The preclinical phase of Alzheimer's disease represents a crucial time window for therapeutic intervention but requires the identification of clinically relevant biomarkers that are sensitive to the effects of disease-modifying drugs. Amyloid peptide and tau proteins, the main histological hallmarks of Alzheimer's disease, have been widely used as biomarkers of anti-amyloid and anti-tau drugs. However, these biomarkers do not fully capture the multiple biological pathways of the brain. Indeed, robust amyloid-target engagement by anti-amyloid monoclonal antibodies has recently translated into modest cognitive and clinical benefits in Alzheimer's disease patients, albeit with potentially life-threatening side effects. Moreover, targeting the tau pathway has yet to result in any positive clinical outcomes. Findings from computational neuroscience have demonstrated that brain regions do not work in isolation but are interconnected within complex network structures. Brain connectivity studies suggest that misfolded proteins can spread through these connections, leading to the hypothesis that Alzheimer's disease is a pathology of network disconnectivity. Based on these assumptions, here we discuss how incorporating brain connectivity outcomes could better capture global brain functionality and, in conjunction with traditional Alzheimer's disease biomarkers, could facilitate the clinical development of new disease-modifying anti-Alzheimer's disease drugs.