Neuronal damage and cognitive impairment in Parkinson's disease: the relation among serum biomarkers, brain atrophy and cognitive performance

Parkinson’s disease (PD) is the second most widespread neurodegenerative pathology after Alzheimer’s one. During the early stages it presents highly heterogeneous features and, as time progresses, it leads patients to present symptoms of a not only motor nature. Among the main motor symptoms, it is possible to find cognitive impairment, apathy, depression, anxiety, impulse control disorders, sleep disturbance, fatigue, pain, visual hallucinations and autonomic dysfunction. At present, the mechanism by which these collateral symptoms can develop during the disease progression and/or what type of physiological factors they are linked to, is not yet well understood. The development of new magnetic resonance imaging techniques has had a great impact on the characterization of many neurodegenerative pathologies, including Parkinson’s disease. The objective of this thesis is to carry out a morphometric analysis on the patients' magnetic resonance images and to investigate the relationship between morphometric results, blood parameters and cognitive status. To do this, it was considered a cohort of 57 PD patients, part of which affected by mild cognitive impairment (MCI), while another part characterized by a normal cognitive state (NC). The structural images considered were acquired using high-field magnetic resonance imaging and the sequences used for the morphometry were mainly T1-weighted and FLAIR T2-weighted. The biomarkers considered were nuerofilament light chain (NFL), Glial fibrillary acidic protein (GFAP), phosporilated Tau 181 and the GFAP/NFL ratio, while the cognitive scales used were the Montreal Cognitive Assessment (MOCA) and the Mini Mental State Examination (MMSE). To obtain brain morphometrics both a surface-based/voxel-based hybrid model on FreeSurfer and a voxel-based model on SPM were used. The morphometry results of the two models were compared. From the statistical parameters of the morphometry, the brain volumes were extracted and, so, the correlations between those of MCI and those of NC patients were calculated. Then, the correlations between volumes and biomarkers were evaluated and comparisons on the values of the biomarkers, dividing the population by gender, cognitive state and score on the cognitive scales were carried out. The results showed that there is a significant linear correlation between FreeSurfer and SPM results for most of the cortical volume estimates, although a scaling factor is present, while the situation in the subcortex is much more heterogenous, presenting areas for which there is little or no correlation. The morphometrics obtained with Freesurfer are in line with the literature. Gray matter atrophy was found in the cingulate, entorhinal, fusiform, parahippocampal and temporal cortex, other than in the orbital regions and in the paracentral lobule. Trends were observed between volumes and blood biomarkers and statistical tests suggested links between biomarker values and demographic parameters like age, gender and cognitive status. However, applying the Bonferroni-Holm multiple testing correction method, the significance values found earlier did not survive. Although the study was limited by the relatively small size of the dataset available, these preliminary results highlight how this line of research could reveal new insights in the specificity of these biomarkers and further studies could lead to a clearer picture of the neurodegenerative phenomenon dictated by Parkinson’s disease in all its aspects.