Neuroscience of Alcohol Addiction David W. Oslin, MD Associate Professor University of Pennsylvania, School of Medicine And Philadelphia, VAMC Hazelden Research Co-Chair on Late Life Addictions
Introduction Alcoholism costs the nation $150 Billion / annum in the US. As such it is the most expensive addictive disorder. Alcoholism leads to increased mortality and morbidity Alcoholism is common with about 7 million Americans afflicted Worldwide alcoholism is the 7 th leading cause of disability 2
Alcohol Effects 30% of inpatients have alcohol related problems Have poorer outcomes Surgical complications 20% of inpatient health care budget 50% fatalities from fire and MVAs 67% of homicides 35% of suicides Responsible for 100,000 deaths /yr 3
What is the current model? 4
Already ate It was there Feel Better I am full Couldn t eat just one Tastes good Was thinking about a snack Don t like Like It tasted good in the past 5
Model of Addiction Emotion Control Tempral - Parietal Orbital Prefrontal Cortex Reward / Euphoria Craving Ventral Tegmentum (dopamine) Nucleus Accumbens Memory Hippocampus 6
Addiction is not one disease What s s the evidence and how do we study this? 7
Alcohol Euphoria / Craving Alcoholics classically lose control after their first drink Their desire for alcohol after the first drink is much greater than before Dopamine and endogenous opioids ( -endorphin, enkephalin) are released by alcohol and mediate alcohol euphoria Genetic factors Vulnerability: Enhance dopamine and natural opioid responses to alcohol and increase its rewarding effect
Alcohol Euphoria / Craving State / Trait Craving Alcohol-Induced Craving Cue-Induced Craving Stress-Induced Craving
Stress and Cue-Induced Alcohol Craving in Abstinent Alcoholics STRESS ALCOHOL CUE NEUTRAL 4 4 4 Alcohol Group Healthy Group Alcohol Group Healthy Group Alcohol Group Healthy Group Raw Values 2 2 2 0 0 0-30 -15-5 0 5 15 30 45 60 75-30 -15-5 0 5 15 30 45 60 75-30 -15-5 0 5 15 30 45 60 75 Time Points Time Points Time Points Group Difference: F{1,55}= 81.13, p<.0001(ad> HC in S and Cue Response, p<.0001) Group X Condition: F{2,110}=12.07, p<.0001 Group X Timepoints: F{7,385}=3.95, p<.0004 Sinha et al., 2005
Craving Hazelden Study To describe the course of alcohol craving and affect (positive and negative) during the first 28 days of recovery from alcohol dependence. To test for associations between the course of alcohol craving and affect during the first 28 days of recovery and continued abstinence over the 6 months post discharge from the center.
Higher Craving m_al c 3. 0 2. 5 Alcohol Craving by Latent Class #1 n=17 2. 0 1. 5 #2 n=43 1. 0 Lower Craving 0. 5 0. 0 #3 n=37 0 5 10 15 20 25 daynum al c Day cr avi ngnumber 1 from 2 start 3 12
6 Month Outcomes 1. 00 0. 75 0. 50 0. 25 0. 00 0 25 50 75 100 125 150 175 200 225 dur dr i nk STRATA: al c_cl ass=1low Censor ed al c_cl ass=1low al c_cl ass=2medi um Censor ed al c_cl ass=2medi um al c_cl ass=3hi gh Censor ed al c_cl ass=3hi gh 13
Deconstructing the phenotype: An experimental medicine approach l Premise: Alcohol reward in part due to dopamine response l Hypothesis: Alcohol (most) rewarding related to genotype
An alcohol endogenous opioid DA cascade (Tanda & Di Chiara 1998) l EtOH releases endogenous opioid peptides in VTA l Activation of OPRM1 receptors removes GABA brake l Disinhibition of DA results l OPRM1 blockade therapeutic by blocking EtOH induced DA release?
