|
In severe dsyplasias, the cells may be either spindle shaped or polygonal and, in the epidermal domain, are usually pigmented. Atypia is marked. The neoplastic cells in the epidermis generally are polygonal; characteristically, they are finely, and uniformly pigmented. They may be spindle shaped. They tend to form rounded nests to the relative exclusion of associated lentiginous components. In addition, individually, and in nests, they migrate upward into the stratum malpighii. The superficial unit of the infiltrated epidermis undergoes a distinctive type of hyperplasia in which individual keratinocytes are hypertrophied; the hypertrophied cells have acidophilic cytoplasm. In this altered domain, the resulting patterns have a pagetoid quality. In toto, the patterns qualify as those of the common final pathway (CFP), as proposed in the concept of minimal deviation melanoma (MDM). The designation, common final pathway (CFP), carries with it the implication that the respective dysplasia has exhausted its neoplastic options. Any new neoplastic transformations would entail alterations in the manner in which nests of neoplastic cells relate to one another, and to their stroma, rather than the emergence of new cytologic aberrations; they would entail a transition from dysplasia to melanoma.
Within the epidermal domain at level I, the “upward” migration of neoplastic cells away from their site of origin is an intriguing, but ill-defined phenomenon. For some examples, the upward movement probably relates to the acquisition of qualities that will eventually allow cells to violate the boundaries which define this level; it anticipates a movement of the neoplastic cells into the papillary dermis. It may be an expression of a loss of “cohesiveness,” or may be evidence of negative cytotaxis (i.e., neoplastic cells move away from their neighbors).
The dermal patterns of severe, and moderately severe dysplasias are similar, but dense, concentric, lamellar fibrosis is not always a uniform feature of thin examples. Some of the nests of cells, which are not associated with patterns of lamellar fibrosis, or appositional fibroplasia, probably have undergone a neoplastic transformation which anticipates the development of a supportive tumor stroma, and the emergence of level III growth (pattern III).
Atypical nevi:
In an atypical nevus, a remnant of typical nevus is associated with patterns of a premalignant melanocytic dysplasia. Zonal variations in patterns, that have been characterized as types A, B, & C in typical nevi, are variably represented in any remnants of typical nevus that are preserved in an atypical nevus. In atypical nevus, as manifested in the premalignant dysplasia-melanoma sequence, the cells of the dysplasia are forever independent of phenomena in any remnant of typical nevus: continuity, which characterizes the transitions from type A, to type B, and then to type C patterns in a typical nevus, is interrupted. In large part, the sclerosis of stroma, as a response to a population of dysplastic melanocytic cells at, or near, the dermal-epidermal interface, becomes a barrier between the dermal cells of any nevic remnant, and those of the dysplasia.
Real images, which are expressive of neoplastic phenomena in certain borderlands of neoplasia (a stage at which the patterns of precursors merge imperceptibly with those of the respective evolved malignancies), include many of the images which are also associated with fully evolved malignancies. In general in defining relevant virtual images for making distinctions between precursor, and its respective malignancy, emphasis often has been placed on the identification of “invasive growth.”
For the melanocytic system, the presence of neoplastic cells, or nests of neoplastic cells in the papillary dermis (i.e., an “invasive” component [micro-invasion]) does not, in itself, provide a definitive separation between dysplasia and melanoma. For other observers, without attention to degree of cytologic atypia, even upward migration of melanocytes into the stratum malpighii is sufficient to characterize the epidermal pattern as “melanoma-in-situ.” The presence of cytologically comparable cells in the papillary dermis would, in turn, qualify the respective lesion as a “melanoma” showing at least level II invasion.
In the setting of premalignant dysplasia-melanoma sequence, some observers (i.e., those who subscribe to the concept of minimal deviaton melanoma [MDM]) define melanoma on the basis of patterns in which nests of atypical cells are distributed in the papillary dermis. In the latter approach, virtual images, which are related to both degrees of cytologic atypia, and host immune response, reinforce those which are related to the interpretation of nesting patterns. In the transformation of real, and virtual images into a definitive amalgam, particular patterns in which nests are distributed, or aggregated define vertical growth. A vertical growth component has been offered as the sine qua non for the definition of melanoma (i.e., melanomas show pattern III or greater).
VERTICAL GROWTH
In placing emphasis on patterns in the dermis, the virtual images, in toto, transmorgrify the real images into phenomena which are related to the survival, and proliferation of neoplastic cells in an immunologically hostile environment. In this approach, nesting patterns, and cytologic atypia of at least a moderate degree in the dermal component provide the distinctions between dysplasia, and melanoma. For thin borderline lesions with vertical growth components (in the common premalignant dysplasia-melanoma sequences), the cells of the vertical growth component often do not manifest as high a degree of atypia as the cells in the lentiginous and junctional patterns in the overlying epidermis (6).
