Integrative Genomic Analysis of Primary Plasma Cell Leukemia Revealed Strong Gene and MicroRNA Dosage Effect

Abstract 4040

Multiple myeloma (MM) is a clonal proliferation of malignant plasma cells (PCs) characterized by a marked genomic instability. Primary plasma-cell leukemia (pPCL) is an aggressive, rare variant of plasma cell dyscrasia characterized by extra-medullary proliferation of PCs, high genomic instability and very poor prognosis. To date, global genomics studies in pPCL are still limited. Highly purified PCs were obtained from 17 previously untreated pPCL patients, recruited in an open-label, exploratory, single-arm, two-stage study from the GIMEMA myeloma network aimed at the evaluation of safety and antitumor activity of the immunomodulatory agent lenalidomide in combination with low dose dexamethasone in previously untreated pPCL. All the samples were characterized for the main chromosomal aberrations by Fluorescence in-situ hybridization (FISH). Specifically, 13q and 17p deletions have been identified in 12/17 (70.6%) and 8/17 (47.1%) cases, respectively; the presence of t(11;14) was found in 4/17 patients (23.5%), t(4;14) in only 1/17 cases (5.9%) and MAF translocations in 8/17 (47.1%). To further investigate the genomic complexity of pPCL, we analyzed a subset of 13 samples by means of an integrative approach using different high-throughput microarray technologies: Human Mapping 250K Nsp SNP-array (Affymetrix) for the detection of copy number alterations (CNA), Gene 1.0 ST array (Affymetrix) for transcriptional profiling and miRNA Microarray v2 (Agilent) for the global miRNA expression. SNP-array data were concordant with FISH results for the detection of the main chromosomal aberrations, i.e. 13q (10/13=77%) and 17p (7/13=58.3%) deletions. The most recurrent CNA specifically identified by SNP-array was represented by 1q gain (8/13=61.5%); in addition. losses involving chromosomes 1p (5/13=38.5%), 8p (4/13=30.8%), 14q (5/13=38.5%), 16q (5/13=38.5%), gains affected 7q (4/13=30.8%) and 19p (4/13=30.8%) and one amplification at 17q21 in 6/13 pPCL (46.2%) were detected. One case displayed a near tetraploid karyotype and, interestingly, another one showed a hyperdiploid pattern. Mapping information was integrated with the gene expression and miRNA profiles of the tumor samples. A non-parametric analysis (Kendall's tau correlation at a p-value<0.005) identifying 199 probes whose expression levels strongly correlated with the occurrence of allelic imbalances. Most of those probes (181/199=91%) were localized in the previously described altered regions; specifically, chromosomes 1p (12.6%), 1q (40.7%), 7q (3.5%), 8p (6.0%), 13q (4.5%), 14q (9.5%), 16q (6.5%) and 17p (7.5%). The same integrative approach has been applied to investigate the correlation between miRNA expression levels and the CNAs: 23 miRNAs had been detected at a p<0.05, most of them (16/23=69.5%) mapped to chromosomes 1p (21.7%), 13q (26.1%) and 19 (21.7%). These results highlight a wide gene-dosage effect suggesting that genomic structural abnormalities in pPCL closely reflect in expression imbalances. Our integrative approach data provide insights into the characterization of novel genetic lesion in pPCL, and suggest that a wide gene- and microRNA-dosage effect is a common characteristic of plasma cell dyscrasias, as previously described by our group in multiple myeloma PCs.