Integrated applications of MRI, 1H-MRS, DWI and Relaxometry in preclinical disease models: a multimodal characterization from preclinical models towards a translational understanding

Abstract This PhD thesis explores the integrated application of Magnetic Resonance–based techniques— Magnetic Resonance Imaging (MRI), Proton Magnetic Resonance Spectroscopy (¹H-MRS), Diffusion Weighted Imaging (DWI), and Relaxometry—in preclinical models of healthy aging, as well as neurodegenerative, psychiatric, and oncological disorders. The main objective is to provide an in vivo multimodal characterization of pathological processes and to support the translational bridge between preclinical research and clinical application. The thesis is organized into two complementary sections. Section I focuses on experimental studies conducted in animal models using advanced multimodal MRI protocols. Section II provides a translational perspective through critical reviews of clinical literature, highlighting the diagnostic and prognostic potential of MR techniques in humans. Section I opens with a theoretical and methodological framework describing the physical principles of MRI, acquisition parameters, sequence optimization, anesthetic protocols, and data analysis pipelines specifically adapted for high-field preclinical systems (7T). MRI is a non-invasive imaging modality that uses magnetic fields and radiofrequency pulses to assess brain morphology, detecting cortical or subcortical atrophy and ventricular enlargement; in oncology, it enables longitudinal monitoring of tumor growth and treatment response. MRS allows the quantification of metabolites such as N-acetylaspartate, choline, and myo-inositol, providing insights into neuronal damage, cellular proliferation, and gliosis. DWI enables the analysis of quantitative parameters such as Fractional Anisotropy (FA) and Mean Diffusivity (MD), which are essential for assessing white matter integrity. Relaxometry quantitatively measures Relaxation Times, such as T2, contributing to tissue characterization and the evaluation of microstructural alterations. The multimodal approach integrates anatomical, metabolic, microstructural, and other quantitative tissue parameters, in compliance with the ethical principles of the 3Rs (Replacement, Reduction, Refinement), through a non-invasive, in vivo, and longitudinal study design. The experimental studies address interconnected pathological conditions and objectives: characterization of preclinical disease models; evaluation of potential diagnostic or treatment response biomarkers (pharmacological or non-pharmacological); and investigation of risk or protective factors involved in the onset and progression of age-related diseases. Specifically, in Early-Life Stress (ELS) models, MRI revealed morphological alterations of the pituitary gland, suggesting that early stress events may program long-term neurobiological effects. In the modulation of Rho-GTPase signaling pathways, the effects of pharmacological treatments (Fasudil and CNF1) were evaluated in 5- and 19-month-old mice, demonstrating an age-dependent role of these proteins in microstructural organization and metabolic profiles. The role of oxidative stress was investigated using transgenic models overexpressing the MTH1 gene, revealing—through diffusion techniques—both acute and chronic paraquat-induced microstructural alterations, as well as neuroprotective mechanisms. In the psychiatric field, the Chronic Mild Stress (CMS) model in Wistar-Kyoto rats enabled the characterization of morphological, microstructural, and metabolic correlates of treatment-resistant depression. The efficacy of accelerated repetitive transcranial magnetic stimulation (arTMS) was longitudinally monitored using imaging biomarkers, demonstrating the value of MR techniques in therapeutic follow-up. In oncology, two triple-negative breast cancer models—one immunocompetent and one immunodeficient—were investigated using a multimodal approach to assess the response to metformin, D609, and their combination. Structural, diffusion, and metabolic parameters were analyzed as potential quantitative indicators of tumor progression, therapeutic efficacy, and immune system response. Furthermore, the thesis contributes to the CoMP-MRS project, aimed at standardizing preclinical ¹H-MRS protocols to optimize acquisitions in the hippocampus and striatum of mice and rats, enabling increasingly robust and reliable metabolic data while improving reproducibility and inter-laboratory transferability. Optimizing preclinical MRI protocols enhances the methodological potential of the technique and helps overcome technical and biological limitations, including long acquisition times, motion artifacts, anesthesia-related effects in animal models, spectral overlap in MRS, and the lack of standardized and reproducible protocols. Section II broadens the perspective to clinical practice, emphasizing the central role of MRI as a non-invasive multimodal diagnostic tool, particularly in the field of dementia. In this context, symptom overlap is common, and MRI-detectable alterations often precede clinical manifestation. The integration of structural MRI, Diffusion Tensor Imaging (DTI), and MRS proves crucial for the early diagnosis of Alzheimer’s disease and Mild Cognitive Impairment (MCI), as well as for differential diagnosis with Parkinson’s disease. Overall, the findings demonstrate that an integrated multimodal approach based on magnetic resonance represents a powerful tool for the in vivo characterization of pathological mechanisms and for the development and validation of translational biomarkers. By combining anatomical, metabolic, and microstructural parameters, this approach provides a more comprehensive understanding of disease processes and strengthens the bridge between preclinical research and clinical application, paving the way toward increasingly personalized medicine.