Exploiting Protein Translation Dependence in Multiple Myeloma with Omacetaxine-Based Therapy

Multiple myeloma is an aggressive hematologic malignancy characterized by overproliferation and tissue invasion by malignant plasma cells that retain the fundamental biologic attributes of antibody production and secretion. Multiple myeloma afflicts more than 30,000 Americans with increasing prevalence (1). In the early 21st century, with the introduction of proteasome inhibitors (PI), immunomodulatory drugs (IMiD), and mAbs, the average life expectancy of patients with multiple myeloma has drastically improved (2). Beginning in 2015, the anti-CD38 mAb, daratumumab, emerged as another important treatment option (3). However, multiple myeloma remains incurable, and drug resistance to all of the above agents is referred to as “penta-refractory,” with a median survival of approximately 9 months (4). Throughout the disease course, patients are debilitated by bone tumors, pathologic fractures, kidney failure, and immunosuppression, eventually becoming fatal. Thus, new therapies that retain their activity in patients with advanced multiple myeloma who have developed IMiD, PI, and daratumumab resistance are urgently needed.

We are focused on the development of new drugs for penta-refractory multiple myeloma, and became interested in omacetaxine as a drug with therapeutic potential. Omacetaxine is FDA-approved for chronic myeloid leukemia (CML) and acts via a unique mechanism among anticancer drugs by binding to the A-site cleft of ribosomes and blocking the initial elongation step of protein synthesis (5). In CML, translation inhibition may act by downregulating key metastable oncoproteins, such as the antiapoptotic protein, MCL1, and the oncogenic transcription factor, c-MYC (6–8). Both MCL1 and c-MYC are known to be important prosurvival proteins frequently overexpressed and activated in multiple myeloma (9, 10). Accordingly, most multiple myeloma cell lines are dependent on MCL1 and c-MYC (11, 12). Notably, genomic abnormalities involving the c-MYC locus occur in a subset of patients with multiple myeloma and upregulate its expression (13–15). In addition, genomic amplification of the MCL1 gene, which resides on chromosome 1q, occurs frequently in multiple myeloma, including approximately 80% of patients with relapsed multiple myeloma (16–18). Thus, MCL1 and c-MYC are attractive targets in multiple myeloma.

Considering the above evidence, its stands to reason that omacetaxine may be effective in targeting multiple myeloma. Work from multiple reports has demonstrated that omacetaxine has potent cytotoxicity against multiple myeloma cell lines in vitro and in vivo (19–22). Further supporting this approach, inhibiting the initiation step of protein translation by knocking down eIF4E hinders multiple myeloma growth and survival (23, 24). In addition, Manier and colleagues demonstrated high rates of protein translation related to c-MYC activity in multiple myeloma and identified the translational initiation inhibiting rocaglates to be detrimental to multiple myeloma survival (25). Multiple myeloma cells are generally sensitive to agents that disrupt their fragile state of proteostasis, including PIs which have a broad effect on protein degradation, and IMiDs, which directly cause the degradation of Ikaros proteins, leading to decreased levels of c-MYC and IRF4 and downregulation of their downstream genetic programs (26–28). Protein translation inhibition is also known to decrease the levels of short half-life proteins, such as c-MYC, MCL1, and other prosurvival proteins in multiple myeloma (20, 21, 25). Thus, multiple groups have supported that protein translation inhibition has the potential to be an effective strategy to target multiple myeloma cells.

Here, we report further preclinical assessments of omacetaxine with a focus on primary samples from patients with multiple myeloma. We first analyzed the antimyeloma activity of omacetaxine and the multiple myeloma cell protein translation rates in samples from a cohort of patients with multiple myeloma ranging from newly diagnosed to having advanced disease. In doing so, we found that high rates of protein translation in multiple myeloma cells served as a biomarker for potent antimyeloma cytotoxicity of omacetaxine treatment. In seeking drug combination partners, we establish the IMiDs, lenalidomide and pomalidomide, were synergistic with omacetaxine in vitro, in vivo, and ex vivo. In evaluating the mechanism for that synergy, we found a double-hit on the key multiple myeloma cell survival factors, IRF4 and c-MYC. Overall, omacetaxine exhibits potent and selective antimyeloma activity that is retained in PI/IMiD refractory disease, and in combination with IMiDs represents a synergistic regimen to test in patients.

Omacetaxine mepesuccinate/homoharringtonine (referred to herein as omacetaxine or Oma) and (+)-JQ1 were purchased from Selleckchem. The PIs, IMiDs, and dexamethasone were purchased from Thermo Fisher Scientific. Four-hydroperoxy cyclophosphamide was purchased from Santa Cruz Biotechnology and Dara was obtained from the University of Colorado Health Pharmacy (Aurora, CO).

Multiple myeloma cell lines, RPMI-8226, U266, NCI-H929, OPM-2, AMO-1, MM.1S, and MM.1R, were obtained from the ATCC, and validated via short tandem repeat polymorphism profile match analysis. Cell lines were cultured at 37°C with 5% CO₂ in RPMI1640 medium supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin (Thermo Fisher Scientific).

Cell proliferation was determined using the CellTiter-Glo Luminescent Cell Proliferation Assay (Promega) per the manufacturer's instructions following drug incubations in 96-well culture plates at 37°C for 96 hours. Results were obtained using a Vmax Kinetic Microplate Reader (Molecular Devices). Cell viability was also measured via flow cytometry using the LIVE/DEAD Fixable Near-IR Dead Cell Stain (Invitrogen).

Drug combination studies were set-up using a full matrix design with myeloma cell lines or primary samples. Synergy was calculated using ZIP delta score via the SynergyFinder online tool (https://synergyfinder.fimm.fi; ref. 29).

Bone marrow aspirates from patients with multiple myeloma or smoldering myeloma were obtained and mononuclear cells (MNC) were isolated from the samples using SepMate Ficoll-Plaque Tubes (Stemcell Technologies) according to the manufacturer's instructions. CD138-positive cell selection was not performed for any of the experiments. Samples were cryopreserved in freezing medium consisting of Iscove's modified Dulbecco's medium, 45% FBS, and 10% DMSO at 10 × 10⁶ cells/mL.

Patient multiple myeloma cell drug sensitivity testing was performed with our myeloma-drug sensitivity testing (My-DST) platform as described previously (30). In brief, deidentified primary myeloma MNCs were cultured in RPMI1640 medium containing l-glutamine with 10% FBS and 100 U/mL penicillin, 100 μg/mL streptomycin (Thermo Fisher Scientific), and 2 ng/mL IL6 (PeproTech) at 37°C. Cells were then transferred to 96-well plates at 4.5 × 10⁵ cells/mL (90,000 MNCs/well) and treated for 48 hours.

After 48-hour treatment, cells were washed in 1 × cold DPBS and resuspended in BD Brilliant Stain Buffer (BD Biosciences) in a 96-well V-bottom plate. Cells were treated with FcR Blocking Reagent (Miltenyi Biotec) for 5 minutes and then surface stained with anti-CD19-BV786 (SJ25C1), anti-CD45-BV510 (HI30), multi-epitope anti-CD38-FITC (American Laboratory Products Company), anti-CD138-BV421 (MI15), and anti-CD319-Alexa647 (235614) or anti-CD46-Alexa647 (E4.3) antibodies for 10 minutes on ice. Intracellular staining with anti-kappa-BV605 and anti-lambda-PE light chains was performed after paraformaldehyde fixation and permeabilization. All flow cytometry antibodies were purchased from BD Biosciences, except where noted. After staining, samples were washed and resuspended with 100 μL DPBS containing 2% FBS (FACS buffer). Viability was determined with LIVE/DEAD Fixable Near-IR Dead Cell Stain Kit (Invitrogen) at 1:1,000. Flow cytometry data were collected using a BD FACSCelesta Multicolor Cell Analyzer (BD Biosciences) equipped with a high-throughput sampler. Analysis was completed using FlowJo Software (FlowJo LLC). Myeloma populations were analyzed by gating for CD19, CD45, CD38, and CD138, based on the cells expressing clonal light chain restriction. Intracellular flow was used to measure MCL1 (22/Mcl-1 Antibody, BD Biosciences), c-MYC (Y69 antibody Alexa-488, Abcam), IRF4 (3E4 antibody, eFluor710, Invitrogen), and BCL2 (BMS1028 antibody, Thermo Fisher Scientific) expression. Intracellular staining of MCL1 and BCL2 was conducted following permeabilization with the Cytofix/Cytoperm Kit (BD Biosciences).

Protein synthesis levels were determined in vitro utilizing the puromycin analogue, O-propargyl-puromycin (OP-puro), in the Protein Synthesis Assay Kit (Cayman Chemical) according to the manufacturer's instructions. For in vivo protein synthesis, engrafted NSG mice were intraperitoneally injected with 500 μg Puromycin (Gibco), then euthanized after 1.5 hours, marrow was flushed, and cells were stained with anti-puromycin antibody (12D10, Alexa-488, Millipore Sigma), followed by flow cytometry.

To evaluate BCL2 family member dependence in myeloma cell lines and patient samples, we adapted the published assay for BH3 profiling (31). Following surface staining, cells were exposed to the various BH3 mimetic peptides to test for BH3 priming, including BIM, BAD, and NOXA. Mitochondria were permeabilized with digitonin, and cells were fixed with paraformaldehyde. BH3 profile was determined using flow cytometry and staining for cytochrome-c loss.

For in vivo assessment of omacetaxine, 5 × 10⁵ MM.1S cells stably expressing firefly luciferase were injected intravenously into NSG mice (The Jackson Laboratories) to generate an orthometastatic multiple myeloma xenograft model, as done previously (32). When untreated, the NSG mice succumbed to orthometastatic myeloma–like disease approximately 50 days after implantation in this model. Bioluminescence imaging (BLI) was used to monitor graft status weekly and mice were stratified into treatment groups to normalize BLI at start of study. Treatment was administered by intraperitoneal injection in a final volume 0.5 mL of PBS with 0.5% FBS. Tumor status was assessed by BLI, and the results were analyzed by Living Image (PerkinElmer).

Multiple myeloma cell lines were incubated with pomalidomide for 24 hours, with omacetaxine for 4 hours, or in staggered combination. The cells were washed and resuspended in Agilent Seahorse XF DMEM Medium, pH 7.4 (Agilent Technologies). Glycolytic flux and the extracellular acidification rate (ECAR) were measured on a Seahorse XFe96 analyzer using the Seahorse XF glycolysis stress test kit according to the manufacturer's instructions (Agilent Technologies).

Cells were lysed with RIPA Buffer (Thermo Fisher Scientific) and digested according to the filter aided sample preparation protocol (33). Recovered peptides were dried, desalted, and concentrated on Thermo Scientific Pierce C18 Tips. Cell lysates were analyzed using an Orbitrap Fusion mass spectrometer connected to an EASY-nLC 1200 System (Thermo Fisher Scientific) with a nanoelectrospray ion source. Data were acquired using the Xcalibur (version 4.1) software. MS-MS spectra were extracted, and Proteome Discoverer Software (ver. 2.1.0.62) was used to convert the raw data into mgf files. The files were then searched against a human database using an in-house Mascot Server (version 2.6, Matrix Science). Scaffold (version 4.8, Proteome Software) was used to validate MS-MS–based peptide and protein identifications. Protein identifications were established using >99% probability cutoff and contained at least two identified unique peptides. Heatmaps, principal component analysis (PCA), and volcano plots were generated using MetaboAnalyst Software (McGill University, Montreal, Quebec, Canada) and R Studio. Western blots to measure specific proteins were performed with anti-c-MYC (9402 Antibody, Cell Signaling Technology), anti-IRF4 (IRF4.3E4, BioLegend), anti-IKZF1 (E-2, Santa Cruz Biotechnology), and anti-IKZF3 (NBP2-24495, Novus Biologicals), with anti-ACTB (C4, Santa Cruz Biotechnology) as a loading control. For Western blotting, cells were lysed with RIPA Buffer (Thermo Fisher Scientific). Protein concentrations were determined with the DC Protein Assay (Bio-Rad). Equal amounts of protein were loaded and resolved by SDS-PAGE and transferred to polyvinylidene difluoride membranes for blotting. Antibodies were incubated overnight at 4°C in 5% BSA TBS (Boston BioProducts). Membranes were incubated with secondary antibody for 2 hours.

Statistics and figures were generated using Prism 6 Software (GraphPad Software). All data are presented as mean and SD. EC₅₀ values were determined by nonlinear regression variable slope inhibitor versus response curves. Two-tailed Student t test was used for comparing two means. When comparing more than two means, ANOVA was used with Tukey correction. Survival analyses were conducted using SAS version 9.4 (SAS Institute) with Cox proportional hazard models to calculate HRs. Levels of statistical significance are shown by: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Bone marrow aspirates were collected from patients at the University of Colorado Blood Cancer and Bone Marrow Transplant Program (Aurora, CO) after obtaining written informed consent and in accordance with the Declaration of Helsinki. Samples from patients with multiple myeloma or smoldering myeloma were obtained from the hematologic malignancies tissue bank with protocol approval from the Colorado Multiple Institutional Review Board (Aurora, CO). All animal studies were conducted in compliance with protocols reviewed and approved by the University of Colorado Institutional Animal Care and Use Committee (Aurora, CO).

We tested omacetaxine in multiple myeloma cell lines in vitro and in bone marrow aspirates from patient with multiple myeloma ex vivo. First, we verified the reported cytotoxicity of omacetaxine in cell culture using U266, H929, MM.1S, MM.1R, and RPMI-8226 multiple myeloma cell lines (20, 21). Omacetaxine inhibited cell proliferation with an EC₅₀ range of 6–28 nmol/L after a 96-hour incubation (Fig. 1A). In a time-course study, omacetaxine-mediated induction of apoptosis started at 2 hours, and cell death began after 24 hours (Fig. 1B and C; Supplementary Fig. S1A). As further evidence of its rapid and drastic effects, omacetaxine treatment reduced glycolytic and oxidative phosphorylation metabolism in multiple myeloma cell lines after 4 hours, as measured using the Seahorse assay (Fig. 1D; Supplementary Fig. S1B and S1C). Thus, omacetaxine rapidly reduced metabolism, and induced apoptosis and cell death in all multiple myeloma cell lines tested.

To evaluate the extent to which the in vitro activity of omacetaxine in multiple myeloma cell lines may be clinically significant, we evaluated omacetaxine activity in primary samples from 6 different patients with multiple myeloma. The effect of omacetaxine in primary samples was tested by incubating MNCs from patient bone marrow aspirates (n = 6) for 48 hours and measuring the number of surviving multiple myeloma cells by multiparameter flow cytometry, as described previously (30). With this assay, we found that omacetaxine specifically decreases multiple myeloma cell viability ex vivo with an EC₅₀ range of 25–225 nmol/L (Fig. 1E and F). Using a single concentration to screen additional primary samples, we found that ex vivo treatment with 50 nmol/L omacetaxine for 48 hours resulted in a >50% reduction in viable multiple myeloma cells in 39 of 50 (78%) patient samples (Fig. 1G). Importantly, patients with relapsed/refractory myeloma resistance to PIs and IMiDs did not show cross-resistance to omacetaxine, which retained highly potent and specific antimyeloma activity in single- and double-class refractory patients (Fig. 1G). On the other hand, sensitivity to omacetaxine among these samples did not correlate with other patient characteristics, including the level of disease involvement in the bone marrow, cytogenetics, and prior lines of therapy (Supplementary Fig. S2A–S2D; Supplementary Table S1). In regards to the effects on normal cells, we have previously found that omacetaxine does not affect the viability of normal hematopoietic stem cells at concentrations tested up to 100 nmol/L (34). Overall, these findings show that omacetaxine has antimyeloma activity in samples from a broad range of patients with multiple myeloma, importantly including those with multi-drug–resistant disease, at doses that do not affect normal hematopoietic stem cells.

Because the antimyeloma effect of omacetaxine was present in a broad range of patient samples, we hypothesized that myeloma cells may require inherently high levels of protein synthesis. We used the OP-puro assay to measure the rate of amino acid incorporation into translating ribosomes by flow cytometry. Primary multiple myeloma cells exhibited significantly higher levels of baseline protein translation compared with the other nonmalignant bone marrow MNCs, with a mean 3.9-fold increased OP-puro signal (n = 17; P ≤ 0.001; Fig. 2A and B). Omacetaxine treatment inhibited protein translation with an EC₅₀ of 9–12 nmol/L and within 30 minutes of dosing, preceding the induction of apoptosis (Fig. 2C; Supplementary Fig. S3A). Omacetaxine inhibited translation across myeloma cell lines, while bortezomib treatment did not (Supplementary Fig. S3B and S3C). In 14 primary samples tested, translation was significantly reduced after 2.5 hours of 50 nmol/L omacetaxine treatment to near normal MNC levels (Fig. 2D). Strikingly, using a cutoff of the bottom quartile of baseline multiple myeloma translation compared with MNCs (2.5-fold or higher), “high translation” was associated with sensitivity to omacetaxine, and “low translation” was associated with relative resistance to omacetaxine (P = 0.0018; Fig. 2E). Myeloma cell protein translation levels were not associated with the patient's sensitivity to PIs or IMiDs (Supplementary Fig. S3D). The high level of translation was also observed in bone marrow plasma cells from healthy donors, and omacetaxine was also cytotoxic to these normal plasma cells (Supplementary Fig. S3E and S3F). Thus, protein translation inhibition was consistent across myeloma cells at concentrations that were also cytotoxic, and the translation rates may serve as a biomarker for response.

