COMPUTER-ASSISTED IMAGE-ANALYSIS OF BLADDER-TUMOR NUCLEI FOR MORPHONUCLEAR AND PLOIDY ASSESSMENT

Morphonuclear analysis using a quantitative image analysis system has been demonstrated to be a potentially useful technique in the prognost ic evaluation of bladder carcinoma. The integrated optical density par ameter permits DNA content evaluation in addition to 15 other morphonu clear parameters. We assessed the reliability of morphonuclear analysi s by image analysis and flow cytometry for 46 bladder carcinomas and 1 4 normal bladder specimens. Frozen sample material was obtained from e ndoscopic resection, radical cystectomy and from cadaveric donors. The grade and staging of the tumours according to the World Health Organi zation was as follows: 8 G1, 18 G2, 20 G3 and 28 T1, 7 T2, 7 T3, 4 T4. Quantitative image analysis was made on imprint smears stained by the Feulgen method. Simultaneously cell suspensions were obtained by mech anical dissociation and stained with propidium iodide for Flow cytomet ry analysis. There was a good agreement between quantitative image ana lysis and flow cytometry for DNA content measurement indicating that i mage analysis is a reliable method for the quantitation of DNA content (P = 0.001). Moreover we found a good correlation between five of the morphonuclear parameters: surface (P < 0.001), chromatin clumps distr ibution (P < 0.001), frequency of small (P < 0.001) to large chromatin clumps (P < 0.001) and the grade of bladder tumours. These results in dicate that morphonuclear analysis may be a valuable method of quantit ating DNA and morphonuclear parameters in a single analysis to provide information which may have some prognostic significance for patients with bladder carcinoma.