These include proliferation markers measuring the only two phases of the cell cycle in which cells are detectable on the basis of morphological or phenomenological aspects or as a result of their capacity to incorporate DNA precursors.
Quantification of cells in the mitotic phase (10) is currently expressed as the number of mitotic figures per 10 high-power fields (mitotic activity index [MAI]) or, when corrected for field size and area fraction of the neoplastic epithelium, as standardized mitotic index (volume fraction-corrected mitotic index, or M/VV index, giving the result in mitotic figures per mm2 of neoplastic epithelium). Both methods of expressing the presence of mitotic figures provide comparable results, but the standardized mitotic index (SMI) has consistently shown smaller interobserver variations. These indices, which have long been employed as diagnostic and prognostic tools in the study of tumor pathology, are important components of all histologi-cal grading systems and are routinely used by pathologists. They do not require special processing or staining procedures or the fragmentation of tumor tissues. However, although an increased mitotic activity is a frequent finding in aggressive tumors, the validity of these measurements as markers of tumor proliferative activity will remain controversial (11) until they are standardized or until interlaboratory reproducibility is guaranteed. In fact, mitotic figure counting represents a simple, rapid, and highly feasible approach even for very small tumors, which, however, can be affected by biological and technical factors, and by intra- and interobserver variability owing to the subjective identification of mitotic figures. Although the latter weaknesses can be virtually eliminated by providing precise descriptive criteria for the morphological identification of mitoses, such as those developed by the Amsterdam group (12), technical aspects, including type and time of fixative, section thickness as well as drawbacks related to definition of high-power fields and total number of tumor cells can compromise the interstudy comparability of results. Finally, in addition to problems related to intratumoral heterogeneity and to the poor resolution of the cell kinetics parameter caused by the relatively short time of the M phase (40-60 min) compared to the duration of the entire cell cycle (40-50 h), metaphase arrest may also represent a final stage of cell life.
The quantitative determination of cells in the S phase initially based on the active incorporation of labeled ([3H]thymidine) or halogenated DNA precursors (bromo- or iododeoxyuridine) (13) was successively paralleled by flow or image cytometry of cells with an S-phase DNA content (14). Incorporation measurements, which are performed with autoradiographic or immunohistochemical techniques, require fresh material, aspirates, and surgical or bioptic specimens, and must therefore be prospectively planned. The fraction of S-phase cells is quantified and expressed as the percentage of DNA precursor-incorporating cells over the total number of tumor cells. The main advantages of these approaches, which are considered complex, are the high accessibility and, as in situ procedures, the possibility to discriminate tumor from nontumor cells to overcome bias related to tumor heterogeneity. Thymidine labeling index (TLI) is not affected by type or time of fixation, gives clear-cut and unequivocally positive images of reduced silver grains, and permits determinations of labeled cells even after lengthy preservation of archival paraffin blocks.
The main limitation of these approaches is the requirement of fresh tumor material with a sufficient number of viable cells, which has been partially overcome by the availability of kits for TLI (distributed by Euroframe, Asti, Italy) and for bromodeoxy uridine (distributed by Amersham) labeling index (BrdULI) determination, which guarantee the standardization of the first methodological steps and facilitate their use in institutions without adequately equipped laboratories.
The cytometric quantification of nuclear DNA content, which generally provides information on total DNA content and gross genomic abnormalities, can be used to quantify cells in the different cell cycle phases, in particular in the S phase, based on the knowledge that S-phase cells have a variable DNA content ranging from the presynthetic phase G0/1 (2n) to the postsynthetic G2 phase (4n). The utilization of dyes that specifically bind DNA, such as propidium iodide, ethidium bromide, mitramycin, 4',6-diamidino-2-phenylindole (DAPI), acridine orange, and Hoechst 33258 allows a quantitation of nuclear DNA content by flow cytometry on isolated nuclei or cell suspensions, or by image static cytometry on cytohistological specimens. Both approaches give a frequency histogram of DNA content, which reflects the cell cycle. The fractions of cells in the different phases are quantified by computerized cell cycle analysis. In addition to S-phase cells, the fraction of cells in the S+G2M phases is also considered by some authors as a more complete proliferation index that defines the proportion of cells in the cell cycle excluding only those in the G0/1 phase. The most diffuse approach for the evaluation of the S-phase cell fraction is flow cytometry (FCM-SPF), the main advantage of which consists in a rapid, potentially objective evaluation of a large number of cells obtained from surgical speci mens, biopsy or fine-needle aspirates, effusions, and bone marrow aspirates. The main drawback, which is common to all the non-in situ techniques, is the impossibility to discriminate tumor from nontumor cells. This automated technique received a major impetus in the late 1980s, with the development of procedures to perform flow cytometry in solid tumors using material from formalin-fixed paraffin-embedded blocks (14) or from frozen tumor specimens. The use of the latter material also guarantees more reproducible information on specimens that have been in storage for some time and accrued from different centers. The feasibility of FCM-SPF is potentially high, but the quality of results can be affected by methodological factors. To make results reproducible and comparable among the different centers, standardization of assay methodologies, cell cycle analysis techniques, and cutoff points for classifying and interpreting FCM-S from DNA histograms, as well as strict quality control, are mandatory.
Recently, a concerted effort was carried out and developed by US, French, and Swedish investigators to optimize the prognostic strength of flow-cytometric DNA measurements and to test the validity of the proposed adjustments (15). This study, performed on about 1400 patients with node-negative breast cancer, emphasized the complexity of the interpretation of DNA ploidy histograms and quantification of S-phase cells, which are so closely related to each other that they provide non-independent prognostic information when considered in association. Following 10 adjustments to the two measurements, which involved both DNA ploidy reclassification and S-phase calculation, the association between the two flow cytometric measurements has been reduced and their confounding technical correlation eliminated, thus permitting them to become independent prognostic factors in a single model.
Was this article helpful?