The Food & Drug Administration has teamed up with the Defense Advanced Research Projects Agency (DARPA) to help accelerate the development and approval of innovative devices for continuously monitoring biomarkers in people. Such biomarkers could serve as early warning signs of diseases—such as diabetes, cardiovascular disease, and influenza—before symptoms occur. They could also help optimize the performance of healthy individuals and athletes.
But detecting biomarkers in people is fraught with challenges. Nearly all sensor materials come with biocompatibility problems, and it is difficult to integrate all of the sensor parts into a practical system. In addition, although many biomarkers have been identified, it’s likely that many important ones have not yet been discovered.
To address these problems, FDA and DARPA cosponsored a workshop in Arlington, Va., last month. The gathering brought together agency representatives and scientists to talk about the state of the field.
DARPA, the research and innovation arm of the Department of Defense, wants to identify biomarkers that could enhance the health of military members and their families. The agency became interested in biomarkers after one of the projects it funded led to the identification of biomarkers related to an increased risk of developing a respiratory viral infection.
That finding prompted the agency to ask whether biomarkers can be measured continuously in people. “As an aspirational goal, we would like the possibility to measure biomarkers on person, in real time,” DARPA Program Manager Daniel J. Wattendorf said at the workshop.
Continuous monitoring devices, however, are notoriously difficult to develop. “The only continuous monitors that we are aware of right now are glucose monitors,” Wattendorf noted. Despite decades of development, continuous glucose monitors still have problems with accuracy, and “they are only measuring one molecule in individuals that already have disease,” Wattendorf pointed out.
One of the reasons DARPA cohosted the workshop was to learn more about what other chemicals or biomarkers can potentially be measured in people. The agency invited a few dozen chemists, engineers, clinicians, and regulators to help identify biomarker projects that are ripe for future investment.
Such biomarkers could include an immunological signature or a particular subset of proteins, nucleic acids, or small molecules. A biomarker may not necessarily be a diagnostic for a specific disease, but it could serve as an indication that there’s been a shift in an individual’s baseline health—a warning that something is awry, Wattendorf said.
FDA cohosted the workshop because it wants to help foster innovative technologies, while at the same time ensuring that they are safe and effective. The agency’s interest in facilitating the development of implantable biosensors is part of a larger initiative launched last month to accelerate the approval of transformative medical devices.
That project, called the Medical Device Innovation Initiative, will establish “a priority review program for eligible, new medical devices that demonstrate the potential to revolutionize disease treatment, diagnosis, or health care delivery and that target unmet medical needs,” Jeffrey E. Shuren, director of FDA’s Center for Devices & Radiological Health (CDRH), said during a briefing last month.
By getting involved in projects and providing guidance earlier in the development process, FDA hopes to make it easier for truly transformative technologies to get to market. “We can reduce unnecessary delays and review these devices for approval in roughly half the time it takes for the typical premarket approval application,” Shuren said.
The first medical device that will be evaluated under the innovation pathway is a robotic arm that moves almost naturally and is controlled by brain activity via a chip implanted on the surface of a patient’s brain. The project was funded by DARPA to help wounded soldiers regain fully functioning arms. Clinical trials to examine the safety of the device are expected to start in about six months.
Implantable biosensors are another candidate for the innovation pathway, but FDA is still trying to judge whether patients and consumers would want such products. “What we are talking about is subjecting people to implantation of some long-term sensor that is reading out information about their physiological state,” Megan Moynahan, who leads a team at CDRH responsible for regulation of pacemakers and defibrillators, emphasized at the workshop. “What do the well people want to measure of themselves? What do athletes want to measure?” she asked.
It is difficult to answer those questions because not all biomarkers have been discovered yet. But identifying new biomarkers comes with what some workshop participants refer to as the chicken-and-egg problem. The chemists, engineers, and other physical scientists developing new biosensors said they look to the clinicians to identify the next biomarker to measure. The clinicians said they can’t identify biomarkers until platforms are developed to detect them.
When a biomarker is identified, the biggest problem isn’t building a miniaturized sensor to detect it, according to chemists and other researchers. Instead, the challenge is implanting that sensor in a person and getting it to work for long periods of time.
For example, one of the major problems continues to be the biocompatibility of sensor materials. The list of potential biocompatibility problems to consider when developing a biosensor is overwhelming, one engineer stressed at the workshop. In addition, implanted biosensors affect the environment around the site of implantation, potentially altering levels of the biomarker of interest.
Many workshop participants noted that implanting biosensors in people will be difficult, if not impossible. Although there has been some success developing the individual sensors, researchers are struggling to integrate those sensors into a practical system for continuous monitoring. Ideally, that system would one day be able to generate its own power, monitor biomarkers, and release therapeutics when needed.
Another challenge is the limited number of molecules that can be used in biosensors to bind to the analyte of interest. Participants discussed the need for high-quality, high-activity enzymes besides glucose oxidase, the enzyme used in nearly all glucose sensors.
Many of those other enzymes are by-products of soybean fermentation and are sold by Japanese companies, according to Peter A. Petillo, chief science officer at Pinnacle Technology, a developer and manufacturer of electronic products. “Those enzymes that we typically get in this country are packaged two or three times, and that is affecting the quality and activity of the enzymes,” he said at the workshop.
Because of the shortage of domestically produced enzymes other than glucose oxidase, some workshop participants suggested that designer enzymes be created through synthetic biology approaches. Such enzymes could be tailored to have increased activity and stability, allowing the sensor to last longer.
Biosensors can be designed to detect just about any analyte, but it is unclear what patients want in terms of the ability to measure biomarkers in their bodies. Several workshop attendees made it clear that patients are at the center of the decision-making process. The key to success, they said, is to enhance communication among chemists, engineers, regulators, and clinicians and to find out what patients want.
It is also important to remember that not all biomarkers are easy to measure, Wattendorf told workshop attendees. “Many things that may be interesting are either in low abundance or we are not measuring them continuously,” he stressed.