The use of the Psychiatric Electroencephalography Evaluation Registry (PEER) to personalize pharmacotherapy

One size does not fit all: notable individual variation in brain activity correlates of antidepressant treatment response

Introduction

To date, no robust electroencephalography (EEG) markers of antidepressant treatment response have been identified. Variable findings may arise from the use of group analyses, which neglect individual variation. Using a combination of group and single-participant analyses, we explored individual variability in EEG characteristics of treatment response.

Methods

Resting-state EEG data and Montgomery-Åsberg Depression Rating Scale (MADRS) symptom scores were collected from 43 patients with depression before, at 1 and 12 weeks of pharmacotherapy. Partial least squares (PLS) was used to: 1) identify group differences in EEG connectivity (weighted phase lag index) and complexity (multiscale entropy) between eventual medication responders and non-responders, and 2) determine whether group patterns could be identified in individual patients.

Results

Responders showed decreased alpha and increased beta connectivity, and early, widespread decreases in complexity over treatment. Non-responders showed an opposite connectivity pattern, and later, spatially confined decreases in complexity. Thus, as in previous studies, our group analyses identified significant differences between groups of patients with different treatment outcomes. These group-level EEG characteristics were only identified in ~40-60% of individual patients, as assessed quantitatively by correlating the spatiotemporal brain patterns between groups and individual results, and by independent raters through visualization.

Discussion

Our single-participant analyses suggest that substantial individual variation exists, and needs to be considered when investigating characteristics of antidepressant treatment response for potential clinical applicability.

Clinical trial registration

https://clinicaltrials.gov, identifier NCT00519428.

[EEG features in young female patients with depressive states at different stages of endogenous mental diseases]

OBJECTIVE

To search for neurophysiological correlates of the characteristics of the brain functional state in patients with endogenous depression with an ultra-high risk of developing psychosis in comparison with EEG parameters of patients without symptoms of a risk of developing psychosis and patients who have suffered psychotic episode.

MATERIAL AND METHODS

The study included 92 female patients, aged 16-26 years, at the stage of remission, divided into three groups: with depression without symptoms of ultra-high risk of developing psychosis (group 1, n=42), with depression and attenuated psychotic symptoms, but without a history of a psychotic episode (group 2, n=32) and with depression that developed after experiencing a psychotic episode (group 3, n=18). In all patients, pre-treatment multichannel background EEG was recorded with spectral power analysis in narrow frequency sub-bands.

RESULTS

According to EEG data, the functional state of the cerebral cortex of patients in group 1 at the stage of remission was approaching normal. The EEG of group 2 and group 3 differed from the EEG of group 1 by significantly lower values of EEG spectral power in the alpha3 sub-band (11-13 Hz) in the occipital leads and a significantly increased content of theta1 (4-6 Hz) activity in the central-parietal areas. Such EEG frequency structure of patients in groups 2 and 3 reflects a reduced functional state of associative areas, and may also indicate dysfunction of the frontal parts of the cerebral cortex. These EEG features of patients in groups 2 and 3 are consistent with a significantly greater severity of their positive and negative symptoms on SAPS and SANS compared to group 1.

CONCLUSION

In patients with depression at the stage of remission who have symptoms of an ultra-high risk of developing psychosis and in those who have suffered a psychotic episode, a reduced functional state of the associative and frontal areas of the cerebral cortex is noted, which may underlie the characteristics of their clinical condition.

Machine learning in biosignals processing for mental health: A narrative review

Machine Learning (ML) offers unique and powerful tools for mental health practitioners to improve evidence-based psychological interventions and diagnoses. Indeed, by detecting and analyzing different biosignals, it is possible to differentiate between typical and atypical functioning and to achieve a high level of personalization across all phases of mental health care. This narrative review is aimed at presenting a comprehensive overview of how ML algorithms can be used to infer the psychological states from biosignals. After that, key examples of how they can be used in mental health clinical activity and research are illustrated. A description of the biosignals typically used to infer cognitive and emotional correlates (e.g., EEG and ECG), will be provided, alongside their application in Diagnostic Precision Medicine, Affective Computing, and brain-computer Interfaces. The contents will then focus on challenges and research questions related to ML applied to mental health and biosignals analysis, pointing out the advantages and possible drawbacks connected to the widespread application of AI in the medical/mental health fields. The integration of mental health research and ML data science will facilitate the transition to personalized and effective medicine, and, to do so, it is important that researchers from psychological/ medical disciplines/health care professionals and data scientists all share a common background and vision of the current research.

[EEG predictors of therapeutic response in psychiatry]

The literature review provides data on one of the types of biomarkers - EEG predictors of the therapeutic response of patients with different types of mental pathology. It has been shown that the quantitative parameters of the electroencephalogram (EEG) recorded before the start of the treatment course reflect not only the current functional state of the patient's brain, but also its adaptive resources in terms of the possibility and magnitude of response to therapy. The identified EEG predictors of the therapeutic response in patients with depression, schizophrenia and some other mental disorders have a sufficiently high prognostic ability, sensitivity and specificity in determining responders and non-responders, make it possible to carry out a quantitative prediction of the patient's condition after a course of treatment, and also to assist the clinician in choosing medications for optimal therapy.

High-Gamma: A biological marker for suicide attempt in patients with depression

BACKGROUND

Quantitative EEG (qEEG) analysis can be used to evaluate brain correlates of human psychology in all aspects. As the gamma oscillations of qEEG rhythms are related to depression, and particularly to treatment resistance, they may also be related to suicidality.

AIM

The present study aimed to identify the neural correlates of suicidal ideation and suicide attempt in depression using qEEG, based on the hypothesis that gamma rhythm in patients with depression would be higher in patients with suicidal ideation and suicide attempt.

METHOD

qEEG were recorded in 533 participants (276 female). Groups were divided into the following: Non-suicidal (n = 218), Suicide Ideation (n = 211), Suicide Attempt (n = 74), and control (n = 30).

RESULTS

High-gamma power at the F4, Fz, C4, Cz, O2, F8, T5 and T6 regions was significantly higher in the Suicide Ideation than the other groups.

CONCLUSION

If confirmed by further studies, high-gamma rhythm has the potential to be used as a biomarker for screening suicidality.

Use of machine learning in predicting clinical response to transcranial magnetic stimulation in comorbid posttraumatic stress disorder and major depression: A resting state electroencephalography study

BACKGROUND

Repetitive transcranial magnetic stimulation (TMS) is clinically effective for major depressive disorder (MDD) and investigational for other conditions including posttraumatic stress disorder (PTSD). Understanding the mechanisms of TMS action and developing biomarkers predicting response remain important goals. We applied a combination of machine learning and electroencephalography (EEG), testing whether machine learning analysis of EEG coherence would (1) predict clinical outcomes in individuals with comorbid MDD and PTSD, and (2) determine whether an individual had received a TMS course.

METHODS

We collected resting-state 8-channel EEG before and after TMS (5 Hz to the left dorsolateral prefrontal cortex). We used Lasso regression and Support Vector Machine (SVM) to test the hypothesis that baseline EEG coherence predicted the outcome and to assess if EEG coherence changed after TMS.

RESULTS

In our sample, clinical response to TMS were predictable based on pretreatment EEG coherence (n = 29). After treatment, 13/29 had more than 50% reduction in MDD self-report score 12/29 had more than 50% reduction in PTSD self-report score. For MDD, area under roc curve was for MDD was 0.83 (95% confidence interval 0.69-0.94) and for PTSD was 0.71 (95% confidence interval 0.54-0.87). SVM classifier was able to accurately assign EEG recordings to pre- and post-TMS treatment. The accuracy for Alpha, Beta, Theta and Delta bands was 75.4 ± 1.5%, 77.4 ± 1.4%, 73.8 ± 1.5%, and 78.6 ± 1.4%, respectively, all significantly better than chance (50%, p < 0.001).

LIMITATION

Limitations of this work include lack of sham condition, modest sample size, and a sparse electrode array. Despite these methodological limitations, we found validated and clinically meaningful results.

