High Resolution Molecular Analysis of 5q- Syndrome Patients at Diagnosis and Following Lenalidomide Therapy

The ‘5q- syndrome’ is characterized by macrocytic anaemia, normal to elevated platelet count, dyserythropoiesis, hypolobulated megakaryocytes and an interstitial deletion of the long arm of chromosome 5(q31–q32) as the sole karyotypic abnormality. In this study we have used the Affymetrix 250K high resolution SNP arrays to refine the 5q- syndrome commonly deleted region (CDR) and to explore undetected chromosomal aberrations in CD34+ cells and in constitutional tissue from 5q- syndrome patients at diagnosis (n=17) and following Lenalidomide treatment at haematological remission (n=9). To correlate genomic data with gene expression, genome-wide gene expression analysis on CD34+ cells was also performed using the Affymetrix U133plus2 arrays. We detected 35 chromosomal aberrations (3 deletions, 1 amplification and 31 regions of uniparental disomy (UPD) >2Mb) in 8 (35%) of patients unidentified by metaphase cytogenetic analysis. The median size of UPD was (4.03MB) (range: 2.49 – 30.31). Copy number changes identified were acquired aberrations, whereas UPD’s were also present in constitutional DNA. There were no common regions of overlap between any of the UPDs and no obvious change in gene expression correlating with the sites of UPD. SNP analysis precisely mapped the proximal and distal break points of the 5q deletion and clearly identified the proximal end of the common deleted region (CDR) to extend from 5q23.3 to the distal breakpoint 5q33.1. Our results did not identify any patients with isolated minimal deletions of 5q31 or the 5q32/33.1 regions as defined by Ebert et al, 2008. Furthermore, there were no regions of UPD identified on chromosome 5q in any of our patients. There were 3 genes located on the deleted 5q region that were >3 fold down-regulated; LOX (5q23–q31, Kif20a (5q31) and SLC22A4 (5q31.2). Gene set enrichment analysis (GSEA) of CD34 expression data from presentation samples identified the IL2 (p<0.001), IL2R (p<0.02), IL3 (p<0.03), IL6 (p<0.03) and TCR (p<0.04) gene sets to be significantly up-regulated in these cells compared to normal controls, emphasizing the contribution of aberrant immune regulation in this disease. Previously identified gene sets (erythroid, megakaryocytic and neutrophil (Ebert et al, 2008)) were not significant in our analysis, suggesting that the 5q deletion in CD34 cells does not pre-program differentiation along these lineages. Nine patients had paired analysis at presentation and following treatment with Lenalidomide. Interestingly, the 5q- deletion was detected in two patients by SNP arrays which were undetected by cytogenetic analysis. The proximal and distal breakpoints of the detected 5q deletion in these two patients post Lenalidomide therapy were identical to their corresponding presentation sample. Furthermore, in three other patients cytogenetic and SNP array analysis showed loss of the 5q- deletion following Lenalidomide therapy, however, deletions observed at presentations on chromosome 21q21, 7q22.1-36.3 and 12q21.2-q22 were retained in CD34+ cells post treatment respectively, suggesting the presence of additional abnormal clones in the bone marrow. GSEA of presentation CD34 cells and post Lenalidomide treated cells identified the IL4 (p<0.04) and IL6 (p<0.04) gene sets as being significant pathways affected by Lenalidomide therapy. Our SNP array data clearly defines the 5q- syndrome CDR and shows a high frequency of UPD in these patients. The identification of clonal aberrations not removed by Lenalidomide treatment suggests the presence of additional abnormal clones which may play a role in transformation of this disease. GSEA analysis of gene expression data in CD34+ cells clearly shows the important role of immune modulatory networks both at presentation of the disease and following Lenalidomide therapy.