Cleavage of the amyloid precursor protein, an essential element of the cell membrane, results in amyloid-beta (Aβ) peptides leaving the cell membrane. Some peptides are cleared by the cerebrospinal fluid (CSF), and others by the vascular system (Brothers et al., 2018). Other peptides are more difficult to clear and have the capacity to change shape and interact with similar molecules prior to self-aggregating into long fibrils that form β-sheets (Brothers et al., 2018, Chen and Mobley, 2019, Hensley et al., 1994). The progression of insoluble fibrillar Aβ plaque deposition in the brain has been linked to aging and increasing severity of Alzheimer’s disease (AD) (Braak & Braak, 1991). Thus, detection of amyloid in aging adults has become a paramount focus of AD research and amyloid is the target of many drug therapeutics (Brothers et al., 2018).

Soluble and insoluble Aβ burden can be measured in vivo in CSF. Insoluble Aβ can also be measured using radioactive ligands, such as the [18F]-labeled florbetapir ligand, with positron emission tomography (PET) (, & For the Alzheimer’s Disease Neuroimaging Initiative, 2013, , & for the Alzheimer’s Disease Neuroimaging Initiative*, 2015). Using established florbetapir PET standardized uptake values (SUVR) thresholds, the presence of Aβ can be categorized as amyloid negative (A-) or amyloid positive (A+) (Jack et al., 2018).

Studies have begun investigating how AD biomarkers and cognition are modified in people with changing Aβ burden (Dubois et al., 2018, Harrison et al., 2021, For the Alzheimer’s Disease Neuroimaging Initiative, 2018, for the A4 Study Team, 2020, van der Kall et al., 2021). In the present study, we sought to longitudinally examine participants in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) who underwent a conversion in amyloid status from negative (A-) to positive (A+) in comparison to a group of participants who remained A- over the same follow-up period. Conversion from A- to A+ was based upon established florbetapir PET SUVR thresholds (Jagust et al., 2015, , & For the Alzheimer’s Disease Neuroimaging Initiative, 2013;). Our hypothesis was that the amyloid converter group would show an increased frequency of apolipoprotein E (APOE) ε4 alleles, greater annualized percent volume loss in selected magnetic resonance imaging (MRI) regions, and lower CSF concentration of the amyloid beta 1-42 peptide (Aβ1-42), in comparison to the non-converter group. Based on previous studies in the literature, we did not expect to see differences in CSF measures of p-tau or cognitive performance (Gordon et al., 2018, Harrison et al., 2021, Ossenkoppele et al., 2019).