Document 2qRqZmKJorYBbYj04EpdV10Q5

VOL 19, NO 1 FEBRUARY 1992 EDITOR-IN-CHIEF John W. Yarbro, MD, PhD ASSOCIATE EDITORS Richard S. Bornstein, MD Michael J. Mastrangelo, MD Myelodysplastic Syndromes Donald C . Doll, MD, anh Alan F. List, MD, Guest Editors VOL 19, NO 1 Seminars in Oncology Myelodysplastic Syndromes FEBRUARY 1992 Table of Contents Myelodysplastic Syndromes: Introduction.....................................Donald C. Doll and Alan F. List Classification and Morphologic Features of the MyelodysplasticSyndromes....Jean E. Goasguen and John M. Bennett 1 4 Biology and Pathogenesis of the Myelodysplastic Syndromes..........Alan F. List and Allan Jacobs 14 Chromosome Abnormalities in Myelodysplastic Syndromes................................... Peter C. Nowell 25 In Vitro Marrow Culture Studies in the MyelodysplasticSyndromes................Peter L. Greenberg 34 Leukemias and Myelodysplastic Syndromes Secondary to Drug, Radiation, and Environmental Exposure..........................................................................Ellis G.Levine and Clara D. Bloomfield 47 Chemotherapy and Bone Marrow Transplantation for MyelodysplasticSyndromes Bruce D. Cheson 85 Differentiation-Inducing Agents in the Treatment of Myelodysplastic Syndromes Masahiro Kizaki and H. Phillip Koefler 95 Treatment of Myelodysplastic SyndromesWith Hemopoietic Growth Factors Peter L. Greenbe% 106 Chromosome Abnormalities in Myelodysplastic Syndromes Peter C. Nowell DURING THE PAST decade, our understanding of the group of diseases defined as "preleukemi~m~~yeloid disorders has increased significantly. Clinically, the myeloproliferative entities have been separated from the myelodysplastic syndromes (MDS), and the latter have been subdivided into a workable FAB classification.'.' Biologically, it is clear that the MDS subtypes (refractory anemia [RA], l2A with ringed sideroblasts [RARS], RA with excess of blasts [RAEB], RAEB in transformation [EUEB-t], chronic myelomonocytic leukemia [CMML]) represent a continuum, with overt leukemia as the ultimate result in a significant proportion of cases. At every stage, these appear to be clonal disorders arising from a marrow stem cell, with the degree of clinical severity reflecting, at least to some degree, the rate at which the clone is expanding. Although MDS-like disorders do occur in children, this is typically a disease of older individuals, making therapy particularly difficult, and perhaps helping to explain, in an aging population, the apparent increase in frequency in recent years. Chromosome data, along with biochemical studies, have helped to establish the clonal nature of these disorders, as well as their relationship to the frank leukemias that they p r e ~ e d e . ~C'y~togenetic information has also proved of diagnostic and prognostic value, generally keeping pace with the more detailed clinical chara~terization.~T~h'~is~ brief review will attempt to summarize our present knowledge concerning specific chromosome alterations in MDS, with, wherever possible, related information on genes that may be involved in the pathogenesis of these disorders. It will also indicate current thinking concerning the prognostic significance of various types of karyotypic abnormalities, in both primary and therapyrelated MDS. More detailed review of the chromosomal data can be found in other recent ~ources~t.h'~e;limited goal of this summary is to provide an indication, including some personal experience, of how cytogenetic studies are now contributing to both clinical and basic efforts to better understand this difficult group of diseases and some prospects for the future. ChromosomeAbnormalities in Primary Myelodysplastic Syndromes In most series reported to date, a chromosomally abnormal clone has been demonstrable in the bone marrow of 40% to 60% of patients with MDS unrelated to previous therapy?." These have ranged from highly aneuploid karyotypes, with multiple structural alterations, to clones characterized by a single chromosome abnormality. In general, clones with complex karyotypes having multiple changes have been more frequent in the more advanced FAB subtypes of MDS (RAEB, RAEB-T, CMML), and have been associated with a poor prognosis, as compared to patients with either a single abnormality in the clone or no demonstrable karyotypic ~hange.2.