Failure to deactivate medial prefrontal cortex in people at high risk for psychosis

G Model

EURPSY-3241; No. of Pages 8 European Psychiatry xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

European Psychiatry journal homepage: http://www.europsy-journal.com

Original article

Failure to deactivate medial prefrontal cortex in people at high risk for psychosis I. Falkenberg a,b,*, C. Chaddock a, R.M. Murray a, C. McDonald d, G. Modinos a, E. Bramon a,e, M. Walshe a, M. Broome a,c, P. McGuire a, P. Allen a a

Department of Psychosis Studies, Institute of Psychiatry, King’s College London, London, United Kingdom Department of Psychiatry and Psychotherapy, Philipps-University of Marburg, Marburg, Germany Division of Mental Health and Wellbeing, Warwick Medical School, University of Warwick, Gibbet Hill, Coventry, CV4 7AL, United Kingdom d Department of Psychiatry, Clinical Science Institute, National University of Ireland, Galway, Galway, Ireland e Department of Clinical Neuroscience, Institute of Psychiatry, King’s College London, London, United Kingdom b c

A R T I C L E I N F O

A B S T R A C T

Article history: Received 27 January 2015 Received in revised form 4 March 2015 Accepted 5 March 2015 Available online xxx

Impaired working memory is a core feature of schizophrenia and is linked with altered engagement the lateral prefrontal cortex. Although altered PFC activation has been reported in people with increased risk of psychosis, at present it is not clear if this neurofunctional alteration differs between familial and clinical risk states and/or increases in line with the level of psychosis risk. We addressed this issue by using functional MRI and a working memory paradigm to study familial and clinical high-risk groups. We recruited 17 subjects at ultra-high-risk (UHR) for psychosis, 10 non-affected siblings of patients with schizophrenia (familial high risk [FHR]) and 15 healthy controls. Subjects were scanned while performing the N-back working memory task. There was a relationship between the level of task-related deactivation in the medial PFC and precuneus and the level of psychosis risk, with deactivation weakest in the UHR group, greatest in healthy controls, and at an intermediate level in the FHR group. In the highrisk groups, activation in the precuneus was associated with the level of negative symptoms. These data suggest that increased vulnerability to psychosis is associated with a failure to deactivate in the medial PFC and precuneus during a working memory task, and appears to be most evident in subjects at clinical, as opposed to familial high risk. ß 2015 Published by Elsevier Masson SAS.

Keywords: Psychosis Working memory Cognition UHR Default mode network fMRI

1. Introduction Impaired working memory (WM) is a robust feature of schizophrenia and is thought to reflect prefrontal cortex (PFC) dysfunction [27]. Although much attention has been devoted to altered engagement of the lateral PFC, presenting as both hypo- and hyperfrontality [15], the disorder is also associated with a failure to deactivate the medial PFC (mPFC) during task performance. The latter is thought to be associated with a dysfunction of the default mode network (DMN) [17], a network of brain regions that are active during a rest or baseline condition and usually deactivate during engagement in a cognitive tasks [5]. PFC dysfunction has also been reported in people at familial high-risk for schizophrenia (FHR) [39] and in people at clinical or ultra-high-risk (UHR; * Corresponding author. Department of Psychosis Studies (PO67), Institute of Psychiatry, King’s College London, 16, De Crespigny Park, London SE5 8AF, United Kingdom. Tel.: +44 20 7848 0801; fax: +44 20 7848 0976. E-mail address: [email protected] (I. Falkenberg).

individuals who present with attenuated psychotic symptoms and/ or a decline in function) for psychosis [1,4,28]. However, it is not clear to what extent PFC dysfunction reflects varying levels of psychosis risk mediated via genetic, non-genetic or illness-related (e.g. symptoms, medication) factors. Moreover, while both FHR and UHR individuals have an increased vulnerability to psychosis compared to the general population, the risk of subsequently developing the disorder is substantially greater in the UHR cohorts (approximately 30% over the next two years [13], compared to 6– 13% over their lifetime in FHR cohorts [16,29]). Whether mPFC dysfunction increases with the level of psychosis risk is not clear. The aim of the present study was to use functional magnetic resonance imaging (fMRI) to explore PFC (dys-)function in relation to different levels of psychosis risk, independent of the manifestation of the disorder. We measured PFC activation during a working memory task in subjects at UHR for psychosis, FHR subjects and healthy control subjects. We first tested the hypothesis that PFC dysfunction would be evident in all people at risk of psychosis (i.e. FHR and UHR groups). We then

http://dx.doi.org/10.1016/j.eurpsy.2015.03.003 0924-9338/ß 2015 Published by Elsevier Masson SAS.

