3
The Nature of Medical Innovation

Presentations in this session of the Workshop provided background and context for the status of innovation in medical devices since the late 1980s, and addressed the invention and development process map for medical device technologies and products. Several case studies were offered to analyze the factors that have led to significant medical device innovations in the past 50 years. Speakers discussed the factors that have supported significant ongoing and emerging technology innovations to reach the development and clinical stage.1

THE INNOVATIVE PROCESS FOR MEDICAL DEVICES: A NASA PERSPECTIVE

John Hines, M.S.

Technology Development Manager

Space Life Sciences Program

and

Joan Vernikos, Ph.D.

Director, Life Sciences Division

National Aeronautics and Space Administration

Since its inception in 1958, NASA has collaborated with many entities on technology R&D. These collaborations have included the development of medical devices in support of astronaut health and biomedical research, both on the

1  

One presenter, Dean Kamen, President of DEKA Research and Development Corporation, played a videotape from a news show that aired on network television.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary 3 The Nature of Medical Innovation Presentations in this session of the Workshop provided background and context for the status of innovation in medical devices since the late 1980s, and addressed the invention and development process map for medical device technologies and products. Several case studies were offered to analyze the factors that have led to significant medical device innovations in the past 50 years. Speakers discussed the factors that have supported significant ongoing and emerging technology innovations to reach the development and clinical stage.1 THE INNOVATIVE PROCESS FOR MEDICAL DEVICES: A NASA PERSPECTIVE John Hines, M.S. Technology Development Manager Space Life Sciences Program and Joan Vernikos, Ph.D. Director, Life Sciences Division National Aeronautics and Space Administration Since its inception in 1958, NASA has collaborated with many entities on technology R&D. These collaborations have included the development of medical devices in support of astronaut health and biomedical research, both on the 1   One presenter, Dean Kamen, President of DEKA Research and Development Corporation, played a videotape from a news show that aired on network television.

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary ground and in space. This collaborative R&D process has been based on the need to utilize a broad range of expertise and experience to meet special requirements, minimize development costs, and exercise the NASA mandate to “provide for the widest practicable…dissemination of information concerning its activities and the results thereof.” From the earliest years of the United States Space Program, NASA has in many cases taken off-the-shelf commercial biomedical instruments and modified them for use in space. This process mainly involved the use of NASA customized components and packaging designed to survive the rigors of space flight, and special considerations for safety and materials composition. NASA’s special requirements overlapped to a significant extent with military requirements for biomedical devices, including several core needs: portability, operation within intravehicular and extravehicular environments, telecommunication of data, minimally invasive sensors and non-encumbering instrumentation, and low-power, 28VDC and/or battery-powered systems. NASA has always had especially challenging requirements for medical devices, including operation in variable pressure environments (space capsules and space suits), high radiation environments, and high vibration and shock environments (launch/reentry). Paramount in these considerations for medical devices has been the safety and well being of the astronaut crew and biological subjects. To this end, devices have been designed with the highest medical device standards in mind, and rigorous testing has been performed to validate their performance. In addition, NASA has high reliability requirements for biomedical devices, since on-orbit repair and/or replacement often is not possible. Some of these requirements overlap in part with those of the emergency medical monitoring and transport industry. More recently, NASA’s Life Sciences Division has established internal Advanced Technology Development (ATD) programs to anticipate needs for medical (and biological) devices and similar technology. Because it can take years to go from initial requirements to having flight-qualified hardware, one aim of the ATD-Biosensors Program has been to collaborate with the future users of biomedical technology and develop and demonstrate modular, prototype systems in anticipation of need. When requirements are more solidly defined, often by multiple users who need similar technology, these prototype systems can be more quickly assembled, tested, and made available for use. NASA also has established multi-disciplinary teams to plan for integration of advanced technologies into the International Space Station (ISS). The ISS provides a special challenge for development of medical devices, as it is a large, international research laboratory built in space to remain for 10 to 15 years. Over that time period, medical technologies will rapidly evolve, and older technology will need to be infused with new modular systems that take advantage of industry-developed technologies to optimize functional performance at minimum cost. These include sensors and instrumentation, analytical tools, and specialty devices. The emerging medical device industry/academic focus on wireless, tele-

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary metric, wearable, automated, “intelligent assistant,” reconfigurable technology fits well with NASA’s needs for the future. The complementary emerging technologies of nanotechnology and biotechnology can greatly facilitate in-flight clinical analyses, and are also of great interest to NASA, due to their small size, low-power, low-cost disposable, and customizable features. Medical Device Case Studies NASA’s transfer of its R&D results to the private sector has been implemented through various methods and mechanisms that provide industry and academia access to spin-off technologies and knowledge, several of which have been applied to commercial biomedical devices. This collaboration is enabled by the provisions of the Space Act Agreement, and is implemented with the assistance of the NASA Commercial Technology Offices (CTOs) by way of a variety of arrangements, including cooperative agreements, reimbursable and non-reimbursable space act agreements, memoranda of agreements, and inter-agency agreements. Additionally, NASA often develops requirements for new biomedical devices that are contracted out to industry or academia for R&D, with encouragement to industry to consider commercial development of the technology, when appropriate. On occasion, NASA biomedical technology developers have left government service to privately develop commercial versions of the technology they helped invent within the Agency. In these cases there has been both a knowledge and technology NASA spin-off. Modification of commercial technology is an option whose practicality is assessed prior to initiating internal R&D. The dual-use and co-development of technology by NASA and other government or private-sector partners is a relatively new but expanding method for technology development, with biomedical devices being attractive candidates. For most medical device development funding is provided by NASA’s Life Sciences Division. However, several medical device technologies developed by industry for the private sector have included technology components (and software) developed within NASA for non-life-sciences research, such as a smart probe for breast cancer diagnosis and treatment and a robotic neurosurgical device. Some examples of biomedical devices that illustrate these NASA Invention and Innovation process categories are described in the following sections. Cardiac Monitor A traditional method of assessing heart function is thermo-dilution, which involves the insertion of a catheter into a pulmonary artery. NASA needed a non-invasive system to monitor astronauts in flight. In 1965, Johnson Space Center contracted with the University of Minnesota to explore the concept of Impedance Cardiography (ICG). This led to the development of the Minnesota Impedance Cardiograph (MIC), an electronic system for measuring impedance changes across the thorax that would be reflective of cardiac function and blood

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary flow from the left ventricle into the aorta. NASA separately contracted with Space Labs, Inc., for construction of space-qualified miniaturized impedance units based on the MIC technology. The system was introduced into service aboard Space Shuttle flight STS-8 in 1983. The ICG had potential for hospital applications, but further development was needed. A number of research institutions and medical equipment companies launched development of their own ICGs, using the MIC technology as a departure point. Heart Imaging System Doctors on the ground need to be able to evaluate the vital signs of astronauts in orbit. NASA has therefore researched and developed sophisticated heart monitoring systems for this purpose. Dr. Jeffrey L.Lacy was a biomedical researcher at Johnson Space Center who developed technology that was later adapted for commercial medical use. His research into heart imaging led to the development of the MultiWire Gamma Camera (MWGC), marketed by Xenos Medical Systems. This imaging system is six times faster than other devices, portable, and provides extremely high-resolution images. One of the key components of this system is its use of the radioisotope Tantalum-178 (Ta-178), which can be optimally imaged only with the MWGC. Use of Ta-178 has a major benefit: its short half-life means that it is only in the body for 9 minutes, while other radioisotopes must remain in a patient’s system for 6 to 72 hours. Thus, Ta-178 provides a 20 to 200% decrease in radiation exposure during the imaging process. Since astronauts are exposed to chronic, low-level radiation during space flight, this reduced exposure is an especially important advantage for NASA. Heart Assist Pump The concept of a heart pump containing NASA technology began with talks between Dr. Michael DeBakey from the Baylor College of Medicine and NASA engineer David Saucier, who happened to one of Dr. DeBakey’s heart transplant patients. Saucier felt compelled to help develop a device to assist the 30,000 people a year who are unable to obtain a donor heart, and he understood the Space Shuttle technology that could be applied to create an effective heart pump. Mechanical heart pumps have three potential problems: destruction of red blood cells, formation of blood clots, and the body’s reaction to a more continuous blood flow rather than the normal pulsed flow of blood. A team from NASA Johnson Space Center and NASA Ames Research Center assisted Dr. DeBakey and his Baylor partners in the development of an effective heart pump by using super computers to analyze how shuttle fuel-flow dynamics could be used to reduce red cell damage to acceptable limits. This improved flow pattern also reduces the tendency for blood clots to form. The pump design was eventually licensed to MicroMed, which successfully ran clinical trials of the device in Europe. Efforts are underway to facilitate use of this device in the United States.

