I read with interest the excellent article about primary dermal melanoma that was recently published in your journal.1 While the authors did not specify the stage of the melanoma case they reported, they seemed to imply that it was stage IV.

The pathogenesis of tumors such as the one described, however, is not clear. There are several mutually nonexclusive hypotheses that can explain the presence of a single melanoma nodule in the dermis. Because not all the hypotheses involve a primary dermal origin, a more correct term for such a nodule would be solitary dermal melanoma (SDM).

SDM could originate from a primary tumor with a completely regressed junctional component, an intradermal melanocytic nevus,2,3 or a dermal melanocytosis.4 It might also be derived from a melanocytic cell that was trapped in the dermis during embryogenesis, or from melanocytes associated with appendageal structures.5 In all these cases, the tumor would be considered localized melanoma (T1-4). The nodule, however, could also be a metastasis, in which case it would be classified as M1a if it were a distant metastasis or as N2c if it were an in-transit metastasis.

The above reflections simply mean that different stages will be assigned depending on how the tumor is considered (presuming that staging studies have not detected disease at other sites). In the cases of localized melanoma described above (T1-4a, as SDM will never be ulcerated), the tumor could be assigned a stage as high as IIB if it exceeds a depth of 4mm, while it would be categorized as stage IV or stage IIIB if it were considered a distant or in-transit metastasis, respectively. Clearly, as shown by the literature, the assignment of one stage or another has a very important bearing on treatment, which can range from excision of the melanoma1 to chemotherapy.6,7

Nevertheless, the above reflections are not what prompted me to write this letter, but rather the fact that the Final Version of the 2009 American Joint Committee on Cancer Melanoma Staging and Classification8 specifies that single dermal nodules should be considered regional. In other words, they should be classified as N2c (stage IIIB). According to this classification, a stage III melanoma with T1-4 N2c M0 would have a 5-year survival rate of 69%, which is similar to the rate of 66% reported by Lee et al.9 in the largest series of SDM published to date.

Although what I propose is probably not the best solution, until we have a better understanding of the pathogenesis of SDM or are able to identify the origin of each tumor, I think that single melanoma nodules in the dermis should be considered regional, classified as N2c, and called SDM.