| 1. |
- Donohue, Michael C., et al.
(författare)
-
Cross-validation of optimized composites for preclinical Alzheimer's disease
- 2017
-
Ingår i: Alzheimer's and Dementia: Translational Research and Clinical Interventions. - : Wiley. - 2352-8737. ; 3:1, s. 123-129
-
Tidskriftsartikel (refereegranskat)abstract
- Introduction We discuss optimization and validation of composite end points for presymptomatic Alzheimer's disease clinical trials. Optimized composites offer hope of substantial gains in statistical power or reduction in sample size. But there is tradeoff between optimization and face validity such that optimization should only be considered if there is a convincing rationale. As with statistically derived regions of interest in neuroimaging, validation on independent data sets is essential. Methods Using four data sets, we consider the optimized weighting of four components of a cognitive composite which includes measures of (1) global cognition, (2) semantic memory, (3) episodic memory, and (4) executive function. Weights are optimized to either discriminate amyloid positivity or maximize power to detect a treatment effect in an amyloid-positive population. We apply repeated 5 × 3-fold cross-validation to quantify the out-of-sample performance of optimized composite end points. Results We found the optimized weights varied greatly across the folds of the cross-validation with either optimization method. Both optimization methods tend to down-weight the measures of global cognition and executive function. However, when these optimized composites were applied to the validation sets, they did not provide consistent improvements in power. In fact, overall, the optimized composites performed worse than those without optimization. Discussion We find that component weight optimization does not yield valid improvements in sensitivity of this composite to detect treatment effects.
|
|
| 2. |
- Insel, Philip S., et al.
(författare)
-
Cognitive and functional changes associated with Aβ pathology and the progression to mild cognitive impairment
- 2016
-
Ingår i: Neurobiology of Aging. - : Elsevier BV. - 0197-4580. ; 48, s. 172-181
-
Tidskriftsartikel (refereegranskat)abstract
- Cognitively-normal people with evidence of β-amyloid (Aβ) pathology and subtle cognitive dysfunction are believed to be at high risk for progression to mild cognitive impairment due to Alzheimer's disease (AD). Clinical trials in later stages of AD typically include a coprimary endpoint to demonstrate efficacy on both cognitive and functional assessments. Recent trials focus on cognitively-normal people, but functional decline has not been explored for trial designs in this group. The goal of this study was therefore to characterize cognitive and functional decline in (1) cognitively-normal people converting to mild cognitive impairment (MCI) and (2) cognitively-normal β-amyloid-positive (Aβ+) people. Specifically, we sought to identify and compare the cognitive and functional assessments and their weighted combinations that maximize the longitudinal decline specific to these 2 groups. We studied 68 people who converted from normal cognition to MCI and 70 nonconverters, as well as 137 Aβ+ and 210 β-amyloid-negative cognitively-normal people. We used bootstrap aggregation and cross-validated mixed-models to estimate the distribution of weights applied to cognitive and functional outcomes to form composites. We also evaluated best subset optimization. Using optimized composites, we estimated statistical power for a variety of clinical trial scenarios. Overall, 55.4% of cognitively-normal to MCI converters were Aβ+. Large gains in power estimates were obtained when requiring participants to have both subtle cognitive dysfunction and Aβ pathology compared with requiring Aβ pathology alone. Additional power resulted when including functional as well as cognitive outcomes as part of the composite. Composites formed by applying equal weights to all measures provided the highest estimates of cross-validated power, although similar to both continuous weight optimization and best subset optimization. Using a composite to detect a 30% slowing of decline, 80% power was obtained for predicted Aβ+ converters with 375 completers/arm for a 30-month trial using a combination of cognitive/ functional measures. In the Aβ+ group, power to approach levels suitable for a phase III clinical trial would require considerably larger sample sizes. Composites incorporating both cognitive and functional measures may substantially increase the power of a trial in a preclinical (Aβ+) AD population with subtle evidence of cognitive dysfunction.
|
|
