Quantitative biomarkers to predict response to clozapine treatment using resting EEG data

Advancing treatment response prediction in first-episode psychosis: integrating clinical and electroencephalography features

AIMS

Prompt diagnosis and intervention are crucial for first-episode psychosis (FEP) outcomes, but predicting the response to antipsychotics remains challenging. We studied whether adding electroencephalography (EEG) characteristics improves clinical prediction models for treatment response and whether EEG-based predictors are influenced by initial treatment.

METHODS

We included 115 antipsychotic-naïve patients with FEP. Positive and Negative Syndrome Scale (PANSS) and sociodemographic items were included as clinical features. Additionally, we analyzed resting-state EEG data (n = 45) for (relative) power, functional connectivity, and network organization. Treatment response, measured as change in PANSS positive subscale scores (∆PANSS+), was predicted using a random forest regression model. We analyzed whether the most predictive EEG characteristics were influenced after treatment.

RESULTS

The clinical model explained 12% variance in symptom reduction in the training set and 32% in the validation set. Including EEG variables in the model led to a nonsignificant increase of 2% (total 34%) explained variance in symptom reduction. High hallucination symptom scores and a more hierarchical organization of alpha band networks (tree hierarchy) were associated with ∆PANSS+ reduction. The tree hierarchy in the alpha band decreased after medication. EEG source analysis revealed that this change was driven by alterations in the degree and centrality of frontal and parietal nodes in the functional brain network.

CONCLUSIONS

Both clinical and EEG characteristics can inform treatment response prediction in patients with FEP, but the combined model may not be beneficial over a clinical model. Nevertheless, adding a more objective marker such as EEG could be valuable in selected cases.

Predictive utility of artificial intelligence on schizophrenia treatment outcomes: A systematic review and meta-analysis

Identifying optimal treatment approaches for schizophrenia is challenging due to varying symptomatology and treatment responses. Artificial intelligence (AI) shows promise in predicting outcomes, prompting this systematic review and meta-analysis to evaluate various AI models' predictive utilities in schizophrenia treatment. A systematic search was conducted, and the risk of bias was evaluated. The pooled sensitivity, specificity, and diagnostic odds ratio with 95 % confidence intervals between AI models and the reference standard for response to treatment were assessed. Diagnostic accuracy measures were calculated, and subgroup analysis was performed based on the input data of AI models. Out of the 21 included studies, AI models achieved a pooled sensitivity of 70 % and specificity of 76 % in predicting schizophrenia treatment response with substantial predictive capacity and a near-to-high level of test accuracy. Subgroup analysis revealed EEG-based models to have the highest sensitivity (89 %) and specificity (94 %), followed by imaging-based models (76 % and 80 %, respectively). However, significant heterogeneity was observed across studies in treatment response definitions, participant characteristics, and therapeutic interventions. Despite methodological variations and small sample sizes in some modalities, this study underscores AI's predictive utility in schizophrenia treatment, offering insights for tailored approaches, improving adherence, and reducing relapse risk.

Clozapine Toxicity Predictor: Deep neural network model predicting clozapine toxicity and its therapeutic dose range

Clozapine is the gold standard for treatment-resistant schizophrenia; however, its superior efficacy is accompanied by potentially serious adverse events (neutropenia, seizures, constipations, pneumonia), many of which are also concentration-dependent. As such, clozapine dose titration should be guided by therapeutic drug monitoring (TDM). However, access to TDM is often limited. The present study describes a new deep neural network that can predict the concentrations, toxicity and therapeutic dose range for clozapine and norclozapine. The model was trained on basic clinical data (biological sex, age, clozapine daily dose, BMI, CRP and number of CYP 1A2 and 3A4 substrates, inhibitors and inducers) from 69 patients with treatment-refractory patients treated with different clozapine doses. Our findings provide the training efficacy data for the model, as well as an analysis of clozapine and norclozapine blood concentrations in a test group of three additional patients, to demonstrate its practical capabilities. The model is licensed on a free and permissive 2-Clause BSD license and is available to all clinicians; it can be accessed as a web application, available at https://csk.umed.pl/clotop.

Using deep learning and pretreatment EEG to predict response to sertraline, bupropion, and placebo

OBJECTIVE

Predicting an individual's response to antidepressant medication remains one of the most challenging tasks in the treatment of major depressive disorder (MDD). Our objective was to use the large EMBARC study database to develop an electroencephalography (EEG)-based method to predict response to antidepressant treatment.

METHODS

Pre-treatment EEG data were collected from study participants treated with either sertraline (N = 105), placebo (N = 119), or bupropion (N = 35). After preprocessing, the robust exact low-resolution electromagnetic tomography (ReLORETA) brain source localization method was used to reconstruct the source signals in 54 brain regions. Connectivity between regions was determined using symbolic transfer entropy (STE). A convolutional neural network (CNN) classified participants as responders or non-responders to each treatment.

RESULTS

Classification accuracy was 91.0%, 95.4%, and 86.8% for sertraline, placebo, and bupropion, respectively. The most highly predictive features were connectivity between i) the anterior cingulate cortex and superior parietal lobule (alpha frequency), ii) the anterior cingulate cortex and orbitofrontal area (beta frequency), and iii) the orbitofrontal area and anterior cingulate cortex (gamma frequency).

CONCLUSION

CNN analysis of EEG connectivity may accurately predict response to sertraline, bupropion, and placebo.

SIGNIFICANCE

The suggested method may offer clinicians an accessible and cost-effective tool for speedy treatment and helps pharmaceutical firms to test new antidepressants efficiently.

Diagnosing Suicidal Ideation from Resting State EEG Data Using a Machine Learning Algorithm

Suicide poses a global health crisis with significant social and economic impact. Prevention may be possible if objective quantitative methods are developed to supplement the often inaccurate interview-based risk assessments. Our research goal is to develop a machine learning algorithm (MLA) to predict the presence of suicide ideation from resting state electroencephalography (EEG) data collected from 224 subjects with major depressive disorder (MDD) in the Establishing Moderators and Biosignatures of Antidepressant Response for Clinical Care for Depression (EMBARC) study. Using the Concise Health Risk Tracking Self-Report (CHRT-SR14) questionnaire, 194 subjects acknowledged having suicidal ideation (group 1) and 30 did not (group 2). We balanced the database by matching 30 subjects from group 1 using propensity score analysis. A four-step prediction algorithm was then applied to the selected data including 1) EEG data preprocessing, 2) brain source localization (BSL) using the robust exact low-resolution electromagnetic tomography (ReLORETA) method, 3) determining the connectivity between the brain regions using symbolic transfer entropy (STE), 4) applying MLA to the STE features. Three common classifiers, Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbor (KNN) were used in this study. Using 70% of the data for training and evaluation and 30% for testing, all three classifiers delivered a high accuracy, where the highest performance belonged to SVM with 88.9% accuracy. These findings support the potential utility of ML analysis of EEG data as a non-verbal way to enhance the accuracy of suicide risk evaluation.

Multimodal workflows optimally predict response to repetitive transcranial magnetic stimulation in patients with schizophrenia: a multisite machine learning analysis

The response variability to repetitive transcranial magnetic stimulation (rTMS) challenges the effective use of this treatment option in patients with schizophrenia. This variability may be deciphered by leveraging predictive information in structural MRI, clinical, sociodemographic, and genetic data using artificial intelligence. We developed and cross-validated rTMS response prediction models in patients with schizophrenia drawn from the multisite RESIS trial. The models incorporated pre-treatment sMRI, clinical, sociodemographic, and polygenic risk score (PRS) data. Patients were randomly assigned to receive active (N = 45) or sham (N = 47) rTMS treatment. The prediction target was individual response, defined as ≥20% reduction in pre-treatment negative symptom sum scores of the Positive and Negative Syndrome Scale. Our multimodal sequential prediction workflow achieved a balanced accuracy (BAC) of 94% (non-responders: 92%, responders: 95%) in the active-treated group and 50% in the sham-treated group. The clinical, clinical + PRS, and sMRI-based classifiers yielded BACs of 65%, 76%, and 80%, respectively. Apparent sadness, inability to feel, educational attainment PRS, and unemployment were most predictive of non-response in the clinical + PRS model, while grey matter density reductions in the default mode, limbic networks, and the cerebellum were most predictive in the sMRI model. Our sequential modelling approach provided superior predictive performance while minimising the diagnostic burden in the clinical setting. Predictive patterns suggest that rTMS responders may have higher levels of brain grey matter in the default mode and salience networks which increases their likelihood of profiting from plasticity-inducing brain stimulation methods, such as rTMS. The future clinical implementation of our models requires findings to be replicated at the international scale using stratified clinical trial designs.

The association between clinical, sociodemographic, familial, and environmental factors and treatment resistance in schizophrenia: A machine-learning-based approach

BACKGROUND

Prediction of treatment resistance in schizophrenia (TRS) would be helpful to reduce the duration of ineffective treatment and avoid delays in clozapine initiation. We applied machine learning to identify clinical, sociodemographic, familial, and environmental variables that are associated with TRS and could potentially predict TRS in the future.

