- Email: [email protected]
Cytogenetics and molecular cytogenetics in Hodgkin's disease
Annals of Oncology 7 (SuppL 4): S49-S53, 1996. O 1996 Kluwer Academic Publishers. Printed in the Netherlands.
Review Cytogenetics and molecular cytogenetics in Hodgkin's disease J. Deerberg-Wittram, K. Weber-Matthiesen & B. Schlegelberger Department of Human Genetics, University of Kiel, Kiel Germany Summary
Introduction
During the last decades tumor cytogenetics has become an important tool to specify the diagnosis and to predict the prognosis of malignant disorders. This holds particularly true for hematological neoplasias. Among these, Hodgkin's disease still is an exception, since, to date, a significant correlation between cytogenetic data, histopathological findings and clinical course has not been found. On the other hand, tumorcytogenetics has contributed greatly in assessing the dignity of Hodgkin's disease. Although the pathognomonic Hodgkin and Reed-Sternberg (HRS) cells were discovered nearly 100 years ago, it was unclear until quite recently, whether Hodgkin's disease is in fact a malignant disorder. On the one hand, the clinical course points to malignancy; on the other hand, the scarcity of the HRS-cells in the involved lymph nodes seems to misfit the current view of malignant disorders. It was the cytogenetic detection of chromosomally aberrant clones that gave striking evidence for Hodgkin's disease representing a malignant proliferation. Cytogenetics Among 250 cases of Hodgkin's disease so far studied by cytogenetic analysis, chromosomally aberrant clones could be demonstrated in about 100 cases. According to the ISCN guidelines, clonality is demonstrated if identical chromosome gains or structural aberrations are present in at least two metaphases [1]. Three aberrant metaphases are required in the case of
Key words: cytogenetics, FICTION, FISH, Hodgkin's disease
chromosome losses. For several reasons the success rate of cytogenetic analyses in Hodgkin's disease is relatively low. Firstly, the low mitotic activity of cultured HRS cells and secondly, the very low number of malignant HRS cells within the affected lymph node contribute to the rather poor success rate. Specific chromosome aberrations as known from some nonHodgkin lymphomas, e.g., t(14;18) in follicular lymphoma or t(8;14) in Burkitt's lymphoma, have not been identified in Hodgkin's disease. The failure to detect specific chromosome aberrations could be due to the extraordinary complexity of the karyotypes found in Hodgkin's disease. In fact, aberrant metaphases in Hodgkin's disease are frequently characterized by numerous structural and numerical chromosome aberrations with chromosome numbers in triploid to tetraploid or even octaploid range [2-9]. Chromosome analysis is further complicated by a remarkable variability of chromosome aberrations among the metaphases of one clone. Nevertheless, hitherto available cytogenetic data on Hodgkin's disease indicate that some chromosomal regions are recurrently affected by structural alterations [3, 7, 8]. These include several regions on chromosome 1, on the long arm of the chromosomes 3, 6, 7, 8,11 and 14 and on the short arm of the chromosomes 12, 13 and 15. Some of these breakpoints could be confirmed by us in a cytogenetic study of 21 cases of Hodgkin's disease [10]. Figure 1 schematically shows the distribution of the breakpoints in these 21 cases: Recurrent breakpoints were located in Ipl3, 7q32/34, 2pl6/21, 19pl3, 4q25/28, 6ql5/21 and 12q22/23. Moreover, the pericentric regions of the chromosomes 1, 4, 7, 9, 16, 17, 21 and 22 were recur-
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
For about 20 years we have known from cytogenetic studies that there is a clonal cell population in Hodgkin's disease. Most karyotypes are complexly aberrant and chromosome numbers typically lie in the hyperploid range. Some chromosome regions seem to be preferentially involved, but a chromosome aberration specific for Hodgkin's disease has not yet been determined. Although the existence of a clonal cell population is evident from these cytogenetic studies, there is still an ongoing debate, whether in all cases the pathogno-
monic Hodgkin and Reed-Sternberg cells belong to one single aberrant clone and thus represent a monoclonal proliferation. This article reviews the current knowledge on cytogenetics in Hodgkin's disease. Moreover, our recent data from simultaneous immunophenotyping and interphase cytogenetics (FICTION) are introduced into the passionate discussion on the monoclonality of the Hodgkin and ReedSternberg cells.
