The Effects of Aripiprazole on the Processing of Rewards in Schizophrenia

TITLE: Aripiprazole effects on reward processing in deficit syndrome schizophrenia Principal Investigator: Erica Duncan, M.D.

Background- This is a study of the way people with schizophrenia may feel when they get rewards. Certain parts of the brain play a part in people feeling happy when they win some sort of reward or have a good thing happen to them. Some people with schizophrenia may have trouble feeling pleasure because part of their brains, the frontal cortex, may not work properly. We are doing this study to understand how the frontal cortex works in people with and without schizophrenia when they are trying to win a reward. We will use a special kind of brain scan called Functional Magnetic Resonance Imaging (fMRI) to show the brain working in action. fMRI scanning uses a big magnet to take special pictures of the brain.

A new medicine called aripiprazole has been approved by the Food and Drug Administration for treating symptoms of schizophrenia. This medicine has a new mechanism of action that helps neural functioning in the part of the brain that is involved with planning and goal setting. We think this new medication may help with brain processing of rewards. We are therefore studying the brain with fMRI scans before and during treatment with this aripiprazole.

Objective- The objective of this study is to determine whether subjects with negative symptoms of schizophrenia have abnormal functioning of brain circuits relevant to reward processing, and to determine whether any such abnormalities are normalized by treatment with aripiprazole.

Research Plan-Aim 1. Define impairments in the neural correlates of reward processing in primary deficit syndrome schizophrenics compared to normal controls.

Ten stable outpatients with primary deficit syndrome schizophrenia will be compared to ten normal controls in a BOLD contrast fMRI experiment constructed to assess in a parametric design the recruitment of reward circuitry in response to increasing monetary reward. We predict that frontal cortical areas important in reward processing such as the OFC and ACC will have reduced activation in the schizophrenic subjects.

Aim 2. Assess the ability of aripiprazole to normalize reward circuit functioning in deficit syndrome schizophrenia, and correlate fMRI changes with clinical changes in negative symptoms.

The ten schizophrenic subjects in Aim 1 will be switched from standard antipsychotic treatment to twelve weeks of open label aripiprazole and retested with the fMRI monetary reward task. We predict that twelve weeks of aripiprazole treatment will normalize the activation of OFC and ACC in response to monetary reward stimuli. We further predict that the degree of normalization in fMRI activation will correlate with negative symptom improvement on aripiprazole.

Methods- Up to 25 volunteers with schizophrenia and 25 volunteers without a psychiatric disorder will participate in this study at the Atlanta VA Medical Center. Subjects will first be interviewed about their medical and psychiatric history, and any current symptoms they are having. They will receive an fMRI scan during which they will play a computer game that rewards correct responses with money. This fMRI session will allow for the assessment of which parts of the brain are functioning during rewarding conditions. The results will be compared for subjects with and without schizophrenia.

The subjects with schizophrenia will have their antipsychotic medication tapered and switched to aripiprazole during a twelve week treatment phase. They will be seen at the end of Week 1, Week 2, Week 4, Week 6, Week 8, Week 10, and Week 12. At the last visit the fMRI scan will be repeated during the same reward task.

Clinical Relevance- The inability to experience pleasure and the inability to pursue goals is an important symptom of schizophrenia. These symptoms probably play a huge role in preventing many people with schizophrenia from leading full and fulfilling lives. A central defect in reward processing pathways in the brain may account for these symptoms. If aripiprazole can be demonstrated to improve functioning in these brain pathways, this would greatly advance our understanding of schizophrenia and could bring greater functioning to patients with schizophrenia.

Complete study Protocol:

ARIPIPRAZOLE EFFECTS ON REWARD PROCESSING IN DEFICIT SYNDROME SCHIZOPHRENIA

Principal Investigator: Erica Duncan, MD

Specific Aims

The chronic negative symptoms of schizophrenia remain poorly responsive to psychopharmacologic treatment. The impaired motivation of primary deficit syndrome patients prevents them from pursuing social and occupational goals and prevents them from developing meaningful and productive lives. This proposal is based on the test of a hypothesis that schizophrenic patients with enduring deficit symptoms suffer from impairments in reward processing and the translation of perceived reward salience into goal-oriented behavior. Preclinical and clinical research has revealed the functioning of a distributed neural network in the processing of reward stimuli. Regions of the prefrontal cortex (PFC) such as orbitofrontal cortex (OFC) and anterior cingulate cortex (ACC) are key frontal cortical elements of a neural circuit that mediates reward salience, reward expectation, and reward approach behavior. Aripiprazole, in virtue of its' activity as a dopamine (DA) partial agonist, is uniquely suited to enhance functioning of these frontal areas crucial to reward processing. We hypothesize that primary deficit syndrome schizophrenia is associated with functional deficits in reward circuitry detectable by fMRI scanning during a monetary reward task. We further hypothesize that treatment with aripiprazole will normalize reward circuit function in these subjects and thus contribute to improvements in negative symptoms. The following specific aims will test these hypotheses:

Aim 1. Define impairments in the neural correlates of reward processing in primary deficit syndrome schizophrenics compared to normal controls.

Ten stable outpatients with primary deficit syndrome schizophrenia will be compared to ten normal controls in a BOLD contrast fMRI experiment constructed to assess in a parametric design the recruitment of reward circuitry in response to increasing monetary reward. We predict that frontal cortical areas important in reward processing such as the OFC and ACC will have reduced activation in the schizophrenic subjects.

Aim 2. Assess the ability of aripiprazole to normalize reward circuit functioning in deficit syndrome schizophrenia, and correlate fMRI changes with clinical changes in negative symptoms.

The ten schizophrenic subjects in Aim 1 will be switched from standard antipsychotic treatment to twelve weeks of open label aripiprazole and retested with the fMRI monetary reward task. We predict that twelve weeks of aripiprazole treatment will normalize the activation of OFC and ACC in response to monetary reward stimuli. We further predict that the degree of normalization in fMRI activation will correlate with negative symptom improvement on aripiprazole.

Background and Significance

Negative Symptoms as a Deficit in Reward Processing Antipsychotic medications are often able to relieve the positive symptoms of schizophrenia such as hallucinations, delusions, and disorganization. However the response of deficit symptoms to psychopharmacology remains disappointing. Many patients with minimal active psychotic symptoms are severely impaired in their functioning because of enduring deficit syndrome features.

A key component of the deficit syndrome is a lack of motivation to pursue goals in educational, occupational, social, and recreational domains. This proposal rests on the hypothesis that a core feature of the amotivational component of the deficit state is a deficiency in reward processing. This deficiency in turn fails to provide sufficient stimulation of conditioning reinforcement of approach behavior serving the pursuit of rewards. In most patients the impairment is not so severe that the patients fail to pursue the primary rewards of food and drink. But motivation to pursue more complex secondary reinforcers such as economic success, stable loving relationships, avocations, intellectual fulfillment and other uniquely human goals are strikingly lacking in deficit syndrome patients.

Frontal Cortical Hypofunction in Negative Symptoms Refinements in the DA hypothesis have emphasized an imbalance between mesocortical and subcortical/mesolimbic DA activity. Hyperactivity of the mesolimbic DA system has been implicated in the positive symptoms of the illness; hypofunctioning of the mesocortical DA pathway has been implicated in the pathophysiology of negative symptoms (Weinberger 1987; Davis et al.1991). The finding of neurophysiologic hypofunction of the frontal cortex during cognitive activation studies in schizophrenic subjects has given rise to the concept of hypofrontality (Weinberger et al. 1986; Berman et al. 1986; for review see Weinberger et al. 1994). Furthermore, frontal hypofunction has been linked specifically to negative symptoms irrespective of the severity of positive symptoms in unmedicated schizophrenics (Wolkin et al. 1992).

Neural Circuitry of Reward Processing Research into the neural correlates of reward processing indicates the importance of midbrain dopamine (DA) neurons and distributed reward circuits to the coding of reward anticipation and salience, and the translation of motivation into approach behavior. Key structures that have been implicated in reward processing on the basis of animal studies include DA neurons of the ventral tegmental area (VTA) and substantia nigra and their target areas such as the nucleus accumbens (NAcc), amygdala, and OFC (Wise 1980; Wise and Hoffman 1992; Koob 1992; Robbins and Everitt 1996; see Schultz 2000, 2001 for review). Circuitry linking these areas is hypothesized to mediate several aspects of reward processing such as the detection of primary or secondary reward stimuli, prediction of expected future rewards, and the use of information to control goal-directed or motivated behavior (Schultz 2000). Recently developed methods are being used to study the reward pathways in normal humans and subjects with substance abuse disorders in the fMRI environment. These studies implicate the mesocorticolimbic DA system and functionally connected areas: OFC, ACC, dorsolateral PFC, amygdala, NAcc, and insula (Grant et al. 1996; Childress et al. 1999; Maas et al. 1998; Garavan et al. 2000; Kilts et al. 2001; Wexler et al. 2001). fMRI paradigms using money as a secondary reinforcer reveal activations in many of these same areas: amygdala, NAcc, ventral striatum, and areas of PFC (Thut et al. 1997; Elliott et al. 2000; Elliott et al. 2003). The fact that very similar areas are activated by primary reinforcer paradigms in animal studies, drug cue studies in substance abusers, and monetary reward paradigms in normal controls suggests that the study of monetary reward processing is a valid strategy to assess the integrity of reward circuitry in experimental paradigms.

Overlap of Reward Circuitry and Areas Implicated in Negative Symptoms As indicated above, dopaminergic hypofunction in the PFC is strongly implicated in the pathophysiology of negative symptoms. The presentation of reward or stimuli that predict reward stimulates the phasic activation of dopaminergic neurons in the midbrain (Shultz 2000) that innervate the PFC. Hence there is a critical overlap of circuitry implicated in negative symptoms and circuitry that subserves reward processing. This overlap supports the use of reward paradigms to explore the functional integrity of mesocortical DA inputs and PFC function.

Aripiprazole and Negative Symptoms The novel antipsychotic, aripiprazole, has a unique pharmacological mechanism of action that makes it potentially unique for the amelioration of negative symptoms. Aripiprazole is thought to increase DA activity in the PFC by virtue of its' activity as a partial agonist at post-synaptic DA2 receptors (Burris et al. 2002). We hypothesize that treatment with aripiprazole will normalize the function of PFC reward circuitry through increased DA functioning in frontal cortical regions (OFC, ACC) that subserve reward processing.

Potential Importance of Proposal for Treatment Indications of Aripiprazole The severe impairment of motivated behavior in patients with deficit syndrome schizophrenia is strongly suggestive of an abnormality in the neural circuits that mediate reward processing and the translation of reward salience to motivated behavior. To date there are no published fMRI studies of the functioning of reward circuits in subjects with deficit syndrome schizophrenia. We hypothesize that negative symptom schizophrenics will exhibit impaired fMRI estimated activation of reward circuitry compared to normal controls, and that aripiprazole treatment will normalize functioning of this circuitry.

The demonstration of improvements in negative symptoms in medication trials has been elusive. Even an agent capable of improving reward processing might not produce detectable decreases in rating scales designed to measure negative symptoms over the short term. This is because patients with chronic deficit states have compromised functioning due to the cumulative effects of social and occupational dysfunction. In patients such as these, the normalization of reward processing circuits by a medication such as aripiprazole would only be the first step. These patients are likely to require considerable psychosocial rehabilitation to recoup occupational and social functioning and achieve meaningful gains in negative symptom rating scales. However, normalization of reward circuitry function would strongly suggest that aripiprazole has potential to improve chronic negative symptoms when combined with a supporting rehabilitative program. Significant changes in reward circuit function would also strongly support the early intervention potential for aripiprazole to prevent chronic deterioration in function if first break patients were treated with this agent early in their course.

Materials and Methods

Subjects Ten male schizophrenic subjects ages 18-60 will be recruited. Diagnosis will be confirmed by structured diagnostic interview (SCID-I). Primary deficit syndrome will be confirmed by rating with the Schedule for the Deficit Syndrome (SDS), a scale that defines schizophrenic subjects according to whether they have chronic negative symptoms even across periods when positive symptoms are in remission (Kirkpatrick et al. 1989). This scale thus allows for the differentiation of those patients who have negative symptoms consistently across clinical states from those whose negative symptoms emerge when they are acutely psychotic and remit when positive symptoms remit. Additionally, subjects must have a minimum score of 30 on the SANS. Subjects will be excluded for clinically significant unstable medical illness, history of neurological disease including head trauma with loss of consciousness >5 minutes, active substance abuse or dependence within the prior three months, any contraindication to fMRI, left handedness, mental retardation, color blindness, antipsychotic treatment resistance, or known allergy to or nonresponse to aripiprazole. Subjects with schizophrenia will be excluded if they do not have a working telephone.

Ten male normal comparison subjects will recruited, matched in mean age and ethnicity to the schizophrenic group. Absence of psychopathology will be confirmed by SCID-I, NP version. Subjects will be excluded for clinically significant unstable medical illness, history of neurological disease including head trauma with loss of consciousness >5 minutes, active substance abuse or dependence within the prior three months, any contraindication to fMRI, left handedness, color blindness, or mental retardation. Normal controls will be matched to schizophrenic subjects with regard to monthly income so that monetary rewards during the task will have comparable reinforcement value.

Baseline Assessment Prior to fMRI scanning all subjects will be assessed for smoking status by means of the Fagerstrom Smoking Tolerance Questionnaire (Fagerstrom 1978) and nicotine dependent subjects will be instructed to smoke their usual amount prior to reporting for scanning. Screening for intact vision (≥ 20/25 uncorrected or corrected with eyeglasses or contact lenses) by means an eye chart will be conducted. Cognitive status will be assessed in the following domains: attention (Continuous Performance Test), psychomotor reaction time (finger tapping test), memory (California Verbal Learning Test), IQ (Weschler Abbreviated Scale of Intelligence, vocabulary and matrices subtests), and executive function (Stroop Task). The schizophrenic subjects will be rated for severity of psychopathology by means of the PANSS, Abrams and Taylor Rating Scale for Emotional Blunting (A\&T), SANS, Clinical Global Impression Scale (CGI), and the Quality of Life Scale. Current side effects to antipsychotics will be assessed by means of the Barnes Akathisia Scale and the Simpson Angus Scale. Schizophrenic subjects will have 30cc of blood drawn for CBC and SMA-12 to ensure stable baseline medical status prior to switching to aripiprazole.

Cognitive Task Meaningful comparisons of schizophrenic and non-schizophrenic subjects by a cognitive task necessitates the use of a simple task to minimize confounds related to group differences in cognitive ability. A parametric manipulation of reward contingency will control for group differences in task performance. The proposed monetary incentive task, adapted from that of Elliott et al. (2003), is chosen because of its simplicity and parametric design. The subjects will be given a simple target detection task in which they will be instructed to squeeze a response bulb when they see a red or blue square. Different colored squares will be presented on a screen for 1.3 sec each. When they correctly signal the target stimuli they will see a reward stimulus informing them of how much money they have won on that trial. For trials with non-targets (paying no reward if the subject squeezes the bulb) the subjects will see a neutral text telling them to wait. The reward amounts will vary from 5¢ to $1 per trial (5¢, 25¢, 50¢, 75¢, or $1). The task will be divided into five blocks separated by 30 sec rest periods. Each block will contain 40 trials. Trial types will be presented in randomized order within each block: 30% of trials will contain one of the two target stimuli. Each block will pay correct responses at one of the five monetary reward levels. The task is designed specifically to allow for both an "on-off" assessment of activations in the presence or absence of reward, but also for a parametric analysis of brain areas that increase their activation with increased reward levels. The subjects will be told that their payment for participation in the fMRI scan will be the total money they accumulate by correct responses during the task. This latter point is important in that the money won during task performance has a real world salience more likely to enhance reward-related activations than if all subjects automatically were paid the same amount regardless of their accuracy. Thus the subjects receive actual rather than simulated reward.

fMRI Scanning Blood oxygen level-dependent (BOLD) fMRI scanning will be performed on a research dedicated Siemens 3T whole body MRI scanner located in the Emory Hospital MRI Center. Foam padding will be used to restrict the subjects' head motion within the magnet. 30 axial slices of 3 mm thickness will be acquired parallel to the AC-PC line with a matrix size of 64 x 64 over a field of view of 22 x 22 cm, using a TE of 30 msec. With this protocol we have successfully imaged the orbitofrontal cortex with minimal magnetic susceptibility artifacts. The functional images will be obtained using a T2* weighted spiral-scan pulse sequence (TR 3000 msec, TE 25 msec, flip angle 60 deg). High resolution anatomical T1-weighted MR images will be acquired for localization of task-related neural activations (TR =10 msec, TE = 4.5 msec, TI=900 msec, FOV=24 cm, 256 x 256 matrix, contiguous 3 mm slices covering the whole brain).

