The ability to diagnose and treat rare diseases begins with data. The growing awareness about the need to collect data and do so in ways that are meaningful and usable to research and drug development communities, has mobilized a number of efforts to capture and make patient data available. AllStripes, formerly known as RDMD, completed a $50 million venture round in August to help it launch 100 new rare disease research programs. We spoke to Nancy Yu, co-founder and CEO of AllStripes, about the growing efforts around the collection of patient data, where AllStripes fits into this emerging landscape, and how data can transform the outlook for diagnosing and treating people with rare diseases.
Daniel Levine: Nancy. Thanks for joining us.
Nancy Yu: Happy to be here. Thanks for having me.
Daniel Levine: We’re going to talk about AllStripes, how data is empowering patients and changing the rare disease landscape for research and drug development. We’ve seen a big surge in data efforts around rare disease in the past few years. I’d like to start by having you give your perspective on the broader context of what’s happening to fuel this and why is it happening now.
Nancy Yu: That’s a really good question. I’ve seen the rise in this focus as well. And I think it’s a couple of things starting with the industry forces. There’s been a lot of need for engaging patients in their own care and drug research and relying on and partnering with communities directly to understand what’s important to them in drug clinical development, as well as the level of evidence that is now needed to support different drug programs in terms of trial design, supporting FDA submissions, and a whole host of other things that are needed to get a drug across the finish line. So, I think the urgency and the demand for all of this information that has been missing for quite a long time is now coming to a head. And that’s really driven by a lot of the science that has come to market. We have these great gene therapies, gene editing technologies, et cetera, that are now facing this need for clinical development. And that’s where we’re getting stuck because we don’t know much about the disease and there’s not much existing data.
Daniel Levine: People may be familiar with your company as RDMD. You change it to AllStripes. Why the name change?
Nancy Yu: It’s a good question. The original name, I don’t think many people know this, but it stood for a research doctor, medical doctor, and it was really us thinking through, you know, as we’re collecting real-world data from routine clinical care that doctors are listing in their visits with patients and making sense of that information for research purposes. We were thinking that wow, medical doctors are actually research doctors in rare disease because they’re treating patients, but also, contributing to the wealth of research knowledge as well. But over time, I think the name was harder to remember and, as we evolved our platform, we wanted to pick a name that really resonated with the patient community and our mission, which is to unlock new treatments for people affected by rare disease. So, the zebra is a very well-known and well-loved universal symbol in the rare disease community. And AllStripes is kind of a nod to that as well as underscoring all of our efforts around inclusive inclusivity and accessibility.
Daniel Levine: We’ve seen the emergence of a number of different data collection efforts. There are a number of different business models people are using. Where does AllStripes fit into these efforts? What do you see as its strengths?
Nancy Yu: Yeah. At the highest level we have two sides to our company. On the one side we’re serving patients directly and the communities directly. So, we’re partnering with patient advocacy groups, we’re writing our own content about patient stories that really need to be out there to create even more awareness about an individual condition, but also the broader challenge within rare disease. And so, we’re really trying to create a product that patients find real value in the short-term as well as in the long-term. For example, patients can get access to all their medical records longitudinally at no cost to them. And so, we’re doing all that legwork behind the scenes there to also return insights back to the community where we are doing research in that disease. Then on the other side, what we serve as is researchers, and primarily drug researchers as our primary mission is to unlock new treatments. So, we’ll work with biopharma companies of all sizes, academic researchers, and we’re interested in building natural history studies, or even just learning about the patient community and which facilities they’re going to so that they [partners] can best select their clinical trial centers in a way that patients find valuable. So really bridging these two different main stakeholders, patients and drug researchers, bringing them to one platform where we don’t have these classic challenges with exclusivity and who gets access to the data—patients own their own data. They can use the data however they want, and they can have their data be used for multiple different research studies.
Daniel Levine: Well, how does AllStripes work? What types of data does it collect and how does it go about collecting that data?
Nancy Yu: Whenever people think about data, there’s so many different types. And, ultimately when we started the company, we knew that in order to get the depth of information that is needed to support drug development, whether it’s trial design or natural history, and some of this data may be ending up in the hands of the FDA, so we knew we needed very high-quality data beyond simple structured fields from an electronic health record. We knew we needed, for example, pre-diagnostic history, what happened to patients in their diagnostic journey leading up to that diagnosis and then really a lot of detail on progression and outcomes and endpoints and biomarkers as they’ve gone through their journey. This is why we need access to the entire medical record, including the physician notes, which if you’ve seen some of this can be really messy. They’re kind of in stream of consciousness sometimes and totally unstructured fields. So, what we’re doing is getting all of this longitudinal information, unstructured data from medical records and patient self-reported information, marrying that together and turning it into a useful format for drug researchers.
