A new study led by researchers at University College London explored a promising new way to potentially slow down Huntington disease (HD) by reducing the levels of a key DNA repair protein, called MSH3. Using antisense oligonucleotides (ASOs)—a type of genetic therapy that targets RNA—the researchers successfully stopped C-A-G repeat expansions in lab-grown brain cells derived from people with HD. Their findings highlight a potential treatment approach that could help delay onset of symptoms and progression of HD.
How Expanding C-A-G Repeats May Drive HD
HD is a genetic brain disorder caused by an expansion of C-A-G repeats in the huntingtin (HTT) gene. Everyone inherits two copies of the HTT gene, one copy from each of their parents. While everyone has some C-A-G repeats in their two HTT genes copies, people with HD inherit a copy HTT with too many C-A-Gs.
Research has shown that in certain brain cells, particularly medium spiny neurons – the cells most affected in HD – these C-A-G repeats can continue to grow over time in a process called somatic expansion.
Many scientists believe that this ongoing C-A-G expansion in specific brain cells plays a role in determining when symptoms first emerge. Because of this, researchers have been exploring ways to slow or stop somatic expansion in hopes of delaying the onset of symptoms and slowing how fast the disease progresses.
What Can We Learn About Drug Targets From Genetic Studies?
Large genetic studies of people with HD have linked certain genes responsible for proofreading the genetic code, including MSH3, to the age at which symptoms first appear. Normally, MSH3 helps fix small mistakes in the genetic code, but in HD, it can actually promote repeat expansion, causing CAG repeats to get bigger and bigger.
Other genetic studies in people with HD also suggest that lowering MSH3 could be a promising treatment. Interestingly, a small natural genetic hiccup in the MSH3 gene, which causes people to produce slightly less MSH3, has been linked to slower disease progression, less CAG repeat expansion, and a delay of about one year in symptom onset in people with HD who have this variant.
Other types of genetic variations of MSH3 that lead to even bigger reductions of MSH3 levels have been associated with delaying HD symptoms by more than 10 years. Because MSH3 is not essential for survival and most people born with lower levels of it generally live normal lives, MSH3 has emerged as a promising drug target for HD.
A Genetic Approach to Reducing MSH3
In this study, researchers tested whether drugs designed to lower levels of the MSH3 protein could slow C-A-G repeat expansion. To do this, they used ASOs, which are short synthetic DNA-like molecules designed to bind to the MSH3 message in the cell and prevent the production of MSH3 protein, causing the levels of this protein to drop.
In dishes in the lab, the researchers grew medium spiny neurons, a type of brain cell found in the striatum – the area most affected in HD. These cells were dosed with the ASO, which led to a strong reduction in MSH3 levels. The effect the scientists saw was dose-dependent, meaning that the more ASO they added, the more the levels of MSH3 were lowered in these cells.
MSH3 Lowering Puts The Brakes On C-A-G Repeat Expansion
Importantly, the study found that lowering MSH3 stopped C-A-G repeat expansions in these neurons. In fact, the more MSH3 was reduced, the more the expansion process slowed down. At very high levels of MSH3 lowering, the CAG repeats not only stopped expanding but even began to shrink. This is a very exciting finding because it suggests that drugs lowering MSH3 might be able to reverse some of the genetic changes that occur in HD, which could be very beneficial.
To explore how this therapy might work in living organisms, the researchers developed a special mouse model that carries the human MSH3 gene. This was essential because the ASO being tested specifically targets the human version of MSH3, so the model needed to accurately reflect the drug’s intended future target – MSH3 in people.
They injected the ASO directly into the brains of these mice and found that MSH3 levels were successfully reduced across multiple brain regions, including the striatum – the area most affected in HD. This means that the drug was able to effectively spread out in the mouse brain, getting into brain cells in many different regions. Most importantly, the ASO was well tolerated, showing no signs of toxicity in these mice related to lowering MSH3.
What This Means for HD Research and Future Treatments
These findings provide strong evidence that targeting MSH3 with ASOs could be a safe and effective way to slow, stop, or even reverse C-A-G repeat expansions in HD. By preventing these repeats from growing, this strategy could potentially delay the onset of symptoms and slow disease progression.
Several biotech companies, including Rgenta Therapeutics, LoQus23, Latus Bio, and Harness Therapeutics, are now working on therapies to target somatic expansion in HD, and MSH3 targeting ASOs could be an important addition to these efforts. We expect to hear more from many of these companies later in the year at the Huntington Study Group meeting, which will be held in Nashville, Tennessee in October this year.
While this study was done in lab-grown neurons and mice, the next step would be to test this approach in human clinical trials to determine whether it could be a viable treatment option for people with HD. Stay tuned for more updates as this research progresses!
Two recent studies offer fresh insights into how antidepressants, often prescribed to help manage mood and anxiety, are prescribed in Huntington’s disease (HD) and might also influence cognitive decline. One study zooms in on medication use in HD, while the other takes a broader look at dementia and antidepressants. Together, they reveal a complex and evolving map of treatment decisions. But this doesn’t mean people living with HD should stop taking antidepressants. Let’s dig into why that is.
