Mapping ADHD Heterogeneity and Biotypes through Topological Deviations in Morphometric Similarity Networks

Area of Science:

• Neuroscience
• Psychiatry
• Computational Biology

Background:

• Attention-deficit/hyperactivity disorder (ADHD) presents significant clinical heterogeneity.
• Identifying distinct ADHD subtypes is crucial for understanding its neurobiological underpinnings and developing targeted interventions.

Purpose of the Study:

• To investigate the utility of normative modeling of brain morphometry similarity networks for ADHD stratification.
• To delineate distinct ADHD biotypes based on topological network properties and examine their clinical, neurochemical, and functional correlates.

Main Methods:

• Constructed morphometric similarity networks from multisite neurodevelopmental datasets.
• Developed normative models for degree centrality, nodal efficiency, and participation coefficient.
• Employed semi-supervised clustering to identify ADHD biotypes and analyzed their clinical and neural profiles.

Main Results:

• ADHD exhibited atypical brain network hub organization, particularly in the orbitofrontal cortex.
• Three distinct ADHD biotypes were identified, each associated with specific symptom profiles (overall severity/emotional dysregulation, hyperactivity/impulsivity, inattention) and distinct topological alterations in brain regions like the medial prefrontal cortex, pallidum, and anterior cingulate cortex.
• Neural profiles of these biotypes demonstrated unique neurochemical and functional correlates, and findings were validated in an independent cohort.

Conclusions:

• The study successfully stratified ADHD into three distinct biotypes with unique clinical-neural signatures.
• This approach advances the understanding of ADHD's neurobiological heterogeneity.
• Findings provide a foundation for personalized treatment strategies in ADHD.