Daniel Levine: And why does this data matter? What can be done with it and how does it change the diagnosis or treatment of a patient with a condition?
Nancy Yu: That’s a great question. So, when you’ve talked to biopharma companies in clinical development, and what 70 percent of rare disease drug companies currently in R&D are saying in early-stage clinical developments of pre-phase 2. Usually companies in that period are trying to figure out what are the end points to look at to design my trial that matter to patients that influence quality of life at the end of the day, as well as something pragmatic for me to design a trial around, what are the gaps in our understanding about the progression of this disease, whether it’s symptoms or different biomarkers that could be really useful to measure to fill in gaps around our understanding and characterizing this disease to the FDA for the first time. For many of these conditions, regulators haven’t had a clear pathway of approval that companies can follow, unlike non-rare conditions where they’ve approved hundreds of drugs in a single condition. So, it’s really difficult to amass what they call the totality of evidence to support a drug’s approval. The types of information that we might help with extend from natural history, which is really looking at what is that patient progression in an entire community and an entire cohort? What does that look like over time? Because only then can you make comparisons against how well your drug is actually doing in a patient community. So that’s really important. And some of these natural histories can be used in lieu of a placebo control arm, which of course is really critical in the rare disease space, because sometimes it’s unethical to subject patients and children to some of these placebo control arms. So that’s a big focus that we hear from our customers. Other things that are less well-known span even to getting the drug reimbursed. As we all know, some of these therapies can cost a lot of money to access. And so drug companies need to tell that story to insurance companies to say, look, we believe that our science can actually save costs for the entire healthcare industry. So Blue Cross, Aetna, you guys got to pay for this. So, generating evidence to support that to make it easier for patients to get reimbursed. And then even more broadly, you know, because so little is known about where patients are, we get a lot of companies that come and say, well, we might just work with these top academic centers at Stanford or Boston Children’s and find all our patients there and serve them there. But we’re seeing that a lot, the vast majority of patients in each disease, they’re actually just scattered across the country and of course the world. And so having even that meta level knowledge about where patients actually are getting care and being treated is really helpful for market planning.
Daniel Levine: In terms of the business model, you work with both patient communities and with pharmaceutical companies. Is it the same business model across the board? Is this a fee for service? How does AllStripes make money?
Nancy Yu: Yeah. And, of course, we need this revenue in order to continue launching our communities. We jumpstart a lot of communities that don’t even have a current life sciences biopharma partner active, because what we want to do is work with a portfolio of conditions where some of them we are hoping to jumpstart interest from drug companies, because if you have the information and you have an engaged group of patients that are really interested in driving forward research, that’s really what gets the interest of drug companies with a new platform technology like gene therapy or something like that. In short, our business model is no cost to patients. We charge life sciences companies who pay to get access to certain software that we provide to them. For example, hosting the data in a compliant database, when they go to the FDA and show the data to them, if it’s supplemental or not, they need to host this data in a compliant manner. So that software that we host that we build and sell, and then also whether it’s surveys or whether it’s consent tools, there’s a whole suite of software tools that companies will pay for. But then beyond that, a lot of the analysis of the data. What I think people often don’t understand is once you get the medical records, a lot of work still needs to be done to extract out information that’s relevant, it’s data that is messy. So just having the data and throwing it over the wall is not how the industry works. It requires our research team going into the data, analyzing it, answering the specific questions for life sciences. And for that reason, we charged that platform fee around research, and into actually analyzing the data.
Daniel Levine: And who owns the data that AllStripes gathers?
Nancy Yu: I liked this topic because at the end of the day, patients own their data. Patients, for example, can opt in to saying, “Hey, I want to contribute all my information to research.” And that’s great. Most patients on the platform do consent. But patients can also remove that access at any time. So, if we are doing something that they don’t believe in, in the future, that’s something that we are really held accountable to. When we started the company, we were thinking, how do we align our business model, which is to serve, to unlock more treatments with our mission, which is to help patients. Right? In that way, because we are relying on the trust and consent and permission of patients to give us access to their information, we wanted to align those incentives. In short, patients own all their data. Of course, they’re giving other researchers the right to use that de-identified information for research, so when I think about ownership, it’s a very interesting topic because yes, you can own the data, but you’re also giving the right to use it to somebody else, which is what I actually think is more important to look at.