Please cite this article in press as: Falkenberg I, et al. Failure to deactivate medial prefrontal cortex in people at high risk for psychosis. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2015.03.003

G Model

EURPSY-3241; No. of Pages 8 I. Falkenberg et al. / European Psychiatry xxx (2015) xxx–xxx

2

the At Risk Mental State [43] and a consensus meeting with the clinical team. All UHR subjects met APS criteria, 3 subjects met both APS and BLIPS criteria and 1 subject met APS criteria and also had a family history of psychosis. UHR subjects’ mean Global Assessment of Functioning scores (GAF) [19] was 57.5 (12.1, range: 35–75). All subjects were drug-naive at the time of scanning. The subjects were representative of the local population of people presenting with an UHR state of psychosis in terms of age, gender, ethnicity, duration and intensity of symptoms [3]. Over the 2-year follow-up period at the OASIS clinic, 3 of the UHR subjects (18%) developed frank psychosis.

explored whether PFC dysfunction increased in line with psychosis risk by comparing FHR and UHR groups. 2. Methods 2.1. Participants The study was approved by the local research ethics committee and was conducted in accordance with the Declaration of Helsinki [41]. All participants gave written informed consent. The sample consisted of 17 ultra-high-risk subjects (UHR), 10 subjects with a familial high-risk for psychosis (non-affected siblings of patients with schizophrenia; FHR) and 15 healthy control subjects (CTRL) without any personal or family history of psychiatric disorders. UHR and CTRL samples have previously been reported by Broome et al. [4]. FHR participants were recruited as part of the Maudsley Family Study [23] but fMRI data (N-back task) from this group has not been used in any previous studies. All participants were righthanded [7] and native speakers of English. Demographic and clinical details (Positive and Negative Symptoms Scales; PANSS [21]) and IQ estimates (assessed using the National Adult Reading Test; NART [30]) are provided in Table 1.

2.1.2. Familial high-risk group (FHR) Subjects in this group were non-affected, non-help-seeking siblings of multiply affected patients with schizophrenia, who participated in the Maudsley Family Study [23]. Subjects were from families in which the index patient had at least one first- or second-degree relative affected with schizophrenia, another nonorganic psychotic disorder, or schizotypal disorder. Apart from their affected sibling, the FHR subjects thus had at least one other first- or second-degree relative with a psychotic disorder, indicating a relatively high putative genetic load. The subjects were recruited through voluntary support groups or by direct referral by the affected sibling’s consultant psychiatrist. Additionally, recruitment advertisements were placed in the newsletters of national and international schizophrenia support groups. Structured diagnostic interviews were performed to enable DSM-IV diagnoses (see [23] for details). Of the 10 unaffected relatives of schizophrenia patients, 5 relatives fulfilled criteria for lifetime DSM-IV axis 1 disorder, 3 for major depressive disorder (DSM-IV: 296.20–296.30) and 2 for anxiety and panic disorders (DSM-IV: 300.01). None of the FHR group met criteria for any schizophrenia spectrum disorder or had experienced a recent decline in functioning. One FHR subject was taking an antidepressant (amitryptiline 50 mg/day), all other FHR subjects were medication free.

2.1.1. Ultra-high-risk group (UHR) Subjects meeting the following criteria for the UHR state of psychosis [42] were recruited from Outreach and Support in South London (OASIS) [3]: age between 18 and 35 years, help-seeking, meet the criteria for one or more of the following three groups: group 1: attenuated psychotic symptoms (APS), group 2: brief limited intermittent psychotic symptoms (BLIPS; a history of one or more episodes of frank psychotic symptoms resolving spontaneously within 1 week in the past year), group (3) a recent decline in function, together with either the presence of schizotypal personality disorder or a family history of psychosis in a firstdegree relative. The diagnosis was based on assessment by two experienced clinicians using the Comprehensive Assessment for

Table 1 Demographic and psychopathological characteristics. UHR (n = 17)

FHR (n = 10)

Controls (n = 15)

Statistic

Age Mean (SD) Range

24.3 (4.2) 20–34

40.3 (10.7) 29–59

25.6 (4.8) 19–35

F(2, 39) = 21.6a

IQ Mean (SD) Range

101.7 (11.7) 95.5–108.0

111.9 (7.5) 105.0–118.8

123.2 (16.2) 110.7–135.7

F(2, 39) = 15.7a

Gender (n, female)

5

5

5

X2(2) = 1.2 (n.s.)

Ethnicity (%) White Black Oriental Mixed Other

76.5 11.8 11.8 0 0

100 0 0 0 0

66.7 26.7 0 0 6.7

X2(6) = 8.7 (n.s.)

PANSS score (mean, SD) Positive Negative General

11.8 (3.3) 11.6 (5.0) 25.3 (7.7)

10.9 (9.9) 7.2 (0.7) 17.2 (5.3)

CAARMS score (mean, SD) Disorders of thought content Range Perceptual abnormalities Range Disorganized speech Range

5.9 (6.3) 2–23 2.2 (1.5) 0–4 2.1 (1.5) 0–4

a b

F(1,23) = 0.1 (n.s.) F(1,23) = 6.5b F(1,23) = 7.7b

P  0.001. P < 0.05.

Please cite this article in press as: Falkenberg I, et al. Failure to deactivate medial prefrontal cortex in people at high risk for psychosis. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2015.03.003

G Model

EURPSY-3241; No. of Pages 8 I. Falkenberg et al. / European Psychiatry xxx (2015) xxx–xxx

2.1.3. Control group (CTRL) Healthy participants were recruited via advertisements in the local media. All individuals lived in the same borough of London as the clinical participants. There were no differences between the UHR, FHR and CTRL groups for gender and ethnicity but the groups did differ in terms of age and IQ estimates (Table 1). The FHR group was significantly older than the UHR and CTRL groups (P < 0.001, Bonferroni corrected), and IQ estimates were lower in the UHR and FHR groups compared to the CTRL group (P < 0.001, P = 0.04, Bonferroni corrected; Table 1). There was no significant difference in IQ between the UHR and the FHR group. Age was used as a covariate in all subsequent behavioural and functional imaging analyses. As an impaired intellectual ability may be phenotypic for psychotic illness [34] and familial and clinical risk states [9,14], we did not covary for premorbid differences in subsequent analyses. 2.2. Clinical assessment and IQ estimates Prior to scanning, all subjects underwent a semi-structured interview to obtain information about their family and personal psychiatric history and their current and past medication use. The following instruments were used to assess UHR and FHR subjects: the Positive and Negative Syndrome Scale (PANSS) [21] and a structured questionnaire to exclude abuse of illicit substances and alcohol (adapted from the Cannabis Experiences Questionnaire [2] and also administered to CTRL subjects). The positive symptom scores of the PANSS did not differ between UHR and FHR subjects, however, the UHR subjects scored significantly higher on the negative (P = 0.02, Bonferroni corrected) and general psychopathology subscale (P = 0.01, Bonferroni corrected) of the PANSS. UHR subjects were assessed with the Comprehensive Assessment of At-Risk Mental States (Table 1) [43]. All subjects were assessed with the National Adult Reading Test [30] and the Lateral Preferences Inventory [7]. 2.3. fMRI-task and image acquisition Images were acquired on a 1.5-T Signa (GE Healthcare, Milwaukee, Wisconsin) system at the Maudsley Hospital, London, England. T2*-weighted images were acquired using a gradient echo sequence (repetition time = 2000 milliseconds and echo time = 40 milliseconds) with 7-mm slices and a 0.7-mm gap in 14 axial planes. Images were acquired while subjects performed an N-back working memory task. Subjects were presented with series of letters that they viewed using a prismatic mirror. The interstimulus interval was 2 seconds. During the 0-back condition, subjects were required to move a joystick to the left when the letter X appeared. During 1-back and 2-back conditions, participants were required to press a button on the joystick with their right index finger if the currently presented letter was the same as that presented 1 or 2 trials beforehand, respectively. The 3 conditions were presented in 10 alternating 30-s blocks, matched for the number of target letters per block (i.e., 2 or 3) in a pseudorandom order. Reaction time and response accuracy were recorded online. To facilitate anatomical localization of activation, a high-resolution inversion recovery image data set was also acquired, with 3-mm contiguous slices and an in-plane resolution of 3 mm (repetition time = 1600 milliseconds, inversion time = 180 milliseconds, echo time = 80 milliseconds). 2.4. fMRI data analysis Functional MRI data were analyzed using Statistical Parametric Mapping (SPM8, Wellcome Department of Cognitive Neurology, London, England) running in MATLAB 7.1 (The Math-Works,

3

Natick, Massachusetts). All volumes in each subject were realigned to the first volume, normalized to a standard MNI template using non-linear-basis functions, and spatially smoothed with an 8 mm full width at half maximum isotropic Gaussian kernel. For each subject mean head, motion (in mm) was evaluated as determined by the realignment procedure. An event-related analysis was performed on the block-design-acquired data [24] which allowed us to model error trials (either missed targets or wrong nontargets) separately. Five experimental conditions comprising the target and non-target events in each of the three task conditions (0-back, 1-back, and 2-back) plus instructions and error trials were modeled by convolving their respective onset times with a canonical hemodynamic response function. A general linear model was used to calculate the parameter estimates for all brain voxels, and contrasts were created for each subject comparing non-target events while performing 1-back and 2-back tasks versus 0-back condition. Non-target (rather than target) events were chosen as their cognitive components of updating and maintaining a memory set allowed us to assess working memory components in the absence of the motor responses that occur for a button press. A second level full factorial ANOVA was specified to perform group comparisons (UHR versus CTRL, FHR versus CTRL, UHR versus FHR, UHR versus FHR versus CTRL) during combined 1 + 2back trials versus 0-back trials. Results are reported at P = 0.001 (uncorrected) in Table 2. Only results surviving P < 0.05 corrected for family wise error (FWE) and an extent threshold of 10 voxels are discussed. 2.5. Analysis of behavioural data A repeated measures ANCOVA was performed to test for differences in task performance between groups. Age was included as a covariate into the analysis and all P-values were Bonferroni corrected. As the data did not meet the assumption of sphericity, a Greenhouse-Geisser correction was applied. This procedure gives rise to a correction factor (e) indicating the degree to which sphericity has been violated. e is used to correct the degrees of freedom of the F-distribution in order to produce a more accurate significance value. Thus, the Greenhouse-Geisser correction compensates for the fact that the repeated measures ANCOVA is too liberal when sphericity is violated.

3. Results 3.1. Behavioural results The main effect of group (F = 1.19, df = 2, P = 0.31), task (F = 0.63, df = 4, P = 0.64) and the group by task interaction (F = 1.13, df = 7, P = 0.35) were all non-significant. PANSS positive symptom scores did not differ between UHR and FHR groups. PANSS negative and general psychopathology scores were significantly higher in UHR compared to FHR subjects (Table 1). 3.2. fMRI results No group differences were found in the extent of head motion during scanning (X: F = 0.69, P = 0.27, Y: F = 0.90, P = 0.42, Z: F = 0.32, P = 0.73). Across all participants, performance of the Nback task (1-back and 2-back > 0-back) was associated with activation in a bilateral fronto-parietal network, as well as the left precuneus, cingulate gyrus and thalamus (Table 2). Task-related deactivation (1-back and 2-back < 0-back) was seen in the left posterior cingulate gyrus (Table 2) although this cluster did not survive FWE correction.

Please cite this article in press as: Falkenberg I, et al. Failure to deactivate medial prefrontal cortex in people at high risk for psychosis. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2015.03.003

G Model

EURPSY-3241; No. of Pages 8 I. Falkenberg et al. / European Psychiatry xxx (2015) xxx–xxx

4

Table 2 Effects of group and risk type on activation during N-back. Group effects

Positive effect of task (1 & 2-back > 0-back) Medial frontal gyrus Precuneus Middle frontal gyrus Precentral gyrus Inferior parietal lobule Inferior frontal gyrus Inferior frontal gyrus Thalamus Cingulate gyrus

MNI coordinates Side

X

Y

Z

Peak Z-score

Cluster size

L L R R R L R L L

0

14 48 4 8 46 28 18 10 30

54 40 54 32 44 6 4 18 32

6.77 6.01 5.97 5.79 4.93 4.89 4.84 4.74 4.59

686* 479* 719* 882* 60* 36* 15* 48* 12*

28 32 40 38 30 56 14 4

Negative effect of task (0-back > 1 & 2-back) Posterior cingulate gyrus L

2

56

26

3.29

42

UHR > CTRL Lingual gyrus Precuneus

R L

26 14

64 48

2 38

4.60 4.28

450* 152

FHR > CTRL Middle frontal gyrus Fusiform gyrus Superior parietal lobule Inferior parietal lobule Inferior parietal lobule

R R L L R

28 40 28 38 32

4 70 56 46 54

56 12 40 42 42

4.20 3.84 3.68 3.66 3.41

61 24 38 20 30

CTRL > FHR Postcentral gyrus

R

52

26

22

3.53

23

UHR > FHR Cingulate gyrus Posterior cingulate Medial frontal gyrus Inferior parietal lobule Lingual gyrus Insula

L L L R R R

14 14 2 52 16 34

46 58 54 30 58 24

38 8 22 24 0 20

4.16 4.12 3.94 3.80 3.76 3.18

194 148 280* 71 159 11

FHR > UHR Middle frontal gyrus Inferior parietal lobule Middle frontal gyrus Middle frontal gyrus

R L L R

30 38 26 46

4 50 4 28

56 40 56 30

4.76 4.03 3.68 3.28

107 95 29 17

UHR > FHR > CTRL Precuneus Lingual gyrus Medial frontal gyrus

L R L

14 22 14

48 66 58

38 0 18

4.70 4.51 3.84

542* 494* 294*

CTRL > UHR No supra-threshold effect

CTRL > FHR > UHR No supra-threshold effect Results are reported at a threshold of P < 001 uncorrected. * P  0.05 FWE.

3.2.1. Group effects (P < 05 FWE) UHR subjects also showed greater activation than CTRL in the right lingual gyrus. The inverse contrast did not reveal significant results. No significant differences in activation were found between FHR and CTRL subjects. The UHR group showed significantly greater activation than FHR subjects in the left medial frontal gyrus. The FHR showed greater task-related activation in the bilateral lateral PFC relative to the UHR group but these results did not survive FWE correction (Table 2). A linear analysis (UHR > FHR > CTRL) showed that activation in the left medial frontal gyrus, left precuneus extending to the left posterior cingulate cortex and right lingual gyrus decreased in line with the psychosis risk i.e. activation was greatest in the UHR group. Deactivation in these regions was seen in the CTRL and FHR groups (Table 2, Fig. 1a). PANSS negative symptom scores were significantly correlated with activation in the left precuneus (rs = 0.48, P = 02; Fig. 2). There was also a trend towards a negative

correlation between activation in the left precuneus and GAF scores in the UHR groups (r = 0.45, P = 0.068).

4. Discussion In the present study, we compared PFC function during a WM task in individuals at UHR and FHR of psychosis and healthy controls. Contrary to our first hypothesis, when we directly compared UHR and FHR groups to healthy controls no group effects were seen in PFC that survived correction for multiple comparisons, although increased mPFC activation was seen in the UHR relative to the FHR group. However, a comparison of all three groups did reveal an effect in the mPFC. In this region, the taskrelated deactivation seen in healthy controls was diminished in FHR subjects and reversed in UHR subjects (Fig. 1b). Similar effects were also observed in the precuneus and lingual gyrus.

Please cite this article in press as: Falkenberg I, et al. Failure to deactivate medial prefrontal cortex in people at high risk for psychosis. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2015.03.003

G Model

EURPSY-3241; No. of Pages 8 I. Falkenberg et al. / European Psychiatry xxx (2015) xxx–xxx

5

Fig. 1. a: statistical parametric maps (SPMs) showing activation (yellow) and deactivation depending on risk of transition (UHR > FHR > CTRL). Areas of significance are reported at P < 05 (FWE) corrected. The left side of the picture corresponds to the left side of the brain; b: plots showing contrasts estimates by group in the left medial frontal gyrus, left precuneus and right lingual gyrus, respectively.

Previous studies have found inefficient PFC activation during WM in non-affected siblings of patients with schizophrenia, in the absence of behavioural differences [33]. This abnormality is similar to the abnormalities seen in schizophrenia patients, in that both patients with schizophrenia and their unaffected siblings displayed greater activation in the dorsolateral prefrontal cortex during an N-back task relative to healthy comparison subjects. Our results in FHR subjects are less clear, possibly due to the small FHR sample size, but increased PFC and parietal activation was seen relative to controls at an uncorrected threshold. Studies in subjects at UHR for psychosis have also identified PFC dysfunction in this group [11,36], which is associated with increased risk for transition to psychosis [1]. Contrary to these previous studies however, we did not find any PFC abnormalities in the direct comparison of UHR and CTRL groups, which might be

related to differences in paradigms between our and previous studies, sample sizes, or duration of risk stage (e.g. early versus late UHR stages [36]). PFC dysfunction in the UHR group did, however, become evident when task-related activation in UHR, FHR and CTRL groups were compared, such that activation in the left mPFC was seen in the UHR relative to deactivation in both FHR and control groups. The mPFC activation seen in the UHR group encompassed BA6/8, areas activated during executive working memory paradigms [37]. Activation in this region might in part reflect a compensatory mechanism, given intact task performance in the UHR relative to the other two groups. Although previous imaging studies of UHR and FHR cohorts have found functional abnormalities associated with both risk types that were qualitatively similar though less severe than those seen in schizophrenia patients [4,11,20,36,39], the risk of

Please cite this article in press as: Falkenberg I, et al. Failure to deactivate medial prefrontal cortex in people at high risk for psychosis. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2015.03.003

G Model

EURPSY-3241; No. of Pages 8 6

I. Falkenberg et al. / European Psychiatry xxx (2015) xxx–xxx

Fig. 2. Correlations between PANSS negative scores and activity in the left precuneus during the N-back task.

transition to psychosis is substantially higher in UHR than in FHR cohorts (approximately 30% within the next two years [13], compared to 6–13% over their lifetime in FHR cohorts [16,29]). However, to date only one other imaging study has directly compared functional activation in FHR and UHR groups. Using a spatial working memory task, Choi et al. [6] report decreased activation in fronto-parietal regions associated with spatial working memory processing in an UHR group, whereas FHR participants showed increased activation in these regions relative to healthy controls. In the Choi et al. study, one limitation was that the UHR subjects were on low-dose antipsychotic medication, whereas all our UHR subjects were antipsychotic-naive. However, unlike in the Choi et al. study, there was a significant age difference between UHR and FHR subjects in the present study. Although we have covaried for the age difference, it has to be considered a limitation of the present study, as FHR subjects’ were past the typical age of onset for schizophrenia, which is in late adolescence or early twenties [18]. We are therefore not able to draw inferences beyond the comparison of two different risk types, for example with regards to potential neurobiological predictors of psychosis transition. Further limitations of the study include that the samples, in particular the FHR sample, were relatively small and the method did not allow us to directly target DMN function, as we did not acquire any resting state data. The definition of the clinical high risk for psychosis partially overlaps with the definition of increased familial risk, which can lead to a person fulfilling both at-risk criteria. For example, if a person at UHR has a first-degree relative with schizophrenia along with a decline in function, as was the case in one subject in our clinical risk-sample, they will meet criteria for an UHR state. However, a recent meta-analysis of 19 structural imaging studies including 896 high-risk subjects and 701 controls [12] provided further evidence that neural substrates of familial and clinical high-risk states are not directly comparable. Only UHR subjects showed reduced gray matter volume in the PFC relative to controls and when UHR and FHR subjects were compared directly, it was found that UHR individuals had reduced gray matter volume in the bilateral anterior cingulate relative to FHR subjects. Conversely, FHR subjects showed gray matter reductions in the left hippocampus, insula and the right superior temporal gyrus relative to UHR subjects. Thus, different risk types seem to be associated with

