Multi-dataset time-reversal approach to clinical trial placebo response and the relationship to natural variability in epilepsy
Introduction from Dr Markus Reuber, editor-in-chief of Seizure
Randomised Controlled Trials (RCTs) are the gold standard method of establishing whether a treatment of interest has a therapeutic effect. Much has been learned about how to conduct such trials since the British physician John Haygarth published a study in 1799 in which he compared the effectiveness of treatment results of the highly-prized “Perkins Patent Tractor” with those of a wooden replica. The “Perkins Patent Tractor”, which had found favour with celebrities including George Washington, was a metallic spike allegedly capable of drawing pathology (including epileptic seizures) from the body. John Haygarth came to the conclusion that treatment with this instrument was no more effective than treatment with a sham device (1).
One key component of RCTs which has attracted considerable attention recently is the placebo response. The placebo response rate describes the degree of improvement seen during the treatment phase of an RCT in the placebo group compared to the baseline observation period (before commencement of study drug or placebo). In modern medication trials, the placebo arm is meant to differentiate between a drug’s pharmacological properties and other non-specific factors such as those related to the interaction between the clinician and the trial participant, attentional and reporting bias, spontaneous remission or natural variability. Research has demonstrated that the placebo response is not “just psychological”. Like possibly all other phenomena often considered “just psychological” it has fascinating neurobiological and –physiological underpinnings. In the case of the placebo response these involve multiple brain systems and neurochemical mediators, including opioids and dopamine (2). However, most of these neurobiological placebo effects have only been observed in short term experiments and not over longer periods, such as the weeks or months over which most antiepileptic drug (AED) trials are continued.
Placebo effects are particularly important in the field of epilepsy because AEDs are the treatment of choice for most patients with this disorder, and because the licensing of new drugs relies, at least in part, on the placebo-corrected response rate (i.e. the response rate in the treatment arm minus the response rate in the placebo arm of an RCT). Interestingly, even within a single condition category such as epilepsy, the placebo response has proven surprisingly variable. Apart from being much higher in children, it has also been shown to have increased over the years and be greater in patients who have tried fewer AEDs previously (3).
My editor’s choice from the current issue of Seizure explores the “enigma of placebo effects” in a new way. Goldenholz et al. test the hypothesis that ‘reversing’ trial direction, with the baseline period as the treatment observation phase of the placebo arm of RCTs, would reveal effects of natural variability. This method should be particularly suited to differentiate between placebo effects (i.e. therapeutic benefits of receiving a pharmacologically inert substance or other sham treatment due to the psychobiological mechanisms described above) and the phenomenon of regression to the mean. In the case of epilepsy studies, regression to the mean is observed when trial participants enter a study at a time when their seizure disorder is particularly bad and then experience a gradual reduction of their seizures back to their longer term average.
In the case of the epilepsy studies they have examined, Goldenholz et al. come to the conclusion that the so-called placebo response may be attributable almost entirely to this variability, rather than the placebo effect. It will be fascinating to see what emerges when their time reversal approach is applied to larger and more diverse epilepsy study datasets – especially some of the AED studies carried out recently in lower income countries which have been associated with particularly large placebo response rates.
1) Kraemer G. The enigma of placebo effects in drug-refractory epilepsies. Epilepsia 2013, 54(Suppl. S2):13–15.
2) Wager TD, Atlas LY. The neuroscience of placebo effects: connecting context, learning and health. Nature Reviews Neuroscience 2015;16:403–418.
3) Rheims S, Perucca E, Cucherat M, Ryvlin P. Factors determining response to antiepileptic drugs in randomized controlled trials. A systematic review and meta-analysis. Epilepsia 2011;52:219–233.
4) Goldenholz D, Strashny A, Cook M, Moss R, Theodore W. A multi-dataset time-reversal approach to clinical trial placebo response and the relationship to natural variability in epilepsy. Seizure 2017