Objective: Early detection of Alzheimer's Disease (AD) is vital, as earlier interventions may prove more efficacious in preventing the devastating effects of the disorder. An array of cognitive symptoms has been associated with preclinical stages of AD in cross-sectional research. The purpose of this study was to examine longitudinal changes in cognition and the degree to which genetic risk for AD explains heterogeneity in neurocognitive change within non-demented older adults via Latent Growth Curve Modeling. Method: Data was collected from 398 relatively healthy older adults (Average Age = 68.62, SD = 6.45; Average Education = 16 years; 94.8% Caucasian; 67.5% female) that were cognitively intact (Mini-Mental Status Exam > 25) at baseline. DNA was extracted from blood samples (96 apolipoprotein E: APOE-e4 carriers vs. 302 Non-carriers). Three consecutive years of neuropsychological test scores from the National Alzheimer's Coordinating Center's Uniform Data Set served as indicator measures for executive attention/language, memory, and working memory latent variables, while age, sex, and APOE genotype were predictor variables. Results: The model was a good fit (CFI = 941; RMSEA = .056). Greater age associated with lower scores across neurocognitive domains at baseline, and a marginal yet significant rate of change in working memory (−.058), ps < .05. Cognitive profiles significantly differed between sexes. Genetic risk (APOE-e4) was a significant predictor of memory at baseline and greater executive attention/language decline (-.340), ps < .05. Conclusion: These findings provide support for a preclinical cognitive endophenotype for AD that in conjunction with genetic risk may serve as sensitive preclinical predictors of AD.