Huntington’s Disease: What Have We Learned from Recent Failures and Ongoing Trials?
Tominersen (Roche) and branaplam (Novartis) brought major disappointments, but recent interim clinical trial results hint at renewed hope for this devastating neurodegenerative disorder.
Huntington’s disease (HD) research is a rollercoaster ride of soaring hopes and crushing setbacks. The journey has been marked by promising therapeutic approaches that, despite early promise, have faltered in clinical trials, when they mattered most. Yet, each failure brings useful insights regarding safety, patient selection, and clinical trial design, that inform the next generation of studies.
In this post, I delve into the clinical trials that recently stumbled, those still fighting forward, and what this all means for the future of HD treatment. I’ll focus exclusively on therapies that target the HTT gene or mRNA, whose expansion causes HD - I think this approach holds the greatest potential for success in such a genetically defined disease.
A prototypical genetic disorder
HD is a genetic neurodegenerative disorder caused by an autosomal dominant mutation in the HTT gene - this means that children of affected individuals have a 50% probability of inheriting the mutation.
The disease manifests between the ages of 40 and 50, although onset can occur earlier or later, depending on the number of CAG repeats in the HTT gene (more on genetics below). Early signs often include subtle changes in mood, cognition, and motor coordination, which gradually give way to more pronounced features such as involuntary movements (chorea), impaired speech and swallowing, and significant cognitive decline. Psychiatric symptoms, including depression, irritability, and apathy, are also common and can be some of the earliest signs of the disease.
The disease course is progressive over a 15-20-year course, with symptoms worsening over time. Patients increasingly lose their ability to manage daily activities. In the later stages, they require full-time care, becoming bedridden and unable to communicate effectively.
The disease’s relentless progression and broad range of symptoms make it one of the most devastating neurological disorders, affecting not only the patient but also their families.
To fully grasp the promise and pitfalls of therapies targeting the HTT gene in HD, it’s crucial to understand the genetic underpinnings of the disease.
Expansion, penetrance, and anticipation
HD is caused by a genetic mutation in the HTT gene, an abnormal expansion of a CAG trinucleotide repeat sequence.
Normally, the HTT gene contains between 10 to 35 CAG repeats. In individuals with HD, this number expands to 36 or more repeats.
The concept of penetrance in HD refers to the likelihood that a person with the mutation will develop the disease. For individuals with 36-39 CAG repeats, penetrance is nearly complete, meaning most—but not all—will eventually develop the disease if they live long enough. When the repeat count reaches 40 or more, penetrance is complete: everyone with this expansion will develop HD if they live long enough.
The length of the CAG expansion correlates directly with disease onset and severity: the more repeats, the earlier the onset. For instance, individuals with 50 or more repeats may develop symptoms before age 40, leading to a more aggressive disease course.
Another important concept in HD genetics is anticipation, where the disease manifests earlier in successive generations. This occurs because CAG repeats can expand further when passed from parent to child, especially when inherited from the father. As a result, the offspring often experience earlier and more severe symptoms.
Wild-Type vs. Mutant HTT
The HTT gene produces a protein called huntingtin (HTT).
The exact function of the HTT in humans remains unclear. Studies in mice have provided some insights: knocking down the HTT gene in postnatal mice is lethal, underscoring its crucial role in development.
However, the consequences of reducing HTT levels in adults are more ambiguous. A decade ago, a study published in PNAS suggested that knocking down HTT in adult mice appeared safe for neurons but led to non-neuronal issues, such as pancreatitis, hinting that HTT might be more critical outside the brain. However, a more recent study published just weeks ago showed the opposite: severe neurodegeneration in adult mice (including marked increases in NfL and GFAP) but no pancreatic problems. Interestingly, these findings mirror some of the safety signals observed in the tominersen and branaplam trials, detailed below.
In HD, the mutant form of the protein, resulting from the expanded CAG repeats, misfolds and accumulates in neurons, resulting in the progressive neurodegeneration characteristic of HD.
This poses a significant therapeutic challenge: how do we reduce the levels of the harmful mutant huntingtin protein without interfering with the normal function of the wild-type huntingtin protein?
The Therapeutic Dilemma: Selectivity and Efficacy
Current therapeutic strategies targeting the HTT gene face this exact dilemma. Selective silencing of the mutant allele while preserving the wild-type allele is a complex and challenging task.
Some therapies aim to knock down both alleles, but this risks disrupting the critical functions of the normal huntingtin protein. Other approaches, like allele-specific gene silencing, strive to selectively target only the mutant allele.
Lessons learned from failed trials
Let’s now delve into the recently failed clinical trials with tominersen and branaplam for HD, both marking significant setbacks for the field. These trials not only failed to achieve their goals but also raised critical questions about the safety of therapeutic strategies targeting HTT.
Tominersen: A Trailblazer Stumbles
Tominersen, developed by Roche and Ionis, was a groundbreaking therapy designed to target and inhibit the translation of the huntingtin protein using an intrathecal antisense oligonucleotide (ASO) approach.
The Phase 3 GENERATION-HD1 trial, launched with much anticipation, to enroll over 800 patients with manifest HD (DCL 4) across the globe. The trial’s design included two dosing regimens: 120 mg every 8 weeks (Q8W), and every 16 weeks (Q16W), compared to placebo.
The primary clinical endpoints were changes in the Unified Huntington’s Disease Rating Scale Total Functional Capacity (UHDRS-TFC) and Composite Unified Huntington’s Disease Rating Scale (cUHDRS) scores over 101 weeks. Biomarker endpoints included the reduction of mutant HTT protein (mHTT) in cerebrospinal fluid (CSF) and neurofilament light (NfL) as a marker of axonal neurodegeneration.
Unfortunately, the trial was halted prematurely in March 2021 due to a lack of efficacy and an unfavorable safety profile.
The results, published a few months ago in the NEJM, revealed that, while tominersen reduced mHTT levels (up to -40%), it led to an unexpected increase in NfL, suggesting enhanced neurodegeneration, neuroinflammation or both. Additionally, patients receiving the more frequent (Q8W) dosing experienced worsening clinical outcomes compared to placebo, as well as a marked increase in brain ventricular volume.
Quite disappointing results.
The cause of the increased ventricular volume is not clear. It could perhaps be due to a reduced absorption of CSF, which in turn could be due to an increased white-cell count and protein levels in CSF (findings observed in tominersen-treated patients). The ventricular enlargement was not associated with an ex-vacuo whole-brain volume loss, thus indicating that the increases in ventricular volume do not reflect increased brain atrophy.
These adverse effects highlighted the complexities of reducing HTT in the adult brain and cast doubt on the dosing strategy and the role of wild-type HTT in maintaining neural integrity.
Branaplam: A Serendipitous Discovery Falls Short
Branaplam, initially designed by Novartis to correct the splicing of the SMN2 gene in spinal muscular atrophy (SMA) was repurposed for HD when it was found to promote the inclusion of an exon that leads to downregulation of HTT mRNAs and protein (both wild-type and mutant forms).
These findings prompted Novartis to initiate the VIBRANT-HD clinical trial in 2021 to assess its safety and efficacy in HD patients.
The study design included a dose-escalation phase, with primary clinical endpoints focusing on changes in the cUHDRS, Total Motor Score (TMS), and other functional measures. The trial also monitored biomarker endpoints, including mHTT levels and NfL, to assess neurodegeneration and treatment efficacy.
In 2022, a letter from Novartis to the HD community announced that the trial was halted after an interim analysis revealed that, although branaplam effectively lowered mHTT, it also led to a concerning rise in NfL levels. Additionally, patients exhibited increased brain ventricular volume, signaling potential neurotoxicity - not unlike tominersen.
The unexpected toxicity might have arisen from the non-selective effects of the splicing modulator, which may have disrupted essential processes beyond HTT, or due to reductions in wild-type HTT.
Novartis has not yet published the branaplam trial results: this would be very useful to the HD community (I know some people at Novartis are reading this).
What These Failures Teach Us
The failures of tominersen and branaplam underscore the difficulties in targeting HTT, especially given the dual role of the wild-type protein. Both trials highlighted that, while reducing mHTT is possible, off-target effects or over-suppression of the wild-type HTT can lead to harmful outcomes like toxicity and clinical worsening.
Note that neither Roche nor Novartis have reported the level of reduction of wild-type HTT in their trials - this would be important to know.
These findings suggest that future approaches must be more nuanced, requiring selective targeting of the mutant allele while sparing the wild-type HTT to avoid neurotoxicity.
Importantly, NfL and ventricular volume have emerged as key safety biomarkers that must be carefully monitored in HD clinical trials.
Ongoing trials for HD targeting HTT
After the disappointing GENERATION HD1 trial results, Roche conducted post-hoc subgroup analyses, revealing a potential silver lining.
Younger participants with less advanced functional decline, as measured by the CAP score (a product of age and CAG repeat length), who received tominersen every 16 weeks (Q16W), showed a trend toward better outcomes vs. placebo. Notably, point estimates for the cUHDRS and its functional, cognitive, and motor subscales were all directionally favorable at 69 weeks, particularly in those with lower drug exposure.
To validate this finding, Roche initiated the GENERATION HD2 trial (NCT05686551). The key differences from the original trial include:
Loading Dose: Unlike GENERATION HD1, where participants received an initial loading dose to boost drug levels, GENERATION HD2 omits this step.
Dosage: GENERATION HD2 is testing lower doses of tominersen. GENERATION HD1 used a 120 mg dose, while GENERATION HD2 is exploring a high dose of 100 mg and a low dose of 60 mg.
Frequency: Tominersen is administered less frequently in GENERATION HD2. While GENERATION HD1 tested dosing every 8 and 16 weeks, the new trial only administers the drug every 16 weeks.
Inclusion of Early Disease: GENERATION HD2 expands eligibility to include participants with prodromal HD, a group that was not eligible for GENERATION HD1.
GENERATION HD2 continues to recruit participants aged 25-50 with early HD symptoms at sites worldwide
In addition to the GENERATION HD2 trial, other companies have ongoing or recently completed clinical trials for HD targeting HTT. The key information is summarized in the following table.
Wave WVE-003
WVE-003 is an allele-selective ASO targeting the SNP3 variant linked to the expanded CAG repeat in HTT pre-mRNA, estimated to be present in about 40% of adults with HD.
By selectively inhibiting mHTT translation while sparing wild-type HTT, WVE-003 offers a targeted approach, but only for this subset of patients.
Wave recently presented the results (investor presentation June 2024) of their SELECT-HD trial, a phase 1b/2a, placebo-controlled, double-blind study in HD (Stages 0-3 in the new HD Integrated Staging System). The trial recruited 70 participants dosed with single doses (placebo, 30, 60, 90 mg) or multiple doses (30 mg) and followed up for 36 weeks.
Multiple 30 mg doses led to a 46% reduction in mHTT without affecting or even slightly increasing wtHTT levels, confirming the allele selectivity of the drug.
The safety profile was acceptable, with minimal increases in NfL in a minority of patients. They did not report on ventricular enlargement.
Clinical endpoints showed motor improvement in the active group only in the last study visit; functional outcomes were not disclosed, leaving the full clinical impact to be determined in future trials.
Takeda has the option to co-develop and co-commercialize WVE-003. Will they choose to do so?
PTC518
Developed by PTC Therapeutics, PTC518 is an oral small-molecule splicing modulator targeting HTT mRNA, currently under investigation in the PIVOT-HD (NCT05358717) placebo-controlled clinical trial for early HD patients.
Interim data from June 2024 (investor presentation) revealed dose-dependent reductions in mHTT levels (up to -40%), with no increases in NfL and no treatment-related adverse events, despite thorough monitoring for peripheral neuropathy (ventricular enlargement was not specifically mentioned).
Clinically, there was a dose-dependent and significant slowing in the progression of the composite Unified Huntington’s Disease Rating Scale (cUHDRS) and the motor composite score. While the Total Functional Capacity (TFC) also showed less worsening, this effect was not dose-dependent.
Overall, these results are promising, particularly regarding safety and target engagement. The clinical endpoints are trending in the right direction, but it remains to be seen if these effects are sustained and translate into meaningful clinical benefits.
uniQure AMT-130
AMT-130 is a recombinant adeno-associated virus (rAAV5) engineered to deliver a microRNA that targets exon 1 of the HTT mRNA (rAAV5-miHTT), aiming to reduce the production of both mutated (expanded) and wild-type HTT. The rAAV5 is surgically administered as a one-time, MRI-guided, convection-enhanced stereotactic injection into six sites (three per side) within the striatum.
To mitigate potential safety events related to immunogenicity against the AAV5, patients require immunosuppression.
Currently, AMT-130 is being evaluated in two clinical trials involving early-stage HD participants (NCT04120493 & NCT05243017).
