Disease-Modifying Therapies for Huntington's Disease
Huntington's disease (HD) is caused by a CAG trinucleotide repeat expansion (≥40 repeats) in the HTT gene on chromosome 4, encoding a mutant huntingtin protein (mHTT) that is toxic to striatal medium spiny neurons and, progressively, to cortical neurons. Unlike Parkinson's disease, where the causative molecular target remains debated, HD has a single, well-defined genetic cause — making it a compelling candidate for disease-modifying therapy. The central hypothesis driving the field is straightforward: if mutant huntingtin is the root cause, lowering its levels should slow or prevent neurodegeneration. Yet translating this logic into effective therapy has proven far more challenging than anticipated. This article reviews the major disease-modifying strategies under investigation, the clinical trial landscape through early 2026, and the critical lessons learned from both successes and failures.
Bottom Line: HD Disease Modification
- AMT-130 (uniQure) is the first therapy to demonstrate statistically significant slowing of HD progression. In the pivotal Phase I/II trial, high-dose AMT-130 (AAV5-delivered microRNA) showed a 75% slowing of disease progression on cUHDRS at 36 months vs external control (p=0.003). Received FDA Breakthrough Therapy designation (April 2025) and RMAT designation (June 2024). However, the FDA reversed course in late 2025, indicating Phase I/II data alone may not support BLA submission.
- Tominersen (Roche) remains alive but chastened. The Phase 3 GENERATION-HD1 trial was halted in March 2021 for futility and safety concerns — higher doses worsened outcomes. Post-hoc analyses suggested possible benefit in younger, milder patients, leading to the redesigned Phase 2 GENERATION-HD2 trial (100 mg Q16W in ages 25–50), expected to complete in 2026.
- Small molecule HTT-lowering drugs are advancing rapidly. PTC518 (votoplam) met its primary endpoint in PIVOT-HD with dose-dependent HTT protein lowering and favorable NfL trends at 24 months. SKY-0515 (Skyhawk) achieved 72% HTT mRNA reduction in Phase 1 and has entered Phase 2/3 (FALCON-HD). Both are oral agents, a major practical advantage.
- Somatic CAG repeat instability has emerged as a transformative new target. DNA mismatch repair genes — especially MSH3 — drive ongoing CAG expansion in striatal neurons, and genetic modifier studies show these genes influence age of onset more than the inherited CAG length itself. ASOs and small molecules targeting MSH3 are in preclinical/early clinical development.
- Allele-selective approaches preserve wild-type HTT. WVE-003 (Wave Life Sciences) demonstrated the first allele-selective mHTT reduction in humans (46% CSF mHTT lowering); a Phase 2/3 registrational study is planned.
- The Huntington's Disease Integrated Staging System (HD-ISS) provides a biological framework for staging from birth (Stage 0), enabling trials in presymptomatic populations and standardizing endpoints across the field.
Huntingtin-Lowering: Rationale and Approaches
The Central Hypothesis
The gain-of-function toxicity of mutant huntingtin protein provides a clear therapeutic rationale: lowering mHTT should be neuroprotective. mHTT causes harm through multiple mechanisms — protein aggregation, transcriptional dysregulation, mitochondrial dysfunction, impaired autophagy, and excitotoxicity. Preclinical studies in animal models have consistently shown that reducing mHTT expression can prevent or reverse neuropathology, even after disease onset.
However, two critical questions have shaped the field:
- Total vs. allele-selective lowering: Wild-type huntingtin has important cellular functions (embryonic development, BDNF transport, vesicle trafficking, transcriptional regulation). Nonselective approaches that lower both mutant and wild-type HTT risk losing these protective functions. Allele-selective approaches target only the mutant allele but are applicable only to subsets of patients carrying specific targetable SNPs
- Degree and timing of lowering: Animal data from Roche suggest that sustained lowering beyond ~50% of total HTT may be detrimental. The concept of a "huntingtin holiday" — periodic rather than continuous lowering — has gained traction after the GENERATION-HD1 experience
Delivery Methods
| Delivery Route | Modality | Advantages | Disadvantages |
|---|---|---|---|
| Intrathecal injection | ASOs (tominersen, WVE-003) | Direct CNS delivery; adjustable dosing; reversible | Repeated lumbar punctures (Q8–16W); limited brain parenchymal penetration; neuroinflammatory potential |
| Intrastriatal injection | Gene therapy (AMT-130) | One-time administration; targets affected brain region; sustained expression | Neurosurgical procedure; irreversible; limited distribution from injection site |
| Oral / systemic | Small molecules (PTC518, SKY-0515) | Non-invasive; adjustable dosing; wide CNS distribution; reversible | Peripheral HTT lowering (unknown long-term effects); potential off-target splicing effects |
Antisense Oligonucleotides (ASOs)
Tominersen (Roche/Genentech) — Allele-Nonselective
Tominersen (formerly IONIS-HTTRx, then RG6042) is an intrathecally delivered ASO that targets total HTT mRNA — both mutant and wild-type alleles. It was the first huntingtin-lowering drug to enter large-scale clinical trials and remains the most thoroughly studied agent in the field.
GENERATION-HD1 (Phase 3) — Halted March 2021
- Design: 791 patients with manifest HD, randomized to tominersen 120 mg Q8W (every 8 weeks), Q16W (every 16 weeks), or placebo via intrathecal injection. Primary endpoint: cUHDRS at 25 months
- Outcome: Stopped early by the independent data monitoring committee (iDMC) based on futility and an unfavorable risk-benefit profile
- Key findings:
- Dose-dependent CSF mHTT reduction was achieved (confirming target engagement)
- The Q8W group performed worse than placebo on clinical rating scales, with higher rates of serious adverse events
- Significant ventricular enlargement was observed, dose-dependently — interpreted as possible neuroinflammation or neurotoxicity rather than simple CSF dynamics
- Transient rises in CSF neurofilament light chain (NfL) were observed with higher doses, a signal that was present even in the Phase 1/2 open-label extension and was, in retrospect, a warning sign
- Published: Results formally published in the New England Journal of Medicine in 2024
Why GENERATION-HD1 Failed: Key Lessons
- Excessive HTT lowering is harmful. Roche's animal data indicated that frequent, high-dose tominersen lowered total huntingtin excessively (>50%), with negative consequences. Wild-type HTT has essential cellular roles that cannot be eliminated without consequence
- Neuroinflammation from ASO exposure. Intrathecal ASOs can trigger innate immune activation independent of target engagement. The transient NfL spikes likely reflected ASO-induced neuronal stress or inflammatory damage to vulnerable neurons
- Patient selection matters. The broad enrollment (manifest HD at any age) may have included patients too advanced to benefit, or patients whose disease biology made them vulnerable to harm
- Dose and frequency were too aggressive. The Q8W arm performed worst, while post-hoc analyses of the Q16W arm suggested possible benefit in specific subgroups
Post-Hoc Analyses and the Path to GENERATION-HD2
- Post-hoc analysis of GENERATION-HD1 identified a subgroup — younger adults (ages 25–50) with less severe symptoms — who may have benefited from tominersen at the Q16W dose
- These findings are inherently exploratory (the trial was not designed to answer this question), but they provided the rationale for a redesigned trial
GENERATION-HD2 (Phase 2) — Ongoing
- Design: Proof-of-concept study in patients aged 25–50 with prodromal or early manifest HD (CAP score 400–500)
- Dosing: Tominersen 100 mg Q16W intrathecally (amended from initial 60 mg/100 mg arms to 100 mg only after interim iDMC review found no safety concerns)
- Status: Ongoing; expected to complete in 2026. The trial tests the "huntingtin holiday" hypothesis — that periodic, lower-dose HTT suppression in younger, less-affected patients may produce net benefit
WVE-003 (Wave Life Sciences) — Allele-Selective
WVE-003 is a stereopure ASO that selectively targets the mutant HTT allele by binding to a single nucleotide polymorphism (SNP rs362307) that is linked in cis to the expanded CAG repeat on the disease allele. This SNP is present in approximately 40% of HD patients, defining the eligible population. The approach preserves wild-type HTT expression.
SELECT-HD (Phase 1b/2a)
- Single-dose results (September 2022): Mean CSF mHTT reduction of 22% (median 30%) at 85 days after single doses of 30 or 60 mg, with wild-type HTT levels preserved (confirming allele selectivity)
- Multi-dose results (June 2023): Mean CSF mHTT reduction of 46% at 8 weeks post-last dose and 44% at 12 weeks post-last dose in the multi-dose cohort (p=0.0007 and p=0.0002 vs placebo, respectively)
- Statistically significant correlation between mHTT reduction and slowing of caudate atrophy was observed
- Safety: Generally well tolerated; no drug-related serious adverse events
Regulatory and Development Status
- FDA granted Orphan Drug Designation to WVE-003 (November 2024)
- Wave has received supportive FDA feedback regarding a potential accelerated approval pathway
- A global, potentially registrational Phase 2/3 study is being planned, with IND submission expected in the second half of 2025
Historical Note: Earlier Wave Life Sciences ASOs
WVE-003 is Wave's third-generation HD ASO. Its predecessors, WVE-120101 and WVE-120102, were allele-selective ASOs tested in the PRECISION-HD1 and PRECISION-HD2 Phase 1b/2a trials. Both were discontinued in March 2021 after showing no dose-dependent mHTT lowering and no statistically significant target engagement. WVE-003 incorporates Wave's proprietary stereopure chemistry (PN backbone modification) that significantly improves potency and tissue distribution compared to first-generation designs.
Gene Therapy and RNA Interference
AMT-130 (uniQure) — AAV5-Delivered microRNA
AMT-130 is an adeno-associated virus serotype 5 (AAV5) vector delivering an engineered microRNA (miHTT) that silences HTT mRNA. It is administered as a one-time bilateral intrastriatal injection via MRI-guided stereotactic neurosurgery. The therapy is designed to produce sustained, long-term HTT silencing from a single procedure, eliminating the need for repeated dosing.
Phase I/II Pivotal Trial — Landmark Results (September 2025)
- Design: Open-label, dose-escalation study with two dose cohorts (low dose: 6×1012 vg; high dose: 6×1013 vg) in patients with early manifest HD. Primary endpoint: cUHDRS change at 36 months vs propensity score-matched external control from the Enroll-HD natural history database
- Primary endpoint met: High-dose AMT-130 demonstrated a statistically significant 75% slowing of disease progression on cUHDRS at 36 months (mean change −0.38 vs −1.52 in external control; p=0.003)
- Key secondary endpoint met: Statistically significant slowing on Total Functional Capacity (TFC) at 36 months vs external control
- Biomarker data: CSF neurofilament light chain (NfL) decreased by 8.2% from baseline in the high-dose group at 36 months — a favorable trend given that NfL typically rises in HD progression
- Safety: Most adverse events were related to the surgical procedure and resolved. No new drug-related serious adverse events observed since December 2022 (as of June 2025 data cutoff)
Clinical Significance of AMT-130 Results
September 2025 marked the first time any therapy demonstrated statistically significant slowing of Huntington's disease progression. This is a watershed moment for the HD community. However, the study used an external natural history control (not a randomized placebo arm), and the sample size was small. The FDA's subsequent reversal on accepting these data for BLA submission underscores the regulatory challenges of single-arm, externally controlled designs — even with breakthrough designations.
Regulatory Journey
- June 2024: FDA granted RMAT (Regenerative Medicine Advanced Therapy) designation
- April 2025: FDA granted Breakthrough Therapy designation
- November 2024: FDA initially agreed that Phase I/II data with external control could support BLA submission under accelerated approval
- October 2025 (pre-BLA meeting): FDA reversed course, indicating Phase I/II data were insufficient for BLA. The agency no longer agreed that external control comparisons from existing studies could serve as primary evidence
- December 2025: Official FDA meeting minutes confirmed the challenging path ahead
- Path forward: uniQure is seeking a follow-up FDA meeting in Q1 2026 to determine next steps. Additional controlled studies may be required
Other Gene Therapy Approaches
| Program | Sponsor | Mechanism | Status |
|---|---|---|---|
| SPK-10001 | Spark Therapeutics | Engineered AAV delivering artificial miRNA targeting HTT mRNA | Preclinical: dose-dependent reductions of up to 40% HTT mRNA and 70% HTT protein in HD mouse models, stable over 12 months |
Small Molecule Huntingtin-Lowering Agents
Oral small molecules that reduce HTT expression represent a major advance in practical drug delivery. Two leading programs — PTC518 and SKY-0515 — use RNA splicing modification to trigger HTT mRNA degradation, offering systemic CNS distribution without invasive procedures.
PTC518 / Votoplam (PTC Therapeutics)
PTC518 is an oral small molecule splicing modifier that promotes inclusion of a novel pseudoexon in HTT pre-mRNA. This pseudoexon contains a premature termination codon, triggering nonsense-mediated mRNA decay (NMD) and consequent reduction in HTT protein levels.
PIVOT-HD (Phase 2) — Primary Endpoint Met (May 2025)
- Design: Randomized, double-blind, placebo-controlled study in Stage 2 and Stage 3 HD patients, evaluating multiple dose levels (5 mg and 10 mg)
- Primary endpoint met: Statistically significant reduction in blood HTT protein levels at Week 12 (p<0.0001)
- 24-month clinical data:
- Favorable dose-dependent trends on cUHDRS, TFC, and Symbol Digit Modalities Test (SDMT) relative to natural history
- Dose-dependent plasma NfL lowering: −8.9% at 5 mg (nominal p=0.12) and −14% at 10 mg (nominal p=0.03) from baseline at Month 24
- Safety: Favorable safety and tolerability across all dose levels and disease stages. No treatment-related serious adverse events. No NfL spikes (a critical differentiator from the tominersen experience)
SKY-0515 (Skyhawk Therapeutics)
SKY-0515 is an orally administered RNA splicing modifier developed using Skyhawk's SKYSTAR platform. It modulates RNA splicing to reduce production of both huntingtin and PMS1 proteins — the latter being a DNA mismatch repair gene implicated in somatic CAG repeat instability (discussed below), potentially offering dual mechanism benefit.
Phase 1 and FALCON-HD (Phase 2/3)
- Phase 1 healthy volunteer data: Dose-dependent HTT mRNA reduction, achieving an average of 72% lowering at the highest dose
- Phase 1 patient data (9-month interim):
- Mean cUHDRS improvement of +0.64 points from baseline (vs expected natural history worsening of −0.73 points over 9 months)
- 62% reduction in blood mHTT protein at the 9 mg dose
- 26% reduction in PMS1 mRNA (a somatic instability modifier — see below)
- FALCON-HD (Phase 2/3): First patient dosed; randomized, double-blind, placebo-controlled dose-ranging study. Enrolling 120 patients in Australia/New Zealand (Stage 2 and early Stage 3 HD) and 400 patients at 40+ global sites. Over 90 patients dosed as of January 2026
Branaplam (Novartis) — Discontinued
Branaplam was originally developed as an oral splicing modifier for spinal muscular atrophy (SMA). HTT-lowering activity was observed as a secondary pharmacological effect, prompting Novartis to pursue HD as a primary indication.
- VIBRANT-HD (Phase 2b): Dosing was suspended in August 2022 after detection of peripheral neuropathy in some participants, along with elevated NfL levels suggesting nerve damage
- Novartis confirmed discontinuation of the entire HD program based on an "overall assessment of the risk-benefit profile"
- The company concluded it would be "very unlikely that lower doses or different dose timings would be safe and lower [mHTT] sufficiently to slow disease progression"
- Lesson: Branaplam's failure highlights the risk of off-target splicing effects with small molecule splicing modifiers — peripheral neuropathy was likely an off-target consequence unrelated to HTT lowering itself
Somatic CAG Repeat Instability: A New Therapeutic Frontier
One of the most transformative conceptual advances in HD research over the past decade is the recognition that the inherited CAG repeat length is not static. In striatal neurons — the cells most vulnerable in HD — the CAG repeat continues to expand somatically throughout life, driven by DNA mismatch repair (MMR) pathway activity. This somatic expansion may be required for pathogenesis: cells may not become sick until the repeat reaches a critical threshold far beyond the inherited length.
Genetic Modifier Studies: The GeM-HD Consortium
The Genetic Modifiers of Huntington's Disease (GeM-HD) Consortium conducted genome-wide association studies (GWAS) in over 4,000 HD patients, identifying genetic variants that modify the age of motor onset independently of inherited CAG repeat length. The key findings converged on DNA repair pathways:
| Gene | Chromosome | Role | Effect on HD |
|---|---|---|---|
| MSH3 | 5 | Component of MutSβ (MSH2–MSH3 heterodimer); recognizes CAG repeat secondary structures (slipped-strand structures) and initiates mismatch repair | Loss-of-function variants associated with slower disease progression. Reducing MSH3 by 41% halved the somatic expansion rate; 83% reduction estimated to halt expansion completely |
| MLH1 | 3 | Component of MutLα endonuclease complex; coordinates excision during MMR | Variants modify age of onset (signal on chromosome 3) |
| PMS1 | 2 | MutL heterodimer component | Modifies disease onset; also targeted by SKY-0515 |
| PMS2 | 7 | Component of MutLα endonuclease complex | Identified as a modifier in GWAS analyses |
| FAN1 | 15 | Fanconi-associated nuclease; prevents repeat expansion through nuclease activity and MLH1 binding | Coding mutation p.Arg507His associated with 6 years earlier onset. FAN1 protects against expansion; loss-of-function accelerates disease |
Mechanism of Somatic Expansion
Expanded CAG repeats form aberrant secondary structures (hairpins, slipped-strand DNA) during normal DNA replication and transcription. The MutSβ complex (MSH2–MSH3) recognizes these structures and recruits MutL endonucleases, but the repair process is error-prone at long repeats, resulting in progressive lengthening rather than correction. This creates a vicious cycle: longer repeats form more aberrant structures, leading to more erroneous repair, leading to further expansion.
Critically, this somatic expansion is tissue-specific: it is most pronounced in striatal neurons (which express high levels of MSH3) and minimal in cerebellum — mirroring the selective vulnerability pattern of HD.
Therapeutic Approaches Targeting Somatic Instability
| Approach | Target | Status |
|---|---|---|
| MSH3-targeting ASOs | MSH3 mRNA | Preclinical proof-of-concept: ASO-mediated MSH3 suppression reduces somatic CAG expansion in HD iPSC-derived striatal neurons (published 2025 in Science Translational Medicine) |
| MSH3-targeting siRNA | MSH3 mRNA | Preclinical: di-valent siRNA achieved dose-dependent reduction of somatic expansions in HdhQ111 mice (2024) |
| FAN1 enhancers | FAN1 expression | Harness Therapeutics: developing ASO modality to boost FAN1 expression by blocking inhibitory microRNA binding sites |
| SKY-0515 (dual mechanism) | HTT + PMS1 | Phase 2/3 (FALCON-HD): the 26% PMS1 mRNA reduction observed represents a potential simultaneous attack on somatic instability alongside HTT lowering |
| Direct CAG repeat binding | CAG repeat DNA | Preclinical: direct binding agents can suppress somatic instability without affecting HTT levels (published 2025 in Nature Communications) |
Timing Is Critical for Somatic Instability Therapies
A landmark 2024 study in Brain identified a CAG repeat threshold for therapeutic intervention: in striatal neurons with (CAG)185 repeats, further expansion did not accelerate the onset of molecular and neuropathological phenotypes — suggesting a ceiling beyond which additional expansion no longer matters. This implies that somatic instability therapies must be administered early, before the repeat crosses the pathogenic threshold in vulnerable neurons. By the time symptoms appear, many neurons may already have reached or exceeded this threshold. This supports the emerging consensus that disease-modifying therapies in HD should ideally begin in presymptomatic stages.
Other Disease-Modifying Approaches
CRISPR/Gene Editing
Gene editing offers the theoretical possibility of permanently correcting the HD genetic defect at the DNA level, rather than continuously suppressing its expression.
- CRISPR-Cas9 approaches: Can excise the expanded CAG repeat, inactivate the mutant allele at the DNA level, or provide allele-selective editing. Preclinical studies have demonstrated increased lifespan and improved motor deficits in HD mouse models
- CRISPR interference (CRISPRi): Uses catalytically dead Cas9 (dCas9) to suppress HTT expression without creating DNA double-strand breaks, reducing off-target mutation risk. Shown to delay disease progression in HD mouse models
- RNA-targeting CRISPR (CasRx): A 2024 study demonstrated efficient HTT mRNA knockdown with therapeutic effects in multiple HD models (HEK 293T cells, HD 140Q-KI mice, HD-KI pigs). Effects were stage-dependent — earlier treatment produced more benefit
- RIDE delivery system: A novel nanoparticle delivery method that achieved high editing efficiency in mice and monkeys with reduced off-target effects compared to viral vectors. Single-injection administration with sustained results
- Status: All CRISPR approaches for HD remain preclinical. Delivery to the CNS, off-target editing, immunogenicity, and durability remain major challenges. No human trials are currently planned
Immunotherapy
- Passive immunization against mHTT: The C6-17 monoclonal antibody (AFFiRiS) targets huntingtin near the caspase 6 cleavage site. Preclinical studies showed it induces phagocytic clearance of extracellular HTT, produces both peripheral and central HTT lowering, and ameliorates motor deficits in the YAC128 mouse model
- Rationale: Gene-based therapies primarily lower intracellular mHTT, but pathological protein is also abundant extracellularly (in CSF, plasma, and extracellular matrix). With growing evidence for prion-like spreading and seeding of mHTT aggregates, targeting extracellular mHTT via immunotherapy could complement intracellular HTT-lowering approaches
- Status: Preclinical; no active clinical trials
NMDA Receptor Modulation
- Dalzanemdor (SAGE-718, Sage Therapeutics): A positive allosteric modulator of the NMDA receptor, tested for cognitive impairment in HD
- DIMENSION trial (Phase 2): 189 participants randomized; 12-week, double-blind, placebo-controlled study targeting cognitive impairment. Dalzanemdor did not demonstrate a statistically significant difference vs placebo on the primary endpoint (SDMT change at Day 84), nor on any secondary endpoints. Well tolerated but development discontinued (November 2024)
Phosphodiesterase Inhibitors
- PDE10A inhibitors: PDE10A is highly enriched in striatal medium spiny neurons (the primary cells affected in HD). Inhibition was hypothesized to restore cortico-basal ganglia signaling
- AMARYLLIS trial (Phase 2): PF-02545920 (Pfizer) was generally safe and showed some rater-independent improvements in quantitative motor measures, but did not meet primary or secondary clinical endpoints
- The selective PDE10A inhibitors demonstrated an unexpectedly substantial dyskinesia side effect, and intervention may have been too late (after significant striatal degeneration)
Sigma-1 Receptor Agonism — Pridopidine (Prilenia)
- Mechanism: Potent sigma-1 receptor agonist; proposed to enhance BDNF signaling, calcium homeostasis, and neuroprotection
- PROOF-HD (Phase 3): 499 participants with adult-onset HD (TFC ≥7). Primary endpoint (TFC change at 65 weeks) was not met in the overall population, nor was the key secondary endpoint (cUHDRS)
- Pre-specified subgroup analysis: In participants not taking antidopaminergic medications (neuroleptics/chorea drugs), pridopidine showed significant improvements in cUHDRS (week 52: Δ0.43, p=0.04), stroop word reading (week 52: Δ4.22, p=0.02), and multiple other clinical endpoints
- Regulatory: The European Medicines Agency accepted Prilenia's marketing authorization application in September 2025; Committee for Medicinal Products for Human Use (CHMP) opinion expected in the second half of 2025
Stem Cell Therapy
- NestaCell (hDPSC therapy): A Phase II trial of allogeneic human dental pulp stem cells in HD showed significant improvement in motor outcomes (UHDRS-TMS) vs placebo. The 2 million cells/kg group showed functional benefits and a non-significant trend toward neuroprotection on MRI. Advancing to Phase III
- Cellavita-HD: Mesenchymal stem cell therapy being studied in Brazil (ADORE-DH dose-response trial; 35 participants assessing multiple administrations)
- Embryonic stem cell-derived neural stem cells (UC Irvine): Leslie Thompson received a $12 million CIRM grant for a first-in-human safety trial, expected to begin mid-2026. This will be the first test of embryonic stem cell-derived cells in HD
Biomarker Development
Reliable biomarkers are essential for measuring disease progression, confirming target engagement, selecting patients for trials, and defining therapeutic endpoints. The HD biomarker landscape has matured substantially.
The Huntington's Disease Integrated Staging System (HD-ISS)
Published by Tabrizi et al. in The Lancet Neurology (2022), the HD-ISS provides a standardized biological classification of HD for research purposes. It uses a genetic case definition (HTT CAG ≥40 repeats) and stages the disease from birth based on prognostic biological, clinical, and functional landmarks:
| Stage | Definition | Key Features |
|---|---|---|
| Stage 0 | Gene-positive, no detectable pathological change | HTT CAG ≥40 with no biomarker abnormalities. Individual has the genetic mutation from birth but no measurable disease |
| Stage 1 | Biomarker changes without clinical symptoms | Measurable changes on quantitative neuroimaging (e.g., caudate/putamen volume loss on MRI) or cognitive biomarkers, but no clinical symptoms. Currently, no wet biomarker (e.g., NfL) has met the stringent criteria to serve as an HD-ISS landmark |
| Stage 2 | Prodromal HD | Subtle motor and/or cognitive signs detectable on clinical assessment, but not sufficient for a clinical diagnosis of HD |
| Stage 3 | Clinical HD | Overt motor, cognitive, and/or psychiatric symptoms meeting criteria for clinical diagnosis |
The HD-ISS is transforming trial design by enabling enrollment of presymptomatic individuals (Stages 0–1) and providing standardized staging across studies. The PIVOT-HD and FALCON-HD trials, for example, use HD-ISS stages to define their patient populations.