different neuroanatomical changes. Our findings of greater PFC abnormalities in the UHR compared to the FHR group fit broadly with these previous and extend them by showing that functional changes in the PFC also differ between familial and clinical groups. In light of the partial overlap between UHR and GHR groups discussed above, further differentiation into subgroups of UHR subjects with and without a family history of psychosis may help to examine the contribution of genetic versus environmental factors on brain abnormalities associated with different risk states. This approach has been chosen in two studies using structural neuroimaging. Wood et al. [40] found more pronounced structural abnormalities in the left hippocampus in UHR subjects without a family history of psychosis relative to those with a family history. Nenadic et al. [31] found reduced regional gray matter in the left superior/middle frontal gyri, the right caudate, the right hippocampus and right amygdala in UHR subjects with a family history of psychosis relative to controls. UHR subjects without such a history showed right middle temporal cortical reductions. However, as in our UHR sample only one subject had a family history of psychosis, further differentiation of the UHR sample into subgroups was not possible. In addition to the mPFC, the UHR group showed increased activation in the left precuneus/posterior cingulate and right lingual gyrus. Along with the mPFC, these regions are key nodes in the default mode network [32]. The DMN is characterized by high activation and high functional connectivity during the rest or baseline conditions of functional imaging tasks, especially when executive and mnemonic paradigms are used. These DMN regions usually show marked deactivation during the task phase of these paradigms [32]. DMN abnormalities such as abnormal functional connectivity or failure to deactivate the DMN during task performance have previously been described in FEP patients [17], and in UHR and FHR populations [20,25,35]. Adaptive task performance requires the ability to redirect resources away from internal thoughts and feelings towards external stimuli. Accordingly, less deactivation of the DMN during task performance has been found to be associated with momentary lapses in attention in healthy subjects, possibly due to less effective suspension of task-irrelevant mental processes such as daydreaming or monitoring other internal or external signals [38]. In patients with schizophrenia as well as UHR and FHR individuals the anti-correlation between DMN and task positive network activity, which is necessary to maintain adequate performance in a number of tasks, has been shown to be reduced [35,39]. DMN dysfunction might therefore reflect the inability of patients with schizophrenia to allocate sufficient neural resources to cognitively demanding tasks and actions, an impairment that is also present, albeit to a lesser degree, in subclinical stages of the disorder [10,14]. This notion is consistent with neuropsychological studies in UHR cohorts that report impaired executive and mnemonic function [14]. In the present study, increased activation in the precuneus during task performance was associated with levels of negative symptoms, which was mainly driven by the UHR subjects. This finding is consistent with a recent resting state fMRI study, which also found that abnormalities in the fronto-polar node of the DMN in patients with chronic schizophrenia were correlated with negative symptoms [26]. Negative symptoms are also closely linked with impaired social and role functioning in UHR subjects who later convert to psychosis [8] and there is growing appreciation of the importance of functional outcome of UHR samples, with poor function being considered just as important an outcome measure as transition to psychosis [22]. Increased activation in the precuneus during task performance was also associated with lower levels of global functioning in UHR subjects (albeit at a trend level). Although this is a marginal finding, the

Please cite this article in press as: Falkenberg I, et al. Failure to deactivate medial prefrontal cortex in people at high risk for psychosis. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2015.03.003

G Model

EURPSY-3241; No. of Pages 8 I. Falkenberg et al. / European Psychiatry xxx (2015) xxx–xxx

associations between negative symptoms, functioning and activation of the precuneus, but not with other areas of the DMN, might reflect the unstable nature of the UHR syndrome. Before the onset of frank psychosis, only certain areas of the DMN might be affected, whereas illness progression and formation of permanent symptoms might involve further DMN structures, such as the mPFC [26]. In order to further underpin this interpretation, we have conducted an additional analysis including a small sample of patients with first episode psychosis (FEP), the results of which are reported in the supplementary material to this article (Table S3 and Fig. S1). In brief, we found that at an uncorrected threshold, patients with FEP displayed deactivation in the left precuneus during task performance, although to a lesser degree than FHR and CTRL subjects. This may be considered a partial normalization of activity patterns, possibly due to the effects of antipsychotic medication. The failure to deactivate the mPFC during task performance, which we found in the UHR GROUP, was also present in the FEP sample. There was a positive correlation between the level of negative symptoms and activation in the right mPFC (rs = 0.42, P < 001) and the left precuneus (rs = 0.3, P = 013). In summary, our results lend further support to the notion that DMN abnormalities are relevant to core psychopathological features of schizophrenia, even at a subclinical stage. To conclude, our results suggest that mPFC dysfunction may not be a robust trait marker of schizophrenia vulnerability, but instead relate to the presence of attenuated negative symptoms. These results clearly point to the importance of future investigation into the longitudinal course of PFC dysfunction in schizophrenia and schizophrenia proneness. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Acknowledgments This work was supported by the G.A. Lienert Foundation, AdolfSchmidtmann-Foundation, FAZIT Foundation and German Academic Exchange Service (to I.F.); a Medical Research Council Studentship (to C.C.); a Medical Research Council Pathfinder Award (to C.M.); a postdoctoral award from the Department of Health (to E.B.); the Guy’s and St Thomas’ Charitable Trust (to M.B.); and the National Alliance of Research into Schizophrenia and Depression (to P.A.).

