|
Typical vertical growth is manifested by the formation of a plaque or nodule (3,6,7). In the plaque or nodule, nests and fascicles of tumor cells are closely, and compactly spaced. In short, the virtual images are those of a clone of cells that forever is independent of phenomena at the dermal-epidermal interface. The proliferating cells cluster in nests, fascicles, and sheets; in close aggregation, the nests, fascicles, and sheets form an expansile nodule, or densely cellular plaque. Typical vertical growth, by definition, is a lesion showing level III invasion (pattern III).
In general, infiltrating neoplasms, in making their own accommodations, may substitute a specific tumor stroma at level IV for the collagenous framework of the reticular dermis, or individual cells and nests of cells may infiltrate the reticular dermsi among the coarse collagen bundles with little supporting tumor stroma. The latter pattern, if manifested as a feature of a melanoma, and associated with a pattern of typical vertical growth in a widened papillary dermis, qualifies as migrant (diffuse) vertical growth (pattern IV). It is not likely to be associated with significant markers for cellular host immune response; it has prognostic significance (if for no other reason than as a marker for the lack of a “brisk host immune response”). The presence of neoplastic cells at level IV signifies that the cells have divorced themselves from a community; henceforth, they are independent of both the primary community, and its communal stroma. The cells of such lesions have acquired universal independence with the capacity to colonize all sites of the body. Cells, with the capacity to infiltrate individually among pre-existing collagen bundles of the reticular dermis, might be characterized as universal migrants, adaptable to all tissues. A melanoma, that has attained level IV in the dermis, commonly involves the adventitia of dermal blood vessels. It should be noted that the quality of “universal independence” is relative; melanomas often manifest patterns of metastasis which evoke images of preferential homing to particular sites.
In all vertical growth components, but particularly in thin lesions (less than 1 mm in vertical dimensions) showing variant vertical growth, the cytologic features of the component cells often are deceptively bland. With rare exceptions, thin melanomas, that have an acceptable vertical growth component, and also measure less than 1 mm in vertical dimensions, will deviate cytologically in the vertical growth component from the pattern of the so-called “common final pathway” as defined in the concept of MDM, and as manifested in the radial growth component of a fully evolved SSM (3,6). Such lesions are properly evaluated by both Breslow’s, and modified Clark’s criteria. Conservative local excisions generally are adequate for the control of the thin, borderline lesions (incipient melanomas) in the setting of dysplastic (atypical) nevi.
Internal organization of melanomas (general observations):
In all the common variants of melanomas, combinations of immune responses, epidermal responses, and cytologic features are basically similar, but anatomic sites, and genetics have a modulating influence on the expressions of these variables. The basic relationships are maintained in such diverse settings as melanoma arising in the setting of the dysplastic nevus syndrome (i.e., “superficial spreading melanoma” [SSM]), melanoma arising in the setting of fair, actinically damaged skin (i.e., “lentigo maligna melanoma” [LMM]), and melanoma involving acral or mucosal surfaces (i.e., “acral lentiginous melanoma”[ALM]). In these examples, atypical cells in lentiginous and junctional patterns are represented in the epidermis adjacent to all vertical growth components. By inference, in all of these variants, neoplastic cells initially arise at the dermal-epidermal interface, and in their evolution acquire the ability to migrate into the dermis. In the process of neoplastic progression, migrants in the dermis eventually acquire the ability to survive, and multiply, thereby accumulating in clustered nests to form vertical growth components.
The modulating influences, such as site, environmental factors, and age, lend credence to the common categories of typical melanomas. On the other hand, the similarities among the more common variants of melanoma are such that some observers simply ignore the distinctions and, in turn, deny the existence of variant melanomas. They then are at liberty to promote the concept that there is only “one” melanoma (but, paradoxically, to also proclaim that there are “no” melanoma cells?).
Malignant melanomas, even in the limited setting of melanomas of the dysplastic nevus syndrome, are histologically variable. At the most basic level, melanomas in vertical growth might be characterized as spindle cell variants, round or polygonal (“epithelioid”) cell variants, mixed variants, or indeterminate variants. The round cell variants tend to form rounded nests in the dermis. The spindle cell variants tend to be manifested in fascicular patterns. The dermal patterns associated with rounded junctional nests are likely to be ascribed to the phenomena of “dropping off.” On the other hand, fascicles tend to extend in continuity by elongation into the dermis, and may even reach into the reticular dermis while maintaining contact with the epidermis. In fascicular, spindle cell variants, individual neoplastic cells are often delivered into the reticular derms from the extremities of the fascicles. In contrast, in round cell variants, it is extremely difficult to identify, by the evaluation of rounded nests at the dermal-epidermal interface, the mechanism by which contact with the epidermis is lost, and the respective nest, or a portion of nest ,“drops off” into the dermis. Most likely, the process is comparable to that in the fascicular variants, but is too subtle to be easily recognized on routine histologic sections.
The evaluation of many variables is required in the histologic interpretation of melanomas. The variables are not simply those of cytologic, and architectural deviations; they include additional qualities such as the nature, and degree of host immune response, and interactions with stroma. Physical dimensions (generally evaluated by a measurement of the dimensions of lesions along an axis that is perpendicular to the surface of the “thickest” portion of a lesion), nuclear grade (size of nuclei, chromatic nuclear aberrations, and size of nucleoli), mitotic activity, and nuclear aberrations, and levels of invasion (Clark’s levels of invasion as modified in the concept of MDM) are additional parameters. Most of these variables, with the rare exception of degree of cytologic atypia (e.g., degree of atypia is a requisite in the diagnosis of SSM), are given recognition in current attempts to provide guidelines to treatment. In these attempts at prognostications, size (vertical dimensions) of the dermal component is the most critical parameter (and, in the concept of MDM, is the morphologic feature that is emphasized in distinguishing melanoma from dysplasias).
Typical melanomas (common types):
The currently most accepted characterizations of common variants of melanoma are derived from the contributions of W. H. Clark, Jr. and co-workers in the late ‘60’s and early ‘70’s. Basically, the variants include SSM, LMM, and nodular melanoma. To this list, ALM, desmoplastic melanoma, and neurotropic melanoma were added in the early, and mid ‘70’s. Myxoid melanoma probably is a variant of the desmoplastic variant. If the classifications were to be limited to these contributions, six specific variants,and a seventh category of melanoma, not other classified, would be identified. At least one additional variant of a somewhat different nature is “malignant blue nevus.” The validity, and utility of these characterizations of common, or typical melanomas has not been universally accepted.
Critically, the following comments are pertinent:
1) Thin SSM (i.e., lesions with a vertical growth component less than 1 mm in vertical dimensions) is the common expression of neoplastic progression in a special segment of a neoplastic continuum. This segment represents a borderland, whether encountered in the setting of either multiple dysplastic nevi, or sporadically. Cytologically, moderately severe to marked atypia (depending on the criteria of an individual observer) is the distinguishing characteristic of thin SSM. Within the framework of the definition of patterns I & II (Clark’s criteria as modified in the concept of MDM, and as additionally modified herein), the distinctions between the patterns of a moderately severe or marked dysplasia, and SSM at levels I & II (as defined by classic Clark’s criteria) are arbitrary; they are basically a learned, rather than a real, distinction.
From a different perspective, SSM at level II is an expression of a common final pathway, variably encountered in the evolution leading from common precursors to the common melanomas. Having entered this pathway, the only option for further neoplastic progression is for the neoplasm to enter vertical growth. SSM at level II basically is a “segment” of a neoplastic continuum, validated by a selection of an arbitrary set of criteria. It is distinguishable from its more benign precursor stages, and from all subsequent stages in which nesting patterns in the dermal component acquire 3 dimensions in the patterns of a plaque, or nodule (patterns III & IV); at level II, dimensionalities are insufficient to characterize the patterns as those of vertical growth.The cells of pattern II SSM are plump, usually rounded or polygonal, and finely pigmented.
Superficial spreading melanoma is not a unique neoplasm; it issues from a common final pathway which, on occasion, is also accessible to neoplastic cells in the clinical, and histologic setting of both lentigo maligna melanoma, and acral, and mucosal lentiginous melanoma. SSM is melanoma vulgaris. By definition, its nuclear grade is high (i.e., grade III - IV).
II. Lentigo maligna melanoma (LMM), as defined herein, is a lesion in vertical growth, and is the common melanoma of the sun-exposed skin of fair complected patients who are usually beyond the age of 40. By definition, a remnant of its precursor is represented in the patterns of a dysplasia (patterns I & II as defined above; radial growth) at the margins of the vertical growth component. In fact, the precursor (lentigo maligna) is a special variant of a dysplasia (and as in the common dysplasias, patterns I & II are acceptable). Both it and other precursor lesions, such as those seen in the dysplastic nevus syndrome, can be evaluated by the same criteria, but often the histologic features of LM are sufficiently distinctive to allow an experienced observer to distinguish a precursor lesion of LM from a precursor lesion of the dysplastic nevus syndrome.
Clark and Mihm proposed the designation, lentigo maligna, for the precursor of LMM, but then restricted the definition to lesions in which the neoplastic cells are confined to the epidermis. In practice, the designation, LM, should embrace dermal patterns in the same manner as the definition of melanocytic dysplasias in the dysplastic nevus syndrome (i.e., patterns I & II). The precursors of both SSM and LMM are basically two dimensional in regard to their dermal patterns. Scattered, randomly spaced nests in the upper portion of the dermis should be incorporated in the collection of virtual images related to a definition of LM. Thin lesions (i.e., less than 1 mm in vertical dimensions) in the LM and LMM categories are representative of a borderland.
The epidermis in lesions of LM and LMM often is remarkably unaffected (and shows the atrophy associated with aged skin). In contrast, the epidermis of SSM at level I & II is hyperplastic in a most distinctive manner. The hyperplasia affects the superficial unit (i.e., the unit comprised of keratinocytes showing terminal differentiation). The differences in the epidermal responses in these two variants suggest that either the epidermis of the elderly is unresponsive to growth factors produced by a clone of neoplastic cells, or that the cells of lentigo maligna do not as regularly produce growth factors as do those of SSM.
The precursor, LM, is expressed in a wide range of degrees of cytologic atypia. Some examples are deceptively bland; in such examples, the cells tend to be spindle shaped; they resemble the cells of pigmented spindle cell “nevus” of Reed. The restrictions on degree of atypia in the radial growth component, that are basic to the definition of SSM, are not imposed in the definition of LMM. Distinctive multinucleated giant cells in the lentiginous components, and in junctional nests at the dermal-epidermal interface are a common feature. The designation, lentigo maligna, is not synonymous with “melanoma-in-situ” but, in common conceptualizations, the two designations often are equated. Certainly, a clinician is likely to afford greater significance to a diagnosis of LM than to that of a common premalignant melanocytic dysplasia showing a comparable degree of atypia. This caution may be justified; there may be a greater tendency for local recurrence in the category of lentigo maligna and lentigo maligna melanoma than in the common premalignant dysplasias.
In low grade dysplasias in the setting of LM, markers for host immune response often are scant. As is common with other lentiginous variants, the emergence of a vertical growth component is likely to be expressed in patterns of spindle cells and fascicles. In some examples, the degree of atypia, even in a vertical growth component is mild. Such lesions, by cytologic criteria, are minimal deviation variants. Thin lesions, in the setting of LMM, are more likely to be level IV lesions (pattern IV, or diffuse [migrant] vertical growth) than are thin lesions, of comparable vertical dimensions, in the category of SSM.
|
