Objective
We investigated potential effects of increased African American participation in Alzheimer’s disease (AD) and mild cognitive impairment (MCI) clinical trials by examining differences in comorbid conditions and treatment outcome affecting trial design.
Methods
Using a meta-database of 18 studies from the Alzheimer’s Disease Cooperative Study and the Alzheimer’s Disease Neuroimaging Initiative, we included a cohort of 5,164 subjects for whom there were baseline demographic data and information on comorbid disorders, grouped by organ system. We used meta-analysis to compare prevalence of comorbidities, dropouts, and rates of change on the Alzheimer’s Disease Assessment Scale (ADAS-cog) by race. We also simulated clinical trial scenarios similar to recent therapeutic trials to determine effects of increased African American participation on statistical power.
Results
Approximately 7% of AD, 4% of MCI, and 11% of normal participants were African American. African American subjects had higher prevalence of cardiovascular disorders (odds ratio [OR] 2.10, 95% confidence interval [CI] 1.71, 2.57) and higher rate of dropouts (OR 1.60, 95% CI 1.15, 2.21) compared to whites, but lower rates of other disorders. There were no significant differences in rate of progression (−0.862 points/year, 95% CI −1.89, 0.162) by race, and little effect on power in simulated trials with sample sizes similar to current AD trial designs.
Conclusions
Increasing African American participation in AD clinical trials will require adaptation of trial protocols to address comorbidities and dropouts. However, increased diversity is unlikely to negatively affect trial outcomes and should be encouraged to promote generalizability of trial results.