RNA-targeted small molecules vs. ASOs in neurodegeneration

How Each Platform Engages RNA Targets

Small-molecule splice modulators and antisense oligonucleotides (ASOs) both act on RNA, but they do so through fundamentally different molecular mechanisms. Risdiplam and Branaplam do not base-pair with their targets — instead, they stabilise the RNA structural duplex formed between the SMN2 pre-mRNA 5′ splice site and the U1 small nuclear ribonucleoprotein (U1 snRNP) complex, potentiating the activity of endogenous splicing factors to promote exon 7 inclusion and increase functional SMN protein production.

ASOs, by contrast, achieve target engagement through Watson–Crick base-pairing with complementary RNA sequences at single-nucleotide resolution. Nusinersen is an 18-mer 2′-MOE phosphorothioate ASO that sterically blocks the ISS-N1 intronic splicing silencer — the same exon 7 inclusion endpoint, reached by an entirely different molecular route. Tofersen operates through a distinct ASO subtype: as a 20-mer 5-10-5 MOE gapmer, it recruits RNase H1 to degrade SOD1 mRNA rather than modulating splicing at all. Chemical modifications — 2′-MOE substitutions and phosphorothioate backbones — are essential for conferring metabolic stability and optimising pharmacokinetic properties in both ASO agents.

Risdiplam’s binding sites encompass both the 5′ splice site and an upstream GA-rich purine-rich element. Branaplam engages distinct binding sites with different sequence preferences from Risdiplam, giving rise to a unique off-target splicing profile — a distinction with significant clinical consequences, as discussed below. According to research published by Nature, structural characterisation of small-molecule–RNA–protein ternary complexes has become a central focus for understanding and improving splice modulator selectivity.

CNS Delivery: Oral Bioavailability vs. Intrathecal Injection

The route of administration represents the starkest practical difference between the two platforms. Risdiplam was the first FDA-approved oral treatment for SMA, administered once daily. Following oral dosing, small molecules readily cross the blood–brain barrier and achieve therapeutic concentrations across the CNS, skeletal muscle, and peripheral organs — without any invasive procedure. This systemic biodistribution allows Risdiplam to address peripheral manifestations of SMA including cardiac, hepatic, and skeletal muscle pathology that intrathecal therapies cannot reach.

ASOs face a fundamental barrier: as large, charged molecules, they cannot efficiently cross the blood–brain barrier following systemic administration. Both Nusinersen and Tofersen are therefore delivered via lumbar puncture directly into the cerebrospinal fluid. Nusinersen’s loading regimen consists of doses at days 0, 14, 28, and 63 (12 mg each), followed by maintenance every 4 months. Tofersen requires loading doses at days 0, 14, and 28 (100 mg each), followed by maintenance every 4–8 weeks. Standards for intrathecal drug delivery are monitored by regulatory bodies including the FDA and the EMA.

“ASOs distribute primarily throughout the spinal cord and superficial cortical regions; drug concentrations in deep brain structures such as the basal ganglia are substantially lower, following a medial-to-lateral concentration gradient.”

This distributional constraint carries important clinical implications for diseases like Huntington’s disease, where pathology is concentrated in the striatum and other deep subcortical structures. Repeated lumbar punctures also impose procedural burden and carry risks of infection, CSF leakage, and post-lumbar puncture headache — factors that affect long-term patient adherence. The WHO has highlighted patient-centricity in treatment design as a key priority in neurodegenerative disease management.

Durability of Effect and Pharmacokinetics: Short Half-Life vs. Months-Long Tissue Retention

ASOs and small-molecule splice modulators sit at opposite ends of the pharmacokinetic spectrum. ASOs have a tissue half-life of approximately 20 days in brain parenchyma, with spinal cord retention extending to several months. Following a single intrathecal dose, suppression of target mRNA can persist for at least 91 days and in some cases beyond 6 months. Repeated dosing leads to progressive drug accumulation in CNS tissue, further prolonging therapeutic effect — enabling Nusinersen’s maintenance schedule of every 4 months.

Small-molecule splice modulators have a comparatively short half-life, necessitating once-daily dosing to maintain therapeutic plasma and tissue concentrations. Pharmacodynamic effects dissipate rapidly upon discontinuation, requiring uninterrupted treatment to sustain SMN protein levels. The corollary advantage is significant dose flexibility: the short half-life permits rapid upward or downward adjustments in response to clinical findings or tolerability concerns. For ASOs, the long tissue residence that enables infrequent dosing also means that adverse effects, once established, are difficult to reverse rapidly.

Selectivity, Off-Target Effects, and Clinical Terminations

Off-target risk profiles diverge sharply between the two platforms, with real-world clinical consequences. Risdiplam and Branaplam can alter the splicing of hundreds to thousands of genes across the transcriptome — a dose-dependent phenomenon involving aberrant exon inclusion, exon skipping, and intron retention. The clinical impact was most acute with Branaplam: the VIBRANT-HD trial in Huntington’s disease was terminated due to drug-induced peripheral neuropathy, attributed mechanistically to p53 pathway activation, nucleolar stress, and upregulation of the pro-apoptotic gene BBC3.

Next-generation splice modulators such as TEC-1 are being developed with improved selectivity profiles, and low-dose combination regimens are being explored to balance efficacy against tolerability. ASOs offer substantially higher sequence specificity — Watson–Crick base-pairing enables discrimination at single-nucleotide resolution, and allele-selective ASOs are under active development for HTT and NEFL mutations to silence mutant alleles while sparing wild-type transcripts. Residual off-target risks for ASOs include homology-dependent engagement of partially complementary non-target transcripts, and acute CNS toxicity with certain sequences — particularly those with high guanine content or propensity to form secondary structures following intraventricular administration. Research from NIH has characterised the structural basis for both on-target and off-target ASO interactions in CNS tissue.

“Small-molecule splice modulators can alter the splicing of hundreds to thousands of genes across the transcriptome — a dose-dependent liability that led to the termination of both the VIBRANT-HD trial (Branaplam) and the Phase III tominersen programme.”

Disease-by-Disease Comparison: SMA, ALS, and Huntington’s Disease

Spinal Muscular Atrophy (SMA)

SMA is the disease where both platforms have achieved the most robust clinical validation. Both Nusinersen and Risdiplam produce significant improvements in motor function and survival, but through different routes and with different tissue coverage profiles. Nusinersen’s ISS-N1 blockade and Risdiplam’s U1 snRNP stabilisation converge on the same exon 7 inclusion endpoint — but Risdiplam’s systemic distribution additionally addresses cardiac, hepatic, and skeletal muscle pathology that intrathecal delivery cannot reach.

Amyotrophic Lateral Sclerosis (ALS)

In ALS, Tofersen is the only approved RNA-targeting therapy, having received accelerated approval based on its ability to reduce CSF and plasma neurofilament light chain (NfL) — a biomarker of neurodegeneration. Tofersen targets SOD1 mRNA for RNase H1-mediated degradation at a dose of 100 mg intrathecally. Although the VALOR trial did not meet its primary endpoint (ALSFRS-R score change), open-label extension data indicate that early treatment initiation attenuates disease progression and may extend survival. No small-molecule splice modulators are currently approved for ALS, though investigational approaches targeting STMN2 splicing modulation are under study. Regulatory frameworks for accelerated approval in ALS are overseen by bodies including the FDA.

Huntington’s Disease

Huntington’s disease has seen setbacks for both platforms. Tominersen — a non-allele-selective ASO targeting both mutant and wild-type HTT — was halted in Phase III due to clinical worsening and progressive ventricular enlargement. Branaplam’s mechanism in Huntington’s disease involved inducing inclusion of a cryptic exon harbouring a premature stop codon to trigger nonsense-mediated decay of HTT mRNA — but the VIBRANT-HD trial was terminated due to peripheral neuropathy. Allele-selective ASO programmes, designed to silence only the mutant HTT allele while sparing the wild-type, remain in active development and represent the most promising next step for ASO-based approaches in this indication.

Next-Generation Strategies and Combination Approaches

Both platforms are converging toward improved selectivity, better delivery, and rational combination design. For small molecules, the primary focus is structural optimisation to reduce transcriptome-wide off-target splicing — exemplified by TEC-1, a next-generation splice modulator designed with a refined specificity profile. Novel delivery systems to enhance CNS penetration are also under investigation. For ASOs, allele-selective engineering is the most clinically urgent priority, alongside advanced delivery platforms including lipid nanoparticles, apolipoprotein-based nanodiscs, intranasal administration routes, and continuous intrathecal infusion systems to optimise pharmacokinetic profiles.

The most intellectually compelling near-term strategy may be combination therapy. Low-dose co-administration of Risdiplam and Nusinersen is being explored as a synergistic approach: because the two platforms operate through mechanistically distinct pathways — splicing modulation versus mRNA degradation — they are theoretically well-positioned for complementary combination use, potentially augmenting SMN protein expression beyond either agent alone while attenuating the respective off-target liabilities of each. Innovation in this space is tracked by global IP bodies including WIPO, which publishes annual trend data on RNA therapeutics patent filings.

Several important caveats should be acknowledged when interpreting the comparative evidence. Patent data are subject to an 18-month publication delay, meaning the most recent developments may not yet be fully captured in the public literature. Substantial variation in trial design, endpoints, and patient populations limits direct cross-study comparisons. Long-term safety profiles — including potential retinal toxicity with Risdiplam and neurotoxicity with certain ASO sequences — require extended follow-up to characterise fully. Patient genotype, disease severity, and timing of treatment initiation are major determinants of therapeutic outcome across both platforms.