By blocking protein translation, omacetaxine is known to downregulate the expression of short-lived oncoproteins, such as MCL1, in multiple cell types (6–8, 20). Thus, we investigated omacetaxine-induced downregulation of proteins known to be important for multiple myeloma cell survival, such as MCL1 and c-MYC. Indeed, by quantitative flow cytometry, we confirmed that MCL1 is overexpressed in multiple myeloma cells and that omacetaxine rapidly downregulates MCL1, but not BCL2, expression in myeloma cells after 2.5-hour incubation (Fig. 2F and G; Supplementary Fig. S4A and S4B). In comparison, the MCL1 inhibitor, S63, was less cytotoxic than omacetaxine (ref. 35; Supplementary Fig. S4C). To determine whether MCL1 dependence may predict ex vivo response to omacetaxine, we established the BH3 profiling for BCL2 family member priming, as demonstrated in Supplementary Fig. S4D, with H929 cells known to be MCL1 primed (11, 31). In a subset of our patient sample cohort, we found that omacetaxine exhibited potent cytotoxicity against both multiple myeloma cells that were primed and multiple myeloma cells that were not primed for MCL1-mediated apoptosis (Supplementary Fig. S4E). Whereas S63 cytotoxicity correlated well with MCL1 priming, omacetaxine demonstrated potent antimyeloma activity in samples that were not MCL1 primed (Supplementary Fig. S4F and S4G). In addition to MCL1, we also demonstrated that omacetaxine rapidly downregulates c-MYC after 4 hours of treatment in MM.1S cells (Fig. 2H). Thus, downregulation of MCL1, although important in myeloma cell survival, did not appear to fully explain the spectrum of omacetaxine activity against the disease.

The benefits of combination therapy in multiple myeloma have been consistently demonstrated in clinical trials and in practice since the early 2000s. As a result, we sought to identify the best clinically available agents to combine with omacetaxine. Using multiple myeloma cell lines, Amo-1, L363, and U266, we combined omacetaxine with various antimyeloma agents in two-drug combination matrices and measured viability after a 96-hour incubation. Omacetaxine combined with lenalidomide or with pomalidomide stood out as synergistic in all cell lines tested (Fig. 3A and C). The δ-scores for synergy of omacetaxine with lenalidomide and pomalidomide were 13.9 and 11.2, respectively, with a broad range of concentrations showing a productive interaction (Fig. 3B and D). Notably, in an IMiD-resistant relapsed patient sample, the combination of omacetaxine and pomalidomide was even more synergistic and resensitized the multiple myeloma cells to the IMiD (δ-score, 22.7; Fig. 3E and F). In contrast, omacetaxine combined with the PI, bortezomib, showed a lack of synergy when screened in the same manner (Supplementary Fig. S5A and S5B). Omacetaxine combined with dexamethasone was synergistic in only one of three multiple myeloma cell lines tested (Supplementary Fig. S5C and S5D). Thus, when combined with omacetaxine, IMiDs displayed unique and consistent synergy that was not observed with other antimyeloma drugs tested.

To extend our findings to an in vivo model, we evaluated omacetaxine using a firefly luciferase–expressing myeloma cell line, MM.1S (luc-MM.1S), to generate xenografts in NSG mice. MM.1S was chosen for its similarity to primary multiple myeloma compared with other available cell lines (36). First, we sought to study the pharmacodynamics of omacetaxine in this model. Mice were injected with luc-MM.1S cells and allowed to establish disease, followed by treatment with 1, 2, or 3 mg/kg omacetaxine by intraperitoneal injection, followed by puromycin injection 1 hour later, and sacrifice 1.5 hours later, with bone marrow harvest and OP-puro incorporation by flow cytometry (Fig. 4A). In this study, we identified a dose-dependent decrease in the protein translation rate in the luc-MM.1S cells, with the greatest effect occurring at 3 mg/kg (Fig. 4B). These results were used to design a study to determine the benefit of combining omacetaxine with pomalidomide treatment in vivo.

