Slone Partners’ 2018 PMC Interview Series enters its 4th week with an exclusive, in-depth interview with Colin Hill, the Chairman, CEO and Co-Founder of GNS Healthcare, an innovator in artificial intelligence (AI) applications in personalized medicine. On a mission to “unravel the complexities of human disease” utilizing causal machine learning, GNS Healthcare serves health plan providers, biopharma companies, academic medical centers and research organizations. This ongoing series of healthcare executive discussions is presented in partnership with Personalized Medicine Coalition.
See Colin Hill participate in the panel discussion Automating Actionable: How Artificial Intelligence May Chart a Course for Personalized Medicine November 14th at the 14th Annual Personalized Medicine Conference in Boston. For more information, visit www.personalizedmedicineconference.org
Slone Partners: You’ve been Chairman and CEO of GNS Healthcare, a trailblazing data-driven company that leverages artificial intelligence and machine learning in personalized healthcare, for almost two decades. What’s your professional journey been like in terms of nurturing people, personal leadership growth, and striving for breakthroughs?
Mr. Hill: When Iya Khalil and I founded GNS twenty years ago, the idea of leveraging artificial intelligence to discover insights around human biology was viewed as both bold and maybe a little crazy. The human genome was about to be mapped, electronic health records were not widely used, and supercomputers were slow, clunky, expensive and not cloud based. The idea of precision medicine had not yet been articulated. Today there are a significant number of businesses founded on AI, precision medicine and machine learning, and the list grows daily. As you can imagine my professional journey has been equal parts tough, humbling and rewarding.
When we started GNS, we had to continually work to prove our value, but perhaps more importantly to educate the market on this strange thing called causal machine learning and simulation (a powerful form of AI that reverse-engineers the system that created the data vs just finding patterns and correlations). We still do a fair bit of evangelizing around causal AI but more and more we let our growing body of results and peer-reviewed science show their impact.
As CEO my role has primarily been external facing because I needed to be out there promoting and explaining GNS to prospects, clients, investors, and media. Frankly, I would have liked the opportunity to spend more time mentoring employees and the chance to focus more on the internal operations of the company.
We have been extremely fortunate to attract and retain the caliber of employees across multiple disciplines that we have at GNS. Our employees are some of the smartest people I know, and they are drawn to our mission of disrupting biomedicine and healthcare and ensuring we develop and apply the best technology to improve people’s lives. The camaraderie of the team never fails to amaze me; there’s a real culture of “How do we figure out this tough problem together?”
In terms of personal leadership growth, I am continually humbled by the mentors I have had, both formal and informal. I have been fortunate to be part of boards whose members and CEOs have taught me so much about running a company. Our GNS board members continually teach me things that I use in my role as CEO. But perhaps where I learn the most is from sitting down with our employees, especially those on the front lines – I always learn something new and important about our business that I can use to make GNS better.
Striving for breakthroughs is in the very DNA of our company. We started by manually creating computer models of cancer cells and tissues to better understand disease pathways. Now we are reverse-engineering disease biology and the drivers of drug response directly from data. And those novel insights are being used to accelerate the drug discovery and development process and match patients to the most effective interventions for them as an individual, both in clinical trials and in real-world patient care. Our whole mission is to unravel the cause and effect mechanisms that drive patient response to drugs and care management interventions throughout the healthcare system. I’m very proud of what we’ve achieved and to the knowledge we continue to generate – we have over 45 peer-reviewed journal articles and abstracts—that document the impact we are making.
Slone Partners: The cryptocurrency bitcoin has enjoyed huge media exposure over the past year, and the word blockchain has entered the public consciousness as the underlying technology of distributed databases. Does blockchain have a significant place in the future of personalized medicine and precision medicine? If so, how does that work, what’s that look like, and who benefits?