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary Telemedicine Instrumentation Pack During space missions, NASA’s astronauts are out of physical reach for doctors and surgeons on Earth. If an astronaut-physician is available onboard a spacecraft, ground-based remote monitoring medical support is essential to augment that available in-flight. Technology had to be developed to give flight surgeons at Mission Control the capability to conference with, diagnose, and treat astronauts in flight. NASA’s intensive research into these systems led to the creation of new telemedicine instrumentation systems. The latest result of NASA’s research into this field is the Telemedicine Instrumentation Pack (TIP), manufactured under contract to NASA by KRUG Life Sciences of Houston, Texas. Developed at Johnson Space Center, the TIP weighs about 25 pounds and is the size of a small suitcase. Designed to record and display video, audio, and biomedical data (such as ECG waveforms, heart rate, and blood pressure) the TIP allows a doctor to make accurate remote diagnoses. Advanced features still in development will include electronic medical information and literature, decision support systems, and computerized patient records. Pill Telemetry Technologies NASA contracted with Konigsberg Instruments in the 1970s to develop pilltype, implantable, multi-channel biotelemetry systems for animal research studies in space, and later an ingestible pill for human studies. In parallel, NASA, by way of a Goddard telemetry program, contracted with the Advanced Physics Lab (APL) at Johns Hopkins to develop an ingestible, temperature-sensing pill for health care applications. There was a direct spin-off of these technologies for battlefield biomedical monitoring by DoD during Operation Desert Storm, resulting in use of both the APL and Konigsberg ingestible pills. Also in the 1990s, the Sensors 2000! Advanced Technology Development program at NASA Ames Research Center developed a pill implant for pH and pressure monitoring during fetal surgery, which can also have applications to small animal research (mice) on the International Space Station. Benefits of the NASA Process NASA R&D in medical devices during the 1960s and 1970s typically included “contracting out” much of the development work to industry and/or academia. Aerospace companies familiar with developing aircraft and military instrumentation were often the developers, and they in turn subcontracted work to industry specialists, as warranted. Hybridization of commercial devices was done whenever feasible, with special packaging and testing for space flight often done by NASA. Intensive biomedical monitoring and research was done on the Skylab, the first space station, in 1973, but only relatively simple measurements could be made. Problems with attachment of surface sensors occurred throughout the mission, and needs for wireless biotelemetry data were apparent.

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary The innovation and invention processes of the early years continued in the 1980s and 1990s with increasing realization that the high costs and long development times for medical devices were becoming prohibitive. NASA increasingly wanted more involvement in defining the requirements and development process to optimize the reuse and upgrade capability of medical devices for the rapidly evolving Space Shuttle Program. Hybridized medical devices were increasingly used for human medical support and biomedical research. The first Spacelab, which allowed significant biomedical monitoring, flew in 1983 with several life sciences experiments. An industry-built miniaturized mass spectrometer for making gas metabolism measurements was developed and flown by NASA. Today, as NASA continues to work on the International Space Station and plan for longer duration space travel, it is essential that the latest advances in technology are developed and applied to the new research objectives. Next-generation medical devices that will utilize emerging technologies—such as nanoscale devices, MEMS (micro-electromechanical systems), biosensors, gene arrays, biomimetics, robotics, advanced optics, and wireless communications— are under active study or are in development. Increasingly, collaboration by NASA with government, industry, and academic partners will be essential for co-development of dual-use medical devices. This process is driven by the need to take advantage of new clinical and biomedical measurements and to meet NASA’s objectives in its Human Exploration and Development of Space program. NASA’s programs for ATD and Technology Infusion will be essential for providing appropriate new medical devices on a schedule that will allow utilization in the era of the International Space Station. ENDOVASCULAR DEVICES Thomas J.Fogarty, M.D. Professor of Surgery Stanford University School of Medicine Technology development as applied to surgical therapeutics, rather than just device technology alone, involves procedures, devices, instruments, techniques, drugs, and services. Transplantation provides a good example, as it certainly would not be possible without the availability of anti-rejection drugs. One area of device development that relates to less invasive approaches is endovascular technology, which can be divided into diagnostics and therapeutics. In 1929, the concept of injecting dye into the vascular system was introduced by Dos Santos, who used the technique to outline tumors to understand their blood supply. Later, Forsman came up with a concept of coaxial catheter. It was introduced in the vascular system essentially to monitor some physiological parameters, and he actually used it on himself. Forsman was followed by Soames, who was interested in the pathology of a particular patient group with heart disease. By accident, he injected dye in the supervalvular position, and when the catheter was in the coronary he ended up doing a selective coronary

