QuesTek Innovations has used advanced computer modeling to produce innovative materials in the aerospace sector. Now it’s looking to recreate the same magic in the medical device space.
Chris Newmarker, Managing Editor
Medical device developers typically turn to off-the-shelf materials and then design based on the properties of the materials. But does it have to be this way? What if it was possible to simply design and then choose and validate materials that meet the design’s criteria when it comes to strength, fatigue life and corrosion resistance?
That’s the tantalizing concept proposed by officials at QuesTek Innovations (Evanston, Ill.). The idea is nothing new, either. QuesTek has years of success in using computer modeling to create innovative alloys for the aerospace sector. The 20-year-old company has managed the creation of Ferrium M54 steel for U.S. Navy aircraft hook shanks, Ferrium C64 steel for the transmission gear boxes in next-generation Bell and Sikorsky helicopters, Ferrium S53 steel for critical components on SpaceX’s Falcon rocket, and Ferrium C61 steel for a robotic rover destined for Venus.
Company officials say they doubled annual revenue over the past three years after a technology transfer deal with an undisclosed Silicon Valley company, which saw the potential for QuesTek’s materials modeling. (ChicagoInno says Apple bought the technology. QuesTek declined to comment.)
“What QuesTek has enabled is that people can come to us and say, ‘We want a material with this set of properties, and this material does not exist. Can you design it?’ It’s whatever is important to them,” explained Jeff Grabowski, manager of applications and product commercialization at QuesTek.
Grabowski and others at QuesTek think the same could be done in the medical device space: “This is about getting the medical device community to stop using off-the-shelf materials and start to think what kind of materials properties they need to achieve a game-changing device.”
What computer modeling can achieve
There are two types of computer modeling QuesTek has worked to perfect over the years: Integrated Computational Materials Engineering (ICME) and Accelerated Insertion of Materials (AIM). ICME provides a more efficient way to create an alloy with the properties needed for a design, while AIM greatly reduces the amount of testing needed to qualify the alloy’s properties.
ICME has been around since the 1980s; QuesTek co-founder and chief science officer Greg Olson played an important role creating the computational models that made it useful, according to the company.
Before ICME, creating an alloy was all about trial and error, explained senior materials design engineer Nick Hatcher. A steel company, for example, would melt 50 to 100 chemistries and heat treat them and test them all.
ICME draws on physics, density functional theory, thermodynamics and phase diagrams of materials, Hatcher said. It starts with databases of thermodynamic and kinetic properties of the elements, and crunches the information with predictive software and models to quickly go through thousands of iterations of chemistries and thousands of subsequent virtual heat treatments to optimally target a set of performance requirements.
Once Hatcher and his colleagues get the results, they then might come up with 2 or 3 or 5 chemistries – versus 50. They take each chemistry and do a quarter-sized melt in the laboratory, heat-treat it and examine the microstructures. What phases have formed? What is the hardness of the material? Can the material achieve what it was designed to do? In most cases, they said, QuesTek is then able to fine tune the design before going to an outside vendor for a much larger sample.
“ICME allows me to design the materials more quickly, at a lower cost, and we arrive at better properties than traditional trial-and-error methods,” Hatcher said.
Meanwhile, the company is also a leading contributor to the AIM program, spearheaded by the U.S. Defense Advanced Research Projects Agency and the Office of Naval Research. AIM is a probabilistic approach to forecasting property variability and material properties – including 1% minimum properties. A 1% minimum for, say, yield strength means that there is a 99% probability that yield strength will be above the 1% minimum, so determining the 1% minimum is crucial when it comes to qualifying a material’s performance, noted Dana Frankel, a materials design engineer with the firm.
Grabowski said AIM allows QuesTek to take a material it designed, with its chemistry and heat treatment, and then make perhaps three full-scale test melts of the material, which are then tested. “And then, based on the data from your three heats, you can use AIM to predict, ‘If I make 10 heats of this material or 100 heats of this material, these are the minimum properties you should expect,’” Grabowski said.
“There’s this added benefit of AIM that it allows you to do fewer melts of a given material and then have a high degree of confidence you will have a certain minimum set of properties,” he added.
ICME and AIM have allowed QuesTek to design new aerospace alloys the company says have displaced other steels used for decades. And company officials think they already have alloys in their wheelhouse that could be attractive to medical device companies. They include the company’s QuesTalloy SMA, a nanodispersion-strengthened, high-performance shape memory alloy that the company is marketing as an improved material for use in stents and catheter lead wires.
Grabowski also expected QuesTek to have titanium alloys for medical additive manufacturing commercially available in the next 12 to 18 months; the company says its titanium alloys have 20% higher strength at equivalent ductility than the Ti-6-4 presently used in medical 3D printing. QuesTek has also designed superior cobalt chrome alloys that could have medical device uses.
Changing old habits
Company officials, however, acknowledged that it has been slow going when it comes to marketing their alloys to medical device OEMs. The predicament is not a surprise: The medtech industry has a reputation for being highly conservative when it comes to novel materials, because they can greatly add to the time it takes to achieve regulatory approval.
Aerospace and medtech are alike in that failure of a material or a component could result in loss of life. But there’s still a difference, said Olson, the company’s co-founder and chief science officer.
“The difference is we designed airplanes, but we didn’t design humans,” he explained. “Working with aerospace engineers, they really know what they need, and if we meet their requirements, it’s going to fly. In medical, because we don’t know what humans are … there’s going to be a lot of testing in an environment that we’re still trying to better understand.”
Olson is optimistic because the company has heard from FDA officials interested in incorporating more computer modeling when it comes to determining the safety of materials intended for medical device use, especially given the rise of 3D printing in the medical device industry.
The FDA, in fact, has already shown more openness to using computer modeling in medical device studies, even issuing a guidance document in September 2016 with suggestions on how to report on such studies. FDA’s Center for Devices & Radiological Health has accepted computational models for a decade, but the new guidance formalizes the reporting recommendations. The guidance document includes a suggested outline for reporting that includes such items as clear identification of the quantities being analyzed, scope and type of analysis, software quality assurance and much more.
“Computational modeling and other approaches that could spur innovation or improve and reduce premarket burden are assessed on a case by case basis,” said FDA spokeswoman Deborah Kotz, when asked about the agency’s interest in using AIM methods and other computer modeling techniques from aerospace.
FDA is also part of the federal government’s multi-agency Materials Genome Initiative, which has the goal of deploying advanced materials twice as fast, at a fraction of the cost.
“We have high hopes that FDA in the near future will take a page out of the AIM playbook and will get to these statistically informed accelerated qualification methods that will allow for these high-performance designer materials to make it into the marketplace,” said QuesTek’s Frankel.
As Siemens PLM life sciences director Kristian Debus said in a recent blog post, interest is growing when it comes to the use of modeling and simulation in life sciences clinical and trial studies. Debus thinks FDA has been especially supportive of using modeling tools in cardiovascular device design.
When it comes to using computational tools to prove the safety of medical devices, it’s important to be careful of using off-the-shelf computational tools and to pay attention to validation – how close the model comes to reality, said medical device regulatory consultant Michael Drues, president of Vascular Sciences (Grafton, Mass.).
Said Drues: “Like any tool, in the hands of someone who knows what they are doing, it’s very useful. In the hands of someone who doesn’t, it can be disastrous.”
[Want to stay more on top of MDO content? Subscribe to our weekly e-newsletter.]