Aripiprazole Treatment After receiving the first scan, the schizophrenic subjects will be started on open label aripiprazole 15 mg po qd. Over the subsequent six weeks they will be titrated to a target dose of 30 mg po qd and maintained on this dose for the next six weeks. The aripiprazole titration schedule will be as follows: 15 mg Weeks 1-2, 20 mg Weeks 3-6, and then 30 mg. During the first six weeks of the treatment phase their prior antipsychotic will be gradually tapered down (75% of total daily dose in Week 1 and 2, 50% of total daily dose in Weeks 3 and 4, 25% of total daily dose in Week 5 and 6) and then discontinued.

The subjects will be brought in for study visits at the end of Week 1, Week 2, Week 4, Week 6, and Week 8, Week 10, and Week 12. At each study visit they will be rated with the PANSS, SANS, CGI, and A\&T. Extrapyramidal side effects and akathisia will be rated with the Simpson Angus Scale and the Barnes Akathisia Scale respectively. The Quality of Life Scale, the CVLT, CPT, and Stroop Test will be repeated at the end of Week 12.

Stopping Rules for Subjects with Schizophrenia There is a risk of symptom exacerbation when antipsychotic medication is switched. For this reason, several safeguards will be put into place.

1. Subjects will be excluded if they have a history of suicidal, or assaultive behavior during psychotic decompensations.
2. The subjects will be given a cross-titration whereby their prior antipsychotic medication is gradually tapered over six weeks while aripiprazole is being titrated to maximum dose.
3. Subjects in the schizophrenia group will be closely monitored during the treatment phase. They will be given reminder calls the day before each study visit. If they do not come for a scheduled visit, the study staff will call and reschedule for the soonest day they can come in. Subjects without a working telephone will be excluded.
4. At each visit subjects will be rated with the CGI. If the CGI Global Improvement Scale is rated 6 (Much Worse) or 7 (Very Much Worse) at any visit, they will be dropped from the study and their previous antipsychotic will be resumed at the dose prior to enrollment. fMRI Image Analysis Imaging data will be analyzed by means of MATLAB and statistical parametric mapping software (SPM99). The images will be resliced and corrected for motion by registration to the first functional image acquired for each subject using a 6 parameter transformation (Friston et al 1995). Images will be spatially normalized to a Montreal Neurological Institute template. A high-pass filter will be used to remove low frequency noise. Image smoothing using a 6 mm Gaussian kernel will be used to enhance signal-to-noise ratios and accommodate differences in neuroanatomy to facilitate group comparisons. The delayed cerebral blood flow response to activation conditions will be modeled using the standard hemodynamic response function. Statistical analysis will proceed using a random effects model to examine outcomes of the parametric design of the experiment (Buchel et al. 1998). We will model both linear and nonlinear hemodynamic responses in regions of interest: Nacc, amygdala, striatum, thalamus, OFC, ACC, and hippocampus.

Analysis of Behavioral Data For Aim 1, cognitive data will be analyzed with a between subjects MANOVA to compare neurocognitive results between schizophrenics and normal controls. For Aim 2, behavioral rating scale data and repeated cognitive test scores (for CVLT, Stroop Test, CPT) will be analyzed with a repeated measures MANOVA model using timepoint as a within subjects factor.

References

Berman KF, Zec RF, Weinberger DR (1986) Physiologic dysfunction of dorsolateral prefrontal cortex in schizophrenia. II. Role of neuroleptic treatment, attention, and mental effort. Arch Gen Psychiatry. 43: 126-35

Buchel C, Holmes AP, Rees G, Friston KJ (1998) Characterizing stimulus-response functions using nonlinear regressors in parametric fMRI experiments. Neuroimage 8:140-8

Burris KD, Molski TF, Xu C, Ryan E, Tottori K, Kikuchi T, Yocca FD, Molinoff PB (2002) Aripiprazole, a novel antipsychotic, is a high-affinity partial agonist at human dopamine D2 receptors. J Pharmacol Exp Therapeutics 302(1):381-9

Childress AR, Mozley PD, McElgin W, Fitzgerald J, Reivich M and O'Brien CP (1999). Limbic activation during cue-induced cocaine craving. Am J Psychiatry 156(1):11-8

