SCC The induction of squamous cell cancers on mouse skin by chemical carcinogens or UV light has been an excellent model to study cancer pathogenesis in general and skin cancer development in particular. These models have shown remarkable phenotypic homology to human SCC development and have provided additional information on the contribution of speci¢c genetic changes to particular stages of tumor progression. The addition of genetically modi¢ed mice to the mix of models available has

VOL. 7, NO. 1 DECEMBER 2002

PROGRESS IN CUTANEOUS CANCER RESEARCH

23

TableV. Experimental models for squamous cell tumor induction Exogenous inducers Agent

Species

Comments

(PAH, NA, AA) þ promoter (PAH, NA, AA) repeated ultraviolet B ultraviolet A

mouse, rat mouse, rat mouse mouse

predominantly papilloma predominantly SCC p53 mutations, immune suppression, SCC papillomas and SCC

Gentically modi¢ed mouse models Modi¢cation

Enhancers

Comments

Tg.AC (z globin-v-rasHa) Tg.AC K1-ras, K10-ras DK5-ras K6-ras K1-TGFa, K14-TGFa, MT-TGFa Inv-c-MycER K1-v-fos K5-E2F1 K5-Igf1 K5-ErbB2 K5-SOS-F K14-HPV16 K6-ODC XP mutant models (A, C, D) Egfr null mutant p53 null mutant p21waf1 null mutant c-fos null mutant K14-PKCe K14-PKCd

promoters, drugs UVB promoters

enhancers upregulate transgene p53 mutations are absent predominantly papillomas papillomas, KA, SCC SCC predominantly papillomas that regress papilloma papilloma papilloma, SCC, BCC papilloma, SCC papilloma, SCC tumors inhibited by Egfr de¢ciency tumors inhibited by di£uoromethylornithine SCC, K-ras mutations enhanced sensitivity reduced tumor size enhanced malignant conversion enhanced papilloma formation papilloma but no SCC enhanced SCC, metastases reduced papilloma development

promoters promoters promoters p53 de¢ciency promoters promoters FVB/N mouse strain DMBA initiation/promotion/UVR v-rasHa DMBA/TPA DMBA/TPA cross with Tg.AC DMBA/TPA DMBA/TPA

further clari¢ed the speci¢c requirements for particular genes and their downstream pathways in benign and malignant tumor formation (Table V). Together with in vitro analysis of keratinocytes, the biochemistry of SCC development is being revealed. Animal model studies indicate that heterozygous activating Ras gene mutations are su⁄cient to induce a benign squamous lesion, and this is coupled to constitutive activation of the EGFR (Yuspa, 1994) (Fig 4). Activation of the keratinocyte EGFR through transgenic targeting of TGFa to the epidermis can also produce a benign tumor phenotype in the absence of Ras mutations, but tumors regressed unless subjected to continuous exposure to tumor promoters (Vassar et al, 1992; Dominey et al, 1993; Jhappan et al, 1994). This suggests that activation of the EGFR is not su⁄cient for autonomous tumor formation, and alterations in other pathways are required. Under conditions of high expression or homozygosity of a mutant Ras gene, progression to malignancy is enhanced (Quintanilla et al, 1986; Greenhalgh and Yuspa, 1988), suggesting that the Ras pathway can recruit additional changes required for progression. The target cell for Ras activation in the epidermis may also determine the tumor phenotype as transgenic targeting of oncogenic Ras to keratinocytes committed to the di¡erentiation program produces terminally benign tumors whereas targeting to less di¡erentiated cells permits progression to SCC (Brown et al, 1998). In contrast, suprabasal targeting of c-Myc produces the papilloma phenotype, possibly through reducing apoptosis, whereas basal cell targeting of c-Myc is not oncogenic (Pelengaris et al, 1999; Waikel et al, 2001). Deletion of the cyclin/ CDK inhibitor p21waf1, a downstream e¡ector of p53, increases the number of benign tumors but does not in£uence the rate of premalignant progression (Missero et al, 1996; Weinberg et al, 1999). In contrast, p53 depletion enhances malignant progression but does not increase benign tumor formation (Kemp et al, 1993). These results suggest that the p53 pathway involving cell cycle

Figure 4. Genetic changes associated with chemically induced mouse cutaneous SCC. The multistage evolution of anaplastic or spindle cell tumors in this model is highly ordered both temporally and genetically. Ras mutations are characteristic of chemical mutagens used to initiate tumor formation. Early upregulation of cyclin D1 and later up-regulation of TGFa1 occur through epigenetic mechanisms and appear to be important components of carcinogenesis.

