Common and Dysplastic Nevi: A New Diagnostic Approach is Possible

To the Editor: I thank Stewart Cramer, M.D., for his comments on my article “Atypical Histologic Features in Melanocytic Nevi”(1). Dr. Cramer agrees that the histologic features proposed as diagnostic of dysplastic nevus are not specific. He also agrees that a new diagnostic approach (analytic diagnosis, i.e., to diagnose nevi as junctional, compound and dermal, reporting any atypical characteristics (2) and the class number) may be opportune. However, Dr. Cramer disagrees with the conclusion that dysplastic nevus is not a distinct histologic entity. I believe that dysplastic nevus cannot be considered a distinct histologic entity because its “left border” is not defined. If we imagine melanocytic lesions as lying on a continuum, dysplastic nevus, which has intermediate histologic characteristics, is located between “common” nevus and melanoma. To be considered a distinct histologic entity, dysplastic nevus should be distinguished clearly from “common” nevi and from melanoma (i.e., its “histologic borders”(3) should be well defined). Actually, although the “right border” of dysplastic nevus (facing melanoma) is rather well defined (4), the “left border” (facing common nevus) is not, because nevi do not form two distinct classes of lesions (common nevi with no atypical features and dysplastic nevi with all atypical features), but they form a complex spectrum of lesions with a progressively increasing incidence of atypical features. However, the fact that dysplastic nevus is not a distinct histologic entity does not imply that all nevi are equal in terms of risk for melanoma (at contiguous and noncontiguous sites) (5). The problem is to identify lesions implying an increased risk, if they exist and if they are histologically recognizable. I believe that they may exist, because a conspicuous proportion of melanomas seems to arise in a preexisting nevus. I am not sure that nevi implying an increased risk for melanoma are recognizable. I believe, however, we can try to identify them. I agree with Dr. Cramer that size of nevi may be an important parameter, although nevi smaller than 5 mm seem to have the same histologic characteristics of nevi greater than 5 mm. I also agree that in the spectrum of nevi, lesions showing a high number of atypical features (Class 5 nevi, and Class 6 nevi, which I have also observed in another series of nevi) may be potential candidates to be lesions implying an increased risk for melanoma. Of course, further studies are needed. First of all, however, it is necessary to diagnose nevi more accurately and to stratify them histologically in objectively defined classes of lesions. Then, it will be possible to investigate the obtained different classes of nevi at all levels and to study corresponding patients clinically and epidemiologically. Carmelo Urso, M.D.