Naltrexone (opioid and alcohol) Functions as an opioid receptor antagonist (mu >> delta or kappa) Alcohol consumption affects the production, release, and activity of opioid peptides (Herz, 1997) Opioid peptides mediate some of alcohol s rewarding effects by enhancing midbrain dopamine release Opioid antagonists suppress alcohol-induced reward (Swift,1999) and general consummatory behaviors (Boyle et al. 1998) 16
Meta-Analysis Oral Naltrexone vs. Placebo: Relapse to Heavy Drinking Srisurapanont & Jarusuraisin, 2006 17
Long-Acting Naltrexone Results: Median Heavy Drinking Days 30 Median Heavy Drinking Days per Month 25 20 15 10 5 19.3 6.0 4.5 3.1 Baseline Placebo L-A Ntx 190 mg L-A Ntx 380 mg 48% 0 n = 624 p < 0.005 18
Clues on a Genetic Contribution % Days of Heavy Drinking 16 14 12 10 8 6 4 2 Naltrexone Placebo 0 Low Density Medium Density High Density Monterosso et al 2001 19
Study Design Participated in 1 of 3 randomized controlled trials of naltrexone Minimum exposure of at least 6 weeks and adherent to medication on at least 50% of study days Consented to genetic study 20
Genetic Polymorphisms and Alcohol Treatment Cumulative Survival (time to relapse) Naltrexone / Asp40 Allele (A/G, G/G) Naltrexone Asn40 Allele (A/A) Placebo / Asp40 Allele (A/G, G/G) Placebo / Asn40 Allele (A/A) Days Oslin DW, et. al. 2003 21
COMBINE Study Good Clinical Outcome (%) Sample size: 31 Asp40 in the naltrexone group 35 Asp 40 in the placebo group
A+118G (Asn40Asp) Asparagine asn amide (neutral) H2N-CO-CH2-CH(NH2)-COOH Aspartic acid asp (negatively charged) HOOC-CH2-CH(NH2)-COOH Asp40 allele frequency of 13-20% (24.3 36% of European Americans have at least one copy) 23
Alcohol-clamp : a controlled way of exposing the brain to alcohol (Ramchandani et al., Mol Psychiat 2010)
Dopamine release in ventral striatum is restricted to OPRM1-118G carriers (Ramchandani et al., Mol Psychiat 2010) (BP alcohol - BP placebo )/BP placebo (%) 15 10 5 0-5 -10-15 -20 * * AA (n = 16) GX (n = 12) AVS PVS Caudate Putamen
Subjective as well (Ramchandani et al., Mol Psychiat 2010) E "Feel Intoxicated" 4 3 2 1 0 AA AG 7 15 22 30 37 45 60 75 Time (minutes)
h/moprm1 118G is sufficient to mimic increased alcohol-induced DA-release found in humans (Ramchandani et al., Mol Psychiat 2010)
Increased alcohol-induced DA-release in 118GG mice predicts increased voluntary alcohol Intake (g/kg/24hrs) 25 20 15 10 5 0 intake (Thorsell et al, in preparation) 3% 5% 7% ** Alcohol Concentration *** ** 9% 11% 13% 15% 20% 20% end GG AA
Conclusions: What it might mean for alcohol related traits l l l l Alcohol leads to release of endogenous opioids This leads to activation of dopamine in ventral striatum Alcohol-reward cascade can be blocked with naltrexone Important in people with the right (or, rather, wrong) genetics l l A biological category: Opioid-sensitive alcohol addiction? For others, other mechanisms may be important
But molecular mystery remains: 118G is e.g. a loss-of of-functionfunction variant for morphine response 45 40 35 30 25 20 15 10 5 0 45 +/- 8 % meiosis 33 +/- 6* 24 +/- 7* AA (40) AG (8) GG (3) OPRM1 GENOTYPE (n) p<0.001 AA vs AG/GG Lotsch et al, 2006 30
Demand and consumed morphine dose in patients who received patient-controlled analgesia post arthroplasty Genotype frequency (%) Demand in first 48 h Dose in first 48 h AA 62 39.0 (24.7) 25.3 (15.5) AG 27 35.3 (23.3) 25.6 (11.7) GG 11 57.8 (24.7) 40.4 (22.1) AA vs. GG *P = 0.026 *P = 0.003 GG vs. AG *P = 0.012 *P = 0.008 The morphine consumed doses are expressed as mean (standard deviation) milligrams. Demand is the dose that represents the number of times the patient pushed the release button of the patientcontrolled analgesia device. 31 Chou, W.-Y., et al Acta Anaesthesiologica Scandinavica 50 (7), 787-792.
Why should there be an alcohol promoting gene? 32
A trait that may be susceptible to selection: Maternal attachment, stronger in rhesus 77G allele carriers (Barr et al. PNAS 2008) Vocalizations 25 C/C C/G 20 15 10 5 0 Pre S1 S2 S3 S4 Week
Human translation: Increased sensitivity to social rejection in OPRM1 118G carriers (Way et al. PNAS 2009)
Conclusions: What it might mean beyond alcohol l Contributes to hawk-like traits: ä bold / exploratory (mouse +; rhesus +; human?) ä more aggressive (mouse?; rhesus +; human?) ä higher pain-threshold (mouse?; rhesus +; human +) l Alcohol related traits coincidental byproducts late in evolution