Daniel Levine: Well, how does AllStripes work with patient communities or individual patients? Let’s say an organization has reached out to you to start a data collection process. What’s the process you go through?
Nancy Yu: Sometimes we’ll start communities because patient advocacy groups reach out to us and say, “Hey, there’s a big need. We need either a registry, or we need to set up ourselves for natural history in the future because we know that there’s going to be research coming.” Sometimes we’ll get inquiries from life sciences saying, “Hey, we really need to unblock our program and get up to speed really fast on what this disease entails.” So, we can jumpstart communities in both ways. We do a lot of proactive launches as well. For example, we partnered with the University of Pennsylvania’s Orphan Disease Center around their efforts in trying to understand a rare disease called Lesch-Nyhan. There we partnered with the patient advocacy group and worked with them to understand what it is that they are interested in learning about from a research perspective, and then aligning that with what the University of Pennsylvania’s Orphan Disease Center also wanted to learn about so that we are not duplicating efforts. And there we jump-started that community, I think within six weeks, the community was really active and supportive. We blew past the recruitment milestones, almost 2X. Now we’re just looking at that natural history and disease progression, looking at the symptoms. In Lesch-Nyhan, for example, the symptoms are pretty devastating and can include self-injury and aggression. So, looking at triggers for these behaviors and methods used to treat them.
Daniel Levine: The idea of pulling data from an electronic health record, it’s particularly seductive in that it’s sitting there not really benefiting anyone. And from a patient perspective, once I give you permission, I take it my work is done. How much of a workload is there on the patient to collect other types of data?
Nancy Yu: That probably is like a sneak peek into the future of the platform as well and that will evolve. We definitely want to lower the burden on families and patients. I mean, they have a lot to deal with every day, so how it works is patients will sign up, they’ll tell us which facilities that they’ve gone to. We also have ways on the back end to see if there are more facilities that I may have missed so we can get that sort of accurate, consistent view of their journey. We’ll go collect all that information from their different hospitals, so they don’t have to call anyone. They don’t have to upload any papers. A lot of these families carry around these binders of CDs of imaging. And so we handle all of that, over time though, because rare disease is a tough challenge where there will always be missing gaps in our knowledge and understanding of any condition, no matter how full and complete the medical records are. We may still need more information in the future, whether in the form of patient reported outcomes and surveys, which we have on the platform, or over time even getting genetic information. A lot of patients have expressed interest in that. So the types and the breadth of information and data types will grow over time. And that’ll be really interesting to work with the community to see what’s the best way to pull in that information with the least burden as well.
Daniel Levine: One of the challenges is not just building the data side, but once you’re doing this collection it’s getting patients to participate. Is that incumbent on your partners or is that something that AllStripes does?
Nancy Yu: When you mean pulling people into clinical trials?
Daniel Levine: In wanting people to participate in the data collection.
Nancy Yu: Yes, absolutely. I think there’s certainly that other piece around participating in future clinical trials, but in rare disease, I wouldn’t say it’s a small piece of the whole pie in getting to that drug approval, but it is a piece of one of many pieces, right? So whether it’s natural history or understanding disease burden, there’s so many different aspects that need to fall into place. So when patients think of participating in research, there’s some education there initially to say, it’s not just participating in a clinical trial, it’s actually just having your data be contributed in order to build their understanding of this condition more broadly. In that case, we usually work with the community groups who continue to spread awareness. When we have ambassadors, we have over 10 percent of our patient community, our ambassadors, who may reach out to their own networks. Sometimes we do Instagram takeovers for a day in the life of a patient just to create more awareness around specific diseases and what it really means to impact somebody’s life. That helps with generating more awareness of the condition, but also attracting more patients to want to participate. And then over time sharing out transparently what those research insights are, whether it’s partnered with a biopharma company or just our internal research that we’re doing with our research team.
Daniel Levine: I wanted to ask you about two of the partnerships you have because I thought they were particularly interesting examples. The first is an agreement you have with Taysha Gene Therapies. What’s the scope of that and can you walk us through what you’re doing with them?
