Diffusion tensor imaging evidence of corticospinal pathway involvement in frontotemporal lobar degeneration

Abstract

Motor neuron dysfunctions (MNDys) in Frontotemporal Lobar Degeneration (FTLD) have been consistently reported. Clinical and neurophysiological findings proved a variable range of pathological changes, also affecting the corticospinal tract (CST).

This study aims to assess white-matter microstructural alterations in a sample of patients with FTLD, and to evaluate the relationship with MNDys.

Fifty-four FTLD patients (21 bvFTD, 16 PPA, 17 CBS) and 36 healthy controls participated in a Diffusion Tensor Imaging (DTI) study. We analyzed distinctive and common microstructural alteration patterns across FTLD subtypes, including those affecting the CST, and performed an association analysis between CST integrity and the presence of clinical and/or neurophysiological signs of MNDys.

The majority of FTLD patients showed microstructural changes in the motor pathway with a high prevalence of CST alterations also in patients not displaying clinical and/or neurophysiological signs of MNDys.

Our results suggest that subtle CST alterations characterize FTLD patients regardless to the subtype. This may be due to the spread of the pathological process to the motor system, even without a clear clinical manifestation of MNDys.

Introduction

Frontotemporal lobar degeneration (FTLD) refers to a multisystem and heterogeneous neuropathological spectrum, encompassing conditions mainly affecting the anatomo-functional integrity of frontal and temporal regions (Seelaar et al., 2011, Seltman and Matthews, 2012), with an overall estimated prevalence of 10.84/100,000 (Coyle-Gilchrist et al., 2016). FTLD gives rise to different neurobehavioral syndromic expressions (e.g., Woollacott & Rohrer, 2016), involving behavior and social cognition, executive functions or language, for the most part included under the umbrella of frontotemporal dementia (FTD) – i.e., the behavioral variant of FTD (bvFTD) (Rascovsky et al., 2011), and the non-fluent (nfv-PPA) and semantic (sv-PPA) variants of primary progressive aphasia (PPA) (Gorno-Tempini et al., 2011). In addition, the corticobasal syndrome (CBS) a complex neurodegenerative syndromes characterised by the clinical combination of motor disorders and cognitive symptoms (Armstrong et al., 2013), has also been encompassed within the FTLD spectrum (Coyle-Gilchrist et al., 2016).

Noteworthy, clinical phenotypes depending on FTLD may share common genetic and pathological substrates with adult-onset motor neuron diseases (MNDs) (Davidson et al., 2016, Davidson et al., 2014, Ferrari et al., 2011, Ling et al., 2013), whose major subtype is amyotrophic lateral sclerosis (ALS). Consistently revealed in both FTLD and MND conditions, the TAR DNA binding protein (TDP-43) inclusions within the central nervous system, and mutations in the hexanucleotide (GGGGCC) repeat expansion of the chromosome 9 open reading frame 72 (C9orf72) gene are now considered the main neuropathological and genetic hallmarks joining these two disorders (Lattante et al., 2015, Ling et al., 2013, Weishaupt et al., 2016) into a unique disease spectrum, ranging from conditions in which cognitive and/or behavioural deficits are predominant to disorders primary affecting the motor pathway (Clark & Forman, 2006). Despite specific differences characterizing the clinical syndromes emerging from the FTLD-MND continuum, a frank overlap of signs and symptoms has been observed in a considerable proportion of cases (Devenney, Vucic, Hodges, & Kiernan, 2015). In particular, about 15% of patients clinically diagnosed as FTD develop motor neuron dysfunctions (MNDys) typical of MND and reaching the criteria for the diagnosis of ALS (Burrell et al., 2011, Lomen-Hoerth, 2011). Of note, about one-third of FTLD patients with subclinical (i.e., neurophysiological) or mild clinical signs of MNDys does not fulfil the clinical criteria for the diagnosis of MND/ALS (Burrell et al., 2011, Cerami et al., 2015).

Clinical and neurophysiological signs of MNDys in FTLD have been attributed to the pathological involvement of the primary motor cortex, whose involvement has been assumed to drive the secondary degeneration of lower motor neurons (Burrell et al., 2011). Neuropathological studies reported in bvFTD a time-sequential propagation of neurostructural impairments due to the dissemination of phosphorylated TDP-43 aggregates, starting from prefrontal cortex and then spreading caudally to involve sensorimotor cortices (Brettschneider et al., 2014). Neuroimaging findings provided support to this hypothesis, by showing the presence of neurostructural damages involving the premotor and motor cortices, as well as the corticospinal pathway, in bvFTD (Crespi et al., 2018, Lillo et al., 2012).

A number of investigations described the typical pattern of neurostructural alterations in different FTLD phenotypes (e.g., Agosta et al., 2015, Agosta et al., 2012, Omer et al., 2017) and reported evidence of subclinical motor pathway damage (e.g., Crespi et al., 2018, McMillan et al., 2016, Omer et al., 2017). However, to our knowledge no studies comprehensively investigated the degree and patterns of motor pathway changes within the FTLD spectrum and its relationship with the presence of clinical and/or subclinical MNDys.

Diffusion tensor imaging (DTI) technique appears particularly suitable to identify subtle neuronal changes due to pathological alterations in major white-matter bundles, including the corticospinal tract (CST), even when clinical signs are too subtle to be detected (e.g., Senda et al., 2011). Thus, DTI can be useful in identifying in vivo markers of changes in microstructural integrity in specific clinical populations, included FTLD (see Kassubek et al., 2018a).

On this ground, the aim of the present study is two-fold. First, we provided a detailed characterization of common and distinctive microstructural white-matter alterations occurring in the three main subtypes of FTLD – i.e., bvFTD, PPA and CBS – compared to healthy controls, trying to underlie common and possible distinctive alterations, focusing the attention to the CST pathway. Most importantly, we assessed the association of measures of microstructural integrity derived from DTI – i.e., fractional anisotropy (FA), mean (MD), axial (AD) and radial (RD) diffusivities – with clinical signs of MNDys. In a sub-group of patients, we additionally evaluated association between DTI metrics and neurophysiological (i.e., needle electromyography) signs of MNDys. Concerning this latter aspect, we hypothesized the presence of microstructural white-matter alteration in one or more DTI metrics along the CST even in patients with no clinical signs and, to a lesser extent, no neurophysiological evidence of MNDys.