Abstract
Jamie Heywood founded the ALS Therapy Development Institute (ALS TDI), the world's first nonprofit biotechnology company, in 1999, after his brother Stephen was diagnosed with amyotrophic lateral sclerosis (ALS) at age 29. His goal was to accelerate the development of new treatments for ALS. The company implemented an industrialized therapeutic validation process to support its drug discovery program. It runs an open research program, posting in real time the results of its studies for patients, doctors, and the research community. Jamie, together with his brother Ben Heywood, and fellow Massachusetts Institute of Technology (MIT) School of Engineering graduate Jeff Cole, co-founded PatientsLikeMe in 2004. Inspired by Stephen's experiences, he conceptualized and built a health data-sharing platform to transform the way patients manage their own conditions, change the way industry conducts research, and improve patient care. PatientsLikeMe gives patients in-depth information on outcomes, treatments, and symptoms they can use to make more effective decisions regarding their care.
J.H.: I don't know if I agree that most people feel powerless. I work with a lot of people who do feel empowered to try and make a change. One thing was that I was really lucky to be at a place in my life and work, and with my family and friends, where I had access to information and resources. Many people are not fortunate enough to have the financial freedom I had—I didn't own a house or have payments or other constraints to prevent me from changing my life quickly and radically. I think if I hadn't been lucky in that way I might not have taken on the challenges I did. I had just sold my house, invested well in the market, and was, overall, doing well financially at the time Stephen got sick.
I was working at The Neurosciences Institute (La Jolla, CA), and I had this crazy cool job, which was to commercialize the biological basis of consciousness. That meant I was surrounded by people whose job it was to teach me their profession, which gave me access to information and teachers that allowed me to learn quickly. I also had a strange, very short experience early on in which my style of learning changed. Stephen's illness focused me in a way that I had never before been focused. I am best at systems integration thinking—that is, distilling a bunch of disconnected ideas into a whole picture that works. It's a skill useful in machine design, for example, and I used to be able to do statics engineering in my head. I could just look at a bridge and know how much of the forces would be transmitted by each of the components. Stephen's illness let me bring that skill to bear on biology. It was that ability combined with access to the great teachers at The Neurosciences Institute, like Joe Gally and Ralph Greenspan, that gave me a sense it was possible to take on molecular biology.
I think there is a little bit of Gerry Edelman in this story as well. He taught me how broken the National Institutes of Health (NIH) and big research institute models of the world were, and how many research opportunities had become corrupted by their funding suppliers. I acquired the spirit of a rule-breaker from him, and he showed me that there was opportunity out there.
I also had the naiveté to think that what I wanted to accomplish shouldn't really be that hard. As we get older and we learn why things won't work, we tend not to try things that we believe won't work, which means that they won't work. I told Jonathan Weiner at the beginning of our talks for his book 27 that I didn't think I had more than a 10% chance of changing what would happen to my brother, and I don't think I changed my perception of those odds over Stephen's lifetime. I didn't end up helping him live significantly longer and better through the means I intended—using molecular biology or a stem cell transplant—but, I did through the use of technology. We improved his quality of life, helped him stay connected to the world, and ensured that he received the best care. I think I extended his health span, and ultimately your health span is more important than your life span.
Stephen gave me this beautiful force that allows you to attack problems hard. I abused that at times, and I may have been too aggressive on occasion and wasted money on things that didn't work. But he was my compass, he gave me the power to break through any barrier I thought was in my way, and he showed me the direction in which to go. He didn't ask me to save him; he gave me the opportunity. It was a gift. He was a skeptical participant in all that we did—one who was transparent and aware of what he was doing and who understood that the odds were against us.
In health care, there is a spectrum of legitimacy, and at some point the system can enslave you; it takes your passion, power, and resources, and it ends up directing you. That is the path most of the groups end up going down because they are so desperate to be respected by the likes of Harvard, MIT, NIH, and industry. But that is not how you solve a problem. There is also the other side of the coin, and the people that go “off the reservation,” set up clinics in Costa Rica, or do things that are illegal and irresponsible. They don't respect that the rules are there for legitimate reasons.
What I think I ended up doing, at least moderately and uniquely well, was to say, “I'm going to listen to why these rules and reasons are here, but I'm not going to accept the conclusions.” I respected the rules, even as I was determined to get to an answer that was compatible with respecting my brother's needs and those of the projects on which we were working. I worked for no person, institution, or peer-reviewed publication; I worked only for Stephen. That's what is really different about me and others who have gone through an experience like this—John Walsh at Alpha One, Kathy Giusti at the Multiple Myeloma Research Foundation, or what is going on now with Josh Summer (an amazing kid who was diagnosed with chordoma). Stephen gave me the confidence to do this, and I had great advisors from the beginning. PatientsLikeMe comes out of that same history, respecting the patients as the focus of what we do.
J.H.: I was divorced and I was using online dating services, looking at women's profiles and descriptions. I realized that the tools we use to assess whom we choose to date, like height, hair color, or personality, are not very different from clinical assessment tools. At the time, ALS TDI was developing a clinical trial and we were designing the data collection. I had a “Reese's peanut butter cup moment”—you known like the old commercial when the guy with the chocolate bumps into the girl with the peanut butter and they realize they taste great together. I had the realization that if you combined the features of a dating website with valid clinical assessment tools you would have something really amazing and transformative.
I recruited my brother Ben and Jeff Cole, and we then began to build it. It was initially called Dr. Wizza as a kind of a code name, but we eventually came up with PatientsLikeMe; we don't remember who came up with it, but we knew it fit as soon as we heard the name. Early on, we struggled with whether it should be a for-profit or nonprofit venture. I understood the limitations of nonprofits and their ability to scale (having run TDI for so many years). If PatientsLikeMe was going to have the impact we believed it could and put patients at the center of health care, we decided it had to be a for-profit.
People often misunderstood why ALS TDI focused on mouse models. ALS TDI was really about using the ALS mouse model in the most effective way to discover whether therapies or targets work. We wanted to use it as a tool to validate that a particular idea would help a human being; actually it was about determining what might work in Stephen. Sadly, what we mainly showed was that others were using the mouse terribly. In fact, we showed and published 28 that most of what was being concluded was just noise. When we ran out of targets from the literature, we then began to look at whether we could use de novo discovery tools like RNA analysis, real-time polymerase chain reaction (PCR), and in situ hybridization to look at what was wrong in ALS through the mouse model without prior hypotheses. Our goal was literally to take apart this model of disease at a deep level and understand what is wrong in a new way, without all the hypothesis-driven nonsense in which you measure things you know or guess in the hopes that you might learn something.
That approach, which was implemented and is led by Steve Perrin, has led to a new generation of drugs that are currently heading into clinical trials. 29 So the focus at ALS TDI has been a combination of rigorously building a validated mouse model and using de novo tools to understand that model. Today, it is even moving into making its own new drugs and biologics, all on a very small budget relative to most discovery programs.
PatientsLikeMe is now pursuing a very similar path. Rather than studying mice, we are going straight to the patients. We are using humans as the model and phenomics as a tool to understand them. The medical industry has been doing this terribly for many reasons. Today, we don't really have true standards of disease; we don't know what depression is, what fibromyalgia is—we don't even know what ALS really is. These are badly defined syndromes rather than formal descriptions in any engineering sense. By formalizing disease measures collaboratively with the patients, PatientsLikeMe is building rigorous models of volunteer humans that will allow us to validate what we know and begin to do de novo learning. Our publication in Nature Biotechnology, 30 which allowed us to show that we could measure whether or not a drug worked in our highly validated population, was the first step toward a new way of doing clinical development. We are able to validate this new tool—the human being in the natural environment—to be an effective development platform, making it more scalable, efficient, and cost effective.
