Concurrent Cognitive Control Training Augments the Antidepressant Efficacy of tDCS: A Pilot Study

Brain Stimulation 7 (2014) 325e331

Contents lists available at ScienceDirect

Brain Stimulation journal homepage: www.brainstimjrnl.com

Concurrent Cognitive Control Training Augments the Antidepressant Efficacy of tDCS: A Pilot Study R.A. Segrave*, S. Arnold, K. Hoy, P.B. Fitzgerald Monash University, Central Clinical School, Monash Alfred Psychiatry Research Centre, Level 4, 607 St Kilda Rd., Prahran, VIC 3004, Australia

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 October 2013 Received in revised form 13 December 2013 Accepted 13 December 2013

Background: Major depressive disorder (MDD) is frequently associated with underactivity of the dorsolateral prefrontal cortex (DLPFC) which has led to this brain region being identified as an important target for the development of neurobiological treatments. Transcranial direct current stimulation (tDCS) administered to the DLPFC has antidepressant efficacy, however the magnitude of antidepressant outcomes are limited. Concurrent cognitive activity has been shown to enhance tDCS induced stimulation effects. Cognitive control training (CCT) is a new cognitive therapy for MDD that aims to enhance DLPFC activity via behavioral methods. Hypothesis: We tested the hypothesis that co-administration of DLPFC tDCS and CCT would result in a greater reduction in depressive symptomology than administration of tDCS or CCT alone. Methods: 27 adult participants with MDD were randomized into a three-arm sham-controlled betweengroups pilot study comparing the efficacy of 2 mA tDCS þ CCT, sham tDCS þ CCT and sham CCT þ 2 mA tDCS (5 sessions administered on consecutive working days). Blinded assessments of depression severity and cognitive control were conducted at baseline, end of treatment and a three week follow up review. Results: All three treatment conditions were associated with a reduction in depression severity at the end of five treatment sessions. However, only administration of tDCS þ CCT resulted in sustained antidepressant response at follow up, the magnitude of which was greater than that observed immediately following conclusion of the treatment course. Conclusions: The results provide preliminary evidence that concurrent CCT enhances antidepressant outcomes from tDCS. In the current sample, participants receiving concurrent tDCS and CCT continued to improve following cessation of treatment. The clinical superiority of a combined therapeutic approach was apparent even in a small sample and following a relatively short treatment course. Ó 2014 Elsevier Inc. All rights reserved.

Keywords: Depression Transcranial direct current stimulation Cognitive training Cognitive control Dorsolateral prefrontal cortex

Background Because of its well-established neuromodulatory properties, transcranial direct current stimulation (tDCS) has been investigated as a therapeutic tool in psychiatry, with much of this work being done in major depressive disorder (MDD). Five open-label [1e5] and seven randomized sham-controlled investigations [6e12] have been conducted and all, but one [8], have described significant antidepressant effects. However, the results of two recent meta-analyses highlight that while the results to date have been statistically significant, the magnitude of antidepressant outcomes

This study was made possible by funding received from Monash University, Department of Medicine, Nursing and Health Sciences and the National Health and Medical Research Council. * Corresponding author. Tel.: þ61 3 9076 5030. E-mail address: [email protected] (R.A. Segrave). 1935-861X/$ e see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.brs.2013.12.008

have been clinically sub-optimal. Kalu et al. [13] examined data from four open-label and six randomized controlled tDCS trials and observed a mean reduction in depression severity of 28.9% and a mean remission rate of 8.5% following a course of active tDCS. More recently, Berlim et al. [14] found that response (i.e. 50% reduction on depression severity) and remission rates were equivocal following courses of active and sham tDCS. Taken together, this evidence suggests that DLPFC tDCS has antidepressant potential, but that effects seen under treatment protocols trialed thus far are of a clinically modest nature. Therapeutic tDCS for MDD has typically consisted of 1e2 mA anodal stimulation of the left DLPFC. Hypoactivity of the DLPFC is a common finding in functional neuroimaging studies of MDD [15] and the DLPFC is a key node within neurocircuits that subsume cognitive and emotional control. Depressogenic underactivity in this area is frequently conceptualized as a failure of the DLPFC to exert appropriate control over negative affective thought content