Mr. Hill: I’m not an expert in bitcoin and blockchain but from what I know it can probably aid in the transferring and secure storage and aggregation of patient data. I think it’s still early days and I am not sure if its potential matches the hype around it. In my view, if blockchain ends up providing patients better access to their own data, then that’s great because patients should have a seat at the table and can aid in self-organizing their data into meaningfully large databases.
Slone Partners: Why is artificial intelligence and causal machine learning so important to the healthcare industry and the patients it serves?
Mr. Hill: Causal machine learning is a powerful form of artificial intelligence that has the ability to not just find patterns in data, which is what a lot of traditional methods such as deep learning do, but to use the data as fuel to reconstruct the underlying mechanisms of the system that created the data in the first place. With these underlying mechanisms unraveled, it then enables “what if?” interventions such as a one drug versus another to be run on the computer to determine the optimal treatment for an individual patient. This is key is solving the matching problem and getting the right treatment to the right patient at the right time versus treating patients as if they were some hypothetical “average patient.” This is critical not just to curing diseases and slowing disease progression, it is also critical to savings hundreds of billions of dollars in interventions not matched to the right patient and the downstream medical cost of prolonged disease.
Allow me to share a few examples of where our work is making an impact. We recently published the results of a joint effort with the Multiple Myeloma Research Foundation, where we discovered a biomarker that identifies which multiple myeloma patients are likely to benefit from stem cell transplantation. Stem cell transplants are painful and expensive so understanding who will respond is game-changing.
We have also been working with the Alliance for Clinical Trials in Oncology and were able to discover the role that tumor location plays as a driver of overall survival in patients with metastatic colorectal cancer. This helps clinicians choose the right treatment earlier in the disease. We have also leveraged Parkinson’s disease patient data from the Michael J. Fox Foundation PPMI dataset to discover genetic and molecular markers of faster motor progression of the disease which has the potential to accelerate the clinical trial process and the development of effective drugs for Parkinson patients.
We are also beginning to generate results on the managed care side with the ability to apply the technology to determine what’s the best care management intervention for a given member at a given point in time in their disease trajectory to slow disease progression and reduce hospitalizations and ER visits.
Causal machine learning is changing our understanding of human disease which is why it’s so important to healthcare outcomes and the patients whose lives are being impacted. I think we are just starting to tap its potential, and I can’t wait to see what else it will reveal.
Slone Partners: You live in Massachusetts, your business is in (Cambridge) Massachusetts, and you’ve served on the Massachusetts Digital Health Council, appointed by Governor Charlie Baker. The organization, and therefore the state, now brands itself as “the world’s first digital health supercluster.” With all the major technology companies like Microsoft, Oracle and Google HQ’d on the west coast, how did Massachusetts become so powerful in the digital health space? Are those companies missing out geographically?
Mr. Hill: While we founded the company in upstate New York after leaving Cornell University, we decided to move our headquarters to Kendall Square in Cambridge in January of 2006. It was clear then that Kendall Square Cambridge was one of the emerging centers for biotech and pharma and medicine, and in the years since it has become THE center of the industry worldwide. Cambridge/Boston is the ultimate convergence of hard sciences, clinical research, genomics, and computing in a way that the west coast and Europe can’t touch. There is more of a culture of hard science here, including in the financing of companies, that creates an ecosystem of multidisciplinary talent, money, and biological and clinical discovery and evidence generation.
Massachusetts, and the Greater Boston area, including Cambridge, is the convergence of clinical medicine with some of the world’s best hospitals—Mass General, Brigham Women’s, Beth Israel, Tufts, Boston Medical, genomic science with the Broad Institute, Koch Center, and all of the university research, and computing and tech and AI. There is a continual influx of students earning their degrees in bioinformatics, physics, bioengineering, biochemistry, machine learning, and computing.
I think the other attribute that Massachusetts has is a pragmatic and rigorous approach to innovation. It is the center of the 4th industrial revolution in AI and machine learning much as it was the center of the industrial revolution with the textile mills in the 19th century. It’s a place where people debate ideas, collaborate, work hard and are dedicated to technological and scientific rigor. I think you’ve seen a lot of the Silicon Valley companies open large offices in and around the Boston area because of the convergence of talent, clinical expertise and innovation.
