Model for the genetic evolution of human solid tumors.

A conceptual model is proposed for the genetic evolution of many human solid tumors that is based on the observations that cancer cells may spontaneously double their chromosome number; that cells with excessive chromosome numbers may be cytogenetically unstable, both losing chromosomes randomly during subsequent cell divisions, and often developing structural abnormalities in the chromosomes that are retained; and that some structural chromosome abnormalities may activate growth-promoting genes. The sequence of tetraploidization with chromosome loss can occur repeatedly in a given tumor. The available evidence supporting the model is reviewed. A computer simulation system that embodies these concepts is described and the model is used to generate distributions of chromosome number/cell under various simulated conditions and in a variety of simulated biological settings. A simulation of the time course of changes in chromosome number per cell that accompany the spontaneous neoplastic transformation of mouse fibroblasts in vitro is described. The best fit to the data was obtained when provision was made for the activation of at least two growth-promoting genes. The conditions for generating discrete aneuploid peaks in cytogenetic and flow cytometric studies were explored; our modeling studies suggest that the activation of a growth promoting gene is required in order to produce a discrete aneuploid peak. Our modeling studies suggest that the overrepresentation of individual oncogene-bearing chromosomes in aneuploid cell lines may require the activation of gene dose-dependent growth-promoting genes and is not likely to occur in cell lines in which at least two copies of each normal chromosome are required for cell survival. Overall, the results obtained using the model are consistent with a wide variety of flow cytometric and cytogenetic studies in human solid tumors.

Karyotypic evolution of a human undifferentiated large cell carcinoma of the lung in tissue culture.

Serial cytogenetic studies were performed on a cell line derived from a pleural effusion from a patient with undifferentiated large cell carcinoma of the lung. The initial sample had a broad range of chromosome numbers per cell, with a hypodiploid/pseudodiploid stem line and a hypotetraploid sideline. A sequence consisting of a doubling of chromosome number per cell followed by chromosome loss was observed repeatedly during 40 culture passages. The presence of metaphase spreads showing evidence of endoreduplication suggested this as a likely mechanism for the doubling of chromosome number per cell. Eleven marker chromosomes were observed in the cells of the primary sample; these markers persisted through all subsequent passages. Chromosomes 1, 2, 6, 7, 8, 11, and 16 were consistently overrepresented; each of these chromosomes was involved in marker formation. Chromosomes 4, 5, 9, 10, 19, 21, and 22 were consistently underrepresented. Every chromosome, either in its normal form and/or as part of a marker, was represented on the average by at least one copy per diploid cell. Eighteen new marker chromosomes were observed during the course of cell cultivation; one of these evolved into a clonal marker over the course of six cell passages. Of the new marker chromosomes that were formed during the observation period, the majority were found in hypotetraploid cells.

Correlations among p53, Her-2/neu, and ras overexpression and aneuploidy by multiparameter flow cytometry in human breast cancer: evidence for a common phenotypic evolutionary pattern in infiltrating ductal carcinomas.

Human solid tumors develop multiple genetic abnormalities that accumulate progressively in individual cells during the course of tumor evolution. We sought to determine whether there are specific sequences of occurrence of these progressive evolutionary changes in human breast cancers by performing correlated cell-by-cell measurements of cell DNA content, p53 protein, Her-2/neu protein, and ras protein by multiparameter flow cytometry in 56 primary tumor samples obtained at surgery. In addition, p53 allelic loss and Her-2/neu gene amplification were determined by fluorescence in situ hybridization in cells from the same samples. We reasoned that if there is a specific order in which genetic changes occur, the same early changes would be found consistently in the cells with the fewest abnormalities. We reasoned further that late-developing abnormalities would not occur alone in individual cells but would almost always be found together with the early changes inherited by the same cells. By these criteria, abnormalities involving p53 generally occurred early in the course of development of invasive breast cancers, whereas ras protein overexpression was found to be a late-occurring phenomenon. Within individual tumors, cellular p53 overexpression was often observed alone in individual cells, whereas ras protein overexpression was rarely observed in the absence of p53 overexpression and/or Her-2/neu overexpression in the same cells. Furthermore, the intracellular level of each abnormally expressed protein was found to increase progressively as new abnormalities were acquired. Infiltrating ductal carcinomas exhibited characteristic phenotypic patterns in which p53 allelic loss and/or p53 protein overexpression, Her-2/neu amplification and/or overexpression, aneuploidy, and ras overexpression accumulated within individual cells. However, this pattern was not a prominent feature of lobular breast cancers. All six lobular breast cancers studied were diploid. p53 allelic loss and/or early p53 overexpression, and late ras cooverexpression in the same cells were less common in lobular breast cancers than in infiltrating ductal carcinomas. Although Her-21neu overexpression was a common finding in lobular breast cancers, Her-2/neu amplification was not observed in these tumors.

