Barnhill, Raymond L. MD*; Cerroni, Lorenzo MD†; Cook, Martin MD‡; Elder, David E. MD§; Kerl, Helmut MD†; LeBoit, Philip E. MD∥; McCarthy, Stanley W. MD¶; Mihm, Martin C. MD♯; Mooi, Wolter J. MD**; Piepkorn, Michael W. MD, PhD††; Prieto, Victor G. MD, PhD‡‡; Scolyer, Richard A. MD¶
A one-day Workshop of the International Melanoma Pathology Study Group devoted to the histology of benign melanocytic nevi (Table 1) was convened on March 3, 2009 in San Francisco, California at the University of California San Francisco with the goal of assessing the current state of knowledge on this subject, with particular reference to nomenclature, points of consensus and controversy, and future directions. The Workshop took place at a multi-headed microscope and was organized as interactive focus sessions on the specific topics that follow below, with the simultaneous viewing of glass slides. The Proceedings of this Workshop, as recorded below, reflects the collective experience of the participants.
COMMON ACQUIRED MELANOCYTIC NEVI
It was emphasized during the Group discussion that some of the factors determining the baseline histologic phenotype of common acquired nevi include age and programmed natural history; regional microanatomy; genetic factors; physiologic, hormonal, and immunologic factors; environmental factors such as sun exposure, trauma, and other insults; and other poorly understood factors.1–8 None of the participants took issue with the standard nomenclature of common acquired nevi as junctional, compound, and dermal melanocytic nevi and the established variants of common nevi (Table 2). The discussion concentrated on the most controversial aspects of common nevi and attention was focused on halo nevi, traumatized nevi, and nevi from particular anatomic sites.
Halo Melanocytic Nevi
There was general agreement that the terminology of halo nevus (or halo melanocytic nevus) historically is based on the presence of a clinical “halo” and associated histologic lymphocytic infiltrates in its fully developed stage and that “nevus with halo reaction” is preferred for nevi lacking a prototypic clinical halo (or when the clinical appearance is unknown).1–9 The participants recognized that halo nevi and nevi with halo reactions may show significant distortion of architecture, cytologic atypia of the junctional and dermal melanocytic components, and mitoses in the dermal component. The progressive effects of lymphocytic reaction may significantly distort the architecture of halo nevi in fairly characteristic patterns. These alterations may make the distinction from melanoma much more difficult, particularly if there is some cytologic atypia of the dermal component. These resulting patterns may include hypercellular (strikingly crowded) junctional nesting of melanocytes and progressive asymmetry resulting from an asymmetrical inflammatory regression of the nevus (Fig. 1). There was discussion among the Group that such cytologic atypia of melanocytes may be indicative of a background atypical or “dysplastic” nevus. The latter melanocytic atypia may potentially be graded as for conventional “dysplastic” nevi, with the understanding that no universally accepted or standardized grading system currently exists (see discussion below concerning “dysplastic” nevi). However, it was also pointed out that relatively low-grade cytologic atypia and possibly dermal mitoses may have a reactive or regenerative basis secondary to the effects of the lymphocytic infiltrate in producing or be a consequence of melanocyte injury and death. Thus, one is left with the problem of how to distinguish neoplastic cytologic atypia in “dysplastic” nevi from reactive atypia in halo nevi. One suggestion offered was that if the halo nevus shows the typical architecture (diameter >5 to 6 mm, irregular junctional nesting, bridging of junctional nests, etc) of a conventional “dysplastic” nevus, then the cytologic atypia in such a nevus is likely to be “dysplastic” rather than reactive. However, it is certain that the latter issues remain controversial.
Traumatized Melanocytic Nevi
There was general consensus that the effects of external trauma to melanocytic nevi probably mimic closely or are indistinguishable from the features observed in recurrent melanocytic nevi as has been emphasized in the literature. Furthermore, it was agreed that the frequency and effects of trauma are probably much more common than has been appreciated, nevi on exposed sites are subject to repetitive trauma, and that it is often difficult to clearly document trauma to melanocytic nevi.7 There was agreement that trauma-induced alterations may render many melanocytic nevi highly suspicious and sometime indistinguishable from melanoma (Fig. 2), and finally that this topic merits much greater scrutiny and study. Pertinent points raised with respect to this subject included the facts that such effects are often fairly focal and they are often in the central zone but may clearly involve any part of the nevus. Traumatic effects at the periphery of a nevus may be more difficult to recognize than in the center of the nevus. The spectrum of alterations is large and often may include any of the following: parakeratosis, hyperkeratosis, hypergranulosis, pagetoid scatter of melanocytes (commonly beneath parakeratosis and focal), ulceration, variation in epidermal thickness (often acanthosis), effacement of the epidermis, asymmetry, poor circumscription, irregular and disordered (and sometimes confluent) proliferation of melanocytes along the dermal-epidermal junction and in the dermis, expansion of the papillary dermis by fibroplasia, focal absence of melanocytes at the junction and in zones of fibroplasia, enlargement of melanocytes, reactive anisokaryosis, and occasional dermal mitoses.
In general, it is apparent that traumatic insult to melanocytic lesions has not received the attention this topic deserves and that more detailed study with clinicopathologic correlation is needed.
Melanocytic Nevi From Particular Anatomic Sites (Melanocytic Nevi From “Special Sites”)
The participants were in general agreement that the anatomic location of melanocytic nevi may significantly influence their histology and, importantly, may introduce particular histologic properties that require distinction from melanoma.1–7,10–19 The Group members believed that some subset of nevi from acral, genital (especially vulvar), other flexural sites such as the umbilicus, and the scalp region may exhibit particular histologic features that raise concern for melanoma. The findings often observed in such nevi may include pagetoid melanocytosis and lentiginous melanocytic proliferation (particularly in acral nevi), junctional nests of melanocytes that are enlarged, irregularity of junctional nesting with horizontal bridging or confluence of nests, hypercellular (crowded) junctional nesting, large round or polygonal melanocytes in the junctional component, anisokaryosis of melanocytes, and papillary dermal fibroplasias (Fig. 3).
Unfortunately, at present, there are no sufficiently large, comprehensive, and rigorously controlled studies that provide any definitive conclusions about the differences in baseline histology among nevi from many different anatomic sites in relationship with age, sex, considerable heterogeneity of nevi, and other factors. As the latter data are lacking, the fundamental problem once again is how common acquired nevi, atypical or “dysplastic” nevi, and nevi from “special sites” differ.7,20–41 Ultimately, the issue is clarifying and quantifying the biologic significance of the various phenotypic properties of melanocytic nevi such as their risk for progression to melanoma or as an indicator of the risk of melanoma for the individual with such a nevus. Many of the studies reported thus far have not addressed inherent biases such as the origin of study participants, for example, population-based versus various referral biases, how nevi have been selected for removal, and insufficient numbers of study participants.
ATYPICAL OR “DYSPLASTIC” MELANOCYTIC NEVI
A focus session at the Workshop addressed the topic of the dysplastic nevus and its relationship to melanoma. Various viewpoints emerged, reflective of the fact that dysplastic nevi have been the subject of ongoing controversy for more than 20 years.2–7,20–41 There was thus little or no expectation, a priori, that uniformity of opinion would emerge on the broader issues in this brief session, given their contentious history, but there did emerge a general recognition of the problems and limitations of dysplastic nevi as a risk indicator for melanoma, as well as discussions of the items for further study.
It was recognized by most participants that melanoma and nevi are related phenotypes, often with overlapping morphologic features, which is likely reflective of shared etiologic factors such as genetic predisposition, sun exposure and susceptibility, and other less defined environmental factors. As such, the nevus phenotype and melanoma could be considered pleiotropic manifestations of shared proximal stimuli. The relationship of nevi and melanoma has most persuasively been shown in the epidemiologic literature, wherein multiple case-control studies show a positive, near-linear relationship between numbers of nevi and lifetime relative risk for melanoma. Similar correlations exist between large nevi and melanoma risk.