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary catheterization. His peers thought it was an anecdotal occurrence without real impact on the field of medicine. The next era involved catheter-mediated therapeutics to address acute arterial occlusions. The concept was developed by a scrub technician who worked on enough cases to know that when somebody had an acute arterial occlusion there are usually three operations. One was the attempt to get the occlusion out, which ultimately failed. The second was the amputation, usually below the knee, and then third was an amputation above the knee. This observation was based on no real critical science. The use and improvement of catheters was not perceived by the medical community or the device industry as significant events. The developers of these early devices were not well acknowledged by traditional medicine at the time, and their papers were not accepted by peer-reviewed journals. What was viewed as insignificant has led to treatment of aneurysms with stent grafts, as well as treatment of trauma and congenital malformations such as A/V fistulas. I do not think anyone really conceived where this was going to go. As technology is applied to surgical therapeutics there rarely is a major paradigm shift; rather, change occurs slowly. Thus, medical technology as applied to surgical therapeutics involves many iterations in which the ultimate utility is inherently unpredictable. Technology is the application knowingly or unknowingly of documented science for practical purposes, for clinical utility. Certainly technology relies on science, and in some ways science does not move forward without certain technological innovations. Science is premised on theory, whereas technology is concerned with applications and utility. Technology may have implications for the company, the patient, or the physician who is employing it, but it is a very individual and personalized thing and science is rarely that. How does one regulate innovation? A subject that currently presents significant confusion between regulatory agencies and drug and device manufacturers is the approach to safety and efficacy evaluation of these two very dissimilar products. There is a need to establish appropriate parameters for evaluating these entities. Recent interest in using an endovascular approach to manage carotid pathology has spurred great debate within the vascular surgical and radiological interventional community. In relation to moving medical device innovation forward researchers must look at the processes currently being employed to evaluate devices as these methods can significantly impact the rate, quantity, and quality of device development. A subject that currently presents significant confusion between regulatory agencies and drug and device manufacturers is the approach to safety and efficacy evaluations of these two very dissimilar products. There is a need to establish appropriate parameters for evaluating these entities. Several years ago the FDA initiated sweeping changes to the regulatory process that effectively buried device evaluations under the identical regulations imposed for approval of drug entities. This attempt to apply a drug testing method to device testing is a seriously erroneous and inappropriate approach. In the development and assessment of drugs versus devices there are numerous

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary distinct differences in these two therapeutic modalities, which highlight the inappropriateness of subjecting them to the same study methods. The influence of technique, significance of in vitro study, rate of technical change and the ability to visualize “real time” performance are all rated low for drugs while these parameters are considered high for devices. Conversely, the resources of the developer, cost development, and the duration of the regulatory cycle are high for drugs and generally low for devices. Device changes are iterative and rapid, unlike drug regimens, which can be titrated then set into concrete dosage forms. Devices, instrumentation, and specialized treatment systems must continually be refined and changed along the process route, which invariable does not lend itself to prospective randomized testing models. In most cases, device evaluations utilizing prospective randomized clinical trials prove not to be the most advantageous method for determining efficacy because prospective randomization, in order to be valid, makes the following assumptions; Case selection. Technical competence and judgment is equal among all. Diagnostic and monitoring modalities are frozen in time and are technician-insensitive, and interpretive skills are equal among all. There is no prior fund of knowledge or reference points that are valid. Post-procedure care is equivalent under all circumstances and the same in all institutions. Intuition is of no importance in determining outcomes. Improvements in technique, instruments, and implants have been optimized and are stabilized. Assumes patient will cooperate in randomized treatment assignment scheme. Enabling technology has stabilized. To add to this list researchers must consider the medical ethical considerations inherent in randomizing patients into a treatment group. Patients who may best benefit from the new treatment modality may be eliminated from the trial protocol if they are not considered suitable candidates for the surgical alternative due to co-morbidities or psychological considerations. Because valid historical controls exist, the technology is not stabilized, case selection is in a state of flux, and significant learning curves can influence results, attempts to use only randomized controlled studies to assess devices become inappropriate. Prospective randomization does not always answer the question when outcomes are unknown and clinical judgment is lacking. Clinical studies should be time- and cost-efficient, and credible. Prospective randomization represents only one method that can meet these criteria, and very often it is not the best method for obtaining an answer.

