Employing connectome-based models to predict working memory in multiple sclerosis

Abstract

Background: Individuals with multiple sclerosis (MS) are vulnerable to deficits in working memory, and the search for neural correlates of working memory in circumscribed areas has yielded inconclusive findings. Given the widespread neural alterations observed in MS, predictive modeling approaches that capitalize on whole-brain connectivity may better capture individual-level working memory in this population. Methods: Here, we applied connectome-based predictive modeling to functional MRI data from working memory tasks in two independent samples with relapsing-remitting MS. In the internal validation sample (ninternal = 36), functional connectivity data were used to train a model through cross-validation to predict accuracy on the Paced Visual Serial Addition Test, a gold-standard measure of working memory in MS. We then tested its ability to predict performance on the N-back working memory task in the external validation sample (nexternal = 36). Results: The resulting model successfully predicted working memory in the internal validation sample but did not extend to the external sample. We also tested the generalizability of an existing model of working memory derived in healthy young adults to people with MS. It showed successful prediction in both MS samples, demonstrating its translational potential. We qualitatively explored differences between the healthy and MS models in intra- and inter-network connectivity amongst canonical networks. Discussion: These findings suggest that connectome-based predictive models derived in people with MS may have limited generalizability. Instead, models identified in healthy individuals may offer superior generalizability to clinical samples, such as MS, and may serve as more useful targets for intervention.

TRACking health behaviors in people with Multiple Sclerosis (TRAC-MS): Study protocol and description of the study sample

“…As such, it is increasingly important to identify neuromarkers—brain-based signatures that reliably predict variance in specific cognitive domains—and rigorously test them using independent samples. Our prior work has established the validity of a working memory neuromarker in predicting working memory performance in PwMS [ 61 ]. Specifically, using connectome-based predictive modeling [ 62 , 63 ], we recently validated a neuromarker of working memory derived from 502 participants from the Human Connectome Project [ 64 ] to predict working memory in two independent samples of PwMS [ 61 ].…”

“…Our prior work has established the validity of a working memory neuromarker in predicting working memory performance in PwMS [ 61 ]. Specifically, using connectome-based predictive modeling [ 62 , 63 ], we recently validated a neuromarker of working memory derived from 502 participants from the Human Connectome Project [ 64 ] to predict working memory in two independent samples of PwMS [ 61 ]. Fig.…”

“…The resulting matrix will be comprised of 71,824 connections, and each cell of the matrix will represent the magnitude of the correlation between the respective functional regions. Using network masks validated in our prior work [ 61 ], we will then derive the network strength of the marker at pre-intervention and post-intervention. To examine whether changes in the network strength of the working memory connectome mediates the effect of the intervention on working memory we will construct three linear mixed models: 1) the impact of Group X Time on the primary outcome (see section 2.5.3); 2) the impact of Group on the slope of change in network scores; and 3) the impact of the network score slope in mediating the effect of Group on the dependent variable of interest.…”

Review of the Brain’s Behaviour after Injury and Disease for Its Application in an Agent-Based Model (ABM)

“…Functional connectivity remains a benchmark for studies about MS nonetheless, with special attention to network efficiency indicators [ 274 , 275 ], e.g., on working memory, subjected to patient heterogeneity [ 276 ]. Most observations have something in common: altered connectivity in deep grey matter areas, lower brain modularity, hemispheric skewness, and task-independency [ 277 ].…”

Review of the Brain Behaviour after Injury and Disease for Its Application in an Agent Based Model (ABM)

“…Functional connectivity remains a benchmark for studies about MS nonetheless, with special attention to network efficiency indicators [215,216] -e.g. on working memory, but subjected to patient heterogeneity [217] -with an apparent common ground altogether: altered connectivity in deep-gray matter areas, lower brain modularity, hemispheric skewness and task-independency [218]. Treatment strategies have rapidly improved in a paliative sense but keep failing to put a remedy to continuous neurodegeneration [219].…”