326

R.A. Segrave et al. / Brain Stimulation 7 (2014) 325e331

[16] and excessive and maladaptive emotion induced by hyperactivity within subcortical limbic regions [17e19]. As anodal tDCS enhances local cortical excitability, the application of anodal stimulation to the DLPFC is intended to increase excitability in this region and remediate this. Clinical trials of tDCS for depression to date have administered stimulation while the patient sits quietly at rest. However, studies have repeatedly shown that greater functional outcomes occur when tDCS is delivered to an active brain region as opposed to a ‘resting’ one (i.e. the impact of tDCS is dependent on the functional state of the underlying cortical region) [20e22]. For example, tDCS induced improvements in cognition have been reported in numerous cognitive domains [20,23,24], however the most robust findings have been obtained when stimulation is administered during execution of a task tapping the cognitive domain in question (i.e. as opposed to being administered at rest) [20] and applied to a neuroanatomical site involved in that cognitive process [25]. As such, it follows that greater functional enhancement of DLPFC activity, and thus greater antidepressant outcomes, could potentially be achieved via delivery of tDCS during execution of a task that engages DLPFC. This may be especially true if the activity in question has independent antidepressant properties. Cognitive control training (CCT) is a new cognitive therapy for MDD. It aims to treat depression in the same way that tDCS does, by enhancing DLPFC activity, but does so via behavioral methods [16]. CCT comprises repeated engagement in two targeted cognitive activities that have been designed (based on neuroimaging evidence) to engage the DLPFC [16,26,27]. The first, a modified Wells Attentional Training (WAT) paradigm, is a guided auditory activity that takes participants through a series of short tasks requiring focused attention, attentional switching and divided attention [28]. The WAT is designed to improve self-direction of attention in the face of distraction, and by extension control over automatic negative thoughts. The second is an adaptive version of the Paced Serial Addition Task (PSAT), a challenging mental arithmetic task that requires patients to exercise a high degree of cognitive control in a slightly frustrating (i.e. cognitively challenging) situation (see Method section for additional detail). A small amount of preliminary clinical evidence supports the utility of CCT for MDD. Adjunctive CCT has been shown to reduce depression severity in a small group of severely depressed inpatients [16] and the severity of depressive ruminations in a larger cohort of depressed outpatients [29]. The current study aimed to investigate whether concurrent CCT would enhance the antidepressant efficacy of a short course of tDCS. A three-arm randomized sham-controlled pilot study was conducted comparing the therapeutic impact of five sessions of either 2 mA tDCS þ CCT, sham tDCS þ CCT or sham CCT þ 2 mA tDCS in participants with MDD. It was hypothesized that at the end of the five sessions participants treated with 2 mA tDCS þ CCT would experience a greater reduction in depression severity than those who underwent either CCT or tDCS alone (i.e. paired with sham). The impact of acute and cumulative treatment sessions on cognitive control over emotive and non-emotive material were also assessed using non-affective (neutral) and affective (positive and negative) versions of a two-back working memory task. Method Participants Twenty seven acutely depressed adults participated in the study (10 female, 24 right-handed, mean age  SD ¼ 40.44  14.52 years). All met criteria for a current DSM-IV defined Major Depressive Episode at study entry as confirmed by clinician administered interview with the Mini International Neuropsychiatric Interview

[30]. While no minimum level of depression severity was required for study entry, baseline Montgomery Asberg Depression Rating Scale (MADRS) [31] scores ranged between 19 and 41, indicative of moderate to severe depression. All reported no history of brain injury, neurological illness or diagnosed learning difficulty, had normal or corrected-to-normal vision and were not color blind. Estimated intellectual functioning, as determined by the Wechsler Test of Adult Reading (WTAR) [32], fell within or above the average range for all participants. No participants had a history of substance abuse or dependence in the preceding year or lifetime history of mania, hypomania, post-traumatic stress disorder or psychosis. All were free from benzodiazepines, mood stabilizers, antipsychotics or other medications known to have an adverse impact on anodal tDCS (i.e. L-dopa, rivastigmine, dextromethorphan, flunarizine [33]) or cognition. Fourteen were taking antidepressant medication (six ¼ SSRI, four ¼ SNRI, one ¼ TCA, one ¼ SNRI þ TeCA, one ¼ SSRI þ TCS, one ¼ modafinil) and had been stable on the same medication at the same dose for a minimum four weeks (and generally much longer) prior to study entry and medication regimen remained constant throughout participation. Participants were instructed not to engage in any additional independent cognitive training during participation in the study. Participants provided written confirmation of informed consent prior to engaging in the study protocol. They were provided with $20 per treatment session to compensate for their time and travel costs. The study received approval from the Alfred Human Research Ethics Committee and the Monash University Human Research Ethics Committee and was registered on the Australian and New Zealand Clinical Trials Registry (ACTRN12613000050752). Study design Using a predetermined randomization schedule constructed using a random number generator, participants were assigned on 1:1:1 basis at their baseline visit to receive five treatment sessions of either: tDCS þ CCT, sham tDCS þ CCT or tDCS þ sham CCT (n ¼ 9 per group). Randomization was conducted by a member of the research team with no role in outcome assessment. Stimulation and cognitive training were delivered concurrently and treatments were delivered on five consecutive week days. Participants were asked to guess which treatment condition they had been receiving at the end of the treatment course. Both participants and clinical raters remained blind to treatment condition until the end of the three week follow up assessment when unmasking occurred. tDCS tDCS was administered via an Eldith DC-stimulator (NeuroConn GmbH, Germany). Twenty four minutes (2 min fade in/out) of 2 mA anodal and cathodal tDCS was delivered via two conductive rubber electrodes encased in saline soaked sponges (surface area ¼ 5 cm  7 cm) held in place with a broad flexible band. Anodal stimulation was delivered to the left DLPFC at F3 (10-20 system) and cathodal to the lateral aspect of the contralateral orbit (F8 according to the 10-20 system). This electrode positioning was selected to be consistent with the majority of prior randomized controlled clinical trials of tDCS for MDD [7e12]. For sham stimulation, an active 2 min 2 mA fade-in period was delivered, followed by cessation of stimulation. Cognitive Training Delivery and investigator facilitation Cognitive training was commenced immediately at the end of the stimulation fade-in period. For both active and sham cognitive

R.A. Segrave et al. / Brain Stimulation 7 (2014) 325e331

training paradigms, participants were shown a graphical representation of their training accuracy at the end of each session. The treater then discussed with each participant their performance for that session and encouraged daily goal setting to maximize ongoing motivation and engagement. CCT CCT comprised two modules, a modified WAT [28] (8 min, no breaks) protocol and an adaptive PSAT (14 min, brief break taken at 5 min intervals). The WAT exercises multiple aspects of attentional processing known to engage the DLPFC [34e37] and is designed to improve selective and self-direction of attention in the face of distraction [28]. Participants must focus on a specified target sound in a naturalistic acoustic environment whilst ignoring competing non-target sounds. They are periodically required to switch their attention to a new sound, and then back and forth between different sounds and finally to expand their focus to encompass multiple sounds, while continuing to ignore ever-present distracters. Auditory stimuli are common bird sounds presented simultaneously via computer and surround sound speakers. To execute the task, participants must maintain focused attention away from depressive internal dialog and/or dysphoria, which exercises cognitive control over automatic thoughts and emotional processes. The PSAT is a challenging mental arithmetic task that also engages the DLPFC [26,27]. Individual digits between one and nine are presented (auditory presentation via computer and surround sound speakers) to participants who are required to add each presented digit with the one that preceded it (i.e. serial addition, calculating a new number with each presentation e not a running total). The numbers 1e18 are presented in a circular array on a computer screen throughout and participants must indicate via mouse click the correct total with each new digit presented. Initial ISI was set at 4000 ms. To promote engagement an adaptive version of the PSAT was employed; the rate of digit presentation increased or decreased by 100 ms with four successive correct/incorrect responses respectively. The value of the PSAT is that it exercises maintenance of cognitive control in the face of mild frustration and/or negative effect.

327

completed the Beck Depression Inventory-II (BDI-II) [39] at these time points. Remission was defined as MADRS < 10 and participants were considered responders if they demonstrated a 50% reduction in MADRS score from baseline. Modulation of DLPFC mediated cognitive control was assessed using matched affective and non-affective versions of the two-back working memory task. The two-back was completed at baseline, immediately following first and last treatments and at three week follow up. A different version of the task was administered at each testing session and version order was counterbalanced across participants. Stimuli were positive, neutral and negative words selected from the Affective Norms for English Words (ANEW) set, and word length and the accompanying normative data for valence and arousal were used to guide stimuli selection [40]. Words were presented consecutively for 600 ms with a 1400 ms ISI. Participants responded via ‘yes’ or ‘no’ button press to indicate whether the current word matched that presented two words ago. Six task blocks were completed at each assessment: two positive, two negative and two neutral, and block condition order was counterbalanced across sessions and participants. Each block comprised a different set of 10 words, with each word presented seven e eight times per block and twice in the target two-back position (75 presentations in total; 20 targets and 55 non-targets). Words were balanced for number of letters within and between each block. The ANEW provides normative data in the form of mean valence and arousal scores (i.e. 1e9 with higher number indicating greater arousal or more positive valence) for each word. For neutral words, valence and arousal were balanced within and between all blocks (all neutral blocks mean valence ¼ 5.00 and mean arousal ¼ 4.00). For positive and negative blocks, arousal was balanced within and between all blocks (all positive and negative blocks: mean arousal ¼ 5.9) and mean valence for both conditions was equidistant from neutral (all positive blocks mean valence ¼ 8.1, all negative blocks mean valence ¼ 1.9). In order to minimize practice effects and the impact of different learning rates, prior to baseline data collection participants completed a neutral practice version of the two-back as often as required until they achieved a minimum of 70% task accuracy. Statistical analysis

Sham CCT: peripheral vision training Peripheral vision training (PVT) was used as a non-active (i.e. non-therapeutic) control cognitive training paradigm. Participants viewed a circular array of gray discs and were directed to keep their eyes focused on a central fixation marker at all times. They were then required to use their peripheral vision to step their focus, but not their eyes, one disc at a time in a clockwise direction in response to repeated presentation of auditory ‘go’ cues. Upon presentation of a distinct auditory ‘stop’ cue each disc changed from gray to one of five colors (blue, yellow, red, pink or green) and participants were required to indicate the color of the target ‘stop’ disc via button press (22 min, brief break taken at 5 min intervals). All participants began with a circular array of 15 discs. To promote engagement and match the adaptive nature of the PSAT, one disc was added to this array with every four consecutive correct answers, and one disc was removed from the array in response to four consecutive incorrect answers. The PVT has been utilized in prior studies as a sham cognitive training protocol for CCT [38]. Clinical and cognitive outcome measures The primary clinical outcome measure was the MADRS, which was completed by a trained rater blind to treatment condition at baseline, following fifth treatment and at a follow up review conducted three weeks after the end of treatment. Participants also

IBM SPSS Statistics 20.0 was used for all statistical analyses. One way ANOVAS were used to compare the three treatment groups on baseline clinical and demographic variables. Group differences in change in MADRS and BDI-II scores over time were assessed using a time [3] by group [3] repeated measures ANOVA. Group differences in two-back task reaction time and accuracy were assessed using a separate time [4] by group [3] repeated measures ANOVA for each of the three task conditions: positive, negative and neutral. Posthoc analyses were conducted using one-way ANOVAs and pairwise comparisons. As we have previously observed a significant relationship between early cognitive change and antidepressant response to DLPFC [9], we additionally examined group differences in neutral two-back task accuracy immediately following first treatment session via a one-way ANOVA, and used Pearson’s r correlation to ascertain the relationship between task performance at this time point and MADRS score at final outcome assessment as a potential predictor of response. Chi-squared analyses were used to assess group differences in clinical response and remission rates and blinding integrity. All results were assessed using two-tailed tests with an alpha level of .05. As the study is a pilot proof-ofprincipal investigation with a relatively modest sample size no adjustments were made for multiple comparisons. A last observation carried forward approach was taken for missing clinical data (one participant only).

