Validation of a mathematical procedure for computer analysis of flow cytometric DNA data in human tumors

Fluorochromes for DNA Staining and Quantitation

In these last few decades the great explosion of the molecular approaches has casted a little shadow on the DNA quantitative analysis. Nevertheless DNA cytochemistry represented a long piece of history in cell biology since the advent of the Feulgen reaction. This discovery was really the milestone of the emerging quantitative cytochemistry, and scientists from all over the world produced a very large literature on this subject. This first era of quantitation (histochemistry followed by cytochemistry) started by means of absorption measurements (histophotometry and cytophotometry). The successive introduction of fluorescence microscopy gave a great boost to quantitation, making easier and faster the determination of cell components by means of cytofluorometry. The development of flow cytometry further contributed to the importance of quantitative cytochemistry. At its beginning, the mission of flow cytometry was still DNA quantitation. For a decade the Feulgen reaction had been the reference methodology for both conventional and flow cytofluorometry; the advent of Shiff-type reagents contributed to expand the variety of possible fluorochromes excitable in the entire visible spectrum as well as in the ultraviolet region. The fluorescence scenario was progressively enriched by new probes among which are the intercalating dyes which made DNA quantitation simple and fast, thus spreading it worldwide. The final explosion of cytofluorometry was made possible by the availability of a large variety of probes directly binding DNA structure. In addition, immunofluorescence allowed to correlate the cell cycle-related DNA content to other cell markers. In the clinical application of flow cytometry, this promoted the introduction of multiparametric analyses aimed at describing the cytokinetic characteristics of a given cell subpopulation defined by a specific immunophenotype setting.

Interaction between hormone-dependent and hormone-independent human breast cancer cells

We developed two different models based on in vitro co-culture of hormone-dependent and hormone-independent cell lines to simulate the cell population heterogeneity of human breast cancer tumours. Oestrogen-dependent (MCF-7, ZR 75.1) and oestrogen-independent cell lines (MDAMB-231 BT-20) were grown under serum-free conditions. Co-culture of hormone-dependent and hormone-independent cell lines resulted in an increased cell yield compared to single cell cultures carried out at the same seeding ratios. Such an increase was not affected by addition of oestradiol and single growth factors (EGF, bFGF and IGF-I). These results allow us to conclude that in a heterogeneous cell population like human breast tumours, interaction between hormone-dependent and hormone-independent cell lines may result in a complex regulation of cell growth.

Improved prognostic impact of S-phase values from paraffin-embedded breast and prostate carcinomas after correcting for nuclear slicing

Nuclear debris may significantly interfere with the analysis of S-phase fraction (SPF) from paraffin-embedded tumors. We used a background subtraction algorithm to compensate for the effects of slicing of tumor cell nuclei during preparation of paraffin-embedded specimens. DNA histograms were analyzed from 88 node-negative breast and from 78 prostatic carcinomas. Median SPFs corrected for nuclear slicing were lower than uncorrected ones in both breast cancer (7.6% vs. 5.7%) and prostate cancer (6.7% vs. 4.2%). The median SPF value in each group was used as a cut-off point in survival studies. As compared with the uncorrected SPFs, corrected SPF levels resulted in a more significant survival difference between breast cancer patients with above and below median SPF (p = 0.0014 vs. p = 0.014) and in a higher relative risk (RR) of death (4.5 vs. 3.1). The same was true for prostate cancer survival (p less than 0.0001 vs. p = 0.002) and RR (5.3 vs. 3.1). Compared with the exponential background subtraction method, the sliced nuclei correction was more reproducible and could be applied in all evaluable histograms without the risk of overcompensation. In conclusion, our results support the use of background correction with the sliced nuclei model in DNA flow cytometric studies of archival tissues.

How reproducible are flow cytometry data from paraffin-embedded blocks?

The purpose of this technical report is to determine the reproducibility of flow cytometry data for ploidy and cell cycle kinetics using paraffin-embedded blocks of breast cancer tissue. One block from each of 39 tumors was studied in this report with each block having multiple sections analyzed independently. All of these sections gave ploidy analyses, while only 34 gave cell kinetic values. The standard deviation for the DNA index value in the multiple analysis study was less than 0.1 in all but three cases. The cell kinetic values gave larger variability, and the actual values were dependent on the method of analysis. Comparison of the variability for each method of analysis could not predict which procedure was superior. These results would indicate that ploidy is a reproducible value, while cell kinetic parameters should only be used as an indicator of proliferative activity that has been normalized to the mean or median of a large set of observations processed and analyzed by the same procedure.

Comparative analysis of different approaches to investigate cell kinetics

The potential of different methods to investigate proliferative activity of cell populations was analysed for non-Hodgkin's lymphomas. Cells in S phase and all cycling cells were determined on cell suspensions obtained from fresh lymph node material by [3H]-thymidine autoradiography [( 3H]TdR LI), a monoclonal antibody to bromodeoxyuridine (BrdU LI), and the monoclonal antibody Ki67. A good correlation was observed between the values of [3H]TdR LI and BrdU LI (rs = 0.90; P less than 0.01), [3H]TdR LI and S phase (rs = 0.62; P less than 0.01) and [3H]TdR LI and Ki67 (rs = 0.64; P less than 0.01) in individual lymphomas. Using the median values obtained from the different approaches as cut-off points to define slowly and rapidly proliferating tumours, the best agreement was observed between [3H]TdR LI and BrdU LI (91%) and poorer agreements, even though statistically significant, were observed between [3H]TdR LI and S phase (73%) or Ki67 (76%). In conclusion, the kinetic information derived from different approaches was more or less concordant and newly proposed approaches should be directly and carefully verified for their prognostic relevance before using them as alternatives to conventional methods.

Implications of disaggregation procedures on biological representation of human solid tumours

This study was designed to define some biological aspects of cell suspensions, obtained by mechanical or enzymatic disaggregations, and to verify whether single cell suspensions are representative of original solid tumours. The study was performed on a series of 25 human solid tumours including breast carcinoma, ovarian carcinoma and malignant melanoma. A higher cell viability and a loss of aneuploid subpopulations, or a lower fraction of aneuploid cells, were observed in enzymatically-released samples than in samples obtained by the mechanical procedure. Moreover, the proliferative activity, which was generally similar for the cell suspensions obtained by the two disaggregation procedures, was always markedly lower in the cell suspensions than in solid samples from the same tumour. In conclusion, the results from this study indicate that many changes, such as selective release of cell populations from the tumour matrix, damage and destruction of aneuploid and proliferating cells can be induced to various extents by different disaggregation procedures.