Fluid Biomarkers
| Biomarker | Source | Role | Clinical Utility |
|---|---|---|---|
| Mutant huntingtin (mHTT) | CSF | Direct measure of target protein | Confirms target engagement of HTT-lowering therapies. Used as primary/secondary endpoint in multiple trials (GENERATION-HD1, SELECT-HD, PIVOT-HD) |
| Neurofilament light chain (NfL) | CSF or plasma | Marker of neuronal/axonal damage and neurodegeneration | Rises with disease progression; used as safety signal (NfL spikes indicated harm in tominersen and branaplam trials). Plasma NfL correlates with disease stage and can enrich HD-ISS Stage 1 populations. A 14-year longitudinal study showed NfL rises years before motor onset and continues to increase through the premanifest-to-manifest transition |
| Huntingtin protein (total HTT) | Blood | Peripheral pharmacodynamic marker | Used in PTC518 and SKY-0515 trials to confirm systemic HTT lowering (primary endpoint of PIVOT-HD) |
Neuroimaging Biomarkers
- Volumetric MRI: Caudate and putamen atrophy are the earliest and most reliable structural MRI markers, detectable years before motor onset. Progressive ventricular enlargement (lateral ventricles) tracks disease progression. Caudate atrophy correlated with mHTT reduction in the SELECT-HD trial of WVE-003
- Ventricular volume: Used as a safety measure after the GENERATION-HD1 experience, where dose-dependent ventricular enlargement was a concerning finding. Distinguishing between neurodegenerative atrophy and drug-related ventricular changes is important
Current Clinical Trial Landscape (2024–2026)
| Trial | Agent | Phase | Mechanism | Status / Expected Readout |
|---|---|---|---|---|
| AMT-130 Phase I/II | AMT-130 (uniQure) | Phase I/II (pivotal) | AAV5-miHTT gene therapy | Primary endpoint met (September 2025). Regulatory path under discussion with FDA (Q1 2026 meeting planned) |
| GENERATION-HD2 | Tominersen (Roche) | Phase 2 | Allele-nonselective ASO | Ongoing; 100 mg Q16W in early HD (ages 25–50). Expected completion 2026 |
| SELECT-HD | WVE-003 (Wave) | Phase 1b/2a | Allele-selective ASO (SNP rs362307) | Completed multi-dose cohort. Phase 2/3 IND submission planned H2 2025 |
| PIVOT-HD | PTC518 / votoplam (PTC Therapeutics) | Phase 2 | Oral HTT splicing modifier | Primary endpoint met (May 2025). Phase 3 planning anticipated |
| FALCON-HD | SKY-0515 (Skyhawk) | Phase 2/3 | Oral RNA splicing modifier (HTT + PMS1) | First patient dosed; enrolling at 40+ global sites. >90 patients dosed as of January 2026 |
| Pridopidine EMA Review | Pridopidine (Prilenia) | Regulatory | Sigma-1 receptor agonist | EMA marketing authorization application accepted September 2025 |
| NestaCell Phase III | hDPSC (Cellavita) | Phase 3 (planned) | Dental pulp stem cell therapy | Advancing to Phase III after positive Phase II motor outcomes |
| ESC Neural Stem Cells | ESC-derived NSCs (UC Irvine) | Phase 1 (planned) | Embryonic stem cell-derived neural stem cells | $12M CIRM grant received; first-in-human trial expected mid-2026 |
Enroll-HD: The Backbone Observational Study
Enroll-HD is the world's largest longitudinal, observational study of HD, integrating the former REGISTRY (Europe) and COHORT (North America/Australasia) registries and expanding to Latin America. As of June 2024:
- 31,824 participants recruited from 189 sites in 23 countries
- 21,630 currently active participants with annual assessments
- >82,000 blood kits collected across 12 biospecimen types
- Enroll-HD provides the natural history data and external control cohorts that are critical for interpreting single-arm trials (such as AMT-130)
- Protocol updates and site start-up activities for enhanced data collection are scheduled for 2025
Failed Approaches and Lessons Learned
| Agent / Approach | What Happened | Key Lesson |
|---|---|---|
| Tominersen Q8W (GENERATION-HD1) | Worse outcomes than placebo; ventricular enlargement; NfL spikes | Total HTT lowering beyond ~50% is harmful. Wild-type HTT has essential functions. Dose intensity and frequency matter critically. The "huntingtin holiday" concept emerged |
| WVE-120101 / WVE-120102 | No dose-dependent mHTT lowering; no target engagement vs placebo (2021) | First-generation ASO chemistry was insufficient. Stereopure chemistry (WVE-003) dramatically improved potency |
| Branaplam (VIBRANT-HD) | Peripheral neuropathy and NfL elevations; discontinued | Off-target splicing effects are a real risk with small molecule splicing modifiers. Peripheral toxicity may limit the therapeutic window |
| Dalzanemdor (DIMENSION) | No benefit on SDMT or any secondary cognitive endpoint | NMDA receptor modulation alone is insufficient for HD cognitive impairment, which likely reflects irreversible structural neurodegeneration rather than pure synaptic dysfunction |
| PDE10A inhibitors (AMARYLLIS) | No clinical efficacy; dyskinesia side effects | Intervention may have been too late (after substantial striatal neuron loss). The target itself may not be tractable in symptomatic patients |
Overarching Lessons for HD Disease Modification
- Earlier intervention is essential. Nearly every failure can be partly attributed to treating patients too late, when neurodegeneration is already advanced. The HD-ISS framework now enables trials in presymptomatic individuals (Stages 0–1)
- Allele-selective > nonselective (probably). The GENERATION-HD1 failure has shifted the field toward selective approaches that spare wild-type HTT, though the optimal degree of total HTT lowering remains debated
- NfL is an early warning system. Transient NfL spikes should be treated as a red flag for neuronal harm, not dismissed as benign pharmacological effects
- Multiple modalities may be needed. Combining HTT lowering (intracellular) with somatic instability targeting and possibly immunotherapy (extracellular mHTT clearance) may be required for optimal efficacy
- Trial design innovation matters. External controls (Enroll-HD), the HD-ISS staging system, plasma NfL as enrichment biomarker, and digital endpoints are reshaping how HD disease modification is tested
The Road Ahead
The HD disease-modification landscape in 2025–2026 is at an inflection point. For the first time, a therapy (AMT-130) has shown statistically significant slowing of disease progression, multiple oral agents (PTC518, SKY-0515) are advancing through efficacy trials, and the discovery of somatic instability as a druggable target has opened an entirely new therapeutic axis. The field is rapidly moving from a single strategy (lower huntingtin) to a multi-pronged approach (lower mHTT + stabilize the CAG repeat + clear extracellular aggregates + neuroprotect). The next 2–3 years will be decisive: GENERATION-HD2, FALCON-HD, and AMT-130 regulatory outcomes will define whether the first disease-modifying therapy for HD reaches patients.
Key References
- Tabrizi SJ, Schobel S, Gantman EC, et al. A biological classification of Huntington's disease: the Integrated Staging System. Lancet Neurol. 2022;21(7):632–644. doi: 10.1016/S1474-4422(22)00120-X
- Tabrizi SJ, Estevez-Fraga C, van Roon-Mom WMC, et al. Potential disease modifying therapies for Huntington's disease: lessons learned and future opportunities. Lancet Neurol. 2022;21(7):645–658
- McColgan P, Thobhani A, Gollwitzer E, et al. Tominersen in adults with manifest Huntington's disease. N Engl J Med. 2024
- GeM-HD Consortium. Identification of genetic factors that modify clinical onset of Huntington's disease. Cell. 2015;162(3):516–526
- GeM-HD Consortium. CAG repeat not polyglutamine length determines timing of Huntington's disease onset. Cell. 2019;178(4):887–900
- Mena F, Tabrizi SJ, Wild EJ. Huntington's disease clinical trials update (March 2024, March 2025, October 2025). J Huntingtons Dis.
- Scahill RI, Zeun P, Osborne-Crowley K, et al. A CAG repeat threshold for therapeutics targeting somatic instability in Huntington's disease. Brain. 2024;147(5):1784–1798
- Bachoud-Lévi AC, Ferreira J, Massart R, et al. International Guidelines for the Treatment of Huntington's Disease. Front Neurol. 2019;10:710