Abstracts

Rationale: Neuroimaging studies have consistently shown distributed brain anomalies in temporal lobe epilepsy with mesiotemporal sclerosis (Caciagli et al., 2014). Conversely, a system-level approach to focal cortical dysplasia (FCD) has been rarely considered. Given the variability in location and heterogeneity in histological traits (Iffland and Crino, 2017), we hypothesized that FCD connectivity is largely determined by the severity of the underlying structural anomalies. In light of recent evidence showing an impact of focal lesions on whole-brain systems (Aerts et al., 2016), we also postulated that patterns of lesional connectivity may exert differential downstream effects on global network topology. Methods: We studied 27 patients with histologically-verified FCD type-II and 34 healthy controls examined on 3T MRI. Our method is summarized in Fig 1. After subdividing every lesion into a set of similarly-sized cortical patches, we computed seed-based resting-state functional connectivity. Given the canonical community structure of human functional connectome (Yeo et al., 2011), we dichotomized connectivity profiles of lesional patches into those belonging to the same functional community as the lesion (i.e., intra-community) and other communities (i.e., inter-community). We then applied a data-driven hierarchical clustering on these community-based profiles to objectively identify classes with distinct functional connectivity. Identified lesion classes were profiled with respect to their patterns of functional connectivity, structural makeup and whole-brain network topology using graph-theory (Bullmore and Sporns, 2009). Finally, in an effort to provide an integrative mechanistic model, we employed multivariate structural equation analysis (Hair et al., 1998) to discover underlying latent factors linking FCD functional connectivity profiles, structural makeup and global network parameters. Results: Our connectome-based biotyping identified three distinct lesion classes (Fig 2A) with decreased intra- and inter-community connectivity (class 1), decreased intra-community, but normal inter-community connectivity (class 2), and increased intra- and inter-community connectivity (class 3). Classes also presented with differential patterns of structural signatures and global network topology (Fig 2A-B), namely marked cortical thickening and tissue blurring, yet mild network anomalies (class 1), moderately altered structural and network parameters (class 2), and subtle structural anomalies but severely disrupted networks (class 3). Finally, multivariate structural equation analyses provided a mechanistic model whereby the lesion structural makeup shapes its functional connectivity, which in turn modulates patterns of whole-brain network topology (Fig 2C). Conclusions: Patterns of functional connectivity of FCD Type-II are diverse and largely determined by the underlying lesional structural signature. Beyond a better understanding of structure-function relationship, a combined use of connectome and neuroimaging-based biotyping may optimize individualized diagnostics. Funding: Canadian Institutes of Health Research, Canadian League Against Epilepsy, Savoy Epilepsy Foundation