Improved immunohistochemical evaluation of micrometastases in sentinel lymph nodes of cutaneous melanoma with 'MCW Melanoma Cocktail' – A mixture of monoclonal antibodies to MART-1, melan-A, and tyrosinase

摘要

Background MART-1, Melan-A, and Tyrosinase have shown encouraging results for evaluation of melanoma micrometastases in sentinel lymph nodes, as compared to conventionally used S-100 protein and HMB-45. To achieve higher sensitivity, some studies recommend evaluation of three sections, each at intervals of 200 μ. This would mean, routine staining of three adjacent sections in each of the three clusters at intervals of 200 μ, requiring nine slides resulting in added expense. If a cocktail of these antibodies could be used, only one section would be required instead of three generating significant cost savings. Methods We prepared a combination of monoclonal antibodies to these three immunomarkers in optimized dilutions (MART-1, clone M2-7C10, dilution 1:500; Melan-A, clone A103, dilution 1:100; and Tyrosinase, clone T311, dilution 1:50) and designated it as 'MCW melanoma cocktail'. Formalin-fixed paraffin-embedded tissue sections of sentinel lymph nodes from patients with cutaneous melanoma, without macro-metastases were evaluated with this cocktail. Results Melanoma micrometastases were easily detectable with the cocktail in 41 out of 188 slices (8/24 cases). The diagnostic accuracy amongst five pathologists did not show statistically significant difference. Out of 188 slices, 78 had adjacent sections immunostained individually with MART-1 and Melan-A during our previous study. Of these 78 slices, 21 were positive for melanoma micrometastases with MART-1 and Melan-A individually. However, the adjacent section of these slices immunostained with the cocktail detected metastases in four additional slices. Thus, MART-1 and Melan-A could not detect melanoma micrometastases individually in 16% (4/25) of slices positive with the cocktail. Benign capsular nevi were immunoreactive for the cocktail in 4.8% (9/188) slices. All 81 slices of negative test controls (sentinel lymph nodes of mammary carcinoma) were interpreted correctly as negative for melanoma micrometastases. Conclusions The melanoma cocktail facilitated easy interpretation of melanoma micrometastases in sentinel lymph nodes with high interobserver agreement. There was improvement in detection rate with the cocktail as compared to MART-1 and Melan-A individually. Furthermore, this approach facilitates cost savings.