328

R.A. Segrave et al. / Brain Stimulation 7 (2014) 325e331

Table 1 Participant demographic and clinical characteristics at Baseline (mean  SD).

Gender (F/M) Handedness (L/R) Current AD (Y/N) Melancholic versus non-melancholic Age, years Years of education WTAR Pred IQ Baseline MADRS Length current episode, months Number prior episodes Age at initial onset n

CCT þ tDCS

Sham CCT þ tDCS

CCT þ sham tDCS

2/7 0/9 5/4 4/5 42.6  18.32 14.56  3.28 111.38  7.54 27.11  5.42 141.63  184.11 2.14  2.04 17.44  9.11 9

4/5 1/8 6/3 4/5 33.8  12.96 14.0  1.73 115.43  6.95 31.22  5.74 37.33  39.12 2.43 (1.13) 19.63  13.22 9

4/5 2/7 3/6 4/5 45.0  10.15 14.89  2.52 111.67  7.55 27.11  4.68 84.11  102.09 2.63  2.00 16.0  5.92 9

F

P

1.56 .27 .70 1.82 1.59 .14 .30

.23 .77 .51 .18 .23 .87 .75

AD ¼ antidepressant medication.

Results Baseline demographics, clinical characteristics and cognitive performance Participants in the three treatment groups did not differ significantly in age, number of year’s education or estimated IQ (WTAR). No baseline group differences were observed in depression severity (MADRS), length of current depressive episode, number of prior episodes, age at initial depression onset, number of prior antidepressant medications trialed (Table 1) or baseline performance (reaction time or accuracy) on any version of the two-back task. Clinical outcomes One participant in the tDCS þ CCT condition withdrew following completion of the full treatment course and end of treatment assessments, but prior to follow up review, due to non-response and desire to commence antidepressant medication without delay. The three treatment groups differed significantly in change in MADRS score over time, F(4,48) ¼ 4.63, P ¼ .003. Post hoc pairwise comparisons revealed that participants’ MADRS scores decreased significantly following five sessions of tDCS þ sham CCT (P ¼ .04) and sham tDCS þ CCT (P ¼ .02), and a decrease that approached significance was seen following tDCS þ CCT (P ¼ .06). At three week follow up, MADRS scores of participants who received tDCS þ sham CCT and sham tDCS þ CCT had increased and were no longer different from baseline, while scores of participants who received tDCS þ CCT had further decreased and were significantly lower than pre-treatment levels (P < .001) (Fig. 1). A trend for a difference in response rates for the three treatment groups was observed following the fifth treatment session, c2(2) ¼ 5.01, P ¼ .08. Thirty three percent of participants who received tDCS þ CCT, forty four percent of participants who received sham tDCS þ CCT and no participants in the tDCS þ sham CCT group met response criteria at this time point. At three week follow up, there was a significant difference in response rates for the three treatment groups, c2(2) ¼ 6.38, P ¼ .04. Forty four percent of tDCS þ CCT participants, eleven percent of sham tDCS þ CCT participants and no tDCS þ sham CCT participants met response criteria at this assessment. Differences between the groups in remission rates did not achieve significance at either outcome assessment. Similarly to MADRS scores, a significant group difference in change in BDI-II over time was observed, F(4,48) ¼ 2.83, P ¼ .03. Post hoc pairwise comparisons demonstrated that participants’ BDI-II scores decreased significantly following tDCS þ sham CCT (P ¼ .005) and sham tDCS þ CCT (P ¼ .006), but did not differ from baseline following tDCS þ CCT (P > .05). At three week follow up, BDI-II scores of participants in the sham tDCS þ CCT group had

returned to baseline level, while those who had received tDCS þ sham CCT (P ¼ .02) and tDCS þ CCT (P ¼ .004) had scores that were significantly lower than baseline. Mean and standard deviation MADRS and BDI-II scores and response and remission rates are depicted in Table 2. Cognitive outcomes Accuracy On the positive and neutral versions of the two-back task, participants in all treatment groups performed significantly more accurately over time (positive: F(3,69) ¼ 8.13 , P < .001; neutral: F(3,69) ¼ 4.77, P ¼ .007). No significant differences were found between the treatment groups in improvement in accuracy over time, on either of these versions of the task. While there was no main effect of time on accuracy for the negative version of the two-back, a significant time by treatment interaction was observed (F(6,69) ¼ 2.47, P ¼ .03). Post hoc analyses indicated that this interaction was driven by a significant group difference in negative two-back accuracy at final assessment (F(2,23) ¼ 3.50, P ¼ .05), with participants in the tDCS þ CCT group more accurate at this time point (P ¼ .02) than those in either of the other two treatment conditions. Across all groups, a significant negative correlation between negative two-back accuracy and final MADRS score was also observed (r ¼ .39, P ¼ .05), with increased task accuracy associated with lower depression severity. Reaction time Participants in all treatment groups responded faster over time on each of the three two-back task conditions (positive: F(3,69) ¼ 7.93, P ¼ .001; neutral: F(3,69) ¼ 10.16, P < .001; negative:

Figure 1. Mean percentage MADRS change from baseline.

R.A. Segrave et al. / Brain Stimulation 7 (2014) 325e331

329

Discussion

Figure 2. Mean percentage neutral two-back task accuracy over time.