~-T" his has been true whether the particular outcome being determined was progression to frank leukemia or overall length of survival. For example, in our own series,".'2 which now includes more than 250 primary MDS patients, median survival with a complex karyotype was only 3 months following initial study, compared to approximately 12months for patients with single abnormalities or a normal karyotype. Also, a higher percentage of patients with complex karyotypes have progressed to acute leukemia (62%) versus the other two groups (45%), but the difference is not as striking, perhaps reflecting the very short survival of many patients with multiple cytogenetic abnormalities. In most series, MDS patients with a normal karyotype have had a slightly better outcome than patients with any chromosomal abnormality, single or m ~ l t i p l e . ~I~t m~a~y' 'be, in fact, that some individuals with refractory anemia (RA, RARS), a normal karyotype, and a very pro- From the Department of Pathology and Laboratory Medicine, School of Medicine, University of Pennsylvania,Philadelphia. Supported in part by Outstanding Investigator Grant CA42232from the National Institutes of Health. Address reprint requests to Peter C.Nowell, MD, Department of Pathology and Laboratory Medicine, School of Medicine, Universityof Pennsylvania, Philadelphia, PA 19104-6082. Copyright 0 1992 by W E . Saunders Company 0093- 775419211901-0OO3$05.OOiO Seminarsin Oncology, Vol 19. No 1 (February), 1992: pp 25-33 25 26 PETER C. NOWELL longed indolent course are not appropriately classified as MDS and grouped with other RA patients in which there is definite clinical and cytogenetic evidence for a dysplastic myeloid clone expanding in the bone marrow.' -e, [.& It is true, however, that some MDS patients \- J?with a normal karyotype have acquired a chro- *r"\A$' osome abnormality during the course of their disease, and additional cytogenetic changes have a!.; L L wlso been documented in serial studies of pa- L , y'& tients who had a karyotypically aberrant clone -r from the In both circumstances, this <\,$-= clonal evolution has typically been associated with clinical progression and early death. Inter- `.k estingly, such karyotypic evolution has generally been less frequent than in some other myeloid disorders, most notably chronic myelogenous le~kemia.'~T.'h~ere is also no clear evidence, in MDS, that the presence of some normal cells in association with a chromosomallv abnormal clone in the marrow is of prognostic cance."" Recently, we and others have begun to inves- tigate in more detail, both clinically and biologi- cally, the circumstances in which a karyotypi- cally abnormal clone in the bone marrow of an MDS patient has a single alteration, either gain, loss, or translocation. This represents approxi- mately 20% to 30% of all patients with primary MDS, o r half of those with an abnormal karyo- itype. Five single abnormalities appear most common: loss of all or part of a chromosome 7, an extra copy of chromosome 8, an isochromosome for the long arm of chromosome 17, and deletion of the long arm of chromosome 5 or chromosome 20. Other less common isolated abnormalities include deletion of the short arm of chromosome 12 (12p-), deletions and translocations involving band llq23, and loss of the Y chromosome.2~"'T2here have also been scattered reports of other small deletions (eg, 2p-, 8p-, 9p-), as well as gain of chromosome 14 and loss of 22, all occurring as single alter- ations.10.16-19 Conversely, a number of the specific chromosome rearrangements commonly associated with various types of myeloid leukemia, such as t(9;22), t(8;21), t(15;17), t(9;11), and inv(l6), are almost never observed in MDS, supporting the view that these forms of acute leukemia are not preceded by a preleukemic phase. Summarized below are some recent findings and suggestions concerning five of the isolated alterations commonly seen in MDS, with respect to prognostic significance (Table 1) and also possible gene involvement that might ultimately have diagnostic or therapeutic implications. Monosomy 7/de1(7q) Several studies now indicate that a marrow clone with this single abnormality in primary MDS indicates a poor prognosis, nearly as ominous as clones with multiple karyotypic alteration^.