Cell DNA content measurements, including determination of tumor ploidy and S-phase fraction, have been performed on a wide variety of human tumors for > 20 years using flow cytometry. During this time many publications have discussed the clinical utility of cell DNA content measurements. A major impediment to the more widespread application of cytometric DNA content measurements has been the lack of agreement among these studies. Whereas some discrepancies can be attributed to poorly designed studies lacking sufficient follow-up or significant numbers of patients, in many cases the discrepancies are due to technical factors in flow or image cytometry.The terminology used to describe the results of flow cytometry studies is often confusing and not universally applied. Although a convention for nomenclature for all DNA cytometry was recommended in 1984 (l), the guidelines suggested are frequently not used in published studies. Cytometric studies should not use cytogenetic terminology (hypodiploid, peritetriploid, etc.), where no direct measurement of changes in the number or composition of individual chromosomes has been made. Rather, the terms DNA diploid and DNA aneuploid should be used, with identification of the degree of DNA content abnormality given by the use of the DNA index, or DI (ratio of mean or mode of sample G,/G, population divided by mean or mode of diploid reference cells).One critical aspect of the clinical applications of these measurements is the use of standardized procedures to prepare and analyze clinical samples and to analyze and interpret cytometry data. A number of studies have demonstrated significant intralaboratory, as well as interlaboratory variation in the results of DNA content analyses (2-5) and in the interpretation of flow cytometric data (6). The purpose of this work is to help increase the reliability and reproducibility of DNA content flow cytometry by pointing out important technical considerations for cytometry, to provide guidelines that logically follow from these considerations, and to provide a framework for the development of standards and standardization of DNA content flow cytometry. Although a number of reported studies have utilized image cytometry to provide DNA content analysis, the technical differences between image and flow cytometry suggest that guidelines for the clinical application of image cytometry be developed independently.In reviewing the published literature, it is clear that many studies fail to provide sufficient information to judge critically the quality of cytometric measurement. It is recommended that all publications using DNA content cytometry provide details of the technique used to isolate and prepare cells, data that indicate that the sample used for cytometry contains representative tumor material, details concerning the techniques used to stain cells or nuclei (dye concentration, enzyme concentrations in units, cell concentrations), and details of the techniques used to analyze DNA content histograms (debris and aggregation correction ...

Cell DNA content measurements, including determination of tumor ploidy and S-phase fraction, have been performed on a wide variety of human tumors for > 20 years using flow cytometry. During this time many publications have discussed the clinical utility of cell DNA content measurements. A major impediment to the more widespread application of cytometric DNA content measurements has been the lack of agreement among these studies. Whereas some discrepancies can be attributed to poorly designed studies lacking sufficient follow-up or significant numbers of patients, in many cases the discrepancies are due to technical factors in flow or image cytometry.The terminology used to describe the results of flow cytometry studies is often confusing and not universally applied. Although a convention for nomenclature for all DNA cytometry was recommended in 1984 (l), the guidelines suggested are frequently not used in published studies. Cytometric studies should not use cytogenetic terminology (hypodiploid, peritetriploid, etc.), where no direct measurement of changes in the number or composition of individual chromosomes has been made. Rather, the terms DNA diploid and DNA aneuploid should be used, with identification of the degree of DNA content abnormality given by the use of the DNA index, or DI (ratio of mean or mode of sample G,/G, population divided by mean or mode of diploid reference cells).One critical aspect of the clinical applications of these measurements is the use of standardized procedures to prepare and analyze clinical samples and to analyze and interpret cytometry data. A number of studies have demonstrated significant intralaboratory, as well as interlaboratory variation in the results of DNA content analyses (2-5) and in the interpretation of flow cytometric data (6). The purpose of this work is to help increase the reliability and reproducibility of DNA content flow cytometry by pointing out important technical considerations for cytometry, to provide guidelines that logically follow from these considerations, and to provide a framework for the development of standards and standardization of DNA content flow cytometry. Although a number of reported studies have utilized image cytometry to provide DNA content analysis, the technical differences between image and flow cytometry suggest that guidelines for the clinical application of image cytometry be developed independently.In reviewing the published literature, it is clear that many studies fail to provide sufficient information to judge critically the quality of cytometric measurement. It is recommended that all publications using DNA content cytometry provide details of the technique used to isolate and prepare cells, data that indicate that the sample used for cytometry contains representative tumor material, details concerning the techniques used to stain cells or nuclei (dye concentration, enzyme concentrations in units, cell concentrations), and details of the techniques used to analyze DNA content histograms (debris and aggregation correction t...

The surface antigens expressed by the cells of chronic lymphocytic leukemia (CLL) are well known. Most CLL are monoclonal B-cell lymphoproliferative disorders characterized by the coexpression of B-cell antigens and CD5, an antigen present predominantly on T cells. Very little attention, however, has been paid to the quantitative characteristics of the expression of B-cell antigens in CLL. In this study, we used flow cytometry to analyze the expression of CD20, a well-known B-cell-associated antigen, in lymphocytes from 42 cases of CLL and its tissue counterpart, small lymphocytic lymphoma (SLL), and compared the results with results obtained from the analysis of 21 follicular lymphomas, 20 hyperplastic reactive nodes, and 26 samples of normal peripheral blood. The intensity of CD20 expression in the CLL/SLL cells was significantly lower than that of B cells in the other categories. This antigen expression abnormality does not appear to be a universal phenomenon in CLL/SLL, since CD19, another pan-B antigen, was expressed in CLL/SLL at levels higher than those in follicular lymphomas and comparable to those in reactive lymph nodes. These results indicate that the low CD20 expression can be used as a marker for CLL/SLL. The few cases exhibiting intense CD20 expression may represent a biologically different disease. CLL/SLL cells faintly expressing CD20 also show concomitant low CD5 expression in a manner not observed in normal CD5-expressing B cells.

Numerous prospective (fresh tissue) and retrospective (archival tissue) studies of DNA flow cytometry have been undertaken to ascertain the biologic behavior of hematopoietic malignancies. We have critically examined most of the studies that relate DNA ploidy and proliferative activity to clinical outcome in nonHodgkin's lymphomas, acute leukemias, and multiple myeloma. Selected studies relating DNA flow cytometry to histologic grade, but not clinical outcome, of nonHodgkin's lymphomas were also considered. The goal was to reach a consensus, based on a review of the literature and on our collective experience, on the value of DNA flow cytometry in the characterization of hematopoietic malignancies.The following review of the literature covers lymphomas, multiple myelomas, and acute leukemias. LYMPHOMAS Hodgkin's DiseaseThere have been a few studies demonstrating subtle DNA ploidy abnormalities in Hodgkin's disease (l), but there does not appear to be a rationale for performing routine flow cytometric DNA ploidy and cell cycle studies in this entity. However, if obvious DNA aneuploidy is encountered in tissues presumed to be Hodgkin's disease, a non-Hodgkin's lymphoma must be considered. Non-Hodgkin's LymphomasTwenty studies that correlated flow cytometric analysis of DNA ploidy and cell cycle with clinical outcome in patients with non-Hodgkin's lymphomas were reviewed (2-21). These studies are summarized in Table 1. Selection was based on the availability of histologic grade and clinical data, including adequate survival information and/or clinical outcome. In seven of the studies (742 patients), the analyses were performed on freshly prepared cell suspensions. In the remaining 13 (1,471 patients), nuclei were extracted from archival specimens embedded in paraffin blocks. There was great variability with respect to numbers of cells analyzed, staining procedures, instrumentation, amount of debris, frequency of DNA-aneuploidy, and coefficient of variation (CV). The CVs were higher in the archival specimens (mean 5.4%; range L3.8-7.0%1) than in the fresh specimens (mean 3.2%; range [2.5-4.4%1). The data analysis of cell cycle kinetics varied from study to study. There were differences in the mathematical strategies for cell cycle phase calculations. Additionally, numerous laboratories excluded cases that had overlapping DNA-diploid and DNA-aneuploid histograms. Some laboratories reported proliferative indices (S + G,M), whereas others reported S-phase fractions.General conclusions are limited by nonuniformity of therapeutic regimens and variability and flaws in the analysis of survival data.Several publications that correlated DNA-ploidy and cell cycle kinetics to histopathologic grade were reviewed (22)(23)(24). These were considered to be informative and of high quality. Although a direct correlation with survival and/or clinical outcome was not provided, it is generally accepted that most histopathologic classifications of non-Hodgkin's lymphomas correlate well with survival. Clinical Utility of DNA Index inNon...