F(3,69) ¼ 13.25, P < .001). However, there were no significant time by treatment interactions in reaction time on any version of the task. Mean and standard deviation two-back reaction time and accuracy rates are depicted in Tables S1 and S2 in Supplementary materials. Early cognitive change and antidepressant outcomes Immediately following the first treatment session, a trend was observed for participants who received tDCS þ CCT to perform more accurately on the two-back task than participants receiving tDCS or CCT paired with sham, F(2,24) ¼ 3.24, P ¼ .06 (Fig. 2). Further, at a whole sample level, a significant negative correlation was observed between two-back accuracy immediately following session one and MADRS score at final follow up (r ¼ .45, P ¼ .03), such that participants who performed better on the two-back following their first treatment were less depressed at study end. Blinding integrity As two participants did not complete the blinding questionnaire at the end of the treatment course (one from the tDCS þ sham CCT group and another from the tDCS þ CCT group), data assessing the blind integrity was available for 25 participants only. Forty four percent of those sampled (11/25) correctly guessed both the stimulation and cognitive training condition they were allocated to. Fifty six percent (14/25) correctly guessed their cognitive training condition and 68% (17/25) correctly guessed the stimulation condition they received. Differences between the groups in correct versus incorrect guess of condition allocation did not reach significance in any of these comparisons: combined stimulation and cognitive training blinding, c2(2) ¼ .26, P ¼ .88; stimulation blinding: c2(2) ¼ 1.00, P ¼ .61; cognitive training blinding: c2(2) ¼ .65, P ¼ .72.

The current pilot study is the first investigation into whether administration of simultaneous cognitive training augments the therapeutic efficacy of tDCS for MDD. The results support the hypothesis that concurrent CCT potentiates antidepressant outcomes from anodal DLPFC tDCS. While five sessions of tDCS þ sham CCT, sham tDCS þ CCT and tDCS þ CCT resulted in varying degrees of acute antidepressant efficacy, only the double active treatment condition was associated with ongoing antidepressant response at three week follow up. Importantly, the clinical superiority of tDCS þ CCT over either intervention alone was apparent even in a small clinical sample and following a relatively brief treatment course. Interestingly, the most prominent therapeutic benefit of tDCS þ CCT was observed three weeks after cessation of treatment, and not immediately following the final treatment session as hypothesized. This is at odds with findings from the only other study to administer a course of cognitive training with simultaneous tDCS, which did not observe evidence of persistent or delayed benefit. Martin et al. [41] investigated the impact of 10 sessions of anodal DLPFC tDCS delivered during execution of a complex multimodal n-back task in a large sample of healthy controls. While active tDCS improved ‘online’ performance of the cognitive training paradigm above sham tDCS, the performance enhancing effect did not generalize to other ‘offline’ cognitive activities and was not persistent at four week follow up. It is likely that the discordant findings stem from the difference in participant populations, ‘i.e. acutely depressed participants versus healthy controls.’ There is evidence that the potential for tDCS induced cognitive augmentation differs substantially between patients with neurological/psychiatric illnesses and healthy individuals [42]. It has been suggested that this may be underpinned by a homeostatic response to tDCS that is present in healthy individuals and perturbed in many neuropsychiatric conditions [42,43]. The delayed onset of maximal therapeutic efficacy of tDCS þ CCT seen in the current study is suggestive of downstream processes that are initiated by, but continue beyond, the one week treatment course and are critical components of clinical response. The design of the current study does not allow for determination of whether these processes may be neurobiological (e.g. downstream neuroplastic modulation of functional connectivity in distant nodes of the neural networks subsuming MDD) or psychosocial (e.g. subtle cognitive changes provoking increased capacity and/or desire to engage in activities of daily living, which are positively reinforcing over time). Of course these possibilities are not mutually exclusive, and it may be that neurophysiological and environmental influences act synergistically to maximize antidepressant outcomes. These mechanistic questions warrant further systematic investigation. The observation of delayed therapeutic effects also has methodological implications, in that it demonstrates a utility in following up both non-responders and responders to treatment. In

Table 2 Mean and standard deviation MADRS and BDI-II scores over time. Baseline

MADRS BDI-II Responda Remitb a b

End treatment course

Three week follow up

CCT þ tDCS

Sham CCT þ tDCS

CCT þ sham tDCS

CCT þ tDCS

Sham CCT þ tDCS

CCT þ sham tDCS

CCT þ tDCS

Sham CCT þ tDCS

CCT þ sham tDCS

27.00  5.57 27.78  5.87

31.22  5.74 34.67  5.24

27.11  4.68 31.89  8.85

22.33  10.81 23.44  12.37 3/9 2/9

26.44  7.52 25.33  10.05 0/9 0/9

19.56  11.36 19.78  9.00 4/9 2/9

14.44  9.93 15.33  10.37 4/9 3/9

28.44  6.88 27.78  8.61 0/9 0/9

22.44  12.18 26.67  16.83 1/9 1/9

Number of participants with 50% reduction in MADRS score from baseline. Number of participants with MADRS score 10.

330

R.A. Segrave et al. / Brain Stimulation 7 (2014) 325e331

our experience with clinical trials of transcranial magnetic stimulation for MDD [44e46], anecdotal reports of delayed response to therapeutic brain stimulation are common. However, as clinical brain stimulation trials have not routinely followed up participants who do not meet response criteria at the end of a treatment course, it is not possible to know how frequent a phenomenon this may be. Negative information processing biases are common in MDD [47], they play a critical role in the onset and maintenance of depressive episodes and correction of these biases is an important component of clinical recovery [48]. The observation that five sessions of tDCS þ CCT resulted in heightened accuracy on the negative two-back task at follow up, and that this was associated with reduced depression severity, suggests that the combined treatment approach was successful in enhancing cognitive control over negative stimuli. In addition, compared with participants who received tDCS or CCT coupled with sham, those receiving tDCS þ CCT demonstrated a near significant (P ¼ .06) increase in accuracy on the non-affective two-back task immediately following the first treatment session. Greater accuracy at this time point was also positively associated with the magnitude of antidepressant response at final assessment. As efficient working memory processing is heavily reliant upon DLPFC activity and intact cognitive control [49], and in the context of the purported mechanisms of the current intervention, these findings suggest that beneficial clinical response may be associated with early neuromodulation of the DLPFC and associated cognitive control neurocircuitry, and laterstage remediation of negative processing biases. These observations are consistent with prior findings of a relationship between early cognitive change and reduction in negative affective bias and longer-term therapeutic outcomes from a range of depression treatments [48,50,51]. The current investigation was a pilot study and the results should be considered with a number of limitations in mind. In light of the modest sample size, brief treatment course and the lack of a sham tDCS þ sham CCT treatment arm the current findings are not definitive, and best considered as preliminary. Replication in a larger clinical cohort, following delivering of an extend treatment course and with the addition of a double sham condition is required before they can be more confidently generalized. Future inclusion of a double sham condition will help to parcel out potential therapeutic impact of daily contact with supportive research staff and any benefit derived from the imposition of routine associated with attending daily outpatient treatment sessions. It should also be noted that while every effort was made to deliver equivocal feedback and encouragement to all participants irrespective of treatment group, the study treater who provided this feedback was not blinded and it cannot be discounted that this could have subtly biases outcomes. Finally, the issue of participant blinding is an important one. While accuracy of participant guesses for their cognitive training (44%) and stimulation þ cognitive training condition (56%) allocation hovered around chance level, accuracy of guesses for allocated stimulation condition was slightly higher (68%). Differences in the integrity of participant blinding between the treatment groups were not statistically significant, however the small sample size did not allow for robustly powered blinding analyses. The study also has a number of strengths, notably the well matched clinical groups and rigorous methodological design. In addition, the findings are significant in that while many single session studies have examined the application of tDCS with concurrent cognitive activity [20,25,52,53], the effect of repeated sessions on consecutive days has been comparatively under investigated, and until now not at all in clinical populations. Moreover, while a considerable number of studies have combined tDCS and cognitive activity with the intent of enhancing cognitive

processing [20,25,41,42,53] the current study is the first to demonstrate treatment of an affective disorder using this approach. From a practical perspective, the tDCS þ CCT paradigm is highly translatable and replication of the current findings in a larger sample would have exciting clinical implications. Both tDCS and CCT are inexpensive, portable and safe. Following relatively brief formal training and with appropriate supervision they can be administered by non-medical personnel via highly structured standardized protocols. This means that they could be delivered in remote and rural areas where novel mental health interventions are typically unavailable. There is also potential for this approach to be trialed for other indications associated with pathophysiological activity with the DLPFC and associated cognitive and emotional control neural networks, such as generalized anxiety disorder and panic disorder [54,55]. In conclusion, the results of the current pilot study suggest that antidepressant outcomes from anodal DLPFC tDCS may be potentiated via delivery of concurrent CCT. MDD is frequently resistant to standard therapeutic approaches [56] and there is a substantial clinical need for new treatment paradigms. The current approach is a promising one and warrants further clinical and mechanistic investigation, both of which are ongoing. Acknowledgments We are extremely grateful to A/Prof Greg Siegle of Pittsburgh University, United States, for generously providing the CCT and PVT paradigms and associated technical assistance. Appendix. Supplementary data Supplementary data related to this article can be found at doi:10. 1016/j.brs.2013.12.008. References [1] Brunoni AR, Ferrucci R, Bortolomasi M, Vergari M, Tadini L, Boggio PS, et al. Transcranial direct current stimulation (tDCS) in unipolar vs. bipolar depressive disorder. Prog Neuropsychopharmacol Biol Psychiatry 2011;35(1): 96e101. [2] Dell’osso B, Zanoni S, Ferrucci R, Vergari M, Castellano F, D’Urso N, et al. Transcranial direct current stimulation for the outpatient treatment of poorresponder depressed patients. Eur Psychiatry 2012;27(7):513e7. [3] Ferrucci R, Bortolomasi M, Vergari M, Tadini L, Salvoro B, Giacopuzzi M, et al. Transcranial direct current stimulation in severe, drug-resistant major depression. J Affect Disord 2009;118(1e3):215e9. [4] Martin DM, Alonzo A, Mitchell PB, Sachdev P, Galvez V, Loo CK. Frontoextracephalic transcranial direct current stimulation as a treatment for major depression: an open-label pilot study. J Affect Disord 2011;134(1e3):459e63. [5] Ferrucci R, Bortolomasi M, Brunoni AR, Vergares M, Tadini L, Giacopuzzi M, et al. Comparative benefits of transcranial direct current stimulation (tDCS) treatment in patients with mild/moderate vs. severe depression. Clin Neuropsychiatry 2009;6(6):246e51. [6] Rignoatti R, Boggio PS, Myczkowski ML, Otta E, Ribeiro RB, Nitsche MA, et al. Transcranial direct stimulation and fluoxetine for the treatment of depression. Eur Psychiatry 2008;23:74e6. [7] Loo CK, Alonzo A, Martin D, Mitchell PB, Galvez V, Sachdev P. Transcranial direct current stimulation for depression: 3-week, randomised, shamcontrolled trial. Br J Psychiatry 2012;200(1):52e9. [8] Loo CK, Sachdev P, Martin D, Pigot M, Alonzo A, Malhi GS, et al. A double-blind, sham-controlled trial of transcranial direct current stimulation for the treatment of depression. Int J Neuropsychopharmacol 2010;13(1):61e9. [9] Boggio PS, Rigonatti SP, Ribeiro RB, Myczkowski ML, Nitsche MA, PascualLeone A, et al. A randomized, double-blind clinical trial on the efficacy of cortical direct current stimulation for the treatment of major depression. Int J Neuropsychopharmacol 2008;11(2):249e54. [10] Fregni F, Boggio PS, Nitsche MA, Marcolin MA, Rigonatti SP, Pascual-Leone A. Treatment of major depression with transcranial direct current stimulation. Bipolar Disord 2006;8(2):203e4. [11] Fregni F, Boggio PS, Nitsche MA, Rigonatti SP, Pascual-Leone A. Cognitive effects of repeated sessions of transcranial direct current stimulation in patients with depression. Depress Anxiety 2006;23(8):482e4. [12] Palm U, Schiller C, Fintescu Z, Obermeier M, Keeser D, Reisinger E, et al. Transcranial direct current stimulation in treatment resistant depression: a

