Introducing an In Vitro Liver Stability Assay Capable of Predicting the In Vivo Pharmacodynamic Efficacy of siRNAs for IVIVC

Tissue pharmacokinetics of antisense oligonucleotides

Pharmacokinetics (PK) of antisense oligonucleotides (ASOs) is characterized by rapid distribution from plasma to tissue and slow terminal plasma elimination driven by re-distribution from tissue. Quantitative understanding of tissue PK and RNA knockdown for various ASO chemistries, conjugations, and administration routes is critical for successful drug discovery. Here, we report concentration-time and RNA knockdown profiles for a gapmer ASO with locked nucleic acid ribose chemistry in mouse liver, kidney, heart, and lung after subcutaneous and intratracheal administration. Additionally, the same ASO with liver targeting conjugation (galactosamine-N-acetyl) is evaluated for subcutaneous administration. Data indicate that exposure and knockdown differ between tissues and strongly depend on administration route and conjugation. In a second study, we show that tissue PK is similar between the three different ribose chemistries locked nucleic acid, constrained ethyl and 2'-O-methoxyethyl, both after subcutaneous and intratracheal administration. Further, we show that the half-life in mouse liver may vary with ASO sequence. Finally, we report less than dose-proportional increase in liver concentration in the dose range of 3-30 μmol/kg. Overall, our studies contribute pivotal data to support design and interpretation of ASO in vivo studies, thereby increasing the probability of delivering novel ASO therapies to patients.

Self-delivering, chemically modified CRISPR RNAs for AAV co-delivery and genome editing in vivo

Guide RNAs offer programmability for CRISPR-Cas9 genome editing but also add challenges for delivery. Chemical modification, which has been key to the success of oligonucleotide therapeutics, can enhance the stability, distribution, cellular uptake, and safety of nucleic acids. Previously, we engineered heavily and fully modified SpyCas9 crRNA and tracrRNA, which showed enhanced stability and retained activity when delivered to cultured cells in the form of the ribonucleoprotein complex. In this study, we report that a short, fully stabilized oligonucleotide (a 'protecting oligo'), which can be displaced by tracrRNA annealing, can significantly enhance the potency and stability of a heavily modified crRNA. Furthermore, protecting oligos allow various bioconjugates to be appended, thereby improving cellular uptake and biodistribution of crRNA in vivo. Finally, we achieved in vivo genome editing in adult mouse liver and central nervous system via co-delivery of unformulated, chemically modified crRNAs with protecting oligos and AAV vectors that express tracrRNA and either SpyCas9 or a base editor derivative. Our proof-of-concept establishment of AAV/crRNA co-delivery offers a route towards transient editing activity, target multiplexing, guide redosing, and vector inactivation.

Metabolically Stable Anomeric Linkages Containing GalNAc-siRNA Conjugates: An Interplay among ASGPR, Glycosidase, and RISC Pathways

Conjugation of synthetic triantennary N-acetyl-d-galactosamine (GalNAc) to small interfering RNA (siRNA) mediates binding to the asialoglycoprotein receptor (ASGPR) on the surface of hepatocytes, facilitating liver-specific uptake and siRNA-mediated gene silencing. The natural β-glycosidic bond of the GalNAc ligand is rapidly cleaved by glycosidases in vivo. Novel GalNAc ligands with S-, and C-glycosides with both α- and β-anomeric linkages, N-glycosides with β-anomeric linkage, and the O-glycoside with α-anomeric linkage were synthesized and conjugated to siRNA either on-column during siRNA synthesis or through a high-throughput, post-synthetic method. Unlike natural GalNAc, modified ligands were resistant to glycosidase activity. The siRNAs conjugated to newly designed ligands had similar affinities for ASGPR and similar silencing activity in mice as the parent GalNAc-siRNA conjugate. These data suggest that other factors, such as protein-nucleic acid interactions and loading of the antisense strand into the RNA-induced silencing complex (RISC), are more critical to the duration of action than the stereochemistry and stability of the anomeric linkage between the GalNAc moiety of the ligand conjugated to the sense strand of the siRNA.

Dendrimer-siRNA Conjugates for Targeted Intracellular Delivery in Glioblastoma Animal Models

Small interfering RNAs (siRNAs) are potent weapons for gene silencing, with an opportunity to correct defective genes and stop the production of undesirable proteins, with many applications in central nervous system (CNS) disorders. However, successful delivery of siRNAs to the brain parenchyma faces obstacles such as the blood-brain barrier (BBB), brain tissue penetration, and targeting of specific cells. In addition, siRNAs are unstable under physiological conditions and are susceptible to protein binding and enzymatic degradation, necessitating a higher dosage to remain effective. To address these issues and advance siRNA delivery, we report the development of covalently conjugated hydroxyl-terminated poly(amidoamine) (PAMAM) dendrimer-siRNA conjugates, demonstrated with a siRNA against GFP (siGFP) conjugate (D-siGFP) utilizing glutathione-sensitive linkers. This allows for precise nucleic acid loading, protects the payload from premature degradation, delivers the siRNA cargo into cells, and achieves significant GFP knockdown in vitro (∼40%) and in vivo (∼30%). Compared to commercially available delivery systems such as RNAi Max and Lipofectamine, D-siGFP retains the potency of the siRNA in vitro. In addition, the dendrimer-siGFP conjugate significantly enhances the half-life of siRNA in the presence of plasma and endonucleases and maintains the passive targeting ability of PAMAM dendrimers to reactive microglia. When administered intratumorally to orthotopic glioblastoma multiform tumors (GBM) in CX3CR-1GFP mice, D-siGFP localizes in tumor-associated macrophages (TAMs) within the tumor parenchyma, minimizing off-target effects in other cell populations. The facile conjugation strategy for dendrimer-siRNA conjugates presented here offers a promising approach for targeted, systemic intracellular delivery of siRNA, serving as a potential bridge for the clinical translation of RNAi therapies.

Pharmacokinetic and Pharmacodynamic Modeling of siRNA Therapeutics - a Minireview

The approval of four small interfering RNA (siRNA) products in the past few years has demonstrated unequivocally the therapeutic potential of this novel modality. Three such products (givosiran, lumasiran and inclisiran) are liver-targeted, using tris N-acetylgalactosamine (GalNAc)₃ as the targeting ligand. Upon subcutaneous administration, GalNAc-conjugated siRNAs rapidly distribute into the liver via asialoglycoprotein receptor (ASGPR) mediated uptake in the hepatocytes, resulting in fast elimination from the systemic circulation. Patisiran, on the other hand, has been formulated in a lipid nanoparticle for optimal delivery to the liver. While several publications have described preclinical and clinical pharmacokinetic (PK) and pharmacodynamic (PD) results, including absorption, distribution, metabolism, and elimination (ADME) profiles in selected species as well as limited modeling efforts for siRNA therapeutics, there is no systematic review of the PK and PD models developed for these agents or work summarizing the utility and application(s) of such models in drug development and regulatory review. Here, we provide a mini-review of the current state of modeling efforts for siRNA therapeutics within the early preclinical, translational, and clinical stages of siRNA development. Diverse modeling methods including simple compartmental, mechanistic and systems PK/PD, physiologically-based PK (PBPK), population PK/PD, and dose-response-time models are introduced and reviewed. The utility of such models in development and regulatory review for siRNA therapeutics is also discussed with examples. Finally, the current knowledge gaps in mechanism of action of siRNA and resulting challenges in model development are summarized. The goal of this minireview is to trigger cross-functional discussion amongst all key stakeholders to generate key experimental datasets and align on current assumptions, model structures, and approaches to facilitate development and application of robust PK/PD models for siRNA therapeutics.

Considerations and recommendations for assessment of plasma protein binding and drug-drug interactions for siRNA therapeutics

At the time of writing, although siRNA therapeutics are approved for human use, no official regulatory guidance specific to this modality is available. In the absence of guidance, preclinical development for siRNA followed a hybrid of the small molecule and biologics guidance documents. However, siRNA differs significantly from small molecules and protein-based biologics in its physicochemical, absorption, distribution, metabolism and excretion properties, and its mechanism of action. Consequently, certain reports typically included in filing packages for small molecule or biologics may benefit from adaption, or even omission, from an siRNA filing. In this white paper, members of the 'siRNA working group' in the IQ Consortium compile a list of reports included in approved siRNA filing packages and discuss the relevance of two in vitro reports-the plasma protein binding evaluation and the drug-drug interaction risk assessment-to support siRNA regulatory filings. Publicly available siRNA approval packages and the literature were systematically reviewed to examine the role of siRNA plasma protein binding and drug-drug interactions in understanding pharmacokinetic/pharmacodynamic relationships, safety and translation. The findings are summarized into two decision trees to help guide industry decide when in vitro siRNA plasma protein binding and drug-drug interaction studies are warranted.

Unlocking the potential of TRPV1 based siRNA therapeutics for the treatment of chemotherapy-induced neuropathic pain

Chemotherapy-induced neuropathic pain (CINP) is among the most common clinical complications associated with the use of anti-cancer drugs. CINP occurs in nearly 68.1% of the cancer patients receiving chemotherapeutic drugs. Most of the clinically available analgesics are ineffective in the case of CINP patients as the pathological mechanisms involved with different chemotherapeutic drugs are distinct from each other. CINP triggers the somatosensory nervous system, increases the neuronal firing and activation of nociceptive mediators including transient receptor protein vanilloid 1 (TRPV1). TRPV1 is widely present in the peripheral nociceptive nerve cells and it has been reported that the higher expression of TRPV1 in DRGs serves a critical role in the potentiation of CINP. The therapeutic glory of TRPV1 is well recognized in clinics which gives a promising insight into the treatment of pain. But the adverse effects associated with some of the antagonists directed the scientists towards RNA interference (RNAi), a tool to silence gene expression. Thus, ongoing research is focused on developing small interfering RNA (siRNA)-based therapeutics targeting TRPV1. In this review, we have discussed the involvement of TRPV1 in the nociceptive signaling associated with CINP and targeting this nociceptor, using siRNA will potentially arm us with effective therapeutic interventions for the clinical management of CINP.

Minimal Physiologically Based Pharmacokinetic-Pharmacodynamic (mPBPK-PD) Model of N-Acetylgalactosamine-Conjugated Small Interfering RNA Disposition and Gene Silencing in Preclinical Species and Humans

Conjugation of small interfering RNA (siRNA) to tris N-acetylgalactosamine [(GalNAc)₃] can enable highly selective, potent, and durable knockdown of targeted proteins in the liver. However, potential knowledge gaps between in vitro experiments, preclinical species, and clinical scenarios remain. A minimal physiologically based pharmacokinetic-pharmacodynamic model for GalNAc-conjugated siRNA (GalNAc-siRNA) was developed using published data for fitusiran (ALN-AT3), an investigational compound targeting liver antithrombin (AT), to delineate putative determinants governing the whole-body-to-cellular pharmacokinetic (PK) and pharmacodynamic (PD) properties of GalNAc-siRNA and facilitate preclinical-to-clinical translation. The model mathematically linked relevant mechanisms: 1) hepatic biodistribution, 2) tris-GalNAc binding to asialoglycoprotein receptors (ASGPRs) on hepatocytes, 3) ASGPR endocytosis and recycling, 4) endosomal transport and escape of siRNA, 5) cytoplasmic RNA-induced silencing complex (RISC) loading, 6) degradation of target mRNA by bound RISC, and 7) knockdown of protein. Physiologic values for 36 out of 48 model parameters were obtained from the literature. Kinetic parameters governing (GalNAc)₃-ASGPR binding and internalization were derived from published studies of uptake in hepatocytes. The proposed model well characterized reported pharmacokinetics, RISC dynamics, and knockdown of AT mRNA and protein by ALN-AT3 in mice. The model bridged multiple PK-PD data sets in preclinical species (mice, rat, monkey) and successfully captured reported plasma pharmacokinetics and AT knockdown in a phase I ascending-dose study. Estimates of in vivo potency were similar (∼2-fold) across species. Subcutaneous absorption and serum AT degradation rate constants scaled across species by body weight with allometric exponents of -0.29 and -0.22. The proposed mechanistic modeling framework characterizes the unique PK-PD properties of GalNAc-siRNA. SIGNIFICANCE STATEMENT: Tris N-acetylgalactosamine (GalNAc)₃-conjugated small interfering RNA (siRNA) therapeutics enable liver-targeted gene therapy and precision medicine. Using a translational and systems-based minimal physiologically based pharmacokinetic-pharmacodynamic (mPBPK-PD) modeling approach, putative determinants influencing GalNAc-conjugated siRNA (GalNAc-siRNA) functionality in three preclinical species and humans were investigated. The developed model successfully integrated and characterized relevant published in vitro-derived biomeasures, mechanistic PK-PD profiles in animals, and observed clinical PK-PD responses for an investigational GalNAc-siRNA (fitusiran). This modeling effort delineates the disposition and liver-targeted pharmacodynamics of GalNAc-siRNA.

Syntheses of Morpholine-Based Nucleotide Analogs for Hepatic siRNA Targeting and Stabilization

A morpholine-based nucleotide analog was developed as a building block for hepatic siRNA targeting and stabilization. Attachment of an asialoglycoprotein-binding GalNAc ligand at the morpholine nitrogen was realized with different linkers. The obtained morpholino GalNAc scaffolds were coupled to the sense strand of a transthyretin-targeting siRNA and tested for their knockdown potency in vitro and in vivo. A clear structure-activity relationship was developed with regard to the linker type and length as well as the attachment site of the morpholino GalNAc moieties at the siRNA sense strand. Further, simple alkylation of the morpholine nitrogen led to a nucleotide analog, which increased siRNA stability, when used as a double 3'-overhang at the sense strand sequence. Combination of the best morpholino GalNAc building blocks as targeting nucleotides with an optimized stabilizing alkyl-substituted morpholine as 3'-overhangs resulted in siRNAs without any phosphorothioate stabilization in the sense strand and clearly improved the duration of action in vivo.

Physiologically based pharmacokinetic (PBPK) modeling of RNAi therapeutics: Opportunities and challenges

Physiologically based pharmacokinetic (PBPK) modeling is a powerful tool with many demonstrated applications in various phases of drug development and regulatory review. RNA interference (RNAi)-based therapeutics are a class of drugs that have unique pharmacokinetic properties and mechanisms of action. With an increasing number of RNAi therapeutics in the pipeline and reaching the market, there is a considerable amount of active research in this area requiring a multidisciplinary approach. The application of PBPK models for RNAi therapeutics is in its infancy and its utility to facilitate the development of this new class of drugs is yet to be fully evaluated. From this perspective, we briefly discuss some of the current computational modeling approaches used in support of efficient development and approval of RNAi therapeutics. Considerations for PBPK model development are highlighted both in a relative context between small molecules and large molecules such as monoclonal antibodies and as it applies to RNAi therapeutics. In addition, the prospects for drawing upon other recognized avenues of PBPK modeling and some of the foreseeable challenges in PBPK model development for these chemical modalities are briefly discussed. Finally, an exploration of the potential application of PBPK model development for RNAi therapeutics is provided. We hope these preliminary thoughts will help initiate a dialogue between scientists in the relevant sectors to examine the value of PBPK modeling for RNAi therapeutics. Such evaluations could help standardize the practice in the future and support appropriate guidance development for strengthening the RNAi therapeutics development program.