Clinical Research: Phase 1 - Phase 4

The Placebo Problem, Part 4: Neurobiological Mechanisms

This is the fourth in our series examining the increasingly high placebo response issues plaguing analgesia and psychiatry clinical trials. Additional posts in the series are located here.

As with many other aspects of placebo research, the majority of research on underlying neurobiological mechanisms has focused on placebo analgesia. In fact, roughly 40 positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) studies over the last 15 years have shed light on the specific central nervous system underpinnings of pain improvement in response to placebo.

Reduced nociception

These functional neuroimaging studies have implicated a number of brain areas in placebo analgesia. Placebo treatment reduces activity in a number of “classical” pain-related brain regions that receive input from spinal nociceptive pathways. Regions displaying diminished activity include primary and secondary somatosensory cortices, thalamus, and insula, and anterior cingulate cortex (ACC). These reductions are evident not only during the anticipation and later reporting of pain, but also during the administration of the painful stimulus itself. A spinal cord fMRI study showed that pain-related activity in the dorsal horn is also reduced by placebo, suggesting that the reduction in nociception-related activity is not limited to the brain.

The descending pain modulatory system, a network of brain regions that modulates analgesia through top-down control, has also been heavily implicated in placebo analgesia. Neuroimaging studies have demonstrated increased functional coupling of several brain areas in this network, including dorsolateral prefrontal cortex (DLPFC), ACC, and periaqueductal grey (PAG), during placebo analgesia. Placebo-induced increases in PAG activity are correlated with the strength of the pain relief provided.

It is important to note that many of these pain-related brain regions are also involved in mediating other placebo-related processes such as perception, decision-making, and emotion. For example, the DLPFC is thought to maintain and update the expectancies that are an important component of placebo analgesia. It is therefore difficult to definitively say which aspects of the placebo response these brain changes sub-serve. In addition, some brain areas such as the ACC and PAG show opposite effects across studies and with respect to the timing of the imaging relative to the noxious stimulus, findings that underscore the complexity of the placebo response and the challenges of studying it.

Endogenous opioids…

A wealth of pharmacological and neuroimaging studies has characterized the role of the endogenous opioid system in placebo analgesia. The first experimental evidence of opioid involvement occurred nearly 40 years ago. In a landmark 1978 study that has since been replicated numerous times, placebo analgesia was used to treat postoperative dental pain, and this effect was reversed by administration of the opioid antagonist naloxone. Endogenous opioids also mediate other effects of analgesia outside of the pain reduction itself, such as a slowing of respiration and heart rate, as these symptoms can also be blocked with naloxone.

Molecular imaging studies suggest that a specific receptor, the mu opioid receptor, mediates many of these effects. For example, placebo analgesia is associated with increased mu opioid receptor activity in a number of brain areas including ACC and PAG, and the strength of his activity is directly related to the size of the placebo effect. The increased PAG activity in response to placebo analgesia can also be blocked by naloxone. Opioid mechanisms of placebo analgesia have been linked to learning-related placebo responses, and are thought to underlie both expectancy and conditioned mechanisms.

…and beyond

Although many studies on the chemical mediators of placebo analgesia have focused on the endogenous opioid system, other neurotransmitter systems have also been implicated. PET studies of dopamine neurotransmission demonstrate that placebo analgesia is associated with dopaminergic release in the basal ganglia, particularly in the nucleus accumbens. This higher dopamine release is associated with greater analgesia, expectation of relief, and perceived effectiveness of the placebo. In fact, dopamine release accounted for a quarter of the variance in placebo analgesia in one study.

More recently, the endocannabinoid system, previously linked to analgesia, reward, and reinforcement, has also been implicated in placebo analgesia. Placebo analgesia was observed following tourniquet pain when participants were preconditioned with a nonsteroidal anti-inflammatory drug, and this effect was reversed after administration of a cannabinoid 1 receptor antagonist. The antagonist had no effect on placebo analgesia preconditioned with morphine, indicating that endocannabinoids function separately from endogenous opioids in mediating placebo effects.

Many effects = many mechanisms

Mirroring the trend in the literature, we’ve focused on the neurobiology of placebo analgesia. We now turn briefly to the findings from the relatively sparse literature examining placebo effects in two other conditions: Parkinson’s disease (PD) and depression.

In PD, the placebo effect can improve specific symptoms such as rigidity as well as decreased overall disability, in some cases in as many as 40 percent of patients. In a PET study, PD patients who responded to placebo showed increased dopamine release in the striatum (a key area of dopamine loss in the disease) compared to non-responders, a finding that was correlated with patients’ therapeutic expectations.

In a PET study of depression, brain activity changes in placebo responders – cortical increases and limbic-paralimbic decreases – mimicked those of subjects receiving a selective serotonin reuptake inhibitor (SSRI). Interestingly, despite similar clinical improvement between the two groups, SSRI responders displayed additional activity changes in subcortical and limbic regions not present in placebo responders.

While more research on this topic is needed, in general it seems that, unsurprisingly, the varied placebo responses observed in different diseases work through disease-specific mechanisms. It will be interesting to see if this trend holds true in other conditions such as anxiety and multiple sclerosis that also have high placebo responses.

Join us next week for the next installment of the Placebo Problem series, where we’ll revisit the endogenous opioid, dopamine, and endocannabinoid neurotransmitter systems as we examine new research on genetic variants associated with the placebo response.

If you can’t wait for next week, and want more information, you can always sign up for our latest industry insights. Psssst, part five, all about the placebome is already available here.