Intracellular coexpression of epidermal growth factor receptor, Her-2/neu, and p21ras in human breast cancers: evidence for the existence of distinctive patterns of genetic evolution that are common to tumors from different patients.

Multiparameter flow cytometry studies were performed on cells from the primary tumors of 94 patients with breast cancer. Correlated cellular measurements of cell DNA content, Her-2/neu, epidermal growth factor receptor (EGFR), and p21ras levels were performed on each of 5,000 to 100,000 cells from each tumor. When criteria for positivity were matched with those in common use for immunohistochemical studies, 28 of 94 (30%) breast cancers were classified as positive for Her-2/neu overexpression. When similar criteria were applied to the EGFR measurements, 23 of 94 (24%) cases were classified as positive for EGFR overexpression. Similarly, 23 of 94 (24%) cases were classified as positive for p21ras overexpression. By conventional flow cytometric criteria for DNA ploidy, 24 cases were diploid, 28 were tetraploid, and 42 were aneuploid. When the measurements were treated as separate sets of data, the only statistically significant correlations noted were the high frequency of diploid tumors, which did not overexpress any of the three oncogenes studied (P < 0.05), and an association between Her-2/neu overexpression and aneuploidy (P < 0.03). When the data were treated as correlated intracellular measurements, 90 of the 94 tumors studied contained a population of cells in which the intracellular levels of Her-2/neu expression were directly correlated with the levels of EGFR expression in the same cells. The ratio of Her-2/neu molecules to EGFR molecules in the same cells exceeded 1 in the majority of tetraploid and aneuploid cases and was close to or less than 1 in the majority of diploid cases. In nearly all tumors, p21ras overexpression was observed only in cells that overexpressed Her-2/neu, EGFR, or both, and p21ras levels per cell were more closely correlated with levels of EGFR per cell in the same cells than with Her-2/neu levels per cell. The data are consistent with a model in which heterodimerization of Her-2/neu and EGFR in individual cells is achieved by one of several genetic evolutionary pathways, all of which commonly lead to p21ras overexpression. The two major genetic evolutionary pathways identified in this study are an aneuploid, Her-2/neu overexpression-driven pathway seen in 59 of 94 tumors, and a diploid, EGFR overexpression-driven pathway seen in 19 of 94 tumors. All tumors with Her-2/neu:EGFR ratios greater than 2 contained an infiltrating ductal carcinoma component, whereas all infiltrating pure lobular carcinomas had Her-2/ neu:EGFR ratios that were less than 2. All of the genetic evolutionary pathways identified in this study were represented among the 11 tumors from patients who experienced early tumor recurrences.

HER2 expression as a potential marker for response to therapy targeted to the EGFR.

Since human epidermal growth factor receptor 2 (HER2) is known to participate with the epidermal growth factor receptor (EGFR) in mitogenic signalling, we hypothesised that HER2 overexpression might indicate responsiveness to EGFR targeted therapies. MCF7 breast cancer cells transfected with the HER2 gene were subcloned to establish a set of genetically related cell lines expressing graded levels of HER2 by immunoblot analysis. The subcloned cell lines and parental MCF7 cells were characterised by their growth characteristics, and cell by cell patterns of EGFR, HER2 and HER3 expression as well as levels of phosphorylated mitogen-activated protein kinase (MAPK) and AKT by laser scanning cytometry (LSC). Growth inhibition assays were used to characterise response to EGFR targeted therapy, and to determine the relationship between therapeutic response and levels of tyrosine kinase expression. The levels of growth inhibition of AG1478 and of the AG1478-trastuzumab combinations were correlated with levels of HER2 expression among the different cell lines. Among EGFR, HER2 and HER3, HER2 overexpression was the best single predictive marker, but combinations of two markers provided additional predictive information.

