RNA therapeutics in the clinic

Abstract Ribonucleic acid (RNA) therapeutics are being actively researched as a therapeutic modality in preclinical and clinical studies. They have become one of the most ubiquitously known and discussed therapeutics in recent years in part due to the ongoing coronavirus pandemic. Since the first approval in 1998, research on RNA therapeutics has progressed to discovering new therapeutic targets and delivery strategies to enhance their safety and efficacy. Here, we provide an overview of the current clinically relevant RNA therapeutics, mechanistic basis of their function, and strategies to improve their clinical use. We discuss the 17 approved RNA therapeutics and perform an in‐depth analysis of the 222 ongoing clinical trials, with an emphasis on their respective mechanisms and disease areas. We also provide perspectives on the challenges for clinical translation of RNA therapeutics and suggest potential strategies to address these challenges.

in 1998 when the Food and Drug Administration (FDA) approved Vitravene (fomivirsen) for the treatment of retinitis caused by cytomegalovirus. There have since been 16 additional approvals with the most recent approval in early 2022 when Spikevax (COVID-19 vaccine, mRNA) was upgraded from emergency use authorization. While RNA therapeutics have been studied in research laboratories for over 30 years, 1,2 the recent approvals of Spikevax and Comirnaty (tozinameran) as COVID-19 vaccines have brought RNA therapeutics, especially mRNA therapeutics, to the forefront of medicine. In this review, we discuss the clinical translation of this rapidly growing class, RNA therapeutics. In particular, we discuss the biological basis of RNA therapeutics, their function, and key approaches to deliver RNA payloads to the target sites. Additionally, we overview the clinical landscape of 17 approved products and 222 ongoing clinical trials. We also provide a perspective on current challenges and future outlook.

| BIOLOGICAL BASIS, APPLICATIONS, POTENTIAL, AND ADVANTAGES OF RNA THERAPEUTICS
The central dogma of RNA therapeutics is the modulation of protein function and/or production, either by directly targeting proteins, interfering with RNAs encoding the relevant proteins, or providing the genetic code for protein production. There are three major ways in which this central goal is accomplished: (i) binding and blocking of proteins using aptamers; (ii) targeting and binding to native RNAs using ASOs, siRNA, and miRNA mimics; and (iii) expressing target proteins using mRNA. 3 A schematic representation of the mechanisms exploited by RNA therapeutics can be found in Figure 1.
mRNAs are currently the most investigated RNA therapeutic type in active clinical studies. 4,5 mRNA in healthy cells is transcribed from DNA and then translated to proteins. Several efforts have been made to deliver exogenous mRNAs to encode particular proteins of interest ( Figure 1). These exogenous mRNAs, which are single-stranded RNA (ssRNA) fragments that range from 2000 to 20,000 base pairs, have largely been used to code for antigens for vaccination or therapeutic proteins for direct disease intervention. 6 Significant work has been done to improve mRNA stability and translation. Major sequence modifications made to the coding strand include addition of untranslated regions and caps to the 5 0 end and untranslated regions flanked by polyadenylation tails to the 3 0 end. 7 Additionally, codon optimization, to remove rare codons from the sequence, is thought to be important for enhancing translation, 8 but it may represent a potential cause for safety concern. 9 While optimization of stability and translation has improved the efficacy of mRNA therapeutics, a major challenge that has stymied their preclinical and clinical success is the difficulty in delivering them into the target cells. A variety of carriers have been explored to solve mRNA delivery challenges including lipid nanoparticles (LNPs), 10,11 protamine conjugates, 12,13 nanoemulsions, 14 liposomes, 15,16, and a variety of other nanoparticles. 17,18 These carriers stabilize and protect mRNAs, augment cellular uptake, and promote endosomal escape of the payloads, which eventually leads to protein production. Extensive reviews of mRNA delivery systems can be found in recently published reviews elsewhere. 19,20 The most successful, in terms of number of approved products, of the RNA therapeutics are ASOs. ASOs are single-stranded oligonucleotides that target endogenous RNA, including noncoding RNA (ncRNA) and mRNA. ASOs are usually between 13 and 30 base pairs long and rely on complementarity for their function. 21 ASOs' mechanisms of action vary depending on the target, but there are four major considerations for mRNA and pre-mRNA therapies including (i) RNA knockdown, (ii) steric translation inhibition, (iii) splice modulation, and (iv) translation modulation ( Figure 1). RNA knockdown and steric translation inhibition aim to decrease protein expression while splice modulation and translation modulation increase protein expression.
For RNA knockdown, ASOs contain small segments of DNA, which are used to directly bind to the target site on target pre-mRNA or mRNA via complementary base pairing. 22 RNase H1 recognizes the DNA-RNA duplex and cleaves the phosphodiester bond by hydrolysis, 23 thus destroying the pre-mRNA or mRNA and lowering protein levels in the cell. 24 Steric translation inhibition occurs when an ASO directly binds mRNA at a location that is close to the start codon of the relevant exon. This event sterically blocks ribosomes and transfer RNAs from associating with the relevant exon and subsequently prevents initiation of translation. 25 Additionally, ASOs have been used to increase protein expression. Splice modulation is a mechanism by which ASOs binding to pre-mRNA near exons blocks splice enhancers or repressors. This results in an alteration of normal splicing function in cells and can be used to skip unwanted mutations or include previously excluded exons. 26 ASOs can also increase protein expression by translation modulation. By binding the 5 0 untranslated region, upstream of the relevant exon, ASOs can decrease the ribosome attachment to that region, resulting in an increased likelihood of ribosome attachment at the start codon. 27 There are additional mechanisms that have been explored to target endogenous translation silencing molecules like microRNA and other ncRNAs. [28][29][30] A major challenge for ASO therapeutics is to overcome the biological barriers that prevent them from reaching target sites. This challenge is largely overcome by the backbone, base, and ribose modifications or substitutions as well as utilizing alternative chemistries. While the details of the chemistries and mechanisms are beyond the scope of this review, several reviews can be found that discuss these drug delivery methodologies in depth. [31][32][33] Extensive studies on the use of nanoparticles 34 or conjugates 35 to improve ASO delivery can also be found in the literature.
siRNA therapeutics share similarities to ASOs. siRNAs are doublestranded RNAs (dsRNAs) that are usually around 20 base pairs long.
They are used for gene downregulation or complete silencing. siRNAs comprise of a "passenger" strand and an "antisense" strand that is complementary to the mRNA sequence of interest, similar to ASOmediated RNA knockdown or steric translation inhibition ( Figure 1).
After entering cells, siRNA associates with the Argonaute 2 (AGO2) component of the RNA-induced silencing complex (RISC). 36 The passenger strand is then disposed and the antisense strand-coupled with AGO2 guides the RISC to the mRNA site that is complementary to the antisense strand. This either results in mRNA destruction, in the case of perfect complementarity, or translation inhibition, in the case of imperfect complementarity. 37 This results in decreased expression of the proteins encoded in the mRNA target. This process is shared by endogenous microRNA (miRNA) and as a result there has been substantial exploration of miRNA mimics. 38,39 miRNA mimics need to go through an additional modification process by an enzyme called Dicer before being loaded into the RISC. 40 Many of the previously mentioned ASO modifications or substitutions can be used for siRNA to enhance the stability and protect from endonucleases. For siRNA and miRNA mimics, however, delivery is one of the most significant challenges associated with their clinical translation. Like mRNA, there have been significant explorations into utilizing nanocarriers including liposomes, nanoparticles, and nucleic acid nanostructures, among others. 41,42 Additionally, bioconjugation to molecules like antibodies, aptamers, peptides, and lipids has also been exploited. 33,43,44 One of the most successful methods for siRNA delivery thus far has been conjugation with N-acetylgalactsamine (GalNAc). GalNAc binds the asialoglycoprotein receptor, which is highly expressed in hepatocytes, and allows for efficient uptake of siRNAs bound to it. 45 The final RNA therapeutics discussed here are aptamers. Aptamers are ssRNAs of 25-80 base pairs that incorporate hairpin folding to form highly specific binding surfaces (Figure 1), analogous to the antigen binding surfaces of antibodies. 46 They fold into favorable conformations based on complementary base pairing within individual oligo strands. Unlike the previously mentioned RNA therapeutics, aptamers do not need to be delivered into the cell cytosol to take affect and generally have extracellular protein targets. 47 Aptamers are being used as antagonists to block extracellular interactions, agonists for disease preventing receptors, and as targeted delivery systems for therapeutic molecules such as siRNA, proteins, and small molecule drugs. 46 While targeted delivery may not be a major challenge for aptamers because of their own high specificity, stability and clearance avoidance in vivo is. Some of the aforementioned nucleobase modifications and substitutions have been proven effective to solving the stability and quick clearance issues. Additionally, conjugation of biocompatible polymers, for example, polyethylene glycol (PEG), to the 5 0 end of the aptamer has helped with the clearance issue. 48 Notably, RNA therapeutics hold several advantages over other therapeutic modalities such as small molecules, antibodies, and DNA therapeutics. One major advantage is their capability to target undruggable targets that conventional therapeutics cannot. 49 RNA therapeutics, particularly ASOs and siRNAs, interact with their target via sequence-specific binding. This unique mechanism renders them capable of targeting both noncoding and coding RNAs, which small molecules and antibodies cannot easily achieve. 33 Because of this, F I G U R E 1 Summary of RNA therapeutic ongoing clinical trial landscape with a schematic representation of the mechanisms of action involved in different classes of RNA therapeutics. Created with BioRender.com RNA therapeutics are well suited to treat a broad spectrum of diseases including some orphan genetic disorders, which have no other effective therapeutic options. In addition, RNA therapeutics can be modular and versatile in the sense that the RNA sequence and/or delivery system can be easily modified to treat other diseases. 50 Unlike small molecules and antibodies, which require a long discovery and production process, new RNA therapeutics can be quickly designed and produced using existing modification methods and delivery technologies. This is best exemplified by the rapid development of mRNA-based COVID-19 vaccines, which employed new mRNA sequences but existing mRNA modification methods and LNPs for fast clinical testing. 51 Further, because RNA therapeutics can modulate the protein production/function from the RNA level, they may achieve longer-lasting effect and reduce administration frequency as compared to conventional therapeutics such as small molecules. This is exemplified by an approved siRNA product, inclisiran, which can maintain its effect for over 6 months following a single-dose administration. 52 Moreover, RNA therapeutics usually do not modify the patients' genome and therefore have relatively low risk of genotoxicity as compared to DNA therapeutics and gene editing technologies.
Gene editing therapies such as clustered regularly interspaced short palindromic repeats (CRISPR) and RNA editing can provide functions similar to ASO, siRNA, and mRNA using guide DNA/RNA coupled with CRISPR associated proteins (Cas) or adenosine deaminases acting on RNA (ADARs), respectively. 53 However, CRISPR's permanent genome editing can lead to genotoxicity when off-target genes are mutated. 54 Additionally, both CRISPR and RNA editing require more macromolecular machinery to be delivered to the right location inside the target cell than the RNA therapeutics do. 55 In contrast, RNA therapeutics can offer a safer means to treat genetic disorders.

| APPROVED PRODUCTS
Seventeen RNA therapeutic products using mRNA, ASOs, siRNA, or aptamers have been approved worldwide ( Table 2). These therapeutics are used to treat three main disease types: genetic, infectious, and physiological (diseases that cause organ dysfunction but do not fall into the genetic or infectious disease category). Notably, 12 of these approved products were granted orphan designation by the FDA.  56,57 The important immune cascades involved have been discussed in greater detail elsewhere. 58,59 Both vaccines code for identical amino acid sequences but differ in antigen-coding nucleic acid sequences. They also use proprietary 5' UTR and 3' UTR sequences 60 and lipid nanoparticle delivery vehicles. 61 LNPs aid in the transfection, endocytosis, and endosomal escape of mRNA in target cells, while the UTR modifications enhance the mRNA translation. Both Comirnaty (tozinameran) and Spikevax were given emergency use authorization by the FDA in December 2020 and granted approval in 2021 and 2022, respectively.

| Approved ASO therapeutics
Ten ASO products have been approved worldwide for treating genetic and infectious diseases ( Table 1). Nine of these products received orphan status by their respective regulatory agencies. The first RNA therapeutic ever approved and the only ASO approved for infectious diseases was Vitravene (fomivirsen). Vitravene was approved for treating cytomegalovirus (CMV) retinitis in immunocompromised patients. CMV-infected healthy patients are generally asymptomatic.
However, immunocompromised acquired immunodeficiency syndrome (AIDS) patients face more substantial systemic infection with severe inflammation in the eye causing blindness. 62 Vitravene was developed as an RNA-knockdown ASO, which targets the mRNA encoding the major immediate-early region of CMV resulting in decreased viral replication and load. Vitravene is injected intravitreally (IVT) to improve its local targeting to the retina. 63 However, it was discontinued due to adverse effects at the injection site and liver toxicity. 64 While Vitravene was unsuccessful in the clinic, its landmark approval introduced ASO therapeutics to the clinic. Nine ASO products have since received approvals, all for treating genetic diseases ( Table 1). Four of the approved ASOs including Exondys 51 (eteplirsen), Vyondys 53 (golodirsen), Viltepso (viltolarsen), and Amondys 45 (casimersen) were approved for treating Duchenne muscular dystrophy (DMD). DMD is an X-linked genetic disorder which affects mostly young boys. Muscle weakness is usually noticed at very young ages, with loss of ambulation occurring by age 12, and death between age 20 and 40. 65 When the gene encoding muscle dystrophin is mutated, the truncated dystrophin prevents proper muscle fiber connection and causes muscular dystrophy. 66 ASOs used to treat DMD all use splice modulation to exclude the mutated exon of interest resulting in a shortened dystrophin isoform, which can partially or fully restore function. 67 All four approved ASOs for DMD are phosphordiamidate morpholino oligonucleotides (PMOs). This alternative chemistry replaces the ribose backbone rings with morpholino rings, through a phosphordiamidate linkage, resulting in a more neutral backbone at the physiological pH. This chemistry helps to stabilize the ASOs and protect against proteolytic degradation and nonspecific protein bindinig. 33,68 Spinraza (nusinersen) is an ASO approved for treating another muscular genetic disorder, spinal muscular atrophy (SMA). 69

| Approved aptamer therapeutic
Macugen (pegaptanib) is the only aptamer therapeutic approved for treating neovascular macular degeneration (wet AMD). Wet AMD arises from angiogenesis toward the outer retina causing fluid accumulation in and around the retina. This can lead to blurred vision and eventually blindness if left untreated. 91 Macugen is administered IVT to bind to vascular endothelial growth factor 165 (VEGF165) blocking its interaction with vascular endothelial cells that would normally lead to neovascularization. As a result, it slows or stops the progression of wet AMD. 92 Macugen has both 2'-OMe and 2'-Fluoro ribose modifications for endonuclease protection and is conjugated to PEG to help prevent bulk clearance. However, it was discontinued in 2020 by the manufacturer due to undisclosed reasons.

| CURRENT CLINICAL TRIALS
We performed a search on clinicaltrials.gov to identify active clinical trials on RNA therapeutics. We performed searches in both the "Other Terms" and "Intervention/Treatment" categories, using the terms "RNA therapeutics," "mRNA," "siRNA," "ASO," "aptamers," "miRNA," and alternative versions (e.g., messenger RNA, small interfering RNA, etc.). Under the "Recruitment Status" section we checked "Not yet recruiting," "recruiting," "enrolling by invitation," and "Active, not recruiting." Also, in the "Study Type" category, we selected "Interventional Studies (Clinical Trials)." We excluded studies that were listed as "Not Applicable" in the "Phase" category. A total of 415 trials were initially identified. We then excluded 43 trials that were regarding cell therapies that incorporated one of the listed RNA therapeutics. Additionally, 189 trials were excluded because they mentioned one of the RNA types but were not regarding treatments,      Table 2.  (Table 3). 96 Table 3). The physiological disease-treating ASOs include trials for liver-targeted lipid down-regulators, and intrathecally injected treatments for Amyotrophic Lateral Sclerosis (ALS) and Progressive Supranuclear Palsy (Table 3). There is only one trial (NCT05018533) with a broad indication.

| Current ASO clinical trials
The administration method for these ASO trials is mostly IV Phase 1 trial is for broad indications but will aim to treat severe asthma in later phases. It utilizes an ASO that limits MEX3B, an RNA binding protein, whose inhibition could have extensive applications in oncology and infectious disease. 100 Modifications and substitutions with no carrier (72.7%) are still heavily favored as delivery systems for ASOs (Figure 5c). A common modification in these ASO trials that we have not previously mentioned is the constrained ethyl bridge nucleic acid substitution (cEt), which provides nuclease protection and enhances complementary binding. 6 We broke down the substitution (a) (c) (b) F I G U R E 5 Landscape of 44 ongoing ASO therapeutic clinical trials. The trials were analyzed based on (a) disease type, (b) administration route, and (c) delivery system and modification methods even further ( Figure S4). Of the ASO trials that utilized modifications or substitutions, 65.6% had a PS backbone modification, 56.3% had a 2 0 ribose substitution (either 2'-MOE or 2'-OMe), and 15.6% had a cEt bridged nucleic acid substitution. Liposomes (6.8%) and conjugates (9.1%) are also being explored in the clinic as ASO delivery carriers (Figure 5c). Selected ongoing ASO therapeutic clinical trials can be found in Table 3.  Table 4.

| Other clinical trials
Six of the identified RNA therapeutic trials involve RNA types beyond mRNA, ASOs, and siRNA ( Table 5) Functioning similar to mRNA, srRNA and saRNA can self-replicate to increase antigen expression and thus elicit better immune responses. (a) (c) (b) F I G U R E 6 Landscape of 54 ongoing siRNA therapeutic clinical trials. The trials were analyzed based on (a) disease type, (b) administration route, and (c) delivery system most stubborn impurities. Affinity chromatography, on the other hand, produces highly purified RNA, but has a low throughput slowing down large-scale manufacturing. 107 A major step that the industry can take to alleviate the purification challenges is to implement continuous manufacturing practices, which have been shown to make a significant impact on therapeutic protein production. 108 While the additional step of LNP encapsulation with microfluidic devices, which has been performed at the largest of scales with Comirnaty and Spikevax, in non-disease-causing organs. As discussed in Section 3, two approved ASO products (Vitravene and Kynamro) were discontinued due to hepatic toxicity. 64,80 Additionally, there have been a number of reports that risk of myocarditis is increased with Comirnaty and Spikevax vaccination, [115][116][117] but additional retrospective analysis with larger sample cohorts and less voluntary reporting should be performed to quantify the true added risk. One mitigation strategy for off-target toxicity is highly selective targeted delivery, which GalNAc-conjugates seem to have achieved for hepatic-associated diseases.

| TRANSLATIONAL CHALLENGES
But the approved GalNAc therapies, Givlaari, Oxlumo, and Leqvio, have only been on the market for a couple of years, so time will tell whether long-term dosing is safe. Better models of human and interorgan biology can help to bridge the gap in safety assessments.
Researchers have long been working toward organ-on-a-chip and eventually body-on-a-chip in vitro models that can aid in understanding the impacts of new medcines, 118 if successful this will prove critical to elucidating the safety and efficacy of RNA therapeutics.
Additional challenges that RNA therapeutics face is the storage and stability issue. Generally, cold chain conditions are used to mitigate degradation and aggregation of RNA therapeutic molecules and their carriers. 119 Currently, for long-term storage, Comirnaty and Spikevax must be kept at À90 to À60 C and À50 to À15 C, respectively. These are temperature requirements that some medical facilities, especially ones in rural areas or developing countries, do not have the means to adhere to. Comirnaty and Spikevax can be stored in a refrigerator (2-8 C) for 30 days after thawing but must be used within 30 min and 24 h, respectively, after removal from cold chain. In contrast, all the approved ASOs and siRNAs are stored in the refrigerator until vial expiry, with some capable of being stored at room temperature (below 25 C) for up to 6 weeks (Tegsedi and Waylivra) or longer (Leqvio). However, the storage-stability profiles for ASOs and siRNAs could change with the approval of different drug carriers.
Moderna's next generation refrigerator-stable (2-5 C) COVID-19 vaccine (mRNA-1283), which is in Phase 2 trials (NCT04813796 and NCT05137236), could be pivotal for mRNA use in communities where refrigerator storage is more accessible than freezer storage. However, with research in the past largely focusing on naked mRNA storage, 120 additional efforts need to be made to push the stability and storage boundaries of encapsulated RNA therapeutics.

| CONCLUSION AND OUTLOOK
Despite all the challenges, RNA therapeutics are emerging as a major therapeutic modality. The ability to control protein expression has therapeutics to the clinic, the 2010s provided only a few approvals, mostly in orphan genetic diseases, the early progress since 2020 and clinical trials highlighted here show that the next decade should prove monumental in the evolution of RNA therapeutics. While this process will be met with challenges in manufacturing, delivery, and safety, it will be an exciting process to see unfold in preclinical and clinical trials.

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed.