NewsJan | 19 | 2023
The process of getting a drug to patients is cumbersome and heroic. For Drs. Sonia Vallabh and Eric Minikel, who founded the Prion (pronounced “pree-on”) Alliance to develop therapeutics for human prion disease, their path has been, they say, like “driving at night in the fog,” quoting E.L. Doctorow: “You can only see as far as your headlights, but you can make the whole trip that way.”
We recently talked with Sonia about where she and Eric find themselves now, more than 10 years since Sonia discovered she had inherited a genetic variant that puts her at high risk for prion disease. We wanted to know not only where are they in identifying drug candidates to treat the disease, but where are they in the process of clinical trials to get drugs to patients?
Drs. Vallabh and Minikel’s lab is based at the Broad Institute of MIT and Harvard, and they are also on the faculty at the McCance Center for Brain Health.
Going back to the beginning, we got married in 2009. Eric and I (Sonia) were training in non-biomedical careers. I was in law school, and Eric was getting his master's in transportation planning.
And then, in 2010, my mom got sick, with what started as not a very clear story. At first she had a mix of miscellaneous symptoms. We didn't know how they related or whether they were a big deal. But then her symptoms began to snowball, and from there she declined incredibly quickly. And what she was losing was her ability to communicate and coordinate activities. Once we had the full story, we could see how the central driving force was that her brain was dying. But early in her illness, we were looking at these isolated “islands” of symptoms and problems, and trying to connect the dots.
What was shocking to us is that no one in the medical system was able to connect the dots for us, and give us a diagnosis in her lifetime. Very naively, perhaps, before that I hadn't considered the possibility that you could have somebody dying in front of your eyes with such a spectacular phenotype and still not be able to get a name for it. Even a very bleak prognosis would have been better than a total mystery. We lived in that place of total mystery for a year.
My mom passed away at the end of 2010, and it was from her autopsy that we finally learned that she had died of genetic prion disease. And the genetic piece came as a surprise to us.
As not much more than newlyweds, this put us in this position of having to figure out, do we want to act on this and get predictive genetic testing? And our answer was “yes,” because I think to us it was really clear that we'd been thrown into limbo, whether we liked it or not. And staying there was a much less tolerable choice than finding out one way or the other how the 50/50 coin flip had gone for us.
So we got the testing at the end of 2011, and learned from my genetic test report that I'd inherited my mom's mutation. And at that point all we knew were the top line facts about the disease, which remain true now: prion disease is a rapidly progressive, universally fatal, neurodegenerative disease.
The tempo of prion disease is somewhat unique. As we would learn, my mom's case was not unusual in terms of how rapidly it had progressed. But before those first symptoms, it appears that people at risk remain healthy for some number of decades. New data from our own lab and others support the idea that it’s really at the point of symptoms, or not much before that, that things start to go wrong in prion disease. One critical point is that the age at which the disease will begin is not predictable for any given person: we can look at averages, but they don't really tell you, with any confidence, when it'll happen for you. All we can say is that there are a handful of mutations in the prion protein gene, including mine, that convey a very high risk that at some point in adulthood this will happen.
To begin with, we were coming at this with very little scientific background. And though of course we grieved for a little while, one thing that happened surprisingly quickly after we got my test report, within just a couple of weeks, is that we started talking to scientists, local friends, and others who worked in research. We were sort of “pulling back the veil,” learning how the research picture was different from the clinical picture.
The clinical picture was just so unremittingly bleak. And we had just been there with my mom, so it was actually a very interesting breath of fresh air to realize that it wasn’t that nobody knows anything, which is certainly what it seemed like from the clinical perspective. And it also wasn’t true that this disease was so rare that no one could ever possibly spend time thinking about it, which is also what it had seemed like from our recent experience with my mom.
Instead, we learned, there are some number of research groups around the world for whom this is their thing. And there are actually a handful of Nobel prizes that have been awarded in prion science. It's not a total unknown. That was so encouraging that we started reading more about what's known, and this took on a certain amount of momentum and became the thing we wanted to be thinking about.
So just a couple months after getting my test report, I was leaving my job to take my first research job at Mass General.
And as we would discover, this was the nature of everything for us. As E.L. Doctorow once said of the writing process, “It’s like driving at night in the fog. You can only see as far as your headlights, but you can make the whole trip that way.” We made each decision as it came up.
It wasn’t that we immediately had a vision that we were going to develop a drug ourselves, but the thing that we could see in the headlights was that we needed to learn more about the science of this disease. And then once we acted on this first step, the new thing that became visible was that we probably could make a go of doing research for a living, and we probably weren't going back to our old careers.
And then within two years, the thing that we could see was we wanted to find a way to be focusing our day job on prion disease in particular. So again we went out and talked to everyone who could possibly advise us. We solicited a lot of advice on how to focus this mission and, somewhat surprisingly, the balance tipped towards enrolling in a PhD program, and then figuring out a way to spend our “trainee” years already always working on our mission. This was a big gamble, because it's not every advisor or institute that would be open to this proposal. But we connected with Stuart Schreiber and Eric Lander at the Broad Institute, and through them we landed here at the Broad, where we were immediately supported to get started . This was an absolutely amazing stroke of luck for us.
We started our PhDs in 2014, defended in 2019, and since then have continued our research in the form our own independent lab physically based at the Broad. All along, the drive for an effective therapy has remained our “north star,” even if the steps to get there are constantly changing. And one of the ways in which our mentorship network has been hugely supportive of us is in supporting us to pivot in real time according to what we learn, and what we best understand our “job” to currently be.
One big insight from our PhD years is that while we had initially conceived of ourselves as the people who needed to discover the molecule, we came to understand that our job was much bigger than that. We saw that to ever do a successful clinical trial, we were going to need a lot of other infrastructure and a lot of other building blocks. We couldn't afford to wait to assemble all of these pieces serially, and we couldn't necessarily count on anyone else to do it.
So we essentially launched a series of parallel programs, with many different orthogonal pieces that had to be moving together. This is how we started working on biomarkers and natural history and a patient registry and regulatory engagement and genetics and genomics — all of these needed to be moving at the same time, along with what we might call “drug discovery.”
Another big, related insight was coming to realize that there are two patient populations in our disease and they need to be treated really differently: symptomatic patients, who decline very quickly, and pre-symptomatic individuals at high risk like myself, whose risk we can know of years or decades in advance of disease onset.
Success could look really different in these two populations. The science of intervention could be really different. And as a consequence, clinical trials might also have to look really different. They might require different tools, or even a totally different framework than what precedent might suggest.
For the first group, symptomatic patients, the mandate is to make a diagnosis early and intervene as quickly as we can. But from that point on, we imagine a clinical trial in that population could have many traditional characteristics. We would treat sick patients and we would ask, do they feel or function better or survive longer? And you'd aim toward a traditional clinical readout. There would be challenges around doing it right and doing it rigorously, and getting to people early enough — but the framework for that kind of clinical trial exists, and it has been done before, even in our field.
For the second population, for me people like me, the path was less obvious at first. We were really lucky to be pushed early on to think about what it would mean to do the most good in this population. The lived experience of going through my mom's disease was important to this thought process; we just had to look at what it had to teach us. And one key learning for us was that with prion disease, if you let the disease get a head start and you’re chasing after it, you are going to be hard pressed to do as much good as you could have by preventing the disease from getting a foothold in the first place.
We work in a biomedical ecosystem that is very focused on the knowledge that we’re assembling, and the arsenal we’re building, to go out and defeat disease. I absolutely love this vision and this energy and couldn’t be more grateful to be here.
At the same time, Eric and I operate within the time horizon of what can be accomplished within our human lifetime. And sometimes this requires us to take a step back and take a sober look at the limits of our knowledge today. To take a humble look at the massive complexity of the brain and just how much we don’t know about it. We are not, right now, in a position to rebuild functionality that is lost. Instead, we can aspire to be good stewards to the brain. For people living at risk today, that's our highest calling. The most good that we can do is to get to them way before the disease does.
Picture the alternative: we can watch someone for 10 or 20 years of good health, wait for them to convert and then chase them during their six month or year-long decline, and try and desperately to salvage any quality of life. If we picture this process prospectively, we can see that this wasn't a strategic use of time. This is the insight that led to a lot of the work we've done specifically with the carrier community: quantifying the penetrance of different variants in the prion protein gene, collecting data on age of onset, bringing carriers into MGH for our natural history study, looking at biomarkers in this population, and talking to FDA scientists to understand what would it mean to do a fully preventive trial.
The word prevention is used in so many different ways, and I think all of them have an important role to play in where neurological interventions need to go. But when I look at which one means the most to me as a patient-scientist, it's primary prevention.
For me, primary prevention is the idea that we're not waiting for a sign of disease to intervene, be it a biochemical signature or imaging signature or the detection of misfolded prions in the spinal fluid. We’re not waiting for any of that because all of that means this runaway disease process has already begun.
When we say “prevention,” I think I think a lot of people imagine a “secondary prevention” paradigm that pivots on detecting a sign of disease — in part because this scenario is more familiar and more often discussed. And there’s a certain comfort to the idea that you're intervening after you have some call to action. But our message is the call to action is in my DNA — it's already there — and this perspective creates a whole new mandate for figuring out what you're going measure in a trial, figuring out how you're going to get the people into your trial, and just thinking about the parameters that a drug needs to have.
