Alzheimer's disease (AD) might be treated with symptomatic, neuroprotective, or neurorestorative therapies. Neuroprotective and neurorestorative interventions are disease-modifying therapies. Disease modification can be defined as treatments or interventions that affect the underlying pathophysiology of the disease and have a beneficial outcome on the course of AD. In a clinical trial the criteria for affecting the underlying cause of the disease can be supported by demonstrating an effect on a biomarker such as medial temporal atrophy on magnetic resonance imaging (MRI) or diminished tau or phospho-tau levels in cerebrospinal fluid. The claim for a beneficial effect on the clinical course of AD is supported by a drug-placebo difference on the primary clinical outcomes of the clinical trial. A statistically significant correlation between the biomarker outcome and the clinical trial outcome would support the claim that these are based on the same underlying mechanism. Delayed start or staggered withdrawal designs might in themselves support a disease-modifying claim but are difficult to implement. A combination of clinical outcomes and biomarker measures is a more likely pathway to a disease-modifying claim. Labeling of disease-modifying agents might refer to slowing of disease progression, delay in reaching predefined disease milestones, or reduction in progression of a biomarker such as cerebral atrophy or ventricular enlargement on MRI. Prevention claims will depend heavily on biomarker outcomes.