Although there are fewer disputes that clinically atypical nevus phenotypes constitute risk indicators for melanoma, the proposition that the dysplastic nevus, defined histologically, represents a risk indicator and a potential precursor lesion of melanoma is significantly contentious. Objective information to assist with resolution of this problem with a reasonable degree of certainty may be assisted by the application of new and emerging molecular technologies to the problem.
At the more practical level of daily dermatopathology practice, discordance between skilled observers in the application of the published criteria for the dysplastic nevus remains a currently intractable issue. During the Workshop, it was clear that the various histologic criteria for dysplasia are identifiable by most, if not all, participants in histologic sections. Where much of the difference of opinion occurs, and where much of the basis for poor interobserver concordance probably resides, is at the level of assigning significance to the various criteria. In part, this is related to the fact that many of the criteria can be found at various degrees of expression in otherwise clinically and histologically banal nevi, reflective of a significant limitation in the specificity and predictive value of the criteria. What to one participant at the Workshop denoted high-grade cytologic or architectural dysplasia was to another participant a variation on the theme of histologic normality in a nevus (Figs. 4, 5). Fundamental differences of opinion regarding the intrinsic significance of the findings, either as a reflection of normal nevus physiology or as indicators of neoplastic progression, will ensure ongoing disparities in viewpoints. The intrinsic subjectivity in the application of the criteria has similar implications.
Although no uniformity of opinion emerged from the Workshop, the majority viewpoint holds that the histologic criteria for dysplasia, in particular the low-grade or “mild” range of expression, is a common occurrence in dermatopathology practice and is unlikely to constitute on its own merit a clinically significant observation regarding relative risk for melanoma. The Group reviewed data and conclusions from a Seattle, WA, case-control study,41 and there was no serious dispute that higher grades of histologic dysplasia (eg, “moderate grade” dysplasia and greater) could constitute statistically significant risk indicators, albeit with a potency less than that conferred by some other established phenotypic traits associated with melanoma risk, for example, green or blue eye color, large nevi, increased numbers of nevi, and clinically dysplastic nevi (in rough order of potency).38,41
The principle of improving histologic observer reproducibility and sensitivity for risk of melanoma was discussed at the meeting in the context of an empirical approach to the problem. The methodology discussed was that of using an iterative method of identifying, a posteriori, those observers whose interpretations are more tightly correlated with melanoma risk categories in case-control data sets and then to “train” a panel of other observers in the specific nuances of histologic criteria and their application; the objective, accordingly, is to improve group performance in the recognition of histologic criteria that in their collective correlate as more robust risk indicators. After the discussion in the Workshop, there emerged no substantive dispute that such an approach could be useful as an investigative technique to establish proof of concept, but there remained uncertainty as to the ultimate utility of the data emerging from that approach in dermatopathology practice.
On review of the microscopic sections showing varying degrees of histologic dysplasia from melanoma patients and from control subjects, the preponderant opinion among the participants is that there is no clear qualitative difference in histologic dysplasia between individuals who have had versus those who have not had (and will likely never have) melanoma. In contrast, there was no serious dispute that dysplasia may be somewhat more common among melanoma patients than among control populations. This association could be stronger for the higher grades of histologic dysplasia, but more rigorous investigative data are necessary for definitive conclusions.
It is theoretically possible, as held by many in the Study Group, that molecular transformation may occur as a stochastic process in dysplastic, as well as in histologically banal nevi, leading to the development of melanoma; the fundamental question, assuming the validity of the premise, however, is the quantitative frequency of occurrences of this type. Although no highly reliable data are available, the consensus of opinion in the Group is that such events, although unpredictable in their occurrence, are rare in any given nevus, and that wholesale excision of nevi is not appropriate as a means of reducing melanoma incidence. A more practical issue, however, stems from the collective experience that certain of the histologic findings of dysplastic nevi are shared with melanoma, and it is fairly common that difficulties arise in the subjective application of what are, fundamentally, “soft” criteria. Thus, the question is generally not that of whether a specific dysplastic nevus is a significant precursor lesion, but rather whether the lesion in question could possibly represent a histologically subtle melanoma closely mimicking a dysplastic nevus. Difficulties in resolving this differential diagnosis with reliable certitude are responsible for many of the recommendations found in dermatopathology reports for the precautionary resection of melanocytic lesions with dysplastic or atypical features, especially those at the more atypical end of the spectrum. Most, although not all, of the participants at the session tacitly acknowledged the problem.
It remains for future investigations to determine whether high degrees of melanocytic dysplasia can correlate under rigorous scientific design with melanoma, either as a risk indicator or as a putative precursor lesion. There was no general dispute among the participants with the concept that pathologic processes, as with biologic processes, are developmental in nature. Accordingly, there is no reason to summarily refute the proposition that future molecular techniques could be fruitfully brought to bear on the problem, with the objective of defining markers with better predictive value than can be achieved by simple histologic assessment, which, after all, is merely an opinion as to diagnosis in a patient as inferred from observation. It is, moreover, altogether plausible that molecular processes occurring within nevi, whether dysplastic or not, are relevant to understanding malignant transformation in melanocytic systems. A detailed understanding of the process is, of course, a prerequisite for the ultimate goal of designing selective pharmaceutic therapies for the disease, which target the specific defects leading to the development of the fully evolved malignant phenotype.
CONGENITAL MELANOCYTIC NEVI AND RELATED LESIONS
This focus session involved discussion of the nomenclature and histologic aspects of congenital nevi. In general, the most controversial aspects concerning congenital nevi relate to their developmental biology and frequency of malignant transformation.2–7,42–52
By definition, a congenital nevus is present at birth. Documentation of its presence at birth therefore constitutes the sole direct evidence that a nevus is congenital. However, as acquired nevi seldom reach a size of more than 1 cm in diameter, whereas congenital nevi often do and may even cover a major part of the body, large nevus size is generally accepted as evidence of the congenital nature of a nevus (Table 3). However, there are occasional problems with size as a substitute parameter. True congenital nevi may stay small, and conversely, some acquired nevi, especially those manifesting in early infancy, may reach a size of several centimeters. This has led some to regard such lesions as “tardive congenital nevi,” that is, congenital nevi, which for some reason were not clinically apparent at birth. This introduces a circular argument and rules out the possibility that some acquired nevi reach a size that is generally associated with the congenital nature of a nevus. In addition, blue nevi and their variants, and nevi arising in blistering skin disorders, may become significantly larger than the usual acquired nevus.
It could be argued that the parameter of clinical manifestation at birth does not provide an ideal parameter for dividing melanocytic nevi into 2 distinct groups. Instead, it could be said that when all melanocytic nevi are taken together, those arising early in life tend to reach a large, sometimes very large, size, whereas those arising later on, tend to reach progressively smaller final sizes. Nevi that are not apparent at birth but manifest in infancy may thus constitute a middle group in terms of age at presentation and final size.
With respect to size, congenital nevi are commonly subdivided into small, medium-sized, and large or giant, nevi. Size subdivisions based on an absolute size scale (the 1.5 and 20 cm size limits being used most commonly to delineate medium-sized nevi from small and large ones) have the disadvantage that patient size increases substantially, so that a 15 cm nevus in a neonate covers a much larger part of the body than a 15 cm lesion in an adult. However, generally, it is accepted that a congenital nevus that would reach 20 cm in size in an adult (even if smaller in an infant) be regarded as a large congenital nevus and has a significant risk (2% to 7%) of malignant transformation that often occurs (in more than 50% of cases) before puberty. In addition, as congenital nevi tend to grow proportionally with the growth of the patient, a division based on absolute size means that individual nevi change from small to middle-sized, or middle-sized to large, as the patient grows.
Some large congenital nevi show symmetrical or otherwise remarkable shapes that are difficult to explain on the basis of local spread only. Perhaps, the oncogenic mutation resulting in its formation has occurred in a melanoblast or neural crest precursor cell before its migration and its descendants, along the dorsolateral pathway to the skin, so that substantial body areas may become seeded with the descendants of the precursor cell harboring the oncogenic mutation, most commonly NRAS, and thus result in nevus outgrowth in that particular area.50,52
The histologic features will be discussed below, but at this stage it should be noted that there is no obvious qualitative difference between congenital nevi of different sizes although the risk of malignant transformation is greater in large congenital nevi. Large lesions tend to extend more deeply and may exhibit greater variation in histologic appearances, but these differences are a matter of degree rather than a firm basis for rational subdivision. It is of interest, though, that large congenital nevi seem to harbor NRAS mutations more frequently than small nevi, whereas the converse seems to be true with regard to BRAF mutations (the mutations far more commonly present in common acquired nevi).
Histology in Early Infancy
For cosmetic reasons, and to minimize the—generally small—chance of malignant transformation later in life, some surgeons resect congenital nevi in early infancy. Resections may be full thickness or may remove only the superficial parts of the nevus.
Histologically, congenital nevi of early infancy usually show the features of a compound nevus of varying and often high cellularity, and with a degree of variation in cell size, shape, pigmentation, and arrangement that exceeds the variation seen in most common acquired nevi. These variations need not lead to a suspicion of malignancy. Even the presence of some pagetoid intraepidermal spread of nevoid melanocytes—a feature most commonly seen at the edges and in satellite lesions of giant congenital nevi, particularly in infancy—is entirely acceptable and should not lead to a diagnosis of melanoma, or in situ melanoma.
The deeper part of the nevus may show a diffuse arrangement of small nevoid melanocytes throughout the dermis and often the subcutis or even deeper tissues, but commonly there are variations, so that some areas are pigmented, others show a nonpigmented spindle cell type, yet other areas show ballooning, or densely packed sheets of small or medium-sized cells with some mitotic activity. Areas of extensive neuroid change commonly feature pseudotactile corpuscles. Some areas of blue nevus or Spitz nevus may be formed. Very occasionally, chondroid or osseous metaplasia may be noted. Some congenital nevi are covered with hairs, presumably resulting from some paracrine effect of the neoplastic nevoid melanocytes on the entrapped vellus hairs.
Dermal Proliferative Nodules in Infancy
Not uncommonly, congenital nevi develop several papules and nodules consisting of mitotically active enlarged nevoid melanocytes. Their mitotic activity together with the larger cell size, which stands out from the background of smaller nondividing nevoid cells, may lead to a histologic suspicion of malignancy. However, almost without exception, such nodules are benign (Fig. 6).6,47–51 Clues for benignity are very young age of the patient (although similar lesions may be seen in older patients), superficial location (the few melanomas that arise in congenital nevi during infancy are generally deeply located, sarcoma-like, massively cellular tumors), multiplicity (melanoma arising multiply is vanishingly rare), gradual transition/merging of the atypical cells in the proliferative nodule with those of the surrounding nevus (rather than 2 distinct cell types in melanoma arising in congenital nevi), absence of an epidermal component with pagetoid spread, absence of high-grade nuclear atypia and, mostly, necrosis. Interestingly, the proliferative nodules were shown to harbor numerical abnormalities of whole chromosomes rather than the far more complex genotypes seen in melanoma.50 Future work will show whether this cytogenetic finding may be applied in a diagnostic setting in diagnostically equivocal problem cases.
Histology Later in Life
Later in life, the congenital nevus tends to become somewhat less heterogenous in histologic appearance, but to some degree, the variations mentioned above may persist. The risk of malignant transformation to melanoma is small—it is estimated at 2% to 7% in the case of a large nevus; the risk in small examples presumably being smaller—but real, so that an area of change in a previously stable congenital nevus should be removed and investigated histologically. Melanoma arising in a congenital nevus usually resembles melanomas arising in acquired nevi or previously normal skin, so that the same histologic criteria can be employed.
SPITZ NEVI AND RELATED VARIANTS
Over more than 60 years, there has been an evolution in terminology that has often reflected the current thinking about the biologic nature of this group of lesions.2–8,53–72 Thus, various terms have included juvenile melanoma, epithelioid cell nevus, spindle cell nevus, spindle and epithelioid cell nevus, large spindle and/or epithelioid cell nevus, Spitz tumor, Spitz's nevus, and Spitz nevus. Although the biologic nature of these lesions, particularly unusual variants, and those lesions difficult or impossible to differentiate from melanoma await more definitive characterization, there was consensus among the Group that Spitz nevus or Spitz tumor is an acceptable term for this general category of lesions for the time being. Although these lesions undoubtedly share biologic properties with “conventional” melanocytic nevi, they are also, at the same time, distinctive melanocytic neoplasms, suggesting the need for some qualification such as spitzoid melanocytic tumor, neoplasm, or melanocytoma, particularly for those with atypical features.
Spitz Nevus/Tumor Variants: Consensus
Although the Spitz nevus is a well-established entity among melanocytic neoplasms, the clinical and histopathologic features that parenthetically define this tumor are largely derived from a relatively limited number of studies that all too frequently have suffered from selectional or referral biases. There is still the need for a large multicenter study that should be, as much as possible, population-based and should involve review by a diverse panel of melanoma pathologists with considerable experience to more precisely define the spectrum of these lesions and various proposed variants (described in the literature).
As outlined in Table 4, there was a general agreement among the Workshop participants that particular variants of Spitz nevus are well established in the literature. Among the conventional or typical Spitz nevi, junctional, compound, and dermal variants exist and to some degree may be related to anatomic site, stage in the natural history of the lesion, and to a lesser degree age of the individual; however, the evidence for the latter observations has not been rigorously documented. Such lesions usually exhibit the characteristics of small diameter (usually less than 5 to 6 mm), symmetry from side to side, sharply delimited peripheral margins, often regular epidermal hyperplasia, vertically oriented junctional nests of melanocytes, dispersion or so-called maturation with descent into the dermis, and the typical cytomorphologic features (Fig. 7).2–8,53–56
Distinctive variants that are generally accepted by the Group include desmoplastic Spitz nevi,2–7,57–59 pigmented spindle cell variants,2–8,60–63 halo Spitz nevi,64 biphasic or combined variants,2–7 recurrent/persistent Spitz nevi,65 and finally Spitz nevi/tumors (or neoplasms) with deviant or atypical features (Table 4).2–8,56,66–72 There is general acceptance that often predominantly or entirely dermal desmoplastic Spitz nevi are characterized by a desmoplasia (sclerosis or collagenization) of the dermis and sometimes subcutaneous fat that contains the spitzoid melanocytes. Such lesions show varying degrees of cellularity, often with individual or small aggregates of melanocytes sequestrated in dense collagen. A proportion of these desmoplastic lesions commonly harbor a concurrent population of small ovoid melanocytes that may represent small conventional nevus cells or small spitzoid melanocytes resulting from maturation. The 2 populations of melanocytes may be closely intermingled or topographically distinct. There is general agreement that pigmented spindle cell variants of Spitz nevus are distinctive and the term pigmented spindle cell nevus is acceptable to designate these lesions (see discussion below). Halo variants of Spitz nevi are defined by a distinct clinical halo and associated dense lymphocytic infiltrates. As discussed above Spitz nevi with 2 or more distinct populations of melanocytes are accepted as legitimate entities but probably require rigorous attention as to how they are defined. Recurrent or persistent Spitz nevi/tumors after apparent incomplete removal are now clearly recognized and accepted by the Group. However, considerable caution should be exercised in these circumstances as to the limitations of histologic interpretation of such “recurrent” lesions without comprehensive evaluation of the individual lesion and rigorous and long-term follow-up of the patient for an adverse outcome. Finally, there is general agreement that Spitz tumors (or neoplasms) with a spectrum of atypical or deviant features exist. However, as discussed below, the challenge to pathologists and clinicians alike is to accumulate sufficient knowledge about this spectrum of lesions so as to better or more fully understand their biologic natures, and develop appropriate terminologies and management strategies for patient care (see below).
