9.2.1 Classes of materials

Hydrophilic polymers

The pharmaceutical industry has shown great interest in the development of controlled release systems based on hydrophilic polymers. Hydrogels represent a common class of polymers used for drug delivery. In essence a hydrogel is a material based on polymers such as poly(vinyl alcohol) or poly(acrylic acid) which would normally be soluble in water. However, either during or after the polymer synthesis, a degree of crosslinking converts the linear polymer chains into a polymer network. This process renders the polymers insoluble; however, the high presence of hydrophilic groups within the network gives the structure a high affinity for water. Although the network does not dissolve, it is capable of adsorbing water with consequent swelling of the polymer matrix. The nature of the polymer and degree of crosslinking affects the swelling behaviour. A network with few crosslinks and a large number of hydrophilic groups will adsorb large amounts of water with a high degree of swelling. Less hydrophilic monomers, incorporation of hydrophobic co-monomers or a high degree of crosslinking all act to reduce water adsorption, usually leading to a firmer, more rigid gel. Due to the fact that they include a high water content, hydrogels often show high degrees of biocompatibility. The polymer network itself can be either bio-inert or be a biodegradable polymer network. Natural polymers can also be used with, for example, hydrogels based on chitosan, alginase or collagen having been utilised. The application of a wide variety of hydrophilic biodegradable polymers to delivery of proteins has been extensively reviewed (Gombotz et al. 1995; Ganji and Vashedghani-Farahani 2009). Frequently the drugs, especially if they are biologically derived (such as proteins) can be quite unstable. Besides enabling the controlled release of the active material, these hydrogels often act as a stabilizing media for these unstable agents. For example, a drug could be incorporated into a hydrogel either as it is synthesised or post-synthesis. This can then be applied to the patient via any of the techniques described above. Once in vivo, the drug is released by a number of means. These might include simple diffusion or alternatively the polymer may be eroded or dissolved, for example, by digestion. In an ideal situation this process will occur at a constant rate, leading to continual release of controlled amounts of the drug. If the polymer is utilised as a transdermal patch or is surgically implanted close to an affected site, the drug can be delivered where it is most needed.

A similar method utilises polymers which are actually degraded rather than just swollen. Examples include polymers which are based on poly(lactide) or poly(glycolide) (Fig. 9.1), and which are slowly hydrolysed in vivo to release an active agent, e.g. leuprolide acetate (Ogawa et al. 1988).

A polymer must address several criteria before it is suitable for use as a carrier system. In the case of polymers that are effective in vivo, they must be both biocompatible and must biodegrade within a reasonable period of time. Irrespective of the manner of application, the polymer itself and any degradation products must be non-toxic and must create neither allergic nor inflammatory response. The method of release is often dependent on the nature of the drug itself; low molecular weight drugs are capable of diffusing out of the polymeric matrix and, if water soluble, will be rapidly released. However, larger molecules such as proteins will not diffuse as readily and often remain within the hydrogel matrix until the polymer itself is either degraded or enzymatic digestion releases them. Release rates depend on several factors including the quantity/dosing of drugs within the composite, the rate of degradation of the polymer, the water content (if it is a hydrogel-type material), and the presence and degree of crosslinking.

A property that makes hydrogels exceedingly suitable for drug encapsulation is that many of the materials used for these systems can be synthesised so as to be responsive to their environment. A change of pH, for example, can lead to protonation or deprotonation of active groups within the polymer. This can change its affinity for water, giving us a material whose swelling is pH responsive, thereby affecting many of its other physical properties such as permeability. This gives us the opportunity to design drug release agents which are selectively triggered by certain conditions, e.g. their responsiveness to pH means they can be designed to release active agents within a selected part of the digestive tract. Similar smart materials can be designed which respond to changes in physiological conditions and therefore only release the drug at times when it is needed. Hydrogels can be designed which respond to other stimuli as well as pH, such as ionic strength, temperature and electric field. They can even be designed to respond to the presence or absence of specific analytes.

As mentioned earlier, polymer-based delivery systems allow the application of active agents in many ways including ingestion, transdermal patches, suppositories, ocular and subcutaneous methods. A few examples are given here: nitroglycerin is often used in the treatment of angina; however, volatilisation of the active component can lead to loss of tablet activity (Markovich et al. 1997). This problem can be mitigated by the use of a acrylic-based hydrogel as a host for the nitroglycerin and incorporation of this composite in the construction of a transdermal patch.

There has been wide research into utilising these materials in the treatment of cancer for, for example, the delivery of ara-C for leukaemia. This drug gives rise to a number of side effects but these are mitigated when a constant infusion of the drug is introduced subcutaneously. As an alternative approach, crosslinked polyhydroxyethyl acrylate can be utilised as a host for this material. Discs of this composite display a steady controllable release of the active material (Teijon et al. 1997). A similar composite based upon polycaprolactone/polyethylene glycol (PCL/PEG) has been used as a matrix for the anticonvulsant drug clonazepam. Stable constant release properties were displayed for over 45 days (Cho et al. 1999).

Similar polymers are suitable for localised delivery of pharmaceuticals. In the treatment of brain cancer, one approach that has been successfully used is surgery to remove as much of the tumour as possible, followed by placing in the surgical site small wafers based on polyanhydrides. These contain the anti-cancer drug carmustine which is slowly released over a one month period to kill any remaining tumour cells (Brem et al. 1995). Alternatively thermosensitive hydrogels based on block copolymers of poly(ethylene oxide) and poly(lactide) have been made which are liquid at 45 °C and can be injected directly to the required site. Upon cooling to body temperature, the gel immediately sets, trapping any pharmaceutical compounds in the solution, so allowing them to be released slowly (Jeong et al. 1997).

9.2.2 Micelles, vesicles and liposomes

Micelles are formed by a wide range of amphiphilic surfactant molecules in aqueous solution. A typical amphiphile contains a long alkyl chain and a polar headgroup. In solution the interactions between the alkyl chains and water are extremely unfavourable and this drives aggregation of the alkyl chains together to minimise this interaction. A structure then forms as shown schematically in Fig. 9.2, where a spherical aggregate spontaneously assembles, with the polar headgroups on the outside of the sphere and a hydrophobic interior, stabilised by van der Waals interactions between the chains.

What makes micelles so useful as carrier agents is that hydrophobic species can be incorporated within the core of the micelle. This enables the transport of hydrophobic active agents at concentrations much higher than would be possible for a simple aqueous solution of these compounds. The hydrophobic environment of the core moreover protects the guest against hydrolysis and enzymatic degradation.

As an alternative to the classical long chain surfactants shown in Fig. 9.2, amphiphilic block copolymers are also very adept at forming micellar structures. For example, a block copolymer containing a hydrophobic block, such as polystyrene, and a hydrophilic block, such as polyethylene oxide, can dissolve in water to give a micellar structure with a polystyrene core surrounded by a polyethylene oxide corona, with the whole structure being typically 5–50 nm in diameter. Block copolymers are especially suitable for this purpose since they are available in a wide variety of molecular weights. Since a wide variety of monomers and the composition ratios can easily be selected, this allows fine control of the size, composition and morphology of the micelles. Incorporation of many functional groups is easily attained; for example, crosslinking groups can be utilised to stabilise the micelle structure and the surface properties, selectivity and biocompatibility of the micelles can all be tailored.

In a similar manner, a variety of compounds can be utilised to form vesicles. These are somewhat more complex structures than micelles (Fig. 9.2) in which a spherical membrane is formed containing an aqueous core. Liposomes are types of vesicles which are widely used within drug transport and consist of spherical phospholipid bilayers with an aqueous core and are the most commonly used of these carrier systems (Langer 1998; Kaparissides et al. 2006). They are capable of transporting either hydrophilic agents within the core or hydrophobic agents which are incorporated into the membrane. The membrane itself can be tailored to prevent access of enzymes to the core, thereby preventing enzymatic degradation, but do, however, allow the diffusion out of the active agent. For example, channel-type proteins may be incorporated within the membrane and retain their activity, thereby allowing transport of ions or other small solutes such as drugs through the membrane.

The use of liposomes can lead to a large increase in drug carrying capacity compared to the use of polymeric systems; however, disadvantages can be encountered such as shorter shelf-life and rapid destruction of the vesicle within the body. The biocompatibility of these systems, however, can be improved by modifying their surface properties, such as by the attachment of PEG units (Park et al. 1997; Lasic et al. 1999; Moghimi and Szebeni 2003). Also, antibodies can be attached to the surfaces of these systems, allowing them to specifically target sites that require treatment. Tumours, for example, can be targeted by substituting liposomes with antibodies against the HER2 proto-oncogene, found in breast and other cancers (Kirpotin et al. 1997; Park et al. 1998). Similarly amylopectin can be attached to the surface of liposomes to enable targeting of lung tissue (Vyas et al. 2004).

More recently there has been much attention paid to using liposomes to deliver protein or peptide-based drugs since they can maintain the guest in an internal aqueous environment whilst protecting from the external environment. For example, vasorestrictive intestinal peptide could be encapsulated into PEG-conjugated liposomes and then delivered via aerosol inhalation, with the liposome protecting the peptide from aggregation and enzymatic degradation (Hajos et al. 2008). Similarly liposomes could be used to deliver therapeutic proteins against Chlamydia into mice and demonstrated reduced rates of infection (Hansen et al. 2008). Insulin can be incorporated into multivesicular liposomes coated with chitosan or carbopol. The resultant liposomes were 26–34 μm in diameter, contained 58–62% protein and when applied by nasal or ocular administration showed sustained protein released as assessed by reduction of blood glucose in diabetic rats (Jain et al. 2007). Other workers have studied chitosan containing liposomes and demonstrated controlled release of calcitonin in rats after oral application (Werle and Takeuchi 2009).

Liposomes have also been widely studied since they offer the potential to introduce drugs into the body via application to the eyes. Much of the work on ocular administration of drugs and other compounds has been recently reviewed (Mishra et al. 2011). Recently there has also been interest in catansomes, vesicles made from oppositely charged surfactants rather than phospholipids. In a recent example, perfluorinated sodium octanoate could be mixed with a hydrogenated surfactant such as dodecyl pyridinium chloride to form vesicles 100–200 nm diameter which could encapsulate a fluorescent marker (calcein) and release it upon application of fatty acids (Rosholm et al. 2012).

9.2.3 Nanotechnology

In recent times, nanotechnology has become an intensely studied field of research. The application of nanosized materials is expected to make significant advances in biomedical applications such as drug delivery and gene therapy (Moghimi et al. 2001; Moghimi and Szeleni 2003; Sahoo and Labhasetwar 2003).

Earlier in this chapter the use of biodegradable polymers as drug carriers was discussed. Nanoparticles of these types of materials with sizes in the range 10–1000 nm can be synthesised and incorporate drug molecules by way of entrapment or binding. Similarly, nanocapsules can be synthesised in which a polymer membrane forms a vesicle-like structure with the drug molecules confined within a central cavity. Because of their small size, these systems are capable of passing through small capillaries and be taken up by cells, allowing efficient drug accumulation at a target size (Desai et al. 1997). In these materials, size and surface properties determine their distribution in the body (Sahoo and Labhasetwar 2003). Localised application of these nanoparticles can be achieved by their tendency to accumulate in tumours due to enhanced permeation effects; this has been extensively reviewed elsewhere (Maeda 2001). Alternatively the nanoparticles can be delivered locally to the size of interest such as within a specific artery after balloon angioplasty (Guzman et al. 1996) to deliver long-term release (14 days) of dexamethasone. More recently nanoparticulate PLGA has been used to deliver zinc phthalocyanine into mouse tumours for use in photodynamic therapy (Fadel et al. 2010). Other workers used PLGA nanoparticles to deliver paxitaxel (Yang et al. 2009) or to simultaneously deliver vincristine and verapamil (Song et al. 2009) into cancer cells. Chitosan nanoparticles substituted with folic acid could be used to selectively deliver oligonucleotides into tumour cells (J. Wang et al. 2010) and similarly modified chitosan nanoparticles could selectively deliver paxitaxel with good cytotoxicity (Sahu et al. 2011). Crossing the barrier between blood and the brain or central nervous system is a major challenge for many drugs; however, modified chitosan/PLGA nanoparticles have been shown to cross the barrier (Z. H. Wang et al. 2010). These and other polymeric nanoparticles for drug delivery have been recently extensively reviewed (Parveen et al. 2012).

Dendrimers are branched polymers grown from a central core with very precisely controlled degrees of polymerisation. Figure 9.3 shows a schematic of a fourth generation dendrimer. This makes their size, composition and molecular substitution controllable to an exact degree and is an alternative form of polymer nanoparticle to those made by classical emulsion-type polymerisations. There has been great interest in utilising dendrimers as drug carrier agents because not only can the bulk of the dendrimer be synthesised exactly but a wide variety of surface groups can be utilised, enabling further surface functionalisation. Dendrimers, for example, can be made with hydrophilic surfaces and hydrophobic interiors or vice versa. Reviews on this subject (Svenson and Tomalia 2005; Gupta et al. 2006) and on the biomedical applications of these materials (Mintzer and Grinstaff 2011) extensively detail many of the most pertinent advances in this field, so only a few highlights will be given here.

Dendrimers have been utilized to carry a variety of small molecule pharmaceuticals. Commercial poly(amido amine) (PAMAM) dendrimers have been used to encapsulate the anticancer drug cisplatin, giving conjugates that exhibited slower release, higher accumulation in solid tumours, and lower toxicity compared to the free drug (Malik et al. 1999). PAMAM dendrimers which had been functionalised with PEG chains had their encapsulation behaviour for the anticancer drugs adriamycin and methotrexate studied. Up to 6.5 adriamycin molecules or 26 methotrexate molecules per dendrimer could be incorporated for one of the materials studied. The drug release from this dendrimer was slow at low ionic strength but fast in isotonic solution (Kojima et al. 2000). Similar dendrimers encapsulated the anticancer drug 5-fluorouracil, showing reasonable drug loading, and reduced release rate and haemolytic toxicity (Bhadra et al. 2003). Up to 78 molecules of the anti-inflammatory drug ibuprofen molecules were complexed by PAMAM dendrimers through electrostatic interactions between the dendrimer amines and the carboxyl group of the drug. The drug was successfully transported into lung epithelial carcinoma cells by the dendrimers (Kolhe et al. 2003). Other recent studies have also indicated that low generation PAMAM dendrimers cross cell membranes (El-Sayed et al. 2003).

One of the main applications recently of dendrimers is as carrier agents for anticancer drugs (Mintzer and Grinstaff 2011). Glycerol succinic acid dendrimers have been shown to encapsulate hydrophobic molecules (Morgan et al. 2006), including the anti-cancer drugs 10-hydroxycamptothecin and 7-butyl-10-aminocamptothecin. The resultant materials display cytotoxicity against breast, lung and colonorectal cancer cells. In mice the hepatotoxicity of methotrexate and 6-mercaptopurine, both Food and Drug Administration (FDA)-approved anticancer drugs, was shown to be reduced by binding to a melamine-based dendrimer (Neerman et al. 2004). Dendrimers could be substituted with glucosamine, which enabled them to cross the blood–brain barrier, and loaded with methotrexate. The resultant composites showing higher efficiency and enhanced permeability compared to the drug alone (Dhanikula et al. 2008).

Covalent binding of the drug species to the dendrimer by a linkage which can then by hydrolysed or otherwise decomposed has also proved an effective means of drug carriage. Doxorubicin could be attached by acid-sensitive cis-aconityl linkages to a commercial PAMAM dendrimer conjugated with various amounts of PEG chains (Zhu et al. 2010). These, when mixed with ovarian cancer cells, showed clear cellular uptake of the dendrimer along with release of the drug. Higher levels of PEG substitution led to higher levels of uptake. An asymmetric biodegradable PEG substituted polyester dendrimer containing 8–10 wt % doxorubicin attached by covalent hydrazone linkages could be prepared (C. C. Lee et al. 2006) and injected into cancerous mice. The dendrimer composite was nine times more effective at being uptaken by tumours than the pure drug and led to 100% tumour regression and survival of the mice over 60 days whereas no regression was seen for the control group treated with doxorubicin alone. Targeted dendrimers such as the recent example substituted with folic acid and containing methotrexate have been shown to display specificity and cytotoxocity towards cancer cells (Zhang et al. 2010). Other very recent work has compared doxorubicin encapsulated in PEGylated dendrimers and PEGylated liposomes and concluded that although the two systems display similar superior anti-tumour efficiency and enhanced accumulation in rat tumours to the native drug, the dendrimer has lower systemic toxicity (Kaminskas et al. 2012).

Ceramic nanoparticles have also been studied due to their inherent advantages of ease of synthesis in a wide variety of sizes, shapes and porosities by methods similar to sol–gel processes. They are available in very small (< 30 nm) sizes making them capable of crossing cell membranes, possess surfaces which are easily chemically modified and do not display swelling processes with changes in pH (Sahoo and Labhasetwar 2003). For example, silica nanoparticles can be constructed with an anti-cancer drug entrapped within their cores (Roy et al. 2003) which are stable in aqueous systems. Tumour cells take up these nanoparticles, which upon irradiation generate singlet oxygen which significantly damages the tumour cells. Similar particles have also been shown to act as non-viral DNA carriers (Bharali et al. 2005) and to be capable of crossing the blood–brain barrier.

Other metal oxides have been studied, especially iron oxides since they are paramagnetic and therefore potentially allow for targeting certain locations in the body by using a magnetic field (Parveen et al. 2012). Hyaluronic acid coated iron oxide nanoparticles were shown to be capable of transferring peptides into cells (Kumar et al. 2007). Other workers demonstrated delivery of doxorubicin into breast and prostate cancer cells using polymer-coated iron oxide nanoparticles (Jain et al. 2005) or delivery of genistein using chitosan-modified nanoparticles (Si et al. 2010). Similar particles coated with polyethyleneimine were shown to be uptaken by cancer cells and when injected into rat carotid arteries, they could then be magnetically targeted into the brain (Chertok et al. 2010).

A wide range of other nanoparticles are being studied as delivery agents such as gold and other metals, and carbon nanotubes. These are outside the scope of this chapter but have been extensively reviewed elsewhere (Parveen et al. 2012). One major issue with any nanoparticle-based delivery system is potential long-term toxicology issues, since the toxicity of many nanoparticles is still unknown. However, there have been a range of studies showing that many materials, which whilst benign in bulk, are highly toxic in a nanosized form, a classical example being asbestos.

9.3.1 Polyanhydrides

As previously mentioned, polyanhydrides are widely used as slow release agents since they biodegrade reproducibly with no toxic by-products. Commercial products include materials such as Decapeptyl SR®, manufactured by Ipsen Ltd which contains the active ingredient triptorelin acetate encapsulated in poly(lactide-co-glycolide). This is utilised in the treatment of prostate cancer. Similar products include Lupron Depot® (Abbott Laboratories, active ingredient leuprolide acetate) and Sandostatin LAR® (Novartis, active ingredient octreotide acetate).

Nitroglycerin is a problematic drug due to its loss of tablet activity, often by volatilisation of the active component (Markovich et al. 1997). This can be avoided by incorporation of the nitroglycerin into an acrylic-based hydrogel which is then incorporated into a transdermal patch, as exemplified by products such as Deponit® (Schwarz Pharma), Minitran® (3 M Pharma) and Nitrodisc® (G.D. Searle Company).

9.3.2 Liposomes

A number of commercial drug delivery systems have been developed, especially in the field of cancer treatment. Daunozome® is a liposomal form of daunorubicin, a chemotherapy drug given to treat AIDS-related Kaposi’s sarcoma. Doxorubicin can also be formulated within PEG-substituted liposomes to give Doxil® which can be used to treat a variety of cancers. Other anti-cancer drugs which have been formulated in this way include cytarabine to give Depocyt®. Various other commercial liposome formulations have also been brought to market (Zhang et al. 2008).

9.4.1 History and format of biosensors

The purpose of this section is to introduce the concept of using biological molecules as the selective recognition elements within biosensors. Most sensors consist of three principal components, as described below and shown in Fig. 9.4:

There are several advantages associated with using biological molecules as the active recognition entity within a sensor. Usually they display unsurpassed selectivities; for example glucose oxidase will interact with glucose and no other sugar, and in this way will act as a highly selective receptor. In the case of glucose oxidase, the electrochemically inactive substrate glucose is oxidised to form gluconolactone along with the concurrent generation of the electroactive species hydrogen peroxide. Enzymes also generally display rapid turnover rates and this is often essential to (a) avoid saturation and (b) to allow sufficient generation of the active species in order to be detectable.

Antibodies bind solely to their antigens and achieve specificity via a complex series of multiple non-covalent bonds. Since the principle of immunoassay was first published by Yalow and Berson in 1959, there has been an exponential growth in both the range of analytes to which the technique has been successfully applied and the number of novel assay designs. Development of enzyme-labelled immunoanalytical techniques, e.g. enzyme-linked immunosorbent assay (ELISA) has provided analytical tests without the safety risks associated with radiolabelling-based techniques.

The rapid measurement of analytes of clinical significance, e.g. towards various disease markers, would permit earlier intervention, which in a medical setting is frequently of utmost importance. There has been much research on the development of direct immunosensors that do not rely on the use of a detectable label. Such a system will lead to simpler assay formats and ideally lower times of detection. A reusable and rapid detection system would, moreover, allow for continuous real-time measurement, so helping to maintain optimal homeostatic conditions.

Unfortunately there are also some disadvantages related to the construction and use of biosensors. Often the biological species can either be extremely expensive or difficult to isolate in sufficient purity. Immobilisation of these species can lead to loss of activity and the presence of various chemical species in the test solution can also cause loss of activity, e.g. enzymes can be easily poisoned by heavy metals. In biological samples such as blood or saliva, there can also be solutes that are electrochemically active and interfere with determinations of the target species or various species may be present which bind to the surface so causing fouling and loss of sensor response.

Antibody–antigen binding is based on multiple non-covalent interactions, and therefore there are no newly formed molecules, protons or electrons that are easily detectable, which has limited the development of direct antibody affinity-type sensors. Also affinity binding constants typically range from 10⁵ to 10¹¹ mol. L^− 1 meaning that the antibody-antigen binding event is often irreversible. As a consequence of this, many contemporary immunosensors are only of use for ‘single-shot’ analyses and must be disposable in nature.

9.4.2 Glucose biosensors

A series of extensive reviews on biosensors and their history have been published elsewhere (Hall 1990; Eggins 1996; Wang 2001) and therefore only a brief history will be given here. Easily the most intensively researched area has been towards the development of glucose biosensors (Wang 2001; Newman et al. 2004). The reason for this is the prevalence of diabetes, which has become a world-wide public health problem. Diabetes represents an increasing epidemic with, at the time of writing, 346 million sufferers worldwide and this is estimated to double between 2005 and 2030 (World Health Organisation, www.who.org). The world market for biosensors is predicted to reach $15–16bn by 2016 and in 2009 approximately 32% of this market was for blood glucose monitoring. Point-of-care applications account for approximately half of the market (Thusu 2010).

These factors have led to the development of a number of inexpensive disposable electrochemical biosensors for glucose, incorporating glucose oxidase bound immobilised at various electrodes. They are generally amperometric sensors, with electrodes polarised at a set potential; the oxidation or reduction of a chosen electroactive species at the surface will then lead to generation of a detectable current. The principal classes of glucose and other types of biosensors are described below.

First generation biosensors

The first electrochemical glucose biosensor was based on an oxygen electrode (Clark and Lyons 1962). A film of immobilised glucose oxidase was laid down upon the oxygen electrode, which was overlaid with a semipermeable dialysis membrane. Upon exposure to glucose, the enzymatically catalysed oxidation reaction occurs, causing a localised consumption of oxygen.

$\mathrm{glucose}+{\mathrm{O}}_2\stackrel{\mathrm{Glucose}\mathrm{oxidase}}{\to }\mathrm{gluconolactone}+{\mathrm{H}}_2{\mathrm{O}}_2$ [9.1]

[9.1]

This then leads to a drop in the current generated at the oxygen electrode. This device was subject to fluctuations caused by variable oxygen levels. However, further work (Updike and Hicks 1967) utilised two oxygen electrodes, one of which was coated with glucose oxidase. Measurement of differential current between the two electrodes served to cancel out these effects. Alternatively the production of hydrogen peroxide can be electrochemically determined instead. The first commercial glucose analyser was the Model 23 YSI analyser, launched by the Yellow Spring Instrument Company in 1975 and based on the Clark electrode. This device was capable of measuring the glucose level in 25 ml of whole blood.

The results from these so-called first generation sensors can be affected by fluctuations in the ambient oxygen concentration or by the presence of electroactive species such as ascorbate, that are capable of being oxidised at + 650 mV – giving rise to an erroneous result. Concentration of interferents at the electrode surface can be minimised by application of a permselective coating to the sensor, thereby reducing interference from electroactive species. Polymeric materials have led the way, with materials such as the fluorinated ionomer Nafion (Turner and Sherwood 1994) and cellulose acetate (Maines et al. 1997) being two of the most commonly used. A beneficial side effect is that these materials can also confer a degree of biocompatibility. An alternative approach has been to electropolymerise suitable monomers to form protective coatings. 1,2-Diaminobenzene (Myler et al. 1997), for example, when deposited at the bioelectrode surface serves to both stabilise the electrode due to its inherent high biocompatibility whilst also imparting selective exclusion of interferents such as ascorbate.

Diabetes as a condition requires regular monitoring of blood glucose levels, so hospital analysis is impractical for a normal lifestyle. The obvious solution has been the development of inexpensive home detection methods where the physiological sample, usually blood, can be analysed by the patient. The problems of cleaning the sensor are moreover negated by using disposable sensor strips.

