Premalignant Melanocytic Dysplasias. Reed RJ, Clark WH Jr, and Mihm MC. In Pathology of Malignant Melanoma. Ackerman AB, ed. Masson, New York, 1981;159–83.

"The concept of melanocytic dysplasia makes clearer the terminology for classifying melanocytic tumors. It is based exclusively on histologic evaluations and complements, rather than supplants existing concepts based on clinicopathologic correlations. A diagnosis of a dysplasia gives recognition to epidermal and dermal patterns, melanocytic configurations (spindle or epithelioid), degree of cytologic dysplasia, associated histologic changes (i.e., actinic damage), host immune response (melanocytic lichenoid reaction), regression, and preexisting benign lesions. On the basis of this concept, most examples of lentigo maligna and lentigo maligna melanoma (level II) qualify as mild or moderate, actinic, lentiginous melanocytic dysplasias."

Atypical (B-K) moles are said by the authors of a chapter in this book (Figs. 11, 12) to possess a lentiginous component of melanocytes that extends beyond the dermal component of the nevus, i.e., a shoulder, a feature which they insist is not possessed by "common acquired nevi." This is the first time that proponents of the concept of "melanocytic dysplasia" set forth a finding that they contend differentiates, consistently, "atypical" moles from "common" ones. If their premise is incorrect, then everything that flows from it also must be wrong.

Reed et al.: "The development of a concept of cutaneous melanocytic dysplasia has been hampered by a peculiarity of the melanocyte: it is not confined to an epithelial layer in the evolution of its hamartomas."

Critique: "Dysplasia" is linked by Reed et al. to hamartomas of melanocytes, but the authors fail to provide definitions for either "dysplasia" or hamartoma (Fig. 13, their Fig. 1). Until this day, the co-workers have not defined "melanocytic dysplasia" in the same way in any two of the scores of articles written by them about the subject. That deficiency prevents readers from comprehending the two major concepts that derive from the notion of "melanocytic dysplasia," to wit, DN and the DNS.

Reed et al.: "The delineation of the histologic patterns in the atypical mole-familial melanoma syndrome has established a basis for the recognition of melanocytic dysplasias. The changes in the atypical moles establish cytologic atypism, aggregation and distribution of atypical melanocytes, and host immune response as features to be evaluated."

Critique: Nuclear atypia is not a feature we are able to appreciate in melanocytes of B-K moles and the notion of assessing host immune response in a B-K mole by microscopy is specious. Those who advocate the validity of judging host immune response by morphologic findings imply that lymphocytes, fibroplasia, and proliferation of blood vessels telegraph information relevant to the immunologic status of a host. But the densest of lymphocytic infiltrates in any melanocytic neoplasm is found in melanomas en route to complete regression, a phenomenon associated with a grave prognosis, i.e., death from metastasis. By contrast, some very thin melanomas with an excellent prognosis are accompanied by scant infiltrates of lymphocytes or no lymphocytes at all. Density of infiltrates of lymphocytes in a melanocytic neoplasm does not convey meaningful information about the actual immunologic status of a person who bears it.

Reed et al.: "It is also necessary to define deviant patterns which distinguish dysplasias from melanomas. As an outgrowth of these definitions, minimal deviation and borderline minimal deviation melanomas emerge as variants."

Critique: Reed and co-workers muddy the waters of melanocytic dysplasia even further by introducing what they consider to be outgrowths from it, namely, "minimal deviation melanoma" and "borderline minimal deviation melanoma." They do not define "deviant patterns" of melanomas, but rather describe various histopathologic findings in an obtuse way, thereby adding still more confusion to the opaque concept of melanocytic dysplasia. What could be more confusing than to diagnose melanoma as dysplasia? (Fig. 14, their Fig. 3) Terms like "minimal deviation" and "borderline" in diagnosis of melanocytic, or any other, neoplasms are evasions from either making a specific diagnosis of nevus, melanoma, or melanoma in association with a nevus, or acknowledging that one does not know the diagnosis.

Reed et al.: "The cytological changes in dysplasia are graded as mild, moderate, or severe. The dysplasia is evaluated by the same criteria as epithelial dysplasia (e.g., nuclear hyperchromatism, irregularities in the nuclear membrane, nuclear size, mitoses, loss of cohesion, etc.) Variations in nuclear size and nuclear hyperchromatism are common features in acquired and congenital melanocytic nevi. In the absence of lymphoid infiltrates, abnormal cellular aggregates, lamellar fibrosis, or mitoses, these cytologic features are of questionable significance with the possible exception of the blastomatous melanomas of infancy."

Critique: Reed et al. do not provide any criteria for distinguishing among mild, moderate and severe dysplasia, but simply give examples of each of them. In Figure 15 (their Fig. 12) of their chapter, the authors even employ the word "moderately" to describe severe "dysplasia." All of this is dizzying. In the quotation, they use "dysplasia" as a synonym for "cytologic atypia," but they acknowledge that cytologic atypia, i.e., variations in nuclear size and nuclear hyperchromatism, is "common" in acquired (and congenital) melanocytic nevi. To which acquired nevi are they referring, to B-K moles (the commonest acquired melanocytic nevi), to what they call "common," "ordinary," or "banal" melanocytic nevi, or to Spitz's nevus? It is impossible for a reader to come to a rational answer to this query on the basis of the information provided by Reed et al.

