Effect of add-on transcranial alternating current stimulation (tACS) in major depressive disorder: A randomized controlled trial.

BACKGROUND: The effect of transcranial alternating current stimulation (tACS) on major depressive disorder (MDD) was not confirmed. OBJECTIVE: To evaluate the feasibility, safety, and efficacy of tACS as an add-on treatment for the symptoms of depression and to understand how tACS affects brain activity. METHODS: The 4-week, double-blind, randomized, sham-controlled trial was performed from January 29, 2023 to December 22, 2023. Sixty-six participants were recruited and randomly assigned to receive 20 40-min sessions of either active (77.5Hz, 15 mA) or sham stimulation, with one electrode on the forehead and two on the mastoid, each day (n = 33 for each group) for four weeks (till Week 4). The participants were followed for 4 more weeks (till Week 8) without stimulation for efficacy/safety assessment. During the 4-week trial, all participants were required to take 10-20 mg of escitalopram daily. The primary efficacy endpoint was the change in HAMD-17 scores from baseline to Week 4 (with 20 treatment sessions completed). Resting-state electroencephalography (EEG) was collected with a 64-channel EEG system (Brain Products, Germany) at baseline and the Week 4 follow-up. The chi-square test, Fisher's exact test, independent-sample t-test, or Wilcoxon rank-sum test were used, as appropriate, to compare the differences in variables between groups. The effect of the intervention on the HAMD-17 score was also evaluated with linear mixed modeling (LMM) as sensitivity analysis. The correlation between the mean reduction in EEG and the mean reduction in the HAMD-17 total score was evaluated using Spearman correlation analysis. RESULTS: A total of 66 patients (mean [SD] age, 28.4 [8.18] years; 52 [78.8 %] female) were randomized, and 57 patients completed the study. Significant differences were found in the reductions in the HAMD-17 scores at Week 4 (t = 3.44, P = 0.001). Response rates at Week 4 were significantly higher in the active tACS group than in the sham tACS group (22 out of 33 patients [66.7 %] versus 11 out of 33 [33.3 %], P = 0.007). In the active tACS group, a correlation between the mean change in alpha power and HAMD-17 scores at Week 4 was found (r = 2.38, P = 0.024), and the mean change in alpha power was significantly bigger for responders (Z = 2.46, P = 0.014). No serious adverse events were observed in this trial. CONCLUSION: The additional antidepressant effect of tACS is significant, and the combination of tACS with antidepressants is a feasible and effective approach for the treatment of MDD. The antidepressant mechanism of tACS may be the reduction in alpha power in the left frontal lobe. Future research directions may include exploring more appropriate treatment parameters of tACS.

Pharmacological treatment of depression in patients with brain tumors.

Patients with brain tumors suffer from intense psychosocial distress. Although the prevalence of depressive symptoms in patients with brain tumors is high, the pharmacological antidepressant treatment of those patients is not well defined and results from clinical trials are largely missing. In this review, we describe the current standard of evidence and clinical guidelines for the pharmacological treatment of depression in brain tumor patients. We present specific side effects and interactions that should guide treatment decisions. Furthermore, we provide evidence for the diagnosis, screening and risk factors for depression in brain tumor patients and we elaborate on potential antineoplastic effects of antidepressant drugs and ongoing clinical trials. Antidepressant drugs should not be withheld from patients with brain tumors. Future clinical trials should explore the effectiveness and side effects of antidepressants in this specific patient population.

Update on the assessment of resistance to antidepressant treatment: Rationale for the Antidepressant Treatment History Form: Short Form-2 (ATHF-SF2).

All definitions of treatment-resistant depression (TRD) require that patients have experienced insufficient benefit from one or more adequate antidepressant trials. Thus, identifying "failed, adequate trials" is key to the assessment of TRD. The Antidepressant Treatment History Form (ATHF) was one of the first and most widely used instruments that provided objective criteria in making these assessments. The original ATHF was updated in 2018 to the ATHF-SF, changing to a checklist format for scoring, and including specific pharmacotherapy, brain stimulation, and psychotherapy interventions as potentially adequate antidepressant treatments. The ATHF-SF2, presented here, is based on the consensus of the ATHF workgroup about the novel interventions introduced since the last revision and which should/should not be considered effective treatments for major depressive episodes. This document describes the rationale for these choices and, for each intervention, the minimal criteria for determining the adequacy of treatment administration. The Supplementary Material that accompanies this article provide the Scoring Checklist, Data Collection Forms (current episode and composite of previous episodes), and Instruction Manual for the ATHF-SF2.

Rumination in patients with major depressive disorder before and after antidepressant treatment.

BACKGROUND: Rumination is a maladaptive response to distress characteristic of Major Depressive Disorder (MDD). It is unclear to what degree rumination is associated with depression severity prior to treatment and how it responds to antidepressant treatment. Therefore, we evaluated the association between rumination and depression severity in 92 untreated patients with MDD and explored the changes in rumination after initiation of antidepressant medication. METHOD: We measured rumination using the Rumination Response Scale (RRS) and depression severity with the Hamilton Depression Rating Scale (HDRS17 or HDRS6) before and after initiation of 12 weeks of antidepressant treatment. The association between RRS and pre-treatment HDRS17 was evaluated using a linear regression model. RRS at week 4, 8, and 12 across treatment response categories (remission vs. non-response) were evaluated using a mixed effect model. RESULTS: RRS was positively associated with depression severity prior to treatment at a trend level (p = 0.06). After initiation of treatment RRS decreased significantly (p < 0.0001) and remitters exhibited lower rumination compared to non-responders at week 4 (p = 0.03), 8 (p = 0.01), and 12 (p = 0.007). LIMITATIONS: The study had no placebo group. CONCLUSIONS: Although pre-treatment rumination did not significantly associate with depressive symptoms, rumination was closely connected to change in depressive symptoms. Tormented patients could be reassured that rumination symptoms may be alleviated over the course of antidepressant treatment.

Peripheral serotonin levels as a predictor of antidepressant treatment response: A systematic review.

There are currently no reliable biomarkers to predict clinical response to pharmacological treatments of depressive disorders. Peripheral blood 5-hydroxytryptamine (5-HT; serotonin) has been suggested as a biomarker of antidepressant treatment response, but there has not been an attempt to systematically summarize and evaluate the scientific evidence of this hypothesis. In this systematic review we searched MEDLINE, Embase, PsycINFO, and the Cochrane Central Register of Controlled Trials. Twenty-six relevant studies investigating peripheral 5-HT as an antidepressant biomarker were identified. In all, we did not find robust support for an association between baseline 5-HT and treatment response. Several larger studies with lower risk of bias, however, showed that higher baseline 5-HT was associated with a greater antidepressant response to SSRIs, prompting future studies to investigate this hypothesis. Our results also confirm previous reports that SSRI treatment is associated with a decrease in peripheral 5-HT levels; however, we were not able to confirm that larger decreases of 5-HT are associated with better treatment outcome as results were inconclusive.

Characterizing the treatment patterns, medication burden, and patient demographics of older adults with major depressive disorder treated with antidepressants with or without selected comorbidities.

OBJECTIVE: Evaluate clinical characteristics, comorbidity burden, major depressive disorder (MDD)-related healthcare resource utilization (HCRU), medication burden, and antidepressant treatment (ADT) patterns among older adults with MDD with and without selected comorbidities. METHODS: Using Komodo's Healthcare Map claims data (1/1/2016-9/30/2022), patients with MDD (≥65 years) treated with ADTs were assessed 24 months preceding (baseline) and 12 months following (follow-up) first observed ADT prescription fill (index). Patients were separated into cohorts of those with ≥1 of 5 selected comorbidities and those without. Clinical characteristics, comorbidities, and MDD-related HCRU were assessed during baseline; treatment patterns were assessed during follow-up. Baseline and follow-up all-cause and comorbidity-specific medication burdens (mean prescription claims/month) were determined. RESULTS: Among the total cohort (N = 417,643), 97.1% had ≥1 of 5 selected comorbidities: hypertension (80.3%), hyperlipidemia (75.4%), diabetes (54.2%), anxiety disorder (39.0%), and chronic obstructive pulmonary disorder (19.5%). Baseline and follow-up all-cause medication burdens per month were 3.8 and 4.5 for patients with selected comorbidities and 1.7 and 2.3 for those without. During baseline, most patients (96.0% with selected comorbidities, 96.2% without) had ≥1 outpatient visit, and a numerically higher percentage of those with vs. without selected comorbidities had MDD-related emergency room (13.9% vs. 6.0%) and inpatient (13.5% vs. 4.1%) visits. The majority of both cohorts (61.0% with selected comorbidities, 59.5% without) underwent treatment pattern changes. CONCLUSION: This study highlights the medication burden and ADT patterns in older adults with MDD, assessing these outcomes among patients with and without comorbidities. Numerically higher medication burdens among those with selected comorbidities suggests future studies could investigate the impact of comorbidities on MDD-related care.

Perspectives in treatment-resistant depression: esketamine and electroconvulsive therapy.

Modern electroconvulsive therapy (ECT) and the approval of nasal esketamine for clinical use have significantly improved the approach to treatment-resistant depression (TRD), which is defined as non-response to at least two different courses of antidepressants with verified adherence to treatment, adequate dosage, and duration of treatment. The goal of this literature review is to present the newest evidence regarding efficacy and safety. Furthermore, we aim to provide an overview of future perspectives in this field of research, for example, regarding structural and molecular effects. Both treatment methods will be critically evaluated for their individual advantages, disadvantages, and response rates. Firstly, we will discuss the well-established method of ECT and its different treatment modalities. Secondly, we will discuss the properties of ketamine, the discovery of its antidepressive effects and the route to clinical approval of the esketamine nasal spray. We will comment on research settings which have evaluated intravenous ketamine against ECT. The decision-making process between esketamine nasal spray or ECT should include the assessment of contraindications, age, severity of disease, presence of psychotic symptoms, patient preference and treatment accessibility. We conclude that both treatment options are highly effective in TRD. If both are indicated, pragmatically esketamine will be chosen before ECT; however, ECT studies in ketamine non-responders are missing.

Mechanisms Involved in the Link between Depression, Antidepressant Treatment, and Associated Weight Change.

Major depressive disorder is a severe mood disorder associated with a marked decrease in quality of life and social functioning, accompanied by a risk of suicidal behavior. Therefore, seeking out and adhering to effective treatment is of great personal and society-wide importance. Weight changes associated with antidepressant therapy are often cited as the reason for treatment withdrawal and thus are an important topic of interest. There indeed exists a significant mechanistic overlap between depression, antidepressant treatment, and the regulation of appetite and body weight. The suggested pathomechanisms include the abnormal functioning of the homeostatic (mostly humoral) and hedonic (mostly dopaminergic) circuits of appetite regulation, as well as causing neuromorphological and neurophysiological changes underlying the development of depressive disorder. However, this issue is still extensively discussed. This review aims to summarize mechanisms linked to depression and antidepressant therapy in the context of weight change.

The clinical perspective on late-onset depression in European real-world treatment settings.

The clinical phenotype of the so-called late-onset depression (LOD) affecting up to 30% of older adults and yielding heterogeneous manifestations concerning symptoms, severity and course has not been fully elucidated yet. This European, cross-sectional, non-interventional, naturalistic multicenter study systematically investigated socio-demographic and clinical correlates of early-onset depression (EOD) and LOD (age of onset ≥ 50 years) in 1410 adult in- and outpatients of both sexes receiving adequate psychopharmacotherapy. In a total of 1329 patients (94.3%) with known age of disease onset, LOD was identified in 23.2% and was associated with unemployment, an ongoing relationship, single major depressive episodes, lower current suicidal risk and higher occurrence of comorbid hypertension. In contrast, EOD was related to higher rates of comorbid migraine and additional psychotherapy. Although the applied study design does not allow to draw any causal conclusions, the present results reflect broad clinical settings and emphasize easily obtainable features which might be characteristic for EOD and LOD. A thoughtful consideration of age of onset might, hence, contribute to optimized diagnostic and therapeutic processes in terms of the globally intended precision medicine, ideally enabling early and adequate treatment allocations and implementation of respective prevention programs.

Depression and coronary heart disease: mechanisms, interventions, and treatments.

Coronary heart disease (CHD), a cardiovascular condition that poses a significant threat to human health and life, has imposed a substantial economic burden on the world. However, in contrast to conventional risk factors, depression emerges as a novel and independent risk factor for CHD. This condition impacts the onset and progression of CHD and elevates the risk of adverse cardiovascular prognostic events in those already affected by CHD. As a result, depression has garnered increasing global attention. Despite this growing awareness, the specific mechanisms through which depression contributes to the development of CHD remain unclear. Existing research suggests that depression primarily influences the inflammatory response, Hypothalamic-pituitary-adrenocortical axis (HPA) and Autonomic Nervous System (ANS) dysfunction, platelet activation, endothelial dysfunction, lipid metabolism disorders, and genetics, all of which play pivotal roles in CHD development. Furthermore, the effectiveness and safety of antidepressant treatment in CHD patients with comorbid depression and its potential impact on the prognosis of CHD patients have become subjects of controversy. Further investigation is warranted to address these unresolved questions.

Links among genetic variants and hierarchical brain structural and functional networks for antidepressant treatment: A multivariate study.

BACKGROUND: Antidepressant treatment effects are strongly heritable and have substantial effects on brain function and structure, but the underlying mechanisms are still poorly understood. In this research, we aimed to evaluate the factors of single nucleotide polymorphisms (SNPs) and hierarchical brain structural and functional networks that were associated with antidepressant treatment. Moreover, we further explored the correlations and mediation pattern among "brain structure-brain function-gene" in major depressive disorder (MDD). METHODS: We analysed 405 SNPs and rich club/feeder/local connections of hierarchical structural and functional networks with three-way parallel independent component analysis in 179 MDD patients. The group-discriminative independent components of the three modalities between responders and non-responders of antidepressant treatment were identified. Pearson correlations and mediation analysis were further utilized to investigate the associations among SNPs and connections of the structural and functional networks. RESULTS: Notably, correlations with antidepressant treatment outcomes were found in structural, functional and SNP modalities simultaneously. The features of group-discriminative independent components included the shared feeder connections of hub regions with the inferior frontal orbital gyrus and amygdala in structural and functional modalities and genes enriched in circadian rhythmic processes and dopaminergic synapse pathways. The structural feeder network displayed close correlations with SNPs and the functional feeder network. Furthermore, the structural feeder network could mediate the association between SNPs and the functional feeder network, implying that genetic variants might influence brain function by affecting brain structure in MDD. CONCLUSIONS: These findings provide potential biomarkers for antidepressant therapy and provide a better grasp of the associations among SNPs and hierarchical structural and functional networks in MDD.

Discontinuation of antidepressant treatment: a retrospective cohort study on more than 20,000 participants.

BACKGROUND: Factors influencing antidepressant treatment discontinuation are poorly understood. In the present study, we aimed to estimate the prevalence of antidepressant treatment discontinuation and identify demographic characteristics, psychiatric comorbidities, and specific side effects associated with treatment discontinuation. METHODS: We leveraged data from the Australian Genetics of Depression Study (AGDS; N = 20,941) to perform a retrospective cohort study on antidepressant treatment discontinuation. Participants were eligible if they were over 18 years of age, had taken antidepressants in the past 4 years, and provided informed consent. RESULTS: Among the ten antidepressants studied, the highest discontinuation rates were observed for Mirtazapine (57.3%) and Amitriptyline (51.6%). Discontinuation rates were comparable across sexes except for Mirtazapine, for which women were more likely to discontinue. The two most common side effects, reduced sexual function and weight gain, were not associated with increased odds of treatment discontinuation. Anxiety, agitation, suicidal thoughts, vomiting, and rashes were associated with higher odds for treatment discontinuation, as were lifetime diagnoses of PTSD, ADHD, and a higher neuroticism score. Educational attainment showed a negative (protective) association with discontinuation across medications. CONCLUSIONS: Our study suggests that not all side effects contribute equally to discontinuation. Common side effects such as reduced sexual function and weight gain may not necessarily increase the risk of treatment discontinuation. Side effects linked to discontinuation can be divided into two groups, psychopathology related and allergy/intolerance.

The Effect of Antidepressant Treatment on Neurocognitive Functions, Redox and Inflammatory Parameters in the Context of COVID-19.

Inflammation is an important component of the etiopathology of depression that uses oxidative and nitrosative stress (O&NS) and elevated inflammatory markers. SARS-CoV-2 infection is also associated with abnormal inflammatory processes, which may impair effective treatment of depression in COVID-19 survivors. In the presented study, thirty-three hospitalized patients with major depressive disorder (MDD) were started on antidepressant treatment, and twenty-one were re-evaluated after 4-6 weeks. The control group consisted of thirty healthy volunteers. All participants underwent neuropsychiatric evaluation, biochemical blood and urine analyses. The results of the research demonstrated positive correlations of the Hamilton Depression Rating Scale (HAM-D) scores with serum catalase (CAT) and urinary S-Nitrosothiols levels, and the Beck Depression Inventory (BDI) scores with serum reduced glutathione (GSH) and superoxide dismutase (SOD) levels. Depressed patients with a history of COVID-19 prior to the treatment had higher urinary nitric oxide (NO) levels and lower serum glutathione peroxidase (GPx) levels. In the control group, COVID-19 survivors had higher levels of urinary N-formylkynurenine (NFK). Our results suggest that the antidepressant treatment has a modulating effect on O&NS, reduces depressive symptoms and improves cognitive functions The present study does not indicate that clinical response to antidepressant treatment is associated with COVID-19 history and baseline SARS-CoV-2 antibody levels. Nevertheless, further research in this area is needed to systematize antidepressant treatment in COVID-19 survivors.

A 6-week, phase IIb, randomized, double-blind, placebo-controlled trial of Anyu Peibo capsules for the treatment of major depressive disorder in adults.

Background: Almost one-third of patients with major depressive disorder (MDD) do not respond to conventional antidepressants, and new treatments for MDD are urgently needed. Objectives: This phase IIb clinical trial was designed to evaluate the efficacy and safety of Anyu Peibo capsules in the treatment of adults with MDD. Design: A multicenter, randomized, double-blind, placebo-controlled, parallel-group, fixed-dose study. Methods: A total of 172 patients with MDD from nine study centers were randomized (1:1) to receive placebo (n = 86) or oral Anyu Peibo capsules (0.8 g) twice per day (n = 86) for 6 weeks. The primary endpoint was the change in the Montgomery Åsberg Depression Rating Scale (MADRS) total score from baseline to week 6, analyzed using an analysis of covariance (ANCOVA) approach with the baseline MADRS score, center effect and center by group interaction as the covariates. Other efficacy endpoints and variables included clinical response and remission rates according to the MADRS and the 17-item Hamilton Depression Rating Scale (HAMD-17) scores, the change in the HAMD-17, Clinical Global Impression - Severity scale and Clinical Global Impression - Improvement scale scores and the reduction in the Hamilton Anxiety Scale from baseline to week 6. Results: The mean baseline MADRS total scores were 29.20 and 29.72 in the Anyu Peibo (n = 82) and placebo groups (n = 81), respectively. The least squares mean change in the MADRS score from baseline to week 6 was 16.59 points in the Anyu Peibo group and 14.51 points in the placebo group. Although there were greater reductions in the MADRS score from baseline to week 6 in the Anyu Peibo capsule group compared to the placebo group, the difference did not reach statistical significance (least-squares mean difference, 2.07 points; 95% confidence interval, -0.27 to 4.41; p = 0.0819). The results of sensitivity analyses by ANCOVA with the last observation carried forward method for missing data indicated that the administration of Anyu Peibo capsules may lead to a significant reduction in depressive symptoms compared to the placebo (least-squares mean difference: 3.29 points; 95% confidence interval: 0.64-5.93; p = 0.0152). Furthermore, Anyu Peibo capsules showed significant benefits over placebo when the change in the HAMD-17 score from baseline to week 6 was evaluated as the secondary analysis (t = 2.01; 95% confidence interval, 0.03-4.23; p = 0.0464). Conclusion: Anyu Peibo capsules may have an effective and safe antidepressant effect, which warrants further research.

