Prediction of mild cognitive impairment that evolves into Alzheimer's disease dementia within two years using a gene expression signature in blood: a pilot study.

BACKGROUND: The focus on Alzheimer's disease (AD) is shifting from dementia to the prodromal stage of the disorder, to a large extent due to increasing efforts in trying to develop disease modifying treatment for the disorder. For development of disease-modifying drugs, a reliable and accurate test for identification of mild cognitive impairment (MCI) due to AD is essential. OBJECTIVE: In the present study, MCI progressing to AD will be predicted using blood-based gene expression. MATERIAL AND METHODS: Gene expression analysis using qPCR was performed on blood RNA from a cohort of patients with amnestic MCI (aMCI; n = 66). Within the aMCI cohort, patients progressing to AD within 1 to 2 years were grouped as MCI converters (n = 34) and the patients remaining at the MCI stage after 2 years were grouped as stable MCI (n = 32). AD and control populations were also included in the study. RESULTS: Multivariate statistical method partial least square regression was used to develop predictive models which later were tested using leave-one-out cross validation. Gene expression signatures that identified aMCI subjects that progressed to AD within 2 years with a prediction accuracy of 74%-77% were identified for the complete dataset and subsets thereof. CONCLUSION: The present pilot study demonstrates for the first time that MCI that evolves into AD dementia within 2 years may be predicted by analyzing gene expression in blood. Further studies will be needed to validate this gene signature as a potential test for AD in the predementia stage.

A novel blood test for the early detection of Alzheimer's disease.

Despite a variety of testing approaches, it is often difficult to make an accurate diagnosis of Alzheimer's disease (AD), especially at an early stage of the disease. Diagnosis is based on clinical criteria as well as exclusion of other causes of dementia but a definitive diagnosis can only be made at autopsy. We have investigated the diagnostic value of a 96-gene expression array for detection of early AD. Gene expression analysis was performed on blood RNA from a cohort of 203 probable AD and 209 cognitively healthy age matched controls. A disease classification algorithm was developed on samples from 208 individuals (AD = 103; controls = 105) and was validated in two steps using an independent initial test set (n = 74; AD = 32; controls = 42) and another second test set (n = 130; AD = 68; controls = 62). In the initial analysis, diagnostic accuracy was 71.6 ± 10.3%, with sensitivity 71.9 ± 15.6% and specificity 71.4 ± 13.7%. Essentially the same level of agreement was achieved in the two independent test sets. High agreement (24/30; 80%) between algorithm prediction and subjects with available cerebrospinal fluid biomarker was found. Assuming a clinical accuracy of 80%, calculations indicate that the agreement with underlying true pathology is in the range 85%-90%. These findings suggest that the gene expression blood test can aid in the diagnosis of mild to moderate AD, but further studies are needed to confirm these findings.