Genetic Analysis of Malignant Melanoma: Do We Still Need Histology?
Victor G. Prieto
University of Texas MD Anderson Cancer Center
The incidence of reported melanoma cases has risen strikingly over the past several decades. Melanoma
is the skin cancer with highest mortality. Since there is no cure for advanced disease, it is crucial to
detect melanoma in early stages, either in-situ (confined to the epidermis) or when it involves the upper
dermis (around 95% of such lesions are cured by complete surgical removal). Exposure to ultraviolet
radiation (UVR) is a major melanoma risk factor, but also fair skin, history of sunburn, blue or green
eyes, red hair, number of typical acquired, congenital, or atypical (dysplastic) nevi, ephelides, and
lentigines. Immunosuppression and personal and family history of melanoma are also associated with
increased risk of melanoma.
Despite multiple studies analyzing the pathogenesis of melanoma, it is still mostly unknown for
several reasons. There appears to be different types of melanoma; as mentioned before, a number of
melanomas is associated with sun / ultraviolet light exposure (mostly lentigo maligna) but melanomas do
occur in sun-protected acral and genital locations. Also, some patients have relatives with melanoma
while most cases do not appear to be familial so it is more difficult to detect any possible genetic
alterations. Finally, skin lesions are easily observed by patient, family, and doctor. Therefore,
lesions that are suspicious for melanoma are likely to undergo biopsy before they reach large size. And
since the determination of prognosis is essentially based upon the histologic examination of the lesion,
very little material can be spared for the non-routine histologic studies currently necessary for
analyzing genetic anomalies.
Although some rare authors disagree, there seems to be some sort of progression between benign and
malignant melanocytic lesions. As such, it has been proposed that mature melanocytes develop in nevi and
that some nevi (common or atypical) develop in melanomas in situ, then invasive, and finally metastatic.
Therefore, before analyzing the possible genetic and molecular changes associated with this purported
progression we will briefly analyze these possible steps.
The first step in melanoma progression may be benign nevus. Number of nevi is an independent risk
factor for melanoma. However, since nevi are so prevalent while melanomas are much less common, it is
obvious that the vast majority of nevi are not premalignant. Precursor nevi are identified in contiguity
with melanoma in 10-40% of all cases, in particular large congenital nevi. Conversely a majority of
melanomas seem to originate from epidermal melanocytes so the nevus is probably not an obligatory stage
in the pathogenesis of melanoma.
The terms atypical/ dysplastic/ Clark nevus are synonymous. After much controversy started against
the original term ("dysplastic nevus") proposed by Clark et el, the NIH Consensus Conference recommended
using "nevus with architectural disorder and cytologic atypia", but this longer name has not completely
caught up. Although much controversy has attended this concept, in essence it is a nevus that resembles
melanoma, clinically and histologically: large size (generally >5 mm), poor circumscription, border
irregularity, a play of colors, and asymmetry. Atypical nevi occur at a prevalence of 2-9% in unselected
white populations in most series. These nevi represent a strong, independent risk factor for development
of melanoma, whether they occur sporadically or in the context of a melanoma-prone familial
predisposition (dysplastic nevus syndrome). Moreover, this association shows a dose-response
relationship, with melanoma risk rising proportionately with number of atypical nevi present.
Histologically, these lesions are described as having cytologic and architectural atypia. Both the
degree of cytologic atypia and the extent of architectural disorder tend to be positively correlated with
each other, so it seems that histologic analysis is successful in detecting these markers for melanoma
The next step in melanoma progression is melanoma. Early melanoma has been arbitrarily defined as
those lesions with Breslow thickness <1 mm (measured from the top of the granular layer to the deepest
tumor cell). Some such lesions also are in radial growth phase (RGP), early phase in which melanomas are
considered to have little if any potential for metastasis. By definition, all MIS (Clark level I) and
some superficially invasive (Clark level II, lesions involving but not filling or expansion of papillary
dermis, and lacking large nests in the dermis or dermal mitotic figures) are in RGP. Beyond this point,
some authors consider that melanoma enters a tumorigenic phase (vertical growth phase, VGP). In this
early melanomas less than 1 mm in Breslow thickness, the overall survival after complete excision is
greater than 90% at 5 years.
