Premium
P1‐352: Plasma Aβ peptides and clinical variables as predictors of short‐ and long‐term response to rivastigmine in Alzheimer's disease
Author(s) -
Sobow Tomasz,
Flirski Marcin,
Liberski Pawel P.,
Kloszewska Iwona
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.2008.05.934
Subject(s) - rivastigmine , medicine , linear regression , disease , regression analysis , dementia , donepezil , statistics , mathematics
Background: Prediction of treatment response to cholinesterase inhibitors is currently an unachievable aim. Some clinical variables, including rate of progression and disease severity have been proposed but not verified in independent studies. Methods: Longitudinal clinical study of 54 carefully selected subjects (37 females) satisfying criteria for mild (N 25) or moderate (N 29) AD. Multiple linear regression was used to establish the predictive value of any baseline characteristics (including A levels and psychometric tests scores) and A levels change after rivastigmine treatment on the treatment effectiveness as measured by ADAS-cog score changes at months 6, 12 and 24. Results: There were no between-the-group differences (responders versus non-responders) in any demographic characteristics including gender, age, and education level. However, responders (at months 6, 12 and 24) had a considerably higher rate of disease progression, shorter disease duration, more advanced disease staging and a more pronounced change in A 42 plasma level following the testing dose of rivastigmine. All baseline characteristics already shown to be significantly different between responders and non-responders groups were further used as potential response predictors (independent variables) in a multiple linear regression model with the ADAS-cog difference at 12 and 24 months used as a dependent variable (model outcome). The initial ADAS-cog score was the only variable which correctly predicted a change in ADAS-cog score following 12 months of treatment, while the initial ADAScog score and a change in A 42 level in plasma after the testing dose of rivastigmine could predict ADAS-cog score change after 24 months of treatment. Finally, apolipoprotein E genotype did not influence response to treatment. Although 4genotype tended to be more prevalent in non-responders than 4 genotype at all time points, the differences did not reach statistical significance, probably because of a relatively small number of cases. Conclusions: Measurement of A 42 level in plasma after the testing dose of rivastigmine coupled with clinical data could be of help to predict treatment response in highly selected, pure AD subjects. Replication of our findings in an unselected AD patients groups is needed to establish real-life clinical utility of the presented results.