inhibition through p21waf1 does not determine the risk for malignant conversion, but another p53 regulated pathway is critical for suppression of premalignant progression. TGFb plays a dual role in experimental SCC development, suppressing premalignant progression while enhancing phenotypic progression from SCC to a spindle cell phenotype (Glick et al, 1994; Portella et al, 1998; Go et al, 1999). Members of the AP-1 transcription factor family also play a dual role in experimental skin tumor development, where c-June is essential for papilloma and c-Fos is essential for SCC development (Saez et al, 1995; Young et al, 1999). Other pathways now implicated in SCC development and progression to spindle cell tumors from experimental studies are cyclin D1, ornithine decarboxylase, p16ink4A, p15ink4A, and E-cadherin (Navarro et al, 1991; Cli¡ord

24

DLUGOSZ ET AL

JID SYMPOSIUM PROCEEDINGS

et al, 1995; Linardopoulos et al, 1995; Robles et al, 1998). These pathways are frequently altered in human SCC, but their contribution to pathogenesis remains to be proven by studies that directly transform human keratinocytes in an in vivo setting. Mouse models promise to reveal another essential aspect of skin cancer pathogenesis, that of individual susceptibility. Inbred mouse strains di¡er in susceptibility to particular exposures by several orders of magnitude (DiGiovanni et al, 1992). Carcinogenesis studies on cultured mouse keratinocytes derived from di¡erent background strains or on skin from speci¢c strains grafted to nude mice indicate that sensitivity is determined by the target tissue rather than systemically (Yuspa and Morgan, 1981; Yuspa et al, 1982; Glick et al, 1999). Tumor induction on F1 hybrid backcrosses between sensitive and resistant strains followed by genome scans using microsatellite markers has revealed that determinants for susceptibility or resistance are multigenic and distinct for benign tumor formation or premalignant progression and malignant conversion (Nagase et al, 1995; Mock et al, 1998). Analysis of congenic strains and other approaches indicate that speci¢c loci are epistatic, and the interactions are speci¢c for benign or malignant tumor formation (Nagase et al, 2001). Genetic loci also determine the survival potential for tumorbearing animals (Nagase et al, 1999). As the speci¢c genes are identi¢ed for these determinants, syntenic sites in the human genome that determine skin cancer susceptibility may be revealed. CONCLUSION

The rapidly accumulating genetic details of cutaneous cancer pathogenesis combined with the illumination of interacting signaling pathways downstream from the genes involved provides an opportunity to model the biochemistry of cutaneous tumors, a scheme that must ultimately be understood for rational therapy or medical intervention. Currently any biochemical model would almost certainly be incomplete. Alternatively, a general biologic scheme that may ¢t the multistage development of both melanoma and squamous cell skin cancers might be constructed with current information. The scheme has value as an organizational foundation to focus on emerging understanding of the biochemistry of the individual biologic processes and to test the biochemical principles in cutaneous cancer models. Such a scheme (Fig 5) assigns aberrations in control of proliferation, normal di¡erentiation, apoptosis or senescence, or developmental processes (in the case of some hereditary syndromes) as early events. As a consequence of these changes or coupled to p53 mutations, resistance to terminal cell death produces a survival advantage for the incipient cancer cells. Under selective pressure (cytotoxicity from sunlight or environmental chemicals being the most obvious exogenous sources), clonal expansion produces a clinically apparent benign lesion (AK) or dysplastic nevus. Premalignant progression is associated with genomic instability that could be inherent in a lesion with enhanced survival and a high proliferation rate. Nevertheless, other important changes must occur in this growing lesion that could be modulated by the host response. An in£ammatory reaction could deposit cytokines in the lesional environment as well as mutagenic reactive oxygen or nitrogen species (Smith et al, 1998; Hussain et al, 2000; Fitzpatrick, 2001). Angiogenesis and immune surveillance may play opposing roles or systemic immunosuppression may act in concert with angiogenesis to enhance progression (Prehn and Prehn, 1996; Smith-McCune et al, 1997). Many cell generations may pass as the positive and negative in£uences of the organismal response interact with the endogenous activities of the neoplastic lesion. Subsequent clones evolve and subclones from these as further survival advantage is acquired and selected. Alternatively, lethal genetic aberrations may arise and account for the spontaneous regression of some AK or melanotic lesions. The tumor environment evolves concomitant with clonal selection, altering the interactions of stromal, epithelial, and in£ammatory components and the tumor matrix itself. It is this balance that ¢nally allows

Figure 5. Stage-speci¢c biologic changes associated with multistage cutaneous carcinogenesis. This scheme is proposed to focus on processes that may be common for multistage tumor development where precise biochemical pathways are still under study. Early events re£ect changes occuring in incipient tumor cells whereas progression incorporates processes at the tissue and organismal level that must be addressed to fully comprehend cutaneous cancer pathogenesis.

for the invasive properties of tumor cells to be displayed. Thus, we see experimental evidence for stromal and matrix determinants of tumor cell behavior (Arias, 2001; Liotta and Kohn, 2001). The advantage the host has over the tumor cell is time. The opportunity for intervention is long. Our goal, in addition to education and lifestyle change, should be reduction in mortality, morbidity, and deformity. The knowledge to achieve this goal is rapidly evolving. Cutaneous cancer intervention and therapy is ideally suited to showcase the opportunities for rational approaches to cancer control and cancer cure. Due to the breadth of this review, we apologize for the unavoidable exclusion of references to work done by many outstanding investigators working in these areas.We wish to acknowledge the contribution of Dr. Luowei Li to the construction of the ¢gures and Ms. Bettie Sugar for outstanding editorial assistance.

