RESUMEN
Flow cytometric analyses of DNA content and proliferative fractions have been found to be important prognostic indicators in a variety of human tumors. However, variability in reported results and interlaboratory differences in single-parameter DNA measurements have impeded the broader use of this methodology. Multiparameter DNA analysis, especially that which allows the ploidy and cell cycle measurements to be targeted specifically to tumor cells, may improve the quality and reliability of these measurements. Cytokeratin labeling simultaneously with DNA allows the identification of the malignant epithelial cells within tumor samples that have been microdissected for tumor enrichment and, thus, can increase the accuracy of tumor ploidy and cell cycle measurements. There are a limited number of reports of cytokeratin labeling of paraffin-extracted cells, and results with standard preparation procedures can be highly variable. We have developed an improved technique for cytokeratin labeling of paraffin-extracted cells that is based on predigestion of tissue with collagenase prior to brief pepsin digestion. This two-step enzymatic digestion produces a superior cytokeratin vs. DNA bivariate analysis, with increased intensity and greater uniformity of cytokeratin labeling. This method should increase the accuracy of ploidy and proliferation measurements from paraffin-embedded tissue both for retrospective studies and in clinical settings in which only fixed tissue is available.