In the early stages, people with HD take an average of 2.5 medications. But as the disease advances, that number more than doubles to 5.2. This really highlights just how much a person’s medical needs change as HD progresses.
So, what medications are people taking? The study found that antipsychotics (used to manage movement symptoms and psychiatric issues), selective serotonin reuptake inhibitors (SSRIs, a common class of antidepressants), and painkillers (for chronic discomfort associated with HD) top the list. All of these, including SSRIs, are a critical tool in the toolkit for people living with HD.
Surprising Factors
But here’s where things get really interesting—prescription patterns vary based on factors like disease stage, gender, and location. For instance, men with HD are more likely to be prescribed antipsychotics, while women tend to use more antidepressants and painkillers.
The geographical divide is equally fascinating: In North America, SSRIs are the go-to choice, whereas in Europe, doctors are more likely to prescribe antipsychotics.
Why? It could be differences in treatment guidelines, cultural attitudes toward medications, or even drug cost and availability. Whatever the reason, this variation suggests that medication choices might be influenced by more than just individual patient needs.
What’s important here is that this study actually looked at what medications people were using, not just what their doctors recommended. So this gives us a much more realistic picture of what’s actually happening. This is valuable because it gives us a peek into the real world, the lived experience of these folks who are dealing with HD on a day-to-day basis.
Treatment Shift
Another crucial takeaway from the study is how medication use shifts over time. Early on, doctors may focus on medications that aim to manage mood and anxiety. But as involuntary movements and challenging behaviors become more prominent, treatment shifts toward managing these more disruptive symptoms.
This shift is particularly evident in the use of antipsychotics, which increase significantly as HD progresses.
Meanwhile, people who develop the rare form of juvenile HD show different medication patterns altogether, often requiring more treatments for aggression and irritability rather than for movement symptoms.
These findings highlight the need for personalized treatment approaches that consider unique disease trajectories and needs of different patient groups, particularly for those with juvenile HD.
Antidepressant Use in People with Dementia
A second study steps back from HD specifically and looks at a broader question: Do antidepressants influence cognitive decline in people with dementia? Antidepressants are often prescribed for people with dementia to help manage the psychological symptoms that come with the disease, like anxiety and depression.
Using data from the Swedish Registry for Cognitive Dementia Disorders, researchers examined whether certain antidepressants might actually accelerate cognitive deterioration. And the findings are raising eyebrows.
Among people with dementia, those taking antidepressants—especially SSRIs—experienced faster cognitive decline. The effect was particularly pronounced in individuals with more severe dementia at the study’s start.
However, it’s critical to note that some other studies have shown conflicting results, which just goes to show how complex this issue is. These findings add layers of complexity for the decision-making process for doctors and patients around the use of these medications, particularly for the most vulnerable groups of people with severe dementia.
More Medicine, Faster Decline?
Interestingly, they also suggest there is a dose-response relationship—meaning that higher doses of SSRIs were linked to an even greater rate of cognitive decline.
Medications like sertraline, citalopram, and escitalopram—widely used SSRIs—were the most strongly associated with cognitive decline. This raises important questions: Are these medications helping more than they’re harming? Should doctors rethink how and when they prescribe them to people with dementia? For HD, the answers aren’t black-and-white and there’s more nuance to these questions.
Another intriguing twist? The study found that men experienced a steeper cognitive decline on antidepressants compared to women, despite the fact that women are more likely to be prescribed these medications. Additionally, people who were not taking anti-anxiety or sleep medications alongside their antidepressants showed a more pronounced decline. Could other medications be offering some kind of protective effect, or is this just a coincidence? The answers remain unclear, highlighting the limitations of this study and the need for further research.
Things to Keep In Mind
There are some critical caveats for the study that links accelerated dementia to antidepressant use that people need to keep in mind, because this study isn’t a one-to-one comparator for people from HD families.
First, depression itself is associated with dementia and cognitive impairment, so we can’t really tease apart the chicken-and-egg problem here. The associations between antidepressant use and cognitive decline could be due to the underlying psychiatric condition rather than the drug itself. In other words, people may be prescribed antidepressants because their symptoms are worse or progressing more rapidly – the underlying cause of decline is the brain disease, not the drug. Although the researchers tried to account for this, it’s not something we can entirely rule out.
Second, dementia severity could itself be contributing to cognitive decline, making it difficult to conclusively say the results they saw were because of the antidepressants. The relationship between antidepressant use and dementia severity is complicated. From the Enroll-HD data described here, we know that treatment and medication use evolves as HD progresses, which should likely be the case for other diseases as well, like dementia.
Third, different forms of dementia have very different biological causes, like Alzheimer’s, Lewy body dementia, or frontotemporal dementia. But this study grouped these various types of dementia together. This could be masking some of the disease-specific effects that may be at play between the effects of antidepressants and these specific types of dementia. To add to this, HD is also a unique disease which likely has its own individual effects with specific medications. For that reason, it’s important to assess medication effects at the individual disease and patient level, rather than drawing conclusions broadly across a group of diseases.
Lastly, and perhaps most importantly, this study looked at association, not causation. These types of study designs that aren’t testing medications in a blinded clinical trial have major limitations. They just don’t have the power or rigor to draw black-and-white conclusions about what is happening biologically. However, they are good at making associations between events, like the use of antidepressants and cognitive decline, that can be examined in more detail in future studies.
