Modeling the heterogeneity in risk of progression to Alzheimer's disease acrosscognitive profiles in mild cognitive impairment

摘要

Introduction Heterogeneity in risk of conversion to Alzheimer's disease (AD) among individualswith mild cognitive impairment (MCI) is well known. Novel statistical methods thatare based on partially ordered set (poset) models can be used to create modelsthat provide detailed and accurate information about performance with specificcognitive functions. This approach allows for the study of direct links betweenspecific cognitive functions and risk of conversion to AD from MCI. It also allowsfor further delineation of multi-domain amnestic MCI, in relation to specificnon-amnestic cognitive deficits, and the modeling of a range of episodic memoryfunctioning levels. Methods From the Alzheimer's Disease Neuroimaging Initiative (ADNI) study, conversion at24 months of 268 MCI subjects was analyzed. It was found that 101 of thosesubjects (37.7%) converted to AD within that time frame. Poset models were thenused to classify cognitive performance for MCI subjects. Respective observedconversion rates to AD were calculated for various cognitive subgroups, and byAPOE e4 allele status. These rates were then compared across subgroups. Results The observed conversion rate for MCI subjects with a relatively lower functioningwith a high level of episodic memory at baseline was 61.2%. In MCI subjects whoadditionally also had relatively lower perceptual motor speed functioning and atleast one APOE e4 allele, the conversion rate was 84.2%. In contrast, the observedconversion rate was 9.8% for MCI subjects with a relatively higher episodic memoryfunctioning level and no APOE e4 allele. Relatively lower functioning withcognitive flexibility and perceptual motor speed by itself also appears to beassociated with higher conversion rates. Conclusions Among MCI subjects, specific baseline cognitive profiles that were derived throughposet modeling methods, are clearly associated with differential rates ofconversion to AD. More precise delineation of MCI by such cognitive functioningprofiles, including notions such as multidomain amnestic MCI, can help in gainingfurther insight into how heterogeneity arises in outcomes. Poset-based modelingmethods may be useful for providing more precise classification of cognitivesubgroups among MCI for imaging and genetics studies, and for developing moreefficient and focused cognitive test batteries.