Y PLACEBO EFFECTS P O & TRANSCRANIAL MAGNETIC C STIMULATION - PowerPoint PPT Presentation

Y PLACEBO EFFECTS P O & TRANSCRANIAL MAGNETIC C STIMULATION T INTENSIVE COURSE IN TRANSCRANIAL MAGNETIC STIMULATION O N O D E S A MATTHEW BURKE, MD FRCPC E COGNITIVE NEUROLOGIST, NEUROPSYCHIATRY PROGRAM L DIRECTOR, TRAUMATIC BRAIN INJURY CLINIC P ASSOCIATE SCIENTIST, HURVITZ BRAIN SCIENCES PROGRAM SUNNYBROOK HEALTH SCIENCES CENTRE ASSISTANT PROFESSOR, UNIVERSITY OF TORONTO

Y P DISCLOSURES O C None T O N O D E S A E L P

Y P CONTEXT O C T O N O D E S A E L P Dr. Alvaro Pascual-Leone Dr. Ted Kaptchuk

Y P PROGRAM IN PLACEBO STUDIES O C T O N O D E S A E L P

Y P OUTLINE O C 1. Neurobiology of Placebo Effects 1. T O ▪ Definitions ▪ Mechanisms of action N Evidence and theories ▪ O 2. 2. “Differential” Placebo Effects D ▪ Historical context E ▪ Meta-analytic approaches S Prospective approaches ▪ A 3. TMS and Placebo Effects 3. E ▪ Sham devices L ▪ Quantifying magnitude P ▪ Implications on clinical trial results

Y P 1 NEUROBIOLOGY OF PLACEBO EFFECTS O C T O N O D E S A E L P

Y P PLACEBO EFFECTS O C T O N O D E S A E L P

Y P PLACEBO TERMINOLOGY O C T  Placebo “Response” vs. Placebo “Effect s ” O N  The latter requires a comparison to “no - treatment” O controls to delineate placebo effects from other D nonspecific changes: ▪ Regression to the mean E ▪ Spontaneous changes S ▪ Elevation bias (higher reported symptom severity at A initial/baseline assessment than actually experienced) E ▪ Hawthorne effects (changes in outcomes associated with the L act of being studied/observed) P

Y P NEUROIMAGING STUDIES O C T O N O D E S A E L P

Y P CURRENT META-ANALYSES AND MODELS O C T O N O D E S A E L P Wager and Atlas 2015, Ashar et al 2017

Y P NEUROPHARMACOLOGICAL STUDIES O C T O N O D E S A E L P

Y P BIOLOGICAL MECHANISMS O C T  Opioid, dopamine, cannabinoid, serotonergic, O neuroendocrine, and neuro-immunological pathways (+ others) have all been implicated in placebo effects N O D E S A E L P

Y P THEORIES OF PLACEBO EFFECTS O C T  Two major theories to explain placebo effects: O N O D E LEARNING/ EXPECTATION EXPECTATION S CONDITIONING A E L P

Y P EXPECTATION O C T O N O D E S “Placebo effects generally correspond to people’s A knowledge or beliefs about the kind of drug they believe they are receiving, and for that reason, a E causal relation between expectancy and placebo L reaction has generally been assumed…” P

Y P OPEN-HIDDEN PARADIGMS O C T O N O D E S A E L P Enck et al. 2013

Y P THEORIES OF PLACEBO EFFECTS O C T  Two major theories to explain placebo effects: O N O D E LEARNING/ LEARNING/ EXPECTATION S CONDITIONING CONDITIONING A E L P

Y P CONDITIONING O C T O N O D E S A E L P

Y P CONDITIONING PARADIGMS O C T O N O D E S A E L P

Y P THEORIES OF PLACEBO EFFECTS O C T  Two major theories to explain placebo effects: O N O D E LEARNING/ EXPECTATION S CONDITIONING A E L P “Rather than being viewed as an alternative to expectancy, classical conditioning can be understood as one method by which expectancies are formed”

Y P SHIFT THE SPIN… O C T O N O D E S A E L P

Y P FROM NUISANCE TO TREATMENT O C T O N O D E S A E L P

Y P CURE ALL? O C T O N O D E S A E L P

Y P ALL DISEASES? O C T O N O D E S A E L P

Y P PLACEBO EFFECTS? O C T O N O D E S A E L P

Y P PLACEBO EFFECTS? O C T O N O D E S A E L P

Y P RESPONDERS AND NON-RESPONDERS O C T O N O D E S A E L P

Y P APPROACHES O C T O N 1) “Deceptive” 2) “Open - label” Placebo Placebo O D E S A 3) Extracting E Placebo L P

Y P HOT TOPIC O C T O N O D E S A E L P

Y P 2 “DIFFERENTIAL” PLACEBO EFFECTS O C T O N O D E S A E L P  The concept that different types of placebos may yield different magnitudes of placebo effects

Y P “DIFFERENTIAL” PLACEBO EFFECTS O C T O N O D E S A E L P

Y P EARLY CONCEPTIONS… O C T O N O D E S A E L P

Y P SHAM-CONTROLLED SURGICAL TRIALS O C T O N O D E S A E L P

Y P SHAM-CONTROLLED SURGICAL TRIALS O C T O N O D E S A E L P

Y P RECENT ATTENTION… O C T O N O D E S A E L P

Y P META-ANALYTIC APPROACHES O C T Drug vs. Inert Pill Device/Procedure vs. Sham O N O D VS. E S A E L P

Y P META-ANALYTIC APPROACHES O C T O N O D E S A E L P

Y P META-ANALYTIC APPROACHES O C T O N O D “Meta -regression analyses showed that larger E effects of placebo interventions were associated with physical placebo S interventions” (e.g. sham devices) A E L P Not head-to-head comparisons

Y P DIRECT APPROACHES O C T O N O D E S A E L P

Y P OTHER FACTORS… O C  Treatment cost, perceived innovation, branding, pill shape/colour… T O N O D E S A E L P

Y P REAL-WORLD DATA O C T O N O D E S A E L P

Y P 3 TMS AND PLACEBO EFFECTS O C T O N O D E S A E L P

Y P INTO THE SPOTLIGHT… O C T O N O D E S A E L P

Y P MANY EXAMPLES O C T O N O D E S A E L P

Y EXEMPLIFICATION OF AN ELABORATE P O THERAPEUTIC TECHNOLOGY C T O N O D E S A E L P Brainsi ght TMS

Y P SHAM TMS O C T ▪ Achieve blinding but avoid meaningful O stimulation to the brain N ▪ Goal: Mimic TMS’s visual and auditory (+/- tactile) experience but shield the O brain from the magnetic fields D ▪ Many different sham device techniques E S A E L P *Include a measure assessing success of blinding!

Y P QUANTIFYING PLACEBO EFFECTS O C T O N O  61 studies, large effect size of D 0.8 (Hedge’s g)  Meta-regression E ▪ Placebo response magnitude was S positively associated with the year A of publication (increasing sham TMS responses over time). E ▪ Studies that included patients with L treatment-resistant depression had P lower placebo responses

Y P VARIABILITY IN PLACEBO RESPONSES O C T O N O D E “41.0% of the veterans in the active S A treatment group achieved remission E of depressive symptoms ”* L P *No difference from sham group (37%)

Y P PLACEBO MODULATION OF AMYGDALA O C T O N O D E S A E L P

Y P EXTREME EXAMPLES O C T O N O D E “Contrary to our primary S hypothesis, the number of headache days decreased A significantly more in the sham E group than in the group treated with active rTMS-DLPFC at L eight weeks. Average decrease in headache days was >50% in P the sham group, indicating a powerful placebo response.”

Y P EVIDENCE FOR “DIFFERENTIAL EFFECT”? O C T O N O D  Compared inert pill group from escitalopram E medication trials to the sham TMS group of TMS S trials A  Reported no significant difference…BUT E  Methodological limitations L ▪ Heterogenous patient populations – “refractory” P ▪ Blinding – double vs single ▪ Dated (only included trials 2002-2008)

Y P FURTHER RESEARCH? O C  No studies comparing sham TMS to “no treatment” control T ▪ Needed to delineate placebo effects from “other” effects O (including activation of coming to hospital for treatment) N O D E S A E L P

Y P IMPLICATIONS O C  Unfavorable impact on statistical power for sham controlled T treatment trials O ▪ RCT investigating a treatment with a large embedded placebo effect N will generally need more subjects to prove efficacy than a treatment with a smaller placebo effect (Kaptchuk et al. 2000) O D Open Label #1 Open Label #2 Pilot RCT 60% E S A E L P Active Active Active Placebo

Y P ONGOING ISSUES… O C T O N O D E S A E L P

Y P THE EFFICACY PARADOX O C T O N O D E S A E L P

Y P ILLUSTRATIVE EXAMPLE - PAIN O C T O N O D E S A E L P

Y P O C T O N O D E S A E L P

Y ISSUES REQUIRING CRITICAL P O REFLECTION… C T What is the best way to O measure efficacy in N this context? O D E S A E How can we leverage enhanced placebo L effects? P

Y P ALTERNATIVE EXPLANATIONS… O C  Every field jumps to assuming that there must be something T “active” about their placebo group… O  Occam’s Razor: When presented with competing hypotheses, N simpler solutions are more likely to be correct than complex ones O D E S A E L P

Y P A FINAL COMPLICATING ISSUE… O C T O N O D E S A E L P

Y P MECHANISMS? O C T O N O D E S A E L P

Y P QUESTIONS O C T O N O D E S A E L P matthew.burke@sunnybrook.ca mburke11@bidmc.harvard.edu