Don’t Toss Your Meds!
Both studies highlight the delicate balancing act of prescribing medications for neurodegenerative diseases based on the individual. For people with HD and other forms of dementia, medications can provide crucial relief from psychiatric and motor symptoms.
A critical takeaway is that these recent findings don’t mean antidepressants should be abandoned for HD! Rather, they underscore the need for a thoughtful, individualized approach through collaborative relationships between clinicians, patients, and caregivers. Often people close to us know us better than we know ourselves, and this is particularly true for caregivers.
For many people with HD, the short-term risk from depression or challenging behaviours is huge – these are symptoms that can all too easily lead to injury, self-harm, and premature death. Balancing short-term and long-term risks, and the potential harms and benefits from treatment options, is a delicate business demanding full engagement between patients, their loved ones, and medical professionals.
Conversations between HD families and doctors should be open and honest, so that clinicians can remain vigilant, adjusting treatment plans based on the latest research and the evolving needs of each patient. This could also include helping people find access to non-drug treatments, like therapy, support groups, and lifestyle changes.
The Road Ahead
The studies discussed here are a reminder that medicine is never one-size-fits-all. Particularly for HD, medication use is incredibly common and just gets more frequent and more complicated as the disease progresses. Treatment patterns can be so different for various groups, which really highlights the need for open and honest dialog between patients and doctors to develop personalized care plans.
This work also highlights how much we still have to learn about the brain and the interplay between medications and neurodegeneration. More research is needed to untangle these complex relationships, but one thing is clear: Whether in HD or broader dementia care, the goal remains the same—to create a smoother, safer journey for those navigating these difficult conditions.
For now, patients and families should stay informed, ask questions, and work closely with their doctors to ensure that treatments align with their individual needs. Because when it comes to the brain’s roadmap, careful navigation is key to getting where we want to go.
Real progress is being made on the road to Huntington’s disease (HD) treatments, but in today’s fast-moving digital world, it can be harder than ever to separate genuine breakthroughs from overhyped headlines or flat-out misinformation.
That’s why HDBuzz has updated our Ten Golden Rules to help you decide whether a piece of news about HD research offers real promise, or whether its claims should be taken with a healthy pinch of salt.
If this article seems familiar, that’s because it is! Ed and Jeff wrote a version of these rules back in 2011. But over the past 14 years, the way we consume news in the age of social media and clickbait-focused news websites has changed a lot. So we hope this updated article provides some clarity and helpful guidelines for navigating research news in 2025.
Snowflakes and glaciers
At HDBuzz, we love science. We like to imagine all the world’s scientific research as a flurry of snowflakes, gently settling on a mountaintop and gradually, over months, years, and decades, compacting into a huge, unstoppable glacier that can carve entire mountains.
No single snowflake could do that, but combined, over time, the power of science to change the world – and improve the lives of people with HD – is immense.
The search for treatments and cures for HD is exactly like that. Most progress is small, incremental, and takes place behind the scenes. But step-by-step, we’re moving closer to therapies that will make a meaningful difference.
How science reaches the public in 2025
Science becomes “official” when it’s published in a peer-reviewed journal – but that’s rarely how most people first hear about it. In 2025, scientific information spreads across a sprawling ecosystem of platforms: news sites, press releases, blogs, Facebook, YouTube explainers, Reddit threads, and increasingly, short-form video on platforms like TikTok and Instagram.
That’s not necessarily a bad thing – more access to information is a win. But it comes with risks. Short videos and viral posts are often created for clicks and engagement, not accuracy. Many creators aren’t scientists, and even well-meaning ones can misunderstand or oversimplify complex findings. Sometimes, content is misleading or, worse, just plain false.
At the same time, researchers and institutions are under more pressure than ever to promote their work and secure funding. Teams will often issue press releases that highlight the long-term potential of early-stage research or niche studies, even when those applications are years (or decades) away or only give insight to a narrow aspect of HD research.
One way to excite people about this type of research is to get them to imagine the whole glacier, rather than just the snowflake. When those press releases are subsequently re-written into news articles or turned into social media snippets, nuance can be lost. A promising experiment in cells or mice can easily become “Scientists close to curing Huntington’s disease!” – even if human trials are a long way off.
What’s the harm?
It’s easy to feel hopeful, and we should! But when stories exaggerate the readiness or relevance of early-stage science, people in the HD community can end up misled, confused, or disappointed. And repeated disappointments can erode trust in science altogether.
We don’t believe this is the fault of individual scientists, journalists, or creators. But in a world where misinformation spreads fast, it’s important to stay curious and critical.
We know that some folks think of us as “HDBuzzkill”, as our articles can be less enthused than other press releases or articles on other platforms. However, for our editorial team, whilst we strive to be peppy and hopeful as much as possible, we also want to prioritise managing expectations and keeping our reporting as accurate as possible as a matter of the utmost importance.
HDBuzz’s Ten Golden Rules for navigating science news
The good news? You don’t need to be a scientist to protect yourself from hype and heartbreak. So, HDBuzz has Ten Golden Rules for reading a press release or scientific news article. They’re here to help you to draw hope from scientific news where it’s warranted – and avoid being let down where it’s not.
1. Be skeptical of anyone promising a “cure” for HD now, or in the near future.
There are promising leads, but no magic bullets yet.
2. If something sounds too good to be true, it probably is.
“Breakthrough,” “miracle,” and “game-changer” are red flags if not backed by details.
3. Has the research been published in a peer-reviewed scientific journal?
If not, it might just be a preliminary result or a publicity push.
4. Is the news about actual research results? Or a new partnership, startup, or funding award?
Hope is good, and investment in HD research from different stakeholders is fantastic, but ultimately results from the lab and the clinic matter most.
5. Has the treatment been tested in people with HD?
If not, no one really knows if it works in people.
6. Has it been tested in an HD animal model?
Even if a treatment seems to have worked in mice, that’s a great first step, but a long way from working in people, and many things fail in the path to the clinic.
7. Has it even been tested in an HD model?
If the research hasn’t yet been tried in a model that mimics HD, it’s still at a very early stage.
8. Watch out for clickbait.
Articles with headlines like “Scientists Stun the World” or “This CRISPR Discovery Changes Everything” are designed to get attention – not tell the full story. Stick with sources that prioritize facts over flash.
9. Look out for overconfident or definitive language.
Be cautious if the language used is absolute – words like “always,” “never,” “guaranteed,” or “proven.” Science is careful and cautious for a reason.
10. When in doubt, ask HDBuzz!
If you’re unsure about something you read or see on social media, drop us a note at editor@hdbuzz.net or use the form at HDBuzz.net. We’re pulling double duty with day jobs as HD scientists and researchers, giving us a unique lens to translate findings from the lab for HD families. And we’re happy to investigate and contact our network of experts to help differentiate hope from hype.
Progress is still progress, even when it’s slow. Every study, even ones that don’t lead directly to a treatment, helps us get closer to our goal. That’s the power of the snowflake – glacier model: steady, collective progress over time.
Using the golden rules in practise
Example 1 – “New drugs cross blood-brain barrier to slow progression and even reverse symptoms of Huntington’s disease”
In this research paper, we learn that a team of researchers from the Weizmann Institute of Science in Israel have identified two small molecules that can reduce levels of the harmful huntingtin protein that causes HD. These molecules work by interfering with part of the machinery that helps coordinate how genetic message molecules are made.
In a mouse model of HD, direct delivery of these drugs into the brain improved many signs and symptoms of HD in this model. Importantly, the molecules were able to cross the blood-brain barrier, and seemed to slow disease progression without noticeable side effects.
This is a really cool study and suggests that these small molecules that target the genetic root of HD could be a promising path toward new treatments.
Unfortunately, in one of the articles covering this research paper, the story comes across quite differently. The headline alone states that these small molecules are drugs that can slow and even reverse symptoms of HD. Let’s go through the rules to see how this article fares. Here is our summary:
Rule 1. Cure claim? ⚠️ Caution – Data suggests reversal of symptoms in mice; not a human cure.
Reversing HD symptoms is akin to claiming a cure. Whilst the data in mice might support reversal of features of HD mouse models, which we use as a surrogate for HD symptoms, this is a long way from a cure for people.
Rule 2. Too good to be true? ⚠️ Caution – Promising results in mice; human applicability uncertain.
This article definitely sounds too good to be true in our opinion. These aren’t drugs, but tool molecules these folks are using in the lab. There is a long road from a tool molecule being tested in mice in the lab to a drug being tested in people in the clinic.
Rule 3. Peer-reviewed? ✅ Pass – Published in EMBO Molecular Medicine.
Rule 4. Actual results? ✅ Pass – Reports on experimental findings in mice.
Rule 5. Tested in HD patients? ❌ Fail – Not yet tested in humans.
This study has not conducted any experiments in people, not that you would know from the headline, which fails to caveat that this work was conducted in mouse models of HD. The data in the paper is supportive that these molecules might help slow symptoms, or stuff scientists can measure in HD mouse models which look somewhat like human symptoms of HD. This is an excellent start, but certainly a long way from showing this is a tractable approach to be investigated in people.
Rule 6. Tested in HD animal model? ✅ Pass – Conducted in mouse models of HD.
Rule 7. Tested in HD model? ✅ Pass – Yes, in relevant animal models.
Rule 8. Clickbait? ⚠️ Caution – Headline may overstate findings.
In our opinion, the title of this article is clickbait. The title doesn’t tally with the subsequent text which goes into more detail about what really happened in the study.
Rule 9. Definitive language? ⚠️ Caution – Language suggests more certainty than warranted.
The title of the article and some of the text therein is definitely peppered with overconfident or definitive language. Claiming to reverse symptoms of HD is a very bold statement, which we don’t believe the data definitively support.
Example 2 – “AMT-130 slows progression in early Huntington’s, 2-year trial data show”
Last July, the HD community had an exciting update from uniQure about their gene therapy drug, AMT-130, currently in clinical trials. The update uniQure provided in their press release was based on data collected from folks treated with the drug over a 2-year time period.
They found that the treatment continued to appear relatively safe, with no new serious side effects since the study was briefly paused in 2022. Most side effects so far are linked to the brain surgery required to deliver the drug. They also reported on a key brain health biomarker, NfL, which typically rises as HD progresses. After an initial spike (likely due to the surgery), people treated with AMT-130 showed a long-term decrease in NfL levels, suggesting the drug might be slowing the disease process. This trend was seen in both low- and high-dose groups at 2 years post-treatment, though the number of participants remains small.
On clinical measures, the high-dose group showed about 80% slower progression by cUHDRS, a sensitive scale that tracks HD symptoms, compared to untreated patients in a natural history study. That’s a potentially big deal, but with only 9 people in that group, we still need to interpret the results cautiously. Other individual clinical scores showed less obvious effects, and data on huntingtin protein levels or brain imaging weren’t shared in this update.
HDBuzz and many others reported on this update. One report from Huntington’s Disease News somewhat missed the mark in our opinion. Let’s go through the rules to see where this piece falls short. Here is our summary:
Rule 1. Cure claim? ✅ Pass – No cure claimed.
Rule 2. Too good to be true? ⚠️ Caution – Significant claims based on small sample sizes.
In our opinion, the article makes claims that are very bold for data from such a small number of trial participants.
Rule 3. Peer-reviewed? ❌ Fail – Data not yet peer-reviewed.
Whilst companies typically work to ensure that all of the data they share in updates on clinical trials are accurate and appropriately interpreted, their reports are not peer-reviewed. In fact these “science-by-press release” updates are often geared towards their financial investors, so will often put a positive lens on their findings. Some healthy skepticism is certainly warranted for updates from all companies in this format.
Rule 5. Tested in HD patients? ✅ Pass – Yes, in early-stage HD patients.
Rule 6. Tested in HD animal model? ✅ Pass – Preclinical testing conducted.
Rule 7. Tested in HD model? ✅ Pass – Yes, in HD models.
Rule 8. Clickbait? ⚠️ Mild Caution – Headline somewhat overstates conclusiveness.
For this particular write up, the title clearly overstates the conclusions of the updates released by the company. AMT-130 certainly seems to be moving things in a positive direction, but to write that it absolutely slows HD progression is an overstatement in our opinion.
Rule 9. Definitive language? ⚠️ Caution – Some overconfident phrasing present.
As laid out with the previous rule, much of the language in the article is too definitive in our opinion. The article claims “Treatment with AMT-130 high dose slowed disease progression 80%”. This is in fact based on data from one clinical measure, cUHDRS, with data derived from just 9 people. The data are encouraging, but this phrasing seems to overstate the facts we have so far.
Example 3 – “AAN 2025: Pridopidine Shows Sustained Benefits on Progression, Cognition, and Motor Function in Patients With HD”
Recently, at the prestigious 2025 American Academy of Neurology Annual Meeting in San Diego, there was a presentation about pridopidine, a drug that activates a brain-protective protein called the sigma-1 receptor. Often, the abstract from talks at meetings like this are published in journals so that there is a record of the meeting and folks can see what research was presented.
From this abstract, we learnt that researchers studying pridopidine had scrutinised over 100 weeks of data to see how pridopidine might affect everyday function (TFC), disease progression (cUHDRS), movement, cognition, and quality of life. They found that for people not taking antidopaminergic medications (like VMAT2 inhibitors or antipsychotics), pridopidine seemed to outperform placebo across all measures, with some benefits lasting over a year. Based on this, the authors suggest that pridopidine could be a promising and safe, long-term treatment for HD.
This abstract was picked up and written about in this article. The HDBuzz team thinks this article falls short of conveying the complete picture. Let’s use the rules to help us work through where this article misses the mark. Here is our summary:
Rule 1. Cure claim? ✅ Pass – No claim of a cure.
Rule 2. Too good to be true? ⚠️ Partial fail – Subtle overhyping.
This article provides little context of the long and complicated journey the drug pridopidine has taken in clinical trials in HD. In its most recent trial, PROOF-HD, pridopidine failed to show benefit in people with HD and did not meet its endpoints. The news article does not provide this historical context and the title of the abstract does not accurately portray the facts of the study.
Rule 3. Peer-reviewed? ❌ Fail – Conference data only.
The conference abstract, although published in the esteemed journal Neurology, is not peer reviewed. This is stated in the abstract itself, but not the news article. This means that no external scientists have scrutinised the study to see if the claims made in the abstract are really supported by the underlying data and should be interpreted with caution until they are.
Rule 4. Actual results? ✅ Pass – Real clinical data, though limited to a subgroup.
Rule 5. Tested in HD patients? ✅ Pass – Phase 3 trial.
Rule 6. Tested in HD animal model? ✅ Pass – Not mentioned but detailed in published literature.
Rule 7. Tested in HD model at all? ✅ Pass – Preclinical work exists.
Rule 8. Clickbait? ⚠️ Borderline – Sensational phrasing in the absence of context.
The article uses phrases such as “Sustained benefits” and “Significant benefits on progression, cognition, and motor function” that imply more certainty than the subgroup analysis supports, and could be deemed clickbait by omission.
Rule 9. Definitive language? ❌ Fail – Overstated benefits not clearly qualified.
There are examples in this article of overconfident or definitive language which lack in our opinion appropriate qualifiers such as “suggests,” “trended toward,” or “needs further validation,” which are standard in cautious scientific reporting.
Final thoughts
In a world flooded with content, from traditional news stories to snappy TikToks, it’s more important than ever to know how to spot good science. At HDBuzz, we’re here to help. We believe in sharing clear, accurate, and hopeful information with the HD community.
With that in mind, don’t forget Rule 10 – if you’re unsure about something you read or see on social media, drop us a note at editor@hdbuzz.net or use the form at HDBuzz.net. We’re happy to investigate!
Science is slow, but it’s moving. And we’re moving with it. Let’s keep learning, questioning, and pushing forward – together.
This week, we heard an update from Roche about their huntingtin-lowering therapy, tominersen, currently being tested in the GENERATION HD2 trial. An independent data monitoring committee (iDMC) that regularly reviews all of the data from the trial recently held their scheduled meeting and made a recommendation to modify the trial design. To cut to the chase and set everyone’s mind at ease – the trial is continuing and there are no major safety concerns. Tominersen still appears to be well tolerated but there are some changes to the trial in regards to the dosing of the drug that the community should know about. Let’s get into it.
Recap – what is tominersen and how does it work?
Tominersen is an experimental drug developed by Roche that is designed to treat Huntington’s disease (HD) by targeting the root cause of the condition: the huntingtin (HTT) gene. People with HD inherit a version of this gene with a mutation at the beginning that leads to the production of a faulty version of the HTT protein. Making this faulty protein is thought to cause damage to brain cells, leading to the progressive symptoms of HD.
One approach being tested in the clinic currently to treat HD is HTT lowering. There are many different HTT-lowering approaches that various companies are testing out, all of which aim to reduce the amount of HTT protein produced in the brain. The idea is simple: if we can reduce levels of the harmful form of the HTT protein made in people with HD, we may be able to slow or even stop the disease’s progression.
Tominersen is a type of therapy known as an antisense oligonucleotide (ASO). These are short strands of synthetic genetic material that bind to the messenger RNA (mRNA) instructions that cells use to make the HTT protein. Once bound, the ASO causes the message to be destroyed, leading to lower levels of the HTT protein overall. Unlike gene therapy approaches, ASOs don’t permanently alter DNA, so their effects wear off over time. While this means that repeated dosing is required, it also means that the dosing of the ASO can be adjusted as needed.
How did we get here with tominersen?
Roche’s tominersen program has had a winding road. Initial trials showed encouraging signs, but a Phase 3 study called GENERATION HD1 was halted early in 2021 after an iDMC found that the safety risks outweighed any potential benefits in the group of trial participants being assessed.
However, an after-the-fact investigation of the data generated in this trial, known as a post hoc analysis, suggested that certain groups of participants, particularly younger individuals with less advanced disease and lower CAG numbers, might benefit from a lower or less frequent dose.
The best way to prove that the findings from the subgroup analysis were real was to do another trial. So even though this trial didn’t give us the results we had hoped for, this important finding sparked renewed interest in studying tominersen and the launch of a new trial called GENERATION HD2.
What was different in the new trial?
GENERATION HD2 has completed enrollment with 301 participants in total. Everyone in the trial is being dosed every 16 weeks over the course of 16+ months. The cohort is divided into three approximately equal groups who are either receiving a placebo, 60 mg, or 100 mg of tominersen by spinal tap. This is less drug, less often than folks in the GENERATION HD1 trial, who received 120 mg of drug every 8 or 16 weeks.
The trial is randomised and double blinded – this means that neither the participants nor the investigators in the clinic know who is getting which dose of the drug or the placebo. This helps to reduce bias in the way the trial is run and data are analysed. Data from people on the drug can be fairly compared to folks given the placebo as they will have all undergone the same procedures and testing.
Now, Roche has shared a new update on the program. Let’s break down what this latest update tells us, what this means for the HD community, and what comes next in the long road toward a potential therapy.
What did Roche share in this latest update?
The iDMC has recently completed a scheduled review of the data generated so far in the ongoing Phase 2 GENERATION HD2 trial. The iDMC is an impartial group that reviews the study data regularly—every 4 to 6 months—to ensure safety and scientific integrity. It’s important to note that the scientists at Roche have NOT seen any of the data yet; this is a completely independent review to make sure things are proceeding appropriately.
The good news from this review? Tominersen continues to appear safe and the trial is continuing. Based on safety and early data, the iDMC had no major concerns with folks in the trial who had received the drug—no new safety issues and no signs of worsening symptoms. This is great news since this was not the case with the very disappointing iDMC review from March of 2021 that ultimately halted the GENERATION HD1 trial.
The big piece of news in the update from Roche is that after the pre-planned interim analysis, the iDMC has recommended stopping the 60 mg dose for the rest of the study and continuing only with the 100 mg dose, which was judged more likely to result in clinical benefit. This doesn’t mean that the 60 mg dose doesn’t work, or that the 100 mg dose does work—it just means that, based on what they can see so far, the higher dose looks like a more promising dose with which to proceed.
What does the data say?
Right now, the only people that know what the data are showing are on the iDMC. While we know that Roche has collected data so far looking at different clinical progression measures (e.g., TFC), brain structure (volumetric MRI), and different biomarkers (e.g., plasma NfL), only the folks on the iDMC know what that data actually looks like. Not even the top researchers on the tominersen project at Roche have access to the data from the GENERATION HD2 trial. This ensures that the trial and the data remain unbiased until the trial concludes.
Because of this, no one knows exactly what data caused the iDMC to make the recommendation that they have to alter dosing. Everyone will have to wait until data is unblinded and released publically.
To maintain the integrity of the blinded trial, participants, researchers, and Roche staff still don’t know who’s receiving what. All participants in the trial will be told about this change, regardless of whether they were previously receiving placebo, 60 mg of drug, or 100 mg of drug. Everyone will be asked whether they consent to carrying on with the amended plan. Those in the placebo group will remain in that group and those on the 60 mg dose will now be moved to the 100 mg dose in a blinded and seamless transition.
What does this mean for the HD community?
While any study change can understandably elicit concern, particularly for those participating in the trial, this is actually a reassuring and encouraging update—and an important one for the HD community. First and foremost, the study is continuing with no major safety red flags. That’s always a critical milestone in drug development, particularly given the history of this drug in the clinic.
It’s also somewhat hopeful that one of the doses—100 mg—was seen as potentially more likely to offer benefit. While it’s still too soon to know for sure what impact tominersen is really having, the data so far are encouraging enough to continue the trial with this potentially more effective dose.
Another point to note is that Roche has taken special care to inform those involved with the study first. Roche communicated the study update earlier this week to investigators so that they could start informing study participants under their care, prior to the broader HD community learning of this update.
What’s next for tominersen?
As mentioned above, all study participants currently on the lower dose will be moved to the higher dose (100 mg to be given every 16 weeks), and everyone will remain blinded to their treatment. Because some participants will be switching doses midway through the study, the final data analysis will need to account for this. But there are well-established statistical methods to ensure the data remains valid and meaningful.
Even with these changes, we still expect the study to end in 2026. So this update to transition everyone to the 100 mg dose is not expected to cause any delays. Roche is working quickly to implement this change to get everyone on the right track as quickly as possible. The iDMC will continue to monitor data from the trial every 4 to 6 months as planned as things proceed forward.
In short: while we don’t yet know if tominersen will work to treat HD, we know that the study is continuing, the higher dose appears to look more promising than the lower dose, and everyone involved is doing their best to move this effort forward responsibly and transparently.
If you are a participant in the GENERATION HD2 study and you have questions, please reach out to your study site directly. Questions from those not in the study can be directed to your neurologist.
Huntington’s disease (HD) is caused by extra repeats of the DNA letters CAG within the genetic code of the huntingtin gene. We used to think that those CAG lengths were stable in most tissues, but we now have a growing understanding that CAG instability contributes to HD. Somatic instability is the concept that CAG repeats expand over time in some types of cells, particularly cells that are vulnerable in HD. Many scientists think this process might play a role in speeding up the development of symptoms.
A current paper delves into some of the machinery behind this phenomenon, asking how CAG expansion is connected with disease, and how science could harness DNA repair genes as a treatment for HD.
From people to animals… and back to people
Our current understanding of the phenomenon of CAG repeat expansion stems from huge human studies in which participants with HD donated samples and clinical data. Their contributions enabled scientists to link subtle differences in people’s genetics with the age when they developed HD symptoms. These studies, known as genome-wide association studies, or GWAS, (pronounced ‘gee-wass’) showed that certain DNA variations could greatly hasten or delay the onset of HD.
Many of the genes identified in these studies that warranted further research belong to a family of genes that help repair DNA. The past few years have seen a flurry of activity leading us to some new conclusions about HD and expansion of CAGs. Here’s a brief recap before we dive into some novel data around this topic:
The DNA repair machinery can slip up when trying to “correct” extra-long CAG repeats – accidentally making them longer and longer!
This does not happen in the vast majority of cells, but seems to occur a lot in a part of the brain called the striatum, which controls mood, movement, and motivation. Studying CAG expansion in the striatum could lead us closer to understanding why these cells are so vulnerable in HD.
Experimental “knockout” or genetic removal of DNA repair genes known to make mistakes on CAGs can slow down or even stop CAG repeat expansion in lab models of HD.
Some of these genes, like Msh3, are the targets of human HD therapies in development – but we still need to understand more about how they influence HD biology, and what are the consequences (positive and negative) of knocking them out.
Setting up the experiment
The authors of a recent paper, led by X. William Yang at The University of California, Los Angeles (UCLA), took a direct and thorough approach to exploring the connection between DNA repair genes, CAG expansion, brain cell health, and even behavior.
They chose a set of nine genes identified in the human GWAS studies and which form part of the machinery that performs a certain type of DNA repair, the kind that drives the “oops” of CAG lengthening. Then they made use of specialized mouse genetics and breeding schemes to create HD mice missing one or both copies of these DNA repair genes.