Several genetic changes are associated with the proposed melanoma progression. There are anomalies in
chromosome 1 (1p36) in patients with the dysplastic nevus syndrome/familial melanoma. Such families also
exhibit abnormalities in a second gene, in chromosome 9, that encodes an inhibitor of a kinase involved
in the cell cycle, named p16INK4a or cyclin-dependent kinase inhibitor-2 (CDKN2A). Loss of
heterozygosity for p16 is an early event in melanomas, having been reported also in atypical nevi. The
addition of activating mutations in N-ras to melanoma cells already containing mutations in p16 may
induce the VGP phenotype. Such N-ras mutations are common in melanomas from sun-exposed skin.
In addition to p16 mutations, atypical nevi may occur due to abnormalities in DNA repair in response
to UV light. These patients with dysplastic nevus syndrome have chromosomal instability, as documented
in normal skin, atypical nevi, and blood lymphocytes. The lesions contain increased amounts of
pheomelanin compared with common nevi, a fact that might increase their susceptibility to UVR damage.
Shifts among these melanin types may reflect changes in pigment regulatory genes such as that for the
A number of genes show relatively consistent mutations in thick melanomas, possibly correlating with a
decreased survival: c-myc (nuclear protein required for transition from G1 to S phase), p53
(gatekeeper protein preventing cell proliferation after DNA damage), bcl-2 (anti-apoptotic factor),
transforming growth factor-b (TGF- b , protein involved in angiogenesis and wound healing), CD40
(receptor that induces escape from apoptosis), and the cyclin-dependent kinase inhibitors p27 and p21
(KIP1 and Waf-1/SDI-1, respectively). Enzymes that degrade collagen have been proposed to facilitate
permeation of the dermis and access to blood vessels; in particular, collagenases-1 and -3 are expressed
in invasive and metastatic but not in in situ melanomas. The Ets-1
transcription factor is overexpressed in melanoma and may play a role in tumor vascularization and
invasion by regulating expression of matrix-degrading proteases in endothelial cells and fibroblasts in
the tumor stroma. Melastatin (TRPM1) is expressed in nevi and melanomas, but it is usually lost in
Similarly, some integrins (molecules involved in cell-to-cell recognition), such as the beta3 subunit
of the vitronectin receptor, are first detected in VGP melanomas, as is interleukin-8 (IL-8). Since
tumors must recruit new vessels in order to grow, it seems obvious that there should be secretion of
angiogenic factors at some point. There is a relationship between numbers of mast cells (which contain
high quantities of angiogenic factors) and invasiveness in melanomas. Also some studies have shown
increased vascularity in VGP versus radial growth phase melanomas. Two candidate proteins are VEGF
(vascular endothelial growth factor) and b-FGF, which have mitogenic effects on endothelial cells.
Supporting this view, by immunohistochemistry or in situ hybridization,
b-FGF is expressed in invasive but not in situ melanomas.
An exciting discovery has been the recent description of activating mutations in 60% to 80% of
cutaneous melanomas and benign melanocytic nevi in the BRAF gene, a component of the RAS/RAF/MAPK
signaling cassette. In particular, several groups are currently studying the possible use of anti-BRAF
compounds as treatment for melanoma. Furthermore, some authors have suggested that there is an inverse
relationship between activation of BRAF and loss of PTEN. Interestingly, Spitz nevi lack the T1796A BRAF
mutation that is commonly detected in other nevi and melanoma.
Due to the need for sufficient tissue for specialized studies, most genetic changes have been
described in advanced lesions, either deeply invasive primary lesions or metastatic lesions, or in cell
lines. Generally, only small quantities of tissue are available for analysis in common nevi, atypical
nevi, and early MM. Breakthrough techniques such as laser microdissection now allow removal and analysis
of minute quantities of tissue, potentially allowing molecular analysis of the earliest stages of
incipient melanoma. This approach has been employed to detect chromosomal differences between RGP and
VGP melanoma: losses of chromosomes 9 and 10 early, and gains of chromosome 7 later. Using similar
techniques we found an alteration in Chr 20q13, in an area where is located a gene involved in chromosome
function (aurora2, STK15, or STK6). This modification is present in superficial melanomas and in
dysplastic nevi, but not in common nevi and may therefore be important in early malignancy
transformation. Similar studies have shown that superficial spreading melanomas are genetically
different from acral lentiginous melanomas, which may explain the different phenotype that these lesions
present. Furthermore, the same genetic alterations seen in acral lentiginous melanoma are also present
in individual, morphologically benign melanocytes located away from the main lesion ("field effect").
Also using microdissection, gene array profiling may be able to detect genetic differences between benign
and malignant melanocytic lesions.