R.A. Segrave et al. / Brain Stimulation 7 (2014) 325e331

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27] [28] [29]

[30]

[31] [32] [33]

randomized double-blind, placebo-controlled study. Brain Stimul 2012; 5(3):242e51. Kalu UG, Sexton CE, Loo CK, Ebmeier KP. Transcranial direct current stimulation in the treatment of major depression: a meta-analysis. Psychol Med 2012:1e10. Berlim MT, Van den Eynde F, Daskalakis ZJ. Clinical utility of transcranial direct current stimulation (tDCS) for treating major depression: a systematic review and meta-analysis of randomized, double-blind and sham-controlled trials. J Psychiatr Res 2013;47(1):1e7. Siegle GJ, Thompson W, Carter CS, Steinhauer SR, Thase ME. Increased amygdala and decreased dorsolateral prefrontal BOLD responses in unipolar depression: related and independent features. Biol Psychiatry 2007;61(2): 198e209. Siegle GJ, Ghinassi F, Thase ME. Neurobehavioral therapies in the 21st century: summary of an emerging field and an extended example of cognitive control training for depression. Cog Ther Res 2007;31(2):235e62. Wang L, LaBar KS, Smoski M, Rosenthal MZ, Dolcos F, Lynch TR, et al. Prefrontal mechanisms for executive control over emotional distraction are altered in major depression. Psychiatry Res 2008;163(2):143e55. Fales CL, Barch DM, Rundle MM, Mintun MA, Snyder AZ, Cohen JD, et al. Altered emotional interference processing in affective and cognitive-control brain circuitry in major depression. Biol Psychiatry 2008;63(4):377e84. Phillips ML, Drevets WC, Rauch SL, Lane R. Neurobiology of emotion perception II: implications for major psychiatric disorders. Biol Psychiatry 2003; 54(5):515e28. Andrews SC, Hoy KE, Enticott PG, Daskalakis ZJ, Fitzgerald PB. Improving working memory: the effect of combining cognitive activity and anodal transcranial direct current stimulation to the left dorsolateral prefrontal cortex. Brain Stimul 2011;4(2):84e9. Nitsche MA, Cohen LG, Wassermann EM, Priori A, Lang N, Antal A, et al. Transcranial direct current stimulation: state of the art 2008. Brain Stimul 2008;1(3):206e23. Antal A, Terney D, Poreisz C, Paulus W. Towards unravelling task-related modulations of neuroplastic changes induced in the human motor cortex. Eur J Neurosci 2007;26(9):2687e91. Cohen Kadosh R, Soskic S, Iuculano T, Kanai R, Walsh V. Modulating neuronal activity produces specific and long-lasting changes in numerical competence. Curr Biol 2010;20(22):2016e20. Vercammen A, Rushby JA, Loo C, Short B, Weickert CS, Weickert TW. Transcranial direct current stimulation influences probabilistic association learning in schizophrenia. Schizophr Res 2011;131(1e3):198e205. Fregni F, Boggio PS, Nitsche M, Bermpohl F, Antal A, Feredoes E, et al. Anodal transcranial direct current stimulation of prefrontal cortex enhances working memory. Exp Brain Res 2005;166(1):23e30. Lockwood AH, Linn RT, Szymanski H, Coad ML, Wack DS. Mapping the neural systems that mediate the Paced Auditory Serial Addition Task (PASAT). J Int Neuropsychol Soc 2004;10(1):26e34. Lazeron RH, Rombouts SA, de Sonneville L, Barkhof F, Scheltens P. A paced visual serial addition test for fMRI. J Neurol Sci 2003;213(1e2):29e34. Wells A. Emotional disorders and metacognition: innovative cognitive therapy. New York: Wiley; 2000. Siegle GJ, Price RB, Jones NP, Ghinassi F, Painter T, Thase ME. You gotta work at it: pupillary indices of task focus are prognostic for response to a neurocognitive intervention for rumination in depression. Clin Psychol Sci; in press. Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al. The MiniInternational Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59(Suppl. 20):22e33. quiz 4e57. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979;134:382e9. Wechsler D. Wechsler test of adult reading. San Antonio, TX: The Psychological Corporation; 2001. Brunoni AR, Nitsche MA, Bolognini N, Bikson M, Wagner T, Merabet L, et al. Clinical research with transcranial direct current stimulation (tDCS): challenges and future directions. Brain Stimul 2012;5(3):175e95.

