Multiple myeloma (MM) is one of the most common hematologic malignancies, characterized by uncontrolled proliferation of plasma cells in the bone marrow. Investigating MM epigenetics at the non-coding RNAs (ncRNAs) level and their interactions can be a key factor in understanding this disease. Circular RNAs (circRNAs) are ncRNAs whose 5’and 3’ends are covalently joined and have been mostly investigated in MM by acting as microRNAs (miRs) sponges, but it is known that they also interact with RNA binding proteins (RBPs). This study aimed to investigate the differentially expressed (DE) circRNAs in MM, their relationship with RBPs and miRs, and their biological implications. Public RNA-Seq datasets (PRJNA634534, PRJNA472759) were downloaded from ENA and the reads were treated to obtain the circRNA count matrix. The DE circRNAs were identified using the CircTest package and then submitted in circAtlas3.0 to obtain their interactions with miRs and RBPs. Furthermore, only miRs that interacted with two or more circRNAs were considered. miRTarBase was used to identify the target genes, selecting those validated by strong evidence. Functional enrichment analyses (FEA) were performed, using ReactomePA, for each circRNA considering both their RBP and miR-targets. Lastly, by combining the circRNAs'expression results and their interaction with RBPs or miRs, genes were selected to be investigated in a bigger RNA-Seq dataset of MM patients, available in TCGA. Gene expression was compared according to the patient's vital status and ISS stage, and the DE genes were selected for a Kaplan-Meier survival analysis. All analyses were performed on R and p-values<0.05 were considered statistically significant. First, 6 circRNAs were found as downregulated in MM compared with health controls (referred to as circs-down), and 4 circRNAs were specifically expressed in MM (referred to as circs-spe). In RBP analyses, 47 RBPs interacted with circs-down, and 51 with circs-spe. FEA resulted in 35 pathways, in which one pathway was exclusively enriched by circs-down, and 7 others by circs-spe. From 12 RBPs that interacted exclusively with circs-down, 5 genes were DE in the vital status analysis, and 2 genes were DE in the ISS stage analysis. And from 16 RBPs exclusively to circs-spe, 8 genes were DE in the vital status analysis, and 2 genes were DE in the ISS stage analysis. In miR analyses, 24 miRs and 172 targets interacted with 5 circs-down, while 18 miRs and 79 targets interacted with 3 circs-spe. FEA resulted in 396 pathways, 242 only for circs-down and 61 exclusively to circs-spe. From 198 exclusive axes formed by 5 circs-down, 7 miRs, and 94 targets; 8 genes were deregulated in both analyses (vital status and ISS stage). 2 axes were formed by 2 circs-spe, 1 miR, and 1 target. Altogether (RBPs and miR-targets), 22 genes (when upregulated) and 4 genes (when downregulated), were significantly associated with poor overall survival. This study highlighted potential molecular interactions in which circRNAs are key players. Through their interactions with RBPs and miRs, circRNAs can be involved in important cancer-related functions. Furthermore, we identified genes that are associated with poor survival in patients with multiple myeloma, which may help in the prognosis assessment and development of new therapies for this disease.