Section 6 -

The Precursors of Malignant Melanoma Neil Crowson, MD Cynthia M. Magro, MD Martin C. Mihm, Jr., MD

Introduction Previously precursors to melanoma were generally considered to be nevi, either acquired or congenital. In recent decades the systematic study of patients with malignant melanoma has revealed the presence in a significant number of patients of precursor lesions which could be more specifically subclassified; a certain group of lesions was found that in themselves were precursors but the presence of which were a clinical sign that the patients harbored a genetic anomaly that placed them at an increased risk for the development of malignant melanoma and were carriers of susceptibility trait for the disease, now known as dysplastic nevi. We intend to elucidate recent advances in our understanding of melanoma precursors from the standpoint of the biological events and the histologic clues which help to predict malignant transformation. We will further discuss other precursors of the disease that include: lentigo maligna, the congenital nevus, mucosal melanocytoses of conjunctival, nasopharyngeal, penile, vulvar and gastrointestinal tract mucosa, and the atypical Spitz tumor (Crowson et al, 2001; Crowson et al., 2002). In the course of this discussion we must also consider those pigmented lesions which mimick melanoma and its precursors. References Crowson AN, Magro CM, Mihm MC, Jr. The melanocytic proliferations : a comprehensive textbook of pigmented lesions. New York : John Wiley and sons, 2001. Crowson AN, Magro CM, Sanchez-Carpintero I, Mihm MC Jr. The precursors of malignant melanoma. Recent Results Cancer Res 2002;160:75-84. Case # 6: Dysplastic Nevus. The Dysplastic Melanocytic Nevus and the Dysplastic Nevus Syndrome Lynch and co-workers (Lynch et al, 1978) coined the term familial multiple atypical mole melanoma syndrome for a symptom complex of multiple atypical nevi which placed family members at increased risk for developing melanoma in 1978. Clark and co-workers (Clark et al, 1978) characterized the dysplastic nevus as a defining element of this syndrome. Due to controversy surrounding the both the clinical and histologic definition of these lesion , the National Institutes of Health (NIH) consensus conference in 1992 recommended supplanting the term dysplastic nevus with the appellation "nevus with architectural disorder and cytologic atypia" (Crowson et al., 2001). In our view the original term dysplastic nevus is preferred because clinicians are familiar with this term suggested by Clark in his original article. Furthermore, it is clear that subgroups of congenital and acquired nevi (such as those in acral and genital sites) manifest architectural disorder and cytologic atypia but have no association with subsequent malignant melanoma. Also, the NIH term does not encompass grading which we believe is integral to assessment of any dysplastic nevus (Shea et al, 1999; Crowson et al., 2001). The incidence of dysplastic nevi is likely in the 5–10% range in Cauacasians (book; Nordlund et al, 1985). The concern raised by the dysplastic nevus relates to the inherent risk of transformation to malignant melanoma and to its being a marker for the development of melanoma at other sites. The risk of progression in an individual dysplastic nevus is unknown, but certain facts have been gleaned with regard to incidence. Melanoma patients with two or more dysplastic nevi may also be at increased risk for a second primary. Up to 92% of melanomas occurring in patients with dysplastic nevus syndrome have evidence of a dysplastic nevus precursor. Dysplastic nevi are precursors for up to 18% of all nonfamilial melanomas (Grob etal, 1988). Roughly 95% of malignant melanomas arising in dysplastic nevi are of superficial spreading type. Patients with two or more dysplastic nevi are said to have the dysplastic nevus syndrome that occurs sporadically or in a familial complex with an autosomal dominant pattern of inheritance, the latter most often encoded by a gene found on chromosome 9 at 9p21 (Cannon-Albright et al., 1994). In a given sibship, if one of the members with dysplastic nevi develops melanoma, other members with dysplastic nevi have a 100% lifetime probability of developing melanoma (Greene et al, 1985). Some patients with sporadic dysplastic nevus syndrome may resemble a person with familial dysplastic nevus syndrome by virtue of numerous large, atypical nevi. These patients may be the spontaneous mutants or be members of a family in which melanoma has not yet appeared or where history is unavailable. Most patients with sporadic dysplastic nevus syndrome present in the 4th-5th decades with only a few atypical moles on sun-exposed areas. UV radiation may play a role in the development of sporadic dysplastic nevi. Mutations and loss of heterozygosity of p16 and its alternate reading frame p14 ARF and p53 genes have been detected in blood lymphocytes from members of kindreds with hereditary cutaneous malignant melanoma, most of the mutations being of the C -> T transitional type known to be a signature for UV light-induced point mutation. One study showed areas of chromosomal loss at regions encoding for p16 (9p21–22) and p53 (17p13) in 