We next tested combination treatment with omacetaxine and pomalidomide using the in vivo model with treatment starting when disseminated disease was detectable. To evaluate the benefit of the drug combination, lower dose levels of omacetaxine (1 mg/kg) and pomalidomide (8 mg/kg) treatment were used in the in vivo experiment. Four study arms were used to initiate the treatments, consisting of vehicle control, omacetaxine monotherapy, pomalidomide monotherapy, and omacetaxine/pomalidomide combination therapy intraperitoneally daily Monday–Friday (Fig. 4C). On the basis of BLI monitoring, the disease development occurred more slowly in the omacetaxine/pomalidomide combination therapy arm (Fig. 4D; Supplementary Fig. S6A and S6B). Mouse survival was also most significantly extended in the combination therapy arm compared with the vehicle control arm (Fig. 4E and F).

As both omacetaxine and pomalidomide directly affect protein levels, with omacetaxine acting by blocking protein production and pomalidomide acting through selective protein degradation, we examined the effects of the combination treatment on the proteome. We started with whole-cell proteomics of multiple myeloma cells comparing the single-agent treatments with the combination treatment at a timepoint before substantial apoptosis or cell death began. Although omacetaxine is rapidly cytotoxic to multiple myeloma cells, the cytotoxicity of pomalidomide occurs later, with apoptosis and cell death beginning at 24 hours (Supplementary Fig 7A). In addition, the glycolysis was reduced most drastically in multiple myeloma cell lines after combination treatment (Supplementary Fig. S7B–S7D). For optimal detection of protein changes by these drugs in proteomic analyses, MM.1S cells were treated in triplicate, with omacetaxine for 2.5 hours, with pomalidomide for 24 hours, or with the staggered combination treatment and compared with vehicle controls. Proteomic analyses identified 2,015 proteins, of which 30 were significantly depleted (P < 0.05; fold change decrease > 2) by omacetaxine, 25 by pomalidomide, and 53 by the combination (Supplementary Tables S2–S4).

We next examined the proteins that were most differentially changed in the treated cells compared with the vehicle control cells. By unsupervised clustering and PCA, the treated cells separated differentially into three groups that were distinct from the vehicle control group and from each other (Supplementary Fig. S8A and S8B). The top 25 most differentially affected proteins are shown by heatmap in Fig. 5A. IGL1 and JCHAIN, which are components of the IgA protein produced by the MM.1S cells, were among the proteins that were most differentially downregulated by omacetaxine alone (Fig. 5B and C). IRF4 was among the proteins that were most differentially downregulated by the combination treatment (Fig. 5D; Supplementary Fig. S8C–S8E). IRF4 stood out because it is critical to multiple myeloma cell survival and a known downstream target of IMiD-mediated Ikaros degradation (28, 37, 38). Because we used a staggered treatment schedule for proteomics in which pomalidomide treatment was administered for 24 hours and omacetaxine was added during the last 2.5 hours, we repeated the cytotoxicity measurement and found that the combinations remained synergistic (Supplementary Fig. S8F). Overall, the stepwise decrease in IRF4 between the single-agent and combination treatments supports the idea that the combination may act as a double hit on the levels of this protein.

In multiple myeloma, IRF4 and c-MYC form a positive feedback loop that propels malignant survival and proliferation (10, 37). Although c-MYC was not detected, it may have been below the resolution of semiquantitative proteomics, which can miss short half-life proteins. However, FUBP1, a transcription factor known to upregulate c-MYC (39), was among the most significantly downregulated proteins by the combination treatment (Fig. 5A; Supplementary Fig. S8G). Thus, we envisioned a model in which omacetaxine and IMiDs synergize to cause an even greater reduction in IRF4 and c-MYC levels (Fig. 6A). To test this model, we incubated MM.1S cells with omacetaxine and pomalidomide alone and in staggered combination, and observed a greater loss of c-MYC and IRF4 compared with single agents, as measured by immunoblot and intracellular flow cytometry (Fig. 6B and C). To further test the model, we evaluated whether omacetaxine and pomalidomide would be synergistic with JQ1, a BET inhibitor, which downregulates c-MYC transcription (40). JQ1 was synergistic with both omacetaxine and pomalidomide in multiple myeloma cell lines (δ-scores, 17.56 and 21.18, respectively; Fig. 6D and E). In summary, omacetaxine and IMiDs are synergistic in multiple myeloma cells and cooperate to elicit a more substantial downregulation of the IRF4/c-MYC pathway than either drug alone, creating an attractive and clinically testable regimen for patients with relapsed/refractory multiple myeloma.