Reed et al.: "If a lentiginous component coexists with any of the above variants [of melanocytic dysplasia], it may manifest asymmetry by irregular extensions beyond the confines of the main portion of the lesions."

Critique: "Irregular extensions" of the lentiginous component refers to the shoulders of a B-K mole, findings that Reed et al. consider to be an evidence of abnormality in a CAN and a sign that "dysplasia" has supervened, i.e., a DN has come into being. In our experience, shoulders are nearly invariable in compound CAN as conceived of by Reed et al. because CAN are simply small DN. The notion of absence of shoulders from a "normal" mole derived from Reed's observations about dome-shaped nevi (designated "Miescher's nevi" by us) that favor the face and not from his study of B-K moles (called "Clark's nevi" by us) that rarely occur on the face*** (Fig. 16). His observations about that nevus on the head were correct, but they are not applicable to B-K moles.

Reed et al.: "The lentiginous component in the epidermis commonly extends beyond the limits of the dermal components (lentiginous asymmetry) . . . The full expression of a lentiginous and compound melanocytic dysplasia is common in the nevi in the atypical mole-familial melanoma syndrome (B-K mole syndrome). Similar lesions are also seen sporadically. They qualify as lentiginous and junctional, compound or dermal dysplasias."

Critique: Reed and co-workers consider shoulders crucial to histopathologic diagnosis of B-K moles and equate shoulders with "dysplasia." We agree that shoulders are nearly invariable in the compound type of those exceedingly common acquired nevi, but disagree that diagnosis of those nevi by conventional microscopy turns on shoulders. Our criteria for diagnosis of compound Clark's nevus are silhouette of a benign neoplasm, e.g., symmetry, sharp circumscription, and maturation of melanocytes, slight elevation of the lesion above the skin surface, and confinement of nests of small oval monomorphous melanocytes to the dermo-epidermal junction and papillary dermis.

Reed et al.: "Atypical melanocytes in nests in the epidermis and closely aggregated fascicles of mildly or moderately atypical melanocytes in the upper portion of the papillary dermis are occasionally confined to a small area in a preexisting nevus."

Critique: A histopathologist is not able to recognize "mildly atypical melanocytes" because criteria have not been established for them. Even if such criteria existed, the changes would be too subtle to identify with repeatability. Last, nuclear characteristics of melanocytes are irrelevant to diagnosis of DN; the diagnosis is made at scanning magnification by attributes referable to silhouette.

Reed et al. imply that "dysplasia" at the shoulders of a B-K mole represents superimposition of those changes on a pre-existing CAN. Our interpretation is different. The entire lesion is a B-K mole from the beginning, the shoulders being as much a part of the nevus as the rest of the intraepidermal and the intradermal components.

Reed et al.: "In the diagnosis of a dysplasia, nevus cell populations mark a precursor phase. If 50% or more of the cells in a lesion are dysplastic, the nevus cell component is relatively inconspicuous and is easily ignored in the diagnosis. If the dysplastic component forms 25% or less of total population of melanocytes, the lesion is classified, preferably, as an atypical nevus (junctional, compound, dermal, congenital, or acquired) and qualified as to degree and types of dysplasia (lentiginous, junctional, compound, or dermal)."

Critique: Not only is the concept of "melanocytic dysplasia" befuddling, but it is entirely arbitrary. How were the numbers (25%, 50%) for dysplastic cells in a lesion arrived at? In this quotation, Reed et al. use "dysplasia" for cytologic features of melanocytes, whereas at other times they employ it to mean atypical hyperplasia, and at still other times to signify the constellation of atypical melanocytic hyperplasia, fibroplasia, neovascularization, and infiltrates of lymphocytes. The authors again imply that dysplasia is a process that develops in a preexisting CAN. There is no evidence whatsoever for that assertion. Types of "dysplasia" are said to be "lentiginous, junctional, compound, and dermal," but junctional, compound, and dermal are histopathologic signs of stages in the evolution of all acquired melanocytic nevi, junctional being relatively early and dermal being late. Lentiginous refers to an increase in number of melanocytes disposed mostly as solitary units at bases of elongated rete ridges. Why are not epithelioid and spindle cell "dysplasias" mentioned in this context? A reader can only be bewildered.

Reed et al.: "The concept of melanocytic dysplasia makes clearer the terminology for classifying melanocytic tumors."

Critique: We see it differently. Not only would general pathology, dermatology, and dermatopathology have been better served without the introduction of the concept and term (melanocytic) "dysplasia," but so, too, would medicine in general. All of this redounds to the disadvantage of patients. The concept of melanocytic dysplasia is impenetrable, hampers classification of "melanocytic tumors," hinders histopathologists from coming to specific diagnoses of nevus, melanoma, or melanoma in association with a nevus, and puts patients at risk for unnecessary surgery (a nevus is a nevus whether or not it is termed dysplastic)10 and for emotional trauma (a single DN was said, repeatedly, to enhance risk markedly for melanoma because a person who bore it had the DNS).