Randomized comparison of cyclophosphamide-total body irradiation versus busulfan-cyclophosphamide conditioning in autologous bone marrow transplantation for acute myeloid leukemia.

PURPOSE: This prospective trial of autologous bone marrow transplantation for acute myeloid leukemia was undertaken to compare the outcome using two different preparative regimens. METHODS AND MATERIALS: Between October 1987 and April 1993, 35 patients with acute myeloid leukemia in first (n = 12) or greater (n = 23) remission were stratified by remission status and randomized to undergo 4-hydroperoxycyclophosphamide purged autologous bone marrow transplantation after either cyclophosphamide (120 mg/kg) and total body irradiation (1320 Gy in eight fractions over 4 days) (CY/TBI), or busulfan (16 mg/kg) and cyclophosphamide (200 mg/kg) (BU/CY) conditioning. RESULTS: At 2 years, overall survival and disease-free survival were 39% (95% confidence intervals (CI) 22-57%) and 36% (95% CI 19-52%), respectively. Patients in first complete remission had a significantly better outcome with a 2-year disease free survival of 57% (95% CI 28-86%) compared to others at 24% (95% CI 6-43%, log rank p = 0.048). For patients conditioned with CY/TBI, the estimated 2-year disease-free survival was 50% compared to 24% for patients conditioned with BU/CY (log rank p = 0.12). Estimated 2-year relapse rates were 43% vs. 70% (log rank p = 0.17), respectively. For patients in first complete remission no differences in disease-free survival (2-year estimates 67% vs. 50%, log rank p = 0.69), between the two regimens were observed. For patients in greater than first complete remission there was a trend towards improved disease-free survival in the CY/TBI arm (2-year estimates 42% vs. 9%, log rank p = 0.06). There were no differences in time to white blood cell count (WBC) engraftment, absolute neutrophil count of > 500, incidence of bacteremias, or median time to hospital discharge between the two regimens. Acute toxicities were similar. Interstitial pneumonitis developed in two patients (one on each arm), while veno occlusive disease developed in three BU/CY patients, but none of the CY/TBI patients (log rank p = 0.07). CONCLUSIONS: Cyclophosphamide-total body irradiation provided an equivalent or better outcome to BU/CY, particularly in advanced patients, and should remain the standard by which new regimens are judged. The high relapse rate with both regimens, especially patients who were in greater than in first complete remission, emphasizes the need for early transplant and for new strategies to improve outcome.

Direct detection of the Philadelphia chromosome in CD20-positive lymphocytes in chronic myeloid leukemia by tri-color immunophenotyping/FISH.

Six patients with previously diagnosed chronic myelogenous leukemia (CML) were studied by a tri-color immunophenotyping/FISH method for direct determination of the Philadelphia (Ph) chromosome in B and T lymphocytes. Two patients had involvement of CD20-positive lymphocytes. CD3-positive lymphocytes in all patients were negative for the Ph chromosome.

Chronic myelogenous leukemia: molecular diagnostic considerations.

Chronic myelogenous leukemia (CML) is associated in 95% of cases with the Philadelphia chromosome (Ph1) by cytogenetic analysis. This acquired karyotypic abnormality is the product of a balanced translocation between the c-abl oncogene on chromosome 9 and the breakpoint cluster region (BCR) gene on chromosome 22. The resulting BCR/c-abl hybrid gene is actively transcribed and is considered essential in the pathogenesis of this disease. Southern blot- and polymerase chain reaction (PCR)-based tests for the detection of this rearrangement have been introduced over the last decade. These molecular tests are useful adjuncts to cytogenetic analysis in CML and may provide useful clinical information in certain instances.

Morphologic and quantitative changes in blood and marrow cells following growth factor therapy.

