Research news

A recently published study in the journal ‘Neuron’ has identified new potential therapeutic targets for the treatment of Huntington’s disease (HD). The work conducted by Professor Myriam Heiman and colleagues used cutting-edge genetic technologies and discovered several genes could modify HD progression in their models in the lab. Many of these genes have not been linked to HD before and could be exciting new targets for researchers to pursue when developing drugs and treatments for the HD patient community.

Ambitious screening of the whole genome

Cells in our bodies contain DNA encoding thousands of genes, each a recipe giving instructions to our cells on how to make a different protein molecule. These recipes are transcribed from our DNA into a message called the messenger RNA. The RNA is then translated by cellular machinery into protein molecules. Scientists can manipulate these processes in the lab to understand the role of different genes in our bodies.

Genetic screens look to understand the role of a single gene in different contexts, in this case, the researchers were interested as to the role of all the different genes in our cells in protecting against the damaging effects of the HD mutation. So the idea is to mess with every single gene in turn, to try and discover whether that gene has any impact on HD symptoms.

Genetic screen technologies can work in lots of different ways, but they all aim to stop or lower the expression of proteins from specific genes. Genes may be targeted directly by editing the genome itself. Other technologies interfere with the message RNA which is transcribed from the gene and is essential for the cells to make the protein which the gene encodes.

This might sound familiar to HDBuzz readers as these are similar technologies to those being used in huntingtin-lowering therapies which are currently being assessed in various clinical trials. Whereas these huntingtin-lowering therapies target just the huntingtin gene, in the case of this genetic screen, the researchers are targeting every gene in the genome, one-by-one, to work out the role they play in HD.

The Broad Institute, where Prof. Heiman is based, is a world leader in developing libraries which can be used for genetic screens. In this study the researchers used two different technologies in their screen, both of which are delivered into brain cells by special types of viruses. Firstly, short-hairpin RNAs which target the messenger RNA and turn down expression of the gene by interrupting the message from being translated into the functional protein molecule. Secondly, CRISPR was used to directly edit the gene sequence in the genome, disrupting its ability to be switched on to make the protein it encodes.

Systematic genetic screens in different animal models have been around for decades for more simple systems such as worms and flies. However, these types of experiments have been much more technically challenging to perform for the brains of mammals which has been a hindrance to scientists interested in conducting these screens to understand different neurodegenerative diseases.

Prof. Heiman’s team were able to overcome these difficulties by finding a way to pool and concentrate the reagents which need to be injected into the brains of the mice in the genetic screen, and were able to directly target the striatum which is the area of the brain they were interested in studying. The striatum is the most heavily impacted brain region in HD patients, which is why this area was chosen.

20,000+ genes were investigated in this one study

Rather than look at familiar genes associated with neurodegeneration in their mouse models, the scientists in this study took an unbiased approach and completed a genome-wide screen to look at the role played by almost every gene. In fact, they screened nearly all of the approximately 22,000 genes found in mice! This was an incredibly ambitious approach and provides a wealth of data to researchers in the HD field and beyond.

As this was the first systematic screen of all genes in the mammalian central nervous system, the researchers used normal mice with no known mutations to work out which genes are important in brain cell survival in normal conditions. Genes which had been previously identified in systematic screens in more simple models such as flies and worms were shown to be important in mice too.

In this study however, many new genes were identified including several which play a role in metabolism in cells. These were not previously identified in other systematic screens in flies or worms, which is probably because the mammalian central nervous system requires more energy and is more dependent on genes which help the cells make energy. These findings are a good reminder of how important it is for scientists to consider their research findings in the context of the animal models being investigated.

In addition to the control mouse model, two different HD mouse models were used in this experiment, R6/2 and zQ175, both of which are extensively described in the HD research literature. By comparing the genes identified in the screen in the HD mouse models to those identified in the control mice, scientists could work out which genes were specifically important for HD, rather than genes which affect brain cell function more generally.

For the genetic screens conducted on the two HD mouse models used in the study, approximately 500 genes were identified as being important in HD progression. Many of these genes play roles in pathways scientists have previously identified in other studies such as the genome-wide association studies (GWAS) which looked for genes which can alter the age-of-onset of HD symptoms in human patients. These include genes involved with DNA damage repair pathways which maintain the integrity of our genetic material as well as genes in transcription pathways which regulate how the messenger RNA is processed in cells and therefore which protein molecules are made.

New gene targets were identified in the screen too, including genes belonging to the Nme family. Nme genes have been previously reported to be linked to spread in some cancers but this is the first time they have been connected to HD. Heiman and colleagues think that targeting the Nme pathway may be important in helping the brain cells get rid of mutant huntingtin protein in HD brains. If we can design therapeutics which modulate this pathway, this could be a potential way to help treat HD.

New leads for making new HD medicines

Even with lots of ground-breaking clinical trials underway testing different therapies for HD patients, it is important that researchers continue to look for alternative ways to potentially make new medicines for HD. This research provides a wealth of data on HD as it works in mouse model brains and also gives us ideas of new targets to pursue as potential drug targets, which may one day end up in the drug discovery pipeline. It will be exciting to see how these new leads are followed up by researchers around the world and also how this technology might be applied to other neurodegenerative diseases.

A new publication used tiny 3D brain models created from human cells to show that the mutation that causes HD could lead to early changes in brain development. However, it’s clear that HD patients can, and do, develop fully mature brain cells that maintain normal function, in most cases, for decades. So let’s put these findings into context and dig into what these developmental changes that have been discovered using human cells in a dish might mean for HD patients.

Getting human brain cells without collecting brain samples

Even though HD is unique to humans, most organisms have a version of the gene that is mutated to cause HD – huntingtin, or Htt for short. A variety of organisms can be used for studying HD and each model can inform different aspects of how the disease works. For example, if a scientist wants to know if an experimental treatment could benefit HD, they could use fruit flies or even worms to get those answers.

While flies and worms are quite different than humans, they have very short lifespans (about 14 days for fruit flies) so scientists can get their answers quickly. If they want to know what will happen in a more complex brain, scientists often choose mice. But to understand the effects that their work will have in humans, scientists need to test their ideas in humans - or at least human cells.

In 2006 two separate scientists showed that you can reverse the biological timeline of a skin cell, priming it to turn into any other cell type in the body. More recently, blood cells have even been used. These primed cells are called “induced pluripotent stem cells”, or iPSCs.

If scientists are interested in studying a brain disease like HD, they can then turn those iPSCs into the cell types of interest, like neurons. And even better, if the skin or blood cells are from an HD patient, scientists then have everything they need to study the neurons of that patient without having to take a brain sample. Not only super cool science, but also great news for HD patients, who would like to hang onto their brains!

Usually, cells are grown on the flat surface of a Petri dish, but recently researchers have devised a way to coax iPSCs to grow into 3 dimensional balls of cells - which resemble a little brain at an early stage of development. These 3D structures are called brain organoids and are akin to a tiny model of a brain.

Growing these cells in 3D allows researchers to study the way that they organize as the organoid grows, informing very early events in development within the brain. But while these tiny brain-like structures seem to have similar early developmental patterns to a human brain, it’s not a working replica and they don’t possess the capacity for cognitive function.

You are a beautiful and unique snowflake

In a recent study, these brain organoids were used to investigate the impact that the mutation that causes HD has on their development. They did this using 4 different cell lines that are identical in every way, except one: the HD gene. But, wait. How can 4 different cell lines be identical and different?

You can think of people as snowflakes – we’re all unique in our own way, not just with obvious physical differences, like different hair color or eye shape, but also at the genetic level. Everyone has a slightly unique makeup in the code of their DNA that makes them different. So while 2 people may have the genetic code necessary for hands, one may have very long fingers and another may have short fingers.

If researchers take cells from 2 people, one with HD and one without, their cells will not only contain the different CAG lengths of that person’s HTT gene, but will also contain all the other genetic differences that make them uniquely them! This can confuse results a bit though because researchers won’t know if any changes they measure are because of differences in their HD gene or if they’re because of another unique alteration in that person’s DNA.

So back to those identically different cells – to prevent any confusion in their study about whether the results are from different CAG lengths in the HD gene or some other unique DNA code a person has, researchers used a series of cell lines originating from a single cell line that has been genetically altered only within the HTT gene so that it contains CAG repeats of different sizes.

In this case, the CAG repeat tract was increased from 30 (to represent someone without a risk for HD) to 45, 65, or 81 (representing adult-, adolescent-, or juvenile-onset HD, respectively) while all other genes in these cells remained identical. So now the researchers can be sure that any differences they observe between these cell lines are explicitly due to the changes they induced in the HD gene. Pretty clever!

Early-onset juvenile HD may not be a purely degenerative disorder

When using all 4 cell lines to create organoids, the first thing the researchers noticed was that even though organoids from all 4 lines were the same size, the HD organoids developed smaller internal structures that developmentally lead to the formation of important brain cells called neurons, suggesting that brain development is blunted. However, this was only observed in organoids that correspond to adolescent- (CAG of 65) and juvenile-onset HD (CAG of 81), while the organoids that represents adult-onset HD (CAG of 45) were similar to the organoids representing someone without HD.

So what does this mean? The authors interpreted their findings to mean that the mutation that causes HD, particularly in cases of juvenile-onset, stunts brain development. However, an alternative idea is that the mutation that causes HD may just delay development.

To test this, the authors examined older organoids – they measured the difference between the organoids that have 30 and 81 CAGs and found that they still had smaller internal structures, even at this later time point. So it appears that, at least for juvenile-onset HD cases, brain development in these organoids is not just delayed, but rather stalled. However, the adolescent- and adult-onset organoids weren’t included in this specific experiment.

Another key finding from this study suggests that the juvenile-onset organoids develop into neurons more quickly than the organoids without the HD mutation. But if you’ve been staying up-to-date on your HD organoid literature, you may find this a bit confusing because a paper that came out about a year ago found the exact opposite – that HD organoids derived from iPSCs develop into neurons more slowly than organoids without HD.

So does this mean that one study is right and the other is wrong? No. Even though the 2 studies found opposite effects in the speed of HD organoid neurodevelopment, each study was performed slightly differently, using different cell lines and measuring effects at different time points.

What both studies agree on is that the mutation that causes HD leads to early changes in neurodevelopment. But, just because results suggest early changes in development doesn’t mean that these changes can’t be compensated for. In fact, the authors of the more recent study identified a drug with the ability to partially restore the lower measurements they observed in the juvenile-onset HD organoids!

But what about the organoids that represent adolescent- and adult-onset HD? If you’re a stickler for details, you may have noticed that most of the findings of this study just focus on organoids that represent juvenile-onset HD, which represent about 5-10% of the HD patient population. This means these experiments are assessing a rare form of an already rare disease. However, the authors of this study are diligent about interpreting their findings in the context of what their data represents, saying, “Overall, these findings suggest that HD, at least in its early-onset juvenile forms, may not be a purely neurodegenerative disorder and that abnormal neurodevelopment may be a component of its pathophysiology”.

Hot off the presses

One thing to note about this study is that it’s currently published in a repository called BioRxiv (pronounced “bio archive”). BioRxiv is a phenomenal resource because it publishes data ahead of print and is available to everyone. While this gets data out to the masses sooner, it also means that it hasn’t undergone the scientific process of “peer review”, which is an unbiased evaluation of the work by other scientists in the field who are unconnected to the project.

Peer review is critical for maintaining the rigor of scientific studies and provides the authors of the work a thoughtful outside perspective from other experts in their field. Because this study hasn’t yet undergone peer review, reviewers might request additional work prior to publication to clarify some of the results or even request further examination of the organoids that represent adolescent- and adult-onset HD. So you can think of this study like an unfinished book at the moment – we’ll have to tune back in after its final publication to get the full story.

Do these developmental changes ever normalize?

While the organoids are very cool because they can tell us about HD-related changes at the cellular level that occur early in development using human cells, we really need data from patients to interpret the effect that any changes may or may not have on a fully developed human.

Another study did just that and examined the sizes of different brain structures of children and adolescents (age 6 to 18) with and without the adult-onset form of the HD mutation using MRI. These are kids with no symptoms of HD, whose parents have agreed to allow them to participate in research to better understand the very earliest changes caused by the HD mutation.

This study reported a larger striatum (one of the primary brain regions affected by the mutation that causes HD) in HD mutation-carrying kids early on, from age 6 to 11, while HD gene-negative kids have a larger striatum later, from age 11 to 18. So it seems that the gene-positive kids have more rapid neurodevelopment, at least of the striatum, but that gene-negative kids eventually catch up and end up having a larger striatum at the ages examined in this study. However, this difference appears to be quite modest, with only about a 1mL swing – about ¼ of one gummy bear for perspective.

Studies like these that use non-invasive methods capable of detecting very small changes are exactly what’s needed to assess the contribution that HD has on brain development. They will help interpret findings from studies that represent very early development, such as the organoid study in a dish, in the context of human patients.

Ultimately, research demonstrating brain developmental changes resulting from HD is new, and while biologically interesting, researchers don’t yet know what it all means in the context of the disease. However, it’s important to remember that researchers are also working discovering mechanisms that can compensate for any brain developmental changes they report.

DNA-based drugs called antisense oligonucleotides, or ASOs, are now in multiple clinical trials in Huntington's disease, aiming to lower production of the harmful mutant huntingtin protein in the brain. Wave Life Sciences has been running parallel trials of two new ASO drugs, administered by injection into the spine. Just before the new year, Wave announced that the drug in the PRECISION-HD2 trial had successfully lowered the concentration of mutant huntingtin in the spinal fluid. The reduction was quite modest, at 12%, so the company will be adding a higher-dose cohort to both its trials. While the investment community seems disappointed that another trial arm is needed, and we need to see the results in full, to us it's good news that there are now multiple huntingtin-lowering drugs in the world.

Lowering huntingtin

The genetic mutation that causes Huntington's disease does damage to the brain by telling cells to make a harmful protein, mutant huntingtin. Reducing production of this protein - or Huntingtin Lowering - is the biggest focus of drug development in HD.

A drug called HTTRx made a big splash a couple of years ago when it was reported that it had successfully lowered the production of mutant huntingtin in the spinal fluid of HD patients. That drug has been renamed RG6042 and is now being tested by Roche/Genentech in the GENERATION-HD1 trial which will hopefully tell us whether lowering huntingtin production slows the progression of the disease.

RG6042 is a drug made from DNA that interrupts the protein production chain. DNA drugs like that are called antisense oligonucleotides or ASOs.

Wave Life Sciences was the second company to start testing ASO drugs for Huntington's disease. Wave wants to achieve the same aim – lowering mutant huntingtin – but with a twist.

Every person has two copies of the huntingtin gene - one inherited from mom, and one from dad. One abnormal copy is enough to cause HD by causing cells to make the mutant protein. But those cells also produce the normal or healthy version of the protein. Scientists call this healthy version of a gene or protein "wild-type" because it's the one most commonly seen "in the wild".

Roche's RG6042 has equal effects on the mutant and healthy version of huntingtin - it cannot distinguish between the two production lines and is expected to lower mutant and wild-type huntingtin equally.

Wave's ASO drugs aim to target just the mutant version of the huntingtin protein, leaving wild-type production relatively unaltered.

This is much harder to do, which is why Wave had to design two different drugs, each targeting a little single-letter genetic spelling differences that are sometimes passed down along with the mutation that causes HD. These spelling differences don't do anything in themselves, but they can be used to steer the drug to the mutant side of the protein production line, in people who have the right genetic markers in the right place. Wave estimates that about two-thirds of the HD population will have one or other of the necessary genetic markers to make them suitable for treatment with one of their two drugs.

Wave's two trials launched in 2017. They were called PRECISION-HD1 and PRECISION-HD2, testing drugs called WVE-120101 and WVE-120102 respectively. Within each trial, patients were allocated randomly to treatment with the drug or placebo (injection without any drug). Four different doses of the drug were tried as the trial proceeded, which is important to remember as we look at the results of this study. The trials were short - about five months' treatment per patient.

The headlines

Wave's latest press release sets out the first results from the PRECISION-HD2 trial. The release announces that WVE-120102 successfully lowered mutant huntingtin in the spinal fluid, when all of the active treatment arms were looked at together and compared against the placebo-treated group. Wave's announcement gives a figure of about 12% for the degree of mutant huntingtin lowering.

If a drug is working, we expect higher doses to produce a bigger effect. This is called a dose-dependent response, and if you can show it in a clinical trial, it strengthens the case that your drug is doing what you intended. Without giving much detail, Wave's announcement states that the huntingtin-lowering did show a dose-dependent response at the highest doses tested when looking across all of the treatment groups together.

To be clear - Wave has not yet released enough information for us to understand exactly how the amount of mutant Huntingtin in the spinal fluid is related to the dose of the drug given in the PRECISION-HD2 study. We expect that, as commonly happens with these small early trials, more data will become available soon and we'll be able to evaluate this relationship.

Important but easily glossed-over is the primary reason behind the trial: safety. From the information given, the short-term safety looks good. 'Adverse events' were no more common in drug-treated patients than in those receiving the placebo. In itself, that is a very solid result from this first-in-human trial.

Apples and oranges?

The first person to climb a mountain has a tough job, but gets lots of cheers. The second person to the summit may have an easier time, thanks to the first person mapping out a route - but is likely to be asked questions like "how did your time compare?" when they get there.

It's similar with drugs. Roche's RG6042 was the first ASO drug to lower huntingtin, and two years down the line, we have much more detail about how they did it and the full results of the trial have been published. It's inevitable that Wave's results will be scrutinised to see how they compare. Such comparisons may not be terribly helpful, because of the important differences between Wave's drugs and Roche's – but let's do it anyway and see what we can learn.

How does Wave's 12% reduction in mutant huntingtin compare? Well, RG6042 reduced mutant huntingtin by roughly 40 to 60% in patients on higher doses. 12% is less than 40%, so that means the Wave drug is less good, right? Not so fast...

Fundamentally, no drug has yet been shown to slow progression of HD, so we don't know how much mutant huntingtin reduction is ideal. Furthermore, we don't yet know whether reducing only mutant Huntingtin, as Wave is trying to do, is going to be more beneficial and safer than RG6042, which targets both forms of Huntingtin. That's why we do these studies - so we can figure out what approach is safest and has the biggest impact on HD symptoms.

Another important wrinkle to keep in mind is that the doses of drug used in the two trials were very different - the highest dose in the RG6042 study was 120 milligrams and the highest dose tested in the PRECISION-HD2 study was 16 milligrams - that's a big difference!

Based on these results showing their drug was safe at lower doses, Wave has already announced it will now add an extra dosing arm to the PRECISION-HD2 trial, to test higher doses – 32 milligrams per injection. That's twice the amount tested at the highest dose in this trial. So the 12% mutant huntingtin reduction they're reporting may well be a stepping stone to a bigger reduction from a higher dose.

Adding extra dosing arms like this is a fairly common strategy in drug development, where it can be very difficult to predict what dose will be ideal, even if very detailed work is done in animals before going into humans. Sometimes it is necessary to keep increasing the dose, guided by some measure of success, until some hint of a problem is seen, then step back to the previous dose and test that in a bigger trial.

Testing a higher dose will help Wave find whether greater reductions in mutant huntingtin can be achieved, and whether doing so is safe. It may be necessary to go even higher, depending on what the results of the new 32 milligram dose show.

Wave has also added a 32 milligram dose to its other trial, PRECISION-HD1. Because of this, the final results of both trials will now arrive later than initially planned, in late 2020.

Mutant, wild-type and total

There's another complication to understanding these results: remember that the Wave drug is trying to lower the mutant form of the protein without reducing the wild-type form, whereas the Roche drug is expected to lower them both equally.

So even if both drugs achieved the same degree of mutant huntingtin reduction, there is more happening behind the scenes that the headline 'mutant huntingtin' percentage doesn't tell us. We don't yet have any clear idea whether lowering wild-type huntingtin alongside the mutant form makes any difference, and until Roche's big trial completes, we are unlikely to find out.

To us, this is another reason to be cautiously pleased that a reduction in mutant huntingtin has been reported, and wait as patiently as we can for more information.

Talking of wild-type huntingtin – what can we say about whether Wave's drug succeeded in leaving it unaltered while lowering the mutant version? So far, not a lot!

For reasons to do with how awkward the protein is, we can measure the level of mutant huntingtin quite accurately, but there is no direct way of measuring how much wild-type huntingtin the spinal fluid contains. We can measure the total amount of huntingtin in spinal fluid – that's the combined pool of mutant and wild-type. When Wave did that, they found that the drug hadn't altered it.

That might seem weird - if they reduced mutant huntingtin by 12%, and didn't change the level of wild-type huntingtin, then surely the total level of protein should fall by 6%? Possibly - but every measurement has error in it, and simple assumptions like that might be built on shaky foundations.

What's certainly true is that with a small reduction in mutant huntingtin, it is very hard to say anything for sure about the drug's effects on wild-type protein. At this point, we don't think any conclusions can be drawn on that front. We need more information, from more people, before we can start to understand the relationship between changes in mutant and total Huntingtin in the spinal fluid of HD patients in these studies.

Life's complicated

One thing we've noticed in the wake of this announcement is a fair amount of speculation on social media and in the news. There seems to be a 'received wisdom' among investment folks that these results should disappointing for Wave.

We don't really agree with that position, which seems to have come from an over-simplistic comparison of the headline percentages in mutant huntingtin reduction, and the potentially expensive addition of a new higher dose arm.

In fact, RG6042 went through exactly the same process when it was first tested in patients by Ionis Pharmaceuticals. Initially, four dosing levels were planned, but then a fifth, higher dose was added when the trial was already well underway. The main difference here is that Wave has announced their initial results at the same time as the decision to add another dosing arm.

Our advice here is – as ever – to take speculation in the news and especially on social media with a large pinch of salt. Try to get your information from many sources, and if things are confusing, it may well be because nobody knows the full answer.

As scientists driven by progress towards effective treatments for HD, we are interested above all in facts and data. Assuming Wave's announcement is an accurate reflection of the trial data, it represents an important milestone: for the first time, there are multiple drugs in the world that can lower mutant huntingtin in the spinal fluid of patients. Critically, we have drugs that target total Huntingtin, and others that target only mutant Huntingtin, allowing us to compare the risks and benefits of both approaches, in the only place that matters, which is HD patients.

Many questions remain unanswered, and for now we have to be OK with that. What's the best dose of Wave's drugs? Will Wave's drugs slow progression of Huntington's disease? How will they compare with other huntingtin-lowering drugs? These questions will take much longer to answer, and we must be patient and determined to get the trials done and hope that clear answers will emerge. For now, we're cautiously pleased that 2020 has begun with a little ray of light.