Print BCTV: Patient Scientists Patient Learning -- How PatientsLikeMe plans to boost impact of its online patient community

Patient Scientists Patient Learning

Transcript of BioCentury This Week TV Episode 142

 

GUESTS

 

Ben Heywood, Co-founder, President of PatientsLikeMe

 

PRODUCTS, COMPANIES, INSTITUTIONS AND PEOPLE MENTIONED

 

23andMe

Robert Wood Johnson Foundation

Nature Biotech

Journal of Epilepsy

 

HOST

Steve Usdin, Senior Editor

 

SEGMENT 1

 

STEVE USDIN: Using social media to create citizen scientists, how PatientsLikeMe is organizing patients to help themselves and contribute to the search for cures. I'm Steve Usdin. Welcome to "BioCentury This Week."

 

NARRATOR: Your trusted source for biotechnology information and analysis, "BioCentury This Week."

 

STEVE USDIN: Stephen Heywood was diagnosed with ALS, Lou Gehrig's Disease, in 1998. His family did everything they could think of to slow the disease. But there's no cure and little to manage its symptoms. It was even difficult to find other ALS patients to share information.

 

The experience led Stephen's brothers Ben and Jamie to create a social media platform for patients. The Heywood brothers and a family friend, Jeff Cole, founded PatientsLikeMe in 2004 as an online community for patients, their families, and caregivers. At first, it was limited to ALS and a handful of other neurological diseases. Today, patients are using the website to share information about over 1,500 diseases and conditions.

 

PatientsLikeMe has thrived because it allows patients to break out of the isolation imposed by disease. Members report on their progress between doctor or hospital visits, document the severity of their symptoms, identify triggers, describe the responses to new treatments, and track side effects. But PatientsLikeMe aspires to be far more than an online support group. The data it collects makes it possible to chart the natural history of diseases, to correlate real world outcomes with medications, and even to run experiments.

 

PatientsLikeMe is a for-profit business. It sells the information its members provide to organizations that are trying to develop new or enhance existing treatments, including some of the world's largest drug companies.

 

I'm pleased to be joined today by Ben Heywood, Co-founder and President of PatientsLikeMe. Ben, why did you start PatientsLikeMe? And what are you trying to achieve with it?

 

BEN HEYWOOD: Thanks, Steve. Thanks for having me. We originally started PatientsLikeMe because I come from a family of three brothers. And my middle brother Stephen was diagnosed with Lou Gehrig's Disease back in 1999.

 

And for those of you who don't know, ALS or Lou Gehrig's Disease is a uniformly fatal illness that patients die in two-to-five years, depending on their age. And our family did a couple of things. One is we started a nonprofit biotech, ALS TDI that is doing hardcore in vivo science around ALS.

 

And the other offshoot of it was PatientsLikeMe. And what we wanted to enable with PatientsLikeMe, which is a free, online service for patients to come and be part of a community of patients just like them, learn from the other patients, and contribute their data and their information and their experience to the collective understanding of disease, as well as receive the social support that we actually are seeing more and more on the web.

 

STEVE USDIN: So you started out around a handful of diseases, neurological diseases, a small number of patients. What's been the trajectory since then?

 

BEN HEYWOOD: Yeah. So we originally started in ALS. And then we moved to some neurological diseases, MS, Parkinson's. And then back in 2011, we opened up the platform to any patient with any condition. And since then, we have added -- we went from having about 40 or so conditions to actually about 1,500 -- plus different conditions in the system, and over 200,000 patients. And we're coming close to doubling that every year.

 

STEVE USDIN: So one of the things that's surprising to me, and I think to most people, about PatientsLikeMe is that it's a for-profit company. And patients, when they sign up for it, they're told over and over again, all of the information that you give besides your name and some identifying information, can be shared, actually can be sold to other companies. How do patients respond to that? Does that change the kind of conversations that people have?

 

BEN HEYWOOD: Yeah. I mean, ultimately I think we are dealing with patients with significant, life-changing illnesses, particularly chronic, life-changing illnesses. And I think what they realize is that we have the front-end network for the patients, where they can engage and interact in a community setting. And on the back end, it's a clinical research platform, where we work with pharmaceutical companies, academics, and as you said, sell them data and research services to understand the patient experience.

 

And the patients with the diseases that we are mostly operating in really actually want to engage the research enterprise, because they feel that those are the places where treatments and interventions for their disease are most likely to come from. So as you've said, we're very aggressive about telling our patients that we utilize the data to monetize it and engage pharma and academics. And that's partially just so there are no surprises. We want them to know what we're doing. But I think overall the primary reaction is that that's a positive, a net positive, that they're engaging in the research and they're engaging in these companies that are actually going to bring them treatments and interventions and that can really affect the care of that disease.

 

STEVE USDIN: So when people think about social media, they usually think about people talking with each other, kind of exchanging anecdotes, support, things like that. You're really talking about data, about generating data. How do you do that in a structured way?

 

BEN HEYWOOD: Yeah. So we have both components, actually. We have what people traditionally think of social media, a place where patients can congregate, have conversations, have qualitative discussions about themselves, both social- and disease-related issues.

 

And then we also have a quantitative side of the site, where patients can track their disease progress. And on the back end, there's a really complex medical ontology that maps the individual diseases around symptoms, treatments, patient-reported outcomes in that disease, ways that they can measure their progress in each of those illnesses. And that's a continuously evolving medical framework that allows them to track and share that information quantitatively as data, so then again, on the research side we can much more agilely use it.

 

STEVE USDIN: So then do the patients actually have access to that data too, and presumably their physicians? Does that help them in monitoring the course of their disease?

 

BEN HEYWOOD: Absolutely. So one of things that patients can do is when they build their profile out, it's public within the community. Any patients in the system can see them individually. Again, not necessarily utilizing identifiable information, but they can see their disease information. And then that is also aggregated up so patients can see the aggregate, what PatientsLikeMe are experiencing for symptoms, what treatments patients are taking for different symptoms. And you can drill down all the way down to the individual level and understand that. So absolutely they can see it.

 

STEVE USDIN: So a patient could look on there and say, I want to find another patient who's had the disease for a similar period of time as I have, who's taking similar drugs, maybe be in the same age bracket or something like that, and what's happened to them.

 

BEN HEYWOOD: Yeah. Exactly right, Steve. What's unique about our site is it's not just a place -- so let's take MS for example, where we have over 30,000 patients in the network. It's not just a place where you can go find another MS patient.

 

But you can go find a male MS patient who's 41 who's had the disease for a few years. And you can see their progress. You can see the treatments they've tried. And you can also see lots of them, so it's a very useful tool. And actually finding a patient like you is one of the main value propositions.

 

We actually published a study in the Journal of Epilepsy about epilepsy patients. And one third of our members had never met another epilepsy patient in their entire life. So you think about such a debilitating condition like epilepsy and never having met someone else with the same condition is just unfathomable.

 

STEVE USDIN: The PatientsLikeMe model relies on voluntary participation. Its value, the patients' and the company's corporate customers, increases with the size of its network. Here's a snapshot of the PatientsLikeMe community.

 

NARRATOR: Now in its 21st year, visit Biocentury.com for the most in-depth biotech news and analysis. And visit Biocenturytv.com for exclusive free content. You're watching "BioCentury This Week."

 

SEGMENT 2

 

STEVE USDIN: We're talking with Ben Heywood about how social media can create citizen scientists.

 

Ben, one of the striking examples from the PatientsLikeMe experience is a trial that patients with ALS, with Lou Gehrig's disease, actually conducted among themselves of a potential therapy. Can you tell us about that?

 

BEN HEYWOOD: Yeah, absolutely. There was a published paper back in 2008 that showed that lithium carbonate dramatically slowed the progress of ALS, which as I said, was a uniformly fatal disease. So there's a lot of excitement. And on our platform, over 10% of the patients with their clinicians started taking lithium and tracking and sharing their progress on the site.

 

We enabled some tools to allow them to characterize the treatments in a little bit more detail, like blood levels of lithium, and dosing, and other things. And we were able to show that in the real world and our population with a higher power than the original study, that the lithium did not actually affect the progress. And three subsequent trials, randomized controlled trials, actually failed to show that as well. So we did that within nine months of the release of the original trial, all through the patients engaging and sharing that information on our platform.

 

STEVE USDIN: How many patients were engaged in that, and was it their idea? Did that bubble up from the bottom or was that something that PatientsLikeMe thought, well, this is a good idea, and let's ask our community if they want to participate?

 

BEN HEYWOOD: No, absolutely. It was all from the bottom. Patients started taking the treatment. And actually, it was an individual patient who wanted to do this experiment so badly that he was going to track it offline using Google Spreadsheets. And we said, well, let's actually create the tools to actually enable the population and the community to actually track and share that data. So it was definitely from the bottom up. And we had over 400 patients who shared their experience of being on lithium. And ultimately, in the trial -- in the paper that we published in Nature Biotech -- that described this trial and our results and the method for determining the effectiveness of the treatment in the real world, we had about a hundred patients in that data set.

 

STEVE USDIN: So do you have a concern, not in this case but in other cases, that you might be encouraging, or patients might be encouraging each other to do things that are dangerous, that are putting their health at risk in this kind of experimentation?

 

BEN HEYWOOD: No. Actually, I really don't. I mean, I think patients are doing what they're going to do regardless of whether there's online tools like ours or whether they're in trials. I think, ultimately, what's more important than what PatientsLikeMe enables is that if patients are trying things that don't work, that experience is captured. And that future patients don't have to go through that cycle of the same bad science or bad treatments that happens, particularly in rare diseases where there's little hope. Patients are grasping at straws in everything. And there's no data to deny that those don't work or do work. So a site like ours begins to collect that information, and then you can actually debunk some of that bad stuff if it's happening.

 

STEVE USDIN: You've recently got an award, a contract, to work with patients to develop what are called patient-reported outcomes. Can you talk a little bit about that?

 

BEN HEYWOOD: Yeah. It's a very exciting project. We just received a $2 million grant from the Robert Wood Johnson Foundation to build an open research exchange, a platform for patients to -- I mean, for patients, disease matter experts, and researchers to rapidly develop patient -- reported outcome measures. So how patients experience their disease. Not many diseases actually have a way for patients to measure and engage in their own health management and measurement. And so this is going to develop a lot of PROs as they're called. We can rapidly develop them on the network, and then we'll also be able to deploy them in the medical infrastructure of the network so that patients will be able to manage their health better.

 

And we hope that measurement paradigm will drive back into the system clinicians, patients. And then, ultimately drug companies and researchers, and clinical trials can start to utilize and leverage those patient-reported outcomes to actually manage and think about what works in the real world.

 

STEVE USDIN: So you might be able to, in the terminology, to validate those patient-reported outcomes so that drug companies could use them in trials, so that FDA might accept them as outcomes?

 

BEN HEYWOOD: Yeah. No, absolutely, we want to validate them. But we also -- one of the important principles that we're going to do on this platform is that we want to actually encourage rapid iteration and better and better improvements. In the traditional method for developing PROs, once it's validated, it's never touched again. And they age over time and there's a lot more data behind them. And so people never want to actually go back and revisit, are they actually measuring as well as they could?

 

And we want to build a system that rapidly allows those patient-reported outcomes to get better and better. And so if you improve a way of measuring walking in MS, well, you can actually think about applying that in other disease states.

 

STEVE USDIN: That's really interesting. Thanks.

 

Most companies have a privacy policy. PatientsLikeMe has an openness policy. We'll talk about that in a moment.

 

First, PatientsLikeMe says it has a not just for-profit business model that has worked with over 20 pharmaceutical companies. Here's a list of some of its drug company partners.

 

[MUSIC PLAYING]

 

SEGMENT 3

 

NARRATOR: Now back to "BioCentury This Week."

 

STEVE USDIN: We're talking about openness, privacy, and the search for better health with Ben Heywood, Co-founder of PatientsLikeMe. Ben, most companies have a privacy policy. You have a privacy policy. But the interesting thing is that it has a link on it to your openness policy. What is that all about?

 

BEN HEYWOOD: Yeah. The openness philosophy came out of this desire to actually get data to flow in the system. And the reality is that there is actually a tremendous amount of health data in our system today. And because of privacy rules and the way that we view it, it's actually locked down. And it's probably harming our ability to actually advance medicine in significant ways.

 

And so we decided that there's tremendous power in the sharing of health information, not just because you can aggregate it up and begin to do interesting research like we're doing, but also because patients can learn from each other. You start to reduce stigmas of disease. And you actually begin to have a dialogue about things that previously the way that the health system is designed those dialogues don't happen and that learning doesn't happen. And think it's been tremendous, I think, the willingness of patients to share and engage in their own health, but through that sharing improve not just themselves but the disease communities they belong to, but also research and contribute to research.