50
Figure 1. Chromosomal breakpoints in Hodgkin's disease as determined in a cytogenetic study on 21 cases.
rently affected. In some cases, breakpoints repeatedly involved in non-Hodgkin-Lymphoma could be identified, e.g., in 14qll and 14q32, the sites of the T-cell receptor (TcR) a and 6 genes and the immunoglobulin heavy chain (IgH) genes, respectively [11]. The hypothesis of a lymphocytic origin of the HRS cells may be supported by these findings. Although karyotypes in Hodgkin's disease are mostly very complex, some metaphases with discrete chromosome aberrations, for example with a single numerical or structural aberration, were reported [5, 12]. According to the ISCN rules, these aberrations are not always clonal. Sometimes, only single metaphases with discrete aberrations are detected [4, 5]. It is not likely that these findings represent artefacts caused by culturing, since they are also found in so-called direct preparations of uncultured lymph node cells. One reason for the occasional finding of single metaphases with discrete aberrations might be the well known generic instability in Hodgkin's disease, which may also be responsible for the relatively high frequency of single aberrant metaphases in some patients' peripheral blood [13]. On the other hand, one has to consider that the unusual metaphases with discrete aberrations belong to the tumor cell clone and might correspond to early-stage HRS cells at the beginning of karyotype evolution.
The most frequent finding in chromosome analyses of Hodgkin's disease are metaphases with a regular karyotype. One might appoint the normal metaphases to the lymphocytic bystander cells. However, it is conceivable that in some cases, HRS cells in fact do not contain cytogenetically detectable chromosome aberrations. Such HRS cells could possibly be typical for a very early stage of Hodgkin's disease characterized by molecular genetic, rather than cytogenetic alterations. Particularly, these putative early HRS cells could be an interesting target for molecular studies. Unfortunately, conventional cytogenetic techniques do not allow the assignment of normal and aberrant metaphases to a certain cell type. Cell membranes and cytoplasm are destroyed by the preparation techniques and morphologic and immunophenotypic features are regularly lost. The first attempt to avoid this dilemma was made by Teerenhovi and coworkers [14]. They introduced a new technique combining classical chromosome banding and immunophenotyping of metaphase cells. The crucial advance of this technique was that - in contrast to the classical cytogenetic analysis - chromosome banding was possible on morphologically intact metaphase cells. Since the procedure was technically expending and time consuming, only a small amount of data was obtained with this technique. At any rate, it could be
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
8
51 demonstrated at a few cases of Hodgkin's disease [15, 16] that the chromosomally aberrant metaphase cells expressed the CD30 antigen, typically present on HRS cells. Thus, it was shown for the first time that in Hodgkin's disease the HRS cells are the clonally proliferating and cytogenetically aberrant cells. Considering the heterogeneity of the histopathologic pattern in Hodgkin's disease these few data were not sufficient to give general evidence to the clonality of the HRS cells. Moreover, the technical expenditure of this method was too high and the success rate too low to examine a reasonable number of cases.
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
does not limit interphase cytogenetic studies by means of FICTION, provided that tumor cells and normal cells have different immunophenotypes. In most cases of Hodgkin's disease the malignant HRS cells express the CD30 antigen. Besides the HRS cells only a few activated lymphocytes and monocytes stain positive for CD30 within the lymph nodes affected by HD. Therefore, immunophenotyping for CD30 gives the opportunity to trace the rare HRS cells amidst the overwhelming lymphocytic infiltrate. Using the FICTION approach with a CD 30 antibody and centromeric DNAprobes, all CD 30 positive HRS cells on the slide can easily be detected and studied with regard to their chromosome numbers. In this way, the Question of Molecular cytogenetics clonality of the HRS cells can effectively be investigated. In a recent study we have scrutinized 30 cases of The diverse difficulties and insufficiencies of the above Hodgkin's disease of different histopathological submentioned method have now been overcome by the types by FICTION [23]. In this study we used the Ki-1 new technique of combined fluorescence immuno- antibody for the CD 30 antigen and different centrophenotyping plus fluorescence in situ hybridization. meric DNA probes specific for the chromosomes 1,8, This method, developed by us, is called 'Fluorescence 12, 15, 17, X and Y. These probes were chosen to Immunophenotyping and Interphase Cytogenetics as a detect numerical chromosome aberrations typically Tool for Investigations of Neoplasms (FICTION)' [17- found in the complexly aberrant clones in Hodgkin's 23]. The most commonly used specific DNA probes disease. Figure 2 shows a typical microscopic image for fluorescence in situ hybridization (FISH) are those obtained by the FICTION technique. We could demthat specifically hybridize with the centromeric regions onstrate that a) all CD30 positive HRS cells in each of of certain chromosomes, the so-called centromeric the 30 cases contained numerical chromosome aberraprobes. Hybridization with these probes results in tions; and b) that