Erythroblastic transformation of chronic granulocytic leukemia: a clinical and cytogenetic case study

Chromosome abnormalities in CML

The Ph chromosome is the hallmark of CML, where it is found in more than 90% of the cases. Cytogenetically, it usually results from a t(9;22)(q34;q11). The Ph arises in a stem cell and in chronic phase is found in all haematopoietic cell lineages, although it causes only increased granulopoiesis, and sometimes increased thrombopoiesis; furthermore blast crisis may occur in all differentiative patterns of the pluripotent stem cell. Recently, molecular investigations of Ph positive CML cases have revealed a consistent genomic recombination between two genes, BCR on chromosome 22 and the ABL oncogene. The latter is translocated from 9q34, its normal site, to the 22q- or Ph chromosome. This molecular rearrangement expresses a unique 8.5 kb BCR-ABL hybrid mRNA transcript, that encodes an altered BCR-ABL protein of approximately 210 kD with enhanced in vitro tyrosine kinase activity. The breakpoints on chromosome 22q- are clustered in a 5 kb DNA fragment, allowing their study using Southern blot analysis. Cytogenetic variant forms of the Ph translocation involving three or more chromosomes are found in about 5% of the cases. Southern blot and in situ hybridization studies have demonstrated that these variants are cytogenetically more complex than the standard t(9;22) but molecularly they show the same essential genomic recombination. This is also true for a small number of cases of Ph negative CML. Clonal progression, indicated by the presence of clonal, non-random chromosome abnormalities, in addition to the Ph is rare during chronic phase but is found in 80% of blast crisis. These additional aberrations may precede BC by weeks or months and have therefore a clear prognostic value. Ph is not restricted to CML, since it is also found in ALL (20% of adult cases) and rarely in AML. Ph in acute leukaemia is cytogenetically indistinguishable from Ph in CML, but molecular studies have shown that in 50% of the cases the breakpoint on chromosome 22 is different from the very consistent and characteristic breakpoint in CML. Nevertheless genomic recombination takes place that results in a novel ABL protein at least in some of the cases. Despite extensive cytogenetic and molecular investigations, the mechanisms underlying the formation of the Ph as well as the pathogenesis of Ph positive CML are still unknown but are now the object of intensive research.

The karyotype of blastic crisis

The nonrandomness of chromosome clonal evolution in blastic crisis of chronic myeloid leukemia is well established, with three major changes [+8, +Ph, i(17q)] occurring alone or in combination in over 70% of the patients. The chromosome changes observed in different tissues may reveal the origin of the abnormal clones, as well as provide evidence for distinct or common evolution by different cell populations. The significance of the chromosome abnormalities and their relationship to blastic conversion are discussed. In general, chromosome evolution may be considered a rather reliable predictive or diagnostic parameter of blastic crisis but both the nature and the subsequent behavior of abnormal clones appear to be of critical value. As to the clinical/chromosome correlations, a few major points have emerged: the i(17q) aberration is mostly associated with signs of myeloid differentiation of blasts and a marked basophilia; it is mainly observed in the late stage of the disease, but overall median survival of patients with this marker is usually long; more atypical or complex changes usually are associated with a worse prognosis; patients with only a Ph in their blasts may have a longer survival, at least in some cytologic subgroups; and d) the loss of the Y chromosome seems to protect the cell against further clonal evolution. Finally, the relevance of the chromosome changes in the multistage process of blastic conversion is discussed, and the breakpoints of secondary changes recorded so far are reviewed and examined. It appears that certain chromosome regions are more often affected; these might contain genes of critical importance for the final malignant progression. Molecular biology may provide insight, in the future, on the nature and expression of involved genes.

Cytogenetics and cell surface marker analysis in chronic myelocytic leukemia. II. Implications for patient management

Chronic myelocytic leukemia (CML) is a model system for the study of many aspects of malignant disease. One aspect that correlates with decreasing therapeutic response is tumor progression. This progression is often accompanied by clonal evolution. In those cases where aggressive therapy does not prevent this evolution, the clinical response to therapy usually proves to be poor and of short duration. Investigators are concentrating their efforts in three primary, but not mutually exclusive, areas with respect to the clinical management of CML. These include: an attempt to distinguish patients at risk for early transformation from those who will have a prolonged chronic phase; the cryopreservation of autologous bone marrow or buffy coat early in chronic phase for subsequent use in the accelerated phase and; endeavors to identify early markers for disease progression allowing intervention before an irreversible blast crisis occurs. This report deals with two types of potential prognostic markers of transformation: chromosomal and cell surface characteristics. The appearance of nonrandom abnormal chromosomal patterns has been correlated with myeloblastic transformation by many investigators. However, there has always been a subset of CML patients who do not undergo clonal evolution. Additionally, the type(s) of transformation in CML may vary depending on the cell lineages involved. Unlike myeloblastic transformants, many of our patients who do not exhibit clonal evolution as a concomitance of disease progression develop a lymphoblastic transformation. Cytofluorometric analysis can distinguish small populations of abnormal cells with lymphoblastic characteristics (HLA DR+). Initial data suggests that patients expressing the HLA DR+ in their "normal" peripheral blood cells are at risk of undergoing lymphoblastic transformation. The combined use of clinical, cytogenetic, and cytofluorometric data to predict an impending transformation and to discriminate between myeloblastic and lymphoblastic populations allows clinicians to manage their patients more effectively.

Translocation (1;17) in accelerating and blast phases of chronic myelogenous leukemia

A translocation (1;17)(p11;q11) has been observed in two chronic myelogenous leukemia (CML) patients studied during the accelerated or acute stages of their diseases. In a review of the literature, four additional cases of t(1;17)(p11;q11) were identified, suggesting that this marker may be specific for the terminal phase of CML.

Chronic granulocytic leukemia: correlation of blastic transformation type with karyotypic evolution

Over a 3-year period, 26 patients with Philadelphia chromosome-positive chronic granulocytic leukemia were studied cytogenetically in both the chronic and blastic transformation phases of the disease; a further three patients were studied only after blastic transformation. Sixteen were considered to have adequate evidence of the type of transformation and form the basis of the report, where chromosome changes have been correlated with the morphological type of blastic transformation. Seven patients developed a myeloblastic transformation, seven a lymphoblastic transformation, and two an erythroblastic transformation. All patients in the myeloid group acquired one or more of the nonrandom changes associated with CGL blastic transformation, viz. +8,i(17q), +19, +22q-. Patients in the lymphoblastic group acquired structural abnormalities, apparently random in nature and usually in a small percentage of cells. The two patients with erythroblastic transformation developed markedly hyperdiploid cells (greater than 50 chromosomes) with both numerical and structural abnormalities. Patients in the lymphoblastic group appeared to have a slightly better prognosis than the myeloid group, whilst the patients with erythroblastic transformation had a very poor prognosis.

Characterization of blast cells in chronic granulocytic leukaemia in transformation, acute myelofibrosis and undifferentiated leukaemia. I. Ultrastructural morphology and cytochemistry

A systematic analysis of the blast cell population was carried out on samples from 50 patients suffering from blast transformation of chronic granulocytic leukaemia (CGL) (31) and of myelofibrosis (4), acute myelofibrosis (AM) (11) and undifferentiated acute leukaemia (4). Transmission electron microscopy (TEM), used in 41 samples, included: morphology and techniques for myeloperoxidase (MPO), platelet-peroxidase (PPO) and acid phosphatase (AP). The majority of cases were also studied by light microscopy cytochemistry and with a battery of cell markers which are reported in the accompanying paper (San Miguel et al, 1985). The characterization of the type(s) of proliferating blasts was made from the integration of ultrastructural and immunological data. TEM morphology allowed the precise recognition of specific granules in basophil and mast-cell precursors and of ferritin particles in blasts of erythroid lineage; these rare cell types were not adequately characterized by other methods. The PPO reaction made possible the identification of pure megakaryoblastic proliferations in 38% of cases, including eight of the 11 with AM; megakaryoblasts were also present in nine of 12 cases with mixed blast cell types. The MPO and AP reactions were useful for the characterization of myeloblasts and monoblasts, respectively. Lymphoblasts could be distinguished from other cell types by TEM morphology and negative MPO and PPO reactions. TEM techniques were valuable for diagnosing correctly the type of blast cell in this study in which only four cases (8%) remained unclassifiable.