Search

Background: Moderate mitochondrial stress induced by multiple mediators but most notably ROS can lead to activation of persistent mito-protective mechanisms termed "Mitohormesis". As a result of massive protein synthesis, malignant plasma cells (PCs) from MM patients (pts) undergo substantial ER stress but in addition high rates of Ig synthesis contributes to overproduction of ROS. We hypothesized that MM cells exploit mitohormesis to maintain ROS in the hometic zone, thereby increasing mitochondrial fitness to avoid apoptosis. We therefore set out to determine if the processes of mitohormesis are activated in MM and whether unmitigated mitochondrial stress can be exploited as a therapeutic strategy in MM. Results: Protective stress mechanisms of mitohormesis include the activation of the mitochondrial UPR (UPR MT),a mitochondrial-to-nuclear signaling pathway mediated by CHOP and ATF5 that upregulates mitochondrial import proteins, chaperones and proteases to maintain mitochondrial proteastasis. We first demonstrated that UPR MT activation occurs with progression from precursor to overt MM. Using a UPR MTgene signature derived from published gene-sets we observed upregulation of UPR MT genes in single-cell RNA sequencing (scRNA-seq) data generated from PCs derived from Vκ*MYC mice (a transgenic mouse model of MM) spanning the spectrum of the disease. UPR MT gene signature scores in PCs from mice increased with disease progression with the highest levels found in late-MM> int-MM> early MM>wild type mice. Similarly, analysis of publicly available gene expression datasets (GSE6477) that includes normal donors, MGUS and newly diagnosed MM (NDMM) revealed higher expression of UPR MT genes in the majority of NDMM, weak expression in MGUS and absence in normal PCs. To assess the impact of UPR MT expression on pt outcomes we calculated a UPR MT index score derived from the median expression of 12 mtUPR classifier genes across the MMRF CoMMpass dataset of NDMM pts. Stratifying pts by UPR MT expression score we found that pts in the top quartile had a significantly shorter PFS and OS compared to pts with the lowest quartile weighted score. Next, we postulated that perturbation of the mitochondrial import protein, Translocase of the Inner Membrane 23 (TIM23) would exaggerate mitochondrial stress as mitochondrial import efficiency is a key regulator of the UPR MT. First, we demonstrated that TIM23 complex genes are enriched in pts from the CoMMpass dataset with poor risk (1q gain and PR gene signature) and that shorter PFS and OS is associated with a higher weighted score of TIM23 complex genes. We then demonstrated that genetic (shRNA) knockdown or pharmacologic inhibition of TIM23 with MB-10, a small molecule inhibitor of TIM23 induced apoptosis of MM cell lines and primary pt PCs. Further non-transformed cell lines, CD138 - non-MM cells and normal donor hematopoietic progenitor cells were less susceptible to the effects of MB-10. Consistent with activation of the UPR MT, treatment of MM cells resulted in increased cytosolic ATF4, CHOP and a shift of ATF5 to the nuclear fraction. Activation of the CHOP-dependent branch of the UPR MT resulted in in upregulation of mitochondrial-targeted proteins, cpn10 and ClpP. Interestingly, MB-10 also induced XBP1 splicing demonstrating that inhibition of TIM23 complex can simultaneously activate the IRE1/XBP1 branch of integrated stress response (ISR), This led us to hypothesize that targeting TIM23 as an alternative means of activating the ISR could overcome acquired resistance to proteosome inhibitors (PIs). Indeed, PI-resistant and parental isogenic cell lines were equally susceptible to MB-10 as measured by IC50 values of cell growth. Finally, we demonstrated that doxycycline inducible knockdown of TIM23 in a mouse xenograft model induced tumor regression with significantly small tumor volumes at the end of 17 days of doxycycline treatment compared to tumors expressing an inducible control vector. Conclusions: These data demonstrate that mitohormesis and UPR MT activation is associated with MM progression and worse clinical outcomes. Further we show that disrupting mitochondrial protein import results in unmitigated mitochondrial stress that switches the UPR MT from an adaptive cytoprotective to cytotoxic proapoptotic response. Thus, targeting mitochondrial import proteins such as TIM23 may represent novel therapeutic targets for MM. Disclosures Schimmer: Takeda Pharmaceuticals: Consultancy, Research Funding; Medivir AB: Research Funding; Novartis: Consultancy, Honoraria; Jazz: Consultancy, Honoraria; Otsuka Pharmaceuticals: Consultancy, Honoraria; UHN: Patents & Royalties. Trudel: Janssen: Honoraria, Research Funding; GlaxoSmithKline: Consultancy, Honoraria, Research Funding; Amgen: Honoraria, Research Funding; Roche: Consultancy; Sanofi: Honoraria; Pfizer: Honoraria, Research Funding; Genentech: Research Funding; BMS/Celgene: Consultancy, Honoraria, Research Funding.

Recent studies have demonstrated the therapeutic potential of targeting BRD2, BRD3 and BRD4 in hematological malignancies including AML. Here, we report a novel orally bioavailable BRD inhibitor, FV-281, that displays preclinical efficacy in AML in vitro and in vivo. In a binding screen assay of 32 bromodomains (BromoScanTM), FV-281 selectively bound BRD2, BRD3 and BRD4 with Kd values of 2.7 to 7.3 nM and was 2-5 fold more potent than the leading bromodomain inhibitor, OTX015. In AML cell lines (OCI-AML2, Tex, HL-60 and MV4-11), FV-281 reduced cell growth and viability after 72 hour incubation with IC50 values < 50 nM and was 6-fold more potent than OTX015. FV-281 was not cytotoxic to normal hematopoietic cells (n=4) with all IC50 values > 10 µM. In contrast, FV-281 induced cell death as measured by Annexin V/PI staining in 3 of 4 primary AML samples with IC50values < 850 nM. Within 4 hours of incubation with FV-281, c-Myc mRNA and protein level in OCI-AML2 cells were reduced by >95% and >80%, respectively. Following FV-281 washout, c-Myc protein expression returned to baseline within 20 hours with no loss of cell viability observed. Thus, the FV-281 is a reversible bromodomain inhibitor and sustained target inhibition is required to induce cell death. Likewise, after 6 and 48 hours of incubation with OCI-AML2 cells, FV-281 reduced Bcl-2 mRNA and protein level by >60% and >90%, respectively. To assess in vivo efficacy and toxicity, MV4-11 AML cells were engrafted subcutaneously in NOD-SCID mice (n=10). Once tumors were palpable, mice were treated with FV-281 orally (30mg/kg/d x 2 weeks) or vehicle control. FV-281 suppressed tumor growth without evidence of overt toxicity. In addition, primary AML cells were engrafted into the femurs of NOD-SCID mice (n=10). Three weeks after implantation, mice were treated with FV-281 orally (30mg/kg, 5 of 7 days, x 4 weeks) or vehicle control. FV-281 decreased AML engraftment in the injected and non-injected femur without evidence of overt toxicity (% human CD45+/CD19-/CD33+ in non-injected femur treated vs control: 20% vs 60%, p Thus, FV-281 is a novel oral reversible bromodomain inhibitor with significant in vivo activity in murine models of AML. These data support evaluation of this agent in upcoming phase I clinical trials. Disclosures Wang: Fluorinov Pharma Inc.: Employment. Dove:Fluorinov Pharma Inc.: Employment. Hadri:Fluorinov Pharma Inc.: Employment. O'Neill:Fluorinov Pharma Inc.: Employment. Slassi:Fluorinov Pharma Inc.: Employment.