at least some of the detected aberramicroscopically analyzable fluorescent spots ('signal') tions were identical throughout all HRS cells of one in the nucleus; the number of the signals corresponds single case. These findings indicate that the HRS cells to the copy number of the studied chromosome. Using in Hodgkin's disease are cytogenetically clonal, i.e., all the new FICTION method, numerically aberrant tu- tumor cells originate from the same aberrant precursor mor cells can be identified by interphase cytogenetics cell. Interestingly, the hybridization signal numbers and subsequently characterized immunophenotypical- corresponded well with numerical chromosome aberly. Since all interphase cells can be analysed, one is not rations detected in the metaphases. If in a given case limited by the number and banding quality of analyz- the copy number of a certain chromsome was constant able mitoses. Chromosome analysis is thus possible on within all metaphases, in situ hybridization also relarge numbers of immunologically identified cells. vealed constant signal numbers in the CD30 positive What is the particular benefit of FICTION in HRS cells. On the other hand, if the copy numbers of Hodgkin's disease? In Hodgkin's disease the number of certain chromosomes were variable in the metaphases, the malignant HRS cells is usually very low, often less in situ hybridization also showed variable signal numthan 1% of total lymph node cells. The detection of tu- bers for the respective chromosomes. mor clones with aneuploid chromosome numbers by In one case of Hodgkin's disease with a relative high means of FISH is restricted to cases with relatively high amount of HRS cells we were able to analyze 1500 percentages of aberrant cells. At least l%-2% of the HRS cells concerning the copy number of different analyzed cells must be trisomic to define the presence chromosomes. In this case, some chromosome aberraof an aberrant clone. In the case of a monosomy, more tions were constant throughout the whole tumor cell than 10% of the cells on the slide must show the chro- population, whereas other chromosomes showed mosome loss. The reason is that normal specimens highly variable copy numbers. For example, the HRS hybridized with centromeric DNA-probes also show cells had two to eight copies of chromosome 1. Cytoconsiderable portions of cells with one or three signals. genetic analysis, too, had revealed this extreme copy This is due to unspecific or ineffective in situ hybridiza- number variability of chromosome 1. These findings tion. Consequently, in samples with low numbers of indicate that karyotype variability is a real phenomtumor cells it is impossible to differentiate true aneu- enon in Hodgkin's disease and not owing to preparaploidy from artificial hybridization results. FICTION is tion artifacts. able to overcome this problem. True aneuploidy is conRemarkably, the signal constellation of the single firmed, if numerical aberrations are detected exclusive- nuclei of the polynucleated Reed-Sternberg (RS) cells ly within cells with tumor cell-associated immuno- was the same as found in the mononucleated Hodgkin phenotype. This way, a low percentage of tumor cells cells. This finding confirms the assumption that Reed-
52
Sternberg cells derived from Hodgkin cells by endomitosis or cell fusion. The fact that we were able to detect supernumerary chromosomes in all cases using only a few probes indicates that the HRS cells were regularly hyperploid. Therefore, one must question whether some recently reported near diploid karyotypes with simple chromosome changes actually represented HRS cells. We must assume that karyotypes with simple chromosome defects simply mirror the genetic instability of lymphocytic cells, which is a well known phenomenon in Hodgkin's disease. The same holds true for the majority of cytogenetically studied cases, in which only normal metaphases are found. Thus, it appears likely that in Hodgkin's disease normal metaphases always represent proliferating lymphocytes. Incidentally, using FICTION with CD3 and CD19 immunostaining we were able to clarify that among the 30 cases studied Tand B-cells were not affected by the chromosome aberrations detected within the HRS cells. At present, the clonality of the HRS cells in Hodgkin's disease is passionately debated. Although it has been clear for at least twenty years that there must be a clonal, cytogenetically aberrant population in Hodgkin's disease, the question for clonality has come
into fashion again. This is due to contradictory data from very recent studied obtained by the new powerful technique of single cell PCR. Whatever the explanation for these inconsistent findings migh be, our FICTION results are well in line with the conclusion cytogeneticists have drawn from their observations. Our FICTION study has revealed that in all cases of Hodgkin's disease numerically aberrant HRS cells may be detected. Most important, all HRS cells belong to the same cytogenetically defined clone. References 1. Mitelman F (ed). ISCN (1991): Guidelines for Cancer Cytogenetics (1st ed.) Supplement to an International System for Human Cytogenetic Nomenclature. Switzerland: Basel, Karger 1991. 2. Rowley JD. Chromosomes in Hodgkin's disease: Cancer Treat Rep 1982; 66:639-43. 3. Fonatsch C, Diehl V, Schaadt M et al. Cytogenetic investigations in Hodgkin's disease: I. Involvement of specific chromosomes in marker formation. Cancer Genet Cytogen 1986; 20: 39-52. 4. Kristoffersson U, Heim S, Mandahl N et al. Cytogenetic studies in Hodgkin's disease. Acta Path Microbiol Immunol Scand Sect A 1987; 95: 289-95.
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
Figure 2. FICTION study: a CD30 positive cheromosomally aberrant Hodgkin cell with three green signals for chromosome 1, indicating a trisomy 1, and one blue signal for the Y chromosome. The digital image was made using the Metasystems ISIS system.
53
18.
19.
20.
21.
22.
23.
al. Simultaneous fluorescence immunophenotyping and interphase cytogenetics: A contribution to the characterization of tumor cells. J Histochem Cytochem 1992; 40:171-6. Weber-Matthiesen K, Deerberg J, Miiller-Hermelink A et al. Rapid immunophenotypic characterization of chromosomally aberrant cells by the new FICTION method. Cytogenet Cell Genet 1993; 63:123-5. Weber-Matthiesen K, Pressl S, Schlegelberger B, Grote W. Combined immunophenotyping and interphase cytogenetics on cryostat sections by the new FICTION method. Leukemia 1993; 7:646-9. Weber-Matthiesen K, Miiller-Hermelink A, Deerberg J et al. Discrimination of distinct subpopulations within a tumor with combined double immunophenotyping and interphase cytogenetics. J Histochem Cytochem 1993; 41:1641-4. Weber-Matthiesen K, Winkemann M, Grote W, Schlegelberger B. Kappa- and lambda-positive cells in centroblastic-centrocytic lymphoma (follicular lymphoma) may share a secondary chromosome aberration: Reflections on early lymphoma development Genes Chromosom Cancer 1995; 14: 22-7. Weber-Matthiesen K, Deerberg J, Poetsch M et al. Clarification of dubious karyotypes in Hodgkin's disease by simultaneous immunophenotyping and interphase cytogenetics (FICTION). Cytogenet Cell Genet 1995; 70:243-5. Weber-Matthiesen K, Deerberg J, Poetsch M et al. Numerical chromosome aberrations are present within the CD30-positive Hodgkin and Reed-Sternberg cells in 100% of analyzed cases of Hodgkin's disease. Blood 1995; 86:1464-8.
Correspondence to: Klaus Weber-Matthiesen, MD. Department of Human Genetics University of Kiel Schwanenweg 24 24105 Kiel Germany
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
5. Schouten HC, Sanger WG, Duggan M et al. Chromosomal abnormalities in Hodgkin's disease. Blood 1989; 73:2149-54. 6. Sandberg A. The Chromosomes in Human Cancer and Leukemia (2nd ed.). New York: Elsevier 1990. 7. Tilly H, Bastard C, Delastre T et al. Cytogenetic studies in untreated Hodgkin's disease. Blood 1991; 77:1298-304. 8. Dohner H, Bloomfield CD, Frizzera G et al. Recurring chromosome abnormalities in Hodgkin's disease. Genes Chromosom Cancer 1992; 5: 392-8. 9. Fonatsch C, Gradl G, Rademacher J. Genetics of Hodgkin's lymphoma. Rec Results Cancer Res 1989; 117: 36-49. 10. Schlegelberger B, Weber-Matthiesen K, Himmler A et al. Cytogenetic findings and results of combined immunophenotyping and karyotyping in Hodgkin's disease. Leukemia 1994; 8: 7280. 11. Cabanillas F, Pathak S, Trujillo J et al. Cytogenetic features of Hodgkin's disease suggest possible origin from a lymphocyte. Blood 1988; 71:1615-7. 12. Dennis TR, Stock AD, Winberg CD et al. Cytogenetic studies of Hodgkin's disease: Analysis of involved lymph nodes from 12 patients. Cancer Genetc Cytogen 1989; 37: 201-8. 13. Barrios L, Caballin MR, Miro R et al. Chromosome abnormalities in peripheral blood lymphocytes from untreated Hodgkin's patients: A possible evidence for chromosome instability. Hum Genet 1988; 78: 320-4. 14. Teerenhovi L, Kunntila S, Ekblom M et al. A method for simultaneous study of karyotype, morphology and immunologic phenotype of mitoric cells in hematologic malignancies. Blood 1984; 64:1116-22. 15. Teerenhovi L, Lindholm C, Pakkala A et al. Unique display of a pathologic karyotype in Hodgkin's disease by Reed-Sternberg cells. Cancer Genet Cytogen 1988; 34: 305-11. 16. Schlegelberger B, Weber-Matthiesen K, Eumann S et al. Detection of structural chromosome aberrations in immunophenotyped mitoses. Cytogenet Cell Genet 1992; 59:10-1. 17. Weber-Matthiesen K, Winkemann M, Miiller-Hermelink A et
Review Cytogenetics and molecular cytogenetics in Hodgkin's disease J. Deerberg-Wittram, K. Weber-Matthiesen & B. Schlegelberger Department of Human Genetics, University of Kiel, Kiel Germany Summary
Introduction