Spitz Tumors: Controversies
There remain certain unresolved problems about the nosology and final definitive characterization of Spitzoid lesions. Although there is agreement about the recognition of pigmented spindle cell variants of Spitz nevus, there is nonetheless controversy about the thresholds for their recognition and distinction from “conventional amelanotic” Spitz nevi. The problem can be briefly stated as follows: a percentage of relatively pigmented Spitz nevi show overlapping topography and cytology with classic pigmented spindle cell nevus (PSCN). Thus, such lesions may show the general flattish papular and mainly junctional configuration associated with PSCN and harbor a spectrum of variably pigmented spindled and epithelioid melanocytes of varying sizes. The latter group of lesions is difficult to definitively classify as PSCN or Spitz nevi because of the features shared in common. One potential solution is a more restrictive definition of PSCN such as inclusion of lesions characterized by a relatively pure population of small uniform pigmented spindle cells only in the typical architectural setting.
Atypical or Ambiguous Variants of Spitz Tumor
Unusual or atypical variants of Spitz tumor may show a number of morphologic characteristics often observed in or mimicking melanoma or other atypical melanocytic lesions (Table 4).2–8,56,66–72 What to make of such abnormal or deviant morphologic characteristics has remained the subject of considerable debate as their biologic significance in many such ambiguous spitzoid lesions. One perspective holds that such abnormal findings are, for the most part, simply variations in the spectrum of Spitz tumors. In contrast, another perspective maintains that the progressive accumulation of certain abnormal features may have some correlation, albeit imperfect, with neoplastic recurrence or progression of Spitz tumors. Thus, methods have been suggested for the risk stratification of Spitz tumors exhibiting a range of abnormal properties.69 It was the general perspective of the Workshop participants that such controversial spitzoid lesions remain so poorly understood as to preclude any definitive statements as to their biologic nature other than that they require rigorous study and, in general, neoplastic progression and an adverse outcome cannot be ruled out.
Unusual, atypical, or deviant Spitz tumors may be characterized with respect to alterations of either the intraepidermal components, dermal components, or both (Table 4) (Figs. 8, 9). Thus, variants with striking pagetoid melanocytosis or pagetoid Spitz nevus/tumor and junctional configurations suggesting conventional atypical or “dysplastic” nevi may be observed on occasion.6–8,66–72 However, abnormalities of the dermal component have received more attention because of their potentially greater biologic import. Consequently, unusual Spitz tumors with significant dermal components manifesting asymmetry, large diameter (>6 and especially >10 mm), significant thickness (particularly subcutaneous extension), lack of “maturation” and nodule formation, and mitotic rates >6/mm2 (or in another study >2/mm2) have been categorized as atypical or biologically ambiguous Spitzoid neoplasms. Other characteristics including age of the patient, ulceration, cytologic atypia, and adjunctive studies such as assessment of mitotic index, other biologic markers, and analysis of chromosomal aberrations may aid in this evaluation.
Among the substantial controversies surrounding Spitz nevi/tumors has been the introduction into the literature in recent years of the term “Spitzoid” melanoma. Although this term is meant to clearly conceptualize a variant of melanoma that closely resembles a Spitz nevus/tumor, the fact of the matter is that the term has largely become meaningless because of its imprecise usage in the literature. For example, the term has been applied to a spectrum of benign, malignant, and poorly characterized “spitzoid lesions” in children and adults. For the time being, it is recommended that the term be abandoned until potentially a subset of melanomas can be rigorously defined by objective phenotypic and genotypic parameters and natural history as Spitzoid melanoma. It may, however, be useful at times for descriptive and nosologic reasons to designate a Spitzoid lesion judged to be a melanoma as having “spitzoid features.” In addition, several panelists advocated the use of descriptive terminology such as “Melanocytic Tumor of Uncertain Malignant Potential” or “MELTUMP” as a means of frankly expressing the uncertainty that is attached to the diagnosis in many of these cases (Fig. 9).2
Conclusions and Future Directions
Although guidelines for evaluating and managing controversial spitzoid lesions have been suggested, considerable controversy and angst surround the approach to such lesions. The overall feeling of the Group is that such spitzoid lesions should be approached as having some potential for recurrence and possibly malignant behavior. In general, these lesions should be completely resected and some consideration be given to managing them as for conventional melanoma of equivalent Breslow thickness with wider margins of up to 1 cm or more for selected lesions. Sentinel lymph node biopsy, although controversial for these lesions in general, and additional measures may, on a case-by-case basis, be discussed with the patient (and parents if a child) for selected lesions. However, the clinical significance of sentinel lymph node deposits, particularly small deposits in such cases, remains unknown, and several studies, albeit as yet with relatively limited follow-up, have shown an excellent survival experience for patients with positive sentinel nodes and a Spitzoid primary tumor.72
The principal challenge to the community is to accumulate sufficient numbers of various types of Spitzoid tumors and to execute comprehensive objective characterization of this material with specific end points and long-term follow-up of patients ideally through a single international cooperative study. The outcome of the latter type of study may clearly allow for a more straightforward nomenclature and risk assessment of individual lesions. Such a study should be funded by national cancer research agencies. Until this is accomplished, histopathologists and clinicians alike will continue to struggle to make sense of spitzoid lesions and how to manage patients with these lesions and the inexplicable adverse outcomes that occur from time to time.
BLUE NEVI AND RELATED LESIONS
A focus session on blue nevi revealed that there was general consensus among many participants about the nomenclature and characteristics of the most frequently encountered variants of blue nevus; for example, the common or Tiéche-Jadassohn variant, cellular blue nevus, etc (Tables 5, 6).2–8,73–89 However, it became clear that there is considerable controversy surrounding other proposed variants of blue nevus, particularly those suggested to have a pigmented epithelioid cell phenotype,87 the recently described entity “pigmented epithelioid melanocytoma,”88,89 deep-penetrating nevus,2–7,90–92 so-called combined variants of blue nevus,93–96 and finally the biologic potential of many of the latter lesions.97,98 In fact, as pointed out by 1 participant, the classification and nomenclature of blue nevi remain controversial as many melanocytic neoplasms (correctly or incorrectly) classified as blue nevi are extremely heterogeneous and may show a diverse spectrum and admixture of melanocytes such as dendritic cells, small nevoid cells, “pagetoid” cells (defined as a large round melanocyte as observed in “pagetoid” or “superficial spreading” melanoma), spindle cells, oval cells with small nuclei, epithelioid cells, and even other cell types. Thus, the essence of the problem is one of definition and how restrictive or liberal are the criteria for classifying candidate melanocytic lesions as “blue nevi.” It was evident that many participants would place many of the latter entities with 2 or more distinct components of melanocytes into the general category of combined nevus (see the section on Combined Nevus and Related Lesions).
The fundamental definitions of a blue nevus suggested by 1 participant and many others historically are (1) clinical: blue coloration and often congenital onset and (2) histologic: the presence of at least some pigmented bipolar dendritic melanocytes in the dermal lesion and often accompanied by melanophages. The latter definition is that of the classic common or dendritic blue nevus (Tiéche-Jadassohn) (Fig. 10). An alternative and controversial classification of blue nevi was proposed by one of the participants (Table 6) who maintained that virtually all blue nevi outside of the common variant show mixed cytomorphologic features and are therefore “combined” variants of blue nevus (Table 7). According to this perspective, the most frequent cell type observed in “combined” blue nevi is an epithelioid melanocyte (Fig. 11). Other investigators have observed that a common participant in a combined nevus is the deep penetrating nevus (see below).
The latter proposed classification of blue nevi clearly accommodates a certain number of melanocytic lesions that would be considered as either combined nevi (see the section on Combined Nevus and Related Lesions) or would remain controversial as to their nomenclature and classification by other members of the Group. An alternative reductionist approach is simply the description of the cellular composition of a melanocytic neoplasm under the rubric of melanocytic nevus or neoplasm with phenotypic heterogeneity and specification of the cellular populations present (see the section on Combined Nevus and Related Lesions below).6
Blue Nevus With Atypical Features or Uncertain Malignant Potential
It was nonetheless emphasized that most melanocytic proliferations with features of blue nevi can be correctly diagnosed and should be called what they are: either a benign blue nevus or a melanoma.2–7,97,98 However, there is a group within this spectrum, which may pose diagnostic problems. These “atypical” blue nevi reveal overlapping features of both blue nevi and blue nevi-like melanomas. All the combined blue nevi, listed in Table 7, may show atypical features and the diagnosis can be difficult.
Atypical Blue Nevi Possibly Represent Low-grade Malignant Melanocytic Tumors
Many cases reveal lymph node involvement; distant metastases are, however, rare. It seems that their prognosis is usually favorable in comparison with “conventional” melanomas. Unfortunately, there is still confusion about the clear definition and the biologic behavior of these atypical blue nevi (Fig. 12).
The outcome from this focus session clearly illustrated that there remain significant problems of nomenclature and classification among this group of lesions that cannot be resolved without protracted efforts to reach consensus at the microscope and perhaps resolution at the molecular level.
Pigmented Epithelioid Melanocytoma
This somewhat controversial term has been proposed as a unification of previously described lesions termed “animal” or “macrophagic” melanomas characterized by heavy pigment synthesis and a resemblance to certain tumors of animals.4,49,88,89 There is also overlap with the epithelioid blue nevus seen in the Carney complex. The lesions seem to have low-grade potential for malignancy in the form of both regional and occasional nonregional metastases, but rarely cause death. It has been proposed that these lesions should be termed “pigment-synthesizing melanomas,” with the recognition that they seem to be “low grade,” but the recent demonstration of a specific mutated oncogene protein kinase A regulatory subunit 1 α in them suggests that they are a distinct entity.89
COMBINED NEVUS AND RELATED LESIONS: MELANOCYTIC NEVUS WITH PHENOTYPIC HETEROGENEITY
A focus session was devoted to melanocytic nevi that appear to harbor 2 or more distinct populations of melanocytes. Historically, the term “combined” nevus has been coined to describe this general category of melanocytic nevi and the participants generally had no objection to the general use of this term (Table 8).2–7,93–96 As mentioned in the above section, others prefer a reductionist approach and the term melanocytic nevus with phenotypic heterogeneity, with specification as to the melanocytic components present (Table 9).6,7 This terminology may be a more accurate reflection of the pathogenesis of these melanocytic tumors. As outlined in Table 9, such an approach may accommodate a large spectrum of both acquired and congenital nevi containing 2 or more populations of melanocytes, any type of topography, and other constituents such as sclerosis of collagen and melanophages. However, caution must be exercised in the use of this terminology. Although the use of “combined” nevus or nevus with phenotypic heterogeneity may seem relatively straightforward, in some cases it may be difficult to be certain whether a particular nevus has more than a single population of melanocytes versus a simple variation within a nevus. Thus one pathologist's conventional dermal or blue nevus becomes another pathologist's combined nevus. These issues have little importance when considered against the primary purpose of the biopsy, which is to rule out melanoma.
In a general sense, combined nevi may be composed of any combination of common acquired nevus, blue nevus, or Spitz nevus and may occur in both cutaneous and mucosal sites (Table 8) (Fig. 13). Variants of blue nevi such as dendritic blue nevi are frequently observed in combined nevi. The more controversial entity “deep penetrating nevus,” whether considered a variant of blue nevus, a variant of Spitz nevus, a hybrid of the latter 2 categories (so-called Blitz nevus), or a separate melanocytic neoplasm is also frequently present in combined nevi.90–92,95 Histopathologically similar tumors that have been described under a variety of other terms including inverted type A nevus,3 clonal nevus,99–101 nevus with focal atypical epithelioid component,99 and atypical dermal nodules in benign melanocytic nevus102 probably represent examples of combined nevi. None of the latter lesions have been associated with malignant behavior in any of the cases reported thus far.
Deep Penetrating Nevus
There was considerable controversy among the Group regarding the entity “deep penetrating nevus” (Table 10).6,90,103 Some recognized this lesion as a distinct entity, which is commonly associated with a second discrete type of nevus, that is, forming part of a combined nevus.90,92,95 They recommended that the term “deep penetrating nevus” be applied to “monophasic” tumors (ie, those with no associated second melanocytic population) and combined nevus to those with 2 cell populations (in the latter instance, the 2 constituent nevus cell components should be described).
Some members of the Group recognized 2 variants of deep penetrating nevus, one with an epithelioid cell phenotype (which usually constitutes a variant of combined nevi and has been termed clonal nevus, inverted type A nevus, atypical dermal nodule in a benign melanocytic nevus, and melanocytic nevus with atypical epithelioid cell component).95,100,101 The second variant was described as showing a spindle cell phenotype (that has been termed inverted or plexiform pigmented spindle cell nevus).91,92 However, such plexiform nevi may also exhibit a cellular composition of both spindle and epithelioid cells.6,7,91 Deep penetrating nevi share histologic features with blue nevi (deep extension, associated stromal sclerosis, plump cells, pigmentation, and HMB45 positivity) and Spitz nevi (enlarged epithelioid and/or spindle cells, enlarged nuclei, and nuclear pseudoinclusions) but most participants were of the opinion that they are sufficiently distinctive to allow their recognition in most cases.90,103 Some participants object to the term deep penetrating nevus altogether and prefer plexiform spindle (and epithelioid cell) nevus as a more accurate depiction of this entity that stands apart from both blue nevus and Spitz nevus.6,91
There was general agreement that occasional cases of “deep penetrating” nevus may involve regional lymph nodes but spread beyond the regional nodes for typical cases has been rare. Some members of the Group recommended that, in view of the latter, all deep penetrating nevi should be reported as melanomas. Others contended that as most of these tumors behave in a clinically benign manner, they should be regarded as nevi and that a note may be added to indicate that there seems to be a low, but definite risk of metastasis (estimated to be less than 2% of cases without unusual features such as increased mitotic activity, expansile growth, or excessive nuclear pleomorphism). Others indicated a preference to categorize them as being of intermediate malignant or indeterminate or uncertain biologic potential, especially if any mitotic activity is present (Fig. 14). There was a consensus of opinion that further study of these lesions correlated with long-term follow-up was necessary before definitive conclusions could be drawn regarding their malignant potential and preferred terminology.104,105
Combined Nevus With Spitzoid Component
For combined nevi that include a Spitzoid component, the latter component should be assessed for atypical features as described above (in the section on Spitz tumors) and reported in concordance with these recommendations. For example, a tumor manifesting a compound or dermal nevus associated with a dermal Spitzoid component with atypical features (but without clear-cut evidence of malignancy) may be reported as a combined melanocytic tumor with atypical Spitzoid tumor and conventional nevus components.
There was also a controversy among the Group about the entity of regressing or “halo” Spitz nevus component occurring as part of a combined nevus. Some thought the latter should be regarded as being of indeterminate or uncertain malignant potential whereas others accepted that this was not an uncommon component of combined nevi. Some members maintained that in many instances, the use of the term “Spitz” nevus component in the latter context of an inflamed or halo melanocytic lesion may not be appropriate and may be a misnomer.6 In fact, the epitheloid cell “Spitzoid” component in the latter lesions may not be a true Spitz component but rather represent induced or reactive changes secondary to the lymphocytic infiltrate. Again, further study of patients with these lesions correlated with long-term follow-up and molecular analysis are needed to clarify many of these issues.
As already mentioned in the above section, there is considerable controversy as to the agreement about whether certain candidate lesions can be reproducibly classified as variants of blue nevus versus variants of Spitz nevus or placed into a third category of “combined” nevus or melanocytic nevus with phenotypic heterogeneity. In effect at present, a wastebasket or “black hole” of melanocytic lesions often with mixed cellular phenotypes, particularly those with pigmented epithelioid cells and also many suggesting Spitz nevi, blue nevi, and “combined” nevi has been created that generally lacks definition or any agreement about nomenclature. To make progress in this area, consensus sessions at the microscope and molecular characterization will be needed to hammer out some agreement about and reproducible definitions of these various melanocytic phenotypes and a nomenclature that makes sense and can be used by the general dermatopathology community.
