Abstract Alzheimer's disease (AD) dementia is a major burden on the healthcare system and a major barrier to human longevity, especially productive human longevity. Genetics of familial early-onset AD, pathological changes in the brain of sporadic late-onset AD patients and pathological changes in the brain of individuals with Down syndrome, who have increased expression of amyloid precursor protein (APP) owing to partial trisomy of chromosome 21, all point to APP and its cleavage product amyloid β (Aβ) peptide as causative factors in the pathogenesis of AD. Adenosine triphosphate-binding cassette transporter subfamily A member 7 (ABCA7) is also a major factor in AD pathogenesis that bears on Aβ metabolism. Unfortunately, the mechanisms of its effect on Aβ turnover and the disease progression have been uncovered insufficiently. One reason for this is that the currently commonly-used Abca7 deletion mouse model have yielded divergent and contradictory findings and may not be suitable for the discovery of ABCA7 functions. We summarize the finding made with the model and point to the weaknesses of this model. Given this, we propose to make a tissue-specific ABCA7 overexpression mouse using the Cre/loxP system. An overexpression model will have several advantages as an ABCA7 investigation tool over the deletion model. It is then proposed to use the overexpression model to test a hypothesis about the role of ABCA7 in astrocytes, in particular that astrocyte ABCA7 drives production of apolipoprotein E-lipoprotein and thus promotes efflux of the neurodegenerative lipid from the cells and clearance of Aβ from the brain. We have recently proposed the existence of the neurodegenerative lipid generated during normal functioning of neural tissue. The proposed model will be valuable for the studies of ABCA7 in the neural tissues, while the proposed experiment will advance our understanding ABCA7 functions.