ICOSR 2011—Cognition in Schizophrenia: Biology and Therapy
As part of our ongoing coverage of the 2011 International Congress on Schizophrenia Research (ICOSR), 2-6 April, in Colorado Springs, Colorado, we bring you session summaries from some of the attendees. For this report, we thank Rajiv Radhakrishnan of Yale University School of Medicine, New Haven, Connecticut.
15 May 2011. Cognitive deficits are considered to be a core feature of schizophrenia. The cognitive deficits seen in patients with schizophrenia include lower IQ and deficits in working memory, executive functioning, verbal memory, cognitive flexibility, planning, and social cognition. The relevance of cognitive impairment in schizophrenia is underscored by the fact that cognitive symptoms are a better predictor of disability and vocational functioning than positive symptoms. Unfortunately, currently available antipsychotic drugs (which are predominantly dopamine D2 receptor antagonists) have limited efficacy against cognitive deficits. Hence, there is a need to develop new treatments to target these symptoms. A morning session on Monday, 4 April 2011, at the International Congress on Schizophrenia Research, chaired by Oliver Howes from the Institute of Psychiatry at Maudsley Hospital, London, addressed both biological substrates and treatment of cognitive symptoms in schizophrenia.
Linda Scoriels from the University of Cambridge, U.K., opened the session with an overview of the pro-cognitive agents which have been tried in schizophrenia as adjuncts to typical and atypical antipsychotics, including clozapine. Among these are cholinergic drugs such as donepezil, galantamine, and rivastigmine, and noradrenergic drugs such as clonidine, guanfacine, and modafinil. These agents have shown small to modest effects on specific domains of cognition, but the effects are neither robust nor sustained. The focus of the talk turned to modafinil, a drug approved by the FDA for the treatment of narcolepsy, shift work sleep disorder, and excessive daytime sleepiness associated with obstructive sleep apnea. Although the mechanism of action of modafinil is not clearly understood, the drug is found to affect various neurotransmitter systems including GABA, glutamate, dopamine, noradrenaline, serotonin, and orexin in the hypothalamus, cortex, and basal ganglia. Modafinil also inhibits MPTP-induced increase in GABAA receptors in the internal globus pallidus, indicating that it may be neuroprotective (Zeng et al., 2004). In clinical studies, modafinil was found to improve attention, concentration, memory, learning, and inhibitory control. An earlier study of the use of modafinil in schizophrenia (Turner et al., 2004) demonstrated a significant improvement in sustained attention (digit span-forwards and -backwards), set-shifting (extra-dimensional shift errors on IDED task) and working memory (move count and latency on the Tower of London task).
Scoriels went on the present data from a randomized, double-blind, placebo-controlled, crossover design trial of adjunctive modafinil in first-episode psychosis. The study comprised 40 patients with first-episode psychosis who were randomized to receive a single dose of modafinil (200 mg) or placebo followed a week later by a single dose of placebo or modafinil (200 mg). Modafinil was found to improve sustained attention (digit span-backwards), spatial working memory, and inhibition control. Patients were better able to detect “sad” faces on an emotional face recognition task, but modafinil had no effect on subjective mood ratings, emotional sensitivity to reward or punishment, or on interference of emotional valence on cognitive function, as measured by the affective go-no go task (Scoriels et al., 2011). The use of modafinil was well tolerated and did not exacerbate psychotic symptoms.
The next to take the podium was Oliver D. Howes, whose talk focused on the neurobiology of cognitive impairment in the prodrome to schizophrenia. We know from functional imaging studies that patients with schizophrenia and healthy controls activate a similar neural network during executive task performance, the difference being in the degree of activation (Minzenberg et al., 2009). Since cognitive deficits are known to predate the onset of psychotic symptoms, it is interesting to know if this pattern is also seen in ultra-high-risk (UHR) or at-risk mental state (ARMS) subjects. Howes went on to present data on functional imaging during executive task performance in a sample of UHR subjects versus healthy controls. UHR subjects showed increased activation in the inferior frontal gyrus during a verbal fluency task and decreased activation of the middle frontal gyrus, medial frontal lobe, and superior parietal lobule during an N-back task, reminiscent of findings frequently seen in schizophrenia.
Are the neurotransmitter abnormalities seen in schizophrenia also seen in UHR? Howes turned his focus to dopamine and the finding that patients with schizophrenia had increased presynaptic dopamine (Abi-Dargham et al., 1998). A similar finding of increased striatal dopamine synthesis has been seen in the UHR group (Howes et al., 2009) using positron emission tomographic (18)F-dopa imaging. In a follow-up of the subjects, it was seen that only UHR subjects who had shown an increase in striatal dopamine synthesis went on to develop psychosis. The increase in striatal dopamine may be a marker of conversion to psychosis, but these findings await replication.
Striatal dopamine has also been linked with dorsolateral prefrontal cortex (DLPFC) function in schizophrenia. Increased striatal dopamine is associated with decreased blood flow in the DLPFC (Meyer-Lindenberg et al., 2002) and an inverse relationship has been shown between 11C raclopride binding in striatum and DLPFC (Bertolino et al., 2000), suggesting a functional link between the two regions. Howes presented data from his group which showed that the decreased activation during the N-back task in the UHR group was inversely correlated with an increase in striatal dopamine (Fusar-Poli et al., 2010), and increased striatal dopamine was negatively correlated with performance on a verbal fluency task. Striatal [18F]DOPA uptake is also found to be negatively correlated with hippocampal glutamate levels in UHR subjects (Stone et al., 2010). The striatum may hence be important in mediating cognitive processes—more so in the context of psychosis, spanning from ultra-high-risk states to schizophrenia.
The third speaker of the morning was Cameron Carter from University of California, Davis, who spoke on neurofunctional substrates of impaired higher cognition in schizophrenia. Although cognitive dysfunction in schizophrenia is viewed as multiple non-overlapping domains of impairment, it can also be viewed as a singular dysfunction of “impaired cognitive control.” Functional imaging studies show that patients with schizophrenia and healthy controls recruit a similar network of brain areas during task performance, but patients show decreased activation in the DLPFC and increased activation in ventrolateral areas compared to controls (Minzenberg et al., 2009). Carter described studies demonstrating this deficit in prefrontal function during three different cognitive tasks: that of memory encoding and retrieval, emotion processing, and language comprehension. A predominant prefrontal dysfunction was also seen during an episodic memory task in which patients with schizophrenia showed less prefrontal activation than did controls in the frontal pole, dorsolateral, and ventrolateral prefrontal cortex during encoding, and the dorsolateral prefrontal cortex and ventrolateral prefrontal cortex during retrieval (Ragland et al., 2009).
Carter went on to describe a slow-event related fMRI paradigm which his group has used to study emotional processing in schizophrenia. Patients with schizophrenia and healthy controls were required to view an affective picture and, after a delay, report their emotional experience. Patients and controls activated brain regions to a similar extent while viewing the affective picture. However, during the delay, patients showed decreased activation in the DLPFC and other prefrontal, limbic, and paralimbic areas (Ursu et al., 2011). The study supports the idea that a deficit in prefrontal circuitry was related to deficits in emotion processing.
In the study of language comprehension, Carter described a semantic priming task where the target word was preceded by a semantically primed word or a sentence to generate a coherent discourse. Patients with schizophrenia showed a smaller N400 amplitude for the discourse compared to controls, but not for words. This supports the idea of the deficit being at the level of integration of cortical activity rather than a specific deficit in one functional brain region.
Matcheri Keshavan and Shaun Eack from Harvard Medical School took the podium next to talk about the effects of treatment on cognition and brain structure in schizophrenia. Keshavan reiterated that cognitive deficits in schizophrenia were present in the prodromal phase, were pervasive and persistent, progress early, and predict functional outcome. Atypical antipsychotics have limited efficacy against cognitive deficits with the possibility that improvement seen in trials is a result of practice effects (Goldberg et al., 2007). Cognitive enhancement therapy (CET), originally developed by the late Gerard Hogarty and Samuel Flesher at the EPICS Program at the University of Pittsburgh Medical School, consists of repetitive, multimodal, computer-based exercises which are individualized and use systematic and sequential training of bottom-up processing and of top-down processing in parallel. The "Cadillac" approach consists of 1-1.5 hours per week of neurocognitive retraining, followed by 1.5 hours per week of social-cognitive group therapy focusing on "gistfulness," non-verbal communication, and emotion perception. In a two-year randomized controlled trial of CET versus enriched supportive therapy in schizophrenia, robust improvement was seen on neurocognition and processing speed composites and in measures of cognitive style, social cognition, and social adjustment (Hogarty et al., 2004). These effects were found to be durable in a one-year follow-up (Hogarty et al., 2006). There is some evidence that CET works better for those earlier in the course of the illness, and this may be because CET can capitalize on neurobiological reserve which is still unaffected by the pathological process of psychosis (Keshavan and Hogarty, 1999).
In a two-year randomized controlled trial of CET versus enriched supportive therapy in early-course schizophrenia, modest improvement on neurocognition and significant improvement on measures of cognitive style, social cognition, and social adjustment were observed with CET. The CET training was also found to translate to lower negative symptom scores (Eack et al., 2009). These improvements were found to be sustained at one-year follow-up (Eack et al., 2010a) and also translate to improved functioning (Eack et al., 2010b). The findings were accompanied by changes in brain morphology functioning. CET was associated with greater preservation of gray matter volume over two years in left fusiform gyrus, left hippocampus, and left parahippocampal gyrus as measured using structural MRI. The increase in the gray matter volume in left amygdala may indicate that CET is neuroprotective. Increase in gray matter density was associated with an improvement in social cognition. An analysis of functional imaging data during the preparing-to-overcome-prepotency (POP) task revealed increased activity in the ventrolateral portion of the DLPFC associated with improved performance and increased differential connectivity of the orbitofrontal cortex. Hence, CET shows promise as a treatment strategy in improving social cognition and functioning.—Rajiv Radhakrishnan.