Pre-symptomatic radiological changes in frontotemporal dementia: propagation characteristics, predictive value and implications for clinical trials

Abstract

Computational imaging and quantitative biomarkers offer invaluable insights in the pre-symptomatic phase of neurodegenerative conditions several years before clinical manifestation. In recent years, there has been a focused effort to characterize pre-symptomatic cerebral changes in familial frontotemporal dementias using computational imaging. Accordingly, a systematic literature review was conducted of original articles investigating pre-symptomatic imaging changes in frontotemporal dementia focusing on study design, imaging modalities, data interpretation, control cohorts and key findings. The review is limited to the most common genotypes: chromosome 9 open reading frame 72 (C9orf72), progranulin (GRN), or microtubule-associated protein tau (MAPT) genotypes. Sixty-eight studies were identified with a median sample size of 15 (3–141) per genotype. Only a minority of studies were longitudinal (28%; 19/68) with a median follow-up of 2 (1–8) years. MRI (97%; 66/68) was the most common imaging modality, and primarily grey matter analyses were conducted (75%; 19/68). Some studies used multimodal analyses 44% (30/68). Genotype-associated imaging signatures are presented, innovative study designs are highlighted, common methodological shortcomings are discussed and lessons for future studies are outlined. Emerging academic observations have potential clinical implications for expediting the diagnosis, tracking disease progression and optimising the timing of pharmaceutical trials.

Thalamic pathology in frontotemporal dementia: Predilection for specific nuclei, phenotype‐specific signatures, clinical correlates, and practical relevance

“…This also supports observations of co‐occurring deficits in interlinked clinical domains (Grossman et al., 2008 ). Emerging evidence from presymptomatic studies confirms that pathological change accrues long before symptom onset (Bede et al., 2020 ; Bertrand et al., 2018 ; McKenna et al., 2022 ; Querin et al., 2019 ; Wen et al., 2019 ) indicating that neurodevelopmental factors may also be at play (Bede et al., 2020 ; Lulé et al., 2020 ). Clustering strategies on large admixed imaging datasets have revealed clinically and radiologically distinct subgroups.…”

“…Thalamic integrity may be evaluated with relative ease and a number of robust open‐source software libraries are available to retrieve a variety of thalamus metrics. The observation that in most FTD subtypes thalamic atrophy is an early feature (McKenna, Lope, et al., 2022 ) and may precede characteristic cortical atrophy provides a strong rationale for quantitative thalamus imaging in FTD. Total thalamus volume and the volumes of specific nuclei can be estimated from high‐resolution three‐dimensional T1‐weighted data (Iglesias et al., 2018 ), which are routinely acquired in clinical protocols as part of the diagnostic workup; therefore, there are no additional time or cost implications for acquiring raw data for post hoc thalamic analyses.…”

The Potential of Applied Brain Imaging in Research and Clinical Settings

“…Although neural plasticity and neurorehabilitation have been widely studied in several neurological and psychiatric conditions [4], adaptive mechanisms in neurodegenerative conditions such as motor neuron disorder (MND) are seldom evaluated [5]. Seminal papers have described pre-symptomatic changes in mutation carriers, revealed propagation patterns in neurodegenerative conditions, and confirmed treatment effects in clinical trials, thereby contributing many novel insights [6].…”

Editorial: The gap between academic advances and therapy development in motor neuron disease

“…Ultimately, the study of non-ALS motor neuron diseases or 'restricted phenotypes' offers important methodological and conceptual lessons and these patients may also serve as 'disease controls' for ALS studies. Similarly, there are ample methodological learning opportunities from other neurodegenerative conditions, especially FTLD; concepts, study-frameworks, multisite designs, data repositories, and so forth, which could be readily adopted in ALS/MND [61,62].…”