Clinical, genetic and neuroimaging aspects of attention-deficit/hyperactivity disorder courses through midlife

Most follow-up studies have been characterizing the course of ADHD from childhood to young adulthood. However, there is a dearth of of knowledgment on adulthood ADHD trajectories and their predisposing factors. The aim of this study was to evaluate possible clinical, genetic and neuroimaging predictors of ADHD trajectories in midlife based on three assessments comprising 13 years of follow-up. A total of 323 subjects diagnosed with ADHD in adulthood were evaluated seven and thirteen years after the first assessment. Based on ADHD diagnostic status at the reassessments, midlife trajectories were characterized as: (i) stable if subjects fulfilled ADHD diagnostic criteria at baseline and at the reassessments; (ii) remission if subjects presented three or less symptoms of ADHD and lacked impairment criterion at both reassessments; and (iii) unstable in the remaining possibilities. Polygenic scores (PGS) were calculated for ADHD and for psychiatric phenotypes genetically correlated with ADHD (depression, schizophrenia and bipolar disorder). Neuroimaging liability scores were calculated for structural brain measures (cortical thickness, surface area, and subcortical volume). These scores represent the similarity to an ADHD brain and have the advantage of increasing statistical power by joining regional effects in the same way as PGS. Finally, functional connectivity between Default Mode Network (DMN) regions were acquired through resting-state fMRI. In the last assessment (13 years after baseline evaluation), mean age was 47 years and retention rate was 62%. Among the patients evaluated three times, 68.8% coursed a stable, 25.5% an unstable, and 5.7% a remission trajectory of ADHD. The remission trajectory was not included in the comparisons due to its low frequency. Women, individuals with more severe syndromes, with higher frequency of comorbidities and/or with higher genetic liability to depression presented a higher probability of a stable trajectory. ADHD, schizophrenia and bipolar disorder PGSs did not differ between stable and unstable trajectories. Stable and unstable trajectories did not differ regarding neuroimaging scores or functional connectivity within DMN nodes. This is the first study evaluating neurobiological aspects in midlife ADHD trajectories. We were able to detect demographic, clinical and genetic differences (for depression PGS) between trajectories. Structural and functional brain aspects might be related to the neurodevelopmental aspects of ADHD and its manifestation. On the other hand, more subtle differences such as fluctuation along lifespan could be associated with clinical and genetic aspects. Nonetheless, increasing sample size may also allow the detection of neuroimaging specificities for stable and unstable trajectory.