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary GENE ARRAYS Stephen P.A.Fodor, Ph.D. Chairman and Chief Executive Officer Affymetrix There are roughly three billion base pairs in the human genome. A huge multinational and multicorporate effort to sequence the genome will produce a first rough draft in 2001. But the sequence alone is not enough. Researchers need to know not just the base-pair sequence, but also the function of those sequences and how variation affects function. Which genes are turned on and off under different circumstances? What are the polymorphisms? What are the differences from individual to individual? The technology developed at Affymetrix is the so-called “DNA chip.” The chip uses some of the tricks of the semiconductor industry, photolithography, and combinatorial chemical synthesis and some detection technologies. A lithographic mask shines light in certain areas of the chip, activating the surface, which is flooded with reagents to build DNA molecules. These chips are not made one at a time, but in wafers, like semiconductors. A chip about the size of a dime can hold up to 400,000 different pieces of DNA at precise locations on the surface. The basic idea of the chip is that one adds a patient sample with a fluorescent tag. As it incubates with a single strand of DNA it will find its complementary structure on the chip and bond. The fluorescent tag allows the DNA to be read using a confocal scanning system made by Hewlett-Packard. The chips are disposable, and there is software that goes with the system. Much of the early development of the technology was funded by small companies and federal research agencies. The technology allows for numerous comparisons. For example, a sequence that matches up with normal wild-type DNA shows one sort of pattern of fluorescence, and if a mutation is present the pattern changes, which can be detected very easily. This allows for the detection of disease-causing or predisposing mutations, such as P53, BRCA-1, and BRCA-2. There is a chip to look at some of the genes in HIV to monitor drug resistance, and others to find the presence of a strain of virus or bacteria, based on a piece of its DNA. Nearly 400,000 probes can be placed on a chip, which allows information to be processed simultaneously, an important tool for the human genome project. This is especially useful for the study of polymorphisms, or genetic differences between people. It turns out that there is about one in every 1,000 base pairs that varies across individuals. These single nucleotide polymorphisms, or SNPs, usually have no clinical implications, but serve as a means to identify similarities and differences among people and populations. Chips and powerful statistical genetics provide the opportunity to pursue SNPs as markers for disease as well as for the application of pharmacogenetics. In 1995, Affymetrix took the entire human mitochondrial genome, 16,500 base pairs, and put it on a chip. The subsequent large-scale polymorphism scan-

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary ning project with Eric Lander at the Whitehead Institute found 3,000 SNPs throughout the genome, which were mapped to different chromosomes. This capability will lead to better prediction and diagnosis of disease, but also the ability to trace population migrations and family history. Subtle variations might also explain human preferences, for example, preference or aversion to specific smells. It also allows one to look at banks of genes, for example, related to hypertension or cancer, and understand which genes turn off or on in disease. One can develop drugs that target that action. INHALED INSULIN Robert B.Chess Chief Executive Officer Inhale Therapeutic Systems, Inc. Many of the drugs being developed through biotechnology—insulin, erythropoietin, growth hormone, and interferon—are essentially protein products, and the common problem with these drugs is that the only way to administer them is by injection. If one takes them as pills they are broken down by the gastrointestinal system. They are too large to enter through the stomach or through the nose or skin. One alternative is to administer them through the lung, and if one gets them down to the deep lung, to the alveoli, most of them flow into the bloodstream. This is the technological premise of advanced inhalables technology from Inhale Therapeutic Systems, Inc. (San Carlos, Calif). Inhale Therapeutic Systems, Inc. is a drug delivery development company with a platform of technologies. The researchers there do not develop the drug products themselves but partner with biotechnology and large pharmaceutical companies. These partners typically lead the clinical development and market the product; Inhale Therapeutic Systems, Inc. provides the technology to allow them to do this and then manufacture it once the product is developed. Inhale has 400 employees, including mechanical and chemical engineers, aerosol scientists, physicists, and protein chemists. In 10 years, Inhale has initiated 12 different financings. Inhale’s lead inhalable product is developing a better system for insulin delivery. Inhalable insulin is less invasive than injections and might increase compliance and adherence to treatment. A large NIH-sponsored trial followed 1,400 people over 9 years: some took insulin twice a day and others took it three to six times a day. The study found that those who took insulin three to six times a day decreased the side effects of diabetes by 35 to 60%. Side effects include blindness, the need for lower body amputations, and coronary failure. One out of seven health care dollars—$100 billion a year—is spent on diabetes care. Despite that study, only 15 percent of patients have adopted the increased dosage, most likely because the injections are painful. Inhalable insulin offers an easy alternative. The reason that inhalable insulin has not been developed to date is the difficulty of getting drugs to reach the deep lung so that they can be absorbed effi-

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary ciently into the bloodstream. Inhale has pioneered the pulmonary delivery technology to enable the delivery of these delicate macromolecules through the lung. The first step was to make the particles stable at 1 to 5 microns in size. The challenge was to formulate the powders so that they are as chemically and physically stable as the day they were made and can deliver the same product six months later. Researchers used glass stabilization to pack the particles so they are not moving around and so that the protein is not coming in contact with the water of the particles. The final product is very stable in a wide range of temperatures, which is a great convenience for those with diabetes as they will not have to refrigerate the drug. To process the powdered drugs, researchers had to modify conventional spray drying used in food processing to make the small particles. It is the first time that any company has used spray drying on this scale to make particles this fine and in the process researchers had to keep the consistency of the particles the same for each level of scaling up to the next output size. Another challenge was filling the particles in individual dose units. Researchers needed different dose strengths, because individuals must titrate their dose. To help drive the drug into the deep lung where it needs to be delivered to enable systemic delivery, researchers developed a unique delivery inhaler. Because drug particles need to be delivered consistently from person to person, researchers could not rely on patients’ inspiratory flow rate so they developed an inhaler that operates independently of a patient’s inhaling flow rate. Clinical trials have indicated that reproducibility is as good as, if not better than, the injection approach. Phase II and Phase III clinical trials have been encouraging and patients have expressed satisfaction with this new delivery method that is breaking new ground. IMAGING THE MICROVASCULATURE Richard Nadeau, Ph.D. Chairman and Chief Executive Officer Cytometrics, Inc. Cytometrics was founded in 1992. The mission of Cytometrics is to commercialize noninvasive, point-of-care products, using the Company’s patented OPS Imaging technology for direct observation and measurement of the microcirculatory system and surrounding tissue. For example, by inserting a probe under the tongue, individual blood cells can be visualized. This allows one to do a complete blood count, hemoglobin, and hematocrit, noninvasively. In newborns the device can be used directly on the skin. When it is applied to the conjunctiva of an adult diabetic, the distinct irregularities in the microvascular structure are clear. Also, the device can be used during surgery to observe the microvasculature of various organs, such as the liver, heart, and lung. The microvascular structure is very characteristic and different for different organs. Irregularities are powerful predictors of disease.

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary The Cytoscan Video Microscope is an OPS imaging platform: it has an optical probe, a computer, a light source, and software to operate the instrument. The device takes two high-quality polarizers and crosses them. A small percentage of the polarized light undergoes multiple scattering events as it penetrates into the material or the subject, and during that process the light becomes depolarized. The only reflected light observed is the depolarized light, which effectively is created inside the material, and which has back illuminated the blood vessels. In effect, this optical design creates a virtual light source inside the subject, even though the light is in front of the subject. The Cytoscan is designed primarily for visualization of the microcirculation. It is Class I FDA exempt, and will be marketed in Europe and the United States. Model II, available in 2001, will have more extensive image analysis capabilities. A desirable computation capability being developed is to measure a functional capillary index, or the amount of blood flowing within a tissue along with its hemoglobin concentration. Clinical observations of the microvascular structure have been made by inserting the optical probe into the rectum for the differential diagnosis of Crohn’s disease and ulcerative colitis. Also, the device has been tested comparing normal subjects sublingually with subjects in cardiogenic or septic shock. In cardiogenic shock, the large vessels are almost empty, and more importantly, in the small vessels there is virtually no flow. Capillary flow is either sluggish or completely stopped; this also occurs following cardiac bypass and in some cases of stroke. In septic shock, there is hypervelocity flow in the larger vessels, and again slow flow or no flow in the small vessels. Further, the Cytoscan has been used in the context of neurosurgery, in individuals being operated on for cerebral hemorrhages. In 4 out of the 12, or 14 cases to date, microvascular spasms were observed; these patients later died, suggesting that the spasms require early aggressive treatment. The company founders put $1.6 million of their own money into the company to get the device through the very early seed phase. That allowed them to do proof of concept in vitro. They then attempted to raise money through contracts, venture capital, and technology incubators, all of which proved to be ineffective. The company was forced to raise its own funds, privately. In the seed phase, raising funds was the greatest obstacle for Cytometrics. The federal and state grants and contract systems were too slow, too bureaucratic, and too much trouble for too little money. Moreover, incubators want to own the intellectual property. A solution would be to improve the government role in seed funding. In order to receive federal funding, universities should be required to establish technology incubators and provide an environment in which technologies can be nurtured, particularly in the early critical phases. Added tax incentives for investment would help raise capital for early-stage, high-risk companies. The development stage poses a different set of problems that are more regulatory and legal. These obstacles are higher in the United States; thus, many

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary companies do their initial testing in Europe. In effect, most U.S. companies are exporting their technology. Some remedy could come if U.S. legislation were enacted limiting liability for clinical trials, and if there were more industry involvement in the design and implementation of government regulations. GENERAL DISCUSSION OF THE NATURE OF MEDICAL INNOVATION Discussion of this session’s talks began with a question about whether the Cytoscan imaging system had been used in the capillary bed in the nose, where one could gain access to two of the four major arteries that serve the brain. Richard Nadeau answered that the current probe is a little shorter and thicker than he would like, but it is only the first generation. Cytometrics would very much like a longer, thinner probe that could be used in the nose or be curved around the back of the heart. If they had $100 million they would definitely develop a second-generation probe like that, but the priority at the moment is to get the basic product launched. Clifford Goodman, chairing the session, then asked Robert Chess to expand on one of the lessons learned that he had mentioned, namely, whether he actually had to stop the innovative process so that he would not upset the regulatory review. Mr. Chess responded that that was correct. He had to continue improvements in parallel, while freezing the initial insulin product to keep the regulatory approval process rolling. For example, he said, they learned in late 1997 that if they made their particles a slightly different way they could improve the efficiency of the system by a factor of two, which would be very important both for the cost of goods for the product and the cost of the product for the patients. It probably would not have made any difference clinically, but they would have needed to repeat probably a year’s worth of trials, so they decided not to do it. They are now hoping to convince their partner, Pfizer, to start some clinical trials with the improved version in about a year or a year and a half, and then introduce a more efficient version a year or two later. The net effect is to slow the stream of innovation and leave patient care lagging significantly behind the capabilities the technology could provide. Mr. Chess also pointed out that the decision to pursue improvements on a parallel track was one a less well-financed company would not have been able to make, delaying innovation still further. Thomas Fogarty from Stanford pointed out that some of the recent changes in FDA have allowed innovations based upon experience in Phase I and Phase II without going back. For example, he said, when some mechanical parameters were obviously marginal very early in Phase I, FDA permitted the necessary changes without mandate that the study be restarted. That was something he would not have expected 5 to 7 years ago. Dean Kamen provided a second example of how regulation can slow the pace of innovation, drawing on his development of a robotic wheelchair substi-

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary tute and noting that he is amazed at how well the regulatory people “get it” when one talks to them individually, but how the process and the rules that they have to follow cause trouble nonetheless. Mr. Kamen recounted that 9 years ago when he started building his robot, he did not care what the gyros cost for his proof of concept, so he bought aerospace gyros at $2,000 apiece. They are big and delicate, but he crammed them in, and only used two, since he did not need redundancy just to check the system. For a production machine he planned to get super reliability through redundant architecture, and put in six gyros, monitoring all of them with redundant processes. He also knew that the price would be brought down by using some solid-state gyros that cost a few hundred dollars, so he started trials. He monitored the auto industry, however, and soon discovered that two major suppliers to the automotive industry built a whole subassembly of solid-state gyros for use in advanced systems like the detonation devices on air bags. Because they are making them by the millions it turns out they are $10 apiece and incredibly robust. It also does not matter if something goes wrong with them because of the redundancy in Kamen’s design—with six gyros, one failure would not negatively impact the machine. When he tried to put these new gyros in, though, he discovered that, since the gyro is listed as a critical component of the device, he would have to redo all their system level testing. That would take a year, cost a few million dollars. As a result, he is about to launch the product with a set of gyros that will add $200 to the final selling price of the machine, knowing for the last 2 years that he must wait for the next generation of product to incorporate an individual component that might legally fit the description of critical, but is not critical at all because of the system architecture. He concluded that the approval system has to evolve to where good engineering judgment is what dictates what makes the product safe rather than a process that was designed at a different time. Kshitij Mohan started the discussion on a second major thread by asking about the value of the degree of exclusivity, the protection of patent-like quality, that FDA approval grants to a technology. Dr. Nadeau replied that he hated to use the regulatory process as a barrier to entry, since everyone here is in the business of helping patients, and if one does that one is not helping patients, because that tends to exclude smaller companies. Steven Fodor proffered that the dilemma is not limited to just the FDA process. Once someone is shown how to do something it is always easier for that person to do it again, and big companies particularly often will not pay for the start-up risk of many of the technologies discussed today. Once other people are shown how to do it, they tend to compete with you, and that brings up the nature of our whole patent system and being able to protect one’s innovations. Dr. Nadeau talked about the pros and cons of the Patent Cooperation Treaty (PCT) in response to a question about the getting a “world patent.” He reported that his company generally filed PCT applications first, because the PCT examiners who do the search are professional searchers, whereas in the United States