So what is the current status in analysis of melanocytic lesions? Are we at a point in which a
clinician may place a piece of skin into a machine and then retrieve a diagnosis of melanoma vs. nevus
and a list of prognostic factors? Despite the existence of a spectrum of melanocytic atypia, ranging
from banal nevi to outright melanoma, we believe that the experienced pathologist can confidently render
a diagnosis of nevus versus melanoma in the majority of cases encountered in routine practice. We make
this assertion despite some studies wherein experts reviewed cases of melanoma or of nevus sharing
features with melanoma, and were asked to classify them as malignant, indeterminate, or benign, then
obtaining complete agreement in only 30% of cases. In part, the discordance may be the result of
case-selection at the high end of the spectrum (where concordance regarding malignancy versus benignity
is most problematical). In addition, when morphologic features may be ambiguous, immunohistochemical
studies may be helpful. The use of antibodies against antigens typically expressed in melanocytic cells,
such as S100 protein, gp100 (detected with HMB45), or MART-1 (melanoma antigen recognized by T-cells)
allows confirmation of melanocytic lineage in a given neoplasm and distinction in small biopsies of
pigmented basal cell carcinoma versus small cell melanoma. Also, when examining sentinel node specimens,
immunohistochemistry may help detect small clusters of or even single melanoma cells within the lymph
node. An exciting but controversial field is the use of immunohistochemistry for the differential
diagnosis of melanoma and nevus. Although some authors reject this approach we strongly feel that, in
selected cases and along with clinical and histologic criteria, immunohistochemical data are helpful in
the diagnosis of melanocytic lesions. An example is the assessment of maturation in the dermal
components. Superficial, round, type A nevus cells share many immunohistochemical markers with neurons
while the deep, spindle, type C nevus cells resemble Schwann cells. This morphology is reflected in the
decreased expression of gp100 by those melanocytes located at the base of the nevus and not in melanomas.
Also, the almost universal lack of mitotic figures in the deep regions of nevi correlates with the very
sparse expression of proliferation markers such as Ki-67 by the more deeply located melanocytes in nevi.
In contrast, melanomas do not show this orderly pattern, but instead had a random pattern of
immunoreactivity, especially concentrated at the deep tumor borders (see table below).
| Pattern of expression: ||Nevus ||Melanoma|
|gp100 (with HMB45) ||Superficial region ||Throughout|
|Tyrosinase ||Superficial region ||Throughout|
|Peripherin ||Superficial region ||Throughout|
|Ki67 (nuclear, proliferation marker) ||Rare at base ||Throughout|
In summary, the take home message of this talk is that despite the very exciting findings described
in the last 10 years, the gold standard (and by far cheapest) method to study melanocytic lesions remains
- Becker B, Roesch A, Hafner C, Stolz W, Dugas M, Landthaler M, Vogt T (2004). Discrimination of melanocytic tumors by cDNA array hybridization of tissues prepared by laser pressure catapulting. J Invest Dermatol 122: 361-8.
- Airola K, Karonen T, Vaalamo M, et al. (1999). Expression of collagenases-1 and -3 and their inhibitors TIMP-1 and -3 correlates with the level of invasion in malignant melanomas. Br J Cancer 80: 733-743.
- Bales ES, Dietrich C, Bandyopadhyay D, et al. (1999). High levels of expression of p27 (KIP1) and cyclin E in invasive primary malignant melanomas. J Invest Derm 113: 1039-1046.
- Bastian BC, LeBoit PE, Hamm H, Brocker EB, and Pinkel D (1998). Chromosomal gains and losses in primary cutaneous melanomas detected by comparative genomic hybridization. Cancer Res 58: 2170-2175.
- Bastian BC (2004). Molecular genetics of melanocytic neoplasia: practical applications for diagnosis. Pathology 36: 458-61.
- Davies H, Bignell GR, Cox C, et al (2002) Mutations of the BRAF gene in human cancer. Nature 417: 949-954.
- Duncan LM, Deeds J, Hunter J, Shao J, Holmgren LM, Woolf EA, Tepper RI, Shyjan AW (1998). Down-regulation of the novel gene melastatin correlates with potential for melanoma metastasis. Cancer Res 58: 1515-1520
- Erhard H, Rietveld FJ, van Altena MC, Brocker EB, Ruiter DJ, and de Waal RM (1997). Transition of horizontal to vertical growth phase melanoma is accompanied by induction of vascular endothelial growth factor expression and angiogenesis. Melan Res 7: S19-26.
