Brain, Vol. 126, No. 7, 1716-1717,
July 2003
© 2003 Guarantors of Brain
doi: 10.1093/brain/awg181
Book Review |
CLINICAL TRIALS IN NEUROLOGY
Department of Clinical Neurosciences, Western General Hospital, Edinburgh, UK
CLINICAL TRIALS IN NEUROLOGY
By Roberto J. Guiloff
2001. Heidelberg: L. Springer-Verlag
Price £135. ISBN 1-85233-239-5.
As a clinical trials aficionado, I was intrigued to see what this multi-author book, edited by Roberto Guiloff, would turn out to be. This list of contributors, drawn from both sides of the Atlantic, certainly includes some very distinguished names. The task set for this group of authors was to ‘comprehensively tackle the methodology and design of clinical trials in neurological disease’. I am not sure that target has necessarily been met, but there are certainly many contributions that stand out: Ethical considerations (Hope and Savulescu); Ethics, Outcome Variables and Clinical Scales (Buskens and van Gijn); The Role of Meta-Analyses (Stewart); Cerebrovascular Disease (Rothwell); Epilepsy (Chadwick and Marson); and Headache (Tfelt-Hanson and Seidelin), to name but a few.
Perhaps not surprisingly, much of the book is devoted to randomized trials of drugs, with particular emphasis on trials sponsored by the pharmaceutical industry. This bias towards industry-sponsored trials coloured both the first half of the book, which deals with general methodological considerations, and the second half of the book, which devotes itself to trials in specific diseases. On the other hand, this is a reality; pharmaceutically sponsored research has been the dominant force, so it is hardly surprising that the ‘industry perspective’ dominates the chapters on design. The chapters on drug regulatory requirements and quality control were viewed exclusively from the perspective of industry-sponsored trials. Quality control is important for trials run independently of industry but has to operate in a different framework with different priorities and methods. Furthermore, a tidal wave of regulatory bureaucracy will hit the clinical trials community when the EU Directive on Clinical Trials comes into force in 2004. This will force an ‘industry-type’ requirement for monitoring on clinical trials. The requirements are complex, burdensome, and for trials of non-commercial drugs and non-pharmacological interventions, will make even the most resolute independent academic trialist quake at the prospect of trying to get an international multi-centre trial off the ground. Some seem to have given up the fight already: ‘The requirements of most trials for large numbers of patients, sophisticated laboratory monitoring and extensive data collection from regulators are beyond the financial means of most academic departments (and increasingly of Government research budgets). Therefore, sponsorship by pharmaceutical companies is necessary’. I would vigorously dispute this. Data collection does not have to be exhaustive, laboratory monitoring may not be necessary and large-scale clinical trials can be conducted without pharmaceutical industry sponsorship.
On the other hand, if our patients with neurological diseases are to benefit from the fruits of new drug developments, neurologists will need to continue to work with the drug industry to test new products. Tfelt-Hanson gives useful advice in the form of a checklist for investigators considering whether or not to participate in a multi-centre randomized controlled trial (page 295). He concludes by recommending strongly ‘that before agreeing to participate in a randomized trial, investigators should insist on a clearly stated publication policy. This should, as a minimum, define how the Publications Committee will be formed and its responsibilities. Furthermore, it should be stated that the trial, whatever the results, will be published in a peer review journal within a reasonable time limit’.
As a methodologist, I was surprised that some of the specific chapters did not cite some of the well-known problems that have occurred with disease-specific trials. For example, the trial of selegiline in Parkinson’s disease, which appeared to show an excess risk of death amongst those allocated to the drug. It later became clear that this was almost certainly not due to a biological effect of the drug, but merely to the play of chance compounded by the fact that the trial was not powered reliably to detect effects on deaths. Another example is the problem that has occurred in the trials of disease-modifying treatments for multiple sclerosis such as beta interferon in which loss to follow-up and drop-outs made true ‘intention-to-treat’ analyses impossible. Indeed, the chapter devoted to intention-to-treat analyses seemed—rather misleadingly in my opinion—to suggest that loss to follow-up was acceptable and could be handled in analyses, and even that significant numbers of post-randomisation exclusions were acceptable. It has been pointed out recently that, whether a trial is designed as ‘pragmatic’ or ‘explanatory’, complete follow-up of all subjects is essential, whether or not they comply with study medication and whether or not they fully meet trial eligibility criteria. It would have been helpful to see this spelled out more clearly. All study subjects must be accounted for at the end of the trial. This has certainly not been the case in many trials in neurological disease.
These remarks may seem unduly critical. Perhaps this is because many of the randomized trials of treatments for neurological diseases have had methodological failings, and this book has documented at least the worst of them. We need more trials, better trials, bigger trials, and trials of interventions other than drugs. For anybody planning trials in a neurological disorder, this book would be a good starting point. It would probably be of greater benefit to somebody from a pharmaceutical background rather than an independent academic trialist, but both types of readers will find relevant background information.
Finally, trialists are beginning to appreciate the value of involving consumers in all stages of clinical trials, from the very earliest stages of design through to public presentation. Greater involvement of patients with neurological disorders might well lead to more appropriate clinical trial designs. I am sure that patients with axonal peripheral neuropathy would prefer their potential treatments assessed by their effects on symptoms rather than nerve conduction studies of the sort recommended by the ‘Peripheral Nerve Society’.
Overall, this book is a testament to the efforts of neuroscientists and clinical neurologists to identify and evaluate treatments to reduce the distressing and disabling effects of neurological diseases. As Chadwick says, ‘There can be no doubt that the quality of studies has improved considerably over the last three decades. In spite of this, the number of high quality randomized controlled trials in epilepsy remains small, the quality of many questionable and the size of most inadequate’. So, this book is a testament to the efforts of the past, but sets a substantial challenge for us to do better in the future.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||