REFERENCES Albino AP: The role of oncogenes and growth factors in progressive melanomagenesis. Pigment Cell Res Supplement 2:199^218, 1992 Arias AM: Epithelial mesenchymal interactions in cancer and development. Cell 105:425^431, 2001 Armstrong BK, Kricker A: Skin cancer. Dermatol Clin 13:583^594, 1995 Aszterbaum M, Epstein J, Oro A, Douglas V, LeBoit PE, Scott MP, Epstein EH Jr: Ultraviolet and ionizing radiation enhance the growth of BCC and trichoblastomas in patched heterozygous knockout mice. Nat Med 5:1285^1291, 1999 Atillasoy ES, Seykora JT, Soballe PW, et al: UVB induces atypical melanocytic lesions and melanoma in human skin. Am J Pathol 152:1179^1186, 1998 Aza-Blanc P, Kornberg TB: Ci: a complex transducer of the hedgehog signal. Trends Genet 15:458^462, 1999 Bale AE, Yu KP: The hedgehog pathway and basal cell carcinomas. Hum Mol Genet 10:757^762, 2001 Bardeesy N, Bastian BC, Hezel A, Pinkel D, DePinho RA, Chin L: Dual inactivation of RB and p53 pathways in RAS-induced melanomas. Mol Cell Biol 21:2144^2153, 2001 Bennett DC: Genetics, development, and malignancy of melanocytes. Int Rev Cytol 146:191^260, 1993 Berking C, Takemoto R, Satyamoorthy K, Elenitsas R, Herlyn M: Basic ¢broblast growth factor and ultraviolet B transform melanocytes in human skin. Am J Pathol 158:943^953, 2001 Bishop DT, Demenais F, Goldstein AM, et al: Geographical variation in the penetrance of CDKN2A mutations for melanoma. J Natl Cancer Inst 94:894^903, 2002 Black HS, Herd JA, Goldberg LH, et al: E¡ect of a low-fat diet on the incidence of actinic keratosis. N Engl J Med 330:1272^1275, 1994 Bradl M, Klein-Szanto A, Porter S, Mintz B: Malignant melanoma in transgenic mice. Proc Natl Acad Sci USA 88:164^168, 1991 Broome PM, Gause PR, Hyman P, Gregus J, Lluria-Prevatt M, Nagle R, Bowden GT: Induction of melanoma in TPras transgenic mice. Carcinogenesis 20:1747^1753, 1999 Brown K, Strathdee D, Bryson S, Lambie W, Balmain A: The malignant capacity of skin tumours induced by expression of a mutant H-ras transgene depends on the cell type targeted. Curr Biol 8:516^524, 1998 Cariello NF, Cui L, Beroud C, Soussi T: Database and software for the analysis of mutations in the human p53 gene. Cancer Res 54:4454^4460, 1994