Initial sham-controlled data from the U.S. trial (n=6 low dose, n=10 high dose, n=10 placebo) disclosed in 2023 (investor presentation) showed trends towards less worsening in motor and functional outcomes, along with approximately a 10% reduction in NfL at 24 months. However, while mHTT initially decreased, it rebounded by the 24-month mark, casting doubts on the sustainability of this therapy.
More recent data from 21 patients (9 high-dose, 12 low-dose) with 24-month follow-up in both the U.S. and Europe (2024 investor presentation) were compared to an external propensity-matched group (n=154).
Remember: external comparisons to natural history controls can be misleading. Patients enrolled in natural history studies often present in worse shape than those selected for clinical trials, even when matched for factors like sex, age, and other characteristics. Therefore, it’s crucial to approach externally matched control data with caution
Results showed no major safety or tolerability issues, with drug-related serious adverse events occurring only in the high-dose group.
Regarding, efficacy, the high-dose group demonstrated significantly slower cUHDRS progression compared to the external control, although this was mainly driven by the cognitive items, with no clear benefits in functional (TFC) or motor (TMC) items.
NfL levels initially spiked dramatically but gradually decreased to baseline (or slightly below) at 24 months. However, the absence of mHTT results or volumetric MRI data leaves some questions unanswered.
In summary, while the safety and tolerability of AMT-130 appear acceptable, the efficacy data are still inconclusive.
Vico VO659
The VO659-CT01 trial (NCT05822908) is a Phase 1/2a study investigating VO659, an intrathecal antisense oligonucleotide (ASO) that blocks CAG repeats. This trial, initiated in April 2023, is actively enrolling patients with spinocerebellar ataxia types 1 and 3 (SCA1, SCA3), and early HD.
Unlike HD-specific approaches, VO659 targets all polyglutamine (polyQ) proteins across several neurodegenerative disorders, potentially posing safety risks by impacting wild-type CAG genes.
ExoRNA ER2001
Chinese company ExoRNA Bio is conducting a Phase 1 study with ER2001 in patients with manifest HD. ER2001 is a double-stranded DNA plasmid carrying genetic information encoding for a rabies virus glycoprotein fusion protein coupled with a small interfering RNA (siRNA) targeting mHTT.
The plasmid is administered intravenously to, somehow, “infect” liver cells, which then use the plasmid DNA to produce a fusion protein composed of a neurotrophic rabies virus protein and mHTT-siRNA. However, this approach raises several questions: Will the plasmid integrate into the liver (and other tissues) genome? Is it immunogenic? How does the rabies virus protein penetrate the blood-brain barrier, and how does the siRNA selectively target mHTT?
The complexity of this method leaves many uncertainties, and only time will reveal its safety and effectiveness.
Conclusion: Navigating the Future of HD
The landscape of HD research is both dynamic and challenging. As we navigate through recent setbacks and emerging trials, several key takeaways emerge:
The complexity of targeting HTT: The tominersen and branaplam trials underscore the intricate balance required in targeting mHTT while preserving the essential functions of wild-type HTT. The safety signals observed in these trials highlight the need for more nuanced approaches that could achieve allele-specific targeting without significantly compromising the physiological functions of wild-type HTT.
Emerging Strategies: Ongoing trials, such as GENERATION HD2 and those involving novel compounds like WVE-003, PTC518, AMT-130, and VO659, offer new hope and insights. WVE-003 successfully showed allele-specific mHTT silencing, and PTC518 had both reductions in mHTT (albeit nonselective) with no increases in NfL. Thus, the field is adapting and innovating.
Safety and Efficacy Considerations: Monitoring biomarkers like NfL and assessing changes in ventricular volume are crucial, but we also need to demonstrate clinically meaningful and sustained efficacy. Special attention should be given to functional endpoints, such as Total Functional Capacity (TFC). Regulatory agencies, including the FDA, emphasize how new treatments impact patients' overall functioning and quality of life, rather than solely focusing on improvements in motor scales.
In summary, while recent setbacks remind us of the complexities involved, the continued research and adaptation in therapeutic strategies are paving the way for potential breakthroughs toward a future where disease-modifying HD treatments offer real hope to patients and their families.
Readers - Did I miss any ongoing HD trial targeting HTT? Do you have thoughts on how to target allele-specific mHTT? Please share your thoughts, questions, and experiences in the comments below!
Thanks for reading. If you found this entry interesting, please, share it so that others can receive more like this in their inbox.
Views expressed here are my own and not necessarily those of my employer. All data mentioned and discussed are publicly available.