Despite improvements in therapies over the past 2 decades, multiple myeloma remains an incurable blood cancer. Nearly all patients with multiple myeloma experience relapse and eventual resistance to current therapies. The relapsed and refractory multiple myeloma setting continues to require new therapeutic options, particularly therapies with unique mechanisms of action that are distinct from those of existing therapies. One way this could be accomplished is through development of new small-molecule classes that exploit cellular processes (e.g., protein synthesis) and/or molecular pathways (e.g., IRF4 and c-MYC) that are uniquely upregulated in malignant plasma cells.

In oncology, omacetaxine has a unique mechanism of action, in which it binds to the A-site cleft of ribosomes and inhibits protein biosynthesis. In the context of multiple myeloma, several independent groups have now shown that the inhibition of protein translation has selective and potent cytotoxic effects (19–21, 24, 25). This approach may be tumoricidal through diminution of short-lived oncoproteins, such as MCL1 and c-MYC (23–25, 41). Here, we extend the previous preclinical studies showing omacetaxine activity in multiple myeloma cell lines by documenting broad efficacy in primary samples across the disease spectrum and describing clinically applicable drug combination with IMiDs. Multiple myeloma is a particularly attractive disease for development of this strategy in light of the significant unmet medical need in the growing patient population that is refractory to currently available drugs. Together with the previous reports, our study further supports clinical testing of this approach.

We first became interested in omacetaxine as a potential means to exploit the MCL1 dependence that has emerged as underlying multiple myeloma pathogenesis. We have developed a high-throughput, flow cytometry–based approach termed My-DST to test primary samples (30). With My-DST, omacetaxine treatment showed specific, dose-dependent cytotoxic effects on patient multiple myeloma cells, with no appreciable effect on the normal bone marrow cells. In a substantial sized cohort that included 51 multiple myeloma patient samples, treatment with 50 nmol/L omacetaxine showed antimyeloma activity in the majority of samples and was independent of resistance to IMiDs or PIs. This ex vivo dose of 50 nmol/L (27.3 ng/mL) is within the achievable human plasma levels of 36.2 ng/mL (42). Surprisingly, we found that both MCL1-dependent and -independent samples were susceptible to omacetaxine. To help understand this broad activity, we examined protein translation and found almost ubiquitously higher levels of protein translation in multiple myeloma cells compared with background MNCs. On the basis of these results using primary samples, our findings show that omacetaxine retains its efficacy in PI/IMiD refractory multiple myeloma, a finding that cannot be gleaned from cell line data.

In the current clinical approach to multiple myeloma, combination therapies are utilized almost exclusively rather than single agents (43). The benefit of this combinatorial approach is likely still due to the independent actions of the drugs, rather than mechanistic synergy in most cases (44). Although true mechanistic drug synergism may be rare in multiple myeloma, it may be very valuable in combating drug resistance. In our evaluation of combination treatments using clinically available agents together with omacetaxine, the IMiD-based combinations stood out as synergetic. On the basis of our proteomics and targeted protein measurements, we found that this synergy made mechanistic sense, because the crucial multiple myeloma survival factors, IRF4 and c-MYC, were most substantially downregulated after the combination treatment. Targeting c-MYC in multiple myeloma via translation inhibition is also independently supported by the previous study by Manier and colleagues (25).

Although the importance of c-MYC and MCL1 for multiple myeloma survival is well-supported, approaches that target these oncoproteins have yet to demonstrate clinical activity. Targeted MCL1 inhibitors have recently emerged and are being tested in phase I clinical trials as single agents (NCT03218683), but one is currently on FDA-placed hold for possible cardiac toxicity (NCT03465540; refs. 35, 45). Pharmacologic targeting of c-MYC has been challenging, but recently BET inhibition was supported as one possible approach (40). A phase I study of BET inhibition in patients with multiple myeloma is also underway (NCT03068351). Whether MCL1 and BET inhibitors will be safe and effective is not yet clear, whereas omacetaxine is FDA approved and tolerable in humans. In this study, we found that MCL1 inhibition had less antimyeloma activity than omacetaxine treatment. For protein translation inhibition, other studies have been published supporting its potential as a new avenue for multiple myeloma treatment (23–25), but no clinical trials have yet opened using this approach. Thus, omacetaxine may have the advantages of being multi-targeted and readily translated into patient testing.

On the basis of the broad preclinical activity of omacetaxine in multiple myeloma, as well as its strong and consistent synergy with IMiDs, a dedicated clinical trial of this approach in patients with multiple myeloma should be conducted. In the clinical trials, the measurement of the protein translation rate in multiple myeloma cells could serve as a biomarker for those patients most likely to respond. On the basis of the prior clinical experience in patients with CML, omacetaxine has favorable bioavailability in humans, with a suitable side effect profile (46). Primary adverse events observed included hematologic events, chiefly thrombocytopenia, grade 1–2 infections, and gastrointestinal side effects (47). In addition, two case reports of patients with multiple myeloma treated with omacetaxine-based combination reported responses (48). Thus, omacetaxine is known to be bioavailable and safe in patients, suggesting that administration will be straight forward in multiple myeloma and may be tolerable in drug combinations as well. To achieve this, we have designed a phase I clinical trial to evaluate omacetaxine as a single agent and in combination with pomalidomide for patients with relapsed/refractory multiple myeloma.

Z.J. Walker reports a patent for U.S. Provisional Patent Application 62/945,204 pending to University of Colorado. B.M. Idler reports a patent for U.S. Provisional Patent Application 62/945,204 pending to University of Colorado. T.M. Mark reports personal fees from Adaptive Inc., Takeda Inc., Genzyme Inc., Amgen Inc., and Karyopharm Inc. outside the submitted work. D.W. Sherbenou reports grants from NCI and Cancer League of Colorado during the conduct of the study, grants from CCTSI CO-Pilot Award, and NCCN Foundation and other from Oncopeptides AB, Fortis Therapeutics, Inc, and Virtuoso Therapeutics outside the submitted work, and has a patent for U.S. Provisional Patent Application 62/945,204 pending to University of Colorado, U.S. Patent Application No. 62/421,113 pending to University of California San Francisco, and U.S. Patent Application No. 62/774,177 pending to University of Colorado. No disclosures were reported by the other authors.

Z.J. Walker: Conceptualization, data curation, investigation, methodology, writing-original draft, writing-review and editing. B.M. Idler: Data curation, methodology, writing-review and editing. L.N. Davis: Data curation, methodology, writing-review and editing. B.M. Stevens: Conceptualization, data curation, investigation, writing-review and editing. M.J. VanWyngarden: Data curation, methodology, writing-review and editing. D. Ohlstrom: Data curation, formal analysis, writing-review and editing. S.C. Bearrows: Data curation, methodology, writing-review and editing. A. Hammes: Formal analysis, writing-review and editing. C.A. Smith: Resources, writing-review and editing. C.T. Jordan: Resources, writing-review and editing. T.M. Mark: Resources, writing-review and editing. P.A. Forsberg: Resources, writing-review and editing. D.W. Sherbenou: Conceptualization, resources, data curation, supervision, funding acquisition, investigation, methodology, writing-original draft, project administration, writing-review and editing.

The authors dedicate this article to the memory of David L. Kessenich, whose support was instrumental in the implementation of this project. This work was also supported by grants from the Cancer League of Colorado (2019 Cancer Research Grant to D.W. Sherbenou), the National Comprehensive Cancer Network Foundation (2016 Young Investigator Award to D.W. Sherbenou), and the NCI (K08CA222704 to D.W. Sherbenou). The authors thank the Hematology Clinical Trials Unit at the University of Colorado for tissue bank and regulatory support. They also thank Courtney Jones and Taylor Mills for instruction on conducting and interpreting Seahorse assays and interpretation, and Eric Pietras and Neelanjan Mukerjee for helpful discussions. The authors acknowledge Dr. Heidi Chial (BioMed Bridge, LLC) for scientific editing of this article.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.