Abstract
This edition of Cephalalgia is focused on the role of placebos in clinical trials of acute treatment and prophylaxis of headaches such as migraine and cluster headache. GlaxoSmithKline (GSK) initiated the Headache Masters Programme (HMP) four years ago. This programme involves 60 young headache researchers and clinicians in the age range of 25–35. This group meets twice a year and spends 2 days on different topics of headache research. The meetings are conducted and supervised by a distinguished group of headache scientists. For one of the meetings ‘placebo’ was chosen as the main topic. The participants of the HMP formed ad hoc groups to work on certain topics such as side-effects of placebos in clinical trials, efficacy of placebo, difficulties in trial design and the ethical aspects of using placebos. The results of their team effort were discussed during a HMP meeting in October 2001 and subsequently papers were written. These papers can be found in this issue of ‘Cephalalgia’. The authors the articles and the editors of Cephalalgia express their gratitude for GSK for supporting this work.
Placebo is one of the most interesting topics in pain research. I came across this problem when we were asked to do a dose finding trial for Zolmitriptan in Germany. At that time the ethics committees of two states in Germany refused to accept a placebo group for this trial. They argued, that it was ethically not justifiable to withhold effective therapy from patients with severe pain. The compromise we finally achieved with ethics committees was to expose as few patients as possible to placebo and we had a 1 : 16 randomization rate between verum and placebo. The result was a spectacular rate of improvement of migraine headache with placebo with no difference against Zolmitriptan and sumatriptan (1,2). This result showed that patient expectation drives the placebo response and it also showed that even randomization between placebo and active drug is necessary. Another example of a powerful placebo response is the data gained in the triptan trials in children and adolescents. With one exception none of these trials showed a statistically significant difference between a triptan and placebo (3–6). This was also true when children with attacks lasting for more than 2 h were included. The response rates to the triptan, however, were comparable with those in adults. This example shows, that trial design in studies on the treatment of paediatric headache has to take into account the different placebo response and new strategies to minimize the placebo response have to be used. One possibility would be to treat consecutive attacks with a decreasing placebo response over time.
Another important aspect is the physiological mechanism by which placebo works in humans. Recent PET studies showed, that placebo activates the same pain modulating brain structures (caudal cingulum) than opioids (7). Similar studies have to be performed in headache patients. The use of placebos is now accepted as a scientific standard for clinical trials. Whether this is mandatory in diseases were mortality is the endpoint remains doubtful. In conditions like pain treatment placebo is mandatory (8). Reading over the last two decades of ‘Cephalalgia’ and ‘Headache’ it is amazing how many treatment options were effective in open trials and failed in a spectacular way in clinical trials. One example is an unpublished open trial on the use of acupuncture in migraine in Germany resulting in a 90% success rate, whereas randomised trials with sham acupuncture so far failed to show efficacy of acupuncture (9).