Multidimensional prediction of treatment response to antidepressants with cognitive control and functional MRI

Abstract

Predicting treatment response for major depressive disorder can provide a tremendous benefit for our overstretched health care system by reducing number of treatments and time to remission, thereby decreasing morbidity. The present study used neural and performance predictors during a cognitive control task to predict treatment response (% change in Hamilton Depression Rating Scale pre- to post-treatment). Forty-nine individuals diagnosed with major depressive disorder were enrolled with intent to treat in the open-label study; 36 completed treatment, had useable data, and were included in most data analyses. Participants included in the data analysis sample received treatment with escitalopram (n = 22) or duloxetine (n = 14) for 10 weeks. Functional MRI and performance during a Parametric Go/No-go test were used to predict per cent reduction in Hamilton Depression Rating Scale scores after treatment. Haemodynamic response function-based contrasts and task-related independent components analysis (subset of sample: n = 29) were predictors. Independent components analysis component beta weights and haemodynamic response function modelling activation during Commission errors in the rostral and dorsal anterior cingulate, mid-cingulate, dorsomedial prefrontal cortex, and lateral orbital frontal cortex predicted treatment response. In addition, more commission errors on the task predicted better treatment response. Together in a regression model, independent component analysis, haemodynamic response function-modelled, and performance measures predicted treatment response with 90% accuracy (compared to 74% accuracy with clinical features alone), with 84% accuracy in 5-fold, leave-one-out cross-validation. Convergence between performance markers and functional magnetic resonance imaging, including novel independent component analysis techniques, achieved high accuracy in prediction of treatment response for major depressive disorder. The strong link to a task paradigm provided by use of independent component analysis is a potential breakthrough that can inform ways in which prediction models can be integrated for use in clinical and experimental medicine studies.

Keywords: duloxetine; escitalopram; functional MRI; independent components analysis; major depressive disorder.

Figure 1

Illustration of the spatial extent of the key components involved in cognitive and inhibitory control. Components 24 and 25 were significantly related to commission errors on the PGNG task and also related to treatment response to escitalpram and duloxetine. Component 11 is composed of areas related to cognitive control, e.g. right lateralized DLPFC, IPL, and dorsal ACC.

Figure 2

Convergence of Component 24 across folds in the 5-fold leave-one-out cross-validation. (A) Areas of highest convergence in ICA estimated beta weights in relation to commission errors in the initial sample of 29 individuals completing treatment and five cross folds (37 − 7 or 8 randomly selected individuals, without replacement, for cross-validation). The 37 participants include those who did not complete treatment but had sufficient events for three runs to accommodate the ICA modelling in GIFT. (B) Areas of highest convergence in ICA estimated beta weights in relation to commission errors in the initial sample of 29 individuals completing treatment and five cross folds (37 − 7 or 8 randomly selected individuals, without replacement, for cross-validation). The 37 participants include those who did not complete treatment but had sufficient events for three runs to accommodate the ICA modelling in GIFT.

Figure 3

Exploratory analyses of differential prediction of treatment response by duloxetine and escitalopram of Component 25 in ICA modelled individuals during rejections. (A) Relations of treatment change with activation within the ICA (beta weights) for individuals for either escitalopram or duloxetine. (B) The full component (yellow) and the clusters within it that are predictive (red).

Figure 4

Areas of Components 11 (A), 24 (B), and 25 (C) predicting treatment response in second level models during Commissions in functional MRI completers. Results were deemed significant at P < 0.01, k = 23 using AlphaSim correction for multiple comparisons. x-, y-, and z-values represent the peak coordinate of the cluster displayed.

Figure 5

Poorer cognitive control performance (A), ICA 25 (B), and decreased mid cingulate activation (C) predicts better treatment response.