The first-in-class alkylating HDAC inhibitor EDO-S101 is highly synergistic with proteasome inhibition against multiple myeloma through activation of multiple pathways

EDO-S101 is a first-in-class alkylating, histone deacetylase inhibitor (HDACi) fusion molecule with dual activity. It structurally combines the strong DNA damaging effect of bendamustine with a fully functional pan-HDAC inhibitor, vorinostat (Supplementary Figure S1) that is intended to simultaneously deliver DNAdamaging activity while inhibiting DNA repair activity, and is currently undergoing phase I clinical testing in hematologic malignancies. EDO-S101 has shown a strong preclinical activity in vivo against multiple myeloma (MM), leukemia and B-cell lymphomas in preclinical models with a toxicity profile similar to bendamustine. EDO-S101 has remarkable activity in vivo in the multi-drug resistant Vk12653 transplant model of relapsed/ refractory MM, in which it was the only drug identified with single agent activity. Bendamustine has substantial activity against B-cell malignancies and vorinostat sensitizes the same type of cancers against alkylators or proteasome inhibitors (PI). Cytotoxicity of PI in MM relies on excess induction of proteotoxic stress and triggering of the unfolded protein response (UPR) upon proteasome inhibition, and HDACi synergize with PI by interfering with the α-tubulinmediated transport of polyubiquitinated proteasome substrates to lysosomal destruction. Thus, combining EDO-S101 with PI is expected to deliver mechanism-based, highly synergistic cytotoxicity, based on combining proteasome inhibition with histone deacetylase inhibition and alkylating activity in one molecule. We here focus on the preclinical in vitro exploration of the potential use of EDO-S101 in combination with PI against MM and B-cellderived malignancies. EDO-S101 in combination with PI bortezomib induced the strongest cytotoxic effect compared with vorinostat or bendamustine alone or their combination with bortezomib (Figure 1a). Further, EDO-S101 showed superior cytotoxicity in MM in vitro compared to melphalan, cyclophosphamide or bendamustine (Supplementary Figure S2A), showed alike cytotoxicity compared to vorinostat+bendamustine combination (Supplementary Figure S2B) and induced more effective synergistic cytotoxicity in combination with bortezomib and second generation PI carfilzomib, compared to bendamustine or melphalan (Supplementary Figure S2C). Likewise, EDO-S101 induced robust synergistic cytotoxicity in combination with all approved proteasome-inhibiting drugs, as well as PI in clinical development (Supplementary Figure S3, Supplementary Table S1). This synergy was observed already at a 4 μM drug concentration of EDO-S101 and yielded highly significant combination indices for both bortezomib and carfilzomib in a variety of MM cell lines, cell lines from hematologic (mantle cell lymphoma, ABC-type and GC-type diffuse large B-cell lymphoma, acute myeloid leukemia) and nonhematologic malignancies, as well as primary cells from hematologic cancers (Supplementary Figures S4–S6; Supplementary Table S2). The combination between EDO-S101 and carfilzomib was synergistic to overcome bortezomib-resistance, as shown using the AMO-BTZ model previously described (Supplementary Figure S4A). The antineoplastic activity of HDAC inhibitors is partly mediated through modulation of nuclear histone acetylation, which results in epigenetic changes, and partly through the modulation of protein acetylation in the cytosol, which functionally controls the activities of several key proteins involved in basic cellular functions as well as oncogenesis. Acetylation of α-tubulin controls the transport of polyubiquitinated protein to aggresomal proteolysis and is a major molecular mechanism for the cytotoxic synergy between proteasome inhibitors and HDAC inhibitors. EDO-S101 resulted in superior histone acetylation compared to vorinostat, as revealed by acetylated H3K9, and in particular induced robust acetylation of α-tubulin, also in contrast to vorinostat, suggesting superior inhibition of aggresomal transport of poyubiquitinated protein (Figure 1b). Consistent with this, we observed increased accumulation of polyubiquitinated protein in cells treated with EDO-S101, compared to controls treated with vorinostat or untreated controls. Combined treatment with bortezomib and EDO-S101 led to a maximum increase in cellular polyubiquitinated protein, highlighting the synergy between both pathways. We can exclude a direct effect of EDO-S101 on proteasome activity because EDO-S101 treatment did not change activity-based labeling of active proteasome subunits in treated, viable cells, in contrast to bortezomib (Supplementary Figure S7). The activation status of the UPR, and in particular the activation levels of IRE1/XBP1 determine proteasome inhibitor sensitivity of MM. EDO-S101 caused phosphorylation of IRE1, the key activator of the UPR, already 1 h post treatment (Figure 1c). The combination of EDO-S101 with bortezomib resulted in more effective IRE1 activation than either agent alone, as evidenced for example, by the induction of CHOP expression already after 1 h (Figure 1c and Supplementary Figure S8C). IRE1 activation leads to splicing and activation of the XBP1 transcription factor, which significantly increased 4 h post treatment with EDO-S101 or bortezomib treatment alone and further increased upon treatment with the combination of EDO-S101 with bortezomib (Supplementary Figure S8A). By contrast, vorinostat at the same dose did not induce IRE1/XBP1 activation. In summary, treatment with EDO-S101 and bortezomib resulted in robust activation of the UPR, as evidenced by induction of binding immunoglobulin protein (BIP) and protein disulfide-isomerase (PDI) and triggered the ATF4/CHOP signaling pathway that connects proteotoxic stress with the induction of apoptosis (Figure 1c and Supplementary Figure S8B and D). These effects were significantly stronger after combination treatment, compared to either agent alone. Given the accumulation of polyubiquitinated protein upon simultaneous inhibition of the proteasomal and aggresomal pathways after combination treatment with bortezomib and EDO-S101, we hypothesized that this might activate alternative cellular proteolytic systems, in particular autophagy. Indeed, accumulation of MAP1LC3A/B protein after EDO-S101 treatment, alone and in particular in combination with proteasome inhibition, Citation: Blood Cancer Journal (2017) 7, e589; doi:10.1038/bcj.2017.69