“Switchboard” malfunction in motor neuron diseases: Selective pathology of thalamic nuclei in amyotrophic lateral sclerosis and primary lateral sclerosis

Abstract

The thalamus is a key cerebral hub relaying a multitude of corticoefferent and corticoafferent connections and mediating distinct extrapyramidal, sensory, cognitive and behavioural functions. While the thalamus consists of dozens of anatomically well-defined nuclei with distinctive physiological roles, existing imaging studies in motor neuron diseases typically evaluate the thalamus as a single structure. Based on the unique cortical signatures observed in ALS and PLS, we hypothesised that similarly focal thalamic involvement may be observed if the nuclei are individually evaluated. A prospective imaging study was undertaken with 100 patients with ALS, 33 patients with PLS and 117 healthy controls to characterise the integrity of thalamic nuclei. ALS patients were further stratified for the presence of GGGGCC hexanucleotide repeat expansions in C9orf72. The thalamus was segmented into individual nuclei to examine their volumetric profile. Additionally, thalamic shape deformations were evaluated by vertex analyses and focal density alterations were examined by region-of-interest morphometry. Our data indicate that C9orf72 negative ALS patients and PLS patients exhibit ventral lateral and ventral anterior involvement, consistent with the ‘motor’ thalamus. Degeneration of the sensory nuclei was also detected in C9orf72 negative ALS and PLS. Both ALS groups and the PLS cohort showed focal changes in the mediodorsal-paratenial-reuniens nuclei, which mediate memory and executive functions. PLS patients exhibited distinctive thalamic changes with marked pulvinar and lateral geniculate atrophy compared to both controls and C9orf72 negative ALS. The considerable ventral lateral and ventral anterior pathology detected in both ALS and PLS support the emerging literature of extrapyramidal dysfunction in MND. The involvement of sensory nuclei is consistent with sporadic reports of sensory impairment in MND. The unique thalamic signature of PLS is in line with the distinctive clinical features of the phenotype. Our data confirm phenotype-specific patterns of thalamus involvement in motor neuron diseases with the preferential involvement of nuclei mediating motor and cognitive functions. Given the selective involvement of thalamic nuclei in ALS and PLS, future biomarker and natural history studies in MND should evaluate individual thalamic regions instead overall thalamic changes.