VOL. 7, NO. 1 DECEMBER 2002

Chiang C, Swan RZ, Grachtchouk M, et al: Essential role for Sonic hedgehog during hair follicle morphogenesis. Dev Biol 205:1^9, 1999 Chin L, Pomerantz J, Polsky D, et al: Cooperative e¡ects of INK4a and ras in melanoma susceptibility in vivo. Genes Dev 11:2822^2834, 1997 Chin L, Merlino G, DePinho RA: Malignant melanoma: modern black plague and genetic black box. Genes Dev 12:3467^3481, 1998 Chin L, Tam A, Pomerantz J, et al: Essential role for oncogenic Ras in tumour maintenance. Nature 400:468^472, 1999 Chuang PT, Kornberg TB: On the range of hedgehog signaling. Curr Opin Genet Dev 10:515^522, 2000 Cli¡ord A, Morgan D, Yuspa SH, Soler AP, Gilmour S: Role of ornithine decarboxylase in epidermal tumorigenesis. Cancer Res 55:1680^1686, 1995 Dahmane N, Lee J, Robins P, Heller P, Altaba A: Activation of the transcription factor Gli1 and the Sonic hedgehog signalling pathway in skin tumours. Nature 389:876^881, 1997 Davies H, Bignell GR, Cox C, et al: Mutations of the BRAF gene in human cancer. Nature 417:949^954, 2002 De Gruijl FR: Skin cancer and solar UV radiation. Eur J Cancer 35:2003^2009, 1999 De Gruijl FR: Photocarcinogenesis. UVA vs UVB. Meth Enzymol 319:359^366, 2000 De Hertog S, Wensveen C, Bastiaens MT, et al: Relation between smoking and skin cancer. J Clin Oncol 19:231^238, 2001 DiGiovanni J, Imamoto A, Naito M, Walker SE, Beltran L, Chenicek KJ, Skow L: Further genetic analyses of skin tumor promoter susceptibility using inbred and recombinant inbred mice. Carcinogenesis 13:525^531, 1992 Dominey AM, Wang XJ, King LE Jr. et al: Targeted overexpression of transforming growth factor a in the epidermis of transgenic mice elicits hyperplasia, hyperkeratosis, and spontaneous squamous papillomas. Cell Growth Di¡er 4:1071^ 1082, 1993 van Elsas A, van der Scheibenbogen CMC, Zerp SF, Keilholz U, Schrier PI: UVinduced N-ras mutations are T-cell targets in human melanoma. Melanoma Res 7(Suppl.2):S107^S113, 1997 Fan H, Oro AE, Scott MP, Khavari PA: Induction of basal cell carcinoma features in transgenic human skin expressing Sonic Hedgehog. Nat Med 3:788^792, 1997 FitzGerald MG, Harkin DP, Silva-Arrieta S, et al: Prevalence of germ-line mutations in p16, p19ARF, and CDK4 in familial melanoma: analysis of a clinic-based population. Proc Natl Acad Sci USA 93:8541^8545, 1996 Fitzpatrick FA: In£ammation, carcinogenesis and cancer. Int Immunopharmacol 1:1651^1667, 2001 Funk JO, Schiller PI, Barrett MT, Wong DJ, Kind P, Sander CA: p16INK4a expression is frequently decreased and associated with 9p21 loss of heterozygosity in sporadic melanoma. J Cutan Pathol 25:291^296, 1998 Gallagher RP, Bajdik CD, Fincham