Correlation of Non-Coding RNA Expression with Response to Proteasome Inhibitors in Multiple Myeloma

Background: Multiple myeloma (MM) is a heterogeneous diseaseand there is an increased need for more accurate risk classification methods to improve treatment decision-making because of its high impact on clinical outcomes. Here, we demonstrate evidence to support the prognostic value of non-coding RNAs (ncRNAs) as newly discovered genetic biomarkers of drug-resistant and/or high-risk forms of MM. NcRNAs, e.g., long ncRNAs (lncRNAs) and microRNAs (miRNAs), act as positive or negative regulators of gene expression to control cell proliferation, apoptosis and drug resistance. NcRNAs have been shown to play a role in both solid and hematological tumors. Stratification of MM based upon cytogenetic abnormalities and protein-coding gene signatures does not adequately correlate with the depth and durability of response to novel agents such as bortezomib. Therefore, ncRNAs as new class of molecular effectors may enhance the basic understanding of myelomagenesis and provide better stratification of myeloma subtypes. To investigate the role of ncRNAs in resistance to proteasome inhibitors (PIs), we compared global ncRNA profiling in drug-naïve cells to cells with acquired resistance to the PIs bortezomib, carfilzomib and ixazomib. We hypothesized that ncRNAs commonly deregulated in the 3 resistant cell lines would yield a ncRNA signature and novel therapeutic targets.

Experimental Procedures: RPMI 8226 cells resistant to PIs were generated through successive exposure to bortezomib, carfilzomib or ixazomib over a period of 6 months. Total RNA was isolated and genome-wide ncRNA expression profiling was performed using Affymetrix3.0 microarray chips that contained nearly 40,000 miRNA and 13,300 lncRNA probes. NcRNA expression profiles from drug-resistant cells were compared to that of drug- naïve parental cells treated with vehicle alone using the same treatment algorithm. Housekeeping genes were used for log expression normalization. MM patients' bone marrow aspirates were obtained from patients after University of Cincinnati Institutional Review Board approval. Results: Bioinformatic analysis of the ncRNA profiles identified a panel of 87 lncRNAs and ~40 miRNAs that were significantly (>100-fold) deregulated in all three drug-resistant cell lines relative to drug- naïve parental cells. Strikingly, ~90% of the deregulated lncRNAs exhibited a similar expression pattern in all 3 PI-resistant cell lines. Twenty lncRNAs were deregulated > 1000-fold in all 3 resistant cell lines (Figure 1). RPMI 8226 cells carry a chromosomal (14,16) translocation. Interestingly, none of the deregulated lncRNAs detected here localized to chromosome 14 or 16, suggestive of a cytogenetic-independent mechanism of drug resistance. The lncRNA COL4A-2A was upregulated >5,000-fold in resistant cells and displayed extensive sequence complementarity to miRNA-29 that was downregulated in resistant cells. Also, our microarray-based studies have identified ncRNAs deregulated in MM patient tumor samples relative to normal plasma cells from healthy age-matched individuals. A significant number of the deregulated ncRNAs between drug- naïve and drug resistant cells were also deregulated in normal plasma cells relative to myeloma cells. Studies are correlating the ncRNA patterns seen in drug-sensitive and drug-resistant cell lines with ncRNA patterns obtained from malignant plasma cells of patients currently receiving bortezomib-based therapy. Updated results to correlate ncRNA expression with myeloma patient response to bortezomib will be presented.Conclusions: Taken together, we have identified a curated panel of ncRNAs deregulated in common within myeloma cells generated with acquired resistant to three different clinically-relevant proteasome inhibitors. Ongoing studies will correlate ncRNA expression patterns from resistant cells with patterns generated from patients with monoclonal gammopathy of unknown significance (MGUS), Smoldering MM, newly diagnosed MM, refractory disease and plasma cell leukemia. In addition, ncRNA patterns will be generated based upon MM patient response to bortezomib. Further investigation is warranted to shed light on the role of these ncRNAs in the development of MM, to identify their targets and to define their role in drug resistance.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal