Medical trials are risky business. Human subjects eager to help researchers find cures or treatments for life-threatening ailments put their trust in medical professionals. But what if these trials could be conducted without human testers?
Biotech company ELEM is doing just that. It built virtual humans—”not made of flesh and bones, but bits and bytes”—powered by supercomputers, which model disease states and, by extension, are “treated” by novel breakthroughs. More recently, its Virtual Heart project has been looking at how certain COVID-19 treatments affect cardiac health.
Is this the future of medicine? And if so, when do we get our own bioidentical virtual human twin who can test personalized cures before we have to pop a pill or take a shot? We spoke with Barcelona-based Christopher Morton, ELEM’s CEO, to find out more.
As we’re still under the occupation of COVID-19, it must be dangerous to conduct large scale human-based medical trials. So I imagine ELEM’s computational avatars, which test medical breakthroughs on disease states, are more necessary than ever? [CM] Yes, that’s right. We are constantly in contact with doctors, businesses, and regulators: recruiting candidates and getting them into the clinics to take part in clinical trials was always an issue and, yes, COVID has made it worse. There is only so much risk people are willing to take, I guess. Another associated problem with clinical trials is the selection of candidates and the retention.
Because you have to be brave to submit yourself to a beta, especially when it comes to your health?[CM] Yes, that and also selection is tough because of exclusion criteria. If people have too many comorbidities, it is harder to understand the true impact of a new treatment. And records show that the dropout rate can be as high as 30%.
But virtual humans are ready and willing to step in.[CM] Right. With our Virtual Human trials, we bypass a lot of these issues. Populations can be created ad-hoc and on-demand. There’s also no gender-bias issue. Today, in cardiac trials, drugs, or devices, many products are dominantly tested in male subjects, usually white and slightly overweight, if possible. But evidence shows that more and more women are suffering from heart conditions. I mention cardiac because it is our first application. Do you know that for children, doctors often have to be creative because cardiac devices are made for adults?
And you can’t put an adult-sized human heart stent into a kid’s body.[CM] No, it’s not acceptable in the 21st century. Coming back to our solutions, for us speed is just a matter of adding computer resources. You may recall a topic in the news at the beginning of the pandemic. Could chloroquine, a drug used to combat malaria, be used to treat COVID patients? A huge aspect of drug development and/or repurposing is safety. There were suspicions that chloroquine and hydroxychloroquine could lead to complications for individuals suffering from heart conditions. Well, one of the consequences of COVID is an acceleration of the heart rate due to respiratory difficulties.
ELEM had the virtual human hearts ready to test chloroquine and other possible interactions?[CM] Yes, within a week, we had created a virtual test and confirmed the risk of cardiac arrest in COVID patients. That is very hard to beat in a real trial scenario. Since then we have expanded our study and tested close to 200 virtual patients in eight scenarios, combining various treatment combinations and patient states.
Are you also able to let the virtual humans take the place of real-life animals in testing? Bringing us into a more humane future?[CM] Definitely. During COVID, one of our clients had planned a series of invivo tests on sheep. They are working on a new type of pacemaker for patients suffering from weak hearts. Despite everything being ready and the animals having reached the size and age required for the tests, lockdown meant they were unable to perform those tests. We offered them an unexpected alternative.
Give us the backstory on ELEM, and how you met your co-founders Mariano Vázquez, Guillaume Houzeaux, and José María Cela.[CM] I studied aeronautical engineering at Imperial College London and was given the opportunity to go to Barcelona on Erasmus. After my graduation, and a compulsory military service, I went back to the research center, where I had studied to work in numerical optimization and in computational fluid dynamics. Mariano and Guillaume arrived soon after to do their PhDs. We have been friends ever since.
So ELEM emerged from the Barcelona Supercomputing Center (BSC)? [CM] In 2005, Mariano and Guillaume joined Jose Maria Cela at the Barcelona Supercomputing Center literally weeks after it was established. Their mission was to build a simulation software that could solve bigger and more complex scientific and engineering problems, much closer to reality. Pushing the accuracy and the boundaries of what could be achieved using normal modelling and simulation tools by leveraging the power of supercomputers. And that is how Alya was born—powerful simulation software capable of multiscale and multiphysics modelling, and used today to model human physiology, amongst other things. Alya is used by a vast community of researchers. It is a benchmark for computational performance.
Had you all been interested in using this computational power for medicine, or did that come later?[CM] Mariano, who is a physicist by the way, and so is Guillaure, was always extremely interested in medicine. He always reminds us how had he not studied physics, he would have become a doctor. When he got the chance, he engaged in research projects to model arteries, veins, and the respiratory system, too. It was exciting because with the power of supercomputers, real human anatomies could be used. Mid-2013 we got together to discuss the possibility of setting up a company. They showed me the first beating heart.
That must have been incredible to see a virtual heart beating on screen.[CM] What a moment. To see something for the first time. Slowing down the heartbeat so you can see with your naked eye how it contracts. What a moment. But despite this excitement, we were still in the aftermath of the financial crisis, [so we waited until] 2017, when the time was right. The technology had matured enough, biomedical companies took an interest in the capability, and when they expressed the desire to use it, that’s when it all happened. How could we put this amazing technology in the hands of those who really need it? In the hands of those who develop new treatments and those who save patients lives? The answer: ELEM.
Today, ELEM is arguably at the forefront of what’s known as exascale computing, moving far beyond today’s high performance computing, as you illustrate in your paperpaper on Alya.[CM] One of the basic premises on the design and development of Alya is portability. Two facts push us in this direction. On one hand, as mentioned earlier, BSC is at a crossroads of hardware, middleware, modelling software, and applications. On the other hand, its MareNostrum 4 is in fact a hybrid interconnected supercomputer, with all kinds of processors, such as Intel, AMD, ARM, IBM’s PowerPC, or Nvidia. Alya ported to all of them. This privileged position allows us to have early access to leading-edge supercomputer architectures with the assurance that our code will run there with the highest efficiency, because a deep understanding on how the hardware works allows us to develop the best strategy to port our modelling tools to current and future computers. This is a very rare case in modelling and simulation.
