Medical Device & Diagnostic Industry Magazine | MDDI Article Index

An MD&DI March 1997 Column

R&D HORIZONS

Hopes are high for alternatives to finger sticks, but so are the technological hurdles.

If blood sugar were better controlled, the complications associated with diabetes might be far less prevalent and far less severe. Yet the average insulin-dependent diabetic exerts relatively poor control, injecting insulin just twice daily, despite conclusive evidence that three to four precisely measured administrations of insulin daily could prevent long-term complications, such as blindness.

The pain and inconvenience of current blood glucose tests, which require finger sticks with lancets to draw blood for analysis on personal glucose monitors, are one reason that the average insulin-dependent diabetic administers the hormone twice a day. But help may be on the way.

Entrepreneurs are racing to develop the ultimate expression of biosensor technology--a fast, painless, and convenient means for testing blood glucose. Ultimately, such a monitor could be tied into an implantable insulin pump that would deliver exact amounts of insulin to the patient. Adequate control could reduce--even eliminate for some patients--complications that impose some $45 billion in health-care costs annually in the United States alone.

In the process, companies making these monitors could reap an impressive financial harvest. Industry sources estimate that the worldwide market for glucose-monitoring products surpassed $1.5 billion in 1994 and continues to grow at a 14% clip annually. Of that market, U.S. sales make up about 59% of total revenues.

About 90% of the total sales today are related to disposable glucose reagent strips for finger stick monitoring. Painless monitors could take away a substantial portion of those sales. "Given the size of the market, anyone who can come up with a viable noninvasive or painless technique is going to make a lot of money," says Gregory Faris, an analyst at SRI International (Menlo Park, CA).

MiniMed's flexible glucose sensor is housed in a tube for subcutaneous placement using a needle introducer.

Faris notes that there has been a virtually endless stream of ideas driving entrepreneurs, including tests using samples of tears, saliva, and urine; an optical technique that scans the eye; and technologies for shining infrared or laser light into and through the body. "It's easy to mislead yourself that you can do it," Faris says. Several companies do, however, appear to be making significant progress.

Cygnus, Inc. (Redwood City, CA), has developed several prototypes of its GlucoWatch, a wrist-worn device that promises to noninvasively monitor glucose levels. The watchlike instrument would use electroosmosis to draw glucose molecules from the patient's skin into a dermal patch, whose contents would be measured and the data interpreted by an integrated circuit.

SpectRx (Norcross, GA) is developing a handheld device that uses a laser to create a micropore about the width of a human hair in the outer, dead layer of skin from which interstitial fluid is collected. This fluid is then measured using off-the-shelf glucose test strip chemistry. Integ (St. Paul, MN) is designing a device that would use a small needle to penetrate the skin and gain a sample of interstitial fluid containing glucose. A handheld battery-operated infrared photometer would then measure the glucose in the sample.

BROKEN PROMISES

But amid the hopes for developing a painless glucose monitor are stories such as that of Futrex Medical Instrumentation, Inc. (Gaithersburg, MD). For years, the firm showcased its DreamBeam, a battery-operated box about the size of a television remote control designed to provide noninvasive glucose measurements with the use of infrared radiation. Last September, the Securities and Exchange Commission (SEC) filed a fraud action alleging that Futrex and its senior officer, Robert D. Rosenthal, made false claims to investors in connection with a $1.85 million private placement of debt securities. The SEC alleges that the company and Rosenthal knowingly deceived investors, presenting false conclusions from clinical studies. During at least one meeting with investors, Rosenthal used the device on himself, and claimed the readings were accurate. But according to the SEC, he allegedly had "directed a Futrex employee to program a DreamBeam to function as if it were giving a glucose reading." Rosenthal was not available to MD&DI for comment.

The Futrex incident has not quelled hopes that a painless glucose monitor can be built. "There's a rich array of technologies supporting biosensor R&D," says Cort Wrotnowski, principal of the consulting firm Amvir Associates (Greenwich, CT), which specializes in the assessment of biosensor technology. "Somewhere there is an answer for what these guys want to do."

Several big names in the medical industry agree. Becton Dickinson (Franklin Lakes, NJ) and Yamanouchi Pharmaceutical (Tokyo, Japan) have signed on to market products under development by Cygnus. Yamanouchi bought marketing and distribution rights for Japan and Korea; Becton Dickinson bought them for the rest of the world. Similarly, Abbott Laboratories (Abbott Park, IL) purchased exclusive worldwide rights to SpectRx technology except in Singapore and the Netherlands, where the company has coexclusive rights.

The leaders in the race to develop a painless glucose monitor are taking either of two tacks. One is the use of infrared--or near infrared -- technology to noninvasively obtain optical signatures indicating the level of glucose. The other collects samples of interstitial fluid for analysis.