Proteasome inhibitors (PIs), such as bortezomib and carfilzomib, are often associated with serious toxicities, poor pharmacokinetics (PK), and the inconvenience of intravenous administration. Therefore, there is a need for safer and more effective orally active PIs. Applying a novel fluorine-based medicinal chemistry technology, we designed, synthesized and tested novel proprietary analogs of epoxyketone-based PIs. Our initial screening identified a lead orally bioavailable PI, FV-162, which had superior potency in proteasome activity inhibition, cytotoxicity in cultured human multiple myeloma (MM) cells and metabolic stability in mouse liver microsomes, compared to an oral PI currently under clinical evaluation, ONX-0912. FV-162 inhibited the β subunits of purified proteasomes of T. acidophilum and specifically inhibited the β5 subunit of the proteasome in 10 human myeloma cell lines with an IC50 range of 12-58 nM. In comparison, IC50 values for the β1 and β2 subunits were consistently greater than 10 µM. FV-162 also disrupted the ubiquitin-proteasome pathway in the LP-1 human myeloma cell line in a time- and dose-dependent manner, inducing intracellular accumulation of ubiquitylated proteins. FV-162 was cytotoxic to 10 human MM cell lines with an IC50 range of 16-611 nM. Furthermore, in 4 MM patient samples tested, at least 80% of primary CD138+ myeloma cells were preferentially killed by FV-162 at concentrations less than or equal to 0.5 µM after 24 hours of incubation, while more than 70% of CD138- normal hematopoietic cells remained viable. FV-162 displayed permeability in murine gastrointestinal epithelial cells, consistent with excellent oral absorption. In addition, cytochrome P450 isozyme assays in pooled human liver microsomes using up to 3 μM FV-162 suggested minimal risk for drug-drug interaction. The toxicity of epoxyketone-based PIs, such as carfilzomib, appears to be related to the Cmax, while efficacy is more dependent on the AUC. We hypothesized that the novel fluorine chemistry of FV-162 would enhance the stability of the drug and thereby improve its PK profile and therapeutic index. Therefore, we measured the oral PK of FV-162 and ONX-0912 in NOD/SCID mice (25 mg/kg). FV-162 reached a peak plasma concentration (Cmax) of 21 ± 5 ng/mL (mean ± SD) after 1 h (Tmax), while ONX-0912 reached a Cmax of 38 ± 46 ng/mL after a Tmax of 5 min. Despite the lower Cmax, the AUC of FV-162 was 6-fold higher compared to ONX-0912 (3117 vs. 513 min*ng/mL). In addition, the oral bioavailability and half-life of FV-162 were 3- and 4.3-fold higher than ONX-0912, respectively. Attenuated Cmax and improved AUC may improve the tolerability profile of FV-162 compared to ONX-0912 and carfilzomib. Supporting this contention, the maximum tolerated dose of ONX-0912 in NOD/SCID mice was 30 mg/kg, while mice tolerated 200 mg/kg FV-162 for 28 days without significant toxicity. Next, we evaluated FV-162 activity in mouse models. Inhibition of β5 subunit activity reached 80% in murine red blood cells within 1 hour of oral administration of FV-162 to NOD/SCID mice (30-100 mg/kg), which was sustained for at least 24 hours. Oral dosing of 100 mg/kg FV-162 achieved similar β5 inhibition as 30 mg/kg ONX-0912. Finally, continuous daily oral administration of FV-162 at 50 or 100 mg/kg significantly inhibited tumor growth in a myeloma MM.1S mouse xenograft model without toxicity and was as effective as intravenous carfilzomib or ONX-0912 administered at their maximal doses. Taken together, our preclinical data suggest that FV-162 has superior PK, an improved safety profile, and high oral potency, identifying it as a promising orally bioavailable PI for myeloma therapy. Further clinical investigation of this novel PI in MM is therefore warranted. Disclosures: Wang: Fluorinov Pharma Inc.: partial salary/fellowship support Other. Dove:Fluorinov Pharma Inc.: Employment. Climie:Fluorinov Pharma Inc.: Employment. O'Neill:Fluorinov Pharma Inc.: Employment. Slassi:Fluorinov Pharma Inc.: Employment.