AIMS: Primary cutaneous desmoplastic melanoma (DM) may be entirely desmoplastic ['pure' DM (pDM)] or exhibit a desmoplastic component admixed with a non-desmoplastic component ['combined' DM (cDM)]. Our aim was to describe the histological features of metastases of primary DM and to determine whether they were predictive of outcome. METHODS: The effect of clinicopathological parameters on overall survival (OS) was analysed, and the correlation between histological features of the primary melanoma and metastases was studied in patients with metastatic DM. RESULTS: Twenty-six patients (18 males; eight females) developed 50 metastases in sentinel nodes (13; 26.0%), regional nodes (10; 20.0%), skin (14; 28.0%), and distant sites (13; 26.0%). The cellular composition of metastases correlated with that of the corresponding primary tumours. Time to development of first non-sentinel lymph node metastasis was shorter in cDM than in pDM (median 11.2 versus 24.9 months, P = 0.075). The only independent predictors of poorer OS were cDM type [hazard ratio 6.17, 95% confidence interval (CI) 1.61-23.81, P = 0.008] and male sex (hazard ratio 5.98, 95% CI 1.34-26.66, P = 0.019). CONCLUSIONS: The cellular composition of primary DM correlates with that of metastatic DM and with outcome. It is important to consider the possibility of primary or metastatic DM when examining tumours composed of spindle cells.
AIMS: Primary cutaneous desmoplastic melanoma (DM) may be entirely desmoplastic ['pure' DM (pDM)] or exhibit a desmoplastic component admixed with a non-desmoplastic component ['combined' DM (cDM)]. Our aim was to describe the histological features of metastases of primary DM and to determine whether they were predictive of outcome. METHODS: The effect of clinicopathological parameters on overall survival (OS) was analysed, and the correlation between histological features of the primary melanoma and metastases was studied in patients with metastatic DM. RESULTS: Twenty-six patients (18 males; eight females) developed 50 metastases in sentinel nodes (13; 26.0%), regional nodes (10; 20.0%), skin (14; 28.0%), and distant sites (13; 26.0%). The cellular composition of metastases correlated with that of the corresponding primary tumours. Time to development of first non-sentinel lymph node metastasis was shorter in cDM than in pDM (median 11.2 versus 24.9 months, P = 0.075). The only independent predictors of poorer OS were cDM type [hazard ratio 6.17, 95% confidence interval (CI) 1.61-23.81, P = 0.008] and male sex (hazard ratio 5.98, 95% CI 1.34-26.66, P = 0.019). CONCLUSIONS: The cellular composition of primary DM correlates with that of metastatic DM and with outcome. It is important to consider the possibility of primary or metastatic DM when examining tumours composed of spindle cells.
Authors: Dale Han; Jonathan S Zager; Daohai Yu; Xiuhua Zhao; Brooke Walls; Suroosh S Marzban; Nikhil G Rao; Vernon K Sondak; Jane L Messina Journal: Ann Surg Oncol Date: 2013-02-07 Impact factor: 5.344
Authors: Dale Han; Gang Han; Xiuhua Zhao; Nikhil G Rao; Jane L Messina; Suroosh S Marzban; Amod A Sarnaik; C Wayne Cruse; Vernon K Sondak; Jonathan S Zager Journal: PLoS One Date: 2015-03-26 Impact factor: 3.240