Towards diagnostic efficacy of magnetic resonance imaging in neurodegenerative diseases

This doctoral thesis investigates biomarkers in neurodegenerative diseases using fully automatic pipelines based on magnetic resonance imaging (MRI) techniques.

Chapter 1 introduces the three MRI techniques and their applications: morphometry in amyotrophic lateral sclerosis (ALS) and Huntington’s disease (HD), diffusion tensor imaging (DTI) in Alzheimer’s disease (AD) and other forms of dementia, and quantitative susceptibility mapping (QSM) in ALS. It also outlines the translational research process, contextualized within international frameworks and the Italian research environment in which this work was conducted.

Chapter 2 focuses on a morphometric investigation of cortical markers of cognitive syndromes in non-demented ALS patients. While cortical thinning in the fronto-parietal network was present regardless of cognitive and behavioral status, a specific thinning in the inferior frontal, temporal, cingulate, and insular regions was shown to be a marker of cognitive and behavioral impairments in ALS.

Chapter 3 presents a multimodal study with morphometric analysis in premanifest and manifest HD subjects. The utility of quantitative brain MRI biomarkers for investigating neurodegeneration in the preclinical phase of HD was demonstrated, and posturography and arithmetic tests showed potential for detecting early preclinical changes.

Chapter 4 focuses on a DTI study of the superficial white matter (SWM) in AD and other neurodegenerative dementias. A correlation was found between SWM alterations and cerebrospinal fluid (CSF) biomarkers as well as cognitive decline in neurodegenerative dementias. Furthermore, parietal lobe alterations distinguished AD from other forms of dementia.

Chapter 5 reports on a QSM study of ALS patients, investigating a potential biomarker of upper motor neuron (UMN) impairment based on the magnetic susceptibility properties of the precentral cortex.

Chapter 6 expands on QSM findings by exploring magnetic susceptibility differences among ALS phenotypes and mimics, revealing phenotype-specific alterations.

Chapter 7 evaluates the diagnostic accuracy of visual qualitative susceptibility-weighted imaging (SWI) and automatic QSM-based quantification of susceptibility alterations in the precentral gyrus of ALS patients. Both SWI and QSM demonstrated high diagnostic accuracy in differentiating ALS with predominant UMN involvement from ALS mimics.

Chapter 8 integrates the results across diseases, discussing them within the framework of translational research. Barriers and facilitators to progressing toward clinical translation are analyzed, highlighting that while automatic MRI methods enable robust and non-invasive biomarkers, real-world validation is necessary to achieve clinical impact.

Future research should focus on validating these findings using real-world data, particularly the automatic quantitative assessment of magnetic susceptibility in the precentral gyrus for distinguishing ALS phenotype with predominant UMN impairment. These measures must be tested across larger cohorts and different MRI scanners to ensure reproducibility and clinical applicability.
The implementation of a translational team is proposed to systematically evaluate evidence, develop implementation strategies, and foster collaboration among researchers, clinicians, patients, and decision-makers. Such a multidisciplinary approach is essential to advance these imaging biomarkers toward clinical use for diagnosis, monitoring, and therapeutic trials in neurodegenerative diseases.