The first conversation we had with the FDA on the subject was in 2017, it went incredibly well. We went in wondering if our proposal for primary prevention was too radical. Instead we learned that some of these scientists had seen prion disease patients in their careers, and they were saying to us, “Of course you have to prevent this disease.” We were thrilled to feel like the high-level buy-in was there. Executing on the idea is what we've been doing ever since that first meeting. And while it’s been a lot of work, I think we've been really lucky that FDA scientists have been willing to advise us on the data we need to collect.
In the years since then, we’ve been able to gather a lot of enabling data, and I’m proud of the progress we’ve made. It’s been a heavy lift in terms of nailing down the relationship between the intervention, the biomarker, the disease process, and survival. The basic therapeutic hypothesis is that this disease is caused by the misfolding of a normal protein, the prion protein, in the brain. So we propose to preventively lower prion protein levels in the brains of healthy people at risk. We further propose to measure the level of prion protein in their cerebrospinal fluid (CSF) before and after treatment. And if we see a treatment-related reduction, we expect this reduction to predict a clinical benefit for these carriers. Because this kind of trial relies on a prediction of clinical benefit as the first step instead of a direct observation, we want to lay a lot of groundwork and make sure we have really closely linked each of the different “nodes” in the causal chain.
This is all doable. It does all take time. And I think there's also a degree to which just being different takes time. So that's been part of the challenge of the past several years.
Nothing is certain in biology, but I think our therapeutic hypothesis is really well supported: based on everything we know today, lowering the amount of prion protein in the brain should be both effective against disease and well tolerated. When we think about testing a prion protein-lowering therapy in healthy people at risk, I think in principle the prospect for benefit outweighs the prospect for on-target risk — that is, the risk associated directly with the goal of the drug, the lowering of prion protein levels.
But then you have the sort of imperfect reality of these different genetically targeted therapies that are all quite new, and that we are still learning about. In the years that we’ve been working with Ionis Pharmaceuticals to develop a prion protein-lowering antisense oligonucleotide (ASO), we have continued to learn from ASOs tested against other brain diseases. And some of the news has been awesome, and some has been disappointing, and some we are still trying to understand. I do think there are platform uncertainties around really all new drug modalities.
To me one key question as we move forward will be therapeutic window. If you dose any drug high enough, this will cause problems. For us, we want to lower prion protein levels as much as possible. We think it's a closely dose-responsive phenomenon where you lower prion protein levels a little bit, you do some good, but the more you lower, the more good you do. So given that these drugs aren't trivially deliverable to the brain, a close relationship emerges between the distribution properties of the drug, how you dose it, potency across the different regions of the brain, and how well the drug is tolerated. At any given dose level, are you lowering prion protein enough on the one hand, while still staying within a dose level that is safe, based on the properties of the drug?
Planning for clinical trials requires a number of these incredibly delicate balancing acts. In some ways this process is its whole own mountain to climb. Having de-risked the science of our therapeutic hypothesis at least to a degree, we now see a whole new vista of challenges ahead that are perhaps best described as a blend of scientific and human systems challenges. There are many different parties that bring their own principles to the table for what a first experiment in humans needs to look like, how it should be designed and run, how long it should go, and how it should read out. It isn’t always easy to keep straight what’s required according to our own judgment and principles and conscience, versus what others bring to the table.
This process is enormously complex and time-consuming, and it can be an unsatisfying place to languish for too long because ultimately all parties involved — academic researchers, pharma collaborators, FDA scientists — want the same thing. Everyone involved in this discussion is devoting their life to the development of new therapies. They all want a safe and effective drug for prion disease.
Ultimately, multi-party decision-making in the face of tremendous uncertainty and high stakes is just hard. And I do fear that in any highly uncertain context, convention can take on an outsize role. And in this way, I fear that you can end up walking through steps that take a lot of time, but weren’t crafted to fit a particular use case, and perhaps don’t perfectly fit the risk-benefit profile of a given disease.
This is a challenge we’re learning how to navigate now. At this point we've been on this quest long enough for me to understand that each phase brings an entirely new retraining process and an entirely new chief challenge. But compared to our starting point 10 years ago I think we're very lucky to now be asking tough questions about human clinical trials, with tools on earth that can at least theoretically do the job we need them to do.
So that's where we are right now. And in a few years it'll be probably something different. Let's hope! While it’s impossible to say with certainty, it’s possible that the first ASO trial will begin in the next year. No matter what comes out of it, I think the first trial of the first prion protein-lowering therapy in humans will be a huge step forward and a tremendous opportunity for new learnings. And we will keep moving forward and keep updating our understanding of what “forward” looks like.
