Abstract 2744

Deletions involving chromosome 7 are frequently encountered in myeloid malignancies, mostly imparting a poor prognosis. Single nucleotide polymorphism array (SNP-A)-based cytogenetics has allowed for an improvement in the mapping of disease-related chromosomal regions through identification of microdeletion that define minimally affected areas for the search for pathogenic mutations. In addition, SNP-A facilitated the detection of somatic uniparental disomy (sUPD), which is frequently found in myeloid malignancies. Regions of sUPD are often associated with homozygous mutations present in the affected area, as shown for example for UPD9p and JAK2 or UPD11q and C-CBL mutations. Based on the precise mapping of genomic lesions on chromosome 7, we compared the outcomes of patients characterized by specific chromosomal abnormalities. Among 1163 patients with various myeloid malignancies studied by SNP-A we identified patients with somatic LOH involving chromosome 7 for whom clinical outcome data was available to assess survival. Regions fulfilling the criteria of germ line-encoded copy number variants and regions of homozygosity present in a control cohort (N=1535) were excluded. The rest of newly observed regions were in/excluded based on their absence or presence in the paired non-clonal DNA. In total, we identified 45 patients with monosomy 7, 37 patients with del(7q), and 21 with UPD7q. The CDR for del(7q) and UPD(7q) spanned nt 115221516–126602492 and nt 127,484,450–158821424, respectively. Among patients with monosomy 7, 17 were primary MDS, 22 had various bone marrow failure disorders including juvenile myelomonocytic leukemia (JMML) and clonal disease in the context of aplastic anemia (AA) and Fanconi anemia (FA), whereas UPD7q was predominantly found in MDS/MPD and MPD (N=10), sAML (N=3), and primary MDS (N=7). Patients with del(7q) included primary MDS and sAML. The most common concomitant cytogenetic abnormalities were seen in chromosome 5q and 17p, including frequent UPD17p. Chromosome 5q and 17p lesions were more common in del(7q), identified in 10 and 7 patients respectively, and were represented equivalently among monosomy 7 (N=5 for each) and UPD7q patients (N=3 and 5 respectively). In addition to the large deletions of the long arm of chromosome 7 ([del(7q)] discussed above, using SNP-A-based karyotyping, 18 patients were identified with overlapping microdeletions involving two distinct loci at 7q22.1 (N=5 and 6, respectively), and one at 7q36.1 (N=2). The first 7q22.1 loci was found in 1 patient with MDS and 5 with MPN, while the second loci was identified in 3 patients with chronic myelomonocytic leukemia (CMML), 1 with MPN, and 1 with sAML. The 7q36.1 minimally affected region was found in 2 patients, 1 each with MDS and CMML. In 15 patients non-recurrent microdeletions were found with average size of 799 kb.

Clinical parameters were compared for each abnormality and Kaplan-Meier plots were generated and evaluated by log-rank test. Comparison of monosomy 7 and del(7q) did not show statistically different impact on outcome (p=0.26), whereas each was found to carry a significantly poorer prognosis (OS) than UPD7 (p=0.05 and 0.02 respectively). Deletion of 7q36.1, whose common deleted region was consistent with our previous identification of mutation within EZH2, carried an overall poor prognosis while 7q22.1 did not affect survival compared to patients without chromosome 7 abnormalities. These data confirm the known poor outcomes of patients with chromosome 7 abnormalities shared among monosomy 7 as well as del(7q). A slightly better albeit poor prognosis was seen in UPD(7q), suggesting that pattern or inheritance of chromosome 7 abnormalities into the dysplastic clone may affect ultimate leukemogenicity. UPD7q may represent a distinct pathogenesis, likely with distinct gene mutations compared to del(7q) and monosomy 7. Somatic microdeletions in chromosome 7 analysis may be nonpathogenic while those with poor prognosis, such as 7q36.1, may harbor pathogeneic genes.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution