A systematic meta-review of predictors of antidepressant treatment outcome in major depressive disorder

Association of CYP2D6 and CYP2C19 metabolizer status with switching and discontinuing antidepressant drugs: an exploratory study

BACKGROUND

Tailoring antidepressant drugs (AD) to patients' genetic drug-metabolism profile is promising. However, literature regarding associations of ADs' treatment effect and/or side effects with drug metabolizing genes CYP2D6 and CYP2C19 has yielded inconsistent results. Therefore, our aim was to longitudinally investigate associations between CYP2D6 (poor, intermediate, and normal) and CYP2C19 (poor, intermediate, normal, and ultrarapid) metabolizer-status, and switching/discontinuing of ADs. Next, we investigated whether the number of perceived side effects differed between metabolizer statuses.

METHODS

Data came from the multi-site naturalistic longitudinal cohort Netherlands Study of Depression and Anxiety (NESDA). We selected depression- and/or anxiety patients, who used AD at some point in the course of the 9 years follow-up period (n = 928). Medication use was followed to assess patterns of AD switching/discontinuation over time. CYP2D6 and CYP2C19 alleles were derived using genome-wide data of the NESDA samples and haplotype data from the PharmGKB database. Logistic regression analyses were conducted to investigate the association of metabolizer status with switching/discontinuing ADs. Mann-Whitney U-tests were conducted to compare the number of patient-perceived side effects between metabolizer statuses.

RESULTS

No significant associations were observed of CYP metabolizer status with switching/discontinuing ADs, nor with the number of perceived side effects.

CONCLUSIONS

We found no evidence for associations between CYP metabolizer statuses and switching/discontinuing AD, nor with side effects of ADs, suggesting that metabolizer status only plays a limited role in switching/discontinuing ADs. Additional studies with larger numbers of PM and UM patients are needed to further determine the potential added value of pharmacogenetics to guide pharmacotherapy.

Predicting Outcomes of Antidepressant Treatment in Community Practice Settings

OBJECTIVE

The authors examined whether machine-learning models could be used to analyze data from electronic health records (EHRs) to predict patients' responses to antidepressant medications.

METHODS

EHR data from a Washington State health system identified patients ages ≥13 years who started an antidepressant medication in 2016 in a community practice setting and had a baseline Patient Health Questionnaire-9 (PHQ-9) score of ≥10 and at least one PHQ-9 score recorded 14-180 days later. Potential predictors of a response to antidepressants were extracted from the EHR and included demographic characteristics, psychiatric and substance use diagnoses, past psychiatric medication use, mental health service use, and past PHQ-9 scores. Random-forest and penalized regression analyses were used to build models predicting follow-up PHQ-9 score and a favorable treatment response (≥50% improvement in score).

RESULTS

Among 2,469 patients starting antidepressant medication treatment, the mean±SD baseline PHQ-9 score was 17.3±4.5, and the mean lowest follow-up score was 9.2±5.9. Outcome data were available for 72% of the patients. About 48% of the patients had a favorable treatment response. The best-fitting random-forest models yielded a correlation between predicted and observed follow-up scores of 0.38 (95% CI=0.32-0.45) and an area under the receiver operating characteristic curve for a favorable response of 0.57 (95% CI=0.52-0.61). Results were similar for penalized regression models and for models predicting last PHQ-9 score during follow-up.

CONCLUSIONS

Prediction models using EHR data were not accurate enough to inform recommendations for or against starting antidepressant medication. Personalization of depression treatment should instead rely on systematic assessment of early outcomes.

Olfactory genes affect major depression in highly educated, emotionally stable, lean women: a bridge between animal models and precision medicine

Most current approaches to establish subgroups of depressed patients for precision medicine aim to rely on biomarkers that require highly specialized assessment. Our present aim was to stratify participants of the UK Biobank cohort based on three readily measurable common independent risk factors, and to investigate depression genomics in each group to discover common and separate biological etiology. Two-step cluster analysis was run separately in males (n = 149,879) and females (n = 174,572), with neuroticism (a tendency to experience negative emotions), body fat percentage, and years spent in education as input variables. Genome-wide association analyses were implemented within each of the resulting clusters, for the lifetime occurrence of either a depressive episode or recurrent depressive disorder as the outcome. Variant-based, gene-based, gene set-based, and tissue-specific gene expression test were applied. Phenotypically distinct clusters with high genetic intercorrelations in depression genomics were found. A two-cluster solution was the best model in each sex with some differences including the less important role of neuroticism in males. In females, in case of a protective pattern of low neuroticism, low body fat percentage, and high level of education, depression was associated with pathways related to olfactory function. While also in females but in a risk pattern of high neuroticism, high body fat percentage, and less years spent in education, depression showed association with complement system genes. Our results, on one hand, indicate that alteration of olfactory pathways, that can be paralleled to the well-known rodent depression models of olfactory bulbectomy, might be a novel target towards precision psychiatry in females with less other risk factors for depression. On the other hand, our results in multi-risk females may provide a special case of immunometabolic depression.

Symptom clusters in adolescent depression and differential responses of clusters to pharmacologic treatment

OBJECTIVE

Symptoms of depression in adolescents are widely variable, but they are often interactive and clustered. The analysis of interactions and clusters among individual symptoms may help predict treatment outcomes. We aimed to determine clusters of individual symptoms in adolescent depression and their changes in the response to pharmacological treatment.

METHOD

A total of 95 adolescents, aged 12-17 years, with major depressive disorder were included. Participants were treated with escitalopram, and depressive symptoms were assessed at baseline (V1) and 1, 2, 4, 6, and 8 weeks (V6). The severity of depression was assessed using the Children's Depression Rating Scale-Revised. To construct network and clustering structures among symptoms, the Gaussian graphical model and Exploratory Graph Analysis with the tuning parameter to minimize the extended Bayesian information criterion were adopted.

RESULTS

Exploratory Graph Analysis revealed that symptoms of depression comprised four clusters: impaired activity, somatic concerns, subjective mood, and observed affect. The main effect of visit with decreased symptom severity was significant in all four clusters; however, the degree of symptom improvement differed among the four clusters. The effect size of score differences from V1 to V6 was the highest in the subjective mood (Cohen's d = 1.075), and lowest in impaired activity (d = 0.501) clusters.

CONCLUSION

The present study identified four symptom clusters associated with adolescent depression and their differential changes related to antidepressant treatment. This finding suggests that escitalopram was the most effective at improving subjective mood among different clusters. However, other therapeutic modalities may be needed to improve other clusters of symptoms, consequently leading to increased overall improvement of depression in adolescents.

Associations of allostatic load with sociodemographic factors, depressive symptoms, lifestyle, and health characteristics in a large general population-based sample

OBJECTIVE

We examined the associations between allostatic load (AL) and sociodemographic factors, depressive symptoms, lifestyle and health characteristics in a population-based sample of 4993 adults in Finland.

METHODS

Thirteen biomarkers were used to construct AL. High AL was defined as scoring highly in ≥4 items.

RESULTS

AL scores of 4 and above were exceeded in the age group of 45-54 years in men and 65-74 years in women. Age was the strongest predictor for belonging to the high AL score group. In addition, elevated depressive symptoms (BDI-6 ≥ 4), male sex, not engaging in physical exercise, high alcohol use and a low level of education were associated with an increased likelihood of belonging to the high AL group.

CONCLUSION

The older the participants were, the greater their AL burden was. However, AL burden increased more steeply as a function of age in men. In addition to lifestyle interventions, effective prevention strategies for depression at the population level could have a major public health impact in reducing the accumulation of AL burden.

Trajectories of remitted psychotic depression: identification of predictors of worsening by machine learning

BACKGROUND

Remitted psychotic depression (MDDPsy) has heterogeneity of outcome. The study's aims were to identify subgroups of persons with remitted MDDPsy with distinct trajectories of depression severity during continuation treatment and to detect predictors of membership to the worsening trajectory.

METHOD

One hundred and twenty-six persons aged 18-85 years participated in a 36-week randomized placebo-controlled trial (RCT) that examined the clinical effects of continuing olanzapine once an episode of MDDPsy had remitted with sertraline plus olanzapine. Latent class mixed modeling was used to identify subgroups of participants with distinct trajectories of depression severity during the RCT. Machine learning was used to predict membership to the trajectories based on participant pre-trajectory characteristics.

RESULTS

Seventy-one (56.3%) participants belonged to a subgroup with a stable trajectory of depression scores and 55 (43.7%) belonged to a subgroup with a worsening trajectory. A random forest model with high prediction accuracy (AUC of 0.812) found that the strongest predictors of membership to the worsening subgroup were residual depression symptoms at onset of remission, followed by anxiety score at RCT baseline and age of onset of the first lifetime depressive episode. In a logistic regression model that examined depression score at onset of remission as the only predictor variable, the AUC (0.778) was close to that of the machine learning model.

CONCLUSIONS

Residual depression at onset of remission has high accuracy in predicting membership to worsening outcome of remitted MDDPsy. Research is needed to determine how best to optimize the outcome of psychotic MDDPsy with residual symptoms.

