In vivo Natural Killer Cell Activation with siRNA-Lipid Nanoparticles for Cancer Immunotherapy

Introduction: Natural killer (NK) cell-based immunotherapy is has recently come to the forefront for cancer treatment, with early data from clinical trials showing promising outcomes. Due to their ability to kill cancer cells without requiring MHC-antigen presentation, lack of graft versus host reactions, and relatively low abundance in many cancers, ex vivo edited NK cell therapies have been employed. NK cell receptor engineering, knock-out of genes involved in NK cell suppression, and insertion of chimeric antigen receptors (CARs) are all currently under investigation to boost the activity of NK cell-therapeutics in immunosuppressive tumor microenvironments. Such ex vivo-edited cell therapies, however, often require patients to undergo lymphodepleting chemotherapy, and there is no clear choice of optimal donor NK cells; cell line, umbilical cord, and iPSC-derived NK cells are all currently in trial. The ability to directly modulate NK cells in vivo, to boost their proliferation and activation in tumor microenvironments, would substantially reduce treatment cost, allow for more rapid treatment following diagnosis, eliminate the need for lymphodepleting chemotherapy, and improve patient outcome. The development of delivery vectors capable of in vivo NK cell delivery, however, has been challenging. Here, we developed an siRNA–LNP capable of delivering RNA to NK cells in vivo, to silence an inhibitor of NK cell activation. By using RNA to target an intracellular negative regulator of IL-15 signaling, this platform technology allows the silencing of cytokines not accessible to other therapeutic modalities like small molecules or antibody therapeutics; and allows for synergy with recombinant IL-15 therapeutics.

Materials and

Methods: Short interfering RNA (siRNA) was designed in silico to target the negative regulator of IL-15 signaling in NK cells. siRNAs with various sequences and chemical modification patterns were screened in vitro to assess silencing efficiency, at the level of transcription, using RT-qPCR. Lipid nanoparticles (LNPs) were formulated with a novel, biodegradable, ionizable lipid (previously selected from a large library of ionizable lipids); used to encapsulated CD45 silencing siRNA, and flow cytometry used to assess in vivo delivery to NK cells, based on CD45 expression level (median fluorescent intensity).

Therapeutic efficacy was assessed in a murine model of peritoneal disseminated melanoma. Two days following intraperitoneal melanoma engraftment, mice were treated with intraperitoneal injections of siRNA-LNP (0.25 mg/kg, 2x weekly), recombinant murine IL-15 (0.1 mg/kg, 2x weekly), and/or a melanoma-opsonizing antibody (TA99; 10 mg/kg, 1x weekly). Flow cytometry was used to assess peritoneal NK cell recruitment and phenotype; IVIS imaging was used to track tumor burden. In order to assess the contribution of NK cells, in a subsequent experiment, anti-NK1.1 antibodies were used to deplete NK cells throughout the course of treatment.

Results, Conclusions, and Discussions: siRNA screening identified a partially modified siRNA, incorporating 2’-OMe and phosphonothioate linkages, that allowed potent silencing (IC50 ~250pM) of the negative regulator of IL-15 signaling in NK cells, in vitro. Our siRNA-LNP formulation demonstrated efficient silencing in peritoneal NK cells, in vivo. IL-15 was administered via intraperitoneal injection into healthy mice (2ug/mouse) or in combination with siRNA-LNPs (0.01mg/kg). Three days following administration, 70% silencing was observed in peritoneal NK cells (RT-qPCR). Silencing the inhibitor of IL-15 signaling increased expression of the IL-15 receptor and boosted NK activation: more than doubling expression of perforin, granzyme A, and granzyme B (flow cytometry), and allowing 4-fold expansion (flow cytometry/cell counts). Such NK expansion is important for therapeutic efficacy of in vivo NK-targeted cancer therapeutics.

To assess therapeutic efficacy, mice were grafted intraperitoneally with B16F10-luciferase melanoma and treated as described in Methods. Day 14 IVIS imaging demonstrated moderate reduction in tumor burden in IL-15 and siRNA-LNP monotherapy groups (although not statically significant for IL-15 monotherapy). IL-15 + siRNA-LNPs synergistically reduced tumor burden (n=6). The addition of tumor opsonizing antibodies to IL-15 + siRNA-LNP did not further reduce tumor burden. Interestingly, treatment with IL-15 + antibody (without siRNA-LNPs) provided no therapeutic advantage compared to saline, suggesting that NK-checkpoint silencing is critical. In a 25-day survival curve, IL-15 + siRNA-LNPs significantly prolonged survival.

In order to assess NK contribution, NK cells were depleted with anti-NK1.1, and tumor burden tracked with IVIS imaging. NK cell specific depletion was confirmed with flow cytometry. Through day 7, NK cell depletion completely abrogated the therapeutic efficacy of IL-15 + siRNA-LNPs. By day 14, IL-15 + siRNA-LNP demonstrated minimally (yet significantly) reduced tumor burden even in the presence of NK cell depletion, suggesting that the IL-15 + siRNA-LNPs may have therapeutic efficacy mediated through additional immune cells recruited at later timepoints. The most substantially reduced tumor burden, through day 14, was observed in the IL-15 + siRNA-LNP group (without NK depletion). Collectively, our data demonstrate that a novel ionizable lipid allows for efficient RNA delivery to NK cells in vivo, and that in vivo NK modulation is a viable therapeutic modality.

Acknowledgements (Optional):