Llorente-Saguer, Isaac;
(2025)
Novel precision biomarker models applied to Alzheimer’s disease.
Doctoral thesis (Ph.D), UCL (University College London).
Preview |
Text
Llorente-Saguer_10214591_Thesis.pdf Download (12MB) | Preview |
Abstract
Biomarkers are essential for diagnosing, monitoring, and developing treatments for Alzheimer's disease (AD). However, conventional biomarkers for tau pathology (PET), brain atrophy (MRI), and amyloid beta peptides often rely on simple, predefined ratios that limit precision, stability, and statistical power, hindering clinical trial efficiency and research progress. This thesis addresses these challenges by introducing and validating BioDisCVR, a novel, modality-agnostic framework for the data-driven discovery of optimised biomarkers. The work first establishes the methodological fragility of the gold-standard Standardised Uptake Value Ratio (SUVR) in tau PET, exposing the instability of commonly used reference regions and the detrimental effects of certain processing techniques. It further reveals the suboptimal performance of traditional volumetric measures for tracking longitudinal change. One central innovation is the Composite Value Ratio (CVR), a data-driven biomarker construct where both numerator and denominator are optimised to maximise statistical power. Applied to tau PET and structural MRI, CVR demonstrates transformative improvements over established methods, with the potential to reduce clinical trial sample sizes by over 79%, and improve detection of pathological changes. The framework is extended to proteomics with a weighted CVR (wCVR), providing novel insights into the pathogenic contributions of different amyloid beta peptides. This culminates in theta, a parameter-free multidimensional ratio that achieved outstanding classification of familial AD mutations (AUC > 0.99), dramatically outperforming all conventional peptide ratios. Overall, this thesis delivers a new paradigm for biomarker discovery. It provides a validated framework and a suite of statistically superior biomarkers to accelerate clinical trials, enhance disease monitoring, and advance the fundamental understanding of Alzheimer's disease.
| Type: | Thesis (Doctoral) |
|---|---|
| Qualification: | Ph.D |
| Title: | Novel precision biomarker models applied to Alzheimer’s disease |
| Open access status: | An open access version is available from UCL Discovery |
| Language: | English |
| Additional information: | Copyright © The Author 2025. Original content in this thesis is licensed under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) Licence (https://creativecommons.org/licenses/by/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request. |
| Keywords: | Alzheimer, biomarkers, modelling, amyloid, tau, PET, MRI, atrophy, classification, monitoring, disease progression, clinical trials, machine learning |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10214591 |
Archive Staff Only
 |
View Item |