During the last decades tumor cytogenetics has become an important tool to specify the diagnosis and to predict the prognosis of malignant disorders. This holds particularly true for hematological neoplasias. Among these, Hodgkin's disease still is an exception, since, to date, a significant correlation between cytogenetic data, histopathological findings and clinical course has not been found. On the other hand, tumorcytogenetics has contributed greatly in assessing the dignity of Hodgkin's disease. Although the pathognomonic Hodgkin and Reed-Sternberg (HRS) cells were discovered nearly 100 years ago, it was unclear until quite recently, whether Hodgkin's disease is in fact a malignant disorder. On the one hand, the clinical course points to malignancy; on the other hand, the scarcity of the HRS-cells in the involved lymph nodes seems to misfit the current view of malignant disorders. It was the cytogenetic detection of chromosomally aberrant clones that gave striking evidence for Hodgkin's disease representing a malignant proliferation. Cytogenetics Among 250 cases of Hodgkin's disease so far studied by cytogenetic analysis, chromosomally aberrant clones could be demonstrated in about 100 cases. According to the ISCN guidelines, clonality is demonstrated if identical chromosome gains or structural aberrations are present in at least two metaphases [1]. Three aberrant metaphases are required in the case of
Key words: cytogenetics, FICTION, FISH, Hodgkin's disease
chromosome losses. For several reasons the success rate of cytogenetic analyses in Hodgkin's disease is relatively low. Firstly, the low mitotic activity of cultured HRS cells and secondly, the very low number of malignant HRS cells within the affected lymph node contribute to the rather poor success rate. Specific chromosome aberrations as known from some nonHodgkin lymphomas, e.g., t(14;18) in follicular lymphoma or t(8;14) in Burkitt's lymphoma, have not been identified in Hodgkin's disease. The failure to detect specific chromosome aberrations could be due to the extraordinary complexity of the karyotypes found in Hodgkin's disease. In fact, aberrant metaphases in Hodgkin's disease are frequently characterized by numerous structural and numerical chromosome aberrations with chromosome numbers in triploid to tetraploid or even octaploid range [2-9]. Chromosome analysis is further complicated by a remarkable variability of chromosome aberrations among the metaphases of one clone. Nevertheless, hitherto available cytogenetic data on Hodgkin's disease indicate that some chromosomal regions are recurrently affected by structural alterations [3, 7, 8]. These include several regions on chromosome 1, on the long arm of the chromosomes 3, 6, 7, 8,11 and 14 and on the short arm of the chromosomes 12, 13 and 15. Some of these breakpoints could be confirmed by us in a cytogenetic study of 21 cases of Hodgkin's disease [10]. Figure 1 schematically shows the distribution of the breakpoints in these 21 cases: Recurrent breakpoints were located in Ipl3, 7q32/34, 2pl6/21, 19pl3, 4q25/28, 6ql5/21 and 12q22/23. Moreover, the pericentric regions of the chromosomes 1, 4, 7, 9, 16, 17, 21 and 22 were recur-
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
For about 20 years we have known from cytogenetic studies that there is a clonal cell population in Hodgkin's disease. Most karyotypes are complexly aberrant and chromosome numbers typically lie in the hyperploid range. Some chromosome regions seem to be preferentially involved, but a chromosome aberration specific for Hodgkin's disease has not yet been determined. Although the existence of a clonal cell population is evident from these cytogenetic studies, there is still an ongoing debate, whether in all cases the pathogno-
monic Hodgkin and Reed-Sternberg cells belong to one single aberrant clone and thus represent a monoclonal proliferation. This article reviews the current knowledge on cytogenetics in Hodgkin's disease. Moreover, our recent data from simultaneous immunophenotyping and interphase cytogenetics (FICTION) are introduced into the passionate discussion on the monoclonality of the Hodgkin and ReedSternberg cells.
