There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
| Principal Investigator | |
| Principal Investigator's Name: | Maryam Shoai |
| Institution: | UCL |
| Department: | Institute of Neurology |
| Country: | |
| Proposed Analysis: | Background and Rationale: Genome wide association analyses and exome sequencing have revolutionized our understanding of Alzheimer’s disease pathogenesis with the identification of microglial genes and lipid metabolism genes. However, the rate of decline in cases with disease is also extremely variable and this variability is poorly understood. The gene with the largest effect on risk, apolipoprotein E, has little or no effect on the rate of decline in Alzheimer cases, suggesting that the processes of disease initiation and progression are largely independent. Given that most clinical trials have been aimed at slowing the rate of decline in AD, identifying those who are genetically predisposed to being at a higher risk for accelerated progression is paramount to future successful clinical trial designs. We aim to elucidate the genes involved in processes that determine the rate of decline in AD (using any available cognitive measures). This will be aided by access to public data as well as clinical trial data and longitudinal data collected by drug companies as part of the placebo arms of their clinical trials. Scientific Hypothesis: It is hypothesised that a core set of genetic variants drive the rate of progression of Disease. GWAS of rate of decline/Survival analysis will elucidate variants which will affect the rate of decline in AD. This is sensitive to the quality of the phenotype employed. Given the high rate of misdiagnosis in AD, we will only use studies that have a marker of Aβ measured at baseline. Furthermore, the majority of the existing clinical trials and publicly available data do not possess enough power to detect rate of decline from small and medium sized effect variants and therefore it is essential to combine data from various clinical trials and publicly available data. To this end, we have approached several pharmaceutical companies, including; Biogen, Eisai, Eli Lily, Janssen, Merck, and Roche. All of the above companies are keen to share data for the purposes of this project as discussed with their senior management and scientists. Furthermore, we will have access to AIBL data and ELSA studies. We believe that the meticulous phenotyping of ADNI data will help us increase the power of the study but most importantly, shape the analysis and selection of the most informative cognitive measures for analysis. Brief Statistical analysis plan: Given the lack of studies looking into the role of genetic variation on the rate of decline in AD, our initial analysis will comprise both descriptive analyses (Survival function/ Kaplan-Meier curves) and time to event regressions. Modelling using mixed effects and principle component analysis will also be considered. We further hope to employ machine learning techniques such as survival random forests in order to identify the effects of various genes involved in rate of decline. Cross validated models will also be employed in order to ensure replicability. Any significant finding will be validated in smaller cohorts that we can access, or alternatively 20% of the total sample number. A more detailed breakdown of training and test set division will be finalised when we have samples confirmed. All analysis will be conducted by Dr M Shoai at UCL. Some guidance on statistical analysis will be given by Professor V Escott Price at Cardiff University (and Dementia Research Institute). Publication plan: Depending on the data received and findings of the study, a descriptive paper explaining the datasets will initially be published. All findings from machine learning and time to event analysis will be published separately. All collaborators will be co-authors on any publications produced. |
| Additional Investigators |