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.eurpsy.2015.03. 003. References [1] Allen P, Luigjes J, Howes OD, Egerton A, Hirao K, Valli I, et al. Transition to psychosis associated with prefrontal and subcortical dysfunction in ultrahigh-risk individuals. Schizophr Bull 2012 [Internet; cited 2012 May 9]http:// www.ncbi.nlm.nih.gov/pubmed/22290265. [2] Barkus EJ, Stirling J, Hopkins RS, Lewis S. Cannabis-induced psychosis-like experiences are associated with high schizotypy. Psychopathology 2006; 39(4):175–8. [3] Broome MR, Woolley JB, Johns LC, Valmaggia LR, Tabraham P, Gafoor R, et al. Outreach and support in south London (OASIS): implementation of a clinical service for prodromal psychosis and the at risk mental state. Eur Psychiatry 2005;20(5–6):372–8. [4] Broome MR, Matthiasson P, Fusar-Poli P, Woolley JB, Johns LC, Tabraham P, et al. Neural correlates of executive function and working memory in the ‘‘atrisk mental state’’. Br J Psychiatry 2009;194(1):25–33.

7

[5] Buckner RL, Andrews-Hanna JR, Schacter DL. The brain’s default network: anatomy, function, and relevance to disease. Ann N Y Acad Sci 2008;1124:1–38. [6] Choi J-S, Park J-Y, Jung MH, Jang JH, Kang D-H, Jung WH, et al. Phase-specific brain change of spatial working memory processing in genetic and ultra-highrisk groups of schizophrenia. Schizophr Bull 2011 [Internet]http://www.ncbi. nlm.nih.gov/pubmed/21518920. [7] Coren S. Measurement of handedness via self-report: the relationship between brief and extended inventories. Percept Mot Skills 1993;76(3 Pt 1):1035–42. [8] Cornblatt BA, Carrio´n RE, Addington J, Seidman L, Walker EF, Cannon TD, et al. Risk factors for psychosis: impaired social and role functioning. Schizophr Bull 2012;38(6):1247–57. [9] Egan MF, Goldberg TE, Gscheidle T, Weirich M, Rawlings R, Hyde TM, et al. Relative risk for cognitive impairments in siblings of patients with schizophrenia. Biol Psychiatry 2001;50(2):98–107. [10] Frommann I, Pukrop R, Brinkmeyer J, Bechdolf A, Ruhrmann S, Berning J, et al. Neuropsychological profiles in different at-risk states of psychosis: executive control impairment in the early – and additional memory dysfunction in the late – prodromal state. Schizophr Bull 2011;37(4):861–73. [11] Fusar-Poli P, Broome MR, Woolley JB, Johns LC, Tabraham P, Bramon E, et al. Altered brain function directly related to structural abnormalities in people at ultra-high-risk of psychosis: longitudinal VBM-fMRI study. J Psychiatr Res 2011;45(2):190–8. [12] Fusar-Poli P, Borgwardt S, Crescini A, Deste G, Kempton MJ, Lawrie S, et al. Neuroanatomy of vulnerability to psychosis: a voxel-based meta-analysis. Neurosci Biobehav Rev 2011;35(5):1175–85. [13] Fusar-Poli P, Bonoldi I, Yung AR, Borgwardt S, Kempton MJ, Valmaggia L, et al. Predicting psychosis: meta-analysis of transition outcomes in individuals at high clinical risk. Arch Gen Psychiatry 2012;69(3):220–9. [14] Fusar-Poli P, Deste G, Smieskova R, Barlati S, Yung AR, Howes O, et al. Cognitive functioning in prodromal psychosis: a meta-analysis. Arch Gen Psychiatry 2012;69(6):562–71. [15] Glahn DC, Ragland JD, Abramoff A, Barrett J, Laird AR, Bearden CE, et al. Beyond hypofrontality: a quantitative meta-analysis of functional neuroimaging studies of working memory in schizophrenia. Hum Brain Mapp 2005;25(1):60–9. [16] Gottesman IL, Wolfgram DL. Schizophrenia genesis: the origins of madness. New York: W H Freeman & Co; 1990 [296 p.]. [17] Guerrero-Pedraza A, McKenna PJ, Gomar JJ, Sarro´ S, Salvador R, Amann B, et al. First-episode psychosis is characterized by failure of deactivation but not by hypo- or hyperfrontality. Psychol Med 2012;42(1):73–84. [18] Ha¨fner H, Maurer K, Lo¨ffler W, Fa¨tkenheuer B, an der Heiden W, RiecherRo¨ssler A, et al. The epidemiology of early schizophrenia. Influence of age and gender on onset and early course. Br J Psychiatry Suppl 1994;(23):29–38. [19] Hall RCW. Global assessment of functioning: a modified scale. Psychosomatics 1995;36(3):267–75. [20] Jang JH, Jung WH, Choi J-S, Choi C-H, Kang D-H, Shin NY, et al. Reduced prefrontal functional connectivity in the default mode network is related to greater psychopathology in subjects with high genetic loading for schizophrenia. Schizophr Res 2011;127(1–3):58–65. [21] Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull 1987;13(2):261. [22] Lin A, Wood SJ, Nelson B, Brewer WJ, Spiliotacopoulos D, Bruxner A, et al. Neurocognitive predictors of functional outcome two to 13 years after identification as ultra-high-risk for psychosis. Schizophr Res 2011;132(1):1–7. [23] McDonald C, Bullmore ET, Sham PC, Chitnis X, Wickham H, Bramon E, et al. Association of genetic risks for schizophrenia and bipolar disorder with specific and generic brain structural endophenotypes. Arch Gen Psychiatry 2004;61(10):974–84. [24] Mechelli A, Henson RNA, Price CJ, Friston KJ. Comparing event-related and epoch analysis in blocked design fMRI. Neuroimage 2003;18(3):806–10. [25] Meda SA, Gill A, Stevens MC, Lorenzoni RP, Glahn DC, Calhoun VD, et al. Differences in resting-state functional magnetic resonance imaging functional network connectivity between schizophrenia and psychotic bipolar probands and their unaffected first-degree relatives. Biol Psychiatry 2012;71(10):881–9. [26] Mingoia G, Wagner G, Langbein K, Maitra R, Smesny S, Dietzek M, et al. Default mode network activity in schizophrenia studied at resting state using probabilistic ICA. Schizophr Res 2012;138(2–3):143–9. [27] Minzenberg MJ, Laird AR, Thelen S, Carter CS, Glahn DC. Meta-analysis of 41 functional neuroimaging studies of executive function in schizophrenia. Arch Gen Psychiatry 2009;66(8):811–22. [28] Morey RA, Inan S, Mitchell TV, Perkins DO, Lieberman JA, Belger A. Imaging frontostriatal function in ultra-high-risk, early, and chronic schizophrenia during executive processing. Arch Gen Psychiatry 2005;62(3):254–62. [29] Mortensen PB, Pedersen MG, Pedersen CB. Psychiatric family history and schizophrenia risk in Denmark: which mental disorders are relevant? Psychol Med 2010;40(2):201–10. [30] Nelson H, Willison J. National adult reading test (NART): test manual. Windsor: NFER-Nelson; 1982. [31] Nenadic I, Dietzek M, Scho¨nfeld N, Lorenz C, Gussew A, Reichenbach JR, et al. Brain structure in people at ultra-high-risk of psychosis, patients with firstepisode schizophrenia, and healthy controls: a VBM study. Schizophr Res 2015;161(2–3):169–76. [32] Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL. A default mode of brain function. Proc Natl Acad Sci U S A 2001;98(2):676–82. [33] Rasetti R, Mattay VS, Wiedholz LM, Kolachana BS, Hariri AR, Callicott JH, et al. Evidence that altered amygdala activity in schizophrenia is related to clinical state and not genetic risk. Am J Psychiatry 2009;166(2):216–25.

