Deoxyribonucleic acid (DNA) ploidy was quantified in 84 ocular melanomas (median follow-up, 11-years) by flow cytometry, CAS 200 interactive image analysis, and Pathology Image Processing Environment (PIPE; Department of Quantitative Pathology, Free University, Amsterdam, The Netherlands) automatic image analysis (75). Overall, 32.1% of the melanomas were aneuploid, 2.3% were tetraploid, and 66.6% were diploid. Pathology Image Processing Environment analysis estimated DNA ploidy in 12 tumors that were unprocessable by flow cytometry. Of 10 tumors that were diploid by flow cytometry, PIPE detected stemline aneuploidy in five and some aneuploid cells in five more. Seven tumors were aneuploid by PIPE but diploid by CAS 200, five of which contained occasional DNA aneuploid cells on the CAS 200 histograms. Pathology Image Processing Environment analysis quantified tumor samples with an average of 500 (250 to 1,050) cells in less than 10 minutes. All cells classified as spindle A according to the Callender system (more than 10,000) were diploid. Spindle B and epithelioid cells occupied both diploid and aneuploid peaks on the DNA histograms. Dioxyribonucleic acid variables did not correlate with established prognosticators, such as Callender cell type, largest tumor dimension, or glaucoma, nor did they reach significance by univariate and multivariate analyses. The value of these findings as a diagnostic support in uveal melanoma, particularly in combination with fine needle aspiration biopsy, is discussed.
- Callender classification
- flow cytometry
- image analysis
ASJC Scopus subject areas
- Pathology and Forensic Medicine