The world has not stopped improving de novo discovery tools, and the ones we have available today are amazing. We, in industry, can now do sequence-based RNA analysis, DNA analysis, and proteomics-based analysis; we can look at exosomes to understand what's going on in tissue-specific cells in blood, and more. We can implement these new tools in the same way ALS TDI does, but the difference is that PatientsLikeMe is about characterizing the human across all life-changing illness and making it valid for discovery. Imagine applying these amazing de novo tools in real time to study this newly validated human. I can envision an experiment in which we would take 100,000 people and measure everything about them, every week or month for years. This would enable us to understand the molecular basis of health; for example, what it actually means to be healthy, what it means to deviate from a healthy state, what it means to have a healthy immune system—one that can defend you but does not attack you? What does it mean to have a healthy brain? Is it inflamed, is it ready to learn? What is true cardiovascular health? What is the balance of human health we are trying to achieve and how does changing things like diet, lifestyle, and exercise in an integrative way bring a person back to his or her own personalized, idealized health at the molecular level?
J.H.: I do think there is a difference in perception between early-onset and late-onset diseases. When ALS strikes a 29 year old like my brother, for example, some have said it was more tragic than someone in their 80s who has just been diagnosed with a disease like Alzheimer and had “lived a full life.” I don't know if that's fair or not. Stephen would have said it sucks regardless, but you have to go sometime. I also agree that preventive geriatrics is a very good description of what we could do with good aging. It is so clear that a well-structured society could give its citizens 10 more years of health span—a society that would encourage and make it easier to live a healthier lifestyle. The productivity benefit of that would be enormous.
J.H.: That is a great, nuanced question. I think it is important to understand who is part of PatientsLikeMe. These are people who are already sick with a significant, life-changing illness. We don't really offer much for healthy people, and that is not our goal. Having said that, my brother was incredibly “healthy” throughout his illness; his skin was bright and clear, his eyes were clear, and we exercised him on a motorized bicycle. But he was dying of ALS. Nothing we did to make him healthier was going to cure his ALS. My belief is that being healthy—which is a very personalized thing—makes a huge difference in disease, but it doesn't end it. I would argue, though, that in some diseases like depression or fibromyalgia you can make a significant therapeutic difference by improving overall health, but not necessarily solve the underlying illness.
I don't know that I would call what we do prevention, even though health is something people deal with a lot in PatientsLikeMe. The goal of our system is to help restore something that is dysfunctional back to health. I think awareness of the value of health is increased by the presence of illness, but for most of our users this will not get rid of their disease.
I tend to break down health care into four categories: (1) Health (fundamentally how healthy and in balance with your environment you are); (2) health-influenced diseases (in which potentially preventable choices a person makes are significant drivers of disease, such as hypertension, type 2 diabetes, lung cancer, and to some degree asthma); (3) no-fault diseases (such as MS, ALS, breast cancer, a heart defect, or infectious disease), which you have due to genetic or environmental dumb luck or some combination of those; and (4) acute health (for example, a heart attack or pneumonia that requires hospitalization). If you evaluate how the health care system addresses these four categories, you get a very real sense of the problems. The no-fault diseases are the ones most amenable to PatientsLikeMe. What we have learned from the patients involved with PatientsLikeMe is that prevention is not the issue; they are looking for disease remission or reduction. I think we can help the world learn about what works in that context. We can run observational comparative effectiveness studies on our human model and evaluate the effects of lifestyle factors, such as diet, holistic living, and meditation on disease.
There are all kinds of great research projects going on out there, such as asking why some people live to be centenarians and looking at the microbiomes of these individuals. What makes a healthy microbiome versus a bad one? What antibodies do people who don't get Alzheimer disease have toward amyloid-β? There is so much to study and it is all fascinating, and we can contribute to that process, but the real value from PatientsLikeMe comes after. Once an antibody that might modulate whether or not a person gets Alzheimer disease is identified, we can begin to look at whether it is present and will have an impact in the real world faster and cheaper than the current choices.
I want to touch on the broader question of prevention. I think so many of us are not cautious enough with this word. Many of us choose to live our lives in a healthy way, but better data about how to do so would be very helpful. I am trying to modify my environment and lifestyle to live healthier, and I admire people who do so, but I don't know if I believe in the imposition of a requirement (or coercion) to pursue healthy living on others. When we talk about prevention, what I hear is the need to restructure the meta-human society so that it serves human health, and not instruct humans to be healthy in an unhealthy environment, which is how we approach it now.
J.H.: Honestly, aging is not in my grid, but that's an awesome question. I love the word “health span” and like to think of antiaging strategies as extending health span rather than life span. I think health care related to aging involves prevention and treatment of disease. But I also think there is something fundamental in the aging process—a clock that is ticking, an imperfect repair mechanism—and that disease is a consequence of, and exacerbated by, that. I also believe that aging and disease are different processes. To get to the heart of those differences will involve looking at the key functional elements of biology at a molecular level and determining whether a person is in a state of balance that is going downhill slowly or quickly. For example, how good is your fast muscle tissue, how quickly can they turn on and off? That will all take a great deal of energy and resources to pull apart scientifically.
My biggest regret about all the things I have decided to undertake is insufficient, early “metricking” of what I was after. In ALS, for example, if I gave an ALS patient (or Stephen) a drug that could stop or slow his disease, I wouldn't know if it had done so for 6 months to a year because of the slow and complicated rate of progression of the disease. In retrospect, I should have first tried to identify a biomarker that could tell me whether a patient, like Stephen, is getting better or worse. If I had done this with Stephen, I could have pursued many more ideas in him much faster.
Similarly, the key to unlocking aging is not just whether resveratrol, fish oil, or aspirin works, but identifying the molecular signature of aging, We can then look at a population and determine whether or not we can adjust that molecular signature up or down. The ability to do that could then be used to test whether all the individual interventions will work. The molecular discovery tools we have now are amazing, and using those tools to power a study to answer these questions—to look at 100,000 well-characterized patients with disease, like we talked about earlier—will change everything in health care.
J.H.: We have spent the last couple of years doing some major upgrades to our system's ability to measure the phenome of human conditions. We have evolved to a deeper understanding that what we are doing is developing a vector measurement of fundamental function, pathology, and health. This is a very different way of thinking about health records; we're not talking about the transactional or procedure-based health records we have been using. The vector model asks you what is important and when you need to know it; it asks you for additional data and tells you what you are missing. We are building a vector definition of health that allows us to compute outcomes in real time in a very new way.
We are not going to be able to do this alone, and so the next big step for PatientsLikeMe is to be a scorekeeper for others. For example, when a researcher wants to look at 100 centenarians (all of whom have shared their genetic information on our site), the assessment should be done on our platform, and the data should belong to the patient who helped create it as well as the researcher. We just announced a new partnership with the RARE Project, an organization dedicated to helping people with rare diseases, designed to drive patients and researchers to use PatientsLikeMe as an open registry for all rare diseases. Anyone, anywhere in the world can use this registry to aggregate patients and then apply the available assessment tools to understand health and what impacts it. We can give these organizations the ability to start preparing their populations for research. We can speed things up for everyone. One of our goals is to be a place where patients can go to give data to the institutions that they respect and care about; the institutions that want to build datasets cost effectively, well, and in a way that is truly patient-centric rather than institution-centric. We want to provide a platform on which the world can build these open, patient-focused registries faster and better.
Stephen gave me a gift. When someone you love is sick and you are empowered to do something about it, you suddenly see the lines and rules that define researchers' decisions and limit their choices—their funding limitations, their interests, their alignments and institutional affiliations. These are the factors that constrain them from doing things the fastest and best way possible for patients. I want a world where we bring those things into alignment with the patients' needs. When Stephen went through his ALS, he gave me the ability to see and understand what held things back; it was an amazing gift. I think, though someone else or some experience has to give you this ability. In general, I don't believe we have it within ourselves.