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary the examiner does both the search and the examination. As a result, the PCT applicant gets a preliminary report and a really good idea of what the patent literature is, which is a terrific advantage for a later filing in the United States. Dean Kamen added that “world patent” is unfortunately somewhat of a misnomer, since U.S. inventors can get patents in India, China, and even places as close as Brazil and Mexico, which happily take your money, but will not enforce the patents, or even let the patent holders enforce them very effectively. As a practical matter though, he said, the PCT has been a big improvement. One can file here in the PCT and, at least for the first year, buy the protection for the initial period of testing and evaluation. It is expensive though, and his company has spent a lot more than $500,000 protecting its core technologies. Another member of the group added that one of the problems a small company faces is raising money to do the development, and a key to that is having clinicians present at medical meetings to describe their experiences with the product. That puts it out in the public arena, however, and hurts chances for a patent. A company can get a basic patent on the PCT or in the United States, but the development process here and the issues of doing the iterative changes all take time. All it takes is a doctor in Europe to say, “Hey, that is a brilliant idea; go get an engineer,” and they will end up getting blocking patents. So, despite spending a fortune, foreign competitors will have a major role when the inventor wants to make that next change in his or her product. The intellectual property issue is a big issue when you tie it to the regulatory issue. Dr. Goodman steered the discussion back to the FDA by asking Steve Fodor whether FDA regulations ever change a decision to pursue innovation, or whether the hurdle of Medicare coverage ever changes a decision about pursuing an innovation or how fast to do so. Dr. Fodor responded that when he first started this pursuing the DNA chip technology he had a lot of thoughts about genetic diagnostics, chips for cystic fibrosis or other diseases, but it turned out that precisely because of the FDA barriers, plus the difficulty of making a profit on those relatively small-market products, he had to make the decision to point the technology toward things that people could just not possibly do otherwise, that is, whole genome scanning, very broad-scale applications. The products that his company now makes are not under any sort of FDA regulatory approval process, except for things such as the HIV product and the p53 and cytochrome P450 chips. These three are focused on specific areas, but for those products he partners with companies like Roche Molecular Systems that have experience with the FDA. Jean Harmon, from NIH, then asked about how to improve the Small Business Innovative Research (SBIR) program at the NIH, to which Dr Nadeau responded that the problem is that if one does the straight-line extension of the obvious one can get funding, but if one proposes anything that is innovative, that is unproven, then one cannot. He suggested that one way of solving that problem is similar to the way companies manage R&D budgets. They take a small percentage

OCR for page 31
Innovation and Invention in Medical Devices: Workshop Summary of their budget and they put it into high-risk areas, some additional percentage into less risky areas, and then finally another percentage into sure bets. Mr. Kamen acknowledged that he had never received an SBIR grant, but recounted his amazement that so many small companies come to his firm with them, particularly people who have left universities. He has had four or five people come to his company over the years who literally have lived on SBIR grants for the last 10 years. They understand the process. They know how to get that money, he said, but the only other thing that is consistent about them is that the Darwinian system of the marketplace would have chewed them up. Kamen’s suggestion for improvement: make the grant bigger and tell people, “You get one and only one. If you do what you said, great, you shouldn’t need us anymore. If you didn’t do what you said, shame on you.” Dr. Fodor took a contrary view, noting that his company had actually found the SBIR programs extremely effective, along with traditional R01 awards and National Institute of Standards and Technology (NIST) grants in the first few years of the company’s existence. Mr. Chess on the other hand, said his company looked into SBIR but decided not to bother, because they figured that if the work was important they should do it, and if it can wait 6 months for them to fill out the application and wait to hear back then it probably is not that important. And, at that time at least, the amount of money was so small it was not worth all the trouble. Dr. Nadeau closed the discussion by repeating the suggestion from his presentation about the idea of matching funds as is done in Germany. If the applicant puts up a dollar, he or she can get two. Maybe, he suggested, NIH should omit the elaborate applications and simply say, “If you have money, we will match it, two for one, and that is it.”