PROGNOSTIC-SIGNIFICANCE OF DNA-PLOIDY AND PROLIFERATION IN 309 COLORECTAL CARCINOMAS AS DETERMINED BY 2-COLOR MULTIPARAMETRIC DNA FLOW-CYTOMETRY

BACKGROUND. Although DNA flow cytometry has been shown to be of indepe ndent value in determining the prognosis of colorectal carcinoma, a nu mber of well-designed studies with contradictory findings have left un resolved the clinical significance of DNA ploidy and proliferation in biologically meaningful subsets of colorectal carcinoma cases. METHODS . To evaluate the prognostic significance of DNA ploidy and proliferat ion as determined by flow cytometry in a prospective series of 309 hum an colorectal carcinomas with 4-6 years of follow-up, fresh tumors wer e mechanically dissociated into whole cell suspensions and dual fluore scence-labeled to allow gated analysis of subpopulations with phenotyp ic markers. Software programs with histogram-dependent algorithms empl oying background, aggregate, and debris correction were used in DNA an d cell cycle quantitation. Data were analyzed according to recommendat ions of the 1992 DNA Flow Cytometry Consensus Conference.RESULTS. None of the clinical, site, or staging parameters, including TNM stage var iables, correlated with any flow cytometric DNA ploidy or proliferatio n measurement. Tumors classified as DNA aneuploid or tetraploid, by an y definition, did not differ in prognosis or correlate with stage or a ny pathologic parameter. Univariate Kaplan-Meier survival analysis sho wed prognostic significance of the following: Dukes staging, individua l components of TNM stage (tumor depth, lymph node status, and metasta sis), vascular invasion, histologic pattern of tumor infiltration, and peritumoral lymphocytic inflammation. DNA ploidy status and prolifera tion measurements were not predictive of survival for the overall grou p or within any particular stage. Only Dukes Stage D (metastasis), vas cular invasion, and depth of invasion (T classification) were signific ant independent predictors of survival in multivariate Cox regression models. CONCLUSIONS. In this analysis, DNA ploidy and proliferation me asurements were not predictive of survival in any stage of colorectal carcinoma. However, clinical and pathologic documentation of staging a nd select histopathologic observations were significant predictors of survival in univariate and multivariate analyses. (C) 1997 American Ca ncer Society.