Multiparameter analysis of human epithelial tumor cell lines by laser scanning cytometry.

Laser scanning cytometry (LSC) is a relatively new slide-based technology developed for commercial use by CompuCyte (Cambridge, MA) for performing multiple fluorescence measurements on individual cells. Because techniques developed for performing four or more measurements on individual lymphoid cells based on light scatter as a triggering parameter for cell identification are not suitable for the identification of fixed epithelial tumor cells, an alternative approach is required for the analysis of such cells by LSC. Methods for sample preparation, event triggering, and the performance of multiple LSC measurements on disaggregated fixed human cells were developed using normal lymphocytes and two human breast cancer cell lines, JC-1939 and MCF-7, as test populations. Optimal conditions for individual cell identification by LSC were found to depend on several factors, including deposited cell density (cells per unit area), the dynamic range of probe fluorescence intensities, and intracellular distribution of the fluorescent probe. Sparsely deposited cells exhibited the least cell overlap and the brightest immunofluorescent staining. Major advantages of using DNA probes over a cytoplasmic immunofluorescent protein marker such as tubulin for event triggering are that the former exhibit greater fluorescence intensity within a relatively sharply demarcated nuclear region. The DNA-binding dye LDS-751 was found to be suboptimal for quantitative DNA measurements but useful as a triggering measurement that permits the performance of simultaneous fluorescein isothiocyanate-, phycoerythrin-, and indodicarbocyanine-based measurements on each cell. A major potential advantage of LSC over flow cytometry is the high yields of analyzable cells by LSC, permitting the performance of multiple panels of multicolor measurements on each tumor. In conclusion, we have developed and optimized a technique for performing multiple fluorescence measurements on fixed epithelial cells by LSC based on event triggering using the DNA-binding dye LDS 751. Although not ideal for quantitative measurements of cell DNA content, the large Stokes shift of this dye permits the performance of three or more additional fluorescence measurements on each cell.

Discrepancies between flow cytometric and cytogenetic studies in the detection of aneuploidy in human solid tumors.

Parallel flow cytometric (FCM) cell DNA studies and cytogenetic studies were performed on clinical samples from twenty human solid tumors of various types and on cell lines established in tissue culture from three of these tumors. Six of twenty clinical samples (30%) showed concordance between flow cytometry and cytogenetics with respect to the presence or absence of aneuploidy. Among the fourteen cases with discrepancies between the two methods, 8 (40% of all cases) showed hypodiploidy by cytogenetics and had diploid DNA histograms. Three cases (15%) had prominent discrete peaks in the triploid to tetraploid region by cytogenetics but had only barely discernible corresponding peaks in the DNA histogram. In two cases (10%) cytogenetic studies revealed diffuse aneuploidy. Cytogenetic studies demonstrated near-tetraploidy in three samples, but only one of these was detected by FCM; all three cases exhibited other numerical chromosomal abnormalities. In one case aneuploidy was demonstrated by FCM and not by cytogenetics. Among the tumor cell lines established in culture, the DNA Index was often higher than the cytogenetic index. Overall, 13/20 or 65% of patients with solid tumors in this study had numerical chromosomal abnormalities that were not detected by flow cytometry. Eleven of these patients had distant metastases at the time of tumor sampling, and nine of these died of their disease within 1-11 months of the time of study.

Preferred genetic evolutionary sequences in human breast cancer: a case study.