 

STEVE USDIN: I want to ask you about something else. When I first read PatientsLikeMe, I thought it might be something like PatientsLikeMe, like something that physicians might be proud of. And I'm wondering, do you have physicians in the network? What do physicians think about the kind of patient empowerment you're doing? Are they excited about it? Are they threatened by it?

 

BEN HEYWOOD: You know, we've been doing this for a long time. This is our eighth, ninth year in business. And I think that attitudes around information on the internet and engaged patients have really changed over the last nine years. And I think more and more generally physicians today feel like an engaged, activated patient is great.

 

And so we do have some physicians in our network. But we actually have not actively to date reached out to them as much as we'd like. However, we do enable the patient through what we call a Doctor Visit Sheet to print out the information and their progress over time and bring that in for doctor visits so they can actually have a better dialogue with their clinician about what's happening to them between visits, which is very important, where all of health sort of takes place, right? It doesn't happen in a doctor's office.

 

STEVE USDIN: And what about family members?

 

BEN HEYWOOD: Caregivers can join. And you can link up in the system. It's a useful resource for patients and their caregivers to actually learn about the disease and learn from each other as well. So absolutely. And we're looking to expand our sort of caregiver and care team functionality over the next couple of years.

 

STEVE USDIN: Something else I'm wondering about -- is there a kind of a synergy between what you are doing and other social-media-oriented personal health care websites. I'm thinking, in particular, for example of the personal genomics sites, like 23andMe.

 

BEN HEYWOOD: Yeah. I mean, absolutely. I think what we're all trying to do is enable patients to better characterize themselves. Genomics is one point of view, which is how 23andMe sort of started from in terms of their platform. We really from a scientific standpoint think of our platform as a sort of phenomic platform, which really is measuring the real-world progress and status of disease.

 

And so we think uniquely patients have the ability to characterize and identify new phenomic opportunities in that space. I think ultimately the marriage of all these technologies, not just genomics and 23andMe, but proteomics, microbiome, all of the sort of omic technologies that are coming down the river are all going to be incredibly valuable. But they're all going to be incredibly valuable when they're networked into the patient experience. And I think that's ultimately where we're starting.

 

STEVE USDIN: And so that also suggests, obviously, networking it into the electronic health records, doesn't it?

 

BEN HEYWOOD: Yes. Absolutely. I think there is useful data in electronic health records, although I think we think there's a little bit more useful data than there probably is, I think, for particularly the diseases we're in. There isn't very much well-quantified data about the disease captured in an EMR today. That will change.

 

STEVE USDIN: One of the things about PatientsLikeMe that I noticed is a lot of your partners seem to be in the MS space. Is PatientsLikeMe useful for creating comparative effectiveness data, both for companies and for patients.

 

BEN HEYWOOD: Absolutely. I think one of the reasons we have such a strong community in MS, our data quality is very high and it's a very exciting market from a pharmaceutical standpoint. There's a lot of companies developing drugs. And so I think we've had great synergy in that in terms of that community and research.

 

STEVE USDIN: One of the things that's interesting to me, really fascinating in looking at your website, is that you don't shy away from some of the really difficult and contentious diseases. And in particular, you've got a patient community around depression. What are you doing there?

 

BEN HEYWOOD: We've been working in mood disorders, depression, bipolar, anxiety, since 2008. And I think it's actually one of our most interesting communities, because I think, one, just the level of conversations around hospital re-admission and dealing with really difficult, challenging issues in the real world.

 

But secondly, I think from a data standpoint we have a really rich collection. And the way we help patients characterize their illness is not just asking them about their depression like with a Beck's depression scale, but really a multifaceted scale that measures depression, anxiety, compulsion, physical symptoms, so that you can really get the blended, interesting, subtypes of depression, bipolar, anxiety, all of these issues, and really characterize them in a quantitative sense which, you talked about EMRs earlier, there's very little data in mental health captured in our medical system today that's actually quantifiable and usable to actually measure the disease. So our 10,000-plus depression patients, 10,000-plus anxiety patients, as well as the data on depression we have on hundreds of thousands of patients, is in my mind one of the richest sources of data on mental health that's quantified and useful to actually study.

