Abstract
Multiple myeloma (MM) is an age-dependent neoplasm characterized by clonal expansion of malignant antibody-producing bone marrow plasma cells derived from post-germinal-center B cells. Despite recent improvements in treatment strategies, myeloma remains a difficult-to-cure disease with a 5-year survival rate of ~62%, frequent disease relapse, and acquisition of resistance to standard-of-care drugs. While the proteasome or the immune system are validated targets of myeloma therapy, our laboratory is interested in the development of drug candidates against new molecular targets using a combination of translational bioinformatics and molecular pharmacogenomics in preclinical models of (relapsed/refractory) RR MM. For this purpose, we have designed a novel optimization-regularization-based computational prediction algorithm called “secDrug” to identify therapeutic targets for the management of treatment-resistant cancers. When applied to B-cell malignancies, the secDrug algorithm predicted the pro-survival proteins XIAP, cIAP2, and MCL1 as the top drug targets.
The DEAD-box RNA helicase protein 5 (DDX5 gene, formerly p68) has been shown to play key roles in the transcriptional regulation of several secDrug-predicted targets in myeloma, including MCL1, XIAP, and cIAP2, as well as c-MYC and mutant-KRAS. DDX5 is a commonly expressed oncoprotein in myeloma (~75% of patients) and is found to be overexpressed in 67-74% of malignant plasma cells compared to normal plasma cells. Furthermore, DDX5 depletion has been shown to induce apoptosis via the selective inhibition of p52 and RNA polymerase II recruitment to p21.
Our group has developed FL118, a novel camptothecin analog, that acts as a molecular degrader that selectively inhibits DDX5. We have designed, synthesized, and validated a series of novel analogs of FL118, including FL77-9, FL77-24, and the fluoroaryl-substituted FL118 derivative 7h (FL77-32), which have shown significant in vitro cytotoxicity and in vivo anti-tumor efficacy in solid tumors and mantle cell lymphoma, with a favorable preclinical toxicity profile in murine models. 7h contains a C-F bond, which can withstand metabolic processes, protecting susceptible sites from oxidative metabolism. Furthermore, incorporation of fluorine atoms enhances fat solubility (LogP=3.81 for FL77-32, compared to LogP=2.83 for FL118), cellular plasma membrane permeability (total polar surface area or tPSA<140 Å2), and target specificity.
We performed in vitro cytotoxicity studies with FL118 and its derivatives (FL77-9, FL77-24, and FL77-32) using a panel of human myeloma cell lines (n>50) representing innate and acquired response/resistance to established drugs, we found that FL118 and its derivatives (FL77-9, FL77-24, and FL77-32) are highly potent in multiple myeloma, with the best potency displayed by FL77-32. Single-agent half-maximal inhibitory concentration (IC50) values were consistently in the sub-nanomolar range (as low as 0.41 nM) and showed high efficacy in combination with proteasome inhibitors/PI (e.g., Bortezomib) and Immunomodulatory drugs/IMiDs (e.g., lenalidomide). We then validated target inhibition, induction of apoptotic markers, including DDX5 and its downstream targets MCL1 and survivin, as well as the downregulation of myeloma drivers like MYC and NFkB, using bulk and single-cell whole-transcriptomics, gene knockdown studies, immunoblotting, and functional genomics. Finally, ex vivo analysis in CD138+ primary bone marrow samples obtained from patients confirmed the dose-dependent depletion of myeloma cells following treatment with FL118 derivatives. Currently, we are evaluating the in vivo anti-tumor efficacy of FL77-32 alone or in combination with Bortezomib in transgenic mouse models of myeloma, including chemo-resistant cells in NSG mouse models.
Our results demonstrate the potential of adding novel DDX5 inhibitors to the armamentarium of clinical trial-ready drug candidates for treating RRMM.
