DETERMINING OF MOLECULAR SIGNATURE FOR CENTROSOME ABNORMALITIES IN MULTIPLE MYELOMA

Sborník

Background. In multiple myeloma (MM), biologic complexity originates from complex oncogenic processes involving somatic acquisition of myriad mutations coupled with genetic variability within the host. This pathogenically determined molecular heterogeneity predetermines clinical heterogeneity. Aims. In this study, we performed gene expression profiling (GEP) focusing on centrosome related genes to determine molecular signature characteristic for centrosome abnormalities in MM patients. Methods. In total, 73 patients were evaluated for current study. The patient baseline characteristics were as follows: males/females 37/36, median age of 69.5 years (range, 46-85 years). Newly diagnosed (38/73) and relapsed (35/73) patients were included in this study; most of them had advanced stage of MM (DS II/III n =52; ISS II/III n=68). CD138+ cells were separated by MACS. All patients were evaluated by GEP using Affymetrix GeneChip Human Gene ST 1.0 array. In total, 111 genes were selected for clustering. All of these genes are involved in regulation of centrosome duplication and have known function in oncogenesis. Interphase FISH with cytoplasmic immunoglobulin light chain staining and qRT-PCR were performed on the same PC samples. Results. We have found one distinct pattern of 35 genes belonging to following ontologies: cell-cycle genes (CDK1, CDK2, E2F, Plk, AURKA, NEK2); kinetochore and microtubule attachment genes (AURKB, BIRC5, CENP); mitotic checkpoint genes (BUB1, MUD2); DNA damage checkpoint genes presented by CA suppressors, integrity and reduplicate regulators (RAD51, XRCC2, MSH2, CHEK1) and just one structural gene TUBG1. Based on expression of these genes, patients can be stratified into three groups - ‘High-’, ‘Mediate-’ and ‘Low-expressed’. The expression levels of genes selected from revealed pattern (AURKA, AURKB, CCNB1, CCNB2, HMMR, PLK4, TACC3, CDK2) were validated by qRT-PCR. PCR-based results clearly corresponded with GEP-based findings in mentioned above three subgroups of MM patients. We have found significant differences in relative quantification coefficient R of all 8 analyzed genes (P<0.05).There were no significant differences between revealed subgroups of MM patients regarding the presence/absence of common for MM cytogenetic aberrations. Patients in “Low expressed’ group showed significantly higher level of hemoglobin (P=0.008) and thrombocytes (P=0.018), otherwise patients in “Mediate-”and “High expressed”group had significantly higher b2m (P=0.004), LDH (p=0.019) and plasma cell infiltration in bone marrow (P=0.042). Although statistical significance was not reached as a result of the small numbers of patients (P=0.098), however, in newly diagnosed patients significant difference between subgroups was observed (P=0.005). There was worse OS of patients in “High”and “Medium” subgroups than in patients in ‘Low’ subgroup. Conclusions. We defined a gene pattern, which was used as a molecular signature of centrosome abnormalities. Defined associations with integral clinical parameters and OS allow us to suggest the impact of revealed functional gene set in MM genesis. Our findings still need to be confirmed and validated on a larger external cohort of patients. Funding. This work was supported by grants: NT11154, NT12130, MSM0021622434 and by project GAP304/10/1395.

Haematologica, 2012; 97(s1):  16