Individuals with MCI retain relatively high cognitive function, and slowing or arresting the disease in this population offers immense benefits. To explore sample size requirements for clinical trials aimed at this population, we examined enrichment strategies based on CSF and MRI biomarkers to identify MCI individuals who are most likely to experience decline over the course of a clinical trial, and examined the relative ability of subregional and whole brain volume MRI outcome measures to enable further sample size reductions. To assess the relative powers for outcome measures and enrichment choices, we performed statistical significance testing for multiple pair-wise comparisons of outcome measures for different enrichment strategies, and for multiple pair-wise comparisons of enrichment strategies for different outcome measures. There is, however, growing concern that by the time individuals experience noticeable cognitive impairment and brain atrophy, therapies may be too late to stop the neurodegenerative cascade. Thus, preventive trials focused on asymptomatic individuals with biomarker evidence of AD pathology �C and who therefore may be in a preclinical phase of the disorder �C are being considered. To determine the feasibility of such trials, we also assessed rate of clinical decline and regional brain atrophy in cognitively healthy individuals who are likely to be in a presymptomatic stage of AD, based on CSF biomarkers. We considered HCs with CSF evidence of both amyloid and tau pathology as those most likely at risk for developing AD since prior studies have shown that CSF Ab is associated with elevated Tulathromycin B entorhinal cortex atrophy rate and elevated clinical decline only in the presence of elevated CSF ptau. We calculated sample sizes based on the observed rates of change in the HC group that tested positive for both measures, relative to the control group of stable HCs who tested negative for CSF Ab. We used baseline MRI measures to stratify MCI participants into high and low risk groups, as previously described in detail. Briefly, in prior work, we performed a discriminant analysis using cortical and subcortical ROIs to differentiate ADNI’s HC from AD participants. We then applied the resulting model, which incorporated measures of atrophy from medial and lateral temporal areas, retrosplenial cortex, and orbitofrontal areas) to MCI participants, classifying them into those whose atrophy 
in these regions more strongly resembled that found in the AD group or that found in the HC group. Methodological bias in image registration, leading to artifactually elevated effect sizes and reduced sample size estimates, remains a concern in the structural neuroimaging literature, especially given recent reporting on earlier methodology and results known to be strongly biased, and recent reports citing follow-up methodology and results that are ostensibly LOUREIRIN-B corrected for bias but in fact, as shown in, remain significantly biased. Several robust approaches to reducing or eliminating bias have been developed. Our explicitly inverseconsistent approach essentially eliminates potential bias by combining forward and reverse image registrations, and has been assessed vis-a`-vis other approaches. Here we show that stratifying MCI participants into dichotomized categories with respect to established AD biomarkers results in subgroups of participants with different rates of clinical decline and brain atrophy, and correspondingly different potentially treatable effect sizes that can be leveraged to increase the efficiency of clinical trials. We further show that power for detecting change due to disease progression varies by outcome measure, so that the most powerful outcome measure-enrichment strategy combination dramatically enhances the ability to detect therapeutic effects of investigational disease-altering treatments.