50
Figure 1. Chromosomal breakpoints in Hodgkin's disease as determined in a cytogenetic study on 21 cases.
rently affected. In some cases, breakpoints repeatedly involved in non-Hodgkin-Lymphoma could be identified, e.g., in 14qll and 14q32, the sites of the T-cell receptor (TcR) a and 6 genes and the immunoglobulin heavy chain (IgH) genes, respectively [11]. The hypothesis of a lymphocytic origin of the HRS cells may be supported by these findings. Although karyotypes in Hodgkin's disease are mostly very complex, some metaphases with discrete chromosome aberrations, for example with a single numerical or structural aberration, were reported [5, 12]. According to the ISCN rules, these aberrations are not always clonal. Sometimes, only single metaphases with discrete aberrations are detected [4, 5]. It is not likely that these findings represent artefacts caused by culturing, since they are also found in so-called direct preparations of uncultured lymph node cells. One reason for the occasional finding of single metaphases with discrete aberrations might be the well known generic instability in Hodgkin's disease, which may also be responsible for the relatively high frequency of single aberrant metaphases in some patients' peripheral blood [13]. On the other hand, one has to consider that the unusual metaphases with discrete aberrations belong to the tumor cell clone and might correspond to early-stage HRS cells at the beginning of karyotype evolution.
The most frequent finding in chromosome analyses of Hodgkin's disease are metaphases with a regular karyotype. One might appoint the normal metaphases to the lymphocytic bystander cells. However, it is conceivable that in some cases, HRS cells in fact do not contain cytogenetically detectable chromosome aberrations. Such HRS cells could possibly be typical for a very early stage of Hodgkin's disease characterized by molecular genetic, rather than cytogenetic alterations. Particularly, these putative early HRS cells could be an interesting target for molecular studies. Unfortunately, conventional cytogenetic techniques do not allow the assignment of normal and aberrant metaphases to a certain cell type. Cell membranes and cytoplasm are destroyed by the preparation techniques and morphologic and immunophenotypic features are regularly lost. The first attempt to avoid this dilemma was made by Teerenhovi and coworkers [14]. They introduced a new technique combining classical chromosome banding and immunophenotyping of metaphase cells. The crucial advance of this technique was that - in contrast to the classical cytogenetic analysis - chromosome banding was possible on morphologically intact metaphase cells. Since the procedure was technically expending and time consuming, only a small amount of data was obtained with this technique. At any rate, it could be
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
8
51 demonstrated at a few cases of Hodgkin's disease [15, 16] that the chromosomally aberrant metaphase cells expressed the CD30 antigen, typically present on HRS cells. Thus, it was shown for the first time that in Hodgkin's disease the HRS cells are the clonally proliferating and cytogenetically aberrant cells. Considering the heterogeneity of the histopathologic pattern in Hodgkin's disease these few data were not sufficient to give general evidence to the clonality of the HRS cells. Moreover, the technical expenditure of this method was too high and the success rate too low to examine a reasonable number of cases.
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
does not limit interphase cytogenetic studies by means of FICTION, provided that tumor cells and normal cells have different immunophenotypes. In most cases of Hodgkin's disease the malignant HRS cells express the CD30 antigen. Besides the HRS cells only a few activated lymphocytes and monocytes stain positive for CD30 within the lymph nodes affected by HD. Therefore, immunophenotyping for CD30 gives the opportunity to trace the rare HRS cells amidst the overwhelming lymphocytic infiltrate. Using the FICTION approach with a CD 30 antibody and centromeric DNAprobes, all CD 30 positive HRS cells on the slide can easily be detected and studied with regard to their chromosome numbers. In this way, the Question of Molecular cytogenetics clonality of the HRS cells can effectively be investigated. In a recent study we have scrutinized 30 cases of The diverse difficulties and insufficiencies of the above Hodgkin's disease of different histopathological submentioned method have now been overcome by the types by FICTION [23]. In this study we used the Ki-1 new technique of combined fluorescence immuno- antibody for the CD 30 antigen and different centrophenotyping plus fluorescence in situ hybridization. meric DNA probes specific for the chromosomes 1,8, This method, developed by us, is called 'Fluorescence 12, 15, 17, X and Y. These probes were chosen to Immunophenotyping and Interphase Cytogenetics as a detect numerical chromosome aberrations typically Tool for Investigations of Neoplasms (FICTION)' [17- found in the complexly aberrant clones in Hodgkin's 23]. The most commonly used specific DNA probes disease. Figure 2 shows a typical microscopic image for fluorescence in situ hybridization (FISH) are those obtained by the FICTION technique. We could demthat specifically hybridize with the centromeric regions onstrate that a) all CD30 positive HRS cells in each of of certain chromosomes, the so-called centromeric the 30 cases contained numerical chromosome aberraprobes. Hybridization with these probes results in tions; and b) that at least some of the detected aberramicroscopically analyzable fluorescent spots ('signal') tions were identical throughout all HRS cells of one in the nucleus; the number of the signals corresponds single case. These findings indicate that the HRS cells to the copy number of the studied chromosome. Using in Hodgkin's disease are cytogenetically clonal, i.e., all the new FICTION method, numerically aberrant tu- tumor cells originate from the same aberrant precursor mor cells can be identified by interphase cytogenetics cell. Interestingly, the hybridization signal numbers and subsequently characterized immunophenotypical- corresponded well with numerical chromosome aberly. Since all interphase cells can be analysed, one is not rations detected in the metaphases. If in a given case limited by the number and banding quality of analyz- the copy number of a certain chromsome was constant able mitoses. Chromosome analysis is thus possible on within all metaphases, in situ hybridization also relarge numbers of immunologically identified cells. vealed constant signal numbers in the CD30 positive What is the particular benefit of FICTION in HRS cells. On the other hand, if the copy numbers of Hodgkin's disease? In Hodgkin's disease the number of certain chromosomes were variable in the metaphases, the malignant HRS cells is usually very low, often less in situ hybridization also showed variable signal numthan 1% of total lymph node cells. The detection of tu- bers for the respective chromosomes. mor clones with aneuploid chromosome numbers by In one case of Hodgkin's disease with a relative high means of FISH is restricted to cases with relatively high amount of HRS cells we were able to analyze 1500 percentages of aberrant cells. At least l%-2% of the HRS cells concerning the copy number of different analyzed cells must be trisomic to define the presence chromosomes. In this case, some chromosome aberraof an aberrant clone. In the case of a monosomy, more tions were constant throughout the whole tumor cell than 10% of the cells on the slide must show the chro- population, whereas other chromosomes showed mosome loss. The reason is that normal specimens highly variable copy numbers. For example, the HRS hybridized with centromeric DNA-probes also show cells had two to eight copies of chromosome 1. Cytoconsiderable portions of cells with one or three signals. genetic analysis, too, had revealed this extreme copy This is due to unspecific or ineffective in situ hybridiza- number variability of chromosome 1. These findings tion. Consequently, in samples with low numbers of indicate that karyotype variability is a real phenomtumor cells it is impossible to differentiate true aneu- enon in Hodgkin's disease and not owing to preparaploidy from artificial hybridization results. FICTION is tion artifacts. able to overcome this problem. True aneuploidy is conRemarkably, the signal constellation of the single firmed, if numerical aberrations are detected exclusive- nuclei of the polynucleated Reed-Sternberg (RS) cells ly within cells with tumor cell-associated immuno- was the same as found in the mononucleated Hodgkin phenotype. This way, a low percentage of tumor cells cells. This finding confirms the assumption that Reed-
52
Sternberg cells derived from Hodgkin cells by endomitosis or cell fusion. The fact that we were able to detect supernumerary chromosomes in all cases using only a few probes indicates that the HRS cells were regularly hyperploid. Therefore, one must question whether some recently reported near diploid karyotypes with simple chromosome changes actually represented HRS cells. We must assume that karyotypes with simple chromosome defects simply mirror the genetic instability of lymphocytic cells, which is a well known phenomenon in Hodgkin's disease. The same holds true for the majority of cytogenetically studied cases, in which only normal metaphases are found. Thus, it appears likely that in Hodgkin's disease normal metaphases always represent proliferating lymphocytes. Incidentally, using FICTION with CD3 and CD19 immunostaining we were able to clarify that among the 30 cases studied Tand B-cells were not affected by the chromosome aberrations detected within the HRS cells. At present, the clonality of the HRS cells in Hodgkin's disease is passionately debated. Although it has been clear for at least twenty years that there must be a clonal, cytogenetically aberrant population in Hodgkin's disease, the question for clonality has come
into fashion again. This is due to contradictory data from very recent studied obtained by the new powerful technique of single cell PCR. Whatever the explanation for these inconsistent findings migh be, our FICTION results are well in line with the conclusion cytogeneticists have drawn from their observations. Our FICTION study has revealed that in all cases of Hodgkin's disease numerically aberrant HRS cells may be detected. Most important, all HRS cells belong to the same cytogenetically defined clone. References 1. Mitelman F (ed). ISCN (1991): Guidelines for Cancer Cytogenetics (1st ed.) Supplement to an International System for Human Cytogenetic Nomenclature. Switzerland: Basel, Karger 1991. 2. Rowley JD. Chromosomes in Hodgkin's disease: Cancer Treat Rep 1982; 66:639-43. 3. Fonatsch C, Diehl V, Schaadt M et al. Cytogenetic investigations in Hodgkin's disease: I. Involvement of specific chromosomes in marker formation. Cancer Genet Cytogen 1986; 20: 39-52. 4. Kristoffersson U, Heim S, Mandahl N et al. Cytogenetic studies in Hodgkin's disease. Acta Path Microbiol Immunol Scand Sect A 1987; 95: 289-95.
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
Figure 2. FICTION study: a CD30 positive cheromosomally aberrant Hodgkin cell with three green signals for chromosome 1, indicating a trisomy 1, and one blue signal for the Y chromosome. The digital image was made using the Metasystems ISIS system.
53
18.
19.
20.
21.
22.
23.
al. Simultaneous fluorescence immunophenotyping and interphase cytogenetics: A contribution to the characterization of tumor cells. J Histochem Cytochem 1992; 40:171-6. Weber-Matthiesen K, Deerberg J, Miiller-Hermelink A et al. Rapid immunophenotypic characterization of chromosomally aberrant cells by the new FICTION method. Cytogenet Cell Genet 1993; 63:123-5. Weber-Matthiesen K, Pressl S, Schlegelberger B, Grote W. Combined immunophenotyping and interphase cytogenetics on cryostat sections by the new FICTION method. Leukemia 1993; 7:646-9. Weber-Matthiesen K, Miiller-Hermelink A, Deerberg J et al. Discrimination of distinct subpopulations within a tumor with combined double immunophenotyping and interphase cytogenetics. J Histochem Cytochem 1993; 41:1641-4. Weber-Matthiesen K, Winkemann M, Grote W, Schlegelberger B. Kappa- and lambda-positive cells in centroblastic-centrocytic lymphoma (follicular lymphoma) may share a secondary chromosome aberration: Reflections on early lymphoma development Genes Chromosom Cancer 1995; 14: 22-7. Weber-Matthiesen K, Deerberg J, Poetsch M et al. Clarification of dubious karyotypes in Hodgkin's disease by simultaneous immunophenotyping and interphase cytogenetics (FICTION). Cytogenet Cell Genet 1995; 70:243-5. Weber-Matthiesen K, Deerberg J, Poetsch M et al. Numerical chromosome aberrations are present within the CD30-positive Hodgkin and Reed-Sternberg cells in 100% of analyzed cases of Hodgkin's disease. Blood 1995; 86:1464-8.
Correspondence to: Klaus Weber-Matthiesen, MD. Department of Human Genetics University of Kiel Schwanenweg 24 24105 Kiel Germany
Downloaded from http://annonc.oxfordjournals.org/ at Mahidol University, Faculty of Medicine, Siriraj Hospital on March 18, 2015
5. Schouten HC, Sanger WG, Duggan M et al. Chromosomal abnormalities in Hodgkin's disease. Blood 1989; 73:2149-54. 6. Sandberg A. The Chromosomes in Human Cancer and Leukemia (2nd ed.). New York: Elsevier 1990. 7. Tilly H, Bastard C, Delastre T et al. Cytogenetic studies in untreated Hodgkin's disease. Blood 1991; 77:1298-304. 8. Dohner H, Bloomfield CD, Frizzera G et al. Recurring chromosome abnormalities in Hodgkin's disease. Genes Chromosom Cancer 1992; 5: 392-8. 9. Fonatsch C, Gradl G, Rademacher J. Genetics of Hodgkin's lymphoma. Rec Results Cancer Res 1989; 117: 36-49. 10. Schlegelberger B, Weber-Matthiesen K, Himmler A et al. Cytogenetic findings and results of combined immunophenotyping and karyotyping in Hodgkin's disease. Leukemia 1994; 8: 7280. 11. Cabanillas F, Pathak S, Trujillo J et al. Cytogenetic features of Hodgkin's disease suggest possible origin from a lymphocyte. Blood 1988; 71:1615-7. 12. Dennis TR, Stock AD, Winberg CD et al. Cytogenetic studies of Hodgkin's disease: Analysis of involved lymph nodes from 12 patients. Cancer Genetc Cytogen 1989; 37: 201-8. 13. Barrios L, Caballin MR, Miro R et al. Chromosome abnormalities in peripheral blood lymphocytes from untreated Hodgkin's patients: A possible evidence for chromosome instability. Hum Genet 1988; 78: 320-4. 14. Teerenhovi L, Kunntila S, Ekblom M et al. A method for simultaneous study of karyotype, morphology and immunologic phenotype of mitoric cells in hematologic malignancies. Blood 1984; 64:1116-22. 15. Teerenhovi L, Lindholm C, Pakkala A et al. Unique display of a pathologic karyotype in Hodgkin's disease by Reed-Sternberg cells. Cancer Genet Cytogen 1988; 34: 305-11. 16. Schlegelberger B, Weber-Matthiesen K, Eumann S et al. Detection of structural chromosome aberrations in immunophenotyped mitoses. Cytogenet Cell Genet 1992; 59:10-1. 17. Weber-Matthiesen K, Winkemann M, Miiller-Hermelink A et