Melanocytic Tumors With Indeterminate Biologic or Uncertain Malignant Potential (MELTUMP)
In the last few years, several reports have highlighted the difficulties in the histopathologic diagnosis and differential diagnosis of a particular group of melanocytic lesions often characterized by a thick dermal component that suggests melanoma.2,6,7,105 In particular, these tumors may exhibit a predominance of epithelioid cells or spindle cells. Perhaps other cellular phenotypes may be present. Terminology reflects the uncertainty in classification and interpretation of these atypical melanocytic tumors, and the many terms proposed in the past (such as malignant Spitz nevus, diagnostically controversial spitzoid melanocytic tumors, atypical epithelioid melanocytic proliferations of uncertain biologic potential, atypical Spitz tumors, deep penetrating nevi, and atypical blue nevi among others) testify to the diagnostic difficulties in differentiating tumors with benign behavior from those with a malignant clinical course. Histologic criteria used routinely for the distinction of benign melanocytic nevi from melanomas are not reliable in this group of tumors, and they are often referred to as “melanocytic neoplasms with indeterminate biologic potential” or “melanocytic tumor of uncertain malignant potential—MELTUMP.” We believe that the difficulty in correctly classifying these cases as “benign” or “malignant” reflects an inherent biologic problem, namely, the fact that they probably represent a single group of low-grade melanocytic tumors with potential for lymph node involvement and rarely for distant metastases.
There currently exists considerable agreement about the nomenclature and nature of many entities in the spectrum of benign melanocytic neoplasms. Nonetheless, certain entities continue to resist standardized classification and definition as to their biologic nature and natural history. The latter problems can clearly be attributed to the lack of sufficiently focused study and application of the appropriate techniques to resolve the relevant questions at hand, among those that can be resolved.
The following are among the most pressing issues with regard to benign melanocytic neoplasms and may be the subjects for future Workshops:
1. The development of broad consensus and standardized terminology for melanocytic cellular phenotypes;
2. The development of more standardized criteria and terminology for melanocytic nevi with unusual, deviant, atypical, or “dysplastic” features versus common nevi and nevi from particular anatomic sites;
3. The development of criteria for the classification or grading of melanocytic nevi that are biologically significant in terms of risk for progression to melanoma or that serve as indicators of melanoma risk in the individual patient;
4. The development of more standardized criteria for the distinction of melanocytic nevi from melanoma and classification schemes that are more clinically and biologically relevant as in other organ systems;
5. The development of consensus and more standardized terminology for Spitzoid lesions, blue nevi, and combined nevi, or nevi with phenotypic heterogeneity.
Practical Consensus Statement From the Workshop
Finally, it is common in dermatopathology practice, as has already been mentioned, to encounter melanocytic lesions that share histologic criteria of dysplastic nevi and melanoma, just as the presence of atypical “spitzoid” lesions that share features of Spitz nevi and melanoma. Gray zone, or so-called “borderline,” lesions of these types will, as a rule, defy uniformity of diagnostic interpretation among experienced observers. That fact can easily be illustrated with their independent interpretation by multiple observers; “blinded” interobserver exercises of this nature will usually evoke a distressing divergence of interpretation, which may, not uncommonly, range from “indubitably benign” at one extreme to “clearly malignant” at the other. Nevertheless, when dealing with a diagnostically difficult melanocytic tumor, it is probably prudent to seek opinion from one or more experienced colleagues.
When there is no uniformity of opinion as to diagnosis, the medicolegal standard of dermatopathologic, as well as clinical, practice is difficult to define in an unambiguous manner. One approach to this complex problem recommended by the Group is to acknowledge the diagnostic and biologic uncertainty of a lesion through the use of terminology such as melanocytic neoplasm with indeterminate biologic potential (or melanocytic tumor with uncertain malignant potential). At a minimum, complete resection of the primary lesion to achieve histologically clear margins is an obvious management choice in the interest of patient safety, given the uncertainty of diagnosis; whether the arbitrarily wide margins that have come to be customary for melanomas of classical histologic phenotypes should be applied to them is a matter of differing perspective, with no uniformity of opinion. Nonetheless, there is no reason to presume, in the instances of diagnostically indeterminate lesions that ultimately prove by their natural histories to be malignant, that patients' chances for survival have been compromised by the application of narrower surgical margins than are customary for melanomas of the common varieties, provided that local recurrences at the treatment sites have not occurred before more distant metastases.
1. Lund HS, Stobbe GD. The natural history of the pigmented nevus: factors of age and anatomic location. Am J Pathol. 1949;25:1117–1155.
2. Elder DE, Murphy GF. Melanocytic tumors of the skin. Atlas of tumor pathology. AFIP. 1990:1–101.
3. Mihm MC Jr, Googe PB. Problematic Pigmented Lesions. Philadelphia: Lea and Feiger; 1990.
4. Crowson AN, Magro CM, Mihm MC Jr,. The Melanocytic Proliferations: A Comprehensive Textbook of Pigmented Lesions. New York: Wiley-Liss; 2001.
5. Mooi WJ, Krausz T. Pathology of Melanocytic Disorders. 2nd ed. Oxford University Press: Oxford, UK; 2007.
6. Barnhill RL, Piepkorn M, Busam KJ. Pathology of Melanocytic Nevi and Malignant Melanoma. 2nd ed. New York: Springer-Verlag; 2004.
7. Barnhill RL. Tumors of Melanocytes in Dermatopathology. In: Barnhill RL, Crowson AN, Magro CM, et al, eds. 3rd ed. New York: McGraw-Hill;2010:615–688.
8. Reed RJ, Ichinose H, Clark WH Jr, et al. Common and uncommon melanocytic nevi and borderline melanomas. Semin Oncol. 1975;2:119–147.
9. Wayte DM, Helwig EB. Halo nevi. Cancer. 1968;22:69–90.
10. Fallowfield ME, Collina G, Cook MG. Melanocytic lesions of the palm and sole. Histopathology. 1994;24:463–467.
11. Signoretti S, Annessi G, Puddu P, et al. Melanocytic nevi of palms and soles: a histological study according to the plane of section. Am J Surg Pathol. 1999;23:283–287.
12. Khalifeh I, Taraif S, Reed JA, et al. A subgroup of melanocytic nevi on the distal lower extremity (ankle) shares features of acral nevi, dysplastic nevi, and melanoma in situ: a potential misdiagnosis of melanoma in situ. Am J Surg Pathol. 2007;31:1130–1136.
13. Clark WH Jr, Hood AJ, Tucker MA, et al. Atypical melanocytic nevi of the genital type with a discussion of reciprocal parenchymal-stromal interactions in the biology of neoplasia. Hum Pathol. 1998;29(suppl 1):S1–S24.
14. Gleason BC, Hirsch MS, Nucci MR, et al. Atypical genital nevi. A clinicopathologic analysis of 56 cases. Am J Surg Pathol. 2008;32:51–57.
15. Rongioletti F, Ball RA, Marcus R, et al. Histopathological features of flexural melanocytic nevi: a Study of 40 Cases. J Cutan Pathol. 2000;27:215–217.
16. Rongioletti F, Urso C, Batolo D, et al. Melanocytic nevi of the breast: A histologic case-control study. J Cutan Pathol. 2004;31:137–140.
17. Tucker MA, Greene MH, Clark WH Jr, et al. Dysplastic nevi on the scalp of prepubertal children from melanoma-prone families. J Pediatr. 1983; 103:65–69.
18. Fernandez M, Raimer SS, Sánchez RL. Dysplastic nevi of the scalp and forehead in children. Pediatr Dermatol. 2001;18:5–8.
19. Fabrizi G, Pagliarello C, Parente P, et al. Atypical nevi of the scalp in adolescents. J Cutan Pathol. 2007;34:365–369.
