INTRODUCTION: Pelareorep is the infusible form of human reovirus (RV). Our single-agent phase 1 RV trial in relapsed multiple myeloma (MM) showed that pelareorep treatment selectively infected MM cells, as viral RNA was found in myeloma cells and not the bone marrow (BM) stroma. However, we did not observe apoptosis. Our ongoing phase 1 trial, which combines the proteasome inhibitor carfilzomib with RV, has demonstrated RV infection, apoptosis, and clinical responses. We investigated the molecular mechanisms behind the role of a PI (carfilzomib) in this setting.

RESULTS: In all MM cell lines we tested (n=4), independently from their sensitivity to RV infection (Stiff et al., Mol Cancer Ther, 2016), viral replication and apoptosis was impaired when MM cells were directly exposed to PIs (carfilzomib and bortezomib). When this experiment was repeated in the setting of the bone marrow cellular fraction or peripheral blood mononuclear cells (PBMCs), it had the opposite effect, as the addition of PI to RV increased RV replication and apoptosis in MM cells.

When we washed PBMCs after overnight exposure to RV+PI or to either single agent, then added MM cells, we observed higher infection and apoptosis in cancer cells co-cultured with RV+PI compared to levels from PBMCs treated with each of the single agents, suggesting that PIs increase the ability of PBMCs to serve as a reservoir for infectious reovirus. Monocytes (CD14+) can engage in phagocytosis of reovirus (Berkeley et al., Cancer Immunol Res, 2018), and accordingly we found that RV genome and capsid protein production were detected in CD14+ cells, but not in CD14-depleted PBMCs, and increased upon PI treatment compared to that in RV-treated CD14+ cells.

Given that the NF-κB complex is a key proinflammatory signaling pathway associated with the early innate-antiviral immune response, and because PIs can block the degradation of the NF-κB inhibitor IκBα upon phosphorylation, we investigated the effect of the specific IκBα inhibitor Bay-11 in RV viral replication. Our data show that either PI or Bay-11 can inhibit RV-induced IκBα phosphorylation and its subsequent degradation upon RV infection in CD14+ cells, an effect associated with higher capsid formation in RV-treated CD14+ cells in combination with PI or Bay-11, compared to levels from RV alone. Cytokine profiling in PBMCs and CD14+ cells treated with RV in combination with either PI or Bay-11 showed a significant decrease in IFN-α and IFN-β (IFNs) levels and a concomitant increase in RV replication, in contrast to levels from RV alone (p<0.001).

We then decided to investigate whether CD14 depletion could affect RV delivery to the cancer cells invivo. Upon intra-femoral injection of 5TGM1 MM cells into syngeneic C57BL/KaLwRij mice, the mice in which the monocytes were depleted by clodronate-liposome treatment before intravenous RV injection showed lower capsid protein formation in the BM MM cells compared to that in mice where the monocytic population was intact. Because monocytes respond to infection by dividing into macrophages to eliminate pathogens, we wanted to investigate whether PI could impair this effect. Intriguingly, although higher levels of viral genome were detected in PI+RV-treated CD14+ cells compared to RV-treated cells (p<0.01), PI abolished ex-vivo RV-induced monocyte differentiation (n=3, p<0.001), without affecting the ability of RV-infected CD14+ cells to induce higher MM cell killing as compared to that from the CD14+ fraction treated with either single agent. Immune killing assays showed that RV-infected MM cells are more susceptible than non-infected cells to T cell effectors (p<0.001), a mechanism that is further potentiated by the addition of PI. In fact, in contrast to the PI-induced blocking of monocyte expansion and activation in response to the virus (n=3, p<0.001), our data suggest that the PI potentiate the expansion of CD8+ T cells, both ex vivo and in two RV+PI treated patients we analyzed so far.

CONCLUSIONS: Here we report for the first time that PIs enhance pelareorep entry, infection, and killing of myeloma cells through its effect on the CD14+ fraction. Reovirus infection and replication within CD14+ cells are augmented by PI-induced NF-κB inhibition of the early innate pro-inflammatory immune response. We also report for the first time that carfilzomib induces direct T cell activation and potentiates T cell killing activity against RV-infected MM cells.

Disclosures

Krishnan:Takeda: Research Funding; Celgene, Z Predicta: Other: Stock Ownership; Amgen, Takeda: Speakers Bureau; Sutro BioPharma, zPredicta: Consultancy; Celgene, Janssen, Sanofi, BMS: Consultancy. Sborov:Celgene: Honoraria; Janssen: Consultancy. Hofmeister:Nektar: Honoraria, Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Imbrium: Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria; Janssen: Membership on an entity's Board of Directors or advisory committees; Oncopeptides: Membership on an entity's Board of Directors or advisory committees.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution