Us visiting Ionis Pharmaceuticals in April 2018 to discuss the project.

an exciting announcement

For the six-plus years since getting my genetic test report and changing our lives to become scientists, we’ve always been optimists that something would work — that we could find a way to stop my disease before it was too late. But for most of that time, we couldn’t have told you what exactly we thought would work, and we have been skeptical of most strategies out there. Now, for the first time, I am optimistic about a specific therapeutic strategy: I believe it is plausible that an antisense oligonucleotide (ASO) will become the first effective drug for prion disease. In our lifetime.

Excitingly, we — Eric and I and our team at the Broad Institute — have a pharmaceutical partner in our effort: we are working with Ionis Pharmaceuticals to develop an ASO to lower prion protein in the human brain as a drug for prion disease. The company has already developed an ASO for a neurological disease and so has a lot of relevant expertise.

Developing safe and effective therapies for any disease is incredibly challenging. The vast majority of promising research ends up falling short — but we are optimistic. The pre-clinical results so far are now promising enough that we think first-in-human clinical trials could begin within five years. Yes, that’s a long time to wait, and there will be many hurdles along the way — believe me, we feel it every day — but at the same time, it’s not a lot of time to do all the things we will need to do to make the right clinical trials possible.

So, we decided it was worth sharing what’s in the pipeline.

the therapeutic strategy: lowering PrP

The goal of the ASO will be to lower the amount of prion protein (PrP) in the human brain. PrP is the cause of prion disease. This protein is present in all of us, but when it misfolds, the misfolding spreads from molecule to molecule, neuron to neuron, and eventually builds up to a point where it kills brain cells. There exists a large amount of evidence that lowering PrP should be both a safe and effective strategy to treat prion disease.

In terms of efficacy, we know that PrP is what makes up a prion, the pathogen that causes prion disease [Prusiner 1998], and PrP is also what allows prions to be toxic to brain cells [Brandner 1996]. If you inject prions into the brain of a mouse, that mouse will develop prion disease and die about five months later. But if the mouse has one of its two copies of the PrP gene deleted, and so produces only about half the normal amount of PrP, it will survive more than twice as long [Bueler 1994]. If it has both copies deleted, so that it produces no PrP, then that mouse is invincible — it can never develop prion disease [Bueler 1993]. Conversely, if the mouse has extra copies of the PrP gene and produces lots of extra PrP, it gets sick even faster than a normal mouse [Fischer 1996]. So across a wide range of amounts of PrP, it holds that the more PrP, the worse, and the less PrP, the better. People have also done experiments to show that if you turn the PrP gene off or down even after a mouse is already infected with prions, you can delay or prevent disease entirely [Mallucci 2003, Safar 2005]. Therefore, a drug that lowers PrP should be expected to be effective against prion disease.

In terms of safety, many lines of PrP knockout mice — mice that produce no PrP — have been created, and they are healthy, behave normally and have normal lifespans. The first characterization of these mice found no differences between normal and knockout mice [Bueler 1992]. In fact, it took 18 years to confidently identify a change in these mice compared to normal mice [Bremer 2010], besides the fact that they could not get prion disease. Today, we know that full knockout mice with no PrP do develop a mild peripheral neuropathy (a defect in sensation and motor activity in peripheral nerves) as they age [Bremer 2010]. This appears to be because PrP is normally cleaved in half in order to release a signaling molecule that promotes the maintenance of myelin sheaths on these nerves [Kuffer & Lakkaraju 2016]. This problem is very mild compared to prion disease, and importantly, it was seen only in full knockout mice have no PrP at all — not in mice with 50% reduced PrP [Bremer 2010]. You can think of these knockout mice as having an on/off switch for prion protein, whereas in a human medical context, drugs generally act more like dimmer switches. In addition, these mice lacked PrP in their entire bodies, and the defect was seen in a particular location — peripheral nerves — but not in the central nervous system, where a prion disease ASO would be targeted. Beyond mice, knockout cows and goats have also been studied, and reported to be healthy [Richt 2007, Yu 2009, Benestad 2012]. And the few humans we’ve been able to find who naturally have one of their two copies of the PrP gene inactivated are healthy as well [Minikel 2016]. Therefore, lowering PrP should be a safe strategy against prion disease.

why antisense technology?

Modern biologists have all sorts of tricks in their toolbox for lowering the amount of one specific protein — for example, you may have heard of RNA interference, and CRISPR. But as of today, the problem with almost all such technologies is that they are very difficult to deliver to the brain. The adult human brain has about 170 billion neurons, and if you want to make an impact on prion disease, a drug is going to need to reach a large proportion of them. The blood-brain barrier makes it hard for most drugs given orally or intravenously to get into the brain in any appreciable quantity, and even when drugs are injected directly into the cerebrospinal fluid (CSF) or into brain tissue itself, most do not diffuse very well across brain tissue.

Advances in antisense drug development over the past several years have demonstrated that ASOs injected into the CSF through a lumbar puncture (spinal tap) do distribute across the brain and can be remarkably safe and effective. In 2016, an ASO for spinal muscular atrophy (SMA), nusinersen, proved effective [Chiriboga 2016, Finkel 2016, Finkel 2017] — the first truly disease-modifying therapy for any neurodegenerative disease — and was approved by FDA. In its most severe form, SMA causes a relentless decline in function and kills children by the age of 2, but children who get the drug actually gain function and those who get the drug before the onset of symptoms so far seem to be developing normally (see updates here). That said, SMA is a disease primarily of the spinal cord, not the brain, and the mechanism of action of nusinersen is a bit different than the mechanism we would want for a prion disease drug (nusinersen changes the processing of its target, the SMN2 RNA, whereas we want to lower the total amount of PRNP RNA). That’s why it is so exciting that ASOs have also shown great promise in Huntington’s disease (HD), which, like prion disease, is an adult-onset neurodegenerative disease where a drug would need to reach most of the brain and lower the amount of one specific RNA molecule. The Phase 1/2 clinical trial results for HTTRx, the ASO for HD, were announced in early 2018 and showed that the ASO was remarkably safe (no serious adverse events in people who received the ASO) and lowered mutant huntingtin (the bad actor in HD) in CSF by 40% at the highest dose.

So what are ASOs? ASOs are chemically modified short pieces of DNA. They are strings of about 15-30 nucleotides: the same A, C, G, and T building blocks as regular DNA, but with a few chemical modifications that make the drug more potent and stable, allowing it to be injected only every few months instead of daily or weekly. The sequence of the A, C, G, and T building blocks is designed to be the reverse complement (antisense) of the RNA sequence you want to target, so that the ASO binds that RNA. Once an ASO binds its target RNA, it can cause the RNA molecule to be chopped up, thus reducing the amount of the RNA, and the amount of the protein that the RNA encodes. Because all ASOs are made of the same basic building blocks, it stands to reason that if an ASO targeted to the HTT gene, which causes HD, can lower mutant huntingtin by 40%, then it should be possible to develop an ASO targeted to the PRNP gene, which causes prion disease, and lower PrP by 40%. The modular, building-block nature of ASOs means that we can learn a lot from other neurodegenerative diseases where ASOs are already being tested in clinical trials — not only SMA and HD, but also ALS, Alzheimer’s, and frontotemporal dementia.

Our goal, then, is an ASO drug that would be delivered by lumbar puncture, and will bind PrP RNA and lower the amount of PrP in the human brain. Our strategy is non-allele-specific, meaning, we just want to lower all PrP, we are not worried about specifically lowering mutant PrP in people who have a genetic mutation.

planning for clinical trials

We’re optimistic that we’ll have a therapy in human clinical trials in about five years. Even as we continue the scientific research, we already need to start thinking now about how to design the clinical trials.

Clinical trials can come in many formats and designs, and the designs of any future ASO trials are not yet decided. All clinical trials in prion disease to date have recruited symptomatic patients and have sought to increase their survival time or improve their scores on cognitive or functional tests. This is what is called a “clinical endpoint” — a measure of how patients feel, function, or survive, that is used to determine whether a drug is working. This is the standard paradigm for clinical trials, and may be one way to test the effectiveness of an ASO for prion disease in humans.

But there also may be more to the story. Having witnessed my mom’s disease course firsthand, we come to work every day knowing that our personal goal is not just to treat disease once it strikes. Where it’s possible to identify at-risk individuals ahead of time, we want to delay disease, and keep people healthy longer. So we also working to establish a paradigm for clinical trials in pre-symptomatic mutation carriers — people like me — based on a “biomarker endpoint” — a laboratory test that measures the effect of the drug on the molecular level. Because this isn’t business as usual, it will require extra effort, thought and preparation — more thoughts on this below.

what do we have to do to make it happen?

For us and our collaborators, this means we have a ton of science to do in the lab. And there is a parallel track outside of the lab as well: there are things that we as a community need to do to prepare for trials.

1. Spread the word, and engage the genetic prion disease community. Genetic prion disease is rare, and recruiting healthy carriers will be one of our biggest and most important challenges. We need to start spreading the word about this project now so that at-risk families have time to gather information and make an informed decision about how and whether they want to be involved in future trials. ASOs are currently delivered to the brain by lumbar puncture. This is a low-risk procedure, but no procedure is zero-risk. For this reason, it’s likely that a trial in healthy pre-symptomatic people would enroll only known carriers — people who stand to benefit from the drug. At-risk individuals who have not been tested may need time to consider whether they want to pursue testing, and it’s important to initiate the conversation now so that no one feels rushed if and when a trial launches.

2. Refer healthy mutation carriers to In July 2017, in collaboration with our partner patient organizations CJD Foundation and CJD International Support Alliance, we launched the Prion Registry ( to connect patients, carriers and families with research studies and trials. The registry is private and secure online tool. Participants can choose to opt out at any time. There is a place for everyone impacted by prion disease, whether or not they know their own risk status. By referring at-risk individuals to this centralized platform, we can reduce barriers to future trial recruitment.

3. Build cohorts of healthy carriers. Last year, we launched a clinical research study at Mass General Hospital in collaboration with Dr. Steven Arnold. This study is bringing healthy people at risk for genetic prion disease to MGH to donate blood and cerebrospinal fluid to research, and to undergo baseline testing. The goal of this study is to help establish biomarkers in pre-symptomatic people that could be used in a future clinical trial. As with the registry but more so, building teams of motivated carriers ahead of time should reduce the time and expense of trial recruitment down the line.

4. Offer forward-looking advice on predictive genetic testing. In our journey to get my predictive genetic test, we spoke with multiple doctors and genetic counselors. We knew from the beginning that we wanted to get tested, but expert after expert tried to talk us out of it, arguing that a positive test result could only harm us, since there was no treatment available. We didn’t believe this was true even at the time — as I wrote years ago, there’s actually a lot you can do with a test report like ours. As ASOs advance towards the clinic, it is more important than ever that we honor the pros as well as the cons of predictive genetic testing.