Looking back and moving forward: Questions and lessons 4 weeks out from Roche’s disappointing news

Four weeks ago, Roche announced through a community statement that they were ending both their tominersen programme and development of RG6496 that selectively targets expanded huntingtin. Understandably, this was a difficult moment for the Huntington’s disease (HD) community. Thousands of people contributed to these studies, hoping these trials would bring us the first disease-modifying treatment for HD. And while we are closer, it seems these drugs won’t be the ones to get us across the finish line.

A question many people might have is; does this mean huntingtin lowering doesn’t work? The truth is that we still don’t have a conclusive answer, yet. What we do know is that these studies have taught researchers an enormous amount about HD biology, drug delivery, biomarkers, and the challenges of changing the course of this complicated disease. Those lessons won’t disappear because these programmes have ended, but they will shape the huntingtin-lowering therapies that follow. So what have we learned from this disappointing news, what outstanding questions remain, and where does the field go from here?

Huntingtin lowering itself doesn’t appear to be a problem

One of the biggest concerns when huntingtin-lowering therapies first entered clinical trials was whether reducing huntingtin would be safe. Huntingtin is an essential protein, meaning animals that completely lack huntingtin don’t survive past the embryoembryo the earliest stage during the development of a baby, when it consists of just a few cells development phase, and the protein continues to perform many important jobs throughout life. Understandably, many researchers worried that lowering huntingtin might cause serious side effects.

The reassuring news is that this fear has become much less prominent over the past decade. Across multiple clinical programmes, including Roche’s tominersen studies, hundreds of people have received therapies that reduce huntingtin levels. While every drug has its own safety profile and side effects, there is now little evidence that partial lowering of huntingtin itself is inherently unsafe in adults with HD.

That represents genuine progress for the field. The challenge now appears to be something much harder: lowering huntingtin enough, in the right places, at the right time, to meaningfully slow disease.

If huntingtin went down in the latest tominersen trial, why didn’t symptoms improve?

This is perhaps the biggest unanswered question. Roche reported in their community statement that tominersen lowered huntingtin and reduced levels of neurofilament light (NfL), a protein released when nerve cells are damaged and is a general indicator of brain health. Yet despite these encouraging biological signals, Roche has reported that the trial did not show convincing improvement of signs and symptoms of HD.

At first glance, that seems contradictory. If biomarkers are improving, shouldn’t we expect to see clinical signs improve? In theory, yes. But researchers have learned from other neurological diseases, including ALS, that changes in biomarkers can occur months or even years before changes in symptoms become detectable. 

In fact, a good example comes from the ALS drug, tofersen. Treatment reduced biomarkers long before differences in clinical progression became clear. Likewise, the longest-running follow-up data from uniQure’s AMT-130 programme have suggested that separation between treated participants and expected disease progression may become more apparent over several years rather than during the first year after treatment.

Several other HD programmes have also reported encouraging early biological or clinical signals over relatively short follow-up periods. Whether those signals persist, and whether they eventually translate into meaningful long-lasting improvements for people with HD, remains one of the biggest questions facing the field.

Roche’s announcement is understandably disappointing. But it doesn’t close the book on huntingtin lowering. Instead, it marks the end of one important chapter and the beginning of the next.

A possible lesson here is that while biomarkers matter, no single biomarkerbiomarker a test of any kind – including blood tests, thinking tests and brain scans – that can measure or predict the progression of a disease like HD. Biomarkers may make clinical trials of new drugs quicker and more reliable. can tell the whole story. Researchers typically rely on multiple pieces of evidence together to come to a conclusion. In this case, molecular biomarkers such as huntingtin and NfLNfL biomarker of brain health, brain imaging, measures of brain volume, clinical scales like cUHDRS and TFC, and longer-term follow-up. Together these provide a much more complete picture than any single measurement alone.

Did the drug reach the parts of the brain that matter most?

Another outstanding question from Roche’s latest update is whether tominersen reached enough of the brain regions most affected by HD. Tominersen is delivered into the cerebrospinal fluidCSF A clear fluid produced by the brain, which surrounds and supports the brain and spinal cord. (CSFCSF A clear fluid produced by the brain, which surrounds and supports the brain and spinal cord.), the liquid that bathes the brain, through a lumbar puncture. From there, it spreads through the central nervous system, but not necessarily evenly. 

Some brain regions may receive higher drug exposure than others. The striatum, the deep brain region most affected early in HD, is particularly challenging to reach with drugs delivered this way. It’s therefore possible that huntingtin was lowered substantially in some parts of the brain but less effectively in others.

A possible lesson here is that while biomarkers matter, no single biomarker can tell the whole story.

This idea could potentially help explain some of the biomarkerbiomarker a test of any kind – including blood tests, thinking tests and brain scans – that can measure or predict the progression of a disease like HD. Biomarkers may make clinical trials of new drugs quicker and more reliable. findings. If NfLNfL biomarker of brain health levels are influenced by injury occurring across many brain regions, reductions in NfLNfL biomarker of brain health might partly reflect changes outside the striatum, such as in the outer wrinkly cortex or spinal cord, rather than indicating equivalent protection of the neuronsneuron Brain cells that store and transmit information most vulnerable in HD. At present, we simply don’t know.

Studies that may eventually help answer this question comes from work led by Dr. Blair Leavitt and colleagues, who analysed brain tissue from a participant in an earlier tominersen study who had passed. Such rare post-mortem analyses are incredibly valuable, and may provide critical information about exactly where the drug reached within the human brain.

Are we lowering the right form of huntingtin?

Another question has come increasingly to the forefront over the past few years. There are actually lots of different types of huntingtin, and they may not equally contribute to the toxic effects we see in HD. 

Some therapies lower both the regular and expanded forms of huntingtin. Others aim to selectively reduce only the expanded, disease-causing version. Researchers are also increasingly interested in whether particular fragments, such as the short HTT1a protein, may be more toxic and therefore a better target for huntingtin-lowering drugs. Whilst we are learning more about these different flavours of huntingtin in lab models of HD, like mice and cells in a dish, their contribution in human disease remains an open scientific question.

Roche’s decision to end development of both tominersen and its drug RG6496 that specifically targets the expanded huntingtin copy also leaves uncertainty around expansion-selective approaches. Likewise, there have been few recent public updates from other companies pursuing expanded huntingtin-selective therapies, like Wave Life Sciences and Vico Therapeutics. That doesn’t necessarily mean these approaches won’t succeed, but it highlights that selectively lowering expanded huntingtin remains technically challenging.

We still haven’t seen the full picture

Roche’s announcement answered the biggest question, whether the programmes would continue, but left many scientific questions unanswered as no data were provided. We have not yet seen detailed clinical datasets, complete biomarkerbiomarker a test of any kind – including blood tests, thinking tests and brain scans – that can measure or predict the progression of a disease like HD. Biomarkers may make clinical trials of new drugs quicker and more reliable. analyses, or brain imaging results from either trial. We don’t yet know exactly how much huntingtin was lowered, how much NfLNfL biomarker of brain health was improved by, how responses varied between participants, or whether certain subgroups benefited more than others. Nor have we seen detailed MRImagnetic resonance A technique using powerful magnetic fields to produce detailed images of the brain in living humans and animals analyses, which may provide important clues about how the brain responded during treatment.

So called negative clinical trials often become some of the most informative datasets in drug development.

These data will be essential for understanding what happened and for designing the next generation of studies. So called negative clinical trials often become some of the most informative datasets in drug development.

Huntingtin lowering is bigger than one company

Although these two programmes from Roche have ended, huntingtin lowering remains one of the most active therapeutic areas in HD research. And for good reason! The data from various companies seems to suggest we may be able to improve signs and symptoms of HD progression with huntingtin lowering. The key is getting the location, timing, dose, and delivery method right. To that end, today’s clinical landscape is remarkably diverse. Different companies are testing antisense oligonucleotidesASOs A type of gene silencing treatment in which specially designed DNA molecules are used to switch off a gene (ASOsASOs A type of gene silencing treatment in which specially designed DNA molecules are used to switch off a gene), virus-delivered gene therapies, RNARNA the chemical, similar to DNA, that makes up the ‘message’ molecules that cells use as working copies of genes, when manufacturing proteins.-targeting medicines, splice-modifying drugs, and entirely new delivery approaches. 

Some lower all huntingtin, while others aim to target only the expanded form. Some are delivered by repeated spinal injections, others through brain surgery, and others can be taken orally. Some affect only the central nervous system, some a specific brain region, and others the whole body. Each programme is testing a slightly different scientific hypothesis.

And that’s just the huntingtin lowering pipeline! There are other ideas and preclinical data advancing toward the clinic that target different aspects of HD, like somatic expansion. That diversity matters if we want to succeed long term. Drug development is rarely a straight line. Every successful field that has developed ground-breaking drugs, from cholesterol-lowering medicines to cystic fibrosis treatments, has experienced disappointing trials before later breakthroughs. The HD field has wisely avoided putting all its hopes on a single approach. Multiple “shots on goal” remain one of our greatest strengths.

Barriers for HD medicines

Roche’s news also reminds us of a broader challenge. Many of today’s genetic therapies are scientifically remarkable, but they are also among the most technically demanding medicines. Repeated lumbar punctures, neurosurgery, and highly specialised treatment centres will ultimately limit access for many people around the world.

HD is a global disease. Ultimately, the field needs therapies that not only work, but can also reach everyone who needs them. That means continuing to pursue a wide range of approaches, including small molecules, RNARNA the chemical, similar to DNA, that makes up the ‘message’ molecules that cells use as working copies of genes, when manufacturing proteins. therapies, gene therapies, and entirely new strategies, rather than relying on any single technology.

One chapter ends, another begins

Roche’s announcement is understandably disappointing. But it doesn’t close the book on huntingtin lowering. Instead, it marks the end of one important chapter and the beginning of the next.

The field now knows that huntingtin can be lowered in people with HD, that doing so appears broadly feasible, and that meaningful changes in biomarkers can be achieved. The questions that remain will help the next generation of huntingtin-lowering drugs advance us closer to a disease-modifying drug: which form of huntingtin should be lowered, by how much, in which cells, at what stage of disease, and using which delivery method?

Those questions won’t be answered by one study alone. They will be answered by the many clinical programmes still underway, and by the remarkable generosity of every family who chooses to take part.

Summary

• Roche’s decision to end two huntingtin-lowering programmes is disappointing, but it does not mean huntingtin lowering has failed as a therapeutic strategy necessarily. 
• The studies answered important safety questions, showing that partially lowering huntingtin in adults appears broadly feasible, while highlighting new questions about dose, timing, delivery and which form of huntingtin should be targeted. 
• Biomarkers such as huntingtin and neurofilament lightNfL biomarker of brain health (NfLNfL biomarker of brain health) may improve long before clinical benefits become apparent, meaning longer follow-up and multiple measures could be needed to understand whether treatments are working. 
• Detailed data from the Roche trials have not yet been released, and these results will be crucial for understanding what happened and for designing the next generation of HD therapies. 
• Huntingtin lowering remains one of the most active areas of HD research. Multiple companies are pursuing different approaches, while other new therapeutic strategies continue to move toward the clinic.