Alterations at the rel locus in human lymphoma.

The rel proto-oncogene has been mapped to chromosome region 2p11.2-14, a site associated with nonrandom rearrangements in non-Hodgkin's lymphoma. We have characterized an abnormal rel mRNA from a cell line derived from a diffuse large cell lymphoma, in which the evolutionarily conserved N-terminal half of the rel coding region was fused with the C-terminal coding region of an unrelated gene. In addition, rearrangement or amplification of the rel locus was found in the lymphomatous tissue of two follicular and one diffuse large cell lymphoma. The findings suggest involvement of rel in the pathogenesis of large cell lymphoma.

Gene order, amplification, and rearrangement of chromosome band 11q23 in hematologic malignancies.

Eleven genes were found to be amplified in a patient with acute myelogenous leukemia and a homogeneous staining region 11q23qter. The gene order of such region was determined by using transverse alternating field electrophoresis of normal cell DNA and Southern blots of DNA from somatic cell hybrids, each containing a single human derivative chromosome 11 from six different chromosomal defects. This in turn allowed us to uncover a breakpoint in band 11q23.3 between the CD3 gamma and the ets-1 genes in genomic rearrangements found in acute myelogenous leukemia, acute lymphocytic leukemia, and B-cell diffuse lymphoma. The breakpoint of a constitutional deletion from a patient whose mother and brother have a heritable 11q23.3 fragile site occurs in the same region.

Mechanisms of ras mutation in myelodysplastic syndrome.

Using synthetic oligonucleotide hybridization, we have found a ras mutation in 11 of 27 patients (41%) with primary or non-therapy related myelodysplastic syndrome (MDS). This high incidence of mutation, mainly of the N-ras oncogene, was generally found in patients with disease progression to acute leukemia (8 of 11 patients = 73%). Two general mechanisms of ras mutation were found. In five patients the ras mutation was present in only a fraction of the cells; sometimes appearing in late stages of the disease, suggesting that it can occur in a differentiated cell clone. In six other patients, the ras mutation was present in the great majority of bone marrow or blood cells. The mutation was detected in mature normal lymphocytes and persisted following a complete clinical remission in two of these patients, implying that the ras mutation can occur in an early stage of cell differentiation or stem cell. Patients with a ras mutation had a median survival of nine months (all patients dead) compared to 16 patients without a ras mutation that had a median follow-up of 16 months (10 patients alive; P less than 0.005). Since 9 of the 11 patients (82%) with a ras mutation were found to have an abnormal monocytic component at diagnosis or during disease evolution, it is possible that in myeloid disorders a ras mutation is preferentially associated with myelomonocytic cell differentiation.

bcl-2 and other genomic alterations in the prognosis of large-cell lymphoma.

Approximately half the patients with diffuse or follicular large-cell or mixed large- and small-cell lymphoma enter a prolonged remission or are cured after receiving combined-drug therapy. It has been unclear, however, why the other half do not respond. We evaluated 54 previously untreated patients with diffuse lymphoma and 20 with follicular lymphoma, all of whom had a large-cell component and Stage II through IV disease, subsequently treated with combined chemotherapy. Different recurrent genomic defects were associated with differences in the response to treatment. Among the 54 patients with diffuse lymphoma, all 12 patients with a duplication of chromosome 3p had a complete clinical remission after a median follow-up of 39 months (11 patients survived). In contrast, all seven patients with a duplication of chromosome 2p had a partial response or no response to treatment and a median survival of six months (all died). Among the 20 patients with follicular lymphoma, all 5 patients with duplication 3p or +3 had a complete clinical remission (all survived), and 3 of 4 patients with duplication 2p or +2 had no response or a partial response to treatment and died. Twenty-three patients with B-cell non-immunoblastic lymphoma or follicular lymphoma who had a bcl-2 oncogene rearrangement had a poorer response to therapy (7 of 23 with complete remission) than the patients without bcl-2 rearrangement (21 of 26 with complete remission). We conclude that in large-cell or mixed-cell lymphoma, duplication of chromosome 3p is associated with a relatively good prognosis and duplication of chromosome 2p or bcl-2 oncogene rearrangement is associated with a relatively poor prognosis. Because such multiple recurrent genomic defects are also common in most other types of cancer, they may have general prognostic importance.

Refined chromosome study helps define prognostic subgroups in most patients with primary myelodysplastic syndrome and acute myelogenous leukaemia.

Based on a 6 1/2-year study of 284 consecutive adult patients with primary myelodysplastic syndrome (MDS) and and acute myelogenous leukaemia (AML), we have found that refined chromosome analysis can be used as an independent prognostic indicator in the great majority of patients with MDS and AML. In MDS, the FAB subtype was also found to have prognostic value and this was enhanced when the chromosomal findings were taken into consideration. In AML, the age of the patient correlated more closely with the chromosomal changes in predicting prognosis in most patients than did the FAB classification. Previously we reported that refined chromosome analysis of bone marrow specimens from 161 adult patients with primary or non-therapy related MDS and AML identified three prognostic chromosomal categories in each disease, representing 40% of all patients (Yunis et al, 1984, 1986). By extending our study to 284 patients, as well as a longer follow-up, it was possible to determine the prognostic implications of two additional chromosomal categories in MDS and five in AML. Since 73% of all patients are now represented in well-defined chromosomal subgroups with prognostic significance, refined chromosome analysis emerges as a tool that could have considerable impact in protocols.

Refined chromosome analysis as an independent prognostic indicator in de novo myelodysplastic syndromes.

In a study of 56 consecutive adult patients with de novo myelodysplastic syndromes (MDS), all cases were successfully analyzed with two refined chromosome banding techniques. Most patients (44 of 56, 79%) were found to have a chromosome defect. The majority of these patients had a recurrent loss of chromosomal material rather than a reciprocal translocation or inversion, as commonly found in acute leukemia. The three largest chromosomal categories found were associated with a wide range of survival. Twelve patients (21%) had normal chromosomes, a stable clinical course, and long survival (median follow-up time of 49 months, with all patients alive). Nine patients had in common a single chromosome defect resulting in either monosomy 7 or deletion 7q. They had a median survival of 12 months, and four died of acute nonlymphocytic leukemia (ANLL). Of 12 patients with complex defects, 11 had a complete or partial loss of a chromosome 5 and a complete or partial loss of the long arm of a chromosome 7 or 20. They had a poor median survival of four months, and six patients died of ANLL. Although the French-American-British (FAB) classification was also found to have some prognostic value, FAB subgroups were chromosomally heterogeneous and showed less dramatic differences in median survival than the larger chromosomal subgroups. We have shown, for the first time, that a refined chromosomal analysis is an independent prognostic indicator in de novo MDS and may be helpful in establishing therapeutic approaches in this difficult group of heterogeneous disorders.