A New Prediction Model for Evaluating Treatment-Resistant Depression

Characterizing Major Depressive Disorder (MDD) using alpha-band activity in resting-state electroencephalogram (EEG) combined with MATRICS Consensus Cognitive Battery (MCCB)

BACKGROUND

The diagnosis of major depressive disorder (MDD) is commonly based on the subjective evaluation by experienced psychiatrists using clinical scales. Hence, it is particularly important to find more objective biomarkers to aid in diagnosis and further treatment. Alpha-band activity (7-13 Hz) is the most prominent component in resting electroencephalogram (EEG), which is also thought to be a potential biomarker. Recent studies have shown the existence of multiple sub-oscillations within the alpha band, with distinct neural underpinnings. However, the specific contribution of these alpha sub-oscillations to the diagnosis and treatment of MDD remains unclear.

METHODS

In this study, we recorded the resting-state EEG from MDD and HC populations in both open and closed-eye state conditions. We also assessed cognitive processing using the MATRICS Consensus Cognitive Battery (MCCB).

RESULTS

We found that the MDD group showed significantly higher power in the high alpha range (10.5-11.5 Hz) and lower power in the low alpha range (7-8.5 Hz) compared to the HC group. Notably, high alpha power in the MDD group is negatively correlated with working memory performance in MCCB, whereas no such correlation was found in the HC group. Furthermore, using five established classification algorithms, we discovered that combining alpha oscillations with MCCB scores as features yielded the highest classification accuracy compared to using EEG or MCCB scores alone.

CONCLUSIONS

Our results demonstrate the potential of sub-oscillations within the alpha frequency band as a potential distinct biomarker. When combined with psychological scales, they may provide guidance relevant for the diagnosis and treatment of MDD.

DNA methylation changes in association with trauma-focused psychotherapy efficacy in treatment-resistant depression patients: a prospective longitudinal study

Background: Stressful events increase the risk for treatment-resistant depression (TRD), and trauma-focused psychotherapy can be useful for TRD patients exposed to early life stress (ELS). Epigenetic processes are known to be related to depression and ELS, but there is no evidence of the effects of trauma-focused psychotherapy on methylation alterations.Objective: We performed the first epigenome-wide association study to investigate methylation changes related to trauma-focused psychotherapies effects in TRD patients.Method: Thirty TRD patients assessed for ELS underwent trauma-focused psychotherapy, of those, 12 received trauma-focused cognitive behavioural therapy, and 18 Eye Movement Desensitization and Reprocessing (EMDR). DNA methylation was profiled with Illumina Infinium EPIC array at T0 (baseline), after 8 weeks (T8, end of psychotherapy) and after 12 weeks (T12 - follow-up). We examined differentially methylated CpG sites and regions, as well as pathways analysis in association with the treatment.Results: Main results obtained have shown 110 differentially methylated regions (DMRs) with a significant adjusted p-value area associated with the effects of trauma-focused psychotherapies in the entire cohort. Several annotated genes are related to inflammatory processes and psychiatric disorders, such as LTA, GFI1, ARID5B, TNFSF13, and LST1. Gene enrichment analyses revealed statistically significant processes related to tumour necrosis factor (TNF) receptor and TNF signalling pathway. Stratified analyses by type of trauma-focused psychotherapy showed statistically significant adjusted p-value area in 141 DMRs only for the group of patients receiving EMDR, with annotated genes related to inflammation and psychiatric disorders, including LTA, GFI1, and S100A8. Gene set enrichment analyses in the EMDR group indicated biological processes related to inflammatory response, particularly the TNF signalling pathway.Conclusion: We provide preliminary valuable insights into global DNA methylation changes associated with trauma-focused psychotherapies effects, in particular with EMDR treatment.

Factors Associated with Antidepressant Effects of Ketamine: A Reanalysis of Double-Blind Randomized Placebo-Controlled Trial of Intravenous Ketamine for Treatment-Resistant Depression

INTRODUCTION

Predictors of treatment response to intravenous ketamine remain unclear in patients with treatment-resistant depression (TRD); therefore, this study aimed to clarify these predictors using the US National Institutes of Health database of clinical trials.

METHODS

Data from a placebo-controlled, double-blind, randomized controlled trial were used to assess the efficacy of intravenous ketamine in adult patients with TRD (NCT01920555). For the analysis, data were used from the participants who had received therapeutic doses of intravenous ketamine (i. e., 0.5 and 1.0 mg/kg). Logistic and multivariable regression analyses were conducted to explore the demographic and clinical factors associated with response to treatment or changes in the Hamilton Depression Rating Scale 6 items (HAM-D-6) total score.

RESULTS

This study included 31 patients with TRD (13 women; mean±standard deviation age, 48.4±10.9 years). Logistic regression analysis showed that the age of onset was positively correlated with treatment response after three days of ketamine administration (β=0.08, p=0.037); however, no association was observed between treatment response and age, sex, baseline HAM-D-6 total score, or dissociative score assessed with the Clinician-Administered Dissociative States Scale 40 min after ketamine infusion. Multiple regression analysis showed that no factors were correlated significantly with the percentage change in the HAM-D-6 total score three days after ketamine administration.

DISCUSSION

Later disease onset correlates with a better treatment response three days after ketamine infusion in patients with TRD. Glutamatergic signal transmission may be impaired in patients with an earlier onset of depression, resulting in decreased neuroplasticity, which diminishes ketamine response.

Irremediability in psychiatric euthanasia: examining the objective standard

BACKGROUND

Irremediability is a key requirement for euthanasia and assisted suicide for psychiatric disorders (psychiatric EAS). Countries like the Netherlands and Belgium ask clinicians to assess irremediability in light of the patient's diagnosis and prognosis and 'according to current medical understanding'. Clarifying the relevance of a default objective standard for irremediability when applied to psychiatric EAS is crucial for solid policymaking. Yet so far, a thorough examination of this standard is lacking.

METHODS

Using treatment-resistant depression (TRD) as a test case, through a scoping review in PubMed, we analyzed the state-of-the-art evidence for whether clinicians can accurately predict individual long-term outcome and single out irremediable cases, by examining the following questions: (1) What is the definition of TRD; (2) What are group-level long-term outcomes of TRD; and (3) Can clinicians make accurate individual outcome predictions in TRD?

RESULTS

A uniform definition of TRD is lacking, with over 150 existing definitions, mostly focused on psychopharmacological research. Available yet limited studies about long-term outcomes indicate that a majority of patients with long-term TRD show significant improvement over time. Finally, evidence about individual predictions in TRD using precision medicine is growing, but methodological shortcomings and varying predictive accuracies pose important challenges for its implementation in clinical practice.

CONCLUSION

Our findings support the claim that, as per available evidence, clinicians cannot accurately predict long-term chances of recovery in a particular patient with TRD. This means that the objective standard for irremediability cannot be met, with implications for policy and practice of psychiatric EAS.

Prediction of medication-related osteonecrosis of the jaws using machine learning methods from estrogen receptor 1 polymorphisms and clinical information

Objective

The purpose of this study was to evaluate the effect of estrogen receptor 1 (ESR1) polymorphisms on the development of medication-related osteonecrosis of the jaws (MRONJ) in women with osteoporosis.

Methods

A total of 125 patients taking bisphosphonates was evaluated the relationship between MRONJ occurrence and single nucleotide polymorphisms (SNPs) of ESR1. Clinical information was collected, including current age, treatment duration, and comorbidity. Univariate and Multivariable regression analyzes were performed to evaluate the independent predictive factors for MRONJ occurrence. Predictive models were constructed using machine learning methods such as Lasso regression, Random forest (RF), and Support vector machine (SVM). The area under the receiver-operating curve (AUROC) was used to evaluate the performance of a binary classifier.

Result

Two SNPs of ESR1 (rs4870056 and rs78177662) were significantly associated with MRONJ development. Patients with variant allele (A) of rs4870056 showed 2.45 times (95% CI, 1.03-5.87) the odds of MRONJ occurrence compared to those with wild-type homozygote (GG) after adjusting covariates. Additionally, carriers with variant allele (T) of rs78177662 had higher odds than those with wild-type homozygote (CC) (adjusted odds ratio (aOR), 2.64, 95% CI, 1.00-6.94). Among demographic variables, age ≥ 72 years (aOR, 3.98, 95% CI, 1.60-9.87) and bisphosphonate exposure ≥48 months (aOR, 3.16, 95% CI, 1.26-7.93) were also significant risk factors for MRONJ occurrence. AUROC values of machine learning methods ranged between 0.756-0.806 in the study.

Conclusion

Our study showed that the MRONJ occurrence was associated with ESR1 polymorphisms in osteoporotic women.