The implications of currently popular linguistic options, as embodiments of amalgamated real and virtual images, are multifold:
1) The cells of a vertical growth component may be representative of an older neoplastic clone than the cells in nests at the dermal-epidermal interface (3); they may have even become independent of the progressive phenomena at the dermal-epidermal interface. These independent neoplastic phenomena occurring in a dermal popularion of neoplastic cells are embodied in the concept of typical vertical growth (a virtual image in which nests of tumor cells are closely clustered in a widened papillary dermis to form an expansile nodule). As few as 5 nests of uniformly atypical cells in two strata satisfy the criteria for the recognition of young, typical vertical growth.
2) Cytologic distinctions, when comparisons are made of the stratified nests in a vertical growth component, are commonly characterized as “maturation.” They may be representative of successive generations of neoplastic clones with succeeding clones, as they “drop off” into the dermis, being more advanced neoplastically, and more atypical cytologically. In this approach, the addition of new clones to the dermal population would be dependent on phenomena at the dermal-epidermal interface; an increase in the number of nests would be dependent on “dropping off” of nests of cells to produce a form of accretive growth at the dermal-epidermal interface. Growth would be accretive, and incremental with the bulk of the vertical growth component being supplied by the migration of populations from the epidermis into the dermis in stratified patterns (30). These phenomena are embraced in the concept of variant vertical growth, but have particular application for thin lesions showing arrested variant vertical growth.
3) A gradient of growth factors may be operative in the cytologic variability of many vertical growth components. The gradient may be oriented with the greatest intensity at the dermal-epidermal intreface, and the weakest intensity at the deep margin of the vertical growth component. Similar phenomena may also be operative in some benign variant nevi; they may provide the most appropriate explanation for the “patterns of maturation” commonly observed in lesions of “Spitz” type.
4) A requisite degree of cytologic atypia is not rigidly related to the emergence of a vertical growth component. By definition, a severe degree of atypia (i.e., the common final pathway in the concept of minimal deviation melanoma) is required for the diagnosis of superficial spreading melanoma (c16t3P1-3 & c27t3P1-6) but, in practice, the cytologic spectrum is broad. Melanomas in the setting of lentigo maligna often are bland, and deceptive in the vertical growth components. These deviant patterns qualify many examples of lentigo maligna melanomas as minimal deviation variants. In fact, some examples in their vertical growth components are difficult to distinguish cytologically from the focal, solid patterns of pigmented spindle cell “nevi.” A requisite degree of atypia, in and of itself, is not the prime discriminator between dysplasia (LM) and melanoma (LMM).
5) The concept of MDM, with its rich stores of virtual images (all related to neoplastic phenomena), and with emphasis on the primacy of a vertical growth component (as the chief discriminator for a diagnosis of melanoma), has application in the interpretation of variant “nevi,” and related melanomas.
6) The character of the stromal responses in variant vertical growth may have prognostic significance. Appositional, and lamellar fibroplasia may be features that are indicative of a favorable host immune response. In this type of stromal response, nests of neoplastic cells are relatively sequestered in the widened papillary dermis (level III pattern). In contrast, in areas in which the stromal response is a delicate fibrous matrix, the nests of cells may have acquired independence from the neoplastic phenomena at the dermal-epidermal interface. This change in stromal response heralds either the emergence of typical vertical growth at level III ( as a community of nests of cells), the emergence of progressive vertical growth at level III, or the emergence of migrant (diffuse) vertical growth at level IV.
Dimensionality is an attribute of melanocytic neoplasia. Histologically, an appreciation for the dimensionality of a malignancy (e.g., melanoma) requires the manipulation of virtual images; the ability to translate the two dimensional histologic patterns of closely aggregated nests of neoplastic cells in the dermis (or any other organ system) into three dimensional images (i.e., virtual images that, for a melanoma, satisfy the criteria for the definition, and provide the amalgam for the recognition of a vertical growth component) is required. The manipulations are concerned with the virtual imaging of the patterns produced by closely, and regularly spaced nests of cells rather than an appreciation for the size of a solitary dermal nest of neoplastic cells. The important community is not a large solitary nest, but is one structured of multiple nests and a responsive stroma.
A review of a group of metastasizing melanomas will easily establish the primacy of vertical growth in the diagnosis of melanoma. For some thin lesions, as few as five closely spaced, and stratified nests of uniformly atypical cells (at least moderate atypia) will be sufficient to identify the patterns as those of an early (young) vertical growth component. The problem then becomes one of assigning lesions to appropriate categories. In variant vertical growth, the loose spacial relationships of fascicles, and nests of cells may be misinterpreted as the patterns of a nevus or a dysplasia, especially if too great a reliance has been placed on pattern analysis.