 

STEVE USDIN: So it's interesting when you say you've got people who have diagnosed or diagnosed themselves as having these mood disorders. But you also survey all of your patients, from what I understand, about some of these issues so you can maybe correlate depression or depression symptoms with other drugs, with other diseases.

 

BEN HEYWOOD: Absolutely. Absolutely. So as an ALS patient who's dealing with major depression, you can actually learn from major depression patients broadly, but you also can learn specifically from ALS patients dealing with depression, which might have different characteristics, might have different treatments. For example, my brother was on an antidepressant that actually dried out the mouth, which is a useful side effect for ALS patients, because they have excess saliva. So actually, that side effect was beneficial in the context -- of that drug -- was beneficial in the context of not just treating their depression, but treating some of the symptoms of ALS.

 

STEVE USDIN: And you also have the ability to survey your patients if one of these questions like that comes up, right?

 

BEN HEYWOOD: Yeah. Absolutely. We have a full survey platform that allows us to go back and engage patients about any question later on. So again, when you have patients engaged in the research, engaged in their own health, they become a very strong advocate effort for discovery and research.

 

STEVE USDIN: Thanks. We're going to have more with Ben Heywood on the concept of citizen scientists in just a moment.

 

NARRATOR: Biocentury. Named the 2012 Commentator of the Year by the European Mediscience Awards for excellence in communications and clear, concise commentary.

 

SEGMENT 4

 

STEVE USDIN: In our remaining time, let's get a few final thoughts from Ben Heywood about becoming citizen scientists. Ben, where do you see yourselves, where do you see this movement going a year from now, two years from now, three years from now?

 

BEN HEYWOOD: Steve, I think it touches on a lot of what we've talked about today, which is expanding the network. So obviously, getting more patients in the system and more diseases. But I think it's also beginning to get the other actors who actually care and want to engage in patients where they're at.

 

And so that means in the non-profits. That means the clinical researchers, the pharma that we're in today, payers. You mentioned EMR data. So they're beginning to connect to the various I would say networks that are booting up across the healthcare, but again, through the unique perspective which we try and bring, which is the patient perspective, and starting with them and having them be the connector.

 

And you also mentioned genomics. I think the other area we see us going is into other omic technologies, so wet biology, whether it's proteomics, microbiome, that's a burgeoning field. And I think again, that data is always going to be best if you have the patient engaged in the science so you can come back to them and collect new data sources as they come online.

 

STEVE USDIN: One of the other things I wanted to kind of maybe taking a step back from this whole thing is obviously you're expanding into areas away from just people who have acute, fatal illnesses into chronic illnesses, into other kinds of illnesses. At some point does it get to be too much, that patients become identified with their disease, they become identified as their disease, rather than as individuals?

 

BEN HEYWOOD: So that's a complicated question. I think today we focus a lot on what we call center-of-plate or life-changing illnesses. So it is actually the predominant focus of that patient's life. But what's interesting is I think there's tremendous learning. What's unique about our platform is it's a unified platform across all diseases.

 

And so something like nausea, we have tremendous data on nausea across hundreds of thousands of patients, because that's something that people report on. And so, yes, we have it in the context of those diseases and the patients reporting it. But there's a lot to be learned about characterizing nausea and what are the different types and what you can do about it in the context of such a broad population. And so lots of different ways of again, with the platform there's lots of different ways of collecting data about what I would call not quite as obvious life-changing illnesses in a new way.

 

STEVE USDIN: And I'm wondering also, you've got a lot of partners listed on your website. Do you have any specific examples of how companies have taken the data that they've purchased from you and used it to make better healthcare?

 

BEN HEYWOOD: Yeah. I mean, I think ultimately the research we've done, I think there's a couple great examples. You mentioned the lithium trial. I think what came out of that work in ALS was a very detailed disease model. And that disease model is very useful for pharma companies to actually begin to think about powering up smaller trials in ALS and think about looking at new data sources in ALS.

 

Similarly in Parkinson's, we've showed that the variability of the disease is actually higher in shorter time intervals than previously thought. And again, that goes to actually trial design for Parkinson's patients in thinking about that noise of the variability that was previously unknown because we hadn't measured in such frequency.

 

STEVE USDIN: Thanks very much for a fascinating conversation. That's this week's show. I'd like to thank Ben Heywood. Remember, you can share your thoughts about today's show on Twitter. Just use the hashtag#BioCenturyTV. I'm Steve Usdin. Thanks for watching.