Recommendations for selection and adaptation of rating scales for clinical studies of rapid-acting antidepressants

The novel mechanisms of action (MOA) derived from some recently introduced molecular targets have led to regulatory approvals for rapid acting antidepressants (RAADs) that can generate responses within hours or days, rather than weeks or months. These novel targets include the N-methyl-D-glutamate receptor antagonist ketamine, along with its enantiomers and various derivatives, and the allosteric modulators of gamma-aminobutyric acid (GABA) receptors. There has also been a strong resurgence in interest in psychedelic compounds that impact a range of receptor sites including D1, 5-HT7, KOR, 5-HT5A, Sigma-1, NMDA, and BDNF. The RAADs developed from these novel targets have enabled successful treatment for difficult to treat depressed individuals and has generated a new wave of innovation in research and treatment. Despite the advances in the neurobiology and clinical treatment of mood disorders, we are still using rating instruments that were created decades ago for drugs from a different era (e.g., The Hamilton and Montgomery-Åsberg depression rating scales, HDRS, and MADRS) continue to be used. These rating instruments were designed to assess mood symptoms over a 7-day time frame. Consequently, the use of these rating instruments often requires modifications to address items that cannot be assessed in short time frames, such as the sleep and appetite items. This review describes the adaptative approaches that have been made with the existing scales to meet this need and examines additional domains such as daily activities, side effects, suicidal ideation and behavior, and role functioning. Recommendations for future studies are described, including the challenges related to implementation of these adapted measures and approaches to mitigation.

Early symptom change contributes to the outcome prediction of cognitive behavioral therapy for depression patients: A machine learning approach

BACKGROUND

Limited evidence exists regarding the association between early symptom change and later outcomes of cognitive behavioral therapy (CBT). This study aimed to apply machine learning algorithms to predict continuous treatment outcomes based on pre-treatment predictors and early symptom changes and to uncover whether additional variance could be explained compared to regression methods. Additionally, the study examined early subscale symptom changes to determine the most significant predictors of treatment outcome.

METHODS

We investigated CBT outcomes in a large naturalistic dataset (N = 1975 depression patients). The sociodemographic profile, pre-treatment predictors, and early symptom change, including total and subscale scores were used to predict the Symptom Questionnaire (SQ)48 score at the 10th session as a continuous outcome. Different machine learners were compared to linear regression.

RESULTS

Early symptom change and baseline symptom score were the only significant predictors. Models with early symptom change explained 22.0 % to 23.3 % more variance than those without early symptom change. Specifically, the baseline total symptom score, and the early symptom score changes of the subscales pertaining to depression and anxiety were the top three predictors of treatment outcome.

LIMITATION

Excluded patients with missing treatment outcomes had slightly higher symptom scores at baseline, indicating possible selection bias.

CONCLUSION

Early symptom change improved the prediction of treatment outcomes. The prediction performance achieved is far from clinical relevance: the best learner could only explain 51.2 % of the variance in outcomes. Compared to linear regression, more sophisticated preprocessing and learning methods did not substantially improve performance.

Temporal multi-step predictive modeling of remission in major depressive disorder using early stage treatment data; STAR*D based machine learning approach

BACKGROUND

Artificial intelligence is currently being used to facilitate early disease detection, better understand disease progression, optimize medication/treatment dosages, and uncover promising novel treatments and potential outcomes.

METHODS

Utilizing the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) dataset, we built a machine learning model to predict depression remission rates using same clinical data as features for each of the first three antidepressant treatment steps in STAR*D. We only used early treatment data (baseline and first follow up) in each STAR*D step to temporally analyze predictive features of remission at the end of the step.

RESULTS

Our model showed significant prediction performance across the three treatment steps, At step 1, Model accuracy was 66 %; sensitivity-65 %, specificity-67 %, positive predictive value (PPV)-65.5 %, and negative predictive value (NPV)-66.6 %. At step 2, model accuracy was 71.3 %, sensitivity-74.3 %, specificity-69 %, PPV-64.5 %, and NPV-77.9 %. At step 3, accuracy reached 84.6 %; sensitivity-69 %, specificity-88.8 %, PPV-67 %, and NPV-91.1 %. Across all three steps, the early Quick Inventory of Depressive Symptomatology-Self-Report (QIDS-SR) scores were key elements in predicting the final treatment outcome. The model also identified key sociodemographic factors that predicted treatment remission at different steps.

LIMITATIONS

The retrospective design, lack of replication in an independent dataset, and the use of "a complete case analysis" model in our analysis.

CONCLUSIONS

This proof-of-concept study showed that using early treatment data, multi-step temporal prediction of depressive symptom remission results in clinically useful accuracy rates. Whether these predictive models are generalizable deserves further study.

The update of self-identity: Importance of assessing autobiographical memory in major depressive disorder

Major depressive disorder is a leading global cause of disability. There is a growing interest for memory in mood disorders since it might constitute an original tool for prevention, diagnosis, and treatment. MDD is associated with impaired autobiographical memory characterized by a tendency to overgeneral memory, rather than vivid episodic self-defining memory, which is mandatory for problem-solving and projection in the future. This memory bias is maintained by three mechanisms: ruminations, avoidance, and impaired executive control. If we adopt a broader and comprehensive perspective, we can hypothesize that all those alterations have the potential to impair self-identity updating. We posit that this update requires a double referencing process: (1) to internalized self-representation and (2) to an externalized framework dealing with the representation of the consequence of actions. Diagnostic and therapeutic implications are discussed in the light of this model and the importance of assessing autobiographical memory in MDD is highlighted. This article is categorized under: Psychology > Memory Psychology > Brain Function and Dysfunction Neuroscience > Clinical.

Gradient boosting decision-tree-based algorithm with neuroimaging for personalized treatment in depression

Introduction

Pretreatment positron emission tomography (PET) with 2-deoxy-2-[18F]fluoro-D-glucose (FDG) and magnetic resonance spectroscopy (MRS) may identify biomarkers for predicting remission (absence of depression). Yet, no such image-based biomarkers have achieved clinical validity. The purpose of this study was to identify biomarkers of remission using machine learning (ML) with pretreatment FDG-PET/MRS neuroimaging, to reduce patient suffering and economic burden from ineffective trials.

Methods

This study used simultaneous PET/MRS neuroimaging from a double-blind, placebo-controlled, randomized antidepressant trial on 60 participants with major depressive disorder (MDD) before initiating treatment. After eight weeks of treatment, those with ≤ 7 on 17-item Hamilton Depression Rating Scale were designated a priori as remitters (free of depression, 37%). Metabolic rate of glucose uptake (metabolism) from 22 brain regions were acquired from PET. Concentrations (mM) of glutamine and glutamate and gamma-aminobutyric acid (GABA) in anterior cingulate cortex were quantified from MRS. The data were randomly split into 67% train and cross-validation (n = 40), and 33% test (n = 20) sets. The imaging features, along with age, sex, handedness, and treatment assignment (selective serotonin reuptake inhibitor or SSRI vs. placebo) were entered into the eXtreme Gradient Boosting (XGBoost) classifier for training.

Results

In test data, the model showed 62% sensitivity, 92% specificity, and 77% weighted accuracy. Pretreatment metabolism of left hippocampus from PET was the most predictive of remission.

Conclusions

The pretreatment neuroimaging takes around 60 minutes but has potential to prevent weeks of failed treatment trials. This study effectively addresses common issues for neuroimaging analysis, such as small sample size, high dimensionality, and class imbalance.

Venlafaxine XR treatment for older patients with major depressive disorder: decision trees for when to change treatment

BACKGROUND

Predictors of antidepressant response in older patients with major depressive disorder (MDD) need to be confirmed before they can guide treatment.

OBJECTIVE

To create decision trees for early identification of older patients with MDD who are unlikely to respond to 12 weeks of antidepressant treatment, we analysed data from 454 older participants treated with venlafaxine XR (150-300 mg/day) for up to 12 weeks in the Incomplete Response in Late-Life Depression: Getting to Remission study.

METHODS

We selected the earliest decision point when we could detect participants who had not yet responded (defined as >50% symptom improvement) but would do so after 12 weeks of treatment. Using receiver operating characteristic models, we created two decision trees to minimise either false identification of future responders (false positives) or false identification of future non-responders (false negatives). These decision trees integrated baseline characteristics and treatment response at the early decision point as predictors.

FINDING

We selected week 4 as the optimal early decision point. Both decision trees shared minimal symptom reduction at week 4, longer episode duration and not having responded to an antidepressant previously as predictors of non-response. Test negative predictive values of the leftmost terminal node of the two trees were 77.4% and 76.6%, respectively.

CONCLUSION

Our decision trees have the potential to guide treatment in older patients with MDD but they require to be validated in other larger samples.

CLINICAL IMPLICATIONS

Once confirmed, our findings may be used to guide changes in antidepressant treatment in older patients with poor early response.

Epidemiology of treatment resistant depression among major depressive disorder patients in Israel

INTRODUCTION

Major depressive disorder (MDD) is one of the most common mental disorders worldwide, estimated to affect 10-15% of the population per year. Treatment resistant depression (TRD) is estimated to affect a third of these patients who show difficulties in social and occupational function, decline of physical health, suicidal thoughts and increased health care utilization. We describe the prevalence of MDD, TRD and associated healthcare resource utilization in Maccabi Healthcare Services (MHS), a 2.5 million-member state-mandated health service in Israel.

METHODS

All MHS members with an MDD diagnosis were identified within the years 2017-2018 and prevalence assessed by age, sex and TRD. To assess the incidence of MDD, members aged 18-65 years at the start of any MDD episode were identified between 1^st January 2016 and 31^st May 2018 with at least one systemic first-line antidepressant treatment within three months before or after the initial episode. Treatment patterns, time on first-line treatment, and healthcare resource utilization were compared by TRD.