^^"^^^'^^^^ For example, in our small series of nine patients with monosomy 7 or del(7q) as the only abnormality, only one sur- vived for more than 15 months (median = 8 months), and six of nine progressed to acute leukemia, all within 1year. r Partial or complete monosomy for chromo- some 7 also occurs commonly in complex karyo- types in both MDS and de novo acute myeloid leukemia,'0*"particularly when the disease is therapy-related (see below). In some cases, this takes the form of an unbalanced t( 1;7) translo- cation,20which also results in the loss of 7q. The relevant portion of chromosome 7 has been narrowed to band 7q22, but none of the genes currently mapped to this region have been shown to be definitely involved." It is interesting that monosomy 7 appears to be the most common abnormality in pediatric MDS, and in some cases has been associated with familial patterns of disease.u-" In several instances, these have been families with the increased chromosomal fragility and other char- acteristics of Fanconi's anemia or its but in other familial cases there was no evidence of a recognized fragility It has been suggested that the latter could reflect either an inherited tumor suppressor gene defect or per- haps a "subclinical" chromosomal fragility syn- drome,2225but there is no firm evidence to support either hypothesis at this time. Further Table 1. Prognostic Significance of Single Chromosome Abnormalities in Primary MDS Favorable del (5q) del (20q) Unfavorable -7/del (7ql +8 is0 (17q) CHROMOSOME ABNORMALITIES IN MYELODYSPLASIA 27 \tud! of these pediatric cases may shed light on thc chromosome 7 defccts in the adult disease. Trisomy N Limited data on this single abnormality in primary MDS suggest that it has only a slightly better prognosis than the monosomy 7/de1(7q) group.'.'' In our series of 14 patients, for exam- ple, the median survival was 10 months after chromosome study, and 6 cases progressed to frank leukemia within a year. -Trisomy 8 is the most common single chromo-- some abnormality seen in de novo acute my- eloid leukemia (AML), and also occurs nonran- domly as an additional chromosome change in the blast crisis of Ph-positive chronic myeloge- nous leukemia (CML).'".'' Although there has been some suggestion, in the latter circum- stance, that trisomy 8 may be associated with altered expression of the C-MYC gene, this has not been confirmed, and there are no data directly relevant to altered expression of this or any other gene in the marrow cells of MDS. -a~ - patients with trisomy 8. \*\L1' P. ' .4 .."t/\ '.,,%,.-- ::.,' . ,.?!. * !so ( I 7q) This karyotypic abnormality, in which a normal chromosome 17is replaced by an isochromosome for the long arm (Fig l), may well carry the same poor prognostic significance in MDS patients as monosomy 7/de1(7q) and trisomy 8. However, even fewer data are currently available.12.26 We have followed five patients with this abnormality as the only karyotypic change, and the outcome appears similar to that of the two previous subgroups. The median survival after chromosome study has been 10 months, with 3 cases progressing to acute leukemia. The iso(17q) alteration occurs quite commonly in a variety of hematopoietic and nonhematopoietic neoplasms, usually in association with additional karyotypic alterations."'." Other rearrangements involving the short arm of chromosome 17 (17p) have also been described in MDS.1fl.15.17.27 In several tumors, including colon cancer and the blast crisis of CML, it has recently been demonstrated that the iso(17q) abnormality is consistently associated with loss 1 23 45 78 9 10 11 12 13 U IS \17 18 nW n 22 XY Fig 1. Karyotype of marrow cells from a patient with primary MDS. An iso(l7q) is the only abnormality (arrow). In other tumors, this cytogenetic change has been associated with loss of function of the p53 gene. 28 PETER C. NOWELL of function of the p53 gene, which maps to band 1 7 ~ 1 3t,hrough deletion of the short arm in the isochromosome and submicroscopic alteration of the gene on the other 17 chromosome.2X'" There is increasing evidence that p53 functions as a tumor suppressor gene in many cell lineages, contributing to abnormal cellular proliferation when, through one mechanism or another, normal function of the gene is abrogated within the cell. It is still unclear in the iso(17q) anomaly whether other genes on the short arm of 17, or on the long arm, which is present in extra dosage, are also involved." Considerable research on defects in the p53 gene in a variety of human tumors is currently under way, including the possibility of replacement of normal gene function as a therapeutic approach. Del(.%) This is the only chromosome abnormality in MDS that is correlated with a specific clinical entity. Patients with the "5q - syndrome" have only this karyotypic change in their marrow cells and typically present with refractory macrocytic anemia, thrombocytosis, and nonlobulated megakaryocyte~.''~M'~any of these patients follow an indolent course for a number of years with little or no evidence of clinical progression. Thus, this karyotypic abnormality, occurring alone and associated with the typical refractory anemia syndrome, appears to be one of the most prognostically favorable forms of MDS.h.9.11).1Z.31.32 We have observed, however, 11 patients with only a 5q- abnormality who presented initially with a more advanced stage of MDS (RARS, RAEB, or RAEB-T). Six of these patients died within 6 months of study, 5 having progressed to acute leukemia. It is unclear whether these cases represent end stages of a previously undiagnosed long-term "5q - syndrome" or whether concomitant submicroscopic genetic alterations in the hemic clone, such as mutation of a RAS resulted in more aggressive disease from the outset. Although not all studies have shown as striking a difference in the clinical course of 5q- patients relative to the FAB subtype,(>.`).Itl.31.3: I`t is clear that some individuals with this isolated cytogenetic abnormality have a much less favorable prognosis than those with the clinically typical "5q- syndrome." Loss of all of chromosome 5 or a portion of the long arm is, like monosomy 7/de1(7q), also a common finding in complex karyotypes in MDS and in AML arising de novo, particularly therapy-related cases."." There has been considerable effort in recent years to identify the relevant gene(s) involved on 5q, particularly at band 5q31." " Many colony stimulating factors and interleukin genes have been mapped to 5q, as well as the genes of other growth factors and their receptors, but the function of none of these has been shown to be consistently altered in the resultant neoplasms. Detailed molecular analysis of the 5q- anomaly has demonstrated submicroscopic inversions and duplications adjacent to the deleted region of the chromo- This has led to speculation that the many myeloid-related genes mapped to this area may not themselves be oncogenic, but may, in some fashion, confer unusual instability on cells differentiating along a myeloid lineage and thus increase the probability of a specific tumorigenic alteration in this portion of the g e n ~ m e . ~ ' The rare reports of a clinically typical "5qsyndrome" preceding a detectable chromosome anomaly' are consistent with this concept. One current candidate for the critical leukemogenic locus is the early growth response gene (EGRl), which is also within this region and codes for a nuclear binding protein.35It is currently being investigated as a possible tumor suppressor gene whose loss of function could be important in myeloid tumorigenesis. Del (20q) This abnormality, which occurs nonrandomly in myeloproliferative preleukemic disorders (eg, polycythemia vera, myelofibrosis) as well as in dysplastic syndromes, appears to be associated with a relatively favorable prognosis when present as an isolated alteration in MDS.6.'2 Again, the published data are very limited, but in our series of 8 patients, for example, the median survival was more than 24 months after study. Three MDS patients with a 20q- abnormality progressed to leukemia within a year, but the remaining five lived for 2 to 4 years, dying of other causes. The 20q- alteration is also observed not only CHROMOSOME ABNORMALITIES IN MYELODYSPLASIA 29 in myeloproliferative disorders but in de novo acute myeloid leukemia, with or without additional karyotypic change^.'^^^^^'^ As with the other abnormalities discussed earlier, attempts have been made to identify the relevant genes involved in this deletion, without definitive results to date. It has recently been suggested that the HCK proto-oncogene, which maps to this region and codes for a tyrosine kinase involved in myeloid differentiation, may be important, because lack of HCK transcripts has been correlated with a 20q- deletion in several cases of AML (C. Willman, personal communication). As mentioned earlier, a number of other karyotypic changes have also been reported to occur as solitary alterations in primary MDS. These include deletions of 12p, translocations and deletions involvingchromosomeband 1lq23, loss of the Y chromosome, and occasionally other alteration^.^^"'^^^-^^ In none of these in- stances are there currently sufficient data to warrant any statement concerning prognostic significance, and although some potentially relevant genes have been mapped to the involved chromosomal regions, there is no definite evidence to associate any of them with the pathogenesis of the neoplasm. Clearly, these various nonrandom single abnormalities warrant further investigation,both to determine their value in patient management and to attempt to identify critical genes involved in the pathogenesis of myeloid and other disorders. ChromosomeAbnormalities in Therapy-Related Myelodysplastic Syndromes As discussed elsewhere in this volume, ther- apy-related myelodysplasticsyndromes (t-MDS) have become increasingly recognized as unwel- come sequelae of successful treatment of a primary malignancy with chemotherapy and/or radiation in both adults and children. Typically, t-MDS is much momggressive clinically than primary MDS, and these biological character@- tics are reflected in the karyotypes.At least 80% .