78% of dysplastic nevi, with no loss of heterozygosity in benign intradermal nevi (Park et al., 1998); loss of heterozygosity at 9p21 appears to be restricted to melanoma and to dysplastic, as opposed to banal nevi (Brindelli et al,2000). The CDKN2A gene responsible for melanoma susceptibility in most families with melanoma linked to 9p encodes a cyclin-dependent kinase inhibitor, the dysfunction of which is also implicated in several sporadic cancers; the second most frequent cancer in such kindreds linked to CDKN2A gene mutations is pancreatic carcinoma, which occurs in up to 17% of patients (Vasen et al, 2000). Clinical Features Patients with familial dysplastic nevi develop multiple large atypical moles distributed everywhere on the body surface including the scalp, doubly covered areas (breasts of women and the bathing trunk area of men and women), and lower legs. Banal acquired nevi usually spare the scalp and are rarely found below the belt line. The characteristic features of the dysplastic nevus include: size greater than 6 mm. In diameter, irregular borders, variegated pigmentation in shades of brown dark brown and black and are either flat or slight elevated with a pebbly surface. Rarely, the lesions are tan and erythematous, a sign of inflammation often associated with more marked cytologic atypia in our experience. There is striking heterogeneity of the lesions, no one lesion looking exactly the same as the other. The nevi also form patterns, such as linear, arciform or even round arrays. Foci of hypopigmentation in the lesion appear in areas of partial regression. Halo dysplastic nevi may also occur. It must be emphasized that dysplastic nevi only a few millimeters in diameter occur as the lesions first appear. Even lesions of this small size show both the other clinical and the histologic features of the larger lesions. All these changes are in stark contrast to common acquired nevi that are less that 6 mm in size, exhibit smooth regular borders and regular patterns of pigmentation so that the lesions all resemble each other. Patients with multiple primary melanomas with or without familial dysplastic nevus syndrome may show a diffuse pattern of irregular pigmentation resembling freckling, but with an histology comprising intraepidermal lentiginous melanocytic dysplasia. Epiluminescence microscopy of dysplastic nevi reveals a pattern of patchy interruptions in the pigment network (the "broken network"), which is distinctive from common banal nevi and melanoma. We have established that the use of near infrared spectroscopy to probe molecular vibration of chemical bonds and so to assay tissue biochemistry non-destructively can be applied in-vivo to distinguish dysplastic nevi from banal nevi and from lentigines with a high degree of accuracy (McIntosh et al, 2001). There is no doubt that dysplastic nevi differ from banal nevi from all of clinical, histological, biochemical and molecular standpoints. Histology The histology of the dysplastic nevus is so reproducible that a diagnosis can usually be rendered on scanning magnification. Interobserver variability lies in the area of grading of atypia, which should not be done at scanning magnification as it requires assessment of cytology which can only be assessed at x40 or higher magnification. The constellation of histological findings in the dysplastic nevus encompasses two broad components: architecture and cytology (Clemente et al,1991). Major Criteria for Diagnosis of a Dysplastic Nevus Asymmetric basilar proliferation of nevomelanocytes along the dermoepidermal junction extending laterally beyond the confines of a preexisting dermal component if present. Cells have 1 or both of 2 characteristic cytological and architectural patterns: Lentiginous dysplasia: randomly disposed single cells are located along and between elongate rete with nests of varying sizes; nuclei are hyperchromatic, angulated and are similar in size to or larger than adjacent keratinocytes. Epithelioid dysplasia: epithelioid melanocytes are disposed in variably sized junctional nests as well as in a single-cell array along the dermoepidermal junction of an often normal or hyperplastic epidermis. The cells have round to oval nuclei with delicate chromatin, nucleoli, thick membranes, and diameters greater than those of adjacent keratinocytes. Rounded cytoplasmic contours encompass cytosols ranging from amelanotic to coarsely melanized with giant melanosomes. Minor Criteria for Diagnosis of a Dysplastic Nevus Papillary dermal collagen shows concentric eosinophilic fibrosis in which a dense zone of hypocellular collagen envelops rete ridges and/or lamellar fibroplasia in which delicate layers of collagen are interspersed with presumptive neural crest-derived facultative fibroblasts laying collagen along the tips of hyperplastic retia in parallel arrays. Lymphocytic infiltrates in the papillary dermis. Telangiectasia and/or vascular proliferation. Fusion of retia by confluent growth between adjacent melanocytic nests. Both major and two of four minor criteria must be met for the diagnosis to be made. Some lesions exhibit cytologic atypia without the architectural changes and vice versa. We suggest that the diagnosis be rendered as a junctional or compound nevus with atypia or a junctional or compound nevus with the architectural changes of the dysplastic nevus, respectively. Changes suggestive of dysplasia may be seen in other lesions (Toussaint and Kamino, 1999). 