Atomoxetine Treatment Strengthens an Anti-Correlated Relationship between Functional Brain Networks in Medication-Naïve Adults with Attention-Deficit Hyperactivity Disorder: A Randomized Double-Blind Placebo-Controlled Clinical Trial

Abstract

Background: Although atomoxetine demonstrates efficacy in individuals with attention-deficit hyperactivity disorder, its treatment effects on brain resting-state functional connectivity remain unknown. Therefore, we aimed to investigate major brain functional networks in medication-naïve adults with attention-deficit hyperactivity disorder and the efficacy of atomoxetine treatment on resting-state functional connectivity.

Methods: After collecting baseline resting-state functional MRI scans from 24 adults with attention-deficit hyperactivity disorder (aged 18-52 years) and 24 healthy controls (matched in demographic characteristics), the participants with attention-deficit hyperactivity disorder were randomly assigned to atomoxetine (n=12) and placebo (n=12) arms in an 8-week, double-blind, placebo-controlled trial. The primary outcome was functional connectivity assessed by a resting-state functional MRI. Seed-based functional connectivity was calculated and compared for the affective, attention, default, and cognitive control networks.

Results: At baseline, we found atypical cross talk between the default, cognitive control, and dorsal attention networks and hypoconnectivity within the dorsal attention and default networks in adults with attention-deficit hyperactivity disorder. Our first-ever placebo-controlled clinical trial incorporating resting-state functional MRI showed that treatment with atomoxetine strengthened an anticorrelated relationship between the default and task-positive networks and modulated all major brain networks. The strengthened anticorrelations were associated with improving clinical symptoms in the atomoxetine-treated adults.

Conclusions: Our results support the idea that atypical default mode network task-positive network interaction plays an important role in the pathophysiology of adult attention-deficit hyperactivity disorder. Strengthening this atypical relationship following atomoxetine treatment suggests an important pathway to treat attention-deficit hyperactivity disorder.

Trial registration: ClinicalTrials.gov NCT00917371.

Keywords: adult; atomoxetine; attention-deficit hyperactivity disorder; randomized double-blind placebo-controlled clinical trial; resting-state fMRI.

Figure 1.

Flow diagram of the procedure of the clinical trial.

Figure 2.

Differences in resting state functional connectivity of the major neural networks between controls and adults with attention-deficit hyperactivity disorder (ADHD) at baseline. Comparisons of the 2 groups demonstrated the controls had stronger positive connectivity in the (A) dorsal attention network between the left frontal eye field (FEF) and right fusiform/inferior temporal gyrus (ITG) and between the right FEF and right dorsolateral prefrontal cortex (DLPFC), and in the (B) default mode network (DMN) between the left precuneus (PRE) and right middle temporal gyrus (MTG). The control group also had greater negative connections in the (C) cognitive control network between the left DLPFC and PRE and between the right DLPFC and medial prefrontal cortex (mPFC). Adults with ADHD had stronger anticorrelations in the (D) dorsal attention network from the right FEF to left DLPFC and left MTG/angular gyrus. Statistical height threshold P<.01, FWE cluster-level corrected P<.05. The green dots represent the seed regions and the red dots indicate the regions showing atypical functional connectivity (peak coordinates). L, left; R, right.

Figure 3.

Connections demonstrating treatment × time interactions in the clinical trial. A mixed model for repeated measures revealed atomoxetine treatment modulated resting state functional connectivity across all the major neural networks investigated. Statistical height threshold P<.01, FWE cluster-level corrected P<.05. The green dots represent the seed regions and the red dots indicate the regions showing treatment × time interactions in the clinical trial (peak coordinates). The color (yellow areas with red edges) in the brain map displayed only the spatial extents of the clusters, but did not represent statistical values (see Table 4 for statistical values and functional connection strength). DLPFC,dorsolateral prefrontal cortex; FEF,frontal eye field; ITG,inferior temporal gyrus; L,left; MOG,middle occipital gyrus; mPFC,medial prefrontal cortex; MTG,middle temporal gyrus; OFC,orbitofrontal cortex; PCC,posterior cingulate cortex; PRE,precuneus; R,right; SubgeACC,subgenual anterior cingulate cortex; TPJ,temporoparietal junction.

Figure 4.

Functional connectivity changes with improvement in clinical symptoms and neuropsychological performances. Regions that showed significant (statistical height threshold P<.01, FWE cluster-level corrected P<.05) alterations in functional connectivity as symptoms and performances of Rapid Visual Information Processing (RVP) improved, in (A) ventral attention network, (B) cognitive control network, (C) dorsal attention network, and (D) default mode network (DMN). Yellow maps corresponded to positive associations, whereas blue maps represented negative associations. DLPFC,dorsolateral prefrontal cortex; FEF,frontal eye field; PCC,posterior cingulate cortex; PRE,precuneus; TPJ,temporo-parietal junction; VFC,ventral frontal cortex.