Use of multiple cytometric markers improves discrimination between benign and malignant melanocytic lesions

Autor: Allan A. Palmer, John F. Thompson, Kerry A. Crotty, Ling-Xi L. Li, Stanley W. McCarthy, Richard A. Scolyer, Jillian J. Kril
Rok vydání: 2003
Předmět:
Zdroj: Melanoma Research. 13:581-586
ISSN: 0960-8931
DOI: 10.1097/00008390-200312000-00007
Popis: Confident separation of benign naevi and malignant melanoma can sometimes be very difficult using conventional microscopy. This study evaluated the combined diagnostic abilities of multiple cytometric markers in separating various types of naevi from melanomas. The lesions studied included 27 benign compound naevi, 20 dysplastic naevi, 10 Spitz naevi and 24 melanomas. The cytometric features investigated were: (i) nuclear DNA content and chromatin compactness, measured by video imaged DNA microdensitometry; (ii) nuclear morphology, measured by nuclear morphometry (karyometry); (iii) transcriptional activity of nucleolar organizer regions, measured as the number and size of argyrophilic staining of nucleolar organizer regions (AgNORs); and (iv) cellular proliferative activity detected by quantifying the immunoreactivity of MIB1-Ki67 antigen. These variables were evaluated in the superficial, middle and deep zones of each lesion. Using multivariate discriminant analysis, a total diagnostic effectiveness of 97% could be achieved in separating the benign and malignant melanocytic lesions by co-evaluating variables for DNA microdensitometry, karyometry and AgNORs. A diagnostic effectiveness of 100% could be achieved if further co-evaluation with MIB1-Ki67 immunoreactivity was performed. Our study suggests that co-evaluation of multiple cytometric markers can improve the diagnostic abilities of individual techniques in separating benign naevi from malignant melanomas. This may be of particular significance in the diagnosis of melanocytic lesions whose biological behaviour cannot be confidently predicted by their histological features using conventional microscopy.
Databáze: OpenAIRE