- Farmer ER, Gonin R, and Hanna MP (1996). Discordance in the histopathologic diagnosis of melanoma and melanocytic nevi between expert pathologists. Hum Pathol 27: 528-531.
- Fujita M, Norris DA, Yagi H, et al. (1999). Overexpression of mutant ras in human melanoma increases invasiveness, proliferation and anchorage-independent growth in vitro and induces tumour formation and cachexia in vivo. Melan Res 9: 279-291.
- Goldstein AM, Goldin LR, Dracopoli NC, Clark WH, Jr., and Tucker MA (1996). Two-locus linkage analysis of cutaneous malignant melanoma/dysplastic nevi. Am J Hum Genet 58: 1050-1056.
- Hashemi J, Linder S, Platz A, and Hansson J (1999). Melanoma development in relation to nonfunctional p16/INK4A protein and dysplastic naevus syndrome in Swedish melanoma kindreds. Melan Res 9: 21-30.
- Hsu MY, Shih DT, Meier FE, et al. (1998). Adenoviral gene transfer of beta3 integrin subunit induces conversion from radial to vertical growth phase in primary human melanoma. Am J Pathol 153: 1435-1442.
- Hussein MR, Roggero E, Sudilovsky EC, Tuthill RJ, Wood GS, Sudilovsky O (2001). Alterations of mismatch repair protein expression in benign melanocytic nevi, melanocytic dysplastic nevi, and cutaneous malignant melanomas. Am J Dermatopathol. 23: 308-314.
- Jiveskog S, Ragnarsson-Olding B, Platz A, and Ringborg U (1998). N-ras mutations are common in melanomas from sun-exposed skin of humans but rare in mucosal membranes or unexposed skin. J Invest Derm 111: 757-761.
- Moriwaki SI, Tarone RE, Tucker MA, Goldstein AM, and Kraemer KH (1997). Hypermutability of UV-treated plasmids in dysplastic nevus/familial melanoma cell lines. Cancer Res 57: 4637-4641.
- Palmedo G, Hantschke M, Rutten A, Mentzel T, Hugel H, Flaig MJ, Yazdi AS, Sander CA, Kutzner H (2004). The T1796A mutation of the BRAF gene is absent in Spitz nevi. J Cutan Pathol. 31: 266-70.
- Prieto VG, McNutt NS, Lugo J, and Reed JA (1997). The intermediate filament peripherin is expressed in cutaneous melanocytic lesions. J Cutan Pathol 24: 145-150.
- Reed JA, Loganzo F, Shea CR, et al. (1995). Loss of expression of the p16/cyclin-dependent kinase inhibitor 2 tumor suppressor gene in melanocytic lesions correlates with invasive stage of tumor progression. Cancer Res 55: 2713-2718.
- Reed JA, McNutt NS, and Albino AP (1994). Differential expression of basic fibroblast growth factor (bFGF) in melanocytic lesions demonstrated by in situ hybridization: implications for tumor progression. Am J Pathol 144: 329-336.
- Rothhammer T, Hahne JC, Florin A, Poser I, Soncin F, Wernert N, Bosserhoff AK (2004). The Ets-1 transcription factor is involved in the development and invasion of malignant melanoma. Cell Mol Life Sci. 61:118-28.
- Rudolph P, Schubert C, Schubert B, and Parwaresch R (1997). Proliferation marker Ki-S5 as a diagnostic tool in melanocytic lesions. J Am Acad Dermatol 37: 169-178.
- Salopek TG, Yamada K, Ito S, and Jimbow K (1991). Dysplastic melanocytic nevi contain high levels of pheomelanin: quantitative comparison of pheomelanin/eumelanin levels between normal skin, common nevi, and dysplastic nevi. Pigment Cell Research 4: 172-179.
- Shea CR, Vollmer RT, and Prieto VG (1999). Grading architectural disorder and cytologic atypia in melanocytic nevi. Hum Pathol 30: 500-505.
- Singh RK, Varney ML, Bucana CD, and Johansson SL (1999). Expression of interleukin-8 in primary and metastatic malignant melanoma of the skin. Melan Res 9: 383-387.
- Tronnier M, Smolle J, and Wolff HH (1995). Ultraviolet irradiation induces acute changes in melanocytic nevi. J Invest Derm 104: 475-478.
- Tsao H, Zhang X, Fowlkes K, Haluska FG (2000) Relative reciprocity of NRAS and PTEN/MMAC1 alterations in cutaneous melanoma cell lines. Cancer Res 60: 1800-1804.