STUDY DESIGN

Baseline and follow-up data on trait(-like) variables from the Genetic Risk and Outcome of Psychosis (GROUP) study were used. For the main analysis, we selected patients with non-affective psychotic disorders who met TRS (n = 200) or antipsychotic-responsive criteria (n = 423) throughout the study. For a sensitivity analysis, we only selected patients who met TRS (n = 76) or antipsychotic-responsive criteria (n = 123) at follow-up but not at baseline. Random forest models were trained to predict TRS in both datasets. SHapley Additive exPlanation values were used to examine the variables' contributions to the prediction.

STUDY RESULTS

Premorbid functioning, age at onset, and educational degree were most consistently associated with TRS across both analyses. Marital status, current household, intelligence quotient, number of moves, and family loading score for substance abuse also consistently contributed to the prediction of TRS in the main or sensitivity analysis. The diagnostic performance of our models was modest (area under the curve: 0.66-0.69).

CONCLUSIONS

We demonstrate that various clinical, sociodemographic, familial, and environmental variables are associated with TRS. Our models only showed modest performance in predicting TRS. Prospective large multi-centre studies are needed to validate our findings and investigate whether the model's performance can be improved by adding data from different modalities.

Machine learning methods to predict outcomes of pharmacological treatment in psychosis

In recent years, machine learning (ML) has been a promising approach in the research of treatment outcome prediction in psychosis. In this study, we reviewed ML studies using different neuroimaging, neurophysiological, genetic, and clinical features to predict antipsychotic treatment outcomes in patients at different stages of schizophrenia. Literature available on PubMed until March 2022 was reviewed. Overall, 28 studies were included, among them 23 using a single-modality approach and 5 combining data from multiple modalities. The majority of included studies considered structural and functional neuroimaging biomarkers as predictive features used in ML models. Specifically, functional magnetic resonance imaging (fMRI) features contributed to antipsychotic treatment response prediction of psychosis with good accuracies. Additionally, several studies found that ML models based on clinical features might present adequate predictive ability. Importantly, by examining the additive effects of combining features, the predictive value might be improved by applying multimodal ML approaches. However, most of the included studies presented several limitations, such as small sample sizes and a lack of replication tests. Moreover, considerable clinical and analytical heterogeneity among included studies posed a challenge in synthesizing findings and generating robust overall conclusions. Despite the complexity and heterogeneity of methodology, prognostic features, clinical presentation, and treatment approaches, studies included in this review suggest that ML tools may have the potential to predict treatment outcomes of psychosis accurately. Future studies need to focus on refining feature characterization, validating prediction models, and evaluate their translation in real-world clinical practice.

Discriminating between bipolar and major depressive disorder using a machine learning approach and resting-state EEG data

OBJECTIVE

Distinguishing major depressive disorder (MDD) from bipolar disorder (BD) is a crucial clinical challenge as effective treatment is quite different for each condition. In this study electroencephalography (EEG) was explored as an objective biomarker for distinguishing MDD from BD using an efficient machine learning algorithm (MLA) trained by a relatively large and balanced dataset.

METHODS

A 3 step MLA was applied: (1) a multi-step preprocessing method was used to improve the quality of the EEG signal, (2) symbolic transfer entropy (STE), an effective connectivity measure, was applied to the resultant EEG and (3) the MLA used the extracted STE features to distinguish MDD (N = 71) from BD (N = 71) subjects.

RESULTS

14 connectivity features were selected by the proposed algorithm. Most of the selected features were related to the frontal, parietal, and temporal lobe electrodes. The major involved regions were the Broca region in the frontal lobe and the somatosensory association cortex in the parietal lobe. These regions are near electrodes FC5 and CPz and are involved in processing language and sensory information, respectively. The resulting classifier delivered an evaluation accuracy of 88.5% and a test accuracy of 89.3%, using 80% of the data for training and evaluation and the remaining 20% for testing, respectively.

CONCLUSIONS

The high evaluation and test accuracies of our algorithm, derived from a large balanced training sample suggests that this method may hold significant promise as a clinical tool.

SIGNIFICANCE

The proposed MLA may provide an inexpensive and readily available tool that clinicians may use to enhance diagnostic accuracy and shorten time to effective treatment.

