From concept to product

What is it that turns a good idea into a good product? It is easy to say that if we all knew the answer, then “this time next year we will all be millionaires”. However, Jeff Newman is certain that it is a worthy question to debate.

The December 2006 edition of Laboratory News carried an interesting article about Biotechnology YES (Young Entrepreneurs Scheme). Three of the ideas mentioned were: grass that does not need cutting; surgical gloves that change colour if they are damaged; and contact lenses that detect glucose concentrations. So, which of these is likely to succeed and what needs to happen to ensure this? Maybe it is a little unfair on the YES candidates, who had these ideas, to pick on them directly in this article, but the last concept is certainly one worth looking at further, since there is an interesting history related to this, that may help answer our question.

Diabetes is a serious world problem. It is currently estimated that there are over 150 million diabetics worldwide. Worse still, incidence of the disease has risen by an alarming 11% over the last five years, and a further doubling of new cases is predicted in the next 25 years. Diabetes is one of the leading causes of death by disease. When left untreated or improperly managed, the high levels of blood sugar associated with diabetes can slowly damage both the small and large blood vessels in the body, resulting in a variety of complications. For example: the average incidence of heart disease is raised between two and four-fold; the condition is a leading cause of adult blindness; it is responsible for up to 30% of all new cases of serious kidney disease; over 50% of all non-traumatic limb amputations are due to diabetes; and it is a major cause of erectile dysfunction. Careful management is the key to avoiding many of these complications.

So, from our entrepreneurial viewpoint, the market is sizable and appears receptive to suitable products for detecting glucose. So how have products developed to date and where are they heading?

The first home use product was the Dextrostix “reagent strip”, which was launched in 1965. It contained glucose oxidase, peroxidase and colour indicators. The size of the pad on the strip where blood was applied was approximately 9 x 6 mm, which meant that quite a large drop of blood was required. As well as the pain associated with obtaining this blood sample, there were numerous other problems with these strips. It was necessary to wash off the blood, which was inconvenient, and to compare the colour with a colour chart, which was not always easy for diabetics with poor eyesight. Nevertheless, it was welcomed by diabetics, who had nothing else available.

Some of the flaws associated with Dextrostix were circumvented by Tom Clemens, who was employed by for Ames, a division of Miles Laboratories (now part of Bayer Corporation). His work led to the launch of the Ames Reflectance Meter which could read strips similar to the Dextrostix. Interestingly, work on the meter was started in 1966, four years after Leyland Clark’s description of the glucose biosensor, which was theoretically a superior design. However, development of the reflectometer was much faster than the biosensor. Several prototypes were built in 1968 for field trials and a patent application was filed in April of that year. The product reached the market in 1969, just three years after the original work started, although the original meter was expensive and required a prescription. It was also rather large and heavy, weighing approximately 1 Kg. Despite this, the instrument was a success and eventually led to a great variety of other products, whose popularity meant that, for many years, biosensors accounted for only a small proportion of commercial sales.

Glucose biosensors evolved from the oxygen electrode developed by Professor Clark and were first described in 1962. The electrode was modified by entrapping glucose oxidase using a piece of dialysis membrane. The decrease in measured oxygen concentration was proportional to the glucose concentration. Although rather clumsy in design, these electrodes worked well and were adopted in many laboratories, by scientists who assembled the devices themselves. However, they did not become commercially-available until the mid 1970s.

Clark's ideas became commercial reality in 1975 with the successful re-launch (first launch 1973) of the Yellow Springs Instrument Company (Ohio) glucose analyser based on the amperometric detection of hydrogen peroxide, which was preferred to the earlier oxygen measurement. This was the first of many biosensor-based laboratory analysers to be built by companies around the world. These instruments were successful in the laboratory, but did not challenge the reflectometer designs in the lucrative home use market.

In order to be competitive with the reflectance devices, the instrument needed to be smaller and cheaper. It was also clear, from the reflectometer, that the commercially lucrative product was the disposable strip. The biosensor needed a re-design to meet these goals. Fortunately, during a fuel cell project involving Cranfield and Oxford Universities it was realised that the electron transfer compounds, which they were using to increase the efficiency of their fuel cells, held a great deal of promise in biosensor applications. These redox couples, known as mediators, are able to shuttle electrons between the redox centre of the enzyme and the electrode. Depending on the compound used, they can also be regenerated at potentials where interference from species such as ascorbate, urate and paracetamol is minimal. It was also possible to design electrodes using this technology that could be mass manufactured using screen-printing.

The design has changed little since the launch of the original MediSense ExacTech biosensor in 1987 and has become a huge commercial success, with sales reaching around $6 billion last year. Medisense has since been bought (for $867 million) by Abbott Laboratories, one of the world’s largest healthcare companies, although the Medisense name is still used. Interestingly, MediSense has been overtaken by Roche Diagnostics and Lifescan as the market leader, although it remains firmly in the "big four" alongside Bayer.

Improvements in ergonomics have led to an easier to read display and simpler to use instruments. Blood requirements have fallen to a microlitre or less, drastically reducing the pain of testing and data storage and handling capabilities are widespread. During this time, the meter costs have plummeted. Nevertheless, profitability remains high, due to the enormous sales of the disposable sensors.

As was mentioned previously, it was some time before biosensors overtook reflectance devices in the market. Many people are surprised to find that this did not happen until the late 1990s. Today, they are almost totally dominant. However, there are a few major problems which remain with the biosensors described above. The most important of these are that:

• they only provide a “snapshot” of the patient’s blood glucose level (upward and downward trends require a second measurement) • they do not provide an alarm (particularly a hypoglycaemic alarm for sleeping patients) • they are inconvenient to use

In the absence of a cure for diabetes, the “holy grail” for treatment is the successful introduction of an artificial pancreas. Continuous subcutaneous insulin infusion requires frequent, or preferably continuous glucose measurements, which requires an implantable or continuous non-invasive glucose measurement system. Unfortunately, this has proven more complicated than was originally envisaged. Sensor stability, biocompatibility and calibration have been the main issues.

MiniMed introduced an implantable insulin pump in 1990, since when approximately 700 patients have received these units through clinical trials. The Company's current implantable pump, the Model 2007, received European approval in 2000 and has a projected 10-year battery life. The company now produce a blood glucose sensor, which is located in a central vein leading to the heart through a minor surgical procedure. It records glucose levels once per minute and is designed to be replaced after one to two years. The sensor is designed to communicate directly with the implantable pump, with the eventual goal of "closed-loop" control using continuous glucose information to automatically regulate insulin delivery.

Spectroscopic determination of glucose in blood has proved difficult because glucose is in relatively low concentration and its spectra overlaps that of other blood constituents, such as proteins, urea, uric acid, haemoglobin and even water. Many attempts to measure glucose through the skin or mucous membranes have failed, due to loss of light energy in the intervening tissues and the problem of extracting the blood spectra from that of intervening tissues.

Other notable failures include the minimally-invasive GlucoWatch, produced by Cygnus Inc., which was a wrist-worn device intended for detecting trends and tracking patterns in glucose levels. The Pendragon Pendra device also failed, despite being non-invasive. Both devices appeared promising, but failed to deliver sufficiently robust results and, particularly in the GlucoWatch case, were certainly not the pain free alternatives that were first suggested.

So where does this leave our young entrepreneur with his contact lens device? It is still not easy to predict, but there are some things that are clear from what we have seen:

The medical diagnostics industry is slow to adopt new technologies. It took 13 years to turn Clark’s biosensor concept into a product. It took 18 years for biosensors to begin to compete in the home testing market. Crucially, it was not until around 2000 before biosensors overtook reflectance as the dominant technology. Reflectance has now been almost totally replaced, so what will happen to the current generation of biosensors?

In the absence of a cure for diabetes, home blood glucose monitoring will undoubtedly need to continue and the current commercial dominance of mediated electrochemical biosensors will not be easily displaced. New technology has largely arisen in universities and small innovative companies and has then been acquired by the major players who have the distribution networks and market muscle to do justice to new inventions.

Investment continues in non-invasive technology, but products to date have disappointed the market and led to a lack of credibility. In vivo monitoring has materialised after a long gestation period, but utility is still somewhat limited.

Companies appear to have reached a consensus, as evidenced by their recent product introductions and publicly announced research programmes, that the most important next range of home blood glucose monitoring products will be integrated devices. Integration will take several forms and include instruments offering multiple glucose tests, multiple analytes, lancing combined with testing, testing combined with injectors and pumps combined with sensors. The most successful formats will address real needs, sizeable patient groups and offer unrivalled convenience in simplicity of operation and portability.

The contact lens will, therefore, need serious backing, both financially and in marketing. Our entrepreneur will also be likely to face many years of struggle, which statistically appears likely to end in failure. However, the potential rewards are huge. The sale of Therasense to Abbott in 2004 raised $1.2 billion. Even more remarkable is the fact the technology acquired from this sale was only a marginal improvement on the state-of-the-art. A radical new technology could really result in a blockbuster!

The author: Jeff Newman. Jeff is a biotechnology consultant for Cranfield Health, Cranfield University. He has a degree in biochemical engineering and a PhD in biosensors.