Second generation biosensors

The reaction of glucose oxidase with glucose gives rise to the formation of gluconolactone and the reduced form of the enzyme, which is then reoxidised by oxygen. Direct transfer of electrons from the electrode would circumnavigate this reaction; however, the active site of glucose oxidase is encased in a protein sheath which inhibits this transfer. To facilitate this transfer, a suitable chemical species can be utilised to ‘shuttle’ electrons back and forth between active site and electrode. This moiety, known as a mediator, must react readily with the enzyme to avoid competition by ambient oxygen and, in both its reduced and oxidised forms, be stable and preferably require as low an over-potential to be oxidised as is feasible. This method sidesteps problems associated with detecting species such as oxygen or peroxide and also lowers the potential required for measurement of the enzyme catalysed reaction, thereby reducing the inference by redox active species present within the sample to be studied. A typical reaction scheme (Fig. 9.5) where a ferrocene compound acts as mediator is shown (Cass et al. 1984). The first home glucose testing kit, the Exactech® glucose biosensor, is based on this chemistry. The actual pen-sized device is produced by Medisense® and utilises a disposable strip upon which a single drop of blood is placed. A wide range of glucose sensors (Wang 2001) based on this method have since become commercially available for home glucose testing. Further work has concentrated on reducing blood volumes and produced devices such as the Pelikan® Sun device which only required microlitre blood volumes (Newman et al. 2004). Pelikan® Sun has been discontinued but other manufacturers such as Tiniboy (www.tiniboy.com) have concentrated on minimising lancet size.

9.7.1 Point-of-care tests

Although the biosensor market is dominated by glucose testing applications, there is great interest in developing simple tests for a wide variety of medical problems which can be easily applied by the patient and detect physical problems which are beginning to occur whilst still in an early stage, thereby enabling the patient to seek early medical help. An example of this was the Multisense® system, initially developed by Oxford Biosensors, which detected early markers of coronary heart disease. Heart disease is a major killer in the western world and the risk of developing this condition is known to be related to factors such as diet, smoking, weight and high blood cholesterol.

The Multisense® systems was based on a hand-held device with disposable electrochemical test strips. Each strip contains several sensors which individually sense for cholesterol (high density and low density lipoprotein) and triglycerides. Testing utilises a single drop of blood in the same manner as a glucose sensor, with the intent of fast and simple screening, diagnosis and monitoring of patients who are at risk of heart disease. Unfortunately Oxford Biosensors went into receivership in 2009 and no product of this type has as yet come onto the market.

9.7.2 Artificial pancreas

For many diabetes type 1 sufferers, life is a constant round of glucose testing and insulin injections. Externally worn insulin pumps have been developed, such as those by Medtronic (www.minimed.com). These have the advantage that they remove the need for injections by introducing insulin subcutaneously though a cannula. When coupled together with a glucose biosensor, there is then the potential for the device to inject insulin ‘on demand’. This was the first device of its type to receive FDA approval (Newman et al. 2004). This has led to the development of such products as the MiniMed Paradigm® Revel™ combined glucose sensor and insulin pump, which can automatically deliver insulin in response to glucose levels. At present, current commercial devices usually can only be used for a few days at a time without maintenance and injection, and sensing sites need to be changed regularly. An implanted device which would automatically respond to changes in glucose levels and display longer-term stabilities would in effect act as an artificial pancreas and improve the quality of life for many millions of people.

Recently the first clinical trial of an wirelessly controlled implantable microchip-based drug delivery device that releases the human parathyroid hormone fragment [hPTH(1–34)], used in the treatment of osteoporosis, showed that the device could deliver up to 20 adminstrations of the drug in vivo rather than the standard practice of daily injection (Farra et al. 2012).

9.7.3 Warfarin monitors

Warfarin is routinely prescribed as an anti-coagulation agent for patients with an increased tendency for thrombosis or as prophylaxis in those individuals who have already formed a blood clot (thrombus) which required treatment. However, warfarin when given in too high a dose inhibits natural clotting, leaving the patient subject to uncontrolled bleeding, both externally and as gastrointestinal bleeding. At present, patients placed on warfarin require their blood to be regularly tested to ensure the clotting ability of their blood falls within required ranges (Poller et al. 2003). This requires withdrawal of blood and testing within a laboratory environment. Since warfarin is one of the world’s most proscribed drugs, a simple home test which could monitor warfarin levels or alternatively monitor blood clotting ability would have a huge global market.