Decades of work on RNA targeting, chemical modification, and tissue-specific delivery are reshaping what programmable genetic medicines can treat and where they may go next.
Review: Therapeutic oligonucleotides as genetic medicines: The current state and future perspective. Image Credit: Love Employee / Shutterstock
In a recent review in press with the journal Cell Reports Medicine, a group of authors examined the mechanisms, chemical modifications, delivery strategies, clinical development, and future applications of therapeutic oligonucleotides as genetic medicines.
Background
At the time of writing, 21 therapeutic oligonucleotides had received regulatory approval, with two later withdrawn, demonstrating a new approach to treating genetic diseases. Therapeutic oligonucleotides are chemically modified nucleic acid molecules developed to alter gene expression by interacting with a specific ribonucleic acid (RNA) sequence. Antisense oligonucleotides (ASOs) can induce RNA degradation or alter pre-messenger RNA (pre-mRNA) processing, whereas small interfering RNAs (siRNAs) promote the degradation of target messenger RNA (mRNA). Advances in chemical modification and delivery have improved their stability, efficacy, durability, and target specificity, but major challenges remain in delivery beyond the liver, safety, and the design of individualized treatments.
Molecular mechanisms and chemical refinement
Therapeutic oligonucleotides generally contain a phosphate backbone, ribose sugar, and nitrogenous base. ASOs are usually single-stranded molecules 15–30 nucleotides long. Some recruit ribonuclease H (RNase H) to degrade targeted RNA. Others act as steric blockers by binding to pre-mRNA and altering splicing, thereby restoring protein production or switching protein isoforms. siRNAs are approximately 21 nucleotides long and work through the RNA-induced silencing complex (RISC). Their guide strand directs Argonaute 2 (Ago2) to complementary mRNA, promoting cleavage. MicroRNA (miRNA) mimics use related RNA interference mechanisms but can regulate multiple targets. Small activating RNAs (saRNAs) and antagomirs are briefly discussed as approaches to increase gene expression.
Chemical modification has been central to clinical progress because unmodified deoxyribonucleic acid (DNA) and RNA are rapidly degraded and cleared. Phosphorothioate (PS) backbone modifications improve nuclease resistance, protein binding, and tissue uptake. Ribose modifications, including 2′-methoxyethyl (2′-MOE), locked nucleic acid (LNA), and constrained ethyl (cEt), can increase stability and target binding. But excessive LNA and cEt use has been associated with liver toxicity.
Phosphorodiamidate morpholino oligomers (PMOs) are highly nuclease-resistant and neutral but can have poor systemic bioavailability because they lack a PS backbone. For siRNAs, 2′-O-methyl (2′-OMe), 2′-fluoro (2′-F), and extended nucleic acid (exNA) modifications improve stability and activity, while specific patterns help preserve RISC function.
Delivery remains a major barrier
Oligonucleotides are relatively large and negatively charged, so they do not readily cross cell membranes. Effective therapy requires protection from degradation, avoidance of clearance, access to target tissues, cellular uptake, and escape from lysosomes. Local administration can bypass some barriers. Examples of approved or previously approved therapies include intravitreal delivery to the eye and intrathecal administration to the central nervous system (CNS). Systemic administration is used when distribution through the circulation can reach the target tissue. Subcutaneous administration is widely used because it is less invasive and does not require specialized personnel.
Delivery systems include lipids, polymers, inorganic materials, peptides, and biologically derived carriers. Lipid-based systems are clinically advanced, and patisiran, an siRNA therapy formulated in a lipid nanoparticle, was approved for transthyretin-mediated amyloidosis. Ligand conjugation can provide selective delivery. N-acetylgalactosamine (GalNAc) binds the asialoglycoprotein receptor (ASGPR), which is highly expressed on hepatocytes, making GalNAc conjugates effective for liver delivery. Antibody-oligonucleotide conjugates have reached clinical trials for muscle delivery, while oligonucleotide approaches for transport across the blood-brain barrier remain preclinical. Lipid conjugates are being investigated for applications in the liver, muscle, CNS, placenta, and skin. Peptide-based approaches have shown potential for cellular uptake and endosomal escape, but toxicity has limited some clinical development.
Clinical progress and expanding applications
At the time of the review, 21 therapeutic oligonucleotides had received regulatory approval from the United States Food and Drug Administration (FDA), the European Medicines Agency (EMA), and/or Japan’s Pharmaceuticals and Medical Devices Agency (PMDA), of which two were later withdrawn. The approval history includes therapies for spinal muscular atrophy, Duchenne muscular dystrophy, hereditary transthyretin amyloidosis, amyotrophic lateral sclerosis, hypercholesterolemia, primary hyperoxaluria, hemophilia, and severe hypertriglyceridemia. Nusinersen increased production of survival motor neuron protein in spinal muscular atrophy and reduced the risk of death or permanent ventilation in severe type 1 disease. Four PMO therapies were approved for eligible Duchenne muscular dystrophy mutations through exon-skipping strategies, but confirmatory evidence that these therapies slow disease progression remained lacking.
Inclisiran, a GalNAc-conjugated siRNA targeting proprotein convertase subtilisin/kexin type 9 (PCSK9), produced approximately 50% reductions in low-density lipoprotein cholesterol (LDL-C) in phase 3 trials, with effects sustained for at least 18 months. MicroRNA mimics and saRNAs are also being evaluated in early clinical trials. Other approved siRNA therapies target transthyretin, lactate dehydrogenase A, antithrombin, and apolipoprotein C3.
Future Directions
Artificial intelligence (AI) may support target selection, sequence design, chemical modification, delivery refinement, manufacturing, and safety assessment. The review emphasizes that AI will not replace empirical testing and will depend on large, high-quality datasets, including information from unsuccessful or unsafe treatments. Therapeutic oligonucleotides are also moving toward individualized treatment. Milasen, developed for one patient with CLN7 Batten disease, provided a proof of concept, and more than 40 different ASOs have since been developed for more than 80 patients. These approaches require accelerated development, regulatory processes, infrastructure, monitoring, and data sharing.
The field is also expanding beyond rare diseases. Inclisiran established a role in common hypercholesterolemia, while additional ASOs and siRNAs directed against PCSK9, ANGPTL3, and apolipoprotein C3 (APOC3) are in development, and the lipoprotein(a)-lowering ASO pelacarsen is being studied in inclisiran-treated patients. Continued expansion will depend on validating useful targets and overcoming delivery barriers to additional tissues.
Conclusions
Therapeutic oligonucleotides have progressed from an experimental concept to an established therapeutic class of genetic medicines, supported by advances in chemistry, delivery, and understanding of RNA biology. Approved ASOs and RNA interference therapies now address several rare and common disorders, while individualized treatments demonstrate the flexibility of programmable genetic medicines.
The review identifies delivery beyond the liver, safety improvement, data quality, regulatory pathways, and target validation as continuing priorities. AI may accelerate several stages of development, but empirical testing is still required. Continued progress in delivery and chemistry could broaden the tissues, diseases, and patients reached by these therapies. Further research is needed to improve extrahepatic delivery and expand applications.