Davis KL, Kahn RS, Ko G, Davidson M (1991) Dopamine in schizophrenia: a review and reconceptualization. Am J Psychiatry 148:1474-1486

Friston KJ, Ashburner J, Frith CD, Poline J-B, Heather JD, Frackowiak RSJ (1995) Spatial registration and normalization of images. Hum Brain Mapp 2:1-25

Garavan H, Pankiewicz J, Bloom A, Cho J-K, Sperry L, Ross T, Salmeron B, Risinger R, Kelly D and Stein E (2000) Cue-induced cocaine craving: neuroanatomical specificity for drug users and drug stimuli. Am J Psychiatry 157:1789-98

Elliott R, Friston KJ, Dolan RJ (2000) Dissociable neural responses in human reward systems. J Neurosci 20:6159-65

Elliott R, Newman JL, Longe OA, Deakin JF (2003) Differential response patterns in the striatum and orbitofrontal cortex to financial reward in humans: a parametric functional magnetic resonance imaging study. J Neurosci 23:303-7

Fagerstrom KO (1978) Measuring degree of physical dependence to tobacco smoking with reference to individualization of treatment. Addict Behav 3:235-41

Grant S, London ED, Newlin DB, Villemagne VL, Liu X, Contoreggi C, Phillips RL, Kimes AS and Margolin A (1996) Activation of memory circuits during cue-elicited cocaine craving. Proc Natl Acad Sci USA 93(21):12040-5

Kilts CD, Schweitzer JB, Quinn CK, Gross RE, Faber TL, Muhammad F, Ely TD, Hoffman JM and Drexler KP (2001) Neural activity related to drug craving in cocaine addiction. Arch Gen Psychiatry 58(4):334-41

Kirkpatrick B, Buchanan RW, McKenney PD, et al. (1989)The schedule for the deficit syndrome: an instrument for research in schizophrenia. Psychiatry Res 30:119-23

Koob GF (1992) Drugs of abuse: anatomy, pharmacology and function of reward pathways. Trends in Pharmacological Sciences. 13: 177-84

Maas LC, Lukas SE, Kaufman MJ, Weiss RD, Daniels SL, Rogers VW, Kukes TJ and Renshaw PF (1998) Functional magnetic resonance imaging of human brain activation during cue-induced cocaine craving. Am J Psychiatry 155(1):124-6

Robbins TW, Everitt BJ (1996) Neurobehavioural mechanisms of reward and motivation. Current Opinion Neurobiol 6: 228-236

Schultz W (2000) Multiple reward signals in the brain. Nature Rev Neurosci 1: 199-207

Schultz W (2001) Reward signaling by dopamine neurons. Neuroscientist 7:293-302 Thut G, Schultz W, Roelcke U, Nienhusmeier M, Missimer J, Maguire RP, Leenders KL (1997) Activation of the human brain by monetary reward. NeuroReport 8: 1225-8

Weinberger DR (1987) Implications of normal brain development for the pathogenesis of schizophrenia. Arch Gen Psychiatry 44:660-669

Weinberger DR, Aloia MS, Goldberg TE, Berman KF (1994) The frontal lobes and schizophrenia. Journal of Neuropsychiatry \& Clinical Neurosciences. 6: 419-27

Weinberger DR, Berman KF, Zec RF (1986) Physiologic dysfunction of dorsolateral prefrontal cortex in schizophrenia. I. Regional cerebral blood flow evidence. Archives of General Psychiatry. 43: 114-24

Wexler BE, Gottschalk CH, Fulbright RK, Prohovnik I, Lacadie CM, Rounsaville BJ and Gore JC (2001) Functional magnetic resonance imaging of cocaine craving. Am J Psychiatry 158(1):86-95

Wise RA (1980) Action of drugs of abuse on brain reward systems. Pharmacol Biochem Behav 13:213-223

Wise RA, Hoffman DC (1992) Localization of drug reward mechanisms by intracranial injections. Synapse. 10: 247-63

Wolkin A, Sanfilipo M, Wolf AP, Angrist B, Brodie JD, Rotrosen J (1992) Negative symptoms and hypofrontality in chronic schizophrenia. Arch Gen Psychiatry. 49: 959-65