S, Hill GB, Keefe AR, Coldman A, McLean DI: Chemical exposures, medical history, and risk of squamous and basal cell carcinoma of the skin. Cancer Epidemiol Biomarkers Prev 5:419^424, 1996 Glick AB, Lee MM, Darwiche N, Kulkarni AB, Karlsson S, Yuspa SH: Targeted deletion of the TGF-b 1 gene causes rapid progression to squamous cell carcinoma. Genes Dev 8:2429^2440, 1994 Glick A, Popescu N, Alexander V, Ueno H, Bottinger E,Yuspa SH: Defects in transforming growth factor-b signaling cooperate with a ras oncogene to cause rapid aneuploidy and malignant transformation of mouse keratinocytes. Proc Natl Acad Sci USA 96:14949^14954, 1999 Go C, Li P,Wang XJ: Blocking transforming growth factor b signaling in transgenic epidermis accelerates chemical carcinogenesis: a mechanism associated with increased angiogenesis. Cancer Res 59:2861^2868, 1999 Goldstein AM, Tucker MA: Etiology, epidemiology, risk factors, and public health issues of melanoma. Curr Opin Oncol 5:358^363, 1993 Gonzalgo ML, Bender CM, You EH, et al: Low frequency of p16/CDKN2A methylation in sporadic melanoma: comparative approaches for methylation analysis of primary tumors. Cancer Res 57:5336^5347, 1997 Goodrich LV, Milenkovic L, Higgins KM, Scott MP: Altered neural cell fates and medulloblastoma in mouse patched mutants. Science 277:1109^1113, 1997 Gorlin RJ: Nevoid basal^cell carcinoma syndrome. Medicine (Baltimore) 66:98^113, 1987 Grachtchouk M, Mo R, Hui CC, Dlugosz AA: The Sonic hedgehog target gene Gli2 plays a pivotal role in hair follicle development. J Invest Dermatol 114:756, 2000a Grachtchouk M, Mo R, Yu S, Zhang X, Sasaki H, Hui CC, Dlugosz AA: Basal cell carcinomas in mice overexpressing Gli2 in skin. Nat Genet 24:216^217, 2000b Greenhalgh DA, Yuspa SH: Malignant conversion of murine squamous papilloma cell lines by transfection with the fos oncogene. Mol Carcinog 1:134^143, 1988 Gri⁄th TS, Anderson RD, Davidson BL, Williams RD, Ratli¡ TL: Adenoviralmediated transfer of the TNF-related apoptosis-inducing ligand/Apo-2 ligand gene induces tumor cell apoptosis. J Immunol 165:2886^2894, 2000 Hahn H, Wicking C, Zaphiropoulous PG, et al: Mutations of the human homolog of Drosophila patched in the nevoid basal cell carcinoma syndrome. Cell 85:841^851, 1996 Hahn H, Wojnowski L, Zimmer AM, Hall J, Miller G, Zimmer A: Rhabdomyosarcomas and radiation hypersensitivity in a mouse model of Gorlin syndrome. Nat Med 4:619^622, 1998 Halaban R: Growth factors and melanomas. Semin Oncol 23:673^681, 1996 Halaban R, Rubin JS, Funasaka Y, et al: Met and hepatocyte growth factor/scatter factor signal transduction in normal melanocytes and melanoma cells. Oncogene 7:2195^2206, 1992

