Premium
P3‐092: Prediction of short‐term and long‐term cognitive decline in moderate Alzheimer's disease patients in the Oxford project to investigate memory and aging (optima) cohort using CSF biomarkers
Author(s) -
Seeburger Jeffrey L.,
Holder Daniel,
Oulhaj Abderrahim,
Wang Yue,
Simon Adam,
Dallob Aimee,
Snyder Karen,
Flynn Mary,
Chappell Derek,
Tanen Michael,
King Elizabeth,
Joachim Catharine,
Seabrook Guy,
Potter William,
Smith A. David
Publication year - 2008
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2008.05.1656
Subject(s) - cognitive decline , cohort , medicine , natural history , gerontology , cohort study , memory clinic , cognitive impairment , disease , psychology , oncology , demography , dementia , sociology
Background: Assessment of changes in the rate of cognitive decline of patients with AD is a primary objective of AD clinical trials. However, the heterogeneity in the presentation of the disease reduces the sensitivity of cognitive endpoints for evaluating drug efficacy. Consequently, the potential for CSF markers to predict the rate of cognitive decline was examined as a strategy for mitigating the effects of such variance. Methods: The expression of amyloid (A 40, A 42, sAPP , sAPP ) and tau (t-tau, p-tau-181) markers was measured in baseline visit CSF collected from patients with clinicallydiagnosed AD (n 15), pathologically-confirmed AD (n 29), and agematched controls (n 37) who were part of the OPTIMA natural history cohort. The relationship between CSF marker expression and decline in CAMCOG and MMSE scores over the subsequent year was explored using linear regression analysis, and the relationship to long-term ( 10 years) decline in CAMCOG scores explored using nonlinear mixed effects modeling. Results: The AD regression modeling suggested that high baseline t-tau (log2 pg/ml) was associated with more rapid decline in CAMCOG scores within the first year from baseline (%R 10, m [95%CI] -4.5 [-8.4, -0.7], p 0.023). The ratio of t-tau/A 42 was also associated with short-term CAMCOG decline (%R 21, m [95%CI] -5.7 [-8.8, -2.6], p 0.00067). Examination of other tau/amyloid ratios, and of MMSE decline, produced similar results. By comparison, common demographic variables, as well as CSF A 42/A 40, were not associated with decline in the first year. The nonlinear modeling suggested that high baseline tau was also associated with more rapid long-term CAMCOG decline. The CAMCOG half-life was reduced by approximately 50% for patients with baseline tau levels at the 97.5% quantile compared to those at the 2.5% quantile. Conclusions: These findings indicate that CSF biomarkers can account for some variance in the analysis of cognitive decline in AD and that tau/amyloid expression ratios are stronger predictors of short-term decline than the individual markers. This could be due to a more complete representation of the disease process (pathogenesis by amyloid and degeneration by tau) by the ratios. The tau marker alone, however, appears to be a strong predictor of both short-term and long-term decline.