Slone Partners: As a founding Board Member of non-profit TMed: The Elizabeth Kauffman Institute, how is your work affected knowing that “over half the medicines used today do not work for half the patients,” as stated in their primary mission? Medical revolutions are necessarily methodically slow and rigorous, yet so many patients need life-saving solutions immediately.
Mr. Hill: The current standard of care is based on the average patient. What this statistic proves is that none of us are average patients. If you further break that statistic out into different disease areas you can be looking at non-response rates of nearly 75% in oncology and 50% in arthritis, among others.
This lack of treatment efficacy is especially disheartening when it comes to life threatening diseases like cancer. I don’t know anyone who hasn’t been affected by cancer in one way or another. I lost my cousin to stage IV cervical cancer at the age of 43 – way too young. My father has battled stage III prostate cancer. And I lost my “adopted mother” Elizabeth Kauffman (my mentor Stu Kauffman’s late wife) to pancreatic cancer, which led to the formation of TMed. So, when TMed was formed with the goal of determining the most effective treatment based on an individual patient’s genetic and molecular profile, it made sense from both a personal and business level.
GNS was formed to make precision medicine a reality by identifying the right treatment for the right patient at the right time. This has been our mission since 2000. Patients don’t want to see results at a population level. They don’t want treatment based on a hypothetical average patient. They want personalized care treatment plans that will take into consideration who they are as a unique individual, from their genomic and genetic data to their social determinants of health.
I see this field leading to something truly transformative—the ability to deliver precise care and predict the right intervention at the right time in a patient’s disease trajectory. If we can do that, we can save people like Elizabeth Kauffman and my cousin from untimely deaths.
Slone Partners: Much is being made about wearables, implantables, portable digital health records, and other technologies that empower patients more and more. As hardware, data and software continue to exponentially evolve and converge, what’s your utopian view of the patient “symptom-to-diagnosis-to-treatment-to-cure” experience 25 or even 50 years from now?
Mr. Hill: Health data is only going to grow and become more refined, complete and tangible. The computing power is only going to become faster and more powerful, and AI technology will continue to get smarter. This means that we have the opportunity to discover more insights about disease, unravel human biology and make a significant impact on the health of patients.
Our ability to tap into patient generated data and fold that into our toolbox is just beginning to show real promise. I joined the board of BioTelemetry, (NASDAQ: BEAT), the largest mobile health information company in the world, because I think this kind of data is becoming increasingly crucial to our understanding of disease. The remote monitoring technology and service that BioTelemetry developed and applies is the current gold standard in detecting irregular heart rhythms and has already expanded to type 2 diabetes. The mobile health information has the potential to enable the earlier detection and prediction of disease events and the ability to intervene and keep the individual out of the hospital, improving health outcomes and saving money. We’re even seeing big consumer tech companies such as Apple collaborating with Biotelemetry and Stanford to impact consumer health on a large scale. I think we will see more of these types of breakthrough partnerships.
In terms of the next 25 or 50 years, I believe we are on a path to cure cancer in the near future. We are getting to the point of being able to precisely target the right treatment intervention based on an individual’s biology and disease. We are seeing more results where patients are living longer, higher quality lives. I think our understanding will continue to grow exponentially over the next half century. And Causal AI with “big data” will get us there.
Slone Partners: What makes you happy personally?
Mr. Hill: As an entrepreneur and former scientist, my work drives me (there are people in my life who think I work too much). I truly love what I do. The progress we continue to make as a company on a big mission that really matters, and the enthusiasm of our team is what gets me out of bed each morning. Other than that, I’ve played tennis competitively for most of my life and still enjoy getting out on the court, as well as biking, hanging out on Martha’s Vineyard or Dominica, and going to Big Boys Camp on Green’s Island Maine.