PROGRESS IN CUTANEOUS CANCER RESEARCH

25

Halpern AC, Altman JF: Genetic predisposition to skin cancer. Curr Opin Oncol 11:132^138, 1999 Holman CD, Armstrong BK, Heenan PJ: A theory of the etiology and pathogenesis of human cutaneous malignant melanoma. J Natl Cancer Inst 71:651^656, 1983 Hussain SP, Harris CC: Molecular epidemiology of human cancer: contribution of mutation spectra studies of tumor suppressor genes. Cancer Res 58:4023^4037, 1998 Hussain SP, Raja K, Amstad PA, et al: Increased p53 mutation load in nontumorous human liver of Wilson disease and hemochromatosis: oxyradical overload diseases. Proc Natl Acad Sci USA 97:12770^12775, 2000 Hussussian CJ, Struewing JP, Goldstein AM, et al: Germline p16 mutations in familial melanoma. Nat Genet 8:15^21, 1994 IARC: Monographs onThe Evaluation of Carcinogenic Risks to Humans: Solar and Ultraviolet Radiation. Lyons: IARC, 1992 Jhappan C, Takayama H, Dickson RB, Merlino G: Transgenic mice provide genetic evidence that transforming growth factor a promotes skin tumorigenesis via H-ras-dependent and H-ras-independent pathways. Cell Growth Di¡er 5:385^ 394, 1994 Jiveskog S, Ragnarsson-Olding B, Platz A, Ringborg U: N-ras mutations are common in melanomas from sun-exposed skin of humans but rare in mucosal membranes or unexposed skin. J Invest Dermatol 111:757^761, 1998 Johnson WD, Robertson KA, Pounds JG, Allen JR: Dehydroretronecine-induced skin tumors in mice. J Natl Cancer Inst 61:85^89, 1978 Johnson RL, Rothman AL, Xie J, et al: Human homolog of patched, a candidate gene for the basal cell nevus syndrome. Science 272:1668^1671, 1996 Kamb A, Shattuck-Eidens D, Eeles R, et al: Analysis of the p16 gene (CDKN2) as a candidate for the chromosome 9p melanoma susceptibility locus. Nat Genet 8:23^26, 1994 Kato M,Takahashi M, Akhand AA, et al: Transgenic mouse model for skin malignant melanoma. Oncogene 17:1885^1888, 1998 Kelsall SR, Mintz B: Metastatic cutaneous melanoma promoted by ultraviolet radiation in mice with transgene-initiated low melanoma susceptibility. Cancer Res 58:4061^4065, 1998 Kemp CJ, Donehower LA, Bradley A, Balmain A: Reduction of p53 gene dosage does not increase initiation or promotion but enhances malignant progression of chemically induced skin tumors. Cell 74:813^822, 1993 Klein-Szanto AJ, Silvers WK, Mintz B: Ultraviolet radiation-induced malignant skin melanoma in melanoma-susceptible transgenic mice. Cancer Res 54:4569^4572, 1994 Kos L, Aronzon A, Takayama H, Maina F, Ponzetto C, Merlino G, Pavan W: Hepatocyte growth factor/scatter factor-MET signaling in neural crest- derived melanocyte development. Pigment Cell Res 12:13^21, 1999 Kreimer-Erlacher H, Seidl H, Back B, Kerl H, Wolf P: High mutation frequency at Ha-ras exons 1^4 in squamous cell carcinomas from PUVA-treated psoriasis patients. Photochem Photobiol 74:323^330, 2001 Krimpenfort P, Quon KC, Mooi WJ, Loonstra A, Berns A: Loss of p16INK4a confers susceptibility to metastatic melanoma in mice. Nature 413:83^86, 2001 Kusewitt DF, Ley RD: Animal models of melanoma. Cancer Surv 26:35^70, 1996 Lam CW, Xie J, To KF, et al: a frequent activated smoothened mutation in sporadic basal cell carcinomas. Oncogene 18:833^836, 1999 Linardopoulos S, Street AJ, Quelle DE, Parry D, Peters G, Sherr CJ, Balmain A: Deletion and altered regulation of p16INK4a and p15INK4b in undi¡erentiated mouse skin tumors. Cancer Res 55:5168^5172, 1995 Liotta LA, Kohn EC: The microenvironment of the tumour^host interface. Nature 411:375^379, 2001 Marks R: Epidemiology of melanoma. Clin Exp Dermatol 25:459^463, 2000 Matise MP, Joyner AL: Gli genes in development and cancer. Oncogene 18:7852^7859, 1999 Miller SJ: Etiology and pathogenesis of basal cell carcinoma. Clin Dermatol 13:527^536, 1995 Missero C, Di Cunto F, Kiyokawa H, Ko¡ A, Dotto GP: The absence of p21Cip1/WAF1 alters keratinocyte growth and di¡erentiation and promotes ras-tumor progression. Genes Dev 10:3065^3075, 1996 Mock BA, Lowry DT, Rehman I, Padlan C, Yuspa SH, Hennings H: Multigenic control of skin tumor susceptibility in SENCAR/Part mice. Carcinogenesis 19:1109^1115, 1998 Moghal N, Sternberg PW: Multiple positive and negative regulators of signaling by the EGF- receptor. Curr Opin Cell Biol 11:190^196, 1999 Morris RJ, Potten CS: Highly persistent label-retaining cells in the hair follicles of mice and their fate following induction of anagen. J Invest Dermatol 112:470^475, 1999 Nagase H, Bryson S, Cordell H, Kemp CJ, Fee F, Balmain A: Distinct genetic loci control development of benign and malignant skin tumours in mice. Nat Genet 10:424^429, 1995 Nagase H, Mao JH, Balmain A: A subset of skin tumor modi¢er loci determines survival time of tumor- bearing mice. Proc Natl Acad Sci USA 96:15032^15037, 1999 Nagase H, Mao JH, de Koning JP, Minami T, Balmain A: Epistatic interactions between skin tumor modi¢er loci in interspeci¢c (spretus/musculus) backcross mice. Cancer Res 61:1305^1308, 2001 Navarro P, G˛mez M, Pizarro A, Gamallo C, Quintanilla M, Cano A: A role for the E-cadherin cell-cell adhesion molecule during tumor progression of mouse epidermal carcinogenesis. J Cell Biol 115:517^533, 1991