Nancy Yu: Yeah. The Taysha collaboration leverages the AllStripes platform to inform understanding of SURF1-associated Leigh syndrome as a condition that first appears in infants and young children and affects children over the course of their lives, and sadly does often result in death after a few years. It’s a really huge unmet need. The collaboration will look at natural history and burden of disease, as well as looking at the patient’s diagnostic journeys because only by looking at this information that really doesn’t exist anywhere, and as you mentioned, they’re scattered across the world in these hospitals and maybe not being put to the research use that it could be. So, the focus is on advancing the development of a gene therapy product candidate and development for the treatment of SURF1-associated Leigh syndrome. Taysha will use the deep clinical insights from patients’ de-identified records and data to better inform selection of endpoints for clinical studies. Currently there’s no treatment for this condition. So it’s an important way for patients and caregivers to participate in moving forward research for that condition.
Daniel Levine: The other partnership that was of interest that you mentioned was the one with the Penn Orphan Drug Center. Is it looking at that single condition or is it broader than that?
Nancy Yu: Yeah, it’s actually broader. There’s a couple of conditions we’re working on. I had mentioned, Lesch-Nyhan, but also, we’re partnered with them on Crigler-Najjar syndrome type I, CN type 1. Here, in both cases, we’re looking at the natural history and disease progression of both conditions, looking at those symptoms and we’re still recruiting patients for the CN1 study. For Lesch-Nyhan, like I mentioned, we’ve exceeded our goal there and we’re working on extracting out these insights and collecting survey information from families through the platform. We’re looking forward to sharing that information, what we’re learning, back to the community with Penn ODC later this year.
Daniel Levine: I know both Taysha and Penn look at a lot of different but related syndromes. Are either of them or anyone else using AllStripes to look across diseases?
Nancy Yu: Yes, I think that is going to be super interesting. Right now in industry, of course, how drug programs are set up and even in academia, you want to focus on an individual condition, but as we know how conditions are named sometimes can be pretty arbitrary. This is why we need to look at the specific manifestations or characteristics of the conditions, breaking down a disease into characterizable building blocks. For example, what is the audiology impact? What is the neurocognitive cognition impact? How does this condition affect mobility and ambulation? So, the research team has developed, dozens of these clinical data modules that are pretty standard across diseases, in looking at, okay, how does neurocognition affect these whole suites of conditions versus just looking at the name of the disease and splitting out treatments that way. So I think you’re onto something there where as we build more conditions in the community and as more and more companies are focused on specific therapeutic areas of interest, they are definitely going to look across conditions.
Daniel Levine: AllStripes recently announced a $50 million venture round. How far will that funding take you and how are you going to be using that money?
Nancy Yu: We think that funding’s going to take us for quite a number of years. And of course, with everything in building a company, things may come up earlier in two, three years where we might decide to make even more investment into the platform if we see that there’s a need from the patient community or from our partners. So, it’s kind of TBD, but we think that’ll take us quite a long time. $50 million is a lot. The main thing that we’re focused on is, of course, continue to build the team. The team is definitely our biggest asset in terms of building the things that we need to get going. But in particular, we want to launch a lot more communities—launch a hundred new communities in different conditions. Another is to continue to expand our data types—sources of data, breadth of data, depth of data, and quality of data so that we can continue to generate more insights. As part of this, a big effort around automation of this information so that we can scale to more and more communities, while not impacting costs that much for all of our partners, lowering the cost of research—a lot of technology investment there and expanding those capabilities even globally. Rare disease is of course a huge global challenge, being able to not just engage and find patients in different countries, but also get that medical information. That’s so critical. That’s a big effort in privacy in research, IRB, ethics, and, of course, just general global expansion around translation.
Daniel Levine: At some point, is there an economy of scale to what you’re doing? Does it incrementally lower the cost to onboard patients or communities as you build out?
Nancy Yu: Yes. I mean our vision for using technology to do all of this, to automate pieces of this. When you hear a drug company is doing it the old way, the traditional way, you kind of slap your head and like, wow, there’s so much opportunity for cost savings here. For example, just getting records from different hospitals, we’ve collected information from over 3,500 facilities at this point, and that list continues to grow. With every additional facility you’re getting information from, you build those economies of scale. Similarly, with structuring the data, with our print technology today with natural language processing and optical character recognition, you can actually make big strides in how you extract information from an unstructured format into a structured format. And the more patients we have on the platform and the more conditions we have on the platform; we get to learn more about how this replicates to every new community while making it still a fit for purpose and making sure that we answer the specific research questions of interest.
Daniel Levine: Nancy Yu, co-founder and CEO of AllStripes. Nancy, thanks so much for your time today.
Nancy Yu: Thank you for having me.
This transcript has been edited for clarity and readability.