No abstract

There are differences between PDM and MDM, the most important of which is the incidence of LN involvement. Our findings support the clinical utility of classifying DM into pure and mixed subtypes because the negligible rate of nodal involvement in PDM does not support the routine performance of sentinel LN biopsy in this subgroup of melanoma patients. In contrast, the incidence of LN involvement in MDM is comparable to that of CM.

Basaloid squamous carcinoma (BSC) of the head and neck has been shown to have a poor prognosis when compared with conventional squamous cell carcinoma (SCC). Pathologically, specimens determined to be BSC can have nearly pure basaloid features (group 1) or a mixture of basaloid and squamous features (group 2). The clinical behavior in these 2 subgroups has not been compared previously. BSC is also commonly confused histologically with poorly differentiated SCC (PDSCC). A retrospective comparison of disease stage at presentation, rate of distant metastasis, rate of local recurrence in those offered surgical resection, and rate of survival is made to compare outcomes of the 2 BSC groups and the PDSCC group. The presence of particular histologic features may be associated with poorer outcomes. Patients with BSC have advanced disease at presentation. Survival in the BSC group was less than half that in the PDSCC groups. Statistical analysis shows the 2 groups to be well matched with regard to stage and site of disease. Presence of neck nodal disease on presentation predicts poor survival. In this study distant metastases occurred in 52% of patients with BSC and in 13% of patients in the PDSCC group. The local recurrence rate is comparable for BSC and conventional SCC, with even early tumors in the BSC group recurring distantly rather than locally or regionally. Considering the high distant metastatic rate of BSC and poorer overall survival rate, a more extensive metastatic survey is indicated in these patients before surgery is recommended. We recommend that patients with a diagnosis of BSC not be included with conventional SCC groups in prospective randomized cancer protocols.

We evaluated the usefulness of immunohistochemical examination for E-cadherin, p16, and cyclin D1 in discriminating melanoma from Spitz tumors. Immunoperoxidase staining was performed on formalin-fixed tissue specimens from 46 Spitz tumors and 42 concurrent melanoma specimens. The percentages of immunoreactive melanocytes in the epidermis and dermis were estimated semiquantitatively. Qualitatively abnormal immunoreactivity patterns were also tabulated. Dermal p16 immunoreactivity was the best quantitative discriminator: decreased nuclear immunoreactivity (<25% of dermal melanocytes) was 3-fold more likely in melanoma than in Spitz tumors (P = .004). Loss of both nuclear and cytoplasmic dermal p16 immunoreactivity was 8-fold more likely in melanoma (P = .01). Qualitative irregularities in the zonal distribution of E-cadherin immunoreactivity were 2-fold higher in melanoma (P = .01), but these were often focal or subtle. There was no statistically significant difference in cyclin D1 immunoreactivity. In atypical Spitz tumors, the dermal p16 immunoreactivity and frequency of qualitative E-cadherin abnormalities were intermediate between those of ordinary Spitz nevi and melanoma. Also, contrasting immunoreactivity patterns were helpful in determining Breslow thickness in specimens containing melanoma and contiguous dermal nevi.

No abstract