*g of individuals with t-MDS have a chromosorqally abnormal clone in the bone marrow, and in the vast majority of cases these contain multiple ~normalities.2.'0.1s~19A3ll8-o4f1the nonrandom alterations discussed earlier are observed, in various combinations, but most characteristic are monosomy 7/de1(7q) and monosomy 5/de1(5q) (Fig 2). The 12p- abnormality noted above is also common, as are the t( 1;7) translo- cation and rearrangements involving chromo- some band llq23. Not infrequently, one also observes in the karyotypes of therapy-related disease additional indications of extensive I fl 3genomic damage and instability, such as ring .,' &.chromosomes, dicentrics, and acentric frag- ; c 3ments. On occasion, there may also be cytoge- 3 netic evidence of more than one clone arising : from separate individual cells in the badly -( 4 -I kvdamaged microenvironment of the bone mar- -, f r0w.I3Any combination of these various types of 4 .- karyotypic alterations strongly suggests previ- ous exposure of the patient to large doses of clastogenic agents, and also indicates, as noted, 3a very poor prognosis ._ with respect to prompt progression to frank leukemia and generally short sUrviva~.L10.1S.19.3~41 b + These observations in patients with known exposure to high doses of genotoxic agents have prompted considerable speculation concerning the possibility that smaller exposures to similar agents, occupational or environmental, might be of etiologic importance in myeloid leukemia and preleukemia unrelated to previous therapy. In studies of AML, it has been possible to elicit a histoy of such &posures, in the workplace, hanodmteh,iosraeplpseewarhsearels,ointomsoca&morerelraotuepws.iothf pcaotmiepnltesx, karyotypes containing of chromosomes 5 and clear with resDect to included a group of age- and sex-matched controls without hematological disease, 46% of primary MDS patients gave a historv of occupational or environmental exuosure to potentially genotoxic agents, but the emosure rate in the control po&lation was similarly high (40%). There was also no correlation, within t h i MDS group, between a history of such exposure and the specific cytogenetic findings.&This question of the role of nontherapeutic genotoxins in the etiology of "primary" MDS needs further investigation, with more detailed epidemiological and molecular studies to complement the chromosomal findings. DISCUSSION AND CONCLUSIONS At present, cytogenetic studies are of value in the continuing efforts to understand and to ` c- !w -J&q k Agy3 j I t 30 PETER C.NOWELL 1 23 \ c T 0 9 lo 5 11 12 1) 20 n 22 xx Fig 2. Representative karyotype from a patient with therapy-related MDS (t-MDS), illustrating abnormalitiesof chromosomes 5 and 7 (arrows), as well as other changes. Patients with such complex karyotypes have a poor proqnosis, in both primary MDS and t-MDS. manage the heterogeneous group of difficult disorders that we now classify as MDS. The clonal nature of these diseases is well established, and the continuum through the various stages of MDS to frank leukemia is widely accepted. A major issue for the near future is the identification of the specific somatic genetic changes that underlie the expansion and evolution of the preneoplastic hemic clone. As in all neoplasia, cytogenetic studies are providing valuable clues to the location of relevant growth regulatory genes, and, in some cases, to the mechanisms by which their function is altered. We now recognize that chromosome abnormalities can indicate important changes in gene dosage, either additions or deletions, as well as changes in location and/or structure that can alter gene f ~ n c t i o n . ~T'o date, as indicated above, there is very little firm evidence concerning the scecific genes involved in MDS, but the nonrandom karyotypic changes are being actively studied at the molecular level. and some recent successes with the human leukemias and lymphomas suggest that important results will also be forthcoming with respect to MDS. The hope is, of course, that such information will not only elucidate the basic mechanisms underlying these disorders but also provide additional use- ful tools for diagnosis, for monitoring disease in treated patients, and, ultimately, for new thera- peutic agents. It must also be recognized, as already sug- gested, that not all of the genetic alterations "-1. ... relevant to these diseases will be visible at the karyotypic level. There is already. e.v. id. ence, for example, that mutations in genes of the RAS family, not associated with demonstrable chro- mosome changes, are relatively common in MDS, and may: in fact. have negative prognostic implications.33.31Complete elucidation of the sequence of molecular changes underlying the development of these disorders will have to include such submicroscopic abnormalities. as well as those readily visiblc under the micro- scope. In addition to these explorations of somatic genetic changes in MDS cells. the possibilit! o! CHROMOSOME ABNORMALITIES IN MYELODYSPLASIA 31 inherited gene defects playing a role should also interest to determine how frequently one can continue to be considered. As noted earlier, demonstrate some "subclinical" constitutional there is limited evidence suggesting that in some defect in DNA synthesis, DNA repair, or the instances of pediatric MDS, often characterized mitotic apparatus itself in individuals who de- by monosomy 7, there may be a familial factor velop MDS.2S,52x53 involved. In some cases this has been associated In addition to contributing to these efforts to with a recognized "chromosomal fragility" syn- understand better the fundamental nature of drome, particularly Fanconi's anemia (FA) or myelodysplastic disorders, chromosome studies one of its variant^.^'^^ In a few instances, pa- also continue to have a useful role in current tients with other "fragility" syndromes (eg, attempts to manage these difficult disorders Bloom's syndrome, ataxia telangiectasia) have clinically. As noted, the karyotype has been had a dysplastic preleukemic phase before pro- shown to be an independent prognostic indica- gressing to acute myeloid leukemia, but lympho- tor, second only to the FAB subtype as a cytic neoplasms, with characteristic chromo- predictor of progression to leukemia and of some changes, are much more common in these overall survival. As the cytogenetic data are patients, perhaps reflecting the more severe extended and refined, every physician attempt- immune dysfunction in these disorders as com- ing to evaluate and utilize current therapies for pared to FA.10s15T34h9ere are also examples of MDS should have the benefit of such informa- familial pediatric MDS, including monosomy 7, tion, as well as other parameters, in making that do not fit any of the recognized instability decisions concerning patient management. syndromes and where increased chromosomal MDS appears to be increasing in our aging fragility has not been demonstrable in standard population, and response to treatment is gener- assays.u,24.50T51hese cases could involve either a ally unsatisfactory. This has stimulated a variety classical tumor suppressor gene or the inheri- of therapeutic approaches ranging from very tance of a gene that confers some degree of aggressive regimens, including bone marrow "subclinical" genomic instability, making ran- transplantation, to less cytotoxic treatments us- dom genetic and karyotypic errors more likely." ing biological agents that influence the growth The two possibilities are not mutually exclusive, and differentiation of hemic cells. The latter and could be different in different families. At may require many months before any effect is present, there is no hard evidence to support At present, although not necessar- either hypothesis. ily predicting response, chromosome studies It is also possible that in adult MDS, either can help to indicate which patients with MDS primary or therapy-induced, some minor inher- are likely to survive long enough to be ade- ited defect in DNA "housekeeping" could pre- quately evaluated with the longer-term thera- dispose certain individuals to developing the pies, and which patients, if to receive more than disease. There is beginning to be evidence that supportive care, must be treated rapidly and some members of the population, for example, aggressively. During therapy, karyotypic data are particularly susceptible to chromosomedam- can also be utilized to monitor the size of the age by alkylating agents such as ble~mycin,~' neoplastic clone in the bone marrow, as one and these individuals may well be at greater risk indicator of response.57Eventually, molecular of developing hematopoietic and other neo- approaches may provide better tools, but for the plasms when exposed to certain genotoxic near term it appears that cytogenetic studies of agents, either therapeutically or under other these disorders will continue to provide useful circumstances. This is a field that is currently information of practical clinicalvalue. under active investigation, and it will be of REFERENCES 1. Bennett JM, Catovsky D, Daniel M-T, et al: Proposals for the classification of the myelodysplastic syndromes. Br J Haematol51:189-199,1982 2. Third Cooperative Study Group: tions for a morphologic, immunologic, and cytogenetic (MIC) working classification of the primary and therapy- related myelodysplastic disorders. Cancer Genet Cytogenet 32:l-10, 1988 3. Nowell PC: Preleukemia: Cytogenetic clues in some confusing disorders, Am J patho] 89:459-476, 1977 4. Raskind WH, Tirumali N, Jacobson RM. et al: Evi- - . - .. .-. . . - .-. . . .. -...... .. - . . V 32 PETER C. NOWELL dence for a multistep pathogenesis of a myelodysplastic syndrome. Blood 63:1318-1323,1984 5. Second International Workshop on Chromosomes in Leukemia 1979: Chromosomes in preleukemia. Cancer Genet Cytogenet 2:108-113,1980 6. Yunis JJ, Lobell M, Arnesen MA, et al: Refined chromosome study helps define prognostic subgroups in most patients with primary myelodysplastic syndrome and acute myelogenous leukaemia. Br J Haematol 68:189-194, 1988 7. Pierre RV, Catovsky D, Mufti GJ, et al: Clinicalcytogenetic correlations in myelodysplasia (preleukemia). Cancer Genet Cytogenet 40:149-161,1989 8. Suciu S, Kuse R, Weh HJ, et al: Results of chromosome studies and their relation to morphology, course, and prognosis in 120 patients with de novo myelodysplastic syndrome. Cancer Genet Cytogenet 44:15-26,1990 9. Heim S, Mitelman F: Chromosome abnormalities in the myelodysplastic syndromes. Clinics in Haematology 15:1003-1021,1986 10. Sandberg AA: The Chromosomes in Human Cancer and Leukemia (ed 2). New York, NY, Elsevier, 1990 11. Nowell PC, Besa EC, Stelmach T, et al: Chromosome studies in preleukemic states. V. Prognostic significance of single versus multiple abnormalities. Cancer 58:2571-2575, 1986 12. Nowell PC. Besa EC: Prognostic significance of single chromosome abnormalities in preleukemic states. Cancer Genet Cytogenet 42:l-7,1989 13. Benitez J, Carbonell F, Sanchez Fayos J, et al: Karyotypic evolution in patients with myelodysplastic syndromes. Cancer Genet Cytogenet 16:157-167,1985 14. Horiike S, Taniwaki M, Misawa S, et al: Chromosome abnormalities and karyotypic evolution in 83 patients with myelodysplastic syndrome and predictive value for prognosis. Cancer 621129-1138,1988 15. Heim S, Mitelman F: Cancer Cytogenetics. New York, NY,Alan R. Liss, 1987 16. Jotterand-Ballamo M, Parlier V, Schmidt PM, et al: Cytogenetic analysis of 54 cases of myelodysplastic syndrome. Cancer Genet Cytogenet 46:157-172, 1990 17. Kerndrup G, Pedersen B, Bendix-Hansen K Specific minor chromosome deletions in myelodysplastic syndromes: Clinical and morphologic correlations. Cancer Genet Cytogenet 26:277-234,1987 18. Haas OA, Jager U, Ambros P, et al: Trisomy 14 in refractory anemia with excess of blasts in transformation. Cancer Genet Cytogenet 29:315-318,1987 19. Musilova J, Michalova K Chromosome study of 85 patients with myelodysplastic syndrome. Cancer Genet Cytogenet 33:39-50,1988 20. Alitalo T, Willard HF, de la Chapelle A: Determination of the breakpoints of 1;7 translocation in myelodysplastic syndrome by in situ hybridization using chromosomespecific alpha satellite DNA from human chromosomes 1 and 7. Cytogenet Cell Genet 50:49-53,1989 21. Kere J, Ruutu T, Davies KA, et al: Chromosome 7 long arm deletion in myeloid disorders: A narrow breakpoint region in 7q22 defined by molecular mapping. Blood 73:230-234,1989 22. Nowell P, Wilmoth D, Lange B: Cytogenetics of childhood preleukemia. Cancer Genet Cytogenet 10261265,1983 23. Stivrins IJ, Davis RBM, Sanger W, et al: Transformation of Fanconi's anemia to acute nonlymphoblastic leukemia associated with emergence of monosomy 7. Blood 64~173-176,1984 24. Paul B, Reid MM, Davison EV, et al: Familial myelodysplasia: Progressive disease associated with emergence of monosomy 7. Br J Haematol65:321-323,1987 25. Hsu TC: Genetic instability in the human population: A working hypothesis. Hereditas 98:l-9,1983 26. Becher R, Carbonell F, Bartram CR: Isochromosome 17q in Ph'-negative leukeia: A clinical, cytogenetic, and molecular study. Blood 75:1679-1685,1990 27. Stelmach T, Nowell P C Two cases of myelodysplastic syndrome with both t(5;20) and 17p+ translocations. Cancer Genet Cytogenet 24:371-373,1987 28. Baker SJ, Fearon ER, Nigro JM, et al: Chromosomal 17 deletions and p53 gene mutations in colorectal carcinomas. Science 244:217-220,1989 29. Kelman Z, Prokocimer M, Peller S, et al: Rearrangements in the p53 gene in Philadelphia chromosome positive chronic myelogenous leukemia. Blood 742318-2323, 1989 30. Marshall CJ:Tumor suppressor genes. Cell 64313- 326,1991 31. van den Berghe H, Vermaelen K, Mecucci C, et al: The 5q- anomaly. Cancer Genet Cytogenet 17:189-199, 1985 32. Nimer SD, Golde D W The 5q- abnormality. Blood 701705-1712,1987 33. Liu E, Hyelle B, Morgan R, et al: Mutations of the Krsten-ras protooncogene in human preleukemia. Nature 330~186-188,1987 34. Bar-Eli M, Cadavid NG, Foti A, et al: Analysis of N-ras exon-1 mutations in myelodysplastic syndromes by polymerase chain reaction and direct sequencing. Blood 73~281-2831, 989 35. LeBeau MM, Lemons RS, Espinosa R, et al: Interleukin-4 and interleukin-5 map to human chromosome 5 in a region encoding growth factors and receptors and are deleted in myeloid leukemias with a del(5q). Blood 73:647650,1989 36. Nagarajan L, Lange B, Cannizzaro L, et al: Molecular anatomy of a 5q interstitial deletion. Blood 75:82-87, 1990 37. Huebner K, Nowell PC, Croce CM: Lineage-specific gene rearrangement/deletion: A nonconservative model. Cancer Res 49:4071-4074,1989 38. Bloomfield CD: Chromosome abnormalities in secondary myelodysplastic syndromes. Scand J Haematol36:8290,1986 39. LeBeau MM. Albain KS, Larson RA, et al: Clinical and cytogenetic correlations in 63 patients with therapyrelated myelodysplastic syndromes and acute non-lymphocyticleukemia: Further evidence for characteristic abnormalities of chromosomes no. 5 and 7. J Clin Oncol 4:325-345, 1986 40. Iurlo A, Mecucci C, van Orshoven A, et at: Cytogenetic and clinical investigations in 76 cases with therapyrelated leukemia and myelodysplastic syndrome. Cancer Genet Cytogenet 43:227-241,1989 CHROMOSOME ABNORMALITIES IN MYELODYSPLASIA 41. Pedersen-Bjergaard J, Philip P, Larsen SO, et al: Chromosome aberrations and prognostic factors in therapyrelated myelodysplasia and acute nonlymphocytic leukemia. Blood 76:1083-1091, 1990 42. Lawler SD, Summersgill BM, Clink HM, et al: Chromosomes, leukemia and occupational exposure to leukemogenic agents. Lancet 1:853-854, 1979 43. Mitelman F, Nilsson PG, Brandt L, et al: Chromosome pattern, occupation and clinical features in patients with acute nonlymphocytic leukemia. Cancer Genet Cytogenet 4:197-214,1981 44. Fourth International Workshop on Chromosomes in Leukemia: The correlation of karyotype and occupational exposure to potential mutagenic carcinogenic agents in acute nonlymphocytic leukemia. Cancer Genet Cytogenet 11~326-331,1984 45. Narod SA, Dub6 ID: Occupational history and involvement of chromosomes 5 and 7 in acute nonlymphocytic leukemia. Cancer Genet Cytogenet 38:261-269,1989 46. Goldberg H, Lusk E, Moore J, et al: Survey of exposure to genotoxic agents in primary myelodysplastic syndrome: Correlation with chromosome patterns and data on patients without hematological disease. Cancer Res 50~6876-6881,1990 47. Nowell PC, Croce CM: Chromosomal approaches to oncogenes and oncogenesis. FASEB J 2:3054-3060,1988 48. Nowell P, Bergman G, Besa E, et al: Progressive preleukemia with a chromosomally abnormal clone in a kindred with the Estren-Dameshek variant of Fanconi's anemia. Blood 64:1135-1138,1984 49. German J (ed): Chromosome Mutation and Neoplasia. New York, NY,Alan R. Liss, 1983 33 50. Nowell P: Cytogenetics of preleukemia. Cancer Genet Cytogenet 5:265-278,1982 51. Weiss K, Stars S, Williams D, et al: Childhood monosomy 7 syndrome: Clinical and in vitro studies. Leukemia 1:97-104.1987 52. Hsu TC, Johnston DA, Cherry LM, et al: Sensitivity to genotoxic effects of bleomycin in humans: Possible relationship to environmental carcinogenesis. Intern J Cancer 43:403-409,1989 53. Liang JC, Pinkel DP, Bailey NM, et al: Mutagen sensitivity and cancer susceptibility. Cancer 64:1474-1479, 1989 54. Clark R, Ismail S, Jacob A, et al: A randomized trial of 13-cis retinoic acid with or without cytosine arabinoside in patients with the myelodysplastic syndrome. Br J Haemato1 66:77-83, 1987 55. Vadhan-Raj S, Keating M, LeMaistre A, et al: Effects of recombinant granulocyte-macrophage colony-stimulating factor in patients with myelodysplastic syndromes. N Engl J Med 317:1545-1552,1987 56. Besa EC, Abrahm JL, Bartholomew MJ, et ai: Treatment with 134s-retinoic acid in transfusion-dependent patients with myelodysplastic syndrome and decreased toxicity with addition of alpha-tocopherol. Am J Med 89:739747,1990 57. Abrahm J, Hyzinski M, Finan J, et ai: Disappearance of cytogenetic abnormalities and clinical remission during therapy with 13-cisretinoic acid in a patient with myelodysplastic syndrome: Inhibition of growth of the patient's malignant monocytoid clone. Blood 67:1323-1327,1986