Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 39
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary 3 Psychiatric and Drug Addiction Disorders In assessing psychiatric and drug addiction disorders, there is a need to move from qualitative to quantitative measures. Biomarkers for psychiatric and drug addiction disorders will provide a valuable resource necessary to expand diagnosis and monitoring beyond the often qualitative categorizations revealed by clinical experiences and the manuals on mental health (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition [DSM-IV] and the International Classification of Diseases, 10th Revision [ICD-10]). Quantitative measurements that could be gleaned from the biomarkers themselves offer better categorization of individuals, target treatments more effectively and earlier for patients, and determine vulnerability to disorders. In Session III the discussion centered on specific areas of psychiatric and drug addiction research where specific biomarkers are currently showing promise, as well as opportunities for further impact.
OCR for page 40
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary OVERVIEW Psychiatric disorders, like most other nervous system disorders, lack biomarkers in clinical use. Instead, diagnosis of psychiatric disorders rests on patients’ reports of their symptoms, signs from their mental status examination, and clinician observations of their behavior. To make a diagnosis, mental health professionals group those clues into distinct diagnostic categories listed in one of two classification systems, DSM-IV and ICD-10 (American Psychiatric Association, 2000; World Health Organization, 2007).The categories listed there are based on expert consensus that draws from both scientific evidence and clinical experience. The diagnostic categories are largely descriptive in orientation, with DSM actively professed to be “neutral with respect to theories of etiology” (American Psychiatric Association, 2000). But should the diagnostic categories of psychiatric disorder drive the search for biomarkers? Are there complementary alternatives to using standard diagnostic classifications? Those were the provocative questions raised by Dr. Steven Hyman, provost of Harvard University. Growing evidence suggests that biomarker research might best be served by focusing elsewhere. Hyman proposed that biomarker research should focus less on current categories of disorder and more on underlying clinical states for which some knowledge of pathophysiology or neurocircuitry is available. Clinical states of this kind often transcend the boundaries of a single category of disorder. For example, the cognitive impairment observed in schizophrenia (including impairment of working memory) is associated with thinning of prefrontal cortex observed by structural MRI (Hyman, 2007b). Although it is responsible for substantial disability, it is not part of the DSM-IV criteria, which date to earlier understandings of schizophrenia as primarily reflecting psychotic symptoms such as hallucinations and delusions. A focus on biomarkers to follow working memory deficits involving prefrontal cortical circuits would seem more likely to succeed than searching for a biomarker of DSM-IV schizophrenia, which is a heterogeneous syndrome defined only by symptoms and course. To understand the change in emphasis, Hyman first traced the intellectual and historical underpinnings of the current diagnostic classification systems and then argued that excessive reliance on current, consensus diagnostic categories—especially for the purpose of biomarker research—may lead researchers down blind alleys.
OCR for page 41
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary The criteria used to guide the current categorization of psychiatric disorders were developed in 1970 by pioneering epidemiologists (Robins and Guze, 1970). On the basis of their empirical research, they proposed that reliable and valid diagnoses should be based on five criteria: clinical description (symptom clusters), laboratory studies, delineation of one disorder from another, follow-up studies, and family studies. These criteria fueled the modern era of psychiatric diagnosis and helped to launch decades of empirical research, as opposed to the earlier emphasis on theory and small, nonrepresentative samples. With disorder classifications came the ability to study individual diagnoses and their causation. One of the most powerful lines of research dealt with family studies. This approach strongly influenced the current classification systems, most notably the DSM system of the American Psychiatric Association. The benefit of the DSM system is the greater likelihood that two observers would agree on the diagnosis of an ill individual (reliability). The DSM diagnostic system facilitated epidemiology, clinical trials, and research on disease mechanisms by producing broad agreement on specific disease entities. The drawback of the broad acceptance of DSM criteria by journal editors, grant reviewers, and regulatory agencies is that boundaries drawn in the 1970s—without the benefit of objective tests, knowledge of pathophysiology, or identification of genetic risk factors—could not possibly mirror nature. Thus, imaging or genetic studies that use accepted criteria might be hobbled by starting with heterogeneous populations. An additional concern with the ability of current criteria to capture disease entities is that the DSM system conceptualizes all disorders as categories that are discontinuous with normal, whereas much evidence suggests that core symptoms of many disorders—including autism, schizophrenia, depression, attention deficit hyperactivity disorder, and personality disorders—might be better captured as dimensional or quantitative traits continuous with normal (Hoekstra et al., 2007; Kendler and Gardner, 1998). These concerns are not meant to imply that the DSM criteria represent arbitrary constructions or chimeras. In fact, the cross-cultural similarity of symptoms for the major disorders (Kendler and Gardner, 1998) and the high rates of heritability that have been established suggest that, however imprecise, the criteria for the major disorders are picking out something real. At the same time, genetic studies also point out the limitations of the current criteria. Studies of twins, starting in the 1970s, strongly implicated a genetic contribution to several psychiatric disorders. The evidence revealed that
OCR for page 42
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary diseases such as autism, schizophrenia, and bipolar disorder, in particular, had significant genetic components of risk. Monozygotic (MZ) twins (twins who share the same genetic endowment) were found more concordant for these disorders than were dizygotic (DZ) (twins whose genetic endowment was, on average, 50 percent similar to that of their siblings) (Figure 3-1). Nevertheless, as further evidence was collected, epidemiologic and genetic studies of families and twins called into question some of the categorical boundaries between disorders. For example, schizophrenia and bipolar disorder were sometimes found to occur in the same family pedigrees in distinct families (Pope and Yurgelun-Todd, 1990; Berrettini, 2000). In another key example, two separate diagnoses that frequently co-occur—major depression and generalized anxiety disorder—were found to share risk genes (Kendler et al., 1987). This example suggests that the high rates of comorbidity that characterize psychiatric disorders may be partly artifactual (Kessler et al., 2005). Finally, symptom clusters of certain psychiatric disorders, such as bipolar disorder and psychosis or bipolar disorder and rapid cycling, failed to cosegregate across generations (Craddock et al., 2005). FIGURE 3-1 DZ and MZ twins concordance rate for schizophrenia and bipolar disorder. NOTE: Monozygotic (MZ); dizygotic (DZ). SOURCE: Gottesman and Wolfgram, 1991.
OCR for page 43
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary Additional problems cropped up to challenge the boundaries separating certain psychiatric disorders. DSM diagnoses, which were largely based on cross-sectional research and observation, may not remain stable over a lifetime, according to more recent longitudinal research. For example, early anxiety disorder may give way to depression (Wittchen et al., 2000). Finally, many patients do not fit DSM-IV criteria. The DSM handles this problem by including, within groupings of related disorders, a category called “not otherwise specified” (NOS). Scrupulous clinicians often find themselves using such catch-all diagnoses (Fairburn and Bohn, 2005). In the long run the pathophysiology of psychiatric disorders will be understood. With modern genomic and genetic tools, such as high-density whole-genome association studies, which are beginning to yield results for other complex disorders, risk genes should be found for psychiatric disorders, assuming that large enough populations can be assembled for analysis (Altshuler and Daly, 2007). The question is how best to find biomarkers and drug targets in the mean time. Hyman argues that the intermediate strategy is not to discard DSM, but to “deconstruct” some of the disorders into symptom complexes that can be related to known neural circuits. For example, schizophrenia could be reconceptualized in dimensional terms with dimensions that captured (1) positive symptoms (e.g., hallucinations and delusions), (2) negative symptoms (e.g., avolition1), (3) cognitive impairments (e.g., deficits in working memory), and (4) mood symptoms (e.g., depressive symptoms). The research community could focus on those aspects for which underlying neural circuits could reasonably be identified (Box 3-1). For example, much is known about the circuitry underlying working memory and cognitive control of behavior, whereas relatively little is known about the neural circuits involved in positive symptoms. Using structural and functional imaging, animal models, genotyping, and neuropharmacological, electrophysiological, and other methods, it might be possible to identify both biomarkers and drug targets. In the case of cognitive impairments in schizophrenia, such approaches are already bearing fruit. For example, gray matter thinning and functional imaging abnormalities associated with working memory deficits have been identified (Cannon et al., 2002; Barch et al., 2001). 1 Avolition describes an individual’s perceived disinterest due to a decreased ability or inability to initiate and maintain goal-directed behavior.
OCR for page 44
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary BOX 3-1 Potential Biomarkers Based on “Deconstruction” of DSM-IV Disorders Executive function/working memory in schizophrenia and other conditions (e.g., frontal-striatal thalamic circuits) Abnormalities of conditioned fear that characterize multiple DSM-IV anxiety disorders (amygdala-based fear circuitry) Addiction, impulse control disorders, and possibly anhedonia in depression (mesotelencephalic and related reward circuitry) Mood regulation (more speculatively, circuits involving subgenual prefrontal cortex and its connections) SOURCE: Hyman, 2007a. The argument for this approach is that currently we have little chance of understanding the overall neurobiology of schizophrenia or other psychiatric disorders, but we can selectively understand important symptoms that might emanate from abnormal structure or function of prefrontal cortical circuits involved in executive function, in fear circuitry involving the amygdala, in mesotelencephalic reward circuits, and perhaps even in circuits involving mood regulation (Mayberg et al., 2005). Once we focus on neural circuits and their component cells and synapses, we are within more familiar biological paradigms with respect to discovering biomarkers and drug targets. BIOMARKERS FOR DEPRESSION Treatment of depression is poised for major advances from biomarker research. New findings have identified at least three genes that might guide depression treatment. Those who possess these genes are more likely to experience a positive treatment response, according to convergent findings drawing on multiple methodologies: large clinical trials, human genotyping and imaging research, and animal models, said Dr. Husseini Manji, director of the Mood and Anxiety Disorders Program at the National Institute of Mental Health (NIMH). Finding biomarkers to predict treatment response is extremely important for this highly prevalent condition, which is notoriously difficult to treat (U.S. Department of Health and Human Services, 1999). Treat-
OCR for page 45
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary ment response is markedly variable, with any given antidepressant only effective in about 50 percent of patients (there is a high placebo response, which lowers this figure considerably). The current state of treatment is left to a trial-and-error process consuming weeks to months—a delay during which many patients experience crippling disability, needless suffering, and, in severe cases, the possibility of suicide (Stewart et al., 2003; U.S. Department of Health and Human Services, 1999). The large variation in patients’ treatment response may be partially attributable to genetics, according to at least two lines of indirect evidence: The outcome of treatment appears to run in families (Franchini et al., 1998), and it seems to vary less across illness episodes than across individuals (Fava et al., 2002; Franchini et al., 1998). One candidate biomarker for treatment response is the gene encoding the serotonin 2A receptor (referred to by the acronym HTR2A). Serotonergic neurons are prime targets for the first-line pharmacologic class of treatment, the selective serotonin reuptake inhibitors (SSRIs). HTR2A is one of several subtypes of the serotonin receptor. HTR2A was identified by a large collaboration involving the National Institutes of Health and extramural teams as a candidate biomarker in a study that took advantage of the rich dataset from a large clinical trial, the Sequenced Treatment Alternatives to Relieve Depression (STAR*D). That major clinical trial of nearly 4,000 patients was a multisite effectiveness study aimed at real-world patients rather than the rarefied and more homogeneous samples of patients used in most clinical trials according to strict inclusionary and exclusionary criteria. The SSRI citalopram was administered in the first step of the STAR*D trial. The collaborative teams subsequently used the data from that first phase, collecting DNA samples from nearly 2,000 patients. They genotyped the samples, specifically sequencing more than 750 single nucleotide polymorphisms (SNPs) near 68 candidate genes (wholegenome scanning was neither financially nor technologically feasible at the time of the study). The study’s specific goal was to find genes associated with a positive treatment outcome in both a test sample and a replication sample of patients. Although a positive outcome was defined by at least a 50 percent reduction in symptom severity with treatment, the investigators were most interested in biomarker genes in patients who became nearly asymptomatic. The analysis yielded a strong association between one SNP located in the HTR2A gene and a positive outcome with antidepressant treatment (McMahon et al., 2006). The association was also found to be stronger in Caucasians than in African Americans, a
OCR for page 46
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary finding that might explain what has been found in STAR*D and other studies, namely that African Americans tend to have a poorer response to SSRIs than do Caucasians. Patients with the HTR2A genotype exhibited greater binding potential of the serotonin transporter with positron emission tomography (PET) scanning, which confirms the importance of the HTR2A genotype (McMahon et al., 2007). The serotonin transporter is the target molecule for most SSRIs. The transporter’s role is, in part, to reduce the concentration of serotonin in the synaptic cleft. In addition to the SNP within the HTR2A gene, two other SNPs were found to be associated with a positive treatment response. Both implicated neurotransmission-related genes: Bcl-2 and GRIK4 (Figure 3-2). Bcl-2 is an oncogene that has been shown to have neurotrophic effects in promoting cell growth and survival in neural circuits that modulate mood, motor, and cognition. In the STAR*D analysis, Manji’s team found that individuals homozygous for the good response allele FIGURE 3-2 Genes associated with treatment response. NOTE: False discovery rate (FDR); single nucleotide polymorphism (SNP); glutamate receptor, ionotropic, kainate (GRIK4); serotonin receptor 2A (HTR2A); B-cell CLL/lymphoma 2 (BCL2). SOURCE: Paddock et al., 2007.
OCR for page 47
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary were 40 percent more likely to go into full remission with treatment. The GRIK4 gene is one of several genes forming subunits of the glutamate kainate receptor, a receptor that regulates the flow of ions across neuronal membranes during excitatory neurotransmission. Considering the robust role of the Bcl-2 protein in treatment response, Manji’s team turned to animal models to corroborate its role in treatment response and to probe its role further. Preliminary studies revealed that mice lacking one copy of the Bcl-2 gene displayed less neurogenesis. Depression treatment does not take effect until several weeks after initiation of treatment. Neurogenesis is one proposed mechanism that may explain the slow time line in treatment response, based on the fact that in separate experiments with knockout mice, mice heterozygous for Bcl-2 more quickly developed a depression-like behavior known as learned helplessness after a series of repeated shocks (Yuan et al., 2007). Not only did the mice develop learned helplessness at a markedly greater rate, but they also failed to respond to chronic treatment with citalopram, compared with wild-type mice. The Bcl-2 heterozygous rats also performed worse on other well-accepted behavioral tests of depression. Manji noted that his group is now investigating the possibility of an interaction between the Bcl-2 gene and the other two genes identified in his SNP analysis. THE ROLE OF GENES IN TREATMENT RESPONSE Do genes also predict the likelihood of experiencing adverse effects with depression treatment? This is an important public health question because of the possibility that SSRIs may increase the risk of suicidal behavior in a small subgroup of children. That concern has led to a black box warning issued by the Food and Drug Administration. On the other hand, the large drop in sales of SSRIs after the warning took effect indicates that the black box warning deters physicians from prescribing SSRIs that many youngsters may desperately need (Gibbons et al., 2007). The STAR*D dataset is being used by Manji’s team to determine whether certain genes increase the risk of suicidal ideation or behavior. His group found a subgroup of patients (n = 120) who reported suicidal ideation after treatment with citalopram but none before. The SNP analysis revealed that these patients are more likely to possess one subtype of a kainate receptor (GluR6), said Manji (Lage et al., 2007). Furthermore,
OCR for page 48
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary knockout experiments are consistent with these clinical findings. GluR6 knockout mice display hyperlocomotion, aggression, and increased exploratory behaviors. Since GluR6 is also a putative bipolar susceptibility gene, these results suggest that individuals experiencing these side effects may be those with a subtle bipolar diathesis (Shaltiel et al., 2007). These behavioral findings fit with the interpretation that antidepressants, in a rare group of patients with a certain genotype, stimulate impulsive or aggressive behaviors, that, in turn, might be precursors to suicidal behavior. More research needs to be performed to explore these intriguing findings, but what has been accomplished thus far points to the seminal role that genotype may play in treatment response. The field has managed to marshal clinical findings and animal models to launch potentially the first generation of genetic predictors of treatment response for a widespread, serious, and disabling disorder. BIOMARKERS FOR SCHIZOPHRENIA Promising biomarkers are being investigated to identify and track schizophrenia’s cognitive symptoms, said Dr. David Lewis, director of the University of Pittsburgh Translational Neuroscience Program. The specific gene discussed during the workshop was CHRNA7; however, a few highly replicable risk genes have been identified for schizophrenia, including DlSC1, G72 (DAOA), and neuregulin. These genes were not discussed in the workshop and are therefore not included in the summary. Cognitive symptoms are generally underrecognized by the general public and include abnormalities in attention, verbal fluency, and working memory. The latter refers to the capacity to hold and manipulate information in the mind to guide behavior or to plan ahead. These and other cognitive characteristics are considered core symptoms of schizophrenia because they appear before disorder onset and, over the course of the disorder, are associated with the greatest level of dysfunction (Green, 1996; Heinrichs and Zakzanis, 1998). Although several effective medications have been marketed for decades to treat schizophrenia’s most recognizable symptoms—hallucinations and delusions—no medications have been developed and marketed for improving cognition. Two potential biomarkers of cognitive dysfunction rely on electrophysiology to detect patterns of activity within regions of the cerebral
OCR for page 49
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary cortex. The identification of these biomarkers has been so promising that it has already led to small clinical trials. Because the underlying defects captured by the biomarkers may actually contribute to the pathophysiology of schizophrenia before its onset, future clinical trials are also contemplated with another outcome in mind: the prevention of full-blown schizophrenia by interrupting its pathological progression. The first biomarker is one of impaired attention, assessed by the P50 evoked potential. Its purpose is to detect the ability to filter (gate) sensory stimuli, without which deficits in sustained attention are produced. People with schizophrenia describe being barraged by an onslaught of sensory stimuli, thereby finding it difficult to sustain focus on any one stimulus. The laboratory of Dr. Robert Freedman at the University of Colorado Health Sciences Center has developed a test for filtering sensory stimuli by measuring auditory evoked potentials. The test introduces a tone and measures the evoked response in the subject via a scalp electrode right after the tone and then 50 milliseconds later. In normal adults, the second identical tone presented 50 milliseconds later produces a blunted response (as measured by amplitude of evoked response) in comparison with the first. But in schizophrenia, the so-called P50 amplitude in response to the second tone is the same as the first, or sometimes is even exacerbated. A similar or even exaggerated response means that the second stimulus is perceived as being as novel as the first, suggesting a defect in sustained attention. Relatedly, animal models have provided some of the molecular underpinnings of this defect in sustained attention. Cholinergic stimulation of the alpha7 nicotinic receptor on hippocampal interneurons is essential for the P50 reduction to the second stimulus. The failure to attenuate the P50 auditory evoked response in schizophrenia is associated with a polymorphism in the gene (CHRNA7) for the alpha7 nicotinic receptor (Leonard et al., 2002). In addition, postmortem studies of people with schizophrenia reveal that alpha7 nicotinic receptor expression is reduced. Piecing these findings together, Freedman’s team undertook a small proof-of-concept clinical trial in which an alpha7 nicotinic agonist was administered to a small group of patients with schizophrenia (Olincy et al., 2006). The trial found significant improvement in P50 inhibition to the second stimulus. It also found some improvement in subjects’ performance on a test battery designed to assess neuropsychological functioning. The proof-of-concept clinical trial has galvanized efforts to conduct larger trials with nicotinic agonists.
OCR for page 50
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary Another potential biomarker for schizophrenia strives to capture electrophysiological measures of the neural abnormalities underlying the working memory impairments in the illness, said Lewis. This biomarker measures electrophysiological oscillations in the gamma band range (30 to 80 Hz) during an activity that requires working memory. Physiological activity at gamma band frequency is influenced by GABA neurons in the cerebral cortex, including one type known as the chandelier cell. A given chandelier cell supplies inhibitory input to the axon initial segment of 200 to 300 pyramidal cells and, by virtue of their connectivity and firing patterns, chandelier cells contribute to the synchronized firing of populations of pyramidal cells. In individuals with schizophrenia, chandelier cells in the dorsolateral prefrontal cortex (DLPFC) have reduced expression of GAD 67, an enzyme responsible for synthesis of GABA, and the presumed resulting deficit in GABA input leads to compensatory changes in pyramidal cell axon initial segments, including upregulation of GABA receptors that contain alpha2 subunits (Lewis, et al., 2005). Thus, the postmortem findings predict that schizophrenia would be associated with a reduced capacity to generate gamma band oscillations in the DLPFC during working memory tasks, and exactly this abnormality has been observed in clinical studies (Cho et al., 2006). Lewis explained that the findings provided the rationale for a clinical trial, now in progress, using a GABAA alpha2 selective agonist to treat the cognitive symptoms of schizophrenia and measuring gamma band activity during working memory tasks as one measure of the drug’s effectiveness. The two examples given here highlight the value of electrophysiology for biomarker development in schizophrenia. Electrophysiology’s growing value draws from advances in understanding the molecular, cellular, and circuitry disturbances present in psychiatric disorders; knowledge of the molecular, cellular, and circuitry bases for particular patterns of electrophysiological activity; and cognitive and behavioral tests used to induce those patterns of activity. This convergence of information makes it possible to identify potential drug targets that are predicted to help normalize both the patterns of electrophysiological activity and the associated cognitive performance.
OCR for page 51
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary BIOMARKERS OF ADDICTION Imaging the brain with PET affords new opportunities for finding biomarkers of addiction, said Dr. Nora Volkow, director of the National Institute on Drug Abuse. One major aim is to identify biomarkers of vulnerability, since none are currently available for clinical use. The value of PET scanning is that numerous drugs and other agents can be labeled and injected and their temporal course tracked in vivo. Establishing temporal course is key to understanding addiction vulnerability because drugs with short- versus longer-term reinforcing effects are more likely to elicit frequent administration and thereby pose greater addiction potential. Cocaine, for example, exerts its reinforcing effects and exits the brain so swiftly that it is more prone to addiction than is the psychostimulant methylphenidate, which has longer pharmacodynamics in the brain. Although both bind to the same protein—the dopamine transporter, which is responsible for reuptake of excess dopamine in the synaptic cleft—their abuse liabilities are different. PET scanning has begun to be harnessed to explore the genetic influences on the neurocircuitry underlying addiction vulnerability. The main neurocircuits depicted in Figure 3-3 are dauntingly complex, relying on multiple neurotransmitters and affecting numerous nuclei of the central nervous system (CNS). That complexity makes it unlikely that there are individual biomarkers of addiction and suggests instead that complex biosignatures will need to be identified. But, as Volkow describes, there is general agreement within the field for at least two major points. First, there are strong animal models used to complement clinical studies. Second, a large body of evidence implicates genetics in the vulnerability to addiction—perhaps accounting for 50 percent of the population variance. Genetics plays a role in developing addiction in three ways: it determines the extent to which individuals are likely to experiment with drugs; it influences drug metabolism and pharmacological response once the drug is taken; and it influences why some people become addicted and others do not, a process that depends on plasticity within neurocircuits of addiction. However, individual genes underlying vulnerability have not been identified. A striking example of genetic vulnerability—that is, how genotype may affect the likelihood of drug experimentation—comes from outside the addiction field. In a newly published study, a team of NIMH researchers used PET scanning in healthy human volunteers to investigate
OCR for page 52
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary FIGURE 3-3 Key common neurocircuitry elements in drug-seeking behavior of addiction. NOTE: Enkephalin (ENK); dopamine (DA); norepinephrine (NE); corticotropin releasing factor (CRF); β-endorphin (β-END); ventral tegmental area (VTA). SOURCE: Koob, 2006. neural mechanisms of genetic risk for impulsivity and violence (Meyer-Lindenberg et al., 2006). A common polymorphism in monoamine oxidase A (MAO-A) was found to exert profound effects on the structure and function of corticolimbic circuitry governing emotional regulation and cognitive control. The polymorphism affected the volume of gray matter of the cingulate gyrus and the amygdala. Subjects with high transcription rates of MAO-A had higher volumes, whereas those with lower transcription rates had smaller volumes. Subjects with high transcription rates also displayed a hyperresponsive amygdala and diminished reactivity of the prefrontal regions. Although the focus of the study was on impulsivity and aggression, these traits overlap with those involved in addiction. Impulsivity contributes to the likelihood of experimentation with drugs. How can expression of a single enzyme, in this case MAO-A, have such profound effects on the structure and function of certain neurocircuits? From this study and other ongoing research in the addiction field,
OCR for page 53
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary Volkow hypothesized that the polymorphism exerts its effects during brain development. The human brain has a long developmental stage, relative to other organisms, during which time it remains vulnerable to genetic or environmental insults that affect structure and function into adulthood. An example of the second role of genetics—that is, by drug metabolism and pharmacological response—comes directly from the addiction field. A large body of clinical studies, as well as preclinical research, has found that the dopamine D2 receptor is extremely important in regulating reinforcing responses to drugs of abuse. The evidence shows that drug abusers tend to have lower levels of dopamine D2 receptors (Kalivas and Volkow, 2005). The finding has been replicated in abusers of cocaine, methamphetamine, alcohol, and steroids. However, a reduced number of D2 receptors in the brain cannot be considered a biomarker of addiction because it lacks specificity; many nonabusers also have lower expression levels of D2 receptors. But it is possible that low expression levels eventually may be part of a biosignature of drug addiction, once other biomarkers are found to cluster with it, noted Volkow. Any biomarkers or biosignatures would be of special clinical utility if peripheral nervous system surrogate markers are found, which would obviate the need to sample CNS tissue. As yet, there is no peripheral surrogate marker for CNS D2 levels. Research is being done, however, to develop neurocognitive tasks that can predict expression of D2 receptors, reported Volkow. PET scanning also has revealed that the levels of D2 receptors in the striatum are linearly related to levels of brain glucose metabolism (Volkow et al., 2006). More specifically, the greater the ratio of striatal D2 receptors, the greater the glucose metabolism in the orbitofrontal cortex and cingulate gyrus (two regions of the cerebral cortex involved in salience attribution, emotional reactivity, and inhibitory control in addiction) (Figure 3-4). The finding has been replicated by three or more different groups, according to Volkow. Although it is tempting to interpret the evidence of an association as representing neuroplasticity occurring at the time of drug abuse and addiction, there is another possible interpretation, said Volkow. A new study by Dr. Eric Kandel’s laboratory raises the possibility of a neurodevelopmental effect being responsible in another disorder affecting dopaminergic function (schizophrenia). He and coauthors performed an elegant study in which they selectively overexpressed the D2
OCR for page 54
Neuroscience Biomarkers and Biosignatures: Converging Technologies, Emerging Partnerships, Workshop Summary FIGURE 3-4 Correlations between striatal D2R and brain glucose metabolism. NOTE: Orbitofrontal cortex (OFC); dopamine (DA). SOURCE: Volkow et al., 2004. transgene in the striatum during a restricted period in fetal development (Kellendonk et al., 2006). Yet they found that abnormalities in dopaminergic function in the prefrontal cortex persisted well into adulthood—long after the transgene had been turned off. This study—and others soon to be published regarding cocaine distribution to the fetus from maternal use—raises the possibility of lifelong structural and functional effects on brain neurocircuitry as a result of fetal exposure to drugs of abuse.