Dataset Information

Predictive modeling of treatment resistant depression using data from STAR*D and an independent clinical study.

ABSTRACT: Identification of risk factors of treatment resistance may be useful to guide treatment selection, avoid inefficient trial-and-error, and improve major depressive disorder (MDD) care. We extended the work in predictive modeling of treatment resistant depression (TRD) via partition of the data from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) cohort into a training and a testing dataset. We also included data from a small yet completely independent cohort RIS-INT-93 as an external test dataset. We used features from enrollment and level 1 treatment (up to week 2 response only) of STAR*D to explore the feature space comprehensively and applied machine learning methods to model TRD outcome at level 2. For TRD defined using QIDS-C16 remission criteria, multiple machine learning models were internally cross-validated in the STAR*D training dataset and externally validated in both the STAR*D testing dataset and RIS-INT-93 independent dataset with an area under the receiver operating characteristic curve (AUC) of 0.70-0.78 and 0.72-0.77, respectively. The upper bound for the AUC achievable with the full set of features could be as high as 0.78 in the STAR*D testing dataset. Model developed using top 30 features identified using feature selection technique (k-means clustering followed by ?2 test) achieved an AUC of 0.77 in the STAR*D testing dataset. In addition, the model developed using overlapping features between STAR*D and RIS-INT-93, achieved an AUC of > 0.70 in both the STAR*D testing and RIS-INT-93 datasets. Among all the features explored in STAR*D and RIS-INT-93 datasets, the most important feature was early or initial treatment response or symptom severity at week 2. These results indicate that prediction of TRD prior to undergoing a second round of antidepressant treatment could be feasible even in the absence of biomarker data.

SUBMITTER: Nie Z

PROVIDER: S-EPMC5991746 | biostudies-literature | 2018

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predictive modeling of treatment resistant depression using data from STAR*D and an independent clinical study.

Nie Zhi Z Vairavan Srinivasan S Narayan Vaibhav A VA Ye Jieping J Li Qingqin S QS

PloS one 20180607 6

Identification of risk factors of treatment resistance may be useful to guide treatment selection, avoid inefficient trial-and-error, and improve major depressive disorder (MDD) care. We extended the work in predictive modeling of treatment resistant depression (TRD) via partition of the data from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) cohort into a training and a testing dataset. We also included data from a small yet completely independent cohort RIS-INT-93 as an e ...[more]

PMID: 29879133

Similar Datasets

Project description:Objective: To describe the clinical characteristics of adolescents with antidepressant treatment-resistant major depressive disorder (MDD) and to examine the utility of the Antidepressant Treatment Record (ATR) in categorizing treatment resistance in this population. Methods: Adolescents with treatment-resistant MDD enrolled in an interventional study underwent a baseline evaluation with the ATR, Children's Depression Rating Scale-Revised (CDRS-R), and Clinical Global Impressions-Severity (CGI-S) scales. Demographic and clinical characteristics were examined with regard to ATR-defined level of resistance (level 1 to ≥3) using analysis of variance and χ2 tests. Results: In adolescents with treatment-resistant MDD (N = 97), aged 12-21 years, most were female (65%), white (89%), and had recurrent illness (78%). Patients were severely ill (median CGI-S score of 5), had a mean CDRS-R score of 63 ± 10, and 17.5% had been hospitalized for depression-related symptoms. Fifty-two patients were classified as ATR 1, whereas 32 were classified as ATR level 2 and 13 patients as ≥3, respectively. For increasing ATR-defined levels, illness duration increased from 12.0 (range: 1.5-31.9) to 14.8 (range: 1.8-31.7) to 19.5 (range: 2.5-36.2) months and the likelihood of treatment with serotonin norepinephrine reuptake inhibitors (SNRIs) and dopamine norepinephrine reuptake inhibitors (DNRIs) similarly increased (p = 0.006 for both SNRIs and DNRIs) as did the likelihood of treatment with mixed dopamine serotonin receptor antagonists (χ2 = 17, p < 0.001). Conclusions: This study underscores the morbidity and chronicity of treatment-resistant MDD in adolescents. The present characterization of related clinical features describes the use of nonselective serotonin reuptake inhibitors in adolescents with treatment-resistant depression and raises the possibility that those with the greatest medication treatment resistance are less likely to have had recurrent episodes. The study also demonstrates the utility of the ATR in categorizing treatment resistance in adolescents with MDD.

Project description:OBJECTIVE:The N-methyl-D-aspartate receptor antagonist ketamine has rapid antidepressant effects in treatment-resistant major depressive disorder (MDD) and bipolar depression. Clinical predictors may identify those more likely to benefit from ketamine within clinically heterogeneous populations. METHOD:Data were analyzed from 4 studies of treatment-resistant inpatients with DSM-IV-TR-diagnosed MDD or bipolar I or II depression. Patients who were currently experiencing a moderate-to-severe major depressive episode were enrolled between November 2004 and March 2013. All subjects received a single subanesthetic (0.5 mg/kg) ketamine infusion over 40 minutes. Patients were analyzed at the 230-minute postinfusion time point (n = 108), at day 1 (n = 82), and at day 7 (n = 71). Univariate Pearson correlations were performed for each variable with percent change from baseline in the 17-item Hamilton Depression Rating Scale (HDRS). Multivariate linear regression was then conducted for statistically significant predictors (P ? .05, 2-tailed). RESULTS:Higher body mass index correlated with greater HDRS improvement at 230 minutes (standardized ? = -0.30, P = .004) and at day 1 (standardized ? = -0.37, P = .001), but not at day 7 (standardized ? = -0.18, P = .10). Family history of an alcohol use disorder in a first-degree relative was associated with greater HDRS improvement at day 1 (standardized ? = -0.27, P = .014) and day 7 (standardized ? = -0.41, P < .001). No prior history of suicide attempt(s) was associated with greater improvement only at day 7 (standardized ? = 0.28, P = .01). The overall statistical model explained 13%, 23%, and 36% of HDRS percent change variance at 230 minutes, day 1, and day 7, respectively. CONCLUSIONS:Despite its post hoc nature, this study identified several clinical correlates of ketamine's rapid and durable antidepressant effects. Further investigation of these relationships is critical for individualized treatment of depression.

Dataset Information

Predictive modeling of treatment resistant depression using data from STAR*D and an independent clinical study.

Publications

Predictive modeling of treatment resistant depression using data from STAR*D and an independent clinical study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets