Cytogenetic study of Brazilian patients with Myelodysplastic Syndrome ( MDS )

Bone marrow cytogenetic studies were performed on 93 patients with primary myelodysplastic syndrome (MDS) diagnosed at the Clinical Hospital of the Federal University of Paraná, Brazil. Chromosomal alterations were observed in 69% of the patients. Monosomy of chromosome 7, deletions of 7q, 5q, 12p and 20q, rearrangements of 11q23 and trisomies of chromosomes 8 and 21 were the most frequent abnormalities observed. Among adult patients the most frequent aberrations were rearrangements of 11q23 and 12p deletions. In the pediatric group, 5q deletions and monosomy of chromosome 7 were the most common alterations.


Introduction
Myelodysplastic syndrome (MDS) represents a heterogeneous group of clonal disorders of hematopoietic stem cells characterized by quantitative and qualitative hematopoietic abnormalities, with cytopenia of one or more peripheral blood cell lineages but usually normal or hypercellular bone marrow (Sanz et al., 1989).Hematopoiesis is ineffective in one or more cell lineage, producing dyserythropoiesis, dysgranulopoiesis and dysmegakaryocytopoiesis (Goasguen and Bennett, 1992).These disorders are associated with a high risk of progression to acute myeloid leukemia (AML) and overall short survival, the death of MDS patients usually being due to cytopenia or progression to AML (Ganser and Hoelzer, 1992).
Conventionally, primary and secondary MDS are defined taking into account the previous history of the patients.According to the World Health Organization (WHO), primary MDS occurs without a known history of toxic exposure while secondary MDS is therapy-related and observed in patients with a known history of exposure to chemotherapeutic agents and/or radiation therapy (Jaffe et al., 2001).Although MDS occurs predominantly in older adults (median age 70 years) with a general incidence of 3 per 100,000 and reaches 20 per 100,000 over age 70 (Jaffe et al., 2001), MDS in children is usually more aggressive than in adults (Chan et al., 1997).
In order to improve the risk assessment of MDS an International Prognostic Score System (IPSS) was proposed by Greenberg et al. (1997), based on matched cytogenetic, morphologic and clinical data from 816 patients with primary MDS.In this system, variables were reevaluated prioritizing a more refined classification of bone marrow cytogenetic data, and three cytogenetic IPSS subgroups were recognized: good (presence of normal karyotypes, or monosomy of chromosome Y, or long arm deletions of chromosome 5 or 20); poor (presence of complex karyotypes, including three or more abnormalities, or aberrations of chromosome 7), and intermediate (presence of other chromosomal abnormalities).
Clonal chromosomal abnormalities have been reported in more than 3,000 MDS patients (Mitelman Database of Chromosome Aberrations in Cancer 2004), mainly adults, supporting the view that this syndrome is neoplastic in nature.The nonrandom distribution of the abnormalities through the different stages of MDS has helped to identify primary (pathogenetically essential for the establishment of the disease) and secondary chromosomal changes (acquired during the evolution of the disease) (Heim and Mitelman, 1995;Mitelman, 2000).Even so, further studies are still necessary for a better characterization of the frequency and nature of the chromosome aberrations in pediatric MDS (Martinez-Climent, 1997).As in the other hematological diseases, the identification of the chromosomal alterations involved in MDS is not only a powerful and essential tool for the clinical management and treatment of patients but is also central to basic research on this syndrome.
In the study described in this paper we determined the spectrum of chromosomal alterations in 93 Brazilian patients with myelodysplastic syndrome and investigated the correlation between clinical and cytogenetic findings.

Material and Methods
This study included 93 patients with primary MDS referred to the Cytogenetic Laboratory of the Clinical Hospital, Federal University of Paraná, Brazil from January 1988 to September 2002.Cytogenetic analysis of bone marrow cells was performed at the time of diagnosis.In our sample, 51 patients were male and 42 were female with a median age of 29 (range 1 to 78 years).These cases were grouped according to the French-American-British (FAB) Co-operative Group classification of Bennett et al. (1982) as: refractory anemia (RA), 39 patients; refractory anemia with ringed sideroblasts (RARS), 6 patients; refractory anemia with excess blasts (RAEB), 17 patients; refractory anemia with excess blast in transformation (RAEBT), 17 patients; and chronic myelomonocytic leukemia (CMML), 14 patients.The sample was subdivided into 66 adult (mean age 40.9 years range 19 to 78 years) and 27 pediatric (mean age 7.9 years, range 1 to 18) patients (Lopes et al., 2002).
Bone marrow cells were cultured for 24 h (Williams et al., 1984) and chromosome analyses performed using a modification of the Giemsa banding technique described by Scheres (1972), the chromosomes being classified according to the International System for Human Cytogenetic Nomenclature (ISCN, 1995).
The differences among the mean values obtained from age and bone marrow blast percentage and the cytogenetic IPSS subgroups (good, intermediate and poor) were analyzed using Fisher's test and homogeneity among the variances was tested by Bartlett's test.The number of patients with different cytopenias (number of lineages involved: erythroid, granulocytic and/or megakaryocytic) in the three subgroups was analyzed by the Chi-square test.The survival curves in the cytogenetic IPSS subgroups were estimated and compared using the Kaplan-Meier method and the log rank test, respectively (Bewich et al., 2004).
The frequency of chromosomal alterations observed in our 93 patients was significantly higher (c 2 = 34.01,d.f.= 6; p < 0.001) than that described among the MDS patients studied by Heim and Mitelman (1995) and San Miguel et al.(1996).

Discussion
We found chromosome abnormalities in the bone marrow of 69% of the 93 MDS cases analyzed, this frequency being higher than the 30 to 50% reported in the literature.The fact that most of our patients were referred to our service in a more advanced stage of the disease, when the frequency of chromosome alterations is usually high, might be an explanation for this difference.The possibility of non-referred exposure of our patients to leukemic agents could also account for the higher frequency observed.Indeed, in therapy-related MDS, higher frequencies (92%) of abnormal karyotypes are usually found (Smith et al., 2003).
In agreement with previous reports, we observed chromosome deletions as the most frequent structural alterations, and monosomies as the most frequent numerical alterations.The most frequent abnormalities in our patients [del(7q), del(5q), del(12p), rearrangements of 11q23, +8, del(20q) and +21] have been previously described in MDS (reviewed in San Miguel et al., 1996;Mhawech and Saleem, 2001), but the frequencies of some of them differed from those found in the literature.For example, 11q23 and 12p rearrangements were found at a higher frequency in our study whereas 5q and 7q deletions, monosomy 7 and trisomy 8 were less frequent (Heim and Mitelman, 1995;Mhawech and Saleem, 2001).It is noteworthy that abnormalities involving 11q23 and 12p have  been reported with high frequencies in therapy-related MDS (Streubel et al.,1998;Bloomfield et al., 2002).
In our pediatric cases the frequencies of chromosome aberrations were similar to those reported in the literature.Martinez-Climent (1997) and Kardos et al. (2003) reported the presence of abnormal karyotypes in about 50% of their patients, with monosomy 7/del(7q) being the most frequent abnormalities.Luna-Fineman et al. (1995) has pointed out that, in contrast to adult cases, in pediatric patients chromosome 7 abnormalities are not frequently involved in complex karyotypes but are found as single chromosomal changes.Our findings agree with this view, in that in our sample we found that five out of eight adult patients but only one out of three pediatric cases with chromosome 7 aberrations presented this abnormality as part of a complex karyotype.Monosomy 7 is more frequently observed in pediatric MDS than in AML (Martinez-Climent, 1997;Kardos et al., 2003).A rapid transformation to AML is a general rule for all pre-leukemic states associated with chromosome 7 monosomy (Haas and Gadner,1996), the short survival in these cases point to the importance of detecting this chromosome abnormality at the time of diagnosis to help in the establishment of the most appropriate therapy.
The 5q deletion, which has been described as occurring at a lower frequency in pediatric MDS patients than in adults (Martinez-Climent, 1997), was unexpectedly observed at a high frequency (20%) in our cases.Chromosome 7 monosomy was also found in our cases at a frequency of 20%, although this frequency is often described for this abnormality in MDS.
The significance of individual cytogenetic aberrations in relation to prognosis deserves careful evaluation.For instance, we found chromosome 8 trisomy and 20q deletion in patients in the RA stage, which is in general agreement with the IPSS classification which includes chromosome 8 trisomy in intermediate prognosis group and 20q deletion in the good prognosis group.However, it should be borne in mind that Fernandez et al. (2000) have described chromosome 8 trisomy and 20q deletion in more advanced stages of the disease (RAEB and RAEB-T).
In our study, histopathological parameters, bone marrow blast percentage and cytopenia were compared in the three cytogenetic prognosis subgroups but no significant differences were detected in spite of an increase of these parameters in the intermediate and poor cytogenetic risk groups.A possible explanation for these statistically nonsignificant findings could be due to the different proportions of patients submitted to bone marrow transplantation in each cytogenetic subgroup (good = 39.4%;intermediated = 47.2% and poor = 46.7%).Considering that the higher risk cytogenetic subgroups (intermediate and poor) had the higher proportion of patients submitted to BMT, we did expect an increased survival for these patients but no significant differences were observed between the survival curves for the different subgroups.

Table 1 -
Clinical data and karyotype of 66 adult MDS patients.

Table 2 -
Clinical data and karyotype of 27 pediatric MDS patients.

Table 3 -
Means of age and % blasts among IPSS subgroups of MDS patients.