Mécanismes sous-tendant l’effet placebopharmacomedicale.org/images/desc2016/DESC_placebo_LCarton.pdf · mechanisms underlying nocebo-induced and/or anxiety-induced hyperalgesia.

Louise Carton U1171 – Département de Pharmacologie Médicale

Service d’addictologie, Hôpital Fontan 2, CHU Lille

DESC de Pharmacologie 23 et 24 novembre 2016

Mécanismes sous-tendant l’effet placebo

+Définition

n  « placere » : plaire / « nocere » : nuire

n  Caractéristiques pharmacologiques n  Placebo « pur » : substance inerte dépourvue d’effet physiologique ou

d’action pharmacologique (lactose per os, serum phy…) n  Placebo « impur » : substance ayant possiblement une action

pharmacologique mais dont les propriétés n expliquent pas l’effet obtenu (vitamine…)

Écart positif constaté entre le résultat thérapeutique observé

et l’effet thérapeutique prévisible en fonction des données strictes de la pharmacologie

+ Les attitudes devant l’effet placebo

WG. Thompson. Am J Gastroenterol 2000

several contemporary nostrums in the cure of scurvy (12). In1801, Haygarth compared the “electromagnetism” of metalrods known as “Perkins tractors” with wooden placebos inrelieving symptoms in five patients (5). He demonstrated nodifference. (No mention here of a type II error!) Bingel, 100years ago, compared diphtheria antitoxin and plain horseserum in the treatment of 937 patients (5). Although he didnot call the horse serum a placebo, that was its function andhe demonstrated no difference. Comparative trials of drugswere in use in the early 1930s (10). The word “placebo” wasapparently used in its modern connotation in 1938 to de-scribe inert controls in a therapeutic trial of cold vaccines(11).

Since World War II, placebo-controlled trials of therapieshave become the preferred, indeed the only acceptablemethod of validating therapy (6). No pharmaceutical firmcan dream of marketing a new drug without such validation.However, in our passion to understand therapeutic effects,we have neglected to study the effects of placebos them-selves.

ATTITUDES (Table 1)

In clinical trials of IBS (13) and dyspepsia (14, 15) theplacebo response is 20–70% with a mean of about 45%.Those conducting such trials may consider this phenomenonto be an obstacle that obscures the true value of a treatmentunder study. Philosophers cannot define placebos: “A pla-cebo is inert. How can an inert substance have an effect?”(16). Surely “placebo” is an oxymoron! Scientists despair toexplain them, ethicists struggle with them, charlatans thriveon them, and health administrators attempt to run programsin ignorance of them. Yet placebos exist, and understandingthem is important to modern medicine. Wise doctors knowthem as a factor in every treatment and an essential part oftheir daily work.

RESPONSE TO TREATMENTS

Contemporary physicians can be forgiven if they associateplacebos only with clinical trials. Indeed it is these trials thathave made the word respectable. It has even been suggestedthat we can no longer speak of a placebo effect in anindividual patient (16). Physicians and surgeons have ex-ploited placebo effects for centuries, and the ethics of theiruse have been long debated (6). However, proof of their

existence awaited the general adoption of placebo-con-trolled trials in the 1930s and 1940s.

There are three components to a therapeutic effect: theeffect of the drug or other treatment being tested, the effectdue to the natural history of the disease, and the placeboeffect. The effect of the treatment is self evident—thatpresumed to be scientifically based on the physiological,pharmacological, or even psychological rationale that led toits employment.

The natural history of a disease may be downhill as interminal cancer, static as in hypertension, or unpredictablyfluctuating as in the irritable bowel syndrome (IBS). In IBStrials, the patients are all symptomatic, or they would not beincluded. Since the disease normally fluctuates in intensity,we can expect that many or most of the subjects willimprove during the course of the trial, a regression towardthe mean (17). This syndrome is even more complicated inthat it is made up of several fluctuating or alternating symp-toms such as loose and hard stools, frequent and infrequentstools, urgency and straining (18, 19). Thus, in clinicaltrials, “improvement” will depend largely upon the endpointselected by the investigators.

The placebo effect is often confused with the naturalhistory of the disease being treated. In a clinical trial, it maybe defined in the treatment arm as the total effect, minus theeffect of the drug, minus the natural history of the disease.In the control arm (often erroneously called the “placeboarm”) of a trial, it is the total effect minus the natural historyof the disease (16). Few trials include an “untreated arm,”and such a control would be different to “blind.” Therefore,we cannot know in a trial how much benefit (or harm) is dueto the placebo and how much to the natural tendency ofmany chronic conditions to change. In a static conditionsuch as hypertension, most nondrug benefit is likely due toplacebo, but also remember “regression towards the mean”(17). In terminal cancer, the nondrug effect is likely to bemostly negative and due to the natural history of the disease.In a fluctuating condition such as IBS, even with untreatedcontrols, the source of the nondrug effect is hard to identify.Nevertheless, the tendency of chronic, functional conditionsto improve is powerful, leading previous generations tobelieve that doctors were successful even if we know thatthe treatments they employed were useless. Said Voltaire, agood “physician is the man who successfully amuses hispatients while nature effects a cure.”

FACTS ABOUT PLACEBOS

We think we know a lot about placebos, but we don’t. Thefollowing “facts” are arresting but incomplete. Perhapssome contain the seeds of our ultimate understanding.

A placebo injection is more powerful than a pill, and alarge pill is more effective than a small pill (20, 21). Pinkpills improve mood, and blue ones depress (22). A placeboadministered by a doctor seems to be more powerful thanone given by a nurse or a secretary. Mailed placebos are

Table 1. Attitudes Toward Placebos

Scientists “incredible”Trialists “obstacle”Philosophers “undefinable”Ethicists “culpable”Charlatans “convertible” ($)Politicians, administrators “incomprehensible”Psychologists “inevitable”Physicians “indispensable”

1638 Thompson AJG – Vol. 95, No. 7, 2000

+ Part réelle du placebo dans l’effet observé ?

+Facteurs déterminants la réponse

MALADIE •  Aspect psychosomatique

•  Réponse d’autant plus grande que souffrance importante

•  Plus important dans les pathologies d’évolution

fluctuante •  Des effets pas seulement sur

des critères subjectifs +++

MALADE

•  Pas de profil type +++ •  Impact du contexte

culturel •  Fluctuation possible au

cours du temps •  Impact de la compliance

Wager & Atlas, Nat. Rev. Neurosc. 2015 ; Waber et al, JAMA, 2008 ; Thompson, Am. J. Gastroenterol., 2000

+Facteurs déterminant la réponse

Wager & Atlas, Nat. Rev. Neurosc 2015

PRODUIT PROCEDURE Nom, couleur, taille, goût,

voie et modalité d’administration, prix,

information ++

+Des effets sur des critères objectifs

Action de l’effet placebo sur les fonctions du SNA

Contraction gastrique, Nausées,

Mal des transports, Motilité intestinale

Tension artérielle, Syncope vaso-vagale,

Flux sanguin coronarien Meissner, Philos. Trans. R. Soc. Lond., 2008

+Quels mécanismes discuter ?

Benedetti et al., Neuropsychopharmacology, 2011

Diverses voies mises en jeu selon le contexte pathologique


Benedetti et al. Neuropsychopharmacology 2011


+Implication de variations génétiques

n Anxiété sociale : -  Diminution activité amygdale liée à : à Polymorphisme d’un transporteur sérotonine à Polymorphisme de la tryptophane hydroxylase-2

Furmark et al, J. Neurosci. 2008

n Dépression et placebo : - Lien entre réponse au placebo et à Polymorphisme de la COMT à Polymorphisme de la MAO-A

Leuchter et al, J Clin Psychopharmacol, 2009


Benedetti et al. Neuropsychopharmacology. 2011


+Effet placebo et attente

n Prépare l’organisme à un événement afin de mieux l’appréhender

n A l’origine de réponse physiologique et émotionnelle

n Liée à la mémoire et à la motivation

+Modulation de l’anxiété par les attentes du sujet

n  Interaction entre anxiété et voie de la douleur

n  Effet nocebo pouvant être à l’origine d’une allodynie et hyperalgésie

Proglumide Antagoniste des

récepteurs CCK-1 et CCK-2 Hyperalgésie induite

par l’anxiété Hyperactivité de l’axe

hypothalamo-hypophysaire Diazepam

Benedetti et al, J Neurosci, 2006

-

-

+Effet placebo/nocebo et douleur

Benedetti et al., Neuroscience, 2007

cating an anti-CCK action in the CNS at the level of affec-tive mechanisms (Harro et al., 1990; Harro and Vasar,1991; van Megen et al., 1994).

The antagonist action of CCK on endogenous opioids(Benedetti, 1997) is particularly interesting in light of theopposing effects of placebos and nocebos. In fact, todaythere is general agreement that placebo analgesia is me-diated by endogenous opioids, specifically the mu-opioidreceptors (Zubieta et al., 2005), at least in some circum-stances (Benedetti et al., 2005; Colloca and Benedetti,2005). Therefore, the findings on the involvement of CCKin nocebo hyperalgesia suggest that the opioidergic andthe CCKergic systems may be activated by oppositeexpectations of either analgesia or hyperalgesia, re-spectively. In other words, as shown in Fig. 3, verbalsuggestions of a positive outcome (pain decrease) acti-vate endogenous mu-opioid neurotransmission, whilesuggestions of a negative outcome (pain increase) activateCCK-A and/or CCK-B receptors. This neurochemical viewof the placebo–nocebo phenomenon, in which two oppo-site systems are activated by opposite expectations aboutpain, is in keeping with the opposite action of opioids andCCK in other studies (Benedetti, 1997; Hebb et al., 2005).

The involvement of CCK in both pain modulation andanxiety is particularly relevant to the nocebo effect. It isworth noting that some CCK-B receptor antagonists, likeL-365,260, have a benzodiazepine-based chemical struc-ture that is similar to the anxiolytic drug diazepam, whichsuggests a similarity of action of CCK-antagonists and

anti-anxiety drugs. However, it should be stressed that thestudy by Benedetti et al. (2006) suggests that nocebosuggestions activate two different and independent bio-chemical pathways, one blocked by proglumide and theother by diazepam (Fig. 2).

On the basis of all these considerations and the in-volvement of CCKergic systems in pain and anxiety mech-anisms, nocebo hyperalgesia represents an interestingmodel to better understand when and how the endogenouspro-nociceptive systems are activated. In the case of CCK,besides the studies described above, the pro-nociceptiveand anti-opioid action of this neuropeptide has been doc-umented more recently in the brainstem. For example, ithas been shown that CCK is capable of reversing opioidanalgesia by acting at the level of the rostral ventromedialmedulla, a region that plays a key role in pain modulation(Mitchell et al., 1998; Heinricher et al., 2001). It has alsobeen shown that CCK activates pain facilitating neuronswithin the rostral ventromedial medulla (Heinricher andNeubert, 2004). The similarity of the pain facilitating ac-tion of CCK on brainstem neurons on the one hand andon nocebo mechanisms on the other hand, can stimulateand guide further research into the neurochemicalmechanisms underlying nocebo-induced and/or anxiety-induced hyperalgesia.

It is also worth noting that CCK has been found to playa role in placebo analgesia. In fact, the CCK-antagonistproglumide has been found to potentiate placebo-inducedanalgesia, an effect that is probably due to the blockade of

Fig. 3. Placebo and nocebo modulation of pain. Whereas placebo suggestions activate mu-opioid neurotransmission which inhibits pain, nocebosuggestions induce anxiety which activates CCK-A and/or CCK-B receptors that, in turn, enhance pain.

F. Benedetti et al. / Neuroscience 147 (2007) 260–271 265

+ Modulation du système de récompense

•  Activation dopaminergique (D2/D3)

•  Corrélation avec Effet antalgique du placebo Activation récepteur µ opioïdes

Scott et al. Arch Gen Psychiatry, 2008


Benedetti et al. Neuropsychopharmacology 2011


+Les effets de l’apprentissage

n Conditionnement pavlovien : histamine et rhinite allergique

n Renforcement des attentes : « anesthésie » locale répétée

n Apprentissage social : transmission sociale de l’effet…

Benedetti et al.,Neuropsychopharmacology, 2011

+Conclusion

•  Effet placebo est bref ?

•  Personnes placebo sensibles ?

•  Toute personne peut à un moment donné être placebo sensible?

•  Prescription placebo est anodine ?

•  Placebo permet de voir la réalité d’un symptôme ?

•  Signes objectifs inaccessibles au placebo ?

•  L’effet placebo peut être relayé par des mécanismes neurobiologiques ?

Mécanismes sous-tendant l’effet placebopharmacomedicale.org/images/desc2016/DESC_placebo_LCarton.pdf · mechanisms underlying nocebo-induced and/or anxiety-induced hyperalgesia.

Documents