Alzheimer’s disease is the most common form of dementia and is characterized by the neuropathological hallmarks amyloid-β and neurofibrillary tangles. Despite tremendous efforts in understanding the molecular, pathophysiological, and diagnostic criteria of Alzheimer’s disease in the past 20 years, all disease-modifying therapies to date have failed to effectively treat patients. As the neuropathological alterations associated with the disease are estimated to appear up to two decades prior to the onset of clinical symptoms, initiating treatment in the preclinical stage may be necessary to halt the Alzheimer’s disease biochemical cascade. While targeting the disease in the preclinical stage is not yet feasible in patients due to a lack of clear biomarkers, the use of preclinical animal models bypasses this limitation. Using a modified push-pull microdialysis methodology for the simultaneous infusion of disease-modifying drugs and collection of cerebrospinal fluid, we found that Fasudil reduces intracellular amyloid-β in young animals and amyloid plaques and cerebrospinal fluid amyloid-β in old animals, while Lonafarnib reduces tau neuropathology and cerebrospinal fluid tau biomarkers in young and old animals. However, an unexpected finding was that Lonafarnib treatment increased amyloid plaques and cerebrospinal fluid amyloid-β in old animals, suggesting that activating the endosomal-lysosomal system may inadvertently increase amyloid-β pathology if administered too late in the AD biochemical cascade. Taken together, these findings offer support for the application of repurposed drugs to attenuate Alzheimer’s disease neuropathology at various timepoints in preclinical models to probe potential biochemical events that result in Alzheimer’s disease.