Multiparameter flow cytometry studies were performed on the cells of an aggressive human breast cancer at the time of diagnosis and at relapse. The aneuploid cells that overexpressed large amounts of both HER-2/neu and ras survived intensive chemotherapy and were responsible for tumor relapse. At relapse, these cells were shown to overexpress simultaneously at least five oncogenes: HER-2/neu, ras, EGF receptor, p53 and c-myc. A partial reconstruction of the genetic evolutionary sequence in this tumor indicated that HER-2/neu overexpression was an early step in the sequence. Subsequent HER-2/neu overexpression, EGF receptor overexpression and p53 protein overexpression were each associated with ras overexpression. The data suggest that ploidy and oncogene overexpression cannot be used as independent clinical prognostic factors. The ability to characterize tumors according to the degree of advancement in the genetic evolutionary might serve as a basis for genetic staging for adjuvant therapy.

Aneuploidy in breast cancer: a fluorescence in situ hybridization study.

Although ploidy is associated with the development and progression of most breast cancers, the value of flow cytometric ploidy as a clinical prognostic factor remains controversial. The technique of fluorescence in situ hybridization (FISH) can be used not only to determine overall ploidy, but also to assess the over-representation or under-representation of specific chromosomes in interphase cells. This information may be of prognostic value. We studied 84 primary breast cancers and 20 metastatic tumors by FISH, using chromosome-specific fluorescent centromeric probes. Of these, 100 cases were also studied by DNA flow cytometry. The FISH studies were concordant with DNA flow cytometry with regard to distinguishing aneuploid from diploid tumors in 78% of cases. The FISH data suggested that aneuploidy arises by a process of chromosome complement doubling with subsequent chromosome loss. In tumors that exhibited evidence of more than one round of chromosome complement doubling, the selective accumulation of multiple copies of specific chromosomes or chromosome segments was common. Multiple copies of chromosomes centromeres 1, 3, and 17 were accumulated selectively in the cells of individual tumors more frequently than chromosomes centromeres 7, 11, and 16. Multiple copies of chromosomes centromeres 10 and 20 were selectively accumulated only rarely, if at all. Aneuploidy in breast cancer can be divided into distinct stages using fluorescence in situ hybridization techniques. The stages of aneuploidy provide potential landmarks in the genetic evolution of this disease with possible links to chromosome-specific evolutionary changes.

Assessing sequential oncogene amplification in human breast cancer.

Studies of amplification and/or overexpression of c-myc, HER-2/neu, and H-ras in breast cancer have shown that each is associated with a poor prognosis. The purpose of this study was to explore the possibility that there is a preferred sequence of amplification of these oncogenes in breast cancer. The frequencies of amplification and patterns of co-amplification of c-myc, HER-2/neu, and H-ras were studied in a group of 84 breast cancers. The data suggested a preferred sequence of amplification that consisted of c-myc amplification-HER-2/neu amplification-H-ras amplification. This model was supported by loglinear analysis. In addition, the levels of amplification of JC-A, a DNA fragment newly isolated from a patient with advanced breast cancer, were studied in 61 of these cases. The data suggested that JC-A amplification occurred early. Loglinear analysis supported a model in which JC-A amplification occurred either before or after c-myc amplification but was unrelated to Her-2/neu or ras amplification.

Sequential paraformaldehyde and methanol fixation for simultaneous flow cytometric analysis of DNA, cell surface proteins, and intracellular proteins.

A cell fixation and permeabilization procedure consisting of sequential paraformaldehyde and methanol was evaluated and found suitable for concomitant flow cytometric quantification of total cellular DNA, immunofluorescence measurements of cell surface proteins, and immunofluorescence measurements of intracellular proteins. Paraformaldehyde/methanol-fixed cells exhibited significantly greater intracellular antitubulin immunofluorescence than cells fixed with paraformaldehyde or methanol alone (p less than 0.002) and significantly greater intracellular antitubulin immunofluorescence than cells fixed with methanol followed by paraformaldehyde (p less than 0.006). With paraformaldehyde/methanol fixation, cell morphology was well preserved and forward and right angle light scatter properties were sufficiently well maintained to permit gating on these parameters. Cell surface marker staining with fluorescent anti-leukocyte antibodies was unaffected by fixation with paraformaldehyde/methanol. Paraformaldehyde effects on the intensity of DNA staining with propidium iodide were dependent on paraformaldehyde concentration and fixation temperature; these effects were least pronounced at low paraformaldehyde concentrations (0.25% or less), and at temperatures lower than 37 degrees C. Paraformaldehyde fixation may result in differences in propidium iodide staining of DNA in some diploid cells, which may produce small spurious aneuploid peaks in normal peripheral blood leukocytes. Paraformaldehyde fixation also produces an apparent increase in the DNA index of aneuploid cell populations in comparison with methanol fixation, particularly when the DNA index exceeds 1.5. Occasionally, this paraformaldehyde fixation-induced effect is useful in identifying biologically distinct near-diploid subpopulations in tumors.

