Advertisement

Combination Therapy with Daratumumab and CAR-NK Targeting CS1 for Multiple Myeloma

Yibo Zhang, Lichao Chen, Yufeng Wang, Xinxin Li, Tiffany Hughes, Hans Klingemann, Don M Benson and Jianhua Yu

Abstract

Daratumumab (Dara), a targeted therapy utilizing a monoclonal antibody against CD38, and its combination with other are becoming a new standard of care treatment in multiple myeloma (MM). Recently, chimeric antigen receptor (CAR) T cell immunotherapy has been successful in the clinic for the treatment of leukemia and lymphoma. Our preliminary data suggest that both CS1-CAR T cells and CS1-CAR NK cells are effective in eradicating MM cells in vitro and in vivo (Chu et al., 2014, Leukemia and Chu et al., 2014, Clinical Cancer Research). In this study, we investigated the combination therapy with Dara and CS1-CAR NK cells for the treatment of relapsed MM.

We first showed that that in MM patients, CD38brightCD138CD34CD20+CD27+ MM cancer stem-like cells (CSCs) express CS1 at levels much higher than any other cells, and are susceptible to being eradicated by CS1-CAR NK cells. However, CD34+hematopoietic stem cells from bone marrow of healthy donors do not express CS1. These data suggest that CS1-CAR NK cells can target MM CSCs, and thus may prevent relapse of MM, as ample evidence shows that relapsed or recurrent tumor cells are derived from CSCs.

We also demonstrated that CD38 is highly expressed on NK and MM cells. Dara triggered IFN-γ and GZMB expression (p< 0.01) in primary human NK cells, even in the absence of crosslinking with tumor cells. Interestingly, the increase IFN-γ expression can be validated in the CD16 (+) haNK-92 (high-affinity natural killer cells), but not in the parental NK-92 cell line. Blocking the recognition between CD16 and Dara (an IgG1 mAb) with an Fc blocking Ab completely impaired Dara-induced IFN-γ and GZMB expression, indicating that Dara-induced NK cell activation is CD16-dependent. Mechanistically, Dara significantly induced phosphorylation of NFkB and STAT1, indicating that Dara induces IFN-γ and GZMB in NK cells, which may occur through CD16 and be mediated downstream by STAT1 and NFkB. We also found that Dara failed to stimulate GZMB and IFN-γ expression in CD38(-) CD16(+) NK cells, while successful in stimulating CD38(+) CD16(+) NK cells, indicating that Dara induces NK cell activation, which requires not only the binding between CD16 and Fc fragment of Dara, but also the CD38 signaling pathway. Furthermore, we found that Dara mediated cytotoxicity of NK cells against MM cells through antibody-dependent cell-mediated cytotoxicity (ADCC) against CD38-positive (e.g., MM1.S), but not CD38-negative (e.g., U266), which can be blocked by CD16 blocking Ab. Moreover, Dara displays ADCC effects in CD16(+) NK cells but not CD16(-) NK cells. When CD16(+) NK cells were armed with the CS1-CAR, ADCC is still observed against CD38(+) MM cells at low effector to target ratios, i.e., Dara still enhances cytotoxicity of CS1-CAR NK cells, which already have enhanced cytotoxicity.

We observed that Dara-induced NK cell ADCC against CD38(+) MM MM1.S cells led to increased T cell proliferation and activation in a co-culture system including dendritic cells. This effect was not observed when MM U266 cells were included as the NK cell target. Out data are consistent with that recent discovery by DiLillo and Ravetch showing that engagement of monoclonal antibody can induce an antitumor vaccine effects (David J et al., Cell, 2015).

To tested Dara affects NK cell survival, immunoblotting was performed with anti-cleaved Caspase-3 and anti-cleaved PARP-1 antibodies. We demonstrated that apoptotic activity was increased in both CD16(+)NK cells (primary NK and haNK-92) and parental CD16(-)NK-92 cells treated with Dara for 24 h in a dose-dependent manner. Unlike Dara's positive effects on CD16(+) NK cells (i.e. stimulating IFN-γ production and ADCC), induction of apoptosis seems to be CD16-independent, as parental NK-92 cells, which are CD16(-), also showed an increased levels of apoptosis induced by Dara. We are testing whether the apoptosis induction is dependent on the antigen for Dara, because as mentioned above, both primary NK cells, and modified as well as unmodified NK-92 cells, that were CD38 (+).

In conclusion, our study demonstrates that the combination of Dara and CS1-CAR NK cells, which target two different tumor-associated antigens, both of which have potent anti-MM efficacy, may show additive or synergistic effects; however due to the positive and negative effects of Dara on NK cells, sequential treatment rather than a concomitant treatment modality should be considered.

Disclosures No relevant conflicts of interest to declare.

  • * Asterisk with author names denotes non-ASH members.