A Pathway to Approval for Additive-Made Devices
Devices made with additive manufacturing techniques often replace a similar, or predicate, device made in a different manner
By Ilene Wolff
Additive manufacturing is now producing all manner of medical devices, and new ideas for the process—ranging from printed surgical tools and bone replacements to human tissue—are coming from designers and engineers daily.
Even the best idea, though, has little value in the United States unless the Food and Drug Administration gives its go-ahead for putting the device on the market.
Because many new devices made with additive manufacturing techniques are, essentially, replacing similar devices simply made with different manufacturing techniques, there is often an easier “predicate” device pathway to federal regulatory approval.
This article examines three medical devices manufactured with 3D technology, one of which obtained government approval in 2013: The OsteoFab Patient-Specific Cranial Device. Scott DeFelice, president and CEO of Oxford Performance Materials (South Windsor, CT), which manufactures the device, shares his experience working with the FDA.
OsteoFab is Oxford’s brand name for polyethylene ketone ketone, or PEKK.
We also look at two other, as-yet unapproved devices. One is less intrusive than the OsteoFab skull patch, and would be worn as an exoskeleton on the outside of the body to immobilize and support a broken limb. The other is more intrusive, and would create a prosthesis for a missing nose or ear from a bioresorbable scaffold and a person’s own stem cells. Two experts weigh in on what the FDA may require before approving these devices.
Patching the Skull
Oxford Performance Medical made headlines in early 2013 when its patient-specific cranial device, or skull prosthesis, became the first FDA-cleared, 3D-manufactured polymer implant. The company produces the one-off devices that are custom fit for individual patients using PEKK-based OsteoFab, its own product, and an EOSINT P 800 laser sintering 3D printer in a computer-aided design process.
DeFelice said Oxford has made the skull prostheses for more than 200 people in the year since gaining government clearance to market the device. The devices are shipped within five days of receiving an order to locations in Europe, the Middle East and South America, in addition to the United States. Oxford was due to break into the Japanese market in January.
“We do them every week now,” he said of the devices.
Oxford obtained clearance to market the device within the FDA’s self-imposed time frame of 90 days, he said, because the government agency already had extensive data on the biocompatibility and purity of the proprietary raw material, OsteoFab; all processes were developed internally; and the company controls the entire supply chain.
“In our cranial device, where things got complicated is that we do patient-specific devices,” said DeFelice.
He explained that the FDA already had addressed the patient-specific issue with metal skull prostheses, so the agency knew what to look for and that moved the process along.
“What it basically means at the end of the day is the FDA will evaluate your ability to design and produce your product,” he said.
Oxford had already done about 30 skull prostheses outside of the United States before submitting its request for clearance to the FDA. DeFelice declined to go into detail regarding clinical trials.
His company recently submitted for approval a mid-face implant (for the bone around the eye and cheek) A device for spine fusion is in the works.
“So, it’s a very broad platform [for OsteoFab],” he said.
While DeFelice has in-house advisers to help with the regulatory process, not everyone is so fortunate. For them, there are independent consultants, some of whom previously worked for the FDA.
One such consultant is Dennis Moore, of The FDA Group (Westborough, MA), who has 30 years’ experience with the government agency, both as an employee and an adviser.
“One thing that people should keep in mind is that it’s not a very well-defined process,” he said. “Nothing is absolute. I’ve seen exceptions to every rule.”
That said, Moore offers his three-step process for FDA 510(k) clearance (for Class II devices) or pre-market approval (for Class III devices).
His step No. 1 is to determine what already-approved item—a predicate device—most resembles your bright idea.
“And that’s not always an easy task,” Moore said.
The goal is to closely define what your device is by its intended use.
“A toothpick becomes a device when you say it removes plaque,” Moore explained. “A cotton swab may be a cosmetic product until it’s used to clean a wound bed.
“The whole game is to ramp down your claim so you can get clearance.”
Attorney James Lawrence, of the Coats & Bennett law firm in the Research Triangle Park area of North Carolina, said of the predicate device process: “Essentially, say, 'Here’s my device, and here’s a predicate device.
“My device has the same intended use and it shares the same technical characteristics as this previous device.’ ”
Lawrence, who has an undergraduate biomedical engineering degree in addition to his law degree, interned at the FDA. He offers an insider tip about the predicate device process.
“FDA reviewers are wary of ‘predicate creep,’ where submissions veer further and further away from something that’s been approved,” Lawrence said.
Moore recommends making a blind call to the FDA’s Division of Small Manufacturers Assistance line (1-800-638-2041 or 301-796-7100) and saying, “We’ve got this widget, what do you think?” Or, for a few thousand dollars, you can send in your device and the FDA will tell you where its staff thinks it fits. The process takes 30–60 days, Moore said.
It can pay off to match your invention to a predicate device carefully, he warns.
“If approached correctly and you can tone down the claim enough you can sometimes avoid a clinical trial,” said Moore. Clinical trials typically require a minimum of 50 patients, and each patient’s costs can tally $10,000 or more, he said.
Some makers get a toehold in foreign markets, where regulatory hurdles are lower, before trying to get their device approved in the United States. This offers the advantage of having patient usage data that can be included in an FDA submission.
“In some cases you can leverage that to a degree,” said Moore.
Lawrence is more optimistic.
“You can rely on data from trials that were conducted outside the US,” said Lawrence. “You just have to make sure it complies with the laws in the country of origin.”
After figuring out what your predicate device is, Moore said the next hurdle is to strategize the steps you need to take, and itemize your costs. A white board and sticky notes—one for each step—come in handy.
“You don’t want to get hit with an unexpected electrical safety test that can cost $100,000,” he warned.
Moore’s Step No. 3 is to file a submission with the FDA and answer any objections brought up by its reviewers. The clock on the government’s 90-day timeframe stops ticking while the applicant works on his answers.
The Emergo Group Inc., global consultants for medical devices, got scientific about how long it takes the FDA to complete the 510(k) process and has done periodic analyses.
“On average, it takes five months for the FDA to review and clear a medical device 510(k) application,” wrote Scott Schorre, vice president of global marketing for Emergo in his 2014 analysis of 24,000 applications cleared from 2006 through 2013. “About two-thirds of all 510(k) submissions are cleared within six months.”
Moore offers a warning.
“The FDA loves to get you wrapped in pre-submission meetings, and there’s no timer on those,” he said. “Pre-submission meetings are not required. If you feel you have enough data to file a submission, then go ahead.”
Lawrence offers tips on what goes into that submission.
“The agency wants to see that in the history of designing and trying to get your device on the market that you’ve gone through a rigorous design process,” he said. “You have to prove you have internal controls so designs are vetted and peer-reviewed, tested and validated.”
Keep track of all changes to the device’s design, he said.
“At the end of the day your design history file will show the evaluation of the designs from the beginning to the end,” said Lawrence. “It’s not something foreign to the way engineering is typically done.”
It’s always better to do your homework before submitting, even going so far as to talk with the people who will be reviewing your device.
You can even go one step further hire an FDA-authorized third-party reviewer who can expedite the process for a price, said Emergo’s Schorre, but the FDA still has final say-so.
“Do the math to see if it makes sense for your situation,” he said.
What to do if the FDA rejects your submission?
“You can challenge those determinations by the FDA and companies have tried to do that,” said Lawrence. “You can appeal up the agency’s chain of command, and as a last resort argue [in court] that the FDA made an arbitrary and capricious decision.
“[But] courts are typically very deferential to the scientific expertise of federal agencies.”
Helping Bones Mend
Like DeFelice, Jake Evill also has an idea for bone problems. But his device would support, rather than replace, bone that’s been damaged. Its Cortex, an exoskeleton made of recyclable plastic that would replace plaster or fiberglass casts. Plaster is heavy, and both it and fiberglass are not water-resistant or recyclable. Casts made of plaster and fiberglass and are itchy and smelly as well.
Cortex looks like a custom-fitted plastic web that fits over an arm or leg: the device is more solid and less “webby” at the site of the break to provide extra support. It opens like a book, and snaps closed for one-time patient use. Once the break is healed, Cortex can be tossed in the recycling bin.
“The Cortex exoskeletal cast provides a highly technical and trauma zone localized support system that is fully ventilated, super light, shower friendly, hygienic, recyclable and stylish,” Evill wrote on his website.
To make Cortex, a patient would be X-rayed and 3D scanned. Then computer-aided design would calculate the device’s pattern for support in the area of the bone break, and an algorithm would be used to 3D print the device.
Evill was an industrial design student in New Zealand when he created Cortex. The James Dyson Foundation, established by the creator of the Dyson vacuum cleaner, thought so highly of Evill’s idea that it named him runner-up in its 2013 design engineering competition.
Attorney Lawrence said of Evill’s device: “That being outside the body it’s more akin to an arm sling or an orthopedic brace. I have a hard time believing it wouldn’t be similar to a predicate device.”
DeFelice and Evill aren’t the only ones making news with their devices.
Two years ago, Dr. Scott Hollister, professor of biomedical and mechanical engineering and associate professor of surgery at the University of Michigan, used 3D printing to make a life-saving device for a baby.
Kaiba Gionfriddo, a 3-year-old from Ohio, passed the two-year survival mark in February after receiving an emergency experimental splint to keep his trachea open. Kaiba has severe tracheobronchomalacia, a rare condition that causes the airway to routinely collapse.
To make the splint, doctors at U-M CT scanned the affected area of Kaiba’s respiratory tract, and used the images to design and print the tube-shaped splint to help keep the area open. When a pediatric otolaryngologist implanted it, Kaiba’s lungs immediately started moving on their own in the operating room.
Unlike the skull prosthesis or broken limb exoskeleton, though, Kaiba’s splint is expected to dissolve in his body as his airway naturally matures. That’s because it’s made of polycaprolactone, or PCL.
Hollister explains that it took him six mnths to adapt PCL for use in his EOS Formiga P 100 laser sintering machine. Since his success, he’s made Kaiba’s splint as well as a scaffold to regenerate bone in the mandibule (jaw) for an adult patient in 2013.
But there’s a twist in the prosthesis for the adult patient—a biologic growth factor was adsorbed onto the PCL. The idea is that, as the man’s own replacement bone grows, the PCL prosthesis will dissolve, just like Kaiba’s splint.
So far in the United States, Hollister is using a PCL scaffold seeded with stem cells and growth factor to grow replacement ears and noses in pre-clinical research. It’s possible to fashion a prosthesis from a patient’s own bone and cartilage, but often the soft tissue around the graft breaks down. His idea would eliminate another issue as well.
“You don’t need the surgeon to be Michangelo in the operating room,” said Hollister.
There are other issues still to be resolved, but Hollister said the patient-specific ears and noses would be classified two ways by the FDA: as a device and as a biologic.
Attorney Lawrence predicts that as scientists and clinicians make progress in 3D biomedical engineering, the FDA will increasingly encounter similar two-in-one devices.
“This is a classic example,” he said. “As far as regulation is concerned, it’s probably [something] the Office of Combined Products would look at.
“The question asked would be what’s driving the overall therapeutic effect? It’s something the agency deals with fairly frequently.” ME
This article was first published in the April 2014 edition of Manufacturing Engineering magazine. Click here for PDF.
Published Date : 4/1/2014