Macroscale EEG characteristics in antipsychotic-naïve patients with first-episode psychosis and healthy controls

Electroencephalography in patients with a first episode of psychosis (FEP) may contribute to the diagnosis and treatment response prediction. Findings in the literature vary due to small sample sizes, medication effects, and variable illness duration. We studied macroscale resting-state EEG characteristics of antipsychotic naïve patients with FEP. We tested (1) for differences between FEP patients and controls, (2) if EEG could be used to classify patients as FEP, and (3) if EEG could be used to predict treatment response to antipsychotic medication. In total, we studied EEG recordings of 62 antipsychotic-naïve patients with FEP and 106 healthy controls. Spectral power, phase-based and amplitude-based functional connectivity, and macroscale network characteristics were analyzed, resulting in 60 EEG variables across four frequency bands. Positive and Negative Symptom Scale (PANSS) were assessed at baseline and 4-6 weeks follow-up after treatment with amisulpride or aripiprazole. Mann-Whitney U tests, a random forest (RF) classifier and RF regression were used for statistical analysis. Our study found that at baseline, FEP patients did not differ from controls in any of the EEG characteristics. A random forest classifier showed chance-level discrimination between patients and controls. The random forest regression explained 23% variance in positive symptom reduction after treatment in the patient group. In conclusion, in this largest antipsychotic- naïve EEG sample to date in FEP patients, we found no differences in macroscale EEG characteristics between patients with FEP and healthy controls. However, these EEG characteristics did show predictive value for positive symptom reduction following treatment with antipsychotic medication.

Machine Learning and Non-Affective Psychosis: Identification, Differential Diagnosis, and Treatment

PURPOSE OF REVIEW

This review will cover the most relevant findings on the use of machine learning (ML) techniques in the field of non-affective psychosis, by summarizing the studies published in the last three years focusing on illness detection and treatment.

RECENT FINDINGS

Multiple ML tools that include mostly supervised approaches such as support vector machine, gradient boosting, and random forest showed promising results by applying these algorithms to various sources of data: socio-demographic information, EEG, language, digital content, blood biomarkers, neuroimaging, and electronic health records. However, the overall performance, in the binary classification case, varied from 0.49, which is to be considered very low (i.e., noise), to over 0.90. These results are fully justified by different factors, some of which may be attributable to the preprocessing of the data, the wide variety of the data, and the a-priori setting of hyperparameters. One of the main limitations of the field is the lack of stratification of results based on biological sex, given that psychosis presents differently in men and women; hence, the necessity to tailor identification tools and data analytic strategies. Timely identification and appropriate treatment are key factors in reducing the consequences of psychotic disorders. In recent years, the emergence of new analytical tools based on artificial intelligence such as supervised ML approaches showed promises as a potential breakthrough in this field. However, ML applications in everyday practice are still in its infancy.

Diagnosing Schizophrenia Using Effective Connectivity of Resting-State EEG Data

Schizophrenia is a serious mental illness associated with neurobiological deficits. Even though the brain activities during tasks (i.e., P300 activities) are considered as biomarkers to diagnose schizophrenia, brain activities at rest have the potential to show an inherent dysfunctionality in schizophrenia and can be used to understand the cognitive deficits in these patients. In this study, we developed a machine learning algorithm (MLA) based on eyes closed resting-state electroencephalogram (EEG) datasets, which record the neural activity in the absence of any tasks or external stimuli given to the subjects, aiming to distinguish schizophrenic patients (SCZs) from healthy controls (HCs). The MLA has two steps. In the first step, symbolic transfer entropy (STE), which is a measure of effective connectivity, is applied to resting-state EEG data. In the second step, the MLA uses the STE matrix to find a set of features that can successfully discriminate SCZ from HC. From the results, we found that the MLA could achieve a total accuracy of 96.92%, with a sensitivity of 95%, a specificity of 98.57%, precision of 98.33%, F1-score of 0.97, and Matthews correlation coefficient (MCC) of 0.94 using only 10 out of 1900 STE features, which implies that the STE matrix extracted from resting-state EEG data may be a promising tool for the clinical diagnosis of schizophrenia.

Pattern classification as decision support tool in antipsychotic treatment algorithms

Pattern classification aims to establish a new approach in personalized treatment. The scope is to tailor treatment on individual characteristics during all phases of care including prevention, diagnosis, treatment, and clinical outcome. In psychotic disorders, this need results from the fact that a third of patients with psychotic symptoms do not respond to antipsychotic treatment and are described as having treatment-resistant disorders. This, in addition to the high variability of treatment responses among patients, enhances the need of applying advanced classification algorithms to identify antipsychotic treatment patterns. This review comprehensively summarizes advancements and challenges of pattern classification in antipsychotic treatment response to date and aims to introduce clinicians and researchers to the challenges of including pattern classification into antipsychotic treatment decision algorithms.