Development of cortical and subcortical asymmetry from mid gestation to old age

1252 

Abstract Submission 

Lena Dorfschmidt¹, Simon White², Jenna Schabdach³, Zhiqiang Sha¹, Margaret Gardner⁴, Shreya Gudapati1¹, Dabriel Zimmerman¹, Logan Williams⁵, Vanessa Kyriakopoulou⁵, Emma Robinson⁵, A. Edwards⁵, Lifespan Brain Chart Consortium¹, Edward Bullmore², Russell Shinohara⁴, Richard Bethlehem⁶, Jakob Seidlitz⁴, Aaron Alexander-Bloch⁴

¹University of Philadelphia, Philadelphia, PA, ²University of Cambridge, Cambridge, UK, ³Children's Hospital of Philadelphia, Philadelphia, PA, ⁴University of Pennsylvania, Philadelphia, PA, ⁵King's College London, London, UK, ⁶Autism Research Centre, Department of Psychiatry, University of Cambridge, Cambridge, UK

Lena Dorfschmidt  
University of Philadelphia
Philadelphia, PA

Simon White  
University of Cambridge
Cambridge, UK

Jenna Schabdach, PhD  
Children's Hospital of Philadelphia
Philadelphia, PA

Zhiqiang Sha  
University of Philadelphia
Philadelphia, PA

Margaret Gardner  
University of Pennsylvania
Philadelphia, PA

Shreya Gudapati1  
University of Philadelphia
Philadelphia, PA

Dabriel Zimmerman  
University of Philadelphia
Philadelphia, PA

Logan Williams  
King's College London
London, UK

Vanessa Kyriakopoulou, Dr  
King's College London
London, UK

Emma Robinson, Dr  
King's College London
London, UK

A. Edwards  
King's College London
London, UK

Lifespan Brain Chart Consortium  
University of Philadelphia
Philadelphia, PA

Edward Bullmore  
University of Cambridge
Cambridge, UK

Russell Shinohara  
University of Pennsylvania
Philadelphia, PA

Richard Bethlehem  
Autism Research Centre, Department of Psychiatry, University of Cambridge
Cambridge, UK

Jakob Seidlitz  
University of Pennsylvania
Philadelphia, PA

Aaron Alexander-Bloch  
University of Pennsylvania
Philadelphia, PA

Asymmetry is a key organizing principle of the brain, and has previously been shown to support healthy cognition [1], with alterations in asymmetry observed across neuropsychiatric disorders [2,3]. While asymmetry emerges during fetal development [4], its lifelong dynamics and variability remain unknown [4]. The recent development of normative reference charts for the human brain have enabled individuals to be benchmarked against population-level norms across the lifespan [5]. This study significantly extends this previous work by including regional (left/right) cortical and subcortical brain areas from over 100,000 participants, and comprehensively mapping left-right asymmetry trajectories across the human lifespan and between clinical cohorts.

We aggregated a dataset of 103 primary studies from 25 countries (Fig. 1A), comprising 129,963 scans from 103,013 subjects spanning 21 post conception weeks (PCW) to 102 years of age (Fig. 1B). We mapped lifespan trajectories of regional subcortical gray matter volume (GM), surface area (SA), cortical thickness (CT) and ventricular volume for 68 cortical regions of the Desikan-Killiany and 12 subcortical regions defined by FreeSurfer [6]. We fit generalized additive models for location, scale and shape (GAMLSS) [7] to estimate the nonlinear effects of age, stratified by sex, on all phenotypes, while accounting for study effects using random effects (Fig. 1C). Key developmental milestones were defined by deriving the peak and peak rate-of-change of the trajectories. Next, we derived an index of regional brain asymmetry in each phenotype as (L-R)/(L+R), where L and R are the respective values of the regional phenotype in the left and right hemisphere. Accordingly, an asymmetry score of > 0 indicates a left-dominant region, whereas a score < 0 indicates right-dominance. We then used similar GAMLSS methods for modeling lifespan trajectories of regional brain asymmetry, deriving (per)centile scores for each phenotype for each individual. Group effects across multiple brain disorders with more than 1000 subjects each were estimated using the (signed) Cohen's d of the pairwise difference versus controls. Lastly, we derived a map of hemispheric asymmetry in case-control differences for each disorder and phenotype as the difference between left and right case-control differences. We then computed the first principal component of the (disorder*phenotypes) x regions map as a measure of cross-disorder effects on brain asymmetry.

We found that L/R regional tissue trajectories largely peaked in (early) childhood and adolescence (Fig. 2A), with left-right age differences in hemispheric peaks of up to 3 years across phenotypes (Fig. 2B), in line with previous bilaterally-combined results [7]. Brain asymmetry trajectories across phenotypes and regions showed the greatest dynamics in the first two years of life, generally stabilizing thereafter (Fig. 2C). Overall, asymmetry was greater in GM and SA compared to CT. We found widespread case-control differences in centile scores across phenotypes and disorders, with the largest decreases in GM, SA and CT in AD and MCI, followed by SCZ. Lastly, we estimated a map of cross-disorder asymmetry effects (Fig. 2E). We found that for SA and GM, most disorders negatively loaded onto this component, indicating that the majority of cortical regions were associated with greater case-control differences in the right compared to the left hemisphere (Fig. 2E).

Our findings highlight changes in brain asymmetry over the course of the lifespan and underscore the potential clinical relevance of assessing brain asymmetry in the context of brain disorders across development and aging. The identification of specific deviations and their cross-disorder implications provides a foundation for future research and clinical applications in the realm of brain development and mental health.

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s)

Neurodevelopmental/ Early Life (eg. ADHD, autism)

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ²

Early life, Adolescence, Aging ¹

Psychiatric Disorders

Other - normative modeling, asymmetry subcortex