RESULTS

A total of 4960 eligible MDD patients were identified (median age = 51 years, 65% female), representing a period prevalence of 0.218%, and of those, a high proportion of patients received drug treatment (92%). Among incident MDD cases (n = 2553), 24.4% had TRD. Factors associated with TRD included increasing age and personality disorder. Median time on treatment was 3.7 months (longer for those without TRD than those with) and 81.9% of patients purchased more than one month's supply of therapy. In the year after index, patients with TRD had a significant increased number of visits to primary care physicians, psychiatrists, emergency room visits, general hospitalizations, and psychiatric hospitalizations.

CONCLUSION

Our study shows that prevalence of MDD in Israel is low compared to other countries, however once diagnosed, patients' are likely to receive drug treatment. Among patients diagnosed with MDD, the proportion of TRD is similar to other countries, increases with age and is associated with increased healthcare utilization, therefore should be a focus of continued research for finding effective long term treatment options.

Efficiently identifying individuals at high risk for treatment resistance in major depressive disorder using electronic health records

BACKGROUND

With the emergence of evidence-based treatments for treatment-resistant depression, strategies to identify individuals at greater risk for treatment resistance early in the course of illness could have clinical utility. We sought to develop and validate a model to predict treatment resistance in major depressive disorder using coded clinical data from the electronic health record.

METHODS

We identified individuals from a large health system with a diagnosis of major depressive disorder receiving an index antidepressant prescription, and used a tree-based machine learning classifier to build a risk stratification model to identify those likely to experience treatment resistance. The resulting model was validated in a second health system.

RESULTS

In the second health system, the extra trees model yielded an AUC of 0.652 (95% CI: 0.623-0.682); with sensitivity constrained at 0.80, specificity was 0.358 (95% CI: 0.300-0.413). Lift in the top quintile was 1.99 (95% CI: 1.76-2.22). Including additional data for the 4 weeks following treatment initiation did not meaningfully improve model performance.

LIMITATIONS

The extent to which these models generalize across additional health systems will require further investigation.

CONCLUSION

Electronic health records facilitated stratification of risk for treatment-resistant depression and demonstrated generalizability to a second health system. Efforts to improve upon such models using additional measures, and to understand their performance in real-world clinical settings, are warranted.

Treatment Response Prediction for Major Depressive Disorder Patients via Multivariate Pattern Analysis of Thalamic Features

Antidepressant treatment, as an important method in clinical practice, is not suitable for all major depressive disorder (MDD) patients. Although magnetic resonance imaging (MRI) studies have found thalamic abnormalities in MDD patients, it is not clear whether the features of the thalamus are suitable to serve as predictive aids for treatment responses at the individual level. Here, we tested the predictive value of gray matter density (GMD), gray matter volume (GMV), amplitude of low-frequency fluctuations (ALFF), and fractional ALFF (fALFF) of the thalamus using multivariate pattern analysis (MVPA). A total of 74 MDD patients and 44 healthy control (HC) subjects were recruited. Thirty-nine MDD patients and 35 HC subjects underwent scanning twice. Between the two scanning sessions, patients in the MDD group received selective serotonin reuptake inhibitor (SSRI) treatment for 3-month, and HC group did not receive any treatment. Gaussian process regression (GPR) was trained to predict the percentage decrease in the Hamilton Depression Scale (HAMD) score after treatment. The percentage decrease in HAMD score after SSRI treatment was predicted by building GPRs trained with baseline thalamic data. The results showed significant correlations between the true percentage of HAMD score decreases and predictions (p < 0.01, r² = 0.11) in GPRs trained with GMD. We did not find significant correlations between the true percentage of HAMD score decreases and predictions in GMV (p = 0.16, r² = 0.00), ALFF (p = 0.125, r² = 0.00), and fALFF (p = 0.485, r² = 0.10). Our results suggest that GMD of the thalamus has good potential as an aid in individualized treatment response predictions of MDD patients.

Machine learning in the prediction of depression treatment outcomes: a systematic review and meta-analysis

BACKGROUND

Multiple treatments are effective for major depressive disorder (MDD), but the outcomes of each treatment vary broadly among individuals. Accurate prediction of outcomes is needed to help select a treatment that is likely to work for a given person. We aim to examine the performance of machine learning methods in delivering replicable predictions of treatment outcomes.

METHODS

Of 7732 non-duplicate records identified through literature search, we retained 59 eligible reports and extracted data on sample, treatment, predictors, machine learning method, and treatment outcome prediction. A minimum sample size of 100 and an adequate validation method were used to identify adequate-quality studies. The effects of study features on prediction accuracy were tested with mixed-effects models. Fifty-four of the studies provided accuracy estimates or other estimates that allowed calculation of balanced accuracy of predicting outcomes of treatment.

RESULTS

Eight adequate-quality studies reported a mean accuracy of 0.63 [95% confidence interval (CI) 0.56-0.71], which was significantly lower than a mean accuracy of 0.75 (95% CI 0.72-0.78) in the other 46 studies. Among the adequate-quality studies, accuracies were higher when predicting treatment resistance (0.69) and lower when predicting remission (0.60) or response (0.56). The choice of machine learning method, feature selection, and the ratio of features to individuals were not associated with reported accuracy.

CONCLUSIONS

The negative relationship between study quality and prediction accuracy, combined with a lack of independent replication, invites caution when evaluating the potential of machine learning applications for personalizing the treatment of depression.

Characterising the severity of treatment resistance in unipolar and bipolar depression

Background

Treatment-resistant depression (TRD) is classically defined according to the number of suboptimal antidepressant responses experienced, but multidimensional assessments of TRD are emerging and may confer some advantages. Patient characteristics have been identified as risk factors for TRD but may also be associated with TRD severity. The identification of individuals at risk of severe TRD would support appropriate prioritisation of intensive and specialist treatments.

Aims

To determine whether TRD risk factors are associated with TRD severity when assessed multidimensionally using the Maudsley Staging Method (MSM), and univariately as the number of antidepressant non-responses, across three cohorts of individuals with depression.

Method

Three cohorts of individuals without significant TRD, with established TRD and with severe TRD, were assessed (n = 528). Preselected characteristics were included in linear regressions to determine their association with each outcome.

Results

Participants with more severe TRD according to the MSM had a lower age at onset, fewer depressive episodes and more physical comorbidities. These associations were not consistent across cohorts. The number of episodes was associated with the number of antidepressant treatment failures, but the direction of association varied across the cohorts studied.

Conclusions

Several risk factors for TRD were associated with the severity of resistance according to the MSM. Fewer were associated with the raw number of inadequate antidepressant responses. Multidimensional definitions may be more useful for identifying patients at risk of severe TRD. The inconsistency of associations across cohorts has potential implications for the characterisation of TRD.

Predicting outcomes of psychotherapy for depression with electronic health record data

Predictive analytics with electronic health record (EHR) data holds promise for improving outcomes of psychiatric care. This study evaluated models for predicting outcomes of psychotherapy for depression in a clinical practice setting. EHR data from two large integrated health systems (Kaiser Permanente Colorado and Washington) included 5,554 new psychotherapy episodes with a baseline Patient Health Questionnaire (PHQ-9) score ≥ 10 and a follow-up PHQ-9 14–180 days after treatment initiation. Baseline predictors included demographics and diagnostic, medication, and encounter history. Prediction models for two outcomes—follow-up PHQ-9 score and treatment response (≥ 50% PHQ-9 reduction)—were trained in a random sample of 70% of episodes and validated in the remaining 30%. Two methods were used for modeling: generalized linear regression models with variable selection and random forests. Sensitivity analyses considered alternate predictor, outcome, and model specifications. Predictions of follow-up PHQ-9 scores poorly estimated observed outcomes (mean squared error = 31 for linear regression, 40 for random forest). Predictions of treatment response had low discrimination (AUC = 0.57 for logistic regression, 0.61 for random forest), low classification accuracy, and poor calibration. Sensitivity analyses showed similar results. We note that prediction model performance may vary for settings with different care or EHR documentation practices. In conclusion, prediction models did not accurately predict depression treatment outcomes despite using rich EHR data and advanced analytic techniques. Health systems should proceed cautiously when considering prediction models for psychiatric outcomes using baseline intake information. Transparent research should be conducted to evaluate performance of any model intended for clinical use.

Machine Learning-Based Definition of Symptom Clusters and Selection of Antidepressants for Depressive Syndrome

The current polythetic and operational criteria for major depression inevitably contribute to the heterogeneity of depressive syndromes. The heterogeneity of depressive syndrome has been criticized using the concept of language game in Wittgensteinian philosophy. Moreover, “a symptom- or endophenotype-based approach, rather than a diagnosis-based approach, has been proposed” as the “next-generation treatment for mental disorders” by Thomas Insel. Understanding the heterogeneity renders promise for personalized medicine to treat cases of depressive syndrome, in terms of both defining symptom clusters and selecting antidepressants. Machine learning algorithms have emerged as a tool for personalized medicine by handling clinical big data that can be used as predictors for subtype classification and treatment outcome prediction. The large clinical cohort data from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D), Combining Medications to Enhance Depression Outcome (CO-MED), and the German Research Network on Depression (GRND) have recently began to be acknowledged as useful sources for machine learning-based depression research with regard to cost effectiveness and generalizability. In addition, noninvasive biological tools such as functional and resting state magnetic resonance imaging techniques are widely combined with machine learning methods to detect intrinsic endophenotypes of depression. This review highlights recent studies that have used clinical cohort or brain imaging data and have addressed machine learning-based approaches to defining symptom clusters and selecting antidepressants. Potentially applicable suggestions to realize machine learning-based personalized medicine for depressive syndrome are also provided herein.

Treatment-Resistant Depression: Approaches to Treatment

Approximately 30% of people treated for a major depressive episode will not achieve remission after two or more treatment trials of first-line antidepressants and are considered to have treatment-resistant depression (TRD). Because the odds of remission decrease with every subsequent medication trial, it is important for clinicians to understand the characteristics and risk factors for TRD, subtypes of major depressive disorder that are more likely to be less responsive to first-line anti-depressants, and the available treatment options. In the current article, we review the approved treatments for TRD, including esketamine, and the evidence for psilocybin and pramipexole. Although limited in specificity, guidelines to help prescribers identify person-centered treatments for TRD are available. [Journal of Psychosocial Nursing and Mental Health Services, 59(9), 7-11.].

Predicting clinical outcome to specialist multimodal inpatient treatment in patients with treatment resistant depression

BACKGROUND

Treatment resistant depression (TRD) poses a significant clinical challenge, despite a range of efficacious specialist treatments. Accurately predicting response a priori may help to alleviate the burden of TRD. This study sought to determine whether outcome prediction can be achieved in a specialist inpatient setting.

METHODS

Patients at the Affective Disorders Unit of the Bethlam Royal Hospital, with current depression and established TRD were included (N = 174). Patients were treated with an individualised combination of pharmacotherapy and specialist psychological therapies. Predictors included clinical and sociodemographic characteristics, and polygenic risk scores for depression and related traits. Logistic regression models examined associations with outcome, and predictive potential was assessed using elastic net regularised logistic regressions with 10-fold nested cross-validation.

RESULTS

47% of patients responded (50% reduction in HAMD-21 score at discharge). Age at onset and number of depressive episodes were positively associated with response, while degree of resistance was negatively associated. All elastic net models had poor performance (AUC<0.6). Illness history characteristics were commonly retained, and the addition of genetic risk scores did not improve performance.

LIMITATIONS

The patient sample was heterogeneous and received a variety of treatments. Some variable associations may be non-linear and therefore not captured.

CONCLUSIONS

This treatment may be most effective for recurrent patients and those with a later age of onset, while patients more severely treatment resistant at admission remain amongst the most difficult to treat. Individual level prediction remains elusive for this complex group. The assessment of homogenous subgroups should be one focus of future investigations.

Optimizing prediction of response to antidepressant medications using machine learning and integrated genetic, clinical, and demographic data

Major depressive disorder (MDD) is complex and multifactorial, posing a major challenge of tailoring the optimal medication for each patient. Current practice for MDD treatment mainly relies on trial and error, with an estimated 42-53% response rates for antidepressant use. Here, we sought to generate an accurate predictor of response to a panel of antidepressants and optimize treatment selection using a data-driven approach analyzing combinations of genetic, clinical, and demographic factors. We analyzed the response patterns of patients to three antidepressant medications in the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study, and employed state-of-the-art machine learning (ML) tools to generate a predictive algorithm. To validate our results, we assessed the algorithm's capacity to predict individualized antidepressant responses on a separate set of 530 patients in STAR*D, consisting of 271 patients in a validation set and 259 patients in the final test set. This assessment yielded an average balanced accuracy rate of 72.3% (SD 8.1) and 70.1% (SD 6.8) across the different medications in the validation and test set, respectively (p < 0.01 for all models). To further validate our design scheme, we obtained data from the Pharmacogenomic Research Network Antidepressant Medication Pharmacogenomic Study (PGRN-AMPS) of patients treated with citalopram, and applied the algorithm's citalopram model. This external validation yielded highly similar results for STAR*D and PGRN-AMPS test sets, with a balanced accuracy of 60.5% and 61.3%, respectively (both p's < 0.01). These findings support the feasibility of using ML algorithms applied to large datasets with genetic, clinical, and demographic features to improve accuracy in antidepressant prescription.

Practical recommendations for the management of treatment-resistant depression with esketamine nasal spray therapy: Basic science, evidence-based knowledge and expert guidance

OBJECTIVES

Despite the available therapies for treatment-resistant depression (TRD), there are a limited number that are evidence-based and effective in this hard-to-treat population. Esketamine nasal spray, an intranasal N-methyl-d-aspartate (NMDA) glutamate receptor antagonist, is a novel, fast-acting option in this patient population. This manuscript provides expert guidance on the practicalities of using esketamine nasal spray.

METHODS

A group of six European experts in major depressive disorder (MDD) and TRD, with clinical experience of treating patients with esketamine nasal spray, first generated practical recommendations, before editing and voting on these to develop consensus statements during an online meeting.

RESULTS

The final consensus statements encompass not only pre-treatment considerations for patients with TRD, but also specific guidelines for clinicians to consider during and post-administration of esketamine nasal spray.

CONCLUSIONS

Esketamine nasal spray is a novel, fast-acting agent that provides an additional treatment option for patients with TRD who have previously failed several therapies. The guidance here is based on the authors' experience and the available literature; however, further real-world use of esketamine nasal spray will add to existing knowledge. The recommendations offer practical guidance to clinicians who are unfamiliar with esketamine nasal spray.

Replication of machine learning methods to predict treatment outcome with antidepressant medications in patients with major depressive disorder from STAR*D and CAN-BIND-1

Objectives

Antidepressants are first-line treatments for major depressive disorder (MDD), but 40–60% of patients will not respond, hence, predicting response would be a major clinical advance. Machine learning algorithms hold promise to predict treatment outcomes based on clinical symptoms and episode features. We sought to independently replicate recent machine learning methodology predicting antidepressant outcomes using the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) dataset, and then externally validate these methods to train models using data from the Canadian Biomarker Integration Network in Depression (CAN-BIND-1) dataset.

Methods

We replicated methodology from Nie et al (2018) using common algorithms based on linear regressions and decision trees to predict treatment-resistant depression (TRD, defined as failing to respond to 2 or more antidepressants) in the STAR*D dataset. We then trained and externally validated models using the clinical features found in both datasets to predict response (≥50% reduction on the Quick Inventory for Depressive Symptomatology, Self-Rated [QIDS-SR]) and remission (endpoint QIDS-SR score ≤5) in the CAN-BIND-1 dataset. We evaluated additional models to investigate how different outcomes and features may affect prediction performance.

Results

Our replicated models predicted TRD in the STAR*D dataset with slightly better balanced accuracy than Nie et al (70%-73% versus 64%-71%, respectively). Prediction performance on our external methodology validation on the CAN-BIND-1 dataset varied depending on outcome; performance was worse for response (best balanced accuracy 65%) compared to remission (77%). Using the smaller set of features found in both datasets generally improved prediction performance when evaluated on the STAR*D dataset.

Conclusion

We successfully replicated prior work predicting antidepressant treatment outcomes using machine learning methods and clinical data. We found similar prediction performance using these methods on an external database, although prediction of remission was better than prediction of response. Future work is needed to improve prediction performance to be clinically useful.

Economic Burden of Treatment-Resistant Depression among Adults with Chronic Non-Cancer Pain Conditions and Major Depressive Disorder in the US

OBJECTIVE

Major depressive disorder (MDD) and chronic non-cancer pain conditions (CNPC) often co-occur and exacerbate one another. Treatment-resistant depression (TRD) in adults with CNPC can amplify the economic burden. This study examined the impact of TRD on direct total and MDD-related healthcare resource utilization (HRU) and costs among commercially insured patients with CNPC and MDD in the US.

METHODS

The retrospective longitudinal cohort study employed a claims-based algorithm to identify adults with TRD from a US claims database (January 2007 to June 2017). Costs (2018 US$) and HRU were compared between patients with and without TRD over a 12-month period after TRD/non-TRD index date. Counterfactual recycled predictions from generalized linear models were used to examine associations between TRD and annual HRU and costs. Post-regression linear decomposition identified differences in patient-level factors between TRD and non-TRD groups that contributed to the excess economic burden of TRD.

RESULTS

Of the 21,180 adults with CNPC and MDD, 10.1% were identified as having TRD. TRD patients had significantly higher HRU, translating into higher average total costs (US$21,015_TRD vs US$14,712_{No TRD}) and MDD-related costs (US$1201_TRD vs US$471_{No TRD}) compared with non-TRD patients (all p < 0.001). Prescription drug costs accounted for 37.6% and inpatient services for 30.7% of the excess total healthcare costs among TRD patients. TRD patients had a significantly higher number of inpatient (incidence rate ratio [IRR] 1.30, 95% CI 1.14-1.47) and emergency room visits (IRR 1.21, 95% CI 1.10-1.34) than non-TRD patients. Overall, 46% of the excess total costs were explained by differences in patient-level characteristics such as polypharmacy, number of CNPC, anxiety, sleep, and substance use disorders between the TRD and non-TRD groups.

CONCLUSION

TRD poses a substantial direct economic burden for adults with CNPC and MDD. Excess healthcare costs may potentially be reduced by providing timely interventions for several modifiable risk factors.

Machine learning to predict clinical remission in depressed patients after acute phase selective serotonin reuptake inhibitor treatment

OBJECTIVE

Selective serotonin reuptake inhibitors (SSRIs) are suggested as the first-line treatment for patients with major depressive disorder (MDD), but the remission rate is unsatisfactory. We aimed to establish machine learning models and explore variables available at baseline to predict the 8-week outcome among patients taking SSRIs.

METHODS

Data from 400 patients were used to build machine learnings. The last observation carried forward approach was used to determine the remitter/non-remitter status of the patients at week 8. Using least absolute shrinkage and selection operator (LASSO) to select features, we built 4 different machine learning algorithms including gradient boosting decision tree, support vector machine (SVM), random forests, and logistic regression with five-fold cross-validation. Then, we adopted Shapley additive explanations (SHAP) values to interpret the model output.

RESULTS

The remission rate is 67.8%. We obtained 78 features from the baseline characteristics, including 25 sociodemographic characteristics, 31 clinical features, 15 psychological traits and 7 neurocognitive functions, and 13 of these features were selected to establish SVM. The accuracy of the SVM prediction is 74.49%, reaching an average area under the curve of 0.734±0.043. The sensitivity is 0.899±0.038 with a positive predictive value of 0.776±0.028. The specificity is 0.422±0.091 with a negative predictive value of 0.674±0.086. According to the SHAP values, neurocognitive functions and anxiety and hypochondriasis symptoms were important predictors.

CONCLUSION

Our study supports the utilization of machine learning approaches with inexpensive and highly accessible variables to accurately predict the 8-week treatment outcome of SSRIs in patients with MDD.

Using a machine learning approach to investigate factors associated with treatment-resistant depression among adults with chronic non-cancer pain conditions and major depressive disorder

OBJECTIVE

Presence of chronic non-cancer pain conditions (CNPC) among adults with major depressive disorder (MDD) may reduce benefits of antidepressant therapy, thereby increasing the possibility of treatment resistance. This study sought to investigate factors associated with treatment-resistant depression (TRD) among adults with MDD and CNPC using machine learning approaches.

METHODS

This retrospective cohort study was conducted using a US claims database which included adults with newly diagnosed MDD and CNPC (January 2007-June 2017). TRD was identified using a clinical staging algorithm for claims data. Random forest (RF), a machine learning method, and logistic regression was used to identify factors associated with TRD. Initial model development included 42 known and/or probable factors that may be associated with TRD. The final refined model included 20 factors.

RESULTS

Included in the sample were 23,645 patients (73% female mean age: 55 years; 78% with ≥2 CNPC, and 91% with joint pain/arthritis). Overall, 11.4% adults (N = 2684) met selected criteria for TRD. The five leading factors associated with TRD were the following: mental health specialist visits, polypharmacy (≥5 medications), psychotherapy use, anxiety, and age. Cross-validated logistic regression model indicated that those with TRD were younger, more likely to have anxiety, mental health specialist visits, polypharmacy, and psychotherapy use with adjusted odds ratios (AORs) ranging from 1.93 to 1.27 (all ps < .001).

CONCLUSION

Machine learning identified several factors that warrant further investigation and may serve as potential targets for clinical intervention to improve treatment outcomes in patients with TRD and CNPC.

A personalized index to inform selection of a trauma-focused or non-trauma-focused treatment for PTSD

PTSD treatment guidelines recommend several treatments with extensive empirical support, including Prolonged Exposure (PE), a trauma-focused treatment and Present-Centered Therapy (PCT), a non-trauma-focused therapy. Research to inform treatment selection has yielded inconsistent findings with single prognostic variables that are difficult to integrate into clinical decision-making. We examined whether a combination of prognostic factors can predict different benefits in a trauma-focused vs. a non-trauma-focused psychotherapy. We applied a multi-method variable selection procedure and developed a prognostic index (PI) with a sample of 267 female veterans and active-duty service members (mean age 45; SD = 9.37; 53% White) with current PTSD who began treatment in a randomized clinical trial comparing PE and PCT. We conducted linear regressions predicting outcomes (Clinician-Administered PTSD Scale score) with treatment condition, the PI, and the interaction between the PI and treatment condition. The interaction between treatment type and PI moderated treatment response, moderated post-treatment symptom severity, b = 0.30, SE_b = 0.15 [95% CI: 0.01, 0.60], p = .049. For the 64% of participants with the best prognoses, PE resulted in better post-treatment outcomes; for the remainder, there was no difference. Use of a PI may lead to optimized patient outcomes and greater confidence when selecting trauma-focused treatments.

Higher Negative Self-Reference Level in Patients With Personality Disorders and Suicide Attempt(s) History During Biological Treatment for Major Depressive Disorder: Clinical Implications

Objective: The aim of the study was to identify clinical variables associated with changes in specific domains of self-reported depression during treatment by antidepressant and/or repetitive Transcranial Magnetic Stimulation (rTMS) in patients with Major Depressive Disorder (MDD).

Methods: Data from a trial involving 170 patients with MDD receiving either venlafaxine, rTMS or both were re-analyzed. Depressive symptoms were assessed each week during the 2 to 6 weeks of treatment with the 13-item Beck Depression Inventory (BDI₁₃). Associations between depression changes on BDI₁₃ domains (Negative Self-Reference, Sad Mood, and Performance Impairment), treatment arm, time, and clinical variables were tested in a mixed linear model.

Results: A significant decrease of self-reported depressive symptoms was observed over time. The main characteristics associated with persistent higher depressive symptomatology on Negative Self-Reference domain of the BDI₁₃ were personality disorders (+2.1 points), a past history of suicide attempt(s) (+1.7 points), age under 65 years old (+1.5 points), and female sex (+1.1 points).

Conclusions: Early cognitive intervention targeting specifically negative self-referencing process could be considered during pharmacological or rTMS treatment for patients with personality disorders and past history of suicide attempt(s).

Data Mining in Healthcare: Applying Strategic Intelligence Techniques to Depict 25 Years of Research Development

In order to identify the strategic topics and the thematic evolution structure of data mining applied to healthcare, in this paper, a bibliometric performance and network analysis (BPNA) was conducted. For this purpose, 6138 articles were sourced from the Web of Science covering the period from 1995 to July 2020 and the SciMAT software was used. Our results present a strategic diagram composed of 19 themes, of which the 8 motor themes ('NEURAL-NETWORKS', 'CANCER', 'ELETRONIC-HEALTH-RECORDS', 'DIABETES-MELLITUS', 'ALZHEIMER'S-DISEASE', 'BREAST-CANCER', 'DEPRESSION', and 'RANDOM-FOREST') are depicted in a thematic network. An in-depth analysis was carried out in order to find hidden patterns and to provide a general perspective of the field. The thematic network structure is arranged thusly that its subjects are organized into two different areas, (i) practices and techniques related to data mining in healthcare, and (ii) health concepts and disease supported by data mining, embodying, respectively, the hotspots related to the data mining and medical scopes, hence demonstrating the field's evolution over time. Such results make it possible to form the basis for future research and facilitate decision-making by researchers and practitioners, institutions, and governments interested in data mining in healthcare.

Inflammation-related microRNAs are involved in stressful life events exposure and in trauma-focused psychotherapy in treatment-resistant depressed patients

BACKGROUND

About 30% of major depressive disorder (MDD) patients are classified as resistant to treatment (treatment-resistant depression, TRD). Among the factors associated with unfavourable treatment outcomes, stressful life events play a relevant role, and trauma-focused psychotherapy has been successfully proposed for the treatment of patients with a history of such events. Stressful experiences are related to enhanced inflammation and, recently, microRNAs (miRNAs) have emerged as potential mediators of the association between these experiences and psychiatric disorders. To date, no study has explored the effects of stressful life events on miRNAs in MDD patients.

OBJECTIVE

The objective of the present study was to assess possible miRNA blood expression alterations in TRD patients induced by the exposure to stressful life events and to investigate the effects of trauma-focused psychotherapy on the expression profiles of the same miRNAs, as well as their possible predictivity in relation to therapy outcome.

METHOD

The basal levels (T0) of seven candidate miRNAs (miR-15a/miR-29a/miR-125b/miR-126/miR-146a/miR-195/let-7f) were measured in the whole blood of 41 TRD patients. A subgroup of patients (n = 21) underwent trauma-focused psychotherapy; for all of them, miRNA levels were also longitudinally assessed (T4: after 4 weeks of treatment; T8: end of treatment; T12: follow-up visit), contextually to clinical evaluations.

RESULTS

miR-146a levels negatively correlated with recent stressful life event scores (p = .001), whereas the levels of miR-15a, miR-29a, miR-126, miR-195, and let-7f changed during the psychotherapy (best p = 1.98*10^-9). miR-29a was also identified as a response predictor, with lower baseline levels predicting non-response (p = .019) or worse improvement in mood symptoms (p = .032).

CONCLUSIONS

The study results could contribute to clarify the underlying molecular mechanisms and to identify novel biomarkers of stressful experiences and response to targeted treatments.

Combining machine learning algorithms for prediction of antidepressant treatment response

OBJECTIVES

Predictors for unfavorable treatment outcome in major depressive disorder (MDD) applicable for treatment selection are still lacking. The database of a longitudinal multicenter study on 1079 acutely depressed patients, performed by the German research network on depression (GRND), allows supervised and unsupervised learning to further elucidate the interplay of clinical and psycho-sociodemographic variables and their predictive impact on treatment outcome phenotypes.

EXPERIMENTAL PROCEDURES

Treatment response was defined by a change of HAM-D 17-item baseline score ≥50% and remission by the established threshold of ≤7, respectively, after up to eight weeks of inpatient treatment. After hierarchical symptom clustering and stratification by treatment subtypes (serotonin reuptake inhibitors, tricyclic antidepressants, antipsychotic, and lithium augmentation), prediction models for different outcome phenotypes were computed with random forest in a cross-center validation design. In total, 88 predictors were implemented.

RESULTS

Clustering revealed four distinct HAM-D subscores related to emotional, anxious, sleep, and appetite symptoms, respectively. After feature selection, classification models reached moderate to high accuracies up to 0.85. Highest accuracies were observed for the SSRI and TCA subgroups and for sleep and appetite symptoms, while anxious symptoms showed poor predictability.

CONCLUSION

Our results support a decisive role for machine learning in the management of antidepressant treatment. Treatment- and symptom-specific algorithms may increase accuracies by reducing heterogeneity. Especially, predictors related to duration of illness, baseline depression severity, anxiety and somatic symptoms, and personality traits moderate treatment success. However, prospectives application of machine learning models will be necessary to prove their value for the clinic.

Predicting Incident Treatment-Resistant Depression: A Model Designed for Health Systems of Care

BACKGROUND

Major depressive disorder (MDD) is a prevalent and debilitating condition. While numerous treatment options are available, low treatment response and high remission rates remain common, leading to the concept of treatment-resistant depression (TRD): a classification applied to patients who fail multiple courses of therapy. A patient with TRD can only be identified after repeated, and often prolonged, therapeutic efforts.

OBJECTIVE

To use data readily available to integrated delivery networks to identify characteristics predictive of TRD among patients initiating pharmacotherapy for MDD.

METHODS

Decision Resources Group Real-World Data, an integrated medical/pharmacy claims and electronic health record dataset, was used to conduct a retrospective, longitudinal cohort study of patients with MDD who initiated antidepressant treatment between July 1, 2014, and December 31, 2015. Individuals were followed for 24 months to determine treatment resistance. Eligible individuals had integrated claims and electronic health record data available, completed at least 1 course of therapy of adequate dose and duration to achieve response, and had 30 months of continuous benefits eligibility (6 months before and 24 months after treatment initiation). Stepwise logistic regression and demographic, health history, health care utilization, medication, provider, and related characteristics were used to predict onset of TRD.

RESULTS

35,246 people met eligibility and 7,098 (20.1%) met TRD criteria after an average of 402 days. Significant predictors of TRD included patient age, diagnosis of insomnia and hypertension, psychiatric office visits, nurse telephonic encounters, anticonvulsant medication use, suicidality, physician specialty associated with index prescription, total prescription drug claims, unique antidepressants attempted, and duration of untreated illness (the lag between diagnosis and index prescription). The final model achieved an area under the curve (AUC) = 0.83. Structured patient-generated health data, specifically, the Patient Health Questionnaire-2 and the Patient Health Questionnaire-9 were only reported for 542 patients (1.5%).

CONCLUSIONS

TRD transition occurs after a prolonged treatment period, suggesting clinical inertia. Using data routinely available to integrated delivery networks and accountable care organizations, it is feasible to identify patients likely to qualify as treatment resistant. Monitoring risk factors may allow health systems to identify patients at risk for TRD earlier, potentially improving outcomes. Early identification of this at-risk population can allow for targeted resources for earlier intervention, more aggressive follow-up, and alternative treatment options. Furthermore, this model can be used to estimate future demand for specialized care resources, such as those delivered by mood disorder clinics.

DISCLOSURES

This project was sponsored by Janssen Scientific Affairs. Pesa, Chow, and Verbanac are employed by Janssen Scientific Affairs and report stock ownership in Johnson & Johnson. Liberman, Davis, Heverly-Fitt, and Ruetsch are employed by Health Analytics, which received funding from Janssen Scientific Affairs for work on this project. This study was presented as a poster at the U.S. Psych Congress; October 3-6, 2019; San Diego, CA.

Is cognitive dysfunction involved in difficult-to-treat depression? Characterizing resistance from a cognitive perspective

BACKGROUND

This study aimed to identify clinical and cognitive factors associated with increased risk for difficult-to-treat depression (DTD) or treatment-resistant depression (TRD).

METHODS

A total of 229 adult outpatients with major depression were recruited from the mental health unit at a public hospital. Participants were subdivided into resistant and nonresistant groups according to their Maudsley Staging Model score. Sociodemographic, clinical, and cognitive (objective and subjective measures) variables were compared between groups, and a logistic regression model was used to identify the factors most associated with TRD risk.

RESULTS

TRD group patients present higher verbal memory impairment than the nonresistant group irrespective of pharmacological treatment or depressive symptom severity. Logistic regression analysis showed that low verbal memory scores (odds ratio [OR]: 2.02; 95% confidence interval [CI]: 1.38-2.95) together with high depressive symptom severity (OR: 1.29; CI95%: 1.01-1.65) were associated with TRD risk.

CONCLUSIONS

Our findings align with neuroprogression models of depression, in which more severe patients, defined by greater verbal memory impairment and depressive symptoms, develop a more resistant profile as a result of increasingly detrimental neuronal changes. Moreover, our results support a more comprehensive approach in the evaluation and treatment of DTD in order to improve illness course. Longitudinal studies are warranted to confirm the predictive value of verbal memory and depression severity in the development of TRD.

The effect of co-morbid anxiety on remission from depression for people participating in a randomised controlled trial of the Friendship Bench intervention in Zimbabwe

BACKGROUND

There is a lack of data from low- and middle-income countries on whether anxiety independently predicts a more chronic course for depression.

METHODS

We undertook secondary data analysis of a cluster randomised controlled trial in Zimbabwe which had tested the effectiveness of the Friendship Bench intervention for common mental disorders compared to enhanced usual care. Inclusion for the current study was participants from the trial who had probable major depression at baseline, defined as scoring => 11 on the locally validated Patient Health Questionnaire (PHQ9). This emerged to be 354 of the original 573 (61.78%) of the original trial sample. Anxiety was measured using the locally validated cut-point on the Generalised Anxiety Disorder scale (GAD-7). Persistent depression was defined as scoring => 11 on the PHQ-9 at six-months follow-up. Analysis in Stata 15 used random-effects logistic regression to adjust for clustering by clinic.

OUTCOMES

Of the 354 participants who were eligible for treatment, 329 (92·9%) completed 6-month follow-up assessment. 37% of the trial sample had persistent depression at 6-months follow-up; 59% in the control arm and 17% in the intervention arm. Co-morbid anxiety present at trial baseline was independently associated with persistent depression after adjusting for age, gender and baseline depression severity (adjusted OR = 2·83, 95% CI 1·32-6·07). There was no evidence of effect modification by trial arm. Baseline depression severity also predicted persistent depression. Interpretation Treatment for depression in low and middle-income countries (LMIC) should be directed towards those with greatest need. This includes people with co-morbid anxiety and greater depression severity at initial assessment who are less likely to remit at six months. Advice on coping with anxiety, psychological treatments which target common anxiety symptoms such as fear, avoidance, excessive worry and intrusive thoughts, and Selective Serotonin Reuptake Inhibitors (SSRIs) should be made more widely available in LMIC and offered to those with persistent mixed depression and anxiety.

The Predictive Validity of Machine Learning Models in the Classification and Treatment of Major Depressive Disorder: State of the Art and Future Directions

Major depressive disorder imposes a substantial disease burden worldwide, ranking as the third leading contributor to global disability. In spite of its ubiquity, classifying and treating depression has proven troublesome. One argument put forward to explain this predicament is the heterogeneity of patients diagnosed with the disorder. Recently, many areas of daily life have witnessed the surge of machine learning techniques, computational approaches to elucidate complex patterns in large datasets, which can be employed to make predictions and detect relevant clusters. Due to the multidimensionality at play in the pathogenesis of depression, it is suggested that machine learning could contribute to improving classification and treatment. In this paper, we investigated literature focusing on the use of machine learning models on datasets with clinical variables of patients diagnosed with depression to predict treatment outcomes or find more homogeneous subgroups. Identified studies based on best practices in the field are evaluated. We found 16 studies predicting outcomes (such as remission) and identifying clusters in patients with depression. The identified studies are mostly still in proof-of-concept phase, with small datasets, lack of external validation, and providing single performance metrics. Larger datasets, and models with similar variables present across these datasets, are needed to develop accurate and generalizable models. We hypothesize that harnessing natural language processing to obtain data 'hidden' in clinical texts might prove useful in improving prediction models. Besides, researchers will need to focus on the conditions to feasibly implement these models to support psychiatrists and patients in their decision-making in practice. Only then we can enter the realm of precision psychiatry.

Prognosis and Improved Outcomes in Major Depression: A Review

(Reprinted from Transl Psychiatry. 2019 Apr 3; 9(1):127. Open access; is licensed under a Creative Commons Attribution 4.0 International License).

The identification, assessment and management of difficult-to-treat depression: An international consensus statement

BACKGROUND

Many depressed patients are not able to achieve or sustain symptom remission despite serial treatment trials - often termed "treatment resistant depression". A broader, perhaps more empathic concept of "difficult-to-treat depression" (DTD) was considered.

METHODS

A consensus group discussed the definition, clinical recognition, assessment and management implications of the DTD heuristic.