In fully evolved melanomas, a vertical growth component (regularly and closely spaced nests of atypical cells in the dermis, a three dimensional pattern), is the dominant histologic feature, and the main determinant of prognosis. In the precursors of such lesions, nests of neoplastic cells are randomly, and sparsely distributed in the dermis (if they are even represented at all, a relatively two dimensional pattern). Thus, in th common melanocytic, neoplastic system, precursors (i.e., SSM at level II, “melanoma-in-situ,” lentigo maligna, and the dysplasias [if distinctions between all these processes are justified]) are basically two dimensional lesions, and fully evolved melanomas are basically three dimensional lesions (dimensionality being a measure of the configuration of multiple nests, rather than a solitary nest).
Vertical growth is manifested in at least two patterns. In typical vertical growth, tumor cells in nests, or fascicles induce a continuous (community) stroma. In aggregate, the nests, and their stroma form a community. The resulting lesion would have an expansile quality, and would appear to have pushed the collagenous framework of the reticular dermis aside, or to have compressed it. A lesion with these characteristics is a level III lesion regardless of its vertical dimensions, or of the close proximity of its deep margin to anatomic markers such as the sweat glands.
Arrested level III growth:
Stratified nests are a common manifestation of patterns in the papillary dermis in melanocytic neoplasias. The patterns are such that, with the imposition of virtual images, a pattern with 3 dimensions can be virtually structured from the examination of histologic sections having only 2 dimensions. In the concept of MDM, such patterns, resulting from the amalgamation of real and virtual images, come to be acceptable as vertical growth. In some exceptional examples, if each nest is sequestered in concentric lamellae of fibrous tissue, and if inflammation is not a prominent feature, the 3 dimensional patterns qualify as arrested growth, even though the spacial relationships are those of pattern III. The patterns of arrested vertical growth are a modification of patterns that are common in premalignant dysplasias. They qualify as “arrested level III growth.” Their presence in a problematic lesion is always troublesome and, on this basis, arrested level III growth in a thin lesion (i.e., a lesion less than 1 mm in height) should be additionally qualified as a borderline pattern (i.e., borderline melanocytic neoplasia of indeterminate malignant potential; pattern of arrested variant vertical growth). Some of these arrested variants are composed of round, or polygonal cells in rounded nests which are loosely, but regularly, spaced. Usually, such lesions are confined to a widened papillary dermis. These lesions generally are associated with patterns at the dermal-epidermal interface that resemble those of the radial growth component of the common melanoma (SSM). In these lesions, imposed virtual images might lead to the conclusion that, with time, this pattern of vertical growth will give way to typical vertical growth. Typical vertical growth would then appear to be sequenced later then variant vertical growth in neoplastic progressions (with the stipulation that typical VG might, on occasion, arise de novo without the necessity for an intermediate variant phase).
In the concept of variant vertical growth (6), it is proposed that a clone of cells at the dermal-epidermal interface acquires the capacity to supply nests of cells to the papillary dermis at a rate which exceeds the capacity of the host immune response to control, and confine, all dermal, neoplastic components. In early, or incipient, variant vertical growth, the nests are loosely stratified, and at least two strata with 2 or 3 nests per row are required (e.g., a total of 5 to 6 such nests in the prescribed pattern would be sufficient). In addition, the images should include markers for host immune response, at least moderate cytologic atypia in the stratified component, and a fairly uniform degree of atypia. In variant vertical growth, the pertinent virtual images define a nidus of proliferating cells at the dermal-epidermal interface, and the transfer of some of these nests into the dermis in a pattern of accretive growth (with a vector perpendicular to the surface of the skin). In such examples, growth of the variant vertical growth component would not have acquired independence from phenomena at the dermal-epidermal interface. The requisite, that the nests accrete in strata at, and near, the dermal-epidermal interface, and are dependent on cellular kinetics along this interface, may correlate with the observation that, in variant vertical growth, the cells in nests near the interface between the papillary dermis, and the reticular dermis are smaller, and less atypical than those in nests at the dermal-epidermal interface. The locus for the release of growth factors may play a more significant role in modifying histologic patterns in variant, than in typical, vertical growth (3,6,7).
Some melanomas enter vertical growth in progressive variant patterns. In such thin lesions, the pattern of arrested variant vertical growth is never a feature, or is fleeting; such lesions tend to be characterized by spindle cells, and by fascicular patterns. Fascicular, spindle cell melanomas are not predictably sequenced to proceed from variant to typical vertical growth. For some fascicular spindle cell melanomas, variant vertical growth is the only option in the papillary dermis. In these examples, there is progression in bulk from variant vertical growth at level III to migrant (diffuse) vertical growth at level IV. In this sequence, a distinction between variant, and migrant vertical growth is defined by identifying fascicles of cells among pre-existing collagen bundles of the reticular dermis. Lesions of this type also tend to be associated with minimal markers for host immune response. The neoplastic cells often are non-descript in type, but such lesions may be erroneously assigned to a “Spitz” category. These fascicular, variant and migrant vertical growth melanomas commoly are organoid with a configuration of an inverted wedge, and with fairly uniform nuclear characteristics (cytologically, low grade tumors); they are of a type that might easily be accommodated by an observer’s virtual images of a “nevoid” melanoma.
|