20. Clark WH Jr, Reimer RR, Greene M, et al. Origin of familial malignant melanomas from heritable melanocytic lesions: the B-K mole syndrome. Arch Dermatol. 1978;114:732–738.
21. Elder DE, Goldman LI, Goldman SC, et al. Dysplastic nevus syndrome: a phenotypic association of sporadic cutaneous melanoma. Cancer. 1980;46:1787–1794.
22. NIH Consensus Conference: precursors to malignant melanoma. JAMA. 1984;251:1864–1866.
23. Elder DE. The dysplastic nevus. Pathology. 1985;17:291–297.
24. Seywright MM, Doherty VR, MacKie RM. Proposed alternative terminology and subclassification of so-called “dysplastic naevi”. J Clin Pathol. 1986;39:189–194.
25. Bergman W, Ruiter DJ, Scheffer E, et al. Melanocytic atypia in dysplastic nevi: immunohistochemical and cytophotometrical analysis. Cancer. 1988;61:1660–1666.
26. Steijlen PM, Bergman W, Hermans J, et al. The efficacy of histopathological criteria required for diagnosing dysplastic naevi. Histopathology. 1988;12:289–300.
27. Piepkorn M, Meyer LJ, Goldgar D. The dysplastic melanocytic nevus—a prevalent lesion that correlates poorly with clinical phenotype. J Am Acad Dermatol. 1989;20:407–415.
28. Barnhill RL, Roush GC, Duray PH. Correlation of histologic and cytoplasmic features with nuclear atypia in atypical (dysplastic) nevomelanocytic nevi. Hum Pathol. 1990;21:51–58.
29. Barnhill RL. Current status of the dysplastic melanocytic nevus. J Cutan Pathol. 1991;18:147–159.
30. Duray PH, DerSimonian R, Barnhill RL, et al. An analysis of interobserver recognition of the histopathologic features of dysplastic nevi from a mixed group of nevomelanocytic lesions. J Am Acad Dermatol. 1992;27:741–749.
31. NIH Consensus Development Panel on Early Melanoma: diagnosis and treatment of early melanoma. JAMA. 1992;268:1314–1319.
32. Duncan LM, Berwick MA, Bruijn JA, et al. Histopathologic recognition and grading of dysplastic melanocytic nevi: an inter-observer agreement study. J Invest Dermatol. 1993;100(suppl 3):318S–321S.
33. Piepkorn MW, Barnhill RL, Rabkin MS, et al. Histologic diagnosis of the dysplastic nevus: an analysis of inter- and intra-observer concordance, correlation with clinical phenotype, and prevalence in population controls. J Am Acad Dermatol. 1994;30:707–714.
34. Schmidt B, Hollister K, Weinberg D, et al. Analysis of dysplastic nevi by DNA image cytometry. Cancer. 1994;73:2971–2977.
35. Kang S, Barnhill RL, Mihm MC Jr, et al. Melanoma risk in individuals with clinically atypical nevi. Arch Dermatol. 1994;130:999–1001.
36. Garbe C, Buttner P, Weiss J, et al. Risk factors for developing cutaneous melanoma and criteria for identifying persons at risk: multicenter case-control study of the central malignant melanoma registry of the German Dermatological Society. J Invest Dermatol. 1994;102:695–699.
37. Garbe C, Buttner P, Weiss J, et al. Associated factors in the prevalence of more than 50 common melanocytic nevi, atypical melanocytic nevi, and actinic lentigines: multicenter case-control study of the central malignant melanoma registry of the German Dermatological Society. J Invest Dermatol. 1994;102:700–705.
38. Tucker MA, Halpern A, Holly EA, et al. Clinically recognized dysplastic nevi: a central risk factor for cutaneous melanoma. JAMA. 1997;277:1439–1444.
39. Blessing K. Benign atypical naevi: diagnostic difficulties and continued controversy:. Histopathology. 1999;34:189–198.
40. Shea CR, Vollmer RT, Prieto VG. Correlating architectural disorder and cytological atypia in Clark (dysplastic) melanocytic nevi. Hum Pathol. 1999;30:500–505.
41. Shors AR, Kim S, White E, et al. Dysplastic naevi with moderate to severe histological dysplasia: a risk factor for melanoma. Br J Dermatol. 2006;155:988–993.
42. Mark GJ, Mihm MC, Liteplo MG, et al. Congenital melanocytic nevi of the small and garment type. Hum Pathol. 1973;4:395–418.
43. Cramer SF. The melanocytic differentiation pathway in congenital melanocytic nevi: theoretical considerations. Pediatr Pathol. 1988;8:253–265.
44. Reed RJ. Giant congenital nevi: a conceptualization of patterns. J Invest Dermatol. 1993;100(suppl):300S–312S.
45. Barnhill RL, Fleischli M. Histologic features of congenital melanocytic nevi in infants less than a year of age. J Am Acad Dermatol. 1995;33:780–785.
46. Silvers DN, Helwig EB. Melanocytic nevi in neonates. J Am Acad Dermatol. 1981;4:166–175.
47. Hendrickson MR, Ross JC. Neoplasms arising in congenital giant nevi: morphologic study of seven cases and a review of the literature. Am J Surg Pathol. 1981;5:109–135.
48. Mancianti ML, Clark WH, Hayes FA, et al. Malignant melanoma simulants arising in congenital melanocytic nevi do not show experimental evidence for a malignant phenotype. Am J Pathol. 1990;136:817–829.
49. Clark WH Jr, Elder DE, Guerry D. IV: Dysplastic nevi and malignant melanoma. In: Farmer ER, Hood AF, eds. Pathology of the Skin. Norwalk, CT: Appleton and Lange; 1990:684–756.
50. Bastian BC, Xiong J, Frieden IJ, et al. Genetic changes in neoplasms arising in congenital melanocytic nevi: differences between nodular proliferations and melanoma. Am J Pathol. 2002;161:1163–1169.
51. Zaal LH, Mooi WJ, Sillevis Smitt JH, et al. Classification of congenital melanocytic naevi and malignant transformation: a review of the literature. Br J Plast Surg. 2004;57:707–719.
52. Bauer J, Curtin JC, Pinkel D, et al. Congenital melanocytic nevi frequently harbor NRAS mutations but no BRAF mutations. J Invest Dermatol. 2007;127:179–182.
53. Spitz S. Melanomas of childhood. Am J Pathol. 1948;24:591–609.
54. Weedon D, Little J. Spindle and epithelioid cell nevi in children and adults. A review of 211 cases of the Spitz nevus. Cancer. 1977;40:217–225.
55. Paniago-Pereira C, Maize J, Ackerman A. Nevus of large spindle and/or epithelioid cells (Spitz's nevus). Arch Dermatol. 1978;114:1811–1823.
56. Busam KJ, Barnhill RL. The spectrum of spitz tumors. In: Kirkham N, Lemoine NR, eds. Progress in Pathology. Vol. 2. Edinburgh: Churchill Livingstone; 1995:31–46.
57. Barr R, Morales R, Graham J. Desmoplastic nevus. A distinct histologic variant of mixed spindle and epithelioid cell nevus. Cancer. 1980;46:557–564.
58. MacKie RM, Doherty VR. The desmoplastic melanocytic naevus: a distinct histological entity. Histopathology. 1992;20:207–211.
59. Harris GR, Shea CR, Horenstein MG, et al. Desmoplastic (sclerotic) nevus an underrecognized entity that resembles dermatofibroma and desmoplastic melanoma. Am J Surg Pathol. 1999;23:786–794.
60. Sagebiel R, Chinn E, Egbert B. Pigmented spindle cell nevus. Clinical and histologic review of 90 cases. Am J Surg Pathol. 1984;8:645–653.
61. Smith N. The pigmented spindle cell tumor of Reed: an underdiagnosed lesion. Semin Diagn Pathol. 1987;4:75–87.
62. Barnhill RL, Mihm MC. Pigmented spindle cell nevus and its variants: distinction from melanoma. Br J Dermatol. 1989;121:717–726.
