Laurence A Goldberg
The main theme of the EAHP Congress was the challenges for hospital pharmacy arising from the introduction of new therapies. Advances in genomics, drug delivery systems and preparation methods are now reaching day-to-day practice.
Childhood acute lymphoblastic leukaemia
The cure rate for childhood acute lymphoblastic leukaemia (ALL) is now more than 90% but treatment still relies on toxic drugs, according to William Evans (Professor of Pharmacy and Paediatrics and Chief Executive Officer, St Jude Children’s Research Hospital, Memphis, Tennessee, USA). Describing how the human genome project will influence hospital pharmacy over the next 10-20 years, he detailed the recent progress in ALL treatment.
For decades, drug treatment was entirely empirical – standard treatment was offered to all patients and some experienced toxicity, some had no response and a middle group responded satisfactorily. Adverse reactions had become the fourth leading cause of hospital admissions and the fifth leading cause of mortality in the USA. Using ALL as an example, Professor Evans pointed out that early treatments had a cure rate of 9% but that by the early 1970s this had risen to 36%. The main ï¿½problem was the variable pharmacokinetics of the drugs, he suggested. Methotrexate, 6-mercaptopurine (6-MP) and cytosine arabinoside respectively showed 5-, 30- and 10-fold variations in drugs exposure from the same doses. It was soon realised that fast drug clearance was associated with early relapse. When doses were adjusted to compensate, better cure rates were obtained.
By 1998, the cure rate for ALL was 80%; treatment “failures” were the result of toxicity or relapse. Turning to the question of whether genetic variations could account for these treatment failures, Professor Evans explained that single nucleotide polymorphisms (SNPs) could cause differences in drug metabolism or disposition and also differences in drug targets, such as enzymes or receptors. So far polymorphisms have been identified for more than 25 drug-metabolising enzymes, seven transporters and 25 drug targets.
One example is thiopurine methyl transferase (TPMT), the enzyme responsible for the inactivation of 6-MP. The absence of TPMT is associated with increased toxicity of 6-MP. Almost 90% of the population has high TPMT activity but about 1% has very low enzyme activity. Those with little active enzyme develop severe bone-marrow toxicity with 6-MP. Just three SNPs are responsible for this variation and a diagnostic test has been developed to identify them. Now every child with ALL is genotyped routinely; 10% of the normal dose of 6-MP is given to TPMT-deficient patients and 50% to the heterozygotes. This dosage reduction does not influence the relapse rate, he noted.
All drug-metabolising enzymes are affected by SNPs but they do not all result in increased toxicity. For example, CYP2D6 converts codeine to its active metabolite, morphine. About 10% of the population is CYP2D6-deficient and cannot respond to codeine, but this is not routinely tested, said Professor Evans.
Drug receptors are also susceptible to genetic variation. SNPs in adrenergic receptors account for variations in response to beta-agonists. However, it is important to remember that drug effects are usually polygenic and it is the overall result that is relevant.
The inclusion of genetic information on labels is now in its early stages; for example, variations in drug-metabolising genes are now mentioned on the labels for 6-MP and irinotecan.
Researchers are now concerned with finding other candidate genes such as those which influence the pharmacokinetics or pharmacodynamics of the drugs used in ALL treatment. Work in Professor Evans’ hospital has found 16 different SNPs in 13 candidate genes involved in ALL therapy and they are now beginning to define the diagnostic array that would be helpful in childhood leukaemia. The advantage of genotyping is that it only has to be done once, unlike blood tests, said Professor Evans. We will no longer simply group together patients who share a textbook diagnosis âˆ’ we will be able to subdivide them according to response and drug toxicity profiles, he predicted.
Human testing of pharmaceuticals
The results of the first-in-man study of TGN1412 call into doubt the relevance of traditional methods of testing therapeutic agents for human use, said Nirmala Bhogal (Science Manager, Fund for the Replacement of Animals in Medical Experiments, Nottingham, UK). The drug known as TGN1412 was a humanised, mouse, anti-human CD28 antibody. It binds to and “switches on” the target protein CD28 on the surface of T-cells and triggers cell activation without the need for the normal T-cell receptor occupation. It was developed as a treatment for rheumatoid arthritis and B-cell chronic lymphocytic leukaemia (B-CLL). It was known to cause expansion and activation of human T-cells, leading to cytokine release and indirectly increasing the numbers of B-cells.
Several aspects of the study design could have contributed to events, explained Dr Bhogal. TGN1412 is a fully humanised antibody and this increases the likelihood of immunogenicity. The method of calculating the starting dose was another point. TGN1412 had been tested in two species of macaque monkeys, both of which have CD28 sites similar to humans. A surrogate molecule (JJ316) had also been tested in rats. In rats, JJ316 had caused fast T-cell expansion and given a “no observed adverse effect level” (NOAEL) of less than 0.3â€‰mg/kg, whereas in monkeys TGN1412 had caused slow T-cell expansion with a NOAEL of 50â€‰mg/kg. Starting doses can be 10% of pharmacologically active dose, 1% of the NOAEL or a fraction of the minimum anticipated biological effect level (MABEL), adjusted for receptor affinity. In this case the MABEL was 0.3â€‰mg/kg and this gave a calculated starting dose of 0.005â€‰mg/kg. The dose that was actually used was 0.1â€‰mg/kg. MABEL is the preferred method where species differences are large or where NOAELs are difficult to define, noted Dr Bhogal. Another point was that in macaque monkeys, it took two hours before the first reactions were seen. This should have prompted investigators to leave a gap of at least two hours between dosing of individual volunteers, she said.
Numerous other aspects of the study had been considered and investigated and the results of deliberations are important, especially for new bioâ€‘technology drugs that stimulate the immune system. In future, a flexible but cautious approach should be adopted, recommended Dr Bhogal.
Nonprofit drug development has a bright future, according to Merrill Goozner (Director, Integrity in Science Project, Center for Science in the Public Interest, Washington DC, USA). Investment in R&D by the pharmaceutical industry has risen steadily in recent years but there have been no corresponding gains in public health. A number of factors have contributed to this situation. Tight controls on reimbursement make it difficult for companies to recoup the money invested in drug development. Moreover, the small-molecule drug revolution is now mature – treatments for high blood pressure, aches and pains, allergies and common infections have been ï¿½developed. “The low-hanging fruits of medicinal chemistry have been picked,” said Mr Goozner. Today’s innovators need to discover cures for ï¿½diseases like diabetes, cancer, dementia and rheumatoid arthritis, he added.
Another reason for declining productivity is that the industry ignores areas where there is great need, for example the neglected diseases of the developing world – tuberculosis, malaria, leishmaniasis, hookworm and Chaga’s disease. Although there are many potential customers, they do not have the money to pay for drug development. The result has been that, in recent years, the pharmaceutical industry has spent money developing new products for its existing market. Esomeprazole and the COX-2 inhibitors were two examples of this type of development.
In order to break out of this situation a number of steps will be needed, said Mr Goozner:
- Governments must increase investment in basic and applied research.
- Incentives for innovation need to be restructured (for example, by extending patent life).
- Nonprofit drug development and generic manufacturing will need to flourish.
Mr Goozner also recommended targeted research campaigns such as the HIV/AIDS campaign funded by the US Government that yielded such spectacular results. Without effective restructuring, new treatments threaten to bankrupt all our systems, he concluded.
The treatment of nonresectable hepatocellular ï¿½carcinoma can be made safer and less painful through the use of doxorubicin-loaded drug-eluting beads, explained Frank Pilleul (Professor of Radiology, Edouard Herriot Hospital, Lyon, France). Until recently, transcatheter arterial chemoâ€‘therapeutic embolisation (TACE) has been the treatment of choice. This involves injection of a suitable chemotherapeutic agent (usually doxorubicin 50â€‰mg) mixed with Lipiodol (an X-ray contrast medium) into the branch of the hepatic artery that feeds the tumour, followed by injection of an embolising agent such as gelatin sponge.
Although effective, the high viscosity of the Lipiodol and cytostatic mixture makes it difficult and painful to inject and a significant amount of cytostatic agent reaches the systemic circulation.
When drug-eluting beads (DEBs) loaded with doxorubicin are injected intra-arterially they become trapped in the tumour vasculature. The drug leaches out of the beads and into the surrounding tissue ï¿½during the next 7-14 days. These provide simultaneous embolisation and local delivery of chemotherapy. The technique is easier and safer to use than conventional TACE, causes little or no pain and is reproducible. In addition, survival is improved and there are fewer complications. This is probably associated with the prolonged high tumour concentrations of doxorubicin combined with negligible systemic exposure to the drug.
The preparation of chemotherapy-loaded beads is a cytostatic drug preparation and as such should be undertaken in a pharmacy by well-trained personnel, Irene Kramer (Pharmacy Director, Johannes Gutenberg University Hospital, Mainz, Germany) told the audience. The beads are polyvinyl alcohol (PVA) hydrogel microspheres. When mixed with a solution of a cationic drug, such as doxorubicin, the drug is taken up on to the surface of the beads by an ion exchange process.
The preparation of doxorubicin-loaded beads using dry powder vials involves nine manipulation steps and a waiting time of 20-100 minutes, whereas preparation from doxorubicin liquid requires eight steps but a waiting time of 8-11 hours. A stable, preloaded doxorubicin bead injection is now available in some countries and this can be ready for use in about six minutes.
If DEBs are prepared outside the pharmacy, then the importance of the waiting time may not be understood and the injection may be administered before loading is complete, cautioned Professor Kramer.
Automated injection preparation
The next generation of pharmacy automated systems is likely to feature injection preparation, according to Michael Cross (Chief Pharmacist, Bart’s and the London NHS Trust, London, UK), speaking at a satellite symposium sponsored by ARX.
Intravenous treatment is associated with a high error rate. The ideal approach would be a ready-to-use product that could be checked against the prescription, the patient and the administration pump. However, centralised preparation of large numbers of injections is associated with a high frequency of repetitive strain injury and upper limb disorders, he explained.
The Intellifil(R) is a robot that can prepare large numbers of injections in a sterile environment. So far it has been installed in 21 hospitals across the USA. Describing the robot, Joel Osborne (President, ForHealth Technologies Inc, Daytona Beach, Florida, USA) described how the Intellifil was designed on a “start sterile – stay sterile” principle.
Doses prepared in the automation deck are not touched by human hands during the compounding process and the compounding area bioburden is virtually eliminated by locating the IntelliFill IV automation deck within an enclosed, ISO 4 (Class 10) cabinet.
Another device, the Cyto-Formulating Machine (CFM), was designed to prepare one-off injectable doses. It is particularly suitable for preparation of cytotoxics, antibiotics, radiopharmaceuticals and gene therapy products because it can be sited inside an isolator and can safely be disinfected using peracetic acid, explained Sander van Vreeland (Chief Executive, Medical Dispensing Systems, Enschede, Netherlands). The device bridges the gap between manual and high-volume automated production, he added.