Examining factors related to low performance of predicting remission in participants with major depressive disorder using neuroimaging data and other clinical features

Major depressive disorder (MDD), a prevalent mental health issue, affects more than 8% of the US population, and almost 17% in the young group of 18-25 years old. Since Covid-19, its prevalence has become even more significant. However, the remission (being free of depression) rates of first-line antidepressant treatments on MDD are only about 30%. To improve treatment outcomes, researchers have built various predictive models for treatment responses and yet none of them have been adopted in clinical use. One reason is that most predictive models are based on data from subjective questionnaires, which are less reliable. Neuroimaging data are promising objective prognostic factors, but they are expensive to obtain and hence predictive models using neuroimaging data are limited and such studies were usually in small scale (N<100). In this paper, we proposed an advanced machine learning (ML) pipeline for small training dataset with large number of features. We implemented multiple imputation for missing data and repeated K-fold cross validation (CV) to robustly estimate predictive performances. Different feature selection methods and stacking methods using 6 general ML models including random forest, gradient boosting decision tree, XGBoost, penalized logistic regression, support vector machine (SVM), and neural network were examined to evaluate the model performances. All predictive models were compared using model performance metrics such as accuracy, balanced accuracy, area under ROC curve (AUC), sensitivity and specificity. Our proposed ML pipeline was applied to a training dataset and obtained an accuracy and AUC above 0.80. But such high performance failed while applying our ML pipeline using an external validation dataset from the EMBARC study which is a multi-center study. We further examined the possible reasons especially the site heterogeneity issue.

Optimizing precision medicine for second-step depression treatment: a machine learning approach

BACKGROUND

Less than a third of patients with depression achieve successful remission with standard first-step antidepressant monotherapy. The process for determining appropriate second-step care is often based on clinical intuition and involves a protracted course of trial and error, resulting in substantial patient burden and unnecessary delay in the provision of optimal treatment. To address this problem, we adopt an ensemble machine learning approach to improve prediction accuracy of remission in response to second-step treatments.

METHOD

Data were derived from the Level 2 stage of the STAR*D dataset, which included 1439 patients who were randomized into one of seven different second-step treatment strategies after failing to achieve remission during first-step antidepressant treatment. Ensemble machine learning models, comprising several individual algorithms, were evaluated using nested cross-validation on 155 predictor variables including clinical and demographic measures.

RESULTS

The ensemble machine learning algorithms exhibited differential classification performance in predicting remission status across the seven second-step treatments. For the full set of predictors, AUC values ranged from 0.51 to 0.82 depending on the second-step treatment type. Predicting remission was most successful for cognitive therapy (AUC = 0.82) and least successful for other medication and combined treatment options (AUCs = 0.51-0.66).

CONCLUSION

Ensemble machine learning has potential to predict second-step treatment. In this study, predictive performance varied by type of treatment, with greater accuracy in predicting remission in response to behavioral treatments than to pharmacotherapy interventions. Future directions include considering more informative predictor modalities to enhance prediction of second-step treatment response.

Amygdala biomarkers of treatment response in major depressive disorder: An fMRI systematic review of SSRI antidepressants

Functional neuroimaging studies have demonstrated abnormal activity and functional connectivity (FC) of the amygdala among individuals with major depressive disorder (MDD), which may be rectified with selective serotonin reuptake inhibitor (SSRI) treatment. This systematic review aimed to identify changes in the amygdala on functional magnetic resonance imaging (fMRI) scans among individuals with MDD who received SSRIs. A search for fMRI studies examining amygdala correlates of SSRI response via fMRI was conducted through OVID (MEDLINE, PsycINFO, and Embase). The end date was April 4th, 2023. In total, 623 records were screened, and 16 studies were included in this review. While the search pertained to SSRIs broadly, the included studies were escitalopram-, citalopram-, fluoxetine-, sertraline-, and paroxetine-specific. Decreases in event-related amygdala activity were found following 6-to-12-week SSRI treatment, particularly in response to negative stimuli. Eight-week courses of SSRI pharmacotherapy were associated with increased event-related amygdala FC (i.e., with the prefrontal [PFC] and anterior cingulate cortices, insula, thalamus, caudate nucleus, and putamen) and decreased resting-state effective connectivity (i.e., amygdala-PFC). Preliminary evidence suggests that SSRIs may alter amygdala activity and FC in MDD. Additional studies are needed to corroborate findings. Future research should employ long-term follow-ups to determine whether effects persist after treatment termination.

Towards Outcome-Driven Patient Subgroups: A Machine Learning Analysis Across Six Depression Treatment Studies

BACKGROUND

Major depressive disorder (MDD) is a heterogeneous condition; multiple underlying neurobiological and behavioral substrates are associated with treatment response variability. Understanding the sources of this variability and predicting outcomes has been elusive. Machine learning (ML) shows promise in predicting treatment response in MDD, but its application is limited by challenges to the clinical interpretability of ML models, and clinicians often lack confidence in model results. In order to improve the interpretability of ML models in clinical practice, our goal was to demonstrate the derivation of treatment-relevant patient profiles comprised of clinical and demographic information using a novel ML approach.

METHODS

We analyzed data from six clinical trials of pharmacological treatment for depression (total n = 5438) using the Differential Prototypes Neural Network (DPNN), a ML model that derives patient prototypes which can be used to derive treatment-relevant patient clusters while learning to generate probabilities for differential treatment response. A model classifying remission and outputting individual remission probabilities for five first-line monotherapies and three combination treatments was trained using clinical and demographic data. Prototypes were evaluated for interpretability by assessing differences in feature distributions (e.g. age, sex, symptom severity) and treatment-specific outcomes.

RESULTS

A 3-prototype model achieved an area under the receiver operating curve of 0.66 and an expected absolute improvement in remission rate for those receiving the best predicted treatment of 6.5% (relative improvement of 15.6%) compared to the population remission rate. We identified three treatment-relevant patient clusters. Cluster A patients tended to be younger, to have increased levels of fatigue, and more severe symptoms. Cluster B patients tended to be older, female, have less severe symptoms, and the highest remission rates. Cluster C patients had more severe symptoms, lower remission rates, more psychomotor agitation, more intense suicidal ideation, and more somatic genital symptoms.

CONCLUSION

It is possible to produce novel treatment-relevant patient profiles using ML models; doing so may improve interpretability of ML models and the quality of precision medicine treatments for MDD.

The effect of inflammation markers on cortical thinning in major depressive disorder: A possible mediator of depression and cortical changes

BACKGROUND

Major depressive disorder (MDD) is a prevalent mental health condition with significant societal impact. Owing to the intricate biological diversity of MDD, treatment efficacy remains limited. Immune biomarkers have emerged as potential predictors of treatment response, underscoring the interaction between the immune system and the brain. This study investigated the relationship between cytokine levels and cortical thickness in patients with MDD, focusing on the corticolimbic circuit, to elucidate the influence of neuroinflammation on structural brain changes and contribute to a deeper understanding of the pathophysiology of MDD.

METHOD

A total of 114 patients with MDD and 101 healthy controls (HC) matched for age, sex, and body mass index (BMI) were recruited. All participants were assessed for depression severity using the Hamilton Depression Rating Scale (HDRS), and 3.0 T T1 weighted brain MRI data were acquired. Additionally, cytokine levels were measured using a highly sensitive bead-based multiplex immunosorbent assay.

RESULTS

Patients diagnosed with MDD exhibited notably elevated levels of interleukin-6 (p = 0.005) and interleukin-8 (p = 0.005), alongside significant cortical thinning in the left anterior cingulate gyrus and left superior frontal gyrus, with these findings maintaining significance even after applying Bonferroni correction. Furthermore, increased interleukin-6 and interleukin-8 levels in patients with MDD are associated with alterations in the left frontomarginal gyrus and right anterior cingulate cortex (ACC).

CONCLUSIONS

This suggests a potential influence of neuroinflammation on right ACC function in MDD patients, warranting longitudinal research to explore interleukin-6 and interleukin-8 mediated neurotoxicity in MDD vulnerability and brain morphology changes.

QEEG Predictors of Treatment Response in Major Depressive Disorder- A Replication Study from Northwest India

Background: Predicting treatment response with antidepressant is a challenging task for clinicians and researchers. An important limitation of an antidepressant trial is the increased time spent before an adequacy of trial can be decided. Quantitative Electroencephalography has shown some evidence in identifying early changes seen with antidepressants. No data has been reported from Indian population on its predictive capabilities. Aim: To examine whether early changes in frontal and prefrontal theta value in QEEG could predict antidepressant treatment response. Methods: Structured clinical assessments were conducted at baseline and after one week in a sample of treatment-seeking adults with major depressive disorder (n = 50). Patients were started on SSRI (Escitalopram, fluoxetine, paroxetine or sertraline) and followed for 8 weeks. QEEG recordings were carried out at baseline and week 1 and its parameters (relative theta power and cordance) were assessed to identify its predictive value for treatment response. Treatment response was assessed using Hamilton depression rating scale with 50% reduction after 8 weeks being considered as response. Results: Mean age of the sample was 39 ± 10 years and majority of them were females (64%). A significant reduction was found in relative frontal theta value (p = 0.021) from baseline to one week in responders. However, linear regression revealed that this change could not predict the treatment response (p = 0.37). Conclusions: QEEG changes are observed in initial phase of antidepressant treatment but these changes can't predict the treatment response.

Use of Serum Biomarkers to Aid Antidepressant Selection in Depressive Patients

Objective

: This study aimed to identify serum biomarkers prospectively associated with remission at 12 weeks in out-patients with depressive disorders receiving stepwise psychopharmacotherapy, according to the main antidepressant used during the treatment period.

Methods

: This study included 1,024 depressive outpatients initially treated using antidepressant monotherapy, followed by alternating pharmacological strategies during the acute phase (3-12 weeks; 3-week interval). Fourteen serum biomarkers, sociodemographics, and clinical characteristics were evaluated at baseline. Based on the use frequency and mechanism of action, four main antidepressant types were distinguished: escitalopram, other selective serotonin reuptake inhibitors (SSRIs), serotonin norepinephrine reuptake inhibitors (SNRIs), and mirtazapine. A Hamilton Depression Rating Scale score ≤ 7 was take to indicate remission.

Results

: Lower high-sensitivity C-reactive protein levels were correlated with remission at 12 weeks for all antidepressant types. Lower interleukin (IL)-6 levels and tumor necrosis factor-alpha levels were associated with remission using escitalopram and other SSRIs respectively. Lower IL-1β and leptin levels, predicted remission in association with SSRIs including escitalopram. For SNRIs, remission at 12 weeks was predicted by lower IL-4 and IL-10 levels. For mirtazapine, remission at 12 weeks was associated with lower leptin levels, and higher serotonin and folate levels.

Conclusion

: Baseline serum status, as estimated by nine serum markers, may help clinicians determine the most appropriate antidepressant to achieve remission in the acute phase of depression.

A common symptom geometry of mood improvement under sertraline and placebo associated with distinct neural patterns

Importance

Understanding the mechanisms of major depressive disorder (MDD) improvement is a key challenge to determine effective personalized treatments.

Objective

To perform a secondary analysis quantifying neural-to-symptom relationships in MDD as a function of antidepressant treatment.

Design

Double blind randomized controlled trial.

Setting

Multicenter.

Participants

Patients with early onset recurrent depression from the public Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care (EMBARC) study.

Interventions

Either sertraline or placebo during 8 weeks (stage 1), and according to response a second line of treatment for 8 additional weeks (stage 2).

Main Outcomes and Measures

To identify a data-driven pattern of symptom variations during these two stages, we performed a Principal Component Analysis (PCA) on the variations of individual items of four clinical scales measuring depression, anxiety, suicidal ideas and manic-like symptoms, resulting in a univariate measure of clinical improvement. We then investigated how initial clinical and neural factors predicted this measure during stage 1. To do so, we extracted resting-state global brain connectivity (GBC) at baseline at the individual level using a whole-brain functional network parcellation. In turn, we computed a linear model for each brain parcel with individual data-driven clinical improvement scores during stage 1 for each group.

Results

192 patients (127 women), age 37.7 years old (standard deviation: 13.5), were included. The first PC (PC1) capturing 20% of clinical variation was similar across treatment groups at stage 1 and stage 2, suggesting a reproducible pattern of symptom improvement. PC1 patients' scores significantly differed according to treatment during stage 1, whereas no difference of response was evidenced between groups with the Clinical Global Impressions (CGI). Baseline GBC correlated to stage 1 PC1 scores in the sertraline, but not in the placebo group.

Conclusions and Relevance

Using data-driven reduction of symptoms scales, we identified a common profile of symptom improvement across placebo and sertraline. However, the neural patterns of baseline that mapped onto symptom improvement distinguished between treatment and placebo. Our results underscore that mapping from data-driven symptom improvement onto neural circuits is vital to detect treatment-responsive neural profiles that may aid in optimal patient selection for future trials.

Prospective study of antidepressant treatment of psychiatric patients with depressive disorders: treatment adequacy and outcomes

BACKGROUND

Despite numerous national depression care guidelines (DCGs), suboptimal antidepressant treatment may occur. We examined DCG concordance and depression treatment outcomes in psychiatric settings.

METHODS

We evaluated treatment received and outcomes of 128 psychiatric out- and inpatients participating in the PEGAD (Pharmacoepidemiology and Pharmacogenetics of Antidepressant Treatment for Depressive Disorders) study at baseline, two weeks, and eight weeks using interviews and questionnaires. Inclusion criteria were ICD-10 diagnosis of a depressive disorder, a Patient Health Questionnaire-9 symptom (PHQ-9) score ≥ 10, and a new antidepressant prescribed. The primary outcome of the study was within-individual change in PHQ-9 scores.

RESULTS

At baseline, patients had predominately recurrent (83%) and in 19% treatment-resistant depression (TRD). The median preceding duration of the current episode was 6.5 months. At eight weeks, 85% of the patients (n = 107) used a DCG-concordant antidepressant dose. However, due to the scarcity of antidepressant combinations and augmentations, fewer TRD than non-TRD patients (25% vs. 84%, p < 0.005) received adequate antidepressant treatment. Additionally, one-third of the patients received inadequate follow-up. Overall, only 53% received treatment compatible with DCG recommendations for adequate pharmacotherapy and follow-up. The mean decline in PHQ-9 scores (-3.8 ± SD 5.7) was significant (p < 0.0005). Nearly 40% of the patients reached a subthreshold level of depression (PHQ-9 < 10), predicted by a lower baseline PHQ-9 score, recurrent depression, and female sex. However, 45% experienced no significant clinical improvement (PHQ-9 score reduction < 20%).

CONCLUSIONS

Our findings suggest that inadequate treatment continues to occur in psychiatric care settings, particularly for TRD patients.

Is human cytomegalovirus a potential risk factor for mood disorders? A systematic review and meta-analysis

Mood disorders are among the common mental disorders worldwide. Because of the persistence of cytomegalovirus (CMV) in the body and nervous system, this virus can be activated when the immune system is weakened and continues to exert its destructive effects throughout life. This study aimed to investigate the seroprevalence and association of human cytomegalovirus with mood disorders. Eligible articles were extracted using online international databases Science Direct, Medline, Web of Science, Scopus, and Google Scholar between 2000 and 2023. After quality assessment and specific inclusion and exclusion criteria, a total of eight eligible articles were included in the meta-analysis. Our finding showed that the seropositivity of CMV in mood disorders was 51.6% (95% CI; 42.8-60.4). There were statistical differences between mood disorders and control groups regarding the seropositivity of CMV 1.327% (95% CI; 13.27-10.45). The results of the publication bias using the Egger test confirmed no publication bias in each sub-group. The results of this meta-analysis study demonstrated that CMV infection might have associations with the incidence of mood disorders. Furthermore, we found that there were statistical differences between mood disorders and control groups regarding the seropositivity of CMV.

ICEBERG study: an indirect adjusted comparison estimating the long-term benefit of esketamine nasal spray when compared with routine treatment of treatment resistant depression in general psychiatry

Background

Treatment resistant depression (TRD) affects 10-30% of patients with major depressive disorder. In 4-week trials, esketamine nasal spray (NS) was efficacious vs. placebo when both were initiated in addition to a new selective serotonin or serotonin norepinephrine reuptake inhibitor. However, comparison with an extended range of real-world treatments (RWT) is lacking.

Methods

ICEBERG was an adjusted indirect treatment comparison using propensity score-based inverse probability weighting, performed on 6-month response and remission data from patients receiving esketamine NS plus oral antidepressant from the SUSTAIN-2 (NCT02497287; clinicaltrials.gov) study, compared with patients receiving other RWT from the European Observational TRD Cohort (EOTC; NCT03373253; clinicaltrials.gov) study. SUSTAIN-2 was a long-term open-label study of esketamine NS, while the EOTC was conducted at a time when esketamine NS was not available as RWT. Threshold and sensitivity analyses were conducted to assess how robust the primary analyses were.

Results

Patients receiving esketamine NS had a higher probability of 6-month response (49.7% [95% confidence interval (CI) 45.6-53.9]) and remission (33.6% [95% CI 29.7-37.6]) vs. patients receiving RWT (26.4% [95% CI 21.5-31.4] and 18.2% [95% CI 13.9-22.5], respectively), according to rescaled average treatment effect among treated estimates. Resulting adjusted odds ratios (OR) and relative risk (RR) favoured esketamine NS over RWT for 6-month response (OR 2.756 [95% CI 2.034-3.733], p < 0.0001; RR 1.882 [95% CI 1.534-2.310], p < 0.0001) and remission (OR 2.276 [95% CI 1.621-3.196], p < 0.0001; RR 1.847 [95% CI 1.418-2.406], p < 0.0001). Threshold analyses suggested that differences between the two studies were robust, and results were consistent across extensive sensitivity analyses.

Conclusion

ICEBERG supports that, at 6 months, esketamine NS has a substantial and significant benefit over RWT for patients with TRD. While results may be affected by unobserved confounding factors, threshold analyses suggested these were unlikely to impact the study conclusions.To view an animated summary of this publication, please click on the Supplementary video.

Treatment outcomes in major depressive disorder in patients with comorbid alcohol use disorder: A STAR*D analysis

INTRODUCTION

Guidance on Major Depressive Disorder (MDD) treatment in those with comorbid Alcohol Use Disorder (AUD) is limited. We performed a secondary analysis on the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study, examining the association between comorbid AUD and depression outcomes.

METHODS

STAR*D was a real-world effectiveness trial starting with citalopram in level 1. Non-responding participants progressed through 3 other sequential treatment levels with different switch or augmentation options. Antidepressant outcomes were compared between MDD (n = 2826) and comorbid MDD and AUD (n = 864). Logistic regressions were performed to evaluate remission and response predictors in the total STAR*D sample and the AUD-comorbidity interaction.

RESULTS

Chi-squared tests showed no significant difference in response or remission rates from depression between groups across treatment levels. Higher Hamilton Rating Scale for Depression (HRSD) score was associated with overall lower odds of remission in treatment level 1 (OR = 0.93, p < 0.001) and 2 (OR = 0.95, p < 0.001), with no significant interaction with comorbid AUD. Higher baseline suicidality had overall lower odds of remission in level 1 (OR = 0.82, p < 0.001) and 2 (OR = 0.1, p < 0.001), but with comorbid AUD compared to no AUD, suicidality increased odds of level 1 remission (OR = 1.30, p = 0.012). In comorbid AUD in level 2, venlafaxine was associated with lower odds of remission (OR = 0.13, p = 0.013) and response (OR = 0.12, p = 0.006); bupropion with lower odds of response (OR = 0.22, p = 0.024).

LIMITATIONS

Open label study design and lack of alcohol use data.

CONCLUSIONS

Comorbid AUD may interact with predictors of antidepressant response in MDD and using venlafaxine or bupropion may be less effective. Addressing this comorbidity requires unique assessment and treatment approaches.

Genetic contributions to brain serotonin transporter levels in healthy adults

The serotonin transporter (5-HTT) critically shapes serotonin neurotransmission by regulating extracellular brain serotonin levels; it remains unclear to what extent 5-HTT levels in the human brain are genetically determined. Here we applied [11C]DASB positron emission tomography to image brain 5-HTT levels and evaluated associations with five common serotonin-related genetic variants that might indirectly regulate 5-HTT levels (BDNF rs6265, SLC6A4 5-HTTLPR, HTR1A rs6295, HTR2A rs7333412, and MAOA rs1137070) in 140 healthy volunteers. In addition, we explored whether these variants could predict in vivo 5-HTT levels using a five-fold cross-validation random forest framework. MAOA rs1137070 T-carriers showed significantly higher brain 5-HTT levels compared to C-homozygotes (2-11% across caudate, putamen, midbrain, thalamus, hippocampus, amygdala and neocortex). We did not observe significant associations for the HTR1A rs6295 and HTR2A rs7333412 genotypes. Our previously observed lower subcortical 5-HTT availability for rs6265 met-carriers remained in the presence of these additional variants. Despite this significant association, our prediction models showed that genotype moderately improved prediction of 5-HTT in caudate, but effects were not statistically significant after adjustment for multiple comparisons. Our observations provide additional evidence that serotonin-related genetic variants modulate adult human brain serotonin neurotransmission.

Development of a model to predict antidepressant treatment response for depression among Veterans

BACKGROUND

Only a limited number of patients with major depressive disorder (MDD) respond to a first course of antidepressant medication (ADM). We investigated the feasibility of creating a baseline model to determine which of these would be among patients beginning ADM treatment in the US Veterans Health Administration (VHA).

METHODS

A 2018-2020 national sample of n = 660 VHA patients receiving ADM treatment for MDD completed an extensive baseline self-report assessment near the beginning of treatment and a 3-month self-report follow-up assessment. Using baseline self-report data along with administrative and geospatial data, an ensemble machine learning method was used to develop a model for 3-month treatment response defined by the Quick Inventory of Depression Symptomatology Self-Report and a modified Sheehan Disability Scale. The model was developed in a 70% training sample and tested in the remaining 30% test sample.

RESULTS

In total, 35.7% of patients responded to treatment. The prediction model had an area under the ROC curve (s.e.) of 0.66 (0.04) in the test sample. A strong gradient in probability (s.e.) of treatment response was found across three subsamples of the test sample using training sample thresholds for high [45.6% (5.5)], intermediate [34.5% (7.6)], and low [11.1% (4.9)] probabilities of response. Baseline symptom severity, comorbidity, treatment characteristics (expectations, history, and aspects of current treatment), and protective/resilience factors were the most important predictors.

CONCLUSIONS

Although these results are promising, parallel models to predict response to alternative treatments based on data collected before initiating treatment would be needed for such models to help guide treatment selection.

Outcome of transcranial magnetic intermittent theta-burst stimulation in the treatment of depression - A Swedish register-based study

BACKGROUND

Repetitive transcranial magnetic stimulation (rTMS) is an established treatment of depression. The more recently introduced intermittent Theta-burst stimulation (iTBS) has shown significant superiority over sham-stimulation and equal effect sizes to a 10 Hz protocol in one clinical trial. The aim of the current study was to investigate the effectiveness and tolerability of iTBS in a naturalistic, clinical setting. Further, we explored demographical and clinical predictors of response.

METHODS

Data was collected from seventeen rTMS-sites in Sweden between January 2018 and May 2021, through the Swedish National Quality register for repetitive Transcranial Magnetic Stimulation (Q-rTMS). We included 542 iTBS-treated patients with unipolar or bipolar depression. Outcome was assessed with Clinical Global Impression Severity and Improvement scores in an intention to treat analysis.

RESULTS

The response rate was 42.1 % and 16.1 % reached remission. The response rate was significantly larger in the oldest age group compared to the youngest (odds ratio 3.46, 95 % confidence interval 1.65-7.22). Less severe level of depression (Montgomery-Åsberg depression rating scale self-assessment < 36) at baseline predicted response and remission. Only <1 % were much or very much worse after treatment. Drop-out rate was 10.9 %. No serious adverse events were reported.

LIMITATIONS

Retrospective analysis of register data. No comparison group.

CONCLUSIONS

In a clinical setting, iTBS was shown to be safe and tolerable and the response rate was similar to that reported from clinical trials. Older age-group and less severe illness predicted response.

Effectiveness of Genotype-Specific Tricyclic Antidepressant Dosing in Patients With Major Depressive Disorder: A Randomized Clinical Trial

Importance

Evidence of the clinical benefit of pharmacogenetics-informed treatment (PIT) with antidepressants is still limited. Especially for tricyclic antidepressants (TCAs), pharmacogenetics may be of interest because therapeutic plasma concentrations are well defined, identification of optimal dosing can be time consuming, and treatment is frequently accompanied by adverse effects.

Objective

To determine whether PIT results in faster attainment of therapeutic TCA plasma concentrations compared with usual treatment in patients with unipolar major depressive disorder (MDD).

Design, Setting, and Participants

This randomized clinical trial compared PIT with usual treatment among 111 patients at 4 centers in the Netherlands. Patients were treated with the TCAs nortriptyline, clomipramine, or imipramine, with clinical follow-up of 7 weeks. Patients were enrolled from June 1, 2018, to January 1, 2022. At inclusion, patients had unipolar nonpsychotic MDD (with a score of ≥19 on the 17-item Hamilton Rating Scale for Depression [HAMD-17]), were aged 18 to 65 years, and were eligible for TCA treatment. Main exclusion criteria were a bipolar or psychotic disorder, substance use disorder, pregnancy, interacting comedications, and concurrent use of psychotropic medications.

Intervention

In the PIT group, the initial TCA dosage was based on CYP2D6 and CYP2C19 genotypes. The control group received usual treatment, which comprised the standard initial TCA dosage.

Main Outcomes and Measures

The primary outcome was days until attainment of a therapeutic TCA plasma concentration. Secondary outcomes were severity of depressive symptoms (measured by HAMD-17 scores) and frequency and severity of adverse effects (measured by Frequency, Intensity, and Burden of Side Effects Rating scores).

Results

Of 125 patients randomized, 111 (mean [SD] age, 41.7 [13.3] years; 69 [62.2%] female) were included in the analysis; of those, 56 were in the PIT group and 55 were in the control group. The PIT group reached therapeutic concentrations faster than the control group (mean [SD], 17.3 [11.2] vs 22.0 [10.2] days; Kaplan-Meier χ21 = 4.30; P = .04). No significant difference in reduction of depressive symptoms was observed. Linear mixed-model analyses showed that the interaction between group and time differed for the frequency (F6,125 = 4.03; P = .001), severity (F6,114 = 3.10; P = .008), and burden (F6,112 = 2.56; P = .02) of adverse effects, suggesting that adverse effects decreased relatively more for those receiving PIT.

Conclusions and Relevance

In this randomized clinical trial, PIT resulted in faster attainment of therapeutic TCA concentrations, with potentially fewer and less severe adverse effects. No effect on depressive symptoms was observed. These findings indicate that pharmacogenetics-informed dosing of TCAs can be safely applied and may be useful in personalizing treatment for patients with MDD.

Trial Registration

ClinicalTrials.gov Identifier: NCT03548675.

Reliability and validity of the difficult to treat depression questionnaire (DTDQ)

It has recently been recommended that treatment resistant depression be reconceptualized and renamed as difficult to treat depression (DTD). A consensus statement by an expert panel identified multiple variables associated with DTD and emphasized the importance of conducting a comprehensive evaluation of patients to identify predictors of inadequate treatment response. For practical reasons, it would be desirable to develop a self-report scale that can be incorporated into clinical practice that identifies patient, clinical, and treatment risk factors for DTD. Nine hundred twenty depressed patients completed the Difficult to Treat Depression Questionnaire (DTDQ). A subset of patients completed the scale a second time and completed the Remission from Depression Questionnaire at admission and discharge from a partial hospital program. The DTDQ demonstrated excellent internal consistency and test-retest reliability. Both the total DTDQ and the number of prior failed medication trials, the metric primarily relied upon to classify treatment resistant depression, predicted outcome. However, the DTDQ continued to be significantly associated with outcome after controlling for the number of failed trials, whereas the number of failed trials did not predict outcome after controlling for DTDQ scores. The DTDQ is a reliable and valid measure of the recently discussed concept of DTD.

Multimodal predictions of treatment outcome in major depression: A comparison of data-driven predictors with importance ratings by clinicians

BACKGROUND

Reliable prediction models of treatment outcome in Major Depressive Disorder (MDD) are currently lacking in clinical practice. Data-driven outcome definitions, combining data from multiple modalities and incorporating clinician expertise might improve predictions.

METHODS

We used unsupervised machine learning to identify treatment outcome classes in 1060 MDD inpatients. Subsequently, classification models were created on clinical and biological baseline information to predict treatment outcome classes and compared to the performance of two widely used classical outcome definitions. We also related the findings to results from an online survey that assessed which information clinicians use for outcome prognosis.

RESULTS

Three and four outcome classes were identified by unsupervised learning. However, data-driven outcome classes did not result in more accurate prediction models. The best prediction model was targeting treatment response in its standard definition and reached accuracies of 63.9 % in the test sample, and 59.5 % and 56.9 % in the validation samples. Top predictors included sociodemographic and clinical characteristics, while biological parameters did not improve prediction accuracies. Treatment history, personality factors, prior course of the disorder, and patient attitude towards treatment were ranked as most important indicators by clinicians.

LIMITATIONS

Missing data limited the power to identify biological predictors of treatment outcome from certain modalities.

CONCLUSIONS

So far, the inclusion of available biological measures in addition to psychometric and clinical information did not improve predictive value of the models, which was overall low. Optimized biomarkers, stratified predictions and the inclusion of clinical expertise may improve future prediction models.

Deep phenotyping towards precision psychiatry of first-episode depression - the Brain Drugs-Depression cohort

BACKGROUND

Major Depressive Disorder (MDD) is a heterogenous brain disorder, with potentially multiple psychosocial and biological disease mechanisms. This is also a plausible explanation for why patients do not respond equally well to treatment with first- or second-line antidepressants, i.e., one-third to one-half of patients do not remit in response to first- or second-line treatment. To map MDD heterogeneity and markers of treatment response to enable a precision medicine approach, we will acquire several possible predictive markers across several domains, e.g., psychosocial, biochemical, and neuroimaging.

METHODS

All patients are examined before receiving a standardised treatment package for adults aged 18-65 with first-episode depression in six public outpatient clinics in the Capital Region of Denmark. From this population, we will recruit a cohort of 800 patients for whom we will acquire clinical, cognitive, psychometric, and biological data. A subgroup (subcohort I, n = 600) will additionally provide neuroimaging data, i.e., Magnetic Resonance Imaging, and Electroencephalogram, and a subgroup of patients from subcohort I unmedicated at inclusion (subcohort II, n = 60) will also undergo a brain Positron Emission Tomography with the [¹¹C]-UCB-J tracer binding to the presynaptic glycoprotein-SV2A. Subcohort allocation is based on eligibility and willingness to participate. The treatment package typically lasts six months. Depression severity is assessed with the Quick Inventory of Depressive Symptomatology (QIDS) at baseline, and 6, 12 and 18 months after treatment initiation. The primary outcome is remission (QIDS ≤ 5) and clinical improvement (≥ 50% reduction in QIDS) after 6 months. Secondary endpoints include remission at 12 and 18 months and %-change in QIDS, 10-item Symptom Checklist, 5-item WHO Well-Being Index, and modified Disability Scale from baseline through follow-up. We also assess psychotherapy and medication side-effects. We will use machine learning to determine a combination of characteristics that best predict treatment outcomes and statistical models to investigate the association between individual measures and clinical outcomes. We will assess associations between patient characteristics, treatment choices, and clinical outcomes using path analysis, enabling us to estimate the effect of treatment choices and timing on the clinical outcome.

DISCUSSION

The BrainDrugs-Depression study is a real-world deep-phenotyping clinical cohort study of first-episode MDD patients.

TRIAL REGISTRATION

Registered at clinicaltrials.gov November 15th, 2022 (NCT05616559).

PROTOCOL: Treatment for depressive disorder among adults: An evidence and gap map of systematic reviews

This is the protocol for a Campbell evidence and gap map. The objective of the map is to map available systematic reviews on the effectiveness of treatments for depressive disorders among adults. Specifically, this EGM includes studies on the effectiveness of treatments across a range of outcome domains.

Multi-omics data integration methods and their applications in psychiatric disorders

To study mental illness and health, in the past researchers have often broken down their complexity into individual subsystems (e.g., genomics, transcriptomics, proteomics, clinical data) and explored the components independently. Technological advancements and decreasing costs of high throughput sequencing has led to an unprecedented increase in data generation. Furthermore, over the years it has become increasingly clear that these subsystems do not act in isolation but instead interact with each other to drive mental illness and health. Consequently, individual subsystems are now analysed jointly to promote a holistic understanding of the underlying biological complexity of health and disease. Complementing the increasing data availability, current research is geared towards developing novel methods that can efficiently combine the information rich multi-omics data to discover biologically meaningful biomarkers for diagnosis, treatment, and prognosis. However, clinical translation of the research is still challenging. In this review, we summarise conventional and state-of-the-art statistical and machine learning approaches for discovery of biomarker, diagnosis, as well as outcome and treatment response prediction through integrating multi-omics and clinical data. In addition, we describe the role of biological model systems and in silico multi-omics model designs in clinical translation of psychiatric research from bench to bedside. Finally, we discuss the current challenges and explore the application of multi-omics integration in future psychiatric research. The review provides a structured overview and latest updates in the field of multi-omics in psychiatry.

The Precision in Psychiatry (PIP) study: Testing an internet-based methodology for accelerating research in treatment prediction and personalisation

BACKGROUND

Evidence-based treatments for depression exist but not all patients benefit from them. Efforts to develop predictive models that can assist clinicians in allocating treatments are ongoing, but there are major issues with acquiring the volume and breadth of data needed to train these models. We examined the feasibility, tolerability, patient characteristics, and data quality of a novel protocol for internet-based treatment research in psychiatry that may help advance this field.

METHODS

A fully internet-based protocol was used to gather repeated observational data from patient cohorts receiving internet-based cognitive behavioural therapy (iCBT) (N = 600) or antidepressant medication treatment (N = 110). At baseline, participants provided > 600 data points of self-report data, spanning socio-demographics, lifestyle, physical health, clinical and other psychological variables and completed 4 cognitive tests. They were followed weekly and completed another detailed clinical and cognitive assessment at week 4. In this paper, we describe our study design, the demographic and clinical characteristics of participants, their treatment adherence, study retention and compliance, the quality of the data gathered, and qualitative feedback from patients on study design and implementation.

RESULTS

Participant retention was 92% at week 3 and 84% for the final assessment. The relatively short study duration of 4 weeks was sufficient to reveal early treatment effects; there were significant reductions in 11 transdiagnostic psychiatric symptoms assessed, with the largest improvement seen for depression. Most participants (66%) reported being distracted at some point during the study, 11% failed 1 or more attention checks and 3% consumed an intoxicating substance. Data quality was nonetheless high, with near perfect 4-week test retest reliability for self-reported height (ICC = 0.97).

CONCLUSIONS

An internet-based methodology can be used efficiently to gather large amounts of detailed patient data during iCBT and antidepressant treatment. Recruitment was rapid, retention was relatively high and data quality was good. This paper provides a template methodology for future internet-based treatment studies, showing that such an approach facilitates data collection at a scale required for machine learning and other data-intensive methods that hope to deliver algorithmic tools that can aid clinical decision-making in psychiatry.

Taking modern psychiatry into the metaverse: Integrating augmented, virtual, and mixed reality technologies into psychiatric care

The landscape of psychiatry is ever evolving and has recently begun to be influenced more heavily by new technologies. One novel technology which may have particular application to psychiatry is the metaverse, a three-dimensional digital social platform accessed via augmented, virtual, and mixed reality (AR/VR/MR). The metaverse allows the interaction of users in a virtual world which can be measured and manipulated, posing at once exciting new possibilities and significant potential challenges and risks. While the final form of the nascent metaverse is not yet clear, the immersive simulation and holographic mixed reality-based worlds made possible by the metaverse have the potential to redefine neuropsychiatric care for both patients and their providers. While a number of applications for this technology can be envisioned, this article will focus on leveraging the metaverse in three specific domains: medical education, brain stimulation, and biofeedback. Within medical education, the metaverse could allow for more precise feedback to students performing patient interviews as well as the ability to more easily disseminate highly specialized technical skills, such as those used in advanced neurostimulation paradigms. Examples of potential applications in brain stimulation and biofeedback range from using AR to improve precision targeting of non-invasive neuromodulation modalities to more innovative practices, such as using physiological and behavioral measures derived from interactions in VR environments to directly inform and personalize treatment parameters for patients. Along with promising future applications, we also discuss ethical implications and data security concerns that arise when considering the introduction of the metaverse and related AR/VR technologies to psychiatric research and care.

Predicting treatment outcome in depression: an introduction into current concepts and challenges

Improving response and remission rates in major depressive disorder (MDD) remains an important challenge. Matching patients to the treatment they will most likely respond to should be the ultimate goal. Even though numerous studies have investigated patient-specific indicators of treatment efficacy, no (bio)markers or empirical tests for use in clinical practice have resulted as of now. Therefore, clinical decisions regarding the treatment of MDD still have to be made on the basis of questionnaire- or interview-based assessments and general guidelines without the support of a (laboratory) test. We conducted a narrative review of current approaches to characterize and predict outcome to pharmacological treatments in MDD. We particularly focused on findings from newer computational studies using machine learning and on the resulting implementation into clinical decision support systems. The main issues seem to rest upon the unavailability of robust predictive variables and the lacking application of empirical findings and predictive models in clinical practice. We outline several challenges that need to be tackled on different stages of the translational process, from current concepts and definitions to generalizable prediction models and their successful implementation into digital support systems. By bridging the addressed gaps in translational psychiatric research, advances in data quantity and new technologies may enable the next steps toward precision psychiatry.

Clinical predictors of antipsychotic treatment resistance: Development and internal validation of a prognostic prediction model by the STRATA-G consortium

INTRODUCTION

Our aim was to, firstly, identify characteristics at first-episode of psychosis that are associated with later antipsychotic treatment resistance (TR) and, secondly, to develop a parsimonious prediction model for TR.

METHODS

We combined data from ten prospective, first-episode psychosis cohorts from across Europe and categorised patients as TR or non-treatment resistant (NTR) after a mean follow up of 4.18 years (s.d. = 3.20) for secondary data analysis. We identified a list of potential predictors from clinical and demographic data recorded at first-episode. These potential predictors were entered in two models: a multivariable logistic regression to identify which were independently associated with TR and a penalised logistic regression, which performed variable selection, to produce a parsimonious prediction model. This model was internally validated using a 5-fold, 50-repeat cross-validation optimism-correction.

RESULTS

Our sample consisted of N = 2216 participants of which 385 (17 %) developed TR. Younger age of psychosis onset and fewer years in education were independently associated with increased odds of developing TR. The prediction model selected 7 out of 17 variables that, when combined, could quantify the risk of being TR better than chance. These included age of onset, years in education, gender, BMI, relationship status, alcohol use, and positive symptoms. The optimism-corrected area under the curve was 0.59 (accuracy = 64 %, sensitivity = 48 %, and specificity = 76 %).

IMPLICATIONS

Our findings show that treatment resistance can be predicted, at first-episode of psychosis. Pending a model update and external validation, we demonstrate the potential value of prediction models for TR.

Infraslow closed-loop brain training for anxiety and depression (ISAD): a protocol for a randomized, double-blind, sham-controlled pilot trial in adult females with internalizing disorders

Background

The core intrinsic connectivity networks (core-ICNs), encompassing the default-mode network (DMN), salience network (SN) and central executive network (CEN), have been shown to be dysfunctional in individuals with internalizing disorders (IDs, e.g. major depressive disorder, MDD; generalized anxiety disorder, GAD; social anxiety disorder, SOC). As such, source-localized, closed-loop brain training of electrophysiological signals, also known as standardized low-resolution electromagnetic tomography (sLORETA) neurofeedback (NFB), targeting key cortical nodes within these networks has the potential to reduce symptoms associated with IDs and restore normal core ICN function. We intend to conduct a randomized, double-blind (participant and assessor), sham-controlled, parallel-group (3-arm) trial of sLORETA infraslow (<0.1 Hz) fluctuation neurofeedback (sLORETA ISF-NFB) 3 times per week over 4 weeks in participants (n=60) with IDs. Our primary objectives will be to examine patient-reported outcomes (PROs) and neurophysiological measures to (1) compare the potential effects of sham ISF-NFB to either genuine 1-region ISF-NFB or genuine 2-region ISF-NFB, and (2) assess for potential associations between changes in PRO scores and modifications of electroencephalographic (EEG) activity/connectivity within/between the trained regions of interest (ROIs). As part of an exploratory analysis, we will investigate the effects of additional training sessions and the potential for the potentiation of the effects over time.

Methods

We will randomly assign participants who meet the criteria for MDD, GAD, and/or SOC per the MINI (Mini International Neuropsychiatric Interview for DSM-5) to one of three groups: (1) 12 sessions of posterior cingulate cortex (PCC) ISF-NFB up-training (n=15), (2) 12 sessions of concurrent PCC ISF up-training and dorsal anterior cingulate cortex (dACC) ISF-NFB down-training (n=15), or (3) 6 sessions of yoked-sham training followed by 6 sessions genuine ISF-NFB (n=30). Transdiagnostic PROs (Hospital Anxiety and Depression Scale, HADS; Inventory of Depression and Anxiety Symptoms – Second Version, IDAS-II; Multidimensional Emotional Disorder Inventory, MEDI; Intolerance of Uncertainty Scale – Short Form, IUS-12; Repetitive Thinking Questionnaire, RTQ-10) as well as resting-state neurophysiological measures (full-band EEG and ECG) will be collected from all subjects during two baseline sessions (approximately 1 week apart) then at post 6 sessions, post 12 sessions, and follow-up (1 month later). We will employ Bayesian methods in R and advanced source-localisation software (i.e. exact low-resolution brain electromagnetic tomography; eLORETA) in our analysis.

Discussion

This protocol will outline the rationale and research methodology for a clinical pilot trial of sLORETA ISF-NFB targeting key nodes within the core-ICNs in a female ID population with the primary aims being to assess its potential efficacy via transdiagnostic PROs and relevant neurophysiological measures.

Trial registration

Our study was prospectively registered with the Australia New Zealand Clinical Trials Registry (ANZCTR; Trial ID: ACTRN12619001428156). Registered on October 15, 2019.

A systematic meta-review of patient-level predictors of psychological therapy outcome in major depressive disorder

BACKGROUND

Psychological therapies are effective for treating major depressive disorder, but current clinical guidelines do not provide guidance on the personalization of treatment choice. Established predictors of psychotherapy treatment response could help inform machine learning models aimed at predicting individual patient responses to different therapy options. Here we sought to comprehensively identify known predictors.

METHODS

EMBASE, Medline, PubMed, PsycINFO were searched for systematic reviews with or without meta-analysis published until June 2020 to identify individual patient-level predictors of response to psychological treatments. 3113 abstracts were identified and 300 articles assessed. We qualitatively synthesized our findings by predictor category (sociodemographic; symptom profile; social support; personality features; affective, cognitive, and behavioural; comorbidities; neuroimaging; genetics) and treatment type. We used the AMSTAR 2 to evaluate the quality of included reviews.

RESULTS

Following screening and full-text assessment, 27 systematic reviews including 12 meta-analyses were eligible for inclusion. 74 predictors emerged for various psychological treatments, primarily cognitive behavioural therapy, interpersonal therapy, and mindfulness-based cognitive therapy.

LIMITATIONS

A paucity of studies examining predictors of psychological treatment outcome, as well as methodological heterogeneities and publication biases limit the strength of the identified predictors.

CONCLUSIONS

The synthesized predictors could be used to supplement clinical decision-making in selecting psychological therapies based on individual patient characteristics. These predictors could also be used as a priori input features for machine learning models aimed at predicting a given patient's likelihood of response to different treatment options for depression, and may contribute toward the development of patient-specific treatment recommendations in clinical guidelines.

Computational approaches to treatment response prediction in major depression using brain activity and behavioral data: A systematic review

Major depressive disorder is a heterogeneous diagnostic category with multiple available treatments. With the goal of optimizing treatment selection, researchers are developing computational models that attempt to predict treatment response based on various pretreatment measures. In this paper, we review studies that use brain activity data to predict treatment response. Our aim is to highlight and clarify important methodological differences between various studies that relate to the incorporation of domain knowledge, specifically within two approaches delineated as data-driven and theory-driven. We argue that theory-driven generative modeling, which explicitly models information processing in the brain and thus can capture disease mechanisms, is a promising emerging approach that is only beginning to be utilized in treatment response prediction. The predictors extracted via such models could improve interpretability, which is critical for clinical decision-making. We also identify several methodological limitations across the reviewed studies and provide suggestions for addressing them. Namely, we consider problems with dichotomizing treatment outcomes, the importance of investigating more than one treatment in a given study for differential treatment response predictions, the need for a patient-centered approach for defining treatment outcomes, and finally, the use of internal and external validation methods for improving model generalizability.

Prediction of remission among patients with a major depressive disorder based on the resting-state functional connectivity of emotion regulation networks

The prediction of antidepressant response is critical for psychiatrists to select the initial antidepressant drug for patients with major depressive disorders (MDD). The implicated brain networks supporting emotion regulation (ER) are critical in the pathophysiology of MDD and the prediction of antidepressant response. Therefore, the primary aim of the current study was to identify the neuroimaging biomarkers for the prediction of remission in patients with MDD based on the resting-state functional connectivity (rsFC) of the ER networks. A total of 81 unmedicated adult MDD patients were investigated and they underwent resting-state functional magnetic resonance imagining (fMRI) scans. The patients were treated with escitalopram for 12 weeks. The 17-item Hamilton depression rating scale was used for assessing remission. The 36 seed regions from predefined ER networks were selected and the rsFC matrix was caculated for each participant. The support vector machine algorithm was employed to construct prediction model, which separated the patients with remission from those with non-remission. And leave-one-out cross-validation and the area under the curve (AUC) of the receiver operating characteristic were used for evaluating the performance of the model. The accuracy of the prediction model was 82.08% (sensitivity = 71.43%, specificity = 89.74%, AUC = 0.86). The rsFC between the left medial superior frontal gyrus and the right inferior frontal gyrus as well as the precuneus were the features with the highest discrimination ability in predicting remission from escitalopram among the MDD patients. Results from our study demonstrated that rsFC of the ER brain networks are potential predictors for the response of antidepressant drugs. The trial name: appropriate technology study of MDD diagnosis and treatment based on objective indicators and measurement. URL: http://www.chictr.org.cn/showproj.aspx?proj=21377 . Registration number: ChiCTR-OOC-17012566.

Shared and unique imaging-derived endo-phenotypes of two typical antidepressant-applicative depressive patients

OBJECTIVES

Determining the clinical homogeneous and heterogeneous sets among depressive patients is the key to facilitate individual-level treatment decision.

METHODS

The diffusion tensor imaging (DTI) data of 62 patients with major depressive disorder (MDD) and 39 healthy controls were used to construct a Latent Dirichlet Allocation (LDA) Bayesian model. Another 48 MDD patients were used to verify the robustness. The LDA model was employed to identify both shared and unique imaging-derived factors of two typically antidepressant-targeted depressive patients, selective serotonin reuptake inhibitors (SSRIs) and serotonin norepinephrine reuptake inhibitors (SNRIs). Furthermore, we applied canonical correlation analysis (CCA) between each factor loading and Hamilton depression rating scale (HAMD) sub-score, to explore the potential neurophysiological significance of each factor.

RESULTS

The results revealed the imaging-derived connectional fingerprint of all patients could be situated along three latent factor dimensions; such results were also verified by the out-of-sample dataset. Factor 1, uniquely expressed by SNRI-targeted patients, was associated with retardation (r = 0.4, p = 0.037) and characterized by coupling patterns between default mode network and cognitive control network. Factor 3, uniquely expressed by SSRI-targeted patients, was associated with cognitive impairment (r = 0.36, p = 0.047) and characterized by coupling patterns within cognitive control and attention network, and the connectivity between threat and reward network. Shared factor 2, characterized by coupling patterns within default mode network, was associated with anxiety (r = 0.54, p = 0.005) and sleep disturbance (r = 0.37, p = 0.032).

CONCLUSIONS

Our findings suggested that quantification of both homogeneity and heterogeneity within MDD may have the potential to inform rational design of pharmacological therapies.

KEY POINTS

• The shared and unique manifestations guiding pharmacotherapy of depressive patients are caused by the homogeneity and heterogeneity of underlying structural connections of the brain. • Both shared and unique factor loadings were found in different antidepressant-targeted patients. • Significant correlations between factor loading and HAMD sub-scores were found.

Prospective associations of multimodal serum biomarkers with 12-week and 12-month remission in patients with depressive disorders receiving stepwise psychopharmacotherapy

Prognostic biomarkers for depression treatment outcomes have yet to be elucidated. This study sought to evaluate whether a multi-modal serum biomarker panel was prospectively associated with 12-week and 12-month remission in outpatients with depressive disorders receiving stepwise psychopharmacotherapy. At baseline, 14 serum biomarkers and socio-demographic/clinical characteristics were evaluated in 1094 patients. They received initial antidepressant monotherapy followed, as required by a protocol of successive alternative pharmacological strategies administered in 3-week steps during the acute (3-12 week) phase (N = 1086), and in 3-month steps during the continuation (6-12 month) phase (N = 884). Remission was defined as a Hamilton Depression Rating Scale score of ≤ 7. Remission was achieved in 490 (45.1%) over the 12-week, and in 625 (70.7%) over the 12-month, treatment periods. Combination scores of four serum biomarkers (high-sensitivity C-reactive protein, interleukin-1 beta, interleukin-6, and leptin) were prospectively associated with 12-week remission; and four (high-sensitivity C-reactive protein, tumor necrosis factor-alpha, interleukin-1 beta, and brain-derived neurotrophic factor) were prospectively associated with 12-month remission in a clear gradient manner (P-values < 0.001) and after adjustment for relevant covariates. These associations were evident after the Step 1 treatment monotherapy but weakened with increasing treatment steps, falling below statistical significance after 4 + treatment steps. Application of combined multiple serum biomarkers, particularly on inflammatory markers, could improve predictability of remission at acute and continuation treatment phases for depressive disorders. Patients with unfavourable biomarkers might require alternative treatment regimes for better outcomes.

Creating sparser prediction models of treatment outcome in depression: a proof-of-concept study using simultaneous feature selection and hyperparameter tuning

Background

Predicting treatment outcome in major depressive disorder (MDD) remains an essential challenge for precision psychiatry. Clinical prediction models (CPMs) based on supervised machine learning have been a promising approach for this endeavor. However, only few CPMs have focused on model sparsity even though sparser models might facilitate the translation into clinical practice and lower the expenses of their application.

Methods

In this study, we developed a predictive modeling pipeline that combines hyperparameter tuning and recursive feature elimination in a nested cross-validation framework. We applied this pipeline to a real-world clinical data set on MDD treatment response and to a second simulated data set using three different classification algorithms. Performance was evaluated by permutation testing and comparison to a reference pipeline without nested feature selection.

Results

Across all models, the proposed pipeline led to sparser CPMs compared to the reference pipeline. Except for one comparison, the proposed pipeline resulted in equally or more accurate predictions. For MDD treatment response, balanced accuracy scores ranged between 61 and 71% when models were applied to hold-out validation data.

Conclusions

The resulting models might be particularly interesting for clinical applications as they could reduce expenses for clinical institutions and stress for patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-022-01926-2.

Biomarkers as predictors of treatment response to tricyclic antidepressants in major depressive disorder: A systematic review

Tricyclic antidepressants (TCAs) are frequently prescribed in case of non-response to first-line antidepressants in Major Depressive Disorder (MDD). Treatment of MDD often entails a trial-and-error process of finding a suitable antidepressant and its appropriate dose. Nowadays, a shift is seen towards a more personalized treatment strategy in MDD to increase treatment efficacy. One of these strategies involves the use of biomarkers for the prediction of antidepressant treatment response. We aimed to summarize biomarkers for prediction of TCA specific (i.e. per agent, not for the TCA as a drug class) treatment response in unipolar nonpsychotic MDD. We performed a systematic search in PubMed and MEDLINE. After full-text screening, 36 papers were included. Seven genetic biomarkers were identified for nortriptyline treatment response. For desipramine, we identified two biomarkers; one genetic and one nongenetic. Three nongenetic biomarkers were identified for imipramine. None of these biomarkers were replicated. Quality assessment demonstrated that biomarker studies vary in endpoint definitions and frequently lack power calculations. None of the biomarkers can be confirmed as a predictor for TCA treatment response. Despite the necessity for TCA treatment optimization, biomarker studies reporting drug-specific results for TCAs are limited and adequate replication studies are lacking. Moreover, biomarker studies generally use small sample sizes. To move forward, larger cohorts, pooled data or biomarkers combined with other clinical characteristics should be used to improve predictive power.

Functional MRI in major depressive disorder: A review of findings, limitations, and future prospects

Objective diagnosis and prognosis in major depressive disorder (MDD) remains a challenge due to the absence of biomarkers based on physiological parameters or medical tests. Numerous studies have been conducted to identify functional magnetic resonance imaging‐based biomarkers of depression that either objectively differentiate patients with depression from healthy subjects, predict personalized treatment outcome, or characterize biological subtypes of depression. While there are some findings of consistent functional biomarkers, there is still lack of robust data acquisition and analysis methodology. According to current findings, primarily, the anterior cingulate cortex, prefrontal cortex, and default mode network play a crucial role in MDD. Yet, there are also less consistent results and the involvement of other regions or networks remains ambiguous. We further discuss image acquisition, processing, and analysis limitations that might underlie these inconsistencies. Finally, the current review aims to address and discuss possible remedies and future opportunities that could improve the search for consistent functional imaging biomarkers of depression. Novel acquisition techniques, such as multiband and multiecho imaging, and neural network‐based cleaning approaches can enhance the signal quality in limbic and frontal regions. More comprehensive analyses, such as directed or dynamic functional features or the identification of biological depression subtypes, can improve objective diagnosis or treatment outcome prediction and mitigate the heterogeneity of MDD. Overall, these improvements in functional MRI imaging techniques, processing, and analysis could advance the search for biomarkers and ultimately aid patients with MDD and their treatment course.

Predictors of 4-week antidepressant outcome in patients with first-episode major depressive disorder: An ROC curve analysis

BACKGROUND

Pretreatment characteristics of patients, symptom and function could be associated with antidepressant treatment outcome, but its predictive ability is not adequate. Our study aimed to identify predictors of acute antidepressant efficacy in patients with first-episode Major Depressive Disorder (MDD).

METHODS

187 patients with first-episode MDD were included and assessed clinical symptoms, cognitive function and global functioning using the 17-item Hamilton Depression Inventory (HAMD-17), MATRICS Consensus Cognitive Battery (MCCB) and Global Assessment of Functioning (GAF). Participants received treatment with a SSRI (escitalopram or venlafaxine) for 4 weeks. Logistic regression was used to analyze the association between patients' characteristics, symptom profiles, cognitive performance, and global functioning and the antidepressant outcome at the end of 4 weeks, and ROC curve analysis was performed for predictive accuracy with area under the receiver operating curve (AUC).

RESULTS

Antidepressant improvement, response and remission rate at week 4 was 87.7%, 64.7% and 42.8%, respectively. The combination of pretreatment clinical profiles, speed of processing and global functioning showed moderate discrimination of acute improvement, response and remission with AUCs of 0.863, 0.812 and 0.734, respectively.

LIMITATIONS

The major limitation of the present study is the study did not combine pharmacogenomics from the perspective of antidepressant drug metabolism.

CONCLUSION

Aside from the baseline clinical symptoms, cognitive function and global functioning could be predictors of acute treatment outcome in first episode MDD using escitalopram or venlafaxine. This relatively simple application based on clinical symptoms and function seems to be cost-effective method to identify individuals who are more likely to respond to antidepressant treatment.

Identifying the Common Genetic Basis of Antidepressant Response

Background

Antidepressants are a first-line treatment for depression. However, only a third of individuals experience remission after the first treatment. Common genetic variation, in part, likely regulates antidepressant response, yet the success of previous genome-wide association studies has been limited by sample size. This study performs the largest genetic analysis of prospectively assessed antidepressant response in major depressive disorder to gain insight into the underlying biology and enable out-of-sample prediction.

Methods

Genome-wide analysis of remission (n remit = 1852, n nonremit = 3299) and percentage improvement (n = 5218) was performed. Single nucleotide polymorphism-based heritability was estimated using genome-wide complex trait analysis. Genetic covariance with eight mental health phenotypes was estimated using polygenic scores/AVENGEME. Out-of-sample prediction of antidepressant response polygenic scores was assessed. Gene-level association analysis was performed using MAGMA and transcriptome-wide association study. Tissue, pathway, and drug binding enrichment were estimated using MAGMA.

Results

Neither genome-wide association study identified genome-wide significant associations. Single nucleotide polymorphism-based heritability was significantly different from zero for remission (h 2 = 0.132, SE = 0.056) but not for percentage improvement (h 2 = -0.018, SE = 0.032). Better antidepressant response was negatively associated with genetic risk for schizophrenia and positively associated with genetic propensity for educational attainment. Leave-one-out validation of antidepressant response polygenic scores demonstrated significant evidence of out-of-sample prediction, though results varied in external cohorts. Gene-based analyses identified ETV4 and DHX8 as significantly associated with antidepressant response.

Conclusions

This study demonstrates that antidepressant response is influenced by common genetic variation, has a genetic overlap schizophrenia and educational attainment, and provides a useful resource for future research. Larger sample sizes are required to attain the potential of genetics for understanding and predicting antidepressant response.

Exploring the potential of thyroid hormones to predict clinical improvements in depressive patients: A machine learning analysis of the real-world based study

BACKGROUND

Although undergoing antidepressant treatments, many patients continue to struggle with chronic depression episode. Seeking the potential biomarkers and establishing a predictive model of clinical improvements is vital to optimize personalized management of depression. Mounting evidence showed thyroid hormones changes are central to leading paradigms of depression.

METHODS

Here, we conducted a real-world based retrospective study using clinical and biochemical data of 2086 depressive inpatients during period of 2014-2020. We first performed regression analyses to evaluate the contributing effect of free triiodothyronine (FT3), free thyroxine (FT4) and thyroid stimulating hormone (TSH) in predicting the clinical outcomes of depression. Then we established 7 predictive models using different combination of such hormones by supervised learning methods and tested the actual prediction efficacy on clinical outcomes, in order to select the one with the best predictive power.

RESULTS

The results showed that lower values of FT3 and FT4 can both predict a poor clinical outcome in depression. Further, a model with the best performance was selected (sensitivity=0.91, specificity=0.79, and ROC-AUC=0.86), including the values of FT3 and FT4, and the scores of Hamilton Depression Scale (HAMD) and Hamilton Anxiety Scale (HAMA) as features.

LIMITATIONS

The predictive model requires further external validation, and multi-center researches to confirm its clinical applicability.

CONCLUSIONS

Our findings present a crucial role of thyroid measurements in predicting clinical outcomes of depression. Assessment of thyroid hormone should be extended to routine practice settings to determine which patients should be most in need of earlier or intensive interventions for preventing continued dysfunction.

Serum amyloid P component level is associated with clinical response to escitalopram treatment in patients with major depressive disorder

Serum amyloid P component (SAP) is a universal constituent of human amyloid deposits, which has been implicated in Alzheimer's disease and major depressive disorder (MDD). However, the relationship between SAP level and depression severity remains obscure. The aims of this study were to investigate how SAP is involved in depression and to explore the association between SAP level and antidepressant treatment response. Patients with MDD (n = 85) who received escitalopram monotherapy for 8-12 weeks were selected from a multicenter open-label randomized clinical trial. The same number of healthy controls was recruited. Depression severity was measured according to the Hamilton Depression Rating Scale (HAMD-17) at baseline and weeks 4, 8, and 12. The plasma levels of SAP were measured at baseline, week 2 and week 12. As a result, baseline levels of SAP were significantly higher in depressed patients than in control subjects (p < 0.001). SAP levels at baseline were negatively associated with depression severity after escitalopram treatment (p < 0.05), and the changes in SAP levels from baseline to week 12 were highly correlated with the severity of depressive symptoms based on the HAMD-17 score (p < 0.05). Interestingly, treatment with escitalopram significantly decreased the plasma levels of SAP in females, but not in males. Altogether, our results suggest that SAP not only involved in the pathobiology of depression but also mediates the action of antidepressant medications.

Evaluating the efficacy and moderators of algorithm-guided antidepressant treatments of major depressive disorder

BACKGROUND

In spite of numerous options, the most efficacious treatment for major depressive disorder (MDD) remains elusive. Algorithm-guided treatments (AGTs) are proposed to address inadequate remission and optimize treatment delivery. This study aimed to evaluate the clinical benefit of AGTs for MDD, and to explore specific moderators of treatment outcomes for individual patients.

METHODS

The study recruited 987 patients with MDD across eight hospitals who were randomly assigned to AGT with escitalopram (AGT-E), AGT with mirtazapine (AGT-M), or treatment-as-usual (TAU). The outcomes were symptom remission, response rate, early improvement rate, subsymptom clusters improvement over time, the mean time to first remission, relapse rate at 6-months posttreatment follow-up, quality of life (QOL), and adverse events.

RESUTLS

No significant differences were observed across groups in outcome, except that TAU showed significantly poorer QOL, higher relapse rates at 6-months posttreatment follow-up, and marginally significantly worse maximal burden of adverse events than the AGT groups. After 6 weeks of treatment initiation, remission rate did not significantly increase with extended treatment. AGT-M outperformed the TAU and AGT-E in treating sleep symptoms. AGT-E was less effective than AGT-M and TAU in patients with severe depression and somatic symptoms (DSSS). The superiority of TAU over AGTs was observed in recurrent MDD patients.

CONCLUSION

Although the superiority of AGTs over TAU was limited by failure of alternative subsequent treatment, AGTs outperformed in QOL and relapse rate. Types of disease episode and DSSS were regarded as specific moderators in treatment of depression. These findings might contribute to future research on targeted antidepressant treatment.

The Impact of Hormonal Contraceptive Use on Serotonergic Neurotransmission and Antidepressant Treatment Response: Results From the NeuroPharm 1 Study

BACKGROUND

Hormonal contraceptive (HC) use has been associated with an increased risk of developing a depressive episode. This might be related to HC's effect on the serotonergic brain system as suggested by recent cross-sectional data from our group, which show that healthy oral contraceptive (OC) users relative to non-users have lower cerebral serotonin 4 receptor (5-HT4R) levels. Here, we determine if cerebral 5-HT4R binding differs between HC non-users, OC users, and hormonal intrauterine device (HIUD) users among women with an untreated depressive episode. Also, we test if antidepressant drug treatment response and its association with pre-treatment 5-HT4R binding depends on HC status.

METHODS

[¹¹C]-SB207145 Positron Emission Tomography imaging data from the NeuroPharm-NP1 Study (NCT02869035) were available from 59 depressed premenopausal women, of which 26 used OCs and 10 used HIUDs. The participants were treated with escitalopram. Treatment response was measured as the relative change in the Hamilton Depression Rating Scale 6 items (rΔHAMD₆) from baseline to week eight. Latent variable models were used to evaluate the association between global 5-HT4R binding and OC and HIUD use as well as rΔHAMD₆.

RESULTS

We found no evidence of a difference in global 5-HT4R binding between depressed HC users and non-users (p≥0.51). A significant crossover interaction (p=0.02) was observed between non-users and OC users in the association between baseline global 5-HT4R binding and week eight rΔHAMD₆; OC users had 3-4% lower binding compared to non-users for every 10% percent less improvement in HAMD₆. Within the groups, we observed a trend towards a positive association in non-users (p_adj=0.10) and a negative association in OC users (p_adj=0.07). We found no strong evidence of a difference in treatment response between the groups (p=0.13).

CONCLUSIONS

We found no difference in 5-HT4R binding between HC users vs. non-users in depressed women, however, it seemed that 5-HT4R settings differed qualitatively in their relation to antidepressant drug treatment response between OC users and non-users. From this we speculate that depressed OC users constitutes a special serotonin subtype of depression, which might have implications for antidepressant drug treatment response.

Attenuated alpha-gamma coupling in emotional dual pathways with right-Amygdala predicting ineffective antidepressant response

Aims

The diversity of treatment outcomes for major depressive disorder (MDD) remains uncertain in neuropathology. The current study aimed at exploring electrophysiological biomarkers associated with treatment response.

Methods

The present study recruited 130 subjects including 100 MDD patients and 30 healthy controls. All subjects participated in a sad expression recognition task while their magnetoencephalography data were recorded. Patients who had a reduction of at least 50% in disorder severity at endpoint (>2 weeks) were considered as responders. Within‐frequency power and phase‐amplitude coupling were measured for the brain regions involved in the emotional visual information processing pathways.

Results

The significant alpha–gamma decoupling from the right thalamus to the right amygdala in unconscious processing and from right orbital frontal cortices to the right amygdala in conscious processing was found in non‐responders relative to responders and healthy controls. These kinds of dysregulation could also predict the potential treatment response.

Conclusion

The attenuated alpha–gamma coupling in dual pathways indicated increased sensitivity to the negative emotional information and reduced moderated effect of the amygdala, which might cause insensitivity to antidepressant treatment and could be regarded as potential neural mechanisms for treatment response prediction.

Validation of Machine Learning-Based Individualized Treatment for Depressive Disorder Using Target Trial Emulation

This study aims to develop and validate the use of machine learning-based prediction models to select individualized pharmacological treatment for patients with depressive disorder. This study used data from Taiwan’s National Health Insurance Research Database. Patients with incident depressive disorders were included in this study. The study outcome was treatment failure, which was defined as psychiatric hospitalization, self-harm hospitalization, emergency visits, or treatment change. Prediction models based on the Super Learner ensemble were trained separately for the initial and the next-step treatments if the previous treatments failed. An individualized treatment strategy was developed for selecting the drug with the lowest probability of treatment failure for each patient as the model-selected regimen. We emulated clinical trials to estimate the effectiveness of individualized treatments. The area under the curve of the prediction model using Super Learner was 0.627 and 0.751 for the initial treatment and the next-step treatment, respectively. Model-selected regimens were associated with reduced treatment failure rates, with a 0.84-fold (95% confidence interval (CI) 0.82–0.86) decrease for the initial treatment and a 0.82-fold (95% CI 0.80–0.83) decrease for the next-step. In emulation of clinical trials, the model-selected regimen was associated with a reduced treatment failure rate.