Junto Profile: Healx
Of late, we have seen Artificial Intelligence (AI) transform the healthcare landscape in ways previously unimaginable. Many believe that AI has the capability to increase both the efficiency and the accuracy of the healthcare industry. Many startups have already explored how AI can be applied to various facets of medical treatment, ranging from pathology, to medical imaging, to genomics, to vision assistance. AI has also been used to help healthcare companies organize and manage their operations.
Under the leadership of Tim Guilliams, Healx is introducing AI to an special group of patients in need of treatment: those with rare diseases. Through their novel AI platform Healnet, Healx uses AI to match patients with the most appropriate treatment. Rare diseases are often difficult to manage, as their associated symptoms and other intricacies can be difficult to track. Furthermore, many rare diseases currently do not have appropriate medications for treatment, so funding and direction for drug discovery and repurposing medications is still needed . Aligning the priorities of Pharma and rare disease patients can be challenging, as rare disease treatment medications may not generate the same revenue as ‘blockbuster’ medications. With rare disease patients in mind, Healx looks to revolutionize the way medical professionals diagnose and treat rare diseases so everyone, no matter what ills them, receives the care they deserve. We spoke with CEO Tim Guilliams to learn more about what Healx has accomplished, and what they are looking to do next.
Junto Health (JH) What was your inspiration for founding Healx? What was your background prior to founding Healx? Why did you choose to focus on rare diseases?
Tim Guilliams (TG): I have always been passionate about using artificial intelligence (AI) and machine learning to accelerate treatments for rare diseases. In addition to my role at Healx, I’m also the Founding Director and Trustee of the Cambridge Rare Disease Network (CRDN), a charity that fosters dialogue around and awareness of rare diseases.
I co-founded Healx in 2014 alongside Dr. David Brown, the co-inventor of Viagra and former Global Head of Drug Discovery at Roche. Following a meeting with a rare disease parent, Nick Sireau, who at the time was relentlessly trying to repurpose a weed killer to save his children from an ultra-rare disease, we realised how big the therapeutic unmet need was.
We’ve since made it our mission to help rare disease patients access life-saving treatments. We achieve this by partnering with patient groups and harnessing the latest in AI and machine learning technology to accelerate the drug discovery process.
There are presently over 7,000 rare diseases affecting 350 million people worldwide, and 95 per cent of these still lack an approved treatment. The limitations of the traditional drug discovery model mean big pharma often prioritises profit-making ‘blockbuster’ drugs for common diseases over finding breakthrough treatments for rare diseases.
Rare disease patients have long been side-lined by the pharmaceutical industry who struggle to justify investment in treatments for smaller populations. At Healx, however, we believe that every disease patient deserves a treatment. This belief helps us to think big. We don’t intend to go after just one disease, we have created a platform that will help us impact 100 rare diseases by 2025.
JH: Can you briefly explain the functions of Healx?
TG: We believe that every disease patient deserves a treatment. It was this mindset that drove us to create a platform and process that would allow us to tackle not just one disease area, but multiple rare diseases in parallel.
The first step towards making rare disease drug discovery a faster and more efficient process was to develop Healnet, the most comprehensive AI platform currently out there for the prediction of rare disease treatments.
Healnet uses machine learning to mine data from scientific literature, patents, clinical trials, disease symptoms, drug targets and underlying chemical structures. From this unique knowledge base of rare disease data, Healnet then maps more than a billion data points connecting diseases, patients and drugs. This approach allows for new connections to be made between related diseases (in terms of biology, the symptoms and drugs that may be effective on them) which then enables Healnet to predict potential treatments. As a by-product of this process, we often uncover new biological mechanisms as well.
The complex algorithms driving Healnet’s predictions are extremely clever, however, they still need drug discovery experts to ensure treatment predictions are made in the most responsible way possible. Our team of world-class pharmacologists review all treatment predictions before selecting the most relevant ones for further testing.
Whilst technology is integral to our work, we’re committed to keeping patients at the very heart of what we do. This patient-centric focus sees us collaborating with a number of patient groups to understand the most urgent challenges patients and their carers face. This is an approach that we feel not only marks us out from our competitors but which, ultimately, helps us do a better job of improving the quality of life of those living with rare diseases.
JH: Who will benefit from your technology?
TG: With one in 17 people suffering from a rare disease, the potential impact of our product is huge. It’s our aim to impact 100 diseases by 2025. By those calculations, our technology has the capacity to change the lives of millions of people.
From the very beginning we felt that forming partnerships with patient groups was the right direction to take. As well as being experts in understanding rare disease symptomology, they are also usually the most determined partner in driving treatments forward. Involving them from the outset closes the discovery research loop earlier. We also recognise that without their support we could not have achieved as much as we have done so far.
Healx is currently working across ten different disease areas, mostly within the fields of rare neurological disorders and rare cancers. The objective is to expand the portfolio 10-fold over the next five years. Our most advanced programmes are now moving into Phase 2a clinical trials. None of the above would have been possible without the help of exceptional patient groups, such as FRAXA, the LouLou Foundation, the Pitt Hopkins Research Foundation (PHRF) and aPODD, amongst others.