RESULTS

The group proposed that DTD be defined as "depression that continues to cause significant burden despite usual treatment efforts". All depression management should include a thorough initial assessment. When DTD is recognized, a regular reassessment that employs a multi-dimensional framework to identify addressable barriers to successful treatment (including patient-, illness- and treatment-related factors) is advised, along with specific recommendations for addressing these factors. The emphasis of treatment, in the first instance, shifts from a goal of remission to optimal symptom control, daily psychosocial functional and quality of life, based on a patient-centred approach with shared decision-making to enhance the timely consideration of all treatment options (including pharmacotherapy, psychotherapy, neurostimulation, etc.) to optimize outcomes when sustained remission is elusive.

LIMITATIONS

The recommended definition and management of DTD is based largely on expert consensus. While DTD would seem to have clinical utility, its specificity and objectivity may be insufficient to define clinical populations for regulatory trial purposes, though DTD could define populations for service provision or phase 4 trials.

CONCLUSIONS

DTD provides a clinically useful conceptualization that implies a search for and remediation of specific patient-, illness- and treatment obstacles to optimizing outcomes of relevance to patients.

Personalized prediction of smartphone-based psychotherapeutic micro-intervention success using machine learning

BACKGROUND

Tailoring healthcare to patients' individual needs is a central goal of precision medicine. Combining smartphone-based interventions with machine learning approaches may help attaining this goal. The aim of our study was to explore the predictability of the success of smartphone-based psychotherapeutic micro-interventions in eliciting mood changes using machine learning.

METHODS

Participants conducted daily smartphone-based psychotherapeutic micro-interventions, guided by short video clips, for 13 consecutive days. Participants chose one of four intervention techniques used in psychotherapeutic approaches. Mood changes were assessed using the Multidimensional Mood State Questionnaire. Micro-intervention success was predicted using random forest (RF) tree-based mixed-effects logistic regression models. Data from 27 participants were used, totaling 324 micro-interventions, randomly split 100 times into training and test samples, using within-subject and between-subject sampling.

RESULTS

Mood improved from pre- to post-intervention in 137 sessions (initial success-rate: 42.3%). The RF approach resulted in predictions of micro-intervention success significantly better than the initial success-rate within and between subjects (positive predictive value: 0.732 (95%-CI: 0.607; 0.820) and 0.698 (95%-CI: 0.564; 0.805), respectively). Prediction quality was highest using the RF approach within subjects (rand accuracy: 0.75 (95%-CI: 0.641; 0.840), Matthew's correlation coefficient: 0.483 (95%-CI: 0.323; 0.723)).

LIMITATIONS

The RF approach does not allow firm conclusions about the exact contribution of each factor to the algorithm's predictions. We included a limited number of predictors and did not compare whether predictability differed between psychotherapeutic techniques.

CONCLUSIONS

Our findings may pave the way for translation and encourage scrutinizing personalized prediction in the psychotherapeutic context to improve treatment efficacy.

Analysis of Features Selected by a Deep Learning Model for Differential Treatment Selection in Depression

Background: Deep learning has utility in predicting differential antidepressant treatment response among patients with major depressive disorder, yet there remains a paucity of research describing how to interpret deep learning models in a clinically or etiologically meaningful way. In this paper, we describe methods for analyzing deep learning models of clinical and demographic psychiatric data, using our recent work on a deep learning model of STAR*D and CO-MED remission prediction.

Methods: Our deep learning analysis with STAR*D and CO-MED yielded four models that predicted response to the four treatments used across the two datasets. Here, we use classical statistics and simple data representations to improve interpretability of the features output by our deep learning model and provide finer grained understanding of their clinical and etiological significance. Specifically, we use representations derived from our model to yield features predicting both treatment non-response and differential treatment response to four standard antidepressants, and use linear regression and t-tests to address questions about the contribution of trauma, education, and somatic symptoms to our models.

Results: Traditional statistics were able to probe the input features of our deep learning models, reproducing results from previous research, while providing novel insights into depression causes and treatments. We found that specific features were predictive of treatment response, and were able to break these down by treatment and non-response categories; that specific trauma indices were differentially predictive of baseline depression severity; that somatic symptoms were significantly different between males and females, and that education and low income proved important psycho-social stressors associated with depression.

Conclusion: Traditional statistics can augment interpretation of deep learning models. Such interpretation can lend us new hypotheses about depression and contribute to building causal models of etiology and prognosis. We discuss dataset-specific effects and ideal clinical samples for machine learning analysis aimed at improving tools to assist in optimizing treatment.

Predictors of Treatment Resistance Across Different Clinical Subtypes of Depression: Comparison of Unipolar vs. Bipolar Cases

OBJECTIVE

Treatment-resistant depression (TRD) and treatment-resistant bipolar depression (TRBD) poses a significant clinical and societal burden, relying on different operational definitions and treatment approaches. The detection of clinical predictors of resistance is elusive, soliciting clinical subtyping of the depressive episodes, which represents the goal of the present study.

METHODS

A hundred and thirty-one depressed outpatients underwent psychopathological evaluation using major rating tools, including the Hamilton Rating Scale for Depression, which served for subsequent principal component analysis, followed-up by cluster analysis, with the ultimate goal to fetch different clinical subtypes of depression.

RESULTS

The cluster analysis identified two clinically interpretable, yet distinctive, groups among 53 bipolar (resistant cases = 15, or 28.3%) and 78 unipolar (resistant cases = 20, or 25.6%) patients. Among the MDD patients, cluster "1" included the following components: "Psychic symptoms, depressed mood, suicide, guilty, insomnia" and "genitourinary, gastrointestinal, weight loss, insight". Altogether, with broadly defined "mixed features," this latter cluster correctly predicted treatment outcome in 80.8% cases of MDD. The same "broadly-defined" mixed features of depression (namely, the standard Diagnostic and Statistical Manual for Mental Disorders, Fifth Edition-DSM-5-specifier plus increased energy, psychomotor activity, irritability) correctly classified 71.7% of BD cases, either as TRBD or not.

LIMITATIONS

Small sample size and high rate of comorbidity.

CONCLUSIONS

Although relying on different operational criteria and treatment history, TRD and TRBD seem to be consistently predicted by broadly defined mixed features among different clinical subtypes of depression, either unipolar or bipolar cases. If replicated by upcoming studies to encompass also biological and neuropsychological measures, the present study may aid in precision medicine and informed pharmacotherapy.

Can Machine Learning help us in dealing with treatment resistant depression? A review

BACKGROUND

About one third of patients treated with antidepressant do not show sufficient symptoms relief and up to 15% of patients remain symptomatic even after multiple trials are applied, configuring a state called treatment resistant depression (TRD). A clear definition of this state and the understanding of underlying mechanisms contributing to chronic disability caused by major depressive disorder is still unknown. Therefore, Machine Learning (ML) techniques emerged in the last years as interesting approaches to deal with such complex problems.

METHODS

We performed a bibliographic search on Pubmed, Google Scholar and Medline of clinical, imaging, genetic and EEG ML classification studies on treatment-responding depression and TRD as well as studies trying to predict response to a specific treatment in already established TRD. The inclusion criteria were met by eleven studies. Seven focused on the definition of predictors of TRD onset while four attempted to predict the response to specific treatments in TRD.

RESULTS

The results showed that it seems possible to classify between responders MDD and TRD with good accuracies based on clinical variables. Moreover, some studies reported the possibility of using EEG measures to predict response to different pharmacological and non-pharmacological treatments in established TRD.

LIMITATIONS

The definition of TRD, the selection of variables together with ML algorithms and pipelines varies across the studies, ultimately determining the unfeasibility to implement these models in clinical practice.

CONCLUSIONS

The findings suggest that ML could be a valid approach to increase our understanding of TRD and to better classify and stratify this disorder, which may ultimately help clinicians in the assessment of major depressive disorders.

Testing the Effectiveness of Implementing a Model of Mental Healthcare Involving Trained Lay Health Workers in Treating Major Mental Disorders Among Youth in a Conflict-Ridden, Low-Middle Income Environment: Part II Results

OBJECTIVES

To report the outcomes of young people (aged 14-30 years) treated for major mental disorders in a lay health worker (LHW) intervention model in a rural district of conflict-ridden Kashmir, India.

METHODS

Over a 12-month follow-up, LHWs collected data on symptoms, functioning, quality of life and disability, and patients' and families' service engagement and satisfaction.

RESULTS

Forty trained LHWs (18 males and 22 females) identified 262 individuals who met the criteria for a diagnosis of a major mental disorder, connected them with specialists for treatment initiation (within 14 days), and provided follow-up and support to patients and families. Significantly more patients (14-30 years) were identified during the 14 months of the project than those in all age groups in the preceding 2 years. At 12 months, 205 patients (78%) remained engaged with the service and perceived it as very helpful. Repeated measures ANOVA showed significant improvements in scores on the global assessment of functioning (GAF) scale (F[df, 3.449] = 104.729, p < 0.001) and all 4 domains of the World Health Organization quality of life (WHOQOL) brief version (WHOQOL-BREF) of the survey-Physical F(df, 1.861) = 40.82; Psychological F(df, 1.845) = 55.490; Social F(df, 1.583) = 25.189; Environment F(df, 1.791) = 40.902, all ps < 0.001-and a decrease in disability (F[df, 1.806] = 4.364, p = 0.016). An interaction effect between time and sex was observed for the physical health domain of the WHOQOL-BREF.

DISCUSSION AND CONCLUSIONS

Our results show that an LHW-based service model, implemented in a rural setting of a low-to-middle income region plagued by long-term conflict, benefits young people with major mental disorders. We discuss the implications of our findings in the context of similar environments and the challenges encountered.

