The goal of this proposal, entitled “Metformin in Alzheimer’s dementia Prevention (MAP)”, is to conduct a phase II/III randomized controlled trial of metformin vs. placebo in the prevention of cognitive decline among persons at risk for Alzheimer’s dementia, in response to PAR-18-028 “Phase III Clinical Trials for the Spectrum of Alzheimer’s Disease and Age-related Cognitive Decline”. This MAP proposal is based on the results of a single site 12-month early Phase II placebo controlled 1:1 randomized trial of short acting metformin 1000 mg twice a day among 80 persons with late amnestic mild cognitive impairment (aMCI), without diabetes, who were overweight or obese (clinicaltrials.gov ID NCT00620191; R01AG026413; ADDF 27091), conducted with the goal of obtaining preliminary data on safety, feasibility, and efficacy. This pilot study showed that the improvement in one of the primary outcomes, changes in total recall in the selective reminding test was significantly greater in the metformin arm compared with the placebo arm (9.7 ± 8.5 vs. 5.3 ± 8.5 words; p = 0.02), and this benefit was highest in those taking the highest dose of metformin. We propose MAP as a follow-up multisite phase II/III 1:1 randomized placebo-controlled trial to test the efficacy of metformin in the prevention of the cognitive decline associated with Alzheimer’s dementia. MAP will have the following innovations compared with the previous pilot study: (a) use of the better tolerated long acting form of metformin (Glucophage XR©), with a maximum dose of 2,000 mg a day; b) exclusion of participants who cannot tolerate at least 1000 mg a day of metformin; (c) repeated cognitive testing during the metformin titration period to account for practice effects; (d) Extension of the definition of aMCI to include early aMCI; (e) Extension of the trial duration from 12 months to 24 months and the cognitive testing interval from 3 months to 6 months; (f) use of the Total Recall Score of the Free and Cued Selective Reminding Test (FC-SRT) as a primary cognitive outcome and the Preclinical Alzheimer Cognitive Composite from the Alzheimer’s Disease Cooperative Study (ADCS-PACC) as a secondary cognitive outcome; (g) include hippocampal volume as a biomarker of neurodegeneration, and white matter hyperintensities (WMH) as a marker of cerebrovascular disease ascertained on brain magnetic resonance imaging (MRI) as secondary subclinical outcomes in half of the sample; (h) a proposed initial sample size of 370 (185 per arm), extending to a larger phase III trial based on a priori defined rules. Our primary aim is to compare changes from baseline to 24 months in verbal memory performance, measured with the Total Recall Score of the FC-SRT, between the metformin and placebo arms, following an ITT approach. We will examine global cognitive performance, measured with the ADCS-PACC, as a secondary outcome. We will also examine APOE-e4 genotype as a modifier of the efficacy of metformin. Our secondary a...