Variable selection in high dimensions for discrete-outcome individualized treatment rules: Reducing severity of depression symptoms.

Despite growing interest in estimating individualized treatment rules, little attention has been given the binary outcome setting. Estimation is challenging with nonlinear link functions, especially when variable selection is needed. We use a new computational approach to solve a recently proposed doubly robust regularized estimating equation to accomplish this difficult task in a case study of depression treatment. We demonstrate an application of this new approach in combination with a weighted and penalized estimating equation to this challenging binary outcome setting. We demonstrate the double robustness of the method and its effectiveness for variable selection. The work is motivated by and applied to an analysis of treatment for unipolar depression using a population of patients treated at Kaiser Permanente Washington.

Levonorgestrel intrauterine device and depression: A Swedish register-based cohort study.

BACKGROUND: The levonorgestrel intrauterine device (LNG-IUD) is traditionally viewed as a safe contraceptive with limited systemic effects. However, three recent studies have indicated an increased risk of depression subsequent to LNG-IUD use. This study aimed to examine the potential associated risk between LNG-IUDs and depression, and determine which women are at risk. METHODS: This longitudinal cohort study was based on data from seven Swedish national population-based registers. All Nordic-born women aged 15-24 years residing in Sweden between 2010 and 2017 were included. Cox regression was implemented to estimate the adjusted hazard ratio (AHR) for developing depression, defined as first depression diagnosis or redeemed prescription for antidepressant treatment. We adjusted for age, education level, parental country of origin, parental psychiatric health, previous hormonal contraceptive use and medical indications for contraceptive use. FINDINGS: 703,157 women were included in the analysis. The LNG-IUD was associated with 57 % increased risk of depression [AHR 1.57 (95 % CI 1.51-1.64)]. The greatest risk increase was seen in adolescent women [AHR 2.57, (95 % CI 2.36-2.80)] and women who used the LNG-IUD as their first hormonal contraceptive method [AHR 1.63, (95 % CI 1.50-1.78)]. The risk of depression decreased at the end of study period [AHR 1.43, (95 % CI 1.36-1.51)], once the LNG-IUD became more widely accessible among nulliparous women. CONCLUSIONS: Adolescent women who use the LNG-IUD as their first-ever hormonal contraceptive are at increased risk of developing depression. However, additional impact from confounding factors is likely as risk estimates decreased over the study period. Further research needs to determine if there is a causal relationship between LNG-IUDs and depression.

The kynurenine pathway in major depressive disorder under different disease states: A systematic review and meta-analysis.

BACKGROUND: A disruption of the kynurenine (KYN) pathway may exist in major depressive disorder (MDD). However, the changing pattern of the KYN pathway across the different disease states in MDD is unclear. Herein, we performed a meta-analysis to examine the differences in KYN metabolites between patients in the current episode of MDD (cMDD) and patients in remission (rMDD), as well as the changes after treatments. METHODS: Literature was systematically searched from electronic databases, from inception up to September 2022. Random-effect models were used to quantify the differences in KYN metabolites between patients with MDD across acute depressive episode and remission phases, as well as the changes after treatments. RESULTS: Fifty-one studies involving 7056 participants were included. Tryptophan (TRP), KYN, kynurenic acid (KYNA), KYNA/quinolinic acid (QA), KYNA/3-hydroxykynurenine (3-HK), and KYNA/KYN were significantly lower, while KYN/TRP was significantly higher in patients with cMDD. Moreover, these effect sizes were generally larger in medication-free patients. No significant differences were found between patients with rMDD and HCs. Additionally, KYNA was found negatively correlated with depression severity and significantly increased after treatments, while the alteration was not found in QA. LIMITATIONS: The number of included studies of patients with rMDD and longitudinal studies investigating the change of the KYN metabolites after treatment with antidepressants was limited. In addition, the heterogeneity across included studies was relatively high. CONCLUSIONS: These findings showed a comprehensive image of the unique dysfunction pattern of the KYN pathway across different MDD states and highlighted KYNA as a potentially sensitive biomarker of MDD.

Early antidepressant treatment response prediction in major depression using clinical and TPH2 DNA methylation features based on machine learning approaches.

OBJECTIVE: To identify DNA methylation and clinical features, and to construct machine learning classifiers to assign the patients with major depressive disorder (MDD) into responders and non-responders after a 2-week treatment into responders and non-responders. METHOD: Han Chinese patients (291 in total) with MDD comprised the study population. Datasets contained demographic information, environment stress factors, and the methylation levels of 38 methylated sites of tryptophan hydroxylase 2 (TPH2) genes in peripheral blood samples. Recursive Feature Elimination (RFE) was employed to select features. Five classification algorithms (logistic regression, classification and regression trees, support vector machine, logitboost and random forests) were used to establish the models. Performance metrics (AUC, F-Measure, G-Mean, accuracy, sensitivity, specificity, positive predictive value and negative predictive value) were computed with 5-fold-cross-validation. Variable importance was evaluated by random forest algorithm. RESULT: RF with RFE outperformed the other models in our samples based on the demographic information and clinical features (AUC = 61.2%, 95%CI: 60.1-62.4%) / TPH2 CpGs features (AUC = 66.6%, 95%CI: 65.4-67.8%) / both clinical and TPH2 CpGs features (AUC = 72.9%, 95%CI: 71.8-74.0%). CONCLUSION: The effects of TPH2 on the early-stage antidepressant response were explored by machine learning algorithms. On the basis of the baseline depression severity and TPH2 CpG sites, machine learning approaches can enhance our ability to predict the early-stage antidepressant response. Some potentially important predictors (e.g., TPH2-10-60 (rs2129575), TPH2-2-163 (rs11178998), age of first onset, age) in early-stage treatment response could be utilized in future fundamental research, drug development and clinical practice.

Prediction of Diagnosis and Treatment Response in Adolescents With Depression by Using a Smartphone App and Deep Learning Approaches: Usability Study.

BACKGROUND: Lack of quantifiable biomarkers is a major obstacle in diagnosing and treating depression. In adolescents, increasing suicidality during antidepressant treatment further complicates the problem. OBJECTIVE: We sought to evaluate digital biomarkers for the diagnosis and treatment response of depression in adolescents through a newly developed smartphone app. METHODS: We developed the Smart Healthcare System for Teens At Risk for Depression and Suicide app for Android-based smartphones. This app passively collected data reflecting the social and behavioral activities of adolescents, such as their smartphone usage time, physical movement distance, and the number of phone calls and text messages during the study period. Our study consisted of 24 adolescents (mean age 15.4 [SD 1.4] years, 17 girls) with major depressive disorder (MDD) diagnosed with Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version and 10 healthy controls (mean age 13.8 [SD 0.6] years, 5 girls). After 1 week's baseline data collection, adolescents with MDD were treated with escitalopram in an 8-week, open-label trial. Participants were monitored for 5 weeks, including the baseline data collection period. Their psychiatric status was measured every week. Depression severity was measured using the Children's Depression Rating Scale-Revised and Clinical Global Impressions-Severity. The Columbia Suicide Severity Rating Scale was administered in order to assess suicide severity. We applied the deep learning approach for the analysis of the data. Deep neural network was employed for diagnosis classification, and neural network with weighted fuzzy membership functions was used for feature selection. RESULTS: We could predict the diagnosis of depression with training accuracy of 96.3% and 3-fold validation accuracy of 77%. Of the 24 adolescents with MDD, 10 responded to antidepressant treatments. We predicted the treatment response of adolescents with MDD with training accuracy of 94.2% and 3-fold validation accuracy of 76%. Adolescents with MDD tended to move longer distances and use smartphones for longer periods of time compared to controls. The deep learning analysis showed that smartphone usage time was the most important feature in distinguishing adolescents with MDD from controls. Prominent differences were not observed in the pattern of each feature between the treatment responders and nonresponders. The deep learning analysis revealed that the total length of calls received as the most important feature predicting antidepressant response in adolescents with MDD. CONCLUSIONS: Our smartphone app demonstrated preliminary evidence of predicting diagnosis and treatment response in depressed adolescents. This is the first study to predict the treatment response of adolescents with MDD by examining smartphone-based objective data with deep learning approaches.