HARNESSING INFRARED LIGHT

Infrared analysis is as tantalizing as it is difficult to achieve. "Infrared can penetrate the skin, so measurements are possible at different depths," says biosensor consultant Wrotnowski. "But the feedback can be extremely complex, which means you need very sophisticated mathematical methods to do the analysis." The complexity of the data is a result of the way infrared interacts with aqueous solutions. Water soluble substances absorb infrared radiation very strongly, he explains, returning a "very messy signal." Ironically, nonaqueous substances return a much better signal. "These guys are trying to make something useful out of what constitutes infrared's greatest weakness," he says.

Ominously, the Futrex DreamBeam supposedly was based on infrared technology that could measure blood glucose levels by passing infrared light through a finger. The SEC complaint against Futrex states that studies of an earlier prototype, the Futrex 9000, as well as a version of the DreamBeam were unsuccessful and that a field study conducted in 1995 with the DreamBeam generated useless results, allegedly because of a manufacturing defect.

Another light-based technology, the Diasensor 1000 by Biocontrol Technologies (Pittsburgh) has had its share of problems. The tabletop spectrophotometer is designed to recognize a person's glucose patterns through the use of a light beam that passes through the skin of the forearm into the blood and is then reflected back to a sensor. A microprocessor is intended to interpret the data and calculate the blood glucose level.

Early last year enthusiasm was running high that FDA would soon clear the Diasensor 1000. But an advisory panel in February 1996 recommended against approval. At the meeting, the company produced successful data on only eight patients in its clinical trials, despite enrolling 85. Twenty-two were eliminated due to malfunction of the machine; two were eliminated because glucose levels did not vary sufficiently to calibrate the machine to them. Of the remaining 61 patients, 47 had the machine successfully calibrated to them. The company chose to follow 23 of them for 30 days, and FDA did not object, according to the company. The eight successes were found among those 23 subjects.

In an open letter to stockholders and diabetics, CEO Fred E. Cooper defended the company's position that eight patients provided sufficient data on efficacy and safety: "It was enough because for those eight patients, 263 data points...were submitted to FDA--that's an average of 32 data points per patient. Firms currently using finger stick technology only submit an average of one data point per patient for devices they are attempting to get cleared. That means 100 data points submitted equals 100 patients studied. Therefore, 263 data points submitted for the Diasensor 1000 is equal to having tested 263 patients--a substantial test size."

In the 10 months following the panel meeting, Biocontrol withdrew, revised, resubmitted, and then withdrew again a 510(k) application for the device. The company is continuing to work toward FDA clearance of the Diasensor 1000, says company spokesperson Susan Taylor, "and we are going to keep working at it." Company officials are now trying to finalize the details for a new study to be conducted in the homes of subjects. When completed, Biocontrol expects to submit a newly revised application to FDA.

Descriptions of the Diasensor 1000 published by the company refer only to "optics, electronics, and detection subsystems; software; and algorithms." Details about the device are not released by the company due to the competitive nature of the industry, says Taylor, who will state only that the device uses "a near infrared spectrum."

Cygnus's GlucoWatch uses electroosmosis to draw glucose molecules from the skin into a dermal patch for analysis.

Whereas Biocontrol Technologies is trying to use light radiation to non-invasively probe the patient, Integ's LifeGuide System uses a small needle to sample interstitial fluid in the upper layer of the skin. "We do not puncture the skin; we go into the dermal layer," says Dave Talen, Integ marketing manager. "The dermal layer has very few capillaries and very few nerve endings, so when the needle probes only to that depth, you don't draw blood and you don't feel pain, in the traditional sense." Pushing the device, which is about the size of a large cellular phone, against the skin forces the fluid into a "read" window positioned between an infrared source and a detector. "The glucose molecules absorb a certain amount of the energy and we measure that absorbence," Talen says. Clinical trials are expected to begin in summer.

SAMPLING INTERSTITIAL FLUID

The SpectRx system also samples interstitial fluid but rather than use a needle, the device fires a laser into the skin, creating a micropore approximately 80 µm across and 20 µm deep. The interstitial fluid that flows into the micropore is sampled and then passed to a test strip analyzer now on the market for conventional finger stick blood glucose testing. SpectRx spokesperson Bill Wells refused to provide more details about the device or its stage of development, noting that "we have developed a handheld prototype that successfully creates micropores." In some subjects, Wells says, the interstitial fluid rushes into the micropore quite readily. "In other people, it has to be coaxed out," he says. "There are some enhancements that are part of the development process that allow us to collect the fluid."

The analysis is much more straightforward than the collection process. As a result of its alliance with Abbott, the company is integrating Abbott's MediSense test strip technology into the device. According to SpectRx, preliminary tests have shown a high correlation between glucose in the interstitial fluid and in blood.

MiniMed (Sylmar, CA) is working on a minimally invasive monitor that would use a small flexible probe to sample interstitial fluid from the subcutaneous tissue between the skin and muscle. The probe is inserted using a needle and then the needle is removed. "There is a temporary immediate pain, the same as you would get from an injection," says John Mastrototaro, director of sensor development at MiniMed. The probe would be replaced with a new one after three or four days to prevent undue irritation and the risk of infection, he says. A sensor in the probe analyzes the fluid for glucose content, passing the data via cable to a microprocessor that might be worn on the belt like a pager. Early models might not provide a quantitative readout of glucose levels, but rather be programmed to emit an alarm if glucose levels exceed a certain range. "If it detects that the glucose is low, an alarm would signal the patient to do a finger stick to determine the glucose level," Mastrototaro explains.

Feasibility studies involving 12 to 20 insulin-dependent diabetics are under way. A trial involving up to 50 subjects is scheduled for midyear.

MiniMed is currently a leading manufacturer of external insulin pumps, holding about 75% of the U.S. market for those pumps. The company has also developed an implantable pump being sold in Europe. Ultimately, an automatic glucose sensor capable of delivering precise glucose measurements might be integrated with a version of the pump. "We have the delivery device; the next step is closing the loop by creating a sensor that will automatically tell the pump how much insulin to deliver and when," says Jim Berg, a MiniMed spokesperson. "That is our grand design--to create an artificial pancreas."

Cygnus, Inc., advocates a noninvasive approach that leverages electroosmosis to obtain interstitial fluid. "When exposed to this low-level electrical energy, charged particles in the interstitial fluid pull the fluid out of the skin and the glucose molecules go along for the ride," explains Craig Carlson, Cygnus vice president for corporate marketing and strategic planning. A small disposable pad, constructed from proprietary material, located between the GlucoWatch hardware and patient skin collects the glucose molecules. Those molecules trigger an electrochemical reaction with a reagent in the GlucoPad, thereby generating an electric current. A sensor in the GlucoWatch measures the resultant current and an application-specific integrated circuit (ASIC) calculates the glucose in the patient's blood.

Levels are calculated automatically at set times throughout the day and night. At the push of a button, those calculations are displayed on the watch dial in the context of trends in glucose levels. Alarms would be designed to go off if the levels exceed specific limits. An electronic memory also would allow downloading of the data to a computer for long-term trend analysis, potentially helpful in managing the disease. "We see ourselves not so much as providing glucose measurements as information managers," Carlson says.

CHALLENGES IN DEVELOPMENT

Common to all developers is the challenge of providing accurate blood glucose measurements, regardless of such variables as patient exercise and the digestion of different kinds of foods. The approach that is being taken by most companies, including MiniMed, is to control for these variables by testing subjects under different conditions and then modifying the device to render accurate measurements.

Each of the companies have other specific challenges to overcome. The invasive nature of the MiniMed device, minimal though it is, presents the danger of skewing data. "If the sensor causes a lot of irritation or trauma at the site of insertion, it can cause the local glucose concentration to change," Mastrototaro says. "Another possible problem is that proteins in the body could foul the sensor or impair its performance."

Noninvasive infrared technology presents special challenges because skin and bone absorb and deflect light. There is the added problem of trying to iden-tify optical signatures of glucose levels, signatures that may be as specific to patients as fingerprints. Therefore these devices may have to be calibrated to individual users.

Calibration proved to be a problem for Biocontrol in its clinical study presented to the FDA advisory panel last year. Other companies have been able to avoid that hurdle. SpectRx, for example, uses an off-the-shelf test strip technology provided by Abbott. "It certainly shortens the development time to use existing technology," Wells says.

Cygnus officials believe their technology has no serious technical challenges ahead. But the company must contend with the pitfalls of being a device integrator. "The task we have now is making sure all of the components that have been optimized and tweaked since the early prototype testing operate effectively as a unit," Carlson says.

Cygnus, whose strength is in the design of transdermal patches, most notably the mass-marketed Nicotrol patch, must rely on contract engineering firms to build hardware for doing the analysis and providing the readings. The pitfalls of such an arrangement became obvious in November 1996 when Cygnus announced that a change in the ASIC that serves as the brains of GlucoWatch would delay delivery of the latest prototype and, consequently, clinical trials. The problem, according to Carlson, was a defect in the design of the chip. "Two wires were touching one another," he says. "So when the chip was put in place, there was a short." At press time, it appeared that a reengineered ASIC would soon be delivered and integrated into the latest prototype.

Cygnus executives have shunned media attention, partly because of the highly competitive nature of the industry, Carlson says, but also because hopes have been unduly raised by competitors. Company policy, he explains, is to make a viable product and then seek publicity --not the other way around. MiniMed has a similar policy. "People's lives are involved and we don't want to suggest that this technology is right around the corner," says MiniMed spokesperson Berg. "This is very tricky, difficult work."

Greg Freiherr is a contributing editor to MD&DI.

Copyright © 1997 Medical Device & Diagnostic Industry