Whenever new research tools become available, scientists get excited about new options to explore, new inventions to make, paradigms to shift, and to kick scientific “saints” off their columns — despite knowing that we all stand on the shoulders of giants, as Robert Merton (1910-2003) has elucidated, and always will.
Such was the case when the first brain imaging paper was published showing that there is a neurobiology to the placebo effect (1) — finally proving that it is not mere imagination, as if imagined symptoms or symptom improvement would not happen in the brain — where else? The same happened after the first paper showing that there is a genetic component to the placebo response (2) — of course, genes are involved when a patient suffering from symptoms improves after having received a pill and having been promised relief or even healing.
The question is not whether, the question is how much genes are involved compared to their natural counterpart, the environment.
Genetic studies on the placebo effect/response
While the first paper investigated the placebo response in anxiety disorders (2), the second looked into depression (3) — and not surprisingly, they identified two different single nucleotide polymorphisms (SNPs) or single-nucleotide variants (SNVs) of genes regulating the central or peripheral serotonin and the monoamine signal transmission, respectively; one more involved in the processing of emotions, the other one in the processing of reward. However, both SNPs had already been identified in a number of other central conditions, ranging from stress to autism, from irritable bowel to chronic fatigue syndrome, from depression to schizophrenia, alcoholism, suicidal ideation and other psychiatric disorders, leaving it open whether these by any means are specific for the placebo response in these conditions.
Within the next few years, more SNPs in more diseases and more experimental conditions were added: genes regulating the opioid system, the endocannabinoid system, and the immune system were added, either in the same conditions (depression, anxiety), or with novel diseases and conditions (pain, Parkinson’s disease, nausea, visceral pain, etc.), so that the clear-cut initial picture (or hope only) of one SNP for all placebo responses, or at least one specific SNP for each condition, fell apart. Summary review articles such as the one by Hall (4) and Colagiuri (5) in 2015 attempted to sort these findings, and proposed a “placebome,” an assembly of 28 genes/SNPs in thus far 42 studies (6), to which more and more studies can be added but without much improvement of the concept.
Two rocky pathways ways to study genetic contributions
There is, of course, a reason for this unsatisfactory situation: Of the 2,000,000 or so existing SNPs/SNVs on the 25,000 genes of the human genome, only a few (candidate genes) are tested each time, and the ones selected for a study are usually either the ones that have a close association with the disease/condition under investigation, e.g. pain, depression etc., or have been proven successful in previous studies. Therefore, such a candidate gene approach must run into a double-trap: for one, the SNPs need to be distributed in the general population in a somewhat equal fashion, to allow their identification in a small cohort of patients and volunteers. As an example, the three variants of the serotonin transport gene (SERT or, more precisely, the variants in the promotor region of the gene), the L, S, and L/S genotypes, are present in the general population in 35, 15, and 50%, respectively, so only a reasonably-sized study population would have a chance to identify statistical deviations from this distribution.
At the same time, if a gene has been found to be associated with a regulatory mechanism in a specific condition, e.g. an endorphin-release regulating genes in pain, it becomes rather likely a priori, that the same gene may also be involved in placebo analgesia — testing it there then becomes somewhat of a self-fulfilling prophecy exercise. The alternative approach — genome-wide association studies (GWA), that tests all or at least a large number of SNPs — requires large cohorts to be included because of the requirement to adjust the significance level for multiple testing (often adjusted at very low levels, e.g. p=0.00001 and lower), that placebo research cannot fulfill (and that placebo researchers often cannot finance). But even if they could, there is no guaranty for success.
The example of functional and inflammatory bowel diseases
The search for genes involved in inflammatory (e.g. Crohn’s, Disease, CD; ulcerative colitis, UC) and functional bowel diseases (irritable bowel syndrome, IBS) may serve as an example. Because of the long clinical tradition in patient management, the relatively homogeneous nature of the CD phenotype, and well-organized patient cohorts for studies in most countries of the Western world, the hope was expressed that the rise of genomics would certainly soon deliver the genes responsible for CD onset, its chronicity, and often fatal course. This was in sharp contrast to functional intestinal diseases such as IBS, where the opposite situation (heterogeneous phenotype, diffuse disease management, poor cohort structure) raised low expectations with respect to relevant regulating genes (7).
After 10 years of genetic research with many studies in CD and a few studies in IBS — both candidate gene as well as GWAS — the outcome is surprising. In CD, not one or a few, but more than 160 SNPs or SNVs — of which about 110 are shared between CD and UC — were identified so far to be involved in the disease, with the overall contribution of genetics to the disease to be considered relatively small (8). In comparison, few GWAS but some candidate gene studies in IBS confirmed low contributions of genes towards functional intestinal diseases (9, 10). And twin studies in both diseases have resulted in rather similar estimates of the relative (low) contribution of genes and the dominant role of the environment (7).
Why would one gene (or a few genes) help to understand the placebo effect?
As outlines above, a “magic bullet” for understanding the placebo effect — a single gene — is as attractive as was searching a “placebo personality” in the past (though unsuccessful as well) or a common biological mechanisms for all and any (the great unifying theory, GUT) — fruitless tasks but not a waste of time, as many things were detected on the way to there: open label placebo interventions, sex differences, and the role of cognitive mindsets, to name a few.
When current knowledge is summarized (11) on where and how placebo effects are mediated in different diseases, with different physiological responses, and in different bodily compartments it became evident that not one biological mechanisms exists that drives the placebo response of an individual but that placebos initiate system/organ-specific physiological responses that e.g. can mimic the effects of effective drugs in the same system; there is not one but there are many placebo mechanisms in-build in the human body, and whether or not and where they are utilized depends, among others, on personalized conditions sex, age, previous medical history, biological constitutions, and the social environment (12, 13). And we can investigate in detail the specifics using different experimental techniques, social learning, associative (Pavlovian, instrumental) learning, manipulation of expectancies, but in real life they all operate simultaneously and in concordance, producing the placebo response — a single gene, or even many — is not of help.
Saving time and money: Twin studies first, instead of genetic studies
Genetic studies are costly, and often disappointing because we overestimate their prospective and belief they bring clarity into a field that assembles divergent results. Much time and money can be saved using a much older research tool that was available to assess the relative influence of genes and the environment on complex human behaviors (or simple clinical phenotypes) at times before the genome was deciphered: conventional twin research. While over the last 50 years, more than 2,500 classical twin studies have been performed across medicine (14), few have addressed pain sensitivity, and none so far placebo analgesia.
A recent study from our laboratory (15) can claim to be the first investigating the relative role of genes and shared and non-shared environment in healthy twins, 25 monozygotic (MZ) and 14 dizygotic (DZ) pairs. To account for a high genetic influence, e.g. on pain sensitivity and placebo analgesia, the correlation of a measure must be significantly higher in MZ than in DZ twins, as MZ twins share 100% of their genes, while DZ twins only have about 50% of genes in common; both MZ and DZ twins usually share 100% of their environment.
As it turns out, the elicited placebo analgesia was significant in both groups, but not different between MZ and DZ twins, indicating no role for genes, and a major role for environmental influences, shared or non-shared between individuals. And since placebo analgesia is the so far best studied placebo response phenomenon, other placebo paradigms are far less likely to produce more positive results. But this will not prevent researchers to continue searching for the holy grail in the human genome or elsewhere.
This is part 7 of a series covering “placebo” provided by Paul Enck and Sibylle Klosterhalfen from the Tübingen University Hospital. Continuous updates on placebo research can be found at www.jips.online.
References:
- Kleinschmidt A, Bruhn H, Kruger G, Merboldt KD, Stoppe G, Frahm J. Effects of sedation, stimulation, and placebo on cerebral blood oxygenation: a magnetic resonance neuroimaging study of psychotropic drug action. NMR in biomedicine. 1999;12(5):286-92.
- Furmark T, Appel L, Henningsson S, Ahs F, Faria V, Linnman C, et al. A link between serotonin-related gene polymorphisms, amygdala activity, and placebo-induced relief from social anxiety. The Journal of neuroscience. 2008;28(49):13066-74.
- Leuchter AF, McCracken JT, Hunter AM, Cook IA, Alpert JE. Monoamine oxidase a and catechol-o-methyltransferase functional polymorphisms and the placebo response in major depressive disorder. Journal of clinical psychopharmacology. 2009;29(4):372-7.
- Hall KT, Loscalzo J, Kaptchuk TJ. Genetics and the placebo effect: the placebome. Trends in molecular medicine. 2015;21(5):285-94.
- Colagiuri B, Schenk LA, Kessler MD, Dorsey SG, Colloca L. The placebo effect: From concepts to genes. Neuroscience. 2015;307:171-90.
- Wang RS, Hall KT, Giulianini F, Passow D, Kaptchuk TJ, Loscalzo J. Network analysis of the genomic basis of the placebo effect. JCI Insight. 2017;2(11).
- Goebel-Stengel M, Holtmann G, Enck P. Opportunities of twin research in gastroenterology. Nature reviews gastroenterology & hepatology. 2018; Jun;15(6):325-326
- Ek WE, D’Amato M, Halfvarson J. The history of genetics in inflammatory bowel disease. Annals of gastroenterology. 2014;27(4):294-303.
- Enck P, Aziz Q, Barbara G, Farmer AD, Fukudo S, Mayer EA, et al. Irritable bowel syndrome. Nature reviews disease primers. 2016;2:16014.
- Enck P, Azpiroz F, Boeckxstaens G, Elsenbruch S, Feinle-Bisset C, Holtmann G, et al. Functional dyspepsia. Nature reviews disease primers. 2017;3:17081.
- Schedlowski M, Enck P, Rief W, Bingel U. Neuro-Bio-Behavioral Mechanisms of Placebo and Nocebo Responses: Implications for Clinical Trials and Clinical Practice. Pharmacological reviews. 2015;67(3):697-730.
- Weimer K, Colloca L, Enck P. Age and sex as moderators of the placebo response – an evaluation of systematic reviews and meta-analyses across medicine. Gerontology. 2015;61(2):97-108.
- Weimer K, Colloca L, Enck P. Placebo effects in psychiatry: mediators and moderators. The Lancet Psychiatry. 2015;2(3):246-57.
- Polderman TJ, Benyamin B, de Leeuw CA, Sullivan PF, van Bochoven A, Visscher PM, et al. Meta-analysis of the heritability of human traits based on fifty years of twin studies. Nature genetics. 2015;47(7):702-9.
- Weimer K MN, Stengel A, Enck P. Genetics, shared, or non-shared environment? An experimental twin study on placebo analgesia. Psychosomatic medicine. 2017;79:A-120.