Histogenesis of cutaneous malignant melanoma: The vast majority do not develop from melanocytic nevus but arise de novo as melanoma in situ.

It has been considered that most cutaneous malignant melanomas arise from pre-existing melanocytic nevus. Many clinical and histopathological studies seem to support this concept. Dysplastic nevus originally proposed by Clark's group is a key entity of the intermediate lesion between benign nevus and melanoma. The latest edition of the WHO Classification of Skin Tumours (2018) has excluded Clark nevus (dysplastic nevus of mild atypia) from dysplastic nevus, the latter being now classified into the low- and high-grade by the degrees of nuclear atypia. The World Health Organization classification regards dysplastic nevus of both grades as the intermediate lesion between common acquired nevus and the radial growth-phase melanoma. An extensive genetic study recently performed by Bastian's group indicated the existence of intermediate lesions between nevus and melanoma. In spite of these findings, some investigators doubt the concept of the intermediate lesions including dysplastic nevus and insist that the majority of melanomas arise de novo as melanoma in situ, not in association with a preceding nevus. The concept of de novo genesis of melanoma may be supported by a recent meta-analysis study revealing that 71% of melanomas likely arose de novo and 29% from pre-existing nevus. In this review article, from the viewpoint of de novo genesis of melanoma, the author critically discusses the relevant findings of melanoma genesis and proposes a new framework to interpret the morphological and genetic data alternatively. Clarification of the oncogenic process of melanoma has great impact not only on clinical dermatology but also on basic oncology.