70% of congenital nevis show elongate rete with coarse periretal fibrosis, but the other changes of atypia, increased vascularity and patterned inflammation are not present. Similarly acral and genital nevi may at first glance resemble dysplasia, but should only be diagnosed as dysplastic when requisite criteria are met. Lentigo maligna often has focal reactive hyperplasia of the epidermal component and may suggest dyplastic nevus. Studies have shown that criteria for grading dysplasia can be learned and reproducibly applied by pathologists with a consistency that is maintained in both the three tier grading system which we employ, namely, mild, moderate and severe, or a two tier system, namely low and high grade dysplasia (Shea et al, 1999; Clemente et al., 1991; Murphy and Mihm, 1999). Some observers have found reasonable concordance in grading of architectural, but not cytologic features, implying that the interpretation of cytology is more challenging (Hastrup et al, 1994; Tannous et al, 2000). We grade atypia on the basis of both architecture and cytology, which are assessed separately, although architectural and cytologic grades of atypia tend to correlate. There is correlation between the degree of architectural and nuclear atypia (Shea et al, 1999). Readers are referred elsewhere for grading criteria (Crowson et al., 2001). Management At the first patient visit, we recommend that a clinically atypical nevus be excised to confirm the impression of dysplastic nevus. Any changing or suspicious lesion should be removed. It is our practice that excisional biopsy with a few millimeters margin is appropriate. If a partial biopsy has been performed, we advise removal of any clinically apparent residuum with a few millimeter margin. If slight atypia is present at a histologic margin without a clinically-evident residuum, we do not advise reexcision. If moderate atypia is present at a margin, we advise conservative reexcision with a few millimeter margin. If severe dysplasia is present at a margin, or if a margin is only clear by a millimeter or two, we advise reexcision to obtain a 5 mm margin of normal skin. Follow-up of any patient is dependent on the number of lesions and the degree of clinical or histological atypia. Photographic documentation is desirable. References Crowson AN, Magro CM, Mihm MC, Jr. The nevomelanocytic proliferations : a comprehensive textbook of pigmented lesions. New York : Wiley-Liss, 2001: 550pp. Lynch HT, Frichot B, Lynch JF. Familial atypical multi-mole melanoma syndrome. J Med Genet 1978;15: 352–356. Clark WH, Reimer RR, Greene M, Ainsworth AM, Mastrangelo MJ. Origin of familial malignant melanomas from heritable melanocytic lesions. "The B-K mole syndrome." Arch Dermatol 1978;114: 732–738. Shea CR, Vollmer RT, Prieto VG. Correlating architectural disorder and cytologic atypia in Clark (dysplastic) melanocytic nevi. Hum Pathol 1999;30:500–505. Nordlund JJ, Kirkwood J, Forget BM et al. Demographic study of clinically atypical (dysplastic) nevi in patients with melanoma and comparison subjects. Cancer Res 1985;45:1855–1861. Grob JJ, Andrac L, Romano MH, et al. Dysplastic naevus in non-familial melanoma. A clinicopathological study of 101 cases. Br J Dermatol 1988;118:745–752. Cannon-Albright LA, Meyer LJ, Lewis CM et al. Penetrance and expressivity of the chromosome 9p melanoma susceptibility locus (MLM). Cancer Res 1994;54: 6041-4. Greene MH, Clark WH, Tucker MA, Kraemer KH, Elder DE, Fraser MC. High risk of malignant melanoma in melanoma-prone families with dysplastic nevi. Ann Intern Med 1985;102: 458–465. Park WS, Vortmeyer AO, Pack S, et al. Allelic deletion at chromosome 9p21(p16) and 17p13(p53) in microdissected sporadic dysplastic nevus. Hum Pathol 1998;29: 127–130. Birindelli S, Tragni G, Bartoli C et al. Detection of microsatellite alterations in the spectrum of melanocytic nevi in patients with or without individual or family history of melanoma. Int J Cancer 2000;86: 255-261. Vasen HF, Gruis NA, Frants RR, van Der Velden PA, Hille ET, Bergman W. Risk of developing pancreatic cancer in families associated with a specific 19 bp deletion of p16 (p16-Leiden). Int J Cancer 2000;87: 809-811. McIntosh LM, Summers R, Jackson M et al. Towards non-invasive screening of skin lesions by near-infrared spectroscopy. J Invest Dermatol 2001;116:175-181. Clemente C, Cochran AJ, Elder DE, et al. Histopathologic diagnosis of dysplastic nevi: concordance among pathologists convened by the World Health Organization Melanoma Programme. Hum Pathol 1991;22: 313–319. Toussaint S, Kamino H Dysplastic changes in different types of melanocytic nevi. A unifying concept. J Cutan Pathol 1999;26: 84–90. Murphy G, Mihm MC Jr. Recognition and evaluation of cytological dysplasia in acquired melanocytic nevi. Hum Pathol 1999;30(5):506-512. Hastrup N, Clemmenson OJ, Spaun E, Sondergaard K. Dysplastic naevus : histopathologic criteria and their inter-observer reproducibility. Histopathology 1994;24: 503–509.