A diagnostic algorithm to distinguish desmoplastic from spindle cell melanoma

Abstract

Spindle cell melanoma and desmoplastic melanoma differ clinically in prognosis and therapeutic implications; however, due to partially overlapping histopathological features, diagnostic distinction of spindle cell from desmoplastic melanoma is not always straightforward. A direct comparison of diagnostic and therapeutic biomarkers has not been performed. Meta-review of the literature discloses key clinicopathological differences between spindle cell and desmoplastic melanoma, including immunophenotypes. Using 50 biomarkers available in routine diagnostics we examined 38 archival cases (n = 16 spindle, 18 desmoplastic, 4 mixed spindle/desmoplastic melanoma). S100 remains as the most reliable routine marker to reach the diagnosis of melanoma in spindle cell and desmoplastic melanoma. We identified 9 distinctly labeling markers with spindle cell melanoma showing positivity for laminin, p75, HMB45, c-kit, and MelanA, and desmoplastic melanoma preferentially labeling with collagen IV, trichrome, CD68, and MDM2. Based on comparisons of test performance measures, MelanA and trichrome were used to devise a 94 % sensitive diagnostic algorithm for the distinction of desmoplastic from spindle cell melanoma. Gene amplification and expression status was assessed for a set of potentially drugable targets (HER2, EGFR, MET, MDM2, TP53, ALK, MYC, FLI-1, KIT). Fluorescent in situ hybridizations did not reveal a significant number of gene abberations/rearrangements; however, protein overexpression for at least one of these markers was identified in 35 of 38 cases (92 %). In addition, we found BRAF mutations in 36 % of spindle cell and 5 % of desmoplastic melanoma with an overall mutation frequency of 16 % (n = 6/38). We present the first comprehensive screening study of diagnostic and therapeutic biomarkers in spindle cell and desmoplastic melanoma. The devised algorithm allows diagnostic distinction of desmoplastic from spindle cell melanoma when routine histology is not decisive.