Results of the European Group for the Study of Resistant Depression (GSRD) - basis for further research and clinical practice

Objectives: The overview outlines two decades of research from the European Group for the Study of Resistant Depression (GSRD) that fundamentally impacted evidence-based algorithms for diagnostics and psychopharmacotherapy of treatment-resistant depression (TRD). Methods: The GSRD staging model characterising response, non-response and resistance to antidepressant (AD) treatment was applied to 2762 patients in eight European countries. Results: In case of non-response, dose escalation and switching between different AD classes did not show superiority over continuation of original AD treatment. Predictors for TRD were symptom severity, duration of the current major depressive episode (MDE), suicidality, psychotic and melancholic features, comorbid anxiety and personality disorders, add-on treatment, non-response to the first AD, adverse effects, high occupational level, recurrent disease course, previous hospitalisations, positive family history of MDD, early age of onset and novel associations of single nucleoid polymorphisms (SNPs) within the PPP3CC, ST8SIA2, CHL1, GAP43 and ITGB3 genes and gene pathways associated with neuroplasticity, intracellular signalling and chromatin silencing. A prediction model reaching accuracy of above 0.7 highlighted symptom severity, suicidality, comorbid anxiety and lifetime MDEs as the most informative predictors for TRD. Applying machine-learning algorithms, a signature of three SNPs of the BDNF, PPP3CC and HTR2A genes and lacking melancholia predicted treatment response. Conclusions: The GSRD findings offer a unique and balanced perspective on TRD representing foundation for further research elaborating on specific clinical and genetic hypotheses and treatment strategies within appropriate study-designs, especially interaction-based models and randomized controlled trials.

The pulvinar nucleus and antidepressant treatment: dynamic modeling of antidepressant response and remission with ultra-high field functional MRI

Functional magnetic resonance imaging (fMRI) successfully disentangled neuronal pathophysiology of major depression (MD), but only a few fMRI studies have investigated correlates and predictors of remission. Moreover, most studies have used clinical outcome parameters from two time points, which do not optimally depict differential response times. Therefore, we aimed to detect neuronal correlates of response and remission in an antidepressant treatment study with 7 T fMRI, potentially harnessing advances in detection power and spatial specificity. Moreover, we modeled outcome parameters from multiple study visits during a 12-week antidepressant fMRI study in 26 acute (aMD) patients compared to 36 stable remitted (rMD) patients and 33 healthy control subjects (HC). During an electrical painful stimulation task, significantly higher baseline activity in aMD compared to HC and rMD in the medial thalamic nuclei of the pulvinar was detected (p = 0.004, FWE-corrected), which was reduced by treatment. Moreover, clinical response followed a sigmoid function with a plateau phase in the beginning, a rapid decline and a further plateau at treatment end. By modeling the dynamic speed of response with fMRI-data, perigenual anterior cingulate activity after treatment was significantly associated with antidepressant response (p < 0.001, FWE-corrected). Temporoparietal junction (TPJ) baseline activity significantly predicted non-remission after 2 antidepressant trials (p = 0.005, FWE-corrected). The results underline the importance of the medial thalamus, attention networks in MD and antidepressant treatment. Moreover, by using a sigmoid model, this study provides a novel method to analyze the dynamic nature of response and remission for future trials.

CA, Kasper S. Prognosis and improved outcomes in major depression: a review

Treatment outcomes for major depressive disorder (MDD) need to be improved. Presently, no clinically relevant tools have been established for stratifying subgroups or predicting outcomes. This literature review sought to investigate factors closely linked to outcome and summarize existing and novel strategies for improvement. The results show that early recognition and treatment are crucial, as duration of untreated depression correlates with worse outcomes. Early improvement is associated with response and remission, while comorbidities prolong course of illness. Potential biomarkers have been explored, including hippocampal volumes, neuronal activity of the anterior cingulate cortex, and levels of brain-derived neurotrophic factor (BDNF) and central and peripheral inflammatory markers (e.g., translocator protein (TSPO), interleukin-6 (IL-6), C-reactive protein (CRP), tumor necrosis factor alpha (TNFα)). However, their integration into routine clinical care has not yet been fully elucidated, and more research is needed in this regard. Genetic findings suggest that testing for CYP450 isoenzyme activity may improve treatment outcomes. Strategies such as managing risk factors, improving clinical trial methodology, and designing structured step-by-step treatments are also beneficial. Finally, drawing on existing guidelines, we outline a sequential treatment optimization paradigm for selecting first-, second-, and third-line treatments for acute and chronically ill patients. Well-established treatments such as electroconvulsive therapy (ECT) are clinically relevant for treatment-resistant populations, and novel transcranial stimulation methods such as theta-burst stimulation (TBS) and magnetic seizure therapy (MST) have shown promising results. Novel rapid-acting antidepressants, such as ketamine, may also constitute a paradigm shift in treatment optimization for MDD.

Clinical factors predicting treatment resistant depression: affirmative results from the European multicenter study

OBJECTIVES

Clinical variables were investigated in the 'treatment resistant depression (TRD)- III' sample to replicate earlier findings by the European research consortium 'Group for the Study of Resistant Depression' (GSRD) and enable cross-sample prediction of treatment outcome in TRD.

EXPERIMENTAL PROCEDURES

TRD was defined by a Montgomery and Åsberg Depression Rating Scale (MADRS) score ≥22 after at least two antidepressive trials. Response was defined by a decline in MADRS score by ≥50% and below a threshold of 22. Logistic regression was applied to replicate predictors for TRD among 16 clinical variables in 916 patients. Elastic net regression was applied for prediction of treatment outcome.

RESULTS

Symptom severity (odds ratio (OR) = 3.31), psychotic symptoms (OR = 2.52), suicidal risk (OR = 1.74), generalized anxiety disorder (OR = 1.68), inpatient status (OR = 1.65), higher number of antidepressants administered previously (OR = 1.23), and lifetime depressive episodes (OR = 1.15) as well as longer duration of the current episode (OR = 1.022) increased the risk of TRD. Prediction of TRD reached an accuracy of 0.86 in the independent validation set, TRD-I.

CONCLUSION

Symptom severity, suicidal risk, higher number of lifetime depressive episodes, and comorbid anxiety disorder were replicated as the most prominent risk factors for TRD. Significant predictors in TRD-III enabled robust prediction of treatment outcome in TRD-I.

International Union of Basic and Clinical Pharmacology CIV: The Neurobiology of Treatment-resistant Depression: From Antidepressant Classifications to Novel Pharmacological Targets

Major depressive disorder is one of the most prevalent and life-threatening forms of mental illnesses and a major cause of morbidity worldwide. Currently available antidepressants are effective for most patients, although around 30% are considered treatment resistant (TRD), a condition that is associated with a significant impairment of cognitive function and poor quality of life. In this respect, the identification of the molecular mechanisms contributing to TRD represents an essential step for the design of novel and more efficacious drugs able to modify the clinical course of this disorder and increase remission rates in clinical practice. New insights into the neurobiology of TRD have shed light on the role of a number of different mechanisms, including the glutamatergic system, immune/inflammatory systems, neurotrophin function, and epigenetics. Advances in drug discovery processes in TRD have also influenced the classification of antidepressant drugs and novel classifications are available, such as the neuroscience-based nomenclature that can incorporate such advances in drug development for TRD. This review aims to provide an up-to-date description of key mechanisms in TRD and describe current therapeutic strategies for TRD before examining novel approaches that may ultimately address important neurobiological mechanisms not targeted by currently available antidepressants. All in all, we suggest that drug targeting different neurobiological systems should be able to restore normal function but must also promote resilience to reduce the long-term vulnerability to recurrent depressive episodes.

An augmentation study of MSI-195 (S-adenosylmethionine) in Major Depressive Disorder

We conducted a 6-week double-blind, placebo-controlled, augmentation study comparing the efficacy and safety of MSI-195 800 mg (a proprietary formulation of S-adenosylmethionine) or placebo added to ongoing antidepressant medication (ADT) in acutely depressed subjects with Major Depressive Disorder (MDD) who had experienced an inadequate response to their ongoing ADT (The Horizon Study, ClinicalTrials.gov NCT01912196). There were 234 eligible subjects randomized to either MSI-195 (n = 118) or placebo (n = 116). There were no overall statistically significant differences found between MSI-195 added to ongoing ADT compared to placebo on any of 3 depression-rating instruments (HamD₁₇, MADRS, IDS-SR₃₀) in the ITT set. MSI-195 was generally safe and well tolerated with predominantly mild gastrointestinal side effects. Post-hoc analyses examined factors that might have affected study outcome. The ITT set was divided into subjects enrolled during the 1st half (first nine months) and 2nd half of the study. MSI-195 added to ongoing ADT was significantly better than placebo on both the HamD₁₇ and MADRS in the 1st half (p = 0.03 and 0.02 respectively), but not in the 2nd half of the study. Several demographic and clinical characteristics were significantly different between the two study segments including body mass index, pre-randomization symptom severity fluctuation, number of lifetime depressive episodes, and anxious depression sub-type. Thus, the characteristics of the enrolled subjects changed between the 1st and 2nd half of the study. These post-hoc findings highlight the inherent challenges encountered for subject selection in double-blind, placebo controlled trials and compel further investigation of enrollment criteria and moderating factors that affect treatment. The favorable safety profile and clinical benefit observed with MSI-195 in the 1st half of this study warrant further investigation in MDD.