The accumulation of multiple genetic abnormalities in individual tumor cells in human breast cancers: clinical prognostic implications.

PURPOSE: Human solid tumors undergo multiple genetic evolutionary changes as they evolve from the normal state to advanced stages of malignancy. This study characterizes the degree of advancement of primary human breast cancers in their genetic evolutionary pathways, and determines if this is of clinical significance. MATERIALS AND METHODS: Correlated cell-by-cell measurements of cell DNA content, HER-2/neu protein content per cell, and H-ras protein content per cell were obtained by means of multiparameter flow cytometry on primary tumors from 95 patients with clinically localized breast cancer. Laboratory findings were correlated with subsequent clinical course in 91 of these patients. RESULTS: Multiple genetic abnormalities were found to accumulate in individual cells in primary human breast cancers. Almost all tumors contained subsets of cells with one, two, or three abnormalities per cell in various combinations. After a median follow-up time of 32 months, 11 of 13 patients with early recurrence had primary tumors in which more than 5% of cells were hypertetraploid, overexpressed HER-2/neu protein, and overexpressed H-ras protein (triple-positive cells). The duration of disease-free survival among patients with primary tumors that contained triple-positive cells was significantly shorter than for patients whose tumors did not contain triple-positive cells. The presence of subpopulations of cells with maximums of only one abnormality per cell or only two abnormalities per cell, in any combination, was of no prognostic significance. Among patients whose nodal status was known, 12 had recurrent disease, and all had positive axillary nodes. Among 36 patients known to have negative axillary nodes, no recurrence has been reported to date. CONCLUSIONS: The number of genetic abnormalities that accumulate in individual cells in primary breast cancers reflects the degree of advancement of a tumor in its genetic evolutionary sequence, and provides useful clinical prognostic information. Because follow-up duration is still relatively short, and because disease in node-negative patients tends to recur later than in node-positive patients, it is still too early to know if three measurements per cell will be sufficient to improve prognosis in node-negative disease.

Distinctive patterns of Her-2/neu, c-myc, and cyclin D1 gene amplification by fluorescence in situ hybridization in primary human breast cancers.

BACKGROUND: Human solid tumors undergo clonal evolution as they progress, but evidence for specific sequences of genetic changes that occur in individual tumors and are recapitulated in other tumors is difficult to obtain. METHODS: Patterns of amplification of Her-2/neu, c-myc, and cyclin D1 were determined by fluorescence in situ hybridization (FISH) in relation to the presence of p53 dysfunction and ploidy in 60 primary human breast cancers. RESULTS: We show that there are clusters of genophenotypic abnormalities that distinguish lobular breast cancers from nonlobular tumors; that cyclin D1 amplification occurs prior to the divergence of lobular breast cancers from nonlobular cancers; that p53 dysfunction, Her-2/neu amplification, and c-myc amplification are characteristic features of nonlobular breast cancers, but not of lobular breast cancers; and that the frequencies of amplification of all three oncogenes examined increase progressively with increasing aneuploidy, but that each gene exhibits a different profile of increasing amplification in relation to tumor progression. Early amplification of c-myc appears to be an especially prominent feature of hypertetraploid/hypertetrasomic tumors. CONCLUSIONS: The data suggest that in tumors containing multiple abnormalities, these abnormalities often accumulate in the same cells within each tumor. Furthermore, the same patterns of accumulation of multiple genophenotypic abnormalities are recapitulated in different tumors.