OS14.4.A The Neuroplastic Potential of the Human Brain before and After Glioma Surgery: Towards “Interventional Neurorehabilitation

Abstract

INTRODUCTION: The human brain is a highly neuroplastic 'complex' network: it self-organises without a hard blueprint, adapts to evolving circumstances, and can withstand insults. However, similar to other naturally occurring networks, brain networks can only endure a finite amount of damage before cognitive processes are affected. In this study, we first sought to establish the brain networks governing domain-general cognition (DGC) in healthy individuals across the lifespan. We then sought to map, track, and potentially rehabilitate networks governing DGC through connectomics and non-invasive brain stimulation (NIBS) when damaged by low-grade gliomas (LGG) and surgical oncology. METHODS: Using MRI and cognitive data from n=629 individuals (aged 18-88, Female= 51%), we assessed the structural, functional, and topological relevance of the spatially-distributed multiple-demand (MD) system for DGC. Next, in n=17 patients undergoing glioma surgery, we longitudinally acquired connectomic and cognitive data at multiple time points: pre-surgery and post-surgery Day 1, Month 3, Month 12. In an independent cohort of n=34 patients, we sought to establish the safety profile for interventional neurorehabilitation : connectome-driven NIBS in the acute post-operative period to accelerate cognitive recovery. RESULTS: In healthy individuals, the MD system across multiple scales of biological organisation was positively associated with higher-order cognition (Catell's fluid intelligence). In our patients, pre-operative LGG infiltration into the structural MD system was negatively associated with the number of long-term cognitive deficits, suggesting a functional reorganisation. Mixed-effects modelling demonstrated the resilience of the functional MD system to infiltration and resection, while the early post-operative period was critical for effective neurorehabilitation. Graph analyses revealed increased perioperative modularity can distinguish patients with long-term cognitive improvements at one-year follow-up. Finally, NIBS within two weeks post-craniotomy had a 90% (n=31/34) recruitment rate into the trial. There were no seizures or serious complications due to NIBS in this patient population. Transient headaches and tingling were reported in a minority of patients. CONCLUSION: For the first time, we elucidate long-term cognitive and network trajectories following LGG surgery while establishing a positive safety-profile for NIBS in the acute post-operative period. We argue that mesoscale brain mapping serves as a robust biomarker for intervention-related plasticity for future clinical trials. While we performed these experiments in the context of neurosurgery, connectomics and NIBS could be adopted across diverse neuro-oncological care pathways (i.e. chemotherapy/radiation).