26

DLUGOSZ ET AL

Nilsson M, Unden AB, Krause D, Malmqwist U, Raza K, Zaphiropoulos PG, Toftgard R: Induction of basal cell carcinomas and trichoepitheliomas in mice overexpressing GLI-1. Proc Natl Acad Sci USA 97:3438^3443, 2000 Noonan FP, Otsuka T, Bang S, Anver MR, Merlino G: Accelerated ultraviolet radiation-induced carcinogenesis in hepatocyte growth factor/scatter factor transgenic mice. Cancer Res 60:3738^3743, 2000 Noonan FP, Recio JA, Takayama H, et al: Neonatal sunburn and melanoma in mice. Nature 413:271^272, 2001 Oro AE, Higgins KM, Hu Z, Bonifas JM, Epstein EH Jr, Scott MP: Basal cell carcinomas in mice overexpressing sonic hedgehog. Science 276:817^821, 1997 Otsuka T, Takayama H, Sharp R, et al: c-Met autocrine activation induces development of malignant melanoma and acquisition of the metastatic phenotype. Cancer Res 58:5157^5167, 1998 Pelengaris S, Littlewood T, Khan M, Elia G, Evan G: Reversible activation of c-Myc in skin: induction of a complex neoplastic phenotype by a single oncogenic lesion. Mol Cell 3:565^577, 1999 Peris K, Chimenti S, Fargnoli MC, Valeri P, Kerl H, Wolf P: UV ¢ngerprint CDKN2a but no p14ARF mutations in sporadic melanomas. J Invest Dermatol 112:825^826, 1999 Pierceall WE, Goldberg LH, Tainsky MA, Mukhopadhyay T, Ananthaswamy HN: Ras gene mutation and ampli¢cation in human nonmelanoma skin cancers. Mol Carcinog 4:196^202, 1991 Pollock PM, Yu F, Qiu L, Parsons PG, Hayward NK: Evidence for u.v. induction of CDKN2 mutations in melanoma cell lines. Oncogene 11:663^668, 1995 Portella G, Cumming SA, Liddell J, Cui W, Ireland H, Akhurst RJ, Balmain A: Transforming growth factor b is essential for spindle cell conversion of mouse skin carcinoma in vivo: implications for tumor invasion. Cell Growth Di¡er 9:393^404, 1998 Prehn RT, Prehn LM: Immunostimulation of cancer versus immunosurveillance. Medicina (B Aires) 56(Suppl.1):65^73, 1996 Quelle DE, Zindy F, Ashmun RA, Sherr CJ: Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell 83:993^1000, 1995 Quinn AG, Sikkink S, Rees JL: Basal cell carcinomas and squamous cell carcinomas of human skin show distinct patterns of chromosome loss. Cancer Res 54:4756^4759, 1994 Quintanilla M, Brown K, Ramsden M, Balmain A: Carcinogen-speci¢c mutation and ampli¢cation of Ha-ras during mouse skin carcinogenesis. Nature 322:78^80, 1986 Rane SG, Cosenza SC, Mettus RV, Reddy EP: Germ line transmission of the Cdk4 (R24C) mutation facilitates tumorigenesis and escape from cellular senescence. Mol Cell Biol 22:644^656, 2002 Rehman I, Quinn AG, Healy E, Rees JL: High frequency of loss of heterozygosity in actinic keratoses, a usually benign disease. Lancet 344:788^789, 1994 Rigel DS, Friedman RJ, Kopf AW: Lifetime risk for development of skin cancer in the U.S. population: current estimate is now 1 in 5. J Am Acad Dermatol 35:1012^1013, 1996 Robles AI, Rodriguez-Puebla ML, Glick AB, et al: Reduced skin tumor development in cyclin D1-de¢cient mice highlights the oncogenic ras pathway in vivo. Genes Dev 12:2469^2474, 1998 Russo AA, Tong L, Lee JO, Je¡rey PD, Pavletich NP: Structural basis for inhibition of the cyclin-dependent kinase Cdk6 by the tumour suppressor p16INK4a. Nature 395:237^243, 1998 Saez E, Rutberg SE, Mueller E, Oppenheim H, Smoluk J,Yuspa SH, Spiegelman BM: c- fos is required for malignant progression of skin tumors. Cell 82:721^732, 1995 Sato N, Leopold PL, Crystal RG: Induction of the hair growth phase in postnatal mice by localized transient expression of Sonic hedgehog. J Clin Invest 104:855^864, 1999 Satyamoorthy K, Meier F, Hsu MY, Berking C, Herlyn M: Human xenografts, human skin and skin reconstructs for studies in melanoma development and progression. Cancer Metastasis Rev 18:401^405, 1999 Schwartz RA: Arsenic and the skin. Int J Dermatol 36:241^250, 1997 Sharpless NE, Bardeesy N, Lee KH, et al: Loss of p16Ink4a with retention of p19 Arf predisposes mice to tumorigenesis. Nature 413:86^91, 2001 Shih IM, Herlyn M: Autocrine and paracrine roles for growth factors in melanoma. In Vivo 8:113^123, 1994 Smith CW, Chen Z, Dong G, Loukinova E, Pegram MY, Nicholas-Figueroa L, Van Waes C: The host environment promotes the development of primary and metastatic squamous cell carcinomas that constitutively express proin£ammatory cytokines IL-1alpha, IL- 6, GM-CSF, and KC. Clin Exp Metastasis 16:655^664, 1998 Smith-McCune K, Zhu YH, Hanahan D, Arbeit J: Cross-species comparison of angiogenesis during the premalignant stages of squamous carcinogenesis in the human cervix and K14 - HPV16 transgenic mice. Cancer Res 57:1294^1300, 1997 Soengas MS, Capodieci P, Polsky D, et al: Inactivation of the apoptosis e¡ector Apaf1 in malignant melanoma. Nature 409:207^211, 2001 Soriano P: The PDGF alpha receptor is required for neural crest cell development and for normal patterning of the somites. Development 124:2691^2700, 1997