CONCLUSIONS

Machine learning successfully predicted non-response to TMS with high specificity, and identified pre- and post-TMS status using EEG coherence. This approach may provide mechanistic insights and may also become a clinically useful screening tool for TMS candidates.

Machine learning and big data in psychiatry: toward clinical applications

Psychiatry is a medical field concerned with the treatment of mental illness. Psychiatric disorders broadly relate to higher functions of the brain, and as such are richly intertwined with social, cultural, and experiential factors. This makes them exquisitely complex phenomena that depend on and interact with a large number of variables. Computational psychiatry provides two ways of approaching this complexity. Theory-driven computational approaches employ mechanistic models to make explicit hypotheses at multiple levels of analysis. Data-driven machine-learning approaches can make predictions from high-dimensional data and are generally agnostic as to the underlying mechanisms. Here, we review recent advances in the use of big data and machine-learning approaches toward the aim of alleviating the suffering that arises from psychiatric disorders.

The Concise Health Risk Tracking-Self Report: Psychometrics within a placebo-controlled antidepressant trial among depressed outpatients

BACKGROUND/AIMS

While substantial prior research has evaluated the psychometric properties of the 12-item Concise Health Risk Tracking-Self Report (CHRT-SR₁₂), a measure of suicide propensity and suicidal thoughts, no prior research has investigated its factor structure, sensitivity to change over time, and other psychometric properties in a placebo-controlled trial of antidepressant medication, nor determined whether symptoms change throughout treatment.

METHODS

Participants in the multi-site Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care (EMBARC) study ( n=278) provided data to evaluate the factor structure and sensitivity to change over time of the CHRT-SR₁₂ through eight weeks of a clinical trial in which participants received either placebo or antidepressant medication (sertraline).

RESULTS/OUTCOMES

Factor analysis confirmed two factors: propensity (comprised of first-order factors including pessimism, helplessness, social support, and despair) and suicidal thoughts. Internal consistency (α's ranged from 0.69-0.92) and external validity were both acceptable, with the total score and propensity factor scores significantly correlated with total scores and single-item suicidal-thoughts scores on the self-report Quick Inventory of Depressive Symptoms and the clinician-rated 17-item Hamilton Rating Scale for Depression. Through analyzing CHRT-SR₁₂ changes over eight treatment weeks, the total score and both the factors decreased regardless of baseline suicidal thoughts. Change in clinician-rated suicidal thoughts was reflected by change in both the total score and propensity factor score.

CONCLUSIONS/INTERPRETATION

These results confirm the reliability, validity, and applicability of the CHRT-SR₁₂ to a placebo-controlled clinical trial of depressed outpatients receiving antidepressant medication.

A Promising New Strategy to Improve Treatment Outcomes for Patients with Depression

Each year, ineffective medical management of patients with mental illness compromises the health and well-being of individuals, and also impacts communities and our society. A variety of interrelated factors have impeded the health system's ability to treat patients with behavior health conditions adequately. A key contributing factor is a lack of objective markers to help predict patient response to specific drugs that has led to patterns of “trial and error” prescribing. For many years, clinicians have sought objective data (eg, a laboratory or imaging test) to assist them in selecting appropriate treatments for individual patients. Electroencephalogram (EEG) findings coupled with medication outcomes data may provide a solution. “Crowdsourced” physician registries that reference clinical outcomes to individual patient physiology have been used successfully for cancers. These techniques are now being explored in the context of behavioral health care. The Psychiatric EEG Evaluation Registry (PEER) is one such approach. PEER is a clinical phenotypic database comprising more than 11,000 baseline EEGs and more than 39,000 outcomes of medication treatment for a variety of mental health diagnoses. Collective findings from 45 studies (3130 patients) provide compelling evidence for PEER as a relatively simple, inexpensive predictor of likely patient response to specific antidepressants and likely treatment-related side effects (including suicidal ideation).

A longitudinal investigation of the Concise Health Risk Tracking Self-Report (CHRT-SR) in suicidal patients during and after hospitalization

The Concise Health Risk Tracking Self-Report (CHRT-SR) scale is a brief self-report instrument to assess suicide risk. Initial investigations have indicated good psychometric properties in psychiatric outpatients. The aims of this paper were to examine the construct validity and factor structure of the twelve- (CHRT-SR12) and seven-item (CHRT-SR7) versions and to test if clinically expected within-person changes in suicide risk over time were measurable using the CHRT-SR in two study cohorts hospitalized for suicidal ideation or behavior: (1) patients with major depressive disorder (MDD) who participated in a psychological intervention trial, n = 65, and (2) participants with bipolar disorder or MDD in an observational study, n = 44. The CHRT-SR12 and self-report measures of hopelessness, depression, and positive psychological states were administered during admission and several times post-discharge. Both versions showed good internal consistency in inpatients and confirmed the three-factor structure (i.e., hopelessness, perceived lack of social support and active suicidal ideation and plans) found in outpatients. CHRT-SR scores had strong correlations with negative and positive affective constructs in the expected directions, and indicated decreases in suicide risk following discharge, in line with clinical expectations. The CHRT-SR12 and CHRT-SR7 are promising self-report measures for assessing suicide risk in very high-risk patient populations.

Quantitative Electroencephalography in Guiding Treatment of Major Depression

This paper reviews significant contributions to the evidence for the use of quantitative electroencephalography features as biomarkers of depression treatment and examines the potential of such technology to guide pharmacotherapy. Frequency band abnormalities such as alpha and theta band abnormalities have shown promise as have combinatorial measures such as cordance (a measure combining alpha and theta power) and the Antidepressant Treatment Response Index in predicting medication treatment response. Nevertheless, studies have been hampered by methodological problems and inconsistencies, and these approaches have ultimately failed to elicit any significant interest in actual clinical practice. More recent machine learning approaches such as the Psychiatric Encephalography Evaluation Registry (PEER) technology and other efforts analyze large datasets to develop variables that may best predict response rather than test a priori hypotheses. PEER is a technology that may go beyond predicting response to a particular antidepressant and help to guide pharmacotherapy.

Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical Data

Background: Individuals with major depressive disorder (MDD) vary in their response to antidepressants. However, identifying objective biomarkers, prior to or early in the course of treatment that can predict antidepressant efficacy, remains a challenge. Methods: Individuals with MDD participated in a 12-week antidepressant pharmacotherapy trial. Electroencephalographic (EEG) data was collected before and 1 week post-treatment initiation in 51 patients. Response status at week 12 was established with the Montgomery-Asberg Depression Scale (MADRS), with a ≥50% decrease characterizing responders (N = 27/24 responders/non-responders). We used a machine learning (ML)-approach for predicting response status. We focused on Random Forests, though other ML methods were compared. First, we used a tree-based estimator to select a relatively small number of significant features from: (a) demographic/clinical data (age, sex, individual item/total MADRS scores at baseline, week 1, change scores); (b) scalp-level EEG power; (c) source-localized current density (via exact low-resolution electromagnetic tomography [eLORETA] software). Second, we applied kernel principal component analysis to reduce and map important features. Third, a set of ML models were constructed to classify response outcome based on mapped features. For each dataset, predictive features were extracted, followed by a model of all predictive features, and finally by a model of the most predictive features. Results: Fifty eLORETA features were predictive of response (across bands, both time-points); alpha₁/theta eLORETA features showed the highest predictive value. Eighty-eight scalp EEG features were predictive of response (across bands, both time-points), with theta/alpha₂ being most predictive. Clinical/demographic data consisted of 31 features, with the most important being week 1 "concentration difficulty" scores. When all features were included into one model, its predictive utility was high (88% accuracy). When the most important features were extracted in the final model, 12 predictive features emerged (78% accuracy), including baseline scalp-EEG frontopolar theta, parietal alpha₂ and frontopolar alpha₁. Conclusions: These findings suggest that ML models of pre- and early treatment-emergent EEG profiles and clinical features can serve as tools for predicting antidepressant response. While this must be replicated using large independent samples, it lays the groundwork for research on personalized, "biomarker"-based treatment approaches.