Sequential blood and bone marrow specimens from 53 patients receiving recombinant granulocyte (G-CSF) or granulocyte macrophage colony stimulating growth factor (GM-CSF) for neutropenia were evaluated. The blood findings were marked by a neutrophilia with a prominent left shift, increased azurophilic granulation, Döhle bodies, and an elevated leukocyte alkaline phosphatase; circulating myeloblasts were observed but did not exceed 2% of the leukocytes. Nuclear segmentation abnormalities consisting of hyposegmentation, hypersegmentation, and ring nuclei were noted but were not a prominent finding. A leukoerythroblastosis was present in 54% of patients. No consistent effect on cell lines other than neutrophils was found. A monocytosis was present in 12 patients, a transient lymphocytosis in 2 and an eosinophilia in 1. No effect was evident on basophils. The morphologic changes in the neutrophils in the bone marrow specimens were most pronounced in the early period of growth factor therapy with a relative neutrophil hyperplasia with a marked increase in promyelocytes and myelocytes. With increasing duration of therapy, the myeloid to erythroid ratio normalized and the percentage of promyelocytes decreased while myelocytes and band neutrophils increased. Thirteen patients had no response to growth factor. The nonresponding patients were clinically diverse; all bone marrow biopsy specimens in this group were virtually acellular. No differences were noted between G-CSF and GM-CSF.

Duplication of small segments within the major breakpoint cluster region in chronic myelogenous leukemia.

The t(9;22) in chronic myelogenous leukemia (CML) may be reciprocal or, in a minority of cases, may result in an extensive deletion of a portion of the major breakpoint cluster region (M-bcr) of the BCR. This report provides evidence of the duplication of small segments within the M-bcr in a small group of patients with CML. Southern blots of Bgl II and Bgl II/BamHI double-digested DNA from the blood or bone marrow of 46 patients with CML were probed with a 5' 1.4-kb Taq I/HindIII M-bcr probe and a 3' 2-kb HindIII/BamHI M-bcr probe. In three patients, rearrangements were noted with both probes in Bgl II-digested DNA, but were not present in Bgl II/BamHI-digested DNA with either probe. Southern analysis of DNA samples double-digested with Bgl II and BspHI from two of these three cases showed no rearrangements with either probe; the M-bcr BspHI site is located 26 bp 3' of the BamHI site in the second intron of the M-bcr. The presence of a rearranged M-bcr with both probes in Bgl II-digested DNA and the lack of rearrangement in Bgl II/BamHI and Bgl II/BspHI double-digested DNA suggest the presence of M-bcr BamHI and BspHI sites on both 9q+ chromosome (9q+) and the Philadelphia chromosome (Ph). This implies a duplication of at least the 26-bp M-bcr BamHI/BspHI fragment in these two samples. Sequence data from one of these two cases confirmed the M-bcr breakpoints to be staggered; the Ph M-bcr breakpoint occurred 258 bp downstream from the 9q+ M-bcr breakpoint. It is concluded that a duplication of small segments within the M-bcr occurs in a small group of patients with CML, which may lead to pseudogermline patterns on Southern blot. Such a duplication may provide insight into the mechanism of some chromosomal translocations in neoplasia.

Methylation status of the major breakpoint cluster region in Philadelphia chromosome negative leukemias.

It has been shown that a 600 bp long cluster of cell lineage specific hypomethylated sites in the major breakpoint cluster region (M-bcr) on chromosome 22 exists in hematopoietic cells. To determine possible relationships between methylation patterns within the M-bcr and the stage of hematopoietic cell development, the M-bcr methylation status of 39 patients with leukemia and lymphoma and two patients with myelodysplastic syndrome with non-rearranged M-bcrs was examined by BgIII-HpaII digestion. In the myeloid malignancies, the presence of a hypermethylated 4.8 kb BgIII-BgIII M-bcr allele was directly proportional to the combined myeloblast and promyelocyte percentage of the specimen, whereas the presence of a 2.5 kb BgIII-HpaII allele was directly proportional to the combined percentage of monocytic cells and neutrophils. All five acute monoblastic leukemias showed a methylation pattern that closely resembled neutrophils. All of thirteen surface immunoglobulin positive B-cell malignancies showed a distinct methylation pattern consisting of three or more BgIII-HpaII restriction fragments of 2.5 kb or less in length. The B-cell precursor leukemias showed heterogeneous M-bcr methylation patterns, with four of seven showing a B-cell pattern and three showing a hypermethylated pattern with 4.8, 3.1/3.0 and/or 2.5 kb BgIII-HpaII M-bcr alleles. It is concluded that the M-bcr methylation status is related to the maturation of the neutrophil series; the surface immunoglobulin positive B-cell malignancies are characterized by a distinct, extreme hypomethylation pattern of the M-bcr; and the B-cell precursor malignancies appear to have a heterogeneous M-bcr methylation pattern.