In Session With David A. Lewis, MD: Cognition in Schizophrenia

David A. Lewis, MD

This interview took place on July 25, 2006, and was conducted by Norman Sussman, MD.

Is schizophrenia a heterogeneous disorder, or is there a universal molecular or biochemical defect in all or most cases?

It is recognized at this point that at the etiologic level schizophrenia is clearly heterogeneous. Genetic factors account for 80% to 85% of the variance in the illness, but there is no single gene that is clearly responsible. It appears that a complex of genetic factors can give rise to schizophrenia.

In concert with that, a wide number of environmental factors, such as malnutrition during the mother’s first trimester of pregnancy and infections in the second trimester, population density of the area in which a person is raised, and cannabis use during adolescence, increase the risk of schizophrenia.

I think the question we are faced with now is whether there are some conserved molecular and biochemical changes that are common across individuals with schizophrenia. Some compelling data suggest that the number of molecular mediators of the illness may be less than the number of etiologic factors or the number of clinical manifestations. In other words, a variety of causes can converge upon a final common molecular pathway that is present across multiple brain regions. Disturbances within those different regions and the functions that they subserve give rise to the different clinical features. Such molecular changes likely include alterations in signalling via the dopamine D₁ receptor, abnormalities in signalling via the N-methyl-D-aspartate (NMDA) receptor, and changes in a subpopulation of γ-aminobutyric acid (GABA) neurons in the cerebral cortex.

In addition, Freedman and colleagues have examined an endophenotype that is mediated by the α₇ nicotinic receptor.¹ There is widespread use of nicotine among people with schizophrenia; they tend to smoke more per day and inhale more nicotine per cigarette. It is possible that individuals with the disorder are transiently self-medicating through the use of nicotine by normalizing the function of this receptor.

What kinds of cognitive dysfunction occur in patients with schizophrenia?

Studies over the last 15 years have clearly reawakened us to the idea that schizophrenia is a disorder of impairments in a variety of cognitive domains. This is not a new idea, as both Kraepelin and Bleuler focused on impairments in cognition as defining features of the illness.

At present, it is clear that virtually every individual with schizophrenia is impaired cognitively. On average, individuals with schizophrenia score between 1 and 1.5 standard deviations below the mean on cognitive tests. Twin studies from Goldberg and colleagues² have showed that in monozygotic twins discordant for the illness, the affected twin performs more poorly on cognitive tasks than the unaffected twin. A recent analysis by Keefe and colleagues³ demonstrated that virtually every individual with schizophrenia underperforms cognitively compared to predictions based on parental performance. Cognitive dysfunction is present during the prodromal and premorbid phases as well as at the first onset of psychosis. Unaffected relatives tend to show performance that is slightly lower than comparison groups. In contrast to psychosis, which can wax and wane over the course of the illness, cognitive impairments tend to be persistent.

There is a lot of debate about whether cognitive impairments are progressive. Some literature suggests that there is cognitive decline early in the course of the illness, while other literature suggests that these deficits are fairly consistent across the course of the illness.

There are a variety of domains of cognition that are impaired in schizophrenia. Attention, working memory, and episodic memory are more impaired, whereas general semantic knowledge and visual perception tend to be less impaired. Often, once psychosis is stabilized and individuals are able to live outside of the hospital, they still have substantial difficulties returning to school or getting and holding a job. Over a decade ago, Green⁴ clearly pointed out that the degree of cognitive impairment in individuals with schizophrenia is the best predictor of their long-term outcome for quality-of-life measures.

What is working memory, and what is its significance in people with schizophrenia?

Working memory is one of the domains of cognition that is impaired in schizophrenia. It can be thought of as the ability to transiently maintain and manipulate a limited amount of information in order to guide thought or behavior. A typical example in daily life is dialing a new phone number. A person keeps it in mind, places a phone call, and then no longer remembers the number. Working memory can be assessed clinically through the use of digit recall and in particular reversal of digits. An individual is given a string of digits and asked to repeat them in the reverse order. This involves both maintenance, where the person has to keep the information in mind, and manipulation, where the person has to reverse the order to get the task correct.

These functions are clearly disturbed in individuals with schizophrenia. The disturbance reflects or is associated with altered activation of the dorsal lateral prefrontal cortex, which is the region of the brain that is above and immediately behind the orbits. This impairment in activation is relatively specific to the syndrome of schizophrenia. Individuals who are in their first episode of psychosis who are subsequently diagnosed with schizophrenia perform poorly on this task and show impaired activation of the prefrontal cortex, whereas individuals who cross-sectionally appear no different but who subsequently turn out to have a nonschizophrenic psychosis, usually bipolar disorder, are not impaired or at least not as severely impaired. Likewise, individuals with major depressive disorder do not show this pattern of cognitive abnormalities. The abnormalities are disruptive in many aspects of life, including being able to follow a logical sequence of conversation or communication. If a person cannot appropriately maintain, use, and then erase information, he or she will have problems with thinking, speech production, and interpreting verbal information from others.

What is the relationship between the frontal cortex and schizophrenia?

The dorsal lateral prefrontal cortex is part of a broad network that involves connections with more posterior regions of the cortex, the parietal and temporal lobes, and subcortical structures like the thalamus and caudate nucleus. There are also relationships to the hippocampus, and therefore we cannot ascribe schizophrenia to an isolated abnormality in a particular brain region. It is really the network activity that gives rise to any normal brain function or any disturbed function in the illness. Some of these brain regions may be more disturbed or their disturbance may be more evident under certain conditions.

The individual components of the circuitry of the prefrontal cortex can be divided into three types. The first component includes excitatory pyramidal neurons that utilize the neurotransmitter glutamate. These are the most numerous class of neurons in the cortex and they are the principal conveyors of information. Their axons communicate with other cells within the prefrontal cortex, and project from the prefrontal cortex to other brain regions. The second class of neurons in the cortex utilize GABA, provide inhibitory control of the excitatory neurons, and play a very important role in the timing of neuronal activity. The inhibitory neurons are critical in determining when pyramidal cells fire and whether they fire in synchrony. The third type of information processing unit in the cortex includes projections from other regions. These are basically divided into two categories—excitatory and modulatory inputs. Excitatory inputs from the thalamus and other cortical regions convey specific information. Modulatory inputs tend to shape activity within the prefrontal cortex. These include inputs that contain the neurotransmitters dopamine, norepinephrine, serotonin, and acetylcholine. Each of these three categories are critical for the normal function of the prefrontal cortex.

There is evidence for disturbances in each of these systems in schizophrenia. For any disturbance in dopamine, glutamate, and GABA neurotransmission, the critical goal is to determine if that particular change is a cause, consequence, compensation (ie, something that the brain is trying to do to make up for deleterious effects), or a confound. A confound is something that is related to the treatment, to the abuse of substances, or other factors unrelated to the disease process.

Based upon the existing literature, one hypothetical scenario suggests that certain risk genes such as the gene neuregulin in schizophrenia could induce, through a cascade of events, hypofunction of the NMDA receptor for glutamate. That in turn seems to be associated with the types of changes in GABA neurons that are seen in schizophrenia. These GABA neurons are important for creating the type of network activity among neurons, particularly the capacity of neurons to be synchronized to oscillate at the frequency that is necessary for working memory function.

What are the implications for treatment strategies?

Though the NMDA receptor is disturbed in schizophrenia, the alteration in GABA neurons might be the most proximal event to the impairment in working memory. Thus, a useful therapeutic target would be to reduce the deficit in the GABA input.

The development of rational forms of treatment intervention for cognitive dysfunction in schizophrenia really depends upon the acquisition of additional information that reveals which of these abnormalities are in fact part of the disease process, how they relate to each other, and which ones are most closely related to the pathophysiology and hence represent the most reasonable target.

Could your findings lead to better interventions in children and adolescents at risk by virtue of family history?

One could argue that it makes sense to treat individuals who are at risk for schizophrenia before they become psychotic by targeting the cognitive impairments associated with the illness. The aforementioned types of rational drug development might in fact have implications for early intervention if enhancing cognitive function prevents or at least delays the onset of psychosis in a way that would be better than the use of antipsychotics.

A number of universities, including the University of Pittsburgh, Harvard, Yale, and the University of California, have developed specialized clinics for the treatment of early onset of psychosis. Individuals who are at the onset of psychosis have particular needs in terms of dealing with the impact that the experience of psychosis has on their lives. These patients need clinicians specialized in recognizing that they are a bit more impacted by the effects of antipsychotics and who will manage antipsychotics appropriately to try to minimize the adverse effects. Clinicians can also try to reduce the loss of educational opportunity as a result of the onset of the illness.

At this point, the therapy used is driven by an appreciation of the distinctive clinical and social components of this stage of the illness, rather than driven by an understanding of the basic biology. We are still waiting to translate the biologic findings into new treatments.

Does using antipsychotics early in the course of illness have a stabilizing effect?

While there are certainly studies that show that untreated psychosis has a negative impact on long-term outcome, other data do not implicate psychosis as a toxic factor in producing a worse outcome. On the other hand, some emerging data raise concerns about whether at least certain antipsychotics early in the course of the illness may have some detrimental effects on cognition.

If you have any questions and/or comments regarding the interview with David A. Lewis, MD, please submit a “Letter to the Editor” to Norman Sussman, MD, at [email protected].

Disclosure: Dr. Lewis is a consultant to AstraZeneca, Eli Lilly, Merck, Pfizer, and Sepracor; and receives grant support from the Bristol-Myers Squibb Foundation, Eli Lilly, Merck, the National Institutes of Health, and Pfizer.

References

1. Martin LF, Kem WR, Freedman R. Alpha-7 nicotinic receptor agonists: potential new candidates for the treatment of schizophrenia. Psychopharmacology (Berl). 2004;174(1):54-64.
2. Goldberg TE, Ragland JD, Torrey EF, Gold JM, Bigelow LB, Weinberger DR. Neuropsychological assessment of monozygotic twins discordant for schizophrenia. Arch Gen Psychiatry. 1990;47(11):1066-1072.
3. Keefe RS, Eesley CE, Poe MP. Defining a cognitive function decrement in schizophrenia. Biol Psychiatry. 2005;57(6):688-691.
4. Green MF. What are the functional consequences of neurocognitive deficits in schizophrenia? Am J Psychiatry. 1996;153(3):321-330.