In each of these HD mice, missing genes like Msh3, Pms1, Mlh1, and others, they could learn more about how messing with DNA repair might affect CAG expansion, RNA message production (the lab’s specialty), toxic huntingtin buildup, and other features of HD. They were surprised to learn that some knockouts had profound positive consequences, whilst others had no effect at all. What they learned is valuable for our understanding of HD biology and the development of therapeutics.
Reversing RNA changes
Our DNA is read or “transcribed” by specialized machinery to make RNA messages, which are eventually used to make proteins, the building blocks of life. There is a whole branch of science that explores the location and amount of RNA message made from different genes – this is the field of transcriptomics.
Scientists can define a healthy mouse “transcriptome” by looking at thousands of genes and asking which are normally turned on and off in different cells, and how much of each RNA message is present. Then they can look at how this changes in an HD mouse over time, or experiment to see what might help restore the mouse’s RNA levels to normal.
The Yang lab was working with one type of mouse that models HD that shows major changes in its transcriptome compared to regular mice. Lots of genes are producing more or less RNA than they are meant to, especially within medium spiny neurons, the cells that are most vulnerable in HD. When the Yang lab “knocked out” half of the Msh3 and Pms1 in their HD mouse model, they saw a partial reversal of the RNA changes in medium spiny neurons. With Msh3 or Pms1 fully gone, the RNA changes were almost fully reversed, often lasting up to a year (half a lifetime for a lab mouse!). Knocking out a couple of other genes – Msh2 and Mlh1 – also had some reversal effects, but these were more moderate. Some gene knockouts had no effect at all.
Members of the Yang lab are world experts in studying HD transcriptomics, and they used multiple cutting-edge laboratory techniques as well as different statistical approaches to confirm their results. They examined RNA levels across many cells, down to the level of single cells, and also looked at how tightly the DNA was wound around its “spool,” known as chromatin. In all cases, knocking out Msh3 and Pms1 seemed to reverse the HD-related changes.
Calming CAGs and clumps
In parallel, the Yang lab measured the amount of somatic instability – the lengthening of CAG repeats – in different parts of the brain and body. In this type of HD mouse, CAG repeats get longer over time, especially within the cells of the striatum. In fact, this group used statistics to define the rate at which CAGs expand in these vulnerable mouse brain cells: it’s about 8.8 CAG repeats per month. (These rates of expansion do NOT apply to humans – these mice start out with 140 repeats and are designed for experimentation.)
The exciting finding is that when the mice had less or no Msh3 or Pms1, that rate went way down. In fact, removing both copies of Msh3 slowed that rate down to 0.3 extra CAG repeats per month, all the way up to 20 months old – that’s basically a stable repeat length, in an old mouse!
At the same time, Yang and colleagues observed that HD mice with half or no Msh3 or Pms1 also had far fewer clumps of huntingtin protein in the striatum. The buildup of these clumps is a classic feature of HD that many scientists suspect could be toxic to brain cells. Removing Msh3 prevented the formation of huntingtin clumps in other areas of the brain as well. The amount of clumped-up huntingtin seemed to correspond with the amount of abnormal RNA changes they’d previously seen.
Furthermore, they were able to confirm results from other labs showing that there seems to be a threshold of CAG repeats – around 150 – above which the cell begins to experience more stress. They connected this threshold to higher levels of RNA changes: CAG expansion speeds up and worsens that stress.
Preventing brain and behavioral changes
We love math and multi-panel rainbow graphs, but it’s even cooler to see a link between genetics and behavioral health. One way in which this work represents a step forward is that the authors show how knockout of the genes that prevent CAG expansion can also have effects on brain cells and mouse movement.
This type of HD mouse tends to show changes in the connections between neurons, known as synapses, as well as enlargement of support cells called astrocytes. The mice also have problems with their gait and movement. However, when the researchers knocked out Msh3, they no longer observed any of these HD-related changes in the mice. This is even further evidence for Msh3’s role in HD, and suggests that it is a good drug target.
Note that this wasn’t a main focus of the paper – they only looked at a few features of brain health and one behavioral task – but it’s still a promising link.
Small steps power future treatments
You have likely noticed that HDBuzz has been harping on somatic instability (loudly and frequently) for a while now, and that we’ve presented lots of similar messaging: CAG repeat expansion seems to be contributing to HD, and scientists have identified some ways to combat it. This work is no exception; once again, genes like Msh3 and Pms1 are culprits that can be “knocked out” to great benefit in the brains of one type of HD mouse.
All these individual advances may appear to be small, but publications like this one represent years of collaborative work among a large team, shaped by frequent input from an international community of HD scientists. We chose to highlight this particular paper because it connects the dots between CAG expansion, abnormal RNA messages, and changes in brain health and behavior.
The authors caution that we need much more information to truly understand the link between Msh3 and Pms1 and the symptoms of HD. They also acknowledge, as we always do, that mice are not people. These mice in particular begin life with 140 CAGs in every cell of their body, which is much higher than even most cases of human juvenile HD. CAG repeats are not expanding anywhere near as quickly in humans as they do in these experimental mice.
Nevertheless, their data, along with that of other labs working tirelessly to understand HD, makes a strong argument for developing therapies based around Msh3 and Pms1. And these efforts are indeed under way!
