Digital design tools — computer modeling and realistic simulation — have emerged as key contributors to success in medical device product design and manufacturing, and increasingly in the pharma industry as well. Many of these tools have been used in the aerospace, automotive, construction, and other industries for decades. While life sciences applications are in the earlier stages of development, they are beginning to produce often astounding results.
Not only are the available tools becoming more sophisticated yet easier to use, FDA is now taking an active interest in encouraging new methods of modeling and simulation to accelerate the pace of innovation in life sciences. This is likely to have far-reaching consequences for product design, development, and regulatory approval.
There are a number of acute challenges facing life sciences and medical device developers and manufacturers today. Mahesh Kailasam, PhD, a vice president at Thornton Tomasetti in charge of that group’s Applied Science life-sciences initiative, sees these as threefold.
The first is the challenge of cost, he says. “How do you develop new products in a cost-effective manner, especially as the expense of medical treatment is rising everywhere?”
The second challenge is to make sure that whatever solutions are developed are applicable for the targeted patient population so that treatments are effective. Historically, solutions have been developed generically, but applied to individuals according to experience and observation. “Now we’re getting to the point where devices and treatments can be personalized to an individual’s characteristics, so that the treatment works as intended both in the near term and in the longer term — but we still need to keep that first challenge of cost in mind,” he says. The third challenge Kailasam sees is the need to develop solutions in a timely manner, of course faster than ever before.
“When you consider these all together, it is clear that the heavy reliance on traditional processes, including bench-top experiments, animal testing, and typical clinical studies, is just not suited for the challenges the industry is facing right now,” he says. “Digital technologies will be key to accelerate the transformation that is needed.”
Steven M. Levine, PhD, senior director of life sciences at Dassault Systèmes SIMULIA, thinks that the single biggest way to reduce the skyrocketing costs in the healthcare industry is to lower the volume of post-treatment care. This has two components: 1) getting the treatment right the first time, and 2) shortening recovery time with less-invasive treatments. “The former often is characterized as ‘precision medicine,’ but it basically means that we need better ways to analyze a given condition and select the best treatment,” Levine says.
“The use of digital technologies will be transformational in this, from using real-world data and in silico models to develop better physiological models of patients, to conducting virtual treatments to optimize outcomes,” he says. The latter involves providing physicians with a more targeted approach in situations where they have less ability to see what is happening inside the body. Once again, virtual reality and realistic digital representations of the patient and procedure are critical to make this happen. “Of course, incorporating real-world behavior as part of the diagnostic or follow-up can dramatically improve success while achieving cost-saving goals,” he notes.
FDA Steps up to the Challenge
FDA has taken a lead for several years now, encouraging the adoption of new modeling and simulation technologies for effectively evaluating different solutions and accelerating the pace of innovation — even to the point of accepting simulation data as part of applications for device approval. Of course, in order to make sure that such adoption is done in a consistent manner, FDA is also working with others to refine emerging guidelines, such as for verification and validation, to ensure that this is done in a systematic and controlled manner rather than an arbitrary way. In addition, programs such as MDDT (Medical Device Development Tools) are helping the industry develop virtual human and animal models to evaluate medical devices with greater confidence than before.
“I understand the challenge FDA has in maintaining their need to oversee the introduction of safe medical devices, while at the same time accepting responsibility to help lower the speed and cost barriers without compromising safety,” Says Levine. “To meet this challenge, FDA has invested, through internal R&D as well as extensive collaborations, in understanding new methods that could achieve both goals. They have evaluated the various sources of evidence, animal models, clinical trials, and computational models and concluded that, in many instances, as much as 50 percent of the time the computational models could be a better source.”
FDA is actively working internally and through collaborations with organizations such as the Medical Device Innovation Forum (MDIC) and projects such as the Living Heart Project (LHP) — Levine is the lead and Kailasam spearheaded the commercialization of the technology — toward a future where half or more of the data submitted for regulatory approval comes from computer modeling, virtual patients, or virtual clinical trials. Moreover, FDA is publicly sharing this mission, publishing guidelines such as the V&V40, and encouraging what they have called a “Simulation Revolution” in medical devices.
“Digital tools are what are allowing us to virtually try out multiple solutions to any challenge and do it all efficiently,” says Kailasam. This is certainly relevant at the earlier stages of design evaluation via virtualized benchtop tests, where companies ranging from the largest medical device makers to hundreds of startups can use simulations to effectively zoom in on designs that have the most promise. “But these tools are also very relevant, and perhaps more valuable, in later stages where digital models of organs like the heart or skin can be used to assess effectiveness of devices in virtual populations and tailor the solutions to different population segments,” he points out. “Information from such virtual studies will allow clinical studies to be more effectively designed, giving device developers greater confidence in their outcomes — and leading toward virtual clinical trials in the near future.”
From Skyscrapers to Scapulae
As noted earlier, digital tools reached maturity in a number of other industries before becoming adopted by life sciences. Kailasam’s company, Thornton Tomasetti, has a long history of using digital tools for modeling and analyzing large and complex systems, including the effects of underwater shock on submarines and ships, ground-borne seismic waves and shock on structures, and environmental loading on structures like supertall skyscrapers, stadiums, and arenas.
Notes Thomas Scarangello, the company’s chairman and CEO, “These tools have created tremendous efficiencies and have been proven vital in speeding the pace of innovation and the delivery of new solutions for our clients in these industries, and they will play the same role in helping our life sciences clients accelerate their success as well.” Scarangello says that many of the methods his engineers have been using for decades are directly applicable to life sciences, particularly when you relate the underlying physics — whether they are structural, thermal, CFD, and so on — to the problems at hand. “Digital tools are perfect for capturing the interactions between different aspects of any complex system, whether it is the behavior of submarines underwater or stents experiencing blood flow inside vascular systems,” he points out.
Only five years ago, few examples existed of simulation successfully moving down the path to commercial success in life sciences. One early showcase example was cardiovascular stents — which are now at the point where simulated fatigue prediction has become necessary to ensure not only the safe lifetime of a new design but also the regulatory approval of FDA. Today, nearly every device can be MRI-safety certified virtually, and FDA’s aforementioned MDDT program is working to precertify computational methods that can speed regulatory approval.
“I believe FDA has sent a clear message to the device community when they joined the LHP to lend their support to introduce this technology into the regulatory process,” says Levine. “More importantly, the LHP, now entering its fourth year, has demonstrated that physics-based simulation, once the exclusive domain of mechanical devices, is equally applicable to biological systems such as the human heart.”
Using a consensus methodology among the now 100+ members of the LHP, models and methods to virtually design and test new devices have been developed. These methods offer the potential to test and refine new device designs in a fraction of the time and cost of current methods that are based on a combination of bench and animal testing, and over time to be more predictive of clinical performance. Using the LHP as a basis, Dassault Systèmes is now working on models for other medically important body systems such as the brain, knee, spine, etc.
In addition to quality assurance, Kailasam sees precision personalization to be of particular benefit to patients whose cases can benefit from the latest advances in digitization of life sciences offerings. “Digital modeling and analysis have been used to develop high-quality implants for some time, but now many solutions (knee or hip replacements among them) are being designed to match an individual’s lifestyle choices, such as athleticism, as well as his or her physical characteristics, on a far more granular level,” he says.
Additive manufacturing (3D printing) in combination with such digital tools enables production of high-quality, patient-customized implants. “These solutions are not only being offered by large medical device companies but also by a whole cadre of smaller startups, including several that are able to recommend strategies for surgery and treatment based on simulated estimates of post-treatment outcomes,” says Kailasam.
Similarly, in other domains such as vasculature and blood flow, digital tools are being used to develop models from imaging data, which can then be used for a variety of applications — ranging from 3D printing of realistic blood vessel network models to simulating the behavior of devices inside these models, and to even assessing various disease conditions that may hinder blood flow or risk the integrity of the vasculature.