Monosomy 7 and activating RAS mutations accompany malignant transformation in patients with congenital neutropenia.

Individuals with severe forms of congenital neutropenia suffer from recurrent infections. The therapeutic use of recombinant human granulocyte colony-stimulating factor (rhG-CSF) to increase the neutrophil count is associated with fewer infections and an improved quality of life. However, the long-term effects of this new therapy are largely unknown. In particular, it is unclear if myeloid leukemia, a known complication of some forms of congenital neutropenia, will occur with increased frequency among patients who receive long-term treatment with hematopoietic growth factors. We report 13 patients with congenital disorders of myelopoiesis who developed leukemic transformation with either myelodysplastic syndrome (MDS) or acute myelogenous leukemia (AML) and 1 who acquired a clonal cytogenetic abnormality without evidence of MDS or AML while receiving rhG-CSF. The bone marrows of 10 patients showed monosomy 7 and 5 had activating RAS mutations. These abnormalities were not detected in pretreatment bone marrows and cessation of rhG-CSF was not associated with either clinical improvement or cytogenetic remission. We conclude that patients with severe forms of congenital neutropenia are at relatively high risk of developing MDS and AML. The occurrence of monosomy 7 and RAS mutations in these cases suggests that the myeloid progenitors of some patients are genetically predisposed to malignant transformation. The relationship between therapeutic rhG-CSF and leukemogenesis in patients with severe chronic neutropenia is unclear.

Molecular analysis at the NF1 locus in astrocytic brain tumors.

BACKGROUND: Patients with neurofibromatosis type 1 (NF1) are at increased risk for developing malignant neural crest tumors and juvenile myeloid leukemia. Although the normal allele of the NF1 tumor-suppressor gene is frequently deleted in some of the malignant tumors that arise in patients with NF1, the role of NF1 alterations in the sporadic forms of these cancers is unclear. METHODS: A series of intragenic sequence polymorphisms was used to investigate lymphocyte and tumor DNA samples from 22 adults with high grade malignant gliomas for loss of heterozygosity (LOH) at NF1. In addition, an assay based on the polymerase chain reaction was used to screen these tumors for point mutations at codon 1423. RESULTS: One recurrent anaplastic astrocytoma showed LOH within NF1 but not with a flanking marker located near the gene. Of 21 informative tumors, none showed point mutations affecting codon 1423 of NF1. CONCLUSION: These data suggest that LOH at NF1 is uncommon in sporadic high grade astrocytoma, and codon 1423 is not a "hot spot" for activating point mutations in these tumors.

Molecular evidence that childhood monosomy 7 syndrome is distinct from juvenile chronic myelogenous leukemia and other childhood myeloproliferative disorders.

The observation that juvenile chronic myelogenous leukemia (JCML) and childhood bone marrow monosomy 7 syndrome (Mo 7) are similar in many clinical and epidemiologic respects suggests a shared pathogenic basis and raises the possibility that the bone marrows of patients with JCML might lose chromosome 7 alleles by mechanisms that do not result in detectable cytogenetic deletions. We used a series of polymorphic markers mapped to chromosome 7 to test this hypothesis in 22 children with MPS and MDS, including 19 with JCML. All MPS and MDS samples demonstrated allelic heterozygosity with at least one chromosome 7 marker; 16 were heterozygous with probes from both 7p and 7q. Furthermore, the percentage of patient bone marrow samples heterozygous at each locus tested was similar to the frequency observed in the normal population. Whereas these data demonstrate that submicroscopic loss of large segments of chromosome 7 alleles is uncommon in children with MPS and MDS who do not have Mo 7, they do not exclude small deletions around an uncharacterized tumor-suppressor locus. Our results suggest that a number of distinct molecular events contribute to leukemogenesis, and we propose a multistep model to explain the similarities and differences between the major subtypes of childhood MPS and MDS.

Genetic analysis is consistent with the hypothesis that NF1 limits myeloid cell growth through p21ras.

Children with neurofibromatosis, type 1 (NF-1) are at increased risk of developing malignant myeloid disorders and their bone marrows frequently show loss of the normal allele of the NF1 tumor-suppressor gene. NF1 encodes a protein called neurofibromin, which accelerates guanosine triphosphate (GTP) hydrolysis on the p21ras (Ras) family of signaling proteins. We used a genetic approach to test the hypothesis that NF1 negatively regulates myeloid cell growth through its effect on Ras. This model predicts that, if RAS mutations and loss of NF1 function deregulate myeloid growth by the same biomechanical mechanism, then activating RAS mutations will be restricted to children with malignant myeloid disorders who do not have NF-1. We studied 71 children, including 28 with bone marrow monosomy 7 syndrome (Mo7), 35with juvenile chronic myelogenous leukemia (JCML), three with other forms of preleukemia, and five with acute myelogenous leukemia (AML), for activating mutations of KRAS and NRAS. The incidence of RAS mutations was 21% (12 of 55) in patients without NF-1 and 0% (zero of 16) in children with NF-1 (P = .04). Among the 55 patients who did not have NF-1, we found RAS mutations in four of 27 with Mo 7, in five of 24 with JCML, in two of 3 with AML, and in a patient with myeloproliferative syndrome (MPS). These data from primary human cancer cells provide strong genetic evidence that NF1 limits the growth of myeloid cells by regulating Ras.

Loss of the normal NF1 allele from the bone marrow of children with type 1 neurofibromatosis and malignant myeloid disorders.

BACKGROUND: Children with type 1 neurofibromatosis (NF-1) are at increased risk for malignant myeloid disorders. Analysis of the NF-1 gene (NF1) suggests that the function of its product, neurofibromin, is reduced in affected persons and that NF1 belongs to the tumor-suppressor class of recessive cancer genes. This model is consistent with evidence that neurofibromin accelerates the intrinsic guanosine triphosphate-hydrolyzing activity of the Ras family of regulatory proteins. Loss of constitutional heterozygosity has not been reported in the benign tumors associated with NF-1, however, and has only been detected in a few malignant neural-crest tumors and in some tumor-derived cell lines. METHODS: We studied DNA extracted from the bone marrow of 11 children with NF-1 in whom malignant myeloid disorders developed and from parental leukocytes. We used a series of polymorphic markers within and near NF1 to determine whether leukemogenesis was associated with structural alterations of the gene. RESULTS: Bone marrow samples from five patients showed loss of heterozygosity. In each case, the NF1 allele was inherited from a parent with NF-1 and the normal allele was deleted. CONCLUSIONS: These data provide evidence of NF1 may function as a tumor-suppressor allele in malignant myeloid diseases in children with NF-1 and that neurofibromin is a regulator of ras in early myelopoiesis.

Parental origins of chromosome 7 loss in childhood monosomy 7 syndrome.

Bone marrow monosomy 7 (Mo 7) is associated with childhood preleukemic myeloproliferative and myelodysplastic syndromes (MPS and MDS). We used a series of polymorphic markers to investigate the parental origins of chromosomes lost from the bone marrows of 12 children with MPS/MDS and Mo 7. Eight Mo 7 bone marrows lost a maternal chromosome 7 and four cases lost the paternal homologue. Our data and the results of previous laboratory and clinical observations in the familial and sporadic forms of childhood Mo 7 suggest that chromosome 7 deletions contribute to leukemogenesis by gene dosage.