I spoke with Dr. David Pittmanspoke with Dr. David Pittman, Director of the ERDC and Chief Scientist for the US Army Corps of Engineers, about virtual construction of civil hardware in computational environments for the Department of Defense. Do you face incredulity from engineers when you point out you’re using similar techniques to map inside the human body?[CM] Excitement more than incredulity. Engineers are very good at trying to solve problems with the tools they possess and know. The models used in civil engineering, aerospace, automotive, and in the development and manufacturing of most consumer products, for that matter, are still full of approximations. Engineers deal with this every day. Why should the body be any different?
The body is, arguably, an engineered system in constant flux after all.[CM] Yes. An artery is just another pipe, a joint another structure, a muscle another material. And up to a point it is true. Remember, also, that the tools used in standard engineering practice were developed over decades. Safety margins compensated for what we didn’t know or could not control. Model-Based Systems Engineering (MBSE) is still quite new. But like everything, the devil is in the details. Biomedical modeling requires a system-based approach most of the time. Because everything is intertwined and working together. So, human body versus a bridge or an airplane? Clearly there are subtle and not subtle differences. One is alive and the other one is not. But can we combine engineering and medical knowledge? Yes, absolutely! Are engineers best placed to do that? Not sure. Rather than crossover, integration is what this is about. And actually, at ELEM, we have medical experts, bioengineers, physicists, engineers, mathematicians, computer scientists. All working together.
Still privately owned, ELEM is VC-backed. How much have you raised so far from investors such as Genesis Biomed and Oracle Startup Cloud Accelerator?[CM] ELEM has secured close to 2 million euros since 2018 in a mix of dilutive and non-dilutive funding. In addition, 1.6 million is being invested in our core technology as part of our partnership with the BSC. Public funding is paramount in Europe to support research and development activities in the early stages of technology creation, and we would not be here without it. However, what you get is a diamond (sometimes) in the rough, which is not market ready. “Au contraire,” as Samuel Beckett said when asked if he was English.
Yeah, the Irish don’t like being mistaken for the English. But we digress. How are you using this investment[CM] We are in the process of raising 3 million euros to finance our operations for the next two years. The funds will help us validate Virtual Heart clinically. We are engaged in a clinical trial with cardiologists from the Sant Pau Hospital of Barcelona. We have also initiated proceedings to achieve regulatory approval, and, last but not least, introducing Virtual Heart to market, which requires ready-ing the product, and protecting it, too.
What’s the endgame?[CM] Making medical testing unbiased, much more efficient, and working for all. It’s a good objective and how well we achieve it will determine our future.
To get there, ELEM has participated in a couple of accelerator programs, including Oracle’s (OSCA) and SETsquared Bristol. How did they help with your goal?[CM] SETsquared has some fantastic entrepreneurs in residence (EIRs) who have first hand experience in deeptech and or biotech. SETsquared being very good at leveraging the local network means we get a lot. In addition, the link to four top British Universities guarantees a constant flow of talent and ideas.
OSCA was an incredible opportunity for ELEM. First of all, because of the access to Oracle Cloud Infrastructure (OCI). Not only did Oracle award $100,00 worth of cloud credit to its winning startups, they also made a strong commitment toward high performance computing in the cloud. ELEM needed to find a cloud partner that would allow us to port and test Virtual Heart in a commercial-like setting, and OSCA made this possible. Beyond OCI, we have and continue to work a lot with Oracle’s marketing as well as commercial teams in Spain, in the UK, and in the US, giving us significant exposure in digital and classical media, through webinars, blogs and direct interviews, too.
Looking to the future, when I covered Aromyxcovered Aromyx in Silicon Valley, I got to see chips that digitize taste and smell. Will ELEM go in this direction next, to commoditize its Virtual Heart on chips?[CM] That’s a great question. There are multiple paths ahead of us. Of course, the idea of tiny chips capable of handling millions of operations enabling real-time medical diagnosis is exciting. Looking ahead, some startups are already working on organ-on-a-chip technology or micro robots that will navigate our body. We are all after the same goal, which is to improve human health. Indeed, thanks to our modelling tools, we are in the perfect position to work together with them to design much better biotech solutions.
What’s the next stage for computational mechanics in the clinical space, after the Virtual Heart?[CM] The human body is full of exciting challenges. We are driven by two forces: society needs and our own vision. Today, although we actively work on products for cardiovascular applications, our R&D team is developing solutions for other systems and organs, notably the respiratory one. But this is not the only domain we are targeting, because some other wonderful examples will very soon be revealed.
Finally, biosimulation is a critical step in improving global health. How many years out do you think we are from being able to map our own DNA blueprint, ingest medical records, and test out drug delivery and/or clinical treatments on our completely unique digital twin? Or is that a pipe dream?[CM] Ha! It depends how accurate you want it and what risks you are prepared to take. Seriously, computer modelling and simulation has transformed everything we touch, every industry sector. At the beginning, the models were simple. They became gradually better, more accurate. Planes land with one engine these days, or with a hole in the fuselage. F1 drivers escape unscathed from crashes at 300 miles per hour. Wind turbine blades are 60 meters long. We just landed on Mars again. It did not happen in a day, but we would not be where we are today had someone not made the leap. Virtual Humans capture knowledge and deliver insights beyond imagination, even today. When will we be able to do it all? It doesn’t really matter. I am not sure I really want it either. There is magic in mystery.
