Diagnosis-independent Alzheimer disease biomarker signature in cognitively normal elderly people

Abstract

Objective: To identify biomarker patterns typical for Alzheimer disease (AD) in an independent, unsupervised way, without using information on the clinical diagnosis.

Design: Mixture modeling approach.

Setting: Alzheimer's Disease Neuroimaging Initiative database.

Patients or other participants: Cognitively normal persons, patients with AD, and individuals with mild cognitive impairment.

Main outcome measures: Cerebrospinal fluid-derived beta-amyloid protein 1-42, total tau protein, and phosphorylated tau(181P) protein concentrations were used as biomarkers on a clinically well-characterized data set. The outcome of the qualification analysis was validated on 2 additional data sets, 1 of which was autopsy confirmed.

Results: Using the US Alzheimer's Disease Neuroimaging Initiative data set, a cerebrospinal fluid beta-amyloid protein 1-42/phosphorylated tau(181P) biomarker mixture model identified 1 feature linked to AD, while the other matched the "healthy" status. The AD signature was found in 90%, 72%, and 36% of patients in the AD, mild cognitive impairment, and cognitively normal groups, respectively. The cognitively normal group with the AD signature was enriched in apolipoprotein E epsilon4 allele carriers. Results were validated on 2 other data sets. In 1 study consisting of 68 autopsy-confirmed AD cases, 64 of 68 patients (94% sensitivity) were correctly classified with the AD feature. In another data set with patients (n = 57) with mild cognitive impairment followed up for 5 years, the model showed a sensitivity of 100% in patients progressing to AD.

Conclusions: The mixture modeling approach, totally independent of clinical AD diagnosis, correctly classified patients with AD. The unexpected presence of the AD signature in more than one-third of cognitively normal subjects suggests that AD pathology is active and detectable earlier than has heretofore been envisioned.

Figure 1

Mixture model classification for cerebrospinal fluid–derived β-amyloid protein 1–42 (CSF Aβ1–42). Results are presented as a histogram of observed counts overlaid with the 2 mixture distributions and the joined distribution based on the mixture proportion (A). The mixture distributions and their overlap is also shown (B).

Figure 2

Cerebrospinal fluid–derived β-amyloid protein 1–42 (CSF Aβ1–42) mixture model applied to the clinically diagnosed subject groups. AD indicates Alzheimer disease; MCI, mild cognitive impairment.

Figure 3

Receiver operating characteristic curve analysis for cerebrospinal fluid–derived β-amyloid protein 1–42 (CSF Aβ1–42)–based identification of Alzheimer disease (AD). Sensitivity (“true-positive rate”) was assessed in the AD group, and specificity (“false-positive rate”) was assessed in the normal group. AUC indicates area under the curve. Numbers with arrows indicate optimal decision boundaries (1) obtained with classic receiver operating characteristic analysis and minimizing the difference between sensitivity and specificity (159 pg/mL) and (2) between 2 components identified by unsupervised mixture modeling (188 pg/mL).

Figure 4

A combined cerebrospinal fluid–derived β-amyloid protein 1–42 (CSF Aβ1–42)/CSF phosphorylated tau_181P (CSF P-Tau_181P) mixture model applied to the subject groups. Densities of each signature are represented with confidence ellipses, and signature membership of the subject based on the mixture is indicated with the corresponding color (signature 1 is the Alzheimer disease [AD] signature [red]; signature 2 is the healthy signature [green]). MCI indicates mild cognitive impairment.

Figure 5

Validation of the combined cerebrospinal fluid–derived β-amyloid protein 1–42 (CSF Aβ1–42)/CSF phosphorylated tau_181P (CSF P-Tau_181P) mixture model in 2 data sets. A, Patients with mild cognitive impairment who developed Alzheimer disease within 5 years after the CSF sample. B, Patients with autopsy-confirmed Alzheimer disease with mostly less than 1 year between CSF sample and autopsy (n = 68). Signature 1 is the Alzheimer disease signature (red); signature 2 is the healthy signature (green).