JID SYMPOSIUM PROCEEDINGS

Sotillo R, Garcia JF, Ortega S, Martin J, Dubus P, Barbacid M, Malumbres M: Invasive melanoma in Cdk4 -targeted mice. Proc Natl Acad Sci USA 98:13312^ 13317, 2001 Sou¢r N, Avril MF, Chompret A, et al: Prevalence of p16 and CDK4 germline mutations in 48 melanoma-prone families in France. The French Familial Melanoma Study Group. Hum Mol Genet 7:209^216, 1998 Sou¢r N, Moles JP, Vilmer C, et al: P16 UV mutations in human skin epithelial tumors. Oncogene 18:5477^5481, 1999 St Jacques B, Dassule HR, Karavanova I, et al: Sonic hedgehog signaling is essential for hair development. Curr Biol 8:1058^1068, 1998 van Steeg H, Kraemer KH: Xeroderma pigmentosum and the role of UV-induced DNA damage in skin cancer. Mol Med Today 5:86^94, 1999 Stephenson DA, Mercola M, Anderson E, Wang CY, Stiles CD, Bowen-Pope DF, Chapman VM: Platelet-derived growth factor receptor alpha-subunit gene (Pdgfra) is deleted in the mouse patch (Ph) mutation. Proc Natl Acad Sci USA 88:6^10, 1991 Taipale J, Chen JK, Cooper MK, et al: E¡ects of oncogenic mutations in Smoothened and Patched can be reversed by cyclopamine. Nature 406:1005^1009, 2000 Takayama H, La Rochelle WJ, Anver M, Bockman DE, Merlino G: Scatter factor/ hepatocyte growth factor as a regulator of skeletal muscle and neural crest development. Proc Natl Acad Sci USA 93:5866^5871, 1996 Takayama H, Larochelle WJ, Sharp R, et al: Diverse tumorigenesis associated with aberrant development in mice overexpressing hepatocyte growth factor/scatter factor. Proc Natl Acad Sci USA 94:701^706, 1997 Taylor RS, Ramirez RD, Ogoshi M, Cha⁄ns M, Piatyszek MA, Shay JW: Detection of telomerase activity in malignant and nonmalignant skin conditions. J Invest Dermatol 106:759^765, 1996 Tietze MK, Chin L: Murine models of malignant melanoma. Mol Med Today 6:408^ 410, 2000 Vassar R, Hutton ME, Fuchs E: Transgenic overexpression of transforming growth factor a bypasses the need for c-Ha-ras mutations in mouse skin tumorigenesis. Mol Cell Biol 12:4643^4653, 1992 van’t Veer LJ, Burgering BM, Versteeg R, et al: N-ras mutations in human cutaneous melanoma from sun-exposed body sites. Mol Cell Biol 9:3114^3116, 1989 Vorechovsky I, Unden AB, Sandstedt B,Toftgard R, Stahle-Backdahl M: Trichoepitheliomas contain somatic mutations in the overexpressed PTCH gene: support for a gatekeeper mechanism in skin tumorigenesis. Cancer Res 57:4677^4681, 1997 Waikel RL, Kawachi Y, Waikel PA, Wang XJ, Roop DR: Deregulated expression of c-Myc depletes epidermal stem cells. Nat Genet 28:165^168, 2001 Wang LC, Liu ZY, Gambardella L, et al: Conditional disruption of hedgehog signaling pathway de¢nes its critical role in hair development and regeneration. J Invest Dermatol 114:901^908, 2000 Wei Q, Matanoski GM, Farmer ER, Hedayati MA, Grossman L: DNA repair related to multiple skin cancers and drug use. Cancer Res 54:437^440, 1994 Weinberg WC, Fernandez-Salas E, Morgan DL, et al: Genetic deletion of p21WAF1 enhances papilloma formation but not malignant conversion in experimental mouse skin carcinogenesis. Cancer Res 59:2050^2054, 1999 Whiteman DC, Whiteman CA, Green AC: Childhood sun exposure as a risk factor for melanoma: a systematic review of epidemiologic studies. Cancer Causes Control 12:69^82, 2001 Witte ON: Steel locus de¢nes new multipotent growth factor. Cell 63:5^6, 1990 Xie J, Murone M, Luoh SM, et al: Activating Smoothened mutations in sporadic basal-cell carcinoma. Nature 391:90^92, 1998 Xin H, Matt D, Qin JZ, Burg G, Boni R: The sebaceous nevus nevus. A with deletions of the PTCH gene. Cancer Res 59:1834^1836, 1999 Young MR, Li JJ, Rincon M, Flavell RA, Sathyanarayana BK, Hunziker R, Colburn N: Transgenic mice demonstrate AP-1 (activator protein-1) transactivation is required for tumor promotion. Proc Natl Acad Sci USA 96:9827^9832, 1999 Yuspa SH: The pathogenesis of squamous cell cancer: lessons learned from studies of skin carcinogenesis ^Thirty-third G.H.A. Clowes Memorial Award Lecture. Cancer Res 54:1178^1189, 1994 Yuspa SH, Morgan DL: Mouse skin cells resistant to terminal di¡erentiation associated with initiation of carcinogenesis. Nature (London) 293:72^74, 1981 Yuspa SH, Spangler EF, Donahoe R, Geusz S, Ferguson E, Wenk M, Hennings H: Sensitivity to two-stage carcinogenesis of SENCAR mouse skin grafted to nude mice. Cancer Res 42:437^439, 1982 Zackheim HS: Experimental basal cell carcinoma in the rat. In: Maibach L ed. Models in Dermatology. Basel: Karger, 1985, pp 89^97 Zaphiropoulos PG, Unden AB, Rahnama F, Hollingsworth RE, Toftgard R: PTCH2, a novel human patched gene, undergoing alternative splicing and up-regulated in basal cell carcinomas. Cancer Res 59:787^792, 1999 Zerp SF, van Elsas A, Peltenburg LT, Schrier PI: p53 mutations in human cutaneous melanoma correlate with sun exposure but are not always involved in melanomagenesis. Br J Cancer 79:921^926, 1999 Zuo L, Weger J, Yang Q, et al: Germline mutations in the p16INK4a binding domain of CDK4 in familial melanoma. Nat Genet 12:97^99, 1996