Please cite this article in press as: Falkenberg I, et al. Failure to deactivate medial prefrontal cortex in people at high risk for psychosis. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2015.03.003

G Model

EURPSY-3241; No. of Pages 8 8

I. Falkenberg et al. / European Psychiatry xxx (2015) xxx–xxx

[34] Resnick SM. Matching for education in studies of schizophrenia. Arch Gen Psychiatry 1992;49(3):246. [35] Shim G, Oh JS, Jung WH, Jang JH, Choi C-H, Kim E, et al. Altered resting-state connectivity in subjects at ultra-high-risk for psychosis: an fMRI study. Behav Brain Funct 2010;6:58. [36] Smieskova R, Allen P, Simon A, Aston J, Bendfeldt K, Drewe J, et al. Different duration of at-risk mental state associated with neurofunctional abnormalities. A multimodal imaging study. Hum Brain Mapp 2011 [Internet]http:// www.ncbi.nlm.nih.gov/pubmed/21922599. [37] Wager TD, Smith EE. Neuroimaging studies of working memory: a metaanalysis. Cogn Affect Behav Neurosci 2003;3(4):255–74. [38] Weissman DH, Roberts KC, Visscher KM, Woldorff MG. The neural bases of momentary lapses in attention. Nat Neurosci 2006;9(7):971–8. [39] Whitfield-Gabrieli S, Thermenos HW, Milanovic S, Tsuang MT, Faraone SV, McCarley RW, et al. Hyperactivity and hyperconnectivity of the default

[40]

[41]

[42]

[43]

network in schizophrenia and in first-degree relatives of persons with schizophrenia. Proc Natl Acad Sci U S A 2009;106(4):1279–84. Wood SJ, Yu¨cel M, Velakoulis D, Phillips LJ, Yung AR, Brewer W, et al. Hippocampal and anterior cingulate morphology in subjects at ultra-high-risk for psychosis: the role of family history of psychotic illness. Schizophr Res 2005;75(2–3):295–301. World Medical Association. World Medical Association declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA 2013;310(20):2191–4. Yung AR, Phillips LJ, McGorry PD, McFarlane CA, Francey S, Harrigan S, et al. Prediction of psychosis. A step towards indicated prevention of schizophrenia. Br J Psychiatry Suppl 1998;172(33):14–20. Yung AR, Yuen HP, McGorry PD, Phillips LJ, Kelly D, Dell’Olio M, et al. Mapping the onset of psychosis: the Comprehensive Assessment of At-Risk Mental States. Aust N Z J Psychiatry 2005;39(11–12):964–71.

Please cite this article in press as: Falkenberg I, et al. Failure to deactivate medial prefrontal cortex in people at high risk for psychosis. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2015.03.003