63. Barnhill RL, Berwick M, Mihm MC Jr. The histologic spectrum of pigmented spindle cell nevus: a review of 120 cases with emphasis on atypical variants. Hum Pathol. 1991;22:52–58.
64. Harvell JD, Meehan SA, LeBoit PE. Spitz's nevi with halo reaction: a histopathological study of 17 cases. J Cutan Pathol. 1997;24:611–619.
65. Harvell JD, Bastian BC, LeBoit PE. Persistent (recurrent) Spitz nevi: a histopathologic, immunohistochemical, and molecular pathologic study of 22 cases. Am J Surg Pathol. 2002;26:654–661.
66. Smith K, Barrett T, Skelton H, et al. Spindle cell and epithelioid cell nevi with atypia and metastasis (malignant Spitz nevus). Am J Surg Pathol. 1989;13:931–939.
67. Busam KJ, Barnhill RL. Pagetoid Spitz nevus. Am J Surg Pathol. 1995;19:1061–1067.
68. Barnhill RL, Argenyi ZB, From L, et al. Atypical Spitz nevi/tumors: lack of consensus for diagnosis, discrimination from melanoma, and prediction of outcome. Hum Pathol. 1999;30:513–520.
69. Spatz A, Calonje E, Handfield-Jones S, et al. Spitz tumors in children: a grading system for risk stratification. Arch Dermatol. 1999;135:282–285.
70. Bastian BC, Wesselman U, Pinkel D, et al. Molecular cytogenetic analysis of Spitz nevus shows clear differences to melanoma. J Invest Dermatol. 1999;113:1065–1069.
71. Bastian BC, LeBoit PE, Pinkel D. Mutations and copy number increase of HRAS in Spitz nevi with distinctive histopathologic features. Am J Pathol. 2000;157:967–972.
72. Ludgate MW, Fullen DR, Lee J, et al. The atypical Spitz tumor of uncertain biologic potential. A series of 67 patients from a single institution. Cancer. 2009;115:631–641.
73. Montgomery H. The blue nevus (Jadassohn-Tiéche): its distinction from ordinary moles and malignant melanomas. Am J Cancer. 1939;36:527–539.
74. Allen AC. A reorientation on the histogenesis and clinical significance of cutaneous nevi and melanomas. Cancer. 1949;2:28–56.
75. Dorsey CS, Montgomery H. Blue nevus and its distinction from Mongolian spot and the nevus of Ota. J Invest Dermatol. 1954;22:225–236.
76. Lund HZ, Kraus JM. Melanotic Tumors of the Skin. Atlas of Tumor Pathology. Washington, DC: Armed Forces Institute of Pathology; 1962;104–106.
77. Gartmann H. Neuronaevus blue Masson—cellular blue nevus Allen. Arch Klin Exp Dermatol. 1965;221:109–121.
78. Leopold JG, Richards DB. Cellular blue nevi. J Pathol Bacteriol. 1967;94:247–255.
79. Leopold JG, Richards DB. The interrelationship of blue and common naevi. J Pathol Bacteriol. 1968;95:37–46.
80. Rodriguez HA, Ackerman LV. Cellular blue nevus. clinicopathologic study of forty-five cases. Cancer. 1968;21:393–405.
81. Merkow LP, Burt RC, Hayeslip DW, et al. A cellular and malignant blue nevus: a light and electron Microscopic Study. Cancer. 1969;24:888–896.
82. Mishima Y, Mevorah B. Nevus of Ota and nevus of Ito in American Negroes. J Invest Dermatol. 1961;36:133–154.
83. Kopf AW, Weidman AI. Nevus of Ota. Arch Dermatol. 1962;85:195–208.
84. Hidano A, Kajima H, Ikeda S, et al. Natural history of nevus of Ota. Arch Dermatol. 1967;95:187–195.
85. Hirayama T, Suzuki T. A new classification of Ota's nevus based on histopathological features. Dermatologica. 1991;183:169–172.
86. Pittman JL, Fisher BK. Plaque-type blue nevus. Arch Dermatol 1976;112:1127–1128.
87. Carney JA, Ferreiro JA. The epithelioid blue nevus: a multicentric familial tumor with important associations, including cardiac myxoma and psammomatous melanotic schwannoma. Am J Surg Pathol. 1996;20:259–272.
88. Zembowicz A, Carney JA, Mihm MC. Pigmented epithelioid melanocytoma: a low-grade melanocytic tumor with metastatic potential indistinguishable from animal-type melanoma and epithelioid blue nevus. Am J Surg Pathol. 2004;28:31–40.
89. Zembowicz A, Knoepp SM, Bei T, et al. Loss of expression of protein kinase a regulatory subunit 1alpha in pigmented epithelioid melanocytoma but not in melanoma or other melanocytic lesions. Am J Surg Pathol. 2007;31:1764–1775.
90. Seab JA Jr, Graham JH, Helwig EB. Deep penetrating nevus. Am J Surg Pathol. 1989;13:39–44.
91. Barnhill RL, Mihm MC Jr, Magro CM. Plexiform pigmented spindle cell naevus: a distinctive variant of plexiform melanocytic naevus. Histopathology. 1991;18:243–247.
92. Cooper PH. Deep penetrating (plexiform spindle cell) nevus. A frequent participant in combined nevus. J Cutan Pathol. 1992;19:172–180.
93. Fletcher V, Sagebiel RW. The combined nevus: mixed patterns of benign melanocytic lesions must be differentiated from malignant melanomas. In: Ackerman AB, ed. Pathology of Malignant Melanoma. New York: Masson; 1981:273–283.
94. Rogers GS, Advani H, Ackerman AB. A combined variant of Spitz's nevi. how to differentiate them from malignant melanomas. Am J Dermatopathol. 1985;7:61–78.
95. Pulitzer DR, Martin PC, Cohen AP, et al. Histologic classification of the combined nevus. Analysis of the variable expression of melanocytic nevi. Am J Surg Pathol. 1991;15:1111–1122.
96. Scolyer RA, Zhuang L, Palmer AA, et al. Combined naevus: a benign lesion frequently misdiagnosed both clinically and pathologically as melanoma. Pathology. 2004;36:419–427.
97. Barnhill RL, Argenyi Z, Berwick M, et al. Atypical cellular blue nevi (cellular blue nevi with atypical features): lack of consensus for diagnosis and distinction from cellular blue nevi and malignant melanoma (“Malignant Blue Nevus”). Am J Surg Pathol. 2008;32:36–34.
98. Maize JC Jr, McCalmont TH, Carlson JA, et al. Genomic analysis of blue nevi and related dermal melanocytic proliferations. Am J Surg Pathol. 2005;29:1214–1224.
99. Ball NJ, Golitz LE. Melanocytic nevi with focal atypical epithelioid cell components: a review of seventy-three cases. J Am Acad Dermatol. 1994;30:724–729.
100. High WA, Alanen KW, Golitz LE. Is melanocytic nevus with focal atypical epithelioid components (clonal nevus) a superficial variant of deep penetrating nevus? J Am Acad Dermatol. 2006;55:460–466.
101. Murali R, McCarthy SW, Thompson JF, et al. Melanocytic nevus with focal atypical epithelioid components (clonal nevus) is a combined nevus. J Am Acad Dermatol. 2007;56:889–890.
102. Collina G, Deen S, Cliff S, et al. Atypical dermal nodules in benign melanocytic naevi. Histopathology. 1997;31:77–101.
103. Robson A, Morley-Quante M, Hempel H, et al. Deep penetrating naevus: clinicopathological study of 31 cases with further delineation of histological features allowing distinction from other pigmented benign melanocytic lesions and melanoma. Histopathology. 2003;43:529–537.
104. Cerroni L, Kerl H. Tutorial on melanocytic lesions. Am J Dermatopathol. 2001;23:237–241.
105. Elder DE, Xu X. The approach to the patient with a difficult melanocytic lesion. Pathology. 2004;36:428–434.
© 2010 Lippincott Williams & Wilkins, Inc.