JH: Can you talk a little bit about the team at Healx? How do your life science professionals combine forces with data scientists and technologists? Is it challenging for these groups to work together?
TG: Our team is one of our strongest assets and we’re very fortunate to be based in Cambridge, UK, which is a real hub for talented individuals. As an AI company in the pharmaceutical industry, we know that the collaboration between our world-class pharmacologists, data mining experts and bioinformaticians is the key to our success.
In a company like ours, it can be quite easy to gravitate towards silos where each specialist is working vertically, but we know that this can become a major issue further down the line. To counteract this, we’ve worked hard to establish effective communication strategies at Healx. We work in dynamic cross-functional teams, which allows us to come up with the solutions to our challenges with the full awareness that we are creating something new and disruptive.
For me, the most important aspect of collaboration is fostering a culture of openness. I strongly believe that everyone in a company should have a voice. The culture at Healx makes it so that anyone can suggest new ideas and improvements. I find this way of working helps keep the whole team engaged in actively seeking the best solutions to the challenges we face.
As a company we have also managed to keep a lean startup culture. This allows us to move extremely quickly, while remaining highly innovative. Keeping a fast pace is critical for any growing organization.
JH: The idea of drug repurposing is something that we haven't seen before as part of a company's mission. What is this process like? How do you identify which drugs have a potential to be repurposed?
TG: Repurposing was only the beginning for Healx, we’re now doing a lot more beyond this strategy. We recently developed technology to predict treatment combinations and are now also launching our precision medicine modules.
Since many diseases exhibit diverse symptoms that involve complex biological mechanisms, the best way of tackling these is to combine drugs that can synergistically impact different pathways and targets. The benefit of these optimized combinations is that they can improve therapeutic outcomes, minimize side effects and reduce drug dosage for patients. However, testing the universe of potential combinations can involve billions of experiments. This is where our AI comes in to its own. With the help of Healnet, we are able to predict which combinations have the best chance of succeeding in the lab and in subsequent trials.
This strategy helps identify safe therapies more quickly and at a larger scale than traditional drug discovery which, in turn, reduces the risk, costs and time required for progressing drugs for rare diseases to the clinic and to patients.
To put this in perspective: the current model takes an average of 12 years and costs over $1bn to bring a treatment to patients. Healx’s model (which allows for research to run in parallel rather than sequentially) can bring new medicines to patients in 3-5 years and at a fraction of the cost.
JH: What feedback have you received on Healx? Who are your current project partners?
TG: Healx’s technology has already been used successfully in a number of projects. Most recently we collaborated with FRAXA, the patient group for fragile X syndrome, to discover and progress a novel treatment for this rare disease. The drug discovered by Healx’s platform was ready for the clinic within 18 months. This cut typical drug development timelines by 80 per cent along with the associated costs.
We received the following feedback from FRAXA’s CSO Dr Mike Tranfaglia;
“After 20 years of research and spending $27 million, our work with Healx represents a major paradigm shift in the way we search for treatments. Now, in 18 months and at relatively little expense, we have validated possible treatments which are now moving to clinical trials.
We have found the expertise of Healx to be an ideal complement to more traditional forms of rare disease research, and we are continuing to expand our collaboration which offers enormous value in helping us accomplish our mission.”
To receive this sort of feedback from an experienced researcher, who has invested both personally and professionally in hundreds of projects over the years, is the best type of validation for our team and the work that we’re doing. It makes us proud!
JH: In general, how has AI transformed the medical industry so far? Do you believe AI will keep improving medical processes in the future? Do you think there is a limit to its potential?
TG: We’re seeing AI used very frequently to automate the structuring and consolidation of facts within medical data. However, machine learning and AI technology have the potential to revolutionise the healthcare industry.
Over the past century (particularly the last two decades) the healthcare industry has been accumulating vast amounts of data. The potential benefits this accumulation of data might bring is only now being realised thanks to the advancements in AI and machine learning that allow us to find patterns, link correlations, and test hypotheses at scale.
We believe that once AI is fully integrated into our healthcare system (in about 10 years or so) preventative medicine that provides protective treatment before the onset of certain life-threatening conditions will take centre stage. This will further help cut costs and transform healthcare for patients.
JH: Are there any ambitious goals that you are hoping to achieve in 2019?
TG: Having proven the capabilities of our Healnet AI platform through its use in a number of successful projects, our mission now is to start deploying it at scale. This year we want to use Healnet to double our portfolio, with the intention of doubling it again by 2020. Stay tuned… 2019 looks set to be a very interesting year for us!
JH: Where do you see Healx as a company five years from now?
TG: By 2025, we aim to have identified potential treatments for 100 rare diseases. This is our key company focus. To help us on our way we will integrate more data layers and develop our capabilities in precision medicine to offer tailored treatments to patients. After all, each patient biology and disease signature are a rare combination that offers us an opportunity to develop our technology to predict individual courses of treatment. Five years from now our technology will be mature enough to offer precision medicine treatments tailored to patients’ individual disease profiles.