331

[34] Wager TD, Jonides J, Reading S. Neuroimaging studies of shifting attention: a meta-analysis. Neuroimage 2004;22(4):1679e93. [35] Miller EK. The prefrontal cortex and cognitive control. Nat Rev Neurosci 2000;1(1):59e65. [36] Duncan J, Owen AM. Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends Neurosci 2000;23(10):475e83. [37] Godefroy O, Rousseaux M. Divided and focused attention in patients with lesion of the prefrontal cortex. Brain Cogn 1996;30(2):155e74. [38] Calkins AW, Deveney CM, Weitzman ML, Hearon BA, Siegle GJ, Otto MW. The effects of prior cognitive control task exposure on responses to emotional tasks in healthy participants. Behav Cogn Psychother 2011;39(2):205e20. [39] Beck AT, Steer RA. Internal consistencies of the original and revised Beck Depression Inventory. J Clin Psychol 1984;40(6):1365e7. [40] Bradley MM, Lang PJ. Affective norms for English words (ANEW): stimuli, instruction manual and affective ratings. Gainesville, FL: The Center for Research in Psychophysiology, University of Florida; 1999. [41] Martin DM, Liu R, Alonzo A, Green M, Player MJ, Sachdev P, et al. Can transcranial direct current stimulation enhance outcomes from cognitive training? A randomized controlled trial in healthy participants. Int J Neuropsychopharmacol 2013;16(9):1927e36. [42] Hoy KE, Emonson MRL, Arnold SL, Thomson RH, Daskalakis ZJ, Fitzgerald PB. Testing the limits: investigating the effect of tDCS dose on working memory enhancement in healthy controls. Neuropsychologia 2013;51(9):1777e84. [43] Thickbroom GW, Mastaglia FL. Plasticity in neurological disorders and challenges for noninvasive brain stimulation (NBS). J Neuroeng Rehabil 2009;6:4. [44] Fitzgerald PB, Hoy K, McQueen S, Maller JJ, Herring S, Segrave R, et al. A randomized trial of rTMS targeted with MRI based neuro-navigation in treatment-resistant depression. Neuropsychopharmacology 2009;34 (5):1255e62. [45] Fitzgerald PB, Hoy K, McQueen S, Herring S, Segrave R, Been G, et al. Priming stimulation enhances the effectiveness of low-frequency right prefrontal cortex transcranial magnetic stimulation in major depression. J Clin Psychopharmacol 2008;28(1):52e8. [46] Fitzgerald PB, Hoy K, Gunewardene R, Slack C, Ibrahim S, Bailey M, et al. A randomized trial of unilateral and bilateral prefrontal cortex transcranial magnetic stimulation in treatment-resistant major depression. Psychol Med 2010:1e10. [47] Gotlib IH, Joormann J. Cognition and depression: current status and future directions. Ann Rev Clin Psychol 2010;6:285e312. [48] Harmer CJ, Goodwin GM, Cowen PJ. Why do antidepressants take so long to work? A cognitive neuropsychological model of antidepressant drug action. Br J Psychiatry 2009;195(2):102e8. [49] D’Esposito M, Postle BR, Rypma B. Prefrontal cortical contributions to working memory: evidence from event-related fMRI studies. Exp Brain Res 2000; 133(1):3e11. [50] Hoy KE, Segrave RA, Daskalakis ZJ, Fitzgerald PB. Investigating the relationship between cognitive change and antidepressant response following rTMS: a large scale retrospective study. Brain Stimul 2012;5(4):539e46. [51] Vanderhasselt MA, De Raedt R, Leyman L, Baeken C. Acute effects of repetitive transcranial magnetic stimulation on attentional control are related to antidepressant outcomes. J Psychiatry Neurosci 2009;34(2):119e26. [52] Wolkenstein L, Plewnia C. Amelioration of cognitive control in depression by transcranial direct current stimulation. Biol Psychiatry 2013;73(7):646e51. [53] Boggio PS, Bermpohl F, Vergara AO, Muniz AL, Nahas FH, Leme PB, et al. Go-nogo task performance improvement after anodal transcranial DC stimulation of the left dorsolateral prefrontal cortex in major depression. J Affect Disord 2007;101(1e3):91e8. [54] Ball TM, Sullivan S, Flagan T, Hitchcock CA, Simmons A, Paulus MP, et al. Selective effects of social anxiety, anxiety sensitivity, and negative affectivity on the neural bases of emotional face processing. Neuroimage 2012;59(2): 1879e87. [55] Etkin A, Prater KE, Schatzberg AF, Menon V, Greicius MD. Disrupted amygdalar subregion functional connectivity and evidence of a compensatory network in generalized anxiety disorder. Arch Gen Psychiatry 2009;66(12):1361e72. [56] Fava M. Diagnosis and definition of treatment-resistant depression. Biol Psychiatry 2003;53(8):649e59.