Drug development, from initial design to commercial production, requires a barrage of quality, safety and efficacy testing. Clinical trials mark one of the final hurdles on the path to authorization for use and mass manufacture, and set the parameters within which a drug should be prescribed. However these trials are largely carried out in adults and the absence of equivalent testing in children has resulted in off-label and unlicensed (OLUL) prescribing of drugs by clinicians to children.
Munir Pirmohamed from the University of Liverpool, UK, Rosalind Smyth from the Institute of Child Health, UK and colleagues carried out an investigation, funded by the National Institute for Health Research, into the incidence of adverse drug reactions (ADRs) in paediatric inpatients. Following on from this, they conducted a nested case control study to investigate whether OLUL prescriptions are associated with subsequent ADRs. Smyth and Pirmohamed told Biome more about the impact of their latest findings, both of which are published in BMC Medicine, and the issues around paediatric prescriptions.
Off-label and unlicensed prescribing is common in paediatric practice, why is this?
MP and RS In the past, when new drugs were being developed for patients the regulations did not require companies to conduct clinical trials in children, even for drugs which might be of benefit to children. Where the condition was much more common in adults, often the trials were only conducted in adult patients and the license was for use in the adult population only. This meant that if doctors wanted to use a drug, which was licensed for adults, in children, they had to use it ‘off license’. In the last 5-10 years, new regulations in the USA and Europe, require pharmaceutical companies to conduct trials of new drugs in children, of all relevant ages, if the drug might possibly have an indication in that age-group, so this problem is likely to improve.
What are the challenges of involving children in research?
RS Generally when inviting children to take part in research, the parents provide consent on behalf of the child, although older children provide assent. This increases the complexity of the consent process, although research that we have done shows that children and families are willing to be involved in research studies, even if this will not directly help the child concerned. The same research study showed that clinicians are more concerned about inviting children to become involved in research studies, as they are worried that the research process will be burdensome for the child and family. So we need to convince paediatricians about the importance of research to the clinical care of children.
To what extent are adverse drug reactions (ADRs) a problem in general?
MP and RS ADRs are a major problem for any healthcare system. Work undertaken in the US by Lazarou and colleagues suggested that ADRs were between the fourth to sixth commonest cause of death. Our data in the UK has shown that for adults, at least 8000 hospital beds in the NHS, are occupied at any one time by patients with ADRs, at a cost to the UK healthcare system of more than £1 billion per annum. The figures for bed usage in children are fortunately lower overall than in adults but nevertheless still a major burden – 2.9 percent of all admissions to children’s hospitals are due to ADRs, while over 15 percent develop ADRs while in hospital.
ADRs are also a problem for the pharmaceutical industry. Many drugs are lost early in development because of unacceptable toxicity from the drug. The commonest reasons for drugs being withdrawn during development are drug-induced liver injury and cardiovascular toxicity. Furthermore, drugs cost many hundreds of millions of dollars to develop, and if the drug is withdrawn soon after licensing, the lost revenue cannot be put back into future research and development, and so future drug development may also suffer.
Why do adverse drug reactions (ADRs) differ between children and adults?
MP and RS The function of many organs in the body, which handle drugs, such as the liver and kidney, are more immature in children. The immune system may also be less mature or more reactive in children, which may make them develop immune reactions to drugs more commonly than seen in adults. In addition, where the drug is off licence, estimation of the correct dose may be imprecise. Adults tend to have more co-morbidities and are likely to be on larger numbers of medications, both of which in general also increase the risk of adverse drug reactions.
What do you think could be done to reduce the prevalence of adverse drug reactions in children?
MP and RS Our work has raised awareness of the problems of adverse drug reactions in children in hospital. Hospital prescribers and pharmacists need to be more vigilant about identifying adverse drug reactions in children, particularly those having anaesthetics and pain relief for surgical procedures, which our study identified as being of particularly high risk.
For those drugs that are used extensively off license in children, we suggest that the companies marketing them take advantage of the incentives provided by the European legislation to undertake detailed studies to assess benefit and risk. An important issue that always needs to be considered in any patient being administered a drug, and this becomes even more relevant in children, is the use of the correct or optimal dose. Further studies are needed for many drugs to ensure that the doses used in children are correct for age and/or weight.
What is being done to minimize possible harm caused by prescribed drugs and how can future research help?
MP and RS Everybody (including patients) have a role to play in minimizing the harm caused by prescribed drugs. Almost every country in the world now has ADR reporting systems (in the UK this is called the yellow card system). It is important that ADRs are reported, including by patients. This allows regulatory authorities to monitor ADRs and take regulatory action to improve the benefit-risk ratio of drugs when appropriate.
Research into ADRs is also extremely important in order to understand mechanisms which may allow prevention through better targeting of patients (for example through the use of genetic tests), by identifying biomarkers that allow for pre-symptomatic detection of drug injury, by identifying functional groups in drugs that may then allow for safer drug design in the future, and by identifying agents that could be co-prescribed with the implicated drug to prevent injury.
What is needed to achieve drug prescriptions tailored to the needs of individual patients?
MP This is complex question without an easy answer – many of the issues required for the development of stratified or personalised medicine have been addressed in a recent report from the Academy of Medical Sciences, UK.
In short, it will be important to gain a better understanding of disease heterogeneity, which will in the future lead to a new taxonomy of disease. This stratification will allow for better drug targeting. A better understanding of individual genomic, proteomic and metabolomic factors that determine the response to a drug will also be needed. Environmental factors will need to be incorporated into the assessment of drug responses, and how these factors interact with genes, proteins and metabolites. Finally, an evidence base should be developed to show in which circumstances pursuing a personalised medical approach is cost-effective, and additionally issues around reimbursement and intellectual property associated with the development of such diagnostics will need to be resolved.
How far away are we from personalised therapies?
MP I think we should really look at this as the next step in the evolution of how we practice medicine. As with all evolutionary developments, changes are likely to be in small steps and thus will take many years, but will affect all areas of medicine. Personalised therapy is already being used in many therapeutic areas.
A few examples include; somatic genome sequencing in cancers to identify aberrant mutations with the development of targeted therapies (e.g. crizotinib for ALK-mutation positive lung cancer), genetic testing to prevent adverse drug reactions (e.g. HLA-B*57:01 genotyping before the prescription of the anti-HIV drug abacavir, to eliminate a feared hypersensitivity to the drug), and a greater understanding of the mutational spectrum in cystic fibrosis (CF) patients resulting in the drug ivacaftor, which specifically targets four percent of CF patients with the G551D mutation.
Questions from Joanna Denyer, Senior Assistant Editor for BMC Medicine and Stephanie Harriman, Deputy Medical Editor for BioMed Central.