EXOSOME MIRNAS PROFILING IN SERUM AND PROGNOSTIC EVALUATION IN PATIENTS WITH MULTIPLE MYELOMA

Abstract

Introduction: Multiple Myeloma (MM) is an incurable hematological malignancy. Tightly crosstalk through microRNAs (miRNAs) carried by exosomes between the bone marrow microenvironment and MM cells plays pivotal roles in MM. Our study aimed to identify the expression profile of exosomal miRNAs (exo-miRNAs) in the serum of MM patients and investigate the regulation network and their potential functions. Methods: Exosomes in serum from 19 newly diagnosed multiple myeloma patients and 9 healthy donors were isolated and the miRNA profile was investigated by small RNA sequencing. Differential expression of exo-miRNAs was calculated and target genes of miRNAs were predicted. CytoHubba was applied to identify the hub miRNAs and core target genes. The LASSO Cox regression model was utilized to develop the prognostic model, and the ESTIMATE immune score was calculated to investigate the correlation between the model and immune status in MM patients. Results: Our study clarified 313 differentially expressed serum exo-miRNAs between MM patients and HD. GO analysis of target genes of differential miRNAs showed that these target genes are mainly involved in the critical biological processes related to MM pathogenesis such as proteasome-mediated ubiquitin-dependent protein catabolic process, indicating the indispensable roles of these miRNAs in MM pathogenesis. Top six hub differentially expressed serum exo-miRNAs (hsa-miR-4728-5p, hsa-miR-455-3p, hsa-miR-6779-5p, hsa-miR-124-3p, hsa-miR-615-3p, and hsa-miR-7106-5p) were identified. 513 target genes of six hub exo-miRNAs were confirmed to be differentially expressed in MM cells in Zhan Myeloma microarray dataset. Functional enrichment analysis indicated that these target genes mainly involved mRNA splicing, cellular response to stress, and deubiquitination. Thirteen core exo-miRNA target genes were applied to create a novel prognostic signature to facilitate the risk stratification of MM patients, and two groups of MM patients with diverse outcomes were recognized (Figure 1A). The high-score patients identified by thirteen core genes signature displayed a higher 70 high-risk gene set (UAMS-70) score and the 56 drug-resistance genes set score, which might partly explain their inferior outcomes (Figure 1B–C). Additionally, our results suggest that the thirteen core genes signature model was highly correlated with the status of the immune microenvironment, and patients in the low-score group had higher immune cell infiltration levels (Figure 1D). Besides, patients in the high-risk group displayed lower HLA family gene expression (Figure 1E). Encore Abstract - previously submitted to EHA 2023 Keywords: Bioinformatics, Computational and Systems Biology, Diagnostic and Prognostic Biomarkers, Risk Models No conflicts of interests pertinent to the abstract.