Hello, today I'm going to talk to you about serial connectome mapping of domain general higher-order cognition in neurosurgery. To begin with, I want to tell you about the human problem our patients with tumors face. Patients with brain tumors, up to 50% of them experience memory problems, one in four experience cognitive problems, and in the UK over 5,000 deaths are annually attributed to brain tumors. When we think about the clinical problem, low-grade gliomas are highly infiltrated brain cancers that integrate within the brain's healthy tissue. Unlike strokes, which acutely damage the brain, gliomas dynamically remodel it to minimize disruption. Thus, the purpose of surgery is to maintain this oncofunctional balance that is maximize tumor removal and minimize cognitive functional impairment. Having said that, clinicians need a common neurobiological framework to think about cognition. Whether you're offering radiation, surgery, chemo, or physiotherapy, we all need to be thinking about cognition across our subspecialties. This is where the connectome comes in. The connectome is a brain structural and functional wiring diagram. The best way to think about this is the human connectome as a highway, where the highways themselves represent the white matter and the traffic on these highways, or the traffic on the white matter, represents the functional connectivity. So from taking this raw high-dimensional neurodata, we can perform various analyses to model the brain as a graph and relate this to cognitive outcomes. In my work, I took the experimental approach of harnessing large healthy data sets from our university and coupling that with a small but deeply sampled clinical trial of our patients undergoing glioma surgery. In this trial, we acquired connectomic and cognitive we've performed cognitive testing at multiple time points that including pre-surgery, post-surgery day one, post-op month three, and post-op month twelve. Today I want to share with you findings for both the healthy individuals and also our surgical relevance. So when we look across 1,200 functional MRI studies, whether a healthy individual is performing an attention task, a language task, or a working memory task, a consistent set of frontal parietal brain regions are consistently activated. And this is the frontal parietal network that the field of connectomics utilizes. So when we analyze the structural integrity of this frontal parietal system across the lifespan, beyond the effect of age and sex, the frontal parietal cortical thickness directly correlates with executive function performance. When analyzing what kind of cells are present in the frontal parietal system, quite fascinatingly oligodendrocyte precursor cells, not oligodendrocytes themselves, preferentially populate this frontal parietal system that coordinates cognition. So that led us to studying in our clinical cohorts when taking 210 patients from the TCGA bank and some of our internal imaging, we found that using spatial autocorrelation analysis that low-grade gliomas significantly and preferentially infiltrate the frontal parietal system and brain areas that are preferentially enriched with oligodendrocyte precursor cells. In our clinical trial, we found that frontal parietal activity decreases immediately post-surgery and recalibrates at month 3 and this led us to believe that the optimal time window for rehabilitation is between the post-operative and month 3 period. And finally, it seems perioperative modularity, a network measure for the brain's integrative and segregative capacity, it seems that when the modularity decreases immediately post-surgery, patients have long-term cognitive deficits at the 12-month mark. Whereas when modularity doesn't decrease, patients end up having perfectly normal healthy cognitive improvements. And this makes sense and is consistent with not damaging major associative tracts and maybe projection pathways in the brain that would minimize the oncofunctional imbalance as earlier described. So what are the implications of this work, especially for cognitive rehab? Well, this work demonstrates that the early post-operative period is the optimal time window for aggressive cognitive rehab and that perioperative modularity can serve as an indicator for long-term cognitive trajectories. And so this can lead us to more systematic approaches to achieving supermaximal surgery and increasing progression-free survival and ultimately thinking about how can we intervene on the post-operative brain to accelerate cognitive outcomes. I want to thank my supervisors, Professor John Suckling and Dr. Rafa Romero-Garcia. I also want to thank my funding buddies, the Allenteering Institute and the Brain Tumor Charity. And I also want to thank the rest of my team for making this work possible. Thank you.

serial connectome mapping

higher-order cognition

neurosurgery

brain tumors

cognitive function