Targeted Delivery of RNAi Therapeutics for Cancer Therapy

Keywords::

• cancer therapy
• nanoparticle
• siRNA

Since the discovery of RNA interference (RNAi) [1], siRNA) has become the focus of attention for pharmaceutical development. siRNA is more potent than antisense oligonucleotides owing to the endogenous mechanism of action catalyzed by the RNA-induced silencing complex [2]. Many oncogenic targets involved in survival, specifically antiapoptosis, angiogenesis and drug resistance, have been tested for siRNA-mediated cancer therapy. However, siRNA can only become an anticancer agent when it is specifically and effectively delivered to the target cells in vivo[3]. An ideal siRNA delivery vehicle for cancer therapy must be able to evade the reticuloendothelial system to be effectively taken up by the tumor cells and to escape from the endosome after endocytosis. The existence of a leaky vasculature in most of the solid tumors supports the enhanced permeability and retention effect for nanoparticle (NP)-mediated delivery of anticancer agents, for example the successful Doxil^® formulation of doxorubicin. A well-designed NP of less than 200 nm in diameter is suitable for siRNA delivery to the tumor. NPs composed of different materials have been developed for siRNA delivery. Two major classes of biomaterials have been employed for siRNA delivery. They will be discussed separately in this article.

Lipid base NPs for targeted siRNA delivery

Cationic lipids have been widely used for siRNA delivery in vitro. Many commercially available transfection reagents are made of cationic lipids but these reagents usually do not support in vivo delivery owing to the presence of excess cationic charges in the lipoplex. Ideally, siRNA should be entrapped in a large quantity inside the unilamellar liposomes that are protected with a polyethylene glycol (PEG) brush layer on the surface. The PEG brush effectively reduces blood protein adsorption to the NP and reduces its uptake by the reticuloendothelial system [4]. A targeting ligand is further tethered at the distal end of the PEG chain for specific binding and internalization mediated by tumor-specific receptors. Lipids with an ionizable headgroup having a pKa of approximately 6 are particularly attractive. siRNA can associate with the lipid at low pH because the lipid is protonated and positively charged. Being entrapped at low pH, substantial amounts of siRNA are still liposome-associated after the pH returns to neutral. The lipid will again be cationic in the acidic endosome. It is well known that cationic lipid binds with anionic lipid to form inverted micelles, or the H_II phase, which is a nonbilayer structure [5]. Lipids with bulky hydrocarbon chains are prone to assume such structure [6]. Several lipids containing a tertiary amine headgroup (pKa ∼6) and two hydrocarbon chains with two cis double bonds (bulky) have been designed and used to deliver siRNA [7]. The activity of the ionizable lipid NPs for siRNA delivery to the liver hepatocytes depends on the circulating ApoE, which facilitates an avid uptake of the NP via the low-density lipoprotein receptor [8]. The 50% effective dose (ED₅₀) of siRNA for silencing factor VII, which is exclusively produced in the liver, was as low as 0.01 mg/kg in rodents and 0.1 mg/kg in nonhuman primates. No noticeable toxicity at these low doses was reported. N-acetyl galactosamine can also be used to target the iLNPs to the liver via the galactose receptor. This is a remarkable achievement, although the formulation has not yet been used for cancer therapy.

The stable nucleic acid lipid particle is a similar ionizable lipid base formulation developed by the same group for systemic siRNA administration [9]. The original study focused on hepatic cells, which demonstrated the downregulation of ApoB in the liver with a dose of 2.5 mg/kg in nonhuman primates [10]. High therapeutic doses, however, usually cause some type of immune response. Recently, a clinical trial was terminated owing to the immunotoxicity revealed in the trial [101].

In an alternative approach, a combinatorial library of lipid-like materials was developed and described with the term ‘lipidoids’ [11]. The one-step synthetic scheme provided some choice of cationic lipid for efficient transfection of siRNA [12]. The selected cationic LNPs was able to achieve siRNA-mediated gene silencing effect for hepatic cell at a dose of 0.01 mg/kg in mice and 0.03 mg/kg in nonhuman primates [13]. This formulated NP is again a remarkable achievement, although the toxicity of these lipidoids has not been thoroughly studied.

Instead of application in hepatic cells, a preclinical study of 2´-O-methyl-modified siRNA formulated in LNPs was reported to show potent antitumor efficacy in both hepatic and subcutaneous tumor models [14]. The siRNA is targeted to the essential cell-cycle proteins polo-like kinase 1 (PLK1) and kinesin spindle protein (KSP) in mice. A very mild immune response was observed in mice with intravenous twice-weekly administration at a dose of 2 mg/kg. A Phase I clinical trial was launched to determine the safety, tolerability, pharmacokinetics and pharmacodynamics of intravenous LNPs in patients with advanced solid tumors involving the liver [102].

Another lipid-base NP for cancer therapy is known as lipid/protamine/DNA (LPD) formulation, which was originally developed for plasmid DNA delivery [15]. LPD was engineered by combining cationic liposomes and polycation to condense plasmid DNA. When mixed, the components spontaneously rearranged to form a virus-like structure with the condensed DNA core located inside the lipid membranes [16]. Recently, siRNA was entrapped in the core of LPD and the surface of the particle was wrapped with cationic lipid and PEG. Anisamide was linked to the terminal of PEG as a targeting ligand of s receptor, which is overexpressed in many human tumor cells [17]. The most important feature of the modified LPD is its ability to evade the reticuloendothelial system, which is due to the fact that a high amount (∼10 mol%) of PEG chains could be grafted on the surface to form a brush protective layer on the NPs [18]. The xenograft model demonstrated that LPD NPs could deliver 60–80% of intravenous injected siRNA per gram of tissue weight to the lung cancer xenograft and effectively silence the expression of EGF receptor in the entire tumor. Two consecutive intravenous administrations significantly reduced the lung metastasis of melanoma (70–80%) at a relatively low dose (0.45 mg/kg) [19]. To reduce the potential toxicity of calf thymus DNA in the formulation, liposome–polycation–hyaluronic acid was developed to deliver siRNA systemically into the tumor, which resulted in very little immunotoxicity in a wide dose range (0.15–11.2 mg/kg) [20].

To improve the efficiency of LPD formulation, an acid-sensitive material, calcium phosphate, was investigated to replace the stable protamine/DNA complex. The formulation contained a PEGylated membrane similar to that of LPD, but the core that encapsulated siRNA is replaced with calcium phosphate amorphous nanoprecipitate [21]. The proposed mechanism of action is described as follows. After entering the endosome, calcium phosphate dissolves rapidly in acidic pH, increasing the osmotic pressure, and resulting in swelling and rupture of the endosome to release siRNA into the cytoplasm. The new formulation is known as lipid–calcium–phosphate and the ED₅₀ is similar to LPD formulation with very mild immune toxicity.

There are several reports using neutral phospholipids to formulate siRNA (neutral nanoliposome) [22]. The formulation has a potent activity in silencing Epha2, an oncogene involved in ovarian cancer progression. Impressive antitumor activity was shown by intravenous administration of the siRNA formulation. The study has already been advanced to a Phase I clinical trial, but it is not clear how siRNA could be entrapped with high efficiency in the neutral liposomes.

Polymer-based NPs for targeted siRNA delivery

In addition to cationic lipid, positively charged polymer is another type of material for the delivery of nucleic acid. The complex is termed polyplex. Polyethyleneimine, a highly cationically charged polymer, is a potent in vitro transfection reagent and has been used to deliver VEGF siRNA for cancer therapy [23]. With grafted PEG, the complex spontaneously forms micelles, which have a polyethyleneimine–siRNA core wrapped with PEG. Tail vein injection of 1.5 nmol (1.1 mg/kg) siRNA in the micelles significantly suppressed tumor growth in a xenograft tumor model. In addition, arg–gly–asp peptide-labeled chitosan NP was employed as a vehicle for siRNA delivery, targeting to integrins overexpressed in the tumor cells and the tumor endothelial cells. The NP significantly increased siRNA delivery in orthotopic animal models of ovarian cancer and the injected dose of siRNA was 0.15 mg/kg [24].

Other cationic polymer-based NPs for siRNA delivery consist of a cyclodextrin-containing polymer for binding with siRNA, a PEG as steric stabilization agent, and transferrin as a targeting ligand for cancer cells [25]. The cationic polymer with siRNA stays inside of the NP and the cyclodextrin serves as an adapter, where different PEG and the target ligand conjugated to adamantane can be ‘plugged’ into the complex. The targeted NPs that systematically deliver siRNA against the EWS-FLI1 gene could inhibit tumor growth in a murine model of metastatic Ewing‘s sarcoma. The most exciting characteristic is that the resulting siRNA NPs (2.5 mg/kg) did not induce immune response. When administered to cynomolgus monkeys at doses of 3 and 9 mg siRNA/kg, the NPs were well tolerated. However, elevated levels of blood urea nitrogen and creatinine were observed, indicating kidney toxicity when the dose was 27 mg siRNA/kg. This targeted NP is currently in a clinical trial with patients with solid tumors [26]. Tumor biopsies from melanoma patients in this trial showed the presence of intracellularly localized NPs. Moreover, expected mRNA degradation fragments by the RNA-induced silencing complex were observed in the biopsies.

Recently, several multifunctional polyconjugates have been developed to avoid the off-target delivery and facilitate the escape of siRNA from the endosome. For example, a pH-sensitive polyconjugate was attached to siRNA through a disulfide linkage [27]. The complex was shielded by PEG and modified with a hepatocyte-targeting ligand. After entering the endosome, the polymer shed both the targeting ligand and the PEG to expose its multiple amino groups to destabilize the endosome membrane. siRNA was released into the cytoplasm due to its reductive environment. Over 80% of the ApoB expression was downregulated in vivo by this method with a siRNA dose of 2.5 mg/kg. Another reducible polyconjugate was also prepared by modifying the siRNA with phosphothioethanol portion via a disulfide bond [28]. Although no tumor model was tested to date, the multifunctional polyconjugate is probably safer than other polymers, but its activity appears to be significantly lower than that of ionizable or cationic LNPs.

In addition to the synthetic polymers, biodegradable and biocompatible proteins have been employed as biopolymers for siRNA delivery. For example, a protamine–antibody fusion protein was designed to deliver siRNA to HIV-infected or envelope-transfected cells [29]. Intratumoral or intravenous injection of the antibody–siRNA complex into mice showed a targeted delivery to melanoma cells. Inhibition of subcutaneous B16F10 tumor growth was also observed. Atelocollagen, a chemically modified collagen, was used to deliver siRNA to bone metastasis of prostate cancer [30]. It is interesting to note that the siRNA–atelocollagen complex could be detected in the tumor 24 h after injection and the target silencing lasted for at least 3 days.

Conclusion

Nonviral vectors for delivering nucleic acid have improved dramatically in the last decade. Their activity is comparable with that of viral vectors in many cases. Their clinical development has also been accelerated recently. Despite these encouraging advances, toxicity of nonviral vectors has not been adequately evaluated, especially in clinical trials. Cationic lipids and polymers induce a rapid increase in cellular reactive oxygen species [31]. This is a primary signal in macrophages and other innate immune cells to secrete proinflammatory cytokines. In addition, siRNA is not without its own immunostimulating toxicity, which often leads to off-target effects [32]. Fortunately, various chemical modifications of siRNA tend to limit this type of unwanted toxicity [33]. Still, the best strategy is to develop potent delivery formulation such that only a small dose is needed to silence the target gene for activity to be therapeutic. Toxicity would not be an issue at low doses. In addition, it will significantly decrease the cost of the drug because both siRNA and the formulation of lipid or polymer can be very expensive. Nevertheless, the field of